From 2ffcb6fc50174d1efc8f98633eb2647d84483c68 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 7 Mar 2023 22:59:21 -0800 Subject: bpf: Refactor codes into bpf_local_storage_destroy This patch first renames bpf_local_storage_unlink_nolock to bpf_local_storage_destroy(). It better reflects that it is only used when the storage's owner (sk/task/cgrp/inode) is being kfree(). All bpf_local_storage_destroy's caller is taking the spin lock and then free the storage. This patch also moves these two steps into the bpf_local_storage_destroy. This is a preparation work for a later patch that uses bpf_mem_cache_alloc/free in the bpf_local_storage. Signed-off-by: Martin KaFai Lau Link: https://lore.kernel.org/r/20230308065936.1550103-3-martin.lau@linux.dev Signed-off-by: Alexei Starovoitov --- net/core/bpf_sk_storage.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'net') diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index 7a36353dbc22..8f56438c104b 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -49,7 +49,6 @@ static int bpf_sk_storage_del(struct sock *sk, struct bpf_map *map) void bpf_sk_storage_free(struct sock *sk) { struct bpf_local_storage *sk_storage; - bool free_sk_storage = false; rcu_read_lock(); sk_storage = rcu_dereference(sk->sk_bpf_storage); @@ -58,13 +57,8 @@ void bpf_sk_storage_free(struct sock *sk) return; } - raw_spin_lock_bh(&sk_storage->lock); - free_sk_storage = bpf_local_storage_unlink_nolock(sk_storage); - raw_spin_unlock_bh(&sk_storage->lock); + bpf_local_storage_destroy(sk_storage); rcu_read_unlock(); - - if (free_sk_storage) - kfree_rcu(sk_storage, rcu); } static void bpf_sk_storage_map_free(struct bpf_map *map) -- cgit v1.2.3 From a47eabf216f77cb6f22ceb38d46f1bb95968579c Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 7 Mar 2023 22:59:25 -0800 Subject: bpf: Repurpose use_trace_rcu to reuse_now in bpf_local_storage This patch re-purpose the use_trace_rcu to mean if the freed memory can be reused immediately or not. The use_trace_rcu is renamed to reuse_now. Other than the boolean test is reversed, it should be a no-op. The following explains the reason for the rename and how it will be used in a later patch. In a later patch, bpf_mem_cache_alloc/free will be used in the bpf_local_storage. The bpf mem allocator will reuse the freed memory immediately. Some of the free paths in bpf_local_storage does not support memory to be reused immediately. These paths are the "delete" elem cases from the bpf_*_storage_delete() helper and the map_delete_elem() syscall. Note that "delete" elem before the owner's (sk/task/cgrp/inode) lifetime ended is not the common usage for the local storage. The common free path, bpf_local_storage_destroy(), can reuse the memory immediately. This common path means the storage stays with its owner until the owner is destroyed. The above mentioned "delete" elem paths that cannot reuse immediately always has the 'use_trace_rcu == true'. The cases that is safe for immediate reuse always have 'use_trace_rcu == false'. Instead of adding another arg in a later patch, this patch re-purpose this arg to reuse_now and have the test logic reversed. In a later patch, 'reuse_now == true' will free to the bpf_mem_cache_free() where the memory can be reused immediately. 'reuse_now == false' will go through the call_rcu_tasks_trace(). Signed-off-by: Martin KaFai Lau Link: https://lore.kernel.org/r/20230308065936.1550103-7-martin.lau@linux.dev Signed-off-by: Alexei Starovoitov --- include/linux/bpf_local_storage.h | 2 +- kernel/bpf/bpf_cgrp_storage.c | 2 +- kernel/bpf/bpf_inode_storage.c | 2 +- kernel/bpf/bpf_local_storage.c | 24 ++++++++++++------------ kernel/bpf/bpf_task_storage.c | 2 +- net/core/bpf_sk_storage.c | 2 +- 6 files changed, 17 insertions(+), 17 deletions(-) (limited to 'net') diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h index 613b1805ed9f..18a31add2255 100644 --- a/include/linux/bpf_local_storage.h +++ b/include/linux/bpf_local_storage.h @@ -143,7 +143,7 @@ int bpf_local_storage_map_check_btf(const struct bpf_map *map, void bpf_selem_link_storage_nolock(struct bpf_local_storage *local_storage, struct bpf_local_storage_elem *selem); -void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool use_trace_rcu); +void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now); void bpf_selem_link_map(struct bpf_local_storage_map *smap, struct bpf_local_storage_elem *selem); diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c index 492594d69a86..c975cacdd16b 100644 --- a/kernel/bpf/bpf_cgrp_storage.c +++ b/kernel/bpf/bpf_cgrp_storage.c @@ -121,7 +121,7 @@ static int cgroup_storage_delete(struct cgroup *cgroup, struct bpf_map *map) if (!sdata) return -ENOENT; - bpf_selem_unlink(SELEM(sdata), true); + bpf_selem_unlink(SELEM(sdata), false); return 0; } diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c index 2d25bcfa371b..ad2ab0187e45 100644 --- a/kernel/bpf/bpf_inode_storage.c +++ b/kernel/bpf/bpf_inode_storage.c @@ -122,7 +122,7 @@ static int inode_storage_delete(struct inode *inode, struct bpf_map *map) if (!sdata) return -ENOENT; - bpf_selem_unlink(SELEM(sdata), true); + bpf_selem_unlink(SELEM(sdata), false); return 0; } diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c index 5585dbfd9c66..70c34a948c3c 100644 --- a/kernel/bpf/bpf_local_storage.c +++ b/kernel/bpf/bpf_local_storage.c @@ -147,7 +147,7 @@ static void bpf_selem_free_trace_rcu(struct rcu_head *rcu) */ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_storage, struct bpf_local_storage_elem *selem, - bool uncharge_mem, bool use_trace_rcu) + bool uncharge_mem, bool reuse_now) { struct bpf_local_storage_map *smap; bool free_local_storage; @@ -201,7 +201,7 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor * any special fields. */ rec = smap->map.record; - if (use_trace_rcu) { + if (!reuse_now) { if (!IS_ERR_OR_NULL(rec)) call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_fields_trace_rcu); else @@ -220,7 +220,7 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor } static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem, - bool use_trace_rcu) + bool reuse_now) { struct bpf_local_storage *local_storage; bool free_local_storage = false; @@ -235,11 +235,11 @@ static void bpf_selem_unlink_storage(struct bpf_local_storage_elem *selem, raw_spin_lock_irqsave(&local_storage->lock, flags); if (likely(selem_linked_to_storage(selem))) free_local_storage = bpf_selem_unlink_storage_nolock( - local_storage, selem, true, use_trace_rcu); + local_storage, selem, true, reuse_now); raw_spin_unlock_irqrestore(&local_storage->lock, flags); if (free_local_storage) { - if (use_trace_rcu) + if (!reuse_now) call_rcu_tasks_trace(&local_storage->rcu, bpf_local_storage_free_rcu); else @@ -284,14 +284,14 @@ void bpf_selem_link_map(struct bpf_local_storage_map *smap, raw_spin_unlock_irqrestore(&b->lock, flags); } -void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool use_trace_rcu) +void bpf_selem_unlink(struct bpf_local_storage_elem *selem, bool reuse_now) { /* Always unlink from map before unlinking from local_storage * because selem will be freed after successfully unlinked from * the local_storage. */ bpf_selem_unlink_map(selem); - bpf_selem_unlink_storage(selem, use_trace_rcu); + bpf_selem_unlink_storage(selem, reuse_now); } /* If cacheit_lockit is false, this lookup function is lockless */ @@ -538,7 +538,7 @@ bpf_local_storage_update(void *owner, struct bpf_local_storage_map *smap, if (old_sdata) { bpf_selem_unlink_map(SELEM(old_sdata)); bpf_selem_unlink_storage_nolock(local_storage, SELEM(old_sdata), - false, true); + false, false); } unlock: @@ -651,7 +651,7 @@ void bpf_local_storage_destroy(struct bpf_local_storage *local_storage) * of the loop will set the free_cgroup_storage to true. */ free_storage = bpf_selem_unlink_storage_nolock( - local_storage, selem, false, false); + local_storage, selem, false, true); } raw_spin_unlock_irqrestore(&local_storage->lock, flags); @@ -745,7 +745,7 @@ void bpf_local_storage_map_free(struct bpf_map *map, migrate_disable(); this_cpu_inc(*busy_counter); } - bpf_selem_unlink(selem, false); + bpf_selem_unlink(selem, true); if (busy_counter) { this_cpu_dec(*busy_counter); migrate_enable(); @@ -783,8 +783,8 @@ void bpf_local_storage_map_free(struct bpf_map *map, /* We cannot skip rcu_barrier() when rcu_trace_implies_rcu_gp() * is true, because while call_rcu invocation is skipped in that * case in bpf_selem_free_fields_trace_rcu (and all local - * storage maps pass use_trace_rcu = true), there can be - * call_rcu callbacks based on use_trace_rcu = false in the + * storage maps pass reuse_now = false), there can be + * call_rcu callbacks based on reuse_now = true in the * while ((selem = ...)) loop above or when owner's free path * calls bpf_local_storage_unlink_nolock. */ diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c index 4dcef28744d1..c88cc04c17c1 100644 --- a/kernel/bpf/bpf_task_storage.c +++ b/kernel/bpf/bpf_task_storage.c @@ -168,7 +168,7 @@ static int task_storage_delete(struct task_struct *task, struct bpf_map *map, if (!nobusy) return -EBUSY; - bpf_selem_unlink(SELEM(sdata), true); + bpf_selem_unlink(SELEM(sdata), false); return 0; } diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index 8f56438c104b..a5f185b8e50a 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -40,7 +40,7 @@ static int bpf_sk_storage_del(struct sock *sk, struct bpf_map *map) if (!sdata) return -ENOENT; - bpf_selem_unlink(SELEM(sdata), true); + bpf_selem_unlink(SELEM(sdata), false); return 0; } -- cgit v1.2.3 From c0d63f309186d8492577c67c67984c714b6b72bc Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Tue, 7 Mar 2023 22:59:28 -0800 Subject: bpf: Add bpf_selem_free() This patch refactors the selem freeing logic into bpf_selem_free(). It is a preparation work for a later patch using bpf_mem_cache_alloc/free. The other kfree(selem) cases are also changed to bpf_selem_free(..., reuse_now = true). Signed-off-by: Martin KaFai Lau Link: https://lore.kernel.org/r/20230308065936.1550103-10-martin.lau@linux.dev Signed-off-by: Alexei Starovoitov --- include/linux/bpf_local_storage.h | 4 ++++ kernel/bpf/bpf_local_storage.c | 21 ++++++++++++++------- net/core/bpf_sk_storage.c | 2 +- 3 files changed, 19 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h index 18a31add2255..a34f61467a2f 100644 --- a/include/linux/bpf_local_storage.h +++ b/include/linux/bpf_local_storage.h @@ -152,6 +152,10 @@ struct bpf_local_storage_elem * bpf_selem_alloc(struct bpf_local_storage_map *smap, void *owner, void *value, bool charge_mem, gfp_t gfp_flags); +void bpf_selem_free(struct bpf_local_storage_elem *selem, + struct bpf_local_storage_map *smap, + bool reuse_now); + int bpf_local_storage_alloc(void *owner, struct bpf_local_storage_map *smap, diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c index 146e9caeda96..512943aac435 100644 --- a/kernel/bpf/bpf_local_storage.c +++ b/kernel/bpf/bpf_local_storage.c @@ -125,6 +125,17 @@ static void bpf_selem_free_trace_rcu(struct rcu_head *rcu) call_rcu(rcu, bpf_selem_free_rcu); } +void bpf_selem_free(struct bpf_local_storage_elem *selem, + struct bpf_local_storage_map *smap, + bool reuse_now) +{ + bpf_obj_free_fields(smap->map.record, SDATA(selem)->data); + if (!reuse_now) + call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu); + else + call_rcu(&selem->rcu, bpf_selem_free_rcu); +} + /* local_storage->lock must be held and selem->local_storage == local_storage. * The caller must ensure selem->smap is still valid to be * dereferenced for its smap->elem_size and smap->cache_idx. @@ -175,11 +186,7 @@ static bool bpf_selem_unlink_storage_nolock(struct bpf_local_storage *local_stor SDATA(selem)) RCU_INIT_POINTER(local_storage->cache[smap->cache_idx], NULL); - bpf_obj_free_fields(smap->map.record, SDATA(selem)->data); - if (!reuse_now) - call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu); - else - call_rcu(&selem->rcu, bpf_selem_free_rcu); + bpf_selem_free(selem, smap, reuse_now); if (rcu_access_pointer(local_storage->smap) == smap) RCU_INIT_POINTER(local_storage->smap, NULL); @@ -423,7 +430,7 @@ bpf_local_storage_update(void *owner, struct bpf_local_storage_map *smap, err = bpf_local_storage_alloc(owner, smap, selem, gfp_flags); if (err) { - kfree(selem); + bpf_selem_free(selem, smap, true); mem_uncharge(smap, owner, smap->elem_size); return ERR_PTR(err); } @@ -517,7 +524,7 @@ unlock_err: raw_spin_unlock_irqrestore(&local_storage->lock, flags); if (selem) { mem_uncharge(smap, owner, smap->elem_size); - kfree(selem); + bpf_selem_free(selem, smap, true); } return ERR_PTR(err); } diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index a5f185b8e50a..24c3dc0d62e5 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -197,7 +197,7 @@ int bpf_sk_storage_clone(const struct sock *sk, struct sock *newsk) } else { ret = bpf_local_storage_alloc(newsk, smap, copy_selem, GFP_ATOMIC); if (ret) { - kfree(copy_selem); + bpf_selem_free(copy_selem, smap, true); atomic_sub(smap->elem_size, &newsk->sk_omem_alloc); bpf_map_put(map); -- cgit v1.2.3 From 5fe5a7586c27a13a42b7946334d41a96c776a188 Mon Sep 17 00:00:00 2001 From: Palmer Dabbelt Date: Fri, 10 Mar 2023 17:07:53 +0100 Subject: Move COMPAT_ATM_ADDPARTY to net/atm/svc.c This used to be behind an #ifdef COMPAT_COMPAT, so most of userspace wouldn't have seen the definition before. Unfortunately this header file became visible to userspace, so the definition has instead been moved to net/atm/svc.c (the only user). Signed-off-by: Palmer Dabbelt Reviewed-by: Andrew Waterman Reviewed-by: Albert Ou Message-Id: <1447119071-19392-4-git-send-email-palmer@dabbelt.com> Reviewed-by: Arnd Bergmann Reviewed-by: Christoph Hellwig Signed-off-by: Thomas Huth Signed-off-by: Arnd Bergmann --- include/uapi/linux/atmdev.h | 4 ---- net/atm/svc.c | 5 +++++ 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/uapi/linux/atmdev.h b/include/uapi/linux/atmdev.h index a5c15cf23bd7..20b0215084fc 100644 --- a/include/uapi/linux/atmdev.h +++ b/include/uapi/linux/atmdev.h @@ -101,10 +101,6 @@ struct atm_dev_stats { /* use backend to make new if */ #define ATM_ADDPARTY _IOW('a', ATMIOC_SPECIAL+4,struct atm_iobuf) /* add party to p2mp call */ -#ifdef CONFIG_COMPAT -/* It actually takes struct sockaddr_atmsvc, not struct atm_iobuf */ -#define COMPAT_ATM_ADDPARTY _IOW('a', ATMIOC_SPECIAL+4,struct compat_atm_iobuf) -#endif #define ATM_DROPPARTY _IOW('a', ATMIOC_SPECIAL+5,int) /* drop party from p2mp call */ diff --git a/net/atm/svc.c b/net/atm/svc.c index 4a02bcaad279..d83556d8beb9 100644 --- a/net/atm/svc.c +++ b/net/atm/svc.c @@ -28,6 +28,11 @@ #include "signaling.h" #include "addr.h" +#ifdef CONFIG_COMPAT +/* It actually takes struct sockaddr_atmsvc, not struct atm_iobuf */ +#define COMPAT_ATM_ADDPARTY _IOW('a', ATMIOC_SPECIAL + 4, struct compat_atm_iobuf) +#endif + static int svc_create(struct net *net, struct socket *sock, int protocol, int kern); -- cgit v1.2.3 From 5f24f41e8ea62a6a9095f9bbafb8b3aebe265c68 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 10 Mar 2023 17:26:05 +0800 Subject: xfrm: Remove inner/outer modes from input path The inner/outer modes were added to abstract out common code that were once duplicated between IPv4 and IPv6. As time went on the abstractions have been removed and we are now left with empty shells that only contain duplicate information. These can be removed one-by-one as the same information is already present elsewhere in the xfrm_state object. Removing them from the input path actually allows certain valid combinations that are currently disallowed. In particular, when a transport mode SA sits beneath a tunnel mode SA that changes address families, at present the transport mode SA cannot have AF_UNSPEC as its selector because it will be erroneously be treated as inter-family itself even though it simply sits beneath one. This is a serious problem because you can't set the selector to non-AF_UNSPEC either as that will cause the selector match to fail as we always match selectors to the inner-most traffic. Signed-off-by: Herbert Xu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_input.c | 66 ++++++++++++++++++--------------------------------- 1 file changed, 23 insertions(+), 43 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index 436d29640ac2..39fb91ff23d9 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -231,9 +231,6 @@ static int xfrm4_remove_tunnel_encap(struct xfrm_state *x, struct sk_buff *skb) { int err = -EINVAL; - if (XFRM_MODE_SKB_CB(skb)->protocol != IPPROTO_IPIP) - goto out; - if (!pskb_may_pull(skb, sizeof(struct iphdr))) goto out; @@ -269,8 +266,6 @@ static int xfrm6_remove_tunnel_encap(struct xfrm_state *x, struct sk_buff *skb) { int err = -EINVAL; - if (XFRM_MODE_SKB_CB(skb)->protocol != IPPROTO_IPV6) - goto out; if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) goto out; @@ -331,22 +326,26 @@ out: */ static int xfrm_inner_mode_encap_remove(struct xfrm_state *x, - const struct xfrm_mode *inner_mode, struct sk_buff *skb) { - switch (inner_mode->encap) { + switch (x->props.mode) { case XFRM_MODE_BEET: - if (inner_mode->family == AF_INET) + switch (XFRM_MODE_SKB_CB(skb)->protocol) { + case IPPROTO_IPIP: + case IPPROTO_BEETPH: return xfrm4_remove_beet_encap(x, skb); - if (inner_mode->family == AF_INET6) + case IPPROTO_IPV6: return xfrm6_remove_beet_encap(x, skb); + } break; case XFRM_MODE_TUNNEL: - if (inner_mode->family == AF_INET) + switch (XFRM_MODE_SKB_CB(skb)->protocol) { + case IPPROTO_IPIP: return xfrm4_remove_tunnel_encap(x, skb); - if (inner_mode->family == AF_INET6) + case IPPROTO_IPV6: return xfrm6_remove_tunnel_encap(x, skb); break; + } } WARN_ON_ONCE(1); @@ -355,9 +354,7 @@ xfrm_inner_mode_encap_remove(struct xfrm_state *x, static int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb) { - const struct xfrm_mode *inner_mode = &x->inner_mode; - - switch (x->outer_mode.family) { + switch (x->props.family) { case AF_INET: xfrm4_extract_header(skb); break; @@ -369,17 +366,12 @@ static int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb) return -EAFNOSUPPORT; } - if (x->sel.family == AF_UNSPEC) { - inner_mode = xfrm_ip2inner_mode(x, XFRM_MODE_SKB_CB(skb)->protocol); - if (!inner_mode) - return -EAFNOSUPPORT; - } - - switch (inner_mode->family) { - case AF_INET: + switch (XFRM_MODE_SKB_CB(skb)->protocol) { + case IPPROTO_IPIP: + case IPPROTO_BEETPH: skb->protocol = htons(ETH_P_IP); break; - case AF_INET6: + case IPPROTO_IPV6: skb->protocol = htons(ETH_P_IPV6); break; default: @@ -387,7 +379,7 @@ static int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb) break; } - return xfrm_inner_mode_encap_remove(x, inner_mode, skb); + return xfrm_inner_mode_encap_remove(x, skb); } /* Remove encapsulation header. @@ -433,17 +425,16 @@ static int xfrm6_transport_input(struct xfrm_state *x, struct sk_buff *skb) } static int xfrm_inner_mode_input(struct xfrm_state *x, - const struct xfrm_mode *inner_mode, struct sk_buff *skb) { - switch (inner_mode->encap) { + switch (x->props.mode) { case XFRM_MODE_BEET: case XFRM_MODE_TUNNEL: return xfrm_prepare_input(x, skb); case XFRM_MODE_TRANSPORT: - if (inner_mode->family == AF_INET) + if (x->props.family == AF_INET) return xfrm4_transport_input(x, skb); - if (inner_mode->family == AF_INET6) + if (x->props.family == AF_INET6) return xfrm6_transport_input(x, skb); break; case XFRM_MODE_ROUTEOPTIMIZATION: @@ -461,7 +452,6 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) { const struct xfrm_state_afinfo *afinfo; struct net *net = dev_net(skb->dev); - const struct xfrm_mode *inner_mode; int err; __be32 seq; __be32 seq_hi; @@ -491,7 +481,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) goto drop; } - family = x->outer_mode.family; + family = x->props.family; /* An encap_type of -1 indicates async resumption. */ if (encap_type == -1) { @@ -676,17 +666,7 @@ resume: XFRM_MODE_SKB_CB(skb)->protocol = nexthdr; - inner_mode = &x->inner_mode; - - if (x->sel.family == AF_UNSPEC) { - inner_mode = xfrm_ip2inner_mode(x, XFRM_MODE_SKB_CB(skb)->protocol); - if (inner_mode == NULL) { - XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEMODEERROR); - goto drop; - } - } - - if (xfrm_inner_mode_input(x, inner_mode, skb)) { + if (xfrm_inner_mode_input(x, skb)) { XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEMODEERROR); goto drop; } @@ -701,7 +681,7 @@ resume: * transport mode so the outer address is identical. */ daddr = &x->id.daddr; - family = x->outer_mode.family; + family = x->props.family; err = xfrm_parse_spi(skb, nexthdr, &spi, &seq); if (err < 0) { @@ -732,7 +712,7 @@ resume: err = -EAFNOSUPPORT; rcu_read_lock(); - afinfo = xfrm_state_afinfo_get_rcu(x->inner_mode.family); + afinfo = xfrm_state_afinfo_get_rcu(x->props.family); if (likely(afinfo)) err = afinfo->transport_finish(skb, xfrm_gro || async); rcu_read_unlock(); -- cgit v1.2.3 From f4796398f21b9844017a2dac883b1dd6ad6edd60 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 10 Mar 2023 17:40:32 +0800 Subject: xfrm: Remove inner/outer modes from output path The inner/outer modes were added to abstract out common code that were once duplicated between IPv4 and IPv6. As time went on the abstractions have been removed and we are now left with empty shells that only contain duplicate information. These can be removed one-by-one as the same information is already present elsewhere in the xfrm_state object. Just like the input-side, removing this from the output code makes it possible to use transport-mode SAs underneath an inter-family tunnel mode SA. Signed-off-by: Herbert Xu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_output.c | 33 +++++++++++---------------------- 1 file changed, 11 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c index ff114d68cc43..369e5de8558f 100644 --- a/net/xfrm/xfrm_output.c +++ b/net/xfrm/xfrm_output.c @@ -412,7 +412,7 @@ static int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb) IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE; skb->protocol = htons(ETH_P_IP); - switch (x->outer_mode.encap) { + switch (x->props.mode) { case XFRM_MODE_BEET: return xfrm4_beet_encap_add(x, skb); case XFRM_MODE_TUNNEL: @@ -435,7 +435,7 @@ static int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb) skb->ignore_df = 1; skb->protocol = htons(ETH_P_IPV6); - switch (x->outer_mode.encap) { + switch (x->props.mode) { case XFRM_MODE_BEET: return xfrm6_beet_encap_add(x, skb); case XFRM_MODE_TUNNEL: @@ -451,22 +451,22 @@ static int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb) static int xfrm_outer_mode_output(struct xfrm_state *x, struct sk_buff *skb) { - switch (x->outer_mode.encap) { + switch (x->props.mode) { case XFRM_MODE_BEET: case XFRM_MODE_TUNNEL: - if (x->outer_mode.family == AF_INET) + if (x->props.family == AF_INET) return xfrm4_prepare_output(x, skb); - if (x->outer_mode.family == AF_INET6) + if (x->props.family == AF_INET6) return xfrm6_prepare_output(x, skb); break; case XFRM_MODE_TRANSPORT: - if (x->outer_mode.family == AF_INET) + if (x->props.family == AF_INET) return xfrm4_transport_output(x, skb); - if (x->outer_mode.family == AF_INET6) + if (x->props.family == AF_INET6) return xfrm6_transport_output(x, skb); break; case XFRM_MODE_ROUTEOPTIMIZATION: - if (x->outer_mode.family == AF_INET6) + if (x->props.family == AF_INET6) return xfrm6_ro_output(x, skb); WARN_ON_ONCE(1); break; @@ -875,21 +875,10 @@ static int xfrm6_extract_output(struct xfrm_state *x, struct sk_buff *skb) static int xfrm_inner_extract_output(struct xfrm_state *x, struct sk_buff *skb) { - const struct xfrm_mode *inner_mode; - - if (x->sel.family == AF_UNSPEC) - inner_mode = xfrm_ip2inner_mode(x, - xfrm_af2proto(skb_dst(skb)->ops->family)); - else - inner_mode = &x->inner_mode; - - if (inner_mode == NULL) - return -EAFNOSUPPORT; - - switch (inner_mode->family) { - case AF_INET: + switch (skb->protocol) { + case htons(ETH_P_IP): return xfrm4_extract_output(x, skb); - case AF_INET6: + case htons(ETH_P_IPV6): return xfrm6_extract_output(x, skb); } -- cgit v1.2.3 From 9c94bbf9a87b264294f42e6cc0f76d87854733ec Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Mon, 13 Mar 2023 22:55:52 +0100 Subject: xdp: recycle Page Pool backed skbs built from XDP frames __xdp_build_skb_from_frame() state(d): /* Until page_pool get SKB return path, release DMA here */ Page Pool got skb pages recycling in April 2021, but missed this function. xdp_release_frame() is relevant only for Page Pool backed frames and it detaches the page from the corresponding page_pool in order to make it freeable via page_frag_free(). It can instead just mark the output skb as eligible for recycling if the frame is backed by a pp. No change for other memory model types (the same condition check as before). cpumap redirect and veth on Page Pool drivers now become zero-alloc (or almost). Signed-off-by: Alexander Lobakin Link: https://lore.kernel.org/r/20230313215553.1045175-4-aleksander.lobakin@intel.com Signed-off-by: Alexei Starovoitov --- net/core/xdp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/xdp.c b/net/core/xdp.c index 8c92fc553317..a2237cfca8e9 100644 --- a/net/core/xdp.c +++ b/net/core/xdp.c @@ -658,8 +658,8 @@ struct sk_buff *__xdp_build_skb_from_frame(struct xdp_frame *xdpf, * - RX ring dev queue index (skb_record_rx_queue) */ - /* Until page_pool get SKB return path, release DMA here */ - xdp_release_frame(xdpf); + if (xdpf->mem.type == MEM_TYPE_PAGE_POOL) + skb_mark_for_recycle(skb); /* Allow SKB to reuse area used by xdp_frame */ xdp_scrub_frame(xdpf); -- cgit v1.2.3 From d4e492338d11937c55841b1279287280d6e35894 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Mon, 13 Mar 2023 22:55:53 +0100 Subject: xdp: remove unused {__,}xdp_release_frame() __xdp_build_skb_from_frame() was the last user of {__,}xdp_release_frame(), which detaches pages from the page_pool. All the consumers now recycle Page Pool skbs and page, except mlx5, stmmac and tsnep drivers, which use page_pool_release_page() directly (might change one day). It's safe to assume this functionality is not needed anymore and can be removed (in favor of recycling). Signed-off-by: Alexander Lobakin Link: https://lore.kernel.org/r/20230313215553.1045175-5-aleksander.lobakin@intel.com Signed-off-by: Alexei Starovoitov --- include/net/xdp.h | 29 ----------------------------- net/core/xdp.c | 15 --------------- 2 files changed, 44 deletions(-) (limited to 'net') diff --git a/include/net/xdp.h b/include/net/xdp.h index d517bfac937b..5393b3ebe56e 100644 --- a/include/net/xdp.h +++ b/include/net/xdp.h @@ -317,35 +317,6 @@ void xdp_flush_frame_bulk(struct xdp_frame_bulk *bq); void xdp_return_frame_bulk(struct xdp_frame *xdpf, struct xdp_frame_bulk *bq); -/* When sending xdp_frame into the network stack, then there is no - * return point callback, which is needed to release e.g. DMA-mapping - * resources with page_pool. Thus, have explicit function to release - * frame resources. - */ -void __xdp_release_frame(void *data, struct xdp_mem_info *mem); -static inline void xdp_release_frame(struct xdp_frame *xdpf) -{ - struct xdp_mem_info *mem = &xdpf->mem; - struct skb_shared_info *sinfo; - int i; - - /* Curr only page_pool needs this */ - if (mem->type != MEM_TYPE_PAGE_POOL) - return; - - if (likely(!xdp_frame_has_frags(xdpf))) - goto out; - - sinfo = xdp_get_shared_info_from_frame(xdpf); - for (i = 0; i < sinfo->nr_frags; i++) { - struct page *page = skb_frag_page(&sinfo->frags[i]); - - __xdp_release_frame(page_address(page), mem); - } -out: - __xdp_release_frame(xdpf->data, mem); -} - static __always_inline unsigned int xdp_get_frame_len(struct xdp_frame *xdpf) { struct skb_shared_info *sinfo; diff --git a/net/core/xdp.c b/net/core/xdp.c index a2237cfca8e9..8d3ad315f18d 100644 --- a/net/core/xdp.c +++ b/net/core/xdp.c @@ -531,21 +531,6 @@ out: } EXPORT_SYMBOL_GPL(xdp_return_buff); -/* Only called for MEM_TYPE_PAGE_POOL see xdp.h */ -void __xdp_release_frame(void *data, struct xdp_mem_info *mem) -{ - struct xdp_mem_allocator *xa; - struct page *page; - - rcu_read_lock(); - xa = rhashtable_lookup(mem_id_ht, &mem->id, mem_id_rht_params); - page = virt_to_head_page(data); - if (xa) - page_pool_release_page(xa->page_pool, page); - rcu_read_unlock(); -} -EXPORT_SYMBOL_GPL(__xdp_release_frame); - void xdp_attachment_setup(struct xdp_attachment_info *info, struct netdev_bpf *bpf) { -- cgit v1.2.3 From 007ae9b268ba7553e479608cf9735d3c4672a2ab Mon Sep 17 00:00:00 2001 From: Alexander Wetzel Date: Tue, 14 Mar 2023 22:11:22 +0100 Subject: wifi: mac80211: Serialize ieee80211_handle_wake_tx_queue() ieee80211_handle_wake_tx_queue must not run concurrent multiple times. It calls ieee80211_txq_schedule_start() and the drivers migrated to iTXQ do not expect overlapping drv_tx() calls. This fixes 'c850e31f79f0 ("wifi: mac80211: add internal handler for wake_tx_queue")', which introduced ieee80211_handle_wake_tx_queue. Drivers started to use it with 'a790cc3a4fad ("wifi: mac80211: add wake_tx_queue callback to drivers")'. But only after fixing an independent bug with '4444bc2116ae ("wifi: mac80211: Proper mark iTXQs for resumption")' problematic concurrent calls really happened and exposed the initial issue. Fixes: c850e31f79f0 ("wifi: mac80211: add internal handler for wake_tx_queue") Reported-by: Thomas Mann Link: https://bugzilla.kernel.org/show_bug.cgi?id=217119 Link: https://lore.kernel.org/r/b8efebc6-4399-d0b8-b2a0-66843314616b@leemhuis.info/ Link: https://lore.kernel.org/r/b7445607128a6b9ed7c17fcdcf3679bfaf4aaea.camel@sipsolutions.net> CC: Signed-off-by: Alexander Wetzel Link: https://lore.kernel.org/r/20230314211122.111688-1-alexander@wetzel-home.de [add missing spin_lock_init() noticed by Felix] Signed-off-by: Johannes Berg --- net/mac80211/ieee80211_i.h | 3 +++ net/mac80211/main.c | 2 ++ net/mac80211/util.c | 3 +++ 3 files changed, 8 insertions(+) (limited to 'net') diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index ecc232eb1ee8..e082582e0aa2 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -1284,6 +1284,9 @@ struct ieee80211_local { struct list_head active_txqs[IEEE80211_NUM_ACS]; u16 schedule_round[IEEE80211_NUM_ACS]; + /* serializes ieee80211_handle_wake_tx_queue */ + spinlock_t handle_wake_tx_queue_lock; + u16 airtime_flags; u32 aql_txq_limit_low[IEEE80211_NUM_ACS]; u32 aql_txq_limit_high[IEEE80211_NUM_ACS]; diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 846528850612..ddf2b7811c55 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -802,6 +802,8 @@ struct ieee80211_hw *ieee80211_alloc_hw_nm(size_t priv_data_len, local->aql_threshold = IEEE80211_AQL_THRESHOLD; atomic_set(&local->aql_total_pending_airtime, 0); + spin_lock_init(&local->handle_wake_tx_queue_lock); + INIT_LIST_HEAD(&local->chanctx_list); mutex_init(&local->chanctx_mtx); diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 1a28fe5cb614..3aceb3b731bf 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -314,6 +314,8 @@ void ieee80211_handle_wake_tx_queue(struct ieee80211_hw *hw, struct ieee80211_sub_if_data *sdata = vif_to_sdata(txq->vif); struct ieee80211_txq *queue; + spin_lock(&local->handle_wake_tx_queue_lock); + /* Use ieee80211_next_txq() for airtime fairness accounting */ ieee80211_txq_schedule_start(hw, txq->ac); while ((queue = ieee80211_next_txq(hw, txq->ac))) { @@ -321,6 +323,7 @@ void ieee80211_handle_wake_tx_queue(struct ieee80211_hw *hw, ieee80211_return_txq(hw, queue, false); } ieee80211_txq_schedule_end(hw, txq->ac); + spin_unlock(&local->handle_wake_tx_queue_lock); } EXPORT_SYMBOL(ieee80211_handle_wake_tx_queue); -- cgit v1.2.3 From aa3d65de4b9004d799f97700751a86d3ebd7d5f9 Mon Sep 17 00:00:00 2001 From: Viktor Malik Date: Fri, 10 Mar 2023 08:41:00 +0100 Subject: bpf/selftests: Test fentry attachment to shadowed functions Adds a new test that tries to attach a program to fentry of two functions of the same name, one located in vmlinux and the other in bpf_testmod. To avoid conflicts with existing tests, a new function "bpf_fentry_shadow_test" was created both in vmlinux and in bpf_testmod. The previous commit fixed a bug which caused this test to fail. The verifier would always use the vmlinux function's address as the target trampoline address, hence trying to create two trampolines for a single address, which is forbidden. The test (similarly to other fentry/fexit tests) is not working on arm64 at the moment. Signed-off-by: Viktor Malik Acked-by: Jiri Olsa Link: https://lore.kernel.org/r/5fe2f364190b6f79b085066ed7c5989c5bc475fa.1678432753.git.vmalik@redhat.com Signed-off-by: Alexei Starovoitov --- net/bpf/test_run.c | 5 + tools/testing/selftests/bpf/DENYLIST.aarch64 | 1 + .../selftests/bpf/bpf_testmod/bpf_testmod.c | 6 + .../bpf/prog_tests/module_fentry_shadow.c | 128 +++++++++++++++++++++ 4 files changed, 140 insertions(+) create mode 100644 tools/testing/selftests/bpf/prog_tests/module_fentry_shadow.c (limited to 'net') diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 6a8b33a103a4..71226f68270d 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -560,6 +560,11 @@ long noinline bpf_kfunc_call_test4(signed char a, short b, int c, long d) return (long)a + (long)b + (long)c + d; } +int noinline bpf_fentry_shadow_test(int a) +{ + return a + 1; +} + struct prog_test_member1 { int a; }; diff --git a/tools/testing/selftests/bpf/DENYLIST.aarch64 b/tools/testing/selftests/bpf/DENYLIST.aarch64 index 99cc33c51eaa..0a6837f97c32 100644 --- a/tools/testing/selftests/bpf/DENYLIST.aarch64 +++ b/tools/testing/selftests/bpf/DENYLIST.aarch64 @@ -44,6 +44,7 @@ lookup_key # test_lookup_key__attach unexp lru_bug # lru_bug__attach unexpected error: -524 (errno 524) modify_return # modify_return__attach failed unexpected error: -524 (errno 524) module_attach # skel_attach skeleton attach failed: -524 +module_fentry_shadow # bpf_link_create unexpected bpf_link_create: actual -524 < expected 0 mptcp/base # run_test mptcp unexpected error: -524 (errno 524) netcnt # packets unexpected packets: actual 10001 != expected 10000 rcu_read_lock # failed to attach: ERROR: strerror_r(-524)=22 diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c index 5e6e85c8d77d..7999476b9446 100644 --- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c +++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c @@ -268,6 +268,12 @@ static const struct btf_kfunc_id_set bpf_testmod_kfunc_set = { .set = &bpf_testmod_check_kfunc_ids, }; +noinline int bpf_fentry_shadow_test(int a) +{ + return a + 2; +} +EXPORT_SYMBOL_GPL(bpf_fentry_shadow_test); + extern int bpf_fentry_test1(int a); static int bpf_testmod_init(void) diff --git a/tools/testing/selftests/bpf/prog_tests/module_fentry_shadow.c b/tools/testing/selftests/bpf/prog_tests/module_fentry_shadow.c new file mode 100644 index 000000000000..c7636e18b1eb --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/module_fentry_shadow.c @@ -0,0 +1,128 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2022 Red Hat */ +#include +#include +#include "bpf/libbpf_internal.h" +#include "cgroup_helpers.h" + +static const char *module_name = "bpf_testmod"; +static const char *symbol_name = "bpf_fentry_shadow_test"; + +static int get_bpf_testmod_btf_fd(void) +{ + struct bpf_btf_info info; + char name[64]; + __u32 id = 0, len; + int err, fd; + + while (true) { + err = bpf_btf_get_next_id(id, &id); + if (err) { + log_err("failed to iterate BTF objects"); + return err; + } + + fd = bpf_btf_get_fd_by_id(id); + if (fd < 0) { + if (errno == ENOENT) + continue; /* expected race: BTF was unloaded */ + err = -errno; + log_err("failed to get FD for BTF object #%d", id); + return err; + } + + len = sizeof(info); + memset(&info, 0, sizeof(info)); + info.name = ptr_to_u64(name); + info.name_len = sizeof(name); + + err = bpf_obj_get_info_by_fd(fd, &info, &len); + if (err) { + err = -errno; + log_err("failed to get info for BTF object #%d", id); + close(fd); + return err; + } + + if (strcmp(name, module_name) == 0) + return fd; + + close(fd); + } + return -ENOENT; +} + +void test_module_fentry_shadow(void) +{ + struct btf *vmlinux_btf = NULL, *mod_btf = NULL; + int err, i; + int btf_fd[2] = {}; + int prog_fd[2] = {}; + int link_fd[2] = {}; + __s32 btf_id[2] = {}; + + LIBBPF_OPTS(bpf_prog_load_opts, load_opts, + .expected_attach_type = BPF_TRACE_FENTRY, + ); + + const struct bpf_insn trace_program[] = { + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }; + + vmlinux_btf = btf__load_vmlinux_btf(); + if (!ASSERT_OK_PTR(vmlinux_btf, "load_vmlinux_btf")) + return; + + btf_fd[1] = get_bpf_testmod_btf_fd(); + if (!ASSERT_GE(btf_fd[1], 0, "get_bpf_testmod_btf_fd")) + goto out; + + mod_btf = btf_get_from_fd(btf_fd[1], vmlinux_btf); + if (!ASSERT_OK_PTR(mod_btf, "btf_get_from_fd")) + goto out; + + btf_id[0] = btf__find_by_name_kind(vmlinux_btf, symbol_name, BTF_KIND_FUNC); + if (!ASSERT_GT(btf_id[0], 0, "btf_find_by_name")) + goto out; + + btf_id[1] = btf__find_by_name_kind(mod_btf, symbol_name, BTF_KIND_FUNC); + if (!ASSERT_GT(btf_id[1], 0, "btf_find_by_name")) + goto out; + + for (i = 0; i < 2; i++) { + load_opts.attach_btf_id = btf_id[i]; + load_opts.attach_btf_obj_fd = btf_fd[i]; + prog_fd[i] = bpf_prog_load(BPF_PROG_TYPE_TRACING, NULL, "GPL", + trace_program, + sizeof(trace_program) / sizeof(struct bpf_insn), + &load_opts); + if (!ASSERT_GE(prog_fd[i], 0, "bpf_prog_load")) + goto out; + + /* If the verifier incorrectly resolves addresses of the + * shadowed functions and uses the same address for both the + * vmlinux and the bpf_testmod functions, this will fail on + * attempting to create two trampolines for the same address, + * which is forbidden. + */ + link_fd[i] = bpf_link_create(prog_fd[i], 0, BPF_TRACE_FENTRY, NULL); + if (!ASSERT_GE(link_fd[i], 0, "bpf_link_create")) + goto out; + } + + err = bpf_prog_test_run_opts(prog_fd[0], NULL); + ASSERT_OK(err, "running test"); + +out: + btf__free(vmlinux_btf); + btf__free(mod_btf); + for (i = 0; i < 2; i++) { + if (btf_fd[i]) + close(btf_fd[i]); + if (prog_fd[i] > 0) + close(prog_fd[i]); + if (link_fd[i] > 0) + close(link_fd[i]); + } +} -- cgit v1.2.3 From c7df4813b149362248d6ef7be41a311e27bf75fe Mon Sep 17 00:00:00 2001 From: Kal Conley Date: Wed, 8 Mar 2023 18:40:13 +0100 Subject: xsk: Add missing overflow check in xdp_umem_reg The number of chunks can overflow u32. Make sure to return -EINVAL on overflow. Also remove a redundant u32 cast assigning umem->npgs. Fixes: bbff2f321a86 ("xsk: new descriptor addressing scheme") Signed-off-by: Kal Conley Signed-off-by: Daniel Borkmann Acked-by: Magnus Karlsson Link: https://lore.kernel.org/bpf/20230308174013.1114745-1-kal.conley@dectris.com --- net/xdp/xdp_umem.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c index 4681e8e8ad94..02207e852d79 100644 --- a/net/xdp/xdp_umem.c +++ b/net/xdp/xdp_umem.c @@ -150,10 +150,11 @@ static int xdp_umem_account_pages(struct xdp_umem *umem) static int xdp_umem_reg(struct xdp_umem *umem, struct xdp_umem_reg *mr) { - u32 npgs_rem, chunk_size = mr->chunk_size, headroom = mr->headroom; bool unaligned_chunks = mr->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG; - u64 npgs, addr = mr->addr, size = mr->len; - unsigned int chunks, chunks_rem; + u32 chunk_size = mr->chunk_size, headroom = mr->headroom; + u64 addr = mr->addr, size = mr->len; + u32 chunks_rem, npgs_rem; + u64 chunks, npgs; int err; if (chunk_size < XDP_UMEM_MIN_CHUNK_SIZE || chunk_size > PAGE_SIZE) { @@ -188,8 +189,8 @@ static int xdp_umem_reg(struct xdp_umem *umem, struct xdp_umem_reg *mr) if (npgs > U32_MAX) return -EINVAL; - chunks = (unsigned int)div_u64_rem(size, chunk_size, &chunks_rem); - if (chunks == 0) + chunks = div_u64_rem(size, chunk_size, &chunks_rem); + if (!chunks || chunks > U32_MAX) return -EINVAL; if (!unaligned_chunks && chunks_rem) @@ -202,7 +203,7 @@ static int xdp_umem_reg(struct xdp_umem *umem, struct xdp_umem_reg *mr) umem->headroom = headroom; umem->chunk_size = chunk_size; umem->chunks = chunks; - umem->npgs = (u32)npgs; + umem->npgs = npgs; umem->pgs = NULL; umem->user = NULL; umem->flags = mr->flags; -- cgit v1.2.3 From e5995bc7e2ba1a0d444f806016d2e4ea91c032d0 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Thu, 16 Mar 2023 18:50:50 +0100 Subject: bpf, test_run: fix crashes due to XDP frame overwriting/corruption MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit syzbot and Ilya faced the splats when %XDP_PASS happens for bpf_test_run after skb PP recycling was enabled for {__,}xdp_build_skb_from_frame(): BUG: kernel NULL pointer dereference, address: 0000000000000d28 RIP: 0010:memset_erms+0xd/0x20 arch/x86/lib/memset_64.S:66 [...] Call Trace: __finalize_skb_around net/core/skbuff.c:321 [inline] __build_skb_around+0x232/0x3a0 net/core/skbuff.c:379 build_skb_around+0x32/0x290 net/core/skbuff.c:444 __xdp_build_skb_from_frame+0x121/0x760 net/core/xdp.c:622 xdp_recv_frames net/bpf/test_run.c:248 [inline] xdp_test_run_batch net/bpf/test_run.c:334 [inline] bpf_test_run_xdp_live+0x1289/0x1930 net/bpf/test_run.c:362 bpf_prog_test_run_xdp+0xa05/0x14e0 net/bpf/test_run.c:1418 [...] This happens due to that it calls xdp_scrub_frame(), which nullifies xdpf->data. bpf_test_run code doesn't reinit the frame when the XDP program doesn't adjust head or tail. Previously, %XDP_PASS meant the page will be released from the pool and returned to the MM layer, but now it does return to the Pool with the nullified xdpf->data, which doesn't get reinitialized then. So, in addition to checking whether the head and/or tail have been adjusted, check also for a potential XDP frame corruption. xdpf->data is 100% affected and also xdpf->flags is the field closest to the metadata / frame start. Checking for these two should be enough for non-extreme cases. Fixes: 9c94bbf9a87b ("xdp: recycle Page Pool backed skbs built from XDP frames") Reported-by: syzbot+e1d1b65f7c32f2a86a9f@syzkaller.appspotmail.com Link: https://lore.kernel.org/bpf/000000000000f1985705f6ef2243@google.com Reported-by: Ilya Leoshkevich Link: https://lore.kernel.org/bpf/e07dd94022ad5731705891b9487cc9ed66328b94.camel@linux.ibm.com Signed-off-by: Alexander Lobakin Acked-by: Toke Høiland-Jørgensen Tested-by: Ilya Leoshkevich Link: https://lore.kernel.org/r/20230316175051.922550-2-aleksander.lobakin@intel.com Signed-off-by: Alexei Starovoitov --- net/bpf/test_run.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 71226f68270d..8d6b31209bd6 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -208,6 +208,16 @@ static void xdp_test_run_teardown(struct xdp_test_data *xdp) kfree(xdp->skbs); } +static bool frame_was_changed(const struct xdp_page_head *head) +{ + /* xdp_scrub_frame() zeroes the data pointer, flags is the last field, + * i.e. has the highest chances to be overwritten. If those two are + * untouched, it's most likely safe to skip the context reset. + */ + return head->frm.data != head->orig_ctx.data || + head->frm.flags != head->orig_ctx.flags; +} + static bool ctx_was_changed(struct xdp_page_head *head) { return head->orig_ctx.data != head->ctx.data || @@ -217,7 +227,7 @@ static bool ctx_was_changed(struct xdp_page_head *head) static void reset_ctx(struct xdp_page_head *head) { - if (likely(!ctx_was_changed(head))) + if (likely(!frame_was_changed(head) && !ctx_was_changed(head))) return; head->ctx.data = head->orig_ctx.data; -- cgit v1.2.3 From 984cfd55e0c99e80b2e5b1dc6b2bf98608af7ff9 Mon Sep 17 00:00:00 2001 From: Dongliang Mu Date: Wed, 8 Mar 2023 16:32:31 +0800 Subject: net: ieee802154: remove an unnecessary null pointer check llsec_parse_seclevel has the null pointer check at its begining. Compared with nl802154_add_llsec_seclevel, nl802154_del_llsec_seclevel has a redundant null pointer check of info->attrs[NL802154_ATTR_SEC_LEVEL] before llsec_parse_seclevel. Fix this issue by removing the null pointer check in nl802154_del_llsec_seclevel. Signed-off-by: Dongliang Mu Link: https://lore.kernel.org/r/20230308083231.460015-1-dzm91@hust.edu.cn Signed-off-by: Stefan Schmidt --- net/ieee802154/nl802154.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/ieee802154/nl802154.c b/net/ieee802154/nl802154.c index d8f4379d4fa6..832e3c50816c 100644 --- a/net/ieee802154/nl802154.c +++ b/net/ieee802154/nl802154.c @@ -2488,8 +2488,7 @@ static int nl802154_del_llsec_seclevel(struct sk_buff *skb, if (wpan_dev->iftype == NL802154_IFTYPE_MONITOR) return -EOPNOTSUPP; - if (!info->attrs[NL802154_ATTR_SEC_LEVEL] || - llsec_parse_seclevel(info->attrs[NL802154_ATTR_SEC_LEVEL], + if (llsec_parse_seclevel(info->attrs[NL802154_ATTR_SEC_LEVEL], &sl) < 0) return -EINVAL; -- cgit v1.2.3 From 10a03c36b7dd7759788ebc613091d313b60f93e0 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 13 Mar 2023 19:18:33 +0100 Subject: drivers: remove struct module * setting from struct class There is no need to manually set the owner of a struct class, as the registering function does it automatically, so remove all of the explicit settings from various drivers that did so as it is unneeded. This allows us to remove this pointer entirely from this structure going forward. Cc: "Rafael J. Wysocki" Link: https://lore.kernel.org/r/20230313181843.1207845-2-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman --- arch/mips/kernel/vpe-mt.c | 1 - drivers/base/core.c | 1 - drivers/base/devcoredump.c | 1 - drivers/block/pktcdvd.c | 1 - drivers/block/zram/zram_drv.c | 1 - drivers/gpio/gpiolib-sysfs.c | 2 -- drivers/hwmon/hwmon.c | 1 - drivers/isdn/mISDN/core.c | 1 - drivers/media/pci/ddbridge/ddbridge-core.c | 1 - drivers/mfd/cros_ec_dev.c | 1 - drivers/misc/enclosure.c | 1 - drivers/mtd/mtdcore.c | 1 - drivers/mtd/ubi/build.c | 1 - drivers/mux/core.c | 1 - drivers/net/ipvlan/ipvtap.c | 1 - drivers/net/macvtap.c | 1 - drivers/nvme/host/fc.c | 1 - drivers/platform/chrome/wilco_ec/event.c | 1 - drivers/platform/chrome/wilco_ec/telemetry.c | 1 - drivers/platform/x86/intel/pmt/class.c | 1 - drivers/platform/x86/intel_scu_ipc.c | 1 - drivers/ptp/ptp_ocp.c | 1 - drivers/pwm/sysfs.c | 1 - drivers/rapidio/rio-driver.c | 1 - drivers/scsi/sd.c | 1 - drivers/soc/qcom/rmtfs_mem.c | 1 - drivers/spi/spi.c | 2 -- drivers/staging/fieldbus/dev_core.c | 1 - drivers/staging/greybus/loopback.c | 1 - drivers/staging/greybus/vibrator.c | 1 - drivers/usb/typec/class.c | 1 - drivers/usb/typec/mux.c | 1 - drivers/usb/typec/pd.c | 1 - drivers/usb/typec/retimer.c | 1 - drivers/watchdog/watchdog_dev.c | 1 - fs/ksmbd/server.c | 1 - net/wireless/sysfs.c | 1 - 37 files changed, 39 deletions(-) (limited to 'net') diff --git a/arch/mips/kernel/vpe-mt.c b/arch/mips/kernel/vpe-mt.c index 223d6274f2e5..667bc75f6420 100644 --- a/arch/mips/kernel/vpe-mt.c +++ b/arch/mips/kernel/vpe-mt.c @@ -316,7 +316,6 @@ static void vpe_device_release(struct device *cd) static struct class vpe_class = { .name = "vpe", - .owner = THIS_MODULE, .dev_release = vpe_device_release, .dev_groups = vpe_groups, }; diff --git a/drivers/base/core.c b/drivers/base/core.c index fe74a786e2c3..57076837b33e 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -537,7 +537,6 @@ static void devlink_dev_release(struct device *dev) static struct class devlink_class = { .name = "devlink", - .owner = THIS_MODULE, .dev_groups = devlink_groups, .dev_release = devlink_dev_release, }; diff --git a/drivers/base/devcoredump.c b/drivers/base/devcoredump.c index 1c06781f7114..59aaf2e1375a 100644 --- a/drivers/base/devcoredump.c +++ b/drivers/base/devcoredump.c @@ -226,7 +226,6 @@ ATTRIBUTE_GROUPS(devcd_class); static struct class devcd_class = { .name = "devcoredump", - .owner = THIS_MODULE, .dev_release = devcd_dev_release, .dev_groups = devcd_dev_groups, .class_groups = devcd_class_groups, diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index 2f1a92509271..642e3377441a 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -417,7 +417,6 @@ static int pkt_sysfs_init(void) if (!class_pktcdvd) return -ENOMEM; class_pktcdvd->name = DRIVER_NAME; - class_pktcdvd->owner = THIS_MODULE; class_pktcdvd->class_release = class_pktcdvd_release; class_pktcdvd->class_groups = class_pktcdvd_groups; ret = class_register(class_pktcdvd); diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index aa490da3cef2..b7bb52f8dfbd 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -2481,7 +2481,6 @@ ATTRIBUTE_GROUPS(zram_control_class); static struct class zram_control_class = { .name = "zram-control", - .owner = THIS_MODULE, .class_groups = zram_control_class_groups, }; diff --git a/drivers/gpio/gpiolib-sysfs.c b/drivers/gpio/gpiolib-sysfs.c index cd27bf173dec..774755052618 100644 --- a/drivers/gpio/gpiolib-sysfs.c +++ b/drivers/gpio/gpiolib-sysfs.c @@ -523,8 +523,6 @@ ATTRIBUTE_GROUPS(gpio_class); static struct class gpio_class = { .name = "gpio", - .owner = THIS_MODULE, - .class_groups = gpio_class_groups, }; diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c index 33edb5c02f7d..d15ef89e5191 100644 --- a/drivers/hwmon/hwmon.c +++ b/drivers/hwmon/hwmon.c @@ -138,7 +138,6 @@ static void hwmon_dev_release(struct device *dev) static struct class hwmon_class = { .name = "hwmon", - .owner = THIS_MODULE, .dev_groups = hwmon_dev_attr_groups, .dev_release = hwmon_dev_release, }; diff --git a/drivers/isdn/mISDN/core.c b/drivers/isdn/mISDN/core.c index 9120be590325..f5989c9907ee 100644 --- a/drivers/isdn/mISDN/core.c +++ b/drivers/isdn/mISDN/core.c @@ -159,7 +159,6 @@ static void mISDN_class_release(struct class *cls) static struct class mISDN_class = { .name = "mISDN", - .owner = THIS_MODULE, .dev_uevent = mISDN_uevent, .dev_groups = mISDN_groups, .dev_release = mISDN_dev_release, diff --git a/drivers/media/pci/ddbridge/ddbridge-core.c b/drivers/media/pci/ddbridge/ddbridge-core.c index ee8087f29b2c..40e6c873c36d 100644 --- a/drivers/media/pci/ddbridge/ddbridge-core.c +++ b/drivers/media/pci/ddbridge/ddbridge-core.c @@ -3117,7 +3117,6 @@ static struct device_attribute ddb_attrs_fanspeed[] = { static struct class ddb_class = { .name = "ddbridge", - .owner = THIS_MODULE, .devnode = ddb_devnode, }; diff --git a/drivers/mfd/cros_ec_dev.c b/drivers/mfd/cros_ec_dev.c index 02d4271dfe06..92f4dfccc3cc 100644 --- a/drivers/mfd/cros_ec_dev.c +++ b/drivers/mfd/cros_ec_dev.c @@ -20,7 +20,6 @@ #define DRV_NAME "cros-ec-dev" static struct class cros_class = { - .owner = THIS_MODULE, .name = "chromeos", }; diff --git a/drivers/misc/enclosure.c b/drivers/misc/enclosure.c index 4ba966529458..76511d279aff 100644 --- a/drivers/misc/enclosure.c +++ b/drivers/misc/enclosure.c @@ -451,7 +451,6 @@ ATTRIBUTE_GROUPS(enclosure_class); static struct class enclosure_class = { .name = "enclosure", - .owner = THIS_MODULE, .dev_release = enclosure_release, .dev_groups = enclosure_class_groups, }; diff --git a/drivers/mtd/mtdcore.c b/drivers/mtd/mtdcore.c index 0feacb9fbdac..63fca6141973 100644 --- a/drivers/mtd/mtdcore.c +++ b/drivers/mtd/mtdcore.c @@ -63,7 +63,6 @@ static SIMPLE_DEV_PM_OPS(mtd_cls_pm_ops, mtd_cls_suspend, mtd_cls_resume); static struct class mtd_class = { .name = "mtd", - .owner = THIS_MODULE, .pm = MTD_CLS_PM_OPS, }; diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 0904eb40c95f..ae6d35e3da9c 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -111,7 +111,6 @@ ATTRIBUTE_GROUPS(ubi_class); /* Root UBI "class" object (corresponds to '//class/ubi/') */ struct class ubi_class = { .name = UBI_NAME_STR, - .owner = THIS_MODULE, .class_groups = ubi_class_groups, }; diff --git a/drivers/mux/core.c b/drivers/mux/core.c index 49bedbe6316c..990e7bc17c85 100644 --- a/drivers/mux/core.c +++ b/drivers/mux/core.c @@ -45,7 +45,6 @@ struct mux_state { static struct class mux_class = { .name = "mux", - .owner = THIS_MODULE, }; static DEFINE_IDA(mux_ida); diff --git a/drivers/net/ipvlan/ipvtap.c b/drivers/net/ipvlan/ipvtap.c index dde272586e80..60944a4beada 100644 --- a/drivers/net/ipvlan/ipvtap.c +++ b/drivers/net/ipvlan/ipvtap.c @@ -38,7 +38,6 @@ static const void *ipvtap_net_namespace(const struct device *d) static struct class ipvtap_class = { .name = "ipvtap", - .owner = THIS_MODULE, .ns_type = &net_ns_type_operations, .namespace = ipvtap_net_namespace, }; diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 031344239f27..bddcc127812e 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -43,7 +43,6 @@ static const void *macvtap_net_namespace(const struct device *d) static struct class macvtap_class = { .name = "macvtap", - .owner = THIS_MODULE, .ns_type = &net_ns_type_operations, .namespace = macvtap_net_namespace, }; diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 456ee42a6133..2ed75923507d 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -3875,7 +3875,6 @@ static const struct attribute_group *nvme_fc_attr_groups[] = { static struct class fc_class = { .name = "fc", .dev_groups = nvme_fc_attr_groups, - .owner = THIS_MODULE, }; static int __init nvme_fc_init_module(void) diff --git a/drivers/platform/chrome/wilco_ec/event.c b/drivers/platform/chrome/wilco_ec/event.c index 69ceead8cdaa..a40f60bcefb6 100644 --- a/drivers/platform/chrome/wilco_ec/event.c +++ b/drivers/platform/chrome/wilco_ec/event.c @@ -58,7 +58,6 @@ #define DRV_NAME EVENT_DEV_NAME #define EVENT_DEV_NAME_FMT (EVENT_DEV_NAME "%d") static struct class event_class = { - .owner = THIS_MODULE, .name = EVENT_CLASS_NAME, }; diff --git a/drivers/platform/chrome/wilco_ec/telemetry.c b/drivers/platform/chrome/wilco_ec/telemetry.c index 60da7a29f2ff..54708aa6c700 100644 --- a/drivers/platform/chrome/wilco_ec/telemetry.c +++ b/drivers/platform/chrome/wilco_ec/telemetry.c @@ -42,7 +42,6 @@ #define DRV_NAME TELEM_DEV_NAME #define TELEM_DEV_NAME_FMT (TELEM_DEV_NAME "%d") static struct class telem_class = { - .owner = THIS_MODULE, .name = TELEM_CLASS_NAME, }; diff --git a/drivers/platform/x86/intel/pmt/class.c b/drivers/platform/x86/intel/pmt/class.c index 46598dcb634a..80292d21a0b4 100644 --- a/drivers/platform/x86/intel/pmt/class.c +++ b/drivers/platform/x86/intel/pmt/class.c @@ -155,7 +155,6 @@ ATTRIBUTE_GROUPS(intel_pmt); static struct class intel_pmt_class = { .name = "intel_pmt", - .owner = THIS_MODULE, .dev_groups = intel_pmt_groups, }; diff --git a/drivers/platform/x86/intel_scu_ipc.c b/drivers/platform/x86/intel_scu_ipc.c index e7a3e3402817..6851d10d6582 100644 --- a/drivers/platform/x86/intel_scu_ipc.c +++ b/drivers/platform/x86/intel_scu_ipc.c @@ -82,7 +82,6 @@ static DEFINE_MUTEX(ipclock); /* lock used to prevent multiple call to SCU */ static struct class intel_scu_ipc_class = { .name = "intel_scu_ipc", - .owner = THIS_MODULE, }; /** diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c index 4bbaccd543ad..93eba1430883 100644 --- a/drivers/ptp/ptp_ocp.c +++ b/drivers/ptp/ptp_ocp.c @@ -34,7 +34,6 @@ #define PCI_DEVICE_ID_OROLIA_ARTCARD 0xa000 static struct class timecard_class = { - .owner = THIS_MODULE, .name = "timecard", }; diff --git a/drivers/pwm/sysfs.c b/drivers/pwm/sysfs.c index e7db8e45001c..1a106ec32939 100644 --- a/drivers/pwm/sysfs.c +++ b/drivers/pwm/sysfs.c @@ -475,7 +475,6 @@ static DEFINE_SIMPLE_DEV_PM_OPS(pwm_class_pm_ops, pwm_class_suspend, pwm_class_r static struct class pwm_class = { .name = "pwm", - .owner = THIS_MODULE, .dev_groups = pwm_chip_groups, .pm = pm_sleep_ptr(&pwm_class_pm_ops), }; diff --git a/drivers/rapidio/rio-driver.c b/drivers/rapidio/rio-driver.c index e60e49769bed..1b3b4c2e015d 100644 --- a/drivers/rapidio/rio-driver.c +++ b/drivers/rapidio/rio-driver.c @@ -223,7 +223,6 @@ static int rio_uevent(const struct device *dev, struct kobj_uevent_env *env) struct class rio_mport_class = { .name = "rapidio_port", - .owner = THIS_MODULE, .dev_groups = rio_mport_groups, }; EXPORT_SYMBOL_GPL(rio_mport_class); diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 4f28dd617eca..f98878776204 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -587,7 +587,6 @@ ATTRIBUTE_GROUPS(sd_disk); static struct class sd_disk_class = { .name = "scsi_disk", - .owner = THIS_MODULE, .dev_release = scsi_disk_release, .dev_groups = sd_disk_groups, }; diff --git a/drivers/soc/qcom/rmtfs_mem.c b/drivers/soc/qcom/rmtfs_mem.c index 2d3ee22b9249..5c321da7bb9d 100644 --- a/drivers/soc/qcom/rmtfs_mem.c +++ b/drivers/soc/qcom/rmtfs_mem.c @@ -126,7 +126,6 @@ static int qcom_rmtfs_mem_release(struct inode *inode, struct file *filp) } static struct class rmtfs_class = { - .owner = THIS_MODULE, .name = "rmtfs", }; diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 44b85a8d47f1..4b9a17f024c0 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -2776,7 +2776,6 @@ static void spi_controller_release(struct device *dev) static struct class spi_master_class = { .name = "spi_master", - .owner = THIS_MODULE, .dev_release = spi_controller_release, .dev_groups = spi_master_groups, }; @@ -2879,7 +2878,6 @@ static const struct attribute_group *spi_slave_groups[] = { static struct class spi_slave_class = { .name = "spi_slave", - .owner = THIS_MODULE, .dev_release = spi_controller_release, .dev_groups = spi_slave_groups, }; diff --git a/drivers/staging/fieldbus/dev_core.c b/drivers/staging/fieldbus/dev_core.c index 5f54f2674bd1..bf1812d8924f 100644 --- a/drivers/staging/fieldbus/dev_core.c +++ b/drivers/staging/fieldbus/dev_core.c @@ -154,7 +154,6 @@ __ATTRIBUTE_GROUPS(fieldbus); static struct class fieldbus_class = { .name = "fieldbus_dev", - .owner = THIS_MODULE, .dev_groups = fieldbus_groups, }; diff --git a/drivers/staging/greybus/loopback.c b/drivers/staging/greybus/loopback.c index 1a61fce98056..d7b39f3bb652 100644 --- a/drivers/staging/greybus/loopback.c +++ b/drivers/staging/greybus/loopback.c @@ -100,7 +100,6 @@ struct gb_loopback { static struct class loopback_class = { .name = "gb_loopback", - .owner = THIS_MODULE, }; static DEFINE_IDA(loopback_ida); diff --git a/drivers/staging/greybus/vibrator.c b/drivers/staging/greybus/vibrator.c index 0e2b188e5ca3..227e18d92a95 100644 --- a/drivers/staging/greybus/vibrator.c +++ b/drivers/staging/greybus/vibrator.c @@ -107,7 +107,6 @@ ATTRIBUTE_GROUPS(vibrator); static struct class vibrator_class = { .name = "vibrator", - .owner = THIS_MODULE, .dev_groups = vibrator_groups, }; diff --git a/drivers/usb/typec/class.c b/drivers/usb/typec/class.c index cc3182f70673..349cc2030c90 100644 --- a/drivers/usb/typec/class.c +++ b/drivers/usb/typec/class.c @@ -22,7 +22,6 @@ static DEFINE_IDA(typec_index_ida); struct class typec_class = { .name = "typec", - .owner = THIS_MODULE, }; /* ------------------------------------------------------------------------- */ diff --git a/drivers/usb/typec/mux.c b/drivers/usb/typec/mux.c index c7177ddd4f12..d9eaf9a0b0bf 100644 --- a/drivers/usb/typec/mux.c +++ b/drivers/usb/typec/mux.c @@ -514,5 +514,4 @@ EXPORT_SYMBOL_GPL(typec_mux_get_drvdata); struct class typec_mux_class = { .name = "typec_mux", - .owner = THIS_MODULE, }; diff --git a/drivers/usb/typec/pd.c b/drivers/usb/typec/pd.c index 59c537a5e600..0bcde1ff4d39 100644 --- a/drivers/usb/typec/pd.c +++ b/drivers/usb/typec/pd.c @@ -15,7 +15,6 @@ static DEFINE_IDA(pd_ida); static struct class pd_class = { .name = "usb_power_delivery", - .owner = THIS_MODULE, }; #define to_pdo(o) container_of(o, struct pdo, dev) diff --git a/drivers/usb/typec/retimer.c b/drivers/usb/typec/retimer.c index 0481e82f6bbc..4a7d1b5c4d86 100644 --- a/drivers/usb/typec/retimer.c +++ b/drivers/usb/typec/retimer.c @@ -157,5 +157,4 @@ EXPORT_SYMBOL_GPL(typec_retimer_get_drvdata); struct class retimer_class = { .name = "retimer", - .owner = THIS_MODULE, }; diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c index 0122e8796879..12a6f020b6b5 100644 --- a/drivers/watchdog/watchdog_dev.c +++ b/drivers/watchdog/watchdog_dev.c @@ -1005,7 +1005,6 @@ static struct miscdevice watchdog_miscdev = { static struct class watchdog_class = { .name = "watchdog", - .owner = THIS_MODULE, .dev_groups = wdt_groups, }; diff --git a/fs/ksmbd/server.c b/fs/ksmbd/server.c index 394b6ceac431..b5af3e43e677 100644 --- a/fs/ksmbd/server.c +++ b/fs/ksmbd/server.c @@ -516,7 +516,6 @@ ATTRIBUTE_GROUPS(ksmbd_control_class); static struct class ksmbd_control_class = { .name = "ksmbd-control", - .owner = THIS_MODULE, .class_groups = ksmbd_control_class_groups, }; diff --git a/net/wireless/sysfs.c b/net/wireless/sysfs.c index cdb638647e0b..268f670835e9 100644 --- a/net/wireless/sysfs.c +++ b/net/wireless/sysfs.c @@ -157,7 +157,6 @@ static const void *wiphy_namespace(const struct device *d) struct class ieee80211_class = { .name = "ieee80211", - .owner = THIS_MODULE, .dev_release = wiphy_dev_release, .dev_groups = ieee80211_groups, .pm = WIPHY_PM_OPS, -- cgit v1.2.3 From 1aaba11da9aa7d7d6b52a74d45b31cac118295a1 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 13 Mar 2023 19:18:35 +0100 Subject: driver core: class: remove module * from class_create() The module pointer in class_create() never actually did anything, and it shouldn't have been requred to be set as a parameter even if it did something. So just remove it and fix up all callers of the function in the kernel tree at the same time. Cc: "Rafael J. Wysocki" Acked-by: Benjamin Tissoires Link: https://lore.kernel.org/r/20230313181843.1207845-4-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman --- arch/mips/kernel/mips-mt.c | 2 +- arch/mips/sibyte/common/sb_tbprof.c | 2 +- arch/powerpc/platforms/book3s/vas-api.c | 2 +- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 2 +- arch/x86/kernel/cpuid.c | 2 +- arch/x86/kernel/msr.c | 2 +- block/bsg.c | 2 +- drivers/accel/drm_accel.c | 2 +- drivers/accel/habanalabs/common/habanalabs_drv.c | 2 +- drivers/base/power/wakeup_stats.c | 2 +- drivers/block/aoe/aoechr.c | 2 +- drivers/block/rnbd/rnbd-clt-sysfs.c | 2 +- drivers/block/rnbd/rnbd-srv-sysfs.c | 2 +- drivers/block/ublk_drv.c | 2 +- drivers/char/bsr.c | 2 +- drivers/char/dsp56k.c | 2 +- drivers/char/ipmi/ipmi_devintf.c | 2 +- drivers/char/lp.c | 2 +- drivers/char/mem.c | 2 +- drivers/char/misc.c | 2 +- drivers/char/pcmcia/cm4000_cs.c | 2 +- drivers/char/pcmcia/cm4040_cs.c | 2 +- drivers/char/pcmcia/scr24x_cs.c | 2 +- drivers/char/ppdev.c | 2 +- drivers/char/tpm/tpm-interface.c | 4 ++-- drivers/char/virtio_console.c | 2 +- drivers/char/xilinx_hwicap/xilinx_hwicap.c | 2 +- drivers/char/xillybus/xillybus_class.c | 2 +- drivers/comedi/comedi_fops.c | 2 +- drivers/comedi/drivers/comedi_test.c | 2 +- drivers/crypto/qat/qat_common/adf_ctl_drv.c | 2 +- drivers/dca/dca-sysfs.c | 2 +- drivers/devfreq/devfreq-event.c | 2 +- drivers/devfreq/devfreq.c | 2 +- drivers/dma-buf/dma-heap.c | 2 +- drivers/extcon/extcon.c | 2 +- drivers/fpga/fpga-bridge.c | 2 +- drivers/fpga/fpga-mgr.c | 2 +- drivers/fpga/fpga-region.c | 2 +- drivers/gnss/core.c | 2 +- drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 2 +- drivers/gpu/drm/display/drm_dp_aux_dev.c | 2 +- drivers/gpu/drm/drm_sysfs.c | 2 +- drivers/hid/hid-roccat-arvo.c | 2 +- drivers/hid/hid-roccat-isku.c | 2 +- drivers/hid/hid-roccat-kone.c | 2 +- drivers/hid/hid-roccat-koneplus.c | 2 +- drivers/hid/hid-roccat-konepure.c | 2 +- drivers/hid/hid-roccat-kovaplus.c | 2 +- drivers/hid/hid-roccat-pyra.c | 2 +- drivers/hid/hid-roccat-ryos.c | 2 +- drivers/hid/hid-roccat-savu.c | 2 +- drivers/hid/hidraw.c | 2 +- drivers/i2c/i2c-dev.c | 2 +- drivers/infiniband/core/uverbs_main.c | 2 +- drivers/infiniband/hw/hfi1/device.c | 4 ++-- drivers/infiniband/hw/qib/qib_file_ops.c | 2 +- drivers/infiniband/ulp/rtrs/rtrs-clt.c | 2 +- drivers/infiniband/ulp/rtrs/rtrs-srv.c | 2 +- drivers/isdn/capi/capi.c | 2 +- drivers/isdn/mISDN/dsp_pipeline.c | 2 +- drivers/leds/led-class.c | 2 +- drivers/macintosh/adb.c | 2 +- drivers/media/dvb-core/dvbdev.c | 2 +- drivers/media/rc/lirc_dev.c | 2 +- drivers/misc/c2port/core.c | 2 +- drivers/misc/cxl/file.c | 2 +- drivers/misc/genwqe/card_base.c | 2 +- drivers/misc/hpilo.c | 2 +- drivers/misc/mei/main.c | 2 +- drivers/misc/ocxl/file.c | 2 +- drivers/misc/phantom.c | 2 +- drivers/misc/uacce/uacce.c | 2 +- drivers/most/most_cdev.c | 2 +- drivers/net/ethernet/hisilicon/hns/hnae.c | 2 +- drivers/net/ppp/ppp_generic.c | 2 +- drivers/net/wireless/mac80211_hwsim.c | 2 +- drivers/net/wwan/wwan_core.c | 2 +- drivers/net/wwan/wwan_hwsim.c | 2 +- drivers/nvdimm/bus.c | 2 +- drivers/nvme/host/core.c | 6 +++--- drivers/nvme/host/fabrics.c | 2 +- drivers/nvme/target/fcloop.c | 2 +- drivers/pci/endpoint/pci-epc-core.c | 2 +- drivers/pci/switch/switchtec.c | 2 +- drivers/phy/phy-core.c | 2 +- drivers/power/supply/power_supply_core.c | 2 +- drivers/pps/pps.c | 2 +- drivers/ptp/ptp_clock.c | 2 +- drivers/rapidio/devices/rio_mport_cdev.c | 2 +- drivers/rapidio/rio_cm.c | 2 +- drivers/rpmsg/rpmsg_core.c | 2 +- drivers/rtc/class.c | 2 +- drivers/s390/char/hmcdrv_dev.c | 2 +- drivers/s390/char/raw3270.c | 2 +- drivers/s390/char/tape_class.c | 2 +- drivers/s390/char/vmlogrdr.c | 2 +- drivers/s390/char/vmur.c | 2 +- drivers/s390/crypto/zcrypt_api.c | 2 +- drivers/sbus/char/oradax.c | 2 +- drivers/scsi/ch.c | 2 +- drivers/scsi/cxlflash/main.c | 2 +- drivers/scsi/pmcraid.c | 2 +- drivers/scsi/sg.c | 2 +- drivers/spi/spidev.c | 2 +- drivers/staging/fieldbus/anybuss/arcx-anybus.c | 2 +- drivers/staging/greybus/authentication.c | 2 +- drivers/staging/greybus/fw-management.c | 2 +- drivers/staging/greybus/raw.c | 2 +- drivers/staging/pi433/pi433_if.c | 2 +- drivers/staging/vme_user/vme_user.c | 2 +- drivers/tee/tee_core.c | 2 +- drivers/tty/tty_io.c | 2 +- drivers/tty/vt/vc_screen.c | 2 +- drivers/tty/vt/vt.c | 2 +- drivers/usb/core/file.c | 2 +- drivers/usb/gadget/function/f_hid.c | 2 +- drivers/usb/gadget/function/f_printer.c | 2 +- drivers/usb/gadget/udc/core.c | 2 +- drivers/usb/mon/mon_bin.c | 2 +- drivers/usb/roles/class.c | 2 +- drivers/vdpa/vdpa_user/vduse_dev.c | 2 +- drivers/vfio/group.c | 2 +- drivers/vfio/vfio_main.c | 2 +- drivers/video/backlight/backlight.c | 2 +- drivers/video/backlight/lcd.c | 2 +- drivers/video/fbdev/core/fbmem.c | 2 +- fs/coda/psdev.c | 2 +- fs/fuse/cuse.c | 2 +- fs/pstore/pmsg.c | 2 +- include/linux/device/class.h | 3 +-- mm/backing-dev.c | 2 +- net/bluetooth/hci_sysfs.c | 2 +- net/netfilter/xt_IDLETIMER.c | 2 +- samples/vfio-mdev/mbochs.c | 2 +- samples/vfio-mdev/mdpy.c | 2 +- samples/vfio-mdev/mtty.c | 2 +- sound/sound_core.c | 2 +- tools/testing/nvdimm/test/ndtest.c | 2 +- tools/testing/nvdimm/test/nfit.c | 2 +- 140 files changed, 144 insertions(+), 145 deletions(-) (limited to 'net') diff --git a/arch/mips/kernel/mips-mt.c b/arch/mips/kernel/mips-mt.c index dc023a979803..f88b7919f11f 100644 --- a/arch/mips/kernel/mips-mt.c +++ b/arch/mips/kernel/mips-mt.c @@ -234,7 +234,7 @@ static int __init mips_mt_init(void) { struct class *mtc; - mtc = class_create(THIS_MODULE, "mt"); + mtc = class_create("mt"); if (IS_ERR(mtc)) return PTR_ERR(mtc); diff --git a/arch/mips/sibyte/common/sb_tbprof.c b/arch/mips/sibyte/common/sb_tbprof.c index bc47681e825a..ac376dfb4e7c 100644 --- a/arch/mips/sibyte/common/sb_tbprof.c +++ b/arch/mips/sibyte/common/sb_tbprof.c @@ -550,7 +550,7 @@ static int __init sbprof_tb_init(void) return -EIO; } - tbc = class_create(THIS_MODULE, "sb_tracebuffer"); + tbc = class_create("sb_tracebuffer"); if (IS_ERR(tbc)) { err = PTR_ERR(tbc); goto out_chrdev; diff --git a/arch/powerpc/platforms/book3s/vas-api.c b/arch/powerpc/platforms/book3s/vas-api.c index 36c21648d19a..77ea9335fd04 100644 --- a/arch/powerpc/platforms/book3s/vas-api.c +++ b/arch/powerpc/platforms/book3s/vas-api.c @@ -581,7 +581,7 @@ int vas_register_coproc_api(struct module *mod, enum vas_cop_type cop_type, pr_devel("%s device allocated, dev [%i,%i]\n", name, MAJOR(coproc_device.devt), MINOR(coproc_device.devt)); - coproc_device.class = class_create(mod, name); + coproc_device.class = class_create(name); if (IS_ERR(coproc_device.class)) { rc = PTR_ERR(coproc_device.class); pr_err("Unable to create %s class %d\n", name, rc); diff --git a/arch/x86/kernel/cpu/resctrl/pseudo_lock.c b/arch/x86/kernel/cpu/resctrl/pseudo_lock.c index 524f8ff3e69c..458cb7419502 100644 --- a/arch/x86/kernel/cpu/resctrl/pseudo_lock.c +++ b/arch/x86/kernel/cpu/resctrl/pseudo_lock.c @@ -1580,7 +1580,7 @@ int rdt_pseudo_lock_init(void) pseudo_lock_major = ret; - pseudo_lock_class = class_create(THIS_MODULE, "pseudo_lock"); + pseudo_lock_class = class_create("pseudo_lock"); if (IS_ERR(pseudo_lock_class)) { ret = PTR_ERR(pseudo_lock_class); unregister_chrdev(pseudo_lock_major, "pseudo_lock"); diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c index 621ba9c0f17a..bdc0d5539b57 100644 --- a/arch/x86/kernel/cpuid.c +++ b/arch/x86/kernel/cpuid.c @@ -154,7 +154,7 @@ static int __init cpuid_init(void) CPUID_MAJOR); return -EBUSY; } - cpuid_class = class_create(THIS_MODULE, "cpuid"); + cpuid_class = class_create("cpuid"); if (IS_ERR(cpuid_class)) { err = PTR_ERR(cpuid_class); goto out_chrdev; diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c index 708751311786..7bb17d37db01 100644 --- a/arch/x86/kernel/msr.c +++ b/arch/x86/kernel/msr.c @@ -263,7 +263,7 @@ static int __init msr_init(void) pr_err("unable to get major %d for msr\n", MSR_MAJOR); return -EBUSY; } - msr_class = class_create(THIS_MODULE, "msr"); + msr_class = class_create("msr"); if (IS_ERR(msr_class)) { err = PTR_ERR(msr_class); goto out_chrdev; diff --git a/block/bsg.c b/block/bsg.c index 30fcc865ef4f..7eca43f33d7f 100644 --- a/block/bsg.c +++ b/block/bsg.c @@ -245,7 +245,7 @@ static int __init bsg_init(void) dev_t devid; int ret; - bsg_class = class_create(THIS_MODULE, "bsg"); + bsg_class = class_create("bsg"); if (IS_ERR(bsg_class)) return PTR_ERR(bsg_class); bsg_class->devnode = bsg_devnode; diff --git a/drivers/accel/drm_accel.c b/drivers/accel/drm_accel.c index 1b69824286fd..4a9baf02439e 100644 --- a/drivers/accel/drm_accel.c +++ b/drivers/accel/drm_accel.c @@ -34,7 +34,7 @@ static char *accel_devnode(const struct device *dev, umode_t *mode) static int accel_sysfs_init(void) { - accel_class = class_create(THIS_MODULE, "accel"); + accel_class = class_create("accel"); if (IS_ERR(accel_class)) return PTR_ERR(accel_class); diff --git a/drivers/accel/habanalabs/common/habanalabs_drv.c b/drivers/accel/habanalabs/common/habanalabs_drv.c index 03dae57dc838..3538e89e4e5e 100644 --- a/drivers/accel/habanalabs/common/habanalabs_drv.c +++ b/drivers/accel/habanalabs/common/habanalabs_drv.c @@ -702,7 +702,7 @@ static int __init hl_init(void) hl_major = MAJOR(dev); - hl_class = class_create(THIS_MODULE, HL_NAME); + hl_class = class_create(HL_NAME); if (IS_ERR(hl_class)) { pr_err("failed to allocate class\n"); rc = PTR_ERR(hl_class); diff --git a/drivers/base/power/wakeup_stats.c b/drivers/base/power/wakeup_stats.c index 924fac493c4f..6732ed2869f9 100644 --- a/drivers/base/power/wakeup_stats.c +++ b/drivers/base/power/wakeup_stats.c @@ -210,7 +210,7 @@ void wakeup_source_sysfs_remove(struct wakeup_source *ws) static int __init wakeup_sources_sysfs_init(void) { - wakeup_class = class_create(THIS_MODULE, "wakeup"); + wakeup_class = class_create("wakeup"); return PTR_ERR_OR_ZERO(wakeup_class); } diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c index 7a368c90467d..4c666f72203f 100644 --- a/drivers/block/aoe/aoechr.c +++ b/drivers/block/aoe/aoechr.c @@ -290,7 +290,7 @@ aoechr_init(void) } init_completion(&emsgs_comp); spin_lock_init(&emsgs_lock); - aoe_class = class_create(THIS_MODULE, "aoe"); + aoe_class = class_create("aoe"); if (IS_ERR(aoe_class)) { unregister_chrdev(AOE_MAJOR, "aoechr"); return PTR_ERR(aoe_class); diff --git a/drivers/block/rnbd/rnbd-clt-sysfs.c b/drivers/block/rnbd/rnbd-clt-sysfs.c index e7c7d9a68168..8c6087949794 100644 --- a/drivers/block/rnbd/rnbd-clt-sysfs.c +++ b/drivers/block/rnbd/rnbd-clt-sysfs.c @@ -646,7 +646,7 @@ int rnbd_clt_create_sysfs_files(void) { int err; - rnbd_dev_class = class_create(THIS_MODULE, "rnbd-client"); + rnbd_dev_class = class_create("rnbd-client"); if (IS_ERR(rnbd_dev_class)) return PTR_ERR(rnbd_dev_class); diff --git a/drivers/block/rnbd/rnbd-srv-sysfs.c b/drivers/block/rnbd/rnbd-srv-sysfs.c index 297a6924ff4e..d5d9267e1fa5 100644 --- a/drivers/block/rnbd/rnbd-srv-sysfs.c +++ b/drivers/block/rnbd/rnbd-srv-sysfs.c @@ -215,7 +215,7 @@ int rnbd_srv_create_sysfs_files(void) { int err; - rnbd_dev_class = class_create(THIS_MODULE, "rnbd-server"); + rnbd_dev_class = class_create("rnbd-server"); if (IS_ERR(rnbd_dev_class)) return PTR_ERR(rnbd_dev_class); diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c index d1d1c8d606c8..2eb46419562b 100644 --- a/drivers/block/ublk_drv.c +++ b/drivers/block/ublk_drv.c @@ -2266,7 +2266,7 @@ static int __init ublk_init(void) if (ret) goto unregister_mis; - ublk_chr_class = class_create(THIS_MODULE, "ublk-char"); + ublk_chr_class = class_create("ublk-char"); if (IS_ERR(ublk_chr_class)) { ret = PTR_ERR(ublk_chr_class); goto free_chrdev_region; diff --git a/drivers/char/bsr.c b/drivers/char/bsr.c index d5f943938427..ff429ba02fa4 100644 --- a/drivers/char/bsr.c +++ b/drivers/char/bsr.c @@ -293,7 +293,7 @@ static int __init bsr_init(void) if (!np) goto out_err; - bsr_class = class_create(THIS_MODULE, "bsr"); + bsr_class = class_create("bsr"); if (IS_ERR(bsr_class)) { printk(KERN_ERR "class_create() failed for bsr_class\n"); ret = PTR_ERR(bsr_class); diff --git a/drivers/char/dsp56k.c b/drivers/char/dsp56k.c index 06749e295ada..b3eaf3e5ef2e 100644 --- a/drivers/char/dsp56k.c +++ b/drivers/char/dsp56k.c @@ -504,7 +504,7 @@ static int __init dsp56k_init_driver(void) printk("DSP56k driver: Unable to register driver\n"); return -ENODEV; } - dsp56k_class = class_create(THIS_MODULE, "dsp56k"); + dsp56k_class = class_create("dsp56k"); if (IS_ERR(dsp56k_class)) { err = PTR_ERR(dsp56k_class); goto out_chrdev; diff --git a/drivers/char/ipmi/ipmi_devintf.c b/drivers/char/ipmi/ipmi_devintf.c index d160fa4c73fe..73e5a9e28f85 100644 --- a/drivers/char/ipmi/ipmi_devintf.c +++ b/drivers/char/ipmi/ipmi_devintf.c @@ -860,7 +860,7 @@ static int __init init_ipmi_devintf(void) pr_info("ipmi device interface\n"); - ipmi_class = class_create(THIS_MODULE, "ipmi"); + ipmi_class = class_create("ipmi"); if (IS_ERR(ipmi_class)) { pr_err("ipmi: can't register device class\n"); return PTR_ERR(ipmi_class); diff --git a/drivers/char/lp.c b/drivers/char/lp.c index 38aad99ebb61..70cfc5140c2c 100644 --- a/drivers/char/lp.c +++ b/drivers/char/lp.c @@ -1049,7 +1049,7 @@ static int __init lp_init(void) return -EIO; } - lp_class = class_create(THIS_MODULE, "printer"); + lp_class = class_create("printer"); if (IS_ERR(lp_class)) { err = PTR_ERR(lp_class); goto out_reg; diff --git a/drivers/char/mem.c b/drivers/char/mem.c index ffb101d349f0..f494d31f2b98 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -762,7 +762,7 @@ static int __init chr_dev_init(void) if (register_chrdev(MEM_MAJOR, "mem", &memory_fops)) printk("unable to get major %d for memory devs\n", MEM_MAJOR); - mem_class = class_create(THIS_MODULE, "mem"); + mem_class = class_create("mem"); if (IS_ERR(mem_class)) return PTR_ERR(mem_class); diff --git a/drivers/char/misc.c b/drivers/char/misc.c index 7a1388b0572b..1c44c29a666e 100644 --- a/drivers/char/misc.c +++ b/drivers/char/misc.c @@ -286,7 +286,7 @@ static int __init misc_init(void) struct proc_dir_entry *ret; ret = proc_create_seq("misc", 0, NULL, &misc_seq_ops); - misc_class = class_create(THIS_MODULE, "misc"); + misc_class = class_create("misc"); err = PTR_ERR(misc_class); if (IS_ERR(misc_class)) goto fail_remove; diff --git a/drivers/char/pcmcia/cm4000_cs.c b/drivers/char/pcmcia/cm4000_cs.c index e656f42a28ac..7f96d8571a53 100644 --- a/drivers/char/pcmcia/cm4000_cs.c +++ b/drivers/char/pcmcia/cm4000_cs.c @@ -1878,7 +1878,7 @@ static int __init cmm_init(void) { int rc; - cmm_class = class_create(THIS_MODULE, "cardman_4000"); + cmm_class = class_create("cardman_4000"); if (IS_ERR(cmm_class)) return PTR_ERR(cmm_class); diff --git a/drivers/char/pcmcia/cm4040_cs.c b/drivers/char/pcmcia/cm4040_cs.c index 827711911da4..11ff59e2b963 100644 --- a/drivers/char/pcmcia/cm4040_cs.c +++ b/drivers/char/pcmcia/cm4040_cs.c @@ -650,7 +650,7 @@ static int __init cm4040_init(void) { int rc; - cmx_class = class_create(THIS_MODULE, "cardman_4040"); + cmx_class = class_create("cardman_4040"); if (IS_ERR(cmx_class)) return PTR_ERR(cmx_class); diff --git a/drivers/char/pcmcia/scr24x_cs.c b/drivers/char/pcmcia/scr24x_cs.c index 1bdce08fae3d..870781f5a08c 100644 --- a/drivers/char/pcmcia/scr24x_cs.c +++ b/drivers/char/pcmcia/scr24x_cs.c @@ -325,7 +325,7 @@ static int __init scr24x_init(void) { int ret; - scr24x_class = class_create(THIS_MODULE, "scr24x"); + scr24x_class = class_create("scr24x"); if (IS_ERR(scr24x_class)) return PTR_ERR(scr24x_class); diff --git a/drivers/char/ppdev.c b/drivers/char/ppdev.c index 38b46c7d1737..81ed58157b15 100644 --- a/drivers/char/ppdev.c +++ b/drivers/char/ppdev.c @@ -841,7 +841,7 @@ static int __init ppdev_init(void) pr_warn(CHRDEV ": unable to get major %d\n", PP_MAJOR); return -EIO; } - ppdev_class = class_create(THIS_MODULE, CHRDEV); + ppdev_class = class_create(CHRDEV); if (IS_ERR(ppdev_class)) { err = PTR_ERR(ppdev_class); goto out_chrdev; diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c index 7e513b771832..8763c820d1f8 100644 --- a/drivers/char/tpm/tpm-interface.c +++ b/drivers/char/tpm/tpm-interface.c @@ -466,13 +466,13 @@ static int __init tpm_init(void) { int rc; - tpm_class = class_create(THIS_MODULE, "tpm"); + tpm_class = class_create("tpm"); if (IS_ERR(tpm_class)) { pr_err("couldn't create tpm class\n"); return PTR_ERR(tpm_class); } - tpmrm_class = class_create(THIS_MODULE, "tpmrm"); + tpmrm_class = class_create("tpmrm"); if (IS_ERR(tpmrm_class)) { pr_err("couldn't create tpmrm class\n"); rc = PTR_ERR(tpmrm_class); diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c index d5ac4d955bc8..b65c809a4e97 100644 --- a/drivers/char/virtio_console.c +++ b/drivers/char/virtio_console.c @@ -2244,7 +2244,7 @@ static int __init virtio_console_init(void) { int err; - pdrvdata.class = class_create(THIS_MODULE, "virtio-ports"); + pdrvdata.class = class_create("virtio-ports"); if (IS_ERR(pdrvdata.class)) { err = PTR_ERR(pdrvdata.class); pr_err("Error %d creating virtio-ports class\n", err); diff --git a/drivers/char/xilinx_hwicap/xilinx_hwicap.c b/drivers/char/xilinx_hwicap/xilinx_hwicap.c index 74a4928aea1d..a46f637da959 100644 --- a/drivers/char/xilinx_hwicap/xilinx_hwicap.c +++ b/drivers/char/xilinx_hwicap/xilinx_hwicap.c @@ -856,7 +856,7 @@ static int __init hwicap_module_init(void) dev_t devt; int retval; - icap_class = class_create(THIS_MODULE, "xilinx_config"); + icap_class = class_create("xilinx_config"); mutex_init(&icap_sem); devt = MKDEV(XHWICAP_MAJOR, XHWICAP_MINOR); diff --git a/drivers/char/xillybus/xillybus_class.c b/drivers/char/xillybus/xillybus_class.c index e9a288e61c15..89926fe9d813 100644 --- a/drivers/char/xillybus/xillybus_class.c +++ b/drivers/char/xillybus/xillybus_class.c @@ -242,7 +242,7 @@ EXPORT_SYMBOL(xillybus_find_inode); static int __init xillybus_class_init(void) { - xillybus_class = class_create(THIS_MODULE, "xillybus"); + xillybus_class = class_create("xillybus"); if (IS_ERR(xillybus_class)) { pr_warn("Failed to register xillybus class\n"); diff --git a/drivers/comedi/comedi_fops.c b/drivers/comedi/comedi_fops.c index b982903aaa46..8e43918d38c4 100644 --- a/drivers/comedi/comedi_fops.c +++ b/drivers/comedi/comedi_fops.c @@ -3383,7 +3383,7 @@ static int __init comedi_init(void) if (retval) goto out_unregister_chrdev_region; - comedi_class = class_create(THIS_MODULE, "comedi"); + comedi_class = class_create("comedi"); if (IS_ERR(comedi_class)) { retval = PTR_ERR(comedi_class); pr_err("failed to create class\n"); diff --git a/drivers/comedi/drivers/comedi_test.c b/drivers/comedi/drivers/comedi_test.c index 0b5c0af1cebf..c02dc19a679b 100644 --- a/drivers/comedi/drivers/comedi_test.c +++ b/drivers/comedi/drivers/comedi_test.c @@ -795,7 +795,7 @@ static int __init comedi_test_init(void) } if (!config_mode) { - ctcls = class_create(THIS_MODULE, CLASS_NAME); + ctcls = class_create(CLASS_NAME); if (IS_ERR(ctcls)) { pr_warn("comedi_test: unable to create class\n"); goto clean3; diff --git a/drivers/crypto/qat/qat_common/adf_ctl_drv.c b/drivers/crypto/qat/qat_common/adf_ctl_drv.c index 9190532b27eb..75a50fde5644 100644 --- a/drivers/crypto/qat/qat_common/adf_ctl_drv.c +++ b/drivers/crypto/qat/qat_common/adf_ctl_drv.c @@ -56,7 +56,7 @@ static int adf_chr_drv_create(void) return -EFAULT; } - adf_ctl_drv.drv_class = class_create(THIS_MODULE, DEVICE_NAME); + adf_ctl_drv.drv_class = class_create(DEVICE_NAME); if (IS_ERR(adf_ctl_drv.drv_class)) { pr_err("QAT: class_create failed for adf_ctl\n"); goto err_chrdev_unreg; diff --git a/drivers/dca/dca-sysfs.c b/drivers/dca/dca-sysfs.c index 21ebd0af268b..fcc83ede0909 100644 --- a/drivers/dca/dca-sysfs.c +++ b/drivers/dca/dca-sysfs.c @@ -74,7 +74,7 @@ int __init dca_sysfs_init(void) idr_init(&dca_idr); spin_lock_init(&dca_idr_lock); - dca_class = class_create(THIS_MODULE, "dca"); + dca_class = class_create("dca"); if (IS_ERR(dca_class)) { idr_destroy(&dca_idr); return PTR_ERR(dca_class); diff --git a/drivers/devfreq/devfreq-event.c b/drivers/devfreq/devfreq-event.c index f041edccd107..3ebac2496679 100644 --- a/drivers/devfreq/devfreq-event.c +++ b/drivers/devfreq/devfreq-event.c @@ -469,7 +469,7 @@ ATTRIBUTE_GROUPS(devfreq_event); static int __init devfreq_event_init(void) { - devfreq_event_class = class_create(THIS_MODULE, "devfreq-event"); + devfreq_event_class = class_create("devfreq-event"); if (IS_ERR(devfreq_event_class)) { pr_err("%s: couldn't create class\n", __FILE__); return PTR_ERR(devfreq_event_class); diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c index 817c71da391a..e36cbb920ec8 100644 --- a/drivers/devfreq/devfreq.c +++ b/drivers/devfreq/devfreq.c @@ -1988,7 +1988,7 @@ DEFINE_SHOW_ATTRIBUTE(devfreq_summary); static int __init devfreq_init(void) { - devfreq_class = class_create(THIS_MODULE, "devfreq"); + devfreq_class = class_create("devfreq"); if (IS_ERR(devfreq_class)) { pr_err("%s: couldn't create class\n", __FILE__); return PTR_ERR(devfreq_class); diff --git a/drivers/dma-buf/dma-heap.c b/drivers/dma-buf/dma-heap.c index c9e41e8a9e27..84ae708fafe7 100644 --- a/drivers/dma-buf/dma-heap.c +++ b/drivers/dma-buf/dma-heap.c @@ -314,7 +314,7 @@ static int dma_heap_init(void) if (ret) return ret; - dma_heap_class = class_create(THIS_MODULE, DEVNAME); + dma_heap_class = class_create(DEVNAME); if (IS_ERR(dma_heap_class)) { unregister_chrdev_region(dma_heap_devt, NUM_HEAP_MINORS); return PTR_ERR(dma_heap_class); diff --git a/drivers/extcon/extcon.c b/drivers/extcon/extcon.c index e1c71359b605..d43ba8e7260d 100644 --- a/drivers/extcon/extcon.c +++ b/drivers/extcon/extcon.c @@ -1013,7 +1013,7 @@ ATTRIBUTE_GROUPS(extcon); static int create_extcon_class(void) { if (!extcon_class) { - extcon_class = class_create(THIS_MODULE, "extcon"); + extcon_class = class_create("extcon"); if (IS_ERR(extcon_class)) return PTR_ERR(extcon_class); extcon_class->dev_groups = extcon_groups; diff --git a/drivers/fpga/fpga-bridge.c b/drivers/fpga/fpga-bridge.c index 5cd40acab5bf..6a521c83a1ed 100644 --- a/drivers/fpga/fpga-bridge.c +++ b/drivers/fpga/fpga-bridge.c @@ -416,7 +416,7 @@ static void fpga_bridge_dev_release(struct device *dev) static int __init fpga_bridge_dev_init(void) { - fpga_bridge_class = class_create(THIS_MODULE, "fpga_bridge"); + fpga_bridge_class = class_create("fpga_bridge"); if (IS_ERR(fpga_bridge_class)) return PTR_ERR(fpga_bridge_class); diff --git a/drivers/fpga/fpga-mgr.c b/drivers/fpga/fpga-mgr.c index 8efa67620e21..eb583f86a0b9 100644 --- a/drivers/fpga/fpga-mgr.c +++ b/drivers/fpga/fpga-mgr.c @@ -971,7 +971,7 @@ static int __init fpga_mgr_class_init(void) { pr_info("FPGA manager framework\n"); - fpga_mgr_class = class_create(THIS_MODULE, "fpga_manager"); + fpga_mgr_class = class_create("fpga_manager"); if (IS_ERR(fpga_mgr_class)) return PTR_ERR(fpga_mgr_class); diff --git a/drivers/fpga/fpga-region.c b/drivers/fpga/fpga-region.c index 27ff9dea04ae..ccf6fdab1360 100644 --- a/drivers/fpga/fpga-region.c +++ b/drivers/fpga/fpga-region.c @@ -293,7 +293,7 @@ static void fpga_region_dev_release(struct device *dev) */ static int __init fpga_region_init(void) { - fpga_region_class = class_create(THIS_MODULE, "fpga_region"); + fpga_region_class = class_create("fpga_region"); if (IS_ERR(fpga_region_class)) return PTR_ERR(fpga_region_class); diff --git a/drivers/gnss/core.c b/drivers/gnss/core.c index 77a4b280c552..48f2ee0f78c4 100644 --- a/drivers/gnss/core.c +++ b/drivers/gnss/core.c @@ -387,7 +387,7 @@ static int __init gnss_module_init(void) return ret; } - gnss_class = class_create(THIS_MODULE, "gnss"); + gnss_class = class_create("gnss"); if (IS_ERR(gnss_class)) { ret = PTR_ERR(gnss_class); pr_err("failed to create class: %d\n", ret); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index a0e30f21e12e..20e75bd74bed 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -93,7 +93,7 @@ int kfd_chardev_init(void) if (err < 0) goto err_register_chrdev; - kfd_class = class_create(THIS_MODULE, kfd_dev_name); + kfd_class = class_create(kfd_dev_name); err = PTR_ERR(kfd_class); if (IS_ERR(kfd_class)) goto err_class_create; diff --git a/drivers/gpu/drm/display/drm_dp_aux_dev.c b/drivers/gpu/drm/display/drm_dp_aux_dev.c index 098e482e65a2..29555b9f03c8 100644 --- a/drivers/gpu/drm/display/drm_dp_aux_dev.c +++ b/drivers/gpu/drm/display/drm_dp_aux_dev.c @@ -330,7 +330,7 @@ int drm_dp_aux_dev_init(void) { int res; - drm_dp_aux_dev_class = class_create(THIS_MODULE, "drm_dp_aux_dev"); + drm_dp_aux_dev_class = class_create("drm_dp_aux_dev"); if (IS_ERR(drm_dp_aux_dev_class)) { return PTR_ERR(drm_dp_aux_dev_class); } diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c index 183130355997..3c22a803201d 100644 --- a/drivers/gpu/drm/drm_sysfs.c +++ b/drivers/gpu/drm/drm_sysfs.c @@ -112,7 +112,7 @@ int drm_sysfs_init(void) { int err; - drm_class = class_create(THIS_MODULE, "drm"); + drm_class = class_create("drm"); if (IS_ERR(drm_class)) return PTR_ERR(drm_class); diff --git a/drivers/hid/hid-roccat-arvo.c b/drivers/hid/hid-roccat-arvo.c index d94ee0539421..ea6b79b3aeeb 100644 --- a/drivers/hid/hid-roccat-arvo.c +++ b/drivers/hid/hid-roccat-arvo.c @@ -433,7 +433,7 @@ static int __init arvo_init(void) { int retval; - arvo_class = class_create(THIS_MODULE, "arvo"); + arvo_class = class_create("arvo"); if (IS_ERR(arvo_class)) return PTR_ERR(arvo_class); arvo_class->dev_groups = arvo_groups; diff --git a/drivers/hid/hid-roccat-isku.c b/drivers/hid/hid-roccat-isku.c index e95d59cd8d07..3903a2cea00c 100644 --- a/drivers/hid/hid-roccat-isku.c +++ b/drivers/hid/hid-roccat-isku.c @@ -435,7 +435,7 @@ static struct hid_driver isku_driver = { static int __init isku_init(void) { int retval; - isku_class = class_create(THIS_MODULE, "isku"); + isku_class = class_create("isku"); if (IS_ERR(isku_class)) return PTR_ERR(isku_class); isku_class->dev_groups = isku_groups; diff --git a/drivers/hid/hid-roccat-kone.c b/drivers/hid/hid-roccat-kone.c index 76da04801ca9..945ae236fb45 100644 --- a/drivers/hid/hid-roccat-kone.c +++ b/drivers/hid/hid-roccat-kone.c @@ -890,7 +890,7 @@ static int __init kone_init(void) int retval; /* class name has to be same as driver name */ - kone_class = class_create(THIS_MODULE, "kone"); + kone_class = class_create("kone"); if (IS_ERR(kone_class)) return PTR_ERR(kone_class); kone_class->dev_groups = kone_groups; diff --git a/drivers/hid/hid-roccat-koneplus.c b/drivers/hid/hid-roccat-koneplus.c index 1896c69ea512..97b83b6f53dd 100644 --- a/drivers/hid/hid-roccat-koneplus.c +++ b/drivers/hid/hid-roccat-koneplus.c @@ -549,7 +549,7 @@ static int __init koneplus_init(void) int retval; /* class name has to be same as driver name */ - koneplus_class = class_create(THIS_MODULE, "koneplus"); + koneplus_class = class_create("koneplus"); if (IS_ERR(koneplus_class)) return PTR_ERR(koneplus_class); koneplus_class->dev_groups = koneplus_groups; diff --git a/drivers/hid/hid-roccat-konepure.c b/drivers/hid/hid-roccat-konepure.c index cf8eeb33a125..a297756f2410 100644 --- a/drivers/hid/hid-roccat-konepure.c +++ b/drivers/hid/hid-roccat-konepure.c @@ -207,7 +207,7 @@ static int __init konepure_init(void) { int retval; - konepure_class = class_create(THIS_MODULE, "konepure"); + konepure_class = class_create("konepure"); if (IS_ERR(konepure_class)) return PTR_ERR(konepure_class); konepure_class->dev_groups = konepure_groups; diff --git a/drivers/hid/hid-roccat-kovaplus.c b/drivers/hid/hid-roccat-kovaplus.c index 6fb9b9563769..1a1d96e11683 100644 --- a/drivers/hid/hid-roccat-kovaplus.c +++ b/drivers/hid/hid-roccat-kovaplus.c @@ -638,7 +638,7 @@ static int __init kovaplus_init(void) { int retval; - kovaplus_class = class_create(THIS_MODULE, "kovaplus"); + kovaplus_class = class_create("kovaplus"); if (IS_ERR(kovaplus_class)) return PTR_ERR(kovaplus_class); kovaplus_class->dev_groups = kovaplus_groups; diff --git a/drivers/hid/hid-roccat-pyra.c b/drivers/hid/hid-roccat-pyra.c index 4fcc8e7d276f..15528c3b013c 100644 --- a/drivers/hid/hid-roccat-pyra.c +++ b/drivers/hid/hid-roccat-pyra.c @@ -585,7 +585,7 @@ static int __init pyra_init(void) int retval; /* class name has to be same as driver name */ - pyra_class = class_create(THIS_MODULE, "pyra"); + pyra_class = class_create("pyra"); if (IS_ERR(pyra_class)) return PTR_ERR(pyra_class); pyra_class->dev_groups = pyra_groups; diff --git a/drivers/hid/hid-roccat-ryos.c b/drivers/hid/hid-roccat-ryos.c index 5bf1971a2b14..0eb17a3b925d 100644 --- a/drivers/hid/hid-roccat-ryos.c +++ b/drivers/hid/hid-roccat-ryos.c @@ -216,7 +216,7 @@ static int __init ryos_init(void) { int retval; - ryos_class = class_create(THIS_MODULE, "ryos"); + ryos_class = class_create("ryos"); if (IS_ERR(ryos_class)) return PTR_ERR(ryos_class); ryos_class->dev_groups = ryos_groups; diff --git a/drivers/hid/hid-roccat-savu.c b/drivers/hid/hid-roccat-savu.c index a784bb4ee651..93be7acef673 100644 --- a/drivers/hid/hid-roccat-savu.c +++ b/drivers/hid/hid-roccat-savu.c @@ -204,7 +204,7 @@ static int __init savu_init(void) { int retval; - savu_class = class_create(THIS_MODULE, "savu"); + savu_class = class_create("savu"); if (IS_ERR(savu_class)) return PTR_ERR(savu_class); savu_class->dev_groups = savu_groups; diff --git a/drivers/hid/hidraw.c b/drivers/hid/hidraw.c index 197b1e7bf029..93e62b161501 100644 --- a/drivers/hid/hidraw.c +++ b/drivers/hid/hidraw.c @@ -618,7 +618,7 @@ int __init hidraw_init(void) hidraw_major = MAJOR(dev_id); - hidraw_class = class_create(THIS_MODULE, "hidraw"); + hidraw_class = class_create("hidraw"); if (IS_ERR(hidraw_class)) { result = PTR_ERR(hidraw_class); goto error_cdev; diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c index 107623c4cc14..ef93064cfdb3 100644 --- a/drivers/i2c/i2c-dev.c +++ b/drivers/i2c/i2c-dev.c @@ -739,7 +739,7 @@ static int __init i2c_dev_init(void) if (res) goto out; - i2c_dev_class = class_create(THIS_MODULE, "i2c-dev"); + i2c_dev_class = class_create("i2c-dev"); if (IS_ERR(i2c_dev_class)) { res = PTR_ERR(i2c_dev_class); goto out_unreg_chrdev; diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index bdb179a09d77..fbace69672ca 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -1264,7 +1264,7 @@ static int __init ib_uverbs_init(void) goto out_alloc; } - uverbs_class = class_create(THIS_MODULE, "infiniband_verbs"); + uverbs_class = class_create("infiniband_verbs"); if (IS_ERR(uverbs_class)) { ret = PTR_ERR(uverbs_class); pr_err("user_verbs: couldn't create class infiniband_verbs\n"); diff --git a/drivers/infiniband/hw/hfi1/device.c b/drivers/infiniband/hw/hfi1/device.c index 1f4496032170..05be0d119f79 100644 --- a/drivers/infiniband/hw/hfi1/device.c +++ b/drivers/infiniband/hw/hfi1/device.c @@ -102,7 +102,7 @@ int __init dev_init(void) goto done; } - class = class_create(THIS_MODULE, class_name()); + class = class_create(class_name()); if (IS_ERR(class)) { ret = PTR_ERR(class); pr_err("Could not create device class (err %d)\n", -ret); @@ -111,7 +111,7 @@ int __init dev_init(void) } class->devnode = hfi1_devnode; - user_class = class_create(THIS_MODULE, class_name_user()); + user_class = class_create(class_name_user()); if (IS_ERR(user_class)) { ret = PTR_ERR(user_class); pr_err("Could not create device class for user accessible files (err %d)\n", diff --git a/drivers/infiniband/hw/qib/qib_file_ops.c b/drivers/infiniband/hw/qib/qib_file_ops.c index 80fe92a21f96..c07c95fd3cfa 100644 --- a/drivers/infiniband/hw/qib/qib_file_ops.c +++ b/drivers/infiniband/hw/qib/qib_file_ops.c @@ -2326,7 +2326,7 @@ int __init qib_dev_init(void) goto done; } - qib_class = class_create(THIS_MODULE, "ipath"); + qib_class = class_create("ipath"); if (IS_ERR(qib_class)) { ret = PTR_ERR(qib_class); pr_err("Could not create device class (err %d)\n", -ret); diff --git a/drivers/infiniband/ulp/rtrs/rtrs-clt.c b/drivers/infiniband/ulp/rtrs/rtrs-clt.c index 80abf45a197a..edb2e3a25880 100644 --- a/drivers/infiniband/ulp/rtrs/rtrs-clt.c +++ b/drivers/infiniband/ulp/rtrs/rtrs-clt.c @@ -3163,7 +3163,7 @@ static int __init rtrs_client_init(void) { rtrs_rdma_dev_pd_init(0, &dev_pd); - rtrs_clt_dev_class = class_create(THIS_MODULE, "rtrs-client"); + rtrs_clt_dev_class = class_create("rtrs-client"); if (IS_ERR(rtrs_clt_dev_class)) { pr_err("Failed to create rtrs-client dev class\n"); return PTR_ERR(rtrs_clt_dev_class); diff --git a/drivers/infiniband/ulp/rtrs/rtrs-srv.c b/drivers/infiniband/ulp/rtrs/rtrs-srv.c index d1703e2c0b82..c38901e2c8f4 100644 --- a/drivers/infiniband/ulp/rtrs/rtrs-srv.c +++ b/drivers/infiniband/ulp/rtrs/rtrs-srv.c @@ -2253,7 +2253,7 @@ static int __init rtrs_server_init(void) err); return err; } - rtrs_dev_class = class_create(THIS_MODULE, "rtrs-server"); + rtrs_dev_class = class_create("rtrs-server"); if (IS_ERR(rtrs_dev_class)) { err = PTR_ERR(rtrs_dev_class); goto out_err; diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c index 0f00be62438d..45a4043c5042 100644 --- a/drivers/isdn/capi/capi.c +++ b/drivers/isdn/capi/capi.c @@ -1393,7 +1393,7 @@ static int __init capi_init(void) kcapi_exit(); return major_ret; } - capi_class = class_create(THIS_MODULE, "capi"); + capi_class = class_create("capi"); if (IS_ERR(capi_class)) { unregister_chrdev(capi_major, "capi20"); kcapi_exit(); diff --git a/drivers/isdn/mISDN/dsp_pipeline.c b/drivers/isdn/mISDN/dsp_pipeline.c index cfbcd9e973c2..09b72f14d4b7 100644 --- a/drivers/isdn/mISDN/dsp_pipeline.c +++ b/drivers/isdn/mISDN/dsp_pipeline.c @@ -131,7 +131,7 @@ EXPORT_SYMBOL(mISDN_dsp_element_unregister); int dsp_pipeline_module_init(void) { - elements_class = class_create(THIS_MODULE, "dsp_pipeline"); + elements_class = class_create("dsp_pipeline"); if (IS_ERR(elements_class)) return PTR_ERR(elements_class); diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c index a6b3adcd044a..9255bc11f99d 100644 --- a/drivers/leds/led-class.c +++ b/drivers/leds/led-class.c @@ -626,7 +626,7 @@ EXPORT_SYMBOL_GPL(devm_led_classdev_unregister); static int __init leds_init(void) { - leds_class = class_create(THIS_MODULE, "leds"); + leds_class = class_create("leds"); if (IS_ERR(leds_class)) return PTR_ERR(leds_class); leds_class->pm = &leds_class_dev_pm_ops; diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c index 23bd0c77ac1a..57e987cf84b2 100644 --- a/drivers/macintosh/adb.c +++ b/drivers/macintosh/adb.c @@ -888,7 +888,7 @@ adbdev_init(void) return; } - adb_dev_class = class_create(THIS_MODULE, "adb"); + adb_dev_class = class_create("adb"); if (IS_ERR(adb_dev_class)) return; device_create(adb_dev_class, NULL, MKDEV(ADB_MAJOR, 0), NULL, "adb"); diff --git a/drivers/media/dvb-core/dvbdev.c b/drivers/media/dvb-core/dvbdev.c index 0ed087caf7f3..e9b3ce09e534 100644 --- a/drivers/media/dvb-core/dvbdev.c +++ b/drivers/media/dvb-core/dvbdev.c @@ -1063,7 +1063,7 @@ static int __init init_dvbdev(void) goto error; } - dvb_class = class_create(THIS_MODULE, "dvb"); + dvb_class = class_create("dvb"); if (IS_ERR(dvb_class)) { retval = PTR_ERR(dvb_class); goto error; diff --git a/drivers/media/rc/lirc_dev.c b/drivers/media/rc/lirc_dev.c index 25ab61dae126..043d23aaa3cb 100644 --- a/drivers/media/rc/lirc_dev.c +++ b/drivers/media/rc/lirc_dev.c @@ -785,7 +785,7 @@ int __init lirc_dev_init(void) { int retval; - lirc_class = class_create(THIS_MODULE, "lirc"); + lirc_class = class_create("lirc"); if (IS_ERR(lirc_class)) { pr_err("class_create failed\n"); return PTR_ERR(lirc_class); diff --git a/drivers/misc/c2port/core.c b/drivers/misc/c2port/core.c index fb9a1b49ff6d..f574c83b82cf 100644 --- a/drivers/misc/c2port/core.c +++ b/drivers/misc/c2port/core.c @@ -977,7 +977,7 @@ static int __init c2port_init(void) printk(KERN_INFO "Silicon Labs C2 port support v. " DRIVER_VERSION " - (C) 2007 Rodolfo Giometti\n"); - c2port_class = class_create(THIS_MODULE, "c2port"); + c2port_class = class_create("c2port"); if (IS_ERR(c2port_class)) { printk(KERN_ERR "c2port: failed to allocate class\n"); return PTR_ERR(c2port_class); diff --git a/drivers/misc/cxl/file.c b/drivers/misc/cxl/file.c index 5878329b011a..144d1f2d78ce 100644 --- a/drivers/misc/cxl/file.c +++ b/drivers/misc/cxl/file.c @@ -678,7 +678,7 @@ int __init cxl_file_init(void) pr_devel("CXL device allocated, MAJOR %i\n", MAJOR(cxl_dev)); - cxl_class = class_create(THIS_MODULE, "cxl"); + cxl_class = class_create("cxl"); if (IS_ERR(cxl_class)) { pr_err("Unable to create CXL class\n"); rc = PTR_ERR(cxl_class); diff --git a/drivers/misc/genwqe/card_base.c b/drivers/misc/genwqe/card_base.c index 5b63d179b24e..02628288cd0f 100644 --- a/drivers/misc/genwqe/card_base.c +++ b/drivers/misc/genwqe/card_base.c @@ -1363,7 +1363,7 @@ static int __init genwqe_init_module(void) { int rc; - class_genwqe = class_create(THIS_MODULE, GENWQE_DEVNAME); + class_genwqe = class_create(GENWQE_DEVNAME); if (IS_ERR(class_genwqe)) { pr_err("[%s] create class failed\n", __func__); return -ENOMEM; diff --git a/drivers/misc/hpilo.c b/drivers/misc/hpilo.c index 8d00df9243c4..2c3a991d6e88 100644 --- a/drivers/misc/hpilo.c +++ b/drivers/misc/hpilo.c @@ -888,7 +888,7 @@ static int __init ilo_init(void) int error; dev_t dev; - ilo_class = class_create(THIS_MODULE, "iLO"); + ilo_class = class_create("iLO"); if (IS_ERR(ilo_class)) { error = PTR_ERR(ilo_class); goto out; diff --git a/drivers/misc/mei/main.c b/drivers/misc/mei/main.c index 632d4ae21e46..76c771a424f7 100644 --- a/drivers/misc/mei/main.c +++ b/drivers/misc/mei/main.c @@ -1275,7 +1275,7 @@ static int __init mei_init(void) { int ret; - mei_class = class_create(THIS_MODULE, "mei"); + mei_class = class_create("mei"); if (IS_ERR(mei_class)) { pr_err("couldn't create class\n"); ret = PTR_ERR(mei_class); diff --git a/drivers/misc/ocxl/file.c b/drivers/misc/ocxl/file.c index 3b058654b45b..6e63f060e4cc 100644 --- a/drivers/misc/ocxl/file.c +++ b/drivers/misc/ocxl/file.c @@ -601,7 +601,7 @@ int ocxl_file_init(void) return rc; } - ocxl_class = class_create(THIS_MODULE, "ocxl"); + ocxl_class = class_create("ocxl"); if (IS_ERR(ocxl_class)) { pr_err("Unable to create ocxl class\n"); unregister_chrdev_region(ocxl_dev, OCXL_NUM_MINORS); diff --git a/drivers/misc/phantom.c b/drivers/misc/phantom.c index ce72e46a2e73..7966a6b8b5b3 100644 --- a/drivers/misc/phantom.c +++ b/drivers/misc/phantom.c @@ -503,7 +503,7 @@ static int __init phantom_init(void) int retval; dev_t dev; - phantom_class = class_create(THIS_MODULE, "phantom"); + phantom_class = class_create("phantom"); if (IS_ERR(phantom_class)) { retval = PTR_ERR(phantom_class); printk(KERN_ERR "phantom: can't register phantom class\n"); diff --git a/drivers/misc/uacce/uacce.c b/drivers/misc/uacce/uacce.c index 07023397afc7..346bd7cf2e94 100644 --- a/drivers/misc/uacce/uacce.c +++ b/drivers/misc/uacce/uacce.c @@ -620,7 +620,7 @@ static int __init uacce_init(void) { int ret; - uacce_class = class_create(THIS_MODULE, UACCE_NAME); + uacce_class = class_create(UACCE_NAME); if (IS_ERR(uacce_class)) return PTR_ERR(uacce_class); diff --git a/drivers/most/most_cdev.c b/drivers/most/most_cdev.c index 4ee536980f71..3ed8f461e01e 100644 --- a/drivers/most/most_cdev.c +++ b/drivers/most/most_cdev.c @@ -491,7 +491,7 @@ static int __init most_cdev_init(void) { int err; - comp.class = class_create(THIS_MODULE, "most_cdev"); + comp.class = class_create("most_cdev"); if (IS_ERR(comp.class)) return PTR_ERR(comp.class); diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.c b/drivers/net/ethernet/hisilicon/hns/hnae.c index 9b26f0f2c748..8a1027ad340d 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.c +++ b/drivers/net/ethernet/hisilicon/hns/hnae.c @@ -448,7 +448,7 @@ EXPORT_SYMBOL(hnae_ae_unregister); static int __init hnae_init(void) { - hnae_class = class_create(THIS_MODULE, "hnae"); + hnae_class = class_create("hnae"); return PTR_ERR_OR_ZERO(hnae_class); } diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index 1d71f5276241..a9beacd552cf 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -1394,7 +1394,7 @@ static int __init ppp_init(void) goto out_net; } - ppp_class = class_create(THIS_MODULE, "ppp"); + ppp_class = class_create("ppp"); if (IS_ERR(ppp_class)) { err = PTR_ERR(ppp_class); goto out_chrdev; diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 4cc4eaf80b14..68a7e09a195e 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -5748,7 +5748,7 @@ static int __init init_mac80211_hwsim(void) if (err) goto out_exit_netlink; - hwsim_class = class_create(THIS_MODULE, "mac80211_hwsim"); + hwsim_class = class_create("mac80211_hwsim"); if (IS_ERR(hwsim_class)) { err = PTR_ERR(hwsim_class); goto out_exit_virtio; diff --git a/drivers/net/wwan/wwan_core.c b/drivers/net/wwan/wwan_core.c index 966d0ccd2276..51bbd6bb74b5 100644 --- a/drivers/net/wwan/wwan_core.c +++ b/drivers/net/wwan/wwan_core.c @@ -1174,7 +1174,7 @@ static int __init wwan_init(void) if (err) return err; - wwan_class = class_create(THIS_MODULE, "wwan"); + wwan_class = class_create("wwan"); if (IS_ERR(wwan_class)) { err = PTR_ERR(wwan_class); goto unregister; diff --git a/drivers/net/wwan/wwan_hwsim.c b/drivers/net/wwan/wwan_hwsim.c index 2397a903d8f5..edb5fc803211 100644 --- a/drivers/net/wwan/wwan_hwsim.c +++ b/drivers/net/wwan/wwan_hwsim.c @@ -511,7 +511,7 @@ static int __init wwan_hwsim_init(void) if (!wwan_wq) return -ENOMEM; - wwan_hwsim_class = class_create(THIS_MODULE, "wwan_hwsim"); + wwan_hwsim_class = class_create("wwan_hwsim"); if (IS_ERR(wwan_hwsim_class)) { err = PTR_ERR(wwan_hwsim_class); goto err_wq_destroy; diff --git a/drivers/nvdimm/bus.c b/drivers/nvdimm/bus.c index 4976a0069e9c..954dbc105fc8 100644 --- a/drivers/nvdimm/bus.c +++ b/drivers/nvdimm/bus.c @@ -1320,7 +1320,7 @@ int __init nvdimm_bus_init(void) goto err_dimm_chrdev; nvdimm_major = rc; - nd_class = class_create(THIS_MODULE, "nd"); + nd_class = class_create("nd"); if (IS_ERR(nd_class)) { rc = PTR_ERR(nd_class); goto err_class; diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index c2730b116dc6..877a61fbd78b 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -5381,14 +5381,14 @@ static int __init nvme_core_init(void) if (result < 0) goto destroy_delete_wq; - nvme_class = class_create(THIS_MODULE, "nvme"); + nvme_class = class_create("nvme"); if (IS_ERR(nvme_class)) { result = PTR_ERR(nvme_class); goto unregister_chrdev; } nvme_class->dev_uevent = nvme_class_uevent; - nvme_subsys_class = class_create(THIS_MODULE, "nvme-subsystem"); + nvme_subsys_class = class_create("nvme-subsystem"); if (IS_ERR(nvme_subsys_class)) { result = PTR_ERR(nvme_subsys_class); goto destroy_class; @@ -5399,7 +5399,7 @@ static int __init nvme_core_init(void) if (result < 0) goto destroy_subsys_class; - nvme_ns_chr_class = class_create(THIS_MODULE, "nvme-generic"); + nvme_ns_chr_class = class_create("nvme-generic"); if (IS_ERR(nvme_ns_chr_class)) { result = PTR_ERR(nvme_ns_chr_class); goto unregister_generic_ns; diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c index bbaa04a0c502..0069ebff85df 100644 --- a/drivers/nvme/host/fabrics.c +++ b/drivers/nvme/host/fabrics.c @@ -1254,7 +1254,7 @@ static int __init nvmf_init(void) if (!nvmf_default_host) return -ENOMEM; - nvmf_class = class_create(THIS_MODULE, "nvme-fabrics"); + nvmf_class = class_create("nvme-fabrics"); if (IS_ERR(nvmf_class)) { pr_err("couldn't register class nvme-fabrics\n"); ret = PTR_ERR(nvmf_class); diff --git a/drivers/nvme/target/fcloop.c b/drivers/nvme/target/fcloop.c index 5c16372f3b53..af0d01797e5e 100644 --- a/drivers/nvme/target/fcloop.c +++ b/drivers/nvme/target/fcloop.c @@ -1562,7 +1562,7 @@ static int __init fcloop_init(void) { int ret; - fcloop_class = class_create(THIS_MODULE, "fcloop"); + fcloop_class = class_create("fcloop"); if (IS_ERR(fcloop_class)) { pr_err("couldn't register class fcloop\n"); ret = PTR_ERR(fcloop_class); diff --git a/drivers/pci/endpoint/pci-epc-core.c b/drivers/pci/endpoint/pci-epc-core.c index 9440d9811eea..46c9a5c3ca14 100644 --- a/drivers/pci/endpoint/pci-epc-core.c +++ b/drivers/pci/endpoint/pci-epc-core.c @@ -860,7 +860,7 @@ EXPORT_SYMBOL_GPL(__devm_pci_epc_create); static int __init pci_epc_init(void) { - pci_epc_class = class_create(THIS_MODULE, "pci_epc"); + pci_epc_class = class_create("pci_epc"); if (IS_ERR(pci_epc_class)) { pr_err("failed to create pci epc class --> %ld\n", PTR_ERR(pci_epc_class)); diff --git a/drivers/pci/switch/switchtec.c b/drivers/pci/switch/switchtec.c index 3d6f17ff2429..d837da055921 100644 --- a/drivers/pci/switch/switchtec.c +++ b/drivers/pci/switch/switchtec.c @@ -1804,7 +1804,7 @@ static int __init switchtec_init(void) if (rc) return rc; - switchtec_class = class_create(THIS_MODULE, "switchtec"); + switchtec_class = class_create("switchtec"); if (IS_ERR(switchtec_class)) { rc = PTR_ERR(switchtec_class); goto err_create_class; diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c index 9951efc03eaa..6464dcb56d56 100644 --- a/drivers/phy/phy-core.c +++ b/drivers/phy/phy-core.c @@ -1233,7 +1233,7 @@ static void phy_release(struct device *dev) static int __init phy_core_init(void) { - phy_class = class_create(THIS_MODULE, "phy"); + phy_class = class_create("phy"); if (IS_ERR(phy_class)) { pr_err("failed to create phy class --> %ld\n", PTR_ERR(phy_class)); diff --git a/drivers/power/supply/power_supply_core.c b/drivers/power/supply/power_supply_core.c index f3d7c1da299f..1bf393239d40 100644 --- a/drivers/power/supply/power_supply_core.c +++ b/drivers/power/supply/power_supply_core.c @@ -1462,7 +1462,7 @@ EXPORT_SYMBOL_GPL(power_supply_get_drvdata); static int __init power_supply_class_init(void) { - power_supply_class = class_create(THIS_MODULE, "power_supply"); + power_supply_class = class_create("power_supply"); if (IS_ERR(power_supply_class)) return PTR_ERR(power_supply_class); diff --git a/drivers/pps/pps.c b/drivers/pps/pps.c index 22a65ad4e46e..5d19baae6a38 100644 --- a/drivers/pps/pps.c +++ b/drivers/pps/pps.c @@ -456,7 +456,7 @@ static int __init pps_init(void) { int err; - pps_class = class_create(THIS_MODULE, "pps"); + pps_class = class_create("pps"); if (IS_ERR(pps_class)) { pr_err("failed to allocate class\n"); return PTR_ERR(pps_class); diff --git a/drivers/ptp/ptp_clock.c b/drivers/ptp/ptp_clock.c index 62d4d29e7c05..790f9250b381 100644 --- a/drivers/ptp/ptp_clock.c +++ b/drivers/ptp/ptp_clock.c @@ -460,7 +460,7 @@ static int __init ptp_init(void) { int err; - ptp_class = class_create(THIS_MODULE, "ptp"); + ptp_class = class_create("ptp"); if (IS_ERR(ptp_class)) { pr_err("ptp: failed to allocate class\n"); return PTR_ERR(ptp_class); diff --git a/drivers/rapidio/devices/rio_mport_cdev.c b/drivers/rapidio/devices/rio_mport_cdev.c index 43db495f1986..deb96c3160a7 100644 --- a/drivers/rapidio/devices/rio_mport_cdev.c +++ b/drivers/rapidio/devices/rio_mport_cdev.c @@ -2603,7 +2603,7 @@ static int __init mport_init(void) int ret; /* Create device class needed by udev */ - dev_class = class_create(THIS_MODULE, DRV_NAME); + dev_class = class_create(DRV_NAME); if (IS_ERR(dev_class)) { rmcd_error("Unable to create " DRV_NAME " class"); return PTR_ERR(dev_class); diff --git a/drivers/rapidio/rio_cm.c b/drivers/rapidio/rio_cm.c index db4c265287ae..acaf9cda014c 100644 --- a/drivers/rapidio/rio_cm.c +++ b/drivers/rapidio/rio_cm.c @@ -2297,7 +2297,7 @@ static int __init riocm_init(void) int ret; /* Create device class needed by udev */ - dev_class = class_create(THIS_MODULE, DRV_NAME); + dev_class = class_create(DRV_NAME); if (IS_ERR(dev_class)) { riocm_error("Cannot create " DRV_NAME " class"); return PTR_ERR(dev_class); diff --git a/drivers/rpmsg/rpmsg_core.c b/drivers/rpmsg/rpmsg_core.c index a2207c0cf432..5039df757127 100644 --- a/drivers/rpmsg/rpmsg_core.c +++ b/drivers/rpmsg/rpmsg_core.c @@ -694,7 +694,7 @@ static int __init rpmsg_init(void) { int ret; - rpmsg_class = class_create(THIS_MODULE, "rpmsg"); + rpmsg_class = class_create("rpmsg"); if (IS_ERR(rpmsg_class)) { pr_err("failed to create rpmsg class\n"); return PTR_ERR(rpmsg_class); diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c index e5b7b48cffac..edfd942f8c54 100644 --- a/drivers/rtc/class.c +++ b/drivers/rtc/class.c @@ -475,7 +475,7 @@ EXPORT_SYMBOL_GPL(devm_rtc_device_register); static int __init rtc_init(void) { - rtc_class = class_create(THIS_MODULE, "rtc"); + rtc_class = class_create("rtc"); if (IS_ERR(rtc_class)) { pr_err("couldn't create class\n"); return PTR_ERR(rtc_class); diff --git a/drivers/s390/char/hmcdrv_dev.c b/drivers/s390/char/hmcdrv_dev.c index cb8fdf057eca..8d50c894711f 100644 --- a/drivers/s390/char/hmcdrv_dev.c +++ b/drivers/s390/char/hmcdrv_dev.c @@ -308,7 +308,7 @@ int hmcdrv_dev_init(void) * /proc/devices), but not under /dev nor /sys/devices/virtual. So * we have to create an associated class (see /sys/class). */ - hmcdrv_dev_class = class_create(THIS_MODULE, HMCDRV_DEV_CLASS); + hmcdrv_dev_class = class_create(HMCDRV_DEV_CLASS); if (IS_ERR(hmcdrv_dev_class)) { rc = PTR_ERR(hmcdrv_dev_class); diff --git a/drivers/s390/char/raw3270.c b/drivers/s390/char/raw3270.c index 09d7570d3b7d..7115c0f85650 100644 --- a/drivers/s390/char/raw3270.c +++ b/drivers/s390/char/raw3270.c @@ -1319,7 +1319,7 @@ static int raw3270_init(void) if (rc == 0) { /* Create attributes for early (= console) device. */ mutex_lock(&raw3270_mutex); - class3270 = class_create(THIS_MODULE, "3270"); + class3270 = class_create("3270"); list_for_each_entry(rp, &raw3270_devices, list) { get_device(&rp->cdev->dev); raw3270_create_attributes(rp); diff --git a/drivers/s390/char/tape_class.c b/drivers/s390/char/tape_class.c index c21dc68e05a0..277a0f903d11 100644 --- a/drivers/s390/char/tape_class.c +++ b/drivers/s390/char/tape_class.c @@ -117,7 +117,7 @@ EXPORT_SYMBOL(unregister_tape_dev); static int __init tape_init(void) { - tape_class = class_create(THIS_MODULE, "tape390"); + tape_class = class_create("tape390"); return 0; } diff --git a/drivers/s390/char/vmlogrdr.c b/drivers/s390/char/vmlogrdr.c index ed970ecfafdf..6946ba9a9de2 100644 --- a/drivers/s390/char/vmlogrdr.c +++ b/drivers/s390/char/vmlogrdr.c @@ -699,7 +699,7 @@ static int vmlogrdr_register_driver(void) if (ret) goto out_iucv; - vmlogrdr_class = class_create(THIS_MODULE, "vmlogrdr"); + vmlogrdr_class = class_create("vmlogrdr"); if (IS_ERR(vmlogrdr_class)) { ret = PTR_ERR(vmlogrdr_class); vmlogrdr_class = NULL; diff --git a/drivers/s390/char/vmur.c b/drivers/s390/char/vmur.c index 131293f7f152..82efdd20ad01 100644 --- a/drivers/s390/char/vmur.c +++ b/drivers/s390/char/vmur.c @@ -1022,7 +1022,7 @@ static int __init ur_init(void) debug_set_level(vmur_dbf, 6); - vmur_class = class_create(THIS_MODULE, "vmur"); + vmur_class = class_create("vmur"); if (IS_ERR(vmur_class)) { rc = PTR_ERR(vmur_class); goto fail_free_dbf; diff --git a/drivers/s390/crypto/zcrypt_api.c b/drivers/s390/crypto/zcrypt_api.c index 6fe05bb82c77..582ac301d315 100644 --- a/drivers/s390/crypto/zcrypt_api.c +++ b/drivers/s390/crypto/zcrypt_api.c @@ -2171,7 +2171,7 @@ static int __init zcdn_init(void) int rc; /* create a new class 'zcrypt' */ - zcrypt_class = class_create(THIS_MODULE, ZCRYPT_NAME); + zcrypt_class = class_create(ZCRYPT_NAME); if (IS_ERR(zcrypt_class)) { rc = PTR_ERR(zcrypt_class); goto out_class_create_failed; diff --git a/drivers/sbus/char/oradax.c b/drivers/sbus/char/oradax.c index e300cf26bc2a..75701b165e91 100644 --- a/drivers/sbus/char/oradax.c +++ b/drivers/sbus/char/oradax.c @@ -323,7 +323,7 @@ static int __init dax_attach(void) goto done; } - cl = class_create(THIS_MODULE, DAX_NAME); + cl = class_create(DAX_NAME); if (IS_ERR(cl)) { dax_err("class_create failed"); ret = PTR_ERR(cl); diff --git a/drivers/scsi/ch.c b/drivers/scsi/ch.c index 72fe6df78bc5..ac648bb8f7e7 100644 --- a/drivers/scsi/ch.c +++ b/drivers/scsi/ch.c @@ -995,7 +995,7 @@ static int __init init_ch_module(void) int rc; printk(KERN_INFO "SCSI Media Changer driver v" VERSION " \n"); - ch_sysfs_class = class_create(THIS_MODULE, "scsi_changer"); + ch_sysfs_class = class_create("scsi_changer"); if (IS_ERR(ch_sysfs_class)) { rc = PTR_ERR(ch_sysfs_class); return rc; diff --git a/drivers/scsi/cxlflash/main.c b/drivers/scsi/cxlflash/main.c index 395b00b942f7..debd36974119 100644 --- a/drivers/scsi/cxlflash/main.c +++ b/drivers/scsi/cxlflash/main.c @@ -3880,7 +3880,7 @@ static int cxlflash_class_init(void) cxlflash_major = MAJOR(devno); - cxlflash_class = class_create(THIS_MODULE, "cxlflash"); + cxlflash_class = class_create("cxlflash"); if (IS_ERR(cxlflash_class)) { rc = PTR_ERR(cxlflash_class); pr_err("%s: class_create failed rc=%d\n", __func__, rc); diff --git a/drivers/scsi/pmcraid.c b/drivers/scsi/pmcraid.c index 836ddc476764..7d57cc92532c 100644 --- a/drivers/scsi/pmcraid.c +++ b/drivers/scsi/pmcraid.c @@ -5346,7 +5346,7 @@ static int __init pmcraid_init(void) } pmcraid_major = MAJOR(dev); - pmcraid_class = class_create(THIS_MODULE, PMCRAID_DEVFILE); + pmcraid_class = class_create(PMCRAID_DEVFILE); if (IS_ERR(pmcraid_class)) { error = PTR_ERR(pmcraid_class); diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index a91049213203..4997f880d4a4 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -1677,7 +1677,7 @@ init_sg(void) SG_MAX_DEVS, "sg"); if (rc) return rc; - sg_sysfs_class = class_create(THIS_MODULE, "scsi_generic"); + sg_sysfs_class = class_create("scsi_generic"); if ( IS_ERR(sg_sysfs_class) ) { rc = PTR_ERR(sg_sysfs_class); goto err_out; diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c index 5a038c667401..539e70c7daed 100644 --- a/drivers/spi/spidev.c +++ b/drivers/spi/spidev.c @@ -877,7 +877,7 @@ static int __init spidev_init(void) if (status < 0) return status; - spidev_class = class_create(THIS_MODULE, "spidev"); + spidev_class = class_create("spidev"); if (IS_ERR(spidev_class)) { unregister_chrdev(SPIDEV_MAJOR, spidev_spi_driver.driver.name); return PTR_ERR(spidev_class); diff --git a/drivers/staging/fieldbus/anybuss/arcx-anybus.c b/drivers/staging/fieldbus/anybuss/arcx-anybus.c index 9af2e63050d1..369400ec30d8 100644 --- a/drivers/staging/fieldbus/anybuss/arcx-anybus.c +++ b/drivers/staging/fieldbus/anybuss/arcx-anybus.c @@ -352,7 +352,7 @@ static int __init controller_init(void) { int err; - controller_class = class_create(THIS_MODULE, "arcx_anybus_controller"); + controller_class = class_create("arcx_anybus_controller"); if (IS_ERR(controller_class)) return PTR_ERR(controller_class); err = platform_driver_register(&controller_driver); diff --git a/drivers/staging/greybus/authentication.c b/drivers/staging/greybus/authentication.c index 297e69f011c7..7e01790a4659 100644 --- a/drivers/staging/greybus/authentication.c +++ b/drivers/staging/greybus/authentication.c @@ -402,7 +402,7 @@ int cap_init(void) { int ret; - cap_class = class_create(THIS_MODULE, "gb_authenticate"); + cap_class = class_create("gb_authenticate"); if (IS_ERR(cap_class)) return PTR_ERR(cap_class); diff --git a/drivers/staging/greybus/fw-management.c b/drivers/staging/greybus/fw-management.c index 3342b84597da..cd9141e4b794 100644 --- a/drivers/staging/greybus/fw-management.c +++ b/drivers/staging/greybus/fw-management.c @@ -696,7 +696,7 @@ int fw_mgmt_init(void) { int ret; - fw_mgmt_class = class_create(THIS_MODULE, "gb_fw_mgmt"); + fw_mgmt_class = class_create("gb_fw_mgmt"); if (IS_ERR(fw_mgmt_class)) return PTR_ERR(fw_mgmt_class); diff --git a/drivers/staging/greybus/raw.c b/drivers/staging/greybus/raw.c index 2a375f407d38..8bca8cb12cc6 100644 --- a/drivers/staging/greybus/raw.c +++ b/drivers/staging/greybus/raw.c @@ -340,7 +340,7 @@ static int raw_init(void) dev_t dev; int retval; - raw_class = class_create(THIS_MODULE, "gb_raw"); + raw_class = class_create("gb_raw"); if (IS_ERR(raw_class)) { retval = PTR_ERR(raw_class); goto error_class; diff --git a/drivers/staging/pi433/pi433_if.c b/drivers/staging/pi433/pi433_if.c index b59f6a4cb611..f08fdf06d566 100644 --- a/drivers/staging/pi433/pi433_if.c +++ b/drivers/staging/pi433/pi433_if.c @@ -1400,7 +1400,7 @@ static int __init pi433_init(void) if (status < 0) return status; - pi433_class = class_create(THIS_MODULE, "pi433"); + pi433_class = class_create("pi433"); if (IS_ERR(pi433_class)) { unregister_chrdev(MAJOR(pi433_dev), pi433_spi_driver.driver.name); diff --git a/drivers/staging/vme_user/vme_user.c b/drivers/staging/vme_user/vme_user.c index 4e533c0bfe6d..b9367b575d00 100644 --- a/drivers/staging/vme_user/vme_user.c +++ b/drivers/staging/vme_user/vme_user.c @@ -614,7 +614,7 @@ static int vme_user_probe(struct vme_dev *vdev) } /* Create sysfs entries - on udev systems this creates the dev files */ - vme_user_sysfs_class = class_create(THIS_MODULE, driver_name); + vme_user_sysfs_class = class_create(driver_name); if (IS_ERR(vme_user_sysfs_class)) { dev_err(&vdev->dev, "Error creating vme_user class.\n"); err = PTR_ERR(vme_user_sysfs_class); diff --git a/drivers/tee/tee_core.c b/drivers/tee/tee_core.c index 452cbb8ad484..0eb342de0b00 100644 --- a/drivers/tee/tee_core.c +++ b/drivers/tee/tee_core.c @@ -1226,7 +1226,7 @@ static int __init tee_init(void) { int rc; - tee_class = class_create(THIS_MODULE, "tee"); + tee_class = class_create("tee"); if (IS_ERR(tee_class)) { pr_err("couldn't create class\n"); return PTR_ERR(tee_class); diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c index 36fb945fdad4..1382d9050ce8 100644 --- a/drivers/tty/tty_io.c +++ b/drivers/tty/tty_io.c @@ -3512,7 +3512,7 @@ static char *tty_devnode(const struct device *dev, umode_t *mode) static int __init tty_class_init(void) { - tty_class = class_create(THIS_MODULE, "tty"); + tty_class = class_create("tty"); if (IS_ERR(tty_class)) return PTR_ERR(tty_class); tty_class->devnode = tty_devnode; diff --git a/drivers/tty/vt/vc_screen.c b/drivers/tty/vt/vc_screen.c index 1dc07f9214d5..498ba9c0ee93 100644 --- a/drivers/tty/vt/vc_screen.c +++ b/drivers/tty/vt/vc_screen.c @@ -804,7 +804,7 @@ int __init vcs_init(void) if (register_chrdev(VCS_MAJOR, "vcs", &vcs_fops)) panic("unable to get major %d for vcs device", VCS_MAJOR); - vc_class = class_create(THIS_MODULE, "vc"); + vc_class = class_create("vc"); device_create(vc_class, NULL, MKDEV(VCS_MAJOR, 0), NULL, "vcs"); device_create(vc_class, NULL, MKDEV(VCS_MAJOR, 64), NULL, "vcsu"); diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 57a5c23b51d4..5496bf4b76ec 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -4241,7 +4241,7 @@ static int __init vtconsole_class_init(void) { int i; - vtconsole_class = class_create(THIS_MODULE, "vtconsole"); + vtconsole_class = class_create("vtconsole"); if (IS_ERR(vtconsole_class)) { pr_warn("Unable to create vt console class; errno = %ld\n", PTR_ERR(vtconsole_class)); diff --git a/drivers/usb/core/file.c b/drivers/usb/core/file.c index da7d88e069e6..c4ed3310e069 100644 --- a/drivers/usb/core/file.c +++ b/drivers/usb/core/file.c @@ -88,7 +88,7 @@ static int init_usb_class(void) } kref_init(&usb_class->kref); - usb_class->class = class_create(THIS_MODULE, "usbmisc"); + usb_class->class = class_create("usbmisc"); if (IS_ERR(usb_class->class)) { result = PTR_ERR(usb_class->class); printk(KERN_ERR "class_create failed for usb devices\n"); diff --git a/drivers/usb/gadget/function/f_hid.c b/drivers/usb/gadget/function/f_hid.c index a8da3b4a2855..9f6b10134121 100644 --- a/drivers/usb/gadget/function/f_hid.c +++ b/drivers/usb/gadget/function/f_hid.c @@ -1325,7 +1325,7 @@ int ghid_setup(struct usb_gadget *g, int count) int status; dev_t dev; - hidg_class = class_create(THIS_MODULE, "hidg"); + hidg_class = class_create("hidg"); if (IS_ERR(hidg_class)) { status = PTR_ERR(hidg_class); hidg_class = NULL; diff --git a/drivers/usb/gadget/function/f_printer.c b/drivers/usb/gadget/function/f_printer.c index 4903d761a872..28db3e336e7d 100644 --- a/drivers/usb/gadget/function/f_printer.c +++ b/drivers/usb/gadget/function/f_printer.c @@ -1512,7 +1512,7 @@ static int gprinter_setup(int count) int status; dev_t devt; - usb_gadget_class = class_create(THIS_MODULE, "usb_printer_gadget"); + usb_gadget_class = class_create("usb_printer_gadget"); if (IS_ERR(usb_gadget_class)) { status = PTR_ERR(usb_gadget_class); usb_gadget_class = NULL; diff --git a/drivers/usb/gadget/udc/core.c b/drivers/usb/gadget/udc/core.c index 23b0629a8774..df930b041c33 100644 --- a/drivers/usb/gadget/udc/core.c +++ b/drivers/usb/gadget/udc/core.c @@ -1758,7 +1758,7 @@ static int __init usb_udc_init(void) { int rc; - udc_class = class_create(THIS_MODULE, "udc"); + udc_class = class_create("udc"); if (IS_ERR(udc_class)) { pr_err("failed to create udc class --> %ld\n", PTR_ERR(udc_class)); diff --git a/drivers/usb/mon/mon_bin.c b/drivers/usb/mon/mon_bin.c index abb1cd35d8a6..952c56789258 100644 --- a/drivers/usb/mon/mon_bin.c +++ b/drivers/usb/mon/mon_bin.c @@ -1379,7 +1379,7 @@ int __init mon_bin_init(void) { int rc; - mon_bin_class = class_create(THIS_MODULE, "usbmon"); + mon_bin_class = class_create("usbmon"); if (IS_ERR(mon_bin_class)) { rc = PTR_ERR(mon_bin_class); goto err_class; diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c index 56814ef80c24..0395bd5dbd3e 100644 --- a/drivers/usb/roles/class.c +++ b/drivers/usb/roles/class.c @@ -392,7 +392,7 @@ EXPORT_SYMBOL_GPL(usb_role_switch_get_drvdata); static int __init usb_roles_init(void) { - role_class = class_create(THIS_MODULE, "usb_role"); + role_class = class_create("usb_role"); return PTR_ERR_OR_ZERO(role_class); } subsys_initcall(usb_roles_init); diff --git a/drivers/vdpa/vdpa_user/vduse_dev.c b/drivers/vdpa/vdpa_user/vduse_dev.c index 0c3b48616a9f..c421b83f6fa2 100644 --- a/drivers/vdpa/vdpa_user/vduse_dev.c +++ b/drivers/vdpa/vdpa_user/vduse_dev.c @@ -1793,7 +1793,7 @@ static int vduse_init(void) int ret; struct device *dev; - vduse_class = class_create(THIS_MODULE, "vduse"); + vduse_class = class_create("vduse"); if (IS_ERR(vduse_class)) return PTR_ERR(vduse_class); diff --git a/drivers/vfio/group.c b/drivers/vfio/group.c index 27d5ba7cf9dc..fc75c1000d74 100644 --- a/drivers/vfio/group.c +++ b/drivers/vfio/group.c @@ -878,7 +878,7 @@ int __init vfio_group_init(void) return ret; /* /dev/vfio/$GROUP */ - vfio.class = class_create(THIS_MODULE, "vfio"); + vfio.class = class_create("vfio"); if (IS_ERR(vfio.class)) { ret = PTR_ERR(vfio.class); goto err_group_class; diff --git a/drivers/vfio/vfio_main.c b/drivers/vfio/vfio_main.c index 43bd6b76e2b6..9ae671b98592 100644 --- a/drivers/vfio/vfio_main.c +++ b/drivers/vfio/vfio_main.c @@ -1408,7 +1408,7 @@ static int __init vfio_init(void) goto err_virqfd; /* /sys/class/vfio-dev/vfioX */ - vfio.device_class = class_create(THIS_MODULE, "vfio-dev"); + vfio.device_class = class_create("vfio-dev"); if (IS_ERR(vfio.device_class)) { ret = PTR_ERR(vfio.device_class); goto err_dev_class; diff --git a/drivers/video/backlight/backlight.c b/drivers/video/backlight/backlight.c index 6eea72aa8dbf..9a885d398c22 100644 --- a/drivers/video/backlight/backlight.c +++ b/drivers/video/backlight/backlight.c @@ -751,7 +751,7 @@ static void __exit backlight_class_exit(void) static int __init backlight_class_init(void) { - backlight_class = class_create(THIS_MODULE, "backlight"); + backlight_class = class_create("backlight"); if (IS_ERR(backlight_class)) { pr_warn("Unable to create backlight class; errno = %ld\n", PTR_ERR(backlight_class)); diff --git a/drivers/video/backlight/lcd.c b/drivers/video/backlight/lcd.c index db56e465aaff..77c5cb2a44e2 100644 --- a/drivers/video/backlight/lcd.c +++ b/drivers/video/backlight/lcd.c @@ -323,7 +323,7 @@ static void __exit lcd_class_exit(void) static int __init lcd_class_init(void) { - lcd_class = class_create(THIS_MODULE, "lcd"); + lcd_class = class_create("lcd"); if (IS_ERR(lcd_class)) { pr_warn("Unable to create backlight class; errno = %ld\n", PTR_ERR(lcd_class)); diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c index 875541ff185b..b073795d99d0 100644 --- a/drivers/video/fbdev/core/fbmem.c +++ b/drivers/video/fbdev/core/fbmem.c @@ -1749,7 +1749,7 @@ fbmem_init(void) goto err_chrdev; } - fb_class = class_create(THIS_MODULE, "graphics"); + fb_class = class_create("graphics"); if (IS_ERR(fb_class)) { ret = PTR_ERR(fb_class); pr_warn("Unable to create fb class; errno = %d\n", ret); diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index b39580ad4ce5..3c3148588491 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -361,7 +361,7 @@ static int __init init_coda_psdev(void) __func__, CODA_PSDEV_MAJOR); return -EIO; } - coda_psdev_class = class_create(THIS_MODULE, "coda"); + coda_psdev_class = class_create("coda"); if (IS_ERR(coda_psdev_class)) { err = PTR_ERR(coda_psdev_class); goto out_chrdev; diff --git a/fs/fuse/cuse.c b/fs/fuse/cuse.c index 179a5c5e28fd..91e89e68177e 100644 --- a/fs/fuse/cuse.c +++ b/fs/fuse/cuse.c @@ -623,7 +623,7 @@ static int __init cuse_init(void) /* CUSE is not prepared for FUSE_DEV_IOC_CLONE */ cuse_channel_fops.unlocked_ioctl = NULL; - cuse_class = class_create(THIS_MODULE, "cuse"); + cuse_class = class_create("cuse"); if (IS_ERR(cuse_class)) return PTR_ERR(cuse_class); diff --git a/fs/pstore/pmsg.c b/fs/pstore/pmsg.c index ab82e5f05346..aa178304bc03 100644 --- a/fs/pstore/pmsg.c +++ b/fs/pstore/pmsg.c @@ -64,7 +64,7 @@ void pstore_register_pmsg(void) goto err; } - pmsg_class = class_create(THIS_MODULE, PMSG_NAME); + pmsg_class = class_create(PMSG_NAME); if (IS_ERR(pmsg_class)) { pr_err("device class file already in use\n"); goto err_class; diff --git a/include/linux/device/class.h b/include/linux/device/class.h index bf736f14f0c1..cda598fc5fa0 100644 --- a/include/linux/device/class.h +++ b/include/linux/device/class.h @@ -256,7 +256,6 @@ extern void class_destroy(struct class *cls); /** * class_create - create a struct class structure - * @owner: dummy pointer, does nothing, will be removed soon. * @name: pointer to a string for the name of this class. * * This is used to create a struct class pointer that can then be used @@ -267,7 +266,7 @@ extern void class_destroy(struct class *cls); * Note, the pointer created here is to be destroyed when finished by * making a call to class_destroy(). */ -#define class_create(owner, name) \ +#define class_create(name) \ ({ \ static struct lock_class_key __key; \ __class_create(name, &__key); \ diff --git a/mm/backing-dev.c b/mm/backing-dev.c index a53b9360b72e..ad011308cebe 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -347,7 +347,7 @@ ATTRIBUTE_GROUPS(bdi_dev); static __init int bdi_class_init(void) { - bdi_class = class_create(THIS_MODULE, "bdi"); + bdi_class = class_create("bdi"); if (IS_ERR(bdi_class)) return PTR_ERR(bdi_class); diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 08542dfc2dc5..2934d7f4d564 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -112,7 +112,7 @@ void hci_init_sysfs(struct hci_dev *hdev) int __init bt_sysfs_init(void) { - bt_class = class_create(THIS_MODULE, "bluetooth"); + bt_class = class_create("bluetooth"); return PTR_ERR_OR_ZERO(bt_class); } diff --git a/net/netfilter/xt_IDLETIMER.c b/net/netfilter/xt_IDLETIMER.c index 8d36303f3935..db720efa811d 100644 --- a/net/netfilter/xt_IDLETIMER.c +++ b/net/netfilter/xt_IDLETIMER.c @@ -490,7 +490,7 @@ static int __init idletimer_tg_init(void) { int err; - idletimer_tg_class = class_create(THIS_MODULE, "xt_idletimer"); + idletimer_tg_class = class_create("xt_idletimer"); err = PTR_ERR(idletimer_tg_class); if (IS_ERR(idletimer_tg_class)) { pr_debug("couldn't register device class\n"); diff --git a/samples/vfio-mdev/mbochs.c b/samples/vfio-mdev/mbochs.c index e54eb752e1ba..a3d3249639d0 100644 --- a/samples/vfio-mdev/mbochs.c +++ b/samples/vfio-mdev/mbochs.c @@ -1418,7 +1418,7 @@ static int __init mbochs_dev_init(void) if (ret) goto err_cdev; - mbochs_class = class_create(THIS_MODULE, MBOCHS_CLASS_NAME); + mbochs_class = class_create(MBOCHS_CLASS_NAME); if (IS_ERR(mbochs_class)) { pr_err("Error: failed to register mbochs_dev class\n"); ret = PTR_ERR(mbochs_class); diff --git a/samples/vfio-mdev/mdpy.c b/samples/vfio-mdev/mdpy.c index e8400fdab71d..ef1630fc5a91 100644 --- a/samples/vfio-mdev/mdpy.c +++ b/samples/vfio-mdev/mdpy.c @@ -705,7 +705,7 @@ static int __init mdpy_dev_init(void) if (ret) goto err_cdev; - mdpy_class = class_create(THIS_MODULE, MDPY_CLASS_NAME); + mdpy_class = class_create(MDPY_CLASS_NAME); if (IS_ERR(mdpy_class)) { pr_err("Error: failed to register mdpy_dev class\n"); ret = PTR_ERR(mdpy_class); diff --git a/samples/vfio-mdev/mtty.c b/samples/vfio-mdev/mtty.c index e887de672c52..0b6c386e620d 100644 --- a/samples/vfio-mdev/mtty.c +++ b/samples/vfio-mdev/mtty.c @@ -1316,7 +1316,7 @@ static int __init mtty_dev_init(void) if (ret) goto err_cdev; - mtty_dev.vd_class = class_create(THIS_MODULE, MTTY_CLASS_NAME); + mtty_dev.vd_class = class_create(MTTY_CLASS_NAME); if (IS_ERR(mtty_dev.vd_class)) { pr_err("Error: failed to register mtty_dev class\n"); diff --git a/sound/sound_core.c b/sound/sound_core.c index 3e7dd6fcb7cf..4f6911274d56 100644 --- a/sound/sound_core.c +++ b/sound/sound_core.c @@ -45,7 +45,7 @@ static int __init init_soundcore(void) if (rc) return rc; - sound_class = class_create(THIS_MODULE, "sound"); + sound_class = class_create("sound"); if (IS_ERR(sound_class)) { cleanup_oss_soundcore(); return PTR_ERR(sound_class); diff --git a/tools/testing/nvdimm/test/ndtest.c b/tools/testing/nvdimm/test/ndtest.c index 01ceb98c15a0..3eba10c1e3e8 100644 --- a/tools/testing/nvdimm/test/ndtest.c +++ b/tools/testing/nvdimm/test/ndtest.c @@ -921,7 +921,7 @@ static __init int ndtest_init(void) nfit_test_setup(ndtest_resource_lookup, NULL); - ndtest_dimm_class = class_create(THIS_MODULE, "nfit_test_dimm"); + ndtest_dimm_class = class_create("nfit_test_dimm"); if (IS_ERR(ndtest_dimm_class)) { rc = PTR_ERR(ndtest_dimm_class); goto err_register; diff --git a/tools/testing/nvdimm/test/nfit.c b/tools/testing/nvdimm/test/nfit.c index c75abb497a1a..45b1c072e349 100644 --- a/tools/testing/nvdimm/test/nfit.c +++ b/tools/testing/nvdimm/test/nfit.c @@ -3282,7 +3282,7 @@ static __init int nfit_test_init(void) if (!nfit_wq) return -ENOMEM; - nfit_test_dimm = class_create(THIS_MODULE, "nfit_test_dimm"); + nfit_test_dimm = class_create("nfit_test_dimm"); if (IS_ERR(nfit_test_dimm)) { rc = PTR_ERR(nfit_test_dimm); goto err_register; -- cgit v1.2.3 From a8eff03545d4cef12ae66a1905627c1818a0f81a Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sat, 18 Mar 2023 01:19:00 +0200 Subject: net: dsa: report rx_bytes unadjusted for ETH_HLEN We collect the software statistics counters for RX bytes (reported to /proc/net/dev and to ethtool -S $dev | grep 'rx_bytes: ") at a time when skb->len has already been adjusted by the eth_type_trans() -> skb_pull_inline(skb, ETH_HLEN) call to exclude the L2 header. This means that when connecting 2 DSA interfaces back to back and sending 1 packet with length 100, the sending interface will report tx_bytes as incrementing by 100, and the receiving interface will report rx_bytes as incrementing by 86. Since accounting for that in scripts is quirky and is something that would be DSA-specific behavior (requiring users to know that they are running on a DSA interface in the first place), the proposal is that we treat it as a bug and fix it. This design bug has always existed in DSA, according to my analysis: commit 91da11f870f0 ("net: Distributed Switch Architecture protocol support") also updates skb->dev->stats.rx_bytes += skb->len after the eth_type_trans() call. Technically, prior to Florian's commit a86d8becc3f0 ("net: dsa: Factor bottom tag receive functions"), each and every vendor-specific tagging protocol driver open-coded the same bug, until the buggy code was consolidated into something resembling what can be seen now. So each and every driver should have its own Fixes: tag, because of their different histories until the convergence point. I'm not going to do that, for the sake of simplicity, but just blame the oldest appearance of buggy code. There are 2 ways to fix the problem. One is the obvious way, and the other is how I ended up doing it. Obvious would have been to move dev_sw_netstats_rx_add() one line above eth_type_trans(), and below skb_push(skb, ETH_HLEN). But DSA processing is not as simple as that. We count the bytes after removing everything DSA-related from the packet, to emulate what the packet's length was, on the wire, when the user port received it. When eth_type_trans() executes, dsa_untag_bridge_pvid() has not run yet, so in case the switch driver requests this behavior - commit 412a1526d067 ("net: dsa: untag the bridge pvid from rx skbs") has the details - the obvious variant of the fix wouldn't have worked, because the positioning there would have also counted the not-yet-stripped VLAN header length, something which is absent from the packet as seen on the wire (there it may be untagged, whereas software will see it as PVID-tagged). Fixes: f613ed665bb3 ("net: dsa: Add support for 64-bit statistics") Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- net/dsa/tag.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/dsa/tag.c b/net/dsa/tag.c index b2fba1a003ce..5105a5ff58fa 100644 --- a/net/dsa/tag.c +++ b/net/dsa/tag.c @@ -114,7 +114,7 @@ static int dsa_switch_rcv(struct sk_buff *skb, struct net_device *dev, skb = nskb; } - dev_sw_netstats_rx_add(skb->dev, skb->len); + dev_sw_netstats_rx_add(skb->dev, skb->len + ETH_HLEN); if (dsa_skb_defer_rx_timestamp(p, skb)) return 0; -- cgit v1.2.3 From 04aae213e719ec2bb310158c4025316ace50589b Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 20 Mar 2023 18:41:13 -0700 Subject: net: skbuff: rename __pkt_vlan_present_offset to __mono_tc_offset vlan_present is gone since commit 354259fa73e2 ("net: remove skb->vlan_present") rename the offset field to what BPF is currently looking for in this byte - mono_delivery_time and tc_at_ingress. Signed-off-by: Jakub Kicinski Link: https://lore.kernel.org/r/20230321014115.997841-2-kuba@kernel.org Signed-off-by: Martin KaFai Lau --- include/linux/skbuff.h | 4 ++-- net/core/filter.c | 8 ++++---- tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c | 6 +++--- 3 files changed, 9 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 3f3a2a82a86b..5a63878a4550 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -955,7 +955,7 @@ struct sk_buff { __u8 csum_valid:1; /* private: */ - __u8 __pkt_vlan_present_offset[0]; + __u8 __mono_tc_offset[0]; /* public: */ __u8 remcsum_offload:1; __u8 csum_complete_sw:1; @@ -1078,7 +1078,7 @@ struct sk_buff { #define TC_AT_INGRESS_MASK (1 << 7) #define SKB_MONO_DELIVERY_TIME_MASK (1 << 5) #endif -#define PKT_VLAN_PRESENT_OFFSET offsetof(struct sk_buff, __pkt_vlan_present_offset) +#define SKB_BF_MONO_TC_OFFSET offsetof(struct sk_buff, __mono_tc_offset) #ifdef __KERNEL__ /* diff --git a/net/core/filter.c b/net/core/filter.c index 50f649f1b4a9..3370efad1dda 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -9185,7 +9185,7 @@ static struct bpf_insn *bpf_convert_tstamp_type_read(const struct bpf_insn *si, __u8 tmp_reg = BPF_REG_AX; *insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, - PKT_VLAN_PRESENT_OFFSET); + SKB_BF_MONO_TC_OFFSET); *insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg, SKB_MONO_DELIVERY_TIME_MASK, 2); *insn++ = BPF_MOV32_IMM(value_reg, BPF_SKB_TSTAMP_UNSPEC); @@ -9232,7 +9232,7 @@ static struct bpf_insn *bpf_convert_tstamp_read(const struct bpf_prog *prog, /* AX is needed because src_reg and dst_reg could be the same */ __u8 tmp_reg = BPF_REG_AX; - *insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, PKT_VLAN_PRESENT_OFFSET); + *insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, SKB_BF_MONO_TC_OFFSET); *insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK); *insn++ = BPF_JMP32_IMM(BPF_JNE, tmp_reg, @@ -9267,14 +9267,14 @@ static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_prog *prog, if (!prog->tstamp_type_access) { __u8 tmp_reg = BPF_REG_AX; - *insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, PKT_VLAN_PRESENT_OFFSET); + *insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, SKB_BF_MONO_TC_OFFSET); /* Writing __sk_buff->tstamp as ingress, goto */ *insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg, TC_AT_INGRESS_MASK, 1); /* goto */ *insn++ = BPF_JMP_A(2); /* : mono_delivery_time */ *insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_MONO_DELIVERY_TIME_MASK); - *insn++ = BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, PKT_VLAN_PRESENT_OFFSET); + *insn++ = BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, SKB_BF_MONO_TC_OFFSET); } #endif diff --git a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c index d5fe3d4b936c..ae7b6e50e405 100644 --- a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c +++ b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c @@ -68,17 +68,17 @@ static struct test_case test_cases[] = { #if defined(__x86_64__) || defined(__aarch64__) { N(SCHED_CLS, struct __sk_buff, tstamp), - .read = "r11 = *(u8 *)($ctx + sk_buff::__pkt_vlan_present_offset);" + .read = "r11 = *(u8 *)($ctx + sk_buff::__mono_tc_offset);" "w11 &= 160;" "if w11 != 0xa0 goto pc+2;" "$dst = 0;" "goto pc+1;" "$dst = *(u64 *)($ctx + sk_buff::tstamp);", - .write = "r11 = *(u8 *)($ctx + sk_buff::__pkt_vlan_present_offset);" + .write = "r11 = *(u8 *)($ctx + sk_buff::__mono_tc_offset);" "if w11 & 0x80 goto pc+1;" "goto pc+2;" "w11 &= -33;" - "*(u8 *)($ctx + sk_buff::__pkt_vlan_present_offset) = r11;" + "*(u8 *)($ctx + sk_buff::__mono_tc_offset) = r11;" "*(u64 *)($ctx + sk_buff::tstamp) = $src;", }, #endif -- cgit v1.2.3 From 032a954061afd4b7426c3eb6bfd2952ef1e9a384 Mon Sep 17 00:00:00 2001 From: Álvaro Fernández Rojas Date: Sun, 19 Mar 2023 10:55:40 +0100 Subject: net: dsa: tag_brcm: legacy: fix daisy-chained switches MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When BCM63xx internal switches are connected to switches with a 4-byte Broadcom tag, it does not identify the packet as VLAN tagged, so it adds one based on its PVID (which is likely 0). Right now, the packet is received by the BCM63xx internal switch and the 6-byte tag is properly processed. The next step would to decode the corresponding 4-byte tag. However, the internal switch adds an invalid VLAN tag after the 6-byte tag and the 4-byte tag handling fails. In order to fix this we need to remove the invalid VLAN tag after the 6-byte tag before passing it to the 4-byte tag decoding. Fixes: 964dbf186eaa ("net: dsa: tag_brcm: add support for legacy tags") Signed-off-by: Álvaro Fernández Rojas Reviewed-by: Michal Swiatkowski Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/20230319095540.239064-1-noltari@gmail.com Signed-off-by: Jakub Kicinski --- net/dsa/tag_brcm.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c index 10239daa5745..cacdafb41200 100644 --- a/net/dsa/tag_brcm.c +++ b/net/dsa/tag_brcm.c @@ -7,6 +7,7 @@ #include #include +#include #include #include @@ -252,6 +253,7 @@ static struct sk_buff *brcm_leg_tag_xmit(struct sk_buff *skb, static struct sk_buff *brcm_leg_tag_rcv(struct sk_buff *skb, struct net_device *dev) { + int len = BRCM_LEG_TAG_LEN; int source_port; u8 *brcm_tag; @@ -266,12 +268,16 @@ static struct sk_buff *brcm_leg_tag_rcv(struct sk_buff *skb, if (!skb->dev) return NULL; + /* VLAN tag is added by BCM63xx internal switch */ + if (netdev_uses_dsa(skb->dev)) + len += VLAN_HLEN; + /* Remove Broadcom tag and update checksum */ - skb_pull_rcsum(skb, BRCM_LEG_TAG_LEN); + skb_pull_rcsum(skb, len); dsa_default_offload_fwd_mark(skb); - dsa_strip_etype_header(skb, BRCM_LEG_TAG_LEN); + dsa_strip_etype_header(skb, len); return skb; } -- cgit v1.2.3 From 8e50ed774554f93d55426039b27b1e38d7fa64d8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 20 Mar 2023 16:34:27 +0000 Subject: erspan: do not use skb_mac_header() in ndo_start_xmit() Drivers should not assume skb_mac_header(skb) == skb->data in their ndo_start_xmit(). Use skb_network_offset() and skb_transport_offset() which better describe what is needed in erspan_fb_xmit() and ip6erspan_tunnel_xmit() syzbot reported: WARNING: CPU: 0 PID: 5083 at include/linux/skbuff.h:2873 skb_mac_header include/linux/skbuff.h:2873 [inline] WARNING: CPU: 0 PID: 5083 at include/linux/skbuff.h:2873 ip6erspan_tunnel_xmit+0x1d9c/0x2d90 net/ipv6/ip6_gre.c:962 Modules linked in: CPU: 0 PID: 5083 Comm: syz-executor406 Not tainted 6.3.0-rc2-syzkaller-00866-gd4671cb96fa3 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/02/2023 RIP: 0010:skb_mac_header include/linux/skbuff.h:2873 [inline] RIP: 0010:ip6erspan_tunnel_xmit+0x1d9c/0x2d90 net/ipv6/ip6_gre.c:962 Code: 04 02 41 01 de 84 c0 74 08 3c 03 0f 8e 1c 0a 00 00 45 89 b4 24 c8 00 00 00 c6 85 77 fe ff ff 01 e9 33 e7 ff ff e8 b4 27 a1 f8 <0f> 0b e9 b6 e7 ff ff e8 a8 27 a1 f8 49 8d bf f0 0c 00 00 48 b8 00 RSP: 0018:ffffc90003b2f830 EFLAGS: 00010293 RAX: 0000000000000000 RBX: 000000000000ffff RCX: 0000000000000000 RDX: ffff888021273a80 RSI: ffffffff88e1bd4c RDI: 0000000000000003 RBP: ffffc90003b2f9d8 R08: 0000000000000003 R09: 000000000000ffff R10: 000000000000ffff R11: 0000000000000000 R12: ffff88802b28da00 R13: 00000000000000d0 R14: ffff88807e25b6d0 R15: ffff888023408000 FS: 0000555556a61300(0000) GS:ffff8880b9800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000055e5b11eb6e8 CR3: 0000000027c1b000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: __netdev_start_xmit include/linux/netdevice.h:4900 [inline] netdev_start_xmit include/linux/netdevice.h:4914 [inline] __dev_direct_xmit+0x504/0x730 net/core/dev.c:4300 dev_direct_xmit include/linux/netdevice.h:3088 [inline] packet_xmit+0x20a/0x390 net/packet/af_packet.c:285 packet_snd net/packet/af_packet.c:3075 [inline] packet_sendmsg+0x31a0/0x5150 net/packet/af_packet.c:3107 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg+0xde/0x190 net/socket.c:747 __sys_sendto+0x23a/0x340 net/socket.c:2142 __do_sys_sendto net/socket.c:2154 [inline] __se_sys_sendto net/socket.c:2150 [inline] __x64_sys_sendto+0xe1/0x1b0 net/socket.c:2150 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0033:0x7f123aaa1039 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 b1 14 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 c0 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007ffc15d12058 EFLAGS: 00000246 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f123aaa1039 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000003 RBP: 0000000000000000 R08: 0000000020000040 R09: 0000000000000014 R10: 0000000000000000 R11: 0000000000000246 R12: 00007f123aa648c0 R13: 431bde82d7b634db R14: 0000000000000000 R15: 0000000000000000 Fixes: 1baf5ebf8954 ("erspan: auto detect truncated packets.") Reported-by: syzbot Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230320163427.8096-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/ip_gre.c | 4 ++-- net/ipv6/ip6_gre.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index ffff46cdcb58..e55a20264960 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -552,7 +552,7 @@ static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev) truncate = true; } - nhoff = skb_network_header(skb) - skb_mac_header(skb); + nhoff = skb_network_offset(skb); if (skb->protocol == htons(ETH_P_IP) && (ntohs(ip_hdr(skb)->tot_len) > skb->len - nhoff)) truncate = true; @@ -561,7 +561,7 @@ static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev) int thoff; if (skb_transport_header_was_set(skb)) - thoff = skb_transport_header(skb) - skb_mac_header(skb); + thoff = skb_transport_offset(skb); else thoff = nhoff + sizeof(struct ipv6hdr); if (ntohs(ipv6_hdr(skb)->payload_len) > skb->len - thoff) diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 89f5f0f3f5d6..a4ecfc9d2593 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -959,7 +959,7 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb, truncate = true; } - nhoff = skb_network_header(skb) - skb_mac_header(skb); + nhoff = skb_network_offset(skb); if (skb->protocol == htons(ETH_P_IP) && (ntohs(ip_hdr(skb)->tot_len) > skb->len - nhoff)) truncate = true; @@ -968,7 +968,7 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb, int thoff; if (skb_transport_header_was_set(skb)) - thoff = skb_transport_header(skb) - skb_mac_header(skb); + thoff = skb_transport_offset(skb); else thoff = nhoff + sizeof(struct ipv6hdr); if (ntohs(ipv6_hdr(skb)->payload_len) > skb->len - thoff) -- cgit v1.2.3 From 61587f1556fec39e8bafc40c8715f560639a4cf2 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 21 Mar 2023 10:12:48 +0100 Subject: wifi: mac80211: add support for letting drivers register tc offload support On newer MediaTek SoCs (e.g. MT7986), WLAN->WLAN or WLAN->Ethernet flows can be offloaded by the SoC. In order to support that, the .ndo_setup_tc op is needed. Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230321091248.30947-1-nbd@nbd.name Signed-off-by: Johannes Berg --- include/net/mac80211.h | 9 +++++++++ net/mac80211/driver-ops.h | 17 +++++++++++++++++ net/mac80211/ieee80211_i.h | 3 ++- net/mac80211/iface.c | 11 +++++++++++ net/mac80211/trace.h | 25 +++++++++++++++++++++++++ 5 files changed, 64 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index f12edca660ba..fcfe3e9aff3d 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -4227,6 +4227,10 @@ struct ieee80211_prep_tx_info { * @set_hw_timestamp: Enable/disable HW timestamping of TM/FTM frames. This is * not restored at HW reset by mac80211 so drivers need to take care of * that. + * @net_setup_tc: Called from .ndo_setup_tc in order to prepare hardware + * flow offloading for flows originating from the vif. + * Note that the driver must not assume that the vif driver_data is valid + * at this point, since the callback can be called during netdev teardown. */ struct ieee80211_ops { void (*tx)(struct ieee80211_hw *hw, @@ -4593,6 +4597,11 @@ struct ieee80211_ops { int (*set_hw_timestamp)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct cfg80211_set_hw_timestamp *hwts); + int (*net_setup_tc)(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct net_device *dev, + enum tc_setup_type type, + void *type_data); }; /** diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index a68d606e6987..0bf208f5bbc5 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -1502,6 +1502,23 @@ static inline int drv_net_fill_forward_path(struct ieee80211_local *local, return ret; } +static inline int drv_net_setup_tc(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct net_device *dev, + enum tc_setup_type type, void *type_data) +{ + int ret = -EOPNOTSUPP; + + sdata = get_bss_sdata(sdata); + trace_drv_net_setup_tc(local, sdata, type); + if (local->ops->net_setup_tc) + ret = local->ops->net_setup_tc(&local->hw, &sdata->vif, dev, + type, type_data); + trace_drv_return_int(local, ret); + + return ret; +} + int drv_change_vif_links(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata, u16 old_links, u16 new_links, diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 3d4edc25a69e..b2535614483e 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -1939,7 +1939,8 @@ void ieee80211_color_collision_detection_work(struct work_struct *work); /* interface handling */ #define MAC80211_SUPPORTED_FEATURES_TX (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | \ NETIF_F_HW_CSUM | NETIF_F_SG | \ - NETIF_F_HIGHDMA | NETIF_F_GSO_SOFTWARE) + NETIF_F_HIGHDMA | NETIF_F_GSO_SOFTWARE | \ + NETIF_F_HW_TC) #define MAC80211_SUPPORTED_FEATURES_RX (NETIF_F_RXCSUM) #define MAC80211_SUPPORTED_FEATURES (MAC80211_SUPPORTED_FEATURES_TX | \ MAC80211_SUPPORTED_FEATURES_RX) diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index 23ed13f15067..bd2c48870add 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -813,6 +813,15 @@ ieee80211_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) dev_fetch_sw_netstats(stats, dev->tstats); } +static int ieee80211_netdev_setup_tc(struct net_device *dev, + enum tc_setup_type type, void *type_data) +{ + struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); + struct ieee80211_local *local = sdata->local; + + return drv_net_setup_tc(local, sdata, dev, type, type_data); +} + static const struct net_device_ops ieee80211_dataif_ops = { .ndo_open = ieee80211_open, .ndo_stop = ieee80211_stop, @@ -821,6 +830,7 @@ static const struct net_device_ops ieee80211_dataif_ops = { .ndo_set_rx_mode = ieee80211_set_multicast_list, .ndo_set_mac_address = ieee80211_change_mac, .ndo_get_stats64 = ieee80211_get_stats64, + .ndo_setup_tc = ieee80211_netdev_setup_tc, }; static u16 ieee80211_monitor_select_queue(struct net_device *dev, @@ -929,6 +939,7 @@ static const struct net_device_ops ieee80211_dataif_8023_ops = { .ndo_set_mac_address = ieee80211_change_mac, .ndo_get_stats64 = ieee80211_get_stats64, .ndo_fill_forward_path = ieee80211_netdev_fill_forward_path, + .ndo_setup_tc = ieee80211_netdev_setup_tc, }; static bool ieee80211_iftype_supports_hdr_offload(enum nl80211_iftype iftype) diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h index 9f4377566c42..e0ccf5fe708a 100644 --- a/net/mac80211/trace.h +++ b/net/mac80211/trace.h @@ -2478,6 +2478,31 @@ DEFINE_EVENT(sta_event, drv_net_fill_forward_path, TP_ARGS(local, sdata, sta) ); +TRACE_EVENT(drv_net_setup_tc, + TP_PROTO(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + u8 type), + + TP_ARGS(local, sdata, type), + + TP_STRUCT__entry( + LOCAL_ENTRY + VIF_ENTRY + __field(u8, type) + ), + + TP_fast_assign( + LOCAL_ASSIGN; + VIF_ASSIGN; + __entry->type = type; + ), + + TP_printk( + LOCAL_PR_FMT VIF_PR_FMT " type:%d\n", + LOCAL_PR_ARG, VIF_PR_ARG, __entry->type + ) +); + TRACE_EVENT(drv_change_vif_links, TP_PROTO(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata, -- cgit v1.2.3 From e626dad92383ca16d1d71e66124a272a0cbfe7bd Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 14 Mar 2023 10:59:51 +0100 Subject: wifi: mac80211: fix race in mesh sequence number assignment Since the sequence number is shared across different tx queues, it needs to be atomic in order to avoid accidental duplicate assignment Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230314095956.62085-2-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/ieee80211_i.h | 2 +- net/mac80211/mesh.c | 6 ++---- 2 files changed, 3 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index b2535614483e..04017d5c435a 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -696,7 +696,7 @@ struct ieee80211_if_mesh { struct mesh_stats mshstats; struct mesh_config mshcfg; atomic_t estab_plinks; - u32 mesh_seqnum; + atomic_t mesh_seqnum; bool accepting_plinks; int num_gates; struct beacon_data __rcu *beacon; diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c index 5a99b8f6e465..0608ed415831 100644 --- a/net/mac80211/mesh.c +++ b/net/mac80211/mesh.c @@ -752,10 +752,8 @@ unsigned int ieee80211_new_mesh_header(struct ieee80211_sub_if_data *sdata, meshhdr->ttl = sdata->u.mesh.mshcfg.dot11MeshTTL; - /* FIXME: racy -- TX on multiple queues can be concurrent */ - put_unaligned(cpu_to_le32(sdata->u.mesh.mesh_seqnum), &meshhdr->seqnum); - sdata->u.mesh.mesh_seqnum++; - + put_unaligned_le32(atomic_inc_return(&sdata->u.mesh.mesh_seqnum), + &meshhdr->seqnum); if (addr4or5 && !addr6) { meshhdr->flags |= MESH_FLAGS_AE_A4; memcpy(meshhdr->eaddr1, addr4or5, ETH_ALEN); -- cgit v1.2.3 From d5edb9ae8d568745f893c5c5fa3837d85311b131 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 14 Mar 2023 10:59:53 +0100 Subject: wifi: mac80211: mesh fast xmit support Previously, fast xmit only worked on interface types where initially a sta lookup is performed, and a cached header can be attached to the sta, requiring only some fields to be updated at runtime. This technique is not directly applicable for a mesh device type due to the dynamic nature of the topology and protocol. There are more addresses that need to be filled, and there is an extra header with a dynamic length based on the addressing mode. Change the code to cache entries contain a copy of the mesh subframe header + bridge tunnel header, as well as an embedded struct ieee80211_fast_tx, which contains the information for building the 802.11 header. Add a mesh specific early fast xmit call, which looks up a cached entry and adds only the mesh subframe header, before passing it over to the generic fast xmit code. To ensure the changes in network are reflected in these cached headers, flush affected cached entries on path changes, as well as other conditions that currently trigger a fast xmit check in other modes (key changes etc.) This code is loosely based on a previous implementation by: Sriram R Cc: Sriram R Signed-off-by: Ryder Lee Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230314095956.62085-4-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/ieee80211_i.h | 20 ++++ net/mac80211/mesh.c | 92 +++++++++++++++ net/mac80211/mesh.h | 44 +++++++ net/mac80211/mesh_hwmp.c | 37 ++++-- net/mac80211/mesh_pathtbl.c | 282 ++++++++++++++++++++++++++++++++++++++++++++ net/mac80211/rx.c | 10 +- net/mac80211/tx.c | 47 +++++--- 7 files changed, 507 insertions(+), 25 deletions(-) (limited to 'net') diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 04017d5c435a..4fd4e241f8cd 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -37,6 +37,7 @@ extern const struct cfg80211_ops mac80211_config_ops; struct ieee80211_local; +struct ieee80211_mesh_fast_tx; /* Maximum number of broadcast/multicast frames to buffer when some of the * associated stations are using power saving. */ @@ -656,6 +657,19 @@ struct mesh_table { atomic_t entries; /* Up to MAX_MESH_NEIGHBOURS */ }; +/** + * struct mesh_tx_cache - mesh fast xmit header cache + * + * @rht: hash table containing struct ieee80211_mesh_fast_tx, using skb DA as key + * @walk_head: linked list containing all ieee80211_mesh_fast_tx objects + * @walk_lock: lock protecting walk_head and rht + */ +struct mesh_tx_cache { + struct rhashtable rht; + struct hlist_head walk_head; + spinlock_t walk_lock; +}; + struct ieee80211_if_mesh { struct timer_list housekeeping_timer; struct timer_list mesh_path_timer; @@ -734,6 +748,7 @@ struct ieee80211_if_mesh { struct mesh_table mpp_paths; /* Store paths for MPP&MAP */ int mesh_paths_generation; int mpp_paths_generation; + struct mesh_tx_cache tx_cache; }; #ifdef CONFIG_MAC80211_MESH @@ -2018,6 +2033,11 @@ int ieee80211_tx_control_port(struct wiphy *wiphy, struct net_device *dev, int link_id, u64 *cookie); int ieee80211_probe_mesh_link(struct wiphy *wiphy, struct net_device *dev, const u8 *buf, size_t len); +void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, + struct sta_info *sta, + struct ieee80211_fast_tx *fast_tx, + struct sk_buff *skb, bool ampdu, + const u8 *da, const u8 *sa); /* HT */ void ieee80211_apply_htcap_overrides(struct ieee80211_sub_if_data *sdata, diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c index 0608ed415831..6b94cf2a4046 100644 --- a/net/mac80211/mesh.c +++ b/net/mac80211/mesh.c @@ -10,6 +10,7 @@ #include #include "ieee80211_i.h" #include "mesh.h" +#include "wme.h" #include "driver-ops.h" static int mesh_allocated; @@ -698,6 +699,95 @@ ieee80211_mesh_update_bss_params(struct ieee80211_sub_if_data *sdata, __le32_to_cpu(he_oper->he_oper_params); } +bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata, + struct sk_buff *skb, u32 ctrl_flags) +{ + struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh; + struct ieee80211_mesh_fast_tx *entry; + struct ieee80211s_hdr *meshhdr; + u8 sa[ETH_ALEN] __aligned(2); + struct tid_ampdu_tx *tid_tx; + struct sta_info *sta; + bool copy_sa = false; + u16 ethertype; + u8 tid; + + if (ctrl_flags & IEEE80211_TX_CTRL_SKIP_MPATH_LOOKUP) + return false; + + if (ifmsh->mshcfg.dot11MeshNolearn) + return false; + + /* Add support for these cases later */ + if (ifmsh->ps_peers_light_sleep || ifmsh->ps_peers_deep_sleep) + return false; + + if (is_multicast_ether_addr(skb->data)) + return false; + + ethertype = (skb->data[12] << 8) | skb->data[13]; + if (ethertype < ETH_P_802_3_MIN) + return false; + + if (skb->sk && skb_shinfo(skb)->tx_flags & SKBTX_WIFI_STATUS) + return false; + + if (skb->ip_summed == CHECKSUM_PARTIAL) { + skb_set_transport_header(skb, skb_checksum_start_offset(skb)); + if (skb_checksum_help(skb)) + return false; + } + + entry = mesh_fast_tx_get(sdata, skb->data); + if (!entry) + return false; + + if (skb_headroom(skb) < entry->hdrlen + entry->fast_tx.hdr_len) + return false; + + sta = rcu_dereference(entry->mpath->next_hop); + if (!sta) + return false; + + tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK; + tid_tx = rcu_dereference(sta->ampdu_mlme.tid_tx[tid]); + if (tid_tx) { + if (!test_bit(HT_AGG_STATE_OPERATIONAL, &tid_tx->state)) + return false; + if (tid_tx->timeout) + tid_tx->last_tx = jiffies; + } + + skb = skb_share_check(skb, GFP_ATOMIC); + if (!skb) + return true; + + skb_set_queue_mapping(skb, ieee80211_select_queue(sdata, sta, skb)); + + meshhdr = (struct ieee80211s_hdr *)entry->hdr; + if ((meshhdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6) { + /* preserve SA from eth header for 6-addr frames */ + ether_addr_copy(sa, skb->data + ETH_ALEN); + copy_sa = true; + } + + memcpy(skb_push(skb, entry->hdrlen - 2 * ETH_ALEN), entry->hdr, + entry->hdrlen); + + meshhdr = (struct ieee80211s_hdr *)skb->data; + put_unaligned_le32(atomic_inc_return(&sdata->u.mesh.mesh_seqnum), + &meshhdr->seqnum); + meshhdr->ttl = sdata->u.mesh.mshcfg.dot11MeshTTL; + if (copy_sa) + ether_addr_copy(meshhdr->eaddr2, sa); + + skb_push(skb, 2 * ETH_ALEN); + __ieee80211_xmit_fast(sdata, sta, &entry->fast_tx, skb, tid_tx, + entry->mpath->dst, sdata->vif.addr); + + return true; +} + /** * ieee80211_fill_mesh_addresses - fill addresses of a locally originated mesh frame * @hdr: 802.11 frame header @@ -780,6 +870,8 @@ static void ieee80211_mesh_housekeeping(struct ieee80211_sub_if_data *sdata) changed = mesh_accept_plinks_update(sdata); ieee80211_mbss_info_change_notify(sdata, changed); + mesh_fast_tx_gc(sdata); + mod_timer(&ifmsh->housekeeping_timer, round_jiffies(jiffies + IEEE80211_MESH_HOUSEKEEPING_INTERVAL)); diff --git a/net/mac80211/mesh.h b/net/mac80211/mesh.h index b2b717a78114..13f394e677ae 100644 --- a/net/mac80211/mesh.h +++ b/net/mac80211/mesh.h @@ -122,11 +122,41 @@ struct mesh_path { u8 rann_snd_addr[ETH_ALEN]; u32 rann_metric; unsigned long last_preq_to_root; + unsigned long fast_tx_check; bool is_root; bool is_gate; u32 path_change_count; }; +#define MESH_FAST_TX_CACHE_MAX_SIZE 512 +#define MESH_FAST_TX_CACHE_THRESHOLD_SIZE 384 +#define MESH_FAST_TX_CACHE_TIMEOUT 8000 /* msecs */ + +/** + * struct ieee80211_mesh_fast_tx - cached mesh fast tx entry + * @rhash: rhashtable pointer + * @addr_key: The Ethernet DA which is the key for this entry + * @fast_tx: base fast_tx data + * @hdr: cached mesh and rfc1042 headers + * @hdrlen: length of mesh + rfc1042 + * @walk_list: list containing all the fast tx entries + * @mpath: mesh path corresponding to the Mesh DA + * @mppath: MPP entry corresponding to this DA + * @timestamp: Last used time of this entry + */ +struct ieee80211_mesh_fast_tx { + struct rhash_head rhash; + u8 addr_key[ETH_ALEN] __aligned(2); + + struct ieee80211_fast_tx fast_tx; + u8 hdr[sizeof(struct ieee80211s_hdr) + sizeof(rfc1042_header)]; + u16 hdrlen; + + struct mesh_path *mpath, *mppath; + struct hlist_node walk_list; + unsigned long timestamp; +}; + /* Recent multicast cache */ /* RMC_BUCKETS must be a power of 2, maximum 256 */ #define RMC_BUCKETS 256 @@ -298,6 +328,20 @@ void mesh_path_discard_frame(struct ieee80211_sub_if_data *sdata, void mesh_path_tx_root_frame(struct ieee80211_sub_if_data *sdata); bool mesh_action_is_path_sel(struct ieee80211_mgmt *mgmt); +struct ieee80211_mesh_fast_tx * +mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata, const u8 *addr); +bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata, + struct sk_buff *skb, u32 ctrl_flags); +void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata, + struct sk_buff *skb, struct mesh_path *mpath); +void mesh_fast_tx_gc(struct ieee80211_sub_if_data *sdata); +void mesh_fast_tx_flush_addr(struct ieee80211_sub_if_data *sdata, + const u8 *addr); +void mesh_fast_tx_flush_mpath(struct mesh_path *mpath); +void mesh_fast_tx_flush_sta(struct ieee80211_sub_if_data *sdata, + struct sta_info *sta); +void mesh_path_refresh(struct ieee80211_sub_if_data *sdata, + struct mesh_path *mpath, const u8 *addr); #ifdef CONFIG_MAC80211_MESH static inline diff --git a/net/mac80211/mesh_hwmp.c b/net/mac80211/mesh_hwmp.c index 9b1ce7c3925a..5217e1d97dd6 100644 --- a/net/mac80211/mesh_hwmp.c +++ b/net/mac80211/mesh_hwmp.c @@ -394,6 +394,7 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata, u32 orig_sn, orig_metric; unsigned long orig_lifetime, exp_time; u32 last_hop_metric, new_metric; + bool flush_mpath = false; bool process = true; u8 hopcount; @@ -491,8 +492,10 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata, } if (fresh_info) { - if (rcu_access_pointer(mpath->next_hop) != sta) + if (rcu_access_pointer(mpath->next_hop) != sta) { mpath->path_change_count++; + flush_mpath = true; + } mesh_path_assign_nexthop(mpath, sta); mpath->flags |= MESH_PATH_SN_VALID; mpath->metric = new_metric; @@ -502,6 +505,8 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata, mpath->hop_count = hopcount; mesh_path_activate(mpath); spin_unlock_bh(&mpath->state_lock); + if (flush_mpath) + mesh_fast_tx_flush_mpath(mpath); ewma_mesh_fail_avg_init(&sta->mesh->fail_avg); /* init it at a low value - 0 start is tricky */ ewma_mesh_fail_avg_add(&sta->mesh->fail_avg, 1); @@ -539,8 +544,10 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata, } if (fresh_info) { - if (rcu_access_pointer(mpath->next_hop) != sta) + if (rcu_access_pointer(mpath->next_hop) != sta) { mpath->path_change_count++; + flush_mpath = true; + } mesh_path_assign_nexthop(mpath, sta); mpath->metric = last_hop_metric; mpath->exp_time = time_after(mpath->exp_time, exp_time) @@ -548,6 +555,8 @@ static u32 hwmp_route_info_get(struct ieee80211_sub_if_data *sdata, mpath->hop_count = 1; mesh_path_activate(mpath); spin_unlock_bh(&mpath->state_lock); + if (flush_mpath) + mesh_fast_tx_flush_mpath(mpath); ewma_mesh_fail_avg_init(&sta->mesh->fail_avg); /* init it at a low value - 0 start is tricky */ ewma_mesh_fail_avg_add(&sta->mesh->fail_avg, 1); @@ -1215,6 +1224,20 @@ static int mesh_nexthop_lookup_nolearn(struct ieee80211_sub_if_data *sdata, return 0; } +void mesh_path_refresh(struct ieee80211_sub_if_data *sdata, + struct mesh_path *mpath, const u8 *addr) +{ + if (mpath->flags & (MESH_PATH_REQ_QUEUED | MESH_PATH_FIXED | + MESH_PATH_RESOLVING)) + return; + + if (time_after(jiffies, + mpath->exp_time - + msecs_to_jiffies(sdata->u.mesh.mshcfg.path_refresh_time)) && + (!addr || ether_addr_equal(sdata->vif.addr, addr))) + mesh_queue_preq(mpath, PREQ_Q_F_START | PREQ_Q_F_REFRESH); +} + /** * mesh_nexthop_lookup - put the appropriate next hop on a mesh frame. Calling * this function is considered "using" the associated mpath, so preempt a path @@ -1242,19 +1265,15 @@ int mesh_nexthop_lookup(struct ieee80211_sub_if_data *sdata, if (!mpath || !(mpath->flags & MESH_PATH_ACTIVE)) return -ENOENT; - if (time_after(jiffies, - mpath->exp_time - - msecs_to_jiffies(sdata->u.mesh.mshcfg.path_refresh_time)) && - ether_addr_equal(sdata->vif.addr, hdr->addr4) && - !(mpath->flags & MESH_PATH_RESOLVING) && - !(mpath->flags & MESH_PATH_FIXED)) - mesh_queue_preq(mpath, PREQ_Q_F_START | PREQ_Q_F_REFRESH); + mesh_path_refresh(sdata, mpath, hdr->addr4); next_hop = rcu_dereference(mpath->next_hop); if (next_hop) { memcpy(hdr->addr1, next_hop->sta.addr, ETH_ALEN); memcpy(hdr->addr2, sdata->vif.addr, ETH_ALEN); ieee80211_mps_set_frame_flags(sdata, next_hop, hdr); + if (ieee80211_hw_check(&sdata->local->hw, SUPPORT_FAST_XMIT)) + mesh_fast_tx_cache(sdata, skb, mpath); return 0; } diff --git a/net/mac80211/mesh_pathtbl.c b/net/mac80211/mesh_pathtbl.c index 3b81e6df3f34..d32e304eeb4b 100644 --- a/net/mac80211/mesh_pathtbl.c +++ b/net/mac80211/mesh_pathtbl.c @@ -14,6 +14,7 @@ #include "wme.h" #include "ieee80211_i.h" #include "mesh.h" +#include static void mesh_path_free_rcu(struct mesh_table *tbl, struct mesh_path *mpath); @@ -32,6 +33,41 @@ static const struct rhashtable_params mesh_rht_params = { .hashfn = mesh_table_hash, }; +static const struct rhashtable_params fast_tx_rht_params = { + .nelem_hint = 10, + .automatic_shrinking = true, + .key_len = ETH_ALEN, + .key_offset = offsetof(struct ieee80211_mesh_fast_tx, addr_key), + .head_offset = offsetof(struct ieee80211_mesh_fast_tx, rhash), + .hashfn = mesh_table_hash, +}; + +static void __mesh_fast_tx_entry_free(void *ptr, void *tblptr) +{ + struct ieee80211_mesh_fast_tx *entry = ptr; + + kfree_rcu(entry, fast_tx.rcu_head); +} + +static void mesh_fast_tx_deinit(struct ieee80211_sub_if_data *sdata) +{ + struct mesh_tx_cache *cache; + + cache = &sdata->u.mesh.tx_cache; + rhashtable_free_and_destroy(&cache->rht, + __mesh_fast_tx_entry_free, NULL); +} + +static void mesh_fast_tx_init(struct ieee80211_sub_if_data *sdata) +{ + struct mesh_tx_cache *cache; + + cache = &sdata->u.mesh.tx_cache; + rhashtable_init(&cache->rht, &fast_tx_rht_params); + INIT_HLIST_HEAD(&cache->walk_head); + spin_lock_init(&cache->walk_lock); +} + static inline bool mpath_expired(struct mesh_path *mpath) { return (mpath->flags & MESH_PATH_ACTIVE) && @@ -381,6 +417,243 @@ struct mesh_path *mesh_path_new(struct ieee80211_sub_if_data *sdata, return new_mpath; } +static void mesh_fast_tx_entry_free(struct mesh_tx_cache *cache, + struct ieee80211_mesh_fast_tx *entry) +{ + hlist_del_rcu(&entry->walk_list); + rhashtable_remove_fast(&cache->rht, &entry->rhash, fast_tx_rht_params); + kfree_rcu(entry, fast_tx.rcu_head); +} + +struct ieee80211_mesh_fast_tx * +mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata, const u8 *addr) +{ + struct ieee80211_mesh_fast_tx *entry; + struct mesh_tx_cache *cache; + + cache = &sdata->u.mesh.tx_cache; + entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params); + if (!entry) + return NULL; + + if (!(entry->mpath->flags & MESH_PATH_ACTIVE) || + mpath_expired(entry->mpath)) { + spin_lock_bh(&cache->walk_lock); + entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params); + if (entry) + mesh_fast_tx_entry_free(cache, entry); + spin_unlock_bh(&cache->walk_lock); + return NULL; + } + + mesh_path_refresh(sdata, entry->mpath, NULL); + if (entry->mppath) + entry->mppath->exp_time = jiffies; + entry->timestamp = jiffies; + + return entry; +} + +void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata, + struct sk_buff *skb, struct mesh_path *mpath) +{ + struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; + struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); + struct ieee80211_mesh_fast_tx *entry, *prev; + struct ieee80211_mesh_fast_tx build = {}; + struct ieee80211s_hdr *meshhdr; + struct mesh_tx_cache *cache; + struct ieee80211_key *key; + struct mesh_path *mppath; + struct sta_info *sta; + u8 *qc; + + if (sdata->noack_map || + !ieee80211_is_data_qos(hdr->frame_control)) + return; + + build.fast_tx.hdr_len = ieee80211_hdrlen(hdr->frame_control); + meshhdr = (struct ieee80211s_hdr *)(skb->data + build.fast_tx.hdr_len); + build.hdrlen = ieee80211_get_mesh_hdrlen(meshhdr); + + cache = &sdata->u.mesh.tx_cache; + if (atomic_read(&cache->rht.nelems) >= MESH_FAST_TX_CACHE_MAX_SIZE) + return; + + sta = rcu_dereference(mpath->next_hop); + if (!sta) + return; + + if ((meshhdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6) { + /* This is required to keep the mppath alive */ + mppath = mpp_path_lookup(sdata, meshhdr->eaddr1); + if (!mppath) + return; + build.mppath = mppath; + } else if (ieee80211_has_a4(hdr->frame_control)) { + mppath = mpath; + } else { + return; + } + + /* rate limit, in case fast xmit can't be enabled */ + if (mppath->fast_tx_check == jiffies) + return; + + mppath->fast_tx_check = jiffies; + + /* + * Same use of the sta lock as in ieee80211_check_fast_xmit, in order + * to protect against concurrent sta key updates. + */ + spin_lock_bh(&sta->lock); + key = rcu_access_pointer(sta->ptk[sta->ptk_idx]); + if (!key) + key = rcu_access_pointer(sdata->default_unicast_key); + build.fast_tx.key = key; + + if (key) { + bool gen_iv, iv_spc; + + gen_iv = key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_IV; + iv_spc = key->conf.flags & IEEE80211_KEY_FLAG_PUT_IV_SPACE; + + if (!(key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) || + (key->flags & KEY_FLAG_TAINTED)) + goto unlock_sta; + + switch (key->conf.cipher) { + case WLAN_CIPHER_SUITE_CCMP: + case WLAN_CIPHER_SUITE_CCMP_256: + if (gen_iv) + build.fast_tx.pn_offs = build.fast_tx.hdr_len; + if (gen_iv || iv_spc) + build.fast_tx.hdr_len += IEEE80211_CCMP_HDR_LEN; + break; + case WLAN_CIPHER_SUITE_GCMP: + case WLAN_CIPHER_SUITE_GCMP_256: + if (gen_iv) + build.fast_tx.pn_offs = build.fast_tx.hdr_len; + if (gen_iv || iv_spc) + build.fast_tx.hdr_len += IEEE80211_GCMP_HDR_LEN; + break; + default: + goto unlock_sta; + } + } + + memcpy(build.addr_key, mppath->dst, ETH_ALEN); + build.timestamp = jiffies; + build.fast_tx.band = info->band; + build.fast_tx.da_offs = offsetof(struct ieee80211_hdr, addr3); + build.fast_tx.sa_offs = offsetof(struct ieee80211_hdr, addr4); + build.mpath = mpath; + memcpy(build.hdr, meshhdr, build.hdrlen); + memcpy(build.hdr + build.hdrlen, rfc1042_header, sizeof(rfc1042_header)); + build.hdrlen += sizeof(rfc1042_header); + memcpy(build.fast_tx.hdr, hdr, build.fast_tx.hdr_len); + + hdr = (struct ieee80211_hdr *)build.fast_tx.hdr; + if (build.fast_tx.key) + hdr->frame_control |= cpu_to_le16(IEEE80211_FCTL_PROTECTED); + + qc = ieee80211_get_qos_ctl(hdr); + qc[1] |= IEEE80211_QOS_CTL_MESH_CONTROL_PRESENT >> 8; + + entry = kmemdup(&build, sizeof(build), GFP_ATOMIC); + if (!entry) + goto unlock_sta; + + spin_lock(&cache->walk_lock); + prev = rhashtable_lookup_get_insert_fast(&cache->rht, + &entry->rhash, + fast_tx_rht_params); + if (unlikely(IS_ERR(prev))) { + kfree(entry); + goto unlock_cache; + } + + /* + * replace any previous entry in the hash table, in case we're + * replacing it with a different type (e.g. mpath -> mpp) + */ + if (unlikely(prev)) { + rhashtable_replace_fast(&cache->rht, &prev->rhash, + &entry->rhash, fast_tx_rht_params); + hlist_del_rcu(&prev->walk_list); + kfree_rcu(prev, fast_tx.rcu_head); + } + + hlist_add_head(&entry->walk_list, &cache->walk_head); + +unlock_cache: + spin_unlock(&cache->walk_lock); +unlock_sta: + spin_unlock_bh(&sta->lock); +} + +void mesh_fast_tx_gc(struct ieee80211_sub_if_data *sdata) +{ + unsigned long timeout = msecs_to_jiffies(MESH_FAST_TX_CACHE_TIMEOUT); + struct mesh_tx_cache *cache; + struct ieee80211_mesh_fast_tx *entry; + struct hlist_node *n; + + cache = &sdata->u.mesh.tx_cache; + if (atomic_read(&cache->rht.nelems) < MESH_FAST_TX_CACHE_THRESHOLD_SIZE) + return; + + spin_lock_bh(&cache->walk_lock); + hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list) + if (!time_is_after_jiffies(entry->timestamp + timeout)) + mesh_fast_tx_entry_free(cache, entry); + spin_unlock_bh(&cache->walk_lock); +} + +void mesh_fast_tx_flush_mpath(struct mesh_path *mpath) +{ + struct ieee80211_sub_if_data *sdata = mpath->sdata; + struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache; + struct ieee80211_mesh_fast_tx *entry; + struct hlist_node *n; + + cache = &sdata->u.mesh.tx_cache; + spin_lock_bh(&cache->walk_lock); + hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list) + if (entry->mpath == mpath) + mesh_fast_tx_entry_free(cache, entry); + spin_unlock_bh(&cache->walk_lock); +} + +void mesh_fast_tx_flush_sta(struct ieee80211_sub_if_data *sdata, + struct sta_info *sta) +{ + struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache; + struct ieee80211_mesh_fast_tx *entry; + struct hlist_node *n; + + cache = &sdata->u.mesh.tx_cache; + spin_lock_bh(&cache->walk_lock); + hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list) + if (rcu_access_pointer(entry->mpath->next_hop) == sta) + mesh_fast_tx_entry_free(cache, entry); + spin_unlock_bh(&cache->walk_lock); +} + +void mesh_fast_tx_flush_addr(struct ieee80211_sub_if_data *sdata, + const u8 *addr) +{ + struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache; + struct ieee80211_mesh_fast_tx *entry; + + cache = &sdata->u.mesh.tx_cache; + spin_lock_bh(&cache->walk_lock); + entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params); + if (entry) + mesh_fast_tx_entry_free(cache, entry); + spin_unlock_bh(&cache->walk_lock); +} + /** * mesh_path_add - allocate and add a new path to the mesh path table * @dst: destination address of the path (ETH_ALEN length) @@ -464,6 +737,8 @@ int mpp_path_add(struct ieee80211_sub_if_data *sdata, if (ret) kfree(new_mpath); + else + mesh_fast_tx_flush_addr(sdata, dst); sdata->u.mesh.mpp_paths_generation++; return ret; @@ -523,6 +798,10 @@ static void __mesh_path_del(struct mesh_table *tbl, struct mesh_path *mpath) { hlist_del_rcu(&mpath->walk_list); rhashtable_remove_fast(&tbl->rhead, &mpath->rhash, mesh_rht_params); + if (tbl == &mpath->sdata->u.mesh.mpp_paths) + mesh_fast_tx_flush_addr(mpath->sdata, mpath->dst); + else + mesh_fast_tx_flush_mpath(mpath); mesh_path_free_rcu(tbl, mpath); } @@ -747,6 +1026,7 @@ void mesh_path_fix_nexthop(struct mesh_path *mpath, struct sta_info *next_hop) mpath->exp_time = 0; mpath->flags = MESH_PATH_FIXED | MESH_PATH_SN_VALID; mesh_path_activate(mpath); + mesh_fast_tx_flush_mpath(mpath); spin_unlock_bh(&mpath->state_lock); ewma_mesh_fail_avg_init(&next_hop->mesh->fail_avg); /* init it at a low value - 0 start is tricky */ @@ -758,6 +1038,7 @@ void mesh_pathtbl_init(struct ieee80211_sub_if_data *sdata) { mesh_table_init(&sdata->u.mesh.mesh_paths); mesh_table_init(&sdata->u.mesh.mpp_paths); + mesh_fast_tx_init(sdata); } static @@ -785,6 +1066,7 @@ void mesh_path_expire(struct ieee80211_sub_if_data *sdata) void mesh_pathtbl_unregister(struct ieee80211_sub_if_data *sdata) { + mesh_fast_tx_deinit(sdata); mesh_table_free(&sdata->u.mesh.mesh_paths); mesh_table_free(&sdata->u.mesh.mpp_paths); } diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 0255c5745e1c..b22d736285bf 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2772,6 +2772,7 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta if (mesh_hdr->flags & MESH_FLAGS_AE) { struct mesh_path *mppath; char *proxied_addr; + bool update = false; if (multicast) proxied_addr = mesh_hdr->eaddr1; @@ -2787,11 +2788,18 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta mpp_path_add(sdata, proxied_addr, eth->h_source); } else { spin_lock_bh(&mppath->state_lock); - if (!ether_addr_equal(mppath->mpp, eth->h_source)) + if (!ether_addr_equal(mppath->mpp, eth->h_source)) { memcpy(mppath->mpp, eth->h_source, ETH_ALEN); + update = true; + } mppath->exp_time = jiffies; spin_unlock_bh(&mppath->state_lock); } + + /* flush fast xmit cache if the address path changed */ + if (update) + mesh_fast_tx_flush_addr(sdata, proxied_addr); + rcu_read_unlock(); } diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 628d60f3db2a..ad6193693363 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -3019,6 +3019,9 @@ void ieee80211_check_fast_xmit(struct sta_info *sta) if (!ieee80211_hw_check(&local->hw, SUPPORT_FAST_XMIT)) return; + if (ieee80211_vif_is_mesh(&sdata->vif)) + mesh_fast_tx_flush_sta(sdata, sta); + /* Locking here protects both the pointer itself, and against concurrent * invocations winning data access races to, e.g., the key pointer that * is used. @@ -3371,7 +3374,8 @@ static bool ieee80211_amsdu_prepare_head(struct ieee80211_sub_if_data *sdata, static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata, struct sta_info *sta, struct ieee80211_fast_tx *fast_tx, - struct sk_buff *skb) + struct sk_buff *skb, + const u8 *da, const u8 *sa) { struct ieee80211_local *local = sdata->local; struct fq *fq = &local->fq; @@ -3400,6 +3404,9 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata, if (sdata->vif.offload_flags & IEEE80211_OFFLOAD_ENCAP_ENABLED) return false; + if (ieee80211_vif_is_mesh(&sdata->vif)) + return false; + if (skb_is_gso(skb)) return false; @@ -3484,7 +3491,8 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata, ret = true; data = skb_push(skb, ETH_ALEN + 2); - memmove(data, data + ETH_ALEN + 2, 2 * ETH_ALEN); + ether_addr_copy(data, da); + ether_addr_copy(data + ETH_ALEN, sa); data += 2 * ETH_ALEN; len = cpu_to_be16(subframe_len); @@ -3632,10 +3640,11 @@ free: return NULL; } -static void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, - struct sta_info *sta, - struct ieee80211_fast_tx *fast_tx, - struct sk_buff *skb, u8 tid, bool ampdu) +void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, + struct sta_info *sta, + struct ieee80211_fast_tx *fast_tx, + struct sk_buff *skb, bool ampdu, + const u8 *da, const u8 *sa) { struct ieee80211_local *local = sdata->local; struct ieee80211_hdr *hdr = (void *)fast_tx->hdr; @@ -3644,14 +3653,13 @@ static void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, ieee80211_tx_result r; int hw_headroom = sdata->local->hw.extra_tx_headroom; int extra_head = fast_tx->hdr_len - (ETH_HLEN - 2); - struct ethhdr eth; skb = skb_share_check(skb, GFP_ATOMIC); if (unlikely(!skb)) return; if ((hdr->frame_control & cpu_to_le16(IEEE80211_STYPE_QOS_DATA)) && - ieee80211_amsdu_aggregate(sdata, sta, fast_tx, skb)) + ieee80211_amsdu_aggregate(sdata, sta, fast_tx, skb, da, sa)) return; /* will not be crypto-handled beyond what we do here, so use false @@ -3664,11 +3672,10 @@ static void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, ENCRYPT_NO))) goto free; - memcpy(ð, skb->data, ETH_HLEN - 2); hdr = skb_push(skb, extra_head); memcpy(skb->data, fast_tx->hdr, fast_tx->hdr_len); - memcpy(skb->data + fast_tx->da_offs, eth.h_dest, ETH_ALEN); - memcpy(skb->data + fast_tx->sa_offs, eth.h_source, ETH_ALEN); + memcpy(skb->data + fast_tx->da_offs, da, ETH_ALEN); + memcpy(skb->data + fast_tx->sa_offs, sa, ETH_ALEN); info = IEEE80211_SKB_CB(skb); memset(info, 0, sizeof(*info)); @@ -3686,7 +3693,8 @@ static void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, #endif if (hdr->frame_control & cpu_to_le16(IEEE80211_STYPE_QOS_DATA)) { - tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK; + u8 tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK; + *ieee80211_get_qos_ctl(hdr) = tid; } @@ -3729,6 +3737,7 @@ static bool ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, struct ieee80211_hdr *hdr = (void *)fast_tx->hdr; struct tid_ampdu_tx *tid_tx = NULL; struct sk_buff *next; + struct ethhdr eth; u8 tid = IEEE80211_NUM_TIDS; /* control port protocol needs a lot of special handling */ @@ -3754,6 +3763,8 @@ static bool ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, } } + memcpy(ð, skb->data, ETH_HLEN - 2); + /* after this point (skb is modified) we cannot return false */ skb = ieee80211_tx_skb_fixup(skb, ieee80211_sdata_netdev_features(sdata)); if (!skb) @@ -3761,7 +3772,8 @@ static bool ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, skb_list_walk_safe(skb, skb, next) { skb_mark_not_on_list(skb); - __ieee80211_xmit_fast(sdata, sta, fast_tx, skb, tid, tid_tx); + __ieee80211_xmit_fast(sdata, sta, fast_tx, skb, tid_tx, + eth.h_dest, eth.h_source); } return true; @@ -4245,8 +4257,15 @@ void __ieee80211_subif_start_xmit(struct sk_buff *skb, return; } + sk_pacing_shift_update(skb->sk, sdata->local->hw.tx_sk_pacing_shift); + rcu_read_lock(); + if (ieee80211_vif_is_mesh(&sdata->vif) && + ieee80211_hw_check(&local->hw, SUPPORT_FAST_XMIT) && + ieee80211_mesh_xmit_fast(sdata, skb, ctrl_flags)) + goto out; + if (ieee80211_lookup_ra_sta(sdata, skb, &sta)) goto out_free; @@ -4256,8 +4275,6 @@ void __ieee80211_subif_start_xmit(struct sk_buff *skb, skb_set_queue_mapping(skb, ieee80211_select_queue(sdata, sta, skb)); ieee80211_aggr_check(sdata, sta, skb); - sk_pacing_shift_update(skb->sk, sdata->local->hw.tx_sk_pacing_shift); - if (sta) { struct ieee80211_fast_tx *fast_tx; -- cgit v1.2.3 From 8b0f5cb6bc7cbbee4d78b3221683dcb4d1ed23d0 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 14 Mar 2023 10:59:54 +0100 Subject: wifi: mac80211: use mesh header cache to speed up mesh forwarding Significantly reduces mesh forwarding path CPU usage and enables the direct use of iTXQ. Signed-off-by: Felix Fietkau Signed-off-by: Ryder Lee Link: https://lore.kernel.org/r/20230314095956.62085-5-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/ieee80211_i.h | 2 ++ net/mac80211/rx.c | 65 +++++++++++++++++++++++++++++++++++++++++++++- net/mac80211/tx.c | 6 ++--- 3 files changed, 68 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 4fd4e241f8cd..c8c037a9e4e2 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2038,6 +2038,8 @@ void __ieee80211_xmit_fast(struct ieee80211_sub_if_data *sdata, struct ieee80211_fast_tx *fast_tx, struct sk_buff *skb, bool ampdu, const u8 *da, const u8 *sa); +void ieee80211_aggr_check(struct ieee80211_sub_if_data *sdata, + struct sta_info *sta, struct sk_buff *skb); /* HT */ void ieee80211_apply_htcap_overrides(struct ieee80211_sub_if_data *sdata, diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index b22d736285bf..f6e9a66046a6 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2701,6 +2701,65 @@ ieee80211_deliver_skb(struct ieee80211_rx_data *rx) } } +#ifdef CONFIG_MAC80211_MESH +static bool +ieee80211_rx_mesh_fast_forward(struct ieee80211_sub_if_data *sdata, + struct sk_buff *skb, int hdrlen) +{ + struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh; + struct ieee80211_mesh_fast_tx *entry = NULL; + struct ieee80211s_hdr *mesh_hdr; + struct tid_ampdu_tx *tid_tx; + struct sta_info *sta; + struct ethhdr eth; + u8 tid; + + mesh_hdr = (struct ieee80211s_hdr *)(skb->data + sizeof(eth)); + if ((mesh_hdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6) + entry = mesh_fast_tx_get(sdata, mesh_hdr->eaddr1); + else if (!(mesh_hdr->flags & MESH_FLAGS_AE)) + entry = mesh_fast_tx_get(sdata, skb->data); + if (!entry) + return false; + + sta = rcu_dereference(entry->mpath->next_hop); + if (!sta) + return false; + + if (skb_linearize(skb)) + return false; + + tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK; + tid_tx = rcu_dereference(sta->ampdu_mlme.tid_tx[tid]); + if (tid_tx) { + if (!test_bit(HT_AGG_STATE_OPERATIONAL, &tid_tx->state)) + return false; + + if (tid_tx->timeout) + tid_tx->last_tx = jiffies; + } + + ieee80211_aggr_check(sdata, sta, skb); + + if (ieee80211_get_8023_tunnel_proto(skb->data + hdrlen, + &skb->protocol)) + hdrlen += ETH_ALEN; + else + skb->protocol = htons(skb->len - hdrlen); + skb_set_network_header(skb, hdrlen + 2); + + skb->dev = sdata->dev; + memcpy(ð, skb->data, ETH_HLEN - 2); + skb_pull(skb, 2); + __ieee80211_xmit_fast(sdata, sta, &entry->fast_tx, skb, tid_tx, + eth.h_dest, eth.h_source); + IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, fwded_unicast); + IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, fwded_frames); + + return true; +} +#endif + static ieee80211_rx_result ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta, struct sk_buff *skb) @@ -2805,6 +2864,10 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta skb_set_queue_mapping(skb, ieee802_1d_to_ac[skb->priority]); + if (!multicast && + ieee80211_rx_mesh_fast_forward(sdata, skb, mesh_hdrlen)) + return RX_QUEUED; + ieee80211_fill_mesh_addresses(&hdr, &hdr.frame_control, eth->h_dest, eth->h_source); hdrlen = ieee80211_hdrlen(hdr.frame_control); @@ -2843,6 +2906,7 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta info->control.flags |= IEEE80211_TX_INTCFL_NEED_TXPROCESSING; info->control.vif = &sdata->vif; info->control.jiffies = jiffies; + fwd_skb->dev = sdata->dev; if (multicast) { IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, fwded_mcast); memcpy(fwd_hdr->addr2, sdata->vif.addr, ETH_ALEN); @@ -2864,7 +2928,6 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta } IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, fwded_frames); - fwd_skb->dev = sdata->dev; ieee80211_add_pending_skb(local, fwd_skb); rx_accept: diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index ad6193693363..de17926484bd 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1189,10 +1189,8 @@ static bool ieee80211_tx_prep_agg(struct ieee80211_tx_data *tx, return queued; } -static void -ieee80211_aggr_check(struct ieee80211_sub_if_data *sdata, - struct sta_info *sta, - struct sk_buff *skb) +void ieee80211_aggr_check(struct ieee80211_sub_if_data *sdata, + struct sta_info *sta, struct sk_buff *skb) { struct rate_control_ref *ref = sdata->local->rate_ctrl; u16 tid; -- cgit v1.2.3 From 3468e1e0c639032a603450f0830ccabfa76f5806 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 14 Mar 2023 10:59:55 +0100 Subject: wifi: mac80211: add mesh fast-rx support This helps bring down rx CPU usage by avoiding calls to the rx handlers in the slow path. Supports forwarding and local rx, including A-MSDU. Signed-off-by: Felix Fietkau Signed-off-by: Ryder Lee Link: https://lore.kernel.org/r/20230314095956.62085-6-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 34 +++++++++++++++++++++++++++++----- 1 file changed, 29 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index f6e9a66046a6..85fb1d3eeb2f 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -4542,6 +4542,12 @@ void ieee80211_check_fast_rx(struct sta_info *sta) fastrx.internal_forward = 0; } + break; + case NL80211_IFTYPE_MESH_POINT: + fastrx.expected_ds_bits = cpu_to_le16(IEEE80211_FCTL_FROMDS | + IEEE80211_FCTL_TODS); + fastrx.da_offs = offsetof(struct ieee80211_hdr, addr3); + fastrx.sa_offs = offsetof(struct ieee80211_hdr, addr4); break; default: goto clear; @@ -4751,6 +4757,7 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, struct sk_buff *skb = rx->skb; struct ieee80211_hdr *hdr = (void *)skb->data; struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); + static ieee80211_rx_result res; int orig_len = skb->len; int hdrlen = ieee80211_hdrlen(hdr->frame_control); int snap_offs = hdrlen; @@ -4812,7 +4819,8 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, snap_offs += IEEE80211_CCMP_HDR_LEN; } - if (!(status->rx_flags & IEEE80211_RX_AMSDU)) { + if (!ieee80211_vif_is_mesh(&rx->sdata->vif) && + !(status->rx_flags & IEEE80211_RX_AMSDU)) { if (!pskb_may_pull(skb, snap_offs + sizeof(*payload))) return false; @@ -4851,13 +4859,29 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, /* do the header conversion - first grab the addresses */ ether_addr_copy(addrs.da, skb->data + fast_rx->da_offs); ether_addr_copy(addrs.sa, skb->data + fast_rx->sa_offs); - skb_postpull_rcsum(skb, skb->data + snap_offs, - sizeof(rfc1042_header) + 2); - /* remove the SNAP but leave the ethertype */ - skb_pull(skb, snap_offs + sizeof(rfc1042_header)); + if (ieee80211_vif_is_mesh(&rx->sdata->vif)) { + skb_pull(skb, snap_offs - 2); + put_unaligned_be16(skb->len - 2, skb->data); + } else { + skb_postpull_rcsum(skb, skb->data + snap_offs, + sizeof(rfc1042_header) + 2); + + /* remove the SNAP but leave the ethertype */ + skb_pull(skb, snap_offs + sizeof(rfc1042_header)); + } /* push the addresses in front */ memcpy(skb_push(skb, sizeof(addrs)), &addrs, sizeof(addrs)); + res = ieee80211_rx_mesh_data(rx->sdata, rx->sta, rx->skb); + switch (res) { + case RX_QUEUED: + return true; + case RX_CONTINUE: + break; + default: + goto drop; + } + ieee80211_rx_8023(rx, fast_rx, orig_len); return true; -- cgit v1.2.3 From fe4a6d2db3bad41e9f22c860596f355af8493ebb Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 14 Mar 2023 10:59:56 +0100 Subject: wifi: mac80211: implement support for yet another mesh A-MSDU format MT7996 hardware supports mesh A-MSDU subframes in hardware, but uses a big-endian length field Signed-off-by: Felix Fietkau Signed-off-by: Ryder Lee Link: https://lore.kernel.org/r/20230314095956.62085-7-nbd@nbd.name Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 11 +++++++---- net/mac80211/rx.c | 22 ++++++++++++++++------ net/mac80211/sta_info.h | 5 ++++- net/wireless/util.c | 36 +++++++++++++++++++++++------------- 4 files changed, 50 insertions(+), 24 deletions(-) (limited to 'net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 7cebba1c4135..86cb048dc924 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -6274,10 +6274,13 @@ static inline int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr, * mesh control field. * * @skb: The input A-MSDU frame without any headers. - * @mesh_hdr: use standard compliant mesh A-MSDU subframe header + * @mesh_hdr: the type of mesh header to test + * 0: non-mesh A-MSDU length field + * 1: big-endian mesh A-MSDU length field + * 2: little-endian mesh A-MSDU length field * Returns: true if subframe header lengths are valid for the @mesh_hdr mode */ -bool ieee80211_is_valid_amsdu(struct sk_buff *skb, bool mesh_hdr); +bool ieee80211_is_valid_amsdu(struct sk_buff *skb, u8 mesh_hdr); /** * ieee80211_amsdu_to_8023s - decode an IEEE 802.11n A-MSDU frame @@ -6294,13 +6297,13 @@ bool ieee80211_is_valid_amsdu(struct sk_buff *skb, bool mesh_hdr); * @extra_headroom: The hardware extra headroom for SKBs in the @list. * @check_da: DA to check in the inner ethernet header, or NULL * @check_sa: SA to check in the inner ethernet header, or NULL - * @mesh_control: A-MSDU subframe header includes the mesh control field + * @mesh_control: see mesh_hdr in ieee80211_is_valid_amsdu */ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list, const u8 *addr, enum nl80211_iftype iftype, const unsigned int extra_headroom, const u8 *check_da, const u8 *check_sa, - bool mesh_control); + u8 mesh_control); /** * ieee80211_get_8023_tunnel_proto - get RFC1042 or bridge tunnel encap protocol diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 85fb1d3eeb2f..1c957194554b 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2983,13 +2983,23 @@ __ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx, u8 data_offset) return RX_DROP_UNUSABLE; if (rx->sta && rx->sta->amsdu_mesh_control < 0) { - bool valid_std = ieee80211_is_valid_amsdu(skb, true); - bool valid_nonstd = ieee80211_is_valid_amsdu(skb, false); + s8 valid = -1; + int i; + + for (i = 0; i <= 2; i++) { + if (!ieee80211_is_valid_amsdu(skb, i)) + continue; + + if (valid >= 0) { + /* ambiguous */ + valid = -1; + break; + } + + valid = i; + } - if (valid_std && !valid_nonstd) - rx->sta->amsdu_mesh_control = 1; - else if (valid_nonstd && !valid_std) - rx->sta->amsdu_mesh_control = 0; + rx->sta->amsdu_mesh_control = valid; } ieee80211_amsdu_to_8023s(skb, &frame_list, dev->dev_addr, diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index e8e482a82d77..f354d470e174 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -623,7 +623,10 @@ struct link_sta_info { * @cparams: CoDel parameters for this station. * @reserved_tid: reserved TID (if any, otherwise IEEE80211_TID_UNRESERVED) * @amsdu_mesh_control: track the mesh A-MSDU format used by the peer - * (-1: not yet known, 0: non-standard [without mesh header], 1: standard) + * (-1: not yet known, + * 0: non-mesh A-MSDU length field + * 1: big-endian mesh A-MSDU length field + * 2: little-endian mesh A-MSDU length field) * @fast_tx: TX fastpath information * @fast_rx: RX fastpath information * @tdls_chandef: a TDLS peer can have a wider chandef that is compatible to diff --git a/net/wireless/util.c b/net/wireless/util.c index d1a89e82ead0..3bc0c3072e78 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -776,7 +776,24 @@ __ieee80211_amsdu_copy(struct sk_buff *skb, unsigned int hlen, return frame; } -bool ieee80211_is_valid_amsdu(struct sk_buff *skb, bool mesh_hdr) +static u16 +ieee80211_amsdu_subframe_length(void *field, u8 mesh_flags, u8 hdr_type) +{ + __le16 *field_le = field; + __be16 *field_be = field; + u16 len; + + if (hdr_type >= 2) + len = le16_to_cpu(*field_le); + else + len = be16_to_cpu(*field_be); + if (hdr_type) + len += __ieee80211_get_mesh_hdrlen(mesh_flags); + + return len; +} + +bool ieee80211_is_valid_amsdu(struct sk_buff *skb, u8 mesh_hdr) { int offset = 0, remaining, subframe_len, padding; @@ -790,12 +807,8 @@ bool ieee80211_is_valid_amsdu(struct sk_buff *skb, bool mesh_hdr) if (skb_copy_bits(skb, offset + 2 * ETH_ALEN, &hdr, sizeof(hdr)) < 0) return false; - if (mesh_hdr) - len = le16_to_cpu(*(__le16 *)&hdr.len) + - __ieee80211_get_mesh_hdrlen(hdr.mesh_flags); - else - len = ntohs(hdr.len); - + len = ieee80211_amsdu_subframe_length(&hdr.len, hdr.mesh_flags, + mesh_hdr); subframe_len = sizeof(struct ethhdr) + len; padding = (4 - subframe_len) & 0x3; remaining = skb->len - offset; @@ -812,7 +825,7 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list, const u8 *addr, enum nl80211_iftype iftype, const unsigned int extra_headroom, const u8 *check_da, const u8 *check_sa, - bool mesh_control) + u8 mesh_control) { unsigned int hlen = ALIGN(extra_headroom, 4); struct sk_buff *frame = NULL; @@ -837,11 +850,8 @@ void ieee80211_amsdu_to_8023s(struct sk_buff *skb, struct sk_buff_head *list, skb_copy_bits(skb, offset, &hdr, copy_len); if (iftype == NL80211_IFTYPE_MESH_POINT) mesh_len = __ieee80211_get_mesh_hdrlen(hdr.flags); - if (mesh_control) - len = le16_to_cpu(*(__le16 *)&hdr.eth.h_proto) + mesh_len; - else - len = ntohs(hdr.eth.h_proto); - + len = ieee80211_amsdu_subframe_length(&hdr.eth.h_proto, hdr.flags, + mesh_control); subframe_len = sizeof(struct ethhdr) + len; padding = (4 - subframe_len) & 0x3; -- cgit v1.2.3 From 4e348c6c6e23491ae6eb5e077848a42d0562339c Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 14 Mar 2023 10:59:50 +0100 Subject: wifi: mac80211: fix qos on mesh interfaces When ieee80211_select_queue is called for mesh, the sta pointer is usually NULL, since the nexthop is looked up much later in the tx path. Explicitly check for unicast address in that case in order to make qos work again. Cc: stable@vger.kernel.org Fixes: 50e2ab392919 ("wifi: mac80211: fix queue selection for mesh/OCB interfaces") Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230314095956.62085-1-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/wme.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c index a12c63638680..1601be576414 100644 --- a/net/mac80211/wme.c +++ b/net/mac80211/wme.c @@ -147,6 +147,7 @@ u16 ieee80211_select_queue_80211(struct ieee80211_sub_if_data *sdata, u16 ieee80211_select_queue(struct ieee80211_sub_if_data *sdata, struct sta_info *sta, struct sk_buff *skb) { + const struct ethhdr *eth = (void *)skb->data; struct mac80211_qos_map *qos_map; bool qos; @@ -154,8 +155,9 @@ u16 ieee80211_select_queue(struct ieee80211_sub_if_data *sdata, skb_get_hash(skb); /* all mesh/ocb stations are required to support WME */ - if (sta && (sdata->vif.type == NL80211_IFTYPE_MESH_POINT || - sdata->vif.type == NL80211_IFTYPE_OCB)) + if ((sdata->vif.type == NL80211_IFTYPE_MESH_POINT && + !is_multicast_ether_addr(eth->h_dest)) || + (sdata->vif.type == NL80211_IFTYPE_OCB && sta)) qos = true; else if (sta) qos = sta->sta.wme; -- cgit v1.2.3 From f355f70145744518ca1d9799b42f4a8da9aa0d36 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Tue, 14 Mar 2023 10:59:52 +0100 Subject: wifi: mac80211: fix mesh path discovery based on unicast packets If a packet has reached its intended destination, it was bumped to the code that accepts it, without first checking if a mesh_path needs to be created based on the discovered source. Fix this by moving the destination address check further down. Cc: stable@vger.kernel.org Fixes: 986e43b19ae9 ("wifi: mac80211: fix receiving A-MSDU frames on mesh interfaces") Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230314095956.62085-3-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index f7fdfe710951..e8de500eb9f3 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2765,17 +2765,6 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta mesh_rmc_check(sdata, eth->h_source, mesh_hdr)) return RX_DROP_MONITOR; - /* Frame has reached destination. Don't forward */ - if (ether_addr_equal(sdata->vif.addr, eth->h_dest)) - goto rx_accept; - - if (!ifmsh->mshcfg.dot11MeshForwarding) { - if (is_multicast_ether_addr(eth->h_dest)) - goto rx_accept; - - return RX_DROP_MONITOR; - } - /* forward packet */ if (sdata->crypto_tx_tailroom_needed_cnt) tailroom = IEEE80211_ENCRYPT_TAILROOM; @@ -2814,6 +2803,17 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta rcu_read_unlock(); } + /* Frame has reached destination. Don't forward */ + if (ether_addr_equal(sdata->vif.addr, eth->h_dest)) + goto rx_accept; + + if (!ifmsh->mshcfg.dot11MeshForwarding) { + if (is_multicast_ether_addr(eth->h_dest)) + goto rx_accept; + + return RX_DROP_MONITOR; + } + skb_set_queue_mapping(skb, ieee802_1d_to_ac[skb->priority]); ieee80211_fill_mesh_addresses(&hdr, &hdr.frame_control, -- cgit v1.2.3 From bd54f3c29077f23dad92ef82a78061b40be30c65 Mon Sep 17 00:00:00 2001 From: Aloka Dixit Date: Mon, 5 Dec 2022 16:50:37 -0800 Subject: wifi: mac80211: generate EMA beacons in AP mode Add APIs to generate an array of beacons for an EMA AP (enhanced multiple BSSID advertisements), each including a single MBSSID element. EMA profile periodicity equals the count of elements. - ieee80211_beacon_get_template_ema_list() - Generate and return all EMA beacon templates. Drivers must call ieee80211_beacon_free_ema_list() to free the memory. No change in the prototype for the existing API, ieee80211_beacon_get_template(), which should be used for non-EMA AP. - ieee80211_beacon_get_template_ema_index() - Generate a beacon which includes the multiple BSSID element at the given index. Drivers can use this function in a loop until NULL is returned which indicates end of available MBSSID elements. - ieee80211_beacon_free_ema_list() - free the memory allocated for the list of EMA beacon templates. Modify existing functions ieee80211_beacon_get_ap(), ieee80211_get_mbssid_beacon_len() and ieee80211_beacon_add_mbssid() to accept a new parameter for EMA index. Signed-off-by: Aloka Dixit Co-developed-by: John Crispin Signed-off-by: John Crispin Link: https://lore.kernel.org/r/20221206005040.3177-2-quic_alokad@quicinc.com Signed-off-by: Johannes Berg --- include/net/mac80211.h | 68 +++++++++++++++++++++++ net/mac80211/cfg.c | 11 ++-- net/mac80211/ieee80211_i.h | 10 +++- net/mac80211/tx.c | 134 +++++++++++++++++++++++++++++++++++++++++---- 4 files changed, 205 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index fcfe3e9aff3d..679421d37a42 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -5285,6 +5285,74 @@ ieee80211_beacon_get_template(struct ieee80211_hw *hw, struct ieee80211_mutable_offsets *offs, unsigned int link_id); +/** + * ieee80211_beacon_get_template_ema_index - EMA beacon template generation + * @hw: pointer obtained from ieee80211_alloc_hw(). + * @vif: &struct ieee80211_vif pointer from the add_interface callback. + * @offs: &struct ieee80211_mutable_offsets pointer to struct that will + * receive the offsets that may be updated by the driver. + * @link_id: the link id to which the beacon belongs (or 0 for a non-MLD AP). + * @ema_index: index of the beacon in the EMA set. + * + * This function follows the same rules as ieee80211_beacon_get_template() + * but returns a beacon template which includes multiple BSSID element at the + * requested index. + * + * Return: The beacon template. %NULL indicates the end of EMA templates. + */ +struct sk_buff * +ieee80211_beacon_get_template_ema_index(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_mutable_offsets *offs, + unsigned int link_id, u8 ema_index); + +/** + * struct ieee80211_ema_beacons - List of EMA beacons + * @cnt: count of EMA beacons. + * + * @bcn: array of EMA beacons. + * @bcn.skb: the skb containing this specific beacon + * @bcn.offs: &struct ieee80211_mutable_offsets pointer to struct that will + * receive the offsets that may be updated by the driver. + */ +struct ieee80211_ema_beacons { + u8 cnt; + struct { + struct sk_buff *skb; + struct ieee80211_mutable_offsets offs; + } bcn[]; +}; + +/** + * ieee80211_beacon_get_template_ema_list - EMA beacon template generation + * @hw: pointer obtained from ieee80211_alloc_hw(). + * @vif: &struct ieee80211_vif pointer from the add_interface callback. + * @link_id: the link id to which the beacon belongs (or 0 for a non-MLD AP) + * + * This function follows the same rules as ieee80211_beacon_get_template() + * but allocates and returns a pointer to list of all beacon templates required + * to cover all profiles in the multiple BSSID set. Each template includes only + * one multiple BSSID element. + * + * Driver must call ieee80211_beacon_free_ema_list() to free the memory. + * + * Return: EMA beacon templates of type struct ieee80211_ema_beacons *. + * %NULL on error. + */ +struct ieee80211_ema_beacons * +ieee80211_beacon_get_template_ema_list(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + unsigned int link_id); + +/** + * ieee80211_beacon_free_ema_list - free an EMA beacon template list + * @ema_beacons: list of EMA beacons of type &struct ieee80211_ema_beacons pointers. + * + * This function will free a list previously acquired by calling + * ieee80211_beacon_get_template_ema_list() + */ +void ieee80211_beacon_free_ema_list(struct ieee80211_ema_beacons *ema_beacons); + /** * ieee80211_beacon_get_tim - beacon generation function * @hw: pointer obtained from ieee80211_alloc_hw(). diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 760ad934f9e1..db5fa334b801 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1122,11 +1122,11 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata, if (params->mbssid_ies) { mbssid = params->mbssid_ies; size += struct_size(new->mbssid_ies, elem, mbssid->cnt); - size += ieee80211_get_mbssid_beacon_len(mbssid); + size += ieee80211_get_mbssid_beacon_len(mbssid, mbssid->cnt); } else if (old && old->mbssid_ies) { mbssid = old->mbssid_ies; size += struct_size(new->mbssid_ies, elem, mbssid->cnt); - size += ieee80211_get_mbssid_beacon_len(mbssid); + size += ieee80211_get_mbssid_beacon_len(mbssid, mbssid->cnt); } new = kzalloc(size, GFP_KERNEL); @@ -3406,8 +3406,11 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon) len = beacon->head_len + beacon->tail_len + beacon->beacon_ies_len + beacon->proberesp_ies_len + beacon->assocresp_ies_len + - beacon->probe_resp_len + beacon->lci_len + beacon->civicloc_len + - ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies); + beacon->probe_resp_len + beacon->lci_len + beacon->civicloc_len; + + if (beacon->mbssid_ies) + len += ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies, + beacon->mbssid_ies->cnt); new_beacon = kzalloc(sizeof(*new_beacon) + len, GFP_KERNEL); if (!new_beacon) diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index c8c037a9e4e2..84d10e993eca 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -1186,13 +1186,17 @@ ieee80211_vif_get_shift(struct ieee80211_vif *vif) } static inline int -ieee80211_get_mbssid_beacon_len(struct cfg80211_mbssid_elems *elems) +ieee80211_get_mbssid_beacon_len(struct cfg80211_mbssid_elems *elems, u8 i) { - int i, len = 0; + int len = 0; - if (!elems) + if (!elems || !elems->cnt || i > elems->cnt) return 0; + if (i < elems->cnt) + return elems->elem[i].len; + + /* i == elems->cnt, calculate total length of all MBSSID elements */ for (i = 0; i < elems->cnt; i++) len += elems->elem[i].len; diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index de17926484bd..139eec6c64da 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -5212,13 +5212,20 @@ ieee80211_beacon_get_finish(struct ieee80211_hw *hw, } static void -ieee80211_beacon_add_mbssid(struct sk_buff *skb, struct beacon_data *beacon) +ieee80211_beacon_add_mbssid(struct sk_buff *skb, struct beacon_data *beacon, + u8 i) { - int i; + if (!beacon->mbssid_ies || !beacon->mbssid_ies->cnt || + i > beacon->mbssid_ies->cnt) + return; - if (!beacon->mbssid_ies) + if (i < beacon->mbssid_ies->cnt) { + skb_put_data(skb, beacon->mbssid_ies->elem[i].data, + beacon->mbssid_ies->elem[i].len); return; + } + /* i == beacon->mbssid_ies->cnt, include all MBSSID elements */ for (i = 0; i < beacon->mbssid_ies->cnt; i++) skb_put_data(skb, beacon->mbssid_ies->elem[i].data, beacon->mbssid_ies->elem[i].len); @@ -5231,7 +5238,8 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw, struct ieee80211_mutable_offsets *offs, bool is_template, struct beacon_data *beacon, - struct ieee80211_chanctx_conf *chanctx_conf) + struct ieee80211_chanctx_conf *chanctx_conf, + u8 ema_index) { struct ieee80211_local *local = hw_to_local(hw); struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif); @@ -5250,7 +5258,9 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw, /* headroom, head length, * tail length, maximum TIM length and multiple BSSID length */ - mbssid_len = ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies); + mbssid_len = ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies, + ema_index); + skb = dev_alloc_skb(local->tx_headroom + beacon->head_len + beacon->tail_len + 256 + local->hw.extra_beacon_tailroom + mbssid_len); @@ -5268,7 +5278,7 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw, offs->cntdwn_counter_offs[0] = beacon->cntdwn_counter_offsets[0]; if (mbssid_len) { - ieee80211_beacon_add_mbssid(skb, beacon); + ieee80211_beacon_add_mbssid(skb, beacon, ema_index); offs->mbssid_off = skb->len - mbssid_len; } @@ -5287,12 +5297,51 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw, return skb; } +static struct ieee80211_ema_beacons * +ieee80211_beacon_get_ap_ema_list(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_link_data *link, + struct ieee80211_mutable_offsets *offs, + bool is_template, struct beacon_data *beacon, + struct ieee80211_chanctx_conf *chanctx_conf) +{ + struct ieee80211_ema_beacons *ema = NULL; + + if (!beacon->mbssid_ies || !beacon->mbssid_ies->cnt) + return NULL; + + ema = kzalloc(struct_size(ema, bcn, beacon->mbssid_ies->cnt), + GFP_ATOMIC); + if (!ema) + return NULL; + + for (ema->cnt = 0; ema->cnt < beacon->mbssid_ies->cnt; ema->cnt++) { + ema->bcn[ema->cnt].skb = + ieee80211_beacon_get_ap(hw, vif, link, + &ema->bcn[ema->cnt].offs, + is_template, beacon, + chanctx_conf, ema->cnt); + if (!ema->bcn[ema->cnt].skb) + break; + } + + if (ema->cnt == beacon->mbssid_ies->cnt) + return ema; + + ieee80211_beacon_free_ema_list(ema); + return NULL; +} + +#define IEEE80211_INCLUDE_ALL_MBSSID_ELEMS -1 + static struct sk_buff * __ieee80211_beacon_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_mutable_offsets *offs, bool is_template, - unsigned int link_id) + unsigned int link_id, + int ema_index, + struct ieee80211_ema_beacons **ema_beacons) { struct ieee80211_local *local = hw_to_local(hw); struct beacon_data *beacon = NULL; @@ -5321,8 +5370,29 @@ __ieee80211_beacon_get(struct ieee80211_hw *hw, if (!beacon) goto out; - skb = ieee80211_beacon_get_ap(hw, vif, link, offs, is_template, - beacon, chanctx_conf); + if (ema_beacons) { + *ema_beacons = + ieee80211_beacon_get_ap_ema_list(hw, vif, link, + offs, + is_template, + beacon, + chanctx_conf); + } else { + if (beacon->mbssid_ies && beacon->mbssid_ies->cnt) { + if (ema_index >= beacon->mbssid_ies->cnt) + goto out; /* End of MBSSID elements */ + + if (ema_index <= IEEE80211_INCLUDE_ALL_MBSSID_ELEMS) + ema_index = beacon->mbssid_ies->cnt; + } else { + ema_index = 0; + } + + skb = ieee80211_beacon_get_ap(hw, vif, link, offs, + is_template, beacon, + chanctx_conf, + ema_index); + } } else if (sdata->vif.type == NL80211_IFTYPE_ADHOC) { struct ieee80211_if_ibss *ifibss = &sdata->u.ibss; struct ieee80211_hdr *hdr; @@ -5410,10 +5480,50 @@ ieee80211_beacon_get_template(struct ieee80211_hw *hw, struct ieee80211_mutable_offsets *offs, unsigned int link_id) { - return __ieee80211_beacon_get(hw, vif, offs, true, link_id); + return __ieee80211_beacon_get(hw, vif, offs, true, link_id, + IEEE80211_INCLUDE_ALL_MBSSID_ELEMS, NULL); } EXPORT_SYMBOL(ieee80211_beacon_get_template); +struct sk_buff * +ieee80211_beacon_get_template_ema_index(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + struct ieee80211_mutable_offsets *offs, + unsigned int link_id, u8 ema_index) +{ + return __ieee80211_beacon_get(hw, vif, offs, true, link_id, ema_index, + NULL); +} +EXPORT_SYMBOL(ieee80211_beacon_get_template_ema_index); + +void ieee80211_beacon_free_ema_list(struct ieee80211_ema_beacons *ema_beacons) +{ + u8 i; + + if (!ema_beacons) + return; + + for (i = 0; i < ema_beacons->cnt; i++) + kfree_skb(ema_beacons->bcn[i].skb); + + kfree(ema_beacons); +} +EXPORT_SYMBOL(ieee80211_beacon_free_ema_list); + +struct ieee80211_ema_beacons * +ieee80211_beacon_get_template_ema_list(struct ieee80211_hw *hw, + struct ieee80211_vif *vif, + unsigned int link_id) +{ + struct ieee80211_ema_beacons *ema_beacons = NULL; + + WARN_ON(__ieee80211_beacon_get(hw, vif, NULL, false, link_id, 0, + &ema_beacons)); + + return ema_beacons; +} +EXPORT_SYMBOL(ieee80211_beacon_get_template_ema_list); + struct sk_buff *ieee80211_beacon_get_tim(struct ieee80211_hw *hw, struct ieee80211_vif *vif, u16 *tim_offset, u16 *tim_length, @@ -5421,7 +5531,9 @@ struct sk_buff *ieee80211_beacon_get_tim(struct ieee80211_hw *hw, { struct ieee80211_mutable_offsets offs = {}; struct sk_buff *bcn = __ieee80211_beacon_get(hw, vif, &offs, false, - link_id); + link_id, + IEEE80211_INCLUDE_ALL_MBSSID_ELEMS, + NULL); struct sk_buff *copy; int shift; -- cgit v1.2.3 From 915efd8a446b74442039d31689d5d863caf82517 Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Tue, 21 Mar 2023 14:52:31 +0100 Subject: xdp: bpf_xdp_metadata use EOPNOTSUPP for no driver support MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When driver doesn't implement a bpf_xdp_metadata kfunc the fallback implementation returns EOPNOTSUPP, which indicate device driver doesn't implement this kfunc. Currently many drivers also return EOPNOTSUPP when the hint isn't available, which is ambiguous from an API point of view. Instead change drivers to return ENODATA in these cases. There can be natural cases why a driver doesn't provide any hardware info for a specific hint, even on a frame to frame basis (e.g. PTP). Lets keep these cases as separate return codes. When describing the return values, adjust the function kernel-doc layout to get proper rendering for the return values. Fixes: ab46182d0dcb ("net/mlx4_en: Support RX XDP metadata") Fixes: bc8d405b1ba9 ("net/mlx5e: Support RX XDP metadata") Fixes: 306531f0249f ("veth: Support RX XDP metadata") Fixes: 3d76a4d3d4e5 ("bpf: XDP metadata RX kfuncs") Signed-off-by: Jesper Dangaard Brouer Acked-by: Stanislav Fomichev Acked-by: Toke Høiland-Jørgensen Acked-by: Tariq Toukan Link: https://lore.kernel.org/r/167940675120.2718408.8176058626864184420.stgit@firesoul Signed-off-by: Alexei Starovoitov --- Documentation/networking/xdp-rx-metadata.rst | 7 +++++-- drivers/net/ethernet/mellanox/mlx4/en_rx.c | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c | 4 ++-- drivers/net/veth.c | 4 ++-- net/core/xdp.c | 10 ++++++++-- 5 files changed, 19 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/Documentation/networking/xdp-rx-metadata.rst b/Documentation/networking/xdp-rx-metadata.rst index aac63fc2d08b..25ce72af81c2 100644 --- a/Documentation/networking/xdp-rx-metadata.rst +++ b/Documentation/networking/xdp-rx-metadata.rst @@ -23,10 +23,13 @@ metadata is supported, this set will grow: An XDP program can use these kfuncs to read the metadata into stack variables for its own consumption. Or, to pass the metadata on to other consumers, an XDP program can store it into the metadata area carried -ahead of the packet. +ahead of the packet. Not all packets will necessary have the requested +metadata available in which case the driver returns ``-ENODATA``. Not all kfuncs have to be implemented by the device driver; when not -implemented, the default ones that return ``-EOPNOTSUPP`` will be used. +implemented, the default ones that return ``-EOPNOTSUPP`` will be used +to indicate the device driver have not implemented this kfunc. + Within an XDP frame, the metadata layout (accessed via ``xdp_buff``) is as follows:: diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 0869d4fff17b..4b5e459b6d49 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -674,7 +674,7 @@ int mlx4_en_xdp_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp) struct mlx4_en_xdp_buff *_ctx = (void *)ctx; if (unlikely(_ctx->ring->hwtstamp_rx_filter != HWTSTAMP_FILTER_ALL)) - return -EOPNOTSUPP; + return -ENODATA; *timestamp = mlx4_en_get_hwtstamp(_ctx->mdev, mlx4_en_get_cqe_ts(_ctx->cqe)); @@ -686,7 +686,7 @@ int mlx4_en_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash) struct mlx4_en_xdp_buff *_ctx = (void *)ctx; if (unlikely(!(_ctx->dev->features & NETIF_F_RXHASH))) - return -EOPNOTSUPP; + return -ENODATA; *hash = be32_to_cpu(_ctx->cqe->immed_rss_invalid); return 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index bcd6370de440..c5dae48b7932 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -162,7 +162,7 @@ static int mlx5e_xdp_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp) const struct mlx5e_xdp_buff *_ctx = (void *)ctx; if (unlikely(!mlx5e_rx_hw_stamp(_ctx->rq->tstamp))) - return -EOPNOTSUPP; + return -ENODATA; *timestamp = mlx5e_cqe_ts_to_ns(_ctx->rq->ptp_cyc2time, _ctx->rq->clock, get_cqe_ts(_ctx->cqe)); @@ -174,7 +174,7 @@ static int mlx5e_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash) const struct mlx5e_xdp_buff *_ctx = (void *)ctx; if (unlikely(!(_ctx->xdp.rxq->dev->features & NETIF_F_RXHASH))) - return -EOPNOTSUPP; + return -ENODATA; *hash = be32_to_cpu(_ctx->cqe->rss_hash_result); return 0; diff --git a/drivers/net/veth.c b/drivers/net/veth.c index 1bb54de7124d..046461ee42ea 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -1610,7 +1610,7 @@ static int veth_xdp_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp) struct veth_xdp_buff *_ctx = (void *)ctx; if (!_ctx->skb) - return -EOPNOTSUPP; + return -ENODATA; *timestamp = skb_hwtstamps(_ctx->skb)->hwtstamp; return 0; @@ -1621,7 +1621,7 @@ static int veth_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash) struct veth_xdp_buff *_ctx = (void *)ctx; if (!_ctx->skb) - return -EOPNOTSUPP; + return -ENODATA; *hash = skb_get_hash(_ctx->skb); return 0; diff --git a/net/core/xdp.c b/net/core/xdp.c index 8c92fc553317..247797168579 100644 --- a/net/core/xdp.c +++ b/net/core/xdp.c @@ -720,7 +720,10 @@ __diag_ignore_all("-Wmissing-prototypes", * @ctx: XDP context pointer. * @timestamp: Return value pointer. * - * Returns 0 on success or ``-errno`` on error. + * Return: + * * Returns 0 on success or ``-errno`` on error. + * * ``-EOPNOTSUPP`` : means device driver does not implement kfunc + * * ``-ENODATA`` : means no RX-timestamp available for this frame */ __bpf_kfunc int bpf_xdp_metadata_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp) { @@ -732,7 +735,10 @@ __bpf_kfunc int bpf_xdp_metadata_rx_timestamp(const struct xdp_md *ctx, u64 *tim * @ctx: XDP context pointer. * @hash: Return value pointer. * - * Returns 0 on success or ``-errno`` on error. + * Return: + * * Returns 0 on success or ``-errno`` on error. + * * ``-EOPNOTSUPP`` : means device driver doesn't implement kfunc + * * ``-ENODATA`` : means no RX-hash available for this frame */ __bpf_kfunc int bpf_xdp_metadata_rx_hash(const struct xdp_md *ctx, u32 *hash) { -- cgit v1.2.3 From 6f56ad1b92328997e1b1792047099df6f8d7acb5 Mon Sep 17 00:00:00 2001 From: Jeremy Sowden Date: Wed, 15 Mar 2023 21:48:01 +0000 Subject: netfilter: nft_redir: use `struct nf_nat_range2` throughout and deduplicate eval call-backs `nf_nat_redirect_ipv4` takes a `struct nf_nat_ipv4_multi_range_compat`, but converts it internally to a `struct nf_nat_range2`. Change the function to take the latter, factor out the code now shared with `nf_nat_redirect_ipv6`, move the conversion to the xt_REDIRECT module, and update the ipv4 range initialization in the nft_redir module. Replace a bare hex constant for 127.0.0.1 with a macro. Remove `WARN_ON`. `nf_nat_setup_info` calls `nf_ct_is_confirmed`: /* Can't setup nat info for confirmed ct. */ if (nf_ct_is_confirmed(ct)) return NF_ACCEPT; This means that `ct` cannot be null or the kernel will crash, and implies that `ctinfo` is `IP_CT_NEW` or `IP_CT_RELATED`. nft_redir has separate ipv4 and ipv6 call-backs which share much of their code, and an inet one switch containing a switch that calls one of the others based on the family of the packet. Merge the ipv4 and ipv6 ones into the inet one in order to get rid of the duplicate code. Const-qualify the `priv` pointer since we don't need to write through it. Assign `priv->flags` to the range instead of OR-ing it in. Set the `NF_NAT_RANGE_PROTO_SPECIFIED` flag once during init, rather than on every eval. Signed-off-by: Jeremy Sowden Signed-off-by: Florian Westphal --- include/net/netfilter/nf_nat_redirect.h | 3 +- net/netfilter/nf_nat_redirect.c | 71 +++++++++++++--------------- net/netfilter/nft_redir.c | 84 ++++++++++++--------------------- net/netfilter/xt_REDIRECT.c | 10 +++- 4 files changed, 72 insertions(+), 96 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_nat_redirect.h b/include/net/netfilter/nf_nat_redirect.h index 2418653a66db..279380de904c 100644 --- a/include/net/netfilter/nf_nat_redirect.h +++ b/include/net/netfilter/nf_nat_redirect.h @@ -6,8 +6,7 @@ #include unsigned int -nf_nat_redirect_ipv4(struct sk_buff *skb, - const struct nf_nat_ipv4_multi_range_compat *mr, +nf_nat_redirect_ipv4(struct sk_buff *skb, const struct nf_nat_range2 *range, unsigned int hooknum); unsigned int nf_nat_redirect_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range, diff --git a/net/netfilter/nf_nat_redirect.c b/net/netfilter/nf_nat_redirect.c index f91579c821e9..6616ba5d0b04 100644 --- a/net/netfilter/nf_nat_redirect.c +++ b/net/netfilter/nf_nat_redirect.c @@ -10,6 +10,7 @@ #include #include +#include #include #include #include @@ -24,54 +25,56 @@ #include #include +static unsigned int +nf_nat_redirect(struct sk_buff *skb, const struct nf_nat_range2 *range, + const union nf_inet_addr *newdst) +{ + struct nf_nat_range2 newrange; + enum ip_conntrack_info ctinfo; + struct nf_conn *ct; + + ct = nf_ct_get(skb, &ctinfo); + + memset(&newrange, 0, sizeof(newrange)); + + newrange.flags = range->flags | NF_NAT_RANGE_MAP_IPS; + newrange.min_addr = *newdst; + newrange.max_addr = *newdst; + newrange.min_proto = range->min_proto; + newrange.max_proto = range->max_proto; + + return nf_nat_setup_info(ct, &newrange, NF_NAT_MANIP_DST); +} + unsigned int -nf_nat_redirect_ipv4(struct sk_buff *skb, - const struct nf_nat_ipv4_multi_range_compat *mr, +nf_nat_redirect_ipv4(struct sk_buff *skb, const struct nf_nat_range2 *range, unsigned int hooknum) { - struct nf_conn *ct; - enum ip_conntrack_info ctinfo; - __be32 newdst; - struct nf_nat_range2 newrange; + union nf_inet_addr newdst = {}; WARN_ON(hooknum != NF_INET_PRE_ROUTING && hooknum != NF_INET_LOCAL_OUT); - ct = nf_ct_get(skb, &ctinfo); - WARN_ON(!(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED))); - /* Local packets: make them go to loopback */ if (hooknum == NF_INET_LOCAL_OUT) { - newdst = htonl(0x7F000001); + newdst.ip = htonl(INADDR_LOOPBACK); } else { const struct in_device *indev; - newdst = 0; - indev = __in_dev_get_rcu(skb->dev); if (indev) { const struct in_ifaddr *ifa; ifa = rcu_dereference(indev->ifa_list); if (ifa) - newdst = ifa->ifa_local; + newdst.ip = ifa->ifa_local; } - if (!newdst) + if (!newdst.ip) return NF_DROP; } - /* Transfer from original range. */ - memset(&newrange.min_addr, 0, sizeof(newrange.min_addr)); - memset(&newrange.max_addr, 0, sizeof(newrange.max_addr)); - newrange.flags = mr->range[0].flags | NF_NAT_RANGE_MAP_IPS; - newrange.min_addr.ip = newdst; - newrange.max_addr.ip = newdst; - newrange.min_proto = mr->range[0].min; - newrange.max_proto = mr->range[0].max; - - /* Hand modified range to generic setup. */ - return nf_nat_setup_info(ct, &newrange, NF_NAT_MANIP_DST); + return nf_nat_redirect(skb, range, &newdst); } EXPORT_SYMBOL_GPL(nf_nat_redirect_ipv4); @@ -81,14 +84,10 @@ unsigned int nf_nat_redirect_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range, unsigned int hooknum) { - struct nf_nat_range2 newrange; - struct in6_addr newdst; - enum ip_conntrack_info ctinfo; - struct nf_conn *ct; + union nf_inet_addr newdst = {}; - ct = nf_ct_get(skb, &ctinfo); if (hooknum == NF_INET_LOCAL_OUT) { - newdst = loopback_addr; + newdst.in6 = loopback_addr; } else { struct inet6_dev *idev; struct inet6_ifaddr *ifa; @@ -98,7 +97,7 @@ nf_nat_redirect_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range, if (idev != NULL) { read_lock_bh(&idev->lock); list_for_each_entry(ifa, &idev->addr_list, if_list) { - newdst = ifa->addr; + newdst.in6 = ifa->addr; addr = true; break; } @@ -109,12 +108,6 @@ nf_nat_redirect_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range, return NF_DROP; } - newrange.flags = range->flags | NF_NAT_RANGE_MAP_IPS; - newrange.min_addr.in6 = newdst; - newrange.max_addr.in6 = newdst; - newrange.min_proto = range->min_proto; - newrange.max_proto = range->max_proto; - - return nf_nat_setup_info(ct, &newrange, NF_NAT_MANIP_DST); + return nf_nat_redirect(skb, range, &newdst); } EXPORT_SYMBOL_GPL(nf_nat_redirect_ipv6); diff --git a/net/netfilter/nft_redir.c b/net/netfilter/nft_redir.c index 67cec56bc84a..a70196ffcb1e 100644 --- a/net/netfilter/nft_redir.c +++ b/net/netfilter/nft_redir.c @@ -64,6 +64,8 @@ static int nft_redir_init(const struct nft_ctx *ctx, } else { priv->sreg_proto_max = priv->sreg_proto_min; } + + priv->flags |= NF_NAT_RANGE_PROTO_SPECIFIED; } if (tb[NFTA_REDIR_FLAGS]) { @@ -99,25 +101,37 @@ nla_put_failure: return -1; } -static void nft_redir_ipv4_eval(const struct nft_expr *expr, - struct nft_regs *regs, - const struct nft_pktinfo *pkt) +static void nft_redir_eval(const struct nft_expr *expr, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) { - struct nft_redir *priv = nft_expr_priv(expr); - struct nf_nat_ipv4_multi_range_compat mr; + const struct nft_redir *priv = nft_expr_priv(expr); + struct nf_nat_range2 range; - memset(&mr, 0, sizeof(mr)); + memset(&range, 0, sizeof(range)); + range.flags = priv->flags; if (priv->sreg_proto_min) { - mr.range[0].min.all = (__force __be16)nft_reg_load16( - ®s->data[priv->sreg_proto_min]); - mr.range[0].max.all = (__force __be16)nft_reg_load16( - ®s->data[priv->sreg_proto_max]); - mr.range[0].flags |= NF_NAT_RANGE_PROTO_SPECIFIED; + range.min_proto.all = (__force __be16) + nft_reg_load16(®s->data[priv->sreg_proto_min]); + range.max_proto.all = (__force __be16) + nft_reg_load16(®s->data[priv->sreg_proto_max]); } - mr.range[0].flags |= priv->flags; - - regs->verdict.code = nf_nat_redirect_ipv4(pkt->skb, &mr, nft_hook(pkt)); + switch (nft_pf(pkt)) { + case NFPROTO_IPV4: + regs->verdict.code = nf_nat_redirect_ipv4(pkt->skb, &range, + nft_hook(pkt)); + break; +#ifdef CONFIG_NF_TABLES_IPV6 + case NFPROTO_IPV6: + regs->verdict.code = nf_nat_redirect_ipv6(pkt->skb, &range, + nft_hook(pkt)); + break; +#endif + default: + WARN_ON_ONCE(1); + break; + } } static void @@ -130,7 +144,7 @@ static struct nft_expr_type nft_redir_ipv4_type; static const struct nft_expr_ops nft_redir_ipv4_ops = { .type = &nft_redir_ipv4_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_redir)), - .eval = nft_redir_ipv4_eval, + .eval = nft_redir_eval, .init = nft_redir_init, .destroy = nft_redir_ipv4_destroy, .dump = nft_redir_dump, @@ -148,28 +162,6 @@ static struct nft_expr_type nft_redir_ipv4_type __read_mostly = { }; #ifdef CONFIG_NF_TABLES_IPV6 -static void nft_redir_ipv6_eval(const struct nft_expr *expr, - struct nft_regs *regs, - const struct nft_pktinfo *pkt) -{ - struct nft_redir *priv = nft_expr_priv(expr); - struct nf_nat_range2 range; - - memset(&range, 0, sizeof(range)); - if (priv->sreg_proto_min) { - range.min_proto.all = (__force __be16)nft_reg_load16( - ®s->data[priv->sreg_proto_min]); - range.max_proto.all = (__force __be16)nft_reg_load16( - ®s->data[priv->sreg_proto_max]); - range.flags |= NF_NAT_RANGE_PROTO_SPECIFIED; - } - - range.flags |= priv->flags; - - regs->verdict.code = - nf_nat_redirect_ipv6(pkt->skb, &range, nft_hook(pkt)); -} - static void nft_redir_ipv6_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr) { @@ -180,7 +172,7 @@ static struct nft_expr_type nft_redir_ipv6_type; static const struct nft_expr_ops nft_redir_ipv6_ops = { .type = &nft_redir_ipv6_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_redir)), - .eval = nft_redir_ipv6_eval, + .eval = nft_redir_eval, .init = nft_redir_init, .destroy = nft_redir_ipv6_destroy, .dump = nft_redir_dump, @@ -199,20 +191,6 @@ static struct nft_expr_type nft_redir_ipv6_type __read_mostly = { #endif #ifdef CONFIG_NF_TABLES_INET -static void nft_redir_inet_eval(const struct nft_expr *expr, - struct nft_regs *regs, - const struct nft_pktinfo *pkt) -{ - switch (nft_pf(pkt)) { - case NFPROTO_IPV4: - return nft_redir_ipv4_eval(expr, regs, pkt); - case NFPROTO_IPV6: - return nft_redir_ipv6_eval(expr, regs, pkt); - } - - WARN_ON_ONCE(1); -} - static void nft_redir_inet_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr) { @@ -223,7 +201,7 @@ static struct nft_expr_type nft_redir_inet_type; static const struct nft_expr_ops nft_redir_inet_ops = { .type = &nft_redir_inet_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_redir)), - .eval = nft_redir_inet_eval, + .eval = nft_redir_eval, .init = nft_redir_init, .destroy = nft_redir_inet_destroy, .dump = nft_redir_dump, diff --git a/net/netfilter/xt_REDIRECT.c b/net/netfilter/xt_REDIRECT.c index 353ca7801251..ff66b56a3f97 100644 --- a/net/netfilter/xt_REDIRECT.c +++ b/net/netfilter/xt_REDIRECT.c @@ -46,7 +46,6 @@ static void redirect_tg_destroy(const struct xt_tgdtor_param *par) nf_ct_netns_put(par->net, par->family); } -/* FIXME: Take multiple ranges --RR */ static int redirect_tg4_check(const struct xt_tgchk_param *par) { const struct nf_nat_ipv4_multi_range_compat *mr = par->targinfo; @@ -65,7 +64,14 @@ static int redirect_tg4_check(const struct xt_tgchk_param *par) static unsigned int redirect_tg4(struct sk_buff *skb, const struct xt_action_param *par) { - return nf_nat_redirect_ipv4(skb, par->targinfo, xt_hooknum(par)); + const struct nf_nat_ipv4_multi_range_compat *mr = par->targinfo; + struct nf_nat_range2 range = { + .flags = mr->range[0].flags, + .min_proto = mr->range[0].min, + .max_proto = mr->range[0].max, + }; + + return nf_nat_redirect_ipv4(skb, &range, xt_hooknum(par)); } static struct xt_target redirect_tg_reg[] __read_mostly = { -- cgit v1.2.3 From f6ca5d5ed7ecee1ff4721565fb5a276884d1e785 Mon Sep 17 00:00:00 2001 From: Jeremy Sowden Date: Wed, 15 Mar 2023 21:48:02 +0000 Subject: netfilter: nft_masq: deduplicate eval call-backs nft_masq has separate ipv4 and ipv6 call-backs which share much of their code, and an inet one switch containing a switch that calls one of the others based on the family of the packet. Merge the ipv4 and ipv6 ones into the inet one in order to get rid of the duplicate code. Const-qualify the `priv` pointer since we don't need to write through it. Signed-off-by: Jeremy Sowden Signed-off-by: Florian Westphal --- net/netfilter/nft_masq.c | 75 +++++++++++++++++++----------------------------- 1 file changed, 29 insertions(+), 46 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_masq.c b/net/netfilter/nft_masq.c index 9544c2f16998..b115d77fbbc7 100644 --- a/net/netfilter/nft_masq.c +++ b/net/netfilter/nft_masq.c @@ -96,23 +96,39 @@ nla_put_failure: return -1; } -static void nft_masq_ipv4_eval(const struct nft_expr *expr, - struct nft_regs *regs, - const struct nft_pktinfo *pkt) +static void nft_masq_eval(const struct nft_expr *expr, + struct nft_regs *regs, + const struct nft_pktinfo *pkt) { - struct nft_masq *priv = nft_expr_priv(expr); + const struct nft_masq *priv = nft_expr_priv(expr); struct nf_nat_range2 range; memset(&range, 0, sizeof(range)); range.flags = priv->flags; if (priv->sreg_proto_min) { - range.min_proto.all = (__force __be16)nft_reg_load16( - ®s->data[priv->sreg_proto_min]); - range.max_proto.all = (__force __be16)nft_reg_load16( - ®s->data[priv->sreg_proto_max]); + range.min_proto.all = (__force __be16) + nft_reg_load16(®s->data[priv->sreg_proto_min]); + range.max_proto.all = (__force __be16) + nft_reg_load16(®s->data[priv->sreg_proto_max]); + } + + switch (nft_pf(pkt)) { + case NFPROTO_IPV4: + regs->verdict.code = nf_nat_masquerade_ipv4(pkt->skb, + nft_hook(pkt), + &range, + nft_out(pkt)); + break; +#ifdef CONFIG_NF_TABLES_IPV6 + case NFPROTO_IPV6: + regs->verdict.code = nf_nat_masquerade_ipv6(pkt->skb, &range, + nft_out(pkt)); + break; +#endif + default: + WARN_ON_ONCE(1); + break; } - regs->verdict.code = nf_nat_masquerade_ipv4(pkt->skb, nft_hook(pkt), - &range, nft_out(pkt)); } static void @@ -125,7 +141,7 @@ static struct nft_expr_type nft_masq_ipv4_type; static const struct nft_expr_ops nft_masq_ipv4_ops = { .type = &nft_masq_ipv4_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_masq)), - .eval = nft_masq_ipv4_eval, + .eval = nft_masq_eval, .init = nft_masq_init, .destroy = nft_masq_ipv4_destroy, .dump = nft_masq_dump, @@ -143,25 +159,6 @@ static struct nft_expr_type nft_masq_ipv4_type __read_mostly = { }; #ifdef CONFIG_NF_TABLES_IPV6 -static void nft_masq_ipv6_eval(const struct nft_expr *expr, - struct nft_regs *regs, - const struct nft_pktinfo *pkt) -{ - struct nft_masq *priv = nft_expr_priv(expr); - struct nf_nat_range2 range; - - memset(&range, 0, sizeof(range)); - range.flags = priv->flags; - if (priv->sreg_proto_min) { - range.min_proto.all = (__force __be16)nft_reg_load16( - ®s->data[priv->sreg_proto_min]); - range.max_proto.all = (__force __be16)nft_reg_load16( - ®s->data[priv->sreg_proto_max]); - } - regs->verdict.code = nf_nat_masquerade_ipv6(pkt->skb, &range, - nft_out(pkt)); -} - static void nft_masq_ipv6_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr) { @@ -172,7 +169,7 @@ static struct nft_expr_type nft_masq_ipv6_type; static const struct nft_expr_ops nft_masq_ipv6_ops = { .type = &nft_masq_ipv6_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_masq)), - .eval = nft_masq_ipv6_eval, + .eval = nft_masq_eval, .init = nft_masq_init, .destroy = nft_masq_ipv6_destroy, .dump = nft_masq_dump, @@ -204,20 +201,6 @@ static inline void nft_masq_module_exit_ipv6(void) {} #endif #ifdef CONFIG_NF_TABLES_INET -static void nft_masq_inet_eval(const struct nft_expr *expr, - struct nft_regs *regs, - const struct nft_pktinfo *pkt) -{ - switch (nft_pf(pkt)) { - case NFPROTO_IPV4: - return nft_masq_ipv4_eval(expr, regs, pkt); - case NFPROTO_IPV6: - return nft_masq_ipv6_eval(expr, regs, pkt); - } - - WARN_ON_ONCE(1); -} - static void nft_masq_inet_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr) { @@ -228,7 +211,7 @@ static struct nft_expr_type nft_masq_inet_type; static const struct nft_expr_ops nft_masq_inet_ops = { .type = &nft_masq_inet_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_masq)), - .eval = nft_masq_inet_eval, + .eval = nft_masq_eval, .init = nft_masq_init, .destroy = nft_masq_inet_destroy, .dump = nft_masq_dump, -- cgit v1.2.3 From bde7170a04d69ab8013d0582e60e414d46d46dd6 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 16 Mar 2023 13:59:48 +0100 Subject: netfilter: xtables: disable 32bit compat interface by default This defaulted to 'y' because before this knob existed the 32bit compat layer was always compiled in if CONFIG_COMPAT was set. 32bit iptables on 64bit kernel isn't common anymore, so remove the default-y now. Signed-off-by: Florian Westphal --- net/netfilter/Kconfig | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 4d6737160857..d0bf630482c1 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -753,7 +753,6 @@ if NETFILTER_XTABLES config NETFILTER_XTABLES_COMPAT bool "Netfilter Xtables 32bit support" depends on COMPAT - default y help This option provides a translation layer to run 32bit arp,ip(6),ebtables binaries on 64bit kernels. -- cgit v1.2.3 From 36ce9982ef2fb63fdf39996900866965d71f5a5e Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 16 Mar 2023 15:44:08 +0100 Subject: xtables: move icmp/icmpv6 logic to xt_tcpudp icmp/icmp6 matches are baked into ip(6)_tables.ko. This means that even if iptables-nft is used, a rule like "-p icmp --icmp-type 1" will load the ip(6)tables modules. Move them to xt_tcpdudp.ko instead to avoid this. This will also allow to eventually add kconfig knobs to build kernels that support iptables-nft but not iptables-legacy (old set/getsockopt interface). Signed-off-by: Florian Westphal --- net/ipv4/netfilter/ip_tables.c | 68 +------------------------ net/ipv6/netfilter/ip6_tables.c | 68 +------------------------ net/netfilter/xt_tcpudp.c | 110 ++++++++++++++++++++++++++++++++++++++++ 3 files changed, 112 insertions(+), 134 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index da5998011ab9..7da1df4997d0 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -14,7 +14,6 @@ #include #include #include -#include #include #include #include @@ -31,7 +30,6 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team "); MODULE_DESCRIPTION("IPv4 packet filter"); -MODULE_ALIAS("ipt_icmp"); void *ipt_alloc_initial_table(const struct xt_table *info) { @@ -1799,52 +1797,6 @@ void ipt_unregister_table_exit(struct net *net, const char *name) __ipt_unregister_table(net, table); } -/* Returns 1 if the type and code is matched by the range, 0 otherwise */ -static inline bool -icmp_type_code_match(u_int8_t test_type, u_int8_t min_code, u_int8_t max_code, - u_int8_t type, u_int8_t code, - bool invert) -{ - return ((test_type == 0xFF) || - (type == test_type && code >= min_code && code <= max_code)) - ^ invert; -} - -static bool -icmp_match(const struct sk_buff *skb, struct xt_action_param *par) -{ - const struct icmphdr *ic; - struct icmphdr _icmph; - const struct ipt_icmp *icmpinfo = par->matchinfo; - - /* Must not be a fragment. */ - if (par->fragoff != 0) - return false; - - ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph); - if (ic == NULL) { - /* We've been asked to examine this packet, and we - * can't. Hence, no choice but to drop. - */ - par->hotdrop = true; - return false; - } - - return icmp_type_code_match(icmpinfo->type, - icmpinfo->code[0], - icmpinfo->code[1], - ic->type, ic->code, - !!(icmpinfo->invflags&IPT_ICMP_INV)); -} - -static int icmp_checkentry(const struct xt_mtchk_param *par) -{ - const struct ipt_icmp *icmpinfo = par->matchinfo; - - /* Must specify no unknown invflags */ - return (icmpinfo->invflags & ~IPT_ICMP_INV) ? -EINVAL : 0; -} - static struct xt_target ipt_builtin_tg[] __read_mostly = { { .name = XT_STANDARD_TARGET, @@ -1875,18 +1827,6 @@ static struct nf_sockopt_ops ipt_sockopts = { .owner = THIS_MODULE, }; -static struct xt_match ipt_builtin_mt[] __read_mostly = { - { - .name = "icmp", - .match = icmp_match, - .matchsize = sizeof(struct ipt_icmp), - .checkentry = icmp_checkentry, - .proto = IPPROTO_ICMP, - .family = NFPROTO_IPV4, - .me = THIS_MODULE, - }, -}; - static int __net_init ip_tables_net_init(struct net *net) { return xt_proto_init(net, NFPROTO_IPV4); @@ -1914,19 +1854,14 @@ static int __init ip_tables_init(void) ret = xt_register_targets(ipt_builtin_tg, ARRAY_SIZE(ipt_builtin_tg)); if (ret < 0) goto err2; - ret = xt_register_matches(ipt_builtin_mt, ARRAY_SIZE(ipt_builtin_mt)); - if (ret < 0) - goto err4; /* Register setsockopt */ ret = nf_register_sockopt(&ipt_sockopts); if (ret < 0) - goto err5; + goto err4; return 0; -err5: - xt_unregister_matches(ipt_builtin_mt, ARRAY_SIZE(ipt_builtin_mt)); err4: xt_unregister_targets(ipt_builtin_tg, ARRAY_SIZE(ipt_builtin_tg)); err2: @@ -1939,7 +1874,6 @@ static void __exit ip_tables_fini(void) { nf_unregister_sockopt(&ipt_sockopts); - xt_unregister_matches(ipt_builtin_mt, ARRAY_SIZE(ipt_builtin_mt)); xt_unregister_targets(ipt_builtin_tg, ARRAY_SIZE(ipt_builtin_tg)); unregister_pernet_subsys(&ip_tables_net_ops); } diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 0ce0ed17c758..fd9f049d6d41 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -18,7 +18,6 @@ #include #include #include -#include #include #include #include @@ -35,7 +34,6 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team "); MODULE_DESCRIPTION("IPv6 packet filter"); -MODULE_ALIAS("ip6t_icmp6"); void *ip6t_alloc_initial_table(const struct xt_table *info) { @@ -1805,52 +1803,6 @@ void ip6t_unregister_table_exit(struct net *net, const char *name) __ip6t_unregister_table(net, table); } -/* Returns 1 if the type and code is matched by the range, 0 otherwise */ -static inline bool -icmp6_type_code_match(u_int8_t test_type, u_int8_t min_code, u_int8_t max_code, - u_int8_t type, u_int8_t code, - bool invert) -{ - return (type == test_type && code >= min_code && code <= max_code) - ^ invert; -} - -static bool -icmp6_match(const struct sk_buff *skb, struct xt_action_param *par) -{ - const struct icmp6hdr *ic; - struct icmp6hdr _icmph; - const struct ip6t_icmp *icmpinfo = par->matchinfo; - - /* Must not be a fragment. */ - if (par->fragoff != 0) - return false; - - ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph); - if (ic == NULL) { - /* We've been asked to examine this packet, and we - * can't. Hence, no choice but to drop. - */ - par->hotdrop = true; - return false; - } - - return icmp6_type_code_match(icmpinfo->type, - icmpinfo->code[0], - icmpinfo->code[1], - ic->icmp6_type, ic->icmp6_code, - !!(icmpinfo->invflags&IP6T_ICMP_INV)); -} - -/* Called when user tries to insert an entry of this type. */ -static int icmp6_checkentry(const struct xt_mtchk_param *par) -{ - const struct ip6t_icmp *icmpinfo = par->matchinfo; - - /* Must specify no unknown invflags */ - return (icmpinfo->invflags & ~IP6T_ICMP_INV) ? -EINVAL : 0; -} - /* The built-in targets: standard (NULL) and error. */ static struct xt_target ip6t_builtin_tg[] __read_mostly = { { @@ -1882,18 +1834,6 @@ static struct nf_sockopt_ops ip6t_sockopts = { .owner = THIS_MODULE, }; -static struct xt_match ip6t_builtin_mt[] __read_mostly = { - { - .name = "icmp6", - .match = icmp6_match, - .matchsize = sizeof(struct ip6t_icmp), - .checkentry = icmp6_checkentry, - .proto = IPPROTO_ICMPV6, - .family = NFPROTO_IPV6, - .me = THIS_MODULE, - }, -}; - static int __net_init ip6_tables_net_init(struct net *net) { return xt_proto_init(net, NFPROTO_IPV6); @@ -1921,19 +1861,14 @@ static int __init ip6_tables_init(void) ret = xt_register_targets(ip6t_builtin_tg, ARRAY_SIZE(ip6t_builtin_tg)); if (ret < 0) goto err2; - ret = xt_register_matches(ip6t_builtin_mt, ARRAY_SIZE(ip6t_builtin_mt)); - if (ret < 0) - goto err4; /* Register setsockopt */ ret = nf_register_sockopt(&ip6t_sockopts); if (ret < 0) - goto err5; + goto err4; return 0; -err5: - xt_unregister_matches(ip6t_builtin_mt, ARRAY_SIZE(ip6t_builtin_mt)); err4: xt_unregister_targets(ip6t_builtin_tg, ARRAY_SIZE(ip6t_builtin_tg)); err2: @@ -1946,7 +1881,6 @@ static void __exit ip6_tables_fini(void) { nf_unregister_sockopt(&ip6t_sockopts); - xt_unregister_matches(ip6t_builtin_mt, ARRAY_SIZE(ip6t_builtin_mt)); xt_unregister_targets(ip6t_builtin_tg, ARRAY_SIZE(ip6t_builtin_tg)); unregister_pernet_subsys(&ip6_tables_net_ops); } diff --git a/net/netfilter/xt_tcpudp.c b/net/netfilter/xt_tcpudp.c index 11ec2abf0c72..e8991130a3de 100644 --- a/net/netfilter/xt_tcpudp.c +++ b/net/netfilter/xt_tcpudp.c @@ -4,6 +4,7 @@ #include #include #include +#include #include #include #include @@ -20,6 +21,8 @@ MODULE_ALIAS("ipt_udp"); MODULE_ALIAS("ipt_tcp"); MODULE_ALIAS("ip6t_udp"); MODULE_ALIAS("ip6t_tcp"); +MODULE_ALIAS("ipt_icmp"); +MODULE_ALIAS("ip6t_icmp6"); /* Returns 1 if the port is matched by the range, 0 otherwise */ static inline bool @@ -161,6 +164,95 @@ static int udp_mt_check(const struct xt_mtchk_param *par) return (udpinfo->invflags & ~XT_UDP_INV_MASK) ? -EINVAL : 0; } +/* Returns 1 if the type and code is matched by the range, 0 otherwise */ +static bool type_code_in_range(u8 test_type, u8 min_code, u8 max_code, + u8 type, u8 code) +{ + return type == test_type && code >= min_code && code <= max_code; +} + +static bool icmp_type_code_match(u8 test_type, u8 min_code, u8 max_code, + u8 type, u8 code, bool invert) +{ + return (test_type == 0xFF || + type_code_in_range(test_type, min_code, max_code, type, code)) + ^ invert; +} + +static bool icmp6_type_code_match(u8 test_type, u8 min_code, u8 max_code, + u8 type, u8 code, bool invert) +{ + return type_code_in_range(test_type, min_code, max_code, type, code) ^ invert; +} + +static bool +icmp_match(const struct sk_buff *skb, struct xt_action_param *par) +{ + const struct icmphdr *ic; + struct icmphdr _icmph; + const struct ipt_icmp *icmpinfo = par->matchinfo; + + /* Must not be a fragment. */ + if (par->fragoff != 0) + return false; + + ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph); + if (!ic) { + /* We've been asked to examine this packet, and we + * can't. Hence, no choice but to drop. + */ + par->hotdrop = true; + return false; + } + + return icmp_type_code_match(icmpinfo->type, + icmpinfo->code[0], + icmpinfo->code[1], + ic->type, ic->code, + !!(icmpinfo->invflags & IPT_ICMP_INV)); +} + +static bool +icmp6_match(const struct sk_buff *skb, struct xt_action_param *par) +{ + const struct icmp6hdr *ic; + struct icmp6hdr _icmph; + const struct ip6t_icmp *icmpinfo = par->matchinfo; + + /* Must not be a fragment. */ + if (par->fragoff != 0) + return false; + + ic = skb_header_pointer(skb, par->thoff, sizeof(_icmph), &_icmph); + if (!ic) { + /* We've been asked to examine this packet, and we + * can't. Hence, no choice but to drop. + */ + par->hotdrop = true; + return false; + } + + return icmp6_type_code_match(icmpinfo->type, + icmpinfo->code[0], + icmpinfo->code[1], + ic->icmp6_type, ic->icmp6_code, + !!(icmpinfo->invflags & IP6T_ICMP_INV)); +} + +static int icmp_checkentry(const struct xt_mtchk_param *par) +{ + const struct ipt_icmp *icmpinfo = par->matchinfo; + + return (icmpinfo->invflags & ~IPT_ICMP_INV) ? -EINVAL : 0; +} + +static int icmp6_checkentry(const struct xt_mtchk_param *par) +{ + const struct ip6t_icmp *icmpinfo = par->matchinfo; + + return (icmpinfo->invflags & ~IP6T_ICMP_INV) ? -EINVAL : 0; +} + static struct xt_match tcpudp_mt_reg[] __read_mostly = { { .name = "tcp", @@ -216,6 +308,24 @@ static struct xt_match tcpudp_mt_reg[] __read_mostly = { .proto = IPPROTO_UDPLITE, .me = THIS_MODULE, }, + { + .name = "icmp", + .match = icmp_match, + .matchsize = sizeof(struct ipt_icmp), + .checkentry = icmp_checkentry, + .proto = IPPROTO_ICMP, + .family = NFPROTO_IPV4, + .me = THIS_MODULE, + }, + { + .name = "icmp6", + .match = icmp6_match, + .matchsize = sizeof(struct ip6t_icmp), + .checkentry = icmp6_checkentry, + .proto = IPPROTO_ICMPV6, + .family = NFPROTO_IPV6, + .me = THIS_MODULE, + }, }; static int __init tcpudp_mt_init(void) -- cgit v1.2.3 From b0e214d212030fe497d4d150bb3474e50ad5d093 Mon Sep 17 00:00:00 2001 From: Madhu Koriginja Date: Tue, 21 Mar 2023 21:28:44 +0530 Subject: netfilter: keep conntrack reference until IPsecv6 policy checks are done Keep the conntrack reference until policy checks have been performed for IPsec V6 NAT support, just like ipv4. The reference needs to be dropped before a packet is queued to avoid having the conntrack module unloadable. Fixes: 58a317f1061c ("netfilter: ipv6: add IPv6 NAT support") Signed-off-by: Madhu Koriginja Signed-off-by: Florian Westphal --- net/dccp/ipv6.c | 1 + net/ipv6/ip6_input.c | 14 ++++++-------- net/ipv6/raw.c | 5 ++--- net/ipv6/tcp_ipv6.c | 2 ++ net/ipv6/udp.c | 2 ++ 5 files changed, 13 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 47fb10834223..93c98990d726 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -784,6 +784,7 @@ lookup: if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) goto discard_and_relse; + nf_reset_ct(skb); return __sk_receive_skb(sk, skb, 1, dh->dccph_doff * 4, refcounted) ? -1 : 0; diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index e1ebf5e42ebe..d94041bb4287 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -404,10 +404,6 @@ resubmit_final: /* Only do this once for first final protocol */ have_final = true; - /* Free reference early: we don't need it any more, - and it may hold ip_conntrack module loaded - indefinitely. */ - nf_reset_ct(skb); skb_postpull_rcsum(skb, skb_network_header(skb), skb_network_header_len(skb)); @@ -430,10 +426,12 @@ resubmit_final: goto discard; } } - if (!(ipprot->flags & INET6_PROTO_NOPOLICY) && - !xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) { - SKB_DR_SET(reason, XFRM_POLICY); - goto discard; + if (!(ipprot->flags & INET6_PROTO_NOPOLICY)) { + if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) { + SKB_DR_SET(reason, XFRM_POLICY); + goto discard; + } + nf_reset_ct(skb); } ret = INDIRECT_CALL_2(ipprot->handler, tcp_v6_rcv, udpv6_rcv, diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 6ac2f2690c44..4ab62a9c5c8e 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -194,10 +194,8 @@ static bool ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC); /* Not releasing hash table! */ - if (clone) { - nf_reset_ct(clone); + if (clone) rawv6_rcv(sk, clone); - } } } rcu_read_unlock(); @@ -391,6 +389,7 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb) kfree_skb_reason(skb, SKB_DROP_REASON_XFRM_POLICY); return NET_RX_DROP; } + nf_reset_ct(skb); if (!rp->checksum) skb->ip_summed = CHECKSUM_UNNECESSARY; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 35cf523c9efd..244cf86c4cbb 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1723,6 +1723,8 @@ process: if (drop_reason) goto discard_and_relse; + nf_reset_ct(skb); + if (tcp_filter(sk, skb)) { drop_reason = SKB_DROP_REASON_SOCKET_FILTER; goto discard_and_relse; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index d350e57c4792..4caa70a1b871 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -704,6 +704,7 @@ static int udpv6_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb) drop_reason = SKB_DROP_REASON_XFRM_POLICY; goto drop; } + nf_reset_ct(skb); if (static_branch_unlikely(&udpv6_encap_needed_key) && up->encap_type) { int (*encap_rcv)(struct sock *sk, struct sk_buff *skb); @@ -1027,6 +1028,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) goto discard; + nf_reset_ct(skb); if (udp_lib_checksum_complete(skb)) goto csum_error; -- cgit v1.2.3 From 5f24a8725fe7bc2c6adf7ce00dd3e818387d8995 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Wed, 22 Mar 2023 15:01:34 -0400 Subject: SUNRPC: Fix a crash in gss_krb5_checksum() Anna says: > KASAN reports [...] a slab-out-of-bounds in gss_krb5_checksum(), > and it can cause my client to panic when running cthon basic > tests with krb5p. > Running faddr2line gives me: > > gss_krb5_checksum+0x4b6/0x630: > ahash_request_free at > /home/anna/Programs/linux-nfs.git/./include/crypto/hash.h:619 > (inlined by) gss_krb5_checksum at > /home/anna/Programs/linux-nfs.git/net/sunrpc/auth_gss/gss_krb5_crypto.c:358 My diagnosis is that the memcpy() at the end of gss_krb5_checksum() reads past the end of the buffer containing the checksum data because the callers have ignored gss_krb5_checksum()'s API contract: * Caller provides the truncation length of the output token (h) in * cksumout.len. Instead they provide the fixed length of the hmac buffer. This length happens to be larger than the value returned by crypto_ahash_digestsize(). Change these errant callers to work like krb5_etm_{en,de}crypt(). As a defensive measure, bound the length of the byte copy at the end of gss_krb5_checksum(). Kunit sez: Testing complete. Ran 68 tests: passed: 68 Elapsed time: 81.680s total, 5.875s configuring, 75.610s building, 0.103s running Reported-by: Anna Schumaker Fixes: 8270dbfcebea ("SUNRPC: Obscure Kerberos integrity keys") Signed-off-by: Chuck Lever --- net/sunrpc/auth_gss/gss_krb5_crypto.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c index 6c7c52eeed4f..212c5d57465a 100644 --- a/net/sunrpc/auth_gss/gss_krb5_crypto.c +++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c @@ -353,7 +353,9 @@ gss_krb5_checksum(struct crypto_ahash *tfm, char *header, int hdrlen, err = crypto_ahash_final(req); if (err) goto out_free_ahash; - memcpy(cksumout->data, checksumdata, cksumout->len); + + memcpy(cksumout->data, checksumdata, + min_t(int, cksumout->len, crypto_ahash_digestsize(tfm))); out_free_ahash: ahash_request_free(req); @@ -809,8 +811,7 @@ gss_krb5_aes_encrypt(struct krb5_ctx *kctx, u32 offset, buf->tail[0].iov_len += GSS_KRB5_TOK_HDR_LEN; buf->len += GSS_KRB5_TOK_HDR_LEN; - /* Do the HMAC */ - hmac.len = GSS_KRB5_MAX_CKSUM_LEN; + hmac.len = kctx->gk5e->cksumlength; hmac.data = buf->tail[0].iov_base + buf->tail[0].iov_len; /* @@ -873,8 +874,7 @@ gss_krb5_aes_decrypt(struct krb5_ctx *kctx, u32 offset, u32 len, if (ret) goto out_err; - /* Calculate our hmac over the plaintext data */ - our_hmac_obj.len = sizeof(our_hmac); + our_hmac_obj.len = kctx->gk5e->cksumlength; our_hmac_obj.data = our_hmac; ret = gss_krb5_checksum(ahash, NULL, 0, &subbuf, 0, &our_hmac_obj); if (ret) -- cgit v1.2.3 From d7ba4cc900bf1eea2d8c807c6b1fc6bd61f41237 Mon Sep 17 00:00:00 2001 From: JP Kobryn Date: Wed, 22 Mar 2023 12:47:54 -0700 Subject: bpf: return long from bpf_map_ops funcs This patch changes the return types of bpf_map_ops functions to long, where previously int was returned. Using long allows for bpf programs to maintain the sign bit in the absence of sign extension during situations where inlined bpf helper funcs make calls to the bpf_map_ops funcs and a negative error is returned. The definitions of the helper funcs are generated from comments in the bpf uapi header at `include/uapi/linux/bpf.h`. The return type of these helpers was previously changed from int to long in commit bdb7b79b4ce8. For any case where one of the map helpers call the bpf_map_ops funcs that are still returning 32-bit int, a compiler might not include sign extension instructions to properly convert the 32-bit negative value a 64-bit negative value. For example: bpf assembly excerpt of an inlined helper calling a kernel function and checking for a specific error: ; err = bpf_map_update_elem(&mymap, &key, &val, BPF_NOEXIST); ... 46: call 0xffffffffe103291c ; htab_map_update_elem ; if (err && err != -EEXIST) { 4b: cmp $0xffffffffffffffef,%rax ; cmp -EEXIST,%rax kernel function assembly excerpt of return value from `htab_map_update_elem` returning 32-bit int: movl $0xffffffef, %r9d ... movl %r9d, %eax ...results in the comparison: cmp $0xffffffffffffffef, $0x00000000ffffffef Fixes: bdb7b79b4ce8 ("bpf: Switch most helper return values from 32-bit int to 64-bit long") Tested-by: Eduard Zingerman Signed-off-by: JP Kobryn Link: https://lore.kernel.org/r/20230322194754.185781-3-inwardvessel@gmail.com Signed-off-by: Alexei Starovoitov --- include/linux/bpf.h | 14 +++++++------- include/linux/filter.h | 6 +++--- kernel/bpf/arraymap.c | 12 ++++++------ kernel/bpf/bloom_filter.c | 12 ++++++------ kernel/bpf/bpf_cgrp_storage.c | 6 +++--- kernel/bpf/bpf_inode_storage.c | 6 +++--- kernel/bpf/bpf_struct_ops.c | 6 +++--- kernel/bpf/bpf_task_storage.c | 6 +++--- kernel/bpf/cpumap.c | 8 ++++---- kernel/bpf/devmap.c | 24 ++++++++++++------------ kernel/bpf/hashtab.c | 36 ++++++++++++++++++------------------ kernel/bpf/local_storage.c | 6 +++--- kernel/bpf/lpm_trie.c | 6 +++--- kernel/bpf/queue_stack_maps.c | 22 +++++++++++----------- kernel/bpf/reuseport_array.c | 2 +- kernel/bpf/ringbuf.c | 6 +++--- kernel/bpf/stackmap.c | 6 +++--- kernel/bpf/verifier.c | 14 +++++++------- net/core/bpf_sk_storage.c | 6 +++--- net/core/sock_map.c | 8 ++++---- net/xdp/xskmap.c | 8 ++++---- 21 files changed, 110 insertions(+), 110 deletions(-) (limited to 'net') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 3ef98fb92987..ec0df059f562 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -96,11 +96,11 @@ struct bpf_map_ops { /* funcs callable from userspace and from eBPF programs */ void *(*map_lookup_elem)(struct bpf_map *map, void *key); - int (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags); - int (*map_delete_elem)(struct bpf_map *map, void *key); - int (*map_push_elem)(struct bpf_map *map, void *value, u64 flags); - int (*map_pop_elem)(struct bpf_map *map, void *value); - int (*map_peek_elem)(struct bpf_map *map, void *value); + long (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags); + long (*map_delete_elem)(struct bpf_map *map, void *key); + long (*map_push_elem)(struct bpf_map *map, void *value, u64 flags); + long (*map_pop_elem)(struct bpf_map *map, void *value); + long (*map_peek_elem)(struct bpf_map *map, void *value); void *(*map_lookup_percpu_elem)(struct bpf_map *map, void *key, u32 cpu); /* funcs called by prog_array and perf_event_array map */ @@ -139,7 +139,7 @@ struct bpf_map_ops { struct bpf_local_storage __rcu ** (*map_owner_storage_ptr)(void *owner); /* Misc helpers.*/ - int (*map_redirect)(struct bpf_map *map, u64 key, u64 flags); + long (*map_redirect)(struct bpf_map *map, u64 key, u64 flags); /* map_meta_equal must be implemented for maps that can be * used as an inner map. It is a runtime check to ensure @@ -157,7 +157,7 @@ struct bpf_map_ops { int (*map_set_for_each_callback_args)(struct bpf_verifier_env *env, struct bpf_func_state *caller, struct bpf_func_state *callee); - int (*map_for_each_callback)(struct bpf_map *map, + long (*map_for_each_callback)(struct bpf_map *map, bpf_callback_t callback_fn, void *callback_ctx, u64 flags); diff --git a/include/linux/filter.h b/include/linux/filter.h index efa5d4a1677e..23c08c31bea9 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -1504,9 +1504,9 @@ static inline bool bpf_sk_lookup_run_v6(struct net *net, int protocol, } #endif /* IS_ENABLED(CONFIG_IPV6) */ -static __always_inline int __bpf_xdp_redirect_map(struct bpf_map *map, u64 index, - u64 flags, const u64 flag_mask, - void *lookup_elem(struct bpf_map *map, u32 key)) +static __always_inline long __bpf_xdp_redirect_map(struct bpf_map *map, u64 index, + u64 flags, const u64 flag_mask, + void *lookup_elem(struct bpf_map *map, u32 key)) { struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info); const u64 action_mask = XDP_ABORTED | XDP_DROP | XDP_PASS | XDP_TX; diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 1588c793a715..2058e89b5ddd 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -307,8 +307,8 @@ static int array_map_get_next_key(struct bpf_map *map, void *key, void *next_key } /* Called from syscall or from eBPF program */ -static int array_map_update_elem(struct bpf_map *map, void *key, void *value, - u64 map_flags) +static long array_map_update_elem(struct bpf_map *map, void *key, void *value, + u64 map_flags) { struct bpf_array *array = container_of(map, struct bpf_array, map); u32 index = *(u32 *)key; @@ -386,7 +386,7 @@ int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value, } /* Called from syscall or from eBPF program */ -static int array_map_delete_elem(struct bpf_map *map, void *key) +static long array_map_delete_elem(struct bpf_map *map, void *key) { return -EINVAL; } @@ -686,8 +686,8 @@ static const struct bpf_iter_seq_info iter_seq_info = { .seq_priv_size = sizeof(struct bpf_iter_seq_array_map_info), }; -static int bpf_for_each_array_elem(struct bpf_map *map, bpf_callback_t callback_fn, - void *callback_ctx, u64 flags) +static long bpf_for_each_array_elem(struct bpf_map *map, bpf_callback_t callback_fn, + void *callback_ctx, u64 flags) { u32 i, key, num_elems = 0; struct bpf_array *array; @@ -871,7 +871,7 @@ int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file, return 0; } -static int fd_array_map_delete_elem(struct bpf_map *map, void *key) +static long fd_array_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_array *array = container_of(map, struct bpf_array, map); void *old_ptr; diff --git a/kernel/bpf/bloom_filter.c b/kernel/bpf/bloom_filter.c index 6350c5d35a9b..db19784601a7 100644 --- a/kernel/bpf/bloom_filter.c +++ b/kernel/bpf/bloom_filter.c @@ -41,7 +41,7 @@ static u32 hash(struct bpf_bloom_filter *bloom, void *value, return h & bloom->bitset_mask; } -static int bloom_map_peek_elem(struct bpf_map *map, void *value) +static long bloom_map_peek_elem(struct bpf_map *map, void *value) { struct bpf_bloom_filter *bloom = container_of(map, struct bpf_bloom_filter, map); @@ -56,7 +56,7 @@ static int bloom_map_peek_elem(struct bpf_map *map, void *value) return 0; } -static int bloom_map_push_elem(struct bpf_map *map, void *value, u64 flags) +static long bloom_map_push_elem(struct bpf_map *map, void *value, u64 flags) { struct bpf_bloom_filter *bloom = container_of(map, struct bpf_bloom_filter, map); @@ -73,12 +73,12 @@ static int bloom_map_push_elem(struct bpf_map *map, void *value, u64 flags) return 0; } -static int bloom_map_pop_elem(struct bpf_map *map, void *value) +static long bloom_map_pop_elem(struct bpf_map *map, void *value) { return -EOPNOTSUPP; } -static int bloom_map_delete_elem(struct bpf_map *map, void *value) +static long bloom_map_delete_elem(struct bpf_map *map, void *value) { return -EOPNOTSUPP; } @@ -177,8 +177,8 @@ static void *bloom_map_lookup_elem(struct bpf_map *map, void *key) return ERR_PTR(-EINVAL); } -static int bloom_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 flags) +static long bloom_map_update_elem(struct bpf_map *map, void *key, + void *value, u64 flags) { /* The eBPF program should use map_push_elem instead */ return -EINVAL; diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c index c975cacdd16b..f5b016a5484d 100644 --- a/kernel/bpf/bpf_cgrp_storage.c +++ b/kernel/bpf/bpf_cgrp_storage.c @@ -93,8 +93,8 @@ static void *bpf_cgrp_storage_lookup_elem(struct bpf_map *map, void *key) return sdata ? sdata->data : NULL; } -static int bpf_cgrp_storage_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags) +static long bpf_cgrp_storage_update_elem(struct bpf_map *map, void *key, + void *value, u64 map_flags) { struct bpf_local_storage_data *sdata; struct cgroup *cgroup; @@ -125,7 +125,7 @@ static int cgroup_storage_delete(struct cgroup *cgroup, struct bpf_map *map) return 0; } -static int bpf_cgrp_storage_delete_elem(struct bpf_map *map, void *key) +static long bpf_cgrp_storage_delete_elem(struct bpf_map *map, void *key) { struct cgroup *cgroup; int err, fd; diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c index ad2ab0187e45..9a5f05151898 100644 --- a/kernel/bpf/bpf_inode_storage.c +++ b/kernel/bpf/bpf_inode_storage.c @@ -91,8 +91,8 @@ static void *bpf_fd_inode_storage_lookup_elem(struct bpf_map *map, void *key) return sdata ? sdata->data : NULL; } -static int bpf_fd_inode_storage_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags) +static long bpf_fd_inode_storage_update_elem(struct bpf_map *map, void *key, + void *value, u64 map_flags) { struct bpf_local_storage_data *sdata; struct file *f; @@ -127,7 +127,7 @@ static int inode_storage_delete(struct inode *inode, struct bpf_map *map) return 0; } -static int bpf_fd_inode_storage_delete_elem(struct bpf_map *map, void *key) +static long bpf_fd_inode_storage_delete_elem(struct bpf_map *map, void *key) { struct file *f; int fd, err; diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c index 38903fb52f98..ba7a94276e3b 100644 --- a/kernel/bpf/bpf_struct_ops.c +++ b/kernel/bpf/bpf_struct_ops.c @@ -349,8 +349,8 @@ int bpf_struct_ops_prepare_trampoline(struct bpf_tramp_links *tlinks, model, flags, tlinks, NULL); } -static int bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 flags) +static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key, + void *value, u64 flags) { struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map; const struct bpf_struct_ops *st_ops = st_map->st_ops; @@ -524,7 +524,7 @@ unlock: return err; } -static int bpf_struct_ops_map_delete_elem(struct bpf_map *map, void *key) +static long bpf_struct_ops_map_delete_elem(struct bpf_map *map, void *key) { enum bpf_struct_ops_state prev_state; struct bpf_struct_ops_map *st_map; diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c index c88cc04c17c1..ab5bd1ef58c4 100644 --- a/kernel/bpf/bpf_task_storage.c +++ b/kernel/bpf/bpf_task_storage.c @@ -120,8 +120,8 @@ out: return ERR_PTR(err); } -static int bpf_pid_task_storage_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags) +static long bpf_pid_task_storage_update_elem(struct bpf_map *map, void *key, + void *value, u64 map_flags) { struct bpf_local_storage_data *sdata; struct task_struct *task; @@ -173,7 +173,7 @@ static int task_storage_delete(struct task_struct *task, struct bpf_map *map, return 0; } -static int bpf_pid_task_storage_delete_elem(struct bpf_map *map, void *key) +static long bpf_pid_task_storage_delete_elem(struct bpf_map *map, void *key) { struct task_struct *task; unsigned int f_flags; diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c index 871809e71b4e..8ec18faa74ac 100644 --- a/kernel/bpf/cpumap.c +++ b/kernel/bpf/cpumap.c @@ -540,7 +540,7 @@ static void __cpu_map_entry_replace(struct bpf_cpu_map *cmap, } } -static int cpu_map_delete_elem(struct bpf_map *map, void *key) +static long cpu_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_cpu_map *cmap = container_of(map, struct bpf_cpu_map, map); u32 key_cpu = *(u32 *)key; @@ -553,8 +553,8 @@ static int cpu_map_delete_elem(struct bpf_map *map, void *key) return 0; } -static int cpu_map_update_elem(struct bpf_map *map, void *key, void *value, - u64 map_flags) +static long cpu_map_update_elem(struct bpf_map *map, void *key, void *value, + u64 map_flags) { struct bpf_cpu_map *cmap = container_of(map, struct bpf_cpu_map, map); struct bpf_cpumap_val cpumap_value = {}; @@ -667,7 +667,7 @@ static int cpu_map_get_next_key(struct bpf_map *map, void *key, void *next_key) return 0; } -static int cpu_map_redirect(struct bpf_map *map, u64 index, u64 flags) +static long cpu_map_redirect(struct bpf_map *map, u64 index, u64 flags) { return __bpf_xdp_redirect_map(map, index, flags, 0, __cpu_map_lookup_elem); diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c index 19b036a228f7..802692fa3905 100644 --- a/kernel/bpf/devmap.c +++ b/kernel/bpf/devmap.c @@ -809,7 +809,7 @@ static void __dev_map_entry_free(struct rcu_head *rcu) kfree(dev); } -static int dev_map_delete_elem(struct bpf_map *map, void *key) +static long dev_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map); struct bpf_dtab_netdev *old_dev; @@ -826,7 +826,7 @@ static int dev_map_delete_elem(struct bpf_map *map, void *key) return 0; } -static int dev_map_hash_delete_elem(struct bpf_map *map, void *key) +static long dev_map_hash_delete_elem(struct bpf_map *map, void *key) { struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map); struct bpf_dtab_netdev *old_dev; @@ -897,8 +897,8 @@ err_out: return ERR_PTR(-EINVAL); } -static int __dev_map_update_elem(struct net *net, struct bpf_map *map, - void *key, void *value, u64 map_flags) +static long __dev_map_update_elem(struct net *net, struct bpf_map *map, + void *key, void *value, u64 map_flags) { struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map); struct bpf_dtab_netdev *dev, *old_dev; @@ -939,15 +939,15 @@ static int __dev_map_update_elem(struct net *net, struct bpf_map *map, return 0; } -static int dev_map_update_elem(struct bpf_map *map, void *key, void *value, - u64 map_flags) +static long dev_map_update_elem(struct bpf_map *map, void *key, void *value, + u64 map_flags) { return __dev_map_update_elem(current->nsproxy->net_ns, map, key, value, map_flags); } -static int __dev_map_hash_update_elem(struct net *net, struct bpf_map *map, - void *key, void *value, u64 map_flags) +static long __dev_map_hash_update_elem(struct net *net, struct bpf_map *map, + void *key, void *value, u64 map_flags) { struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map); struct bpf_dtab_netdev *dev, *old_dev; @@ -999,21 +999,21 @@ out_err: return err; } -static int dev_map_hash_update_elem(struct bpf_map *map, void *key, void *value, - u64 map_flags) +static long dev_map_hash_update_elem(struct bpf_map *map, void *key, void *value, + u64 map_flags) { return __dev_map_hash_update_elem(current->nsproxy->net_ns, map, key, value, map_flags); } -static int dev_map_redirect(struct bpf_map *map, u64 ifindex, u64 flags) +static long dev_map_redirect(struct bpf_map *map, u64 ifindex, u64 flags) { return __bpf_xdp_redirect_map(map, ifindex, flags, BPF_F_BROADCAST | BPF_F_EXCLUDE_INGRESS, __dev_map_lookup_elem); } -static int dev_hash_map_redirect(struct bpf_map *map, u64 ifindex, u64 flags) +static long dev_hash_map_redirect(struct bpf_map *map, u64 ifindex, u64 flags) { return __bpf_xdp_redirect_map(map, ifindex, flags, BPF_F_BROADCAST | BPF_F_EXCLUDE_INGRESS, diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c index 0df4b0c10f59..96b645bba3a4 100644 --- a/kernel/bpf/hashtab.c +++ b/kernel/bpf/hashtab.c @@ -1073,8 +1073,8 @@ static int check_flags(struct bpf_htab *htab, struct htab_elem *l_old, } /* Called from syscall or from eBPF program */ -static int htab_map_update_elem(struct bpf_map *map, void *key, void *value, - u64 map_flags) +static long htab_map_update_elem(struct bpf_map *map, void *key, void *value, + u64 map_flags) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct htab_elem *l_new = NULL, *l_old; @@ -1175,8 +1175,8 @@ static void htab_lru_push_free(struct bpf_htab *htab, struct htab_elem *elem) bpf_lru_push_free(&htab->lru, &elem->lru_node); } -static int htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value, - u64 map_flags) +static long htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value, + u64 map_flags) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct htab_elem *l_new, *l_old = NULL; @@ -1242,9 +1242,9 @@ err: return ret; } -static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags, - bool onallcpus) +static long __htab_percpu_map_update_elem(struct bpf_map *map, void *key, + void *value, u64 map_flags, + bool onallcpus) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct htab_elem *l_new = NULL, *l_old; @@ -1297,9 +1297,9 @@ err: return ret; } -static int __htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags, - bool onallcpus) +static long __htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, + void *value, u64 map_flags, + bool onallcpus) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct htab_elem *l_new = NULL, *l_old; @@ -1364,21 +1364,21 @@ err: return ret; } -static int htab_percpu_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags) +static long htab_percpu_map_update_elem(struct bpf_map *map, void *key, + void *value, u64 map_flags) { return __htab_percpu_map_update_elem(map, key, value, map_flags, false); } -static int htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags) +static long htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, + void *value, u64 map_flags) { return __htab_lru_percpu_map_update_elem(map, key, value, map_flags, false); } /* Called from syscall or from eBPF program */ -static int htab_map_delete_elem(struct bpf_map *map, void *key) +static long htab_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct hlist_nulls_head *head; @@ -1414,7 +1414,7 @@ static int htab_map_delete_elem(struct bpf_map *map, void *key) return ret; } -static int htab_lru_map_delete_elem(struct bpf_map *map, void *key) +static long htab_lru_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct hlist_nulls_head *head; @@ -2134,8 +2134,8 @@ static const struct bpf_iter_seq_info iter_seq_info = { .seq_priv_size = sizeof(struct bpf_iter_seq_hash_map_info), }; -static int bpf_for_each_hash_elem(struct bpf_map *map, bpf_callback_t callback_fn, - void *callback_ctx, u64 flags) +static long bpf_for_each_hash_elem(struct bpf_map *map, bpf_callback_t callback_fn, + void *callback_ctx, u64 flags) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct hlist_nulls_head *head; diff --git a/kernel/bpf/local_storage.c b/kernel/bpf/local_storage.c index a993560f200a..4c7bbec4a9e4 100644 --- a/kernel/bpf/local_storage.c +++ b/kernel/bpf/local_storage.c @@ -141,8 +141,8 @@ static void *cgroup_storage_lookup_elem(struct bpf_map *_map, void *key) return &READ_ONCE(storage->buf)->data[0]; } -static int cgroup_storage_update_elem(struct bpf_map *map, void *key, - void *value, u64 flags) +static long cgroup_storage_update_elem(struct bpf_map *map, void *key, + void *value, u64 flags) { struct bpf_cgroup_storage *storage; struct bpf_storage_buffer *new; @@ -348,7 +348,7 @@ static void cgroup_storage_map_free(struct bpf_map *_map) bpf_map_area_free(map); } -static int cgroup_storage_delete_elem(struct bpf_map *map, void *key) +static long cgroup_storage_delete_elem(struct bpf_map *map, void *key) { return -EINVAL; } diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c index dc23f2ac9cde..e0d3ddf2037a 100644 --- a/kernel/bpf/lpm_trie.c +++ b/kernel/bpf/lpm_trie.c @@ -300,8 +300,8 @@ static struct lpm_trie_node *lpm_trie_node_alloc(const struct lpm_trie *trie, } /* Called from syscall or from eBPF program */ -static int trie_update_elem(struct bpf_map *map, - void *_key, void *value, u64 flags) +static long trie_update_elem(struct bpf_map *map, + void *_key, void *value, u64 flags) { struct lpm_trie *trie = container_of(map, struct lpm_trie, map); struct lpm_trie_node *node, *im_node = NULL, *new_node = NULL; @@ -431,7 +431,7 @@ out: } /* Called from syscall or from eBPF program */ -static int trie_delete_elem(struct bpf_map *map, void *_key) +static long trie_delete_elem(struct bpf_map *map, void *_key) { struct lpm_trie *trie = container_of(map, struct lpm_trie, map); struct bpf_lpm_trie_key *key = _key; diff --git a/kernel/bpf/queue_stack_maps.c b/kernel/bpf/queue_stack_maps.c index 63ecbbcb349d..601609164ef3 100644 --- a/kernel/bpf/queue_stack_maps.c +++ b/kernel/bpf/queue_stack_maps.c @@ -95,7 +95,7 @@ static void queue_stack_map_free(struct bpf_map *map) bpf_map_area_free(qs); } -static int __queue_map_get(struct bpf_map *map, void *value, bool delete) +static long __queue_map_get(struct bpf_map *map, void *value, bool delete) { struct bpf_queue_stack *qs = bpf_queue_stack(map); unsigned long flags; @@ -124,7 +124,7 @@ out: } -static int __stack_map_get(struct bpf_map *map, void *value, bool delete) +static long __stack_map_get(struct bpf_map *map, void *value, bool delete) { struct bpf_queue_stack *qs = bpf_queue_stack(map); unsigned long flags; @@ -156,32 +156,32 @@ out: } /* Called from syscall or from eBPF program */ -static int queue_map_peek_elem(struct bpf_map *map, void *value) +static long queue_map_peek_elem(struct bpf_map *map, void *value) { return __queue_map_get(map, value, false); } /* Called from syscall or from eBPF program */ -static int stack_map_peek_elem(struct bpf_map *map, void *value) +static long stack_map_peek_elem(struct bpf_map *map, void *value) { return __stack_map_get(map, value, false); } /* Called from syscall or from eBPF program */ -static int queue_map_pop_elem(struct bpf_map *map, void *value) +static long queue_map_pop_elem(struct bpf_map *map, void *value) { return __queue_map_get(map, value, true); } /* Called from syscall or from eBPF program */ -static int stack_map_pop_elem(struct bpf_map *map, void *value) +static long stack_map_pop_elem(struct bpf_map *map, void *value) { return __stack_map_get(map, value, true); } /* Called from syscall or from eBPF program */ -static int queue_stack_map_push_elem(struct bpf_map *map, void *value, - u64 flags) +static long queue_stack_map_push_elem(struct bpf_map *map, void *value, + u64 flags) { struct bpf_queue_stack *qs = bpf_queue_stack(map); unsigned long irq_flags; @@ -227,14 +227,14 @@ static void *queue_stack_map_lookup_elem(struct bpf_map *map, void *key) } /* Called from syscall or from eBPF program */ -static int queue_stack_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 flags) +static long queue_stack_map_update_elem(struct bpf_map *map, void *key, + void *value, u64 flags) { return -EINVAL; } /* Called from syscall or from eBPF program */ -static int queue_stack_map_delete_elem(struct bpf_map *map, void *key) +static long queue_stack_map_delete_elem(struct bpf_map *map, void *key) { return -EINVAL; } diff --git a/kernel/bpf/reuseport_array.c b/kernel/bpf/reuseport_array.c index 71cb72f5b733..cbf2d8d784b8 100644 --- a/kernel/bpf/reuseport_array.c +++ b/kernel/bpf/reuseport_array.c @@ -59,7 +59,7 @@ static void *reuseport_array_lookup_elem(struct bpf_map *map, void *key) } /* Called from syscall only */ -static int reuseport_array_delete_elem(struct bpf_map *map, void *key) +static long reuseport_array_delete_elem(struct bpf_map *map, void *key) { struct reuseport_array *array = reuseport_array(map); u32 index = *(u32 *)key; diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c index 0d2a45ff83f1..875ac9b698d9 100644 --- a/kernel/bpf/ringbuf.c +++ b/kernel/bpf/ringbuf.c @@ -242,13 +242,13 @@ static void *ringbuf_map_lookup_elem(struct bpf_map *map, void *key) return ERR_PTR(-ENOTSUPP); } -static int ringbuf_map_update_elem(struct bpf_map *map, void *key, void *value, - u64 flags) +static long ringbuf_map_update_elem(struct bpf_map *map, void *key, void *value, + u64 flags) { return -ENOTSUPP; } -static int ringbuf_map_delete_elem(struct bpf_map *map, void *key) +static long ringbuf_map_delete_elem(struct bpf_map *map, void *key) { return -ENOTSUPP; } diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c index 0f1d8dced933..b25fce425b2c 100644 --- a/kernel/bpf/stackmap.c +++ b/kernel/bpf/stackmap.c @@ -618,14 +618,14 @@ static int stack_map_get_next_key(struct bpf_map *map, void *key, return 0; } -static int stack_map_update_elem(struct bpf_map *map, void *key, void *value, - u64 map_flags) +static long stack_map_update_elem(struct bpf_map *map, void *key, void *value, + u64 map_flags) { return -EINVAL; } /* Called from syscall or from eBPF program */ -static int stack_map_delete_elem(struct bpf_map *map, void *key) +static long stack_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_stack_map *smap = container_of(map, struct bpf_stack_map, map); struct stack_map_bucket *old_bucket; diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 5693e4a92752..50c995697f0e 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -17692,21 +17692,21 @@ static int do_misc_fixups(struct bpf_verifier_env *env) BUILD_BUG_ON(!__same_type(ops->map_lookup_elem, (void *(*)(struct bpf_map *map, void *key))NULL)); BUILD_BUG_ON(!__same_type(ops->map_delete_elem, - (int (*)(struct bpf_map *map, void *key))NULL)); + (long (*)(struct bpf_map *map, void *key))NULL)); BUILD_BUG_ON(!__same_type(ops->map_update_elem, - (int (*)(struct bpf_map *map, void *key, void *value, + (long (*)(struct bpf_map *map, void *key, void *value, u64 flags))NULL)); BUILD_BUG_ON(!__same_type(ops->map_push_elem, - (int (*)(struct bpf_map *map, void *value, + (long (*)(struct bpf_map *map, void *value, u64 flags))NULL)); BUILD_BUG_ON(!__same_type(ops->map_pop_elem, - (int (*)(struct bpf_map *map, void *value))NULL)); + (long (*)(struct bpf_map *map, void *value))NULL)); BUILD_BUG_ON(!__same_type(ops->map_peek_elem, - (int (*)(struct bpf_map *map, void *value))NULL)); + (long (*)(struct bpf_map *map, void *value))NULL)); BUILD_BUG_ON(!__same_type(ops->map_redirect, - (int (*)(struct bpf_map *map, u64 index, u64 flags))NULL)); + (long (*)(struct bpf_map *map, u64 index, u64 flags))NULL)); BUILD_BUG_ON(!__same_type(ops->map_for_each_callback, - (int (*)(struct bpf_map *map, + (long (*)(struct bpf_map *map, bpf_callback_t callback_fn, void *callback_ctx, u64 flags))NULL)); diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index 24c3dc0d62e5..cb0f5a105b89 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -94,8 +94,8 @@ static void *bpf_fd_sk_storage_lookup_elem(struct bpf_map *map, void *key) return ERR_PTR(err); } -static int bpf_fd_sk_storage_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags) +static long bpf_fd_sk_storage_update_elem(struct bpf_map *map, void *key, + void *value, u64 map_flags) { struct bpf_local_storage_data *sdata; struct socket *sock; @@ -114,7 +114,7 @@ static int bpf_fd_sk_storage_update_elem(struct bpf_map *map, void *key, return err; } -static int bpf_fd_sk_storage_delete_elem(struct bpf_map *map, void *key) +static long bpf_fd_sk_storage_delete_elem(struct bpf_map *map, void *key) { struct socket *sock; int fd, err; diff --git a/net/core/sock_map.c b/net/core/sock_map.c index 9b854e236d23..7c189c2e2fbf 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -437,7 +437,7 @@ static void sock_map_delete_from_link(struct bpf_map *map, struct sock *sk, __sock_map_delete(stab, sk, link_raw); } -static int sock_map_delete_elem(struct bpf_map *map, void *key) +static long sock_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_stab *stab = container_of(map, struct bpf_stab, map); u32 i = *(u32 *)key; @@ -587,8 +587,8 @@ out: return ret; } -static int sock_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 flags) +static long sock_map_update_elem(struct bpf_map *map, void *key, + void *value, u64 flags) { struct sock *sk = (struct sock *)value; int ret; @@ -925,7 +925,7 @@ static void sock_hash_delete_from_link(struct bpf_map *map, struct sock *sk, raw_spin_unlock_bh(&bucket->lock); } -static int sock_hash_delete_elem(struct bpf_map *map, void *key) +static long sock_hash_delete_elem(struct bpf_map *map, void *key) { struct bpf_shtab *htab = container_of(map, struct bpf_shtab, map); u32 hash, key_size = map->key_size; diff --git a/net/xdp/xskmap.c b/net/xdp/xskmap.c index 0c38d7175922..2c1427074a3b 100644 --- a/net/xdp/xskmap.c +++ b/net/xdp/xskmap.c @@ -162,8 +162,8 @@ static void *xsk_map_lookup_elem_sys_only(struct bpf_map *map, void *key) return ERR_PTR(-EOPNOTSUPP); } -static int xsk_map_update_elem(struct bpf_map *map, void *key, void *value, - u64 map_flags) +static long xsk_map_update_elem(struct bpf_map *map, void *key, void *value, + u64 map_flags) { struct xsk_map *m = container_of(map, struct xsk_map, map); struct xdp_sock __rcu **map_entry; @@ -223,7 +223,7 @@ out: return err; } -static int xsk_map_delete_elem(struct bpf_map *map, void *key) +static long xsk_map_delete_elem(struct bpf_map *map, void *key) { struct xsk_map *m = container_of(map, struct xsk_map, map); struct xdp_sock __rcu **map_entry; @@ -243,7 +243,7 @@ static int xsk_map_delete_elem(struct bpf_map *map, void *key) return 0; } -static int xsk_map_redirect(struct bpf_map *map, u64 index, u64 flags) +static long xsk_map_redirect(struct bpf_map *map, u64 index, u64 flags) { return __bpf_xdp_redirect_map(map, index, flags, 0, __xsk_map_lookup_elem); -- cgit v1.2.3 From 3c44a431d62bf4a20d7b901f98266ae3f4676d48 Mon Sep 17 00:00:00 2001 From: Zhengping Jiang Date: Tue, 21 Feb 2023 16:17:56 -0800 Subject: Bluetooth: hci_sync: Resume adv with no RPA when active scan The address resolution should be disabled during the active scan, so all the advertisements can reach the host. The advertising has to be paused before disabling the address resolution, because the advertising will prevent any changes to the resolving list and the address resolution status. Skipping this will cause the hci error and the discovery failure. According to the bluetooth specification: "7.8.44 LE Set Address Resolution Enable command This command shall not be used when: - Advertising (other than periodic advertising) is enabled, - Scanning is enabled, or - an HCI_LE_Create_Connection, HCI_LE_Extended_Create_Connection, or HCI_LE_Periodic_Advertising_Create_Sync command is outstanding." If the host is using RPA, the controller needs to generate RPA for the advertising, so the advertising must remain paused during the active scan. If the host is not using RPA, the advertising can be resumed after disabling the address resolution. Fixes: 9afc675edeeb ("Bluetooth: hci_sync: allow advertise when scan without RPA") Signed-off-by: Zhengping Jiang Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 64 +++++++++++++++++++++++++++++++++--------------- 1 file changed, 44 insertions(+), 20 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 117eedb6f709..7e152e912e8c 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -2367,6 +2367,45 @@ static int hci_resume_advertising_sync(struct hci_dev *hdev) return err; } +static int hci_pause_addr_resolution(struct hci_dev *hdev) +{ + int err; + + if (!use_ll_privacy(hdev)) + return 0; + + if (!hci_dev_test_flag(hdev, HCI_LL_RPA_RESOLUTION)) + return 0; + + /* Cannot disable addr resolution if scanning is enabled or + * when initiating an LE connection. + */ + if (hci_dev_test_flag(hdev, HCI_LE_SCAN) || + hci_lookup_le_connect(hdev)) { + bt_dev_err(hdev, "Command not allowed when scan/LE connect"); + return -EPERM; + } + + /* Cannot disable addr resolution if advertising is enabled. */ + err = hci_pause_advertising_sync(hdev); + if (err) { + bt_dev_err(hdev, "Pause advertising failed: %d", err); + return err; + } + + err = hci_le_set_addr_resolution_enable_sync(hdev, 0x00); + if (err) + bt_dev_err(hdev, "Unable to disable Address Resolution: %d", + err); + + /* Return if address resolution is disabled and RPA is not used. */ + if (!err && scan_use_rpa(hdev)) + return err; + + hci_resume_advertising_sync(hdev); + return err; +} + struct sk_buff *hci_read_local_oob_data_sync(struct hci_dev *hdev, bool extended, struct sock *sk) { @@ -2402,7 +2441,7 @@ static u8 hci_update_accept_list_sync(struct hci_dev *hdev) u8 filter_policy; int err; - /* Pause advertising if resolving list can be used as controllers are + /* Pause advertising if resolving list can be used as controllers * cannot accept resolving list modifications while advertising. */ if (use_ll_privacy(hdev)) { @@ -5394,27 +5433,12 @@ static int hci_active_scan_sync(struct hci_dev *hdev, uint16_t interval) cancel_interleave_scan(hdev); - /* Pause advertising since active scanning disables address resolution - * which advertising depend on in order to generate its RPAs. - */ - if (use_ll_privacy(hdev) && hci_dev_test_flag(hdev, HCI_PRIVACY)) { - err = hci_pause_advertising_sync(hdev); - if (err) { - bt_dev_err(hdev, "pause advertising failed: %d", err); - goto failed; - } - } - - /* Disable address resolution while doing active scanning since the - * accept list shall not be used and all reports shall reach the host - * anyway. + /* Pause address resolution for active scan and stop advertising if + * privacy is enabled. */ - err = hci_le_set_addr_resolution_enable_sync(hdev, 0x00); - if (err) { - bt_dev_err(hdev, "Unable to disable Address Resolution: %d", - err); + err = hci_pause_addr_resolution(hdev); + if (err) goto failed; - } /* All active scans will be done with either a resolvable private * address (when privacy feature has been enabled) or non-resolvable -- cgit v1.2.3 From 876e78104f23ce9267822757a63562a609b126c3 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Fri, 24 Feb 2023 15:43:31 -0800 Subject: Bluetooth: hci_core: Detect if an ACL packet is in fact an ISO packet Because some transports don't have a dedicated type for ISO packets (see 14202eff214e1e941fefa0366d4c3bc4b1a0d500) they may use ACL type when in fact they are ISO packets. In the past this was left for the driver to detect such thing but it creates a problem when using the likes of btproxy when used by a VM as the host would not be aware of the connection the guest is doing it won't be able to detect such behavior, so this make bt_recv_frame detect when it happens as it is the common interface to all drivers including guest VMs. Fixes: 14202eff214e ("Bluetooth: btusb: Detect if an ACL packet is in fact an ISO packet") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_core.c | 23 +++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index b65c3aabcd53..334e308451f5 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -2871,10 +2871,25 @@ int hci_recv_frame(struct hci_dev *hdev, struct sk_buff *skb) return -ENXIO; } - if (hci_skb_pkt_type(skb) != HCI_EVENT_PKT && - hci_skb_pkt_type(skb) != HCI_ACLDATA_PKT && - hci_skb_pkt_type(skb) != HCI_SCODATA_PKT && - hci_skb_pkt_type(skb) != HCI_ISODATA_PKT) { + switch (hci_skb_pkt_type(skb)) { + case HCI_EVENT_PKT: + break; + case HCI_ACLDATA_PKT: + /* Detect if ISO packet has been sent as ACL */ + if (hci_conn_num(hdev, ISO_LINK)) { + __u16 handle = __le16_to_cpu(hci_acl_hdr(skb)->handle); + __u8 type; + + type = hci_conn_lookup_type(hdev, hci_handle(handle)); + if (type == ISO_LINK) + hci_skb_pkt_type(skb) = HCI_ISODATA_PKT; + } + break; + case HCI_SCODATA_PKT: + break; + case HCI_ISODATA_PKT: + break; + default: kfree_skb(skb); return -EINVAL; } -- cgit v1.2.3 From 2f10e40a948e8a2abe7f983df3959a333ca8955f Mon Sep 17 00:00:00 2001 From: Pauli Virtanen Date: Mon, 20 Feb 2023 19:38:24 +0000 Subject: Bluetooth: ISO: fix timestamped HCI ISO data packet parsing Use correct HCI ISO data packet header struct when the packet has timestamp. The timestamp, when present, goes before the other fields (Core v5.3 4E 5.4.5), so the structs are not compatible. Fixes: ccf74f2390d6 ("Bluetooth: Add BTPROTO_ISO socket type") Signed-off-by: Pauli Virtanen Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 24444b502e58..8d136a730163 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -1620,7 +1620,6 @@ static void iso_disconn_cfm(struct hci_conn *hcon, __u8 reason) void iso_recv(struct hci_conn *hcon, struct sk_buff *skb, u16 flags) { struct iso_conn *conn = hcon->iso_data; - struct hci_iso_data_hdr *hdr; __u16 pb, ts, len; if (!conn) @@ -1642,6 +1641,8 @@ void iso_recv(struct hci_conn *hcon, struct sk_buff *skb, u16 flags) } if (ts) { + struct hci_iso_ts_data_hdr *hdr; + /* TODO: add timestamp to the packet? */ hdr = skb_pull_data(skb, HCI_ISO_TS_DATA_HDR_SIZE); if (!hdr) { @@ -1649,15 +1650,19 @@ void iso_recv(struct hci_conn *hcon, struct sk_buff *skb, u16 flags) goto drop; } + len = __le16_to_cpu(hdr->slen); } else { + struct hci_iso_data_hdr *hdr; + hdr = skb_pull_data(skb, HCI_ISO_DATA_HDR_SIZE); if (!hdr) { BT_ERR("Frame is too short (len %d)", skb->len); goto drop; } + + len = __le16_to_cpu(hdr->slen); } - len = __le16_to_cpu(hdr->slen); flags = hci_iso_data_flags(len); len = hci_iso_data_len(len); -- cgit v1.2.3 From 1c66bee492a5fe00ae3fe890bb693bfc99f994c6 Mon Sep 17 00:00:00 2001 From: Min Li Date: Sat, 4 Mar 2023 21:50:35 +0800 Subject: Bluetooth: Fix race condition in hci_cmd_sync_clear There is a potential race condition in hci_cmd_sync_work and hci_cmd_sync_clear, and could lead to use-after-free. For instance, hci_cmd_sync_work is added to the 'req_workqueue' after cancel_work_sync The entry of 'cmd_sync_work_list' may be freed in hci_cmd_sync_clear, and causing kernel panic when it is used in 'hci_cmd_sync_work'. Here's the call trace: dump_stack_lvl+0x49/0x63 print_report.cold+0x5e/0x5d3 ? hci_cmd_sync_work+0x282/0x320 kasan_report+0xaa/0x120 ? hci_cmd_sync_work+0x282/0x320 __asan_report_load8_noabort+0x14/0x20 hci_cmd_sync_work+0x282/0x320 process_one_work+0x77b/0x11c0 ? _raw_spin_lock_irq+0x8e/0xf0 worker_thread+0x544/0x1180 ? poll_idle+0x1e0/0x1e0 kthread+0x285/0x320 ? process_one_work+0x11c0/0x11c0 ? kthread_complete_and_exit+0x30/0x30 ret_from_fork+0x22/0x30 Allocated by task 266: kasan_save_stack+0x26/0x50 __kasan_kmalloc+0xae/0xe0 kmem_cache_alloc_trace+0x191/0x350 hci_cmd_sync_queue+0x97/0x2b0 hci_update_passive_scan+0x176/0x1d0 le_conn_complete_evt+0x1b5/0x1a00 hci_le_conn_complete_evt+0x234/0x340 hci_le_meta_evt+0x231/0x4e0 hci_event_packet+0x4c5/0xf00 hci_rx_work+0x37d/0x880 process_one_work+0x77b/0x11c0 worker_thread+0x544/0x1180 kthread+0x285/0x320 ret_from_fork+0x22/0x30 Freed by task 269: kasan_save_stack+0x26/0x50 kasan_set_track+0x25/0x40 kasan_set_free_info+0x24/0x40 ____kasan_slab_free+0x176/0x1c0 __kasan_slab_free+0x12/0x20 slab_free_freelist_hook+0x95/0x1a0 kfree+0xba/0x2f0 hci_cmd_sync_clear+0x14c/0x210 hci_unregister_dev+0xff/0x440 vhci_release+0x7b/0xf0 __fput+0x1f3/0x970 ____fput+0xe/0x20 task_work_run+0xd4/0x160 do_exit+0x8b0/0x22a0 do_group_exit+0xba/0x2a0 get_signal+0x1e4a/0x25b0 arch_do_signal_or_restart+0x93/0x1f80 exit_to_user_mode_prepare+0xf5/0x1a0 syscall_exit_to_user_mode+0x26/0x50 ret_from_fork+0x15/0x30 Fixes: 6a98e3836fa2 ("Bluetooth: Add helper for serialized HCI command execution") Cc: stable@vger.kernel.org Signed-off-by: Min Li Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 7e152e912e8c..5b8dc8fb2e27 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -643,6 +643,7 @@ void hci_cmd_sync_clear(struct hci_dev *hdev) cancel_work_sync(&hdev->cmd_sync_work); cancel_work_sync(&hdev->reenable_adv_work); + mutex_lock(&hdev->cmd_sync_work_lock); list_for_each_entry_safe(entry, tmp, &hdev->cmd_sync_work_list, list) { if (entry->destroy) entry->destroy(hdev, entry->data, -ECANCELED); @@ -650,6 +651,7 @@ void hci_cmd_sync_clear(struct hci_dev *hdev) list_del(&entry->list); kfree(entry); } + mutex_unlock(&hdev->cmd_sync_work_lock); } void __hci_cmd_sync_cancel(struct hci_dev *hdev, int err) -- cgit v1.2.3 From 52dd5e964a55c98c1b0bcf5fc737a5ddd00e7d4d Mon Sep 17 00:00:00 2001 From: Brian Gix Date: Mon, 6 Mar 2023 14:32:21 -0800 Subject: Bluetooth: Remove "Power-on" check from Mesh feature The Bluetooth mesh experimental feature enable was requiring the controller to be powered off in order for the Enable to work. Mesh is supposed to be enablable regardless of the controller state, and created an unintended requirement that the mesh daemon be started before the classic bluetoothd daemon. Fixes: af6bcc1921ff ("Bluetooth: Add experimental wrapper for MGMT based mesh") Signed-off-by: Brian Gix Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/mgmt.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'net') diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 7add66f30e4d..39589f864ea7 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -4639,12 +4639,6 @@ static int set_mgmt_mesh_func(struct sock *sk, struct hci_dev *hdev, MGMT_OP_SET_EXP_FEATURE, MGMT_STATUS_INVALID_INDEX); - /* Changes can only be made when controller is powered down */ - if (hdev_is_powered(hdev)) - return mgmt_cmd_status(sk, hdev->id, - MGMT_OP_SET_EXP_FEATURE, - MGMT_STATUS_REJECTED); - /* Parameters are limited to a single octet */ if (data_len != MGMT_SET_EXP_FEATURE_SIZE + 1) return mgmt_cmd_status(sk, hdev->id, -- cgit v1.2.3 From 8fb1a76a0f35c45a424c9eb84b0f97ffd51e6052 Mon Sep 17 00:00:00 2001 From: Kui-Feng Lee Date: Wed, 22 Mar 2023 20:23:59 -0700 Subject: net: Update an existing TCP congestion control algorithm. This feature lets you immediately transition to another congestion control algorithm or implementation with the same name. Once a name is updated, new connections will apply this new algorithm. The purpose is to update a customized algorithm implemented in BPF struct_ops with a new version on the flight. The following is an example of using the userspace API implemented in later BPF patches. link = bpf_map__attach_struct_ops(skel->maps.ca_update_1); ....... err = bpf_link__update_map(link, skel->maps.ca_update_2); We first load and register an algorithm implemented in BPF struct_ops, then swap it out with a new one using the same name. After that, newly created connections will apply the updated algorithm, while older ones retain the previous version already applied. This patch also takes this chance to refactor the ca validation into the new tcp_validate_congestion_control() function. Cc: netdev@vger.kernel.org, Eric Dumazet Signed-off-by: Kui-Feng Lee Link: https://lore.kernel.org/r/20230323032405.3735486-3-kuifeng@meta.com Signed-off-by: Martin KaFai Lau --- include/net/tcp.h | 3 +++ net/ipv4/tcp_cong.c | 66 +++++++++++++++++++++++++++++++++++++++++++++++------ 2 files changed, 62 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index db9f828e9d1e..2abb755e6a3a 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1117,6 +1117,9 @@ struct tcp_congestion_ops { int tcp_register_congestion_control(struct tcp_congestion_ops *type); void tcp_unregister_congestion_control(struct tcp_congestion_ops *type); +int tcp_update_congestion_control(struct tcp_congestion_ops *type, + struct tcp_congestion_ops *old_type); +int tcp_validate_congestion_control(struct tcp_congestion_ops *ca); void tcp_assign_congestion_control(struct sock *sk); void tcp_init_congestion_control(struct sock *sk); diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index db8b4b488c31..1b34050a7538 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -75,14 +75,8 @@ struct tcp_congestion_ops *tcp_ca_find_key(u32 key) return NULL; } -/* - * Attach new congestion control algorithm to the list - * of available options. - */ -int tcp_register_congestion_control(struct tcp_congestion_ops *ca) +int tcp_validate_congestion_control(struct tcp_congestion_ops *ca) { - int ret = 0; - /* all algorithms must implement these */ if (!ca->ssthresh || !ca->undo_cwnd || !(ca->cong_avoid || ca->cong_control)) { @@ -90,6 +84,20 @@ int tcp_register_congestion_control(struct tcp_congestion_ops *ca) return -EINVAL; } + return 0; +} + +/* Attach new congestion control algorithm to the list + * of available options. + */ +int tcp_register_congestion_control(struct tcp_congestion_ops *ca) +{ + int ret; + + ret = tcp_validate_congestion_control(ca); + if (ret) + return ret; + ca->key = jhash(ca->name, sizeof(ca->name), strlen(ca->name)); spin_lock(&tcp_cong_list_lock); @@ -130,6 +138,50 @@ void tcp_unregister_congestion_control(struct tcp_congestion_ops *ca) } EXPORT_SYMBOL_GPL(tcp_unregister_congestion_control); +/* Replace a registered old ca with a new one. + * + * The new ca must have the same name as the old one, that has been + * registered. + */ +int tcp_update_congestion_control(struct tcp_congestion_ops *ca, struct tcp_congestion_ops *old_ca) +{ + struct tcp_congestion_ops *existing; + int ret; + + ret = tcp_validate_congestion_control(ca); + if (ret) + return ret; + + ca->key = jhash(ca->name, sizeof(ca->name), strlen(ca->name)); + + spin_lock(&tcp_cong_list_lock); + existing = tcp_ca_find_key(old_ca->key); + if (ca->key == TCP_CA_UNSPEC || !existing || strcmp(existing->name, ca->name)) { + pr_notice("%s not registered or non-unique key\n", + ca->name); + ret = -EINVAL; + } else if (existing != old_ca) { + pr_notice("invalid old congestion control algorithm to replace\n"); + ret = -EINVAL; + } else { + /* Add the new one before removing the old one to keep + * one implementation available all the time. + */ + list_add_tail_rcu(&ca->list, &tcp_cong_list); + list_del_rcu(&existing->list); + pr_debug("%s updated\n", ca->name); + } + spin_unlock(&tcp_cong_list_lock); + + /* Wait for outstanding readers to complete before the + * module or struct_ops gets removed entirely. + */ + if (!ret) + synchronize_rcu(); + + return ret; +} + u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca) { const struct tcp_congestion_ops *ca; -- cgit v1.2.3 From 68b04864ca425d1894c96b8141d4fba1181f11cb Mon Sep 17 00:00:00 2001 From: Kui-Feng Lee Date: Wed, 22 Mar 2023 20:24:00 -0700 Subject: bpf: Create links for BPF struct_ops maps. Make bpf_link support struct_ops. Previously, struct_ops were always used alone without any associated links. Upon updating its value, a struct_ops would be activated automatically. Yet other BPF program types required to make a bpf_link with their instances before they could become active. Now, however, you can create an inactive struct_ops, and create a link to activate it later. With bpf_links, struct_ops has a behavior similar to other BPF program types. You can pin/unpin them from their links and the struct_ops will be deactivated when its link is removed while previously need someone to delete the value for it to be deactivated. bpf_links are responsible for registering their associated struct_ops. You can only use a struct_ops that has the BPF_F_LINK flag set to create a bpf_link, while a structs without this flag behaves in the same manner as before and is registered upon updating its value. The BPF_LINK_TYPE_STRUCT_OPS serves a dual purpose. Not only is it used to craft the links for BPF struct_ops programs, but also to create links for BPF struct_ops them-self. Since the links of BPF struct_ops programs are only used to create trampolines internally, they are never seen in other contexts. Thus, they can be reused for struct_ops themself. To maintain a reference to the map supporting this link, we add bpf_struct_ops_link as an additional type. The pointer of the map is RCU and won't be necessary until later in the patchset. Signed-off-by: Kui-Feng Lee Link: https://lore.kernel.org/r/20230323032405.3735486-4-kuifeng@meta.com Signed-off-by: Martin KaFai Lau --- include/linux/bpf.h | 7 ++ include/uapi/linux/bpf.h | 12 +++- kernel/bpf/bpf_struct_ops.c | 143 ++++++++++++++++++++++++++++++++++++++++- kernel/bpf/syscall.c | 23 ++++--- net/ipv4/bpf_tcp_ca.c | 8 ++- tools/include/uapi/linux/bpf.h | 12 +++- 6 files changed, 190 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index f04098468d7a..8552279efe46 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -1518,6 +1518,7 @@ struct bpf_struct_ops { void *kdata, const void *udata); int (*reg)(void *kdata); void (*unreg)(void *kdata); + int (*validate)(void *kdata); const struct btf_type *type; const struct btf_type *value_type; const char *name; @@ -1552,6 +1553,7 @@ static inline void bpf_module_put(const void *data, struct module *owner) else module_put(owner); } +int bpf_struct_ops_link_create(union bpf_attr *attr); #ifdef CONFIG_NET /* Define it here to avoid the use of forward declaration */ @@ -1592,6 +1594,11 @@ static inline int bpf_struct_ops_map_sys_lookup_elem(struct bpf_map *map, { return -EINVAL; } +static inline int bpf_struct_ops_link_create(union bpf_attr *attr) +{ + return -EOPNOTSUPP; +} + #endif #if defined(CONFIG_CGROUP_BPF) && defined(CONFIG_BPF_LSM) diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 13129df937cd..42f40ee083bf 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -1033,6 +1033,7 @@ enum bpf_attach_type { BPF_PERF_EVENT, BPF_TRACE_KPROBE_MULTI, BPF_LSM_CGROUP, + BPF_STRUCT_OPS, __MAX_BPF_ATTACH_TYPE }; @@ -1266,6 +1267,9 @@ enum { /* Create a map that is suitable to be an inner map with dynamic max entries */ BPF_F_INNER_MAP = (1U << 12), + +/* Create a map that will be registered/unregesitered by the backed bpf_link */ + BPF_F_LINK = (1U << 13), }; /* Flags for BPF_PROG_QUERY. */ @@ -1507,7 +1511,10 @@ union bpf_attr { } task_fd_query; struct { /* struct used by BPF_LINK_CREATE command */ - __u32 prog_fd; /* eBPF program to attach */ + union { + __u32 prog_fd; /* eBPF program to attach */ + __u32 map_fd; /* struct_ops to attach */ + }; union { __u32 target_fd; /* object to attach to */ __u32 target_ifindex; /* target ifindex */ @@ -6379,6 +6386,9 @@ struct bpf_link_info { struct { __u32 ifindex; } xdp; + struct { + __u32 map_id; + } struct_ops; }; } __attribute__((aligned(8))); diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c index 2f3c4a0e03ee..3d6b5240c25a 100644 --- a/kernel/bpf/bpf_struct_ops.c +++ b/kernel/bpf/bpf_struct_ops.c @@ -17,6 +17,7 @@ enum bpf_struct_ops_state { BPF_STRUCT_OPS_STATE_INIT, BPF_STRUCT_OPS_STATE_INUSE, BPF_STRUCT_OPS_STATE_TOBEFREE, + BPF_STRUCT_OPS_STATE_READY, }; #define BPF_STRUCT_OPS_COMMON_VALUE \ @@ -59,6 +60,11 @@ struct bpf_struct_ops_map { struct bpf_struct_ops_value kvalue; }; +struct bpf_struct_ops_link { + struct bpf_link link; + struct bpf_map __rcu *map; +}; + #define VALUE_PREFIX "bpf_struct_ops_" #define VALUE_PREFIX_LEN (sizeof(VALUE_PREFIX) - 1) @@ -500,11 +506,29 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key, *(unsigned long *)(udata + moff) = prog->aux->id; } - bpf_map_inc(map); + if (st_map->map.map_flags & BPF_F_LINK) { + err = st_ops->validate(kdata); + if (err) + goto reset_unlock; + set_memory_rox((long)st_map->image, 1); + /* Let bpf_link handle registration & unregistration. + * + * Pair with smp_load_acquire() during lookup_elem(). + */ + smp_store_release(&kvalue->state, BPF_STRUCT_OPS_STATE_READY); + goto unlock; + } set_memory_rox((long)st_map->image, 1); err = st_ops->reg(kdata); if (likely(!err)) { + /* This refcnt increment on the map here after + * 'st_ops->reg()' is secure since the state of the + * map must be set to INIT at this moment, and thus + * bpf_struct_ops_map_delete_elem() can't unregister + * or transition it to TOBEFREE concurrently. + */ + bpf_map_inc(map); /* Pair with smp_load_acquire() during lookup_elem(). * It ensures the above udata updates (e.g. prog->aux->id) * can be seen once BPF_STRUCT_OPS_STATE_INUSE is set. @@ -520,7 +544,6 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key, */ set_memory_nx((long)st_map->image, 1); set_memory_rw((long)st_map->image, 1); - bpf_map_put(map); reset_unlock: bpf_struct_ops_map_put_progs(st_map); @@ -538,6 +561,9 @@ static long bpf_struct_ops_map_delete_elem(struct bpf_map *map, void *key) struct bpf_struct_ops_map *st_map; st_map = (struct bpf_struct_ops_map *)map; + if (st_map->map.map_flags & BPF_F_LINK) + return -EOPNOTSUPP; + prev_state = cmpxchg(&st_map->kvalue.state, BPF_STRUCT_OPS_STATE_INUSE, BPF_STRUCT_OPS_STATE_TOBEFREE); @@ -614,7 +640,7 @@ static void bpf_struct_ops_map_free(struct bpf_map *map) static int bpf_struct_ops_map_alloc_check(union bpf_attr *attr) { if (attr->key_size != sizeof(unsigned int) || attr->max_entries != 1 || - attr->map_flags || !attr->btf_vmlinux_value_type_id) + (attr->map_flags & ~BPF_F_LINK) || !attr->btf_vmlinux_value_type_id) return -EINVAL; return 0; } @@ -638,6 +664,9 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr) if (attr->value_size != vt->size) return ERR_PTR(-EINVAL); + if (attr->map_flags & BPF_F_LINK && !st_ops->validate) + return ERR_PTR(-EOPNOTSUPP); + t = st_ops->type; st_map_size = sizeof(*st_map) + @@ -725,3 +754,111 @@ void bpf_struct_ops_put(const void *kdata) bpf_map_put(&st_map->map); } + +static bool bpf_struct_ops_valid_to_reg(struct bpf_map *map) +{ + struct bpf_struct_ops_map *st_map = (struct bpf_struct_ops_map *)map; + + return map->map_type == BPF_MAP_TYPE_STRUCT_OPS && + map->map_flags & BPF_F_LINK && + /* Pair with smp_store_release() during map_update */ + smp_load_acquire(&st_map->kvalue.state) == BPF_STRUCT_OPS_STATE_READY; +} + +static void bpf_struct_ops_map_link_dealloc(struct bpf_link *link) +{ + struct bpf_struct_ops_link *st_link; + struct bpf_struct_ops_map *st_map; + + st_link = container_of(link, struct bpf_struct_ops_link, link); + st_map = (struct bpf_struct_ops_map *) + rcu_dereference_protected(st_link->map, true); + if (st_map) { + /* st_link->map can be NULL if + * bpf_struct_ops_link_create() fails to register. + */ + st_map->st_ops->unreg(&st_map->kvalue.data); + bpf_map_put(&st_map->map); + } + kfree(st_link); +} + +static void bpf_struct_ops_map_link_show_fdinfo(const struct bpf_link *link, + struct seq_file *seq) +{ + struct bpf_struct_ops_link *st_link; + struct bpf_map *map; + + st_link = container_of(link, struct bpf_struct_ops_link, link); + rcu_read_lock(); + map = rcu_dereference(st_link->map); + seq_printf(seq, "map_id:\t%d\n", map->id); + rcu_read_unlock(); +} + +static int bpf_struct_ops_map_link_fill_link_info(const struct bpf_link *link, + struct bpf_link_info *info) +{ + struct bpf_struct_ops_link *st_link; + struct bpf_map *map; + + st_link = container_of(link, struct bpf_struct_ops_link, link); + rcu_read_lock(); + map = rcu_dereference(st_link->map); + info->struct_ops.map_id = map->id; + rcu_read_unlock(); + return 0; +} + +static const struct bpf_link_ops bpf_struct_ops_map_lops = { + .dealloc = bpf_struct_ops_map_link_dealloc, + .show_fdinfo = bpf_struct_ops_map_link_show_fdinfo, + .fill_link_info = bpf_struct_ops_map_link_fill_link_info, +}; + +int bpf_struct_ops_link_create(union bpf_attr *attr) +{ + struct bpf_struct_ops_link *link = NULL; + struct bpf_link_primer link_primer; + struct bpf_struct_ops_map *st_map; + struct bpf_map *map; + int err; + + map = bpf_map_get(attr->link_create.map_fd); + if (!map) + return -EINVAL; + + st_map = (struct bpf_struct_ops_map *)map; + + if (!bpf_struct_ops_valid_to_reg(map)) { + err = -EINVAL; + goto err_out; + } + + link = kzalloc(sizeof(*link), GFP_USER); + if (!link) { + err = -ENOMEM; + goto err_out; + } + bpf_link_init(&link->link, BPF_LINK_TYPE_STRUCT_OPS, &bpf_struct_ops_map_lops, NULL); + + err = bpf_link_prime(&link->link, &link_primer); + if (err) + goto err_out; + + err = st_map->st_ops->reg(st_map->kvalue.data); + if (err) { + bpf_link_cleanup(&link_primer); + link = NULL; + goto err_out; + } + RCU_INIT_POINTER(link->map, map); + + return bpf_link_settle(&link_primer); + +err_out: + bpf_map_put(map); + kfree(link); + return err; +} + diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index cff0348a2871..21f76698875c 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -2825,16 +2825,19 @@ static void bpf_link_show_fdinfo(struct seq_file *m, struct file *filp) const struct bpf_prog *prog = link->prog; char prog_tag[sizeof(prog->tag) * 2 + 1] = { }; - bin2hex(prog_tag, prog->tag, sizeof(prog->tag)); seq_printf(m, "link_type:\t%s\n" - "link_id:\t%u\n" - "prog_tag:\t%s\n" - "prog_id:\t%u\n", + "link_id:\t%u\n", bpf_link_type_strs[link->type], - link->id, - prog_tag, - prog->aux->id); + link->id); + if (prog) { + bin2hex(prog_tag, prog->tag, sizeof(prog->tag)); + seq_printf(m, + "prog_tag:\t%s\n" + "prog_id:\t%u\n", + prog_tag, + prog->aux->id); + } if (link->ops->show_fdinfo) link->ops->show_fdinfo(link, m); } @@ -4314,7 +4317,8 @@ static int bpf_link_get_info_by_fd(struct file *file, info.type = link->type; info.id = link->id; - info.prog_id = link->prog->aux->id; + if (link->prog) + info.prog_id = link->prog->aux->id; if (link->ops->fill_link_info) { err = link->ops->fill_link_info(link, &info); @@ -4577,6 +4581,9 @@ static int link_create(union bpf_attr *attr, bpfptr_t uattr) if (CHECK_ATTR(BPF_LINK_CREATE)) return -EINVAL; + if (attr->link_create.attach_type == BPF_STRUCT_OPS) + return bpf_struct_ops_link_create(attr); + prog = bpf_prog_get(attr->link_create.prog_fd); if (IS_ERR(prog)) return PTR_ERR(prog); diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index 13fc0c185cd9..bbbd5eb94db2 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -239,8 +239,6 @@ static int bpf_tcp_ca_init_member(const struct btf_type *t, if (bpf_obj_name_cpy(tcp_ca->name, utcp_ca->name, sizeof(tcp_ca->name)) <= 0) return -EINVAL; - if (tcp_ca_find(utcp_ca->name)) - return -EEXIST; return 1; } @@ -266,6 +264,11 @@ static void bpf_tcp_ca_unreg(void *kdata) tcp_unregister_congestion_control(kdata); } +static int bpf_tcp_ca_validate(void *kdata) +{ + return tcp_validate_congestion_control(kdata); +} + struct bpf_struct_ops bpf_tcp_congestion_ops = { .verifier_ops = &bpf_tcp_ca_verifier_ops, .reg = bpf_tcp_ca_reg, @@ -273,6 +276,7 @@ struct bpf_struct_ops bpf_tcp_congestion_ops = { .check_member = bpf_tcp_ca_check_member, .init_member = bpf_tcp_ca_init_member, .init = bpf_tcp_ca_init, + .validate = bpf_tcp_ca_validate, .name = "tcp_congestion_ops", }; diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 13129df937cd..9cf1deaf21f2 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -1033,6 +1033,7 @@ enum bpf_attach_type { BPF_PERF_EVENT, BPF_TRACE_KPROBE_MULTI, BPF_LSM_CGROUP, + BPF_STRUCT_OPS, __MAX_BPF_ATTACH_TYPE }; @@ -1266,6 +1267,9 @@ enum { /* Create a map that is suitable to be an inner map with dynamic max entries */ BPF_F_INNER_MAP = (1U << 12), + +/* Create a map that will be registered/unregesitered by the backed bpf_link */ + BPF_F_LINK = (1U << 13), }; /* Flags for BPF_PROG_QUERY. */ @@ -1507,7 +1511,10 @@ union bpf_attr { } task_fd_query; struct { /* struct used by BPF_LINK_CREATE command */ - __u32 prog_fd; /* eBPF program to attach */ + union { + __u32 prog_fd; /* eBPF program to attach */ + __u32 map_fd; /* eBPF struct_ops to attach */ + }; union { __u32 target_fd; /* object to attach to */ __u32 target_ifindex; /* target ifindex */ @@ -6379,6 +6386,9 @@ struct bpf_link_info { struct { __u32 ifindex; } xdp; + struct { + __u32 map_id; + } struct_ops; }; } __attribute__((aligned(8))); -- cgit v1.2.3 From aef56f2e918bf8fc8de25f0b36e8c2aba44116ec Mon Sep 17 00:00:00 2001 From: Kui-Feng Lee Date: Wed, 22 Mar 2023 20:24:02 -0700 Subject: bpf: Update the struct_ops of a bpf_link. By improving the BPF_LINK_UPDATE command of bpf(), it should allow you to conveniently switch between different struct_ops on a single bpf_link. This would enable smoother transitions from one struct_ops to another. The struct_ops maps passing along with BPF_LINK_UPDATE should have the BPF_F_LINK flag. Signed-off-by: Kui-Feng Lee Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20230323032405.3735486-6-kuifeng@meta.com Signed-off-by: Martin KaFai Lau --- include/linux/bpf.h | 3 +++ include/uapi/linux/bpf.h | 21 +++++++++++++----- kernel/bpf/bpf_struct_ops.c | 48 +++++++++++++++++++++++++++++++++++++++++- kernel/bpf/syscall.c | 34 ++++++++++++++++++++++++++++++ net/ipv4/bpf_tcp_ca.c | 6 ++++++ tools/include/uapi/linux/bpf.h | 21 +++++++++++++----- 6 files changed, 122 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 8552279efe46..2d8f3f639e68 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -1476,6 +1476,8 @@ struct bpf_link_ops { void (*show_fdinfo)(const struct bpf_link *link, struct seq_file *seq); int (*fill_link_info)(const struct bpf_link *link, struct bpf_link_info *info); + int (*update_map)(struct bpf_link *link, struct bpf_map *new_map, + struct bpf_map *old_map); }; struct bpf_tramp_link { @@ -1518,6 +1520,7 @@ struct bpf_struct_ops { void *kdata, const void *udata); int (*reg)(void *kdata); void (*unreg)(void *kdata); + int (*update)(void *kdata, void *old_kdata); int (*validate)(void *kdata); const struct btf_type *type; const struct btf_type *value_type; diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 42f40ee083bf..e3d3b5160d26 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -1555,12 +1555,23 @@ union bpf_attr { struct { /* struct used by BPF_LINK_UPDATE command */ __u32 link_fd; /* link fd */ - /* new program fd to update link with */ - __u32 new_prog_fd; + union { + /* new program fd to update link with */ + __u32 new_prog_fd; + /* new struct_ops map fd to update link with */ + __u32 new_map_fd; + }; __u32 flags; /* extra flags */ - /* expected link's program fd; is specified only if - * BPF_F_REPLACE flag is set in flags */ - __u32 old_prog_fd; + union { + /* expected link's program fd; is specified only if + * BPF_F_REPLACE flag is set in flags. + */ + __u32 old_prog_fd; + /* expected link's map fd; is specified only + * if BPF_F_REPLACE flag is set. + */ + __u32 old_map_fd; + }; } link_update; struct { diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c index 3d6b5240c25a..6401deca3b56 100644 --- a/kernel/bpf/bpf_struct_ops.c +++ b/kernel/bpf/bpf_struct_ops.c @@ -65,6 +65,8 @@ struct bpf_struct_ops_link { struct bpf_map __rcu *map; }; +static DEFINE_MUTEX(update_mutex); + #define VALUE_PREFIX "bpf_struct_ops_" #define VALUE_PREFIX_LEN (sizeof(VALUE_PREFIX) - 1) @@ -664,7 +666,7 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr) if (attr->value_size != vt->size) return ERR_PTR(-EINVAL); - if (attr->map_flags & BPF_F_LINK && !st_ops->validate) + if (attr->map_flags & BPF_F_LINK && (!st_ops->validate || !st_ops->update)) return ERR_PTR(-EOPNOTSUPP); t = st_ops->type; @@ -810,10 +812,54 @@ static int bpf_struct_ops_map_link_fill_link_info(const struct bpf_link *link, return 0; } +static int bpf_struct_ops_map_link_update(struct bpf_link *link, struct bpf_map *new_map, + struct bpf_map *expected_old_map) +{ + struct bpf_struct_ops_map *st_map, *old_st_map; + struct bpf_map *old_map; + struct bpf_struct_ops_link *st_link; + int err = 0; + + st_link = container_of(link, struct bpf_struct_ops_link, link); + st_map = container_of(new_map, struct bpf_struct_ops_map, map); + + if (!bpf_struct_ops_valid_to_reg(new_map)) + return -EINVAL; + + mutex_lock(&update_mutex); + + old_map = rcu_dereference_protected(st_link->map, lockdep_is_held(&update_mutex)); + if (expected_old_map && old_map != expected_old_map) { + err = -EPERM; + goto err_out; + } + + old_st_map = container_of(old_map, struct bpf_struct_ops_map, map); + /* The new and old struct_ops must be the same type. */ + if (st_map->st_ops != old_st_map->st_ops) { + err = -EINVAL; + goto err_out; + } + + err = st_map->st_ops->update(st_map->kvalue.data, old_st_map->kvalue.data); + if (err) + goto err_out; + + bpf_map_inc(new_map); + rcu_assign_pointer(st_link->map, new_map); + bpf_map_put(old_map); + +err_out: + mutex_unlock(&update_mutex); + + return err; +} + static const struct bpf_link_ops bpf_struct_ops_map_lops = { .dealloc = bpf_struct_ops_map_link_dealloc, .show_fdinfo = bpf_struct_ops_map_link_show_fdinfo, .fill_link_info = bpf_struct_ops_map_link_fill_link_info, + .update_map = bpf_struct_ops_map_link_update, }; int bpf_struct_ops_link_create(union bpf_attr *attr) diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 21f76698875c..b4d758fa5981 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -4682,6 +4682,35 @@ out: return ret; } +static int link_update_map(struct bpf_link *link, union bpf_attr *attr) +{ + struct bpf_map *new_map, *old_map = NULL; + int ret; + + new_map = bpf_map_get(attr->link_update.new_map_fd); + if (IS_ERR(new_map)) + return -EINVAL; + + if (attr->link_update.flags & BPF_F_REPLACE) { + old_map = bpf_map_get(attr->link_update.old_map_fd); + if (IS_ERR(old_map)) { + ret = -EINVAL; + goto out_put; + } + } else if (attr->link_update.old_map_fd) { + ret = -EINVAL; + goto out_put; + } + + ret = link->ops->update_map(link, new_map, old_map); + + if (old_map) + bpf_map_put(old_map); +out_put: + bpf_map_put(new_map); + return ret; +} + #define BPF_LINK_UPDATE_LAST_FIELD link_update.old_prog_fd static int link_update(union bpf_attr *attr) @@ -4702,6 +4731,11 @@ static int link_update(union bpf_attr *attr) if (IS_ERR(link)) return PTR_ERR(link); + if (link->ops->update_map) { + ret = link_update_map(link, attr); + goto out_put_link; + } + new_prog = bpf_prog_get(attr->link_update.new_prog_fd); if (IS_ERR(new_prog)) { ret = PTR_ERR(new_prog); diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index bbbd5eb94db2..e8b27826283e 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -264,6 +264,11 @@ static void bpf_tcp_ca_unreg(void *kdata) tcp_unregister_congestion_control(kdata); } +static int bpf_tcp_ca_update(void *kdata, void *old_kdata) +{ + return tcp_update_congestion_control(kdata, old_kdata); +} + static int bpf_tcp_ca_validate(void *kdata) { return tcp_validate_congestion_control(kdata); @@ -273,6 +278,7 @@ struct bpf_struct_ops bpf_tcp_congestion_ops = { .verifier_ops = &bpf_tcp_ca_verifier_ops, .reg = bpf_tcp_ca_reg, .unreg = bpf_tcp_ca_unreg, + .update = bpf_tcp_ca_update, .check_member = bpf_tcp_ca_check_member, .init_member = bpf_tcp_ca_init_member, .init = bpf_tcp_ca_init, diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 9cf1deaf21f2..d6c5a022ae28 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -1555,12 +1555,23 @@ union bpf_attr { struct { /* struct used by BPF_LINK_UPDATE command */ __u32 link_fd; /* link fd */ - /* new program fd to update link with */ - __u32 new_prog_fd; + union { + /* new program fd to update link with */ + __u32 new_prog_fd; + /* new struct_ops map fd to update link with */ + __u32 new_map_fd; + }; __u32 flags; /* extra flags */ - /* expected link's program fd; is specified only if - * BPF_F_REPLACE flag is set in flags */ - __u32 old_prog_fd; + union { + /* expected link's program fd; is specified only if + * BPF_F_REPLACE flag is set in flags. + */ + __u32 old_prog_fd; + /* expected link's map fd; is specified only + * if BPF_F_REPLACE flag is set. + */ + __u32 old_map_fd; + }; } link_update; struct { -- cgit v1.2.3 From 943d045a6d796175e5d08f9973953b1d2c07d797 Mon Sep 17 00:00:00 2001 From: Siddharth Kawar Date: Mon, 20 Mar 2023 20:37:40 +0000 Subject: SUNRPC: fix shutdown of NFS TCP client socket NFS server Duplicate Request Cache (DRC) algorithms rely on NFS clients reconnecting using the same local TCP port. Unique NFS operations are identified by the per-TCP connection set of XIDs. This prevents file corruption when non-idempotent NFS operations are retried. Currently, NFS client TCP connections are using different local TCP ports when reconnecting to NFS servers. After an NFS server initiates shutdown of the TCP connection, the NFS client's TCP socket is set to NULL after the socket state has reached TCP_LAST_ACK(9). When reconnecting, the new socket attempts to reuse the same local port but fails with EADDRNOTAVAIL (99). This forces the socket to use a different local TCP port to reconnect to the remote NFS server. State Transition and Events: TCP_CLOSE_WAIT(8) TCP_LAST_ACK(9) connect(fail EADDRNOTAVAIL(99)) TCP_CLOSE(7) bind on new port connect success dmesg excerpts showing reconnect switching from TCP local port of 926 to 763 after commit 7c81e6a9d75b: [13354.947854] NFS call mkdir testW ... [13405.654781] RPC: xs_tcp_state_change client 00000000037d0f03... [13405.654813] RPC: state 8 conn 1 dead 0 zapped 1 sk_shutdown 1 [13405.654826] RPC: xs_data_ready... [13405.654892] RPC: xs_tcp_state_change client 00000000037d0f03... [13405.654895] RPC: state 9 conn 0 dead 0 zapped 1 sk_shutdown 3 [13405.654899] RPC: xs_tcp_state_change client 00000000037d0f03... [13405.654900] RPC: state 9 conn 0 dead 0 zapped 1 sk_shutdown 3 [13405.654950] RPC: xs_connect scheduled xprt 00000000037d0f03 [13405.654975] RPC: xs_bind 0.0.0.0:926: ok (0) [13405.654980] RPC: worker connecting xprt 00000000037d0f03 via tcp to 10.101.6.228 (port 2049) [13405.654991] RPC: 00000000037d0f03 connect status 99 connected 0 sock state 7 [13405.655001] RPC: xs_tcp_state_change client 00000000037d0f03... [13405.655002] RPC: state 7 conn 0 dead 0 zapped 1 sk_shutdown 3 [13405.655024] RPC: xs_connect scheduled xprt 00000000037d0f03 [13405.655038] RPC: xs_bind 0.0.0.0:763: ok (0) [13405.655041] RPC: worker connecting xprt 00000000037d0f03 via tcp to 10.101.6.228 (port 2049) [13405.655065] RPC: 00000000037d0f03 connect status 115 connected 0 sock state 2 State Transition and Events with patch applied: TCP_CLOSE_WAIT(8) TCP_LAST_ACK(9) TCP_CLOSE(7) connect(reuse of port succeeds) dmesg excerpts showing reconnect on same TCP local port of 936 with patch applied: [ 257.139935] NFS: mkdir(0:59/560857152), testQ [ 257.139937] NFS call mkdir testQ ... [ 307.822702] RPC: state 8 conn 1 dead 0 zapped 1 sk_shutdown 1 [ 307.822714] RPC: xs_data_ready... [ 307.822817] RPC: xs_tcp_state_change client 00000000ce702f14... [ 307.822821] RPC: state 9 conn 0 dead 0 zapped 1 sk_shutdown 3 [ 307.822825] RPC: xs_tcp_state_change client 00000000ce702f14... [ 307.822826] RPC: state 9 conn 0 dead 0 zapped 1 sk_shutdown 3 [ 307.823606] RPC: xs_tcp_state_change client 00000000ce702f14... [ 307.823609] RPC: state 7 conn 0 dead 0 zapped 1 sk_shutdown 3 [ 307.823629] RPC: xs_tcp_state_change client 00000000ce702f14... [ 307.823632] RPC: state 7 conn 0 dead 0 zapped 1 sk_shutdown 3 [ 307.823676] RPC: xs_connect scheduled xprt 00000000ce702f14 [ 307.823704] RPC: xs_bind 0.0.0.0:936: ok (0) [ 307.823709] RPC: worker connecting xprt 00000000ce702f14 via tcp to 10.101.1.30 (port 2049) [ 307.823748] RPC: 00000000ce702f14 connect status 115 connected 0 sock state 2 ... [ 314.916193] RPC: state 7 conn 0 dead 0 zapped 1 sk_shutdown 3 [ 314.916251] RPC: xs_connect scheduled xprt 00000000ce702f14 [ 314.916282] RPC: xs_bind 0.0.0.0:936: ok (0) [ 314.916292] RPC: worker connecting xprt 00000000ce702f14 via tcp to 10.101.1.30 (port 2049) [ 314.916342] RPC: 00000000ce702f14 connect status 115 connected 0 sock state 2 Fixes: 7c81e6a9d75b ("SUNRPC: Tweak TCP socket shutdown in the RPC client") Signed-off-by: Siddharth Rajendra Kawar Signed-off-by: Anna Schumaker --- net/sunrpc/xprtsock.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index adcbedc244d6..6cacd70a15ff 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2158,6 +2158,7 @@ static void xs_tcp_shutdown(struct rpc_xprt *xprt) switch (skst) { case TCP_FIN_WAIT1: case TCP_FIN_WAIT2: + case TCP_LAST_ACK: break; case TCP_ESTABLISHED: case TCP_CLOSE_WAIT: -- cgit v1.2.3 From 9aa9d9473f1550d1936c31259720b3f1f4690576 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Wed, 8 Mar 2023 14:20:34 -0800 Subject: Bluetooth: L2CAP: Fix responding with wrong PDU type L2CAP_ECRED_CONN_REQ shall be responded with L2CAP_ECRED_CONN_RSP not L2CAP_LE_CONN_RSP: L2CAP LE EATT Server - Reject - run Listening for connections New client connection with handle 0x002a Sending L2CAP Request from client Client received response code 0x15 Unexpected L2CAP response code (expected 0x18) L2CAP LE EATT Server - Reject - test failed > ACL Data RX: Handle 42 flags 0x02 dlen 26 LE L2CAP: Enhanced Credit Connection Request (0x17) ident 1 len 18 PSM: 39 (0x0027) MTU: 64 MPS: 64 Credits: 5 Source CID: 65 Source CID: 66 Source CID: 67 Source CID: 68 Source CID: 69 < ACL Data TX: Handle 42 flags 0x00 dlen 16 LE L2CAP: LE Connection Response (0x15) ident 1 len 8 invalid size 00 00 00 00 00 00 06 00 L2CAP LE EATT Server - Reject - run Listening for connections New client connection with handle 0x002a Sending L2CAP Request from client Client received response code 0x18 L2CAP LE EATT Server - Reject - test passed Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 117 ++++++++++++++++++++++++++++++--------------- 1 file changed, 79 insertions(+), 38 deletions(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index adfc3ea06d08..49926f59cc12 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -708,6 +708,17 @@ void l2cap_chan_del(struct l2cap_chan *chan, int err) } EXPORT_SYMBOL_GPL(l2cap_chan_del); +static void __l2cap_chan_list_id(struct l2cap_conn *conn, u16 id, + l2cap_chan_func_t func, void *data) +{ + struct l2cap_chan *chan, *l; + + list_for_each_entry_safe(chan, l, &conn->chan_l, list) { + if (chan->ident == id) + func(chan, data); + } +} + static void __l2cap_chan_list(struct l2cap_conn *conn, l2cap_chan_func_t func, void *data) { @@ -775,23 +786,9 @@ static void l2cap_chan_le_connect_reject(struct l2cap_chan *chan) static void l2cap_chan_ecred_connect_reject(struct l2cap_chan *chan) { - struct l2cap_conn *conn = chan->conn; - struct l2cap_ecred_conn_rsp rsp; - u16 result; - - if (test_bit(FLAG_DEFER_SETUP, &chan->flags)) - result = L2CAP_CR_LE_AUTHORIZATION; - else - result = L2CAP_CR_LE_BAD_PSM; - l2cap_state_change(chan, BT_DISCONN); - memset(&rsp, 0, sizeof(rsp)); - - rsp.result = cpu_to_le16(result); - - l2cap_send_cmd(conn, chan->ident, L2CAP_LE_CONN_RSP, sizeof(rsp), - &rsp); + __l2cap_ecred_conn_rsp_defer(chan); } static void l2cap_chan_connect_reject(struct l2cap_chan *chan) @@ -846,7 +843,7 @@ void l2cap_chan_close(struct l2cap_chan *chan, int reason) break; case L2CAP_MODE_EXT_FLOWCTL: l2cap_chan_ecred_connect_reject(chan); - break; + return; } } } @@ -3938,43 +3935,86 @@ void __l2cap_le_connect_rsp_defer(struct l2cap_chan *chan) &rsp); } -void __l2cap_ecred_conn_rsp_defer(struct l2cap_chan *chan) +static void l2cap_ecred_list_defer(struct l2cap_chan *chan, void *data) { + int *result = data; + + if (*result || test_bit(FLAG_ECRED_CONN_REQ_SENT, &chan->flags)) + return; + + switch (chan->state) { + case BT_CONNECT2: + /* If channel still pending accept add to result */ + (*result)++; + return; + case BT_CONNECTED: + return; + default: + /* If not connected or pending accept it has been refused */ + *result = -ECONNREFUSED; + return; + } +} + +struct l2cap_ecred_rsp_data { struct { struct l2cap_ecred_conn_rsp rsp; - __le16 dcid[5]; + __le16 scid[L2CAP_ECRED_MAX_CID]; } __packed pdu; + int count; +}; + +static void l2cap_ecred_rsp_defer(struct l2cap_chan *chan, void *data) +{ + struct l2cap_ecred_rsp_data *rsp = data; + + if (test_bit(FLAG_ECRED_CONN_REQ_SENT, &chan->flags)) + return; + + /* Reset ident so only one response is sent */ + chan->ident = 0; + + /* Include all channels pending with the same ident */ + if (!rsp->pdu.rsp.result) + rsp->pdu.rsp.dcid[rsp->count++] = cpu_to_le16(chan->scid); + else + l2cap_chan_del(chan, ECONNRESET); +} + +void __l2cap_ecred_conn_rsp_defer(struct l2cap_chan *chan) +{ struct l2cap_conn *conn = chan->conn; - u16 ident = chan->ident; - int i = 0; + struct l2cap_ecred_rsp_data data; + u16 id = chan->ident; + int result = 0; - if (!ident) + if (!id) return; - BT_DBG("chan %p ident %d", chan, ident); + BT_DBG("chan %p id %d", chan, id); - pdu.rsp.mtu = cpu_to_le16(chan->imtu); - pdu.rsp.mps = cpu_to_le16(chan->mps); - pdu.rsp.credits = cpu_to_le16(chan->rx_credits); - pdu.rsp.result = cpu_to_le16(L2CAP_CR_LE_SUCCESS); + memset(&data, 0, sizeof(data)); - mutex_lock(&conn->chan_lock); + data.pdu.rsp.mtu = cpu_to_le16(chan->imtu); + data.pdu.rsp.mps = cpu_to_le16(chan->mps); + data.pdu.rsp.credits = cpu_to_le16(chan->rx_credits); + data.pdu.rsp.result = cpu_to_le16(L2CAP_CR_LE_SUCCESS); - list_for_each_entry(chan, &conn->chan_l, list) { - if (chan->ident != ident) - continue; + /* Verify that all channels are ready */ + __l2cap_chan_list_id(conn, id, l2cap_ecred_list_defer, &result); - /* Reset ident so only one response is sent */ - chan->ident = 0; + if (result > 0) + return; - /* Include all channels pending with the same ident */ - pdu.dcid[i++] = cpu_to_le16(chan->scid); - } + if (result < 0) + data.pdu.rsp.result = cpu_to_le16(L2CAP_CR_LE_AUTHORIZATION); - mutex_unlock(&conn->chan_lock); + /* Build response */ + __l2cap_chan_list_id(conn, id, l2cap_ecred_rsp_defer, &data); - l2cap_send_cmd(conn, ident, L2CAP_ECRED_CONN_RSP, - sizeof(pdu.rsp) + i * sizeof(__le16), &pdu); + l2cap_send_cmd(conn, id, L2CAP_ECRED_CONN_RSP, + sizeof(data.pdu.rsp) + (data.count * sizeof(__le16)), + &data.pdu); } void __l2cap_connect_rsp_defer(struct l2cap_chan *chan) @@ -6078,6 +6118,7 @@ static inline int l2cap_ecred_conn_req(struct l2cap_conn *conn, __set_chan_timer(chan, chan->ops->get_sndtimeo(chan)); chan->ident = cmd->ident; + chan->mode = L2CAP_MODE_EXT_FLOWCTL; if (test_bit(FLAG_DEFER_SETUP, &chan->flags)) { l2cap_state_change(chan, BT_CONNECT2); -- cgit v1.2.3 From 1a0291f81529e8044fb29845a0196ba47af894ce Mon Sep 17 00:00:00 2001 From: Howard Chung Date: Thu, 16 Mar 2023 18:11:38 +0800 Subject: Bluetooth: mgmt: Fix MGMT add advmon with RSSI command The MGMT command: MGMT_OP_ADD_ADV_PATTERNS_MONITOR_RSSI uses variable length argument. This causes host not able to register advmon with rssi. This patch has been locally tested by adding monitor with rssi via btmgmt on a kernel 6.1 machine. Reviewed-by: Archie Pusaka Fixes: b338d91703fa ("Bluetooth: Implement support for Mesh") Signed-off-by: Howard Chung Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/mgmt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 39589f864ea7..249dc6777fb4 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -9357,7 +9357,8 @@ static const struct hci_mgmt_handler mgmt_handlers[] = { { add_ext_adv_data, MGMT_ADD_EXT_ADV_DATA_SIZE, HCI_MGMT_VAR_LEN }, { add_adv_patterns_monitor_rssi, - MGMT_ADD_ADV_PATTERNS_MONITOR_RSSI_SIZE }, + MGMT_ADD_ADV_PATTERNS_MONITOR_RSSI_SIZE, + HCI_MGMT_VAR_LEN }, { set_mesh, MGMT_SET_MESH_RECEIVER_SIZE, HCI_MGMT_VAR_LEN }, { mesh_features, MGMT_MESH_READ_FEATURES_SIZE }, -- cgit v1.2.3 From bce56405201111807cc8e4f47c6de3e10b17c1ac Mon Sep 17 00:00:00 2001 From: Sungwoo Kim Date: Mon, 20 Mar 2023 21:50:18 -0400 Subject: Bluetooth: HCI: Fix global-out-of-bounds To loop a variable-length array, hci_init_stage_sync(stage) considers that stage[i] is valid as long as stage[i-1].func is valid. Thus, the last element of stage[].func should be intentionally invalid as hci_init0[], le_init2[], and others did. However, amp_init1[] and amp_init2[] have no invalid element, letting hci_init_stage_sync() keep accessing amp_init1[] over its valid range. This patch fixes this by adding {} in the last of amp_init1[] and amp_init2[]. ================================================================== BUG: KASAN: global-out-of-bounds in hci_dev_open_sync ( /v6.2-bzimage/net/bluetooth/hci_sync.c:3154 /v6.2-bzimage/net/bluetooth/hci_sync.c:3343 /v6.2-bzimage/net/bluetooth/hci_sync.c:4418 /v6.2-bzimage/net/bluetooth/hci_sync.c:4609 /v6.2-bzimage/net/bluetooth/hci_sync.c:4689) Read of size 8 at addr ffffffffaed1ab70 by task kworker/u5:0/1032 CPU: 0 PID: 1032 Comm: kworker/u5:0 Not tainted 6.2.0 #3 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04 Workqueue: hci1 hci_power_on Call Trace: dump_stack_lvl (/v6.2-bzimage/lib/dump_stack.c:107 (discriminator 1)) print_report (/v6.2-bzimage/mm/kasan/report.c:307 /v6.2-bzimage/mm/kasan/report.c:417) ? hci_dev_open_sync (/v6.2-bzimage/net/bluetooth/hci_sync.c:3154 /v6.2-bzimage/net/bluetooth/hci_sync.c:3343 /v6.2-bzimage/net/bluetooth/hci_sync.c:4418 /v6.2-bzimage/net/bluetooth/hci_sync.c:4609 /v6.2-bzimage/net/bluetooth/hci_sync.c:4689) kasan_report (/v6.2-bzimage/mm/kasan/report.c:184 /v6.2-bzimage/mm/kasan/report.c:519) ? hci_dev_open_sync (/v6.2-bzimage/net/bluetooth/hci_sync.c:3154 /v6.2-bzimage/net/bluetooth/hci_sync.c:3343 /v6.2-bzimage/net/bluetooth/hci_sync.c:4418 /v6.2-bzimage/net/bluetooth/hci_sync.c:4609 /v6.2-bzimage/net/bluetooth/hci_sync.c:4689) hci_dev_open_sync (/v6.2-bzimage/net/bluetooth/hci_sync.c:3154 /v6.2-bzimage/net/bluetooth/hci_sync.c:3343 /v6.2-bzimage/net/bluetooth/hci_sync.c:4418 /v6.2-bzimage/net/bluetooth/hci_sync.c:4609 /v6.2-bzimage/net/bluetooth/hci_sync.c:4689) ? __pfx_hci_dev_open_sync (/v6.2-bzimage/net/bluetooth/hci_sync.c:4635) ? mutex_lock (/v6.2-bzimage/./arch/x86/include/asm/atomic64_64.h:190 /v6.2-bzimage/./include/linux/atomic/atomic-long.h:443 /v6.2-bzimage/./include/linux/atomic/atomic-instrumented.h:1781 /v6.2-bzimage/kernel/locking/mutex.c:171 /v6.2-bzimage/kernel/locking/mutex.c:285) ? __pfx_mutex_lock (/v6.2-bzimage/kernel/locking/mutex.c:282) hci_power_on (/v6.2-bzimage/net/bluetooth/hci_core.c:485 /v6.2-bzimage/net/bluetooth/hci_core.c:984) ? __pfx_hci_power_on (/v6.2-bzimage/net/bluetooth/hci_core.c:969) ? read_word_at_a_time (/v6.2-bzimage/./include/asm-generic/rwonce.h:85) ? strscpy (/v6.2-bzimage/./arch/x86/include/asm/word-at-a-time.h:62 /v6.2-bzimage/lib/string.c:161) process_one_work (/v6.2-bzimage/kernel/workqueue.c:2294) worker_thread (/v6.2-bzimage/./include/linux/list.h:292 /v6.2-bzimage/kernel/workqueue.c:2437) ? __pfx_worker_thread (/v6.2-bzimage/kernel/workqueue.c:2379) kthread (/v6.2-bzimage/kernel/kthread.c:376) ? __pfx_kthread (/v6.2-bzimage/kernel/kthread.c:331) ret_from_fork (/v6.2-bzimage/arch/x86/entry/entry_64.S:314) The buggy address belongs to the variable: amp_init1+0x30/0x60 The buggy address belongs to the physical page: page:000000003a157ec6 refcount:1 mapcount:0 mapping:0000000000000000 ia flags: 0x200000000001000(reserved|node=0|zone=2) raw: 0200000000001000 ffffea0005054688 ffffea0005054688 000000000000000 raw: 0000000000000000 0000000000000000 00000001ffffffff 000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffffffffaed1aa00: f9 f9 f9 f9 00 00 00 00 f9 f9 f9 f9 00 00 00 00 ffffffffaed1aa80: 00 00 00 00 f9 f9 f9 f9 00 00 00 00 00 00 00 00 >ffffffffaed1ab00: 00 f9 f9 f9 f9 f9 f9 f9 00 00 00 00 00 00 f9 f9 ^ ffffffffaed1ab80: f9 f9 f9 f9 00 00 00 00 f9 f9 f9 f9 00 00 00 f9 ffffffffaed1ac00: f9 f9 f9 f9 00 06 f9 f9 f9 f9 f9 f9 00 00 02 f9 This bug is found by FuzzBT, a modified version of Syzkaller. Other contributors for this bug are Ruoyu Wu and Peng Hui. Fixes: d0b137062b2d ("Bluetooth: hci_sync: Rework init stages") Signed-off-by: Sungwoo Kim Reviewed-by: Simon Horman Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 5b8dc8fb2e27..5a6aa1627791 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -3360,6 +3360,7 @@ static const struct hci_init_stage amp_init1[] = { HCI_INIT(hci_read_flow_control_mode_sync), /* HCI_OP_READ_LOCATION_DATA */ HCI_INIT(hci_read_location_data_sync), + {} }; static int hci_init1_sync(struct hci_dev *hdev) @@ -3394,6 +3395,7 @@ static int hci_init1_sync(struct hci_dev *hdev) static const struct hci_init_stage amp_init2[] = { /* HCI_OP_READ_LOCAL_FEATURES */ HCI_INIT(hci_read_local_features_sync), + {} }; /* Read Buffer Size (ACL mtu, max pkt, etc.) */ -- cgit v1.2.3 From 2df9bf4d04d2eb458386a40bac01be22c8808450 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 22 Mar 2023 19:35:41 -0400 Subject: ipv6: prevent router_solicitations for team port The issue fixed for bonding in commit c2edacf80e15 ("bonding / ipv6: no addrconf for slaves separately from master") also exists in team driver. However, we can't just disable ipv6 addrconf for team ports, as 'teamd' will need it when nsns_ping watch is used in the user space. Instead of preventing ipv6 addrconf, this patch only prevents RS packets for team ports, as it did in commit b52e1cce31ca ("ipv6: Don't send rs packets to the interface of ARPHRD_TUNNEL"). Note that we do not prevent DAD packets, to avoid the changes getting intricate / hacky. Also, usually sysctl dad_transmits is set to 1 and only 1 DAD packet will be sent, and by now no libteam user complains about DAD packets on team ports, unlike RS packets. Signed-off-by: Xin Long Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 31e0097878c5..3797917237d0 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -4223,7 +4223,8 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp, bool bump_id, ipv6_accept_ra(ifp->idev) && ifp->idev->cnf.rtr_solicits != 0 && (dev->flags & IFF_LOOPBACK) == 0 && - (dev->type != ARPHRD_TUNNEL); + (dev->type != ARPHRD_TUNNEL) && + !netif_is_team_port(dev); read_unlock_bh(&ifp->idev->lock); /* While dad is in progress mld report's source address is in6_addrany. -- cgit v1.2.3 From 4170f0ef582c1ff4557276b98fb5fad85c782c1f Mon Sep 17 00:00:00 2001 From: Taichi Nishimura Date: Thu, 23 Mar 2023 14:27:13 +0900 Subject: fix typos in net/sched/* files This patch fixes typos in net/sched/* files. Signed-off-by: Taichi Nishimura Reviewed-by: Randy Dunlap Signed-off-by: David S. Miller --- net/sched/cls_flower.c | 2 +- net/sched/em_meta.c | 2 +- net/sched/sch_pie.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 475fe222a855..cc49256d5318 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -1057,7 +1057,7 @@ static void fl_set_key_pppoe(struct nlattr **tb, * because ETH_P_PPP_SES was stored in basic.n_proto * which might get overwritten by ppp_proto * or might be set to 0, the role of key_val::type - * is simmilar to vlan_key::tpid + * is similar to vlan_key::tpid */ key_val->type = htons(ETH_P_PPP_SES); key_mask->type = cpu_to_be16(~0); diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index 49bae3d5006b..af85a73c4c54 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -44,7 +44,7 @@ * be provided for non-numeric types. * * Additionally, type dependent modifiers such as shift operators - * or mask may be applied to extend the functionaliy. As of now, + * or mask may be applied to extend the functionality. As of now, * the variable length type supports shifting the byte string to * the right, eating up any number of octets and thus supporting * wildcard interface name comparisons such as "ppp%" matching diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c index 265c238047a4..2152a56d73f8 100644 --- a/net/sched/sch_pie.c +++ b/net/sched/sch_pie.c @@ -319,7 +319,7 @@ void pie_calculate_probability(struct pie_params *params, struct pie_vars *vars, } /* If qdelay is zero and backlog is not, it means backlog is very small, - * so we do not update probabilty in this round. + * so we do not update probability in this round. */ if (qdelay == 0 && backlog != 0) update_prob = false; -- cgit v1.2.3 From f102424befd3751386f3e2f2c70c5a1948248622 Mon Sep 17 00:00:00 2001 From: Bagas Sanjaya Date: Thu, 23 Mar 2023 16:24:54 +0700 Subject: wifi: mac80211: use bullet list for amsdu_mesh_control formats list Commit fe4a6d2db3ba ("wifi: mac80211: implement support for yet another mesh A-MSDU format") expands amsdu_mesh_control list to multi-line list. However, the expansion triggers Sphinx warning: Documentation/driver-api/80211/mac80211-advanced:214: ./net/mac80211/sta_info.h:628: WARNING: Unexpected indentation. Use bullet list instead to fix the warning. Link: https://lore.kernel.org/linux-next/20230323141548.659479ef@canb.auug.org.au/ Fixes: fe4a6d2db3bad4 ("wifi: mac80211: implement support for yet another mesh A-MSDU format") Reported-by: Stephen Rothwell Signed-off-by: Bagas Sanjaya Reviewed-by: Simon Horman Signed-off-by: Johannes Berg --- net/mac80211/sta_info.h | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index f354d470e174..195b563132d6 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -622,11 +622,13 @@ struct link_sta_info { * taken from HT/VHT capabilities or VHT operating mode notification * @cparams: CoDel parameters for this station. * @reserved_tid: reserved TID (if any, otherwise IEEE80211_TID_UNRESERVED) - * @amsdu_mesh_control: track the mesh A-MSDU format used by the peer - * (-1: not yet known, - * 0: non-mesh A-MSDU length field - * 1: big-endian mesh A-MSDU length field - * 2: little-endian mesh A-MSDU length field) + * @amsdu_mesh_control: track the mesh A-MSDU format used by the peer: + * + * * -1: not yet known + * * 0: non-mesh A-MSDU length field + * * 1: big-endian mesh A-MSDU length field + * * 2: little-endian mesh A-MSDU length field + * * @fast_tx: TX fastpath information * @fast_rx: RX fastpath information * @tdls_chandef: a TDLS peer can have a wider chandef that is compatible to -- cgit v1.2.3 From dbbb27e183b1568d5a907ace1cd144b0709ea52a Mon Sep 17 00:00:00 2001 From: Aloka Dixit Date: Thu, 23 Mar 2023 04:38:00 -0700 Subject: cfg80211: support RNR for EMA AP As per IEEE Std 802.11ax-2021, 11.1.3.8.3 Discovery of a nontransmitted BSSID profile, an EMA AP that transmits a Beacon frame carrying a partial list of nontransmitted BSSID profiles should include in the frame a Reduced Neighbor Report element carrying information for at least the nontransmitted BSSIDs that are not present in the Multiple BSSID element carried in that frame. Add new nested attribute NL80211_ATTR_EMA_RNR_ELEMS to support the above. Number of RNR elements must be more than or equal to the number of MBSSID elements. This attribute can be used only when EMA is enabled. Userspace is responsible for splitting the RNR into multiple elements such that each element excludes the non-transmitting profiles already included in the MBSSID element (%NL80211_ATTR_MBSSID_ELEMS) at the same index. Each EMA beacon will be generated by adding MBSSID and RNR elements at the same index. If the userspace provides more RNR elements than the number of MBSSID elements then these will be added in every EMA beacon. Signed-off-by: Aloka Dixit Link: https://lore.kernel.org/r/20230323113801.6903-2-quic_alokad@quicinc.com [Johannes: validate elements] Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 19 +++++++++++ include/uapi/linux/nl80211.h | 13 ++++++++ net/wireless/nl80211.c | 79 ++++++++++++++++++++++++++++++++++++++++---- 3 files changed, 104 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 86cb048dc924..3cf236520288 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1178,6 +1178,23 @@ struct cfg80211_mbssid_elems { } elem[]; }; +/** + * struct cfg80211_rnr_elems - Reduced neighbor report (RNR) elements + * + * @cnt: Number of elements in array %elems. + * + * @elem: Array of RNR element(s) to be added into Beacon frames. + * @elem.data: Data for RNR elements. + * @elem.len: Length of data. + */ +struct cfg80211_rnr_elems { + u8 cnt; + struct { + const u8 *data; + size_t len; + } elem[]; +}; + /** * struct cfg80211_beacon_data - beacon data * @link_id: the link ID for the AP MLD link sending this beacon @@ -1198,6 +1215,7 @@ struct cfg80211_mbssid_elems { * @probe_resp_len: length of probe response template (@probe_resp) * @probe_resp: probe response template (AP mode only) * @mbssid_ies: multiple BSSID elements + * @rnr_ies: reduced neighbor report elements * @ftm_responder: enable FTM responder functionality; -1 for no change * (which also implies no change in LCI/civic location data) * @lci: Measurement Report element content, starting with Measurement Token @@ -1221,6 +1239,7 @@ struct cfg80211_beacon_data { const u8 *lci; const u8 *civicloc; struct cfg80211_mbssid_elems *mbssid_ies; + struct cfg80211_rnr_elems *rnr_ies; s8 ftm_responder; size_t head_len, tail_len; diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 14e958a32b84..cf4fb981e131 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -2794,6 +2794,17 @@ enum nl80211_commands { * @NL80211_ATTR_HW_TIMESTAMP_ENABLED: Indicates whether HW timestamping should * be enabled or not (flag attribute). * + * @NL80211_ATTR_EMA_RNR_ELEMS: Optional nested attribute for + * reduced neighbor report (RNR) elements. This attribute can be used + * only when NL80211_MBSSID_CONFIG_ATTR_EMA is enabled. + * Userspace is responsible for splitting the RNR into multiple + * elements such that each element excludes the non-transmitting + * profiles already included in the MBSSID element + * (%NL80211_ATTR_MBSSID_ELEMS) at the same index. Each EMA beacon + * will be generated by adding MBSSID and RNR elements at the same + * index. If the userspace includes more RNR elements than number of + * MBSSID elements then these will be added in every EMA beacon. + * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -3328,6 +3339,8 @@ enum nl80211_attrs { NL80211_ATTR_MAX_HW_TIMESTAMP_PEERS, NL80211_ATTR_HW_TIMESTAMP_ENABLED, + NL80211_ATTR_EMA_RNR_ELEMS, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 0a31b1d2845d..80a20d69f285 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -809,6 +809,7 @@ static const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = { [NL80211_ATTR_MAX_HW_TIMESTAMP_PEERS] = { .type = NLA_U16 }, [NL80211_ATTR_HW_TIMESTAMP_ENABLED] = { .type = NLA_FLAG }, + [NL80211_ATTR_EMA_RNR_ELEMS] = { .type = NLA_NESTED }, }; /* policy for the key attributes */ @@ -5425,6 +5426,38 @@ nl80211_parse_mbssid_elems(struct wiphy *wiphy, struct nlattr *attrs) return elems; } +static struct cfg80211_rnr_elems * +nl80211_parse_rnr_elems(struct wiphy *wiphy, struct nlattr *attrs, + struct netlink_ext_ack *extack) +{ + struct nlattr *nl_elems; + struct cfg80211_rnr_elems *elems; + int rem_elems; + u8 i = 0, num_elems = 0; + + nla_for_each_nested(nl_elems, attrs, rem_elems) { + int ret; + + ret = validate_ie_attr(nl_elems, extack); + if (ret) + return ERR_PTR(ret); + + num_elems++; + } + + elems = kzalloc(struct_size(elems, elem, num_elems), GFP_KERNEL); + if (!elems) + return ERR_PTR(-ENOMEM); + + nla_for_each_nested(nl_elems, attrs, rem_elems) { + elems->elem[i].data = nla_data(nl_elems); + elems->elem[i].len = nla_len(nl_elems); + i++; + } + elems->cnt = num_elems; + return elems; +} + static int nl80211_parse_he_bss_color(struct nlattr *attrs, struct cfg80211_he_bss_color *he_bss_color) { @@ -5451,7 +5484,8 @@ static int nl80211_parse_he_bss_color(struct nlattr *attrs, static int nl80211_parse_beacon(struct cfg80211_registered_device *rdev, struct nlattr *attrs[], - struct cfg80211_beacon_data *bcn) + struct cfg80211_beacon_data *bcn, + struct netlink_ext_ack *extack) { bool haveinfo = false; int err; @@ -5548,6 +5582,21 @@ static int nl80211_parse_beacon(struct cfg80211_registered_device *rdev, return PTR_ERR(mbssid); bcn->mbssid_ies = mbssid; + + if (bcn->mbssid_ies && attrs[NL80211_ATTR_EMA_RNR_ELEMS]) { + struct cfg80211_rnr_elems *rnr = + nl80211_parse_rnr_elems(&rdev->wiphy, + attrs[NL80211_ATTR_EMA_RNR_ELEMS], + extack); + + if (IS_ERR(rnr)) + return PTR_ERR(rnr); + + if (rnr && rnr->cnt < bcn->mbssid_ies->cnt) + return -EINVAL; + + bcn->rnr_ies = rnr; + } } return 0; @@ -5866,7 +5915,8 @@ static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info) if (!params) return -ENOMEM; - err = nl80211_parse_beacon(rdev, info->attrs, ¶ms->beacon); + err = nl80211_parse_beacon(rdev, info->attrs, ¶ms->beacon, + info->extack); if (err) goto out; @@ -6096,6 +6146,11 @@ static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info) goto out_unlock; } + if (!params->mbssid_config.ema && params->beacon.rnr_ies) { + err = -EINVAL; + goto out_unlock; + } + err = nl80211_calculate_ap_params(params); if (err) goto out_unlock; @@ -6137,6 +6192,7 @@ out: params->mbssid_config.tx_wdev->netdev && params->mbssid_config.tx_wdev->netdev != dev) dev_put(params->mbssid_config.tx_wdev->netdev); + kfree(params->beacon.rnr_ies); kfree(params); return err; @@ -6161,7 +6217,7 @@ static int nl80211_set_beacon(struct sk_buff *skb, struct genl_info *info) if (!wdev->links[link_id].ap.beacon_interval) return -EINVAL; - err = nl80211_parse_beacon(rdev, info->attrs, ¶ms); + err = nl80211_parse_beacon(rdev, info->attrs, ¶ms, info->extack); if (err) goto out; @@ -6171,6 +6227,7 @@ static int nl80211_set_beacon(struct sk_buff *skb, struct genl_info *info) out: kfree(params.mbssid_ies); + kfree(params.rnr_ies); return err; } @@ -10030,7 +10087,8 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info) if (!need_new_beacon) goto skip_beacons; - err = nl80211_parse_beacon(rdev, info->attrs, ¶ms.beacon_after); + err = nl80211_parse_beacon(rdev, info->attrs, ¶ms.beacon_after, + info->extack); if (err) goto free; @@ -10047,7 +10105,8 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info) if (err) goto free; - err = nl80211_parse_beacon(rdev, csa_attrs, ¶ms.beacon_csa); + err = nl80211_parse_beacon(rdev, csa_attrs, ¶ms.beacon_csa, + info->extack); if (err) goto free; @@ -10167,6 +10226,8 @@ skip_beacons: free: kfree(params.beacon_after.mbssid_ies); kfree(params.beacon_csa.mbssid_ies); + kfree(params.beacon_after.rnr_ies); + kfree(params.beacon_csa.rnr_ies); kfree(csa_attrs); return err; } @@ -15882,7 +15943,8 @@ static int nl80211_color_change(struct sk_buff *skb, struct genl_info *info) params.count = nla_get_u8(info->attrs[NL80211_ATTR_COLOR_CHANGE_COUNT]); params.color = nla_get_u8(info->attrs[NL80211_ATTR_COLOR_CHANGE_COLOR]); - err = nl80211_parse_beacon(rdev, info->attrs, ¶ms.beacon_next); + err = nl80211_parse_beacon(rdev, info->attrs, ¶ms.beacon_next, + info->extack); if (err) return err; @@ -15896,7 +15958,8 @@ static int nl80211_color_change(struct sk_buff *skb, struct genl_info *info) if (err) goto out; - err = nl80211_parse_beacon(rdev, tb, ¶ms.beacon_color_change); + err = nl80211_parse_beacon(rdev, tb, ¶ms.beacon_color_change, + info->extack); if (err) goto out; @@ -15952,6 +16015,8 @@ static int nl80211_color_change(struct sk_buff *skb, struct genl_info *info) out: kfree(params.beacon_next.mbssid_ies); kfree(params.beacon_color_change.mbssid_ies); + kfree(params.beacon_next.rnr_ies); + kfree(params.beacon_color_change.rnr_ies); kfree(tb); return err; } -- cgit v1.2.3 From 68b9bea267bfc1259e195dcac1bf69db0c0c28da Mon Sep 17 00:00:00 2001 From: Aloka Dixit Date: Thu, 23 Mar 2023 04:38:01 -0700 Subject: mac80211: support RNR for EMA AP Generate EMA beacons, each including MBSSID and RNR elements at a given index. If number of stored RNR elements is more than the number of MBSSID elements then add those in every EMA beacon. Signed-off-by: Aloka Dixit Link: https://lore.kernel.org/r/20230323113801.6903-3-quic_alokad@quicinc.com Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 63 ++++++++++++++++++++++++++++++++++++++++++---- net/mac80211/ieee80211_i.h | 21 +++++++++++++--- net/mac80211/tx.c | 10 ++++++++ 3 files changed, 86 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index db5fa334b801..ebbd13d0c29a 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1084,6 +1084,23 @@ ieee80211_copy_mbssid_beacon(u8 *pos, struct cfg80211_mbssid_elems *dst, return offset; } +static int +ieee80211_copy_rnr_beacon(u8 *pos, struct cfg80211_rnr_elems *dst, + struct cfg80211_rnr_elems *src) +{ + int i, offset = 0; + + for (i = 0; i < src->cnt; i++) { + memcpy(pos + offset, src->elem[i].data, src->elem[i].len); + dst->elem[i].len = src->elem[i].len; + dst->elem[i].data = pos + offset; + offset += dst->elem[i].len; + } + dst->cnt = src->cnt; + + return offset; +} + static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata, struct ieee80211_link_data *link, struct cfg80211_beacon_data *params, @@ -1091,6 +1108,7 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata, const struct ieee80211_color_change_settings *cca) { struct cfg80211_mbssid_elems *mbssid = NULL; + struct cfg80211_rnr_elems *rnr = NULL; struct beacon_data *new, *old; int new_head_len, new_tail_len; int size, err; @@ -1122,11 +1140,21 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata, if (params->mbssid_ies) { mbssid = params->mbssid_ies; size += struct_size(new->mbssid_ies, elem, mbssid->cnt); - size += ieee80211_get_mbssid_beacon_len(mbssid, mbssid->cnt); + if (params->rnr_ies) { + rnr = params->rnr_ies; + size += struct_size(new->rnr_ies, elem, rnr->cnt); + } + size += ieee80211_get_mbssid_beacon_len(mbssid, rnr, + mbssid->cnt); } else if (old && old->mbssid_ies) { mbssid = old->mbssid_ies; size += struct_size(new->mbssid_ies, elem, mbssid->cnt); - size += ieee80211_get_mbssid_beacon_len(mbssid, mbssid->cnt); + if (old && old->rnr_ies) { + rnr = old->rnr_ies; + size += struct_size(new->rnr_ies, elem, rnr->cnt); + } + size += ieee80211_get_mbssid_beacon_len(mbssid, rnr, + mbssid->cnt); } new = kzalloc(size, GFP_KERNEL); @@ -1137,7 +1165,7 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata, /* * pointers go into the block we allocated, - * memory is | beacon_data | head | tail | mbssid_ies + * memory is | beacon_data | head | tail | mbssid_ies | rnr_ies */ new->head = ((u8 *) new) + sizeof(*new); new->tail = new->head + new_head_len; @@ -1149,7 +1177,13 @@ static int ieee80211_assign_beacon(struct ieee80211_sub_if_data *sdata, new->mbssid_ies = (void *)pos; pos += struct_size(new->mbssid_ies, elem, mbssid->cnt); - ieee80211_copy_mbssid_beacon(pos, new->mbssid_ies, mbssid); + pos += ieee80211_copy_mbssid_beacon(pos, new->mbssid_ies, + mbssid); + if (rnr) { + new->rnr_ies = (void *)pos; + pos += struct_size(new->rnr_ies, elem, rnr->cnt); + ieee80211_copy_rnr_beacon(pos, new->rnr_ies, rnr); + } /* update bssid_indicator */ link_conf->bssid_indicator = ilog2(__roundup_pow_of_two(mbssid->cnt + 1)); @@ -1507,6 +1541,7 @@ static void ieee80211_free_next_beacon(struct ieee80211_link_data *link) return; kfree(link->u.ap.next_beacon->mbssid_ies); + kfree(link->u.ap.next_beacon->rnr_ies); kfree(link->u.ap.next_beacon); link->u.ap.next_beacon = NULL; } @@ -3410,6 +3445,7 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon) if (beacon->mbssid_ies) len += ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies, + beacon->rnr_ies, beacon->mbssid_ies->cnt); new_beacon = kzalloc(sizeof(*new_beacon) + len, GFP_KERNEL); @@ -3425,6 +3461,18 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon) kfree(new_beacon); return NULL; } + + if (beacon->rnr_ies && beacon->rnr_ies->cnt) { + new_beacon->rnr_ies = + kzalloc(struct_size(new_beacon->rnr_ies, + elem, beacon->rnr_ies->cnt), + GFP_KERNEL); + if (!new_beacon->rnr_ies) { + kfree(new_beacon->mbssid_ies); + kfree(new_beacon); + return NULL; + } + } } pos = (u8 *)(new_beacon + 1); @@ -3464,10 +3512,15 @@ cfg80211_beacon_dup(struct cfg80211_beacon_data *beacon) memcpy(pos, beacon->probe_resp, beacon->probe_resp_len); pos += beacon->probe_resp_len; } - if (beacon->mbssid_ies && beacon->mbssid_ies->cnt) + if (beacon->mbssid_ies && beacon->mbssid_ies->cnt) { pos += ieee80211_copy_mbssid_beacon(pos, new_beacon->mbssid_ies, beacon->mbssid_ies); + if (beacon->rnr_ies && beacon->rnr_ies->cnt) + pos += ieee80211_copy_rnr_beacon(pos, + new_beacon->rnr_ies, + beacon->rnr_ies); + } /* might copy -1, meaning no changes requested */ new_beacon->ftm_responder = beacon->ftm_responder; diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 84d10e993eca..85ecbf57d64e 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -270,6 +270,7 @@ struct beacon_data { u16 cntdwn_counter_offsets[IEEE80211_MAX_CNTDWN_COUNTERS_NUM]; u8 cntdwn_current_counter; struct cfg80211_mbssid_elems *mbssid_ies; + struct cfg80211_rnr_elems *rnr_ies; struct rcu_head rcu_head; }; @@ -1186,20 +1187,34 @@ ieee80211_vif_get_shift(struct ieee80211_vif *vif) } static inline int -ieee80211_get_mbssid_beacon_len(struct cfg80211_mbssid_elems *elems, u8 i) +ieee80211_get_mbssid_beacon_len(struct cfg80211_mbssid_elems *elems, + struct cfg80211_rnr_elems *rnr_elems, + u8 i) { int len = 0; if (!elems || !elems->cnt || i > elems->cnt) return 0; - if (i < elems->cnt) - return elems->elem[i].len; + if (i < elems->cnt) { + len = elems->elem[i].len; + if (rnr_elems) { + len += rnr_elems->elem[i].len; + for (i = elems->cnt; i < rnr_elems->cnt; i++) + len += rnr_elems->elem[i].len; + } + return len; + } /* i == elems->cnt, calculate total length of all MBSSID elements */ for (i = 0; i < elems->cnt; i++) len += elems->elem[i].len; + if (rnr_elems) { + for (i = 0; i < rnr_elems->cnt; i++) + len += rnr_elems->elem[i].len; + } + return len; } diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 139eec6c64da..dfe6b9c9b29e 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -5222,6 +5222,15 @@ ieee80211_beacon_add_mbssid(struct sk_buff *skb, struct beacon_data *beacon, if (i < beacon->mbssid_ies->cnt) { skb_put_data(skb, beacon->mbssid_ies->elem[i].data, beacon->mbssid_ies->elem[i].len); + + if (beacon->rnr_ies && beacon->rnr_ies->cnt) { + skb_put_data(skb, beacon->rnr_ies->elem[i].data, + beacon->rnr_ies->elem[i].len); + + for (i = beacon->mbssid_ies->cnt; i < beacon->rnr_ies->cnt; i++) + skb_put_data(skb, beacon->rnr_ies->elem[i].data, + beacon->rnr_ies->elem[i].len); + } return; } @@ -5259,6 +5268,7 @@ ieee80211_beacon_get_ap(struct ieee80211_hw *hw, * tail length, maximum TIM length and multiple BSSID length */ mbssid_len = ieee80211_get_mbssid_beacon_len(beacon->mbssid_ies, + beacon->rnr_ies, ema_index); skb = dev_alloc_skb(local->tx_headroom + beacon->head_len + -- cgit v1.2.3 From 5097f84437c9bd50b2c65b5f85395c34b2d545db Mon Sep 17 00:00:00 2001 From: Jaewan Kim Date: Wed, 22 Mar 2023 13:16:34 +0000 Subject: wifi: nl80211: make nl80211_send_chandef non-static Expose nl80211_send_chandef functionality for mac80211_hwsim or vendor netlink can use it where needed. Signed-off-by: Jaewan Kim Reviewed-by: Michal Kubiak Link: https://lore.kernel.org/r/20230322131637.2633968-3-jaewan@google.com Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 9 +++++++++ net/wireless/nl80211.c | 4 ++-- 2 files changed, 11 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 3cf236520288..9e04f69712b1 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -950,6 +950,15 @@ int cfg80211_chandef_dfs_required(struct wiphy *wiphy, const struct cfg80211_chan_def *chandef, enum nl80211_iftype iftype); +/** + * nl80211_send_chandef - sends the channel definition. + * @msg: the msg to send channel definition + * @chandef: the channel definition to check + * + * Returns: 0 if sent the channel definition to msg, < 0 on error + **/ +int nl80211_send_chandef(struct sk_buff *msg, const struct cfg80211_chan_def *chandef); + /** * ieee80211_chanwidth_rate_flags - return rate flags for channel width * @width: the channel width of the channel diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 80a20d69f285..f1cd3d9130dd 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -3765,8 +3765,7 @@ out: return result; } -static int nl80211_send_chandef(struct sk_buff *msg, - const struct cfg80211_chan_def *chandef) +int nl80211_send_chandef(struct sk_buff *msg, const struct cfg80211_chan_def *chandef) { if (WARN_ON(!cfg80211_chandef_valid(chandef))) return -EINVAL; @@ -3797,6 +3796,7 @@ static int nl80211_send_chandef(struct sk_buff *msg, return -ENOBUFS; return 0; } +EXPORT_SYMBOL(nl80211_send_chandef); static int nl80211_send_iface(struct sk_buff *msg, u32 portid, u32 seq, int flags, struct cfg80211_registered_device *rdev, -- cgit v1.2.3 From fb2211a57c110b4ced3cb7f8570bd7246acf2d04 Mon Sep 17 00:00:00 2001 From: David Vernet Date: Sat, 25 Mar 2023 16:31:45 -0500 Subject: bpf: Remove now-unnecessary NULL checks for KF_RELEASE kfuncs Now that we're not invoking kfunc destructors when the kptr in a map was NULL, we no longer require NULL checks in many of our KF_RELEASE kfuncs. This patch removes those NULL checks. Signed-off-by: David Vernet Link: https://lore.kernel.org/r/20230325213144.486885-3-void@manifault.com Signed-off-by: Alexei Starovoitov --- drivers/hid/bpf/hid_bpf_dispatch.c | 3 --- kernel/bpf/cpumask.c | 3 --- kernel/bpf/helpers.c | 6 ------ net/bpf/test_run.c | 3 --- net/netfilter/nf_conntrack_bpf.c | 2 -- 5 files changed, 17 deletions(-) (limited to 'net') diff --git a/drivers/hid/bpf/hid_bpf_dispatch.c b/drivers/hid/bpf/hid_bpf_dispatch.c index 8a034a555d4c..d9ef45fcaeab 100644 --- a/drivers/hid/bpf/hid_bpf_dispatch.c +++ b/drivers/hid/bpf/hid_bpf_dispatch.c @@ -342,9 +342,6 @@ hid_bpf_release_context(struct hid_bpf_ctx *ctx) { struct hid_bpf_ctx_kern *ctx_kern; - if (!ctx) - return; - ctx_kern = container_of(ctx, struct hid_bpf_ctx_kern, ctx); kfree(ctx_kern); diff --git a/kernel/bpf/cpumask.c b/kernel/bpf/cpumask.c index db9da2194c1a..e991af7dc13c 100644 --- a/kernel/bpf/cpumask.c +++ b/kernel/bpf/cpumask.c @@ -102,9 +102,6 @@ static void cpumask_free_cb(struct rcu_head *head) */ __bpf_kfunc void bpf_cpumask_release(struct bpf_cpumask *cpumask) { - if (!cpumask) - return; - if (refcount_dec_and_test(&cpumask->usage)) call_rcu(&cpumask->rcu, cpumask_free_cb); } diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c index f753676ef652..8980f6859443 100644 --- a/kernel/bpf/helpers.c +++ b/kernel/bpf/helpers.c @@ -2089,9 +2089,6 @@ __bpf_kfunc struct task_struct *bpf_task_kptr_get(struct task_struct **pp) */ __bpf_kfunc void bpf_task_release(struct task_struct *p) { - if (!p) - return; - put_task_struct(p); } @@ -2148,9 +2145,6 @@ __bpf_kfunc struct cgroup *bpf_cgroup_kptr_get(struct cgroup **cgrpp) */ __bpf_kfunc void bpf_cgroup_release(struct cgroup *cgrp) { - if (!cgrp) - return; - cgroup_put(cgrp); } diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 8d6b31209bd6..27587f1c5f36 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -615,9 +615,6 @@ bpf_kfunc_call_memb_acquire(void) __bpf_kfunc void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) { - if (!p) - return; - refcount_dec(&p->cnt); } diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c index cd99e6dc1f35..002e9d24a1e9 100644 --- a/net/netfilter/nf_conntrack_bpf.c +++ b/net/netfilter/nf_conntrack_bpf.c @@ -401,8 +401,6 @@ __bpf_kfunc struct nf_conn *bpf_ct_insert_entry(struct nf_conn___init *nfct_i) */ __bpf_kfunc void bpf_ct_release(struct nf_conn *nfct) { - if (!nfct) - return; nf_ct_put(nfct); } -- cgit v1.2.3 From 6c831c4684124a544f73f7c9b83bc7b2eb0b23d3 Mon Sep 17 00:00:00 2001 From: David Vernet Date: Sat, 25 Mar 2023 16:31:46 -0500 Subject: bpf: Treat KF_RELEASE kfuncs as KF_TRUSTED_ARGS KF_RELEASE kfuncs are not currently treated as having KF_TRUSTED_ARGS, even though they have a superset of the requirements of KF_TRUSTED_ARGS. Like KF_TRUSTED_ARGS, KF_RELEASE kfuncs require a 0-offset argument, and don't allow NULL-able arguments. Unlike KF_TRUSTED_ARGS which require _either_ an argument with ref_obj_id > 0, _or_ (ref->type & BPF_REG_TRUSTED_MODIFIERS) (and no unsafe modifiers allowed), KF_RELEASE only allows for ref_obj_id > 0. Because KF_RELEASE today doesn't automatically imply KF_TRUSTED_ARGS, some of these requirements are enforced in different ways that can make the behavior of the verifier feel unpredictable. For example, a KF_RELEASE kfunc with a NULL-able argument will currently fail in the verifier with a message like, "arg#0 is ptr_or_null_ expected ptr_ or socket" rather than "Possibly NULL pointer passed to trusted arg0". Our intention is the same, but the semantics are different due to implemenetation details that kfunc authors and BPF program writers should not need to care about. Let's make the behavior of the verifier more consistent and intuitive by having KF_RELEASE kfuncs imply the presence of KF_TRUSTED_ARGS. Our eventual goal is to have all kfuncs assume KF_TRUSTED_ARGS by default anyways, so this takes us a step in that direction. Note that it does not make sense to assume KF_TRUSTED_ARGS for all KF_ACQUIRE kfuncs. KF_ACQUIRE kfuncs can have looser semantics than KF_RELEASE, with e.g. KF_RCU | KF_RET_NULL. We may want to have KF_ACQUIRE imply KF_TRUSTED_ARGS _unless_ KF_RCU is specified, but that can be left to another patch set, and there are no such subtleties to address for KF_RELEASE. Signed-off-by: David Vernet Link: https://lore.kernel.org/r/20230325213144.486885-4-void@manifault.com Signed-off-by: Alexei Starovoitov --- Documentation/bpf/kfuncs.rst | 7 ++++--- kernel/bpf/cpumask.c | 2 +- kernel/bpf/verifier.c | 2 +- net/bpf/test_run.c | 6 ++++++ tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c | 4 ++-- tools/testing/selftests/bpf/progs/task_kfunc_failure.c | 6 +++--- tools/testing/selftests/bpf/verifier/calls.c | 10 +++++++--- tools/testing/selftests/bpf/verifier/ref_tracking.c | 6 +++--- 8 files changed, 27 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst index 69eccf6f98ef..bf1b85941452 100644 --- a/Documentation/bpf/kfuncs.rst +++ b/Documentation/bpf/kfuncs.rst @@ -179,9 +179,10 @@ both are orthogonal to each other. --------------------- The KF_RELEASE flag is used to indicate that the kfunc releases the pointer -passed in to it. There can be only one referenced pointer that can be passed in. -All copies of the pointer being released are invalidated as a result of invoking -kfunc with this flag. +passed in to it. There can be only one referenced pointer that can be passed +in. All copies of the pointer being released are invalidated as a result of +invoking kfunc with this flag. KF_RELEASE kfuncs automatically receive the +protection afforded by the KF_TRUSTED_ARGS flag described below. 2.4.4 KF_KPTR_GET flag ---------------------- diff --git a/kernel/bpf/cpumask.c b/kernel/bpf/cpumask.c index e991af7dc13c..7efdf5d770ca 100644 --- a/kernel/bpf/cpumask.c +++ b/kernel/bpf/cpumask.c @@ -402,7 +402,7 @@ __diag_pop(); BTF_SET8_START(cpumask_kfunc_btf_ids) BTF_ID_FLAGS(func, bpf_cpumask_create, KF_ACQUIRE | KF_RET_NULL) -BTF_ID_FLAGS(func, bpf_cpumask_release, KF_RELEASE | KF_TRUSTED_ARGS) +BTF_ID_FLAGS(func, bpf_cpumask_release, KF_RELEASE) BTF_ID_FLAGS(func, bpf_cpumask_acquire, KF_ACQUIRE | KF_TRUSTED_ARGS) BTF_ID_FLAGS(func, bpf_cpumask_first, KF_RCU) BTF_ID_FLAGS(func, bpf_cpumask_first_zero, KF_RCU) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 64f06f6e16bf..20eb2015842f 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -9307,7 +9307,7 @@ static bool is_kfunc_release(struct bpf_kfunc_call_arg_meta *meta) static bool is_kfunc_trusted_args(struct bpf_kfunc_call_arg_meta *meta) { - return meta->kfunc_flags & KF_TRUSTED_ARGS; + return (meta->kfunc_flags & KF_TRUSTED_ARGS) || is_kfunc_release(meta); } static bool is_kfunc_sleepable(struct bpf_kfunc_call_arg_meta *meta) diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 27587f1c5f36..f1652f5fbd2e 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -606,6 +606,11 @@ bpf_kfunc_call_test_acquire(unsigned long *scalar_ptr) return &prog_test_struct; } +__bpf_kfunc void bpf_kfunc_call_test_offset(struct prog_test_ref_kfunc *p) +{ + WARN_ON_ONCE(1); +} + __bpf_kfunc struct prog_test_member * bpf_kfunc_call_memb_acquire(void) { @@ -800,6 +805,7 @@ BTF_ID_FLAGS(func, bpf_kfunc_call_test_mem_len_fail2) BTF_ID_FLAGS(func, bpf_kfunc_call_test_ref, KF_TRUSTED_ARGS | KF_RCU) BTF_ID_FLAGS(func, bpf_kfunc_call_test_destructive, KF_DESTRUCTIVE) BTF_ID_FLAGS(func, bpf_kfunc_call_test_static_unused_arg) +BTF_ID_FLAGS(func, bpf_kfunc_call_test_offset) BTF_SET8_END(test_sk_check_kfunc_ids) static void *bpf_test_init(const union bpf_attr *kattr, u32 user_size, diff --git a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c index 807fb0ac41e9..48b2034cadb3 100644 --- a/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c +++ b/tools/testing/selftests/bpf/progs/cgrp_kfunc_failure.c @@ -206,7 +206,7 @@ int BPF_PROG(cgrp_kfunc_get_unreleased, struct cgroup *cgrp, const char *path) } SEC("tp_btf/cgroup_mkdir") -__failure __msg("expects refcounted") +__failure __msg("Possibly NULL pointer passed to trusted arg0") int BPF_PROG(cgrp_kfunc_release_untrusted, struct cgroup *cgrp, const char *path) { struct __cgrps_kfunc_map_value *v; @@ -234,7 +234,7 @@ int BPF_PROG(cgrp_kfunc_release_fp, struct cgroup *cgrp, const char *path) } SEC("tp_btf/cgroup_mkdir") -__failure __msg("arg#0 is ptr_or_null_ expected ptr_ or socket") +__failure __msg("Possibly NULL pointer passed to trusted arg0") int BPF_PROG(cgrp_kfunc_release_null, struct cgroup *cgrp, const char *path) { struct __cgrps_kfunc_map_value local, *v; diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c index 27994d6b2914..2c374a7ffece 100644 --- a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c +++ b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c @@ -206,7 +206,7 @@ int BPF_PROG(task_kfunc_get_unreleased, struct task_struct *task, u64 clone_flag } SEC("tp_btf/task_newtask") -__failure __msg("arg#0 is untrusted_ptr_or_null_ expected ptr_ or socket") +__failure __msg("Possibly NULL pointer passed to trusted arg0") int BPF_PROG(task_kfunc_release_untrusted, struct task_struct *task, u64 clone_flags) { struct __tasks_kfunc_map_value *v; @@ -234,7 +234,7 @@ int BPF_PROG(task_kfunc_release_fp, struct task_struct *task, u64 clone_flags) } SEC("tp_btf/task_newtask") -__failure __msg("arg#0 is ptr_or_null_ expected ptr_ or socket") +__failure __msg("Possibly NULL pointer passed to trusted arg0") int BPF_PROG(task_kfunc_release_null, struct task_struct *task, u64 clone_flags) { struct __tasks_kfunc_map_value local, *v; @@ -277,7 +277,7 @@ int BPF_PROG(task_kfunc_release_unacquired, struct task_struct *task, u64 clone_ } SEC("tp_btf/task_newtask") -__failure __msg("arg#0 is ptr_or_null_ expected ptr_ or socket") +__failure __msg("Possibly NULL pointer passed to trusted arg0") int BPF_PROG(task_kfunc_from_pid_no_null_check, struct task_struct *task, u64 clone_flags) { struct task_struct *acquired; diff --git a/tools/testing/selftests/bpf/verifier/calls.c b/tools/testing/selftests/bpf/verifier/calls.c index 5702fc9761ef..1bdf2b43e49e 100644 --- a/tools/testing/selftests/bpf/verifier/calls.c +++ b/tools/testing/selftests/bpf/verifier/calls.c @@ -109,7 +109,7 @@ }, .prog_type = BPF_PROG_TYPE_SCHED_CLS, .result = REJECT, - .errstr = "arg#0 is ptr_or_null_ expected ptr_ or socket", + .errstr = "Possibly NULL pointer passed to trusted arg0", .fixup_kfunc_btf_id = { { "bpf_kfunc_call_test_acquire", 3 }, { "bpf_kfunc_call_test_release", 5 }, @@ -165,19 +165,23 @@ BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8), BPF_ST_MEM(BPF_DW, BPF_REG_1, 0, 0), BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0), + BPF_MOV64_REG(BPF_REG_2, BPF_REG_0), BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1), BPF_EXIT_INSN(), BPF_MOV64_REG(BPF_REG_1, BPF_REG_0), - BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, 16), BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -4), BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0), BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_MOV64_REG(BPF_REG_1, BPF_REG_2), + BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0), + BPF_MOV64_IMM(BPF_REG_0, 0), BPF_EXIT_INSN(), }, .prog_type = BPF_PROG_TYPE_SCHED_CLS, .fixup_kfunc_btf_id = { { "bpf_kfunc_call_test_acquire", 3 }, - { "bpf_kfunc_call_test_release", 9 }, + { "bpf_kfunc_call_test_offset", 9 }, + { "bpf_kfunc_call_test_release", 12 }, }, .result_unpriv = REJECT, .result = REJECT, diff --git a/tools/testing/selftests/bpf/verifier/ref_tracking.c b/tools/testing/selftests/bpf/verifier/ref_tracking.c index 9540164712b7..5a2e154dd1e0 100644 --- a/tools/testing/selftests/bpf/verifier/ref_tracking.c +++ b/tools/testing/selftests/bpf/verifier/ref_tracking.c @@ -142,7 +142,7 @@ .kfunc = "bpf", .expected_attach_type = BPF_LSM_MAC, .flags = BPF_F_SLEEPABLE, - .errstr = "arg#0 is ptr_or_null_ expected ptr_ or socket", + .errstr = "Possibly NULL pointer passed to trusted arg0", .fixup_kfunc_btf_id = { { "bpf_lookup_user_key", 2 }, { "bpf_key_put", 4 }, @@ -163,7 +163,7 @@ .kfunc = "bpf", .expected_attach_type = BPF_LSM_MAC, .flags = BPF_F_SLEEPABLE, - .errstr = "arg#0 is ptr_or_null_ expected ptr_ or socket", + .errstr = "Possibly NULL pointer passed to trusted arg0", .fixup_kfunc_btf_id = { { "bpf_lookup_system_key", 1 }, { "bpf_key_put", 3 }, @@ -182,7 +182,7 @@ .kfunc = "bpf", .expected_attach_type = BPF_LSM_MAC, .flags = BPF_F_SLEEPABLE, - .errstr = "arg#0 pointer type STRUCT bpf_key must point to scalar, or struct with scalar", + .errstr = "Possibly NULL pointer passed to trusted arg0", .fixup_kfunc_btf_id = { { "bpf_key_put", 1 }, }, -- cgit v1.2.3 From 08a7ce384e33e53e0732c500a8af67a73f8fceca Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Wed, 22 Mar 2023 14:52:43 -0700 Subject: bpf: Use bpf_mem_cache_alloc/free in bpf_local_storage_elem This patch uses bpf_mem_alloc for the task and cgroup local storage that the bpf prog can easily get a hold of the storage owner's PTR_TO_BTF_ID. eg. bpf_get_current_task_btf() can be used in some of the kmalloc code path which will cause deadlock/recursion. bpf_mem_cache_alloc is deadlock free and will solve a legit use case in [1]. For sk storage, its batch creation benchmark shows a few percent regression when the sk create/destroy batch size is larger than 32. The sk creation/destruction happens much more often and depends on external traffic. Considering it is hypothetical to be able to cause deadlock with sk storage, it can cross the bridge to use bpf_mem_alloc till a legit (ie. useful) use case comes up. For inode storage, bpf_local_storage_destroy() is called before waiting for a rcu gp and its memory cannot be reused immediately. inode stays with kmalloc/kfree after the rcu [or tasks_trace] gp. A 'bool bpf_ma' argument is added to bpf_local_storage_map_alloc(). Only task and cgroup storage have 'bpf_ma == true' which means to use bpf_mem_cache_alloc/free(). This patch only changes selem to use bpf_mem_alloc for task and cgroup. The next patch will change the local_storage to use bpf_mem_alloc also for task and cgroup. Here is some more details on the changes: * memory allocation: After bpf_mem_cache_alloc(), the SDATA(selem)->data is zero-ed because bpf_mem_cache_alloc() could return a reused selem. It is to keep the existing bpf_map_kzalloc() behavior. Only SDATA(selem)->data is zero-ed. SDATA(selem)->data is the visible part to the bpf prog. No need to use zero_map_value() to do the zeroing because bpf_selem_free(..., reuse_now = true) ensures no bpf prog is using the selem before returning the selem through bpf_mem_cache_free(). For the internal fields of selem, they will be initialized when linking to the new smap and the new local_storage. When 'bpf_ma == false', nothing changes in this patch. It will stay with the bpf_map_kzalloc(). * memory free: The bpf_selem_free() and bpf_selem_free_rcu() are modified to handle the bpf_ma == true case. For the common selem free path where its owner is also being destroyed, the mem is freed in bpf_local_storage_destroy(), the owner (task and cgroup) has gone through a rcu gp. The memory can be reused immediately, so bpf_local_storage_destroy() will call bpf_selem_free(..., reuse_now = true) which will do bpf_mem_cache_free() for immediate reuse consideration. An exception is the delete elem code path. The delete elem code path is called from the helper bpf_*_storage_delete() and the syscall bpf_map_delete_elem(). This path is an unusual case for local storage because the common use case is to have the local storage staying with its owner life time so that the bpf prog and the user space does not have to monitor the owner's destruction. For the delete elem path, the selem cannot be reused immediately because there could be bpf prog using it. It will call bpf_selem_free(..., reuse_now = false) and it will wait for a rcu tasks trace gp before freeing the elem. The rcu callback is changed to do bpf_mem_cache_raw_free() instead of kfree(). When 'bpf_ma == false', it should be the same as before. __bpf_selem_free() is added to do the kfree_rcu and call_tasks_trace_rcu(). A few words on the 'reuse_now == true'. When 'reuse_now == true', it is still racing with bpf_local_storage_map_free which is under rcu protection, so it still needs to wait for a rcu gp instead of kfree(). Otherwise, the selem may be reused by slab for a totally different struct while the bpf_local_storage_map_free() is still using it (as a rcu reader). For the inode case, there may be other rcu readers also. In short, when bpf_ma == false and reuse_now == true => vanilla rcu. [1]: https://lore.kernel.org/bpf/20221118190109.1512674-1-namhyung@kernel.org/ Cc: Namhyung Kim Signed-off-by: Martin KaFai Lau Link: https://lore.kernel.org/r/20230322215246.1675516-3-martin.lau@linux.dev Signed-off-by: Alexei Starovoitov --- include/linux/bpf_local_storage.h | 6 ++- kernel/bpf/bpf_cgrp_storage.c | 2 +- kernel/bpf/bpf_inode_storage.c | 2 +- kernel/bpf/bpf_local_storage.c | 95 +++++++++++++++++++++++++++++++++++---- kernel/bpf/bpf_task_storage.c | 2 +- net/core/bpf_sk_storage.c | 2 +- 6 files changed, 95 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h index a34f61467a2f..30efbcab2798 100644 --- a/include/linux/bpf_local_storage.h +++ b/include/linux/bpf_local_storage.h @@ -13,6 +13,7 @@ #include #include #include +#include #include #define BPF_LOCAL_STORAGE_CACHE_SIZE 16 @@ -55,6 +56,8 @@ struct bpf_local_storage_map { u32 bucket_log; u16 elem_size; u16 cache_idx; + struct bpf_mem_alloc selem_ma; + bool bpf_ma; }; struct bpf_local_storage_data { @@ -122,7 +125,8 @@ int bpf_local_storage_map_alloc_check(union bpf_attr *attr); struct bpf_map * bpf_local_storage_map_alloc(union bpf_attr *attr, - struct bpf_local_storage_cache *cache); + struct bpf_local_storage_cache *cache, + bool bpf_ma); struct bpf_local_storage_data * bpf_local_storage_lookup(struct bpf_local_storage *local_storage, diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c index f5b016a5484d..d17d5b694668 100644 --- a/kernel/bpf/bpf_cgrp_storage.c +++ b/kernel/bpf/bpf_cgrp_storage.c @@ -149,7 +149,7 @@ static int notsupp_get_next_key(struct bpf_map *map, void *key, void *next_key) static struct bpf_map *cgroup_storage_map_alloc(union bpf_attr *attr) { - return bpf_local_storage_map_alloc(attr, &cgroup_cache); + return bpf_local_storage_map_alloc(attr, &cgroup_cache, true); } static void cgroup_storage_map_free(struct bpf_map *map) diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c index 9a5f05151898..e17ad581b9be 100644 --- a/kernel/bpf/bpf_inode_storage.c +++ b/kernel/bpf/bpf_inode_storage.c @@ -199,7 +199,7 @@ static int notsupp_get_next_key(struct bpf_map *map, void *key, static struct bpf_map *inode_storage_map_alloc(union bpf_attr *attr) { - return bpf_local_storage_map_alloc(attr, &inode_cache); + return bpf_local_storage_map_alloc(attr, &inode_cache, false); } static void inode_storage_map_free(struct bpf_map *map) diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c index 351d991694cb..309ea727a5cb 100644 --- a/kernel/bpf/bpf_local_storage.c +++ b/kernel/bpf/bpf_local_storage.c @@ -80,8 +80,24 @@ bpf_selem_alloc(struct bpf_local_storage_map *smap, void *owner, if (charge_mem && mem_charge(smap, owner, smap->elem_size)) return NULL; - selem = bpf_map_kzalloc(&smap->map, smap->elem_size, - gfp_flags | __GFP_NOWARN); + if (smap->bpf_ma) { + migrate_disable(); + selem = bpf_mem_cache_alloc_flags(&smap->selem_ma, gfp_flags); + migrate_enable(); + if (selem) + /* Keep the original bpf_map_kzalloc behavior + * before started using the bpf_mem_cache_alloc. + * + * No need to use zero_map_value. The bpf_selem_free() + * only does bpf_mem_cache_free when there is + * no other bpf prog is using the selem. + */ + memset(SDATA(selem)->data, 0, smap->map.value_size); + } else { + selem = bpf_map_kzalloc(&smap->map, smap->elem_size, + gfp_flags | __GFP_NOWARN); + } + if (selem) { if (value) copy_map_value(&smap->map, SDATA(selem)->data, value); @@ -124,12 +140,34 @@ static void bpf_local_storage_free(struct bpf_local_storage *local_storage, call_rcu(&local_storage->rcu, bpf_local_storage_free_rcu); } +/* rcu tasks trace callback for bpf_ma == false */ +static void __bpf_selem_free_trace_rcu(struct rcu_head *rcu) +{ + struct bpf_local_storage_elem *selem; + + selem = container_of(rcu, struct bpf_local_storage_elem, rcu); + if (rcu_trace_implies_rcu_gp()) + kfree(selem); + else + kfree_rcu(selem, rcu); +} + +/* Handle bpf_ma == false */ +static void __bpf_selem_free(struct bpf_local_storage_elem *selem, + bool vanilla_rcu) +{ + if (vanilla_rcu) + kfree_rcu(selem, rcu); + else + call_rcu_tasks_trace(&selem->rcu, __bpf_selem_free_trace_rcu); +} + static void bpf_selem_free_rcu(struct rcu_head *rcu) { struct bpf_local_storage_elem *selem; selem = container_of(rcu, struct bpf_local_storage_elem, rcu); - kfree(selem); + bpf_mem_cache_raw_free(selem); } static void bpf_selem_free_trace_rcu(struct rcu_head *rcu) @@ -145,10 +183,23 @@ void bpf_selem_free(struct bpf_local_storage_elem *selem, bool reuse_now) { bpf_obj_free_fields(smap->map.record, SDATA(selem)->data); - if (!reuse_now) + + if (!smap->bpf_ma) { + __bpf_selem_free(selem, reuse_now); + return; + } + + if (!reuse_now) { call_rcu_tasks_trace(&selem->rcu, bpf_selem_free_trace_rcu); - else - call_rcu(&selem->rcu, bpf_selem_free_rcu); + } else { + /* Instead of using the vanilla call_rcu(), + * bpf_mem_cache_free will be able to reuse selem + * immediately. + */ + migrate_disable(); + bpf_mem_cache_free(&smap->selem_ma, selem); + migrate_enable(); + } } /* local_storage->lock must be held and selem->local_storage == local_storage. @@ -654,13 +705,25 @@ u64 bpf_local_storage_map_mem_usage(const struct bpf_map *map) return usage; } +/* When bpf_ma == true, the bpf_mem_alloc is used to allocate and free memory. + * A deadlock free allocator is useful for storage that the bpf prog can easily + * get a hold of the owner PTR_TO_BTF_ID in any context. eg. bpf_get_current_task_btf. + * The task and cgroup storage fall into this case. The bpf_mem_alloc reuses + * memory immediately. To be reuse-immediate safe, the owner destruction + * code path needs to go through a rcu grace period before calling + * bpf_local_storage_destroy(). + * + * When bpf_ma == false, the kmalloc and kfree are used. + */ struct bpf_map * bpf_local_storage_map_alloc(union bpf_attr *attr, - struct bpf_local_storage_cache *cache) + struct bpf_local_storage_cache *cache, + bool bpf_ma) { struct bpf_local_storage_map *smap; unsigned int i; u32 nbuckets; + int err; smap = bpf_map_area_alloc(sizeof(*smap), NUMA_NO_NODE); if (!smap) @@ -675,8 +738,8 @@ bpf_local_storage_map_alloc(union bpf_attr *attr, smap->buckets = bpf_map_kvcalloc(&smap->map, sizeof(*smap->buckets), nbuckets, GFP_USER | __GFP_NOWARN); if (!smap->buckets) { - bpf_map_area_free(smap); - return ERR_PTR(-ENOMEM); + err = -ENOMEM; + goto free_smap; } for (i = 0; i < nbuckets; i++) { @@ -687,8 +750,20 @@ bpf_local_storage_map_alloc(union bpf_attr *attr, smap->elem_size = offsetof(struct bpf_local_storage_elem, sdata.data[attr->value_size]); + smap->bpf_ma = bpf_ma; + if (bpf_ma) { + err = bpf_mem_alloc_init(&smap->selem_ma, smap->elem_size, false); + if (err) + goto free_smap; + } + smap->cache_idx = bpf_local_storage_cache_idx_get(cache); return &smap->map; + +free_smap: + kvfree(smap->buckets); + bpf_map_area_free(smap); + return ERR_PTR(err); } void bpf_local_storage_map_free(struct bpf_map *map, @@ -754,6 +829,8 @@ void bpf_local_storage_map_free(struct bpf_map *map, */ synchronize_rcu(); + if (smap->bpf_ma) + bpf_mem_alloc_destroy(&smap->selem_ma); kvfree(smap->buckets); bpf_map_area_free(smap); } diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c index ab5bd1ef58c4..d1af0c8f9ce4 100644 --- a/kernel/bpf/bpf_task_storage.c +++ b/kernel/bpf/bpf_task_storage.c @@ -309,7 +309,7 @@ static int notsupp_get_next_key(struct bpf_map *map, void *key, void *next_key) static struct bpf_map *task_storage_map_alloc(union bpf_attr *attr) { - return bpf_local_storage_map_alloc(attr, &task_cache); + return bpf_local_storage_map_alloc(attr, &task_cache, true); } static void task_storage_map_free(struct bpf_map *map) diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index cb0f5a105b89..085025c7130a 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -68,7 +68,7 @@ static void bpf_sk_storage_map_free(struct bpf_map *map) static struct bpf_map *bpf_sk_storage_map_alloc(union bpf_attr *attr) { - return bpf_local_storage_map_alloc(attr, &sk_cache); + return bpf_local_storage_map_alloc(attr, &sk_cache, false); } static int notsupp_get_next_key(struct bpf_map *map, void *key, -- cgit v1.2.3 From 5f5a7d8d8bd461f515543040ad7d107cc405d30c Mon Sep 17 00:00:00 2001 From: Nuno Gonçalves Date: Fri, 24 Mar 2023 10:02:22 +0000 Subject: xsk: allow remap of fill and/or completion rings MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The remap of fill and completion rings was frowned upon as they control the usage of UMEM which does not support concurrent use. At the same time this would disallow the remap of these rings into another process. A possible use case is that the user wants to transfer the socket/ UMEM ownership to another process (via SYS_pidfd_getfd) and so would need to also remap these rings. This will have no impact on current usages and just relaxes the remap limitation. Signed-off-by: Nuno Gonçalves Reviewed-by: Maciej Fijalkowski Acked-by: Magnus Karlsson Link: https://lore.kernel.org/r/20230324100222.13434-1-nunog@fr24.com Signed-off-by: Alexei Starovoitov --- net/xdp/xsk.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 2ac58b282b5e..cc1e7f15fa73 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -1301,9 +1301,10 @@ static int xsk_mmap(struct file *file, struct socket *sock, loff_t offset = (loff_t)vma->vm_pgoff << PAGE_SHIFT; unsigned long size = vma->vm_end - vma->vm_start; struct xdp_sock *xs = xdp_sk(sock->sk); + int state = READ_ONCE(xs->state); struct xsk_queue *q = NULL; - if (READ_ONCE(xs->state) != XSK_READY) + if (state != XSK_READY && state != XSK_BOUND) return -EBUSY; if (offset == XDP_PGOFF_RX_RING) { @@ -1314,9 +1315,11 @@ static int xsk_mmap(struct file *file, struct socket *sock, /* Matches the smp_wmb() in XDP_UMEM_REG */ smp_rmb(); if (offset == XDP_UMEM_PGOFF_FILL_RING) - q = READ_ONCE(xs->fq_tmp); + q = state == XSK_READY ? READ_ONCE(xs->fq_tmp) : + READ_ONCE(xs->pool->fq); else if (offset == XDP_UMEM_PGOFF_COMPLETION_RING) - q = READ_ONCE(xs->cq_tmp); + q = state == XSK_READY ? READ_ONCE(xs->cq_tmp) : + READ_ONCE(xs->pool->cq); } if (!q) -- cgit v1.2.3 From 46c30cb8f5393586c6ebc7b53a235c85bfac1de8 Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Sun, 18 Dec 2022 18:03:54 +0000 Subject: 9p: Add additional debug flags and open modes Add some additional debug flags to assist with debugging cache changes. Also add some additional open modes so we can track cache state in fids more directly. Signed-off-by: Eric Van Hensbergen --- include/net/9p/9p.h | 6 ++++++ net/9p/client.c | 8 ++++---- 2 files changed, 10 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/net/9p/9p.h b/include/net/9p/9p.h index 429adf6be29c..60cad0d200a4 100644 --- a/include/net/9p/9p.h +++ b/include/net/9p/9p.h @@ -42,6 +42,8 @@ enum p9_debug_flags { P9_DEBUG_PKT = (1<<10), P9_DEBUG_FSC = (1<<11), P9_DEBUG_VPKT = (1<<12), + P9_DEBUG_CACHE = (1<<13), + P9_DEBUG_MMAP = (1<<14), }; #ifdef CONFIG_NET_9P_DEBUG @@ -213,6 +215,10 @@ enum p9_open_mode_t { P9_ORCLOSE = 0x40, P9_OAPPEND = 0x80, P9_OEXCL = 0x1000, + P9L_MODE_MASK = 0x1FFF, /* don't send anything under this to server */ + P9L_DIRECT = 0x2000, /* cache disabled */ + P9L_NOWRITECACHE = 0x4000, /* no write caching */ + P9L_LOOSE = 0x8000, /* loose cache */ }; /** diff --git a/net/9p/client.c b/net/9p/client.c index 2adcb5e7b0e2..a3340268ec8d 100644 --- a/net/9p/client.c +++ b/net/9p/client.c @@ -1230,9 +1230,9 @@ int p9_client_open(struct p9_fid *fid, int mode) return -EINVAL; if (p9_is_proto_dotl(clnt)) - req = p9_client_rpc(clnt, P9_TLOPEN, "dd", fid->fid, mode); + req = p9_client_rpc(clnt, P9_TLOPEN, "dd", fid->fid, mode & P9L_MODE_MASK); else - req = p9_client_rpc(clnt, P9_TOPEN, "db", fid->fid, mode); + req = p9_client_rpc(clnt, P9_TOPEN, "db", fid->fid, mode & P9L_MODE_MASK); if (IS_ERR(req)) { err = PTR_ERR(req); goto error; @@ -1277,7 +1277,7 @@ int p9_client_create_dotl(struct p9_fid *ofid, const char *name, u32 flags, return -EINVAL; req = p9_client_rpc(clnt, P9_TLCREATE, "dsddg", ofid->fid, name, flags, - mode, gid); + mode & P9L_MODE_MASK, gid); if (IS_ERR(req)) { err = PTR_ERR(req); goto error; @@ -1321,7 +1321,7 @@ int p9_client_fcreate(struct p9_fid *fid, const char *name, u32 perm, int mode, return -EINVAL; req = p9_client_rpc(clnt, P9_TCREATE, "dsdb?s", fid->fid, name, perm, - mode, extension); + mode & P9L_MODE_MASK, extension); if (IS_ERR(req)) { err = PTR_ERR(req); goto error; -- cgit v1.2.3 From b465518dc27da1ed74b8cbada4659708aac35adb Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Fri, 24 Mar 2023 12:54:50 +0100 Subject: vsock/loopback: use only sk_buff_head.lock to protect the packet queue pkt_list_lock was used before commit 71dc9ec9ac7d ("virtio/vsock: replace virtio_vsock_pkt with sk_buff") to protect the packet queue. After that commit we switched to sk_buff and we are using sk_buff_head.lock in almost every place to protect the packet queue except in vsock_loopback_work() when we call skb_queue_splice_init(). As reported by syzbot, this caused unlocked concurrent access to the packet queue between vsock_loopback_work() and vsock_loopback_cancel_pkt() since it is not holding pkt_list_lock. With the introduction of sk_buff_head, pkt_list_lock is redundant and can cause confusion, so let's remove it and use sk_buff_head.lock everywhere to protect the packet queue access. Fixes: 71dc9ec9ac7d ("virtio/vsock: replace virtio_vsock_pkt with sk_buff") Cc: bobby.eshleman@bytedance.com Reported-and-tested-by: syzbot+befff0a9536049e7902e@syzkaller.appspotmail.com Signed-off-by: Stefano Garzarella Reviewed-by: Bobby Eshleman Reviewed-by: Arseniy Krasnov Signed-off-by: David S. Miller --- net/vmw_vsock/vsock_loopback.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/vsock_loopback.c b/net/vmw_vsock/vsock_loopback.c index 671e03240fc5..89905c092645 100644 --- a/net/vmw_vsock/vsock_loopback.c +++ b/net/vmw_vsock/vsock_loopback.c @@ -15,7 +15,6 @@ struct vsock_loopback { struct workqueue_struct *workqueue; - spinlock_t pkt_list_lock; /* protects pkt_list */ struct sk_buff_head pkt_queue; struct work_struct pkt_work; }; @@ -32,9 +31,7 @@ static int vsock_loopback_send_pkt(struct sk_buff *skb) struct vsock_loopback *vsock = &the_vsock_loopback; int len = skb->len; - spin_lock_bh(&vsock->pkt_list_lock); skb_queue_tail(&vsock->pkt_queue, skb); - spin_unlock_bh(&vsock->pkt_list_lock); queue_work(vsock->workqueue, &vsock->pkt_work); @@ -113,9 +110,9 @@ static void vsock_loopback_work(struct work_struct *work) skb_queue_head_init(&pkts); - spin_lock_bh(&vsock->pkt_list_lock); + spin_lock_bh(&vsock->pkt_queue.lock); skb_queue_splice_init(&vsock->pkt_queue, &pkts); - spin_unlock_bh(&vsock->pkt_list_lock); + spin_unlock_bh(&vsock->pkt_queue.lock); while ((skb = __skb_dequeue(&pkts))) { virtio_transport_deliver_tap_pkt(skb); @@ -132,7 +129,6 @@ static int __init vsock_loopback_init(void) if (!vsock->workqueue) return -ENOMEM; - spin_lock_init(&vsock->pkt_list_lock); skb_queue_head_init(&vsock->pkt_queue); INIT_WORK(&vsock->pkt_work, vsock_loopback_work); @@ -156,9 +152,7 @@ static void __exit vsock_loopback_exit(void) flush_work(&vsock->pkt_work); - spin_lock_bh(&vsock->pkt_list_lock); virtio_vsock_skb_queue_purge(&vsock->pkt_queue); - spin_unlock_bh(&vsock->pkt_list_lock); destroy_workqueue(vsock->workqueue); } -- cgit v1.2.3 From e5b42483ccce50d5b957f474fd332afd4ef0c27b Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Fri, 24 Mar 2023 23:11:49 +0100 Subject: dev_ioctl: fix a W=1 warning This fixes the following warning when compiled with GCC 12.2.0 and W=1. net/core/dev_ioctl.c:475: warning: Function parameter or member 'data' not described in 'dev_ioctl' Signed-off-by: Heiner Kallweit Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/core/dev_ioctl.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 5cdbfbf9a7dc..846669426236 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -462,6 +462,7 @@ EXPORT_SYMBOL(dev_load); * @net: the applicable net namespace * @cmd: command to issue * @ifr: pointer to a struct ifreq in user space + * @data: data exchanged with userspace * @need_copyout: whether or not copy_to_user() should be called * * Issue ioctl functions to devices. This is normally called by the -- cgit v1.2.3 From d1366b283d94ac4537a4b3a1e8668da4df7ce7e9 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Fri, 24 Mar 2023 14:01:41 +0100 Subject: can: j1939: prevent deadlock by moving j1939_sk_errqueue() This commit addresses a deadlock situation that can occur in certain scenarios, such as when running data TP/ETP transfer and subscribing to the error queue while receiving a net down event. The deadlock involves locks in the following order: 3 j1939_session_list_lock -> active_session_list_lock j1939_session_activate ... j1939_sk_queue_activate_next -> sk_session_queue_lock ... j1939_xtp_rx_eoma_one 2 j1939_sk_queue_drop_all -> sk_session_queue_lock ... j1939_sk_netdev_event_netdown -> j1939_socks_lock j1939_netdev_notify 1 j1939_sk_errqueue -> j1939_socks_lock __j1939_session_cancel -> active_session_list_lock j1939_tp_rxtimer CPU0 CPU1 ---- ---- lock(&priv->active_session_list_lock); lock(&jsk->sk_session_queue_lock); lock(&priv->active_session_list_lock); lock(&priv->j1939_socks_lock); The solution implemented in this commit is to move the j1939_sk_errqueue() call out of the active_session_list_lock context, thus preventing the deadlock situation. Reported-by: syzbot+ee1cd780f69483a8616b@syzkaller.appspotmail.com Fixes: 5b9272e93f2e ("can: j1939: extend UAPI to notify about RX status") Co-developed-by: Hillf Danton Signed-off-by: Hillf Danton Signed-off-by: Oleksij Rempel Link: https://lore.kernel.org/all/20230324130141.2132787-1-o.rempel@pengutronix.de Cc: stable@vger.kernel.org Signed-off-by: Marc Kleine-Budde --- net/can/j1939/transport.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/can/j1939/transport.c b/net/can/j1939/transport.c index fce9b9ebf13f..fb92c3609e17 100644 --- a/net/can/j1939/transport.c +++ b/net/can/j1939/transport.c @@ -1124,8 +1124,6 @@ static void __j1939_session_cancel(struct j1939_session *session, if (session->sk) j1939_sk_send_loop_abort(session->sk, session->err); - else - j1939_sk_errqueue(session, J1939_ERRQUEUE_RX_ABORT); } static void j1939_session_cancel(struct j1939_session *session, @@ -1140,6 +1138,9 @@ static void j1939_session_cancel(struct j1939_session *session, } j1939_session_list_unlock(session->priv); + + if (!session->sk) + j1939_sk_errqueue(session, J1939_ERRQUEUE_RX_ABORT); } static enum hrtimer_restart j1939_tp_txtimer(struct hrtimer *hrtimer) @@ -1253,6 +1254,9 @@ static enum hrtimer_restart j1939_tp_rxtimer(struct hrtimer *hrtimer) __j1939_session_cancel(session, J1939_XTP_ABORT_TIMEOUT); } j1939_session_list_unlock(session->priv); + + if (!session->sk) + j1939_sk_errqueue(session, J1939_ERRQUEUE_RX_ABORT); } j1939_session_put(session); -- cgit v1.2.3 From 2b4c99f7d9a57ecd644eda9b1fb0a1072414959f Mon Sep 17 00:00:00 2001 From: Ivan Orlov Date: Tue, 14 Mar 2023 16:04:45 +0400 Subject: can: bcm: bcm_tx_setup(): fix KMSAN uninit-value in vfs_write Syzkaller reported the following issue: ===================================================== BUG: KMSAN: uninit-value in aio_rw_done fs/aio.c:1520 [inline] BUG: KMSAN: uninit-value in aio_write+0x899/0x950 fs/aio.c:1600 aio_rw_done fs/aio.c:1520 [inline] aio_write+0x899/0x950 fs/aio.c:1600 io_submit_one+0x1d1c/0x3bf0 fs/aio.c:2019 __do_sys_io_submit fs/aio.c:2078 [inline] __se_sys_io_submit+0x293/0x770 fs/aio.c:2048 __x64_sys_io_submit+0x92/0xd0 fs/aio.c:2048 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Uninit was created at: slab_post_alloc_hook mm/slab.h:766 [inline] slab_alloc_node mm/slub.c:3452 [inline] __kmem_cache_alloc_node+0x71f/0xce0 mm/slub.c:3491 __do_kmalloc_node mm/slab_common.c:967 [inline] __kmalloc+0x11d/0x3b0 mm/slab_common.c:981 kmalloc_array include/linux/slab.h:636 [inline] bcm_tx_setup+0x80e/0x29d0 net/can/bcm.c:930 bcm_sendmsg+0x3a2/0xce0 net/can/bcm.c:1351 sock_sendmsg_nosec net/socket.c:714 [inline] sock_sendmsg net/socket.c:734 [inline] sock_write_iter+0x495/0x5e0 net/socket.c:1108 call_write_iter include/linux/fs.h:2189 [inline] aio_write+0x63a/0x950 fs/aio.c:1600 io_submit_one+0x1d1c/0x3bf0 fs/aio.c:2019 __do_sys_io_submit fs/aio.c:2078 [inline] __se_sys_io_submit+0x293/0x770 fs/aio.c:2048 __x64_sys_io_submit+0x92/0xd0 fs/aio.c:2048 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd CPU: 1 PID: 5034 Comm: syz-executor350 Not tainted 6.2.0-rc6-syzkaller-80422-geda666ff2276 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/12/2023 ===================================================== We can follow the call chain and find that 'bcm_tx_setup' function calls 'memcpy_from_msg' to copy some content to the newly allocated frame of 'op->frames'. After that the 'len' field of copied structure being compared with some constant value (64 or 8). However, if 'memcpy_from_msg' returns an error, we will compare some uninitialized memory. This triggers 'uninit-value' issue. This patch will add 'memcpy_from_msg' possible errors processing to avoid uninit-value issue. Tested via syzkaller Reported-by: syzbot+c9bfd85eca611ebf5db1@syzkaller.appspotmail.com Link: https://syzkaller.appspot.com/bug?id=47f897f8ad958bbde5790ebf389b5e7e0a345089 Signed-off-by: Ivan Orlov Fixes: 6f3b911d5f29b ("can: bcm: add support for CAN FD frames") Acked-by: Oliver Hartkopp Link: https://lore.kernel.org/all/20230314120445.12407-1-ivan.orlov0322@gmail.com Signed-off-by: Marc Kleine-Budde --- net/can/bcm.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/can/bcm.c b/net/can/bcm.c index 27706f6ace34..a962ec2b8ba5 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -941,6 +941,8 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, cf = op->frames + op->cfsiz * i; err = memcpy_from_msg((u8 *)cf, msg, op->cfsiz); + if (err < 0) + goto free_op; if (op->flags & CAN_FD_FRAME) { if (cf->len > 64) @@ -950,12 +952,8 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, err = -EINVAL; } - if (err < 0) { - if (op->frames != &op->sframe) - kfree(op->frames); - kfree(op); - return err; - } + if (err < 0) + goto free_op; if (msg_head->flags & TX_CP_CAN_ID) { /* copy can_id into frame */ @@ -1026,6 +1024,12 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, bcm_tx_start_timer(op); return msg_head->nframes * op->cfsiz + MHSIZ; + +free_op: + if (op->frames != &op->sframe) + kfree(op->frames); + kfree(op); + return err; } /* -- cgit v1.2.3 From 3948b05950fdd64002a5f182c65ba5cf2d53cf71 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 23 Mar 2023 09:28:42 -0700 Subject: net: introduce a config option to tweak MAX_SKB_FRAGS Currently, MAX_SKB_FRAGS value is 17. For standard tcp sendmsg() traffic, no big deal because tcp_sendmsg() attempts order-3 allocations, stuffing 32768 bytes per frag. But with zero copy, we use order-0 pages. For BIG TCP to show its full potential, we add a config option to be able to fit up to 45 segments per skb. This is also needed for BIG TCP rx zerocopy, as zerocopy currently does not support skbs with frag list. We have used MAX_SKB_FRAGS=45 value for years at Google before we deployed 4K MTU, with no adverse effect, other than a recent issue in mlx4, fixed in commit 26782aad00cc ("net/mlx4: MLX4_TX_BOUNCE_BUFFER_SIZE depends on MAX_SKB_FRAGS") Back then, goal was to be able to receive full size (64KB) GRO packets without the frag_list overhead. Note that /proc/sys/net/core/max_skb_frags can also be used to limit the number of fragments TCP can use in tx packets. By default we keep the old/legacy value of 17 until we get more coverage for the updated values. Sizes of struct skb_shared_info on 64bit arches MAX_SKB_FRAGS | sizeof(struct skb_shared_info): ============================================== 17 320 21 320+64 = 384 25 320+128 = 448 29 320+192 = 512 33 320+256 = 576 37 320+320 = 640 41 320+384 = 704 45 320+448 = 768 This inflation might cause problems for drivers assuming they could pack both the incoming packet (for MTU=1500) and skb_shared_info in half a page, using build_skb(). v3: fix build error when CONFIG_NET=n v2: fix two build errors assuming MAX_SKB_FRAGS was "unsigned long" Signed-off-by: Eric Dumazet Reviewed-by: Nikolay Aleksandrov Reviewed-by: Jason Xing Link: https://lore.kernel.org/r/20230323162842.1935061-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski --- drivers/scsi/cxgbi/libcxgbi.c | 4 ++-- include/linux/skbuff.h | 16 +++++----------- net/Kconfig | 12 ++++++++++++ net/packet/af_packet.c | 4 ++-- 4 files changed, 21 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/drivers/scsi/cxgbi/libcxgbi.c b/drivers/scsi/cxgbi/libcxgbi.c index af281e271f88..3e1de4c842cc 100644 --- a/drivers/scsi/cxgbi/libcxgbi.c +++ b/drivers/scsi/cxgbi/libcxgbi.c @@ -2314,9 +2314,9 @@ static int cxgbi_sock_tx_queue_up(struct cxgbi_sock *csk, struct sk_buff *skb) frags++; if (frags >= SKB_WR_LIST_SIZE) { - pr_err("csk 0x%p, frags %u, %u,%u >%lu.\n", + pr_err("csk 0x%p, frags %u, %u,%u >%u.\n", csk, skb_shinfo(skb)->nr_frags, skb->len, - skb->data_len, SKB_WR_LIST_SIZE); + skb->data_len, (unsigned int)SKB_WR_LIST_SIZE); return -EINVAL; } diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index fe661011644b..82511b2f61ea 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -345,18 +345,12 @@ struct sk_buff_head { struct sk_buff; -/* To allow 64K frame to be packed as single skb without frag_list we - * require 64K/PAGE_SIZE pages plus 1 additional page to allow for - * buffers which do not start on a page boundary. - * - * Since GRO uses frags we allocate at least 16 regardless of page - * size. - */ -#if (65536/PAGE_SIZE + 1) < 16 -#define MAX_SKB_FRAGS 16UL -#else -#define MAX_SKB_FRAGS (65536/PAGE_SIZE + 1) +#ifndef CONFIG_MAX_SKB_FRAGS +# define CONFIG_MAX_SKB_FRAGS 17 #endif + +#define MAX_SKB_FRAGS CONFIG_MAX_SKB_FRAGS + extern int sysctl_max_skb_frags; /* Set skb_shinfo(skb)->gso_size to this in case you want skb_segment to diff --git a/net/Kconfig b/net/Kconfig index 48c33c222199..f806722bccf4 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -251,6 +251,18 @@ config PCPU_DEV_REFCNT network device refcount are using per cpu variables if this option is set. This can be forced to N to detect underflows (with a performance drop). +config MAX_SKB_FRAGS + int "Maximum number of fragments per skb_shared_info" + range 17 45 + default 17 + help + Having more fragments per skb_shared_info can help GRO efficiency. + This helps BIG TCP workloads, but might expose bugs in some + legacy drivers. + This also increases memory overhead of small packets, + and in drivers using build_skb(). + If unsure, say 17. + config RPS bool depends on SMP && SYSFS diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 497193f73030..568f8d76e3c1 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2622,8 +2622,8 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb, nr_frags = skb_shinfo(skb)->nr_frags; if (unlikely(nr_frags >= MAX_SKB_FRAGS)) { - pr_err("Packet exceed the number of skb frags(%lu)\n", - MAX_SKB_FRAGS); + pr_err("Packet exceed the number of skb frags(%u)\n", + (unsigned int)MAX_SKB_FRAGS); return -EFAULT; } -- cgit v1.2.3 From 233eb4e786b57ea686b51c13a04cc2839fd682fc Mon Sep 17 00:00:00 2001 From: Shay Agroskin Date: Thu, 23 Mar 2023 18:36:05 +0200 Subject: ethtool: Add support for configuring tx_push_buf_len This attribute, which is part of ethtool's ring param configuration allows the user to specify the maximum number of the packet's payload that can be written directly to the device. Example usage: # ethtool -G [interface] tx-push-buf-len [number of bytes] Co-developed-by: Jakub Kicinski Signed-off-by: Shay Agroskin Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/ethtool.yaml | 8 +++++ Documentation/networking/ethtool-netlink.rst | 47 ++++++++++++++++++---------- include/linux/ethtool.h | 14 ++++++--- include/uapi/linux/ethtool_netlink.h | 2 ++ net/ethtool/netlink.h | 2 +- net/ethtool/rings.c | 34 ++++++++++++++++++-- 6 files changed, 84 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml index 4727c067e2ba..6d8ae3d9a680 100644 --- a/Documentation/netlink/specs/ethtool.yaml +++ b/Documentation/netlink/specs/ethtool.yaml @@ -165,6 +165,12 @@ attribute-sets: - name: rx-push type: u8 + - + name: tx-push-buf-len + type: u32 + - + name: tx-push-buf-len-max + type: u32 - name: mm-stat @@ -311,6 +317,8 @@ operations: - cqe-size - tx-push - rx-push + - tx-push-buf-len + - tx-push-buf-len-max dump: *ring-get-op - name: rings-set diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst index e1bc6186d7ea..cd0973d4ba01 100644 --- a/Documentation/networking/ethtool-netlink.rst +++ b/Documentation/networking/ethtool-netlink.rst @@ -860,22 +860,24 @@ Request contents: Kernel response contents: - ==================================== ====== =========================== - ``ETHTOOL_A_RINGS_HEADER`` nested reply header - ``ETHTOOL_A_RINGS_RX_MAX`` u32 max size of RX ring - ``ETHTOOL_A_RINGS_RX_MINI_MAX`` u32 max size of RX mini ring - ``ETHTOOL_A_RINGS_RX_JUMBO_MAX`` u32 max size of RX jumbo ring - ``ETHTOOL_A_RINGS_TX_MAX`` u32 max size of TX ring - ``ETHTOOL_A_RINGS_RX`` u32 size of RX ring - ``ETHTOOL_A_RINGS_RX_MINI`` u32 size of RX mini ring - ``ETHTOOL_A_RINGS_RX_JUMBO`` u32 size of RX jumbo ring - ``ETHTOOL_A_RINGS_TX`` u32 size of TX ring - ``ETHTOOL_A_RINGS_RX_BUF_LEN`` u32 size of buffers on the ring - ``ETHTOOL_A_RINGS_TCP_DATA_SPLIT`` u8 TCP header / data split - ``ETHTOOL_A_RINGS_CQE_SIZE`` u32 Size of TX/RX CQE - ``ETHTOOL_A_RINGS_TX_PUSH`` u8 flag of TX Push mode - ``ETHTOOL_A_RINGS_RX_PUSH`` u8 flag of RX Push mode - ==================================== ====== =========================== + ======================================= ====== =========================== + ``ETHTOOL_A_RINGS_HEADER`` nested reply header + ``ETHTOOL_A_RINGS_RX_MAX`` u32 max size of RX ring + ``ETHTOOL_A_RINGS_RX_MINI_MAX`` u32 max size of RX mini ring + ``ETHTOOL_A_RINGS_RX_JUMBO_MAX`` u32 max size of RX jumbo ring + ``ETHTOOL_A_RINGS_TX_MAX`` u32 max size of TX ring + ``ETHTOOL_A_RINGS_RX`` u32 size of RX ring + ``ETHTOOL_A_RINGS_RX_MINI`` u32 size of RX mini ring + ``ETHTOOL_A_RINGS_RX_JUMBO`` u32 size of RX jumbo ring + ``ETHTOOL_A_RINGS_TX`` u32 size of TX ring + ``ETHTOOL_A_RINGS_RX_BUF_LEN`` u32 size of buffers on the ring + ``ETHTOOL_A_RINGS_TCP_DATA_SPLIT`` u8 TCP header / data split + ``ETHTOOL_A_RINGS_CQE_SIZE`` u32 Size of TX/RX CQE + ``ETHTOOL_A_RINGS_TX_PUSH`` u8 flag of TX Push mode + ``ETHTOOL_A_RINGS_RX_PUSH`` u8 flag of RX Push mode + ``ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN`` u32 size of TX push buffer + ``ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN_MAX`` u32 max size of TX push buffer + ======================================= ====== =========================== ``ETHTOOL_A_RINGS_TCP_DATA_SPLIT`` indicates whether the device is usable with page-flipping TCP zero-copy receive (``getsockopt(TCP_ZEROCOPY_RECEIVE)``). @@ -891,6 +893,18 @@ through MMIO writes, thus reducing the latency. However, enabling this feature may increase the CPU cost. Drivers may enforce additional per-packet eligibility checks (e.g. on packet size). +``ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN`` specifies the maximum number of bytes of a +transmitted packet a driver can push directly to the underlying device +('push' mode). Pushing some of the payload bytes to the device has the +advantages of reducing latency for small packets by avoiding DMA mapping (same +as ``ETHTOOL_A_RINGS_TX_PUSH`` parameter) as well as allowing the underlying +device to process packet headers ahead of fetching its payload. +This can help the device to make fast actions based on the packet's headers. +This is similar to the "tx-copybreak" parameter, which copies the packet to a +preallocated DMA memory area instead of mapping new memory. However, +tx-push-buff parameter copies the packet directly to the device to allow the +device to take faster actions on the packet. + RINGS_SET ========= @@ -908,6 +922,7 @@ Request contents: ``ETHTOOL_A_RINGS_CQE_SIZE`` u32 Size of TX/RX CQE ``ETHTOOL_A_RINGS_TX_PUSH`` u8 flag of TX Push mode ``ETHTOOL_A_RINGS_RX_PUSH`` u8 flag of RX Push mode + ``ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN`` u32 size of TX push buffer ==================================== ====== =========================== Kernel checks that requested ring sizes do not exceed limits reported by diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index 2792185dda22..798d35890118 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -75,6 +75,8 @@ enum { * @tx_push: The flag of tx push mode * @rx_push: The flag of rx push mode * @cqe_size: Size of TX/RX completion queue event + * @tx_push_buf_len: Size of TX push buffer + * @tx_push_buf_max_len: Maximum allowed size of TX push buffer */ struct kernel_ethtool_ringparam { u32 rx_buf_len; @@ -82,6 +84,8 @@ struct kernel_ethtool_ringparam { u8 tx_push; u8 rx_push; u32 cqe_size; + u32 tx_push_buf_len; + u32 tx_push_buf_max_len; }; /** @@ -90,12 +94,14 @@ struct kernel_ethtool_ringparam { * @ETHTOOL_RING_USE_CQE_SIZE: capture for setting cqe_size * @ETHTOOL_RING_USE_TX_PUSH: capture for setting tx_push * @ETHTOOL_RING_USE_RX_PUSH: capture for setting rx_push + * @ETHTOOL_RING_USE_TX_PUSH_BUF_LEN: capture for setting tx_push_buf_len */ enum ethtool_supported_ring_param { - ETHTOOL_RING_USE_RX_BUF_LEN = BIT(0), - ETHTOOL_RING_USE_CQE_SIZE = BIT(1), - ETHTOOL_RING_USE_TX_PUSH = BIT(2), - ETHTOOL_RING_USE_RX_PUSH = BIT(3), + ETHTOOL_RING_USE_RX_BUF_LEN = BIT(0), + ETHTOOL_RING_USE_CQE_SIZE = BIT(1), + ETHTOOL_RING_USE_TX_PUSH = BIT(2), + ETHTOOL_RING_USE_RX_PUSH = BIT(3), + ETHTOOL_RING_USE_TX_PUSH_BUF_LEN = BIT(4), }; #define __ETH_RSS_HASH_BIT(bit) ((u32)1 << (bit)) diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h index d39ce21381c5..1ebf8d455f07 100644 --- a/include/uapi/linux/ethtool_netlink.h +++ b/include/uapi/linux/ethtool_netlink.h @@ -357,6 +357,8 @@ enum { ETHTOOL_A_RINGS_CQE_SIZE, /* u32 */ ETHTOOL_A_RINGS_TX_PUSH, /* u8 */ ETHTOOL_A_RINGS_RX_PUSH, /* u8 */ + ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN, /* u32 */ + ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN_MAX, /* u32 */ /* add new constants above here */ __ETHTOOL_A_RINGS_CNT, diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h index f7b189ed96b2..79424b34b553 100644 --- a/net/ethtool/netlink.h +++ b/net/ethtool/netlink.h @@ -413,7 +413,7 @@ extern const struct nla_policy ethnl_features_set_policy[ETHTOOL_A_FEATURES_WANT extern const struct nla_policy ethnl_privflags_get_policy[ETHTOOL_A_PRIVFLAGS_HEADER + 1]; extern const struct nla_policy ethnl_privflags_set_policy[ETHTOOL_A_PRIVFLAGS_FLAGS + 1]; extern const struct nla_policy ethnl_rings_get_policy[ETHTOOL_A_RINGS_HEADER + 1]; -extern const struct nla_policy ethnl_rings_set_policy[ETHTOOL_A_RINGS_RX_PUSH + 1]; +extern const struct nla_policy ethnl_rings_set_policy[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN_MAX + 1]; extern const struct nla_policy ethnl_channels_get_policy[ETHTOOL_A_CHANNELS_HEADER + 1]; extern const struct nla_policy ethnl_channels_set_policy[ETHTOOL_A_CHANNELS_COMBINED_COUNT + 1]; extern const struct nla_policy ethnl_coalesce_get_policy[ETHTOOL_A_COALESCE_HEADER + 1]; diff --git a/net/ethtool/rings.c b/net/ethtool/rings.c index f358cd57d094..1c4972526142 100644 --- a/net/ethtool/rings.c +++ b/net/ethtool/rings.c @@ -11,6 +11,7 @@ struct rings_reply_data { struct ethnl_reply_data base; struct ethtool_ringparam ringparam; struct kernel_ethtool_ringparam kernel_ringparam; + u32 supported_ring_params; }; #define RINGS_REPDATA(__reply_base) \ @@ -32,6 +33,8 @@ static int rings_prepare_data(const struct ethnl_req_info *req_base, if (!dev->ethtool_ops->get_ringparam) return -EOPNOTSUPP; + + data->supported_ring_params = dev->ethtool_ops->supported_ring_params; ret = ethnl_ops_begin(dev); if (ret < 0) return ret; @@ -57,7 +60,9 @@ static int rings_reply_size(const struct ethnl_req_info *req_base, nla_total_size(sizeof(u8)) + /* _RINGS_TCP_DATA_SPLIT */ nla_total_size(sizeof(u32) + /* _RINGS_CQE_SIZE */ nla_total_size(sizeof(u8)) + /* _RINGS_TX_PUSH */ - nla_total_size(sizeof(u8))); /* _RINGS_RX_PUSH */ + nla_total_size(sizeof(u8))) + /* _RINGS_RX_PUSH */ + nla_total_size(sizeof(u32)) + /* _RINGS_TX_PUSH_BUF_LEN */ + nla_total_size(sizeof(u32)); /* _RINGS_TX_PUSH_BUF_LEN_MAX */ } static int rings_fill_reply(struct sk_buff *skb, @@ -67,6 +72,7 @@ static int rings_fill_reply(struct sk_buff *skb, const struct rings_reply_data *data = RINGS_REPDATA(reply_base); const struct kernel_ethtool_ringparam *kr = &data->kernel_ringparam; const struct ethtool_ringparam *ringparam = &data->ringparam; + u32 supported_ring_params = data->supported_ring_params; WARN_ON(kr->tcp_data_split > ETHTOOL_TCP_DATA_SPLIT_ENABLED); @@ -98,7 +104,12 @@ static int rings_fill_reply(struct sk_buff *skb, (kr->cqe_size && (nla_put_u32(skb, ETHTOOL_A_RINGS_CQE_SIZE, kr->cqe_size))) || nla_put_u8(skb, ETHTOOL_A_RINGS_TX_PUSH, !!kr->tx_push) || - nla_put_u8(skb, ETHTOOL_A_RINGS_RX_PUSH, !!kr->rx_push)) + nla_put_u8(skb, ETHTOOL_A_RINGS_RX_PUSH, !!kr->rx_push) || + ((supported_ring_params & ETHTOOL_RING_USE_TX_PUSH_BUF_LEN) && + (nla_put_u32(skb, ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN_MAX, + kr->tx_push_buf_max_len) || + nla_put_u32(skb, ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN, + kr->tx_push_buf_len)))) return -EMSGSIZE; return 0; @@ -117,6 +128,7 @@ const struct nla_policy ethnl_rings_set_policy[] = { [ETHTOOL_A_RINGS_CQE_SIZE] = NLA_POLICY_MIN(NLA_U32, 1), [ETHTOOL_A_RINGS_TX_PUSH] = NLA_POLICY_MAX(NLA_U8, 1), [ETHTOOL_A_RINGS_RX_PUSH] = NLA_POLICY_MAX(NLA_U8, 1), + [ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN] = { .type = NLA_U32 }, }; static int @@ -158,6 +170,14 @@ ethnl_set_rings_validate(struct ethnl_req_info *req_info, return -EOPNOTSUPP; } + if (tb[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN] && + !(ops->supported_ring_params & ETHTOOL_RING_USE_TX_PUSH_BUF_LEN)) { + NL_SET_ERR_MSG_ATTR(info->extack, + tb[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN], + "setting tx push buf len is not supported"); + return -EOPNOTSUPP; + } + return ops->get_ringparam && ops->set_ringparam ? 1 : -EOPNOTSUPP; } @@ -189,6 +209,8 @@ ethnl_set_rings(struct ethnl_req_info *req_info, struct genl_info *info) tb[ETHTOOL_A_RINGS_TX_PUSH], &mod); ethnl_update_u8(&kernel_ringparam.rx_push, tb[ETHTOOL_A_RINGS_RX_PUSH], &mod); + ethnl_update_u32(&kernel_ringparam.tx_push_buf_len, + tb[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN], &mod); if (!mod) return 0; @@ -209,6 +231,14 @@ ethnl_set_rings(struct ethnl_req_info *req_info, struct genl_info *info) return -EINVAL; } + if (kernel_ringparam.tx_push_buf_len > kernel_ringparam.tx_push_buf_max_len) { + NL_SET_ERR_MSG_ATTR_FMT(info->extack, tb[ETHTOOL_A_RINGS_TX_PUSH_BUF_LEN], + "Requested TX push buffer exceeds the maximum of %u", + kernel_ringparam.tx_push_buf_max_len); + + return -EINVAL; + } + ret = dev->ethtool_ops->set_ringparam(dev, &ringparam, &kernel_ringparam, info->extack); return ret < 0 ? ret : 1; -- cgit v1.2.3 From 96d1c81e6a0478535342dff6c730adb076cd84e8 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Sun, 26 Mar 2023 13:59:11 +0200 Subject: can: isotp: add module parameter for maximum pdu size With ISO 15765-2:2016 the PDU size is not limited to 2^12 - 1 (4095) bytes but can be represented as a 32 bit unsigned integer value which allows 2^32 - 1 bytes (~4GB). The use-cases like automotive unified diagnostic services (UDS) and flashing of ECUs still use the small static buffers which are provided at socket creation time. When a use-case requires to transfer PDUs up to 1025 kByte the maximum PDU size can now be extended by setting the module parameter max_pdu_size. The extended size buffers are only allocated on a per-socket/connection base when needed at run-time. changes since v2: https://lore.kernel.org/all/20230313172510.3851-1-socketcan@hartkopp.net - use ARRAY_SIZE() to reference DEFAULT_MAX_PDU_SIZE only at one place changes since v1: https://lore.kernel.org/all/20230311143446.3183-1-socketcan@hartkopp.net - limit the minimum 'max_pdu_size' to 4095 to maintain the classic behavior before ISO 15765-2:2016 Link: https://github.com/raspberrypi/linux/issues/5371 Signed-off-by: Oliver Hartkopp Link: https://lore.kernel.org/all/20230326115911.15094-1-socketcan@hartkopp.net Signed-off-by: Marc Kleine-Budde --- net/can/isotp.c | 65 +++++++++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 56 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/can/isotp.c b/net/can/isotp.c index 9bc344851704..f1ea91772ae7 100644 --- a/net/can/isotp.c +++ b/net/can/isotp.c @@ -85,10 +85,21 @@ MODULE_ALIAS("can-proto-6"); /* ISO 15765-2:2016 supports more than 4095 byte per ISO PDU as the FF_DL can * take full 32 bit values (4 Gbyte). We would need some good concept to handle - * this between user space and kernel space. For now increase the static buffer - * to something about 64 kbyte to be able to test this new functionality. + * this between user space and kernel space. For now set the static buffer to + * something about 8 kbyte to be able to test this new functionality. */ -#define MAX_MSG_LENGTH 66000 +#define DEFAULT_MAX_PDU_SIZE 8300 + +/* maximum PDU size before ISO 15765-2:2016 extension was 4095 */ +#define MAX_12BIT_PDU_SIZE 4095 + +/* limit the isotp pdu size from the optional module parameter to 1MByte */ +#define MAX_PDU_SIZE (1025 * 1024U) + +static unsigned int max_pdu_size __read_mostly = DEFAULT_MAX_PDU_SIZE; +module_param(max_pdu_size, uint, 0444); +MODULE_PARM_DESC(max_pdu_size, "maximum isotp pdu size (default " + __stringify(DEFAULT_MAX_PDU_SIZE) ")"); /* N_PCI type values in bits 7-4 of N_PCI bytes */ #define N_PCI_SF 0x00 /* single frame */ @@ -123,13 +134,15 @@ enum { }; struct tpcon { - unsigned int idx; + u8 *buf; + unsigned int buflen; unsigned int len; + unsigned int idx; u32 state; u8 bs; u8 sn; u8 ll_dl; - u8 buf[MAX_MSG_LENGTH + 1]; + u8 sbuf[DEFAULT_MAX_PDU_SIZE]; }; struct isotp_sock { @@ -503,7 +516,17 @@ static int isotp_rcv_ff(struct sock *sk, struct canfd_frame *cf, int ae) if (so->rx.len + ae + off + ff_pci_sz < so->rx.ll_dl) return 1; - if (so->rx.len > MAX_MSG_LENGTH) { + /* PDU size > default => try max_pdu_size */ + if (so->rx.len > so->rx.buflen && so->rx.buflen < max_pdu_size) { + u8 *newbuf = kmalloc(max_pdu_size, GFP_ATOMIC); + + if (newbuf) { + so->rx.buf = newbuf; + so->rx.buflen = max_pdu_size; + } + } + + if (so->rx.len > so->rx.buflen) { /* send FC frame with overflow status */ isotp_send_fc(sk, ae, ISOTP_FC_OVFLW); return 1; @@ -807,7 +830,7 @@ static void isotp_create_fframe(struct canfd_frame *cf, struct isotp_sock *so, cf->data[0] = so->opt.ext_address; /* create N_PCI bytes with 12/32 bit FF_DL data length */ - if (so->tx.len > 4095) { + if (so->tx.len > MAX_12BIT_PDU_SIZE) { /* use 32 bit FF_DL notation */ cf->data[ae] = N_PCI_FF; cf->data[ae + 1] = 0; @@ -947,7 +970,17 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) so->tx.state = ISOTP_SENDING; } - if (!size || size > MAX_MSG_LENGTH) { + /* PDU size > default => try max_pdu_size */ + if (size > so->tx.buflen && so->tx.buflen < max_pdu_size) { + u8 *newbuf = kmalloc(max_pdu_size, GFP_KERNEL); + + if (newbuf) { + so->tx.buf = newbuf; + so->tx.buflen = max_pdu_size; + } + } + + if (!size || size > so->tx.buflen) { err = -EINVAL; goto err_out_drop; } @@ -1195,6 +1228,12 @@ static int isotp_release(struct socket *sock) so->ifindex = 0; so->bound = 0; + if (so->rx.buf != so->rx.sbuf) + kfree(so->rx.buf); + + if (so->tx.buf != so->tx.sbuf) + kfree(so->tx.buf); + sock_orphan(sk); sock->sk = NULL; @@ -1591,6 +1630,11 @@ static int isotp_init(struct sock *sk) so->rx.state = ISOTP_IDLE; so->tx.state = ISOTP_IDLE; + so->rx.buf = so->rx.sbuf; + so->tx.buf = so->tx.sbuf; + so->rx.buflen = ARRAY_SIZE(so->rx.sbuf); + so->tx.buflen = ARRAY_SIZE(so->tx.sbuf); + hrtimer_init(&so->rxtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_SOFT); so->rxtimer.function = isotp_rx_timer_handler; hrtimer_init(&so->txtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_SOFT); @@ -1658,7 +1702,10 @@ static __init int isotp_module_init(void) { int err; - pr_info("can: isotp protocol\n"); + max_pdu_size = max_t(unsigned int, max_pdu_size, MAX_12BIT_PDU_SIZE); + max_pdu_size = min_t(unsigned int, max_pdu_size, MAX_PDU_SIZE); + + pr_info("can: isotp protocol (max_pdu_size %d)\n", max_pdu_size); err = can_proto_register(&isotp_can_proto); if (err < 0) -- cgit v1.2.3 From 580031ff9952b7dbf48dedba6b56a100ae002bef Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Mon, 27 Mar 2023 17:42:32 -0700 Subject: bpf: tcp: Use sock_gen_put instead of sock_put in bpf_iter_tcp While reviewing the udp-iter batching patches, noticed the bpf_iter_tcp calling sock_put() is incorrect. It should call sock_gen_put instead because bpf_iter_tcp is iterating the ehash table which has the req sk and tw sk. This patch replaces all sock_put with sock_gen_put in the bpf_iter_tcp codepath. Fixes: 04c7820b776f ("bpf: tcp: Bpf iter batching and lock_sock") Signed-off-by: Martin KaFai Lau Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20230328004232.2134233-1-martin.lau@linux.dev --- net/ipv4/tcp_ipv4.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index ea370afa70ed..b9d55277cb85 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2780,7 +2780,7 @@ static int tcp_prog_seq_show(struct bpf_prog *prog, struct bpf_iter_meta *meta, static void bpf_iter_tcp_put_batch(struct bpf_tcp_iter_state *iter) { while (iter->cur_sk < iter->end_sk) - sock_put(iter->batch[iter->cur_sk++]); + sock_gen_put(iter->batch[iter->cur_sk++]); } static int bpf_iter_tcp_realloc_batch(struct bpf_tcp_iter_state *iter, @@ -2941,7 +2941,7 @@ static void *bpf_iter_tcp_seq_next(struct seq_file *seq, void *v, loff_t *pos) * st->bucket. See tcp_seek_last_pos(). */ st->offset++; - sock_put(iter->batch[iter->cur_sk++]); + sock_gen_put(iter->batch[iter->cur_sk++]); } if (iter->cur_sk < iter->end_sk) -- cgit v1.2.3 From b68ffb1b3bee3762f7579bdcc9ca3d415ed0b9b4 Mon Sep 17 00:00:00 2001 From: Arseniy Krasnov Date: Sun, 26 Mar 2023 01:03:52 +0300 Subject: virtio/vsock: allocate multiple skbuffs on tx This adds small optimization for tx path: instead of allocating single skbuff on every call to transport, allocate multiple skbuff's until credit space allows, thus trying to send as much as possible data without return to af_vsock.c. Signed-off-by: Arseniy Krasnov Reviewed-by: Stefano Garzarella Signed-off-by: Paolo Abeni --- net/vmw_vsock/virtio_transport_common.c | 57 +++++++++++++++++++++++++-------- 1 file changed, 43 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 957cdc01c8e8..8c60fc7f94a5 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -196,7 +196,8 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, const struct virtio_transport *t_ops; struct virtio_vsock_sock *vvs; u32 pkt_len = info->pkt_len; - struct sk_buff *skb; + u32 rest_len; + int ret; info->type = virtio_transport_get_type(sk_vsock(vsk)); @@ -216,10 +217,6 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, vvs = vsk->trans; - /* we can send less than pkt_len bytes */ - if (pkt_len > VIRTIO_VSOCK_MAX_PKT_BUF_SIZE) - pkt_len = VIRTIO_VSOCK_MAX_PKT_BUF_SIZE; - /* virtio_transport_get_credit might return less than pkt_len credit */ pkt_len = virtio_transport_get_credit(vvs, pkt_len); @@ -227,17 +224,49 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW) return pkt_len; - skb = virtio_transport_alloc_skb(info, pkt_len, - src_cid, src_port, - dst_cid, dst_port); - if (!skb) { - virtio_transport_put_credit(vvs, pkt_len); - return -ENOMEM; - } + rest_len = pkt_len; + + do { + struct sk_buff *skb; + size_t skb_len; + + skb_len = min_t(u32, VIRTIO_VSOCK_MAX_PKT_BUF_SIZE, rest_len); + + skb = virtio_transport_alloc_skb(info, skb_len, + src_cid, src_port, + dst_cid, dst_port); + if (!skb) { + ret = -ENOMEM; + break; + } - virtio_transport_inc_tx_pkt(vvs, skb); + virtio_transport_inc_tx_pkt(vvs, skb); - return t_ops->send_pkt(skb); + ret = t_ops->send_pkt(skb); + if (ret < 0) + break; + + /* Both virtio and vhost 'send_pkt()' returns 'skb_len', + * but for reliability use 'ret' instead of 'skb_len'. + * Also if partial send happens (e.g. 'ret' != 'skb_len') + * somehow, we break this loop, but account such returned + * value in 'virtio_transport_put_credit()'. + */ + rest_len -= ret; + + if (WARN_ONCE(ret != skb_len, + "'send_pkt()' returns %i, but %zu expected\n", + ret, skb_len)) + break; + } while (rest_len); + + virtio_transport_put_credit(vvs, rest_len); + + /* Return number of bytes, if any data has been sent. */ + if (rest_len != pkt_len) + ret = pkt_len - rest_len; + + return ret; } static bool virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs, -- cgit v1.2.3 From e3ec366eb0d1d5f7cbe28e747c76180f7d45f00d Mon Sep 17 00:00:00 2001 From: Arseniy Krasnov Date: Sun, 26 Mar 2023 01:04:50 +0300 Subject: virtio/vsock: check argument to avoid no effect call Both of these functions have no effect when input argument is 0, so to avoid useless spinlock access, check argument before it. Signed-off-by: Arseniy Krasnov Reviewed-by: Stefano Garzarella Signed-off-by: Paolo Abeni --- net/vmw_vsock/virtio_transport_common.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 8c60fc7f94a5..7fc178c3ee07 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -302,6 +302,9 @@ u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 credit) { u32 ret; + if (!credit) + return 0; + spin_lock_bh(&vvs->tx_lock); ret = vvs->peer_buf_alloc - (vvs->tx_cnt - vvs->peer_fwd_cnt); if (ret > credit) @@ -315,6 +318,9 @@ EXPORT_SYMBOL_GPL(virtio_transport_get_credit); void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit) { + if (!credit) + return; + spin_lock_bh(&vvs->tx_lock); vvs->tx_cnt -= credit; spin_unlock_bh(&vvs->tx_lock); -- cgit v1.2.3 From d288a162dd1c73507da582966f17dd226e34a0c0 Mon Sep 17 00:00:00 2001 From: Wangyang Guo Date: Thu, 23 Mar 2023 21:55:29 +0100 Subject: net: dst: Prevent false sharing vs. dst_entry:: __refcnt dst_entry::__refcnt is highly contended in scenarios where many connections happen from and to the same IP. The reference count is an atomic_t, so the reference count operations have to take the cache-line exclusive. Aside of the unavoidable reference count contention there is another significant problem which is caused by that: False sharing. perf top identified two affected read accesses. dst_entry::lwtstate and rtable::rt_genid. dst_entry:__refcnt is located at offset 64 of dst_entry, which puts it into a seperate cacheline vs. the read mostly members located at the beginning of the struct. That prevents false sharing vs. the struct members in the first 64 bytes of the structure, but there is also dst_entry::lwtstate which is located after the reference count and in the same cache line. This member is read after a reference count has been acquired. struct rtable embeds a struct dst_entry at offset 0. struct dst_entry has a size of 112 bytes, which means that the struct members of rtable which follow the dst member share the same cache line as dst_entry::__refcnt. Especially rtable::rt_genid is also read by the contexts which have a reference count acquired already. When dst_entry:__refcnt is incremented or decremented via an atomic operation these read accesses stall. This was found when analysing the memtier benchmark in 1:100 mode, which amplifies the problem extremly. Move the rt[6i]_uncached[_list] members out of struct rtable and struct rt6_info into struct dst_entry to provide padding and move the lwtstate member after that so it ends up in the same cache line. The resulting improvement depends on the micro-architecture and the number of CPUs. It ranges from +20% to +120% with a localhost memtier/memcached benchmark. [ tglx: Rearrange struct ] Signed-off-by: Wangyang Guo Signed-off-by: Arjan van de Ven Signed-off-by: Thomas Gleixner Reviewed-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20230323102800.042297517@linutronix.de Signed-off-by: Jakub Kicinski --- include/net/dst.h | 15 ++++++++++++++- include/net/ip6_fib.h | 3 --- include/net/ip6_route.h | 2 +- include/net/route.h | 3 --- net/ipv4/route.c | 20 ++++++++++---------- net/ipv4/xfrm4_policy.c | 4 ++-- net/ipv6/route.c | 26 +++++++++++++------------- net/ipv6/xfrm6_policy.c | 4 ++-- 8 files changed, 42 insertions(+), 35 deletions(-) (limited to 'net') diff --git a/include/net/dst.h b/include/net/dst.h index d67fda89cd0f..81f2279ea911 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -69,15 +69,28 @@ struct dst_entry { #endif int __use; unsigned long lastuse; - struct lwtunnel_state *lwtstate; struct rcu_head rcu_head; short error; short __pad; __u32 tclassid; #ifndef CONFIG_64BIT + struct lwtunnel_state *lwtstate; atomic_t __refcnt; /* 32-bit offset 64 */ #endif netdevice_tracker dev_tracker; + + /* + * Used by rtable and rt6_info. Moves lwtstate into the next cache + * line on 64bit so that lwtstate does not cause false sharing with + * __refcnt under contention of __refcnt. This also puts the + * frequently accessed members of rtable and rt6_info out of the + * __refcnt cache line. + */ + struct list_head rt_uncached; + struct uncached_list *rt_uncached_list; +#ifdef CONFIG_64BIT + struct lwtunnel_state *lwtstate; +#endif }; struct dst_metrics { diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 6268963d9599..79570cb4ea9c 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -217,9 +217,6 @@ struct rt6_info { struct inet6_dev *rt6i_idev; u32 rt6i_flags; - struct list_head rt6i_uncached; - struct uncached_list *rt6i_uncached_list; - /* more non-fragment space at head required */ unsigned short rt6i_nfheader_len; }; diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 81ee387a1fc4..3556595ce59a 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -100,7 +100,7 @@ static inline struct dst_entry *ip6_route_output(struct net *net, static inline void ip6_rt_put_flags(struct rt6_info *rt, int flags) { if (!(flags & RT6_LOOKUP_F_DST_NOREF) || - !list_empty(&rt->rt6i_uncached)) + !list_empty(&rt->dst.rt_uncached)) ip6_rt_put(rt); } diff --git a/include/net/route.h b/include/net/route.h index fe00b0a2e475..bcc367cf3aa2 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -78,9 +78,6 @@ struct rtable { /* Miscellaneous cached information */ u32 rt_mtu_locked:1, rt_pmtu:31; - - struct list_head rt_uncached; - struct uncached_list *rt_uncached_list; }; static inline bool rt_is_input_route(const struct rtable *rt) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 6a0a0bb452e9..2a3d14d95ada 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1508,20 +1508,20 @@ void rt_add_uncached_list(struct rtable *rt) { struct uncached_list *ul = raw_cpu_ptr(&rt_uncached_list); - rt->rt_uncached_list = ul; + rt->dst.rt_uncached_list = ul; spin_lock_bh(&ul->lock); - list_add_tail(&rt->rt_uncached, &ul->head); + list_add_tail(&rt->dst.rt_uncached, &ul->head); spin_unlock_bh(&ul->lock); } void rt_del_uncached_list(struct rtable *rt) { - if (!list_empty(&rt->rt_uncached)) { - struct uncached_list *ul = rt->rt_uncached_list; + if (!list_empty(&rt->dst.rt_uncached)) { + struct uncached_list *ul = rt->dst.rt_uncached_list; spin_lock_bh(&ul->lock); - list_del_init(&rt->rt_uncached); + list_del_init(&rt->dst.rt_uncached); spin_unlock_bh(&ul->lock); } } @@ -1546,13 +1546,13 @@ void rt_flush_dev(struct net_device *dev) continue; spin_lock_bh(&ul->lock); - list_for_each_entry_safe(rt, safe, &ul->head, rt_uncached) { + list_for_each_entry_safe(rt, safe, &ul->head, dst.rt_uncached) { if (rt->dst.dev != dev) continue; rt->dst.dev = blackhole_netdev; netdev_ref_replace(dev, blackhole_netdev, &rt->dst.dev_tracker, GFP_ATOMIC); - list_move(&rt->rt_uncached, &ul->quarantine); + list_move(&rt->dst.rt_uncached, &ul->quarantine); } spin_unlock_bh(&ul->lock); } @@ -1644,7 +1644,7 @@ struct rtable *rt_dst_alloc(struct net_device *dev, rt->rt_uses_gateway = 0; rt->rt_gw_family = 0; rt->rt_gw4 = 0; - INIT_LIST_HEAD(&rt->rt_uncached); + INIT_LIST_HEAD(&rt->dst.rt_uncached); rt->dst.output = ip_output; if (flags & RTCF_LOCAL) @@ -1675,7 +1675,7 @@ struct rtable *rt_dst_clone(struct net_device *dev, struct rtable *rt) new_rt->rt_gw4 = rt->rt_gw4; else if (rt->rt_gw_family == AF_INET6) new_rt->rt_gw6 = rt->rt_gw6; - INIT_LIST_HEAD(&new_rt->rt_uncached); + INIT_LIST_HEAD(&new_rt->dst.rt_uncached); new_rt->dst.input = rt->dst.input; new_rt->dst.output = rt->dst.output; @@ -2859,7 +2859,7 @@ struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_or else if (rt->rt_gw_family == AF_INET6) rt->rt_gw6 = ort->rt_gw6; - INIT_LIST_HEAD(&rt->rt_uncached); + INIT_LIST_HEAD(&rt->dst.rt_uncached); } dst_release(dst_orig); diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 3d0dfa6cf9f9..47861c8b7340 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -91,7 +91,7 @@ static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev, xdst->u.rt.rt_gw6 = rt->rt_gw6; xdst->u.rt.rt_pmtu = rt->rt_pmtu; xdst->u.rt.rt_mtu_locked = rt->rt_mtu_locked; - INIT_LIST_HEAD(&xdst->u.rt.rt_uncached); + INIT_LIST_HEAD(&xdst->u.rt.dst.rt_uncached); rt_add_uncached_list(&xdst->u.rt); return 0; @@ -121,7 +121,7 @@ static void xfrm4_dst_destroy(struct dst_entry *dst) struct xfrm_dst *xdst = (struct xfrm_dst *)dst; dst_destroy_metrics_generic(dst); - if (xdst->u.rt.rt_uncached_list) + if (xdst->u.rt.dst.rt_uncached_list) rt_del_uncached_list(&xdst->u.rt); xfrm_dst_destroy(xdst); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 244df77fac87..12e163dec34e 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -139,20 +139,20 @@ void rt6_uncached_list_add(struct rt6_info *rt) { struct uncached_list *ul = raw_cpu_ptr(&rt6_uncached_list); - rt->rt6i_uncached_list = ul; + rt->dst.rt_uncached_list = ul; spin_lock_bh(&ul->lock); - list_add_tail(&rt->rt6i_uncached, &ul->head); + list_add_tail(&rt->dst.rt_uncached, &ul->head); spin_unlock_bh(&ul->lock); } void rt6_uncached_list_del(struct rt6_info *rt) { - if (!list_empty(&rt->rt6i_uncached)) { - struct uncached_list *ul = rt->rt6i_uncached_list; + if (!list_empty(&rt->dst.rt_uncached)) { + struct uncached_list *ul = rt->dst.rt_uncached_list; spin_lock_bh(&ul->lock); - list_del_init(&rt->rt6i_uncached); + list_del_init(&rt->dst.rt_uncached); spin_unlock_bh(&ul->lock); } } @@ -169,7 +169,7 @@ static void rt6_uncached_list_flush_dev(struct net_device *dev) continue; spin_lock_bh(&ul->lock); - list_for_each_entry_safe(rt, safe, &ul->head, rt6i_uncached) { + list_for_each_entry_safe(rt, safe, &ul->head, dst.rt_uncached) { struct inet6_dev *rt_idev = rt->rt6i_idev; struct net_device *rt_dev = rt->dst.dev; bool handled = false; @@ -188,7 +188,7 @@ static void rt6_uncached_list_flush_dev(struct net_device *dev) handled = true; } if (handled) - list_move(&rt->rt6i_uncached, + list_move(&rt->dst.rt_uncached, &ul->quarantine); } spin_unlock_bh(&ul->lock); @@ -334,7 +334,7 @@ static const struct rt6_info ip6_blk_hole_entry_template = { static void rt6_info_init(struct rt6_info *rt) { memset_after(rt, 0, dst); - INIT_LIST_HEAD(&rt->rt6i_uncached); + INIT_LIST_HEAD(&rt->dst.rt_uncached); } /* allocate dst with ip6_dst_ops */ @@ -2638,7 +2638,7 @@ struct dst_entry *ip6_route_output_flags(struct net *net, dst = ip6_route_output_flags_noref(net, sk, fl6, flags); rt6 = (struct rt6_info *)dst; /* For dst cached in uncached_list, refcnt is already taken. */ - if (list_empty(&rt6->rt6i_uncached) && !dst_hold_safe(dst)) { + if (list_empty(&rt6->dst.rt_uncached) && !dst_hold_safe(dst)) { dst = &net->ipv6.ip6_null_entry->dst; dst_hold(dst); } @@ -2748,7 +2748,7 @@ INDIRECT_CALLABLE_SCOPE struct dst_entry *ip6_dst_check(struct dst_entry *dst, from = rcu_dereference(rt->from); if (from && (rt->rt6i_flags & RTF_PCPU || - unlikely(!list_empty(&rt->rt6i_uncached)))) + unlikely(!list_empty(&rt->dst.rt_uncached)))) dst_ret = rt6_dst_from_check(rt, from, cookie); else dst_ret = rt6_check(rt, from, cookie); @@ -6477,7 +6477,7 @@ static int __net_init ip6_route_net_init(struct net *net) net->ipv6.ip6_null_entry->dst.ops = &net->ipv6.ip6_dst_ops; dst_init_metrics(&net->ipv6.ip6_null_entry->dst, ip6_template_metrics, true); - INIT_LIST_HEAD(&net->ipv6.ip6_null_entry->rt6i_uncached); + INIT_LIST_HEAD(&net->ipv6.ip6_null_entry->dst.rt_uncached); #ifdef CONFIG_IPV6_MULTIPLE_TABLES net->ipv6.fib6_has_custom_rules = false; @@ -6489,7 +6489,7 @@ static int __net_init ip6_route_net_init(struct net *net) net->ipv6.ip6_prohibit_entry->dst.ops = &net->ipv6.ip6_dst_ops; dst_init_metrics(&net->ipv6.ip6_prohibit_entry->dst, ip6_template_metrics, true); - INIT_LIST_HEAD(&net->ipv6.ip6_prohibit_entry->rt6i_uncached); + INIT_LIST_HEAD(&net->ipv6.ip6_prohibit_entry->dst.rt_uncached); net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template, sizeof(*net->ipv6.ip6_blk_hole_entry), @@ -6499,7 +6499,7 @@ static int __net_init ip6_route_net_init(struct net *net) net->ipv6.ip6_blk_hole_entry->dst.ops = &net->ipv6.ip6_dst_ops; dst_init_metrics(&net->ipv6.ip6_blk_hole_entry->dst, ip6_template_metrics, true); - INIT_LIST_HEAD(&net->ipv6.ip6_blk_hole_entry->rt6i_uncached); + INIT_LIST_HEAD(&net->ipv6.ip6_blk_hole_entry->dst.rt_uncached); #ifdef CONFIG_IPV6_SUBTREES net->ipv6.fib6_routes_require_src = 0; #endif diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index ea435eba3053..2b493f8d0091 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -89,7 +89,7 @@ static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev, xdst->u.rt6.rt6i_gateway = rt->rt6i_gateway; xdst->u.rt6.rt6i_dst = rt->rt6i_dst; xdst->u.rt6.rt6i_src = rt->rt6i_src; - INIT_LIST_HEAD(&xdst->u.rt6.rt6i_uncached); + INIT_LIST_HEAD(&xdst->u.rt6.dst.rt_uncached); rt6_uncached_list_add(&xdst->u.rt6); return 0; @@ -121,7 +121,7 @@ static void xfrm6_dst_destroy(struct dst_entry *dst) if (likely(xdst->u.rt6.rt6i_idev)) in6_dev_put(xdst->u.rt6.rt6i_idev); dst_destroy_metrics_generic(dst); - if (xdst->u.rt6.rt6i_uncached_list) + if (xdst->u.rt6.dst.rt_uncached_list) rt6_uncached_list_del(&xdst->u.rt6); xfrm_dst_destroy(xdst); } -- cgit v1.2.3 From bc9d3a9f2afca189a6ae40225b6985e3c775375e Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 23 Mar 2023 21:55:32 +0100 Subject: net: dst: Switch to rcuref_t reference counting Under high contention dst_entry::__refcnt becomes a significant bottleneck. atomic_inc_not_zero() is implemented with a cmpxchg() loop, which goes into high retry rates on contention. Switch the reference count to rcuref_t which results in a significant performance gain. Rename the reference count member to __rcuref to reflect the change. The gain depends on the micro-architecture and the number of concurrent operations and has been measured in the range of +25% to +130% with a localhost memtier/memcached benchmark which amplifies the problem massively. Running the memtier/memcached benchmark over a real (1Gb) network connection the conversion on top of the false sharing fix for struct dst_entry::__refcnt results in a total gain in the 2%-5% range over the upstream baseline. Reported-by: Wangyang Guo Reported-by: Arjan Van De Ven Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/20230307125538.989175656@linutronix.de Link: https://lore.kernel.org/r/20230323102800.215027837@linutronix.de Signed-off-by: Jakub Kicinski --- include/net/dst.h | 19 ++++++++++--------- include/net/sock.h | 2 +- net/bridge/br_nf_core.c | 2 +- net/core/dst.c | 26 +++++--------------------- net/core/rtnetlink.c | 2 +- net/ipv6/route.c | 6 +++--- net/netfilter/ipvs/ip_vs_xmit.c | 4 ++-- 7 files changed, 23 insertions(+), 38 deletions(-) (limited to 'net') diff --git a/include/net/dst.h b/include/net/dst.h index 81f2279ea911..78884429deed 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -61,11 +62,11 @@ struct dst_entry { unsigned short trailer_len; /* space to reserve at tail */ /* - * __refcnt wants to be on a different cache line from + * __rcuref wants to be on a different cache line from * input/output/ops or performance tanks badly */ #ifdef CONFIG_64BIT - atomic_t __refcnt; /* 64-bit offset 64 */ + rcuref_t __rcuref; /* 64-bit offset 64 */ #endif int __use; unsigned long lastuse; @@ -75,16 +76,16 @@ struct dst_entry { __u32 tclassid; #ifndef CONFIG_64BIT struct lwtunnel_state *lwtstate; - atomic_t __refcnt; /* 32-bit offset 64 */ + rcuref_t __rcuref; /* 32-bit offset 64 */ #endif netdevice_tracker dev_tracker; /* * Used by rtable and rt6_info. Moves lwtstate into the next cache * line on 64bit so that lwtstate does not cause false sharing with - * __refcnt under contention of __refcnt. This also puts the + * __rcuref under contention of __rcuref. This also puts the * frequently accessed members of rtable and rt6_info out of the - * __refcnt cache line. + * __rcuref cache line. */ struct list_head rt_uncached; struct uncached_list *rt_uncached_list; @@ -238,10 +239,10 @@ static inline void dst_hold(struct dst_entry *dst) { /* * If your kernel compilation stops here, please check - * the placement of __refcnt in struct dst_entry + * the placement of __rcuref in struct dst_entry */ - BUILD_BUG_ON(offsetof(struct dst_entry, __refcnt) & 63); - WARN_ON(atomic_inc_not_zero(&dst->__refcnt) == 0); + BUILD_BUG_ON(offsetof(struct dst_entry, __rcuref) & 63); + WARN_ON(!rcuref_get(&dst->__rcuref)); } static inline void dst_use_noref(struct dst_entry *dst, unsigned long time) @@ -305,7 +306,7 @@ static inline void skb_dst_copy(struct sk_buff *nskb, const struct sk_buff *oskb */ static inline bool dst_hold_safe(struct dst_entry *dst) { - return atomic_inc_not_zero(&dst->__refcnt); + return rcuref_get(&dst->__rcuref); } /** diff --git a/include/net/sock.h b/include/net/sock.h index 573f2bf7e0de..5edf0038867c 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2131,7 +2131,7 @@ sk_dst_get(struct sock *sk) rcu_read_lock(); dst = rcu_dereference(sk->sk_dst_cache); - if (dst && !atomic_inc_not_zero(&dst->__refcnt)) + if (dst && !rcuref_get(&dst->__rcuref)) dst = NULL; rcu_read_unlock(); return dst; diff --git a/net/bridge/br_nf_core.c b/net/bridge/br_nf_core.c index 8c69f0c95a8e..98aea5485aae 100644 --- a/net/bridge/br_nf_core.c +++ b/net/bridge/br_nf_core.c @@ -73,7 +73,7 @@ void br_netfilter_rtable_init(struct net_bridge *br) { struct rtable *rt = &br->fake_rtable; - atomic_set(&rt->dst.__refcnt, 1); + rcuref_init(&rt->dst.__rcuref, 1); rt->dst.dev = br->dev; dst_init_metrics(&rt->dst, br_dst_default_metrics, true); rt->dst.flags = DST_NOXFRM | DST_FAKE_RTABLE; diff --git a/net/core/dst.c b/net/core/dst.c index 31c08a3386d3..3247e84045ca 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -66,7 +66,7 @@ void dst_init(struct dst_entry *dst, struct dst_ops *ops, dst->tclassid = 0; #endif dst->lwtstate = NULL; - atomic_set(&dst->__refcnt, initial_ref); + rcuref_init(&dst->__rcuref, initial_ref); dst->__use = 0; dst->lastuse = jiffies; dst->flags = flags; @@ -162,31 +162,15 @@ EXPORT_SYMBOL(dst_dev_put); void dst_release(struct dst_entry *dst) { - if (dst) { - int newrefcnt; - - newrefcnt = atomic_dec_return(&dst->__refcnt); - if (WARN_ONCE(newrefcnt < 0, "dst_release underflow")) - net_warn_ratelimited("%s: dst:%p refcnt:%d\n", - __func__, dst, newrefcnt); - if (!newrefcnt) - call_rcu_hurry(&dst->rcu_head, dst_destroy_rcu); - } + if (dst && rcuref_put(&dst->__rcuref)) + call_rcu_hurry(&dst->rcu_head, dst_destroy_rcu); } EXPORT_SYMBOL(dst_release); void dst_release_immediate(struct dst_entry *dst) { - if (dst) { - int newrefcnt; - - newrefcnt = atomic_dec_return(&dst->__refcnt); - if (WARN_ONCE(newrefcnt < 0, "dst_release_immediate underflow")) - net_warn_ratelimited("%s: dst:%p refcnt:%d\n", - __func__, dst, newrefcnt); - if (!newrefcnt) - dst_destroy(dst); - } + if (dst && rcuref_put(&dst->__rcuref)) + dst_destroy(dst); } EXPORT_SYMBOL(dst_release_immediate); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index b7b1661d0d56..906aebdc566b 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -843,7 +843,7 @@ int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id, if (dst) { ci.rta_lastuse = jiffies_delta_to_clock_t(jiffies - dst->lastuse); ci.rta_used = dst->__use; - ci.rta_clntref = atomic_read(&dst->__refcnt); + ci.rta_clntref = rcuref_read(&dst->__rcuref); } if (expires) { unsigned long clock; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 12e163dec34e..35085fc0cf15 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -293,7 +293,7 @@ static const struct fib6_info fib6_null_entry_template = { static const struct rt6_info ip6_null_entry_template = { .dst = { - .__refcnt = ATOMIC_INIT(1), + .__rcuref = RCUREF_INIT(1), .__use = 1, .obsolete = DST_OBSOLETE_FORCE_CHK, .error = -ENETUNREACH, @@ -307,7 +307,7 @@ static const struct rt6_info ip6_null_entry_template = { static const struct rt6_info ip6_prohibit_entry_template = { .dst = { - .__refcnt = ATOMIC_INIT(1), + .__rcuref = RCUREF_INIT(1), .__use = 1, .obsolete = DST_OBSOLETE_FORCE_CHK, .error = -EACCES, @@ -319,7 +319,7 @@ static const struct rt6_info ip6_prohibit_entry_template = { static const struct rt6_info ip6_blk_hole_entry_template = { .dst = { - .__refcnt = ATOMIC_INIT(1), + .__rcuref = RCUREF_INIT(1), .__use = 1, .obsolete = DST_OBSOLETE_FORCE_CHK, .error = -EINVAL, diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 80448885c3d7..99c349c0d968 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -339,7 +339,7 @@ __ip_vs_get_out_rt(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, spin_unlock_bh(&dest->dst_lock); IP_VS_DBG(10, "new dst %pI4, src %pI4, refcnt=%d\n", &dest->addr.ip, &dest_dst->dst_saddr.ip, - atomic_read(&rt->dst.__refcnt)); + rcuref_read(&rt->dst.__rcuref)); } if (ret_saddr) *ret_saddr = dest_dst->dst_saddr.ip; @@ -507,7 +507,7 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, spin_unlock_bh(&dest->dst_lock); IP_VS_DBG(10, "new dst %pI6, src %pI6, refcnt=%d\n", &dest->addr.in6, &dest_dst->dst_saddr.in6, - atomic_read(&rt->dst.__refcnt)); + rcuref_read(&rt->dst.__rcuref)); } if (ret_saddr) *ret_saddr = dest_dst->dst_saddr.in6; -- cgit v1.2.3 From 634f1a7110b439c65fd8a809171c1d2d28bcea6f Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Mon, 27 Mar 2023 19:11:51 +0000 Subject: vsock: support sockmap This patch adds sockmap support for vsock sockets. It is intended to be usable by all transports, but only the virtio and loopback transports are implemented. SOCK_STREAM, SOCK_DGRAM, and SOCK_SEQPACKET are all supported. Signed-off-by: Bobby Eshleman Acked-by: Michael S. Tsirkin Reviewed-by: Stefano Garzarella Signed-off-by: David S. Miller --- drivers/vhost/vsock.c | 1 + include/linux/virtio_vsock.h | 1 + include/net/af_vsock.h | 17 ++++ net/vmw_vsock/Makefile | 1 + net/vmw_vsock/af_vsock.c | 64 ++++++++++-- net/vmw_vsock/virtio_transport.c | 2 + net/vmw_vsock/virtio_transport_common.c | 25 +++++ net/vmw_vsock/vsock_bpf.c | 174 ++++++++++++++++++++++++++++++++ net/vmw_vsock/vsock_loopback.c | 2 + 9 files changed, 281 insertions(+), 6 deletions(-) create mode 100644 net/vmw_vsock/vsock_bpf.c (limited to 'net') diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c index c8e6087769a1..6578db78f0ae 100644 --- a/drivers/vhost/vsock.c +++ b/drivers/vhost/vsock.c @@ -439,6 +439,7 @@ static struct virtio_transport vhost_transport = { .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue, .notify_buffer_size = virtio_transport_notify_buffer_size, + .read_skb = virtio_transport_read_skb, }, .send_pkt = vhost_transport_send_pkt, diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h index 3f9c16611306..c58453699ee9 100644 --- a/include/linux/virtio_vsock.h +++ b/include/linux/virtio_vsock.h @@ -245,4 +245,5 @@ u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 wanted); void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit); void virtio_transport_deliver_tap_pkt(struct sk_buff *skb); int virtio_transport_purge_skbs(void *vsk, struct sk_buff_head *list); +int virtio_transport_read_skb(struct vsock_sock *vsk, skb_read_actor_t read_actor); #endif /* _LINUX_VIRTIO_VSOCK_H */ diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index 568a87c5e0d0..0e7504a42925 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -75,6 +75,7 @@ struct vsock_sock { void *trans; }; +s64 vsock_connectible_has_data(struct vsock_sock *vsk); s64 vsock_stream_has_data(struct vsock_sock *vsk); s64 vsock_stream_has_space(struct vsock_sock *vsk); struct sock *vsock_create_connected(struct sock *parent); @@ -173,6 +174,9 @@ struct vsock_transport { /* Addressing. */ u32 (*get_local_cid)(void); + + /* Read a single skb */ + int (*read_skb)(struct vsock_sock *, skb_read_actor_t); }; /**** CORE ****/ @@ -225,5 +229,18 @@ int vsock_init_tap(void); int vsock_add_tap(struct vsock_tap *vt); int vsock_remove_tap(struct vsock_tap *vt); void vsock_deliver_tap(struct sk_buff *build_skb(void *opaque), void *opaque); +int vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, + int flags); +int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg, + size_t len, int flags); + +#ifdef CONFIG_BPF_SYSCALL +extern struct proto vsock_proto; +int vsock_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore); +void __init vsock_bpf_build_proto(void); +#else +static inline void __init vsock_bpf_build_proto(void) +{} +#endif #endif /* __AF_VSOCK_H__ */ diff --git a/net/vmw_vsock/Makefile b/net/vmw_vsock/Makefile index 6a943ec95c4a..5da74c4a9f1d 100644 --- a/net/vmw_vsock/Makefile +++ b/net/vmw_vsock/Makefile @@ -8,6 +8,7 @@ obj-$(CONFIG_HYPERV_VSOCKETS) += hv_sock.o obj-$(CONFIG_VSOCKETS_LOOPBACK) += vsock_loopback.o vsock-y += af_vsock.o af_vsock_tap.o vsock_addr.o +vsock-$(CONFIG_BPF_SYSCALL) += vsock_bpf.o vsock_diag-y += diag.o diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 19aea7cba26e..5f2dda35c980 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -116,10 +116,13 @@ static void vsock_sk_destruct(struct sock *sk); static int vsock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb); /* Protocol family. */ -static struct proto vsock_proto = { +struct proto vsock_proto = { .name = "AF_VSOCK", .owner = THIS_MODULE, .obj_size = sizeof(struct vsock_sock), +#ifdef CONFIG_BPF_SYSCALL + .psock_update_sk_prot = vsock_bpf_update_proto, +#endif }; /* The default peer timeout indicates how long we will wait for a peer response @@ -865,7 +868,7 @@ s64 vsock_stream_has_data(struct vsock_sock *vsk) } EXPORT_SYMBOL_GPL(vsock_stream_has_data); -static s64 vsock_connectible_has_data(struct vsock_sock *vsk) +s64 vsock_connectible_has_data(struct vsock_sock *vsk) { struct sock *sk = sk_vsock(vsk); @@ -874,6 +877,7 @@ static s64 vsock_connectible_has_data(struct vsock_sock *vsk) else return vsock_stream_has_data(vsk); } +EXPORT_SYMBOL_GPL(vsock_connectible_has_data); s64 vsock_stream_has_space(struct vsock_sock *vsk) { @@ -1131,6 +1135,13 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock, return mask; } +static int vsock_read_skb(struct sock *sk, skb_read_actor_t read_actor) +{ + struct vsock_sock *vsk = vsock_sk(sk); + + return vsk->transport->read_skb(vsk, read_actor); +} + static int vsock_dgram_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) { @@ -1242,18 +1253,42 @@ static int vsock_dgram_connect(struct socket *sock, memcpy(&vsk->remote_addr, remote_addr, sizeof(vsk->remote_addr)); sock->state = SS_CONNECTED; + /* sock map disallows redirection of non-TCP sockets with sk_state != + * TCP_ESTABLISHED (see sock_map_redirect_allowed()), so we set + * TCP_ESTABLISHED here to allow redirection of connected vsock dgrams. + * + * This doesn't seem to be abnormal state for datagram sockets, as the + * same approach can be see in other datagram socket types as well + * (such as unix sockets). + */ + sk->sk_state = TCP_ESTABLISHED; + out: release_sock(sk); return err; } -static int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg, - size_t len, int flags) +int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg, + size_t len, int flags) { - struct vsock_sock *vsk = vsock_sk(sock->sk); +#ifdef CONFIG_BPF_SYSCALL + const struct proto *prot; +#endif + struct vsock_sock *vsk; + struct sock *sk; + + sk = sock->sk; + vsk = vsock_sk(sk); + +#ifdef CONFIG_BPF_SYSCALL + prot = READ_ONCE(sk->sk_prot); + if (prot != &vsock_proto) + return prot->recvmsg(sk, msg, len, flags, NULL); +#endif return vsk->transport->dgram_dequeue(vsk, msg, len, flags); } +EXPORT_SYMBOL_GPL(vsock_dgram_recvmsg); static const struct proto_ops vsock_dgram_ops = { .family = PF_VSOCK, @@ -1272,6 +1307,7 @@ static const struct proto_ops vsock_dgram_ops = { .recvmsg = vsock_dgram_recvmsg, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage, + .read_skb = vsock_read_skb, }; static int vsock_transport_cancel_pkt(struct vsock_sock *vsk) @@ -2086,13 +2122,16 @@ out: return err; } -static int +int vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, int flags) { struct sock *sk; struct vsock_sock *vsk; const struct vsock_transport *transport; +#ifdef CONFIG_BPF_SYSCALL + const struct proto *prot; +#endif int err; sk = sock->sk; @@ -2139,6 +2178,14 @@ vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, goto out; } +#ifdef CONFIG_BPF_SYSCALL + prot = READ_ONCE(sk->sk_prot); + if (prot != &vsock_proto) { + release_sock(sk); + return prot->recvmsg(sk, msg, len, flags, NULL); + } +#endif + if (sk->sk_type == SOCK_STREAM) err = __vsock_stream_recvmsg(sk, msg, len, flags); else @@ -2148,6 +2195,7 @@ out: release_sock(sk); return err; } +EXPORT_SYMBOL_GPL(vsock_connectible_recvmsg); static int vsock_set_rcvlowat(struct sock *sk, int val) { @@ -2188,6 +2236,7 @@ static const struct proto_ops vsock_stream_ops = { .mmap = sock_no_mmap, .sendpage = sock_no_sendpage, .set_rcvlowat = vsock_set_rcvlowat, + .read_skb = vsock_read_skb, }; static const struct proto_ops vsock_seqpacket_ops = { @@ -2209,6 +2258,7 @@ static const struct proto_ops vsock_seqpacket_ops = { .recvmsg = vsock_connectible_recvmsg, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage, + .read_skb = vsock_read_skb, }; static int vsock_create(struct net *net, struct socket *sock, @@ -2348,6 +2398,8 @@ static int __init vsock_init(void) goto err_unregister_proto; } + vsock_bpf_build_proto(); + return 0; err_unregister_proto: diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c index 28b5a8e8e094..e95df847176b 100644 --- a/net/vmw_vsock/virtio_transport.c +++ b/net/vmw_vsock/virtio_transport.c @@ -457,6 +457,8 @@ static struct virtio_transport virtio_transport = { .notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue, .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue, .notify_buffer_size = virtio_transport_notify_buffer_size, + + .read_skb = virtio_transport_read_skb, }, .send_pkt = virtio_transport_send_pkt, diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 7fc178c3ee07..f39639dd6eb5 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -1418,6 +1418,31 @@ int virtio_transport_purge_skbs(void *vsk, struct sk_buff_head *queue) } EXPORT_SYMBOL_GPL(virtio_transport_purge_skbs); +int virtio_transport_read_skb(struct vsock_sock *vsk, skb_read_actor_t recv_actor) +{ + struct virtio_vsock_sock *vvs = vsk->trans; + struct sock *sk = sk_vsock(vsk); + struct sk_buff *skb; + int off = 0; + int copied; + int err; + + spin_lock_bh(&vvs->rx_lock); + /* Use __skb_recv_datagram() for race-free handling of the receive. It + * works for types other than dgrams. + */ + skb = __skb_recv_datagram(sk, &vvs->rx_queue, MSG_DONTWAIT, &off, &err); + spin_unlock_bh(&vvs->rx_lock); + + if (!skb) + return err; + + copied = recv_actor(sk, skb); + kfree_skb(skb); + return copied; +} +EXPORT_SYMBOL_GPL(virtio_transport_read_skb); + MODULE_LICENSE("GPL v2"); MODULE_AUTHOR("Asias He"); MODULE_DESCRIPTION("common code for virtio vsock"); diff --git a/net/vmw_vsock/vsock_bpf.c b/net/vmw_vsock/vsock_bpf.c new file mode 100644 index 000000000000..a3c97546ab84 --- /dev/null +++ b/net/vmw_vsock/vsock_bpf.c @@ -0,0 +1,174 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2022 Bobby Eshleman + * + * Based off of net/unix/unix_bpf.c + */ + +#include +#include +#include +#include +#include +#include +#include + +#define vsock_sk_has_data(__sk, __psock) \ + ({ !skb_queue_empty(&(__sk)->sk_receive_queue) || \ + !skb_queue_empty(&(__psock)->ingress_skb) || \ + !list_empty(&(__psock)->ingress_msg); \ + }) + +static struct proto *vsock_prot_saved __read_mostly; +static DEFINE_SPINLOCK(vsock_prot_lock); +static struct proto vsock_bpf_prot; + +static bool vsock_has_data(struct sock *sk, struct sk_psock *psock) +{ + struct vsock_sock *vsk = vsock_sk(sk); + s64 ret; + + ret = vsock_connectible_has_data(vsk); + if (ret > 0) + return true; + + return vsock_sk_has_data(sk, psock); +} + +static bool vsock_msg_wait_data(struct sock *sk, struct sk_psock *psock, long timeo) +{ + bool ret; + + DEFINE_WAIT_FUNC(wait, woken_wake_function); + + if (sk->sk_shutdown & RCV_SHUTDOWN) + return true; + + if (!timeo) + return false; + + add_wait_queue(sk_sleep(sk), &wait); + sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); + ret = vsock_has_data(sk, psock); + if (!ret) { + wait_woken(&wait, TASK_INTERRUPTIBLE, timeo); + ret = vsock_has_data(sk, psock); + } + sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk); + remove_wait_queue(sk_sleep(sk), &wait); + return ret; +} + +static int __vsock_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags) +{ + struct socket *sock = sk->sk_socket; + int err; + + if (sk->sk_type == SOCK_STREAM || sk->sk_type == SOCK_SEQPACKET) + err = vsock_connectible_recvmsg(sock, msg, len, flags); + else if (sk->sk_type == SOCK_DGRAM) + err = vsock_dgram_recvmsg(sock, msg, len, flags); + else + err = -EPROTOTYPE; + + return err; +} + +static int vsock_bpf_recvmsg(struct sock *sk, struct msghdr *msg, + size_t len, int flags, int *addr_len) +{ + struct sk_psock *psock; + int copied; + + psock = sk_psock_get(sk); + if (unlikely(!psock)) + return __vsock_recvmsg(sk, msg, len, flags); + + lock_sock(sk); + if (vsock_has_data(sk, psock) && sk_psock_queue_empty(psock)) { + release_sock(sk); + sk_psock_put(sk, psock); + return __vsock_recvmsg(sk, msg, len, flags); + } + + copied = sk_msg_recvmsg(sk, psock, msg, len, flags); + while (copied == 0) { + long timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + + if (!vsock_msg_wait_data(sk, psock, timeo)) { + copied = -EAGAIN; + break; + } + + if (sk_psock_queue_empty(psock)) { + release_sock(sk); + sk_psock_put(sk, psock); + return __vsock_recvmsg(sk, msg, len, flags); + } + + copied = sk_msg_recvmsg(sk, psock, msg, len, flags); + } + + release_sock(sk); + sk_psock_put(sk, psock); + + return copied; +} + +/* Copy of original proto with updated sock_map methods */ +static struct proto vsock_bpf_prot = { + .close = sock_map_close, + .recvmsg = vsock_bpf_recvmsg, + .sock_is_readable = sk_msg_is_readable, + .unhash = sock_map_unhash, +}; + +static void vsock_bpf_rebuild_protos(struct proto *prot, const struct proto *base) +{ + *prot = *base; + prot->close = sock_map_close; + prot->recvmsg = vsock_bpf_recvmsg; + prot->sock_is_readable = sk_msg_is_readable; +} + +static void vsock_bpf_check_needs_rebuild(struct proto *ops) +{ + /* Paired with the smp_store_release() below. */ + if (unlikely(ops != smp_load_acquire(&vsock_prot_saved))) { + spin_lock_bh(&vsock_prot_lock); + if (likely(ops != vsock_prot_saved)) { + vsock_bpf_rebuild_protos(&vsock_bpf_prot, ops); + /* Make sure proto function pointers are updated before publishing the + * pointer to the struct. + */ + smp_store_release(&vsock_prot_saved, ops); + } + spin_unlock_bh(&vsock_prot_lock); + } +} + +int vsock_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore) +{ + struct vsock_sock *vsk; + + if (restore) { + sk->sk_write_space = psock->saved_write_space; + sock_replace_proto(sk, psock->sk_proto); + return 0; + } + + vsk = vsock_sk(sk); + if (!vsk->transport) + return -ENODEV; + + if (!vsk->transport->read_skb) + return -EOPNOTSUPP; + + vsock_bpf_check_needs_rebuild(psock->sk_proto); + sock_replace_proto(sk, &vsock_bpf_prot); + return 0; +} + +void __init vsock_bpf_build_proto(void) +{ + vsock_bpf_rebuild_protos(&vsock_bpf_prot, &vsock_proto); +} diff --git a/net/vmw_vsock/vsock_loopback.c b/net/vmw_vsock/vsock_loopback.c index 671e03240fc5..40753b661c13 100644 --- a/net/vmw_vsock/vsock_loopback.c +++ b/net/vmw_vsock/vsock_loopback.c @@ -94,6 +94,8 @@ static struct virtio_transport loopback_transport = { .notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue, .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue, .notify_buffer_size = virtio_transport_notify_buffer_size, + + .read_skb = virtio_transport_read_skb, }, .send_pkt = vsock_loopback_send_pkt, -- cgit v1.2.3 From 8cdc3223e78c43e1b60ea1c536a103e32fdca3c5 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 27 Mar 2023 16:54:54 -0700 Subject: ipv6: Remove in6addr_any alternatives. Some code defines the IPv6 wildcard address as a local variable and use it with memcmp() or ipv6_addr_equal(). Let's use in6addr_any and ipv6_addr_any() instead. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Mark Bloch Reviewed-by: David Ahern Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c | 5 ++--- include/net/ip6_fib.h | 9 +++------ include/trace/events/fib.h | 5 ++--- include/trace/events/fib6.h | 5 +---- net/ethtool/ioctl.c | 10 +++++----- net/ipv4/inet_hashtables.c | 11 ++++------- 6 files changed, 17 insertions(+), 28 deletions(-) (limited to 'net') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c index a108e73c9f66..20c2d2ecaf93 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c @@ -98,7 +98,6 @@ int mlx5e_tc_set_attr_rx_tun(struct mlx5e_tc_flow *flow, #if IS_ENABLED(CONFIG_INET) && IS_ENABLED(CONFIG_IPV6) else if (ip_version == 6) { int ipv6_size = MLX5_FLD_SZ_BYTES(ipv6_layout, ipv6); - struct in6_addr zerov6 = {}; daddr = MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6); @@ -106,8 +105,8 @@ int mlx5e_tc_set_attr_rx_tun(struct mlx5e_tc_flow *flow, outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6); memcpy(&tun_attr->dst_ip.v6, daddr, ipv6_size); memcpy(&tun_attr->src_ip.v6, saddr, ipv6_size); - if (!memcmp(&tun_attr->dst_ip.v6, &zerov6, sizeof(zerov6)) || - !memcmp(&tun_attr->src_ip.v6, &zerov6, sizeof(zerov6))) + if (ipv6_addr_any(&tun_attr->dst_ip.v6) || + ipv6_addr_any(&tun_attr->src_ip.v6)) return 0; } #endif diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 79570cb4ea9c..05e6f756feaf 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -469,13 +469,10 @@ void rt6_get_prefsrc(const struct rt6_info *rt, struct in6_addr *addr) rcu_read_lock(); from = rcu_dereference(rt->from); - if (from) { + if (from) *addr = from->fib6_prefsrc.addr; - } else { - struct in6_addr in6_zero = {}; - - *addr = in6_zero; - } + else + *addr = in6addr_any; rcu_read_unlock(); } diff --git a/include/trace/events/fib.h b/include/trace/events/fib.h index c2300c407f58..76297ecd4935 100644 --- a/include/trace/events/fib.h +++ b/include/trace/events/fib.h @@ -36,7 +36,6 @@ TRACE_EVENT(fib_table_lookup, ), TP_fast_assign( - struct in6_addr in6_zero = {}; struct net_device *dev; struct in6_addr *in6; __be32 *p32; @@ -74,7 +73,7 @@ TRACE_EVENT(fib_table_lookup, *p32 = nhc->nhc_gw.ipv4; in6 = (struct in6_addr *)__entry->gw6; - *in6 = in6_zero; + *in6 = in6addr_any; } else if (nhc->nhc_gw_family == AF_INET6) { p32 = (__be32 *) __entry->gw4; *p32 = 0; @@ -87,7 +86,7 @@ TRACE_EVENT(fib_table_lookup, *p32 = 0; in6 = (struct in6_addr *)__entry->gw6; - *in6 = in6_zero; + *in6 = in6addr_any; } ), diff --git a/include/trace/events/fib6.h b/include/trace/events/fib6.h index 6e821eb79450..4d3e607b3cde 100644 --- a/include/trace/events/fib6.h +++ b/include/trace/events/fib6.h @@ -68,11 +68,8 @@ TRACE_EVENT(fib6_table_lookup, strcpy(__entry->name, "-"); } if (res->f6i == net->ipv6.fib6_null_entry) { - struct in6_addr in6_zero = {}; - in6 = (struct in6_addr *)__entry->gw; - *in6 = in6_zero; - + *in6 = in6addr_any; } else if (res->nh) { in6 = (struct in6_addr *)__entry->gw; *in6 = res->nh->fib_nh_gw6; diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c index 646b3e490c71..59adc4e6e9ee 100644 --- a/net/ethtool/ioctl.c +++ b/net/ethtool/ioctl.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include @@ -3127,7 +3128,6 @@ struct ethtool_rx_flow_rule * ethtool_rx_flow_rule_create(const struct ethtool_rx_flow_spec_input *input) { const struct ethtool_rx_flow_spec *fs = input->fs; - static struct in6_addr zero_addr = {}; struct ethtool_rx_flow_match *match; struct ethtool_rx_flow_rule *flow; struct flow_action_entry *act; @@ -3233,20 +3233,20 @@ ethtool_rx_flow_rule_create(const struct ethtool_rx_flow_spec_input *input) v6_spec = &fs->h_u.tcp_ip6_spec; v6_m_spec = &fs->m_u.tcp_ip6_spec; - if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr))) { + if (!ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6src)) { memcpy(&match->key.ipv6.src, v6_spec->ip6src, sizeof(match->key.ipv6.src)); memcpy(&match->mask.ipv6.src, v6_m_spec->ip6src, sizeof(match->mask.ipv6.src)); } - if (memcmp(v6_m_spec->ip6dst, &zero_addr, sizeof(zero_addr))) { + if (!ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6dst)) { memcpy(&match->key.ipv6.dst, v6_spec->ip6dst, sizeof(match->key.ipv6.dst)); memcpy(&match->mask.ipv6.dst, v6_m_spec->ip6dst, sizeof(match->mask.ipv6.dst)); } - if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr)) || - memcmp(v6_m_spec->ip6dst, &zero_addr, sizeof(zero_addr))) { + if (!ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6src) || + !ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6dst)) { match->dissector.used_keys |= BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS); match->dissector.offset[FLOW_DISSECTOR_KEY_IPV6_ADDRS] = diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 6edae3886885..e7391bf310a7 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -826,13 +826,11 @@ bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const unsigned short port, int l3mdev, const struct sock *sk) { #if IS_ENABLED(CONFIG_IPV6) - struct in6_addr addr_any = {}; - if (sk->sk_family != tb->family) { if (sk->sk_family == AF_INET) return net_eq(ib2_net(tb), net) && tb->port == port && tb->l3mdev == l3mdev && - ipv6_addr_equal(&tb->v6_rcv_saddr, &addr_any); + ipv6_addr_any(&tb->v6_rcv_saddr); return false; } @@ -840,7 +838,7 @@ bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const if (sk->sk_family == AF_INET6) return net_eq(ib2_net(tb), net) && tb->port == port && tb->l3mdev == l3mdev && - ipv6_addr_equal(&tb->v6_rcv_saddr, &addr_any); + ipv6_addr_any(&tb->v6_rcv_saddr); else #endif return net_eq(ib2_net(tb), net) && tb->port == port && @@ -866,11 +864,10 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in { struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk); u32 hash; -#if IS_ENABLED(CONFIG_IPV6) - struct in6_addr addr_any = {}; +#if IS_ENABLED(CONFIG_IPV6) if (sk->sk_family == AF_INET6) - hash = ipv6_portaddr_hash(net, &addr_any, port); + hash = ipv6_portaddr_hash(net, &in6addr_any, port); else #endif hash = ipv4_portaddr_hash(net, 0, port); -- cgit v1.2.3 From be689c719eb6153b0fbd83536ae656dca8981f12 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 27 Mar 2023 16:54:55 -0700 Subject: 6lowpan: Remove redundant initialisation. We'll call memset(&tmp, 0, sizeof(tmp)) later. Signed-off-by: Kuniyuki Iwashima Signed-off-by: David S. Miller --- net/6lowpan/iphc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/6lowpan/iphc.c b/net/6lowpan/iphc.c index 52fad5dad9f7..e116d308a8df 100644 --- a/net/6lowpan/iphc.c +++ b/net/6lowpan/iphc.c @@ -848,7 +848,7 @@ static u8 lowpan_compress_ctx_addr(u8 **hc_ptr, const struct net_device *dev, const struct lowpan_iphc_ctx *ctx, const unsigned char *lladdr, bool sam) { - struct in6_addr tmp = {}; + struct in6_addr tmp; u8 dam; switch (lowpan_dev(dev)->lltype) { -- cgit v1.2.3 From 2bb9a37f0e194ed95c70603b0efc7898a5a0d9b4 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Mon, 27 Mar 2023 12:22:21 +0200 Subject: mptcp: avoid unneeded address copy In the syn_recv fallback path, the msk is unused. We can skip setting the socket address. Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/subflow.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index dadaf85db720..a11f4c525e01 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -821,8 +821,6 @@ create_child: goto dispose_child; } - if (new_msk) - mptcp_copy_inaddrs(new_msk, child); mptcp_subflow_drop_ctx(child); goto out; } -- cgit v1.2.3 From a88d0092b24b8cddce57fe0e88e60a9e29e0b515 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Mon, 27 Mar 2023 12:22:22 +0200 Subject: mptcp: simplify subflow_syn_recv_sock() Postpone the msk cloning to the child process creation so that we can avoid a bunch of conditionals. Link: https://github.com/multipath-tcp/mptcp_net-next/issues/61 Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/subflow.c | 41 +++++++++++++---------------------------- 1 file changed, 13 insertions(+), 28 deletions(-) (limited to 'net') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index a11f4c525e01..33dd27765116 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -696,14 +696,6 @@ static bool subflow_hmac_valid(const struct request_sock *req, return !crypto_memneq(hmac, mp_opt->hmac, MPTCPOPT_HMAC_LEN); } -static void mptcp_force_close(struct sock *sk) -{ - /* the msk is not yet exposed to user-space, and refcount is 2 */ - inet_sk_state_store(sk, TCP_CLOSE); - sk_common_release(sk); - sock_put(sk); -} - static void subflow_ulp_fallback(struct sock *sk, struct mptcp_subflow_context *old_ctx) { @@ -755,7 +747,6 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, struct mptcp_subflow_request_sock *subflow_req; struct mptcp_options_received mp_opt; bool fallback, fallback_is_fatal; - struct sock *new_msk = NULL; struct mptcp_sock *owner; struct sock *child; @@ -784,14 +775,9 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, * options. */ mptcp_get_options(skb, &mp_opt); - if (!(mp_opt.suboptions & OPTIONS_MPTCP_MPC)) { + if (!(mp_opt.suboptions & OPTIONS_MPTCP_MPC)) fallback = true; - goto create_child; - } - new_msk = mptcp_sk_clone(listener->conn, &mp_opt, req); - if (!new_msk) - fallback = true; } else if (subflow_req->mp_join) { mptcp_get_options(skb, &mp_opt); if (!(mp_opt.suboptions & OPTIONS_MPTCP_MPJ) || @@ -820,21 +806,23 @@ create_child: subflow_add_reset_reason(skb, MPTCP_RST_EMPTCP); goto dispose_child; } - - mptcp_subflow_drop_ctx(child); - goto out; + goto fallback; } /* ssk inherits options of listener sk */ ctx->setsockopt_seq = listener->setsockopt_seq; if (ctx->mp_capable) { - owner = mptcp_sk(new_msk); + ctx->conn = mptcp_sk_clone(listener->conn, &mp_opt, req); + if (!ctx->conn) + goto fallback; + + owner = mptcp_sk(ctx->conn); /* this can't race with mptcp_close(), as the msk is * not yet exposted to user-space */ - inet_sk_state_store((void *)new_msk, TCP_ESTABLISHED); + inet_sk_state_store(ctx->conn, TCP_ESTABLISHED); /* record the newly created socket as the first msk * subflow, but don't link it yet into conn_list @@ -844,11 +832,9 @@ create_child: /* new mpc subflow takes ownership of the newly * created mptcp socket */ - mptcp_sk(new_msk)->setsockopt_seq = ctx->setsockopt_seq; + owner->setsockopt_seq = ctx->setsockopt_seq; mptcp_pm_new_connection(owner, child, 1); mptcp_token_accept(subflow_req, owner); - ctx->conn = new_msk; - new_msk = NULL; /* set msk addresses early to ensure mptcp_pm_get_local_id() * uses the correct data @@ -898,11 +884,6 @@ create_child: } } -out: - /* dispose of the left over mptcp master, if any */ - if (unlikely(new_msk)) - mptcp_force_close(new_msk); - /* check for expected invariant - should never trigger, just help * catching eariler subtle bugs */ @@ -920,6 +901,10 @@ dispose_child: /* The last child reference will be released by the caller */ return child; + +fallback: + mptcp_subflow_drop_ctx(child); + return child; } static struct inet_connection_sock_af_ops subflow_specific __ro_after_init; -- cgit v1.2.3 From e925a0322ada48a3bfdfaf37be46945af3ad6e2e Mon Sep 17 00:00:00 2001 From: Matthieu Baerts Date: Mon, 27 Mar 2023 12:22:23 +0200 Subject: mptcp: do not fill info not used by the PM in used Only the in-kernel PM uses the number of address and subflow limits allowed per connection. It then makes more sense not to display such info when other PMs are used not to confuse the userspace by showing limits not being used. While at it, we can get rid of the "val" variable and add indentations instead. It would have been good to have done this modification directly in commit 4d25247d3ae4 ("mptcp: bypass in-kernel PM restrictions for non-kernel PMs") but as we change a bit the behaviour, it is fine not to backport it to stable. Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/sockopt.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 5cef4d3d21ac..b655cebda0f3 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -885,7 +885,6 @@ out: void mptcp_diag_fill_info(struct mptcp_sock *msk, struct mptcp_info *info) { u32 flags = 0; - u8 val; memset(info, 0, sizeof(*info)); @@ -893,12 +892,19 @@ void mptcp_diag_fill_info(struct mptcp_sock *msk, struct mptcp_info *info) info->mptcpi_add_addr_signal = READ_ONCE(msk->pm.add_addr_signaled); info->mptcpi_add_addr_accepted = READ_ONCE(msk->pm.add_addr_accepted); info->mptcpi_local_addr_used = READ_ONCE(msk->pm.local_addr_used); - info->mptcpi_subflows_max = mptcp_pm_get_subflows_max(msk); - val = mptcp_pm_get_add_addr_signal_max(msk); - info->mptcpi_add_addr_signal_max = val; - val = mptcp_pm_get_add_addr_accept_max(msk); - info->mptcpi_add_addr_accepted_max = val; - info->mptcpi_local_addr_max = mptcp_pm_get_local_addr_max(msk); + + /* The following limits only make sense for the in-kernel PM */ + if (mptcp_pm_is_kernel(msk)) { + info->mptcpi_subflows_max = + mptcp_pm_get_subflows_max(msk); + info->mptcpi_add_addr_signal_max = + mptcp_pm_get_add_addr_signal_max(msk); + info->mptcpi_add_addr_accepted_max = + mptcp_pm_get_add_addr_accept_max(msk); + info->mptcpi_local_addr_max = + mptcp_pm_get_local_addr_max(msk); + } + if (test_bit(MPTCP_FALLBACK_DONE, &msk->flags)) flags |= MPTCP_INFO_FLAG_FALLBACK; if (READ_ONCE(msk->can_ack)) -- cgit v1.2.3 From 562dc56a88983421a6c5a46e0feb891873d118a1 Mon Sep 17 00:00:00 2001 From: Yixin Shen Date: Wed, 29 Mar 2023 07:35:57 +0000 Subject: bpf: allow a TCP CC to write app_limited A CC that implements tcp_congestion_ops.cong_control() should be able to write app_limited. A built-in CC or one from a kernel module is already able to write to this member of struct tcp_sock. For a BPF program, write access has not been allowed, yet. Signed-off-by: Yixin Shen Link: https://lore.kernel.org/r/20230329073558.8136-2-bobankhshen@gmail.com Signed-off-by: Martin KaFai Lau --- net/ipv4/bpf_tcp_ca.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index e8b27826283e..ea21c96c03aa 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -113,6 +113,9 @@ static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log, case offsetof(struct tcp_sock, ecn_flags): end = offsetofend(struct tcp_sock, ecn_flags); break; + case offsetof(struct tcp_sock, app_limited): + end = offsetofend(struct tcp_sock, app_limited); + break; default: bpf_log(log, "no write support to tcp_sock at off %d\n", off); return -EACCES; -- cgit v1.2.3 From f7154d967bc4ee25ea1572937550e711b2525474 Mon Sep 17 00:00:00 2001 From: Arseniy Krasnov Date: Tue, 28 Mar 2023 14:31:28 +0300 Subject: virtio/vsock: fix header length on skb merging This fixes appending newly arrived skbuff to the last skbuff of the socket's queue. Problem fires when we are trying to append data to skbuff which was already processed in dequeue callback at least once. Dequeue callback calls function 'skb_pull()' which changes 'skb->len'. In current implementation 'skb->len' is used to update length in header of the last skbuff after new data was copied to it. This is bug, because value in header is used to calculate 'rx_bytes'/'fwd_cnt' and thus must be not be changed during skbuff's lifetime. Bug starts to fire since: commit 077706165717 ("virtio/vsock: don't use skbuff state to account credit") It presents before, but didn't triggered due to a little bit buggy implementation of credit calculation logic. So use Fixes tag for it. Fixes: 077706165717 ("virtio/vsock: don't use skbuff state to account credit") Signed-off-by: Arseniy Krasnov Reviewed-by: Stefano Garzarella Signed-off-by: Paolo Abeni --- net/vmw_vsock/virtio_transport_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 6564192e7f20..11ca9b3362aa 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -1068,7 +1068,7 @@ virtio_transport_recv_enqueue(struct vsock_sock *vsk, memcpy(skb_put(last_skb, skb->len), skb->data, skb->len); free_pkt = true; last_hdr->flags |= hdr->flags; - last_hdr->len = cpu_to_le32(last_skb->len); + le32_add_cpu(&last_hdr->len, len); goto out; } } -- cgit v1.2.3 From b8d2f61fdf2a566f7872158f35e65599aceb90fb Mon Sep 17 00:00:00 2001 From: Arseniy Krasnov Date: Tue, 28 Mar 2023 14:32:12 +0300 Subject: virtio/vsock: WARN_ONCE() for invalid state of socket This adds WARN_ONCE() and return from stream dequeue callback when socket's queue is empty, but 'rx_bytes' still non-zero. This allows the detection of potential bugs due to packet merging (see previous patch). Signed-off-by: Arseniy Krasnov Reviewed-by: Stefano Garzarella Signed-off-by: Paolo Abeni --- net/vmw_vsock/virtio_transport_common.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 11ca9b3362aa..37934dfe72f4 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -363,6 +363,13 @@ virtio_transport_stream_do_dequeue(struct vsock_sock *vsk, u32 free_space; spin_lock_bh(&vvs->rx_lock); + + if (WARN_ONCE(skb_queue_empty(&vvs->rx_queue) && vvs->rx_bytes, + "rx_queue is empty, but rx_bytes is non-zero\n")) { + spin_unlock_bh(&vvs->rx_lock); + return err; + } + while (total < len && !skb_queue_empty(&vvs->rx_queue)) { skb = skb_peek(&vvs->rx_queue); -- cgit v1.2.3 From 4d78e032fee5d532e189cdb2c3c76112094e9751 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 30 Mar 2023 11:00:00 +0200 Subject: wifi: mac80211: drop bogus static keywords in A-MSDU rx These were unintentional copy&paste mistakes. Cc: stable@vger.kernel.org Fixes: 986e43b19ae9 ("wifi: mac80211: fix receiving A-MSDU frames on mesh interfaces") Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230330090001.60750-1-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index e8de500eb9f3..9353616b528f 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2896,7 +2896,7 @@ __ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx, u8 data_offset) struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; __le16 fc = hdr->frame_control; struct sk_buff_head frame_list; - static ieee80211_rx_result res; + ieee80211_rx_result res; struct ethhdr ethhdr; const u8 *check_da = ethhdr.h_dest, *check_sa = ethhdr.h_source; @@ -3037,7 +3037,7 @@ ieee80211_rx_h_data(struct ieee80211_rx_data *rx) struct net_device *dev = sdata->dev; struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)rx->skb->data; __le16 fc = hdr->frame_control; - static ieee80211_rx_result res; + ieee80211_rx_result res; bool port_control; int err; -- cgit v1.2.3 From a16fc38315f2c69c520ee769976ecb9c706b8560 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 30 Mar 2023 11:00:01 +0200 Subject: wifi: mac80211: fix potential null pointer dereference rx->sta->amsdu_mesh_control is being passed to ieee80211_amsdu_to_8023s without checking rx->sta. Since it doesn't make sense to accept A-MSDU packets without a sta, simply add a check earlier. Fixes: 6e4c0d0460bd ("wifi: mac80211: add a workaround for receiving non-standard mesh A-MSDU") Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230330090001.60750-2-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 9353616b528f..78351895c3c6 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2930,7 +2930,7 @@ __ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx, u8 data_offset) data_offset, true)) return RX_DROP_UNUSABLE; - if (rx->sta && rx->sta->amsdu_mesh_control < 0) { + if (rx->sta->amsdu_mesh_control < 0) { bool valid_std = ieee80211_is_valid_amsdu(skb, true); bool valid_nonstd = ieee80211_is_valid_amsdu(skb, false); @@ -3006,7 +3006,7 @@ ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx) } } - if (is_multicast_ether_addr(hdr->addr1)) + if (is_multicast_ether_addr(hdr->addr1) || !rx->sta) return RX_DROP_UNUSABLE; if (rx->key) { -- cgit v1.2.3 From dd01579e5ed922dcfcb8fec53fa03b81c7649a04 Mon Sep 17 00:00:00 2001 From: Ryder Lee Date: Tue, 28 Mar 2023 01:07:41 +0800 Subject: wifi: mac80211: fix the size calculation of ieee80211_ie_len_eht_cap() Here should return the size of ieee80211_eht_cap_elem_fixed, so fix it. Fixes: 820acc810fb6 ("mac80211: Add EHT capabilities to association/probe request") Signed-off-by: Ryder Lee Link: https://lore.kernel.org/r/06c13635fc03bcff58a647b8e03e9f01a74294bd.1679935259.git.ryder.lee@mediatek.com Signed-off-by: Johannes Berg --- net/mac80211/util.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 3aceb3b731bf..8c397650b96f 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -4906,7 +4906,7 @@ u8 ieee80211_ie_len_eht_cap(struct ieee80211_sub_if_data *sdata, u8 iftype) &eht_cap->eht_cap_elem, is_ap); return 2 + 1 + - sizeof(he_cap->he_cap_elem) + n + + sizeof(eht_cap->eht_cap_elem) + n + ieee80211_eht_ppe_size(eht_cap->eht_ppe_thres[0], eht_cap->eht_cap_elem.phy_cap_info); return 0; -- cgit v1.2.3 From e26c0946a5c1aa4d27f8dfe78f2a72b4550df91f Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Sun, 26 Mar 2023 17:17:09 +0200 Subject: wifi: mac80211: fix receiving mesh packets in forwarding=0 networks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When forwarding is set to 0, frames are typically sent with ttl=1. Move the ttl decrement check below the check for local receive in order to fix packet drops. Reported-by: Thomas Hühn Reported-by: Nick Hainke Fixes: 986e43b19ae9 ("wifi: mac80211: fix receiving A-MSDU frames on mesh interfaces") Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230326151709.17743-1-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 78351895c3c6..7bd5aa2d5a95 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2769,14 +2769,6 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta if (sdata->crypto_tx_tailroom_needed_cnt) tailroom = IEEE80211_ENCRYPT_TAILROOM; - if (!--mesh_hdr->ttl) { - if (multicast) - goto rx_accept; - - IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, dropped_frames_ttl); - return RX_DROP_MONITOR; - } - if (mesh_hdr->flags & MESH_FLAGS_AE) { struct mesh_path *mppath; char *proxied_addr; @@ -2807,6 +2799,14 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta if (ether_addr_equal(sdata->vif.addr, eth->h_dest)) goto rx_accept; + if (!--mesh_hdr->ttl) { + if (multicast) + goto rx_accept; + + IEEE80211_IFSTA_MESH_CTR_INC(ifmsh, dropped_frames_ttl); + return RX_DROP_MONITOR; + } + if (!ifmsh->mshcfg.dot11MeshForwarding) { if (is_multicast_ether_addr(eth->h_dest)) goto rx_accept; -- cgit v1.2.3 From 8f0149a8ac59c12cd47271ac625c27dac5621d3a Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Fri, 24 Mar 2023 13:09:22 +0100 Subject: wifi: mac80211: fix mesh forwarding Linearize packets (needed for forwarding A-MSDU subframes). Fixes: 986e43b19ae9 ("wifi: mac80211: fix receiving A-MSDU frames on mesh interfaces") Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230324120924.38412-1-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 7bd5aa2d5a95..10efa429e946 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2833,6 +2833,9 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta if (skb_cow_head(fwd_skb, hdrlen - sizeof(struct ethhdr))) return RX_DROP_UNUSABLE; + + if (skb_linearize(fwd_skb)) + return RX_DROP_UNUSABLE; } fwd_hdr = skb_push(fwd_skb, hdrlen - sizeof(struct ethhdr)); -- cgit v1.2.3 From 899c2c11810cfe38cb01c847d0df98e181ea5728 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Fri, 24 Mar 2023 13:09:23 +0100 Subject: wifi: mac80211: fix flow dissection for forwarded packets Adjust the network header to point at the correct payload offset Fixes: 986e43b19ae9 ("wifi: mac80211: fix receiving A-MSDU frames on mesh interfaces") Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230324120924.38412-2-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 10efa429e946..af57616d2f1d 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -2850,7 +2850,7 @@ ieee80211_rx_mesh_data(struct ieee80211_sub_if_data *sdata, struct sta_info *sta hdrlen += ETH_ALEN; else fwd_skb->protocol = htons(fwd_skb->len - hdrlen); - skb_set_network_header(fwd_skb, hdrlen); + skb_set_network_header(fwd_skb, hdrlen + 2); info = IEEE80211_SKB_CB(fwd_skb); memset(info, 0, sizeof(*info)); -- cgit v1.2.3 From 12b220a6171faf10638ab683a975cadcf1a352d6 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Fri, 24 Mar 2023 13:09:24 +0100 Subject: wifi: mac80211: fix invalid drv_sta_pre_rcu_remove calls for non-uploaded sta Avoid potential data corruption issues caused by uninitialized driver private data structures. Reported-by: Brian Coverstone Fixes: 6a9d1b91f34d ("mac80211: add pre-RCU-sync sta removal driver operation") Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230324120924.38412-3-nbd@nbd.name Signed-off-by: Johannes Berg --- net/mac80211/sta_info.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 7d68dbc872d7..941bda9141fa 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -1264,7 +1264,8 @@ static int __must_check __sta_info_destroy_part1(struct sta_info *sta) list_del_rcu(&sta->list); sta->removed = true; - drv_sta_pre_rcu_remove(local, sta->sdata, sta); + if (sta->uploaded) + drv_sta_pre_rcu_remove(local, sta->sdata, sta); if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN && rcu_access_pointer(sdata->u.vlan.sta) == sta) -- cgit v1.2.3 From 968a768d28530cadd13aab51531f8fec70bf9eae Mon Sep 17 00:00:00 2001 From: Tom Rix Date: Sat, 25 Mar 2023 09:26:10 -0400 Subject: mac80211: minstrel_ht: remove unused n_supported variable clang with W=1 reports net/mac80211/rc80211_minstrel_ht.c:1711:6: error: variable 'n_supported' set but not used [-Werror,-Wunused-but-set-variable] int n_supported = 0; ^ This variable is not used so remove it. Signed-off-by: Tom Rix Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230325132610.1334820-1-trix@redhat.com Signed-off-by: Johannes Berg --- net/mac80211/rc80211_minstrel_ht.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'net') diff --git a/net/mac80211/rc80211_minstrel_ht.c b/net/mac80211/rc80211_minstrel_ht.c index 762346598338..b34c80522047 100644 --- a/net/mac80211/rc80211_minstrel_ht.c +++ b/net/mac80211/rc80211_minstrel_ht.c @@ -1708,7 +1708,6 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband, struct sta_info *sta_info; bool ldpc, erp; int use_vht; - int n_supported = 0; int ack_dur; int stbc; int i; @@ -1791,8 +1790,6 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband, continue; mi->supported[i] = mcs->rx_mask[nss - 1]; - if (mi->supported[i]) - n_supported++; continue; } @@ -1819,9 +1816,6 @@ minstrel_ht_update_caps(void *priv, struct ieee80211_supported_band *sband, mi->supported[i] = minstrel_get_valid_vht_rates(bw, nss, vht_cap->vht_mcs.tx_mcs_map); - - if (mi->supported[i]) - n_supported++; } sta_info = container_of(sta, struct sta_info, sta); -- cgit v1.2.3 From 0333a81bc83431d7f90391d38aa09e856c5e5b25 Mon Sep 17 00:00:00 2001 From: Kieran Frewen Date: Thu, 23 Feb 2023 16:25:12 +1300 Subject: wifi: mac80211: S1G capabilities information element in probe request Add the missing S1G capabilities information element to probe requests. Signed-off-by: Kieran Frewen Co-developed-by: Gilad Itzkovitch Signed-off-by: Gilad Itzkovitch Link: https://lore.kernel.org/r/20230223032512.3848105-1-gilad.itzkovitch@virscient.com Signed-off-by: Johannes Berg --- net/mac80211/ieee80211_i.h | 1 + net/mac80211/util.c | 23 +++++++++++++++++++++++ 2 files changed, 24 insertions(+) (limited to 'net') diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 85ecbf57d64e..cdc80285efd4 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2491,6 +2491,7 @@ void ieee80211_add_s1g_capab_ie(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb); void ieee80211_add_aid_request_ie(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb); +u8 *ieee80211_ie_build_s1g_cap(u8 *pos, struct ieee80211_sta_s1g_cap *s1g_cap); /* channel management */ bool ieee80211_chandef_ht_oper(const struct ieee80211_ht_operation *ht_oper, diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 1a28fe5cb614..10ac9819ed91 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -1959,6 +1959,14 @@ static int ieee80211_build_preq_ies_band(struct ieee80211_sub_if_data *sdata, rate_flags = ieee80211_chandef_rate_flags(chandef); shift = ieee80211_chandef_get_shift(chandef); + /* For direct scan add S1G IE and consider its override bits */ + if (band == NL80211_BAND_S1GHZ) { + if (end - pos < 2 + sizeof(struct ieee80211_s1g_cap)) + goto out_err; + pos = ieee80211_ie_build_s1g_cap(pos, &sband->s1g_cap); + goto done; + } + num_rates = 0; for (i = 0; i < sband->n_bitrates; i++) { if ((BIT(i) & rate_mask) == 0) @@ -3020,6 +3028,21 @@ size_t ieee80211_ie_split_vendor(const u8 *ies, size_t ielen, size_t offset) return pos; } +u8 *ieee80211_ie_build_s1g_cap(u8 *pos, struct ieee80211_sta_s1g_cap *s1g_cap) +{ + *pos++ = WLAN_EID_S1G_CAPABILITIES; + *pos++ = sizeof(struct ieee80211_s1g_cap); + memset(pos, 0, sizeof(struct ieee80211_s1g_cap)); + + memcpy(pos, &s1g_cap->cap, sizeof(s1g_cap->cap)); + pos += sizeof(s1g_cap->cap); + + memcpy(pos, &s1g_cap->nss_mcs, sizeof(s1g_cap->nss_mcs)); + pos += sizeof(s1g_cap->nss_mcs); + + return pos; +} + u8 *ieee80211_ie_build_ht_cap(u8 *pos, struct ieee80211_sta_ht_cap *ht_cap, u16 cap) { -- cgit v1.2.3 From 9a8aac92eba90b3b7c71d0531db535f5588388f5 Mon Sep 17 00:00:00 2001 From: Kieran Frewen Date: Fri, 24 Feb 2023 10:29:17 +1300 Subject: wifi: nl80211: support advertising S1G capabilities Include S1G capabilities in netlink band info messages. Signed-off-by: Kieran Frewen Co-developed-by: Gilad Itzkovitch Signed-off-by: Gilad Itzkovitch Link: https://lore.kernel.org/r/20230223212917.4010246-1-gilad.itzkovitch@virscient.com Signed-off-by: Johannes Berg --- include/uapi/linux/nl80211.h | 7 +++++++ net/wireless/nl80211.c | 10 ++++++++++ 2 files changed, 17 insertions(+) (limited to 'net') diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index cf4fb981e131..c59fec406da5 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -4061,6 +4061,10 @@ enum nl80211_band_iftype_attr { * @NL80211_BAND_ATTR_EDMG_BW_CONFIG: Channel BW Configuration subfield encodes * the allowed channel bandwidth configurations. * Defined by IEEE P802.11ay/D4.0 section 9.4.2.251, Table 13. + * @NL80211_BAND_ATTR_S1G_MCS_NSS_SET: S1G capabilities, supported S1G-MCS and NSS + * set subfield, as in the S1G information IE, 5 bytes + * @NL80211_BAND_ATTR_S1G_CAPA: S1G capabilities information subfield as in the + * S1G information IE, 10 bytes * @NL80211_BAND_ATTR_MAX: highest band attribute currently defined * @__NL80211_BAND_ATTR_AFTER_LAST: internal use */ @@ -4081,6 +4085,9 @@ enum nl80211_band_attr { NL80211_BAND_ATTR_EDMG_CHANNELS, NL80211_BAND_ATTR_EDMG_BW_CONFIG, + NL80211_BAND_ATTR_S1G_MCS_NSS_SET, + NL80211_BAND_ATTR_S1G_CAPA, + /* keep last */ __NL80211_BAND_ATTR_AFTER_LAST, NL80211_BAND_ATTR_MAX = __NL80211_BAND_ATTR_AFTER_LAST - 1 diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index f1cd3d9130dd..2c9edb015652 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -1961,6 +1961,16 @@ static int nl80211_send_band_rateinfo(struct sk_buff *msg, nla_nest_end(msg, nl_rates); + /* S1G capabilities */ + if (sband->band == NL80211_BAND_S1GHZ && sband->s1g_cap.s1g && + (nla_put(msg, NL80211_BAND_ATTR_S1G_CAPA, + sizeof(sband->s1g_cap.cap), + sband->s1g_cap.cap) || + nla_put(msg, NL80211_BAND_ATTR_S1G_MCS_NSS_SET, + sizeof(sband->s1g_cap.nss_mcs), + sband->s1g_cap.nss_mcs))) + return -ENOBUFS; + return 0; } -- cgit v1.2.3 From 8fcb76b934daff12cde76adeab3d502eeb0734b1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 28 Mar 2023 23:50:18 +0000 Subject: net: napi_schedule_rps() cleanup napi_schedule_rps() return value is ignored, remove it. Change the comment to clarify the intent. Signed-off-by: Eric Dumazet Reviewed-by: Jason Xing Tested-by: Jason Xing Signed-off-by: Paolo Abeni --- net/core/dev.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 7172334a418f..f7050b95d125 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4582,11 +4582,18 @@ static void trigger_rx_softirq(void *data) } /* - * Check if this softnet_data structure is another cpu one - * If yes, queue it to our IPI list and return 1 - * If no, return 0 + * After we queued a packet into sd->input_pkt_queue, + * we need to make sure this queue is serviced soon. + * + * - If this is another cpu queue, link it to our rps_ipi_list, + * and make sure we will process rps_ipi_list from net_rx_action(). + * As we do not know yet if we are called from net_rx_action(), + * we have to raise NET_RX_SOFTIRQ. This might change in the future. + * + * - If this is our own queue, NAPI schedule our backlog. + * Note that this also raises NET_RX_SOFTIRQ. */ -static int napi_schedule_rps(struct softnet_data *sd) +static void napi_schedule_rps(struct softnet_data *sd) { struct softnet_data *mysd = this_cpu_ptr(&softnet_data); @@ -4596,11 +4603,10 @@ static int napi_schedule_rps(struct softnet_data *sd) mysd->rps_ipi_list = sd; __raise_softirq_irqoff(NET_RX_SOFTIRQ); - return 1; + return; } #endif /* CONFIG_RPS */ __napi_schedule_irqoff(&mysd->backlog); - return 0; } #ifdef CONFIG_NET_FLOW_LIMIT -- cgit v1.2.3 From c59647c0dc679008886756a888368da1c6d4ccd3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 28 Mar 2023 23:50:19 +0000 Subject: net: add softnet_data.in_net_rx_action We want to make two optimizations in napi_schedule_rps() and ____napi_schedule() which require to know if these helpers are called from net_rx_action(), instead of being called from other contexts. sd.in_net_rx_action is only read/written by the owning cpu. Signed-off-by: Eric Dumazet Reviewed-by: Jason Xing Tested-by: Jason Xing Signed-off-by: Paolo Abeni --- include/linux/netdevice.h | 1 + net/core/dev.c | 4 ++++ 2 files changed, 5 insertions(+) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 18a5be6ddd0f..c8c634091a65 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3188,6 +3188,7 @@ struct softnet_data { #ifdef CONFIG_RPS struct softnet_data *rps_ipi_list; #endif + bool in_net_rx_action; #ifdef CONFIG_NET_FLOW_LIMIT struct sd_flow_limit __rcu *flow_limit; #endif diff --git a/net/core/dev.c b/net/core/dev.c index f7050b95d125..15331edbacf4 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6646,6 +6646,7 @@ static __latent_entropy void net_rx_action(struct softirq_action *h) LIST_HEAD(list); LIST_HEAD(repoll); + sd->in_net_rx_action = true; local_irq_disable(); list_splice_init(&sd->poll_list, &list); local_irq_enable(); @@ -6656,6 +6657,7 @@ static __latent_entropy void net_rx_action(struct softirq_action *h) skb_defer_free_flush(sd); if (list_empty(&list)) { + sd->in_net_rx_action = false; if (!sd_has_rps_ipi_waiting(sd) && list_empty(&repoll)) goto end; break; @@ -6682,6 +6684,8 @@ static __latent_entropy void net_rx_action(struct softirq_action *h) list_splice(&list, &sd->poll_list); if (!list_empty(&sd->poll_list)) __raise_softirq_irqoff(NET_RX_SOFTIRQ); + else + sd->in_net_rx_action = false; net_rps_action_and_irq_enable(sd); end:; -- cgit v1.2.3 From 821eba962d95806beb0440742c4062a9da8a386b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 28 Mar 2023 23:50:20 +0000 Subject: net: optimize napi_schedule_rps() Based on initial patch from Jason Xing. Idea is to not raise NET_RX_SOFTIRQ from napi_schedule_rps() when we queued a packet into another cpu backlog. We can do this only in the context of us being called indirectly from net_rx_action(), to have the guarantee our rps_ipi_list will be processed before we exit from net_rx_action(). Link: https://lore.kernel.org/lkml/20230325152417.5403-1-kerneljasonxing@gmail.com/ Signed-off-by: Eric Dumazet Cc: Jason Xing Reviewed-by: Jason Xing Tested-by: Jason Xing Signed-off-by: Paolo Abeni --- net/core/dev.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 15331edbacf4..f34ce93f2f02 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4587,8 +4587,6 @@ static void trigger_rx_softirq(void *data) * * - If this is another cpu queue, link it to our rps_ipi_list, * and make sure we will process rps_ipi_list from net_rx_action(). - * As we do not know yet if we are called from net_rx_action(), - * we have to raise NET_RX_SOFTIRQ. This might change in the future. * * - If this is our own queue, NAPI schedule our backlog. * Note that this also raises NET_RX_SOFTIRQ. @@ -4602,7 +4600,11 @@ static void napi_schedule_rps(struct softnet_data *sd) sd->rps_ipi_next = mysd->rps_ipi_list; mysd->rps_ipi_list = sd; - __raise_softirq_irqoff(NET_RX_SOFTIRQ); + /* If not called from net_rx_action() + * we have to raise NET_RX_SOFTIRQ. + */ + if (!mysd->in_net_rx_action) + __raise_softirq_irqoff(NET_RX_SOFTIRQ); return; } #endif /* CONFIG_RPS */ -- cgit v1.2.3 From 8b43fd3d1d7d88293eb15e92090826e6b7cc13e4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 28 Mar 2023 23:50:21 +0000 Subject: net: optimize ____napi_schedule() to avoid extra NET_RX_SOFTIRQ ____napi_schedule() adds a napi into current cpu softnet_data poll_list, then raises NET_RX_SOFTIRQ to make sure net_rx_action() will process it. Idea of this patch is to not raise NET_RX_SOFTIRQ when being called indirectly from net_rx_action(), because we can process poll_list from this point, without going to full softirq loop. This needs a change in net_rx_action() to make sure we restart its main loop if sd->poll_list was updated without NET_RX_SOFTIRQ being raised. Signed-off-by: Eric Dumazet Cc: Jason Xing Reviewed-by: Jason Xing Tested-by: Jason Xing Signed-off-by: Paolo Abeni --- net/core/dev.c | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index f34ce93f2f02..0c4b21291348 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4360,7 +4360,11 @@ static inline void ____napi_schedule(struct softnet_data *sd, } list_add_tail(&napi->poll_list, &sd->poll_list); - __raise_softirq_irqoff(NET_RX_SOFTIRQ); + /* If not called from net_rx_action() + * we have to raise NET_RX_SOFTIRQ. + */ + if (!sd->in_net_rx_action) + __raise_softirq_irqoff(NET_RX_SOFTIRQ); } #ifdef CONFIG_RPS @@ -6648,6 +6652,7 @@ static __latent_entropy void net_rx_action(struct softirq_action *h) LIST_HEAD(list); LIST_HEAD(repoll); +start: sd->in_net_rx_action = true; local_irq_disable(); list_splice_init(&sd->poll_list, &list); @@ -6659,9 +6664,18 @@ static __latent_entropy void net_rx_action(struct softirq_action *h) skb_defer_free_flush(sd); if (list_empty(&list)) { - sd->in_net_rx_action = false; - if (!sd_has_rps_ipi_waiting(sd) && list_empty(&repoll)) - goto end; + if (list_empty(&repoll)) { + sd->in_net_rx_action = false; + barrier(); + /* We need to check if ____napi_schedule() + * had refilled poll_list while + * sd->in_net_rx_action was true. + */ + if (!list_empty(&sd->poll_list)) + goto start; + if (!sd_has_rps_ipi_waiting(sd)) + goto end; + } break; } -- cgit v1.2.3 From 64fdc5f341db01200e33105265d4b8450122a82e Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 29 Mar 2023 18:18:21 +0300 Subject: net: dsa: sync unicast and multicast addresses for VLAN filters too If certain conditions are met, DSA can install all necessary MAC addresses on the CPU ports as FDB entries and disable flooding towards the CPU (we call this RX filtering). There is one corner case where this does not work. ip link add br0 type bridge vlan_filtering 1 && ip link set br0 up ip link set swp0 master br0 && ip link set swp0 up ip link add link swp0 name swp0.100 type vlan id 100 ip link set swp0.100 up && ip addr add 192.168.100.1/24 dev swp0.100 Traffic through swp0.100 is broken, because the bridge turns on VLAN filtering in the swp0 port (causing RX packets to be classified to the FDB database corresponding to the VID from their 802.1Q header), and although the 8021q module does call dev_uc_add() towards the real device, that API is VLAN-unaware, so it only contains the MAC address, not the VID; and DSA's current implementation of ndo_set_rx_mode() is only for VID 0 (corresponding to FDB entries which are installed in an FDB database which is only hit when the port is VLAN-unaware). It's interesting to understand why the bridge does not turn on IFF_PROMISC for its swp0 bridge port, and it may appear at first glance that this is a regression caused by the logic in commit 2796d0c648c9 ("bridge: Automatically manage port promiscuous mode."). After all, a bridge port needs to have IFF_PROMISC by its very nature - it needs to receive and forward frames with a MAC DA different from the bridge ports' MAC addresses. While that may be true, when the bridge is VLAN-aware *and* it has a single port, there is no real reason to enable promiscuity even if that is an automatic port, with flooding and learning (there is nowhere for packets to go except to the BR_FDB_LOCAL entries), and this is how the corner case appears. Adding a second automatic interface to the bridge would make swp0 promisc as well, and would mask the corner case. Given the dev_uc_add() / ndo_set_rx_mode() API is what it is (it doesn't pass a VLAN ID), the only way to address that problem is to install host FDB entries for the cartesian product of RX filtering MAC addresses and VLAN RX filters. Fixes: 7569459a52c9 ("net: dsa: manage flooding on the CPU ports") Signed-off-by: Vladimir Oltean Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/20230329151821.745752-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/slave.c | 121 +++++++++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 116 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/dsa/slave.c b/net/dsa/slave.c index cac17183589f..165bb2cb8431 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -57,6 +57,12 @@ struct dsa_standalone_event_work { u16 vid; }; +struct dsa_host_vlan_rx_filtering_ctx { + struct net_device *dev; + const unsigned char *addr; + enum dsa_standalone_event event; +}; + static bool dsa_switch_supports_uc_filtering(struct dsa_switch *ds) { return ds->ops->port_fdb_add && ds->ops->port_fdb_del && @@ -155,18 +161,37 @@ static int dsa_slave_schedule_standalone_work(struct net_device *dev, return 0; } +static int dsa_slave_host_vlan_rx_filtering(struct net_device *vdev, int vid, + void *arg) +{ + struct dsa_host_vlan_rx_filtering_ctx *ctx = arg; + + return dsa_slave_schedule_standalone_work(ctx->dev, ctx->event, + ctx->addr, vid); +} + static int dsa_slave_sync_uc(struct net_device *dev, const unsigned char *addr) { struct net_device *master = dsa_slave_to_master(dev); struct dsa_port *dp = dsa_slave_to_port(dev); + struct dsa_host_vlan_rx_filtering_ctx ctx = { + .dev = dev, + .addr = addr, + .event = DSA_UC_ADD, + }; + int err; dev_uc_add(master, addr); if (!dsa_switch_supports_uc_filtering(dp->ds)) return 0; - return dsa_slave_schedule_standalone_work(dev, DSA_UC_ADD, addr, 0); + err = dsa_slave_schedule_standalone_work(dev, DSA_UC_ADD, addr, 0); + if (err) + return err; + + return vlan_for_each(dev, dsa_slave_host_vlan_rx_filtering, &ctx); } static int dsa_slave_unsync_uc(struct net_device *dev, @@ -174,13 +199,23 @@ static int dsa_slave_unsync_uc(struct net_device *dev, { struct net_device *master = dsa_slave_to_master(dev); struct dsa_port *dp = dsa_slave_to_port(dev); + struct dsa_host_vlan_rx_filtering_ctx ctx = { + .dev = dev, + .addr = addr, + .event = DSA_UC_DEL, + }; + int err; dev_uc_del(master, addr); if (!dsa_switch_supports_uc_filtering(dp->ds)) return 0; - return dsa_slave_schedule_standalone_work(dev, DSA_UC_DEL, addr, 0); + err = dsa_slave_schedule_standalone_work(dev, DSA_UC_DEL, addr, 0); + if (err) + return err; + + return vlan_for_each(dev, dsa_slave_host_vlan_rx_filtering, &ctx); } static int dsa_slave_sync_mc(struct net_device *dev, @@ -188,13 +223,23 @@ static int dsa_slave_sync_mc(struct net_device *dev, { struct net_device *master = dsa_slave_to_master(dev); struct dsa_port *dp = dsa_slave_to_port(dev); + struct dsa_host_vlan_rx_filtering_ctx ctx = { + .dev = dev, + .addr = addr, + .event = DSA_MC_ADD, + }; + int err; dev_mc_add(master, addr); if (!dsa_switch_supports_mc_filtering(dp->ds)) return 0; - return dsa_slave_schedule_standalone_work(dev, DSA_MC_ADD, addr, 0); + err = dsa_slave_schedule_standalone_work(dev, DSA_MC_ADD, addr, 0); + if (err) + return err; + + return vlan_for_each(dev, dsa_slave_host_vlan_rx_filtering, &ctx); } static int dsa_slave_unsync_mc(struct net_device *dev, @@ -202,13 +247,23 @@ static int dsa_slave_unsync_mc(struct net_device *dev, { struct net_device *master = dsa_slave_to_master(dev); struct dsa_port *dp = dsa_slave_to_port(dev); + struct dsa_host_vlan_rx_filtering_ctx ctx = { + .dev = dev, + .addr = addr, + .event = DSA_MC_DEL, + }; + int err; dev_mc_del(master, addr); if (!dsa_switch_supports_mc_filtering(dp->ds)) return 0; - return dsa_slave_schedule_standalone_work(dev, DSA_MC_DEL, addr, 0); + err = dsa_slave_schedule_standalone_work(dev, DSA_MC_DEL, addr, 0); + if (err) + return err; + + return vlan_for_each(dev, dsa_slave_host_vlan_rx_filtering, &ctx); } void dsa_slave_sync_ha(struct net_device *dev) @@ -1702,6 +1757,8 @@ static int dsa_slave_vlan_rx_add_vid(struct net_device *dev, __be16 proto, .flags = 0, }; struct netlink_ext_ack extack = {0}; + struct dsa_switch *ds = dp->ds; + struct netdev_hw_addr *ha; int ret; /* User port... */ @@ -1721,6 +1778,30 @@ static int dsa_slave_vlan_rx_add_vid(struct net_device *dev, __be16 proto, return ret; } + if (!dsa_switch_supports_uc_filtering(ds) && + !dsa_switch_supports_mc_filtering(ds)) + return 0; + + netif_addr_lock_bh(dev); + + if (dsa_switch_supports_mc_filtering(ds)) { + netdev_for_each_synced_mc_addr(ha, dev) { + dsa_slave_schedule_standalone_work(dev, DSA_MC_ADD, + ha->addr, vid); + } + } + + if (dsa_switch_supports_uc_filtering(ds)) { + netdev_for_each_synced_uc_addr(ha, dev) { + dsa_slave_schedule_standalone_work(dev, DSA_UC_ADD, + ha->addr, vid); + } + } + + netif_addr_unlock_bh(dev); + + dsa_flush_workqueue(); + return 0; } @@ -1733,13 +1814,43 @@ static int dsa_slave_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, /* This API only allows programming tagged, non-PVID VIDs */ .flags = 0, }; + struct dsa_switch *ds = dp->ds; + struct netdev_hw_addr *ha; int err; err = dsa_port_vlan_del(dp, &vlan); if (err) return err; - return dsa_port_host_vlan_del(dp, &vlan); + err = dsa_port_host_vlan_del(dp, &vlan); + if (err) + return err; + + if (!dsa_switch_supports_uc_filtering(ds) && + !dsa_switch_supports_mc_filtering(ds)) + return 0; + + netif_addr_lock_bh(dev); + + if (dsa_switch_supports_mc_filtering(ds)) { + netdev_for_each_synced_mc_addr(ha, dev) { + dsa_slave_schedule_standalone_work(dev, DSA_MC_DEL, + ha->addr, vid); + } + } + + if (dsa_switch_supports_uc_filtering(ds)) { + netdev_for_each_synced_uc_addr(ha, dev) { + dsa_slave_schedule_standalone_work(dev, DSA_UC_DEL, + ha->addr, vid); + } + } + + netif_addr_unlock_bh(dev); + + dsa_flush_workqueue(); + + return 0; } static int dsa_slave_restore_vlan(struct net_device *vdev, int vid, void *arg) -- cgit v1.2.3 From 356e2adb3f3b9fe4427224eec2f8819a3078665f Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 28 Mar 2023 18:44:15 +0200 Subject: netfilter: nfnetlink_log: remove rcu_bh usage structure is free'd via call_rcu, so its safe to use rcu_read_lock only. While at it, skip rcu_read_lock for lookup from packet path, its always called with rcu held. Signed-off-by: Florian Westphal --- net/netfilter/nfnetlink_log.c | 36 +++++++++++++++++++++++------------- 1 file changed, 23 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index d97eb280cb2e..e57eb168ee13 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -103,9 +103,9 @@ static inline u_int8_t instance_hashfn(u_int16_t group_num) } static struct nfulnl_instance * -__instance_lookup(struct nfnl_log_net *log, u_int16_t group_num) +__instance_lookup(const struct nfnl_log_net *log, u16 group_num) { - struct hlist_head *head; + const struct hlist_head *head; struct nfulnl_instance *inst; head = &log->instance_table[instance_hashfn(group_num)]; @@ -123,15 +123,25 @@ instance_get(struct nfulnl_instance *inst) } static struct nfulnl_instance * -instance_lookup_get(struct nfnl_log_net *log, u_int16_t group_num) +instance_lookup_get_rcu(const struct nfnl_log_net *log, u16 group_num) { struct nfulnl_instance *inst; - rcu_read_lock_bh(); inst = __instance_lookup(log, group_num); if (inst && !refcount_inc_not_zero(&inst->use)) inst = NULL; - rcu_read_unlock_bh(); + + return inst; +} + +static struct nfulnl_instance * +instance_lookup_get(const struct nfnl_log_net *log, u16 group_num) +{ + struct nfulnl_instance *inst; + + rcu_read_lock(); + inst = instance_lookup_get_rcu(log, group_num); + rcu_read_unlock(); return inst; } @@ -698,7 +708,7 @@ nfulnl_log_packet(struct net *net, else li = &default_loginfo; - inst = instance_lookup_get(log, li->u.ulog.group); + inst = instance_lookup_get_rcu(log, li->u.ulog.group); if (!inst) return; @@ -1030,7 +1040,7 @@ static struct hlist_node *get_first(struct net *net, struct iter_state *st) struct hlist_head *head = &log->instance_table[st->bucket]; if (!hlist_empty(head)) - return rcu_dereference_bh(hlist_first_rcu(head)); + return rcu_dereference(hlist_first_rcu(head)); } return NULL; } @@ -1038,7 +1048,7 @@ static struct hlist_node *get_first(struct net *net, struct iter_state *st) static struct hlist_node *get_next(struct net *net, struct iter_state *st, struct hlist_node *h) { - h = rcu_dereference_bh(hlist_next_rcu(h)); + h = rcu_dereference(hlist_next_rcu(h)); while (!h) { struct nfnl_log_net *log; struct hlist_head *head; @@ -1048,7 +1058,7 @@ static struct hlist_node *get_next(struct net *net, struct iter_state *st, log = nfnl_log_pernet(net); head = &log->instance_table[st->bucket]; - h = rcu_dereference_bh(hlist_first_rcu(head)); + h = rcu_dereference(hlist_first_rcu(head)); } return h; } @@ -1066,9 +1076,9 @@ static struct hlist_node *get_idx(struct net *net, struct iter_state *st, } static void *seq_start(struct seq_file *s, loff_t *pos) - __acquires(rcu_bh) + __acquires(rcu) { - rcu_read_lock_bh(); + rcu_read_lock(); return get_idx(seq_file_net(s), s->private, *pos); } @@ -1079,9 +1089,9 @@ static void *seq_next(struct seq_file *s, void *v, loff_t *pos) } static void seq_stop(struct seq_file *s, void *v) - __releases(rcu_bh) + __releases(rcu) { - rcu_read_unlock_bh(); + rcu_read_unlock(); } static int seq_show(struct seq_file *s, void *v) -- cgit v1.2.3 From 28c1b6df436819a7ed8a781835766e45139771a3 Mon Sep 17 00:00:00 2001 From: Eric Sage Date: Mon, 27 Mar 2023 13:44:49 -0400 Subject: netfilter: nfnetlink_queue: enable classid socket info retrieval This enables associating a socket with a v1 net_cls cgroup. Useful for applying a per-cgroup policy when processing packets in userspace. Signed-off-by: Eric Sage Signed-off-by: Florian Westphal --- include/uapi/linux/netfilter/nfnetlink_queue.h | 1 + net/netfilter/nfnetlink_queue.c | 20 ++++++++++++++++++++ 2 files changed, 21 insertions(+) (limited to 'net') diff --git a/include/uapi/linux/netfilter/nfnetlink_queue.h b/include/uapi/linux/netfilter/nfnetlink_queue.h index ef7c97f21a15..efcb7c044a74 100644 --- a/include/uapi/linux/netfilter/nfnetlink_queue.h +++ b/include/uapi/linux/netfilter/nfnetlink_queue.h @@ -62,6 +62,7 @@ enum nfqnl_attr_type { NFQA_VLAN, /* nested attribute: packet vlan info */ NFQA_L2HDR, /* full L2 header */ NFQA_PRIORITY, /* skb->priority */ + NFQA_CGROUP_CLASSID, /* __u32 cgroup classid */ __NFQA_MAX }; diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 87a9009d5234..e311462f6d98 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include #include @@ -301,6 +302,19 @@ nla_put_failure: return -1; } +static int nfqnl_put_sk_classid(struct sk_buff *skb, struct sock *sk) +{ +#if IS_ENABLED(CONFIG_CGROUP_NET_CLASSID) + if (sk && sk_fullsock(sk)) { + u32 classid = sock_cgroup_classid(&sk->sk_cgrp_data); + + if (classid && nla_put_be32(skb, NFQA_CGROUP_CLASSID, htonl(classid))) + return -1; + } +#endif + return 0; +} + static u32 nfqnl_get_sk_secctx(struct sk_buff *skb, char **secdata) { u32 seclen = 0; @@ -406,6 +420,9 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue, + nla_total_size(sizeof(u_int32_t)) /* priority */ + nla_total_size(sizeof(struct nfqnl_msg_packet_hw)) + nla_total_size(sizeof(u_int32_t)) /* skbinfo */ +#if IS_ENABLED(CONFIG_CGROUP_NET_CLASSID) + + nla_total_size(sizeof(u_int32_t)) /* classid */ +#endif + nla_total_size(sizeof(u_int32_t)); /* cap_len */ tstamp = skb_tstamp_cond(entskb, false); @@ -599,6 +616,9 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue, nfqnl_put_sk_uidgid(skb, entskb->sk) < 0) goto nla_put_failure; + if (nfqnl_put_sk_classid(skb, entskb->sk) < 0) + goto nla_put_failure; + if (seclen && nla_put(skb, NFQA_SECCTX, seclen, secdata)) goto nla_put_failure; -- cgit v1.2.3 From 9b7c68b3911aef84afa4cbfc31bce20f10570d51 Mon Sep 17 00:00:00 2001 From: Paul Blakey Date: Wed, 22 Mar 2023 09:35:32 +0200 Subject: netfilter: ctnetlink: Support offloaded conntrack entry deletion Currently, offloaded conntrack entries (flows) can only be deleted after they are removed from offload, which is either by timeout, tcp state change or tc ct rule deletion. This can cause issues for users wishing to manually delete or flush existing entries. Support deletion of offloaded conntrack entries. Example usage: # Delete all offloaded (and non offloaded) conntrack entries # whose source address is 1.2.3.4 $ conntrack -D -s 1.2.3.4 # Delete all entries $ conntrack -F Signed-off-by: Paul Blakey Reviewed-by: Simon Horman Acked-by: Pablo Neira Ayuso Signed-off-by: Florian Westphal --- net/netfilter/nf_conntrack_netlink.c | 8 -------- 1 file changed, 8 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index bfc3aaa2c872..fbc47e4b7bc3 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1554,9 +1554,6 @@ static const struct nla_policy ct_nla_policy[CTA_MAX+1] = { static int ctnetlink_flush_iterate(struct nf_conn *ct, void *data) { - if (test_bit(IPS_OFFLOAD_BIT, &ct->status)) - return 0; - return ctnetlink_filter_match(ct, data); } @@ -1626,11 +1623,6 @@ static int ctnetlink_del_conntrack(struct sk_buff *skb, ct = nf_ct_tuplehash_to_ctrack(h); - if (test_bit(IPS_OFFLOAD_BIT, &ct->status)) { - nf_ct_put(ct); - return -EBUSY; - } - if (cda[CTA_ID]) { __be32 id = nla_get_be32(cda[CTA_ID]); -- cgit v1.2.3 From eb1ab7650d358b553cc946035fd7c7bdda1856e3 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Wed, 29 Mar 2023 16:38:19 +0300 Subject: net: dsa: fix db type confusion in host fdb/mdb add/del We have the following code paths: Host FDB (unicast RX filtering): dsa_port_standalone_host_fdb_add() dsa_port_bridge_host_fdb_add() | | +--------------+ +------------+ | | v v dsa_port_host_fdb_add() dsa_port_standalone_host_fdb_del() dsa_port_bridge_host_fdb_del() | | +--------------+ +------------+ | | v v dsa_port_host_fdb_del() Host MDB (multicast RX filtering): dsa_port_standalone_host_mdb_add() dsa_port_bridge_host_mdb_add() | | +--------------+ +------------+ | | v v dsa_port_host_mdb_add() dsa_port_standalone_host_mdb_del() dsa_port_bridge_host_mdb_del() | | +--------------+ +------------+ | | v v dsa_port_host_mdb_del() The logic added by commit 5e8a1e03aa4d ("net: dsa: install secondary unicast and multicast addresses as host FDB/MDB") zeroes out db.bridge.num if the switch doesn't support ds->fdb_isolation (the majority doesn't). This is done for a reason explained in commit c26933639b54 ("net: dsa: request drivers to perform FDB isolation"). Taking a single code path as example - dsa_port_host_fdb_add() - the others are similar - the problem is that this function handles: - DSA_DB_PORT databases, when called from dsa_port_standalone_host_fdb_add() - DSA_DB_BRIDGE databases, when called from dsa_port_bridge_host_fdb_add() So, if dsa_port_host_fdb_add() were to make any change on the "bridge.num" attribute of the database, this would only be correct for a DSA_DB_BRIDGE, and a type confusion for a DSA_DB_PORT bridge. However, this bug is without consequences, for 2 reasons: - dsa_port_standalone_host_fdb_add() is only called from code which is (in)directly guarded by dsa_switch_supports_uc_filtering(ds), and that function only returns true if ds->fdb_isolation is set. So, the code only executed for DSA_DB_BRIDGE databases. - Even if the code was not dead for DSA_DB_PORT, we have the following memory layout: struct dsa_bridge { struct net_device *dev; unsigned int num; bool tx_fwd_offload; refcount_t refcount; }; struct dsa_db { enum dsa_db_type type; union { const struct dsa_port *dp; // DSA_DB_PORT struct dsa_lag lag; struct dsa_bridge bridge; // DSA_DB_BRIDGE }; }; So, the zeroization of dsa_db :: bridge :: num on a dsa_db structure of type DSA_DB_PORT would access memory which is unused, because we only use dsa_db :: dp for DSA_DB_PORT, and this is mapped at the same address with dsa_db :: dev for DSA_DB_BRIDGE, thanks to the union definition. It is correct to fix up dsa_db :: bridge :: num only from code paths that come from the bridge / switchdev, so move these there. Signed-off-by: Vladimir Oltean Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/20230329133819.697642-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski --- net/dsa/port.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/dsa/port.c b/net/dsa/port.c index 67ad1adec2a2..15cee17769e9 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -1028,9 +1028,6 @@ static int dsa_port_host_fdb_add(struct dsa_port *dp, .db = db, }; - if (!dp->ds->fdb_isolation) - info.db.bridge.num = 0; - return dsa_port_notify(dp, DSA_NOTIFIER_HOST_FDB_ADD, &info); } @@ -1055,6 +1052,9 @@ int dsa_port_bridge_host_fdb_add(struct dsa_port *dp, }; int err; + if (!dp->ds->fdb_isolation) + db.bridge.num = 0; + /* Avoid a call to __dev_set_promiscuity() on the master, which * requires rtnl_lock(), since we can't guarantee that is held here, * and we can't take it either. @@ -1079,9 +1079,6 @@ static int dsa_port_host_fdb_del(struct dsa_port *dp, .db = db, }; - if (!dp->ds->fdb_isolation) - info.db.bridge.num = 0; - return dsa_port_notify(dp, DSA_NOTIFIER_HOST_FDB_DEL, &info); } @@ -1106,6 +1103,9 @@ int dsa_port_bridge_host_fdb_del(struct dsa_port *dp, }; int err; + if (!dp->ds->fdb_isolation) + db.bridge.num = 0; + if (master->priv_flags & IFF_UNICAST_FLT) { err = dev_uc_del(master, addr); if (err) @@ -1210,9 +1210,6 @@ static int dsa_port_host_mdb_add(const struct dsa_port *dp, .db = db, }; - if (!dp->ds->fdb_isolation) - info.db.bridge.num = 0; - return dsa_port_notify(dp, DSA_NOTIFIER_HOST_MDB_ADD, &info); } @@ -1237,6 +1234,9 @@ int dsa_port_bridge_host_mdb_add(const struct dsa_port *dp, }; int err; + if (!dp->ds->fdb_isolation) + db.bridge.num = 0; + err = dev_mc_add(master, mdb->addr); if (err) return err; @@ -1254,9 +1254,6 @@ static int dsa_port_host_mdb_del(const struct dsa_port *dp, .db = db, }; - if (!dp->ds->fdb_isolation) - info.db.bridge.num = 0; - return dsa_port_notify(dp, DSA_NOTIFIER_HOST_MDB_DEL, &info); } @@ -1281,6 +1278,9 @@ int dsa_port_bridge_host_mdb_del(const struct dsa_port *dp, }; int err; + if (!dp->ds->fdb_isolation) + db.bridge.num = 0; + err = dev_mc_del(master, mdb->addr); if (err) return err; -- cgit v1.2.3 From 2384127e98db52a6ac2577924ad9cae25f3e7472 Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Wed, 29 Mar 2023 11:54:52 +0200 Subject: net/sched: act_tunnel_key: add support for "don't fragment" extend "act_tunnel_key" to allow specifying TUNNEL_DONT_FRAGMENT. Suggested-by: Ilya Maximets Reviewed-by: Pedro Tammela Acked-by: Jamal Hadi Salim Signed-off-by: Davide Caratti Signed-off-by: Jakub Kicinski --- include/uapi/linux/tc_act/tc_tunnel_key.h | 1 + net/sched/act_tunnel_key.c | 5 +++++ 2 files changed, 6 insertions(+) (limited to 'net') diff --git a/include/uapi/linux/tc_act/tc_tunnel_key.h b/include/uapi/linux/tc_act/tc_tunnel_key.h index 49ad4033951b..37c6f612f161 100644 --- a/include/uapi/linux/tc_act/tc_tunnel_key.h +++ b/include/uapi/linux/tc_act/tc_tunnel_key.h @@ -34,6 +34,7 @@ enum { */ TCA_TUNNEL_KEY_ENC_TOS, /* u8 */ TCA_TUNNEL_KEY_ENC_TTL, /* u8 */ + TCA_TUNNEL_KEY_NO_FRAG, /* flag */ __TCA_TUNNEL_KEY_MAX, }; diff --git a/net/sched/act_tunnel_key.c b/net/sched/act_tunnel_key.c index 2d12d2626415..0c8aa7e686ea 100644 --- a/net/sched/act_tunnel_key.c +++ b/net/sched/act_tunnel_key.c @@ -420,6 +420,9 @@ static int tunnel_key_init(struct net *net, struct nlattr *nla, nla_get_u8(tb[TCA_TUNNEL_KEY_NO_CSUM])) flags &= ~TUNNEL_CSUM; + if (nla_get_flag(tb[TCA_TUNNEL_KEY_NO_FRAG])) + flags |= TUNNEL_DONT_FRAGMENT; + if (tb[TCA_TUNNEL_KEY_ENC_DST_PORT]) dst_port = nla_get_be16(tb[TCA_TUNNEL_KEY_ENC_DST_PORT]); @@ -747,6 +750,8 @@ static int tunnel_key_dump(struct sk_buff *skb, struct tc_action *a, key->tp_dst)) || nla_put_u8(skb, TCA_TUNNEL_KEY_NO_CSUM, !(key->tun_flags & TUNNEL_CSUM)) || + ((key->tun_flags & TUNNEL_DONT_FRAGMENT) && + nla_put_flag(skb, TCA_TUNNEL_KEY_NO_FRAG)) || tunnel_key_opts_dump(skb, info)) goto nla_put_failure; -- cgit v1.2.3 From f9d2b1e146e0f82f3d04629afd92698522058361 Mon Sep 17 00:00:00 2001 From: Bobby Eshleman Date: Wed, 29 Mar 2023 16:51:58 +0000 Subject: virtio/vsock: fix leaks due to missing skb owner This patch sets the skb owner in the recv and send path for virtio. For the send path, this solves the leak caused when virtio_transport_purge_skbs() finds skb->sk is always NULL and therefore never matches it with the current socket. Setting the owner upon allocation fixes this. For the recv path, this ensures correctness of accounting and also correct transfer of ownership in vsock_loopback (when skbs are sent from one socket and received by another). Fixes: 71dc9ec9ac7d ("virtio/vsock: replace virtio_vsock_pkt with sk_buff") Signed-off-by: Bobby Eshleman Reported-by: Cong Wang Link: https://lore.kernel.org/all/ZCCbATwov4U+GBUv@pop-os.localdomain/ Reviewed-by: Stefano Garzarella Signed-off-by: David S. Miller --- net/vmw_vsock/virtio_transport_common.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index 37934dfe72f4..ee78b4082ef9 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -94,6 +94,11 @@ virtio_transport_alloc_skb(struct virtio_vsock_pkt_info *info, info->op, info->flags); + if (info->vsk && !skb_set_owner_sk_safe(skb, sk_vsock(info->vsk))) { + WARN_ONCE(1, "failed to allocate skb on vsock socket with sk_refcnt == 0\n"); + goto out; + } + return skb; out: @@ -1303,6 +1308,11 @@ void virtio_transport_recv_pkt(struct virtio_transport *t, goto free_pkt; } + if (!skb_set_owner_sk_safe(skb, sk)) { + WARN_ONCE(1, "receiving vsock socket has sk_refcnt == 0\n"); + goto free_pkt; + } + vsk = vsock_sk(sk); lock_sock(sk); -- cgit v1.2.3 From ee05d90d0ac7f1f6e8e3403f02a39d83e0ce6729 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 29 Mar 2023 13:13:48 -0700 Subject: tcp: Refine SYN handling for PAWS. Our Network Load Balancer (NLB) [0] has multiple nodes with different IP addresses, and each node forwards TCP flows from clients to backend targets. NLB has an option to preserve the client's source IP address and port when routing packets to backend targets. [1] When a client connects to two different NLB nodes, they may select the same backend target. Then, if the client has used the same source IP and port, the two flows at the backend side will have the same 4-tuple. While testing around such cases, I saw these sequences on the backend target. IP 10.0.0.215.60000 > 10.0.3.249.10000: Flags [S], seq 2819965599, win 62727, options [mss 8365,sackOK,TS val 1029816180 ecr 0,nop,wscale 7], length 0 IP 10.0.3.249.10000 > 10.0.0.215.60000: Flags [S.], seq 3040695044, ack 2819965600, win 62643, options [mss 8961,sackOK,TS val 1224784076 ecr 1029816180,nop,wscale 7], length 0 IP 10.0.0.215.60000 > 10.0.3.249.10000: Flags [.], ack 1, win 491, options [nop,nop,TS val 1029816181 ecr 1224784076], length 0 IP 10.0.0.215.60000 > 10.0.3.249.10000: Flags [S], seq 2681819307, win 62727, options [mss 8365,sackOK,TS val 572088282 ecr 0,nop,wscale 7], length 0 IP 10.0.3.249.10000 > 10.0.0.215.60000: Flags [.], ack 1, win 490, options [nop,nop,TS val 1224794914 ecr 1029816181,nop,nop,sack 1 {4156821004:4156821005}], length 0 It seems to be working correctly, but the last ACK was generated by tcp_send_dupack() and PAWSEstab was increased. This is because the second connection has a smaller timestamp than the first one. In this case, we should send a dup ACK in tcp_send_challenge_ack() to increase the correct counter and rate-limit it properly. Let's check the SYN flag after the PAWS tests to avoid adding unnecessary overhead for most packets. Link: https://docs.aws.amazon.com/elasticloadbalancing/latest/network/introduction.html [0] Link: https://docs.aws.amazon.com/elasticloadbalancing/latest/network/load-balancer-target-groups.html#client-ip-preservation [1] Signed-off-by: Kuniyuki Iwashima Reviewed-by: Jason Xing Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 2b75cd9e2e92..a057330d6f59 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5714,6 +5714,8 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, tp->rx_opt.saw_tstamp && tcp_paws_discard(sk, skb)) { if (!th->rst) { + if (unlikely(th->syn)) + goto syn_challenge; NET_INC_STATS(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED); if (!tcp_oow_rate_limited(sock_net(sk), skb, LINUX_MIB_TCPACKSKIPPEDPAWS, -- cgit v1.2.3 From 44d807320000db0d0013372ad39b53e12d52f758 Mon Sep 17 00:00:00 2001 From: Ziyang Xuan Date: Thu, 30 Mar 2023 09:25:32 +0800 Subject: net: qrtr: Fix a refcount bug in qrtr_recvmsg() Syzbot reported a bug as following: refcount_t: addition on 0; use-after-free. ... RIP: 0010:refcount_warn_saturate+0x17c/0x1f0 lib/refcount.c:25 ... Call Trace: __refcount_add include/linux/refcount.h:199 [inline] __refcount_inc include/linux/refcount.h:250 [inline] refcount_inc include/linux/refcount.h:267 [inline] kref_get include/linux/kref.h:45 [inline] qrtr_node_acquire net/qrtr/af_qrtr.c:202 [inline] qrtr_node_lookup net/qrtr/af_qrtr.c:398 [inline] qrtr_send_resume_tx net/qrtr/af_qrtr.c:1003 [inline] qrtr_recvmsg+0x85f/0x990 net/qrtr/af_qrtr.c:1070 sock_recvmsg_nosec net/socket.c:1017 [inline] sock_recvmsg+0xe2/0x160 net/socket.c:1038 qrtr_ns_worker+0x170/0x1700 net/qrtr/ns.c:688 process_one_work+0x991/0x15c0 kernel/workqueue.c:2390 worker_thread+0x669/0x1090 kernel/workqueue.c:2537 It occurs in the concurrent scenario of qrtr_recvmsg() and qrtr_endpoint_unregister() as following: cpu0 cpu1 qrtr_recvmsg qrtr_endpoint_unregister qrtr_send_resume_tx qrtr_node_release qrtr_node_lookup mutex_lock(&qrtr_node_lock) spin_lock_irqsave(&qrtr_nodes_lock, ) refcount_dec_and_test(&node->ref) [node->ref == 0] radix_tree_lookup [node != NULL] __qrtr_node_release qrtr_node_acquire spin_lock_irqsave(&qrtr_nodes_lock, ) kref_get(&node->ref) [WARNING] ... mutex_unlock(&qrtr_node_lock) Use qrtr_node_lock to protect qrtr_node_lookup() implementation, this is actually improving the protection of node reference. Fixes: 0a7e0d0ef054 ("net: qrtr: Migrate node lookup tree to spinlock") Reported-by: syzbot+a7492efaa5d61b51db23@syzkaller.appspotmail.com Link: https://syzkaller.appspot.com/bug?extid=a7492efaa5d61b51db23 Signed-off-by: Ziyang Xuan Signed-off-by: David S. Miller --- net/qrtr/af_qrtr.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/qrtr/af_qrtr.c b/net/qrtr/af_qrtr.c index 5c2fb992803b..3a70255c8d02 100644 --- a/net/qrtr/af_qrtr.c +++ b/net/qrtr/af_qrtr.c @@ -393,10 +393,12 @@ static struct qrtr_node *qrtr_node_lookup(unsigned int nid) struct qrtr_node *node; unsigned long flags; + mutex_lock(&qrtr_node_lock); spin_lock_irqsave(&qrtr_nodes_lock, flags); node = radix_tree_lookup(&qrtr_nodes, nid); node = qrtr_node_acquire(node); spin_unlock_irqrestore(&qrtr_nodes_lock, flags); + mutex_unlock(&qrtr_node_lock); return node; } -- cgit v1.2.3 From 154e07c164859fc90bf4e8143f2f6c1af9f3a35e Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Thu, 30 Mar 2023 11:54:42 +0200 Subject: l2tp: generate correct module alias strings Commit 65b32f801bfb ("uapi: move IPPROTO_L2TP to in.h") moved the definition of IPPROTO_L2TP from a define to an enum, but since __stringify doesn't work properly with enums, we ended up breaking the modalias strings for the l2tp modules: $ modinfo l2tp_ip l2tp_ip6 | grep alias alias: net-pf-2-proto-IPPROTO_L2TP alias: net-pf-2-proto-2-type-IPPROTO_L2TP alias: net-pf-10-proto-IPPROTO_L2TP alias: net-pf-10-proto-2-type-IPPROTO_L2TP Use the resolved number directly in MODULE_ALIAS_*() macros (as we already do with SOCK_DGRAM) to fix the alias strings: $ modinfo l2tp_ip l2tp_ip6 | grep alias alias: net-pf-2-proto-115 alias: net-pf-2-proto-115-type-2 alias: net-pf-10-proto-115 alias: net-pf-10-proto-115-type-2 Moreover, fix the ordering of the parameters passed to MODULE_ALIAS_NET_PF_PROTO_TYPE() by switching proto and type. Fixes: 65b32f801bfb ("uapi: move IPPROTO_L2TP to in.h") Link: https://lore.kernel.org/lkml/ZCQt7hmodtUaBlCP@righiandr-XPS-13-7390 Signed-off-by: Guillaume Nault Signed-off-by: Andrea Righi Reviewed-by: Wojciech Drewek Tested-by: Wojciech Drewek Signed-off-by: David S. Miller --- net/l2tp/l2tp_ip.c | 8 ++++---- net/l2tp/l2tp_ip6.c | 8 ++++---- 2 files changed, 8 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/l2tp/l2tp_ip.c b/net/l2tp/l2tp_ip.c index 4db5a554bdbd..41a74fc84ca1 100644 --- a/net/l2tp/l2tp_ip.c +++ b/net/l2tp/l2tp_ip.c @@ -677,8 +677,8 @@ MODULE_AUTHOR("James Chapman "); MODULE_DESCRIPTION("L2TP over IP"); MODULE_VERSION("1.0"); -/* Use the value of SOCK_DGRAM (2) directory, because __stringify doesn't like - * enums +/* Use the values of SOCK_DGRAM (2) as type and IPPROTO_L2TP (115) as protocol, + * because __stringify doesn't like enums */ -MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET, 2, IPPROTO_L2TP); -MODULE_ALIAS_NET_PF_PROTO(PF_INET, IPPROTO_L2TP); +MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET, 115, 2); +MODULE_ALIAS_NET_PF_PROTO(PF_INET, 115); diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c index 2478aa60145f..5137ea1861ce 100644 --- a/net/l2tp/l2tp_ip6.c +++ b/net/l2tp/l2tp_ip6.c @@ -806,8 +806,8 @@ MODULE_AUTHOR("Chris Elston "); MODULE_DESCRIPTION("L2TP IP encapsulation for IPv6"); MODULE_VERSION("1.0"); -/* Use the value of SOCK_DGRAM (2) directory, because __stringify doesn't like - * enums +/* Use the values of SOCK_DGRAM (2) as type and IPPROTO_L2TP (115) as protocol, + * because __stringify doesn't like enums */ -MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET6, 2, IPPROTO_L2TP); -MODULE_ALIAS_NET_PF_PROTO(PF_INET6, IPPROTO_L2TP); +MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET6, 115, 2); +MODULE_ALIAS_NET_PF_PROTO(PF_INET6, 115); -- cgit v1.2.3 From ffa5395a7901e83a68d88207c4592962906641bd Mon Sep 17 00:00:00 2001 From: Arseniy Krasnov Date: Fri, 31 Mar 2023 10:56:41 +0300 Subject: vsock/vmci: convert VMCI error code to -ENOMEM on send This adds conversion of VMCI specific error code to general -ENOMEM. It is needed, because af_vsock.c passes error value returned from transport to the user, which does not expect to get VMCI_ERROR_* values. Fixes: c43170b7e157 ("vsock: return errors other than -ENOMEM to socket") Signed-off-by: Arseniy Krasnov Reviewed-by: Vishnu Dasa Reviewed-by: Stefano Garzarella Signed-off-by: David S. Miller --- net/vmw_vsock/vmci_transport.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c index 36eb16a40745..95cc4d79ba29 100644 --- a/net/vmw_vsock/vmci_transport.c +++ b/net/vmw_vsock/vmci_transport.c @@ -1842,7 +1842,13 @@ static ssize_t vmci_transport_stream_enqueue( struct msghdr *msg, size_t len) { - return vmci_qpair_enquev(vmci_trans(vsk)->qpair, msg, len, 0); + ssize_t err; + + err = vmci_qpair_enquev(vmci_trans(vsk)->qpair, msg, len, 0); + if (err < 0) + err = -ENOMEM; + + return err; } static s64 vmci_transport_stream_has_data(struct vsock_sock *vsk) -- cgit v1.2.3 From df1875c4f345092716f1529d6e8c43e7fc7f250a Mon Sep 17 00:00:00 2001 From: Ryder Lee Date: Tue, 28 Mar 2023 01:07:42 +0800 Subject: wifi: mac80211: enable EHT mesh support Similar to AP beacon, this enables the basic mesh EHT mode, including EHT operation IE and the fixed field of EHT operation information IE. As for the optional part (i.e. preamble puncturing bitmap) will be added in future patch. Tested-by: Lian Chen Signed-off-by: Ryder Lee Link: https://lore.kernel.org/r/1e0ddb9001312451c3e99c4eed2072caf8075f61.1679935259.git.ryder.lee@mediatek.com Signed-off-by: Johannes Berg --- net/mac80211/ieee80211_i.h | 2 ++ net/mac80211/mesh.c | 73 ++++++++++++++++++++++++++++++++++++++++++++-- net/mac80211/mesh.h | 4 +++ net/mac80211/mesh_plink.c | 16 ++++++++-- net/mac80211/util.c | 71 ++++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 162 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 8c058351e629..9b7e184430b8 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2479,6 +2479,8 @@ void ieee80211_ie_build_he_6ghz_cap(struct ieee80211_sub_if_data *sdata, enum ieee80211_smps_mode smps_mode, struct sk_buff *skb); u8 *ieee80211_ie_build_he_oper(u8 *pos, struct cfg80211_chan_def *chandef); +u8 *ieee80211_ie_build_eht_oper(u8 *pos, struct cfg80211_chan_def *chandef, + const struct ieee80211_sta_eht_cap *eht_cap); int ieee80211_parse_bitrates(enum nl80211_chan_width width, const struct ieee80211_supported_band *sband, const u8 *srates, int srates_len, u32 *rates); diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c index 6b94cf2a4046..f72333201903 100644 --- a/net/mac80211/mesh.c +++ b/net/mac80211/mesh.c @@ -105,7 +105,7 @@ bool mesh_matches_local(struct ieee80211_sub_if_data *sdata, ieee80211_chandef_vht_oper(&sdata->local->hw, vht_cap_info, ie->vht_operation, ie->ht_operation, &sta_chan_def); - ieee80211_chandef_he_6ghz_oper(sdata, ie->he_operation, NULL, + ieee80211_chandef_he_6ghz_oper(sdata, ie->he_operation, ie->eht_operation, &sta_chan_def); if (!cfg80211_chandef_compatible(&sdata->vif.bss_conf.chandef, @@ -639,6 +639,65 @@ int mesh_add_he_6ghz_cap_ie(struct ieee80211_sub_if_data *sdata, return 0; } +int mesh_add_eht_cap_ie(struct ieee80211_sub_if_data *sdata, + struct sk_buff *skb, u8 ie_len) +{ + const struct ieee80211_sta_he_cap *he_cap; + const struct ieee80211_sta_eht_cap *eht_cap; + struct ieee80211_supported_band *sband; + u8 *pos; + + sband = ieee80211_get_sband(sdata); + if (!sband) + return -EINVAL; + + he_cap = ieee80211_get_he_iftype_cap(sband, NL80211_IFTYPE_MESH_POINT); + eht_cap = ieee80211_get_eht_iftype_cap(sband, NL80211_IFTYPE_MESH_POINT); + if (!he_cap || !eht_cap || + sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_20_NOHT || + sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_5 || + sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_10) + return 0; + + if (skb_tailroom(skb) < ie_len) + return -ENOMEM; + + pos = skb_put(skb, ie_len); + ieee80211_ie_build_eht_cap(pos, he_cap, eht_cap, pos + ie_len, false); + + return 0; +} + +int mesh_add_eht_oper_ie(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb) +{ + const struct ieee80211_sta_eht_cap *eht_cap; + struct ieee80211_supported_band *sband; + u32 len; + u8 *pos; + + sband = ieee80211_get_sband(sdata); + if (!sband) + return -EINVAL; + + eht_cap = ieee80211_get_eht_iftype_cap(sband, NL80211_IFTYPE_MESH_POINT); + if (!eht_cap || + sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_20_NOHT || + sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_5 || + sdata->vif.bss_conf.chandef.width == NL80211_CHAN_WIDTH_10) + return 0; + + len = 2 + 1 + offsetof(struct ieee80211_eht_operation, optional) + + offsetof(struct ieee80211_eht_operation_info, optional); + + if (skb_tailroom(skb) < len) + return -ENOMEM; + + pos = skb_put(skb, len); + ieee80211_ie_build_eht_oper(pos, &sdata->vif.bss_conf.chandef, eht_cap); + + return 0; +} + static void ieee80211_mesh_path_timer(struct timer_list *t) { struct ieee80211_sub_if_data *sdata = @@ -697,6 +756,9 @@ ieee80211_mesh_update_bss_params(struct ieee80211_sub_if_data *sdata, if (he_oper) sdata->vif.bss_conf.he_oper.params = __le32_to_cpu(he_oper->he_oper_params); + + sdata->vif.bss_conf.eht_support = + !!ieee80211_get_eht_iftype_cap(sband, NL80211_IFTYPE_MESH_POINT); } bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata, @@ -903,7 +965,7 @@ ieee80211_mesh_build_beacon(struct ieee80211_if_mesh *ifmsh) struct ieee80211_chanctx_conf *chanctx_conf; struct mesh_csa_settings *csa; enum nl80211_band band; - u8 ie_len_he_cap; + u8 ie_len_he_cap, ie_len_eht_cap; u8 *pos; struct ieee80211_sub_if_data *sdata; int hdr_len = offsetofend(struct ieee80211_mgmt, u.beacon); @@ -916,6 +978,8 @@ ieee80211_mesh_build_beacon(struct ieee80211_if_mesh *ifmsh) ie_len_he_cap = ieee80211_ie_len_he_cap(sdata, NL80211_IFTYPE_MESH_POINT); + ie_len_eht_cap = ieee80211_ie_len_eht_cap(sdata, + NL80211_IFTYPE_MESH_POINT); head_len = hdr_len + 2 + /* NULL SSID */ /* Channel Switch Announcement */ @@ -939,6 +1003,9 @@ ieee80211_mesh_build_beacon(struct ieee80211_if_mesh *ifmsh) 2 + 1 + sizeof(struct ieee80211_he_operation) + sizeof(struct ieee80211_he_6ghz_oper) + 2 + 1 + sizeof(struct ieee80211_he_6ghz_capa) + + ie_len_eht_cap + + 2 + 1 + offsetof(struct ieee80211_eht_operation, optional) + + offsetof(struct ieee80211_eht_operation_info, optional) + ifmsh->ie_len; bcn = kzalloc(sizeof(*bcn) + head_len + tail_len, GFP_KERNEL); @@ -1059,6 +1126,8 @@ ieee80211_mesh_build_beacon(struct ieee80211_if_mesh *ifmsh) mesh_add_he_cap_ie(sdata, skb, ie_len_he_cap) || mesh_add_he_oper_ie(sdata, skb) || mesh_add_he_6ghz_cap_ie(sdata, skb) || + mesh_add_eht_cap_ie(sdata, skb, ie_len_eht_cap) || + mesh_add_eht_oper_ie(sdata, skb) || mesh_add_vendor_ies(sdata, skb)) goto out_free; diff --git a/net/mac80211/mesh.h b/net/mac80211/mesh.h index 13f394e677ae..022f41292a05 100644 --- a/net/mac80211/mesh.h +++ b/net/mac80211/mesh.h @@ -234,6 +234,10 @@ int mesh_add_he_oper_ie(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb); int mesh_add_he_6ghz_cap_ie(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb); +int mesh_add_eht_cap_ie(struct ieee80211_sub_if_data *sdata, + struct sk_buff *skb, u8 ie_len); +int mesh_add_eht_oper_ie(struct ieee80211_sub_if_data *sdata, + struct sk_buff *skb); void mesh_rmc_free(struct ieee80211_sub_if_data *sdata); int mesh_rmc_init(struct ieee80211_sub_if_data *sdata); void ieee80211s_init(void); diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c index ddfe5102b9a4..8f168bc4e4b8 100644 --- a/net/mac80211/mesh_plink.c +++ b/net/mac80211/mesh_plink.c @@ -219,12 +219,14 @@ static int mesh_plink_frame_tx(struct ieee80211_sub_if_data *sdata, bool include_plid = false; u16 peering_proto = 0; u8 *pos, ie_len = 4; - u8 ie_len_he_cap; + u8 ie_len_he_cap, ie_len_eht_cap; int hdr_len = offsetofend(struct ieee80211_mgmt, u.action.u.self_prot); int err = -ENOMEM; ie_len_he_cap = ieee80211_ie_len_he_cap(sdata, NL80211_IFTYPE_MESH_POINT); + ie_len_eht_cap = ieee80211_ie_len_eht_cap(sdata, + NL80211_IFTYPE_MESH_POINT); skb = dev_alloc_skb(local->tx_headroom + hdr_len + 2 + /* capability info */ @@ -241,6 +243,9 @@ static int mesh_plink_frame_tx(struct ieee80211_sub_if_data *sdata, 2 + 1 + sizeof(struct ieee80211_he_operation) + sizeof(struct ieee80211_he_6ghz_oper) + 2 + 1 + sizeof(struct ieee80211_he_6ghz_capa) + + ie_len_eht_cap + + 2 + 1 + offsetof(struct ieee80211_eht_operation, optional) + + offsetof(struct ieee80211_eht_operation_info, optional) + 2 + 8 + /* peering IE */ sdata->u.mesh.ie_len); if (!skb) @@ -332,7 +337,9 @@ static int mesh_plink_frame_tx(struct ieee80211_sub_if_data *sdata, mesh_add_vht_oper_ie(sdata, skb) || mesh_add_he_cap_ie(sdata, skb, ie_len_he_cap) || mesh_add_he_oper_ie(sdata, skb) || - mesh_add_he_6ghz_cap_ie(sdata, skb)) + mesh_add_he_6ghz_cap_ie(sdata, skb) || + mesh_add_eht_cap_ie(sdata, skb, ie_len_eht_cap) || + mesh_add_eht_oper_ie(sdata, skb)) goto free; } @@ -451,6 +458,11 @@ static void mesh_sta_info_init(struct ieee80211_sub_if_data *sdata, elems->he_6ghz_capa, &sta->deflink); + ieee80211_eht_cap_ie_to_sta_eht_cap(sdata, sband, elems->he_cap, + elems->he_cap_len, + elems->eht_cap, elems->eht_cap_len, + &sta->deflink); + if (bw != sta->sta.deflink.bandwidth) changed |= IEEE80211_RC_BW_CHANGED; diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 9f5cc16538f3..1527d6aafc14 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -3485,6 +3485,77 @@ out: return pos; } +u8 *ieee80211_ie_build_eht_oper(u8 *pos, struct cfg80211_chan_def *chandef, + const struct ieee80211_sta_eht_cap *eht_cap) + +{ + const struct ieee80211_eht_mcs_nss_supp_20mhz_only *eht_mcs_nss = + &eht_cap->eht_mcs_nss_supp.only_20mhz; + struct ieee80211_eht_operation *eht_oper; + struct ieee80211_eht_operation_info *eht_oper_info; + u8 eht_oper_len = offsetof(struct ieee80211_eht_operation, optional); + u8 eht_oper_info_len = + offsetof(struct ieee80211_eht_operation_info, optional); + u8 chan_width = 0; + + *pos++ = WLAN_EID_EXTENSION; + *pos++ = 1 + eht_oper_len + eht_oper_info_len; + *pos++ = WLAN_EID_EXT_EHT_OPERATION; + + eht_oper = (struct ieee80211_eht_operation *)pos; + + memcpy(&eht_oper->basic_mcs_nss, eht_mcs_nss, sizeof(*eht_mcs_nss)); + eht_oper->params |= IEEE80211_EHT_OPER_INFO_PRESENT; + pos += eht_oper_len; + + eht_oper_info = + (struct ieee80211_eht_operation_info *)eht_oper->optional; + + eht_oper_info->ccfs0 = + ieee80211_frequency_to_channel(chandef->center_freq1); + if (chandef->center_freq2) + eht_oper_info->ccfs1 = + ieee80211_frequency_to_channel(chandef->center_freq2); + else + eht_oper_info->ccfs1 = 0; + + switch (chandef->width) { + case NL80211_CHAN_WIDTH_320: + chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_320MHZ; + eht_oper_info->ccfs1 = eht_oper_info->ccfs0; + if (chandef->chan->center_freq < chandef->center_freq1) + eht_oper_info->ccfs0 -= 16; + else + eht_oper_info->ccfs0 += 16; + break; + case NL80211_CHAN_WIDTH_160: + eht_oper_info->ccfs1 = eht_oper_info->ccfs0; + if (chandef->chan->center_freq < chandef->center_freq1) + eht_oper_info->ccfs0 -= 8; + else + eht_oper_info->ccfs0 += 8; + fallthrough; + case NL80211_CHAN_WIDTH_80P80: + chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_160MHZ; + break; + case NL80211_CHAN_WIDTH_80: + chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_80MHZ; + break; + case NL80211_CHAN_WIDTH_40: + chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_40MHZ; + break; + default: + chan_width = IEEE80211_EHT_OPER_CHAN_WIDTH_20MHZ; + break; + } + eht_oper_info->control = chan_width; + pos += eht_oper_info_len; + + /* TODO: eht_oper_info->optional */ + + return pos; +} + bool ieee80211_chandef_ht_oper(const struct ieee80211_ht_operation *ht_oper, struct cfg80211_chan_def *chandef) { -- cgit v1.2.3 From 7d63b67125382ff0ffdfca434acbc94a38bd092b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 30 Mar 2023 17:45:02 +0000 Subject: icmp: guard against too small mtu syzbot was able to trigger a panic [1] in icmp_glue_bits(), or more exactly in skb_copy_and_csum_bits() There is no repro yet, but I think the issue is that syzbot manages to lower device mtu to a small value, fooling __icmp_send() __icmp_send() must make sure there is enough room for the packet to include at least the headers. We might in the future refactor skb_copy_and_csum_bits() and its callers to no longer crash when something bad happens. [1] kernel BUG at net/core/skbuff.c:3343 ! invalid opcode: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 15766 Comm: syz-executor.0 Not tainted 6.3.0-rc4-syzkaller-00039-gffe78bbd5121 #0 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014 RIP: 0010:skb_copy_and_csum_bits+0x798/0x860 net/core/skbuff.c:3343 Code: f0 c1 c8 08 41 89 c6 e9 73 ff ff ff e8 61 48 d4 f9 e9 41 fd ff ff 48 8b 7c 24 48 e8 52 48 d4 f9 e9 c3 fc ff ff e8 c8 27 84 f9 <0f> 0b 48 89 44 24 28 e8 3c 48 d4 f9 48 8b 44 24 28 e9 9d fb ff ff RSP: 0018:ffffc90000007620 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 00000000000001e8 RCX: 0000000000000100 RDX: ffff8880276f6280 RSI: ffffffff87fdd138 RDI: 0000000000000005 RBP: 0000000000000000 R08: 0000000000000005 R09: 0000000000000000 R10: 00000000000001e8 R11: 0000000000000001 R12: 000000000000003c R13: 0000000000000000 R14: ffff888028244868 R15: 0000000000000b0e FS: 00007fbc81f1c700(0000) GS:ffff88802ca00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b2df43000 CR3: 00000000744db000 CR4: 0000000000150ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: icmp_glue_bits+0x7b/0x210 net/ipv4/icmp.c:353 __ip_append_data+0x1d1b/0x39f0 net/ipv4/ip_output.c:1161 ip_append_data net/ipv4/ip_output.c:1343 [inline] ip_append_data+0x115/0x1a0 net/ipv4/ip_output.c:1322 icmp_push_reply+0xa8/0x440 net/ipv4/icmp.c:370 __icmp_send+0xb80/0x1430 net/ipv4/icmp.c:765 ipv4_send_dest_unreach net/ipv4/route.c:1239 [inline] ipv4_link_failure+0x5a9/0x9e0 net/ipv4/route.c:1246 dst_link_failure include/net/dst.h:423 [inline] arp_error_report+0xcb/0x1c0 net/ipv4/arp.c:296 neigh_invalidate+0x20d/0x560 net/core/neighbour.c:1079 neigh_timer_handler+0xc77/0xff0 net/core/neighbour.c:1166 call_timer_fn+0x1a0/0x580 kernel/time/timer.c:1700 expire_timers+0x29b/0x4b0 kernel/time/timer.c:1751 __run_timers kernel/time/timer.c:2022 [inline] Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot+d373d60fddbdc915e666@syzkaller.appspotmail.com Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20230330174502.1915328-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/icmp.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net') diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 8cebb476b3ab..b8607763d113 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -749,6 +749,11 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, room = 576; room -= sizeof(struct iphdr) + icmp_param.replyopts.opt.opt.optlen; room -= sizeof(struct icmphdr); + /* Guard against tiny mtu. We need to include at least one + * IP network header for this message to make any sense. + */ + if (room <= (int)sizeof(struct iphdr)) + goto ende; icmp_param.data_len = skb_in->len - icmp_param.offset; if (icmp_param.data_len > room) -- cgit v1.2.3 From ea4f1009408efb4989a0f139b70fb338e7f687d0 Mon Sep 17 00:00:00 2001 From: Zheng Wang Date: Mon, 13 Mar 2023 22:43:25 +0800 Subject: 9p/xen : Fix use after free bug in xen_9pfs_front_remove due to race condition In xen_9pfs_front_probe, it calls xen_9pfs_front_alloc_dataring to init priv->rings and bound &ring->work with p9_xen_response. When it calls xen_9pfs_front_event_handler to handle IRQ requests, it will finally call schedule_work to start the work. When we call xen_9pfs_front_remove to remove the driver, there may be a sequence as follows: Fix it by finishing the work before cleanup in xen_9pfs_front_free. Note that, this bug is found by static analysis, which might be false positive. CPU0 CPU1 |p9_xen_response xen_9pfs_front_remove| xen_9pfs_front_free| kfree(priv) | //free priv | |p9_tag_lookup |//use priv->client Fixes: 71ebd71921e4 ("xen/9pfs: connect to the backend") Signed-off-by: Zheng Wang Reviewed-by: Michal Swiatkowski Signed-off-by: Eric Van Hensbergen --- net/9p/trans_xen.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/9p/trans_xen.c b/net/9p/trans_xen.c index c64050e839ac..1fffe2bed5b0 100644 --- a/net/9p/trans_xen.c +++ b/net/9p/trans_xen.c @@ -280,6 +280,10 @@ static void xen_9pfs_front_free(struct xen_9pfs_front_priv *priv) write_unlock(&xen_9pfs_lock); for (i = 0; i < priv->num_rings; i++) { + struct xen_9pfs_dataring *ring = &priv->rings[i]; + + cancel_work_sync(&ring->work); + if (!priv->rings[i].intf) break; if (priv->rings[i].irq > 0) -- cgit v1.2.3 From 275b471e3d2daf1472ae8fa70dc1b50c9e0b9e75 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 30 Mar 2023 19:21:44 -0700 Subject: net: don't let netpoll invoke NAPI if in xmit context Commit 0db3dc73f7a3 ("[NETPOLL]: tx lock deadlock fix") narrowed down the region under netif_tx_trylock() inside netpoll_send_skb(). (At that point in time netif_tx_trylock() would lock all queues of the device.) Taking the tx lock was problematic because driver's cleanup method may take the same lock. So the change made us hold the xmit lock only around xmit, and expected the driver to take care of locking within ->ndo_poll_controller(). Unfortunately this only works if netpoll isn't itself called with the xmit lock already held. Netpoll code is careful and uses trylock(). The drivers, however, may be using plain lock(). Printing while holding the xmit lock is going to result in rare deadlocks. Luckily we record the xmit lock owners, so we can scan all the queues, the same way we scan NAPI owners. If any of the xmit locks is held by the local CPU we better not attempt any polling. It would be nice if we could narrow down the check to only the NAPIs and the queue we're trying to use. I don't see a way to do that now. Reported-by: Roman Gushchin Fixes: 0db3dc73f7a3 ("[NETPOLL]: tx lock deadlock fix") Signed-off-by: Jakub Kicinski Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/netpoll.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/netpoll.c b/net/core/netpoll.c index a089b704b986..e6a739b1afa9 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -137,6 +137,20 @@ static void queue_process(struct work_struct *work) } } +static int netif_local_xmit_active(struct net_device *dev) +{ + int i; + + for (i = 0; i < dev->num_tx_queues; i++) { + struct netdev_queue *txq = netdev_get_tx_queue(dev, i); + + if (READ_ONCE(txq->xmit_lock_owner) == smp_processor_id()) + return 1; + } + + return 0; +} + static void poll_one_napi(struct napi_struct *napi) { int work; @@ -183,7 +197,10 @@ void netpoll_poll_dev(struct net_device *dev) if (!ni || down_trylock(&ni->dev_lock)) return; - if (!netif_running(dev)) { + /* Some drivers will take the same locks in poll and xmit, + * we can't poll if local CPU is already in xmit. + */ + if (!netif_running(dev) || netif_local_xmit_active(dev)) { up(&ni->dev_lock); return; } -- cgit v1.2.3 From 2584024b23552c00d95b50255e47bd18d306d31a Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 1 Apr 2023 19:09:57 -0400 Subject: sctp: check send stream number after wait_for_sndbuf This patch fixes a corner case where the asoc out stream count may change after wait_for_sndbuf. When the main thread in the client starts a connection, if its out stream count is set to N while the in stream count in the server is set to N - 2, another thread in the client keeps sending the msgs with stream number N - 1, and waits for sndbuf before processing INIT_ACK. However, after processing INIT_ACK, the out stream count in the client is shrunk to N - 2, the same to the in stream count in the server. The crash occurs when the thread waiting for sndbuf is awake and sends the msg in a non-existing stream(N - 1), the call trace is as below: KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f] Call Trace: sctp_cmd_send_msg net/sctp/sm_sideeffect.c:1114 [inline] sctp_cmd_interpreter net/sctp/sm_sideeffect.c:1777 [inline] sctp_side_effects net/sctp/sm_sideeffect.c:1199 [inline] sctp_do_sm+0x197d/0x5310 net/sctp/sm_sideeffect.c:1170 sctp_primitive_SEND+0x9f/0xc0 net/sctp/primitive.c:163 sctp_sendmsg_to_asoc+0x10eb/0x1a30 net/sctp/socket.c:1868 sctp_sendmsg+0x8d4/0x1d90 net/sctp/socket.c:2026 inet_sendmsg+0x9d/0xe0 net/ipv4/af_inet.c:825 sock_sendmsg_nosec net/socket.c:722 [inline] sock_sendmsg+0xde/0x190 net/socket.c:745 The fix is to add an unlikely check for the send stream number after the thread wakes up from the wait_for_sndbuf. Fixes: 5bbbbe32a431 ("sctp: introduce stream scheduler foundations") Reported-by: syzbot+47c24ca20a2fa01f082e@syzkaller.appspotmail.com Signed-off-by: Xin Long Signed-off-by: David S. Miller --- net/sctp/socket.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/sctp/socket.c b/net/sctp/socket.c index b91616f819de..218e0982c370 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1830,6 +1830,10 @@ static int sctp_sendmsg_to_asoc(struct sctp_association *asoc, err = sctp_wait_for_sndbuf(asoc, &timeo, msg_len); if (err) goto err; + if (unlikely(sinfo->sinfo_stream >= asoc->stream.outcnt)) { + err = -EINVAL; + goto err; + } } if (sctp_state(asoc, CLOSED)) { -- cgit v1.2.3 From 00d521b39307c3232bdc685c2e9fd82ed973ac24 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 2 Apr 2023 15:37:49 +0300 Subject: net: don't abuse "default" case for unknown ioctl in dev_ifsioc() The "switch (cmd)" block from dev_ifsioc() gained a bit too much unnecessary manual handling of "cmd" in the "default" case, starting with the private ioctls. Clean that up by using the "ellipsis" gcc extension, adding separate cases for the rest of the ioctls, and letting the default case only return -EINVAL. Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/core/dev_ioctl.c | 42 +++++++++++++++++++----------------------- 1 file changed, 19 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 846669426236..1c0256ea522f 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -391,36 +391,32 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, rtnl_lock(); return err; + case SIOCDEVPRIVATE ... SIOCDEVPRIVATE + 15: + return dev_siocdevprivate(dev, ifr, data, cmd); + case SIOCSHWTSTAMP: err = net_hwtstamp_validate(ifr); if (err) return err; fallthrough; - /* - * Unknown or private ioctl - */ - default: - if (cmd >= SIOCDEVPRIVATE && - cmd <= SIOCDEVPRIVATE + 15) - return dev_siocdevprivate(dev, ifr, data, cmd); - - if (cmd == SIOCGMIIPHY || - cmd == SIOCGMIIREG || - cmd == SIOCSMIIREG || - cmd == SIOCSHWTSTAMP || - cmd == SIOCGHWTSTAMP) { - err = dev_eth_ioctl(dev, ifr, cmd); - } else if (cmd == SIOCBONDENSLAVE || - cmd == SIOCBONDRELEASE || - cmd == SIOCBONDSETHWADDR || - cmd == SIOCBONDSLAVEINFOQUERY || - cmd == SIOCBONDINFOQUERY || - cmd == SIOCBONDCHANGEACTIVE) { - err = dev_siocbond(dev, ifr, cmd); - } else - err = -EINVAL; + case SIOCGHWTSTAMP: + case SIOCGMIIPHY: + case SIOCGMIIREG: + case SIOCSMIIREG: + return dev_eth_ioctl(dev, ifr, cmd); + + case SIOCBONDENSLAVE: + case SIOCBONDRELEASE: + case SIOCBONDSETHWADDR: + case SIOCBONDSLAVEINFOQUERY: + case SIOCBONDINFOQUERY: + case SIOCBONDCHANGEACTIVE: + return dev_siocbond(dev, ifr, cmd); + /* Unknown ioctl */ + default: + err = -EINVAL; } return err; } -- cgit v1.2.3 From 1193db2a55b6b04f296f03affdfa80c16ecc3814 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 2 Apr 2023 15:37:50 +0300 Subject: net: simplify handling of dsa_ndo_eth_ioctl() return code In the expression "x == 0 || x != -95", the term "x == 0" does not change the expression's logical value, because 0 != -95, and so, if x is 0, the expression would still be true by virtue of the second term. If x is non-zero, the expression depends on the truth value of the second term anyway. As such, the first term is redundant and can be deleted. Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/core/dev_ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 1c0256ea522f..b299fb23fcfa 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -249,7 +249,7 @@ static int dev_eth_ioctl(struct net_device *dev, int err; err = dsa_ndo_eth_ioctl(dev, ifr, cmd); - if (err == 0 || err != -EOPNOTSUPP) + if (err != -EOPNOTSUPP) return err; if (ops->ndo_eth_ioctl) { -- cgit v1.2.3 From 4ee58e1e56800b589afe31c34547e2bc0c59f586 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 2 Apr 2023 15:37:51 +0300 Subject: net: promote SIOCSHWTSTAMP and SIOCGHWTSTAMP ioctls to dedicated handlers DSA does not want to intercept all ioctls handled by dev_eth_ioctl(), only SIOCSHWTSTAMP. This can be seen from commit f685e609a301 ("net: dsa: Deny PTP on master if switch supports it"). However, the way in which the dsa_ndo_eth_ioctl() is called would suggest otherwise. Split the handling of SIOCSHWTSTAMP and SIOCGHWTSTAMP ioctls into separate case statements of dev_ifsioc(), and make each one call its own sub-function. This also removes the dsa_ndo_eth_ioctl() call from dev_eth_ioctl(), which from now on exclusively handles PHY ioctls. Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/core/dev_ioctl.c | 39 ++++++++++++++++++++++++++------------- 1 file changed, 26 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index b299fb23fcfa..3b1402f6897c 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -246,20 +246,34 @@ static int dev_eth_ioctl(struct net_device *dev, struct ifreq *ifr, unsigned int cmd) { const struct net_device_ops *ops = dev->netdev_ops; + + if (!ops->ndo_eth_ioctl) + return -EOPNOTSUPP; + + if (!netif_device_present(dev)) + return -ENODEV; + + return ops->ndo_eth_ioctl(dev, ifr, cmd); +} + +static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr) +{ + return dev_eth_ioctl(dev, ifr, SIOCGHWTSTAMP); +} + +static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr) +{ int err; - err = dsa_ndo_eth_ioctl(dev, ifr, cmd); - if (err != -EOPNOTSUPP) + err = net_hwtstamp_validate(ifr); + if (err) return err; - if (ops->ndo_eth_ioctl) { - if (netif_device_present(dev)) - err = ops->ndo_eth_ioctl(dev, ifr, cmd); - else - err = -ENODEV; - } + err = dsa_ndo_eth_ioctl(dev, ifr, SIOCSHWTSTAMP); + if (err != -EOPNOTSUPP) + return err; - return err; + return dev_eth_ioctl(dev, ifr, SIOCSHWTSTAMP); } static int dev_siocbond(struct net_device *dev, @@ -395,12 +409,11 @@ static int dev_ifsioc(struct net *net, struct ifreq *ifr, void __user *data, return dev_siocdevprivate(dev, ifr, data, cmd); case SIOCSHWTSTAMP: - err = net_hwtstamp_validate(ifr); - if (err) - return err; - fallthrough; + return dev_set_hwtstamp(dev, ifr); case SIOCGHWTSTAMP: + return dev_get_hwtstamp(dev, ifr); + case SIOCGMIIPHY: case SIOCGMIIREG: case SIOCSMIIREG: -- cgit v1.2.3 From d5d5fd8f2552f6b9c50d3937c77b6783f99fbe83 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 2 Apr 2023 15:37:52 +0300 Subject: net: move copy_from_user() out of net_hwtstamp_validate() The kernel will want to start using the more meaningful struct hwtstamp_config pointer in more places, so move the copy_from_user() at the beginning of dev_set_hwtstamp() in order to get to that, and pass this argument to net_hwtstamp_validate(). Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/core/dev_ioctl.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 3b1402f6897c..34a0da5fbcfc 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -183,22 +183,18 @@ static int dev_ifsioc_locked(struct net *net, struct ifreq *ifr, unsigned int cm return err; } -static int net_hwtstamp_validate(struct ifreq *ifr) +static int net_hwtstamp_validate(const struct hwtstamp_config *cfg) { - struct hwtstamp_config cfg; enum hwtstamp_tx_types tx_type; enum hwtstamp_rx_filters rx_filter; int tx_type_valid = 0; int rx_filter_valid = 0; - if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) - return -EFAULT; - - if (cfg.flags & ~HWTSTAMP_FLAG_MASK) + if (cfg->flags & ~HWTSTAMP_FLAG_MASK) return -EINVAL; - tx_type = cfg.tx_type; - rx_filter = cfg.rx_filter; + tx_type = cfg->tx_type; + rx_filter = cfg->rx_filter; switch (tx_type) { case HWTSTAMP_TX_OFF: @@ -263,9 +259,13 @@ static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr) static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr) { + struct hwtstamp_config cfg; int err; - err = net_hwtstamp_validate(ifr); + if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) + return -EFAULT; + + err = net_hwtstamp_validate(&cfg); if (err) return err; -- cgit v1.2.3 From c4bffeaa8d50b7279c5a76597efa4b06e709df63 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 2 Apr 2023 15:37:53 +0300 Subject: net: add struct kernel_hwtstamp_config and make net_hwtstamp_validate() use it Jakub Kicinski suggested that we may want to add new UAPI for controlling hardware timestamping through netlink in the future, and in that case, we will be limited to the struct hwtstamp_config that is currently passed in fixed binary format through the SIOCGHWTSTAMP and SIOCSHWTSTAMP ioctls. It would be good if new kernel code already started operating on an extensible kernel variant of that structure, similar in concept to struct kernel_ethtool_coalesce vs struct ethtool_coalesce. Since struct hwtstamp_config is in include/uapi/linux/net_tstamp.h, here we introduce include/linux/net_tstamp.h which shadows that other header, but also includes it, so that existing includers of this header work as before. In addition to that, we add the definition for the kernel-only structure, and a helper which translates all fields by manual copying. I am doing a manual copy in order to not force the alignment (or type) of the fields of struct kernel_hwtstamp_config to be the same as of struct hwtstamp_config, even though now, they are the same. Link: https://lore.kernel.org/netdev/20230330223519.36ce7d23@kernel.org/ Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/linux/net_tstamp.h | 33 +++++++++++++++++++++++++++++++++ net/core/dev_ioctl.c | 7 +++++-- 2 files changed, 38 insertions(+), 2 deletions(-) create mode 100644 include/linux/net_tstamp.h (limited to 'net') diff --git a/include/linux/net_tstamp.h b/include/linux/net_tstamp.h new file mode 100644 index 000000000000..fd67f3cc0c4b --- /dev/null +++ b/include/linux/net_tstamp.h @@ -0,0 +1,33 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _LINUX_NET_TIMESTAMPING_H_ +#define _LINUX_NET_TIMESTAMPING_H_ + +#include + +/** + * struct kernel_hwtstamp_config - Kernel copy of struct hwtstamp_config + * + * @flags: see struct hwtstamp_config + * @tx_type: see struct hwtstamp_config + * @rx_filter: see struct hwtstamp_config + * + * Prefer using this structure for in-kernel processing of hardware + * timestamping configuration, over the inextensible struct hwtstamp_config + * exposed to the %SIOCGHWTSTAMP and %SIOCSHWTSTAMP ioctl UAPI. + */ +struct kernel_hwtstamp_config { + int flags; + int tx_type; + int rx_filter; +}; + +static inline void hwtstamp_config_to_kernel(struct kernel_hwtstamp_config *kernel_cfg, + const struct hwtstamp_config *cfg) +{ + kernel_cfg->flags = cfg->flags; + kernel_cfg->tx_type = cfg->tx_type; + kernel_cfg->rx_filter = cfg->rx_filter; +} + +#endif /* _LINUX_NET_TIMESTAMPING_H_ */ diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 34a0da5fbcfc..c532ef4d5dff 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -183,7 +183,7 @@ static int dev_ifsioc_locked(struct net *net, struct ifreq *ifr, unsigned int cm return err; } -static int net_hwtstamp_validate(const struct hwtstamp_config *cfg) +static int net_hwtstamp_validate(const struct kernel_hwtstamp_config *cfg) { enum hwtstamp_tx_types tx_type; enum hwtstamp_rx_filters rx_filter; @@ -259,13 +259,16 @@ static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr) static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr) { + struct kernel_hwtstamp_config kernel_cfg; struct hwtstamp_config cfg; int err; if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) return -EFAULT; - err = net_hwtstamp_validate(&cfg); + hwtstamp_config_to_kernel(&kernel_cfg, &cfg); + + err = net_hwtstamp_validate(&kernel_cfg); if (err) return err; -- cgit v1.2.3 From ff6ac4d013e680a5e7a38ee83ca59ffe1846915d Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 2 Apr 2023 15:37:54 +0300 Subject: net: dsa: make dsa_port_supports_hwtstamp() construct a fake ifreq dsa_master_ioctl() is in the process of getting converted to a different API, where we won't have access to a struct ifreq * anymore, but rather, to a struct kernel_hwtstamp_config. Since ds->ops->port_hwtstamp_get() still uses struct ifreq *, this creates a difficult situation where we have to make up such a dummy pointer. The conversion is a bit messy, because it forces a "good" implementation of ds->ops->port_hwtstamp_get() to return -EFAULT in copy_to_user() because of the NULL ifr->ifr_data pointer. However, it works, and it is only a transient step until ds->ops->port_hwtstamp_get() gets converted to the new API which passes struct kernel_hwtstamp_config and does not call copy_to_user(). Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/master.c | 2 +- net/dsa/port.c | 10 ++++++---- net/dsa/port.h | 2 +- 3 files changed, 8 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/dsa/master.c b/net/dsa/master.c index 22d3f16b0e6d..e397641382ca 100644 --- a/net/dsa/master.c +++ b/net/dsa/master.c @@ -212,7 +212,7 @@ static int dsa_master_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) * switch in the tree that is PTP capable. */ list_for_each_entry(dp, &dst->ports, list) - if (dsa_port_supports_hwtstamp(dp, ifr)) + if (dsa_port_supports_hwtstamp(dp)) return -EBUSY; break; } diff --git a/net/dsa/port.c b/net/dsa/port.c index 15cee17769e9..71ba30538411 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -114,19 +114,21 @@ static bool dsa_port_can_configure_learning(struct dsa_port *dp) return !err; } -bool dsa_port_supports_hwtstamp(struct dsa_port *dp, struct ifreq *ifr) +bool dsa_port_supports_hwtstamp(struct dsa_port *dp) { struct dsa_switch *ds = dp->ds; + struct ifreq ifr = {}; int err; if (!ds->ops->port_hwtstamp_get || !ds->ops->port_hwtstamp_set) return false; /* "See through" shim implementations of the "get" method. - * This will clobber the ifreq structure, but we will either return an - * error, or the master will overwrite it with proper values. + * Since we can't cook up a complete ioctl request structure, this will + * fail in copy_to_user() with -EFAULT, which hopefully is enough to + * detect a valid implementation. */ - err = ds->ops->port_hwtstamp_get(ds, dp->index, ifr); + err = ds->ops->port_hwtstamp_get(ds, dp->index, &ifr); return err != -EOPNOTSUPP; } diff --git a/net/dsa/port.h b/net/dsa/port.h index 9c218660d223..dc812512fd0e 100644 --- a/net/dsa/port.h +++ b/net/dsa/port.h @@ -15,7 +15,7 @@ struct switchdev_obj_port_mdb; struct switchdev_vlan_msti; struct phy_device; -bool dsa_port_supports_hwtstamp(struct dsa_port *dp, struct ifreq *ifr); +bool dsa_port_supports_hwtstamp(struct dsa_port *dp); void dsa_port_set_tag_protocol(struct dsa_port *cpu_dp, const struct dsa_device_ops *tag_ops); int dsa_port_set_state(struct dsa_port *dp, u8 state, bool do_fast_age); -- cgit v1.2.3 From 88c0a6b503b7f4fffb68a8d49c3987870c5b1d6b Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Sun, 2 Apr 2023 15:37:55 +0300 Subject: net: create a netdev notifier for DSA to reject PTP on DSA master The fact that PTP 2-step TX timestamping is broken on DSA switches if the master also timestamps the same packets is documented by commit f685e609a301 ("net: dsa: Deny PTP on master if switch supports it"). We attempt to help the users avoid shooting themselves in the foot by making DSA reject the timestamping ioctls on an interface that is a DSA master, and the switch tree beneath it contains switches which are aware of PTP. The only problem is that there isn't an established way of intercepting ndo_eth_ioctl calls, so DSA creates avoidable burden upon the network stack by creating a struct dsa_netdevice_ops with overlaid function pointers that are manually checked from the relevant call sites. There used to be 2 such dsa_netdevice_ops, but now, ndo_eth_ioctl is the only one left. There is an ongoing effort to migrate driver-visible hardware timestamping control from the ndo_eth_ioctl() based API to a new ndo_hwtstamp_set() model, but DSA actively prevents that migration, since dsa_master_ioctl() is currently coded to manually call the master's legacy ndo_eth_ioctl(), and so, whenever a network device driver would be converted to the new API, DSA's restrictions would be circumvented, because any device could be used as a DSA master. The established way for unrelated modules to react on a net device event is via netdevice notifiers. So we create a new notifier which gets called whenever there is an attempt to change hardware timestamping settings on a device. Finally, there is another reason why a netdev notifier will be a good idea, besides strictly DSA, and this has to do with PHY timestamping. With ndo_eth_ioctl(), all MAC drivers must manually call phy_has_hwtstamp() before deciding whether to act upon SIOCSHWTSTAMP, otherwise they must pass this ioctl to the PHY driver via phy_mii_ioctl(). With the new ndo_hwtstamp_set() API, it will be desirable to simply not make any calls into the MAC device driver when timestamping should be performed at the PHY level. But there exist drivers, such as the lan966x switch, which need to install packet traps for PTP regardless of whether they are the layer that provides the hardware timestamps, or the PHY is. That would be impossible to support with the new API. The proposal there, too, is to introduce a netdev notifier which acts as a better cue for switching drivers to add or remove PTP packet traps, than ndo_hwtstamp_set(). The one introduced here "almost" works there as well, except for the fact that packet traps should only be installed if the PHY driver succeeded to enable hardware timestamping, whereas here, we need to deny hardware timestamping on the DSA master before it actually gets enabled. This is why this notifier is called "PRE_", and the notifier that would get used for PHY timestamping and packet traps would be called NETDEV_CHANGE_HWTSTAMP. This isn't a new concept, for example NETDEV_CHANGEUPPER and NETDEV_PRECHANGEUPPER do the same thing. In expectation of future netlink UAPI, we also pass a non-NULL extack pointer to the netdev notifier, and we make DSA populate it with an informative reason for the rejection. To avoid making it go to waste, we make the ioctl-based dev_set_hwtstamp() create a fake extack and print the message to the kernel log. Link: https://lore.kernel.org/netdev/20230401191215.tvveoi3lkawgg6g4@skbuf/ Link: https://lore.kernel.org/netdev/20230310164451.ls7bbs6pdzs4m6pw@skbuf/ Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/linux/netdevice.h | 9 ++++++++- include/net/dsa.h | 51 ----------------------------------------------- net/core/dev.c | 8 +++----- net/core/dev_ioctl.c | 16 +++++++++++++-- net/dsa/master.c | 50 ++++++++++++++-------------------------------- net/dsa/master.h | 3 +++ net/dsa/slave.c | 11 ++++++++++ 7 files changed, 54 insertions(+), 94 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 62e093a6d6d1..a740be3bb911 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2878,6 +2878,7 @@ enum netdev_cmd { NETDEV_OFFLOAD_XSTATS_REPORT_USED, NETDEV_OFFLOAD_XSTATS_REPORT_DELTA, NETDEV_XDP_FEAT_CHANGE, + NETDEV_PRE_CHANGE_HWTSTAMP, }; const char *netdev_cmd_to_name(enum netdev_cmd cmd); @@ -2928,6 +2929,11 @@ struct netdev_notifier_pre_changeaddr_info { const unsigned char *dev_addr; }; +struct netdev_notifier_hwtstamp_info { + struct netdev_notifier_info info; /* must be first */ + struct kernel_hwtstamp_config *config; +}; + enum netdev_offload_xstats_type { NETDEV_OFFLOAD_XSTATS_TYPE_L3 = 1, }; @@ -2984,7 +2990,8 @@ netdev_notifier_info_to_extack(const struct netdev_notifier_info *info) } int call_netdevice_notifiers(unsigned long val, struct net_device *dev); - +int call_netdevice_notifiers_info(unsigned long val, + struct netdev_notifier_info *info); extern rwlock_t dev_base_lock; /* Device list lock */ diff --git a/include/net/dsa.h b/include/net/dsa.h index a15f17a38eca..8903053fa5aa 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -109,16 +109,6 @@ struct dsa_device_ops { bool promisc_on_master; }; -/* This structure defines the control interfaces that are overlayed by the - * DSA layer on top of the DSA CPU/management net_device instance. This is - * used by the core net_device layer while calling various net_device_ops - * function pointers. - */ -struct dsa_netdevice_ops { - int (*ndo_eth_ioctl)(struct net_device *dev, struct ifreq *ifr, - int cmd); -}; - struct dsa_lag { struct net_device *dev; unsigned int id; @@ -317,11 +307,6 @@ struct dsa_port { */ const struct ethtool_ops *orig_ethtool_ops; - /* - * Original copy of the master netdev net_device_ops - */ - const struct dsa_netdevice_ops *netdev_ops; - /* List of MAC addresses that must be forwarded on this port. * These are only valid on CPU ports and DSA links. */ @@ -1339,42 +1324,6 @@ static inline void dsa_tag_generic_flow_dissect(const struct sk_buff *skb, #endif } -#if IS_ENABLED(CONFIG_NET_DSA) -static inline int __dsa_netdevice_ops_check(struct net_device *dev) -{ - int err = -EOPNOTSUPP; - - if (!dev->dsa_ptr) - return err; - - if (!dev->dsa_ptr->netdev_ops) - return err; - - return 0; -} - -static inline int dsa_ndo_eth_ioctl(struct net_device *dev, struct ifreq *ifr, - int cmd) -{ - const struct dsa_netdevice_ops *ops; - int err; - - err = __dsa_netdevice_ops_check(dev); - if (err) - return err; - - ops = dev->dsa_ptr->netdev_ops; - - return ops->ndo_eth_ioctl(dev, ifr, cmd); -} -#else -static inline int dsa_ndo_eth_ioctl(struct net_device *dev, struct ifreq *ifr, - int cmd) -{ - return -EOPNOTSUPP; -} -#endif - void dsa_unregister_switch(struct dsa_switch *ds); int dsa_register_switch(struct dsa_switch *ds); void dsa_switch_shutdown(struct dsa_switch *ds); diff --git a/net/core/dev.c b/net/core/dev.c index 0c4b21291348..7ce5985be84b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -160,8 +160,6 @@ struct list_head ptype_base[PTYPE_HASH_SIZE] __read_mostly; struct list_head ptype_all __read_mostly; /* Taps */ static int netif_rx_internal(struct sk_buff *skb); -static int call_netdevice_notifiers_info(unsigned long val, - struct netdev_notifier_info *info); static int call_netdevice_notifiers_extack(unsigned long val, struct net_device *dev, struct netlink_ext_ack *extack); @@ -1614,7 +1612,7 @@ const char *netdev_cmd_to_name(enum netdev_cmd cmd) N(SVLAN_FILTER_PUSH_INFO) N(SVLAN_FILTER_DROP_INFO) N(PRE_CHANGEADDR) N(OFFLOAD_XSTATS_ENABLE) N(OFFLOAD_XSTATS_DISABLE) N(OFFLOAD_XSTATS_REPORT_USED) N(OFFLOAD_XSTATS_REPORT_DELTA) - N(XDP_FEAT_CHANGE) + N(XDP_FEAT_CHANGE) N(PRE_CHANGE_HWTSTAMP) } #undef N return "UNKNOWN_NETDEV_EVENT"; @@ -1919,8 +1917,8 @@ static void move_netdevice_notifiers_dev_net(struct net_device *dev, * are as for raw_notifier_call_chain(). */ -static int call_netdevice_notifiers_info(unsigned long val, - struct netdev_notifier_info *info) +int call_netdevice_notifiers_info(unsigned long val, + struct netdev_notifier_info *info) { struct net *net = dev_net(info->dev); int ret; diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index c532ef4d5dff..6d772837eb3f 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -259,7 +259,11 @@ static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr) static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr) { + struct netdev_notifier_hwtstamp_info info = { + .info.dev = dev, + }; struct kernel_hwtstamp_config kernel_cfg; + struct netlink_ext_ack extack = {}; struct hwtstamp_config cfg; int err; @@ -272,9 +276,17 @@ static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr) if (err) return err; - err = dsa_ndo_eth_ioctl(dev, ifr, SIOCSHWTSTAMP); - if (err != -EOPNOTSUPP) + info.info.extack = &extack; + info.config = &kernel_cfg; + + err = call_netdevice_notifiers_info(NETDEV_PRE_CHANGE_HWTSTAMP, + &info.info); + err = notifier_to_errno(err); + if (err) { + if (extack._msg) + netdev_err(dev, "%s\n", extack._msg); return err; + } return dev_eth_ioctl(dev, ifr, SIOCSHWTSTAMP); } diff --git a/net/dsa/master.c b/net/dsa/master.c index e397641382ca..c2cabe6248b1 100644 --- a/net/dsa/master.c +++ b/net/dsa/master.c @@ -195,38 +195,31 @@ static void dsa_master_get_strings(struct net_device *dev, uint32_t stringset, } } -static int dsa_master_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +/* Deny PTP operations on master if there is at least one switch in the tree + * that is PTP capable. + */ +int dsa_master_pre_change_hwtstamp(struct net_device *dev, + const struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) { struct dsa_port *cpu_dp = dev->dsa_ptr; struct dsa_switch *ds = cpu_dp->ds; struct dsa_switch_tree *dst; - int err = -EOPNOTSUPP; struct dsa_port *dp; dst = ds->dst; - switch (cmd) { - case SIOCGHWTSTAMP: - case SIOCSHWTSTAMP: - /* Deny PTP operations on master if there is at least one - * switch in the tree that is PTP capable. - */ - list_for_each_entry(dp, &dst->ports, list) - if (dsa_port_supports_hwtstamp(dp)) - return -EBUSY; - break; + list_for_each_entry(dp, &dst->ports, list) { + if (dsa_port_supports_hwtstamp(dp)) { + NL_SET_ERR_MSG(extack, + "HW timestamping not allowed on DSA master when switch supports the operation"); + return -EBUSY; + } } - if (dev->netdev_ops->ndo_eth_ioctl) - err = dev->netdev_ops->ndo_eth_ioctl(dev, ifr, cmd); - - return err; + return 0; } -static const struct dsa_netdevice_ops dsa_netdev_ops = { - .ndo_eth_ioctl = dsa_master_ioctl, -}; - static int dsa_master_ethtool_setup(struct net_device *dev) { struct dsa_port *cpu_dp = dev->dsa_ptr; @@ -267,15 +260,6 @@ static void dsa_master_ethtool_teardown(struct net_device *dev) cpu_dp->orig_ethtool_ops = NULL; } -static void dsa_netdev_ops_set(struct net_device *dev, - const struct dsa_netdevice_ops *ops) -{ - if (netif_is_lag_master(dev)) - return; - - dev->dsa_ptr->netdev_ops = ops; -} - /* Keep the master always promiscuous if the tagging protocol requires that * (garbles MAC DA) or if it doesn't support unicast filtering, case in which * it would revert to promiscuous mode as soon as we call dev_uc_add() on it @@ -414,16 +398,13 @@ int dsa_master_setup(struct net_device *dev, struct dsa_port *cpu_dp) if (ret) goto out_err_reset_promisc; - dsa_netdev_ops_set(dev, &dsa_netdev_ops); - ret = sysfs_create_group(&dev->dev.kobj, &dsa_group); if (ret) - goto out_err_ndo_teardown; + goto out_err_ethtool_teardown; return ret; -out_err_ndo_teardown: - dsa_netdev_ops_set(dev, NULL); +out_err_ethtool_teardown: dsa_master_ethtool_teardown(dev); out_err_reset_promisc: dsa_master_set_promiscuity(dev, -1); @@ -433,7 +414,6 @@ out_err_reset_promisc: void dsa_master_teardown(struct net_device *dev) { sysfs_remove_group(&dev->dev.kobj, &dsa_group); - dsa_netdev_ops_set(dev, NULL); dsa_master_ethtool_teardown(dev); dsa_master_reset_mtu(dev); dsa_master_set_promiscuity(dev, -1); diff --git a/net/dsa/master.h b/net/dsa/master.h index 3fc0e610b5b5..80842f4e27f7 100644 --- a/net/dsa/master.h +++ b/net/dsa/master.h @@ -15,5 +15,8 @@ int dsa_master_lag_setup(struct net_device *lag_dev, struct dsa_port *cpu_dp, struct netlink_ext_ack *extack); void dsa_master_lag_teardown(struct net_device *lag_dev, struct dsa_port *cpu_dp); +int dsa_master_pre_change_hwtstamp(struct net_device *dev, + const struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); #endif diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 165bb2cb8431..8abc1658ac47 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -3289,6 +3289,7 @@ static int dsa_master_changeupper(struct net_device *dev, static int dsa_slave_netdevice_event(struct notifier_block *nb, unsigned long event, void *ptr) { + struct netlink_ext_ack *extack = netdev_notifier_info_to_extack(ptr); struct net_device *dev = netdev_notifier_info_to_dev(ptr); switch (event) { @@ -3418,6 +3419,16 @@ static int dsa_slave_netdevice_event(struct notifier_block *nb, return NOTIFY_OK; } + case NETDEV_PRE_CHANGE_HWTSTAMP: { + struct netdev_notifier_hwtstamp_info *info = ptr; + int err; + + if (!netdev_uses_dsa(dev)) + return NOTIFY_DONE; + + err = dsa_master_pre_change_hwtstamp(dev, info->config, extack); + return notifier_from_errno(err); + } default: break; } -- cgit v1.2.3 From 839349d13905927d8a567ca4d21d88c82028e31d Mon Sep 17 00:00:00 2001 From: Sricharan Ramabadhran Date: Mon, 3 Apr 2023 12:28:51 +0530 Subject: net: qrtr: Do not do DEL_SERVER broadcast after DEL_CLIENT On the remote side, when QRTR socket is removed, af_qrtr will call qrtr_port_remove() which broadcasts the DEL_CLIENT packet to all neighbours including local NS. NS upon receiving the DEL_CLIENT packet, will remove the lookups associated with the node:port and broadcasts the DEL_SERVER packet. But on the host side, due to the arrival of the DEL_CLIENT packet, the NS would've already deleted the server belonging to that port. So when the remote's NS again broadcasts the DEL_SERVER for that port, it throws below error message on the host: "failed while handling packet from 2:-2" So fix this error by not broadcasting the DEL_SERVER packet when the DEL_CLIENT packet gets processed." Fixes: 0c2204a4ad71 ("net: qrtr: Migrate nameservice to kernel from userspace") Reviewed-by: Manivannan Sadhasivam Signed-off-by: Ram Kumar Dharuman Signed-off-by: Sricharan Ramabadhran Signed-off-by: David S. Miller --- net/qrtr/ns.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/qrtr/ns.c b/net/qrtr/ns.c index 722936f7dd98..0f25a386138c 100644 --- a/net/qrtr/ns.c +++ b/net/qrtr/ns.c @@ -274,7 +274,7 @@ err: return NULL; } -static int server_del(struct qrtr_node *node, unsigned int port) +static int server_del(struct qrtr_node *node, unsigned int port, bool bcast) { struct qrtr_lookup *lookup; struct qrtr_server *srv; @@ -287,7 +287,7 @@ static int server_del(struct qrtr_node *node, unsigned int port) radix_tree_delete(&node->servers, port); /* Broadcast the removal of local servers */ - if (srv->node == qrtr_ns.local_node) + if (srv->node == qrtr_ns.local_node && bcast) service_announce_del(&qrtr_ns.bcast_sq, srv); /* Announce the service's disappearance to observers */ @@ -373,7 +373,7 @@ static int ctrl_cmd_bye(struct sockaddr_qrtr *from) } slot = radix_tree_iter_resume(slot, &iter); rcu_read_unlock(); - server_del(node, srv->port); + server_del(node, srv->port, true); rcu_read_lock(); } rcu_read_unlock(); @@ -459,10 +459,13 @@ static int ctrl_cmd_del_client(struct sockaddr_qrtr *from, kfree(lookup); } - /* Remove the server belonging to this port */ + /* Remove the server belonging to this port but don't broadcast + * DEL_SERVER. Neighbours would've already removed the server belonging + * to this port due to the DEL_CLIENT broadcast from qrtr_port_remove(). + */ node = node_get(node_id); if (node) - server_del(node, port); + server_del(node, port, false); /* Advertise the removal of this client to all local servers */ local_node = node_get(qrtr_ns.local_node); @@ -567,7 +570,7 @@ static int ctrl_cmd_del_server(struct sockaddr_qrtr *from, if (!node) return -ENOENT; - return server_del(node, port); + return server_del(node, port, true); } static int ctrl_cmd_new_lookup(struct sockaddr_qrtr *from, -- cgit v1.2.3 From ea30388baebcce37fd594d425a65037ca35e59e8 Mon Sep 17 00:00:00 2001 From: Ziyang Xuan Date: Mon, 3 Apr 2023 15:34:17 +0800 Subject: ipv6: Fix an uninit variable access bug in __ip6_make_skb() Syzbot reported a bug as following: ===================================================== BUG: KMSAN: uninit-value in arch_atomic64_inc arch/x86/include/asm/atomic64_64.h:88 [inline] BUG: KMSAN: uninit-value in arch_atomic_long_inc include/linux/atomic/atomic-long.h:161 [inline] BUG: KMSAN: uninit-value in atomic_long_inc include/linux/atomic/atomic-instrumented.h:1429 [inline] BUG: KMSAN: uninit-value in __ip6_make_skb+0x2f37/0x30f0 net/ipv6/ip6_output.c:1956 arch_atomic64_inc arch/x86/include/asm/atomic64_64.h:88 [inline] arch_atomic_long_inc include/linux/atomic/atomic-long.h:161 [inline] atomic_long_inc include/linux/atomic/atomic-instrumented.h:1429 [inline] __ip6_make_skb+0x2f37/0x30f0 net/ipv6/ip6_output.c:1956 ip6_finish_skb include/net/ipv6.h:1122 [inline] ip6_push_pending_frames+0x10e/0x550 net/ipv6/ip6_output.c:1987 rawv6_push_pending_frames+0xb12/0xb90 net/ipv6/raw.c:579 rawv6_sendmsg+0x297e/0x2e60 net/ipv6/raw.c:922 inet_sendmsg+0x101/0x180 net/ipv4/af_inet.c:827 sock_sendmsg_nosec net/socket.c:714 [inline] sock_sendmsg net/socket.c:734 [inline] ____sys_sendmsg+0xa8e/0xe70 net/socket.c:2476 ___sys_sendmsg+0x2a1/0x3f0 net/socket.c:2530 __sys_sendmsg net/socket.c:2559 [inline] __do_sys_sendmsg net/socket.c:2568 [inline] __se_sys_sendmsg net/socket.c:2566 [inline] __x64_sys_sendmsg+0x367/0x540 net/socket.c:2566 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Uninit was created at: slab_post_alloc_hook mm/slab.h:766 [inline] slab_alloc_node mm/slub.c:3452 [inline] __kmem_cache_alloc_node+0x71f/0xce0 mm/slub.c:3491 __do_kmalloc_node mm/slab_common.c:967 [inline] __kmalloc_node_track_caller+0x114/0x3b0 mm/slab_common.c:988 kmalloc_reserve net/core/skbuff.c:492 [inline] __alloc_skb+0x3af/0x8f0 net/core/skbuff.c:565 alloc_skb include/linux/skbuff.h:1270 [inline] __ip6_append_data+0x51c1/0x6bb0 net/ipv6/ip6_output.c:1684 ip6_append_data+0x411/0x580 net/ipv6/ip6_output.c:1854 rawv6_sendmsg+0x2882/0x2e60 net/ipv6/raw.c:915 inet_sendmsg+0x101/0x180 net/ipv4/af_inet.c:827 sock_sendmsg_nosec net/socket.c:714 [inline] sock_sendmsg net/socket.c:734 [inline] ____sys_sendmsg+0xa8e/0xe70 net/socket.c:2476 ___sys_sendmsg+0x2a1/0x3f0 net/socket.c:2530 __sys_sendmsg net/socket.c:2559 [inline] __do_sys_sendmsg net/socket.c:2568 [inline] __se_sys_sendmsg net/socket.c:2566 [inline] __x64_sys_sendmsg+0x367/0x540 net/socket.c:2566 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd It is because icmp6hdr does not in skb linear region under the scenario of SOCK_RAW socket. Access icmp6_hdr(skb)->icmp6_type directly will trigger the uninit variable access bug. Use a local variable icmp6_type to carry the correct value in different scenarios. Fixes: 14878f75abd5 ("[IPV6]: Add ICMPMsgStats MIB (RFC 4293) [rev 2]") Reported-by: syzbot+8257f4dcef79de670baf@syzkaller.appspotmail.com Link: https://syzkaller.appspot.com/bug?id=3d605ec1d0a7f2a269a1a6936ac7f2b85975ee9c Signed-off-by: Ziyang Xuan Signed-off-by: David S. Miller --- net/ipv6/ip6_output.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index c314fdde0097..95a55c6630ad 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1965,8 +1965,13 @@ struct sk_buff *__ip6_make_skb(struct sock *sk, IP6_UPD_PO_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len); if (proto == IPPROTO_ICMPV6) { struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); + u8 icmp6_type; - ICMP6MSGOUT_INC_STATS(net, idev, icmp6_hdr(skb)->icmp6_type); + if (sk->sk_socket->type == SOCK_RAW && !inet_sk(sk)->hdrincl) + icmp6_type = fl6->fl6_icmp_type; + else + icmp6_type = icmp6_hdr(skb)->icmp6_type; + ICMP6MSGOUT_INC_STATS(net, idev, icmp6_type); ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTMSGS); } -- cgit v1.2.3 From f59f3006ca7b2fc9bf38fbf44e3dc4b1fbd686c1 Mon Sep 17 00:00:00 2001 From: Arseniy Krasnov Date: Mon, 3 Apr 2023 14:24:17 +0300 Subject: vsock/vmci: convert VMCI error code to -ENOMEM on receive This adds conversion of VMCI specific error code to general -ENOMEM. It is preparation for the next patch, which changes af_vsock.c behaviour on receive to pass value returned from transport to the user. Signed-off-by: Arseniy Krasnov Reviewed-by: Vishnu Dasa Reviewed-by: Stefano Garzarella Signed-off-by: Paolo Abeni --- net/vmw_vsock/vmci_transport.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c index 36eb16a40745..a5375c97f5b0 100644 --- a/net/vmw_vsock/vmci_transport.c +++ b/net/vmw_vsock/vmci_transport.c @@ -1831,10 +1831,17 @@ static ssize_t vmci_transport_stream_dequeue( size_t len, int flags) { + ssize_t err; + if (flags & MSG_PEEK) - return vmci_qpair_peekv(vmci_trans(vsk)->qpair, msg, len, 0); + err = vmci_qpair_peekv(vmci_trans(vsk)->qpair, msg, len, 0); else - return vmci_qpair_dequev(vmci_trans(vsk)->qpair, msg, len, 0); + err = vmci_qpair_dequev(vmci_trans(vsk)->qpair, msg, len, 0); + + if (err < 0) + err = -ENOMEM; + + return err; } static ssize_t vmci_transport_stream_enqueue( -- cgit v1.2.3 From 02ab696febab622c4c837d509f6825884c952c22 Mon Sep 17 00:00:00 2001 From: Arseniy Krasnov Date: Mon, 3 Apr 2023 14:25:12 +0300 Subject: vsock: return errors other than -ENOMEM to socket This removes behaviour, where error code returned from any transport was always switched to ENOMEM. This works in the same way as: commit c43170b7e157 ("vsock: return errors other than -ENOMEM to socket"), but for receive calls. Signed-off-by: Arseniy Krasnov Reviewed-by: Stefano Garzarella Signed-off-by: Paolo Abeni --- net/vmw_vsock/af_vsock.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 5f2dda35c980..413407bb646c 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -2043,7 +2043,7 @@ static int __vsock_stream_recvmsg(struct sock *sk, struct msghdr *msg, read = transport->stream_dequeue(vsk, msg, len - copied, flags); if (read < 0) { - err = -ENOMEM; + err = read; break; } @@ -2094,7 +2094,7 @@ static int __vsock_seqpacket_recvmsg(struct sock *sk, struct msghdr *msg, msg_len = transport->seqpacket_dequeue(vsk, msg, flags); if (msg_len < 0) { - err = -ENOMEM; + err = msg_len; goto out; } -- cgit v1.2.3 From 5085e41f9e83a1bec51da1f20b54f2ec3a13a3fe Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Thu, 30 Mar 2023 14:24:27 -0400 Subject: sunrpc: only free unix grouplist after RCU settles While the unix_gid object is rcu-freed, the group_info list that it contains is not. Ensure that we only put the group list reference once we are really freeing the unix_gid object. Reported-by: Zhi Li Link: https://bugzilla.redhat.com/show_bug.cgi?id=2183056 Signed-off-by: Jeff Layton Fixes: fd5d2f78261b ("SUNRPC: Make server side AUTH_UNIX use lockless lookups") Signed-off-by: Chuck Lever --- net/sunrpc/svcauth_unix.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c index 983c5891cb56..4246363cb095 100644 --- a/net/sunrpc/svcauth_unix.c +++ b/net/sunrpc/svcauth_unix.c @@ -416,14 +416,23 @@ static int unix_gid_hash(kuid_t uid) return hash_long(from_kuid(&init_user_ns, uid), GID_HASHBITS); } -static void unix_gid_put(struct kref *kref) +static void unix_gid_free(struct rcu_head *rcu) { - struct cache_head *item = container_of(kref, struct cache_head, ref); - struct unix_gid *ug = container_of(item, struct unix_gid, h); + struct unix_gid *ug = container_of(rcu, struct unix_gid, rcu); + struct cache_head *item = &ug->h; + if (test_bit(CACHE_VALID, &item->flags) && !test_bit(CACHE_NEGATIVE, &item->flags)) put_group_info(ug->gi); - kfree_rcu(ug, rcu); + kfree(ug); +} + +static void unix_gid_put(struct kref *kref) +{ + struct cache_head *item = container_of(kref, struct cache_head, ref); + struct unix_gid *ug = container_of(item, struct unix_gid, h); + + call_rcu(&ug->rcu, unix_gid_free); } static int unix_gid_match(struct cache_head *corig, struct cache_head *cnew) -- cgit v1.2.3 From 7d64c513284408fee5178a0953a686e9410f2399 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Mon, 3 Apr 2023 21:50:22 -0700 Subject: bpf: Invoke btf_struct_access() callback only for writes. Remove duplicated if (atype == BPF_READ) btf_struct_access() from btf_struct_access() callback and invoke it only for writes. This is possible to do because currently btf_struct_access() custom callback always delegates to generic btf_struct_access() helper for BPF_READ accesses. Signed-off-by: Alexei Starovoitov Signed-off-by: Andrii Nakryiko Acked-by: David Vernet Link: https://lore.kernel.org/bpf/20230404045029.82870-2-alexei.starovoitov@gmail.com --- kernel/bpf/verifier.c | 2 +- net/bpf/bpf_dummy_struct_ops.c | 2 +- net/core/filter.c | 6 ------ net/ipv4/bpf_tcp_ca.c | 3 --- 4 files changed, 2 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index eaf9c5291cf0..83984568ccb4 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -5504,7 +5504,7 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env, return -EACCES; } - if (env->ops->btf_struct_access && !type_is_alloc(reg->type)) { + if (env->ops->btf_struct_access && !type_is_alloc(reg->type) && atype == BPF_WRITE) { if (!btf_is_kernel(reg->btf)) { verbose(env, "verifier internal error: reg->btf must be kernel btf\n"); return -EFAULT; diff --git a/net/bpf/bpf_dummy_struct_ops.c b/net/bpf/bpf_dummy_struct_ops.c index ff4f89a2b02a..9535c8506cda 100644 --- a/net/bpf/bpf_dummy_struct_ops.c +++ b/net/bpf/bpf_dummy_struct_ops.c @@ -198,7 +198,7 @@ static int bpf_dummy_ops_btf_struct_access(struct bpf_verifier_log *log, if (err < 0) return err; - return atype == BPF_READ ? err : NOT_INIT; + return NOT_INIT; } static const struct bpf_verifier_ops bpf_dummy_verifier_ops = { diff --git a/net/core/filter.c b/net/core/filter.c index 3370efad1dda..8b9f409a2ec3 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -8753,9 +8753,6 @@ static int tc_cls_act_btf_struct_access(struct bpf_verifier_log *log, { int ret = -EACCES; - if (atype == BPF_READ) - return btf_struct_access(log, reg, off, size, atype, next_btf_id, flag); - mutex_lock(&nf_conn_btf_access_lock); if (nfct_btf_struct_access) ret = nfct_btf_struct_access(log, reg, off, size, atype, next_btf_id, flag); @@ -8830,9 +8827,6 @@ static int xdp_btf_struct_access(struct bpf_verifier_log *log, { int ret = -EACCES; - if (atype == BPF_READ) - return btf_struct_access(log, reg, off, size, atype, next_btf_id, flag); - mutex_lock(&nf_conn_btf_access_lock); if (nfct_btf_struct_access) ret = nfct_btf_struct_access(log, reg, off, size, atype, next_btf_id, flag); diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index ea21c96c03aa..d6465876bbf6 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -78,9 +78,6 @@ static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log, const struct btf_type *t; size_t end; - if (atype == BPF_READ) - return btf_struct_access(log, reg, off, size, atype, next_btf_id, flag); - t = btf_type_by_id(reg->btf, reg->btf_id); if (t != tcp_sock_type) { bpf_log(log, "only read is supported\n"); -- cgit v1.2.3 From b7e852a9ec96635168c04204fb7cf1f7390b9a8c Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Mon, 3 Apr 2023 21:50:23 -0700 Subject: bpf: Remove unused arguments from btf_struct_access(). Remove unused arguments from btf_struct_access() callback. Signed-off-by: Alexei Starovoitov Signed-off-by: Andrii Nakryiko Acked-by: David Vernet Link: https://lore.kernel.org/bpf/20230404045029.82870-3-alexei.starovoitov@gmail.com --- include/linux/bpf.h | 3 +-- include/linux/filter.h | 3 +-- kernel/bpf/verifier.c | 4 ++-- net/bpf/bpf_dummy_struct_ops.c | 12 +++++------- net/core/filter.c | 13 +++++-------- net/ipv4/bpf_tcp_ca.c | 3 +-- net/netfilter/nf_conntrack_bpf.c | 3 +-- 7 files changed, 16 insertions(+), 25 deletions(-) (limited to 'net') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 2d8f3f639e68..4f689dda748f 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -893,8 +893,7 @@ struct bpf_verifier_ops { struct bpf_prog *prog, u32 *target_size); int (*btf_struct_access)(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off, int size, enum bpf_access_type atype, - u32 *next_btf_id, enum bpf_type_flag *flag); + int off, int size); }; struct bpf_prog_offload_ops { diff --git a/include/linux/filter.h b/include/linux/filter.h index 23c08c31bea9..5364b0c52c1d 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -571,8 +571,7 @@ DECLARE_STATIC_KEY_FALSE(bpf_stats_enabled_key); extern struct mutex nf_conn_btf_access_lock; extern int (*nfct_btf_struct_access)(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off, int size, enum bpf_access_type atype, - u32 *next_btf_id, enum bpf_type_flag *flag); + int off, int size); typedef unsigned int (*bpf_dispatcher_fn)(const void *ctx, const struct bpf_insn *insnsi, diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 83984568ccb4..5ca520e5eddf 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -5459,7 +5459,7 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env, const struct btf_type *t = btf_type_by_id(reg->btf, reg->btf_id); const char *tname = btf_name_by_offset(reg->btf, t->name_off); enum bpf_type_flag flag = 0; - u32 btf_id; + u32 btf_id = 0; int ret; if (!env->allow_ptr_leaks) { @@ -5509,7 +5509,7 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env, verbose(env, "verifier internal error: reg->btf must be kernel btf\n"); return -EFAULT; } - ret = env->ops->btf_struct_access(&env->log, reg, off, size, atype, &btf_id, &flag); + ret = env->ops->btf_struct_access(&env->log, reg, off, size); } else { /* Writes are permitted with default btf_struct_access for * program allocated objects (which always have ref_obj_id > 0), diff --git a/net/bpf/bpf_dummy_struct_ops.c b/net/bpf/bpf_dummy_struct_ops.c index 9535c8506cda..5918d1b32e19 100644 --- a/net/bpf/bpf_dummy_struct_ops.c +++ b/net/bpf/bpf_dummy_struct_ops.c @@ -173,14 +173,11 @@ static int bpf_dummy_ops_check_member(const struct btf_type *t, static int bpf_dummy_ops_btf_struct_access(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off, int size, enum bpf_access_type atype, - u32 *next_btf_id, - enum bpf_type_flag *flag) + int off, int size) { const struct btf_type *state; const struct btf_type *t; s32 type_id; - int err; type_id = btf_find_by_name_kind(reg->btf, "bpf_dummy_ops_state", BTF_KIND_STRUCT); @@ -194,9 +191,10 @@ static int bpf_dummy_ops_btf_struct_access(struct bpf_verifier_log *log, return -EACCES; } - err = btf_struct_access(log, reg, off, size, atype, next_btf_id, flag); - if (err < 0) - return err; + if (off + size > sizeof(struct bpf_dummy_ops_state)) { + bpf_log(log, "write access at off %d with size %d\n", off, size); + return -EACCES; + } return NOT_INIT; } diff --git a/net/core/filter.c b/net/core/filter.c index 8b9f409a2ec3..1f2abf0f60e6 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -8742,20 +8742,18 @@ EXPORT_SYMBOL_GPL(nf_conn_btf_access_lock); int (*nfct_btf_struct_access)(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off, int size, enum bpf_access_type atype, - u32 *next_btf_id, enum bpf_type_flag *flag); + int off, int size); EXPORT_SYMBOL_GPL(nfct_btf_struct_access); static int tc_cls_act_btf_struct_access(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off, int size, enum bpf_access_type atype, - u32 *next_btf_id, enum bpf_type_flag *flag) + int off, int size) { int ret = -EACCES; mutex_lock(&nf_conn_btf_access_lock); if (nfct_btf_struct_access) - ret = nfct_btf_struct_access(log, reg, off, size, atype, next_btf_id, flag); + ret = nfct_btf_struct_access(log, reg, off, size); mutex_unlock(&nf_conn_btf_access_lock); return ret; @@ -8822,14 +8820,13 @@ EXPORT_SYMBOL_GPL(bpf_warn_invalid_xdp_action); static int xdp_btf_struct_access(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off, int size, enum bpf_access_type atype, - u32 *next_btf_id, enum bpf_type_flag *flag) + int off, int size) { int ret = -EACCES; mutex_lock(&nf_conn_btf_access_lock); if (nfct_btf_struct_access) - ret = nfct_btf_struct_access(log, reg, off, size, atype, next_btf_id, flag); + ret = nfct_btf_struct_access(log, reg, off, size); mutex_unlock(&nf_conn_btf_access_lock); return ret; diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index d6465876bbf6..4406d796cc2f 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -72,8 +72,7 @@ static bool bpf_tcp_ca_is_valid_access(int off, int size, static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off, int size, enum bpf_access_type atype, - u32 *next_btf_id, enum bpf_type_flag *flag) + int off, int size) { const struct btf_type *t; size_t end; diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c index 002e9d24a1e9..3f821b7ba646 100644 --- a/net/netfilter/nf_conntrack_bpf.c +++ b/net/netfilter/nf_conntrack_bpf.c @@ -192,8 +192,7 @@ BTF_ID(struct, nf_conn___init) /* Check writes into `struct nf_conn` */ static int _nf_conntrack_btf_struct_access(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off, int size, enum bpf_access_type atype, - u32 *next_btf_id, enum bpf_type_flag *flag) + int off, int size) { const struct btf_type *ncit, *nct, *t; size_t end; -- cgit v1.2.3 From 91571a515d1bcdc280bb46423bb697ea7eb42ff3 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Mon, 3 Apr 2023 21:50:25 -0700 Subject: bpf: Teach verifier that certain helpers accept NULL pointer. bpf_[sk|inode|task|cgrp]_storage_[get|delete]() and bpf_get_socket_cookie() helpers perform run-time check that sk|inode|task|cgrp pointer != NULL. Teach verifier about this fact and allow bpf programs to pass PTR_TO_BTF_ID | PTR_MAYBE_NULL into such helpers. It will be used in the subsequent patch that will do bpf_sk_storage_get(.., skb->sk, ...); Even when 'skb' pointer is trusted the 'sk' pointer may be NULL. Signed-off-by: Alexei Starovoitov Signed-off-by: Andrii Nakryiko Acked-by: David Vernet Link: https://lore.kernel.org/bpf/20230404045029.82870-5-alexei.starovoitov@gmail.com --- kernel/bpf/bpf_cgrp_storage.c | 4 ++-- kernel/bpf/bpf_inode_storage.c | 4 ++-- kernel/bpf/bpf_task_storage.c | 8 ++++---- net/core/bpf_sk_storage.c | 4 ++-- net/core/filter.c | 2 +- 5 files changed, 11 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c index d17d5b694668..d44fe8dd9732 100644 --- a/kernel/bpf/bpf_cgrp_storage.c +++ b/kernel/bpf/bpf_cgrp_storage.c @@ -224,7 +224,7 @@ const struct bpf_func_proto bpf_cgrp_storage_get_proto = { .gpl_only = false, .ret_type = RET_PTR_TO_MAP_VALUE_OR_NULL, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &bpf_cgroup_btf_id[0], .arg3_type = ARG_PTR_TO_MAP_VALUE_OR_NULL, .arg4_type = ARG_ANYTHING, @@ -235,6 +235,6 @@ const struct bpf_func_proto bpf_cgrp_storage_delete_proto = { .gpl_only = false, .ret_type = RET_INTEGER, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &bpf_cgroup_btf_id[0], }; diff --git a/kernel/bpf/bpf_inode_storage.c b/kernel/bpf/bpf_inode_storage.c index e17ad581b9be..a4d93df78c75 100644 --- a/kernel/bpf/bpf_inode_storage.c +++ b/kernel/bpf/bpf_inode_storage.c @@ -229,7 +229,7 @@ const struct bpf_func_proto bpf_inode_storage_get_proto = { .gpl_only = false, .ret_type = RET_PTR_TO_MAP_VALUE_OR_NULL, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &bpf_inode_storage_btf_ids[0], .arg3_type = ARG_PTR_TO_MAP_VALUE_OR_NULL, .arg4_type = ARG_ANYTHING, @@ -240,6 +240,6 @@ const struct bpf_func_proto bpf_inode_storage_delete_proto = { .gpl_only = false, .ret_type = RET_INTEGER, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &bpf_inode_storage_btf_ids[0], }; diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c index d1af0c8f9ce4..adf6dfe0ba68 100644 --- a/kernel/bpf/bpf_task_storage.c +++ b/kernel/bpf/bpf_task_storage.c @@ -338,7 +338,7 @@ const struct bpf_func_proto bpf_task_storage_get_recur_proto = { .gpl_only = false, .ret_type = RET_PTR_TO_MAP_VALUE_OR_NULL, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &btf_tracing_ids[BTF_TRACING_TYPE_TASK], .arg3_type = ARG_PTR_TO_MAP_VALUE_OR_NULL, .arg4_type = ARG_ANYTHING, @@ -349,7 +349,7 @@ const struct bpf_func_proto bpf_task_storage_get_proto = { .gpl_only = false, .ret_type = RET_PTR_TO_MAP_VALUE_OR_NULL, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &btf_tracing_ids[BTF_TRACING_TYPE_TASK], .arg3_type = ARG_PTR_TO_MAP_VALUE_OR_NULL, .arg4_type = ARG_ANYTHING, @@ -360,7 +360,7 @@ const struct bpf_func_proto bpf_task_storage_delete_recur_proto = { .gpl_only = false, .ret_type = RET_INTEGER, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &btf_tracing_ids[BTF_TRACING_TYPE_TASK], }; @@ -369,6 +369,6 @@ const struct bpf_func_proto bpf_task_storage_delete_proto = { .gpl_only = false, .ret_type = RET_INTEGER, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &btf_tracing_ids[BTF_TRACING_TYPE_TASK], }; diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index 085025c7130a..d4172534dfa8 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -412,7 +412,7 @@ const struct bpf_func_proto bpf_sk_storage_get_tracing_proto = { .gpl_only = false, .ret_type = RET_PTR_TO_MAP_VALUE_OR_NULL, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &btf_sock_ids[BTF_SOCK_TYPE_SOCK_COMMON], .arg3_type = ARG_PTR_TO_MAP_VALUE_OR_NULL, .arg4_type = ARG_ANYTHING, @@ -424,7 +424,7 @@ const struct bpf_func_proto bpf_sk_storage_delete_tracing_proto = { .gpl_only = false, .ret_type = RET_INTEGER, .arg1_type = ARG_CONST_MAP_PTR, - .arg2_type = ARG_PTR_TO_BTF_ID, + .arg2_type = ARG_PTR_TO_BTF_ID_OR_NULL, .arg2_btf_id = &btf_sock_ids[BTF_SOCK_TYPE_SOCK_COMMON], .allowed = bpf_sk_storage_tracing_allowed, }; diff --git a/net/core/filter.c b/net/core/filter.c index 1f2abf0f60e6..727c5269867d 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -4998,7 +4998,7 @@ const struct bpf_func_proto bpf_get_socket_ptr_cookie_proto = { .func = bpf_get_socket_ptr_cookie, .gpl_only = false, .ret_type = RET_INTEGER, - .arg1_type = ARG_PTR_TO_BTF_ID_SOCK_COMMON, + .arg1_type = ARG_PTR_TO_BTF_ID_SOCK_COMMON | PTR_MAYBE_NULL, }; BPF_CALL_1(bpf_get_socket_cookie_sock_ops, struct bpf_sock_ops_kern *, ctx) -- cgit v1.2.3 From 0a78cf7264d29abeca098eae0b188a10aabc8a32 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 3 Apr 2023 12:49:58 -0700 Subject: raw: Fix NULL deref in raw_get_next(). Dae R. Jeong reported a NULL deref in raw_get_next() [0]. It seems that the repro was running these sequences in parallel so that one thread was iterating on a socket that was being freed in another netns. unshare(0x40060200) r0 = syz_open_procfs(0x0, &(0x7f0000002080)='net/raw\x00') socket$inet_icmp_raw(0x2, 0x3, 0x1) pread64(r0, &(0x7f0000000000)=""/10, 0xa, 0x10000000007f) After commit 0daf07e52709 ("raw: convert raw sockets to RCU"), we use RCU and hlist_nulls_for_each_entry() to iterate over SOCK_RAW sockets. However, we should use spinlock for slow paths to avoid the NULL deref. Also, SOCK_RAW does not use SLAB_TYPESAFE_BY_RCU, and the slab object is not reused during iteration in the grace period. In fact, the lockless readers do not check the nulls marker with get_nulls_value(). So, SOCK_RAW should use hlist instead of hlist_nulls. Instead of adding an unnecessary barrier by sk_nulls_for_each_rcu(), let's convert hlist_nulls to hlist and use sk_for_each_rcu() for fast paths and sk_for_each() and spinlock for /proc/net/raw. [0]: general protection fault, probably for non-canonical address 0xdffffc0000000005: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f] CPU: 2 PID: 20952 Comm: syz-executor.0 Not tainted 6.2.0-g048ec869bafd-dirty #7 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014 RIP: 0010:read_pnet include/net/net_namespace.h:383 [inline] RIP: 0010:sock_net include/net/sock.h:649 [inline] RIP: 0010:raw_get_next net/ipv4/raw.c:974 [inline] RIP: 0010:raw_get_idx net/ipv4/raw.c:986 [inline] RIP: 0010:raw_seq_start+0x431/0x800 net/ipv4/raw.c:995 Code: ef e8 33 3d 94 f7 49 8b 6d 00 4c 89 ef e8 b7 65 5f f7 49 89 ed 49 83 c5 98 0f 84 9a 00 00 00 48 83 c5 c8 48 89 e8 48 c1 e8 03 <42> 80 3c 30 00 74 08 48 89 ef e8 00 3d 94 f7 4c 8b 7d 00 48 89 ef RSP: 0018:ffffc9001154f9b0 EFLAGS: 00010206 RAX: 0000000000000005 RBX: 1ffff1100302c8fd RCX: 0000000000000000 RDX: 0000000000000028 RSI: ffffc9001154f988 RDI: ffffc9000f77a338 RBP: 0000000000000029 R08: ffffffff8a50ffb4 R09: fffffbfff24b6bd9 R10: fffffbfff24b6bd9 R11: 0000000000000000 R12: ffff88801db73b78 R13: fffffffffffffff9 R14: dffffc0000000000 R15: 0000000000000030 FS: 00007f843ae8e700(0000) GS:ffff888063700000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000055bb9614b35f CR3: 000000003c672000 CR4: 00000000003506e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: seq_read_iter+0x4c6/0x10f0 fs/seq_file.c:225 seq_read+0x224/0x320 fs/seq_file.c:162 pde_read fs/proc/inode.c:316 [inline] proc_reg_read+0x23f/0x330 fs/proc/inode.c:328 vfs_read+0x31e/0xd30 fs/read_write.c:468 ksys_pread64 fs/read_write.c:665 [inline] __do_sys_pread64 fs/read_write.c:675 [inline] __se_sys_pread64 fs/read_write.c:672 [inline] __x64_sys_pread64+0x1e9/0x280 fs/read_write.c:672 do_syscall_x64 arch/x86/entry/common.c:51 [inline] do_syscall_64+0x4e/0xa0 arch/x86/entry/common.c:82 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0033:0x478d29 Code: f7 d8 64 89 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f843ae8dbe8 EFLAGS: 00000246 ORIG_RAX: 0000000000000011 RAX: ffffffffffffffda RBX: 0000000000791408 RCX: 0000000000478d29 RDX: 000000000000000a RSI: 0000000020000000 RDI: 0000000000000003 RBP: 00000000f477909a R08: 0000000000000000 R09: 0000000000000000 R10: 000010000000007f R11: 0000000000000246 R12: 0000000000791740 R13: 0000000000791414 R14: 0000000000791408 R15: 00007ffc2eb48a50 Modules linked in: ---[ end trace 0000000000000000 ]--- RIP: 0010:read_pnet include/net/net_namespace.h:383 [inline] RIP: 0010:sock_net include/net/sock.h:649 [inline] RIP: 0010:raw_get_next net/ipv4/raw.c:974 [inline] RIP: 0010:raw_get_idx net/ipv4/raw.c:986 [inline] RIP: 0010:raw_seq_start+0x431/0x800 net/ipv4/raw.c:995 Code: ef e8 33 3d 94 f7 49 8b 6d 00 4c 89 ef e8 b7 65 5f f7 49 89 ed 49 83 c5 98 0f 84 9a 00 00 00 48 83 c5 c8 48 89 e8 48 c1 e8 03 <42> 80 3c 30 00 74 08 48 89 ef e8 00 3d 94 f7 4c 8b 7d 00 48 89 ef RSP: 0018:ffffc9001154f9b0 EFLAGS: 00010206 RAX: 0000000000000005 RBX: 1ffff1100302c8fd RCX: 0000000000000000 RDX: 0000000000000028 RSI: ffffc9001154f988 RDI: ffffc9000f77a338 RBP: 0000000000000029 R08: ffffffff8a50ffb4 R09: fffffbfff24b6bd9 R10: fffffbfff24b6bd9 R11: 0000000000000000 R12: ffff88801db73b78 R13: fffffffffffffff9 R14: dffffc0000000000 R15: 0000000000000030 FS: 00007f843ae8e700(0000) GS:ffff888063700000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f92ff166000 CR3: 000000003c672000 CR4: 00000000003506e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Fixes: 0daf07e52709 ("raw: convert raw sockets to RCU") Reported-by: syzbot Reported-by: Dae R. Jeong Link: https://lore.kernel.org/netdev/ZCA2mGV_cmq7lIfV@dragonet/ Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Signed-off-by: Jakub Kicinski --- include/net/raw.h | 4 ++-- net/ipv4/raw.c | 36 +++++++++++++++++++----------------- net/ipv4/raw_diag.c | 10 ++++------ net/ipv6/raw.c | 10 ++++------ 4 files changed, 29 insertions(+), 31 deletions(-) (limited to 'net') diff --git a/include/net/raw.h b/include/net/raw.h index 2c004c20ed99..3af5289fdead 100644 --- a/include/net/raw.h +++ b/include/net/raw.h @@ -37,7 +37,7 @@ int raw_rcv(struct sock *, struct sk_buff *); struct raw_hashinfo { spinlock_t lock; - struct hlist_nulls_head ht[RAW_HTABLE_SIZE] ____cacheline_aligned; + struct hlist_head ht[RAW_HTABLE_SIZE] ____cacheline_aligned; }; static inline u32 raw_hashfunc(const struct net *net, u32 proto) @@ -51,7 +51,7 @@ static inline void raw_hashinfo_init(struct raw_hashinfo *hashinfo) spin_lock_init(&hashinfo->lock); for (i = 0; i < RAW_HTABLE_SIZE; i++) - INIT_HLIST_NULLS_HEAD(&hashinfo->ht[i], i); + INIT_HLIST_HEAD(&hashinfo->ht[i]); } #ifdef CONFIG_PROC_FS diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 94df935ee0c5..8088a5011e7d 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -91,12 +91,12 @@ EXPORT_SYMBOL_GPL(raw_v4_hashinfo); int raw_hash_sk(struct sock *sk) { struct raw_hashinfo *h = sk->sk_prot->h.raw_hash; - struct hlist_nulls_head *hlist; + struct hlist_head *hlist; hlist = &h->ht[raw_hashfunc(sock_net(sk), inet_sk(sk)->inet_num)]; spin_lock(&h->lock); - __sk_nulls_add_node_rcu(sk, hlist); + sk_add_node_rcu(sk, hlist); sock_set_flag(sk, SOCK_RCU_FREE); spin_unlock(&h->lock); sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); @@ -110,7 +110,7 @@ void raw_unhash_sk(struct sock *sk) struct raw_hashinfo *h = sk->sk_prot->h.raw_hash; spin_lock(&h->lock); - if (__sk_nulls_del_node_init_rcu(sk)) + if (sk_del_node_init_rcu(sk)) sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); spin_unlock(&h->lock); } @@ -163,16 +163,15 @@ static int icmp_filter(const struct sock *sk, const struct sk_buff *skb) static int raw_v4_input(struct net *net, struct sk_buff *skb, const struct iphdr *iph, int hash) { - struct hlist_nulls_head *hlist; - struct hlist_nulls_node *hnode; int sdif = inet_sdif(skb); + struct hlist_head *hlist; int dif = inet_iif(skb); int delivered = 0; struct sock *sk; hlist = &raw_v4_hashinfo.ht[hash]; rcu_read_lock(); - sk_nulls_for_each(sk, hnode, hlist) { + sk_for_each_rcu(sk, hlist) { if (!raw_v4_match(net, sk, iph->protocol, iph->saddr, iph->daddr, dif, sdif)) continue; @@ -264,10 +263,9 @@ static void raw_err(struct sock *sk, struct sk_buff *skb, u32 info) void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info) { struct net *net = dev_net(skb->dev); - struct hlist_nulls_head *hlist; - struct hlist_nulls_node *hnode; int dif = skb->dev->ifindex; int sdif = inet_sdif(skb); + struct hlist_head *hlist; const struct iphdr *iph; struct sock *sk; int hash; @@ -276,7 +274,7 @@ void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info) hlist = &raw_v4_hashinfo.ht[hash]; rcu_read_lock(); - sk_nulls_for_each(sk, hnode, hlist) { + sk_for_each_rcu(sk, hlist) { iph = (const struct iphdr *)skb->data; if (!raw_v4_match(net, sk, iph->protocol, iph->daddr, iph->saddr, dif, sdif)) @@ -950,14 +948,13 @@ static struct sock *raw_get_first(struct seq_file *seq, int bucket) { struct raw_hashinfo *h = pde_data(file_inode(seq->file)); struct raw_iter_state *state = raw_seq_private(seq); - struct hlist_nulls_head *hlist; - struct hlist_nulls_node *hnode; + struct hlist_head *hlist; struct sock *sk; for (state->bucket = bucket; state->bucket < RAW_HTABLE_SIZE; ++state->bucket) { hlist = &h->ht[state->bucket]; - sk_nulls_for_each(sk, hnode, hlist) { + sk_for_each(sk, hlist) { if (sock_net(sk) == seq_file_net(seq)) return sk; } @@ -970,7 +967,7 @@ static struct sock *raw_get_next(struct seq_file *seq, struct sock *sk) struct raw_iter_state *state = raw_seq_private(seq); do { - sk = sk_nulls_next(sk); + sk = sk_next(sk); } while (sk && sock_net(sk) != seq_file_net(seq)); if (!sk) @@ -989,9 +986,12 @@ static struct sock *raw_get_idx(struct seq_file *seq, loff_t pos) } void *raw_seq_start(struct seq_file *seq, loff_t *pos) - __acquires(RCU) + __acquires(&h->lock) { - rcu_read_lock(); + struct raw_hashinfo *h = pde_data(file_inode(seq->file)); + + spin_lock(&h->lock); + return *pos ? raw_get_idx(seq, *pos - 1) : SEQ_START_TOKEN; } EXPORT_SYMBOL_GPL(raw_seq_start); @@ -1010,9 +1010,11 @@ void *raw_seq_next(struct seq_file *seq, void *v, loff_t *pos) EXPORT_SYMBOL_GPL(raw_seq_next); void raw_seq_stop(struct seq_file *seq, void *v) - __releases(RCU) + __releases(&h->lock) { - rcu_read_unlock(); + struct raw_hashinfo *h = pde_data(file_inode(seq->file)); + + spin_unlock(&h->lock); } EXPORT_SYMBOL_GPL(raw_seq_stop); diff --git a/net/ipv4/raw_diag.c b/net/ipv4/raw_diag.c index 999321834b94..da3591a66a16 100644 --- a/net/ipv4/raw_diag.c +++ b/net/ipv4/raw_diag.c @@ -57,8 +57,7 @@ static bool raw_lookup(struct net *net, struct sock *sk, static struct sock *raw_sock_get(struct net *net, const struct inet_diag_req_v2 *r) { struct raw_hashinfo *hashinfo = raw_get_hashinfo(r); - struct hlist_nulls_head *hlist; - struct hlist_nulls_node *hnode; + struct hlist_head *hlist; struct sock *sk; int slot; @@ -68,7 +67,7 @@ static struct sock *raw_sock_get(struct net *net, const struct inet_diag_req_v2 rcu_read_lock(); for (slot = 0; slot < RAW_HTABLE_SIZE; slot++) { hlist = &hashinfo->ht[slot]; - sk_nulls_for_each(sk, hnode, hlist) { + sk_for_each_rcu(sk, hlist) { if (raw_lookup(net, sk, r)) { /* * Grab it and keep until we fill @@ -142,9 +141,8 @@ static void raw_diag_dump(struct sk_buff *skb, struct netlink_callback *cb, struct raw_hashinfo *hashinfo = raw_get_hashinfo(r); struct net *net = sock_net(skb->sk); struct inet_diag_dump_data *cb_data; - struct hlist_nulls_head *hlist; - struct hlist_nulls_node *hnode; int num, s_num, slot, s_slot; + struct hlist_head *hlist; struct sock *sk = NULL; struct nlattr *bc; @@ -161,7 +159,7 @@ static void raw_diag_dump(struct sk_buff *skb, struct netlink_callback *cb, num = 0; hlist = &hashinfo->ht[slot]; - sk_nulls_for_each(sk, hnode, hlist) { + sk_for_each_rcu(sk, hlist) { struct inet_sock *inet = inet_sk(sk); if (!net_eq(sock_net(sk), net)) diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index bac9ba747bde..a327aa481df4 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -141,10 +141,9 @@ EXPORT_SYMBOL(rawv6_mh_filter_unregister); static bool ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) { struct net *net = dev_net(skb->dev); - struct hlist_nulls_head *hlist; - struct hlist_nulls_node *hnode; const struct in6_addr *saddr; const struct in6_addr *daddr; + struct hlist_head *hlist; struct sock *sk; bool delivered = false; __u8 hash; @@ -155,7 +154,7 @@ static bool ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) hash = raw_hashfunc(net, nexthdr); hlist = &raw_v6_hashinfo.ht[hash]; rcu_read_lock(); - sk_nulls_for_each(sk, hnode, hlist) { + sk_for_each_rcu(sk, hlist) { int filtered; if (!raw_v6_match(net, sk, nexthdr, daddr, saddr, @@ -333,15 +332,14 @@ void raw6_icmp_error(struct sk_buff *skb, int nexthdr, u8 type, u8 code, int inner_offset, __be32 info) { struct net *net = dev_net(skb->dev); - struct hlist_nulls_head *hlist; - struct hlist_nulls_node *hnode; + struct hlist_head *hlist; struct sock *sk; int hash; hash = raw_hashfunc(net, nexthdr); hlist = &raw_v6_hashinfo.ht[hash]; rcu_read_lock(); - sk_nulls_for_each(sk, hnode, hlist) { + sk_for_each_rcu(sk, hlist) { /* Note: ipv6_hdr(skb) != skb->data */ const struct ipv6hdr *ip6h = (const struct ipv6hdr *)skb->data; -- cgit v1.2.3 From ab5fb73ffa01072b4d8031cc05801fa1cb653bee Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 3 Apr 2023 12:49:59 -0700 Subject: ping: Fix potentail NULL deref for /proc/net/icmp. After commit dbca1596bbb0 ("ping: convert to RCU lookups, get rid of rwlock"), we use RCU for ping sockets, but we should use spinlock for /proc/net/icmp to avoid a potential NULL deref mentioned in the previous patch. Let's go back to using spinlock there. Note we can convert ping sockets to use hlist instead of hlist_nulls because we do not use SLAB_TYPESAFE_BY_RCU for ping sockets. Fixes: dbca1596bbb0 ("ping: convert to RCU lookups, get rid of rwlock") Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Signed-off-by: Jakub Kicinski --- net/ipv4/ping.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index 409ec2a1f95b..5178a3f3cb53 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -1089,13 +1089,13 @@ static struct sock *ping_get_idx(struct seq_file *seq, loff_t pos) } void *ping_seq_start(struct seq_file *seq, loff_t *pos, sa_family_t family) - __acquires(RCU) + __acquires(ping_table.lock) { struct ping_iter_state *state = seq->private; state->bucket = 0; state->family = family; - rcu_read_lock(); + spin_lock(&ping_table.lock); return *pos ? ping_get_idx(seq, *pos-1) : SEQ_START_TOKEN; } @@ -1121,9 +1121,9 @@ void *ping_seq_next(struct seq_file *seq, void *v, loff_t *pos) EXPORT_SYMBOL_GPL(ping_seq_next); void ping_seq_stop(struct seq_file *seq, void *v) - __releases(RCU) + __releases(ping_table.lock) { - rcu_read_unlock(); + spin_unlock(&ping_table.lock); } EXPORT_SYMBOL_GPL(ping_seq_stop); -- cgit v1.2.3 From e847c7675e19ef344913724dc68f83df31ad6a17 Mon Sep 17 00:00:00 2001 From: Andy Roulin Date: Mon, 3 Apr 2023 14:20:53 -0700 Subject: ethtool: reset #lanes when lanes is omitted If the number of lanes was forced and then subsequently the user omits this parameter, the ksettings->lanes is reset. The driver should then reset the number of lanes to the device's default for the specified speed. However, although the ksettings->lanes is set to 0, the mod variable is not set to true to indicate the driver and userspace should be notified of the changes. The consequence is that the same ethtool operation will produce different results based on the initial state. If the initial state is: $ ethtool swp1 | grep -A 3 'Speed: ' Speed: 500000Mb/s Lanes: 2 Duplex: Full Auto-negotiation: on then executing 'ethtool -s swp1 speed 50000 autoneg off' will yield: $ ethtool swp1 | grep -A 3 'Speed: ' Speed: 500000Mb/s Lanes: 2 Duplex: Full Auto-negotiation: off While if the initial state is: $ ethtool swp1 | grep -A 3 'Speed: ' Speed: 500000Mb/s Lanes: 1 Duplex: Full Auto-negotiation: off executing the same 'ethtool -s swp1 speed 50000 autoneg off' results in: $ ethtool swp1 | grep -A 3 'Speed: ' Speed: 500000Mb/s Lanes: 1 Duplex: Full Auto-negotiation: off This patch fixes this behavior. Omitting lanes will always results in the driver choosing the default lane width for the chosen speed. In this scenario, regardless of the initial state, the end state will be, e.g., $ ethtool swp1 | grep -A 3 'Speed: ' Speed: 500000Mb/s Lanes: 2 Duplex: Full Auto-negotiation: off Fixes: 012ce4dd3102 ("ethtool: Extend link modes settings uAPI with lanes") Signed-off-by: Andy Roulin Reviewed-by: Danielle Ratson Reviewed-by: Ido Schimmel Link: https://lore.kernel.org/r/ac238d6b-8726-8156-3810-6471291dbc7f@nvidia.com Signed-off-by: Jakub Kicinski --- net/ethtool/linkmodes.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ethtool/linkmodes.c b/net/ethtool/linkmodes.c index fab66c169b9f..20165e07ef90 100644 --- a/net/ethtool/linkmodes.c +++ b/net/ethtool/linkmodes.c @@ -270,11 +270,12 @@ static int ethnl_update_linkmodes(struct genl_info *info, struct nlattr **tb, "lanes configuration not supported by device"); return -EOPNOTSUPP; } - } else if (!lsettings->autoneg) { - /* If autoneg is off and lanes parameter is not passed from user, - * set the lanes parameter to 0. + } else if (!lsettings->autoneg && ksettings->lanes) { + /* If autoneg is off and lanes parameter is not passed from user but + * it was defined previously then set the lanes parameter to 0. */ ksettings->lanes = 0; + *mod = true; } ret = ethnl_update_bitset(ksettings->link_modes.advertising, -- cgit v1.2.3 From a1865f2e7d10dde00d35a2122b38d2e469ae67ed Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 3 Apr 2023 21:46:43 +0000 Subject: netlink: annotate lockless accesses to nlk->max_recvmsg_len syzbot reported a data-race in data-race in netlink_recvmsg() [1] Indeed, netlink_recvmsg() can be run concurrently, and netlink_dump() also needs protection. [1] BUG: KCSAN: data-race in netlink_recvmsg / netlink_recvmsg read to 0xffff888141840b38 of 8 bytes by task 23057 on cpu 0: netlink_recvmsg+0xea/0x730 net/netlink/af_netlink.c:1988 sock_recvmsg_nosec net/socket.c:1017 [inline] sock_recvmsg net/socket.c:1038 [inline] __sys_recvfrom+0x1ee/0x2e0 net/socket.c:2194 __do_sys_recvfrom net/socket.c:2212 [inline] __se_sys_recvfrom net/socket.c:2208 [inline] __x64_sys_recvfrom+0x78/0x90 net/socket.c:2208 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd write to 0xffff888141840b38 of 8 bytes by task 23037 on cpu 1: netlink_recvmsg+0x114/0x730 net/netlink/af_netlink.c:1989 sock_recvmsg_nosec net/socket.c:1017 [inline] sock_recvmsg net/socket.c:1038 [inline] ____sys_recvmsg+0x156/0x310 net/socket.c:2720 ___sys_recvmsg net/socket.c:2762 [inline] do_recvmmsg+0x2e5/0x710 net/socket.c:2856 __sys_recvmmsg net/socket.c:2935 [inline] __do_sys_recvmmsg net/socket.c:2958 [inline] __se_sys_recvmmsg net/socket.c:2951 [inline] __x64_sys_recvmmsg+0xe2/0x160 net/socket.c:2951 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd value changed: 0x0000000000000000 -> 0x0000000000001000 Reported by Kernel Concurrency Sanitizer on: CPU: 1 PID: 23037 Comm: syz-executor.2 Not tainted 6.3.0-rc4-syzkaller-00195-g5a57b48fdfcb #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/02/2023 Fixes: 9063e21fb026 ("netlink: autosize skb lengthes") Reported-by: syzbot Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230403214643.768555-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/netlink/af_netlink.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index c64277659753..f365dfdd672d 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1952,7 +1952,7 @@ static int netlink_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, struct scm_cookie scm; struct sock *sk = sock->sk; struct netlink_sock *nlk = nlk_sk(sk); - size_t copied; + size_t copied, max_recvmsg_len; struct sk_buff *skb, *data_skb; int err, ret; @@ -1985,9 +1985,10 @@ static int netlink_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, #endif /* Record the max length of recvmsg() calls for future allocations */ - nlk->max_recvmsg_len = max(nlk->max_recvmsg_len, len); - nlk->max_recvmsg_len = min_t(size_t, nlk->max_recvmsg_len, - SKB_WITH_OVERHEAD(32768)); + max_recvmsg_len = max(READ_ONCE(nlk->max_recvmsg_len), len); + max_recvmsg_len = min_t(size_t, max_recvmsg_len, + SKB_WITH_OVERHEAD(32768)); + WRITE_ONCE(nlk->max_recvmsg_len, max_recvmsg_len); copied = data_skb->len; if (len < copied) { @@ -2236,6 +2237,7 @@ static int netlink_dump(struct sock *sk) struct netlink_ext_ack extack = {}; struct netlink_callback *cb; struct sk_buff *skb = NULL; + size_t max_recvmsg_len; struct module *module; int err = -ENOBUFS; int alloc_min_size; @@ -2258,8 +2260,9 @@ static int netlink_dump(struct sock *sk) cb = &nlk->cb; alloc_min_size = max_t(int, cb->min_dump_alloc, NLMSG_GOODSIZE); - if (alloc_min_size < nlk->max_recvmsg_len) { - alloc_size = nlk->max_recvmsg_len; + max_recvmsg_len = READ_ONCE(nlk->max_recvmsg_len); + if (alloc_min_size < max_recvmsg_len) { + alloc_size = max_recvmsg_len; skb = alloc_skb(alloc_size, (GFP_KERNEL & ~__GFP_DIRECT_RECLAIM) | __GFP_NOWARN | __GFP_NORETRY); -- cgit v1.2.3 From b45193cb4df556fe6251b285a5ce44046dd36b4a Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Tue, 4 Apr 2023 09:31:28 +0200 Subject: can: j1939: j1939_tp_tx_dat_new(): fix out-of-bounds memory access In the j1939_tp_tx_dat_new() function, an out-of-bounds memory access could occur during the memcpy() operation if the size of skb->cb is larger than the size of struct j1939_sk_buff_cb. This is because the memcpy() operation uses the size of skb->cb, leading to a read beyond the struct j1939_sk_buff_cb. Updated the memcpy() operation to use the size of struct j1939_sk_buff_cb instead of the size of skb->cb. This ensures that the memcpy() operation only reads the memory within the bounds of struct j1939_sk_buff_cb, preventing out-of-bounds memory access. Additionally, add a BUILD_BUG_ON() to check that the size of skb->cb is greater than or equal to the size of struct j1939_sk_buff_cb. This ensures that the skb->cb buffer is large enough to hold the j1939_sk_buff_cb structure. Fixes: 9d71dd0c7009 ("can: add support of SAE J1939 protocol") Reported-by: Shuangpeng Bai Tested-by: Shuangpeng Bai Signed-off-by: Oleksij Rempel Link: https://groups.google.com/g/syzkaller/c/G_LL-C3plRs/m/-8xCi6dCAgAJ Link: https://lore.kernel.org/all/20230404073128.3173900-1-o.rempel@pengutronix.de Cc: stable@vger.kernel.org [mkl: rephrase commit message] Signed-off-by: Marc Kleine-Budde --- net/can/j1939/transport.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/can/j1939/transport.c b/net/can/j1939/transport.c index fb92c3609e17..fe3df23a2595 100644 --- a/net/can/j1939/transport.c +++ b/net/can/j1939/transport.c @@ -604,7 +604,10 @@ sk_buff *j1939_tp_tx_dat_new(struct j1939_priv *priv, /* reserve CAN header */ skb_reserve(skb, offsetof(struct can_frame, data)); - memcpy(skb->cb, re_skcb, sizeof(skb->cb)); + /* skb->cb must be large enough to hold a j1939_sk_buff_cb structure */ + BUILD_BUG_ON(sizeof(skb->cb) < sizeof(*re_skcb)); + + memcpy(skb->cb, re_skcb, sizeof(*re_skcb)); skcb = j1939_skb_to_cb(skb); if (swap_src_dst) j1939_skbcb_swap(skcb); -- cgit v1.2.3 From 0145462fc802cd447ef5d029758043c7f15b4b1e Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Thu, 30 Mar 2023 19:02:48 +0200 Subject: can: isotp: isotp_recvmsg(): use sock_recv_cmsgs() to get SOCK_RXQ_OVFL infos isotp.c was still using sock_recv_timestamp() which does not provide control messages to detect dropped PDUs in the receive path. Fixes: e057dd3fc20f ("can: add ISO 15765-2:2016 transport protocol") Signed-off-by: Oliver Hartkopp Link: https://lore.kernel.org/all/20230330170248.62342-1-socketcan@hartkopp.net Cc: stable@vger.kernel.org Signed-off-by: Marc Kleine-Budde --- net/can/isotp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/can/isotp.c b/net/can/isotp.c index 9bc344851704..47c2ebad10ed 100644 --- a/net/can/isotp.c +++ b/net/can/isotp.c @@ -1120,7 +1120,7 @@ static int isotp_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, if (ret < 0) goto out_err; - sock_recv_timestamp(msg, sk, skb); + sock_recv_cmsgs(msg, sk, skb); if (msg->msg_name) { __sockaddr_check_size(ISOTP_MIN_NAMELEN); -- cgit v1.2.3 From 79e19fa79cb5d5f1b3bf3e3ae24989ccb93c7b7b Mon Sep 17 00:00:00 2001 From: Michal Sojka Date: Fri, 31 Mar 2023 14:55:11 +0200 Subject: can: isotp: isotp_ops: fix poll() to not report false EPOLLOUT events When using select()/poll()/epoll() with a non-blocking ISOTP socket to wait for when non-blocking write is possible, a false EPOLLOUT event is sometimes returned. This can happen at least after sending a message which must be split to multiple CAN frames. The reason is that isotp_sendmsg() returns -EAGAIN when tx.state is not equal to ISOTP_IDLE and this behavior is not reflected in datagram_poll(), which is used in isotp_ops. This is fixed by introducing ISOTP-specific poll function, which suppresses the EPOLLOUT events in that case. v2: https://lore.kernel.org/all/20230302092812.320643-1-michal.sojka@cvut.cz v1: https://lore.kernel.org/all/20230224010659.48420-1-michal.sojka@cvut.cz https://lore.kernel.org/all/b53a04a2-ba1f-3858-84c1-d3eb3301ae15@hartkopp.net Signed-off-by: Michal Sojka Reported-by: Jakub Jira Tested-by: Oliver Hartkopp Acked-by: Oliver Hartkopp Fixes: e057dd3fc20f ("can: add ISO 15765-2:2016 transport protocol") Link: https://lore.kernel.org/all/20230331125511.372783-1-michal.sojka@cvut.cz Cc: stable@vger.kernel.org Signed-off-by: Marc Kleine-Budde --- net/can/isotp.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/can/isotp.c b/net/can/isotp.c index 47c2ebad10ed..281b7766c54e 100644 --- a/net/can/isotp.c +++ b/net/can/isotp.c @@ -1608,6 +1608,21 @@ static int isotp_init(struct sock *sk) return 0; } +static __poll_t isotp_poll(struct file *file, struct socket *sock, poll_table *wait) +{ + struct sock *sk = sock->sk; + struct isotp_sock *so = isotp_sk(sk); + + __poll_t mask = datagram_poll(file, sock, wait); + poll_wait(file, &so->wait, wait); + + /* Check for false positives due to TX state */ + if ((mask & EPOLLWRNORM) && (so->tx.state != ISOTP_IDLE)) + mask &= ~(EPOLLOUT | EPOLLWRNORM); + + return mask; +} + static int isotp_sock_no_ioctlcmd(struct socket *sock, unsigned int cmd, unsigned long arg) { @@ -1623,7 +1638,7 @@ static const struct proto_ops isotp_ops = { .socketpair = sock_no_socketpair, .accept = sock_no_accept, .getname = isotp_getname, - .poll = datagram_poll, + .poll = isotp_poll, .ioctl = isotp_sock_no_ioctlcmd, .gettstamp = sock_gettstamp, .listen = sock_no_listen, -- cgit v1.2.3 From 051737439eaee5bdd03d3c2ef5510d54a478fd05 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 31 Mar 2023 15:19:35 +0200 Subject: can: isotp: fix race between isotp_sendsmg() and isotp_release() As discussed with Dae R. Jeong and Hillf Danton here [1] the sendmsg() function in isotp.c might get into a race condition when restoring the former tx.state from the old_state. Remove the old_state concept and implement proper locking for the ISOTP_IDLE transitions in isotp_sendmsg(), inspired by a simplification idea from Hillf Danton. Introduce a new tx.state ISOTP_SHUTDOWN and use the same locking mechanism from isotp_release() which resolves a potential race between isotp_sendsmg() and isotp_release(). [1] https://lore.kernel.org/linux-can/ZB%2F93xJxq%2FBUqAgG@dragonet v1: https://lore.kernel.org/all/20230331102114.15164-1-socketcan@hartkopp.net v2: https://lore.kernel.org/all/20230331123600.3550-1-socketcan@hartkopp.net take care of signal interrupts for wait_event_interruptible() in isotp_release() v3: https://lore.kernel.org/all/20230331130654.9886-1-socketcan@hartkopp.net take care of signal interrupts for wait_event_interruptible() in isotp_sendmsg() in the wait_tx_done case v4: https://lore.kernel.org/all/20230331131935.21465-1-socketcan@hartkopp.net take care of signal interrupts for wait_event_interruptible() in isotp_sendmsg() in ALL cases Cc: Dae R. Jeong Cc: Hillf Danton Signed-off-by: Oliver Hartkopp Fixes: 4f027cba8216 ("can: isotp: split tx timer into transmission and timeout") Link: https://lore.kernel.org/all/20230331131935.21465-1-socketcan@hartkopp.net Cc: stable@vger.kernel.org [mkl: rephrase commit message] Signed-off-by: Marc Kleine-Budde --- net/can/isotp.c | 55 +++++++++++++++++++++++++++++++------------------------ 1 file changed, 31 insertions(+), 24 deletions(-) (limited to 'net') diff --git a/net/can/isotp.c b/net/can/isotp.c index 281b7766c54e..5761d4ab839d 100644 --- a/net/can/isotp.c +++ b/net/can/isotp.c @@ -119,7 +119,8 @@ enum { ISOTP_WAIT_FIRST_FC, ISOTP_WAIT_FC, ISOTP_WAIT_DATA, - ISOTP_SENDING + ISOTP_SENDING, + ISOTP_SHUTDOWN, }; struct tpcon { @@ -880,8 +881,8 @@ static enum hrtimer_restart isotp_tx_timer_handler(struct hrtimer *hrtimer) txtimer); struct sock *sk = &so->sk; - /* don't handle timeouts in IDLE state */ - if (so->tx.state == ISOTP_IDLE) + /* don't handle timeouts in IDLE or SHUTDOWN state */ + if (so->tx.state == ISOTP_IDLE || so->tx.state == ISOTP_SHUTDOWN) return HRTIMER_NORESTART; /* we did not get any flow control or echo frame in time */ @@ -918,7 +919,6 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) { struct sock *sk = sock->sk; struct isotp_sock *so = isotp_sk(sk); - u32 old_state = so->tx.state; struct sk_buff *skb; struct net_device *dev; struct canfd_frame *cf; @@ -928,23 +928,24 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) int off; int err; - if (!so->bound) + if (!so->bound || so->tx.state == ISOTP_SHUTDOWN) return -EADDRNOTAVAIL; +wait_free_buffer: /* we do not support multiple buffers - for now */ - if (cmpxchg(&so->tx.state, ISOTP_IDLE, ISOTP_SENDING) != ISOTP_IDLE || - wq_has_sleeper(&so->wait)) { - if (msg->msg_flags & MSG_DONTWAIT) { - err = -EAGAIN; - goto err_out; - } + if (wq_has_sleeper(&so->wait) && (msg->msg_flags & MSG_DONTWAIT)) + return -EAGAIN; - /* wait for complete transmission of current pdu */ - err = wait_event_interruptible(so->wait, so->tx.state == ISOTP_IDLE); - if (err) - goto err_out; + /* wait for complete transmission of current pdu */ + err = wait_event_interruptible(so->wait, so->tx.state == ISOTP_IDLE); + if (err) + goto err_event_drop; - so->tx.state = ISOTP_SENDING; + if (cmpxchg(&so->tx.state, ISOTP_IDLE, ISOTP_SENDING) != ISOTP_IDLE) { + if (so->tx.state == ISOTP_SHUTDOWN) + return -EADDRNOTAVAIL; + + goto wait_free_buffer; } if (!size || size > MAX_MSG_LENGTH) { @@ -1074,7 +1075,9 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) if (wait_tx_done) { /* wait for complete transmission of current pdu */ - wait_event_interruptible(so->wait, so->tx.state == ISOTP_IDLE); + err = wait_event_interruptible(so->wait, so->tx.state == ISOTP_IDLE); + if (err) + goto err_event_drop; if (sk->sk_err) return -sk->sk_err; @@ -1082,13 +1085,15 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) return size; +err_event_drop: + /* got signal: force tx state machine to be idle */ + so->tx.state = ISOTP_IDLE; + hrtimer_cancel(&so->txfrtimer); + hrtimer_cancel(&so->txtimer); err_out_drop: /* drop this PDU and unlock a potential wait queue */ - old_state = ISOTP_IDLE; -err_out: - so->tx.state = old_state; - if (so->tx.state == ISOTP_IDLE) - wake_up_interruptible(&so->wait); + so->tx.state = ISOTP_IDLE; + wake_up_interruptible(&so->wait); return err; } @@ -1150,10 +1155,12 @@ static int isotp_release(struct socket *sock) net = sock_net(sk); /* wait for complete transmission of current pdu */ - wait_event_interruptible(so->wait, so->tx.state == ISOTP_IDLE); + while (wait_event_interruptible(so->wait, so->tx.state == ISOTP_IDLE) == 0 && + cmpxchg(&so->tx.state, ISOTP_IDLE, ISOTP_SHUTDOWN) != ISOTP_IDLE) + ; /* force state machines to be idle also when a signal occurred */ - so->tx.state = ISOTP_IDLE; + so->tx.state = ISOTP_SHUTDOWN; so->rx.state = ISOTP_IDLE; spin_lock(&isotp_notifier_lock); -- cgit v1.2.3 From aef3b8b8dd55a8bdfd7ef96cfde0aa71fd82ed10 Mon Sep 17 00:00:00 2001 From: "Uladzislau Rezki (Sony)" Date: Wed, 1 Feb 2023 16:08:12 +0100 Subject: net/sysctl: Rename kvfree_rcu() to kvfree_rcu_mightsleep() The kfree_rcu() and kvfree_rcu() macros' single-argument forms are deprecated. Therefore switch to the new kfree_rcu_mightsleep() and kvfree_rcu_mightsleep() variants. The goal is to avoid accidental use of the single-argument forms, which can introduce functionality bugs in atomic contexts and latency bugs in non-atomic contexts. Acked-by: Jakub Kicinski Cc: Eric Dumazet Cc: David S. Miller Signed-off-by: Uladzislau Rezki (Sony) Signed-off-by: Paul E. McKenney Signed-off-by: Joel Fernandes (Google) --- net/core/sysctl_net_core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 74842b453407..782273bb93c2 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -177,7 +177,7 @@ static int rps_sock_flow_sysctl(struct ctl_table *table, int write, if (orig_sock_table) { static_branch_dec(&rps_needed); static_branch_dec(&rfs_needed); - kvfree_rcu(orig_sock_table); + kvfree_rcu_mightsleep(orig_sock_table); } } } @@ -215,7 +215,7 @@ static int flow_limit_cpu_sysctl(struct ctl_table *table, int write, lockdep_is_held(&flow_limit_update_mutex)); if (cur && !cpumask_test_cpu(i, mask)) { RCU_INIT_POINTER(sd->flow_limit, NULL); - kfree_rcu(cur); + kfree_rcu_mightsleep(cur); } else if (!cur && cpumask_test_cpu(i, mask)) { cur = kzalloc_node(len, GFP_KERNEL, cpu_to_node(i)); -- cgit v1.2.3 From eb56a4cbc3ea348dc07fbe82531c27cce83df3b9 Mon Sep 17 00:00:00 2001 From: "Joel Fernandes (Google)" Date: Fri, 10 Mar 2023 01:18:12 +0000 Subject: mac802154: Rename kfree_rcu() to kvfree_rcu_mightsleep() The k[v]free_rcu() macro's single-argument form is deprecated. Therefore switch to the new k[v]free_rcu_mightsleep() variant. The goal is to avoid accidental use of the single-argument forms, which can introduce functionality bugs in atomic contexts and latency bugs in non-atomic contexts. The callers are holding a mutex so the context allows blocking. Hence using the API with a single argument will be fine, but use its new name. There is no functionality change with this patch. Fixes: 57588c71177f ("mac802154: Handle passive scanning") Acked-by: Stefan Schmidt Reviewed-by: Paul E. McKenney Signed-off-by: Joel Fernandes (Google) --- net/mac802154/scan.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac802154/scan.c b/net/mac802154/scan.c index 9b0933a185eb..5c191bedd72c 100644 --- a/net/mac802154/scan.c +++ b/net/mac802154/scan.c @@ -52,7 +52,7 @@ static int mac802154_scan_cleanup_locked(struct ieee802154_local *local, request = rcu_replace_pointer(local->scan_req, NULL, 1); if (!request) return 0; - kfree_rcu(request); + kvfree_rcu_mightsleep(request); /* Advertize first, while we know the devices cannot be removed */ if (aborted) @@ -403,7 +403,7 @@ int mac802154_stop_beacons_locked(struct ieee802154_local *local, request = rcu_replace_pointer(local->beacon_req, NULL, 1); if (!request) return 0; - kfree_rcu(request); + kvfree_rcu_mightsleep(request); nl802154_beaconing_done(wpan_dev); -- cgit v1.2.3 From 8b0f256530d97f2a2310c4f8336ea2c477c8e6c4 Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Tue, 4 Apr 2023 17:34:49 -0300 Subject: net/sched: sch_mqprio: use netlink payload helpers For the sake of readability, use the netlink payload helpers from the 'nla_get_*()' family to parse the attributes. tdc results: 1..5 ok 1 9903 - Add mqprio Qdisc to multi-queue device (8 queues) ok 2 453a - Delete nonexistent mqprio Qdisc ok 3 5292 - Delete mqprio Qdisc twice ok 4 45a9 - Add mqprio Qdisc to single-queue device ok 5 2ba9 - Show mqprio class Acked-by: Jamal Hadi Salim Signed-off-by: Pedro Tammela Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230404203449.1627033-1-pctammela@mojatatu.com Signed-off-by: Jakub Kicinski --- net/sched/sch_mqprio.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index 48ed87b91086..fdd6a6575a54 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -178,12 +178,12 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt, if (tb[TCA_MQPRIO_MODE]) { priv->flags |= TC_MQPRIO_F_MODE; - priv->mode = *(u16 *)nla_data(tb[TCA_MQPRIO_MODE]); + priv->mode = nla_get_u16(tb[TCA_MQPRIO_MODE]); } if (tb[TCA_MQPRIO_SHAPER]) { priv->flags |= TC_MQPRIO_F_SHAPER; - priv->shaper = *(u16 *)nla_data(tb[TCA_MQPRIO_SHAPER]); + priv->shaper = nla_get_u16(tb[TCA_MQPRIO_SHAPER]); } if (tb[TCA_MQPRIO_MIN_RATE64]) { @@ -196,7 +196,7 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt, return -EINVAL; if (i >= qopt->num_tc) break; - priv->min_rate[i] = *(u64 *)nla_data(attr); + priv->min_rate[i] = nla_get_u64(attr); i++; } priv->flags |= TC_MQPRIO_F_MIN_RATE; @@ -212,7 +212,7 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt, return -EINVAL; if (i >= qopt->num_tc) break; - priv->max_rate[i] = *(u64 *)nla_data(attr); + priv->max_rate[i] = nla_get_u64(attr); i++; } priv->flags |= TC_MQPRIO_F_MAX_RATE; -- cgit v1.2.3 From 23baf831a32c04f9a968812511540b1b3e648bf5 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 15 Mar 2023 14:31:33 +0300 Subject: mm, treewide: redefine MAX_ORDER sanely MAX_ORDER currently defined as number of orders page allocator supports: user can ask buddy allocator for page order between 0 and MAX_ORDER-1. This definition is counter-intuitive and lead to number of bugs all over the kernel. Change the definition of MAX_ORDER to be inclusive: the range of orders user can ask from buddy allocator is 0..MAX_ORDER now. [kirill@shutemov.name: fix min() warning] Link: https://lkml.kernel.org/r/20230315153800.32wib3n5rickolvh@box [akpm@linux-foundation.org: fix another min_t warning] [kirill@shutemov.name: fixups per Zi Yan] Link: https://lkml.kernel.org/r/20230316232144.b7ic4cif4kjiabws@box.shutemov.name [akpm@linux-foundation.org: fix underlining in docs] Link: https://lore.kernel.org/oe-kbuild-all/202303191025.VRCTk6mP-lkp@intel.com/ Link: https://lkml.kernel.org/r/20230315113133.11326-11-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov Reviewed-by: Michael Ellerman [powerpc] Cc: "Kirill A. Shutemov" Cc: Zi Yan Signed-off-by: Andrew Morton --- Documentation/admin-guide/kdump/vmcoreinfo.rst | 6 ++-- Documentation/admin-guide/kernel-parameters.txt | 2 +- arch/arc/Kconfig | 4 +-- arch/arm/Kconfig | 9 ++---- arch/arm/configs/imx_v6_v7_defconfig | 2 +- arch/arm/configs/milbeaut_m10v_defconfig | 2 +- arch/arm/configs/oxnas_v6_defconfig | 2 +- arch/arm/configs/pxa_defconfig | 2 +- arch/arm/configs/sama7_defconfig | 2 +- arch/arm/configs/sp7021_defconfig | 2 +- arch/arm64/Kconfig | 27 ++++++++---------- arch/arm64/include/asm/sparsemem.h | 2 +- arch/arm64/kvm/hyp/include/nvhe/gfp.h | 2 +- arch/arm64/kvm/hyp/nvhe/page_alloc.c | 10 +++---- arch/csky/Kconfig | 2 +- arch/ia64/Kconfig | 8 +++--- arch/ia64/include/asm/sparsemem.h | 4 +-- arch/ia64/mm/hugetlbpage.c | 2 +- arch/loongarch/Kconfig | 15 ++++------ arch/m68k/Kconfig.cpu | 5 +--- arch/mips/Kconfig | 19 ++++++------- arch/nios2/Kconfig | 7 ++--- arch/powerpc/Kconfig | 27 ++++++++---------- arch/powerpc/configs/85xx/ge_imp3a_defconfig | 2 +- arch/powerpc/configs/fsl-emb-nonhw.config | 2 +- arch/powerpc/mm/book3s64/iommu_api.c | 2 +- arch/powerpc/mm/hugetlbpage.c | 2 +- arch/powerpc/platforms/powernv/pci-ioda.c | 2 +- arch/sh/configs/ecovec24_defconfig | 2 +- arch/sh/mm/Kconfig | 17 +++++------ arch/sparc/Kconfig | 5 +--- arch/sparc/kernel/pci_sun4v.c | 2 +- arch/sparc/kernel/traps_64.c | 2 +- arch/sparc/mm/tsb.c | 4 +-- arch/um/kernel/um_arch.c | 4 +-- arch/xtensa/Kconfig | 5 +--- drivers/base/regmap/regmap-debugfs.c | 8 +++--- drivers/block/floppy.c | 2 +- drivers/crypto/ccp/sev-dev.c | 2 +- drivers/crypto/hisilicon/sgl.c | 6 ++-- drivers/gpu/drm/i915/gem/i915_gem_internal.c | 2 +- drivers/gpu/drm/i915/gem/selftests/huge_pages.c | 2 +- drivers/gpu/drm/ttm/ttm_pool.c | 22 +++++++------- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 2 +- drivers/iommu/dma-iommu.c | 2 +- drivers/irqchip/irq-gic-v3-its.c | 4 +-- drivers/md/dm-bufio.c | 2 +- drivers/misc/genwqe/card_dev.c | 2 +- drivers/misc/genwqe/card_utils.c | 4 +-- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 2 +- drivers/net/ethernet/ibm/ibmvnic.h | 2 +- drivers/video/fbdev/hyperv_fb.c | 4 +-- drivers/video/fbdev/vermilion/vermilion.c | 2 +- drivers/virtio/virtio_balloon.c | 2 +- drivers/virtio/virtio_mem.c | 12 ++++---- fs/ramfs/file-nommu.c | 2 +- include/drm/ttm/ttm_pool.h | 2 +- include/linux/hugetlb.h | 2 +- include/linux/mmzone.h | 10 +++---- include/linux/pageblock-flags.h | 4 +-- include/linux/slab.h | 6 ++-- kernel/crash_core.c | 2 +- kernel/dma/pool.c | 6 ++-- kernel/events/ring_buffer.c | 6 ++-- mm/Kconfig | 10 +++---- mm/compaction.c | 8 +++--- mm/debug_vm_pgtable.c | 4 +-- mm/huge_memory.c | 2 +- mm/hugetlb.c | 4 +-- mm/kmsan/init.c | 6 ++-- mm/memblock.c | 2 +- mm/memory_hotplug.c | 4 +-- mm/page_alloc.c | 38 ++++++++++++------------- mm/page_isolation.c | 12 ++++---- mm/page_owner.c | 6 ++-- mm/page_reporting.c | 6 ++-- mm/shuffle.h | 2 +- mm/slab.c | 2 +- mm/slub.c | 6 ++-- mm/vmscan.c | 2 +- mm/vmstat.c | 14 ++++----- net/smc/smc_ib.c | 2 +- security/integrity/ima/ima_crypto.c | 2 +- tools/testing/memblock/linux/mmzone.h | 6 ++-- 84 files changed, 223 insertions(+), 253 deletions(-) (limited to 'net') diff --git a/Documentation/admin-guide/kdump/vmcoreinfo.rst b/Documentation/admin-guide/kdump/vmcoreinfo.rst index 86fd88492870..c18d94fa6470 100644 --- a/Documentation/admin-guide/kdump/vmcoreinfo.rst +++ b/Documentation/admin-guide/kdump/vmcoreinfo.rst @@ -172,7 +172,7 @@ variables. Offset of the free_list's member. This value is used to compute the number of free pages. -Each zone has a free_area structure array called free_area[MAX_ORDER]. +Each zone has a free_area structure array called free_area[MAX_ORDER + 1]. The free_list represents a linked list of free page blocks. (list_head, next|prev) @@ -189,8 +189,8 @@ Offsets of the vmap_area's members. They carry vmalloc-specific information. Makedumpfile gets the start address of the vmalloc region from this. -(zone.free_area, MAX_ORDER) ---------------------------- +(zone.free_area, MAX_ORDER + 1) +------------------------------- Free areas descriptor. User-space tools use this value to iterate the free_area ranges. MAX_ORDER is used by the zone buddy allocator. diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 6221a1d057dd..50da4f26fad5 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -3969,7 +3969,7 @@ [KNL] Minimal page reporting order Format: Adjust the minimal page reporting order. The page - reporting is disabled when it exceeds (MAX_ORDER-1). + reporting is disabled when it exceeds MAX_ORDER. panic= [KNL] Kernel behaviour on panic: delay timeout > 0: seconds before rebooting diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig index d9a13ccf89a3..ab6d701365bb 100644 --- a/arch/arc/Kconfig +++ b/arch/arc/Kconfig @@ -556,7 +556,7 @@ endmenu # "ARC Architecture Configuration" config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - default "12" if ARC_HUGEPAGE_16M - default "11" + default "11" if ARC_HUGEPAGE_16M + default "10" source "kernel/power/Kconfig" diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index e24a9820e12f..929e646e84b9 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -1355,9 +1355,9 @@ config ARM_MODULE_PLTS config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - default "12" if SOC_AM33XX - default "9" if SA1111 - default "11" + default "11" if SOC_AM33XX + default "8" if SA1111 + default "10" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -1366,9 +1366,6 @@ config ARCH_FORCE_MAX_ORDER blocks of physically contiguous memory, then you may need to increase this value. - This config option is actually maximum order plus one. For example, - a value of 11 means that the largest free memory block is 2^10 pages. - config ALIGNMENT_TRAP def_bool CPU_CP15_MMU select HAVE_PROC_CPU if PROC_FS diff --git a/arch/arm/configs/imx_v6_v7_defconfig b/arch/arm/configs/imx_v6_v7_defconfig index 6dc6fed12af8..345a67e67dbd 100644 --- a/arch/arm/configs/imx_v6_v7_defconfig +++ b/arch/arm/configs/imx_v6_v7_defconfig @@ -31,7 +31,7 @@ CONFIG_SOC_VF610=y CONFIG_SMP=y CONFIG_ARM_PSCI=y CONFIG_HIGHMEM=y -CONFIG_ARCH_FORCE_MAX_ORDER=14 +CONFIG_ARCH_FORCE_MAX_ORDER=13 CONFIG_CMDLINE="noinitrd console=ttymxc0,115200" CONFIG_KEXEC=y CONFIG_CPU_FREQ=y diff --git a/arch/arm/configs/milbeaut_m10v_defconfig b/arch/arm/configs/milbeaut_m10v_defconfig index bd29e5012cb0..385ad0f391a8 100644 --- a/arch/arm/configs/milbeaut_m10v_defconfig +++ b/arch/arm/configs/milbeaut_m10v_defconfig @@ -26,7 +26,7 @@ CONFIG_THUMB2_KERNEL=y # CONFIG_THUMB2_AVOID_R_ARM_THM_JUMP11 is not set # CONFIG_ARM_PATCH_IDIV is not set CONFIG_HIGHMEM=y -CONFIG_ARCH_FORCE_MAX_ORDER=12 +CONFIG_ARCH_FORCE_MAX_ORDER=11 CONFIG_SECCOMP=y CONFIG_KEXEC=y CONFIG_EFI=y diff --git a/arch/arm/configs/oxnas_v6_defconfig b/arch/arm/configs/oxnas_v6_defconfig index 70a67b3fc91b..90779812c6dd 100644 --- a/arch/arm/configs/oxnas_v6_defconfig +++ b/arch/arm/configs/oxnas_v6_defconfig @@ -12,7 +12,7 @@ CONFIG_ARCH_OXNAS=y CONFIG_MACH_OX820=y CONFIG_SMP=y CONFIG_NR_CPUS=16 -CONFIG_ARCH_FORCE_MAX_ORDER=12 +CONFIG_ARCH_FORCE_MAX_ORDER=11 CONFIG_SECCOMP=y CONFIG_ARM_APPENDED_DTB=y CONFIG_ARM_ATAG_DTB_COMPAT=y diff --git a/arch/arm/configs/pxa_defconfig b/arch/arm/configs/pxa_defconfig index e656d3af2266..b46e39369dbb 100644 --- a/arch/arm/configs/pxa_defconfig +++ b/arch/arm/configs/pxa_defconfig @@ -20,7 +20,7 @@ CONFIG_PXA_SHARPSL=y CONFIG_MACH_AKITA=y CONFIG_MACH_BORZOI=y CONFIG_AEABI=y -CONFIG_ARCH_FORCE_MAX_ORDER=9 +CONFIG_ARCH_FORCE_MAX_ORDER=8 CONFIG_CMDLINE="root=/dev/ram0 ro" CONFIG_KEXEC=y CONFIG_CPU_FREQ=y diff --git a/arch/arm/configs/sama7_defconfig b/arch/arm/configs/sama7_defconfig index 0d964c613d71..954112041403 100644 --- a/arch/arm/configs/sama7_defconfig +++ b/arch/arm/configs/sama7_defconfig @@ -19,7 +19,7 @@ CONFIG_ATMEL_CLOCKSOURCE_TCB=y # CONFIG_CACHE_L2X0 is not set # CONFIG_ARM_PATCH_IDIV is not set # CONFIG_CPU_SW_DOMAIN_PAN is not set -CONFIG_ARCH_FORCE_MAX_ORDER=15 +CONFIG_ARCH_FORCE_MAX_ORDER=14 CONFIG_UACCESS_WITH_MEMCPY=y # CONFIG_ATAGS is not set CONFIG_CMDLINE="console=ttyS0,115200 earlyprintk ignore_loglevel" diff --git a/arch/arm/configs/sp7021_defconfig b/arch/arm/configs/sp7021_defconfig index 5bca2eb59b86..c6448ac860b6 100644 --- a/arch/arm/configs/sp7021_defconfig +++ b/arch/arm/configs/sp7021_defconfig @@ -17,7 +17,7 @@ CONFIG_ARCH_SUNPLUS=y # CONFIG_VDSO is not set CONFIG_SMP=y CONFIG_THUMB2_KERNEL=y -CONFIG_ARCH_FORCE_MAX_ORDER=12 +CONFIG_ARCH_FORCE_MAX_ORDER=11 CONFIG_VFP=y CONFIG_NEON=y CONFIG_MODULES=y diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 1023e896d46b..cb5c6aa3254e 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -1476,22 +1476,22 @@ config XEN # include/linux/mmzone.h requires the following to be true: # -# MAX_ORDER - 1 + PAGE_SHIFT <= SECTION_SIZE_BITS +# MAX_ORDER + PAGE_SHIFT <= SECTION_SIZE_BITS # -# so the maximum value of MAX_ORDER is SECTION_SIZE_BITS + 1 - PAGE_SHIFT: +# so the maximum value of MAX_ORDER is SECTION_SIZE_BITS - PAGE_SHIFT: # # | SECTION_SIZE_BITS | PAGE_SHIFT | max MAX_ORDER | default MAX_ORDER | # ----+-------------------+--------------+-----------------+--------------------+ -# 4K | 27 | 12 | 16 | 11 | -# 16K | 27 | 14 | 14 | 12 | -# 64K | 29 | 16 | 14 | 14 | +# 4K | 27 | 12 | 15 | 10 | +# 16K | 27 | 14 | 13 | 11 | +# 64K | 29 | 16 | 13 | 13 | config ARCH_FORCE_MAX_ORDER int "Maximum zone order" if ARM64_4K_PAGES || ARM64_16K_PAGES - default "14" if ARM64_64K_PAGES - range 12 14 if ARM64_16K_PAGES - default "12" if ARM64_16K_PAGES - range 11 16 if ARM64_4K_PAGES - default "11" + default "13" if ARM64_64K_PAGES + range 11 13 if ARM64_16K_PAGES + default "11" if ARM64_16K_PAGES + range 10 15 if ARM64_4K_PAGES + default "10" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -1500,14 +1500,11 @@ config ARCH_FORCE_MAX_ORDER blocks of physically contiguous memory, then you may need to increase this value. - This config option is actually maximum order plus one. For example, - a value of 11 means that the largest free memory block is 2^10 pages. - We make sure that we can allocate up to a HugePage size for each configuration. Hence we have : - MAX_ORDER = (PMD_SHIFT - PAGE_SHIFT) + 1 => PAGE_SHIFT - 2 + MAX_ORDER = PMD_SHIFT - PAGE_SHIFT => PAGE_SHIFT - 3 - However for 4K, we choose a higher default value, 11 as opposed to 10, giving us + However for 4K, we choose a higher default value, 10 as opposed to 9, giving us 4M allocations matching the default size used by generic code. config UNMAP_KERNEL_AT_EL0 diff --git a/arch/arm64/include/asm/sparsemem.h b/arch/arm64/include/asm/sparsemem.h index 4b73463423c3..5f5437621029 100644 --- a/arch/arm64/include/asm/sparsemem.h +++ b/arch/arm64/include/asm/sparsemem.h @@ -10,7 +10,7 @@ /* * Section size must be at least 512MB for 64K base * page size config. Otherwise it will be less than - * (MAX_ORDER - 1) and the build process will fail. + * MAX_ORDER and the build process will fail. */ #ifdef CONFIG_ARM64_64K_PAGES #define SECTION_SIZE_BITS 29 diff --git a/arch/arm64/kvm/hyp/include/nvhe/gfp.h b/arch/arm64/kvm/hyp/include/nvhe/gfp.h index 0a048dc06a7d..fe5472a184a3 100644 --- a/arch/arm64/kvm/hyp/include/nvhe/gfp.h +++ b/arch/arm64/kvm/hyp/include/nvhe/gfp.h @@ -16,7 +16,7 @@ struct hyp_pool { * API at EL2. */ hyp_spinlock_t lock; - struct list_head free_area[MAX_ORDER]; + struct list_head free_area[MAX_ORDER + 1]; phys_addr_t range_start; phys_addr_t range_end; unsigned short max_order; diff --git a/arch/arm64/kvm/hyp/nvhe/page_alloc.c b/arch/arm64/kvm/hyp/nvhe/page_alloc.c index 803ba3222e75..b1e392186a0f 100644 --- a/arch/arm64/kvm/hyp/nvhe/page_alloc.c +++ b/arch/arm64/kvm/hyp/nvhe/page_alloc.c @@ -110,7 +110,7 @@ static void __hyp_attach_page(struct hyp_pool *pool, * after coalescing, so make sure to mark it HYP_NO_ORDER proactively. */ p->order = HYP_NO_ORDER; - for (; (order + 1) < pool->max_order; order++) { + for (; (order + 1) <= pool->max_order; order++) { buddy = __find_buddy_avail(pool, p, order); if (!buddy) break; @@ -203,9 +203,9 @@ void *hyp_alloc_pages(struct hyp_pool *pool, unsigned short order) hyp_spin_lock(&pool->lock); /* Look for a high-enough-order page */ - while (i < pool->max_order && list_empty(&pool->free_area[i])) + while (i <= pool->max_order && list_empty(&pool->free_area[i])) i++; - if (i >= pool->max_order) { + if (i > pool->max_order) { hyp_spin_unlock(&pool->lock); return NULL; } @@ -228,8 +228,8 @@ int hyp_pool_init(struct hyp_pool *pool, u64 pfn, unsigned int nr_pages, int i; hyp_spin_lock_init(&pool->lock); - pool->max_order = min(MAX_ORDER, get_order((nr_pages + 1) << PAGE_SHIFT)); - for (i = 0; i < pool->max_order; i++) + pool->max_order = min(MAX_ORDER, get_order(nr_pages << PAGE_SHIFT)); + for (i = 0; i <= pool->max_order; i++) INIT_LIST_HEAD(&pool->free_area[i]); pool->range_start = phys; pool->range_end = phys + (nr_pages << PAGE_SHIFT); diff --git a/arch/csky/Kconfig b/arch/csky/Kconfig index dba02da6fa34..c694fac43bed 100644 --- a/arch/csky/Kconfig +++ b/arch/csky/Kconfig @@ -334,7 +334,7 @@ config HIGHMEM config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - default "11" + default "10" config DRAM_BASE hex "DRAM start addr (the same with memory-section in dts)" diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index d7e4a24e8644..0d2f41fa56ee 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -202,10 +202,10 @@ config IA64_CYCLONE If you're unsure, answer N. config ARCH_FORCE_MAX_ORDER - int "MAX_ORDER (11 - 17)" if !HUGETLB_PAGE - range 11 17 if !HUGETLB_PAGE - default "17" if HUGETLB_PAGE - default "11" + int "MAX_ORDER (10 - 16)" if !HUGETLB_PAGE + range 10 16 if !HUGETLB_PAGE + default "16" if HUGETLB_PAGE + default "10" config SMP bool "Symmetric multi-processing support" diff --git a/arch/ia64/include/asm/sparsemem.h b/arch/ia64/include/asm/sparsemem.h index 84e8ce387b69..a58f8b466d96 100644 --- a/arch/ia64/include/asm/sparsemem.h +++ b/arch/ia64/include/asm/sparsemem.h @@ -12,9 +12,9 @@ #define SECTION_SIZE_BITS (30) #define MAX_PHYSMEM_BITS (50) #ifdef CONFIG_ARCH_FORCE_MAX_ORDER -#if ((CONFIG_ARCH_FORCE_MAX_ORDER - 1 + PAGE_SHIFT) > SECTION_SIZE_BITS) +#if (CONFIG_ARCH_FORCE_MAX_ORDER + PAGE_SHIFT > SECTION_SIZE_BITS) #undef SECTION_SIZE_BITS -#define SECTION_SIZE_BITS (CONFIG_ARCH_FORCE_MAX_ORDER - 1 + PAGE_SHIFT) +#define SECTION_SIZE_BITS (CONFIG_ARCH_FORCE_MAX_ORDER + PAGE_SHIFT) #endif #endif diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index 380d2f3966c9..e8dd4323fb86 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -170,7 +170,7 @@ static int __init hugetlb_setup_sz(char *str) size = memparse(str, &str); if (*str || !is_power_of_2(size) || !(tr_pages & size) || size <= PAGE_SIZE || - size >= (1UL << PAGE_SHIFT << MAX_ORDER)) { + size > (1UL << PAGE_SHIFT << MAX_ORDER)) { printk(KERN_WARNING "Invalid huge page size specified\n"); return 1; } diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index 7fd51257e0ed..272a3a12c98d 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -420,12 +420,12 @@ config NODES_SHIFT config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - range 14 64 if PAGE_SIZE_64KB - default "14" if PAGE_SIZE_64KB - range 12 64 if PAGE_SIZE_16KB - default "12" if PAGE_SIZE_16KB - range 11 64 - default "11" + range 13 63 if PAGE_SIZE_64KB + default "13" if PAGE_SIZE_64KB + range 11 63 if PAGE_SIZE_16KB + default "11" if PAGE_SIZE_16KB + range 10 63 + default "10" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -434,9 +434,6 @@ config ARCH_FORCE_MAX_ORDER blocks of physically contiguous memory, then you may need to increase this value. - This config option is actually maximum order plus one. For example, - a value of 11 means that the largest free memory block is 2^10 pages. - The page size is not necessarily 4KB. Keep this in mind when choosing a value for this option. diff --git a/arch/m68k/Kconfig.cpu b/arch/m68k/Kconfig.cpu index 9380f6e3bb66..c9df6572133f 100644 --- a/arch/m68k/Kconfig.cpu +++ b/arch/m68k/Kconfig.cpu @@ -400,7 +400,7 @@ config SINGLE_MEMORY_CHUNK config ARCH_FORCE_MAX_ORDER int "Maximum zone order" if ADVANCED depends on !SINGLE_MEMORY_CHUNK - default "11" + default "10" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -413,9 +413,6 @@ config ARCH_FORCE_MAX_ORDER value also defines the minimal size of the hole that allows freeing unused memory map. - This config option is actually maximum order plus one. For example, - a value of 11 means that the largest free memory block is 2^10 pages. - config 060_WRITETHROUGH bool "Use write-through caching for 68060 supervisor accesses" depends on ADVANCED && M68060 diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index e2f3ca73f40d..3e8b765b8c7b 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -2137,14 +2137,14 @@ endchoice config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - range 14 64 if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_64KB - default "14" if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_64KB - range 13 64 if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_32KB - default "13" if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_32KB - range 12 64 if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_16KB - default "12" if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_16KB - range 0 64 - default "11" + range 13 63 if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_64KB + default "13" if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_64KB + range 12 63 if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_32KB + default "12" if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_32KB + range 11 63 if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_16KB + default "11" if MIPS_HUGE_TLB_SUPPORT && PAGE_SIZE_16KB + range 0 63 + default "10" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -2153,9 +2153,6 @@ config ARCH_FORCE_MAX_ORDER blocks of physically contiguous memory, then you may need to increase this value. - This config option is actually maximum order plus one. For example, - a value of 11 means that the largest free memory block is 2^10 pages. - The page size is not necessarily 4KB. Keep this in mind when choosing a value for this option. diff --git a/arch/nios2/Kconfig b/arch/nios2/Kconfig index a582f72104f3..89708b95978c 100644 --- a/arch/nios2/Kconfig +++ b/arch/nios2/Kconfig @@ -46,8 +46,8 @@ source "kernel/Kconfig.hz" config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - range 9 20 - default "11" + range 8 19 + default "10" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -56,9 +56,6 @@ config ARCH_FORCE_MAX_ORDER blocks of physically contiguous memory, then you may need to increase this value. - This config option is actually maximum order plus one. For example, - a value of 11 means that the largest free memory block is 2^10 pages. - endmenu source "arch/nios2/platform/Kconfig.platform" diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 49c6d36b2b3e..24d56536b269 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -897,18 +897,18 @@ config DATA_SHIFT config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - range 8 9 if PPC64 && PPC_64K_PAGES - default "9" if PPC64 && PPC_64K_PAGES - range 13 13 if PPC64 && !PPC_64K_PAGES - default "13" if PPC64 && !PPC_64K_PAGES - range 9 64 if PPC32 && PPC_16K_PAGES - default "9" if PPC32 && PPC_16K_PAGES - range 7 64 if PPC32 && PPC_64K_PAGES - default "7" if PPC32 && PPC_64K_PAGES - range 5 64 if PPC32 && PPC_256K_PAGES - default "5" if PPC32 && PPC_256K_PAGES - range 11 64 - default "11" + range 7 8 if PPC64 && PPC_64K_PAGES + default "8" if PPC64 && PPC_64K_PAGES + range 12 12 if PPC64 && !PPC_64K_PAGES + default "12" if PPC64 && !PPC_64K_PAGES + range 8 63 if PPC32 && PPC_16K_PAGES + default "8" if PPC32 && PPC_16K_PAGES + range 6 63 if PPC32 && PPC_64K_PAGES + default "6" if PPC32 && PPC_64K_PAGES + range 4 63 if PPC32 && PPC_256K_PAGES + default "4" if PPC32 && PPC_256K_PAGES + range 10 63 + default "10" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -917,9 +917,6 @@ config ARCH_FORCE_MAX_ORDER blocks of physically contiguous memory, then you may need to increase this value. - This config option is actually maximum order plus one. For example, - a value of 11 means that the largest free memory block is 2^10 pages. - The page size is not necessarily 4KB. For example, on 64-bit systems, 64KB pages can be enabled via CONFIG_PPC_64K_PAGES. Keep this in mind when choosing a value for this option. diff --git a/arch/powerpc/configs/85xx/ge_imp3a_defconfig b/arch/powerpc/configs/85xx/ge_imp3a_defconfig index ea719898b581..6cb7e90d52c1 100644 --- a/arch/powerpc/configs/85xx/ge_imp3a_defconfig +++ b/arch/powerpc/configs/85xx/ge_imp3a_defconfig @@ -30,7 +30,7 @@ CONFIG_PREEMPT=y # CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set CONFIG_BINFMT_MISC=m CONFIG_MATH_EMULATION=y -CONFIG_ARCH_FORCE_MAX_ORDER=17 +CONFIG_ARCH_FORCE_MAX_ORDER=16 CONFIG_PCI=y CONFIG_PCIEPORTBUS=y CONFIG_PCI_MSI=y diff --git a/arch/powerpc/configs/fsl-emb-nonhw.config b/arch/powerpc/configs/fsl-emb-nonhw.config index ab8a8c4530d9..3009b0efaf34 100644 --- a/arch/powerpc/configs/fsl-emb-nonhw.config +++ b/arch/powerpc/configs/fsl-emb-nonhw.config @@ -41,7 +41,7 @@ CONFIG_FIXED_PHY=y CONFIG_FONT_8x16=y CONFIG_FONT_8x8=y CONFIG_FONTS=y -CONFIG_ARCH_FORCE_MAX_ORDER=13 +CONFIG_ARCH_FORCE_MAX_ORDER=12 CONFIG_FRAMEBUFFER_CONSOLE=y CONFIG_FRAME_WARN=1024 CONFIG_FTL=y diff --git a/arch/powerpc/mm/book3s64/iommu_api.c b/arch/powerpc/mm/book3s64/iommu_api.c index 7fcfba162e0d..81d7185e2ae8 100644 --- a/arch/powerpc/mm/book3s64/iommu_api.c +++ b/arch/powerpc/mm/book3s64/iommu_api.c @@ -97,7 +97,7 @@ static long mm_iommu_do_alloc(struct mm_struct *mm, unsigned long ua, } mmap_read_lock(mm); - chunk = (1UL << (PAGE_SHIFT + MAX_ORDER - 1)) / + chunk = (1UL << (PAGE_SHIFT + MAX_ORDER)) / sizeof(struct vm_area_struct *); chunk = min(chunk, entries); for (entry = 0; entry < entries; entry += chunk) { diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index f1ba8d1e8c1a..b900933507da 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -615,7 +615,7 @@ void __init gigantic_hugetlb_cma_reserve(void) order = mmu_psize_to_shift(MMU_PAGE_16G) - PAGE_SHIFT; if (order) { - VM_WARN_ON(order < MAX_ORDER); + VM_WARN_ON(order <= MAX_ORDER); hugetlb_cma_reserve(order); } } diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index 4f6e20a35aa1..5a81f106068e 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -1740,7 +1740,7 @@ static long pnv_pci_ioda2_setup_default_config(struct pnv_ioda_pe *pe) * DMA window can be larger than available memory, which will * cause errors later. */ - const u64 maxblock = 1UL << (PAGE_SHIFT + MAX_ORDER - 1); + const u64 maxblock = 1UL << (PAGE_SHIFT + MAX_ORDER); /* * We create the default window as big as we can. The constraint is diff --git a/arch/sh/configs/ecovec24_defconfig b/arch/sh/configs/ecovec24_defconfig index b52e14ccb450..4d655e8d4d74 100644 --- a/arch/sh/configs/ecovec24_defconfig +++ b/arch/sh/configs/ecovec24_defconfig @@ -8,7 +8,7 @@ CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y # CONFIG_BLK_DEV_BSG is not set CONFIG_CPU_SUBTYPE_SH7724=y -CONFIG_ARCH_FORCE_MAX_ORDER=12 +CONFIG_ARCH_FORCE_MAX_ORDER=11 CONFIG_MEMORY_SIZE=0x10000000 CONFIG_FLATMEM_MANUAL=y CONFIG_SH_ECOVEC=y diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig index 411fdc0901f7..40271090bd7d 100644 --- a/arch/sh/mm/Kconfig +++ b/arch/sh/mm/Kconfig @@ -20,13 +20,13 @@ config PAGE_OFFSET config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - range 9 64 if PAGE_SIZE_16KB - default "9" if PAGE_SIZE_16KB - range 7 64 if PAGE_SIZE_64KB - default "7" if PAGE_SIZE_64KB - range 11 64 - default "14" if !MMU - default "11" + range 8 63 if PAGE_SIZE_16KB + default "8" if PAGE_SIZE_16KB + range 6 63 if PAGE_SIZE_64KB + default "6" if PAGE_SIZE_64KB + range 10 63 + default "13" if !MMU + default "10" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -35,9 +35,6 @@ config ARCH_FORCE_MAX_ORDER blocks of physically contiguous memory, then you may need to increase this value. - This config option is actually maximum order plus one. For example, - a value of 11 means that the largest free memory block is 2^10 pages. - The page size is not necessarily 4KB. Keep this in mind when choosing a value for this option. diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 84437a4c6545..e3242bf5a8df 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -271,7 +271,7 @@ config ARCH_SPARSEMEM_DEFAULT config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - default "13" + default "12" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -280,9 +280,6 @@ config ARCH_FORCE_MAX_ORDER blocks of physically contiguous memory, then you may need to increase this value. - This config option is actually maximum order plus one. For example, - a value of 13 means that the largest free memory block is 2^12 pages. - if SPARC64 || COMPILE_TEST source "kernel/power/Kconfig" endif diff --git a/arch/sparc/kernel/pci_sun4v.c b/arch/sparc/kernel/pci_sun4v.c index 384480971805..7d91ca6aa675 100644 --- a/arch/sparc/kernel/pci_sun4v.c +++ b/arch/sparc/kernel/pci_sun4v.c @@ -193,7 +193,7 @@ static void *dma_4v_alloc_coherent(struct device *dev, size_t size, size = IO_PAGE_ALIGN(size); order = get_order(size); - if (unlikely(order >= MAX_ORDER)) + if (unlikely(order > MAX_ORDER)) return NULL; npages = size >> IO_PAGE_SHIFT; diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c index 5b4de4a89dec..08ffd17d5ec3 100644 --- a/arch/sparc/kernel/traps_64.c +++ b/arch/sparc/kernel/traps_64.c @@ -897,7 +897,7 @@ void __init cheetah_ecache_flush_init(void) /* Now allocate error trap reporting scoreboard. */ sz = NR_CPUS * (2 * sizeof(struct cheetah_err_info)); - for (order = 0; order < MAX_ORDER; order++) { + for (order = 0; order <= MAX_ORDER; order++) { if ((PAGE_SIZE << order) >= sz) break; } diff --git a/arch/sparc/mm/tsb.c b/arch/sparc/mm/tsb.c index dba8dffe2113..5e2931a18409 100644 --- a/arch/sparc/mm/tsb.c +++ b/arch/sparc/mm/tsb.c @@ -402,8 +402,8 @@ void tsb_grow(struct mm_struct *mm, unsigned long tsb_index, unsigned long rss) unsigned long new_rss_limit; gfp_t gfp_flags; - if (max_tsb_size > (PAGE_SIZE << (MAX_ORDER - 1))) - max_tsb_size = (PAGE_SIZE << (MAX_ORDER - 1)); + if (max_tsb_size > PAGE_SIZE << MAX_ORDER) + max_tsb_size = PAGE_SIZE << MAX_ORDER; new_cache_index = 0; for (new_size = 8192; new_size < max_tsb_size; new_size <<= 1UL) { diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index 5e5a9c8e0e5d..8dcda617b8bf 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -368,10 +368,10 @@ int __init linux_main(int argc, char **argv) max_physmem = TASK_SIZE - uml_physmem - iomem_size - MIN_VMALLOC; /* - * Zones have to begin on a 1 << MAX_ORDER-1 page boundary, + * Zones have to begin on a 1 << MAX_ORDER page boundary, * so this makes sure that's true for highmem */ - max_physmem &= ~((1 << (PAGE_SHIFT + MAX_ORDER - 1)) - 1); + max_physmem &= ~((1 << (PAGE_SHIFT + MAX_ORDER)) - 1); if (physmem_size + iomem_size > max_physmem) { highmem = physmem_size + iomem_size - max_physmem; physmem_size -= highmem; diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig index bcb0c5d2abc2..3eee334ba873 100644 --- a/arch/xtensa/Kconfig +++ b/arch/xtensa/Kconfig @@ -773,7 +773,7 @@ config HIGHMEM config ARCH_FORCE_MAX_ORDER int "Maximum zone order" - default "11" + default "10" help The kernel memory allocator divides physically contiguous memory blocks into "zones", where each zone is a power of two number of @@ -782,9 +782,6 @@ config ARCH_FORCE_MAX_ORDER blocks of physically contiguous memory, then you may need to increase this value. - This config option is actually maximum order plus one. For example, - a value of 11 means that the largest free memory block is 2^10 pages. - endmenu menu "Power management options" diff --git a/drivers/base/regmap/regmap-debugfs.c b/drivers/base/regmap/regmap-debugfs.c index 817eda2075aa..c491fabe3617 100644 --- a/drivers/base/regmap/regmap-debugfs.c +++ b/drivers/base/regmap/regmap-debugfs.c @@ -226,8 +226,8 @@ static ssize_t regmap_read_debugfs(struct regmap *map, unsigned int from, if (*ppos < 0 || !count) return -EINVAL; - if (count > (PAGE_SIZE << (MAX_ORDER - 1))) - count = PAGE_SIZE << (MAX_ORDER - 1); + if (count > (PAGE_SIZE << MAX_ORDER)) + count = PAGE_SIZE << MAX_ORDER; buf = kmalloc(count, GFP_KERNEL); if (!buf) @@ -373,8 +373,8 @@ static ssize_t regmap_reg_ranges_read_file(struct file *file, if (*ppos < 0 || !count) return -EINVAL; - if (count > (PAGE_SIZE << (MAX_ORDER - 1))) - count = PAGE_SIZE << (MAX_ORDER - 1); + if (count > (PAGE_SIZE << MAX_ORDER)) + count = PAGE_SIZE << MAX_ORDER; buf = kmalloc(count, GFP_KERNEL); if (!buf) diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 90d2dfb6448e..cec2c20f5e59 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -3079,7 +3079,7 @@ static void raw_cmd_free(struct floppy_raw_cmd **ptr) } } -#define MAX_LEN (1UL << (MAX_ORDER - 1) << PAGE_SHIFT) +#define MAX_LEN (1UL << MAX_ORDER << PAGE_SHIFT) static int raw_cmd_copyin(int cmd, void __user *param, struct floppy_raw_cmd **rcmd) diff --git a/drivers/crypto/ccp/sev-dev.c b/drivers/crypto/ccp/sev-dev.c index e2f25926eb51..bf095baca244 100644 --- a/drivers/crypto/ccp/sev-dev.c +++ b/drivers/crypto/ccp/sev-dev.c @@ -886,7 +886,7 @@ static int sev_ioctl_do_get_id2(struct sev_issue_cmd *argp) /* * The length of the ID shouldn't be assumed by software since * it may change in the future. The allocation size is limited - * to 1 << (PAGE_SHIFT + MAX_ORDER - 1) by the page allocator. + * to 1 << (PAGE_SHIFT + MAX_ORDER) by the page allocator. * If the allocation fails, simply return ENOMEM rather than * warning in the kernel log. */ diff --git a/drivers/crypto/hisilicon/sgl.c b/drivers/crypto/hisilicon/sgl.c index 09586a837b1e..3df7a256e919 100644 --- a/drivers/crypto/hisilicon/sgl.c +++ b/drivers/crypto/hisilicon/sgl.c @@ -70,11 +70,11 @@ struct hisi_acc_sgl_pool *hisi_acc_create_sgl_pool(struct device *dev, HISI_ACC_SGL_ALIGN_SIZE); /* - * the pool may allocate a block of memory of size PAGE_SIZE * 2^(MAX_ORDER - 1), + * the pool may allocate a block of memory of size PAGE_SIZE * 2^MAX_ORDER, * block size may exceed 2^31 on ia64, so the max of block size is 2^31 */ - block_size = 1 << (PAGE_SHIFT + MAX_ORDER <= 32 ? - PAGE_SHIFT + MAX_ORDER - 1 : 31); + block_size = 1 << (PAGE_SHIFT + MAX_ORDER < 32 ? + PAGE_SHIFT + MAX_ORDER : 31); sgl_num_per_block = block_size / sgl_size; block_num = count / sgl_num_per_block; remain_sgl = count % sgl_num_per_block; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_internal.c b/drivers/gpu/drm/i915/gem/i915_gem_internal.c index eae9e9f6d3bf..6bc26b4b06b8 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_internal.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_internal.c @@ -36,7 +36,7 @@ static int i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj) struct sg_table *st; struct scatterlist *sg; unsigned int npages; /* restricted by sg_alloc_table */ - int max_order = MAX_ORDER - 1; + int max_order = MAX_ORDER; unsigned int max_segment; gfp_t gfp; diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c index defece0bcb81..99f39a5feca1 100644 --- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c +++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c @@ -115,7 +115,7 @@ static int get_huge_pages(struct drm_i915_gem_object *obj) do { struct page *page; - GEM_BUG_ON(order >= MAX_ORDER); + GEM_BUG_ON(order > MAX_ORDER); page = alloc_pages(GFP | __GFP_ZERO, order); if (!page) goto err; diff --git a/drivers/gpu/drm/ttm/ttm_pool.c b/drivers/gpu/drm/ttm/ttm_pool.c index aa116a7bbae3..6c8585abe08d 100644 --- a/drivers/gpu/drm/ttm/ttm_pool.c +++ b/drivers/gpu/drm/ttm/ttm_pool.c @@ -65,11 +65,11 @@ module_param(page_pool_size, ulong, 0644); static atomic_long_t allocated_pages; -static struct ttm_pool_type global_write_combined[MAX_ORDER]; -static struct ttm_pool_type global_uncached[MAX_ORDER]; +static struct ttm_pool_type global_write_combined[MAX_ORDER + 1]; +static struct ttm_pool_type global_uncached[MAX_ORDER + 1]; -static struct ttm_pool_type global_dma32_write_combined[MAX_ORDER]; -static struct ttm_pool_type global_dma32_uncached[MAX_ORDER]; +static struct ttm_pool_type global_dma32_write_combined[MAX_ORDER + 1]; +static struct ttm_pool_type global_dma32_uncached[MAX_ORDER + 1]; static spinlock_t shrinker_lock; static struct list_head shrinker_list; @@ -405,7 +405,7 @@ int ttm_pool_alloc(struct ttm_pool *pool, struct ttm_tt *tt, else gfp_flags |= GFP_HIGHUSER; - for (order = min_t(unsigned int, MAX_ORDER - 1, __fls(num_pages)); + for (order = min_t(unsigned int, MAX_ORDER, __fls(num_pages)); num_pages; order = min_t(unsigned int, order, __fls(num_pages))) { struct ttm_pool_type *pt; @@ -542,7 +542,7 @@ void ttm_pool_init(struct ttm_pool *pool, struct device *dev, if (use_dma_alloc) { for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i) - for (j = 0; j < MAX_ORDER; ++j) + for (j = 0; j <= MAX_ORDER; ++j) ttm_pool_type_init(&pool->caching[i].orders[j], pool, i, j); } @@ -562,7 +562,7 @@ void ttm_pool_fini(struct ttm_pool *pool) if (pool->use_dma_alloc) { for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i) - for (j = 0; j < MAX_ORDER; ++j) + for (j = 0; j <= MAX_ORDER; ++j) ttm_pool_type_fini(&pool->caching[i].orders[j]); } @@ -616,7 +616,7 @@ static void ttm_pool_debugfs_header(struct seq_file *m) unsigned int i; seq_puts(m, "\t "); - for (i = 0; i < MAX_ORDER; ++i) + for (i = 0; i <= MAX_ORDER; ++i) seq_printf(m, " ---%2u---", i); seq_puts(m, "\n"); } @@ -627,7 +627,7 @@ static void ttm_pool_debugfs_orders(struct ttm_pool_type *pt, { unsigned int i; - for (i = 0; i < MAX_ORDER; ++i) + for (i = 0; i <= MAX_ORDER; ++i) seq_printf(m, " %8u", ttm_pool_type_count(&pt[i])); seq_puts(m, "\n"); } @@ -736,7 +736,7 @@ int ttm_pool_mgr_init(unsigned long num_pages) spin_lock_init(&shrinker_lock); INIT_LIST_HEAD(&shrinker_list); - for (i = 0; i < MAX_ORDER; ++i) { + for (i = 0; i <= MAX_ORDER; ++i) { ttm_pool_type_init(&global_write_combined[i], NULL, ttm_write_combined, i); ttm_pool_type_init(&global_uncached[i], NULL, ttm_uncached, i); @@ -769,7 +769,7 @@ void ttm_pool_mgr_fini(void) { unsigned int i; - for (i = 0; i < MAX_ORDER; ++i) { + for (i = 0; i <= MAX_ORDER; ++i) { ttm_pool_type_fini(&global_write_combined[i]); ttm_pool_type_fini(&global_uncached[i]); diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h index 8d772ea8a583..b574c58a3487 100644 --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h @@ -182,7 +182,7 @@ #ifdef CONFIG_CMA_ALIGNMENT #define Q_MAX_SZ_SHIFT (PAGE_SHIFT + CONFIG_CMA_ALIGNMENT) #else -#define Q_MAX_SZ_SHIFT (PAGE_SHIFT + MAX_ORDER - 1) +#define Q_MAX_SZ_SHIFT (PAGE_SHIFT + MAX_ORDER) #endif /* diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c index ac996fd6bd9c..7a9f0b0bddbd 100644 --- a/drivers/iommu/dma-iommu.c +++ b/drivers/iommu/dma-iommu.c @@ -736,7 +736,7 @@ static struct page **__iommu_dma_alloc_pages(struct device *dev, struct page **pages; unsigned int i = 0, nid = dev_to_node(dev); - order_mask &= GENMASK(MAX_ORDER - 1, 0); + order_mask &= GENMASK(MAX_ORDER, 0); if (!order_mask) return NULL; diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index 586271b8aa39..85790b870877 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -2440,8 +2440,8 @@ static bool its_parse_indirect_baser(struct its_node *its, * feature is not supported by hardware. */ new_order = max_t(u32, get_order(esz << ids), new_order); - if (new_order >= MAX_ORDER) { - new_order = MAX_ORDER - 1; + if (new_order > MAX_ORDER) { + new_order = MAX_ORDER; ids = ilog2(PAGE_ORDER_TO_SIZE(new_order) / (int)esz); pr_warn("ITS@%pa: %s Table too large, reduce ids %llu->%u\n", &its->phys_base, its_base_type_string[type], diff --git a/drivers/md/dm-bufio.c b/drivers/md/dm-bufio.c index cf077f9b30c3..733053c2eaa0 100644 --- a/drivers/md/dm-bufio.c +++ b/drivers/md/dm-bufio.c @@ -408,7 +408,7 @@ static void __cache_size_refresh(void) * If the allocation may fail we use __get_free_pages. Memory fragmentation * won't have a fatal effect here, but it just causes flushes of some other * buffers and more I/O will be performed. Don't use __get_free_pages if it - * always fails (i.e. order >= MAX_ORDER). + * always fails (i.e. order > MAX_ORDER). * * If the allocation shouldn't fail we use __vmalloc. This is only for the * initial reserve allocation, so there's no risk of wasting all vmalloc diff --git a/drivers/misc/genwqe/card_dev.c b/drivers/misc/genwqe/card_dev.c index d0e27438a73c..55fc5b80e649 100644 --- a/drivers/misc/genwqe/card_dev.c +++ b/drivers/misc/genwqe/card_dev.c @@ -443,7 +443,7 @@ static int genwqe_mmap(struct file *filp, struct vm_area_struct *vma) if (vsize == 0) return -EINVAL; - if (get_order(vsize) >= MAX_ORDER) + if (get_order(vsize) > MAX_ORDER) return -ENOMEM; dma_map = kzalloc(sizeof(struct dma_mapping), GFP_KERNEL); diff --git a/drivers/misc/genwqe/card_utils.c b/drivers/misc/genwqe/card_utils.c index ac29698d085a..1c798d6b2dfb 100644 --- a/drivers/misc/genwqe/card_utils.c +++ b/drivers/misc/genwqe/card_utils.c @@ -210,7 +210,7 @@ u32 genwqe_crc32(u8 *buff, size_t len, u32 init) void *__genwqe_alloc_consistent(struct genwqe_dev *cd, size_t size, dma_addr_t *dma_handle) { - if (get_order(size) >= MAX_ORDER) + if (get_order(size) > MAX_ORDER) return NULL; return dma_alloc_coherent(&cd->pci_dev->dev, size, dma_handle, @@ -308,7 +308,7 @@ int genwqe_alloc_sync_sgl(struct genwqe_dev *cd, struct genwqe_sgl *sgl, sgl->write = write; sgl->sgl_size = genwqe_sgl_size(sgl->nr_pages); - if (get_order(sgl->sgl_size) >= MAX_ORDER) { + if (get_order(sgl->sgl_size) > MAX_ORDER) { dev_err(&pci_dev->dev, "[%s] err: too much memory requested!\n", __func__); return ret; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index 25be7f8ac7cd..3973ca6adf4c 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -1041,7 +1041,7 @@ static void hns3_init_tx_spare_buffer(struct hns3_enet_ring *ring) return; order = get_order(alloc_size); - if (order >= MAX_ORDER) { + if (order > MAX_ORDER) { if (net_ratelimit()) dev_warn(ring_to_dev(ring), "failed to allocate tx spare buffer, exceed to max order\n"); return; diff --git a/drivers/net/ethernet/ibm/ibmvnic.h b/drivers/net/ethernet/ibm/ibmvnic.h index b35c9b6f913b..4e18b4cefa97 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.h +++ b/drivers/net/ethernet/ibm/ibmvnic.h @@ -75,7 +75,7 @@ * pool for the 4MB. Thus the 16 Rx and Tx queues require 32 * 5 = 160 * plus 16 for the TSO pools for a total of 176 LTB mappings per VNIC. */ -#define IBMVNIC_ONE_LTB_MAX ((u32)((1 << (MAX_ORDER - 1)) * PAGE_SIZE)) +#define IBMVNIC_ONE_LTB_MAX ((u32)((1 << MAX_ORDER) * PAGE_SIZE)) #define IBMVNIC_ONE_LTB_SIZE min((u32)(8 << 20), IBMVNIC_ONE_LTB_MAX) #define IBMVNIC_LTB_SET_SIZE (38 << 20) diff --git a/drivers/video/fbdev/hyperv_fb.c b/drivers/video/fbdev/hyperv_fb.c index ec3f6cf05f8c..34781dec3856 100644 --- a/drivers/video/fbdev/hyperv_fb.c +++ b/drivers/video/fbdev/hyperv_fb.c @@ -946,7 +946,7 @@ static phys_addr_t hvfb_get_phymem(struct hv_device *hdev, if (request_size == 0) return -1; - if (order < MAX_ORDER) { + if (order <= MAX_ORDER) { /* Call alloc_pages if the size is less than 2^MAX_ORDER */ page = alloc_pages(GFP_KERNEL | __GFP_ZERO, order); if (!page) @@ -977,7 +977,7 @@ static void hvfb_release_phymem(struct hv_device *hdev, { unsigned int order = get_order(size); - if (order < MAX_ORDER) + if (order <= MAX_ORDER) __free_pages(pfn_to_page(paddr >> PAGE_SHIFT), order); else dma_free_coherent(&hdev->device, diff --git a/drivers/video/fbdev/vermilion/vermilion.c b/drivers/video/fbdev/vermilion/vermilion.c index 0374ee6b6d03..32e74e02a02f 100644 --- a/drivers/video/fbdev/vermilion/vermilion.c +++ b/drivers/video/fbdev/vermilion/vermilion.c @@ -197,7 +197,7 @@ static int vmlfb_alloc_vram(struct vml_info *vinfo, va = &vinfo->vram[i]; order = 0; - while (requested > (PAGE_SIZE << order) && order < MAX_ORDER) + while (requested > (PAGE_SIZE << order) && order <= MAX_ORDER) order++; err = vmlfb_alloc_vram_area(va, order, 0); diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c index 3f78a3a1eb75..5b15936a5214 100644 --- a/drivers/virtio/virtio_balloon.c +++ b/drivers/virtio/virtio_balloon.c @@ -33,7 +33,7 @@ #define VIRTIO_BALLOON_FREE_PAGE_ALLOC_FLAG (__GFP_NORETRY | __GFP_NOWARN | \ __GFP_NOMEMALLOC) /* The order of free page blocks to report to host */ -#define VIRTIO_BALLOON_HINT_BLOCK_ORDER (MAX_ORDER - 1) +#define VIRTIO_BALLOON_HINT_BLOCK_ORDER MAX_ORDER /* The size of a free page block in bytes */ #define VIRTIO_BALLOON_HINT_BLOCK_BYTES \ (1 << (VIRTIO_BALLOON_HINT_BLOCK_ORDER + PAGE_SHIFT)) diff --git a/drivers/virtio/virtio_mem.c b/drivers/virtio/virtio_mem.c index 0c2892ec6817..835f6cc2fb66 100644 --- a/drivers/virtio/virtio_mem.c +++ b/drivers/virtio/virtio_mem.c @@ -1120,13 +1120,13 @@ static void virtio_mem_clear_fake_offline(unsigned long pfn, */ static void virtio_mem_fake_online(unsigned long pfn, unsigned long nr_pages) { - unsigned long order = MAX_ORDER - 1; + unsigned long order = MAX_ORDER; unsigned long i; /* * We might get called for ranges that don't cover properly aligned - * MAX_ORDER - 1 pages; however, we can only online properly aligned - * pages with an order of MAX_ORDER - 1 at maximum. + * MAX_ORDER pages; however, we can only online properly aligned + * pages with an order of MAX_ORDER at maximum. */ while (!IS_ALIGNED(pfn | nr_pages, 1 << order)) order--; @@ -1237,9 +1237,9 @@ static void virtio_mem_online_page(struct virtio_mem *vm, bool do_online; /* - * We can get called with any order up to MAX_ORDER - 1. If our - * subblock size is smaller than that and we have a mixture of plugged - * and unplugged subblocks within such a page, we have to process in + * We can get called with any order up to MAX_ORDER. If our subblock + * size is smaller than that and we have a mixture of plugged and + * unplugged subblocks within such a page, we have to process in * smaller granularity. In that case we'll adjust the order exactly once * within the loop. */ diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c index 2f67516bb9bf..9fbb9b5256f7 100644 --- a/fs/ramfs/file-nommu.c +++ b/fs/ramfs/file-nommu.c @@ -70,7 +70,7 @@ int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize) /* make various checks */ order = get_order(newsize); - if (unlikely(order >= MAX_ORDER)) + if (unlikely(order > MAX_ORDER)) return -EFBIG; ret = inode_newsize_ok(inode, newsize); diff --git a/include/drm/ttm/ttm_pool.h b/include/drm/ttm/ttm_pool.h index ef09b23d29e3..8ce14f9d202a 100644 --- a/include/drm/ttm/ttm_pool.h +++ b/include/drm/ttm/ttm_pool.h @@ -72,7 +72,7 @@ struct ttm_pool { bool use_dma32; struct { - struct ttm_pool_type orders[MAX_ORDER]; + struct ttm_pool_type orders[MAX_ORDER + 1]; } caching[TTM_NUM_CACHING_TYPES]; }; diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 7c977d234aba..8fb7d91cd0b1 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -818,7 +818,7 @@ static inline unsigned huge_page_shift(struct hstate *h) static inline bool hstate_is_gigantic(struct hstate *h) { - return huge_page_order(h) >= MAX_ORDER; + return huge_page_order(h) > MAX_ORDER; } static inline unsigned int pages_per_huge_page(const struct hstate *h) diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index bf8786d45b31..35b11cc210d9 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -26,11 +26,11 @@ /* Free memory management - zoned buddy allocator. */ #ifndef CONFIG_ARCH_FORCE_MAX_ORDER -#define MAX_ORDER 11 +#define MAX_ORDER 10 #else #define MAX_ORDER CONFIG_ARCH_FORCE_MAX_ORDER #endif -#define MAX_ORDER_NR_PAGES (1 << (MAX_ORDER - 1)) +#define MAX_ORDER_NR_PAGES (1 << MAX_ORDER) /* * PAGE_ALLOC_COSTLY_ORDER is the order at which allocations are deemed @@ -93,7 +93,7 @@ static inline bool migratetype_is_mergeable(int mt) } #define for_each_migratetype_order(order, type) \ - for (order = 0; order < MAX_ORDER; order++) \ + for (order = 0; order <= MAX_ORDER; order++) \ for (type = 0; type < MIGRATE_TYPES; type++) extern int page_group_by_mobility_disabled; @@ -922,7 +922,7 @@ struct zone { CACHELINE_PADDING(_pad1_); /* free areas of different sizes */ - struct free_area free_area[MAX_ORDER]; + struct free_area free_area[MAX_ORDER + 1]; /* zone flags, see below */ unsigned long flags; @@ -1745,7 +1745,7 @@ static inline bool movable_only_nodes(nodemask_t *nodes) #define SECTION_BLOCKFLAGS_BITS \ ((1UL << (PFN_SECTION_SHIFT - pageblock_order)) * NR_PAGEBLOCK_BITS) -#if (MAX_ORDER - 1 + PAGE_SHIFT) > SECTION_SIZE_BITS +#if (MAX_ORDER + PAGE_SHIFT) > SECTION_SIZE_BITS #error Allocator MAX_ORDER exceeds SECTION_SIZE #endif diff --git a/include/linux/pageblock-flags.h b/include/linux/pageblock-flags.h index 5f1ae07d724b..e83c4c095041 100644 --- a/include/linux/pageblock-flags.h +++ b/include/linux/pageblock-flags.h @@ -41,14 +41,14 @@ extern unsigned int pageblock_order; * Huge pages are a constant size, but don't exceed the maximum allocation * granularity. */ -#define pageblock_order min_t(unsigned int, HUGETLB_PAGE_ORDER, MAX_ORDER - 1) +#define pageblock_order min_t(unsigned int, HUGETLB_PAGE_ORDER, MAX_ORDER) #endif /* CONFIG_HUGETLB_PAGE_SIZE_VARIABLE */ #else /* CONFIG_HUGETLB_PAGE */ /* If huge pages are not used, group by MAX_ORDER_NR_PAGES */ -#define pageblock_order (MAX_ORDER-1) +#define pageblock_order MAX_ORDER #endif /* CONFIG_HUGETLB_PAGE */ diff --git a/include/linux/slab.h b/include/linux/slab.h index 45af70315a94..aa4575ef2965 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -284,7 +284,7 @@ static inline unsigned int arch_slab_minalign(void) * (PAGE_SIZE*2). Larger requests are passed to the page allocator. */ #define KMALLOC_SHIFT_HIGH (PAGE_SHIFT + 1) -#define KMALLOC_SHIFT_MAX (MAX_ORDER + PAGE_SHIFT - 1) +#define KMALLOC_SHIFT_MAX (MAX_ORDER + PAGE_SHIFT) #ifndef KMALLOC_SHIFT_LOW #define KMALLOC_SHIFT_LOW 5 #endif @@ -292,7 +292,7 @@ static inline unsigned int arch_slab_minalign(void) #ifdef CONFIG_SLUB #define KMALLOC_SHIFT_HIGH (PAGE_SHIFT + 1) -#define KMALLOC_SHIFT_MAX (MAX_ORDER + PAGE_SHIFT - 1) +#define KMALLOC_SHIFT_MAX (MAX_ORDER + PAGE_SHIFT) #ifndef KMALLOC_SHIFT_LOW #define KMALLOC_SHIFT_LOW 3 #endif @@ -305,7 +305,7 @@ static inline unsigned int arch_slab_minalign(void) * be allocated from the same page. */ #define KMALLOC_SHIFT_HIGH PAGE_SHIFT -#define KMALLOC_SHIFT_MAX (MAX_ORDER + PAGE_SHIFT - 1) +#define KMALLOC_SHIFT_MAX (MAX_ORDER + PAGE_SHIFT) #ifndef KMALLOC_SHIFT_LOW #define KMALLOC_SHIFT_LOW 3 #endif diff --git a/kernel/crash_core.c b/kernel/crash_core.c index 755f5f08ab38..90ce1dfd591c 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -474,7 +474,7 @@ static int __init crash_save_vmcoreinfo_init(void) VMCOREINFO_OFFSET(list_head, prev); VMCOREINFO_OFFSET(vmap_area, va_start); VMCOREINFO_OFFSET(vmap_area, list); - VMCOREINFO_LENGTH(zone.free_area, MAX_ORDER); + VMCOREINFO_LENGTH(zone.free_area, MAX_ORDER + 1); log_buf_vmcoreinfo_setup(); VMCOREINFO_LENGTH(free_area.free_list, MIGRATE_TYPES); VMCOREINFO_NUMBER(NR_FREE_PAGES); diff --git a/kernel/dma/pool.c b/kernel/dma/pool.c index 4d40dcce7604..1acec2e22827 100644 --- a/kernel/dma/pool.c +++ b/kernel/dma/pool.c @@ -84,8 +84,8 @@ static int atomic_pool_expand(struct gen_pool *pool, size_t pool_size, void *addr; int ret = -ENOMEM; - /* Cannot allocate larger than MAX_ORDER-1 */ - order = min(get_order(pool_size), MAX_ORDER-1); + /* Cannot allocate larger than MAX_ORDER */ + order = min(get_order(pool_size), MAX_ORDER); do { pool_size = 1 << (PAGE_SHIFT + order); @@ -190,7 +190,7 @@ static int __init dma_atomic_pool_init(void) /* * If coherent_pool was not used on the command line, default the pool - * sizes to 128KB per 1GB of memory, min 128KB, max MAX_ORDER-1. + * sizes to 128KB per 1GB of memory, min 128KB, max MAX_ORDER. */ if (!atomic_pool_size) { unsigned long pages = totalram_pages() / (SZ_1G / SZ_128K); diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index d6bbdb7830b2..a0433f37b024 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -609,8 +609,8 @@ static struct page *rb_alloc_aux_page(int node, int order) { struct page *page; - if (order >= MAX_ORDER) - order = MAX_ORDER - 1; + if (order > MAX_ORDER) + order = MAX_ORDER; do { page = alloc_pages_node(node, PERF_AUX_GFP, order); @@ -814,7 +814,7 @@ struct perf_buffer *rb_alloc(int nr_pages, long watermark, int cpu, int flags) size = sizeof(struct perf_buffer); size += nr_pages * sizeof(void *); - if (order_base_2(size) >= PAGE_SHIFT+MAX_ORDER) + if (order_base_2(size) > PAGE_SHIFT+MAX_ORDER) goto fail; node = (cpu == -1) ? cpu : cpu_to_node(cpu); diff --git a/mm/Kconfig b/mm/Kconfig index ca98b2072df5..969286ab14a1 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -346,9 +346,9 @@ config SHUFFLE_PAGE_ALLOCATOR the presence of a memory-side-cache. There are also incidental security benefits as it reduces the predictability of page allocations to compliment SLAB_FREELIST_RANDOM, but the - default granularity of shuffling on the "MAX_ORDER - 1" i.e, - 10th order of pages is selected based on cache utilization - benefits on x86. + default granularity of shuffling on the MAX_ORDER i.e, 10th + order of pages is selected based on cache utilization benefits + on x86. While the randomization improves cache utilization it may negatively impact workloads on platforms without a cache. For @@ -666,8 +666,8 @@ config HUGETLB_PAGE_SIZE_VARIABLE HUGETLB_PAGE_ORDER when there are multiple HugeTLB page sizes available on a platform. - Note that the pageblock_order cannot exceed MAX_ORDER - 1 and will be - clamped down to MAX_ORDER - 1. + Note that the pageblock_order cannot exceed MAX_ORDER and will be + clamped down to MAX_ORDER. config CONTIG_ALLOC def_bool (MEMORY_ISOLATION && COMPACTION) || CMA diff --git a/mm/compaction.c b/mm/compaction.c index 5a9501e0ae01..709136556b9e 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -583,7 +583,7 @@ static unsigned long isolate_freepages_block(struct compact_control *cc, if (PageCompound(page)) { const unsigned int order = compound_order(page); - if (likely(order < MAX_ORDER)) { + if (likely(order <= MAX_ORDER)) { blockpfn += (1UL << order) - 1; cursor += (1UL << order) - 1; } @@ -938,7 +938,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn, * a valid page order. Consider only values in the * valid order range to prevent low_pfn overflow. */ - if (freepage_order > 0 && freepage_order < MAX_ORDER) + if (freepage_order > 0 && freepage_order <= MAX_ORDER) low_pfn += (1UL << freepage_order) - 1; continue; } @@ -954,7 +954,7 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn, if (PageCompound(page) && !cc->alloc_contig) { const unsigned int order = compound_order(page); - if (likely(order < MAX_ORDER)) + if (likely(order <= MAX_ORDER)) low_pfn += (1UL << order) - 1; goto isolate_fail; } @@ -2124,7 +2124,7 @@ static enum compact_result __compact_finished(struct compact_control *cc) /* Direct compactor: Is a suitable page free? */ ret = COMPACT_NO_SUITABLE_PAGE; - for (order = cc->order; order < MAX_ORDER; order++) { + for (order = cc->order; order <= MAX_ORDER; order++) { struct free_area *area = &cc->zone->free_area[order]; bool can_steal; diff --git a/mm/debug_vm_pgtable.c b/mm/debug_vm_pgtable.c index 4362021b1ce7..c54177aabebd 100644 --- a/mm/debug_vm_pgtable.c +++ b/mm/debug_vm_pgtable.c @@ -1086,7 +1086,7 @@ debug_vm_pgtable_alloc_huge_page(struct pgtable_debug_args *args, int order) struct page *page = NULL; #ifdef CONFIG_CONTIG_ALLOC - if (order >= MAX_ORDER) { + if (order > MAX_ORDER) { page = alloc_contig_pages((1 << order), GFP_KERNEL, first_online_node, NULL); if (page) { @@ -1096,7 +1096,7 @@ debug_vm_pgtable_alloc_huge_page(struct pgtable_debug_args *args, int order) } #endif - if (order < MAX_ORDER) + if (order <= MAX_ORDER) page = alloc_pages(GFP_KERNEL, order); return page; diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 2d860e70fe88..1df386d13d24 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -467,7 +467,7 @@ static int __init hugepage_init(void) /* * hugepages can't be allocated by the buddy allocator */ - MAYBE_BUILD_BUG_ON(HPAGE_PMD_ORDER >= MAX_ORDER); + MAYBE_BUILD_BUG_ON(HPAGE_PMD_ORDER > MAX_ORDER); /* * we use page->mapping and page->index in second tail page * as list_head: assuming THP order >= 2 diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 712e32b38295..9122e50ae02a 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -2090,7 +2090,7 @@ pgoff_t hugetlb_basepage_index(struct page *page) pgoff_t index = page_index(page_head); unsigned long compound_idx; - if (compound_order(page_head) >= MAX_ORDER) + if (compound_order(page_head) > MAX_ORDER) compound_idx = page_to_pfn(page) - page_to_pfn(page_head); else compound_idx = page - page_head; @@ -4497,7 +4497,7 @@ static int __init default_hugepagesz_setup(char *s) * The number of default huge pages (for this size) could have been * specified as the first hugetlb parameter: hugepages=X. If so, * then default_hstate_max_huge_pages is set. If the default huge - * page size is gigantic (>= MAX_ORDER), then the pages must be + * page size is gigantic (> MAX_ORDER), then the pages must be * allocated here from bootmem allocator. */ if (default_hstate_max_huge_pages) { diff --git a/mm/kmsan/init.c b/mm/kmsan/init.c index 7fb794242fad..ffedf4dbc49d 100644 --- a/mm/kmsan/init.c +++ b/mm/kmsan/init.c @@ -96,7 +96,7 @@ void __init kmsan_init_shadow(void) struct metadata_page_pair { struct page *shadow, *origin; }; -static struct metadata_page_pair held_back[MAX_ORDER] __initdata; +static struct metadata_page_pair held_back[MAX_ORDER + 1] __initdata; /* * Eager metadata allocation. When the memblock allocator is freeing pages to @@ -211,8 +211,8 @@ static void kmsan_memblock_discard(void) * order=N-1, * - repeat. */ - collect.order = MAX_ORDER - 1; - for (int i = MAX_ORDER - 1; i >= 0; i--) { + collect.order = MAX_ORDER; + for (int i = MAX_ORDER; i >= 0; i--) { if (held_back[i].shadow) smallstack_push(&collect, held_back[i].shadow); if (held_back[i].origin) diff --git a/mm/memblock.c b/mm/memblock.c index 25fd0626a9e7..7911224b1ed3 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -2043,7 +2043,7 @@ static void __init __free_pages_memory(unsigned long start, unsigned long end) int order; while (start < end) { - order = min(MAX_ORDER - 1UL, __ffs(start)); + order = min_t(int, MAX_ORDER, __ffs(start)); while (start + (1UL << order) > end) order--; diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index db3b270254f1..c8f0a8c2d049 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -596,7 +596,7 @@ static void online_pages_range(unsigned long start_pfn, unsigned long nr_pages) unsigned long pfn; /* - * Online the pages in MAX_ORDER - 1 aligned chunks. The callback might + * Online the pages in MAX_ORDER aligned chunks. The callback might * decide to not expose all pages to the buddy (e.g., expose them * later). We account all pages as being online and belonging to this * zone ("present"). @@ -605,7 +605,7 @@ static void online_pages_range(unsigned long start_pfn, unsigned long nr_pages) * this and the first chunk to online will be pageblock_nr_pages. */ for (pfn = start_pfn; pfn < end_pfn;) { - int order = min(MAX_ORDER - 1UL, __ffs(pfn)); + int order = min_t(int, MAX_ORDER, __ffs(pfn)); (*online_page_callback)(pfn_to_page(pfn), order); pfn += (1UL << order); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 0936bde1d486..c3e49d028a7a 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1063,7 +1063,7 @@ buddy_merge_likely(unsigned long pfn, unsigned long buddy_pfn, unsigned long higher_page_pfn; struct page *higher_page; - if (order >= MAX_ORDER - 2) + if (order >= MAX_ORDER - 1) return false; higher_page_pfn = buddy_pfn & pfn; @@ -1118,7 +1118,7 @@ static inline void __free_one_page(struct page *page, VM_BUG_ON_PAGE(pfn & ((1 << order) - 1), page); VM_BUG_ON_PAGE(bad_range(zone, page), page); - while (order < MAX_ORDER - 1) { + while (order < MAX_ORDER) { if (compaction_capture(capc, page, order, migratetype)) { __mod_zone_freepage_state(zone, -(1 << order), migratetype); @@ -2499,7 +2499,7 @@ struct page *__rmqueue_smallest(struct zone *zone, unsigned int order, struct page *page; /* Find a page of the appropriate size in the preferred list */ - for (current_order = order; current_order < MAX_ORDER; ++current_order) { + for (current_order = order; current_order <= MAX_ORDER; ++current_order) { area = &(zone->free_area[current_order]); page = get_page_from_free_area(area, migratetype); if (!page) @@ -2871,7 +2871,7 @@ static bool unreserve_highatomic_pageblock(const struct alloc_context *ac, continue; spin_lock_irqsave(&zone->lock, flags); - for (order = 0; order < MAX_ORDER; order++) { + for (order = 0; order <= MAX_ORDER; order++) { struct free_area *area = &(zone->free_area[order]); page = get_page_from_free_area(area, MIGRATE_HIGHATOMIC); @@ -2955,7 +2955,7 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype, * approximates finding the pageblock with the most free pages, which * would be too costly to do exactly. */ - for (current_order = MAX_ORDER - 1; current_order >= min_order; + for (current_order = MAX_ORDER; current_order >= min_order; --current_order) { area = &(zone->free_area[current_order]); fallback_mt = find_suitable_fallback(area, current_order, @@ -2981,7 +2981,7 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype, return false; find_smallest: - for (current_order = order; current_order < MAX_ORDER; + for (current_order = order; current_order <= MAX_ORDER; current_order++) { area = &(zone->free_area[current_order]); fallback_mt = find_suitable_fallback(area, current_order, @@ -2994,7 +2994,7 @@ find_smallest: * This should not happen - we already found a suitable fallback * when looking for the largest page. */ - VM_BUG_ON(current_order == MAX_ORDER); + VM_BUG_ON(current_order > MAX_ORDER); do_steal: page = get_page_from_free_area(area, fallback_mt); @@ -3955,7 +3955,7 @@ bool __zone_watermark_ok(struct zone *z, unsigned int order, unsigned long mark, return true; /* For a high-order request, check at least one suitable page is free */ - for (o = order; o < MAX_ORDER; o++) { + for (o = order; o <= MAX_ORDER; o++) { struct free_area *area = &z->free_area[o]; int mt; @@ -5475,7 +5475,7 @@ struct page *__alloc_pages(gfp_t gfp, unsigned int order, int preferred_nid, * There are several places where we assume that the order value is sane * so bail out early if the request is out of bound. */ - if (WARN_ON_ONCE_GFP(order >= MAX_ORDER, gfp)) + if (WARN_ON_ONCE_GFP(order > MAX_ORDER, gfp)) return NULL; gfp &= gfp_allowed_mask; @@ -6205,8 +6205,8 @@ void __show_free_areas(unsigned int filter, nodemask_t *nodemask, int max_zone_i for_each_populated_zone(zone) { unsigned int order; - unsigned long nr[MAX_ORDER], flags, total = 0; - unsigned char types[MAX_ORDER]; + unsigned long nr[MAX_ORDER + 1], flags, total = 0; + unsigned char types[MAX_ORDER + 1]; if (zone_idx(zone) > max_zone_idx) continue; @@ -6216,7 +6216,7 @@ void __show_free_areas(unsigned int filter, nodemask_t *nodemask, int max_zone_i printk(KERN_CONT "%s: ", zone->name); spin_lock_irqsave(&zone->lock, flags); - for (order = 0; order < MAX_ORDER; order++) { + for (order = 0; order <= MAX_ORDER; order++) { struct free_area *area = &zone->free_area[order]; int type; @@ -6230,7 +6230,7 @@ void __show_free_areas(unsigned int filter, nodemask_t *nodemask, int max_zone_i } } spin_unlock_irqrestore(&zone->lock, flags); - for (order = 0; order < MAX_ORDER; order++) { + for (order = 0; order <= MAX_ORDER; order++) { printk(KERN_CONT "%lu*%lukB ", nr[order], K(1UL) << order); if (nr[order]) @@ -7581,7 +7581,7 @@ static inline void setup_usemap(struct zone *zone) {} /* Initialise the number of pages represented by NR_PAGEBLOCK_BITS */ void __init set_pageblock_order(void) { - unsigned int order = MAX_ORDER - 1; + unsigned int order = MAX_ORDER; /* Check that pageblock_nr_pages has not already been setup */ if (pageblock_order) @@ -9076,7 +9076,7 @@ void *__init alloc_large_system_hash(const char *tablename, else table = memblock_alloc_raw(size, SMP_CACHE_BYTES); - } else if (get_order(size) >= MAX_ORDER || hashdist) { + } else if (get_order(size) > MAX_ORDER || hashdist) { table = vmalloc_huge(size, gfp_flags); virt = true; if (table) @@ -9290,7 +9290,7 @@ int alloc_contig_range(unsigned long start, unsigned long end, order = 0; outer_start = start; while (!PageBuddy(pfn_to_page(outer_start))) { - if (++order >= MAX_ORDER) { + if (++order > MAX_ORDER) { outer_start = start; break; } @@ -9540,7 +9540,7 @@ bool is_free_buddy_page(struct page *page) unsigned long pfn = page_to_pfn(page); unsigned int order; - for (order = 0; order < MAX_ORDER; order++) { + for (order = 0; order <= MAX_ORDER; order++) { struct page *page_head = page - (pfn & ((1 << order) - 1)); if (PageBuddy(page_head) && @@ -9548,7 +9548,7 @@ bool is_free_buddy_page(struct page *page) break; } - return order < MAX_ORDER; + return order <= MAX_ORDER; } EXPORT_SYMBOL(is_free_buddy_page); @@ -9599,7 +9599,7 @@ bool take_page_off_buddy(struct page *page) bool ret = false; spin_lock_irqsave(&zone->lock, flags); - for (order = 0; order < MAX_ORDER; order++) { + for (order = 0; order <= MAX_ORDER; order++) { struct page *page_head = page - (pfn & ((1 << order) - 1)); int page_order = buddy_order(page_head); diff --git a/mm/page_isolation.c b/mm/page_isolation.c index 47fbc1696466..c6f3605e37ab 100644 --- a/mm/page_isolation.c +++ b/mm/page_isolation.c @@ -226,7 +226,7 @@ static void unset_migratetype_isolate(struct page *page, int migratetype) */ if (PageBuddy(page)) { order = buddy_order(page); - if (order >= pageblock_order && order < MAX_ORDER - 1) { + if (order >= pageblock_order && order < MAX_ORDER) { buddy = find_buddy_page_pfn(page, page_to_pfn(page), order, NULL); if (buddy && !is_migrate_isolate_page(buddy)) { @@ -290,11 +290,11 @@ __first_valid_page(unsigned long pfn, unsigned long nr_pages) * isolate_single_pageblock() * @migratetype: migrate type to set in error recovery. * - * Free and in-use pages can be as big as MAX_ORDER-1 and contain more than one + * Free and in-use pages can be as big as MAX_ORDER and contain more than one * pageblock. When not all pageblocks within a page are isolated at the same * time, free page accounting can go wrong. For example, in the case of - * MAX_ORDER-1 = pageblock_order + 1, a MAX_ORDER-1 page has two pagelbocks. - * [ MAX_ORDER-1 ] + * MAX_ORDER = pageblock_order + 1, a MAX_ORDER page has two pagelbocks. + * [ MAX_ORDER ] * [ pageblock0 | pageblock1 ] * When either pageblock is isolated, if it is a free page, the page is not * split into separate migratetype lists, which is supposed to; if it is an @@ -451,7 +451,7 @@ static int isolate_single_pageblock(unsigned long boundary_pfn, int flags, * the free page to the right migratetype list. * * head_pfn is not used here as a hugetlb page order - * can be bigger than MAX_ORDER-1, but after it is + * can be bigger than MAX_ORDER, but after it is * freed, the free page order is not. Use pfn within * the range to find the head of the free page. */ @@ -459,7 +459,7 @@ static int isolate_single_pageblock(unsigned long boundary_pfn, int flags, outer_pfn = pfn; while (!PageBuddy(pfn_to_page(outer_pfn))) { /* stop if we cannot find the free page */ - if (++order >= MAX_ORDER) + if (++order > MAX_ORDER) goto failed; outer_pfn &= ~0UL << order; } diff --git a/mm/page_owner.c b/mm/page_owner.c index 220cdeddc295..31169b3e7f06 100644 --- a/mm/page_owner.c +++ b/mm/page_owner.c @@ -315,7 +315,7 @@ void pagetypeinfo_showmixedcount_print(struct seq_file *m, unsigned long freepage_order; freepage_order = buddy_order_unsafe(page); - if (freepage_order < MAX_ORDER) + if (freepage_order <= MAX_ORDER) pfn += (1UL << freepage_order) - 1; continue; } @@ -549,7 +549,7 @@ read_page_owner(struct file *file, char __user *buf, size_t count, loff_t *ppos) if (PageBuddy(page)) { unsigned long freepage_order = buddy_order_unsafe(page); - if (freepage_order < MAX_ORDER) + if (freepage_order <= MAX_ORDER) pfn += (1UL << freepage_order) - 1; continue; } @@ -657,7 +657,7 @@ static void init_pages_in_zone(pg_data_t *pgdat, struct zone *zone) if (PageBuddy(page)) { unsigned long order = buddy_order_unsafe(page); - if (order > 0 && order < MAX_ORDER) + if (order > 0 && order <= MAX_ORDER) pfn += (1UL << order) - 1; continue; } diff --git a/mm/page_reporting.c b/mm/page_reporting.c index 275b466de37b..b021f482a4cb 100644 --- a/mm/page_reporting.c +++ b/mm/page_reporting.c @@ -20,7 +20,7 @@ static int page_order_update_notify(const char *val, const struct kernel_param * * If param is set beyond this limit, order is set to default * pageblock_order value */ - return param_set_uint_minmax(val, kp, 0, MAX_ORDER-1); + return param_set_uint_minmax(val, kp, 0, MAX_ORDER); } static const struct kernel_param_ops page_reporting_param_ops = { @@ -276,7 +276,7 @@ page_reporting_process_zone(struct page_reporting_dev_info *prdev, return err; /* Process each free list starting from lowest order/mt */ - for (order = page_reporting_order; order < MAX_ORDER; order++) { + for (order = page_reporting_order; order <= MAX_ORDER; order++) { for (mt = 0; mt < MIGRATE_TYPES; mt++) { /* We do not pull pages from the isolate free list */ if (is_migrate_isolate(mt)) @@ -370,7 +370,7 @@ int page_reporting_register(struct page_reporting_dev_info *prdev) */ if (page_reporting_order == -1) { - if (prdev->order > 0 && prdev->order < MAX_ORDER) + if (prdev->order > 0 && prdev->order <= MAX_ORDER) page_reporting_order = prdev->order; else page_reporting_order = pageblock_order; diff --git a/mm/shuffle.h b/mm/shuffle.h index cec62984f7d3..a6bdf54f96f1 100644 --- a/mm/shuffle.h +++ b/mm/shuffle.h @@ -4,7 +4,7 @@ #define _MM_SHUFFLE_H #include -#define SHUFFLE_ORDER (MAX_ORDER-1) +#define SHUFFLE_ORDER MAX_ORDER #ifdef CONFIG_SHUFFLE_PAGE_ALLOCATOR DECLARE_STATIC_KEY_FALSE(page_alloc_shuffle_key); diff --git a/mm/slab.c b/mm/slab.c index edbe722fb906..6b7c172158e5 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -465,7 +465,7 @@ static int __init slab_max_order_setup(char *str) { get_option(&str, &slab_max_order); slab_max_order = slab_max_order < 0 ? 0 : - min(slab_max_order, MAX_ORDER - 1); + min(slab_max_order, MAX_ORDER); slab_max_order_set = true; return 1; diff --git a/mm/slub.c b/mm/slub.c index 32eb6b50fe18..f49d669ff604 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -4171,8 +4171,8 @@ static inline int calculate_order(unsigned int size) /* * Doh this slab cannot be placed using slub_max_order. */ - order = calc_slab_order(size, 1, MAX_ORDER - 1, 1); - if (order < MAX_ORDER) + order = calc_slab_order(size, 1, MAX_ORDER, 1); + if (order <= MAX_ORDER) return order; return -ENOSYS; } @@ -4697,7 +4697,7 @@ __setup("slub_min_order=", setup_slub_min_order); static int __init setup_slub_max_order(char *str) { get_option(&str, (int *)&slub_max_order); - slub_max_order = min(slub_max_order, (unsigned int)MAX_ORDER - 1); + slub_max_order = min_t(unsigned int, slub_max_order, MAX_ORDER); return 1; } diff --git a/mm/vmscan.c b/mm/vmscan.c index 8faac4310cb5..98719e72b5e2 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -7002,7 +7002,7 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order, * scan_control uses s8 fields for order, priority, and reclaim_idx. * Confirm they are large enough for max values. */ - BUILD_BUG_ON(MAX_ORDER > S8_MAX); + BUILD_BUG_ON(MAX_ORDER >= S8_MAX); BUILD_BUG_ON(DEF_PRIORITY > S8_MAX); BUILD_BUG_ON(MAX_NR_ZONES > S8_MAX); diff --git a/mm/vmstat.c b/mm/vmstat.c index 1ea6a5ce1c41..b7307627772d 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -1055,7 +1055,7 @@ static void fill_contig_page_info(struct zone *zone, info->free_blocks_total = 0; info->free_blocks_suitable = 0; - for (order = 0; order < MAX_ORDER; order++) { + for (order = 0; order <= MAX_ORDER; order++) { unsigned long blocks; /* @@ -1088,7 +1088,7 @@ static int __fragmentation_index(unsigned int order, struct contig_page_info *in { unsigned long requested = 1UL << order; - if (WARN_ON_ONCE(order >= MAX_ORDER)) + if (WARN_ON_ONCE(order > MAX_ORDER)) return 0; if (!info->free_blocks_total) @@ -1462,7 +1462,7 @@ static void frag_show_print(struct seq_file *m, pg_data_t *pgdat, int order; seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name); - for (order = 0; order < MAX_ORDER; ++order) + for (order = 0; order <= MAX_ORDER; ++order) /* * Access to nr_free is lockless as nr_free is used only for * printing purposes. Use data_race to avoid KCSAN warning. @@ -1491,7 +1491,7 @@ static void pagetypeinfo_showfree_print(struct seq_file *m, pgdat->node_id, zone->name, migratetype_names[mtype]); - for (order = 0; order < MAX_ORDER; ++order) { + for (order = 0; order <= MAX_ORDER; ++order) { unsigned long freecount = 0; struct free_area *area; struct list_head *curr; @@ -1531,7 +1531,7 @@ static void pagetypeinfo_showfree(struct seq_file *m, void *arg) /* Print header */ seq_printf(m, "%-43s ", "Free pages count per migrate type at order"); - for (order = 0; order < MAX_ORDER; ++order) + for (order = 0; order <= MAX_ORDER; ++order) seq_printf(m, "%6d ", order); seq_putc(m, '\n'); @@ -2153,7 +2153,7 @@ static void unusable_show_print(struct seq_file *m, seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name); - for (order = 0; order < MAX_ORDER; ++order) { + for (order = 0; order <= MAX_ORDER; ++order) { fill_contig_page_info(zone, order, &info); index = unusable_free_index(order, &info); seq_printf(m, "%d.%03d ", index / 1000, index % 1000); @@ -2205,7 +2205,7 @@ static void extfrag_show_print(struct seq_file *m, seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name); - for (order = 0; order < MAX_ORDER; ++order) { + for (order = 0; order <= MAX_ORDER; ++order) { fill_contig_page_info(zone, order, &info); index = __fragmentation_index(order, &info); seq_printf(m, "%2d.%03d ", index / 1000, index % 1000); diff --git a/net/smc/smc_ib.c b/net/smc/smc_ib.c index 854772dd52fd..9b66d6aeeb1a 100644 --- a/net/smc/smc_ib.c +++ b/net/smc/smc_ib.c @@ -843,7 +843,7 @@ long smc_ib_setup_per_ibdev(struct smc_ib_device *smcibdev) goto out; /* the calculated number of cq entries fits to mlx5 cq allocation */ cqe_size_order = cache_line_size() == 128 ? 7 : 6; - smc_order = MAX_ORDER - cqe_size_order - 1; + smc_order = MAX_ORDER - cqe_size_order; if (SMC_MAX_CQE + 2 > (0x00000001 << smc_order) * PAGE_SIZE) cqattr.cqe = (0x00000001 << smc_order) * PAGE_SIZE - 2; smcibdev->roce_cq_send = ib_create_cq(smcibdev->ibdev, diff --git a/security/integrity/ima/ima_crypto.c b/security/integrity/ima/ima_crypto.c index 64499056648a..51ad29940f05 100644 --- a/security/integrity/ima/ima_crypto.c +++ b/security/integrity/ima/ima_crypto.c @@ -38,7 +38,7 @@ static int param_set_bufsize(const char *val, const struct kernel_param *kp) size = memparse(val, NULL); order = get_order(size); - if (order >= MAX_ORDER) + if (order > MAX_ORDER) return -EINVAL; ima_maxorder = order; ima_bufsize = PAGE_SIZE << order; diff --git a/tools/testing/memblock/linux/mmzone.h b/tools/testing/memblock/linux/mmzone.h index e65f89b12f1c..134f8eab0768 100644 --- a/tools/testing/memblock/linux/mmzone.h +++ b/tools/testing/memblock/linux/mmzone.h @@ -17,10 +17,10 @@ enum zone_type { }; #define MAX_NR_ZONES __MAX_NR_ZONES -#define MAX_ORDER 11 -#define MAX_ORDER_NR_PAGES (1 << (MAX_ORDER - 1)) +#define MAX_ORDER 10 +#define MAX_ORDER_NR_PAGES (1 << MAX_ORDER) -#define pageblock_order (MAX_ORDER - 1) +#define pageblock_order MAX_ORDER #define pageblock_nr_pages BIT(pageblock_order) #define pageblock_align(pfn) ALIGN((pfn), pageblock_nr_pages) #define pageblock_start_pfn(pfn) ALIGN_DOWN((pfn), pageblock_nr_pages) -- cgit v1.2.3 From 3e1c957f9a3b248f47f8b39b607002d948fd17d4 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Thu, 30 Mar 2023 11:02:26 +0300 Subject: xfrm: don't require advance ESN callback for packet offload In packet offload mode, the hardware is responsible to manage replay window and advance ESN. In that mode, there won't any call to .xdo_dev_state_advance_esn callback. So relax current check for existence of that callback. Link: https://lore.kernel.org/r/9f3dfc3fef2cfcd191f0c5eee7cf0aa74e7f7786.1680162300.git.leonro@nvidia.com Reviewed-by: Raed Salem Acked-by: Steffen Klassert Signed-off-by: Leon Romanovsky --- net/xfrm/xfrm_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c index 95f1436bf6a2..bef28c6187eb 100644 --- a/net/xfrm/xfrm_device.c +++ b/net/xfrm/xfrm_device.c @@ -287,7 +287,7 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x, return (is_packet_offload) ? -EINVAL : 0; } - if (x->props.flags & XFRM_STATE_ESN && + if (!is_packet_offload && x->props.flags & XFRM_STATE_ESN && !dev->xfrmdev_ops->xdo_dev_state_advance_esn) { NL_SET_ERR_MSG(extack, "Device doesn't support offload with ESN"); xso->dev = NULL; -- cgit v1.2.3 From 94623f579ce338b5fa61b5acaa5beb8aa657fb9e Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 3 Apr 2023 13:54:37 +0200 Subject: netfilter: br_netfilter: fix recent physdev match breakage Recent attempt to ensure PREROUTING hook is executed again when a decrypted ipsec packet received on a bridge passes through the network stack a second time broke the physdev match in INPUT hook. We can't discard the nf_bridge info strct from sabotage_in hook, as this is needed by the physdev match. Keep the struct around and handle this with another conditional instead. Fixes: 2b272bb558f1 ("netfilter: br_netfilter: disable sabotage_in hook after first suppression") Reported-and-tested-by: Farid BENAMROUCHE Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/skbuff.h | 1 + net/bridge/br_netfilter_hooks.c | 17 +++++++++++------ 2 files changed, 12 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index ff7ad331fb82..1ec3530c8191 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -294,6 +294,7 @@ struct nf_bridge_info { u8 pkt_otherhost:1; u8 in_prerouting:1; u8 bridged_dnat:1; + u8 sabotage_in_done:1; __u16 frag_max_size; struct net_device *physindev; diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c index 638a4d5359db..4bc6761517bb 100644 --- a/net/bridge/br_netfilter_hooks.c +++ b/net/bridge/br_netfilter_hooks.c @@ -868,12 +868,17 @@ static unsigned int ip_sabotage_in(void *priv, { struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb); - if (nf_bridge && !nf_bridge->in_prerouting && - !netif_is_l3_master(skb->dev) && - !netif_is_l3_slave(skb->dev)) { - nf_bridge_info_free(skb); - state->okfn(state->net, state->sk, skb); - return NF_STOLEN; + if (nf_bridge) { + if (nf_bridge->sabotage_in_done) + return NF_ACCEPT; + + if (!nf_bridge->in_prerouting && + !netif_is_l3_master(skb->dev) && + !netif_is_l3_slave(skb->dev)) { + nf_bridge->sabotage_in_done = 1; + state->okfn(state->net, state->sk, skb); + return NF_STOLEN; + } } return NF_ACCEPT; -- cgit v1.2.3 From af0acf22aea359e04412237d68787401f96bb583 Mon Sep 17 00:00:00 2001 From: Chen Aotian Date: Thu, 6 Apr 2023 12:01:51 +0800 Subject: netfilter: nf_tables: Modify nla_memdup's flag to GFP_KERNEL_ACCOUNT For memory alloc that store user data from nla[NFTA_OBJ_USERDATA], use GFP_KERNEL_ACCOUNT is more suitable. Fixes: 33758c891479 ("memcg: enable accounting for nft objects") Signed-off-by: Chen Aotian Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 6004d4b24451..cd52109e674a 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -7052,7 +7052,7 @@ static int nf_tables_newobj(struct sk_buff *skb, const struct nfnl_info *info, } if (nla[NFTA_OBJ_USERDATA]) { - obj->udata = nla_memdup(nla[NFTA_OBJ_USERDATA], GFP_KERNEL); + obj->udata = nla_memdup(nla[NFTA_OBJ_USERDATA], GFP_KERNEL_ACCOUNT); if (obj->udata == NULL) goto err_userdata; -- cgit v1.2.3 From 430cac487400494c19a8b85299e979bb07b4671f Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Tue, 4 Apr 2023 15:12:16 +0200 Subject: xfrm: don't check the default policy if the policy allows the packet The current code doesn't let a simple "allow" policy counteract a default policy blocking all incoming packets: ip x p setdefault in block ip x p a src 192.168.2.1/32 dst 192.168.2.2/32 dir in action allow At this stage, we have an allow policy (with or without transforms) for this packet. It doesn't matter what the default policy says, since the policy we looked up lets the packet through. The case of a blocking policy is already handled separately, so we can remove this check. Fixes: 2d151d39073a ("xfrm: Add possibility to set the default to block if we have no policy") Signed-off-by: Sabrina Dubroca Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 5c61ec04b839..62be042f2ebc 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -3712,12 +3712,6 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, } xfrm_nr = ti; - if (net->xfrm.policy_default[dir] == XFRM_USERPOLICY_BLOCK && - !xfrm_nr) { - XFRM_INC_STATS(net, LINUX_MIB_XFRMINNOSTATES); - goto reject; - } - if (npols > 1) { xfrm_tmpl_sort(stp, tpp, xfrm_nr, family); tpp = stp; -- cgit v1.2.3 From d769ccaf957fe7391f357c0a923de71f594b8a2b Mon Sep 17 00:00:00 2001 From: Kal Conley Date: Thu, 6 Apr 2023 01:59:18 +0200 Subject: xsk: Fix unaligned descriptor validation Make sure unaligned descriptors that straddle the end of the UMEM are considered invalid. Currently, descriptor validation is broken for zero-copy mode which only checks descriptors at page granularity. For example, descriptors in zero-copy mode that overrun the end of the UMEM but not a page boundary are (incorrectly) considered valid. The UMEM boundary check needs to happen before the page boundary and contiguity checks in xp_desc_crosses_non_contig_pg(). Do this check in xp_unaligned_validate_desc() instead like xp_check_unaligned() already does. Fixes: 2b43470add8c ("xsk: Introduce AF_XDP buffer allocation API") Signed-off-by: Kal Conley Acked-by: Magnus Karlsson Link: https://lore.kernel.org/r/20230405235920.7305-2-kal.conley@dectris.com Signed-off-by: Martin KaFai Lau --- include/net/xsk_buff_pool.h | 9 ++------- net/xdp/xsk_queue.h | 1 + 2 files changed, 3 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/net/xsk_buff_pool.h b/include/net/xsk_buff_pool.h index 3e952e569418..d318c769b445 100644 --- a/include/net/xsk_buff_pool.h +++ b/include/net/xsk_buff_pool.h @@ -180,13 +180,8 @@ static inline bool xp_desc_crosses_non_contig_pg(struct xsk_buff_pool *pool, if (likely(!cross_pg)) return false; - if (pool->dma_pages_cnt) { - return !(pool->dma_pages[addr >> PAGE_SHIFT] & - XSK_NEXT_PG_CONTIG_MASK); - } - - /* skb path */ - return addr + len > pool->addrs_cnt; + return pool->dma_pages_cnt && + !(pool->dma_pages[addr >> PAGE_SHIFT] & XSK_NEXT_PG_CONTIG_MASK); } static inline u64 xp_aligned_extract_addr(struct xsk_buff_pool *pool, u64 addr) diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h index bfb2a7e50c26..66c6f57c9c44 100644 --- a/net/xdp/xsk_queue.h +++ b/net/xdp/xsk_queue.h @@ -162,6 +162,7 @@ static inline bool xp_unaligned_validate_desc(struct xsk_buff_pool *pool, return false; if (base_addr >= pool->addrs_cnt || addr >= pool->addrs_cnt || + addr + desc->len > pool->addrs_cnt || xp_desc_crosses_non_contig_pg(pool, addr, desc->len)) return false; -- cgit v1.2.3 From 75c7124ef3bafe0e9d1801d6449196dc3105df24 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Wed, 5 Apr 2023 15:27:17 -0500 Subject: net: rfkill-gpio: Add explicit include for of.h With linux/acpi.h no longer implicitly including of.h, add an explicit include of of.h to fix the following error: net/rfkill/rfkill-gpio.c:181:21: error: implicit declaration of function 'of_match_ptr' is invalid in C99 [-Werror,-Wimplicit-function-declaration] Acked-by: Johannes Berg Reviewed-by: Jacob Keller Reviewed-by: Greg Kroah-Hartman Signed-off-by: Rob Herring Signed-off-by: Rafael J. Wysocki --- net/rfkill/rfkill-gpio.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/rfkill/rfkill-gpio.c b/net/rfkill/rfkill-gpio.c index 786dbfdad772..e9d1b2f2ff0a 100644 --- a/net/rfkill/rfkill-gpio.c +++ b/net/rfkill/rfkill-gpio.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From dc5110c2d959c1707e12df5f792f41d90614adaa Mon Sep 17 00:00:00 2001 From: YueHaibing Date: Thu, 6 Apr 2023 14:34:50 +0800 Subject: tcp: restrict net.ipv4.tcp_app_win UBSAN: shift-out-of-bounds in net/ipv4/tcp_input.c:555:23 shift exponent 255 is too large for 32-bit type 'int' CPU: 1 PID: 7907 Comm: ssh Not tainted 6.3.0-rc4-00161-g62bad54b26db-dirty #206 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 Call Trace: dump_stack_lvl+0x136/0x150 __ubsan_handle_shift_out_of_bounds+0x21f/0x5a0 tcp_init_transfer.cold+0x3a/0xb9 tcp_finish_connect+0x1d0/0x620 tcp_rcv_state_process+0xd78/0x4d60 tcp_v4_do_rcv+0x33d/0x9d0 __release_sock+0x133/0x3b0 release_sock+0x58/0x1b0 'maxwin' is int, shifting int for 32 or more bits is undefined behaviour. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: YueHaibing Reviewed-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.rst | 2 ++ net/ipv4/sysctl_net_ipv4.c | 3 +++ 2 files changed, 5 insertions(+) (limited to 'net') diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index 87dd1c5283e6..58a78a316697 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -340,6 +340,8 @@ tcp_app_win - INTEGER Reserve max(window/2^tcp_app_win, mss) of window for application buffer. Value 0 is special, it means that nothing is reserved. + Possible values are [0, 31], inclusive. + Default: 31 tcp_autocorking - BOOLEAN diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 0d0cc4ef2b85..40fe70fc2015 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -25,6 +25,7 @@ static int ip_local_port_range_min[] = { 1, 1 }; static int ip_local_port_range_max[] = { 65535, 65535 }; static int tcp_adv_win_scale_min = -31; static int tcp_adv_win_scale_max = 31; +static int tcp_app_win_max = 31; static int tcp_min_snd_mss_min = TCP_MIN_SND_MSS; static int tcp_min_snd_mss_max = 65535; static int ip_privileged_port_min; @@ -1198,6 +1199,8 @@ static struct ctl_table ipv4_net_table[] = { .maxlen = sizeof(u8), .mode = 0644, .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = &tcp_app_win_max, }, { .procname = "tcp_adv_win_scale", -- cgit v1.2.3 From 066b86787fa3d97b7aefb5ac0a99a22dad2d15f8 Mon Sep 17 00:00:00 2001 From: Felix Huettner Date: Wed, 5 Apr 2023 07:53:41 +0000 Subject: net: openvswitch: fix race on port output assume the following setup on a single machine: 1. An openvswitch instance with one bridge and default flows 2. two network namespaces "server" and "client" 3. two ovs interfaces "server" and "client" on the bridge 4. for each ovs interface a veth pair with a matching name and 32 rx and tx queues 5. move the ends of the veth pairs to the respective network namespaces 6. assign ip addresses to each of the veth ends in the namespaces (needs to be the same subnet) 7. start some http server on the server network namespace 8. test if a client in the client namespace can reach the http server when following the actions below the host has a chance of getting a cpu stuck in a infinite loop: 1. send a large amount of parallel requests to the http server (around 3000 curls should work) 2. in parallel delete the network namespace (do not delete interfaces or stop the server, just kill the namespace) there is a low chance that this will cause the below kernel cpu stuck message. If this does not happen just retry. Below there is also the output of bpftrace for the functions mentioned in the output. The series of events happening here is: 1. the network namespace is deleted calling `unregister_netdevice_many_notify` somewhere in the process 2. this sets first `NETREG_UNREGISTERING` on both ends of the veth and then runs `synchronize_net` 3. it then calls `call_netdevice_notifiers` with `NETDEV_UNREGISTER` 4. this is then handled by `dp_device_event` which calls `ovs_netdev_detach_dev` (if a vport is found, which is the case for the veth interface attached to ovs) 5. this removes the rx_handlers of the device but does not prevent packages to be sent to the device 6. `dp_device_event` then queues the vport deletion to work in background as a ovs_lock is needed that we do not hold in the unregistration path 7. `unregister_netdevice_many_notify` continues to call `netdev_unregister_kobject` which sets `real_num_tx_queues` to 0 8. port deletion continues (but details are not relevant for this issue) 9. at some future point the background task deletes the vport If after 7. but before 9. a packet is send to the ovs vport (which is not deleted at this point in time) which forwards it to the `dev_queue_xmit` flow even though the device is unregistering. In `skb_tx_hash` (which is called in the `dev_queue_xmit`) path there is a while loop (if the packet has a rx_queue recorded) that is infinite if `dev->real_num_tx_queues` is zero. To prevent this from happening we update `do_output` to handle devices without carrier the same as if the device is not found (which would be the code path after 9. is done). Additionally we now produce a warning in `skb_tx_hash` if we will hit the infinite loop. bpftrace (first word is function name): __dev_queue_xmit server: real_num_tx_queues: 1, cpu: 2, pid: 28024, tid: 28024, skb_addr: 0xffff9edb6f207000, reg_state: 1 netdev_core_pick_tx server: addr: 0xffff9f0a46d4a000 real_num_tx_queues: 1, cpu: 2, pid: 28024, tid: 28024, skb_addr: 0xffff9edb6f207000, reg_state: 1 dp_device_event server: real_num_tx_queues: 1 cpu 9, pid: 21024, tid: 21024, event 2, reg_state: 1 synchronize_rcu_expedited: cpu 9, pid: 21024, tid: 21024 synchronize_rcu_expedited: cpu 9, pid: 21024, tid: 21024 synchronize_rcu_expedited: cpu 9, pid: 21024, tid: 21024 synchronize_rcu_expedited: cpu 9, pid: 21024, tid: 21024 dp_device_event server: real_num_tx_queues: 1 cpu 9, pid: 21024, tid: 21024, event 6, reg_state: 2 ovs_netdev_detach_dev server: real_num_tx_queues: 1 cpu 9, pid: 21024, tid: 21024, reg_state: 2 netdev_rx_handler_unregister server: real_num_tx_queues: 1, cpu: 9, pid: 21024, tid: 21024, reg_state: 2 synchronize_rcu_expedited: cpu 9, pid: 21024, tid: 21024 netdev_rx_handler_unregister ret server: real_num_tx_queues: 1, cpu: 9, pid: 21024, tid: 21024, reg_state: 2 dp_device_event server: real_num_tx_queues: 1 cpu 9, pid: 21024, tid: 21024, event 27, reg_state: 2 dp_device_event server: real_num_tx_queues: 1 cpu 9, pid: 21024, tid: 21024, event 22, reg_state: 2 dp_device_event server: real_num_tx_queues: 1 cpu 9, pid: 21024, tid: 21024, event 18, reg_state: 2 netdev_unregister_kobject: real_num_tx_queues: 1, cpu: 9, pid: 21024, tid: 21024 synchronize_rcu_expedited: cpu 9, pid: 21024, tid: 21024 ovs_vport_send server: real_num_tx_queues: 0, cpu: 2, pid: 28024, tid: 28024, skb_addr: 0xffff9edb6f207000, reg_state: 2 __dev_queue_xmit server: real_num_tx_queues: 0, cpu: 2, pid: 28024, tid: 28024, skb_addr: 0xffff9edb6f207000, reg_state: 2 netdev_core_pick_tx server: addr: 0xffff9f0a46d4a000 real_num_tx_queues: 0, cpu: 2, pid: 28024, tid: 28024, skb_addr: 0xffff9edb6f207000, reg_state: 2 broken device server: real_num_tx_queues: 0, cpu: 2, pid: 28024, tid: 28024 ovs_dp_detach_port server: real_num_tx_queues: 0 cpu 9, pid: 9124, tid: 9124, reg_state: 2 synchronize_rcu_expedited: cpu 9, pid: 33604, tid: 33604 stuck message: watchdog: BUG: soft lockup - CPU#5 stuck for 26s! [curl:1929279] Modules linked in: veth pktgen bridge stp llc ip_set_hash_net nft_counter xt_set nft_compat nf_tables ip_set_hash_ip ip_set nfnetlink_cttimeout nfnetlink openvswitch nsh nf_conncount nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 tls binfmt_misc nls_iso8859_1 input_leds joydev serio_raw dm_multipath scsi_dh_rdac scsi_dh_emc scsi_dh_alua sch_fq_codel drm efi_pstore virtio_rng ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear hid_generic usbhid hid crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel virtio_net ahci net_failover crypto_simd cryptd psmouse libahci virtio_blk failover CPU: 5 PID: 1929279 Comm: curl Not tainted 5.15.0-67-generic #74-Ubuntu Hardware name: OpenStack Foundation OpenStack Nova, BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 RIP: 0010:netdev_pick_tx+0xf1/0x320 Code: 00 00 8d 48 ff 0f b7 c1 66 39 ca 0f 86 e9 01 00 00 45 0f b7 ff 41 39 c7 0f 87 5b 01 00 00 44 29 f8 41 39 c7 0f 87 4f 01 00 00 f2 0f 1f 44 00 00 49 8b 94 24 28 04 00 00 48 85 d2 0f 84 53 01 RSP: 0018:ffffb78b40298820 EFLAGS: 00000246 RAX: 0000000000000000 RBX: ffff9c8773adc2e0 RCX: 000000000000083f RDX: 0000000000000000 RSI: ffff9c8773adc2e0 RDI: ffff9c870a25e000 RBP: ffffb78b40298858 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: ffff9c870a25e000 R13: ffff9c870a25e000 R14: ffff9c87fe043480 R15: 0000000000000000 FS: 00007f7b80008f00(0000) GS:ffff9c8e5f740000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f7b80f6a0b0 CR3: 0000000329d66000 CR4: 0000000000350ee0 Call Trace: netdev_core_pick_tx+0xa4/0xb0 __dev_queue_xmit+0xf8/0x510 ? __bpf_prog_exit+0x1e/0x30 dev_queue_xmit+0x10/0x20 ovs_vport_send+0xad/0x170 [openvswitch] do_output+0x59/0x180 [openvswitch] do_execute_actions+0xa80/0xaa0 [openvswitch] ? kfree+0x1/0x250 ? kfree+0x1/0x250 ? kprobe_perf_func+0x4f/0x2b0 ? flow_lookup.constprop.0+0x5c/0x110 [openvswitch] ovs_execute_actions+0x4c/0x120 [openvswitch] ovs_dp_process_packet+0xa1/0x200 [openvswitch] ? ovs_ct_update_key.isra.0+0xa8/0x120 [openvswitch] ? ovs_ct_fill_key+0x1d/0x30 [openvswitch] ? ovs_flow_key_extract+0x2db/0x350 [openvswitch] ovs_vport_receive+0x77/0xd0 [openvswitch] ? __htab_map_lookup_elem+0x4e/0x60 ? bpf_prog_680e8aff8547aec1_kfree+0x3b/0x714 ? trace_call_bpf+0xc8/0x150 ? kfree+0x1/0x250 ? kfree+0x1/0x250 ? kprobe_perf_func+0x4f/0x2b0 ? kprobe_perf_func+0x4f/0x2b0 ? __mod_memcg_lruvec_state+0x63/0xe0 netdev_port_receive+0xc4/0x180 [openvswitch] ? netdev_port_receive+0x180/0x180 [openvswitch] netdev_frame_hook+0x1f/0x40 [openvswitch] __netif_receive_skb_core.constprop.0+0x23d/0xf00 __netif_receive_skb_one_core+0x3f/0xa0 __netif_receive_skb+0x15/0x60 process_backlog+0x9e/0x170 __napi_poll+0x33/0x180 net_rx_action+0x126/0x280 ? ttwu_do_activate+0x72/0xf0 __do_softirq+0xd9/0x2e7 ? rcu_report_exp_cpu_mult+0x1b0/0x1b0 do_softirq+0x7d/0xb0 __local_bh_enable_ip+0x54/0x60 ip_finish_output2+0x191/0x460 __ip_finish_output+0xb7/0x180 ip_finish_output+0x2e/0xc0 ip_output+0x78/0x100 ? __ip_finish_output+0x180/0x180 ip_local_out+0x5e/0x70 __ip_queue_xmit+0x184/0x440 ? tcp_syn_options+0x1f9/0x300 ip_queue_xmit+0x15/0x20 __tcp_transmit_skb+0x910/0x9c0 ? __mod_memcg_state+0x44/0xa0 tcp_connect+0x437/0x4e0 ? ktime_get_with_offset+0x60/0xf0 tcp_v4_connect+0x436/0x530 __inet_stream_connect+0xd4/0x3a0 ? kprobe_perf_func+0x4f/0x2b0 ? aa_sk_perm+0x43/0x1c0 inet_stream_connect+0x3b/0x60 __sys_connect_file+0x63/0x70 __sys_connect+0xa6/0xd0 ? setfl+0x108/0x170 ? do_fcntl+0xe8/0x5a0 __x64_sys_connect+0x18/0x20 do_syscall_64+0x5c/0xc0 ? __x64_sys_fcntl+0xa9/0xd0 ? exit_to_user_mode_prepare+0x37/0xb0 ? syscall_exit_to_user_mode+0x27/0x50 ? do_syscall_64+0x69/0xc0 ? __sys_setsockopt+0xea/0x1e0 ? exit_to_user_mode_prepare+0x37/0xb0 ? syscall_exit_to_user_mode+0x27/0x50 ? __x64_sys_setsockopt+0x1f/0x30 ? do_syscall_64+0x69/0xc0 ? irqentry_exit+0x1d/0x30 ? exc_page_fault+0x89/0x170 entry_SYSCALL_64_after_hwframe+0x61/0xcb RIP: 0033:0x7f7b8101c6a7 Code: 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2a 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 18 89 54 24 0c 48 89 34 24 89 RSP: 002b:00007ffffd6b2198 EFLAGS: 00000246 ORIG_RAX: 000000000000002a RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f7b8101c6a7 RDX: 0000000000000010 RSI: 00007ffffd6b2360 RDI: 0000000000000005 RBP: 0000561f1370d560 R08: 00002795ad21d1ac R09: 0030312e302e302e R10: 00007ffffd73f080 R11: 0000000000000246 R12: 0000561f1370c410 R13: 0000000000000000 R14: 0000000000000005 R15: 0000000000000000 Fixes: 7f8a436eaa2c ("openvswitch: Add conntrack action") Co-developed-by: Luca Czesla Signed-off-by: Luca Czesla Signed-off-by: Felix Huettner Reviewed-by: Eric Dumazet Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/ZC0pBXBAgh7c76CA@kernel-bug-kernel-bug Signed-off-by: Jakub Kicinski --- net/core/dev.c | 1 + net/openvswitch/actions.c | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 253584777101..48067321c0db 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3199,6 +3199,7 @@ static u16 skb_tx_hash(const struct net_device *dev, } if (skb_rx_queue_recorded(skb)) { + DEBUG_NET_WARN_ON_ONCE(qcount == 0); hash = skb_get_rx_queue(skb); if (hash >= qoffset) hash -= qoffset; diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index ca3ebfdb3023..a8cf9a88758e 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -913,7 +913,7 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, { struct vport *vport = ovs_vport_rcu(dp, out_port); - if (likely(vport)) { + if (likely(vport && netif_carrier_ok(vport->dev))) { u16 mru = OVS_CB(skb)->mru; u32 cutlen = OVS_CB(skb)->cutlen; -- cgit v1.2.3 From 48b7ea1d22ddf657e2692fb2b399138a02d31c17 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 Apr 2023 19:46:34 +0000 Subject: net: make SO_BUSY_POLL available to all users After commit 217f69743681 ("net: busy-poll: allow preemption in sk_busy_loop()"), a thread willing to use busy polling is not hurting other threads anymore in a non preempt kernel. I think it is safe to remove CAP_NET_ADMIN check. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20230406194634.1804691-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/sock.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/core/sock.c b/net/core/sock.c index c25888795390..5440e67bcfe3 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1396,15 +1396,10 @@ set_sndbuf: #ifdef CONFIG_NET_RX_BUSY_POLL case SO_BUSY_POLL: - /* allow unprivileged users to decrease the value */ - if ((val > sk->sk_ll_usec) && !sockopt_capable(CAP_NET_ADMIN)) - ret = -EPERM; - else { - if (val < 0) - ret = -EINVAL; - else - WRITE_ONCE(sk->sk_ll_usec, val); - } + if (val < 0) + ret = -EINVAL; + else + WRITE_ONCE(sk->sk_ll_usec, val); break; case SO_PREFER_BUSY_POLL: if (valbool && !sockopt_capable(CAP_NET_ADMIN)) -- cgit v1.2.3 From 5a17818682cf43ad0fdd6035945f3b7a8c9dc5e9 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Thu, 6 Apr 2023 14:42:46 +0300 Subject: net: dsa: replace NETDEV_PRE_CHANGE_HWTSTAMP notifier with a stub There was a sort of rush surrounding commit 88c0a6b503b7 ("net: create a netdev notifier for DSA to reject PTP on DSA master"), due to a desire to convert DSA's attempt to deny TX timestamping on a DSA master to something that doesn't block the kernel-wide API conversion from ndo_eth_ioctl() to ndo_hwtstamp_set(). What was required was a mechanism that did not depend on ndo_eth_ioctl(), and what was provided was a mechanism that did not depend on ndo_eth_ioctl(), while at the same time introducing something that wasn't absolutely necessary - a new netdev notifier. There have been objections from Jakub Kicinski that using notifiers in general when they are not absolutely necessary creates complications to the control flow and difficulties to maintainers who look at the code. So there is a desire to not use notifiers. In addition to that, the notifier chain gets called even if there is no DSA in the system and no one is interested in applying any restriction. Take the model of udp_tunnel_nic_ops and introduce a stub mechanism, through which net/core/dev_ioctl.c can call into DSA even when CONFIG_NET_DSA=m. Compared to the code that existed prior to the notifier conversion, aka what was added in commits: - 4cfab3566710 ("net: dsa: Add wrappers for overloaded ndo_ops") - 3369afba1e46 ("net: Call into DSA netdevice_ops wrappers") this is different because we are not overloading any struct net_device_ops of the DSA master anymore, but rather, we are exposing a rather specific functionality which is orthogonal to which API is used to enable it - ndo_eth_ioctl() or ndo_hwtstamp_set(). Also, what is similar is that both approaches use function pointers to get from built-in code to DSA. There is no point in replicating the function pointers towards __dsa_master_hwtstamp_validate() once for every CPU port (dev->dsa_ptr). Instead, it is sufficient to introduce a singleton struct dsa_stubs, built into the kernel, which contains a single function pointer to __dsa_master_hwtstamp_validate(). I find this approach preferable to what we had originally, because dev->dsa_ptr->netdev_ops->ndo_do_ioctl() used to require going through struct dsa_port (dev->dsa_ptr), and so, this was incompatible with any attempts to add any data encapsulation and hide DSA data structures from the outside world. Link: https://lore.kernel.org/netdev/20230403083019.120b72fd@kernel.org/ Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/linux/netdevice.h | 6 ------ include/net/dsa_stubs.h | 48 +++++++++++++++++++++++++++++++++++++++++++++++ net/Makefile | 2 +- net/core/dev.c | 2 +- net/core/dev_ioctl.c | 12 ++---------- net/dsa/Makefile | 6 ++++++ net/dsa/dsa.c | 19 +++++++++++++++++++ net/dsa/master.c | 2 +- net/dsa/master.h | 2 +- net/dsa/slave.c | 11 ----------- net/dsa/stubs.c | 10 ++++++++++ 11 files changed, 89 insertions(+), 31 deletions(-) create mode 100644 include/net/dsa_stubs.h create mode 100644 net/dsa/stubs.c (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index a740be3bb911..1c25b39681b3 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2878,7 +2878,6 @@ enum netdev_cmd { NETDEV_OFFLOAD_XSTATS_REPORT_USED, NETDEV_OFFLOAD_XSTATS_REPORT_DELTA, NETDEV_XDP_FEAT_CHANGE, - NETDEV_PRE_CHANGE_HWTSTAMP, }; const char *netdev_cmd_to_name(enum netdev_cmd cmd); @@ -2929,11 +2928,6 @@ struct netdev_notifier_pre_changeaddr_info { const unsigned char *dev_addr; }; -struct netdev_notifier_hwtstamp_info { - struct netdev_notifier_info info; /* must be first */ - struct kernel_hwtstamp_config *config; -}; - enum netdev_offload_xstats_type { NETDEV_OFFLOAD_XSTATS_TYPE_L3 = 1, }; diff --git a/include/net/dsa_stubs.h b/include/net/dsa_stubs.h new file mode 100644 index 000000000000..361811750a54 --- /dev/null +++ b/include/net/dsa_stubs.h @@ -0,0 +1,48 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * include/net/dsa_stubs.h - Stubs for the Distributed Switch Architecture framework + */ + +#include +#include +#include +#include + +#if IS_ENABLED(CONFIG_NET_DSA) + +extern const struct dsa_stubs *dsa_stubs; + +struct dsa_stubs { + int (*master_hwtstamp_validate)(struct net_device *dev, + const struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack); +}; + +static inline int dsa_master_hwtstamp_validate(struct net_device *dev, + const struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + if (!netdev_uses_dsa(dev)) + return 0; + + /* rtnl_lock() is a sufficient guarantee, because as long as + * netdev_uses_dsa() returns true, the dsa_core module is still + * registered, and so, dsa_unregister_stubs() couldn't have run. + * For netdev_uses_dsa() to start returning false, it would imply that + * dsa_master_teardown() has executed, which requires rtnl_lock(). + */ + ASSERT_RTNL(); + + return dsa_stubs->master_hwtstamp_validate(dev, config, extack); +} + +#else + +static inline int dsa_master_hwtstamp_validate(struct net_device *dev, + const struct kernel_hwtstamp_config *config, + struct netlink_ext_ack *extack) +{ + return 0; +} + +#endif diff --git a/net/Makefile b/net/Makefile index 0914bea9c335..87592009366f 100644 --- a/net/Makefile +++ b/net/Makefile @@ -24,7 +24,7 @@ obj-$(CONFIG_PACKET) += packet/ obj-$(CONFIG_NET_KEY) += key/ obj-$(CONFIG_BRIDGE) += bridge/ obj-$(CONFIG_NET_DEVLINK) += devlink/ -obj-$(CONFIG_NET_DSA) += dsa/ +obj-y += dsa/ obj-$(CONFIG_ATALK) += appletalk/ obj-$(CONFIG_X25) += x25/ obj-$(CONFIG_LAPB) += lapb/ diff --git a/net/core/dev.c b/net/core/dev.c index 7ce5985be84b..480600a075ce 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1612,7 +1612,7 @@ const char *netdev_cmd_to_name(enum netdev_cmd cmd) N(SVLAN_FILTER_PUSH_INFO) N(SVLAN_FILTER_DROP_INFO) N(PRE_CHANGEADDR) N(OFFLOAD_XSTATS_ENABLE) N(OFFLOAD_XSTATS_DISABLE) N(OFFLOAD_XSTATS_REPORT_USED) N(OFFLOAD_XSTATS_REPORT_DELTA) - N(XDP_FEAT_CHANGE) N(PRE_CHANGE_HWTSTAMP) + N(XDP_FEAT_CHANGE) } #undef N return "UNKNOWN_NETDEV_EVENT"; diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c index 6d772837eb3f..3730945ee294 100644 --- a/net/core/dev_ioctl.c +++ b/net/core/dev_ioctl.c @@ -7,7 +7,7 @@ #include #include #include -#include +#include #include #include "dev.h" @@ -259,9 +259,6 @@ static int dev_get_hwtstamp(struct net_device *dev, struct ifreq *ifr) static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr) { - struct netdev_notifier_hwtstamp_info info = { - .info.dev = dev, - }; struct kernel_hwtstamp_config kernel_cfg; struct netlink_ext_ack extack = {}; struct hwtstamp_config cfg; @@ -276,12 +273,7 @@ static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr) if (err) return err; - info.info.extack = &extack; - info.config = &kernel_cfg; - - err = call_netdevice_notifiers_info(NETDEV_PRE_CHANGE_HWTSTAMP, - &info.info); - err = notifier_to_errno(err); + err = dsa_master_hwtstamp_validate(dev, &kernel_cfg, &extack); if (err) { if (extack._msg) netdev_err(dev, "%s\n", extack._msg); diff --git a/net/dsa/Makefile b/net/dsa/Makefile index cc7e93a562fe..3835de286116 100644 --- a/net/dsa/Makefile +++ b/net/dsa/Makefile @@ -1,4 +1,10 @@ # SPDX-License-Identifier: GPL-2.0 + +# the stubs are built-in whenever DSA is built-in or module +ifdef CONFIG_NET_DSA +obj-y := stubs.o +endif + # the core obj-$(CONFIG_NET_DSA) += dsa_core.o dsa_core-y += \ diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index e5f156940c67..ab1afe67fd18 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include "devlink.h" @@ -1702,6 +1703,20 @@ bool dsa_mdb_present_in_other_db(struct dsa_switch *ds, int port, } EXPORT_SYMBOL_GPL(dsa_mdb_present_in_other_db); +static const struct dsa_stubs __dsa_stubs = { + .master_hwtstamp_validate = __dsa_master_hwtstamp_validate, +}; + +static void dsa_register_stubs(void) +{ + dsa_stubs = &__dsa_stubs; +} + +static void dsa_unregister_stubs(void) +{ + dsa_stubs = NULL; +} + static int __init dsa_init_module(void) { int rc; @@ -1721,6 +1736,8 @@ static int __init dsa_init_module(void) if (rc) goto netlink_register_fail; + dsa_register_stubs(); + return 0; netlink_register_fail: @@ -1735,6 +1752,8 @@ module_init(dsa_init_module); static void __exit dsa_cleanup_module(void) { + dsa_unregister_stubs(); + rtnl_link_unregister(&dsa_link_ops); dsa_slave_unregister_notifier(); diff --git a/net/dsa/master.c b/net/dsa/master.c index c2cabe6248b1..6be89ab0cc01 100644 --- a/net/dsa/master.c +++ b/net/dsa/master.c @@ -198,7 +198,7 @@ static void dsa_master_get_strings(struct net_device *dev, uint32_t stringset, /* Deny PTP operations on master if there is at least one switch in the tree * that is PTP capable. */ -int dsa_master_pre_change_hwtstamp(struct net_device *dev, +int __dsa_master_hwtstamp_validate(struct net_device *dev, const struct kernel_hwtstamp_config *config, struct netlink_ext_ack *extack) { diff --git a/net/dsa/master.h b/net/dsa/master.h index 80842f4e27f7..76e39d3ec909 100644 --- a/net/dsa/master.h +++ b/net/dsa/master.h @@ -15,7 +15,7 @@ int dsa_master_lag_setup(struct net_device *lag_dev, struct dsa_port *cpu_dp, struct netlink_ext_ack *extack); void dsa_master_lag_teardown(struct net_device *lag_dev, struct dsa_port *cpu_dp); -int dsa_master_pre_change_hwtstamp(struct net_device *dev, +int __dsa_master_hwtstamp_validate(struct net_device *dev, const struct kernel_hwtstamp_config *config, struct netlink_ext_ack *extack); diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 8abc1658ac47..165bb2cb8431 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -3289,7 +3289,6 @@ static int dsa_master_changeupper(struct net_device *dev, static int dsa_slave_netdevice_event(struct notifier_block *nb, unsigned long event, void *ptr) { - struct netlink_ext_ack *extack = netdev_notifier_info_to_extack(ptr); struct net_device *dev = netdev_notifier_info_to_dev(ptr); switch (event) { @@ -3419,16 +3418,6 @@ static int dsa_slave_netdevice_event(struct notifier_block *nb, return NOTIFY_OK; } - case NETDEV_PRE_CHANGE_HWTSTAMP: { - struct netdev_notifier_hwtstamp_info *info = ptr; - int err; - - if (!netdev_uses_dsa(dev)) - return NOTIFY_DONE; - - err = dsa_master_pre_change_hwtstamp(dev, info->config, extack); - return notifier_from_errno(err); - } default: break; } diff --git a/net/dsa/stubs.c b/net/dsa/stubs.c new file mode 100644 index 000000000000..2ed8a6c85fbf --- /dev/null +++ b/net/dsa/stubs.c @@ -0,0 +1,10 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Stubs for DSA functionality called by the core network stack. + * These are necessary because CONFIG_NET_DSA can be a module, and built-in + * code cannot directly call symbols exported by modules. + */ +#include + +const struct dsa_stubs *dsa_stubs; +EXPORT_SYMBOL_GPL(dsa_stubs); -- cgit v1.2.3 From 19cf60bf63cbaf5262eac400c707966e19999b83 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Fri, 24 Mar 2023 10:57:55 -0700 Subject: Bluetooth: hci_conn: Fix not cleaning up on LE Connection failure hci_connect_le_scan_cleanup shall always be invoked to cleanup the states and re-enable passive scanning if necessary, otherwise it may cause the pending action to stay active causing multiple attempts to connect. Fixes: 9b3628d79b46 ("Bluetooth: hci_sync: Cleanup hci_conn if it cannot be aborted") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 52 ++++++++++++++++++++++-------------------------- 1 file changed, 24 insertions(+), 28 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 17b946f9ba31..5af3f6b011c9 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -68,7 +68,7 @@ static const struct sco_param esco_param_msbc[] = { }; /* This function requires the caller holds hdev->lock */ -static void hci_connect_le_scan_cleanup(struct hci_conn *conn) +static void hci_connect_le_scan_cleanup(struct hci_conn *conn, u8 status) { struct hci_conn_params *params; struct hci_dev *hdev = conn->hdev; @@ -88,9 +88,28 @@ static void hci_connect_le_scan_cleanup(struct hci_conn *conn) params = hci_pend_le_action_lookup(&hdev->pend_le_conns, bdaddr, bdaddr_type); - if (!params || !params->explicit_connect) + if (!params) return; + if (params->conn) { + hci_conn_drop(params->conn); + hci_conn_put(params->conn); + params->conn = NULL; + } + + if (!params->explicit_connect) + return; + + /* If the status indicates successful cancellation of + * the attempt (i.e. Unknown Connection Id) there's no point of + * notifying failure since we'll go back to keep trying to + * connect. The only exception is explicit connect requests + * where a timeout + cancel does indicate an actual failure. + */ + if (status && status != HCI_ERROR_UNKNOWN_CONN_ID) + mgmt_connect_failed(hdev, &conn->dst, conn->type, + conn->dst_type, status); + /* The connection attempt was doing scan for new RPA, and is * in scan phase. If params are not associated with any other * autoconnect action, remove them completely. If they are, just unmark @@ -178,7 +197,7 @@ static void le_scan_cleanup(struct work_struct *work) rcu_read_unlock(); if (c == conn) { - hci_connect_le_scan_cleanup(conn); + hci_connect_le_scan_cleanup(conn, 0x00); hci_conn_cleanup(conn); } @@ -1179,31 +1198,8 @@ EXPORT_SYMBOL(hci_get_route); static void hci_le_conn_failed(struct hci_conn *conn, u8 status) { struct hci_dev *hdev = conn->hdev; - struct hci_conn_params *params; - params = hci_pend_le_action_lookup(&hdev->pend_le_conns, &conn->dst, - conn->dst_type); - if (params && params->conn) { - hci_conn_drop(params->conn); - hci_conn_put(params->conn); - params->conn = NULL; - } - - /* If the status indicates successful cancellation of - * the attempt (i.e. Unknown Connection Id) there's no point of - * notifying failure since we'll go back to keep trying to - * connect. The only exception is explicit connect requests - * where a timeout + cancel does indicate an actual failure. - */ - if (status != HCI_ERROR_UNKNOWN_CONN_ID || - (params && params->explicit_connect)) - mgmt_connect_failed(hdev, &conn->dst, conn->type, - conn->dst_type, status); - - /* Since we may have temporarily stopped the background scanning in - * favor of connection establishment, we should restart it. - */ - hci_update_passive_scan(hdev); + hci_connect_le_scan_cleanup(conn, status); /* Enable advertising in case this was a failed connection * attempt as a peripheral. @@ -1240,7 +1236,7 @@ static void create_le_conn_complete(struct hci_dev *hdev, void *data, int err) hci_dev_lock(hdev); if (!err) { - hci_connect_le_scan_cleanup(conn); + hci_connect_le_scan_cleanup(conn, 0x00); goto done; } -- cgit v1.2.3 From b62e72200eaad523f08d8319bba50fc652e032a8 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Fri, 24 Mar 2023 13:18:20 -0700 Subject: Bluetooth: Fix printing errors if LE Connection times out This fixes errors like bellow when LE Connection times out since that is actually not a controller error: Bluetooth: hci0: Opcode 0x200d failed: -110 Bluetooth: hci0: request failed to create LE connection: err -110 Instead the code shall properly detect if -ETIMEDOUT is returned and send HCI_OP_LE_CREATE_CONN_CANCEL to give up on the connection. Link: https://github.com/bluez/bluez/issues/340 Fixes: 8e8b92ee60de ("Bluetooth: hci_sync: Add hci_le_create_conn_sync") Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 1 + net/bluetooth/hci_conn.c | 7 +++++-- net/bluetooth/hci_event.c | 16 ++++++---------- net/bluetooth/hci_sync.c | 13 ++++++++++--- 4 files changed, 22 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 6ed9b4d546a7..d5311ceb21c6 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -954,6 +954,7 @@ enum { HCI_CONN_STK_ENCRYPT, HCI_CONN_AUTH_INITIATOR, HCI_CONN_DROP, + HCI_CONN_CANCEL, HCI_CONN_PARAM_REMOVAL_PEND, HCI_CONN_NEW_LINK_KEY, HCI_CONN_SCANNING, diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 5af3f6b011c9..e4aee5950c36 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1233,6 +1233,8 @@ static void create_le_conn_complete(struct hci_dev *hdev, void *data, int err) { struct hci_conn *conn = data; + bt_dev_dbg(hdev, "err %d", err); + hci_dev_lock(hdev); if (!err) { @@ -1240,8 +1242,6 @@ static void create_le_conn_complete(struct hci_dev *hdev, void *data, int err) goto done; } - bt_dev_err(hdev, "request failed to create LE connection: err %d", err); - /* Check if connection is still pending */ if (conn != hci_lookup_le_connect(hdev)) goto done; @@ -2771,6 +2771,9 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason) { int r = 0; + if (test_and_set_bit(HCI_CONN_CANCEL, &conn->flags)) + return 0; + switch (conn->state) { case BT_CONNECTED: case BT_CONFIG: diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index ad92a4be5851..e68f2a7d863a 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -2881,16 +2881,6 @@ static void cs_le_create_conn(struct hci_dev *hdev, bdaddr_t *peer_addr, conn->resp_addr_type = peer_addr_type; bacpy(&conn->resp_addr, peer_addr); - - /* We don't want the connection attempt to stick around - * indefinitely since LE doesn't have a page timeout concept - * like BR/EDR. Set a timer for any connection that doesn't use - * the accept list for connecting. - */ - if (filter_policy == HCI_LE_USE_PEER_ADDR) - queue_delayed_work(conn->hdev->workqueue, - &conn->le_conn_timeout, - conn->conn_timeout); } static void hci_cs_le_create_conn(struct hci_dev *hdev, u8 status) @@ -5902,6 +5892,12 @@ static void le_conn_complete_evt(struct hci_dev *hdev, u8 status, if (status) goto unlock; + /* Drop the connection if it has been aborted */ + if (test_bit(HCI_CONN_CANCEL, &conn->flags)) { + hci_conn_drop(conn); + goto unlock; + } + if (conn->dst_type == ADDR_LE_DEV_PUBLIC) addr_type = BDADDR_LE_PUBLIC; else diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 5a6aa1627791..632be1267288 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -246,8 +246,9 @@ int __hci_cmd_sync_status_sk(struct hci_dev *hdev, u16 opcode, u32 plen, skb = __hci_cmd_sync_sk(hdev, opcode, plen, param, event, timeout, sk); if (IS_ERR(skb)) { - bt_dev_err(hdev, "Opcode 0x%4x failed: %ld", opcode, - PTR_ERR(skb)); + if (!event) + bt_dev_err(hdev, "Opcode 0x%4x failed: %ld", opcode, + PTR_ERR(skb)); return PTR_ERR(skb); } @@ -5126,8 +5127,11 @@ static int hci_le_connect_cancel_sync(struct hci_dev *hdev, if (test_bit(HCI_CONN_SCANNING, &conn->flags)) return 0; + if (test_and_set_bit(HCI_CONN_CANCEL, &conn->flags)) + return 0; + return __hci_cmd_sync_status(hdev, HCI_OP_LE_CREATE_CONN_CANCEL, - 6, &conn->dst, HCI_CMD_TIMEOUT); + 0, NULL, HCI_CMD_TIMEOUT); } static int hci_connect_cancel_sync(struct hci_dev *hdev, struct hci_conn *conn) @@ -6102,6 +6106,9 @@ int hci_le_create_conn_sync(struct hci_dev *hdev, struct hci_conn *conn) conn->conn_timeout, NULL); done: + if (err == -ETIMEDOUT) + hci_le_connect_cancel_sync(hdev, conn); + /* Re-enable advertising after the connection attempt is finished. */ hci_resume_advertising_sync(hdev); return err; -- cgit v1.2.3 From c95930abd687fcd1aa040dc4fe90dff947916460 Mon Sep 17 00:00:00 2001 From: Min Li Date: Sat, 4 Mar 2023 22:23:30 +0800 Subject: Bluetooth: Fix race condition in hidp_session_thread There is a potential race condition in hidp_session_thread that may lead to use-after-free. For instance, the timer is active while hidp_del_timer is called in hidp_session_thread(). After hidp_session_put, then 'session' will be freed, causing kernel panic when hidp_idle_timeout is running. The solution is to use del_timer_sync instead of del_timer. Here is the call trace: ? hidp_session_probe+0x780/0x780 call_timer_fn+0x2d/0x1e0 __run_timers.part.0+0x569/0x940 hidp_session_probe+0x780/0x780 call_timer_fn+0x1e0/0x1e0 ktime_get+0x5c/0xf0 lapic_next_deadline+0x2c/0x40 clockevents_program_event+0x205/0x320 run_timer_softirq+0xa9/0x1b0 __do_softirq+0x1b9/0x641 __irq_exit_rcu+0xdc/0x190 irq_exit_rcu+0xe/0x20 sysvec_apic_timer_interrupt+0xa1/0xc0 Cc: stable@vger.kernel.org Signed-off-by: Min Li Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hidp/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index bed1a7b9205c..707f229f896a 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -433,7 +433,7 @@ static void hidp_set_timer(struct hidp_session *session) static void hidp_del_timer(struct hidp_session *session) { if (session->idle_to > 0) - del_timer(&session->timer); + del_timer_sync(&session->timer); } static void hidp_process_report(struct hidp_session *session, int type, -- cgit v1.2.3 From 9a8ec9e8ebb5a7c0cfbce2d6b4a6b67b2b78e8f3 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Thu, 30 Mar 2023 14:15:50 -0700 Subject: Bluetooth: SCO: Fix possible circular locking dependency on sco_connect_cfm This attempts to fix the following trace: ====================================================== WARNING: possible circular locking dependency detected 6.3.0-rc2-g0b93eeba4454 #4703 Not tainted ------------------------------------------------------ kworker/u3:0/46 is trying to acquire lock: ffff888001fd9130 (sk_lock-AF_BLUETOOTH-BTPROTO_SCO){+.+.}-{0:0}, at: sco_connect_cfm+0x118/0x4a0 but task is already holding lock: ffffffff831e3340 (hci_cb_list_lock){+.+.}-{3:3}, at: hci_sync_conn_complete_evt+0x1ad/0x3d0 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #2 (hci_cb_list_lock){+.+.}-{3:3}: __mutex_lock+0x13b/0xcc0 hci_sync_conn_complete_evt+0x1ad/0x3d0 hci_event_packet+0x55c/0x7c0 hci_rx_work+0x34c/0xa00 process_one_work+0x575/0x910 worker_thread+0x89/0x6f0 kthread+0x14e/0x180 ret_from_fork+0x2b/0x50 -> #1 (&hdev->lock){+.+.}-{3:3}: __mutex_lock+0x13b/0xcc0 sco_sock_connect+0xfc/0x630 __sys_connect+0x197/0x1b0 __x64_sys_connect+0x37/0x50 do_syscall_64+0x42/0x90 entry_SYSCALL_64_after_hwframe+0x70/0xda -> #0 (sk_lock-AF_BLUETOOTH-BTPROTO_SCO){+.+.}-{0:0}: __lock_acquire+0x18cc/0x3740 lock_acquire+0x151/0x3a0 lock_sock_nested+0x32/0x80 sco_connect_cfm+0x118/0x4a0 hci_sync_conn_complete_evt+0x1e6/0x3d0 hci_event_packet+0x55c/0x7c0 hci_rx_work+0x34c/0xa00 process_one_work+0x575/0x910 worker_thread+0x89/0x6f0 kthread+0x14e/0x180 ret_from_fork+0x2b/0x50 other info that might help us debug this: Chain exists of: sk_lock-AF_BLUETOOTH-BTPROTO_SCO --> &hdev->lock --> hci_cb_list_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(hci_cb_list_lock); lock(&hdev->lock); lock(hci_cb_list_lock); lock(sk_lock-AF_BLUETOOTH-BTPROTO_SCO); *** DEADLOCK *** 4 locks held by kworker/u3:0/46: #0: ffff8880028d1130 ((wq_completion)hci0#2){+.+.}-{0:0}, at: process_one_work+0x4c0/0x910 #1: ffff8880013dfde0 ((work_completion)(&hdev->rx_work)){+.+.}-{0:0}, at: process_one_work+0x4c0/0x910 #2: ffff8880025d8070 (&hdev->lock){+.+.}-{3:3}, at: hci_sync_conn_complete_evt+0xa6/0x3d0 #3: ffffffffb79e3340 (hci_cb_list_lock){+.+.}-{3:3}, at: hci_sync_conn_complete_evt+0x1ad/0x3d0 Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/sco.c | 69 +++++++++++++++++++++++++++++++---------------------- 1 file changed, 40 insertions(+), 29 deletions(-) (limited to 'net') diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index 1111da4e2f2b..f3a5ab9e4fa4 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -235,27 +235,41 @@ static int sco_chan_add(struct sco_conn *conn, struct sock *sk, return err; } -static int sco_connect(struct hci_dev *hdev, struct sock *sk) +static int sco_connect(struct sock *sk) { struct sco_conn *conn; struct hci_conn *hcon; + struct hci_dev *hdev; int err, type; BT_DBG("%pMR -> %pMR", &sco_pi(sk)->src, &sco_pi(sk)->dst); + hdev = hci_get_route(&sco_pi(sk)->dst, &sco_pi(sk)->src, BDADDR_BREDR); + if (!hdev) + return -EHOSTUNREACH; + + hci_dev_lock(hdev); + if (lmp_esco_capable(hdev) && !disable_esco) type = ESCO_LINK; else type = SCO_LINK; if (sco_pi(sk)->setting == BT_VOICE_TRANSPARENT && - (!lmp_transp_capable(hdev) || !lmp_esco_capable(hdev))) - return -EOPNOTSUPP; + (!lmp_transp_capable(hdev) || !lmp_esco_capable(hdev))) { + err = -EOPNOTSUPP; + goto unlock; + } hcon = hci_connect_sco(hdev, type, &sco_pi(sk)->dst, sco_pi(sk)->setting, &sco_pi(sk)->codec); - if (IS_ERR(hcon)) - return PTR_ERR(hcon); + if (IS_ERR(hcon)) { + err = PTR_ERR(hcon); + goto unlock; + } + + hci_dev_unlock(hdev); + hci_dev_put(hdev); conn = sco_conn_add(hcon); if (!conn) { @@ -263,13 +277,15 @@ static int sco_connect(struct hci_dev *hdev, struct sock *sk) return -ENOMEM; } - /* Update source addr of the socket */ - bacpy(&sco_pi(sk)->src, &hcon->src); - err = sco_chan_add(conn, sk, NULL); if (err) return err; + lock_sock(sk); + + /* Update source addr of the socket */ + bacpy(&sco_pi(sk)->src, &hcon->src); + if (hcon->state == BT_CONNECTED) { sco_sock_clear_timer(sk); sk->sk_state = BT_CONNECTED; @@ -278,6 +294,13 @@ static int sco_connect(struct hci_dev *hdev, struct sock *sk) sco_sock_set_timer(sk, sk->sk_sndtimeo); } + release_sock(sk); + + return err; + +unlock: + hci_dev_unlock(hdev); + hci_dev_put(hdev); return err; } @@ -565,7 +588,6 @@ static int sco_sock_connect(struct socket *sock, struct sockaddr *addr, int alen { struct sockaddr_sco *sa = (struct sockaddr_sco *) addr; struct sock *sk = sock->sk; - struct hci_dev *hdev; int err; BT_DBG("sk %p", sk); @@ -574,37 +596,26 @@ static int sco_sock_connect(struct socket *sock, struct sockaddr *addr, int alen addr->sa_family != AF_BLUETOOTH) return -EINVAL; - lock_sock(sk); - if (sk->sk_state != BT_OPEN && sk->sk_state != BT_BOUND) { - err = -EBADFD; - goto done; - } + if (sk->sk_state != BT_OPEN && sk->sk_state != BT_BOUND) + return -EBADFD; - if (sk->sk_type != SOCK_SEQPACKET) { + if (sk->sk_type != SOCK_SEQPACKET) err = -EINVAL; - goto done; - } - - hdev = hci_get_route(&sa->sco_bdaddr, &sco_pi(sk)->src, BDADDR_BREDR); - if (!hdev) { - err = -EHOSTUNREACH; - goto done; - } - hci_dev_lock(hdev); + lock_sock(sk); /* Set destination address and psm */ bacpy(&sco_pi(sk)->dst, &sa->sco_bdaddr); + release_sock(sk); - err = sco_connect(hdev, sk); - hci_dev_unlock(hdev); - hci_dev_put(hdev); + err = sco_connect(sk); if (err) - goto done; + return err; + + lock_sock(sk); err = bt_sock_wait_state(sk, BT_CONNECTED, sock_sndtimeo(sk, flags & O_NONBLOCK)); -done: release_sock(sk); return err; } -- cgit v1.2.3 From 975abc0c90fc485ff9b4a6afa475c3b1398d5d47 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Thu, 30 Mar 2023 14:45:03 -0700 Subject: Bluetooth: SCO: Fix possible circular locking dependency sco_sock_getsockopt This attempts to fix the following trace: ====================================================== WARNING: possible circular locking dependency detected 6.3.0-rc2-g68fcb3a7bf97 #4706 Not tainted ------------------------------------------------------ sco-tester/31 is trying to acquire lock: ffff8880025b8070 (&hdev->lock){+.+.}-{3:3}, at: sco_sock_getsockopt+0x1fc/0xa90 but task is already holding lock: ffff888001eeb130 (sk_lock-AF_BLUETOOTH-BTPROTO_SCO){+.+.}-{0:0}, at: sco_sock_getsockopt+0x104/0xa90 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #2 (sk_lock-AF_BLUETOOTH-BTPROTO_SCO){+.+.}-{0:0}: lock_sock_nested+0x32/0x80 sco_connect_cfm+0x118/0x4a0 hci_sync_conn_complete_evt+0x1e6/0x3d0 hci_event_packet+0x55c/0x7c0 hci_rx_work+0x34c/0xa00 process_one_work+0x575/0x910 worker_thread+0x89/0x6f0 kthread+0x14e/0x180 ret_from_fork+0x2b/0x50 -> #1 (hci_cb_list_lock){+.+.}-{3:3}: __mutex_lock+0x13b/0xcc0 hci_sync_conn_complete_evt+0x1ad/0x3d0 hci_event_packet+0x55c/0x7c0 hci_rx_work+0x34c/0xa00 process_one_work+0x575/0x910 worker_thread+0x89/0x6f0 kthread+0x14e/0x180 ret_from_fork+0x2b/0x50 -> #0 (&hdev->lock){+.+.}-{3:3}: __lock_acquire+0x18cc/0x3740 lock_acquire+0x151/0x3a0 __mutex_lock+0x13b/0xcc0 sco_sock_getsockopt+0x1fc/0xa90 __sys_getsockopt+0xe9/0x190 __x64_sys_getsockopt+0x5b/0x70 do_syscall_64+0x42/0x90 entry_SYSCALL_64_after_hwframe+0x70/0xda other info that might help us debug this: Chain exists of: &hdev->lock --> hci_cb_list_lock --> sk_lock-AF_BLUETOOTH-BTPROTO_SCO Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(sk_lock-AF_BLUETOOTH-BTPROTO_SCO); lock(hci_cb_list_lock); lock(sk_lock-AF_BLUETOOTH-BTPROTO_SCO); lock(&hdev->lock); *** DEADLOCK *** 1 lock held by sco-tester/31: #0: ffff888001eeb130 (sk_lock-AF_BLUETOOTH-BTPROTO_SCO){+.+.}-{0:0}, at: sco_sock_getsockopt+0x104/0xa90 Fixes: 248733e87d50 ("Bluetooth: Allow querying of supported offload codecs over SCO socket") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/sco.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index f3a5ab9e4fa4..cd1a27ac555d 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -1140,6 +1140,8 @@ static int sco_sock_getsockopt(struct socket *sock, int level, int optname, break; } + release_sock(sk); + /* find total buffer size required to copy codec + caps */ hci_dev_lock(hdev); list_for_each_entry(c, &hdev->local_codecs, list) { @@ -1157,15 +1159,13 @@ static int sco_sock_getsockopt(struct socket *sock, int level, int optname, buf_len += sizeof(struct bt_codecs); if (buf_len > len) { hci_dev_put(hdev); - err = -ENOBUFS; - break; + return -ENOBUFS; } ptr = optval; if (put_user(num_codecs, ptr)) { hci_dev_put(hdev); - err = -EFAULT; - break; + return -EFAULT; } ptr += sizeof(num_codecs); @@ -1205,12 +1205,14 @@ static int sco_sock_getsockopt(struct socket *sock, int level, int optname, ptr += len; } - if (!err && put_user(buf_len, optlen)) - err = -EFAULT; - hci_dev_unlock(hdev); hci_dev_put(hdev); + lock_sock(sk); + + if (!err && put_user(buf_len, optlen)) + err = -EFAULT; + break; default: -- cgit v1.2.3 From 5dc7d23e167e2882ef118456ceccd57873e876d8 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 3 Apr 2023 14:19:14 -0700 Subject: Bluetooth: hci_conn: Fix possible UAF This fixes the following trace: ================================================================== BUG: KASAN: slab-use-after-free in hci_conn_del+0xba/0x3a0 Write of size 8 at addr ffff88800208e9c8 by task iso-tester/31 CPU: 0 PID: 31 Comm: iso-tester Not tainted 6.3.0-rc2-g991aa4a69a47 #4716 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.1-2.fc36 04/01/2014 Call Trace: dump_stack_lvl+0x1d/0x70 print_report+0xce/0x610 ? __virt_addr_valid+0xd4/0x150 ? hci_conn_del+0xba/0x3a0 kasan_report+0xdd/0x110 ? hci_conn_del+0xba/0x3a0 hci_conn_del+0xba/0x3a0 hci_conn_hash_flush+0xf2/0x120 hci_dev_close_sync+0x388/0x920 hci_unregister_dev+0x122/0x260 vhci_release+0x4f/0x90 __fput+0x102/0x430 task_work_run+0xf1/0x160 ? __pfx_task_work_run+0x10/0x10 ? mark_held_locks+0x24/0x90 exit_to_user_mode_prepare+0x170/0x180 syscall_exit_to_user_mode+0x19/0x50 do_syscall_64+0x4e/0x90 entry_SYSCALL_64_after_hwframe+0x70/0xda Fixes: 0f00cd322d22 ("Bluetooth: Free potentially unfreed SCO connection") Link: https://syzkaller.appspot.com/bug?extid=8bb72f86fc823817bc5d Cc: Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 30 ++++++++++++++++++++++++------ 1 file changed, 24 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index e4aee5950c36..8455ba141ee6 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1068,6 +1068,17 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, return conn; } +static bool hci_conn_unlink(struct hci_conn *conn) +{ + if (!conn->link) + return false; + + conn->link->link = NULL; + conn->link = NULL; + + return true; +} + int hci_conn_del(struct hci_conn *conn) { struct hci_dev *hdev = conn->hdev; @@ -1079,15 +1090,16 @@ int hci_conn_del(struct hci_conn *conn) cancel_delayed_work_sync(&conn->idle_work); if (conn->type == ACL_LINK) { - struct hci_conn *sco = conn->link; - if (sco) { - sco->link = NULL; + struct hci_conn *link = conn->link; + + if (link) { + hci_conn_unlink(conn); /* Due to race, SCO connection might be not established * yet at this point. Delete it now, otherwise it is * possible for it to be stuck and can't be deleted. */ - if (sco->handle == HCI_CONN_HANDLE_UNSET) - hci_conn_del(sco); + if (link->handle == HCI_CONN_HANDLE_UNSET) + hci_conn_del(link); } /* Unacked frames */ @@ -1103,7 +1115,7 @@ int hci_conn_del(struct hci_conn *conn) struct hci_conn *acl = conn->link; if (acl) { - acl->link = NULL; + hci_conn_unlink(conn); hci_conn_drop(acl); } @@ -2434,6 +2446,12 @@ void hci_conn_hash_flush(struct hci_dev *hdev) c->state = BT_CLOSED; hci_disconn_cfm(c, HCI_ERROR_LOCAL_HOST_TERM); + + /* Unlink before deleting otherwise it is possible that + * hci_conn_del removes the link which may cause the list to + * contain items already freed. + */ + hci_conn_unlink(c); hci_conn_del(c); } } -- cgit v1.2.3 From d2e4f1b1cba8742db66aaf77374cab7c0c7c8656 Mon Sep 17 00:00:00 2001 From: Claudia Draghicescu Date: Wed, 5 Apr 2023 14:19:18 +0300 Subject: Bluetooth: Set ISO Data Path on broadcast sink This patch enables ISO data rx on broadcast sink. Fixes: eca0ae4aea66 ("Bluetooth: Add initial implementation of BIS connections") Signed-off-by: Claudia Draghicescu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_event.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index e68f2a7d863a..e87c928c9e17 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -6991,7 +6991,7 @@ static void hci_le_big_sync_established_evt(struct hci_dev *hdev, void *data, bis->iso_qos.in.latency = le16_to_cpu(ev->interval) * 125 / 100; bis->iso_qos.in.sdu = le16_to_cpu(ev->max_pdu); - hci_connect_cfm(bis, ev->status); + hci_iso_setup_path(bis); } hci_dev_unlock(hdev); -- cgit v1.2.3 From a2a9339e1c9deb7e1e079e12e27a0265aea8421a Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Thu, 6 Apr 2023 09:33:09 -0700 Subject: Bluetooth: L2CAP: Fix use-after-free in l2cap_disconnect_{req,rsp} Similar to commit d0be8347c623 ("Bluetooth: L2CAP: Fix use-after-free caused by l2cap_chan_put"), just use l2cap_chan_hold_unless_zero to prevent referencing a channel that is about to be destroyed. Cc: stable@kernel.org Signed-off-by: Luiz Augusto von Dentz Signed-off-by: Min Li --- net/bluetooth/l2cap_core.c | 24 ++++++------------------ 1 file changed, 6 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 49926f59cc12..55a7226233f9 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -4652,33 +4652,27 @@ static inline int l2cap_disconnect_req(struct l2cap_conn *conn, BT_DBG("scid 0x%4.4x dcid 0x%4.4x", scid, dcid); - mutex_lock(&conn->chan_lock); - - chan = __l2cap_get_chan_by_scid(conn, dcid); + chan = l2cap_get_chan_by_scid(conn, dcid); if (!chan) { - mutex_unlock(&conn->chan_lock); cmd_reject_invalid_cid(conn, cmd->ident, dcid, scid); return 0; } - l2cap_chan_hold(chan); - l2cap_chan_lock(chan); - rsp.dcid = cpu_to_le16(chan->scid); rsp.scid = cpu_to_le16(chan->dcid); l2cap_send_cmd(conn, cmd->ident, L2CAP_DISCONN_RSP, sizeof(rsp), &rsp); chan->ops->set_shutdown(chan); + mutex_lock(&conn->chan_lock); l2cap_chan_del(chan, ECONNRESET); + mutex_unlock(&conn->chan_lock); chan->ops->close(chan); l2cap_chan_unlock(chan); l2cap_chan_put(chan); - mutex_unlock(&conn->chan_lock); - return 0; } @@ -4698,33 +4692,27 @@ static inline int l2cap_disconnect_rsp(struct l2cap_conn *conn, BT_DBG("dcid 0x%4.4x scid 0x%4.4x", dcid, scid); - mutex_lock(&conn->chan_lock); - - chan = __l2cap_get_chan_by_scid(conn, scid); + chan = l2cap_get_chan_by_scid(conn, scid); if (!chan) { mutex_unlock(&conn->chan_lock); return 0; } - l2cap_chan_hold(chan); - l2cap_chan_lock(chan); - if (chan->state != BT_DISCONN) { l2cap_chan_unlock(chan); l2cap_chan_put(chan); - mutex_unlock(&conn->chan_lock); return 0; } + mutex_lock(&conn->chan_lock); l2cap_chan_del(chan, 0); + mutex_unlock(&conn->chan_lock); chan->ops->close(chan); l2cap_chan_unlock(chan); l2cap_chan_put(chan); - mutex_unlock(&conn->chan_lock); - return 0; } -- cgit v1.2.3 From e3e0ca32cf478e78c579b02cd9c1657d93c97add Mon Sep 17 00:00:00 2001 From: Aloka Dixit Date: Mon, 10 Apr 2023 13:03:32 -0700 Subject: wifi: mac80211: set EHT support flag in AP mode Set 'eht_support' flag if EHT capabilities are present. Signed-off-by: Aloka Dixit Link: https://lore.kernel.org/r/20230410200332.32265-1-quic_alokad@quicinc.com Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 473915606715..7317e4a5d1ff 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1342,6 +1342,10 @@ static int ieee80211_start_ap(struct wiphy *wiphy, struct net_device *dev, } if (params->eht_cap) { + if (!link_conf->he_support) + return -EOPNOTSUPP; + + link_conf->eht_support = true; link_conf->eht_puncturing = params->punct_bitmap; changed |= BSS_CHANGED_EHT_PUNCTURING; -- cgit v1.2.3 From 17c6d0ce83406aebc98f8ce16c66f5416106963d Mon Sep 17 00:00:00 2001 From: Luis Chamberlain Date: Sat, 11 Mar 2023 15:39:41 -0800 Subject: sunrpc: simplify one-level sysctl registration for xr_tunables_table There is no need to declare an extra tables to just create directory, this can be easily be done with a prefix path with register_sysctl(). Simplify this registration. Signed-off-by: Luis Chamberlain Reviewed-by: Jeff Layton Signed-off-by: Anna Schumaker --- net/sunrpc/xprtrdma/transport.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) (limited to 'net') diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c index 10bb2b929c6d..29b0562d62e7 100644 --- a/net/sunrpc/xprtrdma/transport.c +++ b/net/sunrpc/xprtrdma/transport.c @@ -140,15 +140,6 @@ static struct ctl_table xr_tunables_table[] = { { }, }; -static struct ctl_table sunrpc_table[] = { - { - .procname = "sunrpc", - .mode = 0555, - .child = xr_tunables_table - }, - { }, -}; - #endif static const struct rpc_xprt_ops xprt_rdma_procs; @@ -799,7 +790,7 @@ int xprt_rdma_init(void) #if IS_ENABLED(CONFIG_SUNRPC_DEBUG) if (!sunrpc_table_header) - sunrpc_table_header = register_sysctl_table(sunrpc_table); + sunrpc_table_header = register_sysctl("sunrpc", xr_tunables_table); #endif return 0; } -- cgit v1.2.3 From c946cb69f238b2bd3e42790b8450f76b62b491f4 Mon Sep 17 00:00:00 2001 From: Luis Chamberlain Date: Sat, 11 Mar 2023 15:39:42 -0800 Subject: sunrpc: simplify one-level sysctl registration for xs_tunables_table There is no need to declare an extra tables to just create directory, this can be easily be done with a prefix path with register_sysctl(). Simplify this registration. Signed-off-by: Luis Chamberlain Reviewed-by: Jeff Layton Signed-off-by: Anna Schumaker --- net/sunrpc/xprtsock.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 6cacd70a15ff..5f9030b81c9e 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -78,7 +78,7 @@ static unsigned int xs_tcp_fin_timeout __read_mostly = XS_TCP_LINGER_TO; /* * We can register our own files under /proc/sys/sunrpc by - * calling register_sysctl_table() again. The files in that + * calling register_sysctl() again. The files in that * directory become the union of all files registered there. * * We simply need to make sure that we don't collide with @@ -158,15 +158,6 @@ static struct ctl_table xs_tunables_table[] = { { }, }; -static struct ctl_table sunrpc_table[] = { - { - .procname = "sunrpc", - .mode = 0555, - .child = xs_tunables_table - }, - { }, -}; - /* * Wait duration for a reply from the RPC portmapper. */ @@ -3178,7 +3169,7 @@ static struct xprt_class xs_bc_tcp_transport = { int init_socket_xprt(void) { if (!sunrpc_table_header) - sunrpc_table_header = register_sysctl_table(sunrpc_table); + sunrpc_table_header = register_sysctl("sunrpc", xs_tunables_table); xprt_register_transport(&xs_local_transport); xprt_register_transport(&xs_udp_transport); -- cgit v1.2.3 From 32e356be32b6676d2d641ed4a4fe088c6d211d92 Mon Sep 17 00:00:00 2001 From: Luis Chamberlain Date: Sat, 11 Mar 2023 15:39:43 -0800 Subject: sunrpc: move sunrpc_table and proc routines above No need to do a forward declaration for sunrpc_table, just move the sysctls up as everyone else does it. This will make the next change easier to read. This change produces no functional changes. Signed-off-by: Luis Chamberlain Reviewed-by: Jeff Layton Signed-off-by: Anna Schumaker --- net/sunrpc/sysctl.c | 35 ++++++++++++++++------------------- 1 file changed, 16 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/sunrpc/sysctl.c b/net/sunrpc/sysctl.c index 3aad6ef18504..afdfcc5403af 100644 --- a/net/sunrpc/sysctl.c +++ b/net/sunrpc/sysctl.c @@ -40,25 +40,6 @@ EXPORT_SYMBOL_GPL(nlm_debug); #if IS_ENABLED(CONFIG_SUNRPC_DEBUG) -static struct ctl_table_header *sunrpc_table_header; -static struct ctl_table sunrpc_table[]; - -void -rpc_register_sysctl(void) -{ - if (!sunrpc_table_header) - sunrpc_table_header = register_sysctl_table(sunrpc_table); -} - -void -rpc_unregister_sysctl(void) -{ - if (sunrpc_table_header) { - unregister_sysctl_table(sunrpc_table_header); - sunrpc_table_header = NULL; - } -} - static int proc_do_xprt(struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { @@ -142,6 +123,7 @@ done: return 0; } +static struct ctl_table_header *sunrpc_table_header; static struct ctl_table debug_table[] = { { @@ -190,4 +172,19 @@ static struct ctl_table sunrpc_table[] = { { } }; +void +rpc_register_sysctl(void) +{ + if (!sunrpc_table_header) + sunrpc_table_header = register_sysctl_table(sunrpc_table); +} + +void +rpc_unregister_sysctl(void) +{ + if (sunrpc_table_header) { + unregister_sysctl_table(sunrpc_table_header); + sunrpc_table_header = NULL; + } +} #endif -- cgit v1.2.3 From 703c6d03f165183b97526c182aa1d701b40f0453 Mon Sep 17 00:00:00 2001 From: Luis Chamberlain Date: Sat, 11 Mar 2023 15:39:44 -0800 Subject: sunrpc: simplify one-level sysctl registration for debug_table There is no need to declare an extra tables to just create directory, this can be easily be done with a prefix path with register_sysctl(). Simplify this registration. Signed-off-by: Luis Chamberlain Reviewed-by: Jeff Layton Signed-off-by: Anna Schumaker --- net/sunrpc/sysctl.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) (limited to 'net') diff --git a/net/sunrpc/sysctl.c b/net/sunrpc/sysctl.c index afdfcc5403af..93941ab12549 100644 --- a/net/sunrpc/sysctl.c +++ b/net/sunrpc/sysctl.c @@ -163,20 +163,11 @@ static struct ctl_table debug_table[] = { { } }; -static struct ctl_table sunrpc_table[] = { - { - .procname = "sunrpc", - .mode = 0555, - .child = debug_table - }, - { } -}; - void rpc_register_sysctl(void) { if (!sunrpc_table_header) - sunrpc_table_header = register_sysctl_table(sunrpc_table); + sunrpc_table_header = register_sysctl("sunrpc", debug_table); } void -- cgit v1.2.3 From 75dcef8d3609d0b1d3497d6ed4809096513e0b83 Mon Sep 17 00:00:00 2001 From: Feng Zhou Date: Mon, 10 Apr 2023 16:59:08 +0800 Subject: selftests/bpf: Add test to access u32 ptr argument in tracing program Adding verifier test for accessing u32 pointer argument in tracing programs. The test program loads 1nd argument of bpf_fentry_test9 function which is u32 pointer and checks that verifier allows that. Co-developed-by: Chengming Zhou Signed-off-by: Chengming Zhou Signed-off-by: Feng Zhou Signed-off-by: Daniel Borkmann Acked-by: Jiri Olsa Link: https://lore.kernel.org/bpf/20230410085908.98493-3-zhoufeng.zf@bytedance.com --- net/bpf/test_run.c | 8 +++++++- tools/testing/selftests/bpf/verifier/btf_ctx_access.c | 13 +++++++++++++ 2 files changed, 20 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index f1652f5fbd2e..68bdfc041a7b 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -541,6 +541,11 @@ int noinline bpf_fentry_test8(struct bpf_fentry_test_t *arg) return (long)arg->a; } +__bpf_kfunc u32 bpf_fentry_test9(u32 *a) +{ + return *a; +} + __bpf_kfunc int bpf_modify_return_test(int a, int *b) { *b += 1; @@ -855,7 +860,8 @@ int bpf_prog_test_run_tracing(struct bpf_prog *prog, bpf_fentry_test5(11, (void *)12, 13, 14, 15) != 65 || bpf_fentry_test6(16, (void *)17, 18, 19, (void *)20, 21) != 111 || bpf_fentry_test7((struct bpf_fentry_test_t *)0) != 0 || - bpf_fentry_test8(&arg) != 0) + bpf_fentry_test8(&arg) != 0 || + bpf_fentry_test9(&retval) != 0) goto out; break; case BPF_MODIFY_RETURN: diff --git a/tools/testing/selftests/bpf/verifier/btf_ctx_access.c b/tools/testing/selftests/bpf/verifier/btf_ctx_access.c index 6340db6b46dc..0484d3de040d 100644 --- a/tools/testing/selftests/bpf/verifier/btf_ctx_access.c +++ b/tools/testing/selftests/bpf/verifier/btf_ctx_access.c @@ -10,3 +10,16 @@ .expected_attach_type = BPF_TRACE_FENTRY, .kfunc = "bpf_modify_return_test", }, + +{ + "btf_ctx_access u32 pointer accept", + .insns = { + BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 0), /* load 1nd argument value (u32 pointer) */ + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }, + .result = ACCEPT, + .prog_type = BPF_PROG_TYPE_TRACING, + .expected_attach_type = BPF_TRACE_FENTRY, + .kfunc = "bpf_fentry_test9", +}, -- cgit v1.2.3 From 9538ebce88ffa074202d592d468521995cb1e714 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 7 Apr 2023 17:14:50 +0300 Subject: net: dsa: add trace points for FDB/MDB operations DSA performs non-trivial housekeeping of unicast and multicast addresses on shared (CPU and DSA) ports, and puts a bit of pressure on higher layers, requiring them to behave correctly (remove these addresses exactly as many times as they were added). Otherwise, either addresses linger around forever, or DSA returns -ENOENT complaining that entries that were already deleted must be deleted again. To aid debugging, introduce some trace points specifically for FDB and MDB - that's where some of the bugs still are right now. Some bugs I have seen were also due to race conditions, see: - 630fd4822af2 ("net: dsa: flush switchdev workqueue on bridge join error path") - a2614140dc0f ("net: dsa: mv88e6xxx: flush switchdev FDB workqueue before removing VLAN") so it would be good to not disturb the timing too much, hence the choice to use trace points vs regular dev_dbg(). I've had these for some time on my computer in a less polished form, and they've proven useful. What I found most useful was to enable CONFIG_BOOTTIME_TRACING, add "trace_event=dsa" to the kernel cmdline, and run "cat /sys/kernel/debug/tracing/trace". This is to debug more complex environments with network managers started by the init system, things like that. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- net/dsa/Makefile | 6 +- net/dsa/switch.c | 61 +++++++++-- net/dsa/trace.c | 39 +++++++ net/dsa/trace.h | 329 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 423 insertions(+), 12 deletions(-) create mode 100644 net/dsa/trace.c create mode 100644 net/dsa/trace.h (limited to 'net') diff --git a/net/dsa/Makefile b/net/dsa/Makefile index 3835de286116..12e305824a96 100644 --- a/net/dsa/Makefile +++ b/net/dsa/Makefile @@ -16,7 +16,8 @@ dsa_core-y += \ slave.o \ switch.o \ tag.o \ - tag_8021q.o + tag_8021q.o \ + trace.o # tagging formats obj-$(CONFIG_NET_DSA_TAG_AR9331) += tag_ar9331.o @@ -37,3 +38,6 @@ obj-$(CONFIG_NET_DSA_TAG_RZN1_A5PSW) += tag_rzn1_a5psw.o obj-$(CONFIG_NET_DSA_TAG_SJA1105) += tag_sja1105.o obj-$(CONFIG_NET_DSA_TAG_TRAILER) += tag_trailer.o obj-$(CONFIG_NET_DSA_TAG_XRS700X) += tag_xrs700x.o + +# for tracing framework to find trace.h +CFLAGS_trace.o := -I$(src) diff --git a/net/dsa/switch.c b/net/dsa/switch.c index d5bc4bb7310d..ff1b5d980e37 100644 --- a/net/dsa/switch.c +++ b/net/dsa/switch.c @@ -18,6 +18,7 @@ #include "slave.h" #include "switch.h" #include "tag_8021q.h" +#include "trace.h" static unsigned int dsa_switch_fastest_ageing_time(struct dsa_switch *ds, unsigned int ageing_time) @@ -164,14 +165,20 @@ static int dsa_port_do_mdb_add(struct dsa_port *dp, int err = 0; /* No need to bother with refcounting for user ports */ - if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) - return ds->ops->port_mdb_add(ds, port, mdb, db); + if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) { + err = ds->ops->port_mdb_add(ds, port, mdb, db); + trace_dsa_mdb_add_hw(dp, mdb->addr, mdb->vid, &db, err); + + return err; + } mutex_lock(&dp->addr_lists_lock); a = dsa_mac_addr_find(&dp->mdbs, mdb->addr, mdb->vid, db); if (a) { refcount_inc(&a->refcount); + trace_dsa_mdb_add_bump(dp, mdb->addr, mdb->vid, &db, + &a->refcount); goto out; } @@ -182,6 +189,7 @@ static int dsa_port_do_mdb_add(struct dsa_port *dp, } err = ds->ops->port_mdb_add(ds, port, mdb, db); + trace_dsa_mdb_add_hw(dp, mdb->addr, mdb->vid, &db, err); if (err) { kfree(a); goto out; @@ -209,21 +217,30 @@ static int dsa_port_do_mdb_del(struct dsa_port *dp, int err = 0; /* No need to bother with refcounting for user ports */ - if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) - return ds->ops->port_mdb_del(ds, port, mdb, db); + if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) { + err = ds->ops->port_mdb_del(ds, port, mdb, db); + trace_dsa_mdb_del_hw(dp, mdb->addr, mdb->vid, &db, err); + + return err; + } mutex_lock(&dp->addr_lists_lock); a = dsa_mac_addr_find(&dp->mdbs, mdb->addr, mdb->vid, db); if (!a) { + trace_dsa_mdb_del_not_found(dp, mdb->addr, mdb->vid, &db); err = -ENOENT; goto out; } - if (!refcount_dec_and_test(&a->refcount)) + if (!refcount_dec_and_test(&a->refcount)) { + trace_dsa_mdb_del_drop(dp, mdb->addr, mdb->vid, &db, + &a->refcount); goto out; + } err = ds->ops->port_mdb_del(ds, port, mdb, db); + trace_dsa_mdb_del_hw(dp, mdb->addr, mdb->vid, &db, err); if (err) { refcount_set(&a->refcount, 1); goto out; @@ -247,14 +264,19 @@ static int dsa_port_do_fdb_add(struct dsa_port *dp, const unsigned char *addr, int err = 0; /* No need to bother with refcounting for user ports */ - if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) - return ds->ops->port_fdb_add(ds, port, addr, vid, db); + if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) { + err = ds->ops->port_fdb_add(ds, port, addr, vid, db); + trace_dsa_fdb_add_hw(dp, addr, vid, &db, err); + + return err; + } mutex_lock(&dp->addr_lists_lock); a = dsa_mac_addr_find(&dp->fdbs, addr, vid, db); if (a) { refcount_inc(&a->refcount); + trace_dsa_fdb_add_bump(dp, addr, vid, &db, &a->refcount); goto out; } @@ -265,6 +287,7 @@ static int dsa_port_do_fdb_add(struct dsa_port *dp, const unsigned char *addr, } err = ds->ops->port_fdb_add(ds, port, addr, vid, db); + trace_dsa_fdb_add_hw(dp, addr, vid, &db, err); if (err) { kfree(a); goto out; @@ -291,21 +314,29 @@ static int dsa_port_do_fdb_del(struct dsa_port *dp, const unsigned char *addr, int err = 0; /* No need to bother with refcounting for user ports */ - if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) - return ds->ops->port_fdb_del(ds, port, addr, vid, db); + if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) { + err = ds->ops->port_fdb_del(ds, port, addr, vid, db); + trace_dsa_fdb_del_hw(dp, addr, vid, &db, err); + + return err; + } mutex_lock(&dp->addr_lists_lock); a = dsa_mac_addr_find(&dp->fdbs, addr, vid, db); if (!a) { + trace_dsa_fdb_del_not_found(dp, addr, vid, &db); err = -ENOENT; goto out; } - if (!refcount_dec_and_test(&a->refcount)) + if (!refcount_dec_and_test(&a->refcount)) { + trace_dsa_fdb_del_drop(dp, addr, vid, &db, &a->refcount); goto out; + } err = ds->ops->port_fdb_del(ds, port, addr, vid, db); + trace_dsa_fdb_del_hw(dp, addr, vid, &db, err); if (err) { refcount_set(&a->refcount, 1); goto out; @@ -332,6 +363,8 @@ static int dsa_switch_do_lag_fdb_add(struct dsa_switch *ds, struct dsa_lag *lag, a = dsa_mac_addr_find(&lag->fdbs, addr, vid, db); if (a) { refcount_inc(&a->refcount); + trace_dsa_lag_fdb_add_bump(lag->dev, addr, vid, &db, + &a->refcount); goto out; } @@ -342,6 +375,7 @@ static int dsa_switch_do_lag_fdb_add(struct dsa_switch *ds, struct dsa_lag *lag, } err = ds->ops->lag_fdb_add(ds, *lag, addr, vid, db); + trace_dsa_lag_fdb_add_hw(lag->dev, addr, vid, &db, err); if (err) { kfree(a); goto out; @@ -370,14 +404,19 @@ static int dsa_switch_do_lag_fdb_del(struct dsa_switch *ds, struct dsa_lag *lag, a = dsa_mac_addr_find(&lag->fdbs, addr, vid, db); if (!a) { + trace_dsa_lag_fdb_del_not_found(lag->dev, addr, vid, &db); err = -ENOENT; goto out; } - if (!refcount_dec_and_test(&a->refcount)) + if (!refcount_dec_and_test(&a->refcount)) { + trace_dsa_lag_fdb_del_drop(lag->dev, addr, vid, &db, + &a->refcount); goto out; + } err = ds->ops->lag_fdb_del(ds, *lag, addr, vid, db); + trace_dsa_lag_fdb_del_hw(lag->dev, addr, vid, &db, err); if (err) { refcount_set(&a->refcount, 1); goto out; diff --git a/net/dsa/trace.c b/net/dsa/trace.c new file mode 100644 index 000000000000..1b107165d331 --- /dev/null +++ b/net/dsa/trace.c @@ -0,0 +1,39 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Copyright 2022-2023 NXP + */ + +#define CREATE_TRACE_POINTS +#include "trace.h" + +void dsa_db_print(const struct dsa_db *db, char buf[DSA_DB_BUFSIZ]) +{ + switch (db->type) { + case DSA_DB_PORT: + sprintf(buf, "port %s", db->dp->name); + break; + case DSA_DB_LAG: + sprintf(buf, "lag %s id %d", db->lag.dev->name, db->lag.id); + break; + case DSA_DB_BRIDGE: + sprintf(buf, "bridge %s num %d", db->bridge.dev->name, + db->bridge.num); + break; + default: + sprintf(buf, "unknown"); + break; + } +} + +const char *dsa_port_kind(const struct dsa_port *dp) +{ + switch (dp->type) { + case DSA_PORT_TYPE_USER: + return "user"; + case DSA_PORT_TYPE_CPU: + return "cpu"; + case DSA_PORT_TYPE_DSA: + return "dsa"; + default: + return "unused"; + } +} diff --git a/net/dsa/trace.h b/net/dsa/trace.h new file mode 100644 index 000000000000..42c8bbc7d472 --- /dev/null +++ b/net/dsa/trace.h @@ -0,0 +1,329 @@ +/* SPDX-License-Identifier: GPL-2.0 + * Copyright 2022-2023 NXP + */ + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM dsa + +#if !defined(_NET_DSA_TRACE_H) || defined(TRACE_HEADER_MULTI_READ) +#define _NET_DSA_TRACE_H + +#include +#include +#include +#include + +/* Enough to fit "bridge %s num %d" where num has 3 digits */ +#define DSA_DB_BUFSIZ (IFNAMSIZ + 16) + +void dsa_db_print(const struct dsa_db *db, char buf[DSA_DB_BUFSIZ]); +const char *dsa_port_kind(const struct dsa_port *dp); + +DECLARE_EVENT_CLASS(dsa_port_addr_op_hw, + + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, u16 vid, + const struct dsa_db *db, int err), + + TP_ARGS(dp, addr, vid, db, err), + + TP_STRUCT__entry( + __string(dev, dev_name(dp->ds->dev)) + __string(kind, dsa_port_kind(dp)) + __field(int, port) + __array(unsigned char, addr, ETH_ALEN) + __field(u16, vid) + __array(char, db_buf, DSA_DB_BUFSIZ) + __field(int, err) + ), + + TP_fast_assign( + __assign_str(dev, dev_name(dp->ds->dev)); + __assign_str(kind, dsa_port_kind(dp)); + __entry->port = dp->index; + ether_addr_copy(__entry->addr, addr); + __entry->vid = vid; + dsa_db_print(db, __entry->db_buf); + __entry->err = err; + ), + + TP_printk("%s %s port %d addr %pM vid %u db \"%s\" err %d", + __get_str(dev), __get_str(kind), __entry->port, __entry->addr, + __entry->vid, __entry->db_buf, __entry->err) +); + +/* Add unicast/multicast address to hardware, either on user ports + * (where no refcounting is kept), or on shared ports when the entry + * is first seen and its refcount is 1. + */ +DEFINE_EVENT(dsa_port_addr_op_hw, dsa_fdb_add_hw, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db, int err), + TP_ARGS(dp, addr, vid, db, err)); + +DEFINE_EVENT(dsa_port_addr_op_hw, dsa_mdb_add_hw, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db, int err), + TP_ARGS(dp, addr, vid, db, err)); + +/* Delete unicast/multicast address from hardware, either on user ports or + * when the refcount on shared ports reaches 0 + */ +DEFINE_EVENT(dsa_port_addr_op_hw, dsa_fdb_del_hw, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db, int err), + TP_ARGS(dp, addr, vid, db, err)); + +DEFINE_EVENT(dsa_port_addr_op_hw, dsa_mdb_del_hw, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db, int err), + TP_ARGS(dp, addr, vid, db, err)); + +DECLARE_EVENT_CLASS(dsa_port_addr_op_refcount, + + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, u16 vid, + const struct dsa_db *db, const refcount_t *refcount), + + TP_ARGS(dp, addr, vid, db, refcount), + + TP_STRUCT__entry( + __string(dev, dev_name(dp->ds->dev)) + __string(kind, dsa_port_kind(dp)) + __field(int, port) + __array(unsigned char, addr, ETH_ALEN) + __field(u16, vid) + __array(char, db_buf, DSA_DB_BUFSIZ) + __field(unsigned int, refcount) + ), + + TP_fast_assign( + __assign_str(dev, dev_name(dp->ds->dev)); + __assign_str(kind, dsa_port_kind(dp)); + __entry->port = dp->index; + ether_addr_copy(__entry->addr, addr); + __entry->vid = vid; + dsa_db_print(db, __entry->db_buf); + __entry->refcount = refcount_read(refcount); + ), + + TP_printk("%s %s port %d addr %pM vid %u db \"%s\" refcount %u", + __get_str(dev), __get_str(kind), __entry->port, __entry->addr, + __entry->vid, __entry->db_buf, __entry->refcount) +); + +/* Bump the refcount of an existing unicast/multicast address on shared ports */ +DEFINE_EVENT(dsa_port_addr_op_refcount, dsa_fdb_add_bump, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db, + const refcount_t *refcount), + TP_ARGS(dp, addr, vid, db, refcount)); + +DEFINE_EVENT(dsa_port_addr_op_refcount, dsa_mdb_add_bump, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db, + const refcount_t *refcount), + TP_ARGS(dp, addr, vid, db, refcount)); + +/* Drop the refcount of a multicast address that we still keep on + * shared ports + */ +DEFINE_EVENT(dsa_port_addr_op_refcount, dsa_fdb_del_drop, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db, + const refcount_t *refcount), + TP_ARGS(dp, addr, vid, db, refcount)); + +DEFINE_EVENT(dsa_port_addr_op_refcount, dsa_mdb_del_drop, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db, + const refcount_t *refcount), + TP_ARGS(dp, addr, vid, db, refcount)); + +DECLARE_EVENT_CLASS(dsa_port_addr_del_not_found, + + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, u16 vid, + const struct dsa_db *db), + + TP_ARGS(dp, addr, vid, db), + + TP_STRUCT__entry( + __string(dev, dev_name(dp->ds->dev)) + __string(kind, dsa_port_kind(dp)) + __field(int, port) + __array(unsigned char, addr, ETH_ALEN) + __field(u16, vid) + __array(char, db_buf, DSA_DB_BUFSIZ) + ), + + TP_fast_assign( + __assign_str(dev, dev_name(dp->ds->dev)); + __assign_str(kind, dsa_port_kind(dp)); + __entry->port = dp->index; + ether_addr_copy(__entry->addr, addr); + __entry->vid = vid; + dsa_db_print(db, __entry->db_buf); + ), + + TP_printk("%s %s port %d addr %pM vid %u db \"%s\"", + __get_str(dev), __get_str(kind), __entry->port, + __entry->addr, __entry->vid, __entry->db_buf) +); + +/* Attempt to delete a unicast/multicast address on shared ports for which + * the delete operation was called more times than the addition + */ +DEFINE_EVENT(dsa_port_addr_del_not_found, dsa_fdb_del_not_found, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db), + TP_ARGS(dp, addr, vid, db)); + +DEFINE_EVENT(dsa_port_addr_del_not_found, dsa_mdb_del_not_found, + TP_PROTO(const struct dsa_port *dp, const unsigned char *addr, + u16 vid, const struct dsa_db *db), + TP_ARGS(dp, addr, vid, db)); + +TRACE_EVENT(dsa_lag_fdb_add_hw, + + TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr, + u16 vid, const struct dsa_db *db, int err), + + TP_ARGS(lag_dev, addr, vid, db, err), + + TP_STRUCT__entry( + __string(dev, lag_dev->name) + __array(unsigned char, addr, ETH_ALEN) + __field(u16, vid) + __array(char, db_buf, DSA_DB_BUFSIZ) + __field(int, err) + ), + + TP_fast_assign( + __assign_str(dev, lag_dev->name); + ether_addr_copy(__entry->addr, addr); + __entry->vid = vid; + dsa_db_print(db, __entry->db_buf); + __entry->err = err; + ), + + TP_printk("%s addr %pM vid %u db \"%s\" err %d", + __get_str(dev), __entry->addr, __entry->vid, + __entry->db_buf, __entry->err) +); + +TRACE_EVENT(dsa_lag_fdb_add_bump, + + TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr, + u16 vid, const struct dsa_db *db, const refcount_t *refcount), + + TP_ARGS(lag_dev, addr, vid, db, refcount), + + TP_STRUCT__entry( + __string(dev, lag_dev->name) + __array(unsigned char, addr, ETH_ALEN) + __field(u16, vid) + __array(char, db_buf, DSA_DB_BUFSIZ) + __field(unsigned int, refcount) + ), + + TP_fast_assign( + __assign_str(dev, lag_dev->name); + ether_addr_copy(__entry->addr, addr); + __entry->vid = vid; + dsa_db_print(db, __entry->db_buf); + __entry->refcount = refcount_read(refcount); + ), + + TP_printk("%s addr %pM vid %u db \"%s\" refcount %u", + __get_str(dev), __entry->addr, __entry->vid, + __entry->db_buf, __entry->refcount) +); + +TRACE_EVENT(dsa_lag_fdb_del_hw, + + TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr, + u16 vid, const struct dsa_db *db, int err), + + TP_ARGS(lag_dev, addr, vid, db, err), + + TP_STRUCT__entry( + __string(dev, lag_dev->name) + __array(unsigned char, addr, ETH_ALEN) + __field(u16, vid) + __array(char, db_buf, DSA_DB_BUFSIZ) + __field(int, err) + ), + + TP_fast_assign( + __assign_str(dev, lag_dev->name); + ether_addr_copy(__entry->addr, addr); + __entry->vid = vid; + dsa_db_print(db, __entry->db_buf); + __entry->err = err; + ), + + TP_printk("%s addr %pM vid %u db \"%s\" err %d", + __get_str(dev), __entry->addr, __entry->vid, + __entry->db_buf, __entry->err) +); + +TRACE_EVENT(dsa_lag_fdb_del_drop, + + TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr, + u16 vid, const struct dsa_db *db, const refcount_t *refcount), + + TP_ARGS(lag_dev, addr, vid, db, refcount), + + TP_STRUCT__entry( + __string(dev, lag_dev->name) + __array(unsigned char, addr, ETH_ALEN) + __field(u16, vid) + __array(char, db_buf, DSA_DB_BUFSIZ) + __field(unsigned int, refcount) + ), + + TP_fast_assign( + __assign_str(dev, lag_dev->name); + ether_addr_copy(__entry->addr, addr); + __entry->vid = vid; + dsa_db_print(db, __entry->db_buf); + __entry->refcount = refcount_read(refcount); + ), + + TP_printk("%s addr %pM vid %u db \"%s\" refcount %u", + __get_str(dev), __entry->addr, __entry->vid, + __entry->db_buf, __entry->refcount) +); + +TRACE_EVENT(dsa_lag_fdb_del_not_found, + + TP_PROTO(const struct net_device *lag_dev, const unsigned char *addr, + u16 vid, const struct dsa_db *db), + + TP_ARGS(lag_dev, addr, vid, db), + + TP_STRUCT__entry( + __string(dev, lag_dev->name) + __array(unsigned char, addr, ETH_ALEN) + __field(u16, vid) + __array(char, db_buf, DSA_DB_BUFSIZ) + ), + + TP_fast_assign( + __assign_str(dev, lag_dev->name); + ether_addr_copy(__entry->addr, addr); + __entry->vid = vid; + dsa_db_print(db, __entry->db_buf); + ), + + TP_printk("%s addr %pM vid %u db \"%s\"", + __get_str(dev), __entry->addr, __entry->vid, __entry->db_buf) +); + +#endif /* _NET_DSA_TRACE_H */ + +/* We don't want to use include/trace/events */ +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE trace +/* This part must be outside protection */ +#include -- cgit v1.2.3 From 02020bd70fa6abcb1c2a8525ce7c1500dd4f44a8 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 7 Apr 2023 17:14:51 +0300 Subject: net: dsa: add trace points for VLAN operations These are not as critical as the FDB/MDB trace points (I'm not aware of outstanding VLAN related bugs), but maybe they are useful to somebody, either debugging something or simply trying to learn more. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- net/dsa/switch.c | 24 ++++++++--- net/dsa/trace.h | 118 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 137 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/dsa/switch.c b/net/dsa/switch.c index ff1b5d980e37..8c9a9f94b756 100644 --- a/net/dsa/switch.c +++ b/net/dsa/switch.c @@ -695,8 +695,12 @@ static int dsa_port_do_vlan_add(struct dsa_port *dp, int err = 0; /* No need to bother with refcounting for user ports. */ - if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) - return ds->ops->port_vlan_add(ds, port, vlan, extack); + if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) { + err = ds->ops->port_vlan_add(ds, port, vlan, extack); + trace_dsa_vlan_add_hw(dp, vlan, err); + + return err; + } /* No need to propagate on shared ports the existing VLANs that were * re-notified after just the flags have changed. This would cause a @@ -711,6 +715,7 @@ static int dsa_port_do_vlan_add(struct dsa_port *dp, v = dsa_vlan_find(&dp->vlans, vlan); if (v) { refcount_inc(&v->refcount); + trace_dsa_vlan_add_bump(dp, vlan, &v->refcount); goto out; } @@ -721,6 +726,7 @@ static int dsa_port_do_vlan_add(struct dsa_port *dp, } err = ds->ops->port_vlan_add(ds, port, vlan, extack); + trace_dsa_vlan_add_hw(dp, vlan, err); if (err) { kfree(v); goto out; @@ -745,21 +751,29 @@ static int dsa_port_do_vlan_del(struct dsa_port *dp, int err = 0; /* No need to bother with refcounting for user ports */ - if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) - return ds->ops->port_vlan_del(ds, port, vlan); + if (!(dsa_port_is_cpu(dp) || dsa_port_is_dsa(dp))) { + err = ds->ops->port_vlan_del(ds, port, vlan); + trace_dsa_vlan_del_hw(dp, vlan, err); + + return err; + } mutex_lock(&dp->vlans_lock); v = dsa_vlan_find(&dp->vlans, vlan); if (!v) { + trace_dsa_vlan_del_not_found(dp, vlan); err = -ENOENT; goto out; } - if (!refcount_dec_and_test(&v->refcount)) + if (!refcount_dec_and_test(&v->refcount)) { + trace_dsa_vlan_del_drop(dp, vlan, &v->refcount); goto out; + } err = ds->ops->port_vlan_del(ds, port, vlan); + trace_dsa_vlan_del_hw(dp, vlan, err); if (err) { refcount_set(&v->refcount, 1); goto out; diff --git a/net/dsa/trace.h b/net/dsa/trace.h index 42c8bbc7d472..567f29a39707 100644 --- a/net/dsa/trace.h +++ b/net/dsa/trace.h @@ -9,7 +9,9 @@ #define _NET_DSA_TRACE_H #include +#include #include +#include #include #include @@ -318,6 +320,122 @@ TRACE_EVENT(dsa_lag_fdb_del_not_found, __get_str(dev), __entry->addr, __entry->vid, __entry->db_buf) ); +DECLARE_EVENT_CLASS(dsa_vlan_op_hw, + + TP_PROTO(const struct dsa_port *dp, + const struct switchdev_obj_port_vlan *vlan, int err), + + TP_ARGS(dp, vlan, err), + + TP_STRUCT__entry( + __string(dev, dev_name(dp->ds->dev)) + __string(kind, dsa_port_kind(dp)) + __field(int, port) + __field(u16, vid) + __field(u16, flags) + __field(bool, changed) + __field(int, err) + ), + + TP_fast_assign( + __assign_str(dev, dev_name(dp->ds->dev)); + __assign_str(kind, dsa_port_kind(dp)); + __entry->port = dp->index; + __entry->vid = vlan->vid; + __entry->flags = vlan->flags; + __entry->changed = vlan->changed; + __entry->err = err; + ), + + TP_printk("%s %s port %d vid %u%s%s%s", + __get_str(dev), __get_str(kind), __entry->port, __entry->vid, + __entry->flags & BRIDGE_VLAN_INFO_PVID ? " pvid" : "", + __entry->flags & BRIDGE_VLAN_INFO_UNTAGGED ? " untagged" : "", + __entry->changed ? " (changed)" : "") +); + +DEFINE_EVENT(dsa_vlan_op_hw, dsa_vlan_add_hw, + TP_PROTO(const struct dsa_port *dp, + const struct switchdev_obj_port_vlan *vlan, int err), + TP_ARGS(dp, vlan, err)); + +DEFINE_EVENT(dsa_vlan_op_hw, dsa_vlan_del_hw, + TP_PROTO(const struct dsa_port *dp, + const struct switchdev_obj_port_vlan *vlan, int err), + TP_ARGS(dp, vlan, err)); + +DECLARE_EVENT_CLASS(dsa_vlan_op_refcount, + + TP_PROTO(const struct dsa_port *dp, + const struct switchdev_obj_port_vlan *vlan, + const refcount_t *refcount), + + TP_ARGS(dp, vlan, refcount), + + TP_STRUCT__entry( + __string(dev, dev_name(dp->ds->dev)) + __string(kind, dsa_port_kind(dp)) + __field(int, port) + __field(u16, vid) + __field(u16, flags) + __field(bool, changed) + __field(unsigned int, refcount) + ), + + TP_fast_assign( + __assign_str(dev, dev_name(dp->ds->dev)); + __assign_str(kind, dsa_port_kind(dp)); + __entry->port = dp->index; + __entry->vid = vlan->vid; + __entry->flags = vlan->flags; + __entry->changed = vlan->changed; + __entry->refcount = refcount_read(refcount); + ), + + TP_printk("%s %s port %d vid %u%s%s%s refcount %u", + __get_str(dev), __get_str(kind), __entry->port, __entry->vid, + __entry->flags & BRIDGE_VLAN_INFO_PVID ? " pvid" : "", + __entry->flags & BRIDGE_VLAN_INFO_UNTAGGED ? " untagged" : "", + __entry->changed ? " (changed)" : "", __entry->refcount) +); + +DEFINE_EVENT(dsa_vlan_op_refcount, dsa_vlan_add_bump, + TP_PROTO(const struct dsa_port *dp, + const struct switchdev_obj_port_vlan *vlan, + const refcount_t *refcount), + TP_ARGS(dp, vlan, refcount)); + +DEFINE_EVENT(dsa_vlan_op_refcount, dsa_vlan_del_drop, + TP_PROTO(const struct dsa_port *dp, + const struct switchdev_obj_port_vlan *vlan, + const refcount_t *refcount), + TP_ARGS(dp, vlan, refcount)); + +TRACE_EVENT(dsa_vlan_del_not_found, + + TP_PROTO(const struct dsa_port *dp, + const struct switchdev_obj_port_vlan *vlan), + + TP_ARGS(dp, vlan), + + TP_STRUCT__entry( + __string(dev, dev_name(dp->ds->dev)) + __string(kind, dsa_port_kind(dp)) + __field(int, port) + __field(u16, vid) + ), + + TP_fast_assign( + __assign_str(dev, dev_name(dp->ds->dev)); + __assign_str(kind, dsa_port_kind(dp)); + __entry->port = dp->index; + __entry->vid = vlan->vid; + ), + + TP_printk("%s %s port %d vid %u", + __get_str(dev), __get_str(kind), __entry->port, __entry->vid) +); + #endif /* _NET_DSA_TRACE_H */ /* We don't want to use include/trace/events */ -- cgit v1.2.3 From 9744d2bf19762703704ecba885b7ac282c02eacf Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 8 Apr 2023 11:49:43 -0700 Subject: smc: Fix use-after-free in tcp_write_timer_handler(). With Eric's ref tracker, syzbot finally found a repro for use-after-free in tcp_write_timer_handler() by kernel TCP sockets. [0] If SMC creates a kernel socket in __smc_create(), the kernel socket is supposed to be freed in smc_clcsock_release() by calling sock_release() when we close() the parent SMC socket. However, at the end of smc_clcsock_release(), the kernel socket's sk_state might not be TCP_CLOSE. This means that we have not called inet_csk_destroy_sock() in __tcp_close() and have not stopped the TCP timers. The kernel socket's TCP timers can be fired later, so we need to hold a refcnt for net as we do for MPTCP subflows in mptcp_subflow_create_socket(). [0]: leaked reference. sk_alloc (./include/net/net_namespace.h:335 net/core/sock.c:2108) inet_create (net/ipv4/af_inet.c:319 net/ipv4/af_inet.c:244) __sock_create (net/socket.c:1546) smc_create (net/smc/af_smc.c:3269 net/smc/af_smc.c:3284) __sock_create (net/socket.c:1546) __sys_socket (net/socket.c:1634 net/socket.c:1618 net/socket.c:1661) __x64_sys_socket (net/socket.c:1672) do_syscall_64 (arch/x86/entry/common.c:50 arch/x86/entry/common.c:80) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120) ================================================================== BUG: KASAN: slab-use-after-free in tcp_write_timer_handler (net/ipv4/tcp_timer.c:378 net/ipv4/tcp_timer.c:624 net/ipv4/tcp_timer.c:594) Read of size 1 at addr ffff888052b65e0d by task syzrepro/18091 CPU: 0 PID: 18091 Comm: syzrepro Tainted: G W 6.3.0-rc4-01174-gb5d54eb5899a #7 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.0-1.amzn2022.0.1 04/01/2014 Call Trace: dump_stack_lvl (lib/dump_stack.c:107) print_report (mm/kasan/report.c:320 mm/kasan/report.c:430) kasan_report (mm/kasan/report.c:538) tcp_write_timer_handler (net/ipv4/tcp_timer.c:378 net/ipv4/tcp_timer.c:624 net/ipv4/tcp_timer.c:594) tcp_write_timer (./include/linux/spinlock.h:390 net/ipv4/tcp_timer.c:643) call_timer_fn (./arch/x86/include/asm/jump_label.h:27 ./include/linux/jump_label.h:207 ./include/trace/events/timer.h:127 kernel/time/timer.c:1701) __run_timers.part.0 (kernel/time/timer.c:1752 kernel/time/timer.c:2022) run_timer_softirq (kernel/time/timer.c:2037) __do_softirq (./arch/x86/include/asm/jump_label.h:27 ./include/linux/jump_label.h:207 ./include/trace/events/irq.h:142 kernel/softirq.c:572) __irq_exit_rcu (kernel/softirq.c:445 kernel/softirq.c:650) irq_exit_rcu (kernel/softirq.c:664) sysvec_apic_timer_interrupt (arch/x86/kernel/apic/apic.c:1107 (discriminator 14)) Fixes: ac7138746e14 ("smc: establish new socket family") Reported-by: syzbot+7e1e1bdb852961150198@syzkaller.appspotmail.com Link: https://lore.kernel.org/netdev/000000000000a3f51805f8bcc43a@google.com/ Signed-off-by: Kuniyuki Iwashima Reviewed-by: Tony Lu Signed-off-by: David S. Miller --- net/smc/af_smc.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index c6b4a62276f6..50c38b624f77 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -3270,6 +3270,17 @@ static int __smc_create(struct net *net, struct socket *sock, int protocol, sk_common_release(sk); goto out; } + + /* smc_clcsock_release() does not wait smc->clcsock->sk's + * destruction; its sk_state might not be TCP_CLOSE after + * smc->sk is close()d, and TCP timers can be fired later, + * which need net ref. + */ + sk = smc->clcsock->sk; + __netns_tracker_free(net, &sk->ns_tracker, false); + sk->sk_net_refcnt = 1; + get_net_track(net, &sk->ns_tracker, GFP_KERNEL); + sock_inuse_add(net, 1); } else { smc->clcsock = clcsock; } -- cgit v1.2.3 From ed17aa92dc56b6d8883e4b7a8f1c6fbf5ed6cd29 Mon Sep 17 00:00:00 2001 From: Xin Liu Date: Thu, 6 Apr 2023 20:26:22 +0800 Subject: bpf, sockmap: fix deadlocks in the sockhash and sockmap When huang uses sched_switch tracepoint, the tracepoint does only one thing in the mounted ebpf program, which deletes the fixed elements in sockhash ([0]) It seems that elements in sockhash are rarely actively deleted by users or ebpf program. Therefore, we do not pay much attention to their deletion. Compared with hash maps, sockhash only provides spin_lock_bh protection. This causes it to appear to have self-locking behavior in the interrupt context. [0]:https://lore.kernel.org/all/CABcoxUayum5oOqFMMqAeWuS8+EzojquSOSyDA3J_2omY=2EeAg@mail.gmail.com/ Reported-by: Hsin-Wei Hung Fixes: 604326b41a6f ("bpf, sockmap: convert to generic sk_msg interface") Signed-off-by: Xin Liu Acked-by: John Fastabend Link: https://lore.kernel.org/r/20230406122622.109978-1-liuxin350@huawei.com Signed-off-by: Alexei Starovoitov --- net/core/sock_map.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/core/sock_map.c b/net/core/sock_map.c index 7c189c2e2fbf..66305b7bf8b7 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -414,8 +414,9 @@ static int __sock_map_delete(struct bpf_stab *stab, struct sock *sk_test, { struct sock *sk; int err = 0; + unsigned long flags; - raw_spin_lock_bh(&stab->lock); + raw_spin_lock_irqsave(&stab->lock, flags); sk = *psk; if (!sk_test || sk_test == sk) sk = xchg(psk, NULL); @@ -425,7 +426,7 @@ static int __sock_map_delete(struct bpf_stab *stab, struct sock *sk_test, else err = -EINVAL; - raw_spin_unlock_bh(&stab->lock); + raw_spin_unlock_irqrestore(&stab->lock, flags); return err; } @@ -932,11 +933,12 @@ static long sock_hash_delete_elem(struct bpf_map *map, void *key) struct bpf_shtab_bucket *bucket; struct bpf_shtab_elem *elem; int ret = -ENOENT; + unsigned long flags; hash = sock_hash_bucket_hash(key, key_size); bucket = sock_hash_select_bucket(htab, hash); - raw_spin_lock_bh(&bucket->lock); + raw_spin_lock_irqsave(&bucket->lock, flags); elem = sock_hash_lookup_elem_raw(&bucket->head, hash, key, key_size); if (elem) { hlist_del_rcu(&elem->node); @@ -944,7 +946,7 @@ static long sock_hash_delete_elem(struct bpf_map *map, void *key) sock_hash_free_elem(htab, elem); ret = 0; } - raw_spin_unlock_bh(&bucket->lock); + raw_spin_unlock_irqrestore(&bucket->lock, flags); return ret; } -- cgit v1.2.3 From ac931d4cdec3df8b6eac3bc40a6871123021f078 Mon Sep 17 00:00:00 2001 From: Christian Ehrig Date: Fri, 7 Apr 2023 15:38:53 +0200 Subject: ipip,ip_tunnel,sit: Add FOU support for externally controlled ipip devices Today ipip devices in collect-metadata mode don't allow for sending FOU or GUE encapsulated packets. This patch lifts the restriction by adding a struct ip_tunnel_encap to the tunnel metadata. On the egress path, the members of this struct can be set by the bpf_skb_set_fou_encap kfunc via a BPF tc-hook. Instead of dropping packets wishing to use additional UDP encapsulation, ip_md_tunnel_xmit now evaluates the contents of this struct and adds the corresponding FOU or GUE header. Furthermore, it is making sure that additional header bytes are taken into account for PMTU discovery. On the ingress path, an ipip device in collect-metadata mode will fill this struct and a BPF tc-hook can obtain the information via a call to the bpf_skb_get_fou_encap kfunc. The minor change to ip_tunnel_encap, which now takes a pointer to struct ip_tunnel_encap instead of struct ip_tunnel, allows us to control FOU encap type and parameters on a per packet-level. Signed-off-by: Christian Ehrig Link: https://lore.kernel.org/r/cfea47de655d0f870248abf725932f851b53960a.1680874078.git.cehrig@cloudflare.com Signed-off-by: Alexei Starovoitov --- include/net/ip_tunnels.h | 28 +++++++++++++++------------- net/ipv4/ip_tunnel.c | 22 ++++++++++++++++++++-- net/ipv4/ipip.c | 1 + net/ipv6/sit.c | 2 +- 4 files changed, 37 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index fca357679816..7912f53caae0 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -57,6 +57,13 @@ struct ip_tunnel_key { __u8 flow_flags; }; +struct ip_tunnel_encap { + u16 type; + u16 flags; + __be16 sport; + __be16 dport; +}; + /* Flags for ip_tunnel_info mode. */ #define IP_TUNNEL_INFO_TX 0x01 /* represents tx tunnel parameters */ #define IP_TUNNEL_INFO_IPV6 0x02 /* key contains IPv6 addresses */ @@ -66,9 +73,9 @@ struct ip_tunnel_key { #define IP_TUNNEL_OPTS_MAX \ GENMASK((sizeof_field(struct ip_tunnel_info, \ options_len) * BITS_PER_BYTE) - 1, 0) - struct ip_tunnel_info { struct ip_tunnel_key key; + struct ip_tunnel_encap encap; #ifdef CONFIG_DST_CACHE struct dst_cache dst_cache; #endif @@ -86,13 +93,6 @@ struct ip_tunnel_6rd_parm { }; #endif -struct ip_tunnel_encap { - u16 type; - u16 flags; - __be16 sport; - __be16 dport; -}; - struct ip_tunnel_prl_entry { struct ip_tunnel_prl_entry __rcu *next; __be32 addr; @@ -293,6 +293,7 @@ struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn, __be32 remote, __be32 local, __be32 key); +void ip_tunnel_md_udp_encap(struct sk_buff *skb, struct ip_tunnel_info *info); int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb, const struct tnl_ptk_info *tpi, struct metadata_dst *tun_dst, bool log_ecn_error); @@ -371,22 +372,23 @@ static inline int ip_encap_hlen(struct ip_tunnel_encap *e) return hlen; } -static inline int ip_tunnel_encap(struct sk_buff *skb, struct ip_tunnel *t, +static inline int ip_tunnel_encap(struct sk_buff *skb, + struct ip_tunnel_encap *e, u8 *protocol, struct flowi4 *fl4) { const struct ip_tunnel_encap_ops *ops; int ret = -EINVAL; - if (t->encap.type == TUNNEL_ENCAP_NONE) + if (e->type == TUNNEL_ENCAP_NONE) return 0; - if (t->encap.type >= MAX_IPTUN_ENCAP_OPS) + if (e->type >= MAX_IPTUN_ENCAP_OPS) return -EINVAL; rcu_read_lock(); - ops = rcu_dereference(iptun_encaps[t->encap.type]); + ops = rcu_dereference(iptun_encaps[e->type]); if (likely(ops && ops->build_header)) - ret = ops->build_header(skb, &t->encap, protocol, fl4); + ret = ops->build_header(skb, e, protocol, fl4); rcu_read_unlock(); return ret; diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index de90b09dfe78..add437f710fc 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -359,6 +359,20 @@ err_dev_set_mtu: return ERR_PTR(err); } +void ip_tunnel_md_udp_encap(struct sk_buff *skb, struct ip_tunnel_info *info) +{ + const struct iphdr *iph = ip_hdr(skb); + const struct udphdr *udph; + + if (iph->protocol != IPPROTO_UDP) + return; + + udph = (struct udphdr *)((__u8 *)iph + (iph->ihl << 2)); + info->encap.sport = udph->source; + info->encap.dport = udph->dest; +} +EXPORT_SYMBOL(ip_tunnel_md_udp_encap); + int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb, const struct tnl_ptk_info *tpi, struct metadata_dst *tun_dst, bool log_ecn_error) @@ -572,7 +586,11 @@ void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, tunnel_id_to_key32(key->tun_id), RT_TOS(tos), dev_net(dev), 0, skb->mark, skb_get_hash(skb), key->flow_flags); - if (tunnel->encap.type != TUNNEL_ENCAP_NONE) + + if (!tunnel_hlen) + tunnel_hlen = ip_encap_hlen(&tun_info->encap); + + if (ip_tunnel_encap(skb, &tun_info->encap, &proto, &fl4) < 0) goto tx_error; use_cache = ip_tunnel_dst_cache_usable(skb, tun_info); @@ -732,7 +750,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, dev_net(dev), tunnel->parms.link, tunnel->fwmark, skb_get_hash(skb), 0); - if (ip_tunnel_encap(skb, tunnel, &protocol, &fl4) < 0) + if (ip_tunnel_encap(skb, &tunnel->encap, &protocol, &fl4) < 0) goto tx_error; if (connected && md) { diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index abea77759b7e..27b8f83c6ea2 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -241,6 +241,7 @@ static int ipip_tunnel_rcv(struct sk_buff *skb, u8 ipproto) tun_dst = ip_tun_rx_dst(skb, 0, 0, 0); if (!tun_dst) return 0; + ip_tunnel_md_udp_encap(skb, &tun_dst->u.tun_info); } skb_reset_mac_header(skb); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 70d81bba5093..063560e2cb1a 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -1024,7 +1024,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, ttl = iph6->hop_limit; tos = INET_ECN_encapsulate(tos, ipv6_get_dsfield(iph6)); - if (ip_tunnel_encap(skb, tunnel, &protocol, &fl4) < 0) { + if (ip_tunnel_encap(skb, &tunnel->encap, &protocol, &fl4) < 0) { ip_rt_put(rt); goto tx_error; } -- cgit v1.2.3 From c50e96099edb134bf107fafc02715fbc4aa2277f Mon Sep 17 00:00:00 2001 From: Christian Ehrig Date: Fri, 7 Apr 2023 15:38:54 +0200 Subject: bpf,fou: Add bpf_skb_{set,get}_fou_encap kfuncs Add two new kfuncs that allow a BPF tc-hook, installed on an ipip device in collect-metadata mode, to control FOU encap parameters on a per-packet level. The set of kfuncs is registered with the fou module. The bpf_skb_set_fou_encap kfunc is supposed to be used in tandem and after a successful call to the bpf_skb_set_tunnel_key bpf-helper. UDP source and destination ports can be controlled by passing a struct bpf_fou_encap. A source port of zero will auto-assign a source port. enum bpf_fou_encap_type is used to specify if the egress path should FOU or GUE encap the packet. On the ingress path bpf_skb_get_fou_encap can be used to read UDP source and destination ports from the receiver's point of view and allows for packet multiplexing across different destination ports within a single BPF program and ipip device. Signed-off-by: Christian Ehrig Link: https://lore.kernel.org/r/e17c94a646b63e78ce0dbf3f04b2c33dc948a32d.1680874078.git.cehrig@cloudflare.com Signed-off-by: Alexei Starovoitov --- include/net/fou.h | 2 + net/ipv4/Makefile | 2 +- net/ipv4/fou_bpf.c | 119 ++++++++++++++++++++++++++++++++++++++++++++++++++++ net/ipv4/fou_core.c | 5 +++ 4 files changed, 127 insertions(+), 1 deletion(-) create mode 100644 net/ipv4/fou_bpf.c (limited to 'net') diff --git a/include/net/fou.h b/include/net/fou.h index 80f56e275b08..824eb4b231fd 100644 --- a/include/net/fou.h +++ b/include/net/fou.h @@ -17,4 +17,6 @@ int __fou_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e, int __gue_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e, u8 *protocol, __be16 *sport, int type); +int register_fou_bpf(void); + #endif diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 880277c9fd07..b18ba8ef93ad 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -26,7 +26,7 @@ obj-$(CONFIG_IP_MROUTE) += ipmr.o obj-$(CONFIG_IP_MROUTE_COMMON) += ipmr_base.o obj-$(CONFIG_NET_IPIP) += ipip.o gre-y := gre_demux.o -fou-y := fou_core.o fou_nl.o +fou-y := fou_core.o fou_nl.o fou_bpf.o obj-$(CONFIG_NET_FOU) += fou.o obj-$(CONFIG_NET_IPGRE_DEMUX) += gre.o obj-$(CONFIG_NET_IPGRE) += ip_gre.o diff --git a/net/ipv4/fou_bpf.c b/net/ipv4/fou_bpf.c new file mode 100644 index 000000000000..3760a14b6b57 --- /dev/null +++ b/net/ipv4/fou_bpf.c @@ -0,0 +1,119 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Unstable Fou Helpers for TC-BPF hook + * + * These are called from SCHED_CLS BPF programs. Note that it is + * allowed to break compatibility for these functions since the interface they + * are exposed through to BPF programs is explicitly unstable. + */ + +#include +#include + +#include +#include + +struct bpf_fou_encap { + __be16 sport; + __be16 dport; +}; + +enum bpf_fou_encap_type { + FOU_BPF_ENCAP_FOU, + FOU_BPF_ENCAP_GUE, +}; + +__diag_push(); +__diag_ignore_all("-Wmissing-prototypes", + "Global functions as their definitions will be in BTF"); + +/* bpf_skb_set_fou_encap - Set FOU encap parameters + * + * This function allows for using GUE or FOU encapsulation together with an + * ipip device in collect-metadata mode. + * + * It is meant to be used in BPF tc-hooks and after a call to the + * bpf_skb_set_tunnel_key helper, responsible for setting IP addresses. + * + * Parameters: + * @skb_ctx Pointer to ctx (__sk_buff) in TC program. Cannot be NULL + * @encap Pointer to a `struct bpf_fou_encap` storing UDP src and + * dst ports. If sport is set to 0 the kernel will auto-assign a + * port. This is similar to using `encap-sport auto`. + * Cannot be NULL + * @type Encapsulation type for the packet. Their definitions are + * specified in `enum bpf_fou_encap_type` + */ +__bpf_kfunc int bpf_skb_set_fou_encap(struct __sk_buff *skb_ctx, + struct bpf_fou_encap *encap, int type) +{ + struct sk_buff *skb = (struct sk_buff *)skb_ctx; + struct ip_tunnel_info *info = skb_tunnel_info(skb); + + if (unlikely(!encap)) + return -EINVAL; + + if (unlikely(!info || !(info->mode & IP_TUNNEL_INFO_TX))) + return -EINVAL; + + switch (type) { + case FOU_BPF_ENCAP_FOU: + info->encap.type = TUNNEL_ENCAP_FOU; + break; + case FOU_BPF_ENCAP_GUE: + info->encap.type = TUNNEL_ENCAP_GUE; + break; + default: + info->encap.type = TUNNEL_ENCAP_NONE; + } + + if (info->key.tun_flags & TUNNEL_CSUM) + info->encap.flags |= TUNNEL_ENCAP_FLAG_CSUM; + + info->encap.sport = encap->sport; + info->encap.dport = encap->dport; + + return 0; +} + +/* bpf_skb_get_fou_encap - Get FOU encap parameters + * + * This function allows for reading encap metadata from a packet received + * on an ipip device in collect-metadata mode. + * + * Parameters: + * @skb_ctx Pointer to ctx (__sk_buff) in TC program. Cannot be NULL + * @encap Pointer to a struct bpf_fou_encap storing UDP source and + * destination port. Cannot be NULL + */ +__bpf_kfunc int bpf_skb_get_fou_encap(struct __sk_buff *skb_ctx, + struct bpf_fou_encap *encap) +{ + struct sk_buff *skb = (struct sk_buff *)skb_ctx; + struct ip_tunnel_info *info = skb_tunnel_info(skb); + + if (unlikely(!info)) + return -EINVAL; + + encap->sport = info->encap.sport; + encap->dport = info->encap.dport; + + return 0; +} + +__diag_pop() + +BTF_SET8_START(fou_kfunc_set) +BTF_ID_FLAGS(func, bpf_skb_set_fou_encap) +BTF_ID_FLAGS(func, bpf_skb_get_fou_encap) +BTF_SET8_END(fou_kfunc_set) + +static const struct btf_kfunc_id_set fou_bpf_kfunc_set = { + .owner = THIS_MODULE, + .set = &fou_kfunc_set, +}; + +int register_fou_bpf(void) +{ + return register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS, + &fou_bpf_kfunc_set); +} diff --git a/net/ipv4/fou_core.c b/net/ipv4/fou_core.c index cafec9b4eee0..0c41076e31ed 100644 --- a/net/ipv4/fou_core.c +++ b/net/ipv4/fou_core.c @@ -1236,10 +1236,15 @@ static int __init fou_init(void) if (ret < 0) goto unregister; + ret = register_fou_bpf(); + if (ret < 0) + goto kfunc_failed; + ret = ip_tunnel_encap_add_fou_ops(); if (ret == 0) return 0; +kfunc_failed: genl_unregister_family(&fou_nl_family); unregister: unregister_pernet_device(&fou_net_ops); -- cgit v1.2.3 From 59d3efd27c11c59b32291e5ebc307bed2edb65ee Mon Sep 17 00:00:00 2001 From: Martin Willi Date: Tue, 11 Apr 2023 09:43:19 +0200 Subject: rtnetlink: Restore RTM_NEW/DELLINK notification behavior The commits referenced below allows userspace to use the NLM_F_ECHO flag for RTM_NEW/DELLINK operations to receive unicast notifications for the affected link. Prior to these changes, applications may have relied on multicast notifications to learn the same information without specifying the NLM_F_ECHO flag. For such applications, the mentioned commits changed the behavior for requests not using NLM_F_ECHO. Multicast notifications are still received, but now use the portid of the requester and the sequence number of the request instead of zero values used previously. For the application, this message may be unexpected and likely handled as a response to the NLM_F_ACKed request, especially if it uses the same socket to handle requests and notifications. To fix existing applications relying on the old notification behavior, set the portid and sequence number in the notification only if the request included the NLM_F_ECHO flag. This restores the old behavior for applications not using it, but allows unicasted notifications for others. Fixes: f3a63cce1b4f ("rtnetlink: Honour NLM_F_ECHO flag in rtnl_delete_link") Fixes: d88e136cab37 ("rtnetlink: Honour NLM_F_ECHO flag in rtnl_newlink_create") Signed-off-by: Martin Willi Acked-by: Guillaume Nault Acked-by: Hangbin Liu Link: https://lore.kernel.org/r/20230411074319.24133-1-martin@strongswan.org Signed-off-by: Jakub Kicinski --- include/linux/rtnetlink.h | 3 ++- net/core/dev.c | 2 +- net/core/rtnetlink.c | 11 +++++++++-- 3 files changed, 12 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 92ad75549e9c..b6e6378dcbbd 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -25,7 +25,8 @@ void rtmsg_ifinfo_newnet(int type, struct net_device *dev, unsigned int change, struct sk_buff *rtmsg_ifinfo_build_skb(int type, struct net_device *dev, unsigned change, u32 event, gfp_t flags, int *new_nsid, - int new_ifindex, u32 portid, u32 seq); + int new_ifindex, u32 portid, + const struct nlmsghdr *nlh); void rtmsg_ifinfo_send(struct sk_buff *skb, struct net_device *dev, gfp_t flags, u32 portid, const struct nlmsghdr *nlh); diff --git a/net/core/dev.c b/net/core/dev.c index 48067321c0db..1488f700bf81 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -10847,7 +10847,7 @@ void unregister_netdevice_many_notify(struct list_head *head, dev->rtnl_link_state == RTNL_LINK_INITIALIZED) skb = rtmsg_ifinfo_build_skb(RTM_DELLINK, dev, ~0U, 0, GFP_KERNEL, NULL, 0, - portid, nlmsg_seq(nlh)); + portid, nlh); /* * Flush the unicast and multicast chains diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 5d8eb57867a9..6e44e92ebdf5 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -3972,16 +3972,23 @@ static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb) struct sk_buff *rtmsg_ifinfo_build_skb(int type, struct net_device *dev, unsigned int change, u32 event, gfp_t flags, int *new_nsid, - int new_ifindex, u32 portid, u32 seq) + int new_ifindex, u32 portid, + const struct nlmsghdr *nlh) { struct net *net = dev_net(dev); struct sk_buff *skb; int err = -ENOBUFS; + u32 seq = 0; skb = nlmsg_new(if_nlmsg_size(dev, 0), flags); if (skb == NULL) goto errout; + if (nlmsg_report(nlh)) + seq = nlmsg_seq(nlh); + else + portid = 0; + err = rtnl_fill_ifinfo(skb, dev, dev_net(dev), type, portid, seq, change, 0, 0, event, new_nsid, new_ifindex, -1, flags); @@ -4017,7 +4024,7 @@ static void rtmsg_ifinfo_event(int type, struct net_device *dev, return; skb = rtmsg_ifinfo_build_skb(type, dev, change, event, flags, new_nsid, - new_ifindex, portid, nlmsg_seq(nlh)); + new_ifindex, portid, nlh); if (skb) rtmsg_ifinfo_send(skb, dev, flags, portid, nlh); } -- cgit v1.2.3 From 6417070918de3bcdbe0646e7256dae58fd8083ba Mon Sep 17 00:00:00 2001 From: Ziyang Xuan Date: Mon, 10 Apr 2023 09:23:52 +0800 Subject: net: qrtr: Fix an uninit variable access bug in qrtr_tx_resume() Syzbot reported a bug as following: ===================================================== BUG: KMSAN: uninit-value in qrtr_tx_resume+0x185/0x1f0 net/qrtr/af_qrtr.c:230 qrtr_tx_resume+0x185/0x1f0 net/qrtr/af_qrtr.c:230 qrtr_endpoint_post+0xf85/0x11b0 net/qrtr/af_qrtr.c:519 qrtr_tun_write_iter+0x270/0x400 net/qrtr/tun.c:108 call_write_iter include/linux/fs.h:2189 [inline] aio_write+0x63a/0x950 fs/aio.c:1600 io_submit_one+0x1d1c/0x3bf0 fs/aio.c:2019 __do_sys_io_submit fs/aio.c:2078 [inline] __se_sys_io_submit+0x293/0x770 fs/aio.c:2048 __x64_sys_io_submit+0x92/0xd0 fs/aio.c:2048 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Uninit was created at: slab_post_alloc_hook mm/slab.h:766 [inline] slab_alloc_node mm/slub.c:3452 [inline] __kmem_cache_alloc_node+0x71f/0xce0 mm/slub.c:3491 __do_kmalloc_node mm/slab_common.c:967 [inline] __kmalloc_node_track_caller+0x114/0x3b0 mm/slab_common.c:988 kmalloc_reserve net/core/skbuff.c:492 [inline] __alloc_skb+0x3af/0x8f0 net/core/skbuff.c:565 __netdev_alloc_skb+0x120/0x7d0 net/core/skbuff.c:630 qrtr_endpoint_post+0xbd/0x11b0 net/qrtr/af_qrtr.c:446 qrtr_tun_write_iter+0x270/0x400 net/qrtr/tun.c:108 call_write_iter include/linux/fs.h:2189 [inline] aio_write+0x63a/0x950 fs/aio.c:1600 io_submit_one+0x1d1c/0x3bf0 fs/aio.c:2019 __do_sys_io_submit fs/aio.c:2078 [inline] __se_sys_io_submit+0x293/0x770 fs/aio.c:2048 __x64_sys_io_submit+0x92/0xd0 fs/aio.c:2048 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd It is because that skb->len requires at least sizeof(struct qrtr_ctrl_pkt) in qrtr_tx_resume(). And skb->len equals to size in qrtr_endpoint_post(). But size is less than sizeof(struct qrtr_ctrl_pkt) when qrtr_cb->type equals to QRTR_TYPE_RESUME_TX in qrtr_endpoint_post() under the syzbot scenario. This triggers the uninit variable access bug. Add size check when qrtr_cb->type equals to QRTR_TYPE_RESUME_TX in qrtr_endpoint_post() to fix the bug. Fixes: 5fdeb0d372ab ("net: qrtr: Implement outgoing flow control") Reported-by: syzbot+4436c9630a45820fda76@syzkaller.appspotmail.com Link: https://syzkaller.appspot.com/bug?id=c14607f0963d27d5a3d5f4c8639b500909e43540 Suggested-by: Manivannan Sadhasivam Signed-off-by: Ziyang Xuan Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230410012352.3997823-1-william.xuanziyang@huawei.com Signed-off-by: Paolo Abeni --- net/qrtr/af_qrtr.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/qrtr/af_qrtr.c b/net/qrtr/af_qrtr.c index 3a70255c8d02..76f0434d3d06 100644 --- a/net/qrtr/af_qrtr.c +++ b/net/qrtr/af_qrtr.c @@ -498,6 +498,11 @@ int qrtr_endpoint_post(struct qrtr_endpoint *ep, const void *data, size_t len) if (!size || len != ALIGN(size, 4) + hdrlen) goto err; + if ((cb->type == QRTR_TYPE_NEW_SERVER || + cb->type == QRTR_TYPE_RESUME_TX) && + size < sizeof(struct qrtr_ctrl_pkt)) + goto err; + if (cb->dst_port != QRTR_PORT_CTRL && cb->type != QRTR_TYPE_DATA && cb->type != QRTR_TYPE_RESUME_TX) goto err; @@ -510,9 +515,6 @@ int qrtr_endpoint_post(struct qrtr_endpoint *ep, const void *data, size_t len) /* Remote node endpoint can bridge other distant nodes */ const struct qrtr_ctrl_pkt *pkt; - if (size < sizeof(*pkt)) - goto err; - pkt = data + hdrlen; qrtr_node_assign(node, le32_to_cpu(pkt->server.node)); } -- cgit v1.2.3 From 32832a2caf82663870126c5186cf8f86c8b2a649 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Mon, 10 Apr 2023 15:43:30 -0400 Subject: sctp: fix a potential overflow in sctp_ifwdtsn_skip Currently, when traversing ifwdtsn skips with _sctp_walk_ifwdtsn, it only checks the pos against the end of the chunk. However, the data left for the last pos may be < sizeof(struct sctp_ifwdtsn_skip), and dereference it as struct sctp_ifwdtsn_skip may cause coverflow. This patch fixes it by checking the pos against "the end of the chunk - sizeof(struct sctp_ifwdtsn_skip)" in sctp_ifwdtsn_skip, similar to sctp_fwdtsn_skip. Fixes: 0fc2ea922c8a ("sctp: implement validate_ftsn for sctp_stream_interleave") Signed-off-by: Xin Long Link: https://lore.kernel.org/r/2a71bffcd80b4f2c61fac6d344bb2f11c8fd74f7.1681155810.git.lucien.xin@gmail.com Signed-off-by: Paolo Abeni --- net/sctp/stream_interleave.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sctp/stream_interleave.c b/net/sctp/stream_interleave.c index 94727feb07b3..b046b11200c9 100644 --- a/net/sctp/stream_interleave.c +++ b/net/sctp/stream_interleave.c @@ -1154,7 +1154,8 @@ static void sctp_generate_iftsn(struct sctp_outq *q, __u32 ctsn) #define _sctp_walk_ifwdtsn(pos, chunk, end) \ for (pos = chunk->subh.ifwdtsn_hdr->skip; \ - (void *)pos < (void *)chunk->subh.ifwdtsn_hdr->skip + (end); pos++) + (void *)pos <= (void *)chunk->subh.ifwdtsn_hdr->skip + (end) - \ + sizeof(struct sctp_ifwdtsn_skip); pos++) #define sctp_walk_ifwdtsn(pos, ch) \ _sctp_walk_ifwdtsn((pos), (ch), ntohs((ch)->chunk_hdr->length) - \ -- cgit v1.2.3 From 0c5f48599bed55c13d6f979ea824554bdebdf2ad Mon Sep 17 00:00:00 2001 From: Kal Conley Date: Mon, 10 Apr 2023 14:18:41 +0200 Subject: xsk: Simplify xp_aligned_validate_desc implementation Perform the chunk boundary check like the page boundary check in xp_desc_crosses_non_contig_pg(). This simplifies the implementation and reduces the number of branches. Signed-off-by: Kal Conley Signed-off-by: Daniel Borkmann Acked-by: Magnus Karlsson Link: https://lore.kernel.org/bpf/20230410121841.643254-1-kal.conley@dectris.com --- net/xdp/xsk_queue.h | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h index 66c6f57c9c44..76b574bffd4a 100644 --- a/net/xdp/xsk_queue.h +++ b/net/xdp/xsk_queue.h @@ -133,16 +133,12 @@ static inline bool xskq_cons_read_addr_unchecked(struct xsk_queue *q, u64 *addr) static inline bool xp_aligned_validate_desc(struct xsk_buff_pool *pool, struct xdp_desc *desc) { - u64 chunk, chunk_end; + u64 offset = desc->addr & (pool->chunk_size - 1); - chunk = xp_aligned_extract_addr(pool, desc->addr); - if (likely(desc->len)) { - chunk_end = xp_aligned_extract_addr(pool, desc->addr + desc->len - 1); - if (chunk != chunk_end) - return false; - } + if (offset + desc->len > pool->chunk_size) + return false; - if (chunk >= pool->addrs_cnt) + if (desc->addr >= pool->addrs_cnt) return false; if (desc->options) -- cgit v1.2.3 From 1ba83f505c53d35eda892ac7f108ef1189da6fa8 Mon Sep 17 00:00:00 2001 From: Kal Conley Date: Tue, 11 Apr 2023 15:00:25 +0200 Subject: xsk: Elide base_addr comparison in xp_unaligned_validate_desc Remove redundant (base_addr >= pool->addrs_cnt) comparison from the conditional. In particular, addr is computed as: addr = base_addr + offset ... where base_addr and offset are stored as 48-bit and 16-bit unsigned integers, respectively. The above sum cannot overflow u64 since base_addr has a maximum value of 0x0000ffffffffffff and offset has a maximum value of 0xffff (implying a maximum sum of 0x000100000000fffe). Since overflow is impossible, it follows that addr >= base_addr. Now if (base_addr >= pool->addrs_cnt), then clearly: addr >= base_addr >= pool->addrs_cnt Thus, (base_addr >= pool->addrs_cnt) implies (addr >= pool->addrs_cnt). Subsequently, the former comparison is unnecessary in the conditional since for any boolean expressions A and B, (A || B) && (A -> B) is equivalent to B. Signed-off-by: Kal Conley Signed-off-by: Daniel Borkmann Acked-by: Magnus Karlsson Link: https://lore.kernel.org/bpf/20230411130025.19704-1-kal.conley@dectris.com --- net/xdp/xsk_queue.h | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h index 76b574bffd4a..6d40a77fccbe 100644 --- a/net/xdp/xsk_queue.h +++ b/net/xdp/xsk_queue.h @@ -149,16 +149,12 @@ static inline bool xp_aligned_validate_desc(struct xsk_buff_pool *pool, static inline bool xp_unaligned_validate_desc(struct xsk_buff_pool *pool, struct xdp_desc *desc) { - u64 addr, base_addr; - - base_addr = xp_unaligned_extract_addr(desc->addr); - addr = xp_unaligned_add_offset_to_addr(desc->addr); + u64 addr = xp_unaligned_add_offset_to_addr(desc->addr); if (desc->len > pool->chunk_size) return false; - if (base_addr >= pool->addrs_cnt || addr >= pool->addrs_cnt || - addr + desc->len > pool->addrs_cnt || + if (addr >= pool->addrs_cnt || addr + desc->len > pool->addrs_cnt || xp_desc_crosses_non_contig_pg(pool, addr, desc->len)) return false; -- cgit v1.2.3 From c8bc34660628769f71e8b230144a2a4d86ab0f91 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 13 Apr 2023 14:51:56 +0100 Subject: sunrpc: Fix RFC6803 encryption test The usage_data[] array in rfc6803_encrypt_case() is uninitialised, so clear it as it may cause the tests to fail otherwise. Fixes: b958cff6b27b ("SUNRPC: Add encryption KUnit tests for the RFC 6803 encryption types") Link: https://lore.kernel.org/r/380323.1681314997@warthog.procyon.org.uk/ Signed-off-by: David Howells cc: Chuck Lever cc: Scott Mayhew cc: Herbert Xu cc: linux-nfs@vger.kernel.org cc: linux-crypto@vger.kernel.org Signed-off-by: Chuck Lever --- net/sunrpc/auth_gss/gss_krb5_test.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/sunrpc/auth_gss/gss_krb5_test.c b/net/sunrpc/auth_gss/gss_krb5_test.c index ce0541e32fc9..aa6ec4e858aa 100644 --- a/net/sunrpc/auth_gss/gss_krb5_test.c +++ b/net/sunrpc/auth_gss/gss_krb5_test.c @@ -1327,6 +1327,7 @@ static void rfc6803_encrypt_case(struct kunit *test) if (!gk5e) kunit_skip(test, "Encryption type is not available"); + memset(usage_data, 0, sizeof(usage_data)); usage.data[3] = param->constant; Ke.len = gk5e->Ke_length; -- cgit v1.2.3 From 2c9abe653bc5134eeab411c46dde008d8a1c37b0 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 31 Mar 2023 16:59:08 +0200 Subject: wifi: ieee80211: correctly mark FTM frames non-bufferable The checks of whether or not a frame is bufferable were not taking into account that some action frames aren't, such as FTM. Check this, which requires some changes to the function ieee80211_is_bufferable_mmpdu() since we need the whole skb for the checks now. Reviewed-by: Ilan Peer Signed-off-by: Johannes Berg --- drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 7 ++-- drivers/net/wireless/mediatek/mt76/tx.c | 2 +- include/linux/ieee80211.h | 52 +++++++++++++++++++++-------- net/mac80211/tx.c | 4 +-- 4 files changed, 45 insertions(+), 20 deletions(-) (limited to 'net') diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c index 51f21cbf7a20..478442e16d43 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c @@ -604,8 +604,9 @@ static void iwl_mvm_skb_prepare_status(struct sk_buff *skb, static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm, struct iwl_mvm_vif_link_info *link, struct ieee80211_tx_info *info, - struct ieee80211_hdr *hdr) + struct sk_buff *skb) { + struct ieee80211_hdr *hdr = (void *)skb->data; __le16 fc = hdr->frame_control; switch (info->control.vif->type) { @@ -622,7 +623,7 @@ static int iwl_mvm_get_ctrl_vif_queue(struct iwl_mvm *mvm, * reason 7 ("Class 3 frame received from nonassociated STA"). */ if (ieee80211_is_mgmt(fc) && - (!ieee80211_is_bufferable_mmpdu(fc) || + (!ieee80211_is_bufferable_mmpdu(skb) || ieee80211_is_deauth(fc) || ieee80211_is_disassoc(fc))) return link->mgmt_queue; @@ -755,7 +756,7 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb) sta_id = link->mcast_sta.sta_id; queue = iwl_mvm_get_ctrl_vif_queue(mvm, link, &info, - hdr); + skb); } else if (info.control.vif->type == NL80211_IFTYPE_MONITOR) { queue = mvm->snif_queue; sta_id = mvm->snif_sta.sta_id; diff --git a/drivers/net/wireless/mediatek/mt76/tx.c b/drivers/net/wireless/mediatek/mt76/tx.c index 1f309d05380a..3ad9742364ba 100644 --- a/drivers/net/wireless/mediatek/mt76/tx.c +++ b/drivers/net/wireless/mediatek/mt76/tx.c @@ -330,7 +330,7 @@ mt76_tx(struct mt76_phy *phy, struct ieee80211_sta *sta, if ((dev->drv->drv_flags & MT_DRV_HW_MGMT_TXQ) && !(info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP) && !ieee80211_is_data(hdr->frame_control) && - !ieee80211_is_bufferable_mmpdu(hdr->frame_control)) { + !ieee80211_is_bufferable_mmpdu(skb)) { qid = MT_TXQ_PSD; } diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 0583b2b0ce1f..c4cf296e7eaf 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -782,20 +782,6 @@ static inline bool ieee80211_is_any_nullfunc(__le16 fc) return (ieee80211_is_nullfunc(fc) || ieee80211_is_qos_nullfunc(fc)); } -/** - * ieee80211_is_bufferable_mmpdu - check if frame is bufferable MMPDU - * @fc: frame control field in little-endian byteorder - */ -static inline bool ieee80211_is_bufferable_mmpdu(__le16 fc) -{ - /* IEEE 802.11-2012, definition of "bufferable management frame"; - * note that this ignores the IBSS special case. */ - return ieee80211_is_mgmt(fc) && - (ieee80211_is_action(fc) || - ieee80211_is_disassoc(fc) || - ieee80211_is_deauth(fc)); -} - /** * ieee80211_is_first_frag - check if IEEE80211_SCTL_FRAG is not set * @seq_ctrl: frame sequence control bytes in little-endian byteorder @@ -4132,6 +4118,44 @@ static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr) return hdr->addr1; } +/** + * ieee80211_is_bufferable_mmpdu - check if frame is bufferable MMPDU + * @skb: the skb to check, starting with the 802.11 header + */ +static inline bool ieee80211_is_bufferable_mmpdu(struct sk_buff *skb) +{ + struct ieee80211_mgmt *mgmt = (void *)skb->data; + __le16 fc = mgmt->frame_control; + + /* + * IEEE 802.11 REVme D2.0 definition of bufferable MMPDU; + * note that this ignores the IBSS special case. + */ + if (!ieee80211_is_mgmt(fc)) + return false; + + if (ieee80211_is_disassoc(fc) || ieee80211_is_deauth(fc)) + return true; + + if (!ieee80211_is_action(fc)) + return false; + + if (skb->len < offsetofend(typeof(*mgmt), u.action.u.ftm.action_code)) + return true; + + /* action frame - additionally check for non-bufferable FTM */ + + if (mgmt->u.action.category != WLAN_CATEGORY_PUBLIC && + mgmt->u.action.category != WLAN_CATEGORY_PROTECTED_DUAL_OF_ACTION) + return true; + + if (mgmt->u.action.u.ftm.action_code == WLAN_PUB_ACTION_FTM_REQUEST || + mgmt->u.action.u.ftm.action_code == WLAN_PUB_ACTION_FTM_RESPONSE) + return false; + + return true; +} + /** * _ieee80211_is_robust_mgmt_frame - check if frame is a robust management frame * @hdr: the frame (buffer must include at least the first octet of payload) diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index dfe6b9c9b29e..1a3327407552 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -488,7 +488,7 @@ ieee80211_tx_h_unicast_ps_buf(struct ieee80211_tx_data *tx) int ac = skb_get_queue_mapping(tx->skb); if (ieee80211_is_mgmt(hdr->frame_control) && - !ieee80211_is_bufferable_mmpdu(hdr->frame_control)) { + !ieee80211_is_bufferable_mmpdu(tx->skb)) { info->flags |= IEEE80211_TX_CTL_NO_PS_BUFFER; return TX_CONTINUE; } @@ -1323,7 +1323,7 @@ static struct txq_info *ieee80211_get_txq(struct ieee80211_local *local, if (!(info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP) && unlikely(!ieee80211_is_data_present(hdr->frame_control))) { if ((!ieee80211_is_mgmt(hdr->frame_control) || - ieee80211_is_bufferable_mmpdu(hdr->frame_control) || + ieee80211_is_bufferable_mmpdu(skb) || vif->type == NL80211_IFTYPE_STATION) && sta && sta->uploaded) { /* -- cgit v1.2.3 From 0b75a1b1e42e07ae84e3a11d2368b418546e2bec Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 31 Mar 2023 16:59:16 +0200 Subject: wifi: mac80211: flush queues on STA removal When we remove a station, we first make it unreachable, then we (must) remove its keys, and then remove the station itself. Depending on the hardware design, if we have hardware crypto at all, frames still sitting on hardware queues may then be transmitted without a valid key, possibly unencrypted or with a fixed key. Fix this by flushing the queues when removing stations so this cannot happen. Signed-off-by: Johannes Berg --- net/mac80211/sta_info.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'net') diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 941bda9141fa..ce7c3b997269 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -1294,6 +1294,14 @@ static void __sta_info_destroy_part2(struct sta_info *sta) WARN_ON_ONCE(ret); } + /* Flush queues before removing keys, as that might remove them + * from hardware, and then depending on the offload method, any + * frames sitting on hardware queues might be sent out without + * any encryption at all. + */ + if (local->ops->set_key) + ieee80211_flush_queues(local, sta->sdata, false); + /* now keys can no longer be reached */ ieee80211_free_sta_keys(local, sta); -- cgit v1.2.3 From d00800a289c9349bb659a698cbd7bc04521dc927 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 31 Mar 2023 16:59:17 +0200 Subject: wifi: mac80211: add flush_sta method Some drivers like iwlwifi might have per-STA queues, so we may want to flush/drop just those queues rather than all when removing a station. Add a separate method for that. Signed-off-by: Johannes Berg --- include/net/mac80211.h | 6 ++++++ net/mac80211/driver-ops.h | 15 +++++++++++++++ net/mac80211/sta_info.c | 8 ++++++-- net/mac80211/trace.h | 7 +++++++ 4 files changed, 34 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 679421d37a42..a8dadbd83d95 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -3950,6 +3950,10 @@ struct ieee80211_prep_tx_info { * Note that vif can be NULL. * The callback can sleep. * + * @flush_sta: Flush or drop all pending frames from the hardware queue(s) for + * the given station, as it's about to be removed. + * The callback can sleep. + * * @channel_switch: Drivers that need (or want) to offload the channel * switch operation for CSAs received from the AP may implement this * callback. They must then call ieee80211_chswitch_done() to indicate @@ -4415,6 +4419,8 @@ struct ieee80211_ops { #endif void (*flush)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, u32 queues, bool drop); + void (*flush_sta)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct ieee80211_sta *sta); void (*channel_switch)(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_channel_switch *ch_switch); diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h index 0bf208f5bbc5..45d3e53c7383 100644 --- a/net/mac80211/driver-ops.h +++ b/net/mac80211/driver-ops.h @@ -649,6 +649,21 @@ static inline void drv_flush(struct ieee80211_local *local, trace_drv_return_void(local); } +static inline void drv_flush_sta(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct sta_info *sta) +{ + might_sleep(); + + if (sdata && !check_sdata_in_driver(sdata)) + return; + + trace_drv_flush_sta(local, sdata, &sta->sta); + if (local->ops->flush_sta) + local->ops->flush_sta(&local->hw, &sdata->vif, &sta->sta); + trace_drv_return_void(local); +} + static inline void drv_channel_switch(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata, struct ieee80211_channel_switch *ch_switch) diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index ce7c3b997269..1400512e0dde 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -1299,8 +1299,12 @@ static void __sta_info_destroy_part2(struct sta_info *sta) * frames sitting on hardware queues might be sent out without * any encryption at all. */ - if (local->ops->set_key) - ieee80211_flush_queues(local, sta->sdata, false); + if (local->ops->set_key) { + if (local->ops->flush_sta) + drv_flush_sta(local, sta->sdata, sta); + else + ieee80211_flush_queues(local, sta->sdata, false); + } /* now keys can no longer be reached */ ieee80211_free_sta_keys(local, sta); diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h index e0ccf5fe708a..de5d69f21306 100644 --- a/net/mac80211/trace.h +++ b/net/mac80211/trace.h @@ -1177,6 +1177,13 @@ TRACE_EVENT(drv_flush, ) ); +DEFINE_EVENT(sta_event, drv_flush_sta, + TP_PROTO(struct ieee80211_local *local, + struct ieee80211_sub_if_data *sdata, + struct ieee80211_sta *sta), + TP_ARGS(local, sdata, sta) +); + TRACE_EVENT(drv_channel_switch, TP_PROTO(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata, -- cgit v1.2.3 From a5cb752b125766524c921faab1a45cc96065b0a7 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 11 Apr 2023 22:42:09 +0200 Subject: mptcp: use mptcp_schedule_work instead of open-coding it Beyond reducing code duplication this also avoids scheduling the mptcp_worker on a closed socket on some edge scenarios. The addressed issue is actually older than the blamed commit below, but this fix needs it as a pre-requisite. Fixes: ba8f48f7a4d7 ("mptcp: introduce mptcp_schedule_work") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: Jakub Kicinski --- net/mptcp/options.c | 5 ++--- net/mptcp/subflow.c | 18 ++++++------------ 2 files changed, 8 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/mptcp/options.c b/net/mptcp/options.c index b30cea2fbf3f..355f798d575a 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -1192,9 +1192,8 @@ bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) */ if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) { if (mp_opt.data_fin && mp_opt.data_len == 1 && - mptcp_update_rcv_data_fin(msk, mp_opt.data_seq, mp_opt.dsn64) && - schedule_work(&msk->work)) - sock_hold(subflow->conn); + mptcp_update_rcv_data_fin(msk, mp_opt.data_seq, mp_opt.dsn64)) + mptcp_schedule_work((struct sock *)msk); return true; } diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index a0041360ee9d..d34588850545 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -408,9 +408,8 @@ void mptcp_subflow_reset(struct sock *ssk) tcp_send_active_reset(ssk, GFP_ATOMIC); tcp_done(ssk); - if (!test_and_set_bit(MPTCP_WORK_CLOSE_SUBFLOW, &mptcp_sk(sk)->flags) && - schedule_work(&mptcp_sk(sk)->work)) - return; /* worker will put sk for us */ + if (!test_and_set_bit(MPTCP_WORK_CLOSE_SUBFLOW, &mptcp_sk(sk)->flags)) + mptcp_schedule_work(sk); sock_put(sk); } @@ -1118,8 +1117,8 @@ static enum mapping_status get_mapping_status(struct sock *ssk, skb_ext_del(skb, SKB_EXT_MPTCP); return MAPPING_OK; } else { - if (updated && schedule_work(&msk->work)) - sock_hold((struct sock *)msk); + if (updated) + mptcp_schedule_work((struct sock *)msk); return MAPPING_DATA_FIN; } @@ -1222,17 +1221,12 @@ static void mptcp_subflow_discard_data(struct sock *ssk, struct sk_buff *skb, /* sched mptcp worker to remove the subflow if no more data is pending */ static void subflow_sched_work_if_closed(struct mptcp_sock *msk, struct sock *ssk) { - struct sock *sk = (struct sock *)msk; - if (likely(ssk->sk_state != TCP_CLOSE)) return; if (skb_queue_empty(&ssk->sk_receive_queue) && - !test_and_set_bit(MPTCP_WORK_CLOSE_SUBFLOW, &msk->flags)) { - sock_hold(sk); - if (!schedule_work(&msk->work)) - sock_put(sk); - } + !test_and_set_bit(MPTCP_WORK_CLOSE_SUBFLOW, &msk->flags)) + mptcp_schedule_work((struct sock *)msk); } static bool subflow_can_fallback(struct mptcp_subflow_context *subflow) -- cgit v1.2.3 From d6a0443733434408f2cbd4c53fea6910599bab9e Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 11 Apr 2023 22:42:10 +0200 Subject: mptcp: stricter state check in mptcp_worker As reported by Christoph, the mptcp protocol can run the worker when the relevant msk socket is in an unexpected state: connect() // incoming reset + fastclose // the mptcp worker is scheduled mptcp_disconnect() // msk is now CLOSED listen() mptcp_worker() Leading to the following splat: divide error: 0000 [#1] PREEMPT SMP CPU: 1 PID: 21 Comm: kworker/1:0 Not tainted 6.3.0-rc1-gde5e8fd0123c #11 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014 Workqueue: events mptcp_worker RIP: 0010:__tcp_select_window+0x22c/0x4b0 net/ipv4/tcp_output.c:3018 RSP: 0018:ffffc900000b3c98 EFLAGS: 00010293 RAX: 000000000000ffd7 RBX: 000000000000ffd7 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffffffff8214ce97 RDI: 0000000000000004 RBP: 000000000000ffd7 R08: 0000000000000004 R09: 0000000000010000 R10: 000000000000ffd7 R11: ffff888005afa148 R12: 000000000000ffd7 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff88803ed00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000405270 CR3: 000000003011e006 CR4: 0000000000370ee0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: tcp_select_window net/ipv4/tcp_output.c:262 [inline] __tcp_transmit_skb+0x356/0x1280 net/ipv4/tcp_output.c:1345 tcp_transmit_skb net/ipv4/tcp_output.c:1417 [inline] tcp_send_active_reset+0x13e/0x320 net/ipv4/tcp_output.c:3459 mptcp_check_fastclose net/mptcp/protocol.c:2530 [inline] mptcp_worker+0x6c7/0x800 net/mptcp/protocol.c:2705 process_one_work+0x3bd/0x950 kernel/workqueue.c:2390 worker_thread+0x5b/0x610 kernel/workqueue.c:2537 kthread+0x138/0x170 kernel/kthread.c:376 ret_from_fork+0x2c/0x50 arch/x86/entry/entry_64.S:308 This change addresses the issue explicitly checking for bad states before running the mptcp worker. Fixes: e16163b6e2b7 ("mptcp: refactor shutdown and close") Cc: stable@vger.kernel.org Reported-by: Christoph Paasch Link: https://github.com/multipath-tcp/mptcp_net-next/issues/374 Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Tested-by: Christoph Paasch Signed-off-by: Matthieu Baerts Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 60b23b2716c4..06c5872e3b00 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2626,7 +2626,7 @@ static void mptcp_worker(struct work_struct *work) lock_sock(sk); state = sk->sk_state; - if (unlikely(state == TCP_CLOSE)) + if (unlikely((1 << state) & (TCPF_CLOSE | TCPF_LISTEN))) goto unlock; mptcp_check_data_fin_ack(sk); -- cgit v1.2.3 From c0ff6f6da66a7791a32c0234388b1bdc00244917 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 11 Apr 2023 22:42:11 +0200 Subject: mptcp: fix NULL pointer dereference on fastopen early fallback In case of early fallback to TCP, subflow_syn_recv_sock() deletes the subflow context before returning the newly allocated sock to the caller. The fastopen path does not cope with the above unconditionally dereferencing the subflow context. Fixes: 36b122baf6a8 ("mptcp: add subflow_v(4,6)_send_synack()") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: Jakub Kicinski --- net/mptcp/fastopen.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mptcp/fastopen.c b/net/mptcp/fastopen.c index d237d142171c..bceaab8dd8e4 100644 --- a/net/mptcp/fastopen.c +++ b/net/mptcp/fastopen.c @@ -9,11 +9,18 @@ void mptcp_fastopen_subflow_synack_set_params(struct mptcp_subflow_context *subflow, struct request_sock *req) { - struct sock *ssk = subflow->tcp_sock; - struct sock *sk = subflow->conn; + struct sock *sk, *ssk; struct sk_buff *skb; struct tcp_sock *tp; + /* on early fallback the subflow context is deleted by + * subflow_syn_recv_sock() + */ + if (!subflow) + return; + + ssk = subflow->tcp_sock; + sk = subflow->conn; tp = tcp_sk(ssk); subflow->is_mptfo = 1; -- cgit v1.2.3 From 1c5950fc6fe996235f1d18539b9c6b64b597f50f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 12 Apr 2023 13:03:08 +0000 Subject: udp6: fix potential access to stale information MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit lena wang reported an issue caused by udpv6_sendmsg() mangling msg->msg_name and msg->msg_namelen, which are later read from ____sys_sendmsg() : /* * If this is sendmmsg() and sending to current destination address was * successful, remember it. */ if (used_address && err >= 0) { used_address->name_len = msg_sys->msg_namelen; if (msg_sys->msg_name) memcpy(&used_address->name, msg_sys->msg_name, used_address->name_len); } udpv6_sendmsg() wants to pretend the remote address family is AF_INET in order to call udp_sendmsg(). A fix would be to modify the address in-place, instead of using a local variable, but this could have other side effects. Instead, restore initial values before we return from udpv6_sendmsg(). Fixes: c71d8ebe7a44 ("net: Fix security_socket_sendmsg() bypass problem.") Reported-by: lena wang Signed-off-by: Eric Dumazet Reviewed-by: Maciej Żenczykowski Link: https://lore.kernel.org/r/20230412130308.1202254-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/udp.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 9fb2f33ee3a7..a675acfb901d 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -1395,9 +1395,11 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) msg->msg_name = &sin; msg->msg_namelen = sizeof(sin); do_udp_sendmsg: - if (ipv6_only_sock(sk)) - return -ENETUNREACH; - return udp_sendmsg(sk, msg, len); + err = ipv6_only_sock(sk) ? + -ENETUNREACH : udp_sendmsg(sk, msg, len); + msg->msg_name = sin6; + msg->msg_namelen = addr_len; + return err; } } -- cgit v1.2.3 From 0646dc31ca886693274df5749cd0c8c1eaaeb5ca Mon Sep 17 00:00:00 2001 From: Liang Chen Date: Thu, 13 Apr 2023 17:03:53 +0800 Subject: skbuff: Fix a race between coalescing and releasing SKBs Commit 1effe8ca4e34 ("skbuff: fix coalescing for page_pool fragment recycling") allowed coalescing to proceed with non page pool page and page pool page when @from is cloned, i.e. to->pp_recycle --> false from->pp_recycle --> true skb_cloned(from) --> true However, it actually requires skb_cloned(@from) to hold true until coalescing finishes in this situation. If the other cloned SKB is released while the merging is in process, from_shinfo->nr_frags will be set to 0 toward the end of the function, causing the increment of frag page _refcount to be unexpectedly skipped resulting in inconsistent reference counts. Later when SKB(@to) is released, it frees the page directly even though the page pool page is still in use, leading to use-after-free or double-free errors. So it should be prohibited. The double-free error message below prompted us to investigate: BUG: Bad page state in process swapper/1 pfn:0e0d1 page:00000000c6548b28 refcount:-1 mapcount:0 mapping:0000000000000000 index:0x2 pfn:0xe0d1 flags: 0xfffffc0000000(node=0|zone=1|lastcpupid=0x1fffff) raw: 000fffffc0000000 0000000000000000 ffffffff00000101 0000000000000000 raw: 0000000000000002 0000000000000000 ffffffffffffffff 0000000000000000 page dumped because: nonzero _refcount CPU: 1 PID: 0 Comm: swapper/1 Tainted: G E 6.2.0+ Call Trace: dump_stack_lvl+0x32/0x50 bad_page+0x69/0xf0 free_pcp_prepare+0x260/0x2f0 free_unref_page+0x20/0x1c0 skb_release_data+0x10b/0x1a0 napi_consume_skb+0x56/0x150 net_rx_action+0xf0/0x350 ? __napi_schedule+0x79/0x90 __do_softirq+0xc8/0x2b1 __irq_exit_rcu+0xb9/0xf0 common_interrupt+0x82/0xa0 asm_common_interrupt+0x22/0x40 RIP: 0010:default_idle+0xb/0x20 Fixes: 53e0961da1c7 ("page_pool: add frag page recycling support in page pool") Signed-off-by: Liang Chen Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20230413090353.14448-1-liangchen.linux@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 1a31815104d6..4c0879798eb8 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -5599,18 +5599,18 @@ bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from, if (skb_cloned(to)) return false; - /* In general, avoid mixing slab allocated and page_pool allocated - * pages within the same SKB. However when @to is not pp_recycle and - * @from is cloned, we can transition frag pages from page_pool to - * reference counted. - * - * On the other hand, don't allow coalescing two pp_recycle SKBs if - * @from is cloned, in case the SKB is using page_pool fragment + /* In general, avoid mixing page_pool and non-page_pool allocated + * pages within the same SKB. Additionally avoid dealing with clones + * with page_pool pages, in case the SKB is using page_pool fragment * references (PP_FLAG_PAGE_FRAG). Since we only take full page * references for cloned SKBs at the moment that would result in * inconsistent reference counts. + * In theory we could take full references if @from is cloned and + * !@to->pp_recycle but its tricky (due to potential race with + * the clone disappearing) and rare, so not worth dealing with. */ - if (to->pp_recycle != (from->pp_recycle && !skb_cloned(from))) + if (to->pp_recycle != from->pp_recycle || + (from->pp_recycle && skb_cloned(from))) return false; if (len <= skb_tailroom(to)) { -- cgit v1.2.3 From 0cd917a4a8ace70ff9082d797c899f6bf10de910 Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Wed, 12 Apr 2023 21:48:40 +0200 Subject: xdp: rss hash types representation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The RSS hash type specifies what portion of packet data NIC hardware used when calculating RSS hash value. The RSS types are focused on Internet traffic protocols at OSI layers L3 and L4. L2 (e.g. ARP) often get hash value zero and no RSS type. For L3 focused on IPv4 vs. IPv6, and L4 primarily TCP vs UDP, but some hardware supports SCTP. Hardware RSS types are differently encoded for each hardware NIC. Most hardware represent RSS hash type as a number. Determining L3 vs L4 often requires a mapping table as there often isn't a pattern or sorting according to ISO layer. The patch introduce a XDP RSS hash type (enum xdp_rss_hash_type) that contains both BITs for the L3/L4 types, and combinations to be used by drivers for their mapping tables. The enum xdp_rss_type_bits get exposed to BPF via BTF, and it is up to the BPF-programmer to match using these defines. This proposal change the kfunc API bpf_xdp_metadata_rx_hash() adding a pointer value argument for provide the RSS hash type. Change signature for all xmo_rx_hash calls in drivers to make it compile. The RSS type implementations for each driver comes as separate patches. Fixes: 3d76a4d3d4e5 ("bpf: XDP metadata RX kfuncs") Signed-off-by: Jesper Dangaard Brouer Acked-by: Toke Høiland-Jørgensen Acked-by: Stanislav Fomichev Link: https://lore.kernel.org/r/168132892042.340624.582563003880565460.stgit@firesoul Signed-off-by: Alexei Starovoitov --- drivers/net/ethernet/mellanox/mlx4/en_rx.c | 3 +- drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 3 +- drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c | 3 +- drivers/net/veth.c | 3 +- include/linux/netdevice.h | 3 +- include/net/xdp.h | 45 ++++++++++++++++++++++++ net/core/xdp.c | 10 +++++- 7 files changed, 64 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 4b5e459b6d49..73d10aa4c503 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -681,7 +681,8 @@ int mlx4_en_xdp_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp) return 0; } -int mlx4_en_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash) +int mlx4_en_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash, + enum xdp_rss_hash_type *rss_type) { struct mlx4_en_xdp_buff *_ctx = (void *)ctx; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 544e09b97483..4ac4d883047b 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -798,7 +798,8 @@ int mlx4_en_netdev_event(struct notifier_block *this, struct xdp_md; int mlx4_en_xdp_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp); -int mlx4_en_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash); +int mlx4_en_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash, + enum xdp_rss_hash_type *rss_type); /* * Functions for time stamping diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index c5dae48b7932..efe609f8e3aa 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -169,7 +169,8 @@ static int mlx5e_xdp_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp) return 0; } -static int mlx5e_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash) +static int mlx5e_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash, + enum xdp_rss_hash_type *rss_type) { const struct mlx5e_xdp_buff *_ctx = (void *)ctx; diff --git a/drivers/net/veth.c b/drivers/net/veth.c index c1178915496d..424e8876a16b 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -1648,7 +1648,8 @@ static int veth_xdp_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp) return 0; } -static int veth_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash) +static int veth_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash, + enum xdp_rss_hash_type *rss_type) { struct veth_xdp_buff *_ctx = (void *)ctx; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 470085b121d3..c35f04f636f1 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1624,7 +1624,8 @@ struct net_device_ops { struct xdp_metadata_ops { int (*xmo_rx_timestamp)(const struct xdp_md *ctx, u64 *timestamp); - int (*xmo_rx_hash)(const struct xdp_md *ctx, u32 *hash); + int (*xmo_rx_hash)(const struct xdp_md *ctx, u32 *hash, + enum xdp_rss_hash_type *rss_type); }; /** diff --git a/include/net/xdp.h b/include/net/xdp.h index 41c57b8b1671..a76c4ea203ea 100644 --- a/include/net/xdp.h +++ b/include/net/xdp.h @@ -8,6 +8,7 @@ #include /* skb_shared_info */ #include +#include /** * DOC: XDP RX-queue information @@ -425,6 +426,50 @@ XDP_METADATA_KFUNC_xxx MAX_XDP_METADATA_KFUNC, }; +enum xdp_rss_hash_type { + /* First part: Individual bits for L3/L4 types */ + XDP_RSS_L3_IPV4 = BIT(0), + XDP_RSS_L3_IPV6 = BIT(1), + + /* The fixed (L3) IPv4 and IPv6 headers can both be followed by + * variable/dynamic headers, IPv4 called Options and IPv6 called + * Extension Headers. HW RSS type can contain this info. + */ + XDP_RSS_L3_DYNHDR = BIT(2), + + /* When RSS hash covers L4 then drivers MUST set XDP_RSS_L4 bit in + * addition to the protocol specific bit. This ease interaction with + * SKBs and avoids reserving a fixed mask for future L4 protocol bits. + */ + XDP_RSS_L4 = BIT(3), /* L4 based hash, proto can be unknown */ + XDP_RSS_L4_TCP = BIT(4), + XDP_RSS_L4_UDP = BIT(5), + XDP_RSS_L4_SCTP = BIT(6), + XDP_RSS_L4_IPSEC = BIT(7), /* L4 based hash include IPSEC SPI */ + + /* Second part: RSS hash type combinations used for driver HW mapping */ + XDP_RSS_TYPE_NONE = 0, + XDP_RSS_TYPE_L2 = XDP_RSS_TYPE_NONE, + + XDP_RSS_TYPE_L3_IPV4 = XDP_RSS_L3_IPV4, + XDP_RSS_TYPE_L3_IPV6 = XDP_RSS_L3_IPV6, + XDP_RSS_TYPE_L3_IPV4_OPT = XDP_RSS_L3_IPV4 | XDP_RSS_L3_DYNHDR, + XDP_RSS_TYPE_L3_IPV6_EX = XDP_RSS_L3_IPV6 | XDP_RSS_L3_DYNHDR, + + XDP_RSS_TYPE_L4_ANY = XDP_RSS_L4, + XDP_RSS_TYPE_L4_IPV4_TCP = XDP_RSS_L3_IPV4 | XDP_RSS_L4 | XDP_RSS_L4_TCP, + XDP_RSS_TYPE_L4_IPV4_UDP = XDP_RSS_L3_IPV4 | XDP_RSS_L4 | XDP_RSS_L4_UDP, + XDP_RSS_TYPE_L4_IPV4_SCTP = XDP_RSS_L3_IPV4 | XDP_RSS_L4 | XDP_RSS_L4_SCTP, + + XDP_RSS_TYPE_L4_IPV6_TCP = XDP_RSS_L3_IPV6 | XDP_RSS_L4 | XDP_RSS_L4_TCP, + XDP_RSS_TYPE_L4_IPV6_UDP = XDP_RSS_L3_IPV6 | XDP_RSS_L4 | XDP_RSS_L4_UDP, + XDP_RSS_TYPE_L4_IPV6_SCTP = XDP_RSS_L3_IPV6 | XDP_RSS_L4 | XDP_RSS_L4_SCTP, + + XDP_RSS_TYPE_L4_IPV6_TCP_EX = XDP_RSS_TYPE_L4_IPV6_TCP | XDP_RSS_L3_DYNHDR, + XDP_RSS_TYPE_L4_IPV6_UDP_EX = XDP_RSS_TYPE_L4_IPV6_UDP | XDP_RSS_L3_DYNHDR, + XDP_RSS_TYPE_L4_IPV6_SCTP_EX = XDP_RSS_TYPE_L4_IPV6_SCTP | XDP_RSS_L3_DYNHDR, +}; + #ifdef CONFIG_NET u32 bpf_xdp_metadata_kfunc_id(int id); bool bpf_dev_bound_kfunc_id(u32 btf_id); diff --git a/net/core/xdp.c b/net/core/xdp.c index 528d4b37983d..fb85aca81961 100644 --- a/net/core/xdp.c +++ b/net/core/xdp.c @@ -734,13 +734,21 @@ __bpf_kfunc int bpf_xdp_metadata_rx_timestamp(const struct xdp_md *ctx, u64 *tim * bpf_xdp_metadata_rx_hash - Read XDP frame RX hash. * @ctx: XDP context pointer. * @hash: Return value pointer. + * @rss_type: Return value pointer for RSS type. + * + * The RSS hash type (@rss_type) specifies what portion of packet headers NIC + * hardware used when calculating RSS hash value. The RSS type can be decoded + * via &enum xdp_rss_hash_type either matching on individual L3/L4 bits + * ``XDP_RSS_L*`` or by combined traditional *RSS Hashing Types* + * ``XDP_RSS_TYPE_L*``. * * Return: * * Returns 0 on success or ``-errno`` on error. * * ``-EOPNOTSUPP`` : means device driver doesn't implement kfunc * * ``-ENODATA`` : means no RX-hash available for this frame */ -__bpf_kfunc int bpf_xdp_metadata_rx_hash(const struct xdp_md *ctx, u32 *hash) +__bpf_kfunc int bpf_xdp_metadata_rx_hash(const struct xdp_md *ctx, u32 *hash, + enum xdp_rss_hash_type *rss_type) { return -EOPNOTSUPP; } -- cgit v1.2.3 From 8c5c2a4898e3d6bad86e29d471e023c8a19ba799 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 13 Apr 2023 20:28:42 +0200 Subject: bpf, sockmap: Revert buggy deadlock fix in the sockhash and sockmap syzbot reported a splat and bisected it to recent commit ed17aa92dc56 ("bpf, sockmap: fix deadlocks in the sockhash and sockmap"): [...] WARNING: CPU: 1 PID: 9280 at kernel/softirq.c:376 __local_bh_enable_ip+0xbe/0x130 kernel/softirq.c:376 Modules linked in: CPU: 1 PID: 9280 Comm: syz-executor.1 Not tainted 6.2.0-syzkaller-13249-gd319f344561d #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/30/2023 RIP: 0010:__local_bh_enable_ip+0xbe/0x130 kernel/softirq.c:376 [...] Call Trace: spin_unlock_bh include/linux/spinlock.h:395 [inline] sock_map_del_link+0x2ea/0x510 net/core/sock_map.c:165 sock_map_unref+0xb0/0x1d0 net/core/sock_map.c:184 sock_hash_delete_elem+0x1ec/0x2a0 net/core/sock_map.c:945 map_delete_elem kernel/bpf/syscall.c:1536 [inline] __sys_bpf+0x2edc/0x53e0 kernel/bpf/syscall.c:5053 __do_sys_bpf kernel/bpf/syscall.c:5166 [inline] __se_sys_bpf kernel/bpf/syscall.c:5164 [inline] __x64_sys_bpf+0x79/0xc0 kernel/bpf/syscall.c:5164 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0033:0x7fe8f7c8c169 [...] Revert for now until we have a proper solution. Fixes: ed17aa92dc56 ("bpf, sockmap: fix deadlocks in the sockhash and sockmap") Reported-by: syzbot+49f6cef45247ff249498@syzkaller.appspotmail.com Cc: Hsin-Wei Hung Cc: Xin Liu Cc: John Fastabend Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/000000000000f1db9605f939720e@google.com/ --- net/core/sock_map.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/core/sock_map.c b/net/core/sock_map.c index 66305b7bf8b7..7c189c2e2fbf 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -414,9 +414,8 @@ static int __sock_map_delete(struct bpf_stab *stab, struct sock *sk_test, { struct sock *sk; int err = 0; - unsigned long flags; - raw_spin_lock_irqsave(&stab->lock, flags); + raw_spin_lock_bh(&stab->lock); sk = *psk; if (!sk_test || sk_test == sk) sk = xchg(psk, NULL); @@ -426,7 +425,7 @@ static int __sock_map_delete(struct bpf_stab *stab, struct sock *sk_test, else err = -EINVAL; - raw_spin_unlock_irqrestore(&stab->lock, flags); + raw_spin_unlock_bh(&stab->lock); return err; } @@ -933,12 +932,11 @@ static long sock_hash_delete_elem(struct bpf_map *map, void *key) struct bpf_shtab_bucket *bucket; struct bpf_shtab_elem *elem; int ret = -ENOENT; - unsigned long flags; hash = sock_hash_bucket_hash(key, key_size); bucket = sock_hash_select_bucket(htab, hash); - raw_spin_lock_irqsave(&bucket->lock, flags); + raw_spin_lock_bh(&bucket->lock); elem = sock_hash_lookup_elem_raw(&bucket->head, hash, key, key_size); if (elem) { hlist_del_rcu(&elem->node); @@ -946,7 +944,7 @@ static long sock_hash_delete_elem(struct bpf_map *map, void *key) sock_hash_free_elem(htab, elem); ret = 0; } - raw_spin_unlock_irqrestore(&bucket->lock, flags); + raw_spin_unlock_bh(&bucket->lock); return ret; } -- cgit v1.2.3 From d54151aa0f4b5c89561705a00d8a5ebb4230028c Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 11 Apr 2023 21:01:49 +0300 Subject: net: ethtool: create and export ethtool_dev_mm_supported() Create a wrapper over __ethtool_dev_mm_supported() which also calls ethnl_ops_begin() and ethnl_ops_complete(). It can be used by other code layers, such as tc, to make sure that preemptible TCs are supported (this is true if an underlying MAC Merge layer exists). Signed-off-by: Vladimir Oltean Reviewed-by: Ferenc Fejes Reviewed-by: Simon Horman Signed-off-by: Jakub Kicinski --- include/linux/ethtool_netlink.h | 6 ++++++ net/ethtool/mm.c | 23 +++++++++++++++++++++++ 2 files changed, 29 insertions(+) (limited to 'net') diff --git a/include/linux/ethtool_netlink.h b/include/linux/ethtool_netlink.h index 17003b385756..fae0dfb9a9c8 100644 --- a/include/linux/ethtool_netlink.h +++ b/include/linux/ethtool_netlink.h @@ -39,6 +39,7 @@ void ethtool_aggregate_pause_stats(struct net_device *dev, struct ethtool_pause_stats *pause_stats); void ethtool_aggregate_rmon_stats(struct net_device *dev, struct ethtool_rmon_stats *rmon_stats); +bool ethtool_dev_mm_supported(struct net_device *dev); #else static inline int ethnl_cable_test_alloc(struct phy_device *phydev, u8 cmd) @@ -112,5 +113,10 @@ ethtool_aggregate_rmon_stats(struct net_device *dev, { } +static inline bool ethtool_dev_mm_supported(struct net_device *dev) +{ + return false; +} + #endif /* IS_ENABLED(CONFIG_ETHTOOL_NETLINK) */ #endif /* _LINUX_ETHTOOL_NETLINK_H_ */ diff --git a/net/ethtool/mm.c b/net/ethtool/mm.c index fce3cc2734f9..e00d7d5cea7e 100644 --- a/net/ethtool/mm.c +++ b/net/ethtool/mm.c @@ -249,3 +249,26 @@ bool __ethtool_dev_mm_supported(struct net_device *dev) return !ret; } + +bool ethtool_dev_mm_supported(struct net_device *dev) +{ + const struct ethtool_ops *ops = dev->ethtool_ops; + bool supported; + int ret; + + ASSERT_RTNL(); + + if (!ops) + return false; + + ret = ethnl_ops_begin(dev); + if (ret < 0) + return false; + + supported = __ethtool_dev_mm_supported(dev); + + ethnl_ops_complete(dev); + + return supported; +} +EXPORT_SYMBOL_GPL(ethtool_dev_mm_supported); -- cgit v1.2.3 From 3dd0c16ec93e2bf5212758baea34cfed90d9e6bd Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 11 Apr 2023 21:01:50 +0300 Subject: net/sched: mqprio: simplify handling of nlattr portion of TCA_OPTIONS In commit 4e8b86c06269 ("mqprio: Introduce new hardware offload mode and shaper in mqprio"), the TCA_OPTIONS format of mqprio was extended to contain a fixed portion (of size NLA_ALIGN(sizeof struct tc_mqprio_qopt)) and a variable portion of other nlattrs (in the TCA_MQPRIO_* type space) following immediately afterwards. In commit feb2cf3dcfb9 ("net/sched: mqprio: refactor nlattr parsing to a separate function"), we've moved the nlattr handling to a smaller function, but yet, a small parse_attr() still remains, and the larger mqprio_parse_nlattr() still does not have access to the beginning, and the length, of the TCA_OPTIONS region containing these other nlattrs. In a future change, the mqprio qdisc will need to iterate through this nlattr region to discover other attributes, so eliminate parse_attr() and add 2 variables in mqprio_parse_nlattr() which hold the beginning and the length of the nlattr range. We avoid the need to memset when nlattr_opt_len has insufficient length by pre-initializing the table "tb". Signed-off-by: Vladimir Oltean Reviewed-by: Ferenc Fejes Reviewed-by: Simon Horman Acked-by: Jamal Hadi Salim Signed-off-by: Jakub Kicinski --- net/sched/sch_mqprio.c | 32 +++++++++++++------------------- 1 file changed, 13 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index fdd6a6575a54..89e5dfb22db6 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -146,32 +146,26 @@ static const struct nla_policy mqprio_policy[TCA_MQPRIO_MAX + 1] = { [TCA_MQPRIO_MAX_RATE64] = { .type = NLA_NESTED }, }; -static int parse_attr(struct nlattr *tb[], int maxtype, struct nlattr *nla, - const struct nla_policy *policy, int len) -{ - int nested_len = nla_len(nla) - NLA_ALIGN(len); - - if (nested_len >= nla_attr_size(0)) - return nla_parse_deprecated(tb, maxtype, - nla_data(nla) + NLA_ALIGN(len), - nested_len, policy, NULL); - - memset(tb, 0, sizeof(struct nlattr *) * (maxtype + 1)); - return 0; -} - +/* Parse the other netlink attributes that represent the payload of + * TCA_OPTIONS, which are appended right after struct tc_mqprio_qopt. + */ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt, struct nlattr *opt) { + struct nlattr *nlattr_opt = nla_data(opt) + NLA_ALIGN(sizeof(*qopt)); + int nlattr_opt_len = nla_len(opt) - NLA_ALIGN(sizeof(*qopt)); struct mqprio_sched *priv = qdisc_priv(sch); - struct nlattr *tb[TCA_MQPRIO_MAX + 1]; + struct nlattr *tb[TCA_MQPRIO_MAX + 1] = {}; struct nlattr *attr; int i, rem, err; - err = parse_attr(tb, TCA_MQPRIO_MAX, opt, mqprio_policy, - sizeof(*qopt)); - if (err < 0) - return err; + if (nlattr_opt_len >= nla_attr_size(0)) { + err = nla_parse_deprecated(tb, TCA_MQPRIO_MAX, nlattr_opt, + nlattr_opt_len, mqprio_policy, + NULL); + if (err < 0) + return err; + } if (!qopt->hw) return -EINVAL; -- cgit v1.2.3 From 57f21bf85400abadac0cb2a4db5de1d663f8863f Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 11 Apr 2023 21:01:51 +0300 Subject: net/sched: mqprio: add extack to mqprio_parse_nlattr() Netlink attribute parsing in mqprio is a minesweeper game, with many options having the possibility of being passed incorrectly and the user being none the wiser. Try to make errors less sour by giving user space some information regarding what went wrong. Signed-off-by: Vladimir Oltean Reviewed-by: Ferenc Fejes Reviewed-by: Simon Horman Acked-by: Jamal Hadi Salim Signed-off-by: Jakub Kicinski --- net/sched/sch_mqprio.c | 30 +++++++++++++++++++++++------- 1 file changed, 23 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index 89e5dfb22db6..8e8151ca8307 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -150,7 +150,8 @@ static const struct nla_policy mqprio_policy[TCA_MQPRIO_MAX + 1] = { * TCA_OPTIONS, which are appended right after struct tc_mqprio_qopt. */ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt, - struct nlattr *opt) + struct nlattr *opt, + struct netlink_ext_ack *extack) { struct nlattr *nlattr_opt = nla_data(opt) + NLA_ALIGN(sizeof(*qopt)); int nlattr_opt_len = nla_len(opt) - NLA_ALIGN(sizeof(*qopt)); @@ -167,8 +168,11 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt, return err; } - if (!qopt->hw) + if (!qopt->hw) { + NL_SET_ERR_MSG(extack, + "mqprio TCA_OPTIONS can only contain netlink attributes in hardware mode"); return -EINVAL; + } if (tb[TCA_MQPRIO_MODE]) { priv->flags |= TC_MQPRIO_F_MODE; @@ -181,13 +185,19 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt, } if (tb[TCA_MQPRIO_MIN_RATE64]) { - if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) + if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) { + NL_SET_ERR_MSG_ATTR(extack, tb[TCA_MQPRIO_MIN_RATE64], + "min_rate accepted only when shaper is in bw_rlimit mode"); return -EINVAL; + } i = 0; nla_for_each_nested(attr, tb[TCA_MQPRIO_MIN_RATE64], rem) { - if (nla_type(attr) != TCA_MQPRIO_MIN_RATE64) + if (nla_type(attr) != TCA_MQPRIO_MIN_RATE64) { + NL_SET_ERR_MSG_ATTR(extack, attr, + "Attribute type expected to be TCA_MQPRIO_MIN_RATE64"); return -EINVAL; + } if (i >= qopt->num_tc) break; priv->min_rate[i] = nla_get_u64(attr); @@ -197,13 +207,19 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt, } if (tb[TCA_MQPRIO_MAX_RATE64]) { - if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) + if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) { + NL_SET_ERR_MSG_ATTR(extack, tb[TCA_MQPRIO_MAX_RATE64], + "max_rate accepted only when shaper is in bw_rlimit mode"); return -EINVAL; + } i = 0; nla_for_each_nested(attr, tb[TCA_MQPRIO_MAX_RATE64], rem) { - if (nla_type(attr) != TCA_MQPRIO_MAX_RATE64) + if (nla_type(attr) != TCA_MQPRIO_MAX_RATE64) { + NL_SET_ERR_MSG_ATTR(extack, attr, + "Attribute type expected to be TCA_MQPRIO_MAX_RATE64"); return -EINVAL; + } if (i >= qopt->num_tc) break; priv->max_rate[i] = nla_get_u64(attr); @@ -252,7 +268,7 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt, len = nla_len(opt) - NLA_ALIGN(sizeof(*qopt)); if (len > 0) { - err = mqprio_parse_nlattr(sch, qopt, opt); + err = mqprio_parse_nlattr(sch, qopt, opt, extack); if (err) return err; } -- cgit v1.2.3 From ab277d2084ba59956b177a7d417791a27361e283 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 11 Apr 2023 21:01:52 +0300 Subject: net/sched: mqprio: add an extack message to mqprio_parse_opt() Ferenc reports that a combination of poor iproute2 defaults and obscure cases where the kernel returns -EINVAL make it difficult to understand what is wrong with this command: $ ip link add veth0 numtxqueues 8 numrxqueues 8 type veth peer name veth1 $ tc qdisc add dev veth0 root mqprio num_tc 8 map 0 1 2 3 4 5 6 7 \ queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 RTNETLINK answers: Invalid argument Hopefully with this patch, the cause is clearer: Error: Device does not support hardware offload. The kernel was (and still is) rejecting this because iproute2 defaults to "hw 1" if this command line option is not specified. Link: https://lore.kernel.org/netdev/ede5e9a2f27bf83bfb86d3e8c4ca7b34093b99e2.camel@inf.elte.hu/ Signed-off-by: Vladimir Oltean Reviewed-by: Ferenc Fejes Reviewed-by: Simon Horman Acked-by: Jamal Hadi Salim Signed-off-by: Jakub Kicinski --- net/sched/sch_mqprio.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index 8e8151ca8307..aae4d64dbf3f 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -133,8 +133,11 @@ static int mqprio_parse_opt(struct net_device *dev, struct tc_mqprio_qopt *qopt, /* If ndo_setup_tc is not present then hardware doesn't support offload * and we should return an error. */ - if (qopt->hw && !dev->netdev_ops->ndo_setup_tc) + if (qopt->hw && !dev->netdev_ops->ndo_setup_tc) { + NL_SET_ERR_MSG(extack, + "Device does not support hardware offload"); return -EINVAL; + } return 0; } -- cgit v1.2.3 From c54876cd5961ce0f8e74807f79a6739cd6b35ddf Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 11 Apr 2023 21:01:53 +0300 Subject: net/sched: pass netlink extack to mqprio and taprio offload With the multiplexed ndo_setup_tc() model which lacks a first-class struct netlink_ext_ack * argument, the only way to pass the netlink extended ACK message down to the device driver is to embed it within the offload structure. Do this for struct tc_mqprio_qopt_offload and struct tc_taprio_qopt_offload. Since struct tc_taprio_qopt_offload also contains a tc_mqprio_qopt_offload structure, and since device drivers might effectively reuse their mqprio implementation for the mqprio portion of taprio, we make taprio set the extack in both offload structures to point at the same netlink extack message. In fact, the taprio handling is a bit more tricky, for 2 reasons. First is because the offload structure has a longer lifetime than the extack structure. The driver is supposed to populate the extack synchronously from ndo_setup_tc() and leave it alone afterwards. To not have any use-after-free surprises, we zero out the extack pointer when we leave taprio_enable_offload(). The second reason is because taprio does overwrite the extack message on ndo_setup_tc() error. We need to switch to the weak form of setting an extack message, which preserves a potential message set by the driver. Signed-off-by: Vladimir Oltean Reviewed-by: Simon Horman Acked-by: Jamal Hadi Salim Signed-off-by: Jakub Kicinski --- include/net/pkt_sched.h | 2 ++ net/sched/sch_mqprio.c | 5 ++++- net/sched/sch_taprio.c | 12 ++++++++++-- 3 files changed, 16 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index bb0bd69fb655..b43ed4733455 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -166,6 +166,7 @@ struct tc_mqprio_caps { struct tc_mqprio_qopt_offload { /* struct tc_mqprio_qopt must always be the first element */ struct tc_mqprio_qopt qopt; + struct netlink_ext_ack *extack; u16 mode; u16 shaper; u32 flags; @@ -193,6 +194,7 @@ struct tc_taprio_sched_entry { struct tc_taprio_qopt_offload { struct tc_mqprio_qopt_offload mqprio; + struct netlink_ext_ack *extack; u8 enable; ktime_t base_time; u64 cycle_time; diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index aae4d64dbf3f..67d77495c8fd 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -33,9 +33,12 @@ static int mqprio_enable_offload(struct Qdisc *sch, const struct tc_mqprio_qopt *qopt, struct netlink_ext_ack *extack) { - struct tc_mqprio_qopt_offload mqprio = {.qopt = *qopt}; struct mqprio_sched *priv = qdisc_priv(sch); struct net_device *dev = qdisc_dev(sch); + struct tc_mqprio_qopt_offload mqprio = { + .qopt = *qopt, + .extack = extack, + }; int err, i; switch (priv->mode) { diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c index 1f469861eae3..cbad43019172 100644 --- a/net/sched/sch_taprio.c +++ b/net/sched/sch_taprio.c @@ -1520,7 +1520,9 @@ static int taprio_enable_offload(struct net_device *dev, return -ENOMEM; } offload->enable = 1; + offload->extack = extack; mqprio_qopt_reconstruct(dev, &offload->mqprio.qopt); + offload->mqprio.extack = extack; taprio_sched_to_offload(dev, sched, offload, &caps); for (tc = 0; tc < TC_MAX_QUEUE; tc++) @@ -1528,14 +1530,20 @@ static int taprio_enable_offload(struct net_device *dev, err = ops->ndo_setup_tc(dev, TC_SETUP_QDISC_TAPRIO, offload); if (err < 0) { - NL_SET_ERR_MSG(extack, - "Device failed to setup taprio offload"); + NL_SET_ERR_MSG_WEAK(extack, + "Device failed to setup taprio offload"); goto done; } q->offloaded = true; done: + /* The offload structure may linger around via a reference taken by the + * device driver, so clear up the netlink extack pointer so that the + * driver isn't tempted to dereference data which stopped being valid + */ + offload->extack = NULL; + offload->mqprio.extack = NULL; taprio_offload_free(offload); return err; -- cgit v1.2.3 From f62af20bed2d9e824f51cfc97ff01bc261f40e58 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 11 Apr 2023 21:01:54 +0300 Subject: net/sched: mqprio: allow per-TC user input of FP adminStatus IEEE 802.1Q-2018 clause 6.7.2 Frame preemption specifies that each packet priority can be assigned to a "frame preemption status" value of either "express" or "preemptible". Express priorities are transmitted by the local device through the eMAC, and preemptible priorities through the pMAC (the concepts of eMAC and pMAC come from the 802.3 MAC Merge layer). The FP adminStatus is defined per packet priority, but 802.1Q clause 12.30.1.1.1 framePreemptionAdminStatus also says that: | Priorities that all map to the same traffic class should be | constrained to use the same value of preemption status. It is impossible to ignore the cognitive dissonance in the standard here, because it practically means that the FP adminStatus only takes distinct values per traffic class, even though it is defined per priority. I can see no valid use case which is prevented by having the kernel take the FP adminStatus as input per traffic class (what we do here). In addition, this also enforces the above constraint by construction. User space network managers which wish to expose FP adminStatus per priority are free to do so; they must only observe the prio_tc_map of the netdev (which presumably is also under their control, when constructing the mqprio netlink attributes). The reason for configuring frame preemption as a property of the Qdisc layer is that the information about "preemptible TCs" is closest to the place which handles the num_tc and prio_tc_map of the netdev. If the UAPI would have been any other layer, it would be unclear what to do with the FP information when num_tc collapses to 0. A key assumption is that only mqprio/taprio change the num_tc and prio_tc_map of the netdev. Not sure if that's a great assumption to make. Having FP in tc-mqprio can be seen as an implementation of the use case defined in 802.1Q Annex S.2 "Preemption used in isolation". There will be a separate implementation of FP in tc-taprio, for the other use cases. Signed-off-by: Vladimir Oltean Reviewed-by: Ferenc Fejes Reviewed-by: Simon Horman Signed-off-by: Jakub Kicinski --- include/net/pkt_sched.h | 1 + include/uapi/linux/pkt_sched.h | 16 ++++++ net/sched/sch_mqprio.c | 128 ++++++++++++++++++++++++++++++++++++++++- net/sched/sch_mqprio_lib.c | 14 +++++ net/sched/sch_mqprio_lib.h | 2 + 5 files changed, 160 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index b43ed4733455..f436688b6efc 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -172,6 +172,7 @@ struct tc_mqprio_qopt_offload { u32 flags; u64 min_rate[TC_QOPT_MAX_QUEUE]; u64 max_rate[TC_QOPT_MAX_QUEUE]; + unsigned long preemptible_tcs; }; struct tc_taprio_caps { diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h index 000eec106856..b8d29be91b62 100644 --- a/include/uapi/linux/pkt_sched.h +++ b/include/uapi/linux/pkt_sched.h @@ -719,6 +719,11 @@ enum { #define __TC_MQPRIO_SHAPER_MAX (__TC_MQPRIO_SHAPER_MAX - 1) +enum { + TC_FP_EXPRESS = 1, + TC_FP_PREEMPTIBLE = 2, +}; + struct tc_mqprio_qopt { __u8 num_tc; __u8 prio_tc_map[TC_QOPT_BITMASK + 1]; @@ -732,12 +737,23 @@ struct tc_mqprio_qopt { #define TC_MQPRIO_F_MIN_RATE 0x4 #define TC_MQPRIO_F_MAX_RATE 0x8 +enum { + TCA_MQPRIO_TC_ENTRY_UNSPEC, + TCA_MQPRIO_TC_ENTRY_INDEX, /* u32 */ + TCA_MQPRIO_TC_ENTRY_FP, /* u32 */ + + /* add new constants above here */ + __TCA_MQPRIO_TC_ENTRY_CNT, + TCA_MQPRIO_TC_ENTRY_MAX = (__TCA_MQPRIO_TC_ENTRY_CNT - 1) +}; + enum { TCA_MQPRIO_UNSPEC, TCA_MQPRIO_MODE, TCA_MQPRIO_SHAPER, TCA_MQPRIO_MIN_RATE64, TCA_MQPRIO_MAX_RATE64, + TCA_MQPRIO_TC_ENTRY, __TCA_MQPRIO_MAX, }; diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index 67d77495c8fd..dc5a0ff50b14 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -5,6 +5,7 @@ * Copyright (c) 2010 John Fastabend */ +#include #include #include #include @@ -27,6 +28,7 @@ struct mqprio_sched { u32 flags; u64 min_rate[TC_QOPT_MAX_QUEUE]; u64 max_rate[TC_QOPT_MAX_QUEUE]; + u32 fp[TC_QOPT_MAX_QUEUE]; }; static int mqprio_enable_offload(struct Qdisc *sch, @@ -63,6 +65,8 @@ static int mqprio_enable_offload(struct Qdisc *sch, return -EINVAL; } + mqprio_fp_to_offload(priv->fp, &mqprio); + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_MQPRIO, &mqprio); if (err) @@ -145,13 +149,95 @@ static int mqprio_parse_opt(struct net_device *dev, struct tc_mqprio_qopt *qopt, return 0; } +static const struct +nla_policy mqprio_tc_entry_policy[TCA_MQPRIO_TC_ENTRY_MAX + 1] = { + [TCA_MQPRIO_TC_ENTRY_INDEX] = NLA_POLICY_MAX(NLA_U32, + TC_QOPT_MAX_QUEUE), + [TCA_MQPRIO_TC_ENTRY_FP] = NLA_POLICY_RANGE(NLA_U32, + TC_FP_EXPRESS, + TC_FP_PREEMPTIBLE), +}; + static const struct nla_policy mqprio_policy[TCA_MQPRIO_MAX + 1] = { [TCA_MQPRIO_MODE] = { .len = sizeof(u16) }, [TCA_MQPRIO_SHAPER] = { .len = sizeof(u16) }, [TCA_MQPRIO_MIN_RATE64] = { .type = NLA_NESTED }, [TCA_MQPRIO_MAX_RATE64] = { .type = NLA_NESTED }, + [TCA_MQPRIO_TC_ENTRY] = { .type = NLA_NESTED }, }; +static int mqprio_parse_tc_entry(u32 fp[TC_QOPT_MAX_QUEUE], + struct nlattr *opt, + unsigned long *seen_tcs, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[TCA_MQPRIO_TC_ENTRY_MAX + 1]; + int err, tc; + + err = nla_parse_nested(tb, TCA_MQPRIO_TC_ENTRY_MAX, opt, + mqprio_tc_entry_policy, extack); + if (err < 0) + return err; + + if (NL_REQ_ATTR_CHECK(extack, opt, tb, TCA_MQPRIO_TC_ENTRY_INDEX)) { + NL_SET_ERR_MSG(extack, "TC entry index missing"); + return -EINVAL; + } + + tc = nla_get_u32(tb[TCA_MQPRIO_TC_ENTRY_INDEX]); + if (*seen_tcs & BIT(tc)) { + NL_SET_ERR_MSG_ATTR(extack, tb[TCA_MQPRIO_TC_ENTRY_INDEX], + "Duplicate tc entry"); + return -EINVAL; + } + + *seen_tcs |= BIT(tc); + + if (tb[TCA_MQPRIO_TC_ENTRY_FP]) + fp[tc] = nla_get_u32(tb[TCA_MQPRIO_TC_ENTRY_FP]); + + return 0; +} + +static int mqprio_parse_tc_entries(struct Qdisc *sch, struct nlattr *nlattr_opt, + int nlattr_opt_len, + struct netlink_ext_ack *extack) +{ + struct mqprio_sched *priv = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + bool have_preemption = false; + unsigned long seen_tcs = 0; + u32 fp[TC_QOPT_MAX_QUEUE]; + struct nlattr *n; + int tc, rem; + int err = 0; + + for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) + fp[tc] = priv->fp[tc]; + + nla_for_each_attr(n, nlattr_opt, nlattr_opt_len, rem) { + if (nla_type(n) != TCA_MQPRIO_TC_ENTRY) + continue; + + err = mqprio_parse_tc_entry(fp, n, &seen_tcs, extack); + if (err) + goto out; + } + + for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) { + priv->fp[tc] = fp[tc]; + if (fp[tc] == TC_FP_PREEMPTIBLE) + have_preemption = true; + } + + if (have_preemption && !ethtool_dev_mm_supported(dev)) { + NL_SET_ERR_MSG(extack, "Device does not support preemption"); + return -EOPNOTSUPP; + } +out: + return err; +} + /* Parse the other netlink attributes that represent the payload of * TCA_OPTIONS, which are appended right after struct tc_mqprio_qopt. */ @@ -234,6 +320,13 @@ static int mqprio_parse_nlattr(struct Qdisc *sch, struct tc_mqprio_qopt *qopt, priv->flags |= TC_MQPRIO_F_MAX_RATE; } + if (tb[TCA_MQPRIO_TC_ENTRY]) { + err = mqprio_parse_tc_entries(sch, nlattr_opt, nlattr_opt_len, + extack); + if (err) + return err; + } + return 0; } @@ -247,7 +340,7 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt, int i, err = -EOPNOTSUPP; struct tc_mqprio_qopt *qopt = NULL; struct tc_mqprio_caps caps; - int len; + int len, tc; BUILD_BUG_ON(TC_MAX_QUEUE != TC_QOPT_MAX_QUEUE); BUILD_BUG_ON(TC_BITMASK != TC_QOPT_BITMASK); @@ -265,6 +358,9 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt, if (!opt || nla_len(opt) < sizeof(*qopt)) return -EINVAL; + for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) + priv->fp[tc] = TC_FP_EXPRESS; + qdisc_offload_query_caps(dev, TC_SETUP_QDISC_MQPRIO, &caps, sizeof(caps)); @@ -415,6 +511,33 @@ nla_put_failure: return -1; } +static int mqprio_dump_tc_entries(struct mqprio_sched *priv, + struct sk_buff *skb) +{ + struct nlattr *n; + int tc; + + for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) { + n = nla_nest_start(skb, TCA_MQPRIO_TC_ENTRY); + if (!n) + return -EMSGSIZE; + + if (nla_put_u32(skb, TCA_MQPRIO_TC_ENTRY_INDEX, tc)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_MQPRIO_TC_ENTRY_FP, priv->fp[tc])) + goto nla_put_failure; + + nla_nest_end(skb, n); + } + + return 0; + +nla_put_failure: + nla_nest_cancel(skb, n); + return -EMSGSIZE; +} + static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb) { struct net_device *dev = qdisc_dev(sch); @@ -465,6 +588,9 @@ static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb) (dump_rates(priv, &opt, skb) != 0)) goto nla_put_failure; + if (mqprio_dump_tc_entries(priv, skb)) + goto nla_put_failure; + return nla_nest_end(skb, nla); nla_put_failure: nlmsg_trim(skb, nla); diff --git a/net/sched/sch_mqprio_lib.c b/net/sched/sch_mqprio_lib.c index c58a533b8ec5..83b3793c4012 100644 --- a/net/sched/sch_mqprio_lib.c +++ b/net/sched/sch_mqprio_lib.c @@ -114,4 +114,18 @@ void mqprio_qopt_reconstruct(struct net_device *dev, struct tc_mqprio_qopt *qopt } EXPORT_SYMBOL_GPL(mqprio_qopt_reconstruct); +void mqprio_fp_to_offload(u32 fp[TC_QOPT_MAX_QUEUE], + struct tc_mqprio_qopt_offload *mqprio) +{ + unsigned long preemptible_tcs = 0; + int tc; + + for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) + if (fp[tc] == TC_FP_PREEMPTIBLE) + preemptible_tcs |= BIT(tc); + + mqprio->preemptible_tcs = preemptible_tcs; +} +EXPORT_SYMBOL_GPL(mqprio_fp_to_offload); + MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_mqprio_lib.h b/net/sched/sch_mqprio_lib.h index 63f725ab8761..079f597072e3 100644 --- a/net/sched/sch_mqprio_lib.h +++ b/net/sched/sch_mqprio_lib.h @@ -14,5 +14,7 @@ int mqprio_validate_qopt(struct net_device *dev, struct tc_mqprio_qopt *qopt, struct netlink_ext_ack *extack); void mqprio_qopt_reconstruct(struct net_device *dev, struct tc_mqprio_qopt *qopt); +void mqprio_fp_to_offload(u32 fp[TC_QOPT_MAX_QUEUE], + struct tc_mqprio_qopt_offload *mqprio); #endif -- cgit v1.2.3 From a721c3e54b80e45cd9202e7fca29ef018bed9069 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 11 Apr 2023 21:01:55 +0300 Subject: net/sched: taprio: allow per-TC user input of FP adminStatus This is a duplication of the FP adminStatus logic introduced for tc-mqprio. Offloading is done through the tc_mqprio_qopt_offload structure embedded within tc_taprio_qopt_offload. So practically, if a device driver is written to treat the mqprio portion of taprio just like standalone mqprio, it gets unified handling of frame preemption. I would have reused more code with taprio, but this is mostly netlink attribute parsing, which is hard to transform into generic code without having something that stinks as a result. We have the same variables with the same semantics, just different nlattr type values (TCA_MQPRIO_TC_ENTRY=5 vs TCA_TAPRIO_ATTR_TC_ENTRY=12; TCA_MQPRIO_TC_ENTRY_FP=2 vs TCA_TAPRIO_TC_ENTRY_FP=3, etc) and consequently, different policies for the nest. Every time nla_parse_nested() is called, an on-stack table "tb" of nlattr pointers is allocated statically, up to the maximum understood nlattr type. That array size is hardcoded as a constant, but when transforming this into a common parsing function, it would become either a VLA (which the Linux kernel rightfully doesn't like) or a call to the allocator. Having FP adminStatus in tc-taprio can be seen as addressing the 802.1Q Annex S.3 "Scheduling and preemption used in combination, no HOLD/RELEASE" and S.4 "Scheduling and preemption used in combination with HOLD/RELEASE" use cases. HOLD and RELEASE events are emitted towards the underlying MAC Merge layer when the schedule hits a Set-And-Hold-MAC or a Set-And-Release-MAC gate operation. So within the tc-taprio UAPI space, one can distinguish between the 2 use cases by choosing whether to use the TC_TAPRIO_CMD_SET_AND_HOLD and TC_TAPRIO_CMD_SET_AND_RELEASE gate operations within the schedule, or just TC_TAPRIO_CMD_SET_GATES. A small part of the change is dedicated to refactoring the max_sdu nlattr parsing to put all logic under the "if" that tests for presence of that nlattr. Signed-off-by: Vladimir Oltean Reviewed-by: Ferenc Fejes Reviewed-by: Simon Horman Signed-off-by: Jakub Kicinski --- include/uapi/linux/pkt_sched.h | 1 + net/sched/sch_taprio.c | 65 +++++++++++++++++++++++++++++++++--------- 2 files changed, 53 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h index b8d29be91b62..51a7addc56c6 100644 --- a/include/uapi/linux/pkt_sched.h +++ b/include/uapi/linux/pkt_sched.h @@ -1252,6 +1252,7 @@ enum { TCA_TAPRIO_TC_ENTRY_UNSPEC, TCA_TAPRIO_TC_ENTRY_INDEX, /* u32 */ TCA_TAPRIO_TC_ENTRY_MAX_SDU, /* u32 */ + TCA_TAPRIO_TC_ENTRY_FP, /* u32 */ /* add new constants above here */ __TCA_TAPRIO_TC_ENTRY_CNT, diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c index cbad43019172..76db9a10ef50 100644 --- a/net/sched/sch_taprio.c +++ b/net/sched/sch_taprio.c @@ -7,6 +7,7 @@ */ #include +#include #include #include #include @@ -96,6 +97,7 @@ struct taprio_sched { struct list_head taprio_list; int cur_txq[TC_MAX_QUEUE]; u32 max_sdu[TC_MAX_QUEUE]; /* save info from the user */ + u32 fp[TC_QOPT_MAX_QUEUE]; /* only for dump and offloading */ u32 txtime_delay; }; @@ -1002,6 +1004,9 @@ static const struct nla_policy entry_policy[TCA_TAPRIO_SCHED_ENTRY_MAX + 1] = { static const struct nla_policy taprio_tc_policy[TCA_TAPRIO_TC_ENTRY_MAX + 1] = { [TCA_TAPRIO_TC_ENTRY_INDEX] = { .type = NLA_U32 }, [TCA_TAPRIO_TC_ENTRY_MAX_SDU] = { .type = NLA_U32 }, + [TCA_TAPRIO_TC_ENTRY_FP] = NLA_POLICY_RANGE(NLA_U32, + TC_FP_EXPRESS, + TC_FP_PREEMPTIBLE), }; static const struct nla_policy taprio_policy[TCA_TAPRIO_ATTR_MAX + 1] = { @@ -1524,6 +1529,7 @@ static int taprio_enable_offload(struct net_device *dev, mqprio_qopt_reconstruct(dev, &offload->mqprio.qopt); offload->mqprio.extack = extack; taprio_sched_to_offload(dev, sched, offload, &caps); + mqprio_fp_to_offload(q->fp, &offload->mqprio); for (tc = 0; tc < TC_MAX_QUEUE; tc++) offload->max_sdu[tc] = q->max_sdu[tc]; @@ -1671,13 +1677,14 @@ out: static int taprio_parse_tc_entry(struct Qdisc *sch, struct nlattr *opt, u32 max_sdu[TC_QOPT_MAX_QUEUE], + u32 fp[TC_QOPT_MAX_QUEUE], unsigned long *seen_tcs, struct netlink_ext_ack *extack) { struct nlattr *tb[TCA_TAPRIO_TC_ENTRY_MAX + 1] = { }; struct net_device *dev = qdisc_dev(sch); - u32 val = 0; int err, tc; + u32 val; err = nla_parse_nested(tb, TCA_TAPRIO_TC_ENTRY_MAX, opt, taprio_tc_policy, extack); @@ -1702,15 +1709,18 @@ static int taprio_parse_tc_entry(struct Qdisc *sch, *seen_tcs |= BIT(tc); - if (tb[TCA_TAPRIO_TC_ENTRY_MAX_SDU]) + if (tb[TCA_TAPRIO_TC_ENTRY_MAX_SDU]) { val = nla_get_u32(tb[TCA_TAPRIO_TC_ENTRY_MAX_SDU]); + if (val > dev->max_mtu) { + NL_SET_ERR_MSG_MOD(extack, "TC max SDU exceeds device max MTU"); + return -ERANGE; + } - if (val > dev->max_mtu) { - NL_SET_ERR_MSG_MOD(extack, "TC max SDU exceeds device max MTU"); - return -ERANGE; + max_sdu[tc] = val; } - max_sdu[tc] = val; + if (tb[TCA_TAPRIO_TC_ENTRY_FP]) + fp[tc] = nla_get_u32(tb[TCA_TAPRIO_TC_ENTRY_FP]); return 0; } @@ -1720,29 +1730,51 @@ static int taprio_parse_tc_entries(struct Qdisc *sch, struct netlink_ext_ack *extack) { struct taprio_sched *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); u32 max_sdu[TC_QOPT_MAX_QUEUE]; + bool have_preemption = false; unsigned long seen_tcs = 0; + u32 fp[TC_QOPT_MAX_QUEUE]; struct nlattr *n; int tc, rem; int err = 0; - for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) + for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) { max_sdu[tc] = q->max_sdu[tc]; + fp[tc] = q->fp[tc]; + } nla_for_each_nested(n, opt, rem) { if (nla_type(n) != TCA_TAPRIO_ATTR_TC_ENTRY) continue; - err = taprio_parse_tc_entry(sch, n, max_sdu, &seen_tcs, + err = taprio_parse_tc_entry(sch, n, max_sdu, fp, &seen_tcs, extack); if (err) - goto out; + return err; } - for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) + for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) { q->max_sdu[tc] = max_sdu[tc]; + q->fp[tc] = fp[tc]; + if (fp[tc] != TC_FP_EXPRESS) + have_preemption = true; + } + + if (have_preemption) { + if (!FULL_OFFLOAD_IS_ENABLED(q->flags)) { + NL_SET_ERR_MSG(extack, + "Preemption only supported with full offload"); + return -EOPNOTSUPP; + } + + if (!ethtool_dev_mm_supported(dev)) { + NL_SET_ERR_MSG(extack, + "Device does not support preemption"); + return -EOPNOTSUPP; + } + } -out: return err; } @@ -2023,7 +2055,7 @@ static int taprio_init(struct Qdisc *sch, struct nlattr *opt, { struct taprio_sched *q = qdisc_priv(sch); struct net_device *dev = qdisc_dev(sch); - int i; + int i, tc; spin_lock_init(&q->current_entry_lock); @@ -2080,6 +2112,9 @@ static int taprio_init(struct Qdisc *sch, struct nlattr *opt, q->qdiscs[i] = qdisc; } + for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) + q->fp[tc] = TC_FP_EXPRESS; + taprio_detect_broken_mqprio(q); return taprio_change(sch, opt, extack); @@ -2223,6 +2258,7 @@ error_nest: } static int taprio_dump_tc_entries(struct sk_buff *skb, + struct taprio_sched *q, struct sched_gate_list *sched) { struct nlattr *n; @@ -2240,6 +2276,9 @@ static int taprio_dump_tc_entries(struct sk_buff *skb, sched->max_sdu[tc])) goto nla_put_failure; + if (nla_put_u32(skb, TCA_TAPRIO_TC_ENTRY_FP, q->fp[tc])) + goto nla_put_failure; + nla_nest_end(skb, n); } @@ -2281,7 +2320,7 @@ static int taprio_dump(struct Qdisc *sch, struct sk_buff *skb) nla_put_u32(skb, TCA_TAPRIO_ATTR_TXTIME_DELAY, q->txtime_delay)) goto options_error; - if (oper && taprio_dump_tc_entries(skb, oper)) + if (oper && taprio_dump_tc_entries(skb, q, oper)) goto options_error; if (oper && dump_schedule(skb, oper)) -- cgit v1.2.3 From 3037933448f60f9acb705997eae62013ecb81e0d Mon Sep 17 00:00:00 2001 From: Gwangun Jung Date: Thu, 13 Apr 2023 19:35:54 +0900 Subject: net: sched: sch_qfq: prevent slab-out-of-bounds in qfq_activate_agg If the TCA_QFQ_LMAX value is not offered through nlattr, lmax is determined by the MTU value of the network device. The MTU of the loopback device can be set up to 2^31-1. As a result, it is possible to have an lmax value that exceeds QFQ_MIN_LMAX. Due to the invalid lmax value, an index is generated that exceeds the QFQ_MAX_INDEX(=24) value, causing out-of-bounds read/write errors. The following reports a oob access: [ 84.582666] BUG: KASAN: slab-out-of-bounds in qfq_activate_agg.constprop.0 (net/sched/sch_qfq.c:1027 net/sched/sch_qfq.c:1060 net/sched/sch_qfq.c:1313) [ 84.583267] Read of size 4 at addr ffff88810f676948 by task ping/301 [ 84.583686] [ 84.583797] CPU: 3 PID: 301 Comm: ping Not tainted 6.3.0-rc5 #1 [ 84.584164] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 [ 84.584644] Call Trace: [ 84.584787] [ 84.584906] dump_stack_lvl (lib/dump_stack.c:107 (discriminator 1)) [ 84.585108] print_report (mm/kasan/report.c:320 mm/kasan/report.c:430) [ 84.585570] kasan_report (mm/kasan/report.c:538) [ 84.585988] qfq_activate_agg.constprop.0 (net/sched/sch_qfq.c:1027 net/sched/sch_qfq.c:1060 net/sched/sch_qfq.c:1313) [ 84.586599] qfq_enqueue (net/sched/sch_qfq.c:1255) [ 84.587607] dev_qdisc_enqueue (net/core/dev.c:3776) [ 84.587749] __dev_queue_xmit (./include/net/sch_generic.h:186 net/core/dev.c:3865 net/core/dev.c:4212) [ 84.588763] ip_finish_output2 (./include/net/neighbour.h:546 net/ipv4/ip_output.c:228) [ 84.589460] ip_output (net/ipv4/ip_output.c:430) [ 84.590132] ip_push_pending_frames (./include/net/dst.h:444 net/ipv4/ip_output.c:126 net/ipv4/ip_output.c:1586 net/ipv4/ip_output.c:1606) [ 84.590285] raw_sendmsg (net/ipv4/raw.c:649) [ 84.591960] sock_sendmsg (net/socket.c:724 net/socket.c:747) [ 84.592084] __sys_sendto (net/socket.c:2142) [ 84.593306] __x64_sys_sendto (net/socket.c:2150) [ 84.593779] do_syscall_64 (arch/x86/entry/common.c:50 arch/x86/entry/common.c:80) [ 84.593902] entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120) [ 84.594070] RIP: 0033:0x7fe568032066 [ 84.594192] Code: 0e 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 41 89 ca 64 8b 04 25 18 00 00 00 85 c09[ 84.594796] RSP: 002b:00007ffce388b4e8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c Code starting with the faulting instruction =========================================== [ 84.595047] RAX: ffffffffffffffda RBX: 00007ffce388cc70 RCX: 00007fe568032066 [ 84.595281] RDX: 0000000000000040 RSI: 00005605fdad6d10 RDI: 0000000000000003 [ 84.595515] RBP: 00005605fdad6d10 R08: 00007ffce388eeec R09: 0000000000000010 [ 84.595749] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000040 [ 84.595984] R13: 00007ffce388cc30 R14: 00007ffce388b4f0 R15: 0000001d00000001 [ 84.596218] [ 84.596295] [ 84.596351] Allocated by task 291: [ 84.596467] kasan_save_stack (mm/kasan/common.c:46) [ 84.596597] kasan_set_track (mm/kasan/common.c:52) [ 84.596725] __kasan_kmalloc (mm/kasan/common.c:384) [ 84.596852] __kmalloc_node (./include/linux/kasan.h:196 mm/slab_common.c:967 mm/slab_common.c:974) [ 84.596979] qdisc_alloc (./include/linux/slab.h:610 ./include/linux/slab.h:731 net/sched/sch_generic.c:938) [ 84.597100] qdisc_create (net/sched/sch_api.c:1244) [ 84.597222] tc_modify_qdisc (net/sched/sch_api.c:1680) [ 84.597357] rtnetlink_rcv_msg (net/core/rtnetlink.c:6174) [ 84.597495] netlink_rcv_skb (net/netlink/af_netlink.c:2574) [ 84.597627] netlink_unicast (net/netlink/af_netlink.c:1340 net/netlink/af_netlink.c:1365) [ 84.597759] netlink_sendmsg (net/netlink/af_netlink.c:1942) [ 84.597891] sock_sendmsg (net/socket.c:724 net/socket.c:747) [ 84.598016] ____sys_sendmsg (net/socket.c:2501) [ 84.598147] ___sys_sendmsg (net/socket.c:2557) [ 84.598275] __sys_sendmsg (./include/linux/file.h:31 net/socket.c:2586) [ 84.598399] do_syscall_64 (arch/x86/entry/common.c:50 arch/x86/entry/common.c:80) [ 84.598520] entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120) [ 84.598688] [ 84.598744] The buggy address belongs to the object at ffff88810f674000 [ 84.598744] which belongs to the cache kmalloc-8k of size 8192 [ 84.599135] The buggy address is located 2664 bytes to the right of [ 84.599135] allocated 7904-byte region [ffff88810f674000, ffff88810f675ee0) [ 84.599544] [ 84.599598] The buggy address belongs to the physical page: [ 84.599777] page:00000000e638567f refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x10f670 [ 84.600074] head:00000000e638567f order:3 entire_mapcount:0 nr_pages_mapped:0 pincount:0 [ 84.600330] flags: 0x200000000010200(slab|head|node=0|zone=2) [ 84.600517] raw: 0200000000010200 ffff888100043180 dead000000000122 0000000000000000 [ 84.600764] raw: 0000000000000000 0000000080020002 00000001ffffffff 0000000000000000 [ 84.601009] page dumped because: kasan: bad access detected [ 84.601187] [ 84.601241] Memory state around the buggy address: [ 84.601396] ffff88810f676800: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 84.601620] ffff88810f676880: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 84.601845] >ffff88810f676900: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 84.602069] ^ [ 84.602243] ffff88810f676980: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 84.602468] ffff88810f676a00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 84.602693] ================================================================== [ 84.602924] Disabling lock debugging due to kernel taint Fixes: 3015f3d2a3cd ("pkt_sched: enable QFQ to support TSO/GSO") Reported-by: Gwangun Jung Signed-off-by: Gwangun Jung Acked-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- net/sched/sch_qfq.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index cf5ebe43b3b4..02098a02943e 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -421,15 +421,16 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, } else weight = 1; - if (tb[TCA_QFQ_LMAX]) { + if (tb[TCA_QFQ_LMAX]) lmax = nla_get_u32(tb[TCA_QFQ_LMAX]); - if (lmax < QFQ_MIN_LMAX || lmax > (1UL << QFQ_MTU_SHIFT)) { - pr_notice("qfq: invalid max length %u\n", lmax); - return -EINVAL; - } - } else + else lmax = psched_mtu(qdisc_dev(sch)); + if (lmax < QFQ_MIN_LMAX || lmax > (1UL << QFQ_MTU_SHIFT)) { + pr_notice("qfq: invalid max length %u\n", lmax); + return -EINVAL; + } + inv_w = ONE_FP / weight; weight = ONE_FP / inv_w; -- cgit v1.2.3 From eaaa4e92397981c65dae7b9a38425b8fa984d674 Mon Sep 17 00:00:00 2001 From: Arseniy Krasnov Date: Thu, 13 Apr 2023 12:17:19 +0300 Subject: vsock/loopback: don't disable irqs for queue access This replaces 'skb_queue_tail()' with 'virtio_vsock_skb_queue_tail()'. The first one uses 'spin_lock_irqsave()', second uses 'spin_lock_bh()'. There is no need to disable interrupts in the loopback transport as there is no access to the queue with skbs from interrupt context. Both virtio and vhost transports work in the same way. Signed-off-by: Arseniy Krasnov Reviewed-by: Stefano Garzarella Signed-off-by: David S. Miller --- net/vmw_vsock/vsock_loopback.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/vmw_vsock/vsock_loopback.c b/net/vmw_vsock/vsock_loopback.c index e3afc0c866f5..5c6360df1f31 100644 --- a/net/vmw_vsock/vsock_loopback.c +++ b/net/vmw_vsock/vsock_loopback.c @@ -31,8 +31,7 @@ static int vsock_loopback_send_pkt(struct sk_buff *skb) struct vsock_loopback *vsock = &the_vsock_loopback; int len = skb->len; - skb_queue_tail(&vsock->pkt_queue, skb); - + virtio_vsock_skb_queue_tail(&vsock->pkt_queue, skb); queue_work(vsock->workqueue, &vsock->pkt_work); return len; -- cgit v1.2.3 From c39ef2130491b2382f2c81774c825527d4bdf4e2 Mon Sep 17 00:00:00 2001 From: Kevin Brodsky Date: Thu, 13 Apr 2023 12:47:03 +0100 Subject: net: Ensure ->msg_control_user is used for user buffers Since commit 1f466e1f15cf ("net: cleanly handle kernel vs user buffers for ->msg_control"), pointers to user buffers should be stored in struct msghdr::msg_control_user, instead of the msg_control field. Most users of msg_control have already been converted (where user buffers are involved), but not all of them. This patch attempts to address the remaining cases. An exception is made for null checks, as it should be safe to use msg_control unconditionally for that purpose. Cc: Christoph Hellwig Cc: Eric Dumazet Cc: "David S. Miller" Cc: Jakub Kicinski Signed-off-by: Kevin Brodsky Reviewed-by: Christoph Hellwig Signed-off-by: David S. Miller --- net/compat.c | 12 ++++++------ net/core/scm.c | 9 ++++++--- net/ipv4/tcp.c | 4 ++-- 3 files changed, 14 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/compat.c b/net/compat.c index 161b7bea1f62..000a2e054d4c 100644 --- a/net/compat.c +++ b/net/compat.c @@ -113,7 +113,7 @@ int get_compat_msghdr(struct msghdr *kmsg, #define CMSG_COMPAT_FIRSTHDR(msg) \ (((msg)->msg_controllen) >= sizeof(struct compat_cmsghdr) ? \ - (struct compat_cmsghdr __user *)((msg)->msg_control) : \ + (struct compat_cmsghdr __user *)((msg)->msg_control_user) : \ (struct compat_cmsghdr __user *)NULL) #define CMSG_COMPAT_OK(ucmlen, ucmsg, mhdr) \ @@ -126,7 +126,7 @@ static inline struct compat_cmsghdr __user *cmsg_compat_nxthdr(struct msghdr *ms struct compat_cmsghdr __user *cmsg, int cmsg_len) { char __user *ptr = (char __user *)cmsg + CMSG_COMPAT_ALIGN(cmsg_len); - if ((unsigned long)(ptr + 1 - (char __user *)msg->msg_control) > + if ((unsigned long)(ptr + 1 - (char __user *)msg->msg_control_user) > msg->msg_controllen) return NULL; return (struct compat_cmsghdr __user *)ptr; @@ -225,7 +225,7 @@ Efault: int put_cmsg_compat(struct msghdr *kmsg, int level, int type, int len, void *data) { - struct compat_cmsghdr __user *cm = (struct compat_cmsghdr __user *) kmsg->msg_control; + struct compat_cmsghdr __user *cm = (struct compat_cmsghdr __user *) kmsg->msg_control_user; struct compat_cmsghdr cmhdr; struct old_timeval32 ctv; struct old_timespec32 cts[3]; @@ -274,7 +274,7 @@ int put_cmsg_compat(struct msghdr *kmsg, int level, int type, int len, void *dat cmlen = CMSG_COMPAT_SPACE(len); if (kmsg->msg_controllen < cmlen) cmlen = kmsg->msg_controllen; - kmsg->msg_control += cmlen; + kmsg->msg_control_user += cmlen; kmsg->msg_controllen -= cmlen; return 0; } @@ -289,7 +289,7 @@ static int scm_max_fds_compat(struct msghdr *msg) void scm_detach_fds_compat(struct msghdr *msg, struct scm_cookie *scm) { struct compat_cmsghdr __user *cm = - (struct compat_cmsghdr __user *)msg->msg_control; + (struct compat_cmsghdr __user *)msg->msg_control_user; unsigned int o_flags = (msg->msg_flags & MSG_CMSG_CLOEXEC) ? O_CLOEXEC : 0; int fdmax = min_t(int, scm_max_fds_compat(msg), scm->fp->count); int __user *cmsg_data = CMSG_COMPAT_DATA(cm); @@ -313,7 +313,7 @@ void scm_detach_fds_compat(struct msghdr *msg, struct scm_cookie *scm) cmlen = CMSG_COMPAT_SPACE(i * sizeof(int)); if (msg->msg_controllen < cmlen) cmlen = msg->msg_controllen; - msg->msg_control += cmlen; + msg->msg_control_user += cmlen; msg->msg_controllen -= cmlen; } } diff --git a/net/core/scm.c b/net/core/scm.c index acb7d776fa6e..3cd7dd377e53 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -250,7 +250,10 @@ int put_cmsg(struct msghdr * msg, int level, int type, int len, void *data) } cmlen = min(CMSG_SPACE(len), msg->msg_controllen); - msg->msg_control += cmlen; + if (msg->msg_control_is_user) + msg->msg_control_user += cmlen; + else + msg->msg_control += cmlen; msg->msg_controllen -= cmlen; return 0; @@ -299,7 +302,7 @@ static int scm_max_fds(struct msghdr *msg) void scm_detach_fds(struct msghdr *msg, struct scm_cookie *scm) { struct cmsghdr __user *cm = - (__force struct cmsghdr __user *)msg->msg_control; + (__force struct cmsghdr __user *)msg->msg_control_user; unsigned int o_flags = (msg->msg_flags & MSG_CMSG_CLOEXEC) ? O_CLOEXEC : 0; int fdmax = min_t(int, scm_max_fds(msg), scm->fp->count); int __user *cmsg_data = CMSG_USER_DATA(cm); @@ -332,7 +335,7 @@ void scm_detach_fds(struct msghdr *msg, struct scm_cookie *scm) cmlen = CMSG_SPACE(i * sizeof(int)); if (msg->msg_controllen < cmlen) cmlen = msg->msg_controllen; - msg->msg_control += cmlen; + msg->msg_control_user += cmlen; msg->msg_controllen -= cmlen; } } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index fd68d49490f2..20db115c38c4 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2165,7 +2165,7 @@ static void tcp_zc_finalize_rx_tstamp(struct sock *sk, struct msghdr cmsg_dummy; msg_control_addr = (unsigned long)zc->msg_control; - cmsg_dummy.msg_control = (void *)msg_control_addr; + cmsg_dummy.msg_control_user = (void __user *)msg_control_addr; cmsg_dummy.msg_controllen = (__kernel_size_t)zc->msg_controllen; cmsg_dummy.msg_flags = in_compat_syscall() @@ -2176,7 +2176,7 @@ static void tcp_zc_finalize_rx_tstamp(struct sock *sk, zc->msg_controllen == cmsg_dummy.msg_controllen) { tcp_recv_timestamp(&cmsg_dummy, sk, tss); zc->msg_control = (__u64) - ((uintptr_t)cmsg_dummy.msg_control); + ((uintptr_t)cmsg_dummy.msg_control_user); zc->msg_controllen = (__u64)cmsg_dummy.msg_controllen; zc->msg_flags = (__u32)cmsg_dummy.msg_flags; -- cgit v1.2.3 From 60daf8d40b80ccbd593930235aea9ee82ea8dbc2 Mon Sep 17 00:00:00 2001 From: Kevin Brodsky Date: Thu, 13 Apr 2023 12:47:04 +0100 Subject: net/compat: Update msg_control_is_user when setting a kernel pointer cmsghdr_from_user_compat_to_kern() is an unusual case w.r.t. how the kmsg->msg_control* fields are used. The input struct msghdr holds a pointer to a user buffer, i.e. ksmg->msg_control_user is active. However, upon success, a kernel pointer is stored in kmsg->msg_control. kmsg->msg_control_is_user should therefore be updated accordingly. Cc: Christoph Hellwig Cc: Eric Dumazet Cc: "David S. Miller" Cc: Jakub Kicinski Signed-off-by: Kevin Brodsky Reviewed-by: Christoph Hellwig Signed-off-by: David S. Miller --- net/compat.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/compat.c b/net/compat.c index 000a2e054d4c..6564720f32b7 100644 --- a/net/compat.c +++ b/net/compat.c @@ -211,6 +211,7 @@ int cmsghdr_from_user_compat_to_kern(struct msghdr *kmsg, struct sock *sk, goto Einval; /* Ok, looks like we made it. Hook it up and return success. */ + kmsg->msg_control_is_user = false; kmsg->msg_control = kcmsg_base; kmsg->msg_controllen = kcmlen; return 0; -- cgit v1.2.3 From b6d85cf5bd1433c5dd6bf6bb3a176537184c630c Mon Sep 17 00:00:00 2001 From: Kevin Brodsky Date: Thu, 13 Apr 2023 12:47:05 +0100 Subject: net/ipv6: Initialise msg_control_is_user do_ipv6_setsockopt() makes use of struct msghdr::msg_control in the IPV6_2292PKTOPTIONS case. Make sure to initialise msg_control_is_user accordingly. Cc: Christoph Hellwig Cc: Eric Dumazet Cc: "David S. Miller" Cc: Jakub Kicinski Signed-off-by: Kevin Brodsky Reviewed-by: Christoph Hellwig Signed-off-by: David S. Miller --- net/ipv6/ipv6_sockglue.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 2917dd8d198c..ae818ff46224 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -716,6 +716,7 @@ int do_ipv6_setsockopt(struct sock *sk, int level, int optname, goto done; msg.msg_controllen = optlen; + msg.msg_control_is_user = false; msg.msg_control = (void *)(opt+1); ipc6.opt = opt; -- cgit v1.2.3 From b07a2d97ba5ef154fe736aa510e43a3299eee5f8 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 12 Apr 2023 21:26:03 -0700 Subject: net: skb: plumb napi state thru skb freeing paths We maintain a NAPI-local cache of skbs which is fed by napi_consume_skb(). Going forward we will also try to cache head and data pages. Plumb the "are we in a normal NAPI context" information thru deeper into the freeing path, up to skb_release_data() and skb_free_head()/skb_pp_recycle(). The "not normal NAPI context" comes from netpoll which passes budget of 0 to try to reap the Tx completions but not perform any Rx. Use "bool napi_safe" rather than bare "int budget", the further we get from NAPI the more confusing the budget argument may seem (particularly whether 0 or MAX is the correct value to pass in when not in NAPI). Reviewed-by: Tariq Toukan Tested-by: Dragos Tatulea Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 38 ++++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 78238a13dbcf..2b5a98c5cb49 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -839,7 +839,7 @@ static void skb_clone_fraglist(struct sk_buff *skb) skb_get(list); } -static bool skb_pp_recycle(struct sk_buff *skb, void *data) +static bool skb_pp_recycle(struct sk_buff *skb, void *data, bool napi_safe) { if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle) return false; @@ -856,12 +856,12 @@ static void skb_kfree_head(void *head, unsigned int end_offset) kfree(head); } -static void skb_free_head(struct sk_buff *skb) +static void skb_free_head(struct sk_buff *skb, bool napi_safe) { unsigned char *head = skb->head; if (skb->head_frag) { - if (skb_pp_recycle(skb, head)) + if (skb_pp_recycle(skb, head, napi_safe)) return; skb_free_frag(head); } else { @@ -869,7 +869,8 @@ static void skb_free_head(struct sk_buff *skb) } } -static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason) +static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason, + bool napi_safe) { struct skb_shared_info *shinfo = skb_shinfo(skb); int i; @@ -894,7 +895,7 @@ free_head: if (shinfo->frag_list) kfree_skb_list_reason(shinfo->frag_list, reason); - skb_free_head(skb); + skb_free_head(skb, napi_safe); exit: /* When we clone an SKB we copy the reycling bit. The pp_recycle * bit is only set on the head though, so in order to avoid races @@ -955,11 +956,12 @@ void skb_release_head_state(struct sk_buff *skb) } /* Free everything but the sk_buff shell. */ -static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason) +static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason, + bool napi_safe) { skb_release_head_state(skb); if (likely(skb->head)) - skb_release_data(skb, reason); + skb_release_data(skb, reason, napi_safe); } /** @@ -973,7 +975,7 @@ static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason) void __kfree_skb(struct sk_buff *skb) { - skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED); + skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED, false); kfree_skbmem(skb); } EXPORT_SYMBOL(__kfree_skb); @@ -1027,7 +1029,7 @@ static void kfree_skb_add_bulk(struct sk_buff *skb, return; } - skb_release_all(skb, reason); + skb_release_all(skb, reason, false); sa->skb_array[sa->skb_count++] = skb; if (unlikely(sa->skb_count == KFREE_SKB_BULK_SIZE)) { @@ -1201,7 +1203,7 @@ EXPORT_SYMBOL(consume_skb); void __consume_stateless_skb(struct sk_buff *skb) { trace_consume_skb(skb, __builtin_return_address(0)); - skb_release_data(skb, SKB_CONSUMED); + skb_release_data(skb, SKB_CONSUMED, false); kfree_skbmem(skb); } @@ -1226,7 +1228,7 @@ static void napi_skb_cache_put(struct sk_buff *skb) void __kfree_skb_defer(struct sk_buff *skb) { - skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED); + skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED, true); napi_skb_cache_put(skb); } @@ -1264,7 +1266,7 @@ void napi_consume_skb(struct sk_buff *skb, int budget) return; } - skb_release_all(skb, SKB_CONSUMED); + skb_release_all(skb, SKB_CONSUMED, !!budget); napi_skb_cache_put(skb); } EXPORT_SYMBOL(napi_consume_skb); @@ -1395,7 +1397,7 @@ EXPORT_SYMBOL_GPL(alloc_skb_for_msg); */ struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src) { - skb_release_all(dst, SKB_CONSUMED); + skb_release_all(dst, SKB_CONSUMED, false); return __skb_clone(dst, src); } EXPORT_SYMBOL_GPL(skb_morph); @@ -2018,9 +2020,9 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, if (skb_has_frag_list(skb)) skb_clone_fraglist(skb); - skb_release_data(skb, SKB_CONSUMED); + skb_release_data(skb, SKB_CONSUMED, false); } else { - skb_free_head(skb); + skb_free_head(skb, false); } off = (data + nhead) - skb->head; @@ -6389,12 +6391,12 @@ static int pskb_carve_inside_header(struct sk_buff *skb, const u32 off, skb_frag_ref(skb, i); if (skb_has_frag_list(skb)) skb_clone_fraglist(skb); - skb_release_data(skb, SKB_CONSUMED); + skb_release_data(skb, SKB_CONSUMED, false); } else { /* we can reuse existing recount- all we did was * relocate values */ - skb_free_head(skb); + skb_free_head(skb, false); } skb->head = data; @@ -6529,7 +6531,7 @@ static int pskb_carve_inside_nonlinear(struct sk_buff *skb, const u32 off, skb_kfree_head(data, size); return -ENOMEM; } - skb_release_data(skb, SKB_CONSUMED); + skb_release_data(skb, SKB_CONSUMED, false); skb->head = data; skb->head_frag = 0; -- cgit v1.2.3 From 8c48eea3adf3119e0a3fc57bd31f6966f26ee784 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 12 Apr 2023 21:26:04 -0700 Subject: page_pool: allow caching from safely localized NAPI Recent patches to mlx5 mentioned a regression when moving from driver local page pool to only using the generic page pool code. Page pool has two recycling paths (1) direct one, which runs in safe NAPI context (basically consumer context, so producing can be lockless); and (2) via a ptr_ring, which takes a spin lock because the freeing can happen from any CPU; producer and consumer may run concurrently. Since the page pool code was added, Eric introduced a revised version of deferred skb freeing. TCP skbs are now usually returned to the CPU which allocated them, and freed in softirq context. This places the freeing (producing of pages back to the pool) enticingly close to the allocation (consumer). If we can prove that we're freeing in the same softirq context in which the consumer NAPI will run - lockless use of the cache is perfectly fine, no need for the lock. Let drivers link the page pool to a NAPI instance. If the NAPI instance is scheduled on the same CPU on which we're freeing - place the pages in the direct cache. With that and patched bnxt (XDP enabled to engage the page pool, sigh, bnxt really needs page pool work :() I see a 2.6% perf boost with a TCP stream test (app on a different physical core than softirq). The CPU use of relevant functions decreases as expected: page_pool_refill_alloc_cache 1.17% -> 0% _raw_spin_lock 2.41% -> 0.98% Only consider lockless path to be safe when NAPI is scheduled - in practice this should cover majority if not all of steady state workloads. It's usually the NAPI kicking in that causes the skb flush. The main case we'll miss out on is when application runs on the same CPU as NAPI. In that case we don't use the deferred skb free path. Reviewed-by: Tariq Toukan Acked-by: Jesper Dangaard Brouer Tested-by: Dragos Tatulea Signed-off-by: Jakub Kicinski --- Documentation/networking/page_pool.rst | 1 + include/linux/netdevice.h | 3 +++ include/linux/skbuff.h | 20 +++++++++++++------- include/net/page_pool.h | 3 ++- net/core/dev.c | 3 +++ net/core/page_pool.c | 15 +++++++++++++-- net/core/skbuff.c | 4 ++-- 7 files changed, 37 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/Documentation/networking/page_pool.rst b/Documentation/networking/page_pool.rst index 30f1344e7cca..873efd97f822 100644 --- a/Documentation/networking/page_pool.rst +++ b/Documentation/networking/page_pool.rst @@ -165,6 +165,7 @@ Registration pp_params.pool_size = DESC_NUM; pp_params.nid = NUMA_NO_NODE; pp_params.dev = priv->dev; + pp_params.napi = napi; /* only if locking is tied to NAPI */ pp_params.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE; page_pool = page_pool_create(&pp_params); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 96d27d558b0c..203c0df2046c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -360,8 +360,11 @@ struct napi_struct { unsigned long gro_bitmask; int (*poll)(struct napi_struct *, int); #ifdef CONFIG_NETPOLL + /* CPU actively polling if netpoll is configured */ int poll_owner; #endif + /* CPU on which NAPI has been scheduled for processing */ + int list_owner; struct net_device *dev; struct gro_list gro_hash[GRO_HASH_BUCKETS]; struct sk_buff *skb; diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 494a23a976b0..a823ec3aa326 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3386,6 +3386,18 @@ static inline void skb_frag_ref(struct sk_buff *skb, int f) __skb_frag_ref(&skb_shinfo(skb)->frags[f]); } +static inline void +napi_frag_unref(skb_frag_t *frag, bool recycle, bool napi_safe) +{ + struct page *page = skb_frag_page(frag); + +#ifdef CONFIG_PAGE_POOL + if (recycle && page_pool_return_skb_page(page, napi_safe)) + return; +#endif + put_page(page); +} + /** * __skb_frag_unref - release a reference on a paged fragment. * @frag: the paged fragment @@ -3396,13 +3408,7 @@ static inline void skb_frag_ref(struct sk_buff *skb, int f) */ static inline void __skb_frag_unref(skb_frag_t *frag, bool recycle) { - struct page *page = skb_frag_page(frag); - -#ifdef CONFIG_PAGE_POOL - if (recycle && page_pool_return_skb_page(page)) - return; -#endif - put_page(page); + napi_frag_unref(frag, recycle, false); } /** diff --git a/include/net/page_pool.h b/include/net/page_pool.h index ddfa0b328677..91b808dade82 100644 --- a/include/net/page_pool.h +++ b/include/net/page_pool.h @@ -77,6 +77,7 @@ struct page_pool_params { unsigned int pool_size; int nid; /* Numa node id to allocate from pages from */ struct device *dev; /* device, for DMA pre-mapping purposes */ + struct napi_struct *napi; /* Sole consumer of pages, otherwise NULL */ enum dma_data_direction dma_dir; /* DMA mapping direction */ unsigned int max_len; /* max DMA sync memory size */ unsigned int offset; /* DMA addr offset */ @@ -239,7 +240,7 @@ inline enum dma_data_direction page_pool_get_dma_dir(struct page_pool *pool) return pool->p.dma_dir; } -bool page_pool_return_skb_page(struct page *page); +bool page_pool_return_skb_page(struct page *page, bool napi_safe); struct page_pool *page_pool_create(const struct page_pool_params *params); diff --git a/net/core/dev.c b/net/core/dev.c index c7f13742b56c..8aea68275172 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4359,6 +4359,7 @@ static inline void ____napi_schedule(struct softnet_data *sd, } list_add_tail(&napi->poll_list, &sd->poll_list); + WRITE_ONCE(napi->list_owner, smp_processor_id()); /* If not called from net_rx_action() * we have to raise NET_RX_SOFTIRQ. */ @@ -6069,6 +6070,7 @@ bool napi_complete_done(struct napi_struct *n, int work_done) list_del_init(&n->poll_list); local_irq_restore(flags); } + WRITE_ONCE(n->list_owner, -1); val = READ_ONCE(n->state); do { @@ -6384,6 +6386,7 @@ void netif_napi_add_weight(struct net_device *dev, struct napi_struct *napi, #ifdef CONFIG_NETPOLL napi->poll_owner = -1; #endif + napi->list_owner = -1; set_bit(NAPI_STATE_SCHED, &napi->state); set_bit(NAPI_STATE_NPSVC, &napi->state); list_add_rcu(&napi->dev_list, &dev->napi_list); diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 193c18799865..2f6bf422ed30 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -19,6 +19,7 @@ #include /* for put_page() */ #include #include +#include #include @@ -874,9 +875,11 @@ void page_pool_update_nid(struct page_pool *pool, int new_nid) } EXPORT_SYMBOL(page_pool_update_nid); -bool page_pool_return_skb_page(struct page *page) +bool page_pool_return_skb_page(struct page *page, bool napi_safe) { + struct napi_struct *napi; struct page_pool *pp; + bool allow_direct; page = compound_head(page); @@ -892,12 +895,20 @@ bool page_pool_return_skb_page(struct page *page) pp = page->pp; + /* Allow direct recycle if we have reasons to believe that we are + * in the same context as the consumer would run, so there's + * no possible race. + */ + napi = pp->p.napi; + allow_direct = napi_safe && napi && + READ_ONCE(napi->list_owner) == smp_processor_id(); + /* Driver set this to memory recycling info. Reset it on recycle. * This will *not* work for NIC using a split-page memory model. * The page will be returned to the pool here regardless of the * 'flipped' fragment being in use or not. */ - page_pool_put_full_page(pp, page, false); + page_pool_put_full_page(pp, page, allow_direct); return true; } diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 2b5a98c5cb49..ef81452759be 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -843,7 +843,7 @@ static bool skb_pp_recycle(struct sk_buff *skb, void *data, bool napi_safe) { if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle) return false; - return page_pool_return_skb_page(virt_to_page(data)); + return page_pool_return_skb_page(virt_to_page(data), napi_safe); } static void skb_kfree_head(void *head, unsigned int end_offset) @@ -889,7 +889,7 @@ static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason, } for (i = 0; i < shinfo->nr_frags; i++) - __skb_frag_unref(&shinfo->frags[i], skb->pp_recycle); + napi_frag_unref(&shinfo->frags[i], skb->pp_recycle, napi_safe); free_head: if (shinfo->frag_list) -- cgit v1.2.3 From 09b501d905217a38f03c0f07d5a66e0b5c8c1644 Mon Sep 17 00:00:00 2001 From: David Vernet Date: Sun, 16 Apr 2023 03:49:26 -0500 Subject: bpf: Remove bpf_kfunc_call_test_kptr_get() test kfunc We've managed to improve the UX for kptrs significantly over the last 9 months. All of the prior main use cases, struct bpf_cpumask *, struct task_struct *, and struct cgroup *, have all been updated to be synchronized mainly using RCU. In other words, their KF_ACQUIRE kfunc calls are all KF_RCU, and the pointers themselves are MEM_RCU and can be accessed in an RCU read region in BPF. In a follow-on change, we'll be removing the KF_KPTR_GET kfunc flag. This patch prepares for that by removing the bpf_kfunc_call_test_kptr_get() kfunc, and all associated selftests. Signed-off-by: David Vernet Link: https://lore.kernel.org/r/20230416084928.326135-2-void@manifault.com Signed-off-by: Alexei Starovoitov --- net/bpf/test_run.c | 12 ---- tools/testing/selftests/bpf/progs/map_kptr.c | 40 ++---------- tools/testing/selftests/bpf/progs/map_kptr_fail.c | 78 ----------------------- tools/testing/selftests/bpf/verifier/map_kptr.c | 27 -------- 4 files changed, 5 insertions(+), 152 deletions(-) (limited to 'net') diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 0b9bd9b39990..f170e8a17974 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -679,17 +679,6 @@ __bpf_kfunc void bpf_kfunc_call_int_mem_release(int *p) { } -__bpf_kfunc struct prog_test_ref_kfunc * -bpf_kfunc_call_test_kptr_get(struct prog_test_ref_kfunc **pp, int a, int b) -{ - struct prog_test_ref_kfunc *p = READ_ONCE(*pp); - - if (!p) - return NULL; - refcount_inc(&p->cnt); - return p; -} - struct prog_test_pass1 { int x0; struct { @@ -804,7 +793,6 @@ BTF_ID_FLAGS(func, bpf_kfunc_call_test_get_rdwr_mem, KF_RET_NULL) BTF_ID_FLAGS(func, bpf_kfunc_call_test_get_rdonly_mem, KF_RET_NULL) BTF_ID_FLAGS(func, bpf_kfunc_call_test_acq_rdonly_mem, KF_ACQUIRE | KF_RET_NULL) BTF_ID_FLAGS(func, bpf_kfunc_call_int_mem_release, KF_RELEASE) -BTF_ID_FLAGS(func, bpf_kfunc_call_test_kptr_get, KF_ACQUIRE | KF_RET_NULL | KF_KPTR_GET) BTF_ID_FLAGS(func, bpf_kfunc_call_test_pass_ctx) BTF_ID_FLAGS(func, bpf_kfunc_call_test_pass1) BTF_ID_FLAGS(func, bpf_kfunc_call_test_pass2) diff --git a/tools/testing/selftests/bpf/progs/map_kptr.c b/tools/testing/selftests/bpf/progs/map_kptr.c index dae5dab1bbf7..d7150041e5d1 100644 --- a/tools/testing/selftests/bpf/progs/map_kptr.c +++ b/tools/testing/selftests/bpf/progs/map_kptr.c @@ -115,8 +115,6 @@ DEFINE_MAP_OF_MAP(BPF_MAP_TYPE_HASH_OF_MAPS, hash_malloc_map, hash_of_hash_mallo DEFINE_MAP_OF_MAP(BPF_MAP_TYPE_HASH_OF_MAPS, lru_hash_map, hash_of_lru_hash_maps); extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym; -extern struct prog_test_ref_kfunc * -bpf_kfunc_call_test_kptr_get(struct prog_test_ref_kfunc **p, int a, int b) __ksym; extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym; void bpf_kfunc_call_test_ref(struct prog_test_ref_kfunc *p) __ksym; @@ -187,25 +185,10 @@ static void test_kptr_ref(struct map_value *v) bpf_kfunc_call_test_release(p); } -static void test_kptr_get(struct map_value *v) -{ - struct prog_test_ref_kfunc *p; - - p = bpf_kfunc_call_test_kptr_get(&v->ref_ptr, 0, 0); - if (!p) - return; - if (p->a + p->b > 100) { - bpf_kfunc_call_test_release(p); - return; - } - bpf_kfunc_call_test_release(p); -} - static void test_kptr(struct map_value *v) { test_kptr_unref(v); test_kptr_ref(v); - test_kptr_get(v); } SEC("tc") @@ -338,38 +321,25 @@ int test_map_kptr_ref_pre(struct map_value *v) if (p_st->cnt.refs.counter != ref) return 4; - p = bpf_kfunc_call_test_kptr_get(&v->ref_ptr, 0, 0); - if (!p) - return 5; - ref++; - if (p_st->cnt.refs.counter != ref) { - ret = 6; - goto end; - } - bpf_kfunc_call_test_release(p); - ref--; - if (p_st->cnt.refs.counter != ref) - return 7; - p = bpf_kptr_xchg(&v->ref_ptr, NULL); if (!p) - return 8; + return 5; bpf_kfunc_call_test_release(p); ref--; if (p_st->cnt.refs.counter != ref) - return 9; + return 6; p = bpf_kfunc_call_test_acquire(&arg); if (!p) - return 10; + return 7; ref++; p = bpf_kptr_xchg(&v->ref_ptr, p); if (p) { - ret = 11; + ret = 8; goto end; } if (p_st->cnt.refs.counter != ref) - return 12; + return 9; /* Leave in map */ return 0; diff --git a/tools/testing/selftests/bpf/progs/map_kptr_fail.c b/tools/testing/selftests/bpf/progs/map_kptr_fail.c index 15bf3127dba3..da8c724f839b 100644 --- a/tools/testing/selftests/bpf/progs/map_kptr_fail.c +++ b/tools/testing/selftests/bpf/progs/map_kptr_fail.c @@ -21,8 +21,6 @@ struct array_map { extern struct prog_test_ref_kfunc *bpf_kfunc_call_test_acquire(unsigned long *sp) __ksym; extern void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym; -extern struct prog_test_ref_kfunc * -bpf_kfunc_call_test_kptr_get(struct prog_test_ref_kfunc **p, int a, int b) __ksym; SEC("?tc") __failure __msg("kptr access size must be BPF_DW") @@ -220,67 +218,6 @@ int reject_kptr_xchg_on_unref(struct __sk_buff *ctx) return 0; } -SEC("?tc") -__failure __msg("arg#0 expected pointer to map value") -int reject_kptr_get_no_map_val(struct __sk_buff *ctx) -{ - bpf_kfunc_call_test_kptr_get((void *)&ctx, 0, 0); - return 0; -} - -SEC("?tc") -__failure __msg("arg#0 expected pointer to map value") -int reject_kptr_get_no_null_map_val(struct __sk_buff *ctx) -{ - bpf_kfunc_call_test_kptr_get(bpf_map_lookup_elem(&array_map, &(int){0}), 0, 0); - return 0; -} - -SEC("?tc") -__failure __msg("arg#0 no referenced kptr at map value offset=0") -int reject_kptr_get_no_kptr(struct __sk_buff *ctx) -{ - struct map_value *v; - int key = 0; - - v = bpf_map_lookup_elem(&array_map, &key); - if (!v) - return 0; - - bpf_kfunc_call_test_kptr_get((void *)v, 0, 0); - return 0; -} - -SEC("?tc") -__failure __msg("arg#0 no referenced kptr at map value offset=8") -int reject_kptr_get_on_unref(struct __sk_buff *ctx) -{ - struct map_value *v; - int key = 0; - - v = bpf_map_lookup_elem(&array_map, &key); - if (!v) - return 0; - - bpf_kfunc_call_test_kptr_get(&v->unref_ptr, 0, 0); - return 0; -} - -SEC("?tc") -__failure __msg("kernel function bpf_kfunc_call_test_kptr_get args#0") -int reject_kptr_get_bad_type_match(struct __sk_buff *ctx) -{ - struct map_value *v; - int key = 0; - - v = bpf_map_lookup_elem(&array_map, &key); - if (!v) - return 0; - - bpf_kfunc_call_test_kptr_get((void *)&v->ref_memb_ptr, 0, 0); - return 0; -} - SEC("?tc") __failure __msg("R1 type=rcu_ptr_or_null_ expected=percpu_ptr_") int mark_ref_as_untrusted_or_null(struct __sk_buff *ctx) @@ -428,21 +365,6 @@ int kptr_xchg_ref_state(struct __sk_buff *ctx) return 0; } -SEC("?tc") -__failure __msg("Unreleased reference id=3 alloc_insn=") -int kptr_get_ref_state(struct __sk_buff *ctx) -{ - struct map_value *v; - int key = 0; - - v = bpf_map_lookup_elem(&array_map, &key); - if (!v) - return 0; - - bpf_kfunc_call_test_kptr_get(&v->ref_ptr, 0, 0); - return 0; -} - SEC("?tc") __failure __msg("Possibly NULL pointer passed to helper arg2") int kptr_xchg_possibly_null(struct __sk_buff *ctx) diff --git a/tools/testing/selftests/bpf/verifier/map_kptr.c b/tools/testing/selftests/bpf/verifier/map_kptr.c index d775ccb01989..a0cfc06d75bc 100644 --- a/tools/testing/selftests/bpf/verifier/map_kptr.c +++ b/tools/testing/selftests/bpf/verifier/map_kptr.c @@ -288,33 +288,6 @@ .result = REJECT, .errstr = "off=0 kptr isn't referenced kptr", }, -{ - "map_kptr: unref: bpf_kfunc_call_test_kptr_get rejected", - .insns = { - BPF_MOV64_REG(BPF_REG_2, BPF_REG_10), - BPF_LD_MAP_FD(BPF_REG_6, 0), - BPF_MOV64_REG(BPF_REG_1, BPF_REG_6), - BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), - BPF_MOV64_IMM(BPF_REG_0, 0), - BPF_ST_MEM(BPF_W, BPF_REG_2, 0, 0), - BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem), - BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1), - BPF_EXIT_INSN(), - BPF_MOV64_REG(BPF_REG_1, BPF_REG_0), - BPF_MOV64_IMM(BPF_REG_2, 0), - BPF_MOV64_IMM(BPF_REG_3, 0), - BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0), - BPF_MOV64_IMM(BPF_REG_0, 0), - BPF_EXIT_INSN(), - }, - .prog_type = BPF_PROG_TYPE_SCHED_CLS, - .fixup_map_kptr = { 1 }, - .result = REJECT, - .errstr = "arg#0 no referenced kptr at map value offset=0", - .fixup_kfunc_btf_id = { - { "bpf_kfunc_call_test_kptr_get", 13 }, - } -}, /* Tests for referenced PTR_TO_BTF_ID */ { "map_kptr: ref: loaded pointer marked as untrusted", -- cgit v1.2.3 From 7a486c443c89bd949f7a64e0040f704e02710b3c Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 14 Apr 2023 16:08:00 +0200 Subject: mptcp: drop unneeded argument After commit 3a236aef280e ("mptcp: refactor passive socket initialization"), every mptcp_pm_fully_established() call is always invoked with a GFP_ATOMIC argument. We can then drop it. Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/options.c | 2 +- net/mptcp/pm.c | 4 ++-- net/mptcp/protocol.h | 2 +- net/mptcp/subflow.c | 2 +- 4 files changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 355f798d575a..cd3b885c8faa 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -1001,7 +1001,7 @@ check_notify: clear_3rdack_retransmission(ssk); mptcp_pm_subflow_established(msk); } else { - mptcp_pm_fully_established(msk, ssk, GFP_ATOMIC); + mptcp_pm_fully_established(msk, ssk); } return true; diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 70f0ced3ca86..78c924506e83 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -126,7 +126,7 @@ static bool mptcp_pm_schedule_work(struct mptcp_sock *msk, return true; } -void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk, gfp_t gfp) +void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk) { struct mptcp_pm_data *pm = &msk->pm; bool announce = false; @@ -150,7 +150,7 @@ void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk, spin_unlock_bh(&pm->lock); if (announce) - mptcp_event(MPTCP_EVENT_ESTABLISHED, msk, ssk, gfp); + mptcp_event(MPTCP_EVENT_ESTABLISHED, msk, ssk, GFP_ATOMIC); } void mptcp_pm_connection_closed(struct mptcp_sock *msk) diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index e1310bc113be..a9eb0e428a6b 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -782,7 +782,7 @@ bool mptcp_pm_addr_families_match(const struct sock *sk, void mptcp_pm_subflow_chk_stale(const struct mptcp_sock *msk, struct sock *ssk); void mptcp_pm_nl_subflow_chk_stale(const struct mptcp_sock *msk, struct sock *ssk); void mptcp_pm_new_connection(struct mptcp_sock *msk, const struct sock *ssk, int server_side); -void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk, gfp_t gfp); +void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk); bool mptcp_pm_allow_new_subflow(struct mptcp_sock *msk); void mptcp_pm_connection_closed(struct mptcp_sock *msk); void mptcp_pm_subflow_established(struct mptcp_sock *msk); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index f46d8f6c40aa..80bbe96c0694 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -850,7 +850,7 @@ create_child: */ if (mp_opt.suboptions & OPTION_MPTCP_MPC_ACK) { mptcp_subflow_fully_established(ctx, &mp_opt); - mptcp_pm_fully_established(owner, child, GFP_ATOMIC); + mptcp_pm_fully_established(owner, child); ctx->pm_notified = 1; } } else if (ctx->mp_join) { -- cgit v1.2.3 From 617612316953093bc859890e405e1b550c27d840 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 14 Apr 2023 16:08:01 +0200 Subject: mptcp: avoid unneeded __mptcp_nmpc_socket() usage In a few spots, the mptcp code invokes the __mptcp_nmpc_socket() helper multiple times under the same socket lock scope. Additionally, in such places, the socket status ensures that there is no MP capable handshake running. Under the above condition we can replace the later __mptcp_nmpc_socket() helper invocation with direct access to the msk->subflow pointer and better document such access is not supposed to fail with WARN(). Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/protocol.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index e6cb36784a68..9cdcfdb44aee 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3143,7 +3143,7 @@ static struct sock *mptcp_accept(struct sock *sk, int flags, int *err, struct socket *listener; struct sock *newsk; - listener = __mptcp_nmpc_socket(msk); + listener = msk->subflow; if (WARN_ON_ONCE(!listener)) { *err = -EINVAL; return NULL; @@ -3363,7 +3363,7 @@ static int mptcp_get_port(struct sock *sk, unsigned short snum) struct mptcp_sock *msk = mptcp_sk(sk); struct socket *ssock; - ssock = __mptcp_nmpc_socket(msk); + ssock = msk->subflow; pr_debug("msk=%p, subflow=%p", msk, ssock); if (WARN_ON_ONCE(!ssock)) return -EINVAL; @@ -3709,7 +3709,10 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, pr_debug("msk=%p", msk); - ssock = __mptcp_nmpc_socket(msk); + /* buggy applications can call accept on socket states other then LISTEN + * but no need to allocate the first subflow just to error out. + */ + ssock = msk->subflow; if (!ssock) return -EINVAL; -- cgit v1.2.3 From a2702a076e73787db660fb8fd07d26a1a3108358 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 14 Apr 2023 16:08:02 +0200 Subject: mptcp: move fastopen subflow check inside mptcp_sendmsg_fastopen() So that we can avoid a bunch of check in fastpath. Additionally we can specialize such check according to the specific fastopen method - defer_connect vs MSG_FASTOPEN. The latter bits will simplify the next patches. Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/protocol.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 9cdcfdb44aee..22e073b373af 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -1662,13 +1662,27 @@ static void mptcp_set_nospace(struct sock *sk) static int mptcp_disconnect(struct sock *sk, int flags); -static int mptcp_sendmsg_fastopen(struct sock *sk, struct sock *ssk, struct msghdr *msg, +static int mptcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, size_t len, int *copied_syn) { unsigned int saved_flags = msg->msg_flags; struct mptcp_sock *msk = mptcp_sk(sk); + struct sock *ssk; int ret; + /* on flags based fastopen the mptcp is supposed to create the + * first subflow right now. Otherwise we are in the defer_connect + * path, and the first subflow must be already present. + * Since the defer_connect flag is cleared after the first succsful + * fastopen attempt, no need to check for additional subflow status. + */ + if (msg->msg_flags & MSG_FASTOPEN && !__mptcp_nmpc_socket(msk)) + return -EINVAL; + if (!msk->first) + return -EINVAL; + + ssk = msk->first; + lock_sock(ssk); msg->msg_flags |= MSG_DONTWAIT; msk->connect_flags = O_NONBLOCK; @@ -1691,6 +1705,7 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct sock *ssk, struct msgh } else if (ret && ret != -EINPROGRESS) { mptcp_disconnect(sk, 0); } + inet_sk(sk)->defer_connect = 0; return ret; } @@ -1699,7 +1714,6 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) { struct mptcp_sock *msk = mptcp_sk(sk); struct page_frag *pfrag; - struct socket *ssock; size_t copied = 0; int ret = 0; long timeo; @@ -1709,12 +1723,10 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) lock_sock(sk); - ssock = __mptcp_nmpc_socket(msk); - if (unlikely(ssock && (inet_sk(ssock->sk)->defer_connect || - msg->msg_flags & MSG_FASTOPEN))) { + if (unlikely(inet_sk(sk)->defer_connect || msg->msg_flags & MSG_FASTOPEN)) { int copied_syn = 0; - ret = mptcp_sendmsg_fastopen(sk, ssock->sk, msg, len, &copied_syn); + ret = mptcp_sendmsg_fastopen(sk, msg, len, &copied_syn); copied += copied_syn; if (ret == -EINPROGRESS && copied_syn > 0) goto out; -- cgit v1.2.3 From ddb1a072f858704b3555876877ca38c5b103a215 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 14 Apr 2023 16:08:03 +0200 Subject: mptcp: move first subflow allocation at mpc access time In the long run this will simplify the mptcp code and will allow for more consistent behavior. Move the first subflow allocation out of the sock->init ops into the __mptcp_nmpc_socket() helper. Since the first subflow creation can now happen after the first setsockopt() we additionally need to invoke mptcp_sockopt_sync() on it. Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/pm_netlink.c | 4 ++-- net/mptcp/protocol.c | 61 ++++++++++++++++++++++++++++++-------------------- net/mptcp/protocol.h | 2 +- net/mptcp/sockopt.c | 24 +++++++++++--------- 4 files changed, 54 insertions(+), 37 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c index 1c42bebca39e..bc343dab5e3f 100644 --- a/net/mptcp/pm_netlink.c +++ b/net/mptcp/pm_netlink.c @@ -1035,8 +1035,8 @@ static int mptcp_pm_nl_create_listen_socket(struct sock *sk, lock_sock(newsk); ssock = __mptcp_nmpc_socket(mptcp_sk(newsk)); release_sock(newsk); - if (!ssock) - return -EINVAL; + if (IS_ERR(ssock)) + return PTR_ERR(ssock); mptcp_info2sockaddr(&entry->addr, &addr, entry->addr.family); #if IS_ENABLED(CONFIG_MPTCP_IPV6) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 22e073b373af..a676ac1bb9f1 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -49,18 +49,6 @@ static void __mptcp_check_send_data_fin(struct sock *sk); DEFINE_PER_CPU(struct mptcp_delegated_action, mptcp_delegated_actions); static struct net_device mptcp_napi_dev; -/* If msk has an initial subflow socket, and the MP_CAPABLE handshake has not - * completed yet or has failed, return the subflow socket. - * Otherwise return NULL. - */ -struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk) -{ - if (!msk->subflow || READ_ONCE(msk->can_ack)) - return NULL; - - return msk->subflow; -} - /* Returns end sequence number of the receiver's advertised window */ static u64 mptcp_wnd_end(const struct mptcp_sock *msk) { @@ -116,6 +104,31 @@ static int __mptcp_socket_create(struct mptcp_sock *msk) return 0; } +/* If the MPC handshake is not started, returns the first subflow, + * eventually allocating it. + */ +struct socket *__mptcp_nmpc_socket(struct mptcp_sock *msk) +{ + struct sock *sk = (struct sock *)msk; + int ret; + + if (!((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN))) + return ERR_PTR(-EINVAL); + + if (!msk->subflow) { + if (msk->first) + return ERR_PTR(-EINVAL); + + ret = __mptcp_socket_create(msk); + if (ret) + return ERR_PTR(ret); + + mptcp_sockopt_sync(msk, msk->first); + } + + return msk->subflow; +} + static void mptcp_drop(struct sock *sk, struct sk_buff *skb) { sk_drops_add(sk, skb); @@ -1667,6 +1680,7 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, { unsigned int saved_flags = msg->msg_flags; struct mptcp_sock *msk = mptcp_sk(sk); + struct socket *ssock; struct sock *ssk; int ret; @@ -1676,8 +1690,11 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, * Since the defer_connect flag is cleared after the first succsful * fastopen attempt, no need to check for additional subflow status. */ - if (msg->msg_flags & MSG_FASTOPEN && !__mptcp_nmpc_socket(msk)) - return -EINVAL; + if (msg->msg_flags & MSG_FASTOPEN) { + ssock = __mptcp_nmpc_socket(msk); + if (IS_ERR(ssock)) + return PTR_ERR(ssock); + } if (!msk->first) return -EINVAL; @@ -2740,10 +2757,6 @@ static int mptcp_init_sock(struct sock *sk) if (unlikely(!net->mib.mptcp_statistics) && !mptcp_mib_alloc(net)) return -ENOMEM; - ret = __mptcp_socket_create(mptcp_sk(sk)); - if (ret) - return ret; - set_bit(SOCK_CUSTOM_SOCKOPT, &sk->sk_socket->flags); /* fetch the ca name; do it outside __mptcp_init_sock(), so that clone will @@ -3563,8 +3576,8 @@ static int mptcp_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) int err = -EINVAL; ssock = __mptcp_nmpc_socket(msk); - if (!ssock) - return -EINVAL; + if (IS_ERR(ssock)) + return PTR_ERR(ssock); mptcp_token_destroy(msk); inet_sk_state_store(sk, TCP_SYN_SENT); @@ -3652,8 +3665,8 @@ static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) lock_sock(sock->sk); ssock = __mptcp_nmpc_socket(msk); - if (!ssock) { - err = -EINVAL; + if (IS_ERR(ssock)) { + err = PTR_ERR(ssock); goto unlock; } @@ -3689,8 +3702,8 @@ static int mptcp_listen(struct socket *sock, int backlog) lock_sock(sk); ssock = __mptcp_nmpc_socket(msk); - if (!ssock) { - err = -EINVAL; + if (IS_ERR(ssock)) { + err = PTR_ERR(ssock); goto unlock; } diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index a9eb0e428a6b..21eda9cd0c52 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -627,7 +627,7 @@ void mptcp_close_ssk(struct sock *sk, struct sock *ssk, void __mptcp_subflow_send_ack(struct sock *ssk); void mptcp_subflow_reset(struct sock *ssk); void mptcp_sock_graft(struct sock *sk, struct socket *parent); -struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk); +struct socket *__mptcp_nmpc_socket(struct mptcp_sock *msk); bool __mptcp_close(struct sock *sk, long timeout); void mptcp_cancel_work(struct sock *sk); void mptcp_set_owner_r(struct sk_buff *skb, struct sock *sk); diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index b655cebda0f3..d4258869ac48 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -301,9 +301,9 @@ static int mptcp_setsockopt_sol_socket(struct mptcp_sock *msk, int optname, case SO_BINDTOIFINDEX: lock_sock(sk); ssock = __mptcp_nmpc_socket(msk); - if (!ssock) { + if (IS_ERR(ssock)) { release_sock(sk); - return -EINVAL; + return PTR_ERR(ssock); } ret = sock_setsockopt(ssock, SOL_SOCKET, optname, optval, optlen); @@ -396,9 +396,9 @@ static int mptcp_setsockopt_v6(struct mptcp_sock *msk, int optname, case IPV6_FREEBIND: lock_sock(sk); ssock = __mptcp_nmpc_socket(msk); - if (!ssock) { + if (IS_ERR(ssock)) { release_sock(sk); - return -EINVAL; + return PTR_ERR(ssock); } ret = tcp_setsockopt(ssock->sk, SOL_IPV6, optname, optval, optlen); @@ -693,9 +693,9 @@ static int mptcp_setsockopt_sol_ip_set_transparent(struct mptcp_sock *msk, int o lock_sock(sk); ssock = __mptcp_nmpc_socket(msk); - if (!ssock) { + if (IS_ERR(ssock)) { release_sock(sk); - return -EINVAL; + return PTR_ERR(ssock); } issk = inet_sk(ssock->sk); @@ -762,13 +762,15 @@ static int mptcp_setsockopt_first_sf_only(struct mptcp_sock *msk, int level, int { struct sock *sk = (struct sock *)msk; struct socket *sock; - int ret = -EINVAL; + int ret; /* Limit to first subflow, before the connection establishment */ lock_sock(sk); sock = __mptcp_nmpc_socket(msk); - if (!sock) + if (IS_ERR(sock)) { + ret = PTR_ERR(sock); goto unlock; + } ret = tcp_setsockopt(sock->sk, level, optname, optval, optlen); @@ -861,7 +863,7 @@ static int mptcp_getsockopt_first_sf_only(struct mptcp_sock *msk, int level, int { struct sock *sk = (struct sock *)msk; struct socket *ssock; - int ret = -EINVAL; + int ret; struct sock *ssk; lock_sock(sk); @@ -872,8 +874,10 @@ static int mptcp_getsockopt_first_sf_only(struct mptcp_sock *msk, int level, int } ssock = __mptcp_nmpc_socket(msk); - if (!ssock) + if (IS_ERR(ssock)) { + ret = PTR_ERR(ssock); goto out; + } ret = tcp_getsockopt(ssock->sk, level, optname, optval, optlen); -- cgit v1.2.3 From 8d547809a5d74aacf022d1df7a508c631088aaec Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 14 Apr 2023 16:08:04 +0200 Subject: mptcp: fastclose msk when cleaning unaccepted sockets When cleaning up unaccepted mptcp socket still laying inside the listener queue at listener close time, such sockets will go through a regular close, waiting for a timeout before shutting down the subflows. There is no need to keep the kernel resources in use for such a possibly long time: short-circuit to fast-close. Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/protocol.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index a676ac1bb9f1..1926b81a9538 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2953,10 +2953,13 @@ bool __mptcp_close(struct sock *sk, long timeout) goto cleanup; } - if (mptcp_check_readable(msk)) { - /* the msk has read data, do the MPTCP equivalent of TCP reset */ + if (mptcp_check_readable(msk) || timeout < 0) { + /* If the msk has read data, or the caller explicitly ask it, + * do the MPTCP equivalent of TCP reset, aka MPTCP fastclose + */ inet_sk_state_store(sk, TCP_CLOSE); mptcp_do_fastclose(sk); + timeout = 0; } else if (mptcp_close_state(sk)) { __mptcp_wr_shutdown(sk); } -- cgit v1.2.3 From aa5887dca2d236fc50000e27023d4d78dce3af30 Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Fri, 14 Apr 2023 17:47:06 +0200 Subject: mptcp: make userspace_pm_append_new_local_addr static mptcp_userspace_pm_append_new_local_addr() has always exclusively been used in pm_userspace.c since its introduction in commit 4638de5aefe5 ("mptcp: handle local addrs announced by userspace PMs"). So make it static. Signed-off-by: Geliang Tang Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/pm_userspace.c | 4 ++-- net/mptcp/protocol.h | 2 -- 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c index a02d3cbf2a1b..27a275805c06 100644 --- a/net/mptcp/pm_userspace.c +++ b/net/mptcp/pm_userspace.c @@ -25,8 +25,8 @@ void mptcp_free_local_addr_list(struct mptcp_sock *msk) } } -int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk, - struct mptcp_pm_addr_entry *entry) +static int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk, + struct mptcp_pm_addr_entry *entry) { DECLARE_BITMAP(id_bitmap, MPTCP_PM_MAX_ADDR_ID + 1); struct mptcp_pm_addr_entry *match = NULL; diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 21eda9cd0c52..d91b852c2405 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -830,8 +830,6 @@ int mptcp_pm_remove_subflow(struct mptcp_sock *msk, const struct mptcp_rm_list * void mptcp_pm_remove_addrs_and_subflows(struct mptcp_sock *msk, struct list_head *rm_list); -int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk, - struct mptcp_pm_addr_entry *entry); void mptcp_free_local_addr_list(struct mptcp_sock *msk); int mptcp_nl_cmd_announce(struct sk_buff *skb, struct genl_info *info); int mptcp_nl_cmd_remove(struct sk_buff *skb, struct genl_info *info); -- cgit v1.2.3 From ce395d0e3ad585839245a1f6a43cc144617a0163 Mon Sep 17 00:00:00 2001 From: Matthieu Baerts Date: Fri, 14 Apr 2023 17:47:08 +0200 Subject: mptcp: remove unused 'remaining' variable In some functions, 'remaining' variable was given in argument and/or set but never read. net/mptcp/options.c:779:3: warning: Value stored to 'remaining' is never read [clang-analyzer-deadcode.DeadStores]. net/mptcp/options.c:547:3: warning: Value stored to 'remaining' is never read [clang-analyzer-deadcode.DeadStores]. The issue has been reported internally by Alibaba CI. Reported-by: Abaci Robot Suggested-by: Mat Martineau Co-developed-by: Jiapeng Chong Signed-off-by: Jiapeng Chong Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/options.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mptcp/options.c b/net/mptcp/options.c index cd3b885c8faa..19a01b6566f1 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -442,7 +442,6 @@ static void clear_3rdack_retransmission(struct sock *sk) static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb, bool snd_data_fin_enable, unsigned int *size, - unsigned int remaining, struct mptcp_out_options *opts) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); @@ -556,7 +555,6 @@ static void mptcp_write_data_fin(struct mptcp_subflow_context *subflow, static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, bool snd_data_fin_enable, unsigned int *size, - unsigned int remaining, struct mptcp_out_options *opts) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); @@ -580,7 +578,6 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, opts->ext_copy = *mpext; } - remaining -= map_size; dss_size = map_size; if (skb && snd_data_fin_enable) mptcp_write_data_fin(subflow, skb, &opts->ext_copy); @@ -851,9 +848,9 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb, } snd_data_fin = mptcp_data_fin_enabled(msk); - if (mptcp_established_options_mp(sk, skb, snd_data_fin, &opt_size, remaining, opts)) + if (mptcp_established_options_mp(sk, skb, snd_data_fin, &opt_size, opts)) ret = true; - else if (mptcp_established_options_dss(sk, skb, snd_data_fin, &opt_size, remaining, opts)) { + else if (mptcp_established_options_dss(sk, skb, snd_data_fin, &opt_size, opts)) { unsigned int mp_fail_size; ret = true; -- cgit v1.2.3 From bd4b28189469492df2b962d737842c311ce2659c Mon Sep 17 00:00:00 2001 From: Xin Long Date: Fri, 14 Apr 2023 17:21:15 -0400 Subject: sctp: delete the obsolete code for the host name address param In the latest RFC9260, the Host Name Address param has been deprecated. For INIT chunk: Note 3: An INIT chunk MUST NOT contain the Host Name Address parameter. The receiver of an INIT chunk containing a Host Name Address parameter MUST send an ABORT chunk and MAY include an "Unresolvable Address" error cause. For Supported Address Types: The value indicating the Host Name Address parameter MUST NOT be used when sending this parameter and MUST be ignored when receiving this parameter. Currently Linux SCTP doesn't really support Host Name Address param, but only saves some flag and print debug info, which actually won't even be triggered due to the verification in sctp_verify_param(). This patch is to delete those dead code. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 1 - net/sctp/sm_make_chunk.c | 10 +--------- net/sctp/socket.c | 2 +- 3 files changed, 2 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index a0933efd93c3..070c9458fff4 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1711,7 +1711,6 @@ struct sctp_association { __u16 ecn_capable:1, /* Can peer do ECN? */ ipv4_address:1, /* Peer understands IPv4 addresses? */ ipv6_address:1, /* Peer understands IPv6 addresses? */ - hostname_address:1, /* Peer understands DNS addresses? */ asconf_capable:1, /* Does peer support ADDIP? */ prsctp_capable:1, /* Can peer do PR-SCTP? */ reconf_capable:1, /* Can peer do RE-CONFIG? */ diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index c7503fd64915..c8f4ec5d5f98 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -2207,7 +2207,7 @@ static enum sctp_ierror sctp_verify_param(struct net *net, break; case SCTP_PARAM_HOST_NAME_ADDRESS: - /* Tell the peer, we won't support this param. */ + /* This param has been Deprecated, send ABORT. */ sctp_process_hn_param(asoc, param, chunk, err_chunk); retval = SCTP_IERROR_ABORT; break; @@ -2589,10 +2589,6 @@ do_addr_param: asoc->cookie_life = ktime_add_ms(asoc->cookie_life, stale); break; - case SCTP_PARAM_HOST_NAME_ADDRESS: - pr_debug("%s: unimplemented SCTP_HOST_NAME_ADDRESS\n", __func__); - break; - case SCTP_PARAM_SUPPORTED_ADDRESS_TYPES: /* Turn off the default values first so we'll know which * ones are really set by the peer. @@ -2624,10 +2620,6 @@ do_addr_param: asoc->peer.ipv6_address = 1; break; - case SCTP_PARAM_HOST_NAME_ADDRESS: - asoc->peer.hostname_address = 1; - break; - default: /* Just ignore anything else. */ break; } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 218e0982c370..079e726909b4 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5195,7 +5195,7 @@ int sctp_get_sctp_info(struct sock *sk, struct sctp_association *asoc, mask = asoc->peer.ecn_capable << 1; mask = (mask | asoc->peer.ipv4_address) << 1; mask = (mask | asoc->peer.ipv6_address) << 1; - mask = (mask | asoc->peer.hostname_address) << 1; + mask = mask << 1; mask = (mask | asoc->peer.asconf_capable) << 1; mask = (mask | asoc->peer.prsctp_capable) << 1; mask = (mask | asoc->peer.auth_capable); -- cgit v1.2.3 From ab4f1e28c941b2554456bfe091db0caf25c33044 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Fri, 14 Apr 2023 17:21:16 -0400 Subject: sctp: add intl_capable and reconf_capable in ss peer_capable There are two new peer capables have been added since sctp_diag was introduced into SCTP. When dumping the peer capables, these two new peer capables should also be included. To not break the old capables, reconf_capable takes the old hostname_address bit, and intl_capable uses the higher available bit in sctpi_peer_capable. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- net/sctp/socket.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 079e726909b4..cda8c2874691 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5192,10 +5192,11 @@ int sctp_get_sctp_info(struct sock *sk, struct sctp_association *asoc, info->sctpi_peer_rwnd = asoc->peer.rwnd; info->sctpi_peer_tag = asoc->c.peer_vtag; - mask = asoc->peer.ecn_capable << 1; + mask = asoc->peer.intl_capable << 1; + mask = (mask | asoc->peer.ecn_capable) << 1; mask = (mask | asoc->peer.ipv4_address) << 1; mask = (mask | asoc->peer.ipv6_address) << 1; - mask = mask << 1; + mask = (mask | asoc->peer.reconf_capable) << 1; mask = (mask | asoc->peer.asconf_capable) << 1; mask = (mask | asoc->peer.prsctp_capable) << 1; mask = (mask | asoc->peer.auth_capable); -- cgit v1.2.3 From 338469d677e5d426f5ada88761f16f6d2c7c1981 Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Sat, 15 Apr 2023 12:33:09 -0300 Subject: net/sched: clear actions pointer in miss cookie init fail Palash reports a UAF when using a modified version of syzkaller[1]. When 'tcf_exts_miss_cookie_base_alloc()' fails in 'tcf_exts_init_ex()' a call to 'tcf_exts_destroy()' is made to free up the tcf_exts resources. In flower, a call to '__fl_put()' when 'tcf_exts_init_ex()' fails is made; Then calling 'tcf_exts_destroy()', which triggers an UAF since the already freed tcf_exts action pointer is lingering in the struct. Before the offending patch, this was not an issue since there was no case where the tcf_exts action pointer could linger. Therefore, restore the old semantic by clearing the action pointer in case of a failure to initialize the miss_cookie. [1] https://github.com/cmu-pasta/linux-kernel-enriched-corpus v1->v2: Fix compilation on configs without tc actions (kernel test robot) Fixes: 80cd22c35c90 ("net/sched: cls_api: Support hardware miss to tc action") Reported-by: Palash Oswal Acked-by: Jamal Hadi Salim Signed-off-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/cls_api.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 2a6b6be0811b..35785a36c802 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -3235,6 +3235,9 @@ int tcf_exts_init_ex(struct tcf_exts *exts, struct net *net, int action, err_miss_alloc: tcf_exts_destroy(exts); +#ifdef CONFIG_NET_CLS_ACT + exts->actions = NULL; +#endif return err; } EXPORT_SYMBOL(tcf_exts_init_ex); -- cgit v1.2.3 From d51425190805d47aecc1910b272e65476dd3b937 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Sun, 16 Apr 2023 13:05:06 -0400 Subject: SUNRPC: Fix failures of checksum Kunit tests Scott reports that when the new GSS krb5 Kunit tests are built as a separate module and loaded, the RFC 6803 and RFC 8009 checksum tests all fail, even though they pass when run under kunit.py. It appears that passing a buffer backed by static const memory to gss_krb5_checksum() is a problem. A printk in checksum_case() shows the correct plaintext, but by the time the buffer has been converted to a scatterlist and arrives at checksummer(), it contains all zeroes. Replacing this buffer with one that is dynamically allocated fixes the issue. Reported-by: Scott Mayhew Fixes: 02142b2ca8fc ("SUNRPC: Add checksum KUnit tests for the RFC 6803 encryption types") Tested-by: Scott Mayhew Signed-off-by: Chuck Lever --- net/sunrpc/auth_gss/gss_krb5_test.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sunrpc/auth_gss/gss_krb5_test.c b/net/sunrpc/auth_gss/gss_krb5_test.c index aa6ec4e858aa..95ca783795c5 100644 --- a/net/sunrpc/auth_gss/gss_krb5_test.c +++ b/net/sunrpc/auth_gss/gss_krb5_test.c @@ -73,7 +73,6 @@ static void checksum_case(struct kunit *test) { const struct gss_krb5_test_param *param = test->param_value; struct xdr_buf buf = { - .head[0].iov_base = param->plaintext->data, .head[0].iov_len = param->plaintext->len, .len = param->plaintext->len, }; @@ -99,6 +98,10 @@ static void checksum_case(struct kunit *test) err = crypto_ahash_setkey(tfm, Kc.data, Kc.len); KUNIT_ASSERT_EQ(test, err, 0); + buf.head[0].iov_base = kunit_kzalloc(test, buf.head[0].iov_len, GFP_KERNEL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, buf.head[0].iov_base); + memcpy(buf.head[0].iov_base, param->plaintext->data, buf.head[0].iov_len); + checksum.len = gk5e->cksumlength; checksum.data = kunit_kzalloc(test, checksum.len, GFP_KERNEL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, checksum.data); -- cgit v1.2.3 From 59e498a3289f685261c076b998a8a2f8a516874f Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Mon, 17 Apr 2023 15:49:15 +0200 Subject: bpf: Set skb redirect and from_ingress info in __bpf_tx_skb MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There are some use-cases where it is desirable to use bpf_redirect() in combination with ifb device, which currently is not supported, for example, around filtering inbound traffic with BPF to then push it to ifb which holds the qdisc for shaping in contrast to doing that on the egress device. Toke mentions the following case related to OpenWrt: Because there's not always a single egress on the other side. These are mainly home routers, which tend to have one or more WiFi devices bridged to one or more ethernet ports on the LAN side, and a single upstream WAN port. And the objective is to control the total amount of traffic going over the WAN link (in both directions), to deal with bufferbloat in the ISP network (which is sadly still all too prevalent). In this setup, the traffic can be split arbitrarily between the links on the LAN side, and the only "single bottleneck" is the WAN link. So we install both egress and ingress shapers on this, configured to something like 95-98% of the true link bandwidth, thus moving the queues into the qdisc layer in the router. It's usually necessary to set the ingress bandwidth shaper a bit lower than the egress due to being "downstream" of the bottleneck link, but it does work surprisingly well. We usually use something like a matchall filter to put all ingress traffic on the ifb, so doing the redirect from BPF has not been an immediate requirement thus far. However, it does seem a bit odd that this is not possible, and we do have a BPF-based filter that layers on top of this kind of setup, which currently uses u32 as the ingress filter and so it could presumably be improved to use BPF instead if that was available. Reported-by: Toke Høiland-Jørgensen Reported-by: Yafang Shao Reported-by: Tonghao Zhang Signed-off-by: Daniel Borkmann Acked-by: Yafang Shao Acked-by: Toke Høiland-Jørgensen Link: https://git.openwrt.org/?p=project/qosify.git;a=blob;f=README Link: https://lore.kernel.org/bpf/875y9yzbuy.fsf@toke.dk Link: https://lore.kernel.org/r/8cebc8b2b6e967e10cbafe2ffd6795050e74accd.1681739137.git.daniel@iogearbox.net Signed-off-by: Alexei Starovoitov --- include/linux/skbuff.h | 9 +++++++++ net/core/filter.c | 1 + 2 files changed, 10 insertions(+) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 494a23a976b0..9ff2e3d57329 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -5041,6 +5041,15 @@ static inline void skb_reset_redirect(struct sk_buff *skb) skb->redirected = 0; } +static inline void skb_set_redirected_noclear(struct sk_buff *skb, + bool from_ingress) +{ + skb->redirected = 1; +#ifdef CONFIG_NET_REDIRECT + skb->from_ingress = from_ingress; +#endif +} + static inline bool skb_csum_is_sctp(struct sk_buff *skb) { return skb->csum_not_inet; diff --git a/net/core/filter.c b/net/core/filter.c index df0df59814ae..44fb997434ad 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2122,6 +2122,7 @@ static inline int __bpf_tx_skb(struct net_device *dev, struct sk_buff *skb) } skb->dev = dev; + skb_set_redirected_noclear(skb, skb_at_tc_ingress(skb)); skb_clear_tstamp(skb); dev_xmit_recursion_inc(); -- cgit v1.2.3 From d46fc894147cf98dd6e8210aa99ed46854191840 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 17 Apr 2023 12:14:29 +0200 Subject: netfilter: nf_tables: validate catch-all set elements catch-all set element might jump/goto to chain that uses expressions that require validation. Fixes: aaa31047a6d2 ("netfilter: nftables: add catch-all set element support") Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 4 +++ net/netfilter/nf_tables_api.c | 64 +++++++++++++++++++++++++++++++++++---- net/netfilter/nft_lookup.c | 36 +++------------------- 3 files changed, 66 insertions(+), 38 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 9430128aae99..1b8e305bb54a 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -1085,6 +1085,10 @@ struct nft_chain { }; int nft_chain_validate(const struct nft_ctx *ctx, const struct nft_chain *chain); +int nft_setelem_validate(const struct nft_ctx *ctx, struct nft_set *set, + const struct nft_set_iter *iter, + struct nft_set_elem *elem); +int nft_set_catchall_validate(const struct nft_ctx *ctx, struct nft_set *set); enum nft_chain_types { NFT_CHAIN_T_DEFAULT = 0, diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index cd52109e674a..98043e83af71 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -3447,6 +3447,64 @@ static int nft_table_validate(struct net *net, const struct nft_table *table) return 0; } +int nft_setelem_validate(const struct nft_ctx *ctx, struct nft_set *set, + const struct nft_set_iter *iter, + struct nft_set_elem *elem) +{ + const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv); + struct nft_ctx *pctx = (struct nft_ctx *)ctx; + const struct nft_data *data; + int err; + + if (nft_set_ext_exists(ext, NFT_SET_EXT_FLAGS) && + *nft_set_ext_flags(ext) & NFT_SET_ELEM_INTERVAL_END) + return 0; + + data = nft_set_ext_data(ext); + switch (data->verdict.code) { + case NFT_JUMP: + case NFT_GOTO: + pctx->level++; + err = nft_chain_validate(ctx, data->verdict.chain); + if (err < 0) + return err; + pctx->level--; + break; + default: + break; + } + + return 0; +} + +struct nft_set_elem_catchall { + struct list_head list; + struct rcu_head rcu; + void *elem; +}; + +int nft_set_catchall_validate(const struct nft_ctx *ctx, struct nft_set *set) +{ + u8 genmask = nft_genmask_next(ctx->net); + struct nft_set_elem_catchall *catchall; + struct nft_set_elem elem; + struct nft_set_ext *ext; + int ret = 0; + + list_for_each_entry_rcu(catchall, &set->catchall_list, list) { + ext = nft_set_elem_ext(set, catchall->elem); + if (!nft_set_elem_active(ext, genmask)) + continue; + + elem.priv = catchall->elem; + ret = nft_setelem_validate(ctx, set, NULL, &elem); + if (ret < 0) + return ret; + } + + return ret; +} + static struct nft_rule *nft_rule_lookup_byid(const struct net *net, const struct nft_chain *chain, const struct nlattr *nla); @@ -4759,12 +4817,6 @@ err_set_name: return err; } -struct nft_set_elem_catchall { - struct list_head list; - struct rcu_head rcu; - void *elem; -}; - static void nft_set_catchall_destroy(const struct nft_ctx *ctx, struct nft_set *set) { diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index cae5a6724163..cecf8ab90e58 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -199,37 +199,6 @@ nla_put_failure: return -1; } -static int nft_lookup_validate_setelem(const struct nft_ctx *ctx, - struct nft_set *set, - const struct nft_set_iter *iter, - struct nft_set_elem *elem) -{ - const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv); - struct nft_ctx *pctx = (struct nft_ctx *)ctx; - const struct nft_data *data; - int err; - - if (nft_set_ext_exists(ext, NFT_SET_EXT_FLAGS) && - *nft_set_ext_flags(ext) & NFT_SET_ELEM_INTERVAL_END) - return 0; - - data = nft_set_ext_data(ext); - switch (data->verdict.code) { - case NFT_JUMP: - case NFT_GOTO: - pctx->level++; - err = nft_chain_validate(ctx, data->verdict.chain); - if (err < 0) - return err; - pctx->level--; - break; - default: - break; - } - - return 0; -} - static int nft_lookup_validate(const struct nft_ctx *ctx, const struct nft_expr *expr, const struct nft_data **d) @@ -245,9 +214,12 @@ static int nft_lookup_validate(const struct nft_ctx *ctx, iter.skip = 0; iter.count = 0; iter.err = 0; - iter.fn = nft_lookup_validate_setelem; + iter.fn = nft_setelem_validate; priv->set->ops->walk(ctx, priv->set, &iter); + if (!iter.err) + iter.err = nft_set_catchall_validate(ctx, priv->set); + if (iter.err < 0) return iter.err; -- cgit v1.2.3 From d4eb7e39929a3b1ff30fb751b4859fc2410702a0 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 17 Apr 2023 17:50:28 +0200 Subject: netfilter: nf_tables: tighten netlink attribute requirements for catch-all elements If NFT_SET_ELEM_CATCHALL is set on, then userspace provides no set element key. Otherwise, bail out with -EINVAL. Fixes: aaa31047a6d2 ("netfilter: nftables: add catch-all set element support") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 98043e83af71..e48ab8dfb541 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -6108,7 +6108,8 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set, if (err < 0) return err; - if (!nla[NFTA_SET_ELEM_KEY] && !(flags & NFT_SET_ELEM_CATCHALL)) + if (((flags & NFT_SET_ELEM_CATCHALL) && nla[NFTA_SET_ELEM_KEY]) || + (!(flags & NFT_SET_ELEM_CATCHALL) && !nla[NFTA_SET_ELEM_KEY])) return -EINVAL; if (flags != 0) { -- cgit v1.2.3 From 9ae708f00161e1d789268fa9cc05bf6bec2af474 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Mon, 17 Apr 2023 15:37:51 +0200 Subject: wifi: mac80211: remove ieee80211_tx_status_8023 It is unused and should not be used. In order to avoid limitations in 4-address mode, the driver should always use ieee80211_tx_status_ext for 802.3 frames with a valid sta pointer. Signed-off-by: Felix Fietkau Link: https://lore.kernel.org/r/20230417133751.79160-1-nbd@nbd.name Signed-off-by: Johannes Berg --- include/net/mac80211.h | 20 -------------------- net/mac80211/status.c | 24 ------------------------ 2 files changed, 44 deletions(-) (limited to 'net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index a8dadbd83d95..ac0370e76874 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -5215,26 +5215,6 @@ static inline void ieee80211_tx_status_ni(struct ieee80211_hw *hw, void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw, struct sk_buff *skb); -/** - * ieee80211_tx_status_8023 - transmit status callback for 802.3 frame format - * - * Call this function for all transmitted data frames after their transmit - * completion. This callback should only be called for data frames which - * are using driver's (or hardware's) offload capability of encap/decap - * 802.11 frames. - * - * This function may not be called in IRQ context. Calls to this function - * for a single hardware must be synchronized against each other and all - * calls in the same tx status family. - * - * @hw: the hardware the frame was transmitted by - * @vif: the interface for which the frame was transmitted - * @skb: the frame that was transmitted, owned by mac80211 after this call - */ -void ieee80211_tx_status_8023(struct ieee80211_hw *hw, - struct ieee80211_vif *vif, - struct sk_buff *skb); - /** * ieee80211_report_low_ack - report non-responding station * diff --git a/net/mac80211/status.c b/net/mac80211/status.c index 3f9ddd7f04b6..2b13a52ce96c 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -1244,30 +1244,6 @@ void ieee80211_tx_rate_update(struct ieee80211_hw *hw, } EXPORT_SYMBOL(ieee80211_tx_rate_update); -void ieee80211_tx_status_8023(struct ieee80211_hw *hw, - struct ieee80211_vif *vif, - struct sk_buff *skb) -{ - struct ieee80211_sub_if_data *sdata; - struct ieee80211_tx_status status = { - .skb = skb, - .info = IEEE80211_SKB_CB(skb), - }; - struct sta_info *sta; - - sdata = vif_to_sdata(vif); - - rcu_read_lock(); - - if (!ieee80211_lookup_ra_sta(sdata, skb, &sta) && !IS_ERR(sta)) - status.sta = &sta->sta; - - ieee80211_tx_status_ext(hw, &status); - - rcu_read_unlock(); -} -EXPORT_SYMBOL(ieee80211_tx_status_8023); - void ieee80211_report_low_ack(struct ieee80211_sta *pubsta, u32 num_packets) { struct sta_info *sta = container_of(pubsta, struct sta_info, sta); -- cgit v1.2.3 From 48380368dec14859723b9e3fbd43e042638d9a76 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 29 Mar 2023 12:14:42 +0200 Subject: Change DEFINE_SEMAPHORE() to take a number argument Fundamentally semaphores are a counted primitive, but DEFINE_SEMAPHORE() does not expose this and explicitly creates a binary semaphore. Change DEFINE_SEMAPHORE() to take a number argument and use that in the few places that open-coded it using __SEMAPHORE_INITIALIZER(). Signed-off-by: Peter Zijlstra (Intel) [mcgrof: add some tribal knowledge about why some folks prefer binary sempahores over mutexes] Reviewed-by: Sergey Senozhatsky Reviewed-by: Davidlohr Bueso Signed-off-by: Luis Chamberlain --- arch/mips/cavium-octeon/setup.c | 2 +- arch/x86/kernel/cpu/intel.c | 2 +- drivers/firmware/efi/runtime-wrappers.c | 2 +- drivers/firmware/efi/vars.c | 2 +- drivers/macintosh/adb.c | 2 +- drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 2 +- drivers/platform/x86/intel/ifs/sysfs.c | 2 +- drivers/scsi/esas2r/esas2r_ioctl.c | 2 +- drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c | 2 +- include/linux/semaphore.h | 10 ++++++++-- kernel/printk/printk.c | 2 +- net/rxrpc/call_object.c | 6 ++---- 12 files changed, 20 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index a71727f7a608..c5561016f577 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -72,7 +72,7 @@ extern void pci_console_init(const char *arg); static unsigned long long max_memory = ULLONG_MAX; static unsigned long long reserve_low_mem; -DEFINE_SEMAPHORE(octeon_bootbus_sem); +DEFINE_SEMAPHORE(octeon_bootbus_sem, 1); EXPORT_SYMBOL(octeon_bootbus_sem); static struct octeon_boot_descriptor *octeon_boot_desc_ptr; diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 291d4167fab8..12bad63822f0 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -1177,7 +1177,7 @@ static const struct { static struct ratelimit_state bld_ratelimit; static unsigned int sysctl_sld_mitigate = 1; -static DEFINE_SEMAPHORE(buslock_sem); +static DEFINE_SEMAPHORE(buslock_sem, 1); #ifdef CONFIG_PROC_SYSCTL static struct ctl_table sld_sysctls[] = { diff --git a/drivers/firmware/efi/runtime-wrappers.c b/drivers/firmware/efi/runtime-wrappers.c index 1fba4e09cdcf..a400c4312c82 100644 --- a/drivers/firmware/efi/runtime-wrappers.c +++ b/drivers/firmware/efi/runtime-wrappers.c @@ -158,7 +158,7 @@ void efi_call_virt_check_flags(unsigned long flags, const char *call) * none of the remaining functions are actually ever called at runtime. * So let's just use a single lock to serialize all Runtime Services calls. */ -static DEFINE_SEMAPHORE(efi_runtime_lock); +static DEFINE_SEMAPHORE(efi_runtime_lock, 1); /* * Expose the EFI runtime lock to the UV platform diff --git a/drivers/firmware/efi/vars.c b/drivers/firmware/efi/vars.c index bd75b87f5fc1..bfc5fa6aa47b 100644 --- a/drivers/firmware/efi/vars.c +++ b/drivers/firmware/efi/vars.c @@ -21,7 +21,7 @@ /* Private pointer to registered efivars */ static struct efivars *__efivars; -static DEFINE_SEMAPHORE(efivars_lock); +static DEFINE_SEMAPHORE(efivars_lock, 1); static efi_status_t check_var_size(bool nonblocking, u32 attributes, unsigned long size) diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c index 23bd0c77ac1a..56599515d51a 100644 --- a/drivers/macintosh/adb.c +++ b/drivers/macintosh/adb.c @@ -80,7 +80,7 @@ static struct adb_driver *adb_controller; BLOCKING_NOTIFIER_HEAD(adb_client_list); static int adb_got_sleep; static int adb_inited; -static DEFINE_SEMAPHORE(adb_probe_mutex); +static DEFINE_SEMAPHORE(adb_probe_mutex, 1); static int sleepy_trackpad; static int autopoll_devs; int __adb_probe_sync; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 5d1e4fe335aa..5a105bab4387 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -298,7 +298,7 @@ const u32 dmae_reg_go_c[] = { /* Global resources for unloading a previously loaded device */ #define BNX2X_PREV_WAIT_NEEDED 1 -static DEFINE_SEMAPHORE(bnx2x_prev_sem); +static DEFINE_SEMAPHORE(bnx2x_prev_sem, 1); static LIST_HEAD(bnx2x_prev_list); /* Forward declaration */ diff --git a/drivers/platform/x86/intel/ifs/sysfs.c b/drivers/platform/x86/intel/ifs/sysfs.c index ee636a76b083..4c3c642ee19a 100644 --- a/drivers/platform/x86/intel/ifs/sysfs.c +++ b/drivers/platform/x86/intel/ifs/sysfs.c @@ -13,7 +13,7 @@ * Protects against simultaneous tests on multiple cores, or * reloading can file while a test is in progress */ -static DEFINE_SEMAPHORE(ifs_sem); +static DEFINE_SEMAPHORE(ifs_sem, 1); /* * The sysfs interface to check additional details of last test diff --git a/drivers/scsi/esas2r/esas2r_ioctl.c b/drivers/scsi/esas2r/esas2r_ioctl.c index e003d923acbf..055d2e87a2c8 100644 --- a/drivers/scsi/esas2r/esas2r_ioctl.c +++ b/drivers/scsi/esas2r/esas2r_ioctl.c @@ -56,7 +56,7 @@ dma_addr_t esas2r_buffered_ioctl_addr; u32 esas2r_buffered_ioctl_size; struct pci_dev *esas2r_buffered_ioctl_pcid; -static DEFINE_SEMAPHORE(buffered_ioctl_semaphore); +static DEFINE_SEMAPHORE(buffered_ioctl_semaphore, 1); typedef int (*BUFFERED_IOCTL_CALLBACK)(struct esas2r_adapter *, struct esas2r_request *, struct esas2r_sg_context *, diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c index cddcd3c596c9..1a656fdc9445 100644 --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c @@ -149,7 +149,7 @@ static char *g_fragments_base; static char *g_free_fragments; static struct semaphore g_free_fragments_sema; -static DEFINE_SEMAPHORE(g_free_fragments_mutex); +static DEFINE_SEMAPHORE(g_free_fragments_mutex, 1); static int vchiq_blocking_bulk_transfer(struct vchiq_instance *instance, unsigned int handle, void *data, diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h index 6694d0019a68..04655faadc2d 100644 --- a/include/linux/semaphore.h +++ b/include/linux/semaphore.h @@ -25,8 +25,14 @@ struct semaphore { .wait_list = LIST_HEAD_INIT((name).wait_list), \ } -#define DEFINE_SEMAPHORE(name) \ - struct semaphore name = __SEMAPHORE_INITIALIZER(name, 1) +/* + * Unlike mutexes, binary semaphores do not have an owner, so up() can + * be called in a different thread from the one which called down(). + * It is also safe to call down_trylock() and up() from interrupt + * context. + */ +#define DEFINE_SEMAPHORE(_name, _n) \ + struct semaphore _name = __SEMAPHORE_INITIALIZER(_name, _n) static inline void sema_init(struct semaphore *sem, int val) { diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index fd0c9f913940..76987aaa5a45 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -89,7 +89,7 @@ static DEFINE_MUTEX(console_mutex); * console_sem protects updates to console->seq and console_suspended, * and also provides serialization for console printing. */ -static DEFINE_SEMAPHORE(console_sem); +static DEFINE_SEMAPHORE(console_sem, 1); HLIST_HEAD(console_list); EXPORT_SYMBOL_GPL(console_list); DEFINE_STATIC_SRCU(console_srcu); diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index e9f1f49d18c2..3e5cc70884dd 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -40,10 +40,8 @@ const char *const rxrpc_call_completions[NR__RXRPC_CALL_COMPLETIONS] = { struct kmem_cache *rxrpc_call_jar; -static struct semaphore rxrpc_call_limiter = - __SEMAPHORE_INITIALIZER(rxrpc_call_limiter, 1000); -static struct semaphore rxrpc_kernel_call_limiter = - __SEMAPHORE_INITIALIZER(rxrpc_kernel_call_limiter, 1000); +static DEFINE_SEMAPHORE(rxrpc_call_limiter, 1000); +static DEFINE_SEMAPHORE(rxrpc_kernel_call_limiter, 1000); void rxrpc_poke_call(struct rxrpc_call *call, enum rxrpc_call_poke_trace what) { -- cgit v1.2.3 From 4e006c7a6dac0ead4c1bf606000aa90a372fc253 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 17 Apr 2023 09:00:52 -0400 Subject: net: rpl: fix rpl header size calculation This patch fixes a missing 8 byte for the header size calculation. The ipv6_rpl_srh_size() is used to check a skb_pull() on skb->data which points to skb_transport_header(). Currently we only check on the calculated addresses fields using CmprI and CmprE fields, see: https://www.rfc-editor.org/rfc/rfc6554#section-3 there is however a missing 8 byte inside the calculation which stands for the fields before the addresses field. Those 8 bytes are represented by sizeof(struct ipv6_rpl_sr_hdr) expression. Fixes: 8610c7c6e3bd ("net: ipv6: add support for rpl sr exthdr") Signed-off-by: Alexander Aring Reported-by: maxpl0it Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/rpl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/rpl.c b/net/ipv6/rpl.c index 488aec9e1a74..d1876f192225 100644 --- a/net/ipv6/rpl.c +++ b/net/ipv6/rpl.c @@ -32,7 +32,8 @@ static void *ipv6_rpl_segdata_pos(const struct ipv6_rpl_sr_hdr *hdr, int i) size_t ipv6_rpl_srh_size(unsigned char n, unsigned char cmpri, unsigned char cmpre) { - return (n * IPV6_PFXTAIL_LEN(cmpri)) + IPV6_PFXTAIL_LEN(cmpre); + return sizeof(struct ipv6_rpl_sr_hdr) + (n * IPV6_PFXTAIL_LEN(cmpri)) + + IPV6_PFXTAIL_LEN(cmpre); } void ipv6_rpl_srh_decompress(struct ipv6_rpl_sr_hdr *outhdr, -- cgit v1.2.3 From 2cdaa3eefed83082923cf219c8b6a314e622da74 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 18 Apr 2023 23:31:26 +0200 Subject: netfilter: conntrack: restore IPS_CONFIRMED out of nf_conntrack_hash_check_insert() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit e6d57e9ff0ae ("netfilter: conntrack: fix rmmod double-free race") consolidates IPS_CONFIRMED bit set in nf_conntrack_hash_check_insert(). However, this breaks ctnetlink: # conntrack -I -p tcp --timeout 123 --src 1.2.3.4 --dst 5.6.7.8 --state ESTABLISHED --sport 1 --dport 4 -u SEEN_REPLY conntrack v1.4.6 (conntrack-tools): Operation failed: Device or resource busy This is a partial revert of the aforementioned commit to restore IPS_CONFIRMED. Fixes: e6d57e9ff0ae ("netfilter: conntrack: fix rmmod double-free race") Reported-by: Stéphane Graber Tested-by: Stéphane Graber Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_bpf.c | 1 + net/netfilter/nf_conntrack_core.c | 1 - net/netfilter/nf_conntrack_netlink.c | 3 +++ 3 files changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c index cd99e6dc1f35..34913521c385 100644 --- a/net/netfilter/nf_conntrack_bpf.c +++ b/net/netfilter/nf_conntrack_bpf.c @@ -381,6 +381,7 @@ __bpf_kfunc struct nf_conn *bpf_ct_insert_entry(struct nf_conn___init *nfct_i) struct nf_conn *nfct = (struct nf_conn *)nfct_i; int err; + nfct->status |= IPS_CONFIRMED; err = nf_conntrack_hash_check_insert(nfct); if (err < 0) { nf_conntrack_free(nfct); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index c6a6a6099b4e..7ba6ab9b54b5 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -932,7 +932,6 @@ nf_conntrack_hash_check_insert(struct nf_conn *ct) goto out; } - ct->status |= IPS_CONFIRMED; smp_wmb(); /* The caller holds a reference to this object */ refcount_set(&ct->ct_general.use, 2); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index bfc3aaa2c872..d3ee18854698 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -2316,6 +2316,9 @@ ctnetlink_create_conntrack(struct net *net, nfct_seqadj_ext_add(ct); nfct_synproxy_ext_add(ct); + /* we must add conntrack extensions before confirmation. */ + ct->status |= IPS_CONFIRMED; + if (cda[CTA_STATUS]) { err = ctnetlink_change_status(ct, cda); if (err < 0) -- cgit v1.2.3 From 2a6a870e44dd88f1a6a2893c65ef756a9edfb4c7 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Mon, 17 Apr 2023 16:00:40 +0200 Subject: mptcp: stops worker on unaccepted sockets at listener close This is a partial revert of the blamed commit, with a relevant change: mptcp_subflow_queue_clean() now just change the msk socket status and stop the worker, so that the UaF issue addressed by the blamed commit is not re-introduced. The above prevents the mptcp worker from running concurrently with inet_csk_listen_stop(), as such race would trigger a warning, as reported by Christoph: RSP: 002b:00007f784fe09cd8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e WARNING: CPU: 0 PID: 25807 at net/ipv4/inet_connection_sock.c:1387 inet_csk_listen_stop+0x664/0x870 net/ipv4/inet_connection_sock.c:1387 RAX: ffffffffffffffda RBX: 00000000006bc050 RCX: 00007f7850afd6a9 RDX: 0000000000000000 RSI: 0000000020000340 RDI: 0000000000000004 Modules linked in: RBP: 0000000000000002 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 00000000006bc05c R13: fffffffffffffea8 R14: 00000000006bc050 R15: 000000000001fe40 CPU: 0 PID: 25807 Comm: syz-executor.7 Not tainted 6.2.0-g778e54711659 #7 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014 RIP: 0010:inet_csk_listen_stop+0x664/0x870 net/ipv4/inet_connection_sock.c:1387 RAX: 0000000000000000 RBX: ffff888100dfbd40 RCX: 0000000000000000 RDX: ffff8881363aab80 RSI: ffffffff81c494f4 RDI: 0000000000000005 RBP: ffff888126dad080 R08: 0000000000000005 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000000 R12: ffff888100dfe040 R13: 0000000000000001 R14: 0000000000000000 R15: ffff888100dfbdd8 FS: 00007f7850a2c800(0000) GS:ffff88813bc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b32d26000 CR3: 000000012fdd8006 CR4: 0000000000770ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: __tcp_close+0x5b2/0x620 net/ipv4/tcp.c:2875 __mptcp_close_ssk+0x145/0x3d0 net/mptcp/protocol.c:2427 mptcp_destroy_common+0x8a/0x1c0 net/mptcp/protocol.c:3277 mptcp_destroy+0x41/0x60 net/mptcp/protocol.c:3304 __mptcp_destroy_sock+0x56/0x140 net/mptcp/protocol.c:2965 __mptcp_close+0x38f/0x4a0 net/mptcp/protocol.c:3057 mptcp_close+0x24/0xe0 net/mptcp/protocol.c:3072 inet_release+0x53/0xa0 net/ipv4/af_inet.c:429 __sock_release+0x4e/0xf0 net/socket.c:651 sock_close+0x15/0x20 net/socket.c:1393 __fput+0xff/0x420 fs/file_table.c:321 task_work_run+0x8b/0xe0 kernel/task_work.c:179 resume_user_mode_work include/linux/resume_user_mode.h:49 [inline] exit_to_user_mode_loop kernel/entry/common.c:171 [inline] exit_to_user_mode_prepare+0x113/0x120 kernel/entry/common.c:203 __syscall_exit_to_user_mode_work kernel/entry/common.c:285 [inline] syscall_exit_to_user_mode+0x1d/0x40 kernel/entry/common.c:296 do_syscall_64+0x46/0x90 arch/x86/entry/common.c:86 entry_SYSCALL_64_after_hwframe+0x72/0xdc RIP: 0033:0x7f7850af70dc RAX: 0000000000000000 RBX: 0000000000000004 RCX: 00007f7850af70dc RDX: 00007f7850a2c800 RSI: 0000000000000002 RDI: 0000000000000003 RBP: 00000000006bd980 R08: 0000000000000000 R09: 00000000000018a0 R10: 00000000316338a4 R11: 0000000000000293 R12: 0000000000211e31 R13: 00000000006bc05c R14: 00007f785062c000 R15: 0000000000211af0 Fixes: 0a3f4f1f9c27 ("mptcp: fix UaF in listener shutdown") Cc: stable@vger.kernel.org Reported-by: Christoph Paasch Link: https://github.com/multipath-tcp/mptcp_net-next/issues/371 Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/protocol.c | 6 ++++- net/mptcp/protocol.h | 1 + net/mptcp/subflow.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 78 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 06c5872e3b00..5181fb91595b 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2365,8 +2365,12 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, mptcp_subflow_drop_ctx(ssk); } else { /* otherwise tcp will dispose of the ssk and subflow ctx */ - if (ssk->sk_state == TCP_LISTEN) + if (ssk->sk_state == TCP_LISTEN) { + tcp_set_state(ssk, TCP_CLOSE); + mptcp_subflow_queue_clean(sk, ssk); + inet_csk_listen_stop(ssk); mptcp_event_pm_listener(ssk, MPTCP_EVENT_LISTENER_CLOSED); + } __tcp_close(ssk, 0); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 339a6f072989..3a2db1b862dd 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -629,6 +629,7 @@ void mptcp_close_ssk(struct sock *sk, struct sock *ssk, struct mptcp_subflow_context *subflow); void __mptcp_subflow_send_ack(struct sock *ssk); void mptcp_subflow_reset(struct sock *ssk); +void mptcp_subflow_queue_clean(struct sock *sk, struct sock *ssk); void mptcp_sock_graft(struct sock *sk, struct socket *parent); struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk); bool __mptcp_close(struct sock *sk, long timeout); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index d34588850545..bf5e5c72b5ee 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -1819,6 +1819,78 @@ static void subflow_state_change(struct sock *sk) } } +void mptcp_subflow_queue_clean(struct sock *listener_sk, struct sock *listener_ssk) +{ + struct request_sock_queue *queue = &inet_csk(listener_ssk)->icsk_accept_queue; + struct mptcp_sock *msk, *next, *head = NULL; + struct request_sock *req; + + /* build a list of all unaccepted mptcp sockets */ + spin_lock_bh(&queue->rskq_lock); + for (req = queue->rskq_accept_head; req; req = req->dl_next) { + struct mptcp_subflow_context *subflow; + struct sock *ssk = req->sk; + + if (!sk_is_mptcp(ssk)) + continue; + + subflow = mptcp_subflow_ctx(ssk); + if (!subflow || !subflow->conn) + continue; + + /* skip if already in list */ + msk = mptcp_sk(subflow->conn); + if (msk->dl_next || msk == head) + continue; + + sock_hold(subflow->conn); + msk->dl_next = head; + head = msk; + } + spin_unlock_bh(&queue->rskq_lock); + if (!head) + return; + + /* can't acquire the msk socket lock under the subflow one, + * or will cause ABBA deadlock + */ + release_sock(listener_ssk); + + for (msk = head; msk; msk = next) { + struct sock *sk = (struct sock *)msk; + + lock_sock_nested(sk, SINGLE_DEPTH_NESTING); + next = msk->dl_next; + msk->dl_next = NULL; + + /* prevent the stack from later re-schedule the worker for + * this socket + */ + inet_sk_state_store(sk, TCP_CLOSE); + release_sock(sk); + + /* lockdep will report a false positive ABBA deadlock + * between cancel_work_sync and the listener socket. + * The involved locks belong to different sockets WRT + * the existing AB chain. + * Using a per socket key is problematic as key + * deregistration requires process context and must be + * performed at socket disposal time, in atomic + * context. + * Just tell lockdep to consider the listener socket + * released here. + */ + mutex_release(&listener_sk->sk_lock.dep_map, _RET_IP_); + mptcp_cancel_work(sk); + mutex_acquire(&listener_sk->sk_lock.dep_map, 0, 0, _RET_IP_); + + sock_put(sk); + } + + /* we are still under the listener msk socket lock */ + lock_sock_nested(listener_ssk, SINGLE_DEPTH_NESTING); +} + static int subflow_ulp_init(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); -- cgit v1.2.3 From 63740448a32eb662e05894425b47bcc5814136f4 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Mon, 17 Apr 2023 16:00:41 +0200 Subject: mptcp: fix accept vs worker race The mptcp worker and mptcp_accept() can race, as reported by Christoph: refcount_t: addition on 0; use-after-free. WARNING: CPU: 1 PID: 14351 at lib/refcount.c:25 refcount_warn_saturate+0x105/0x1b0 lib/refcount.c:25 Modules linked in: CPU: 1 PID: 14351 Comm: syz-executor.2 Not tainted 6.3.0-rc1-gde5e8fd0123c #11 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014 RIP: 0010:refcount_warn_saturate+0x105/0x1b0 lib/refcount.c:25 Code: 02 31 ff 89 de e8 1b f0 a7 ff 84 db 0f 85 6e ff ff ff e8 3e f5 a7 ff 48 c7 c7 d8 c7 34 83 c6 05 6d 2d 0f 02 01 e8 cb 3d 90 ff <0f> 0b e9 4f ff ff ff e8 1f f5 a7 ff 0f b6 1d 54 2d 0f 02 31 ff 89 RSP: 0018:ffffc90000a47bf8 EFLAGS: 00010282 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: ffff88802eae98c0 RSI: ffffffff81097d4f RDI: 0000000000000001 RBP: ffff88802e712180 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000001 R11: ffff88802eaea148 R12: ffff88802e712100 R13: ffff88802e712a88 R14: ffff888005cb93a8 R15: ffff88802e712a88 FS: 0000000000000000(0000) GS:ffff88803ed00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f277fd89120 CR3: 0000000035486002 CR4: 0000000000370ee0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: __refcount_add include/linux/refcount.h:199 [inline] __refcount_inc include/linux/refcount.h:250 [inline] refcount_inc include/linux/refcount.h:267 [inline] sock_hold include/net/sock.h:775 [inline] __mptcp_close+0x4c6/0x4d0 net/mptcp/protocol.c:3051 mptcp_close+0x24/0xe0 net/mptcp/protocol.c:3072 inet_release+0x56/0xa0 net/ipv4/af_inet.c:429 __sock_release+0x51/0xf0 net/socket.c:653 sock_close+0x18/0x20 net/socket.c:1395 __fput+0x113/0x430 fs/file_table.c:321 task_work_run+0x96/0x100 kernel/task_work.c:179 exit_task_work include/linux/task_work.h:38 [inline] do_exit+0x4fc/0x10c0 kernel/exit.c:869 do_group_exit+0x51/0xf0 kernel/exit.c:1019 get_signal+0x12b0/0x1390 kernel/signal.c:2859 arch_do_signal_or_restart+0x25/0x260 arch/x86/kernel/signal.c:306 exit_to_user_mode_loop kernel/entry/common.c:168 [inline] exit_to_user_mode_prepare+0x131/0x1a0 kernel/entry/common.c:203 __syscall_exit_to_user_mode_work kernel/entry/common.c:285 [inline] syscall_exit_to_user_mode+0x19/0x40 kernel/entry/common.c:296 do_syscall_64+0x46/0x90 arch/x86/entry/common.c:86 entry_SYSCALL_64_after_hwframe+0x72/0xdc RIP: 0033:0x7fec4b4926a9 Code: Unable to access opcode bytes at 0x7fec4b49267f. RSP: 002b:00007fec49f9dd78 EFLAGS: 00000246 ORIG_RAX: 00000000000000ca RAX: fffffffffffffe00 RBX: 00000000006bc058 RCX: 00007fec4b4926a9 RDX: 0000000000000000 RSI: 0000000000000080 RDI: 00000000006bc058 RBP: 00000000006bc050 R08: 00000000007df998 R09: 00000000007df998 R10: 0000000000000000 R11: 0000000000000246 R12: 00000000006bc05c R13: fffffffffffffea8 R14: 000000000000000b R15: 000000000001fe40 The root cause is that the worker can force fallback to TCP the first mptcp subflow, actually deleting the unaccepted msk socket. We can explicitly prevent the race delaying the unaccepted msk deletion at listener shutdown time. In case the closed subflow is later accepted, just drop the mptcp context and let the user-space deal with the paired mptcp socket. Fixes: b6985b9b8295 ("mptcp: use the workqueue to destroy unaccepted sockets") Cc: stable@vger.kernel.org Reported-by: Christoph Paasch Link: https://github.com/multipath-tcp/mptcp_net-next/issues/375 Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Tested-by: Christoph Paasch Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller --- net/mptcp/protocol.c | 68 ++++++++++++++++++++++++++++++++++------------------ net/mptcp/protocol.h | 1 + net/mptcp/subflow.c | 22 +++++++++-------- 3 files changed, 58 insertions(+), 33 deletions(-) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 5181fb91595b..b998e9df53ce 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2315,7 +2315,26 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, unsigned int flags) { struct mptcp_sock *msk = mptcp_sk(sk); - bool need_push, dispose_it; + bool dispose_it, need_push = false; + + /* If the first subflow moved to a close state before accept, e.g. due + * to an incoming reset, mptcp either: + * - if either the subflow or the msk are dead, destroy the context + * (the subflow socket is deleted by inet_child_forget) and the msk + * - otherwise do nothing at the moment and take action at accept and/or + * listener shutdown - user-space must be able to accept() the closed + * socket. + */ + if (msk->in_accept_queue && msk->first == ssk) { + if (!sock_flag(sk, SOCK_DEAD) && !sock_flag(ssk, SOCK_DEAD)) + return; + + /* ensure later check in mptcp_worker() will dispose the msk */ + sock_set_flag(sk, SOCK_DEAD); + lock_sock_nested(ssk, SINGLE_DEPTH_NESTING); + mptcp_subflow_drop_ctx(ssk); + goto out_release; + } dispose_it = !msk->subflow || ssk != msk->subflow->sk; if (dispose_it) @@ -2351,18 +2370,6 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, if (!inet_csk(ssk)->icsk_ulp_ops) { WARN_ON_ONCE(!sock_flag(ssk, SOCK_DEAD)); kfree_rcu(subflow, rcu); - } else if (msk->in_accept_queue && msk->first == ssk) { - /* if the first subflow moved to a close state, e.g. due to - * incoming reset and we reach here before inet_child_forget() - * the TCP stack could later try to close it via - * inet_csk_listen_stop(), or deliver it to the user space via - * accept(). - * We can't delete the subflow - or risk a double free - nor let - * the msk survive - or will be leaked in the non accept scenario: - * fallback and let TCP cope with the subflow cleanup. - */ - WARN_ON_ONCE(sock_flag(ssk, SOCK_DEAD)); - mptcp_subflow_drop_ctx(ssk); } else { /* otherwise tcp will dispose of the ssk and subflow ctx */ if (ssk->sk_state == TCP_LISTEN) { @@ -2377,6 +2384,8 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, /* close acquired an extra ref */ __sock_put(ssk); } + +out_release: release_sock(ssk); sock_put(ssk); @@ -2431,21 +2440,14 @@ static void __mptcp_close_subflow(struct sock *sk) mptcp_close_ssk(sk, ssk, subflow); } - /* if the MPC subflow has been closed before the msk is accepted, - * msk will never be accept-ed, close it now - */ - if (!msk->first && msk->in_accept_queue) { - sock_set_flag(sk, SOCK_DEAD); - inet_sk_state_store(sk, TCP_CLOSE); - } } -static bool mptcp_check_close_timeout(const struct sock *sk) +static bool mptcp_should_close(const struct sock *sk) { s32 delta = tcp_jiffies32 - inet_csk(sk)->icsk_mtup.probe_timestamp; struct mptcp_subflow_context *subflow; - if (delta >= TCP_TIMEWAIT_LEN) + if (delta >= TCP_TIMEWAIT_LEN || mptcp_sk(sk)->in_accept_queue) return true; /* if all subflows are in closed status don't bother with additional @@ -2653,7 +2655,7 @@ static void mptcp_worker(struct work_struct *work) * even if it is orphaned and in FIN_WAIT2 state */ if (sock_flag(sk, SOCK_DEAD)) { - if (mptcp_check_close_timeout(sk)) { + if (mptcp_should_close(sk)) { inet_sk_state_store(sk, TCP_CLOSE); mptcp_do_fastclose(sk); } @@ -2899,6 +2901,14 @@ static void __mptcp_destroy_sock(struct sock *sk) sock_put(sk); } +void __mptcp_unaccepted_force_close(struct sock *sk) +{ + sock_set_flag(sk, SOCK_DEAD); + inet_sk_state_store(sk, TCP_CLOSE); + mptcp_do_fastclose(sk); + __mptcp_destroy_sock(sk); +} + static __poll_t mptcp_check_readable(struct mptcp_sock *msk) { /* Concurrent splices from sk_receive_queue into receive_queue will @@ -3737,6 +3747,18 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, if (!ssk->sk_socket) mptcp_sock_graft(ssk, newsock); } + + /* Do late cleanup for the first subflow as necessary. Also + * deal with bad peers not doing a complete shutdown. + */ + if (msk->first && + unlikely(inet_sk_state_load(msk->first) == TCP_CLOSE)) { + __mptcp_close_ssk(newsk, msk->first, + mptcp_subflow_ctx(msk->first), 0); + if (unlikely(list_empty(&msk->conn_list))) + inet_sk_state_store(newsk, TCP_CLOSE); + } + release_sock(newsk); } diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 3a2db1b862dd..d6469b6ab38e 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -634,6 +634,7 @@ void mptcp_sock_graft(struct sock *sk, struct socket *parent); struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk); bool __mptcp_close(struct sock *sk, long timeout); void mptcp_cancel_work(struct sock *sk); +void __mptcp_unaccepted_force_close(struct sock *sk); void mptcp_set_owner_r(struct sk_buff *skb, struct sock *sk); bool mptcp_addresses_equal(const struct mptcp_addr_info *a, diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index bf5e5c72b5ee..281c1cc8dc8d 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -723,9 +723,12 @@ void mptcp_subflow_drop_ctx(struct sock *ssk) if (!ctx) return; - subflow_ulp_fallback(ssk, ctx); - if (ctx->conn) - sock_put(ctx->conn); + list_del(&mptcp_subflow_ctx(ssk)->node); + if (inet_csk(ssk)->icsk_ulp_ops) { + subflow_ulp_fallback(ssk, ctx); + if (ctx->conn) + sock_put(ctx->conn); + } kfree_rcu(ctx, rcu); } @@ -1824,6 +1827,7 @@ void mptcp_subflow_queue_clean(struct sock *listener_sk, struct sock *listener_s struct request_sock_queue *queue = &inet_csk(listener_ssk)->icsk_accept_queue; struct mptcp_sock *msk, *next, *head = NULL; struct request_sock *req; + struct sock *sk; /* build a list of all unaccepted mptcp sockets */ spin_lock_bh(&queue->rskq_lock); @@ -1839,11 +1843,12 @@ void mptcp_subflow_queue_clean(struct sock *listener_sk, struct sock *listener_s continue; /* skip if already in list */ - msk = mptcp_sk(subflow->conn); + sk = subflow->conn; + msk = mptcp_sk(sk); if (msk->dl_next || msk == head) continue; - sock_hold(subflow->conn); + sock_hold(sk); msk->dl_next = head; head = msk; } @@ -1857,16 +1862,13 @@ void mptcp_subflow_queue_clean(struct sock *listener_sk, struct sock *listener_s release_sock(listener_ssk); for (msk = head; msk; msk = next) { - struct sock *sk = (struct sock *)msk; + sk = (struct sock *)msk; lock_sock_nested(sk, SINGLE_DEPTH_NESTING); next = msk->dl_next; msk->dl_next = NULL; - /* prevent the stack from later re-schedule the worker for - * this socket - */ - inet_sk_state_store(sk, TCP_CLOSE); + __mptcp_unaccepted_force_close(sk); release_sock(sk); /* lockdep will report a false positive ABBA deadlock -- cgit v1.2.3 From 73db1b8f2bb6725b7391e85aab41fdf592b3c0c1 Mon Sep 17 00:00:00 2001 From: Tzung-Bi Shih Date: Wed, 19 Apr 2023 13:15:26 +0800 Subject: netfilter: conntrack: fix wrong ct->timeout value (struct nf_conn)->timeout is an interval before the conntrack confirmed. After confirmed, it becomes a timestamp. It is observed that timeout of an unconfirmed conntrack: - Set by calling ctnetlink_change_timeout(). As a result, `nfct_time_stamp` was wrongly added to `ct->timeout` twice. - Get by calling ctnetlink_dump_timeout(). As a result, `nfct_time_stamp` was wrongly subtracted. Call Trace: dump_stack_lvl ctnetlink_dump_timeout __ctnetlink_glue_build ctnetlink_glue_build __nfqnl_enqueue_packet nf_queue nf_hook_slow ip_mc_output ? __pfx_ip_finish_output ip_send_skb ? __pfx_dst_output udp_send_skb udp_sendmsg ? __pfx_ip_generic_getfrag sock_sendmsg Separate the 2 cases in: - Setting `ct->timeout` in __nf_ct_set_timeout(). - Getting `ct->timeout` in ctnetlink_dump_timeout(). Pablo appends: Update ctnetlink to set up the timeout _after_ the IPS_CONFIRMED flag is set on, otherwise conntrack creation via ctnetlink breaks. Note that the problem described in this patch occurs since the introduction of the nfnetlink_queue conntrack support, select a sufficiently old Fixes: tag for -stable kernel to pick up this fix. Fixes: a4b4766c3ceb ("netfilter: nfnetlink_queue: rename related to nfqueue attaching conntrack info") Signed-off-by: Tzung-Bi Shih Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_core.h | 6 +++++- net/netfilter/nf_conntrack_netlink.c | 13 +++++++++---- 2 files changed, 14 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index 71d1269fe4d4..3384859a8921 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -89,7 +89,11 @@ static inline void __nf_ct_set_timeout(struct nf_conn *ct, u64 timeout) { if (timeout > INT_MAX) timeout = INT_MAX; - WRITE_ONCE(ct->timeout, nfct_time_stamp + (u32)timeout); + + if (nf_ct_is_confirmed(ct)) + WRITE_ONCE(ct->timeout, nfct_time_stamp + (u32)timeout); + else + ct->timeout = (u32)timeout; } int __nf_ct_change_timeout(struct nf_conn *ct, u64 cta_timeout); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index d3ee18854698..6f3b23a6653c 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -176,7 +176,12 @@ nla_put_failure: static int ctnetlink_dump_timeout(struct sk_buff *skb, const struct nf_conn *ct, bool skip_zero) { - long timeout = nf_ct_expires(ct) / HZ; + long timeout; + + if (nf_ct_is_confirmed(ct)) + timeout = nf_ct_expires(ct) / HZ; + else + timeout = ct->timeout / HZ; if (skip_zero && timeout == 0) return 0; @@ -2253,9 +2258,6 @@ ctnetlink_create_conntrack(struct net *net, if (!cda[CTA_TIMEOUT]) goto err1; - timeout = (u64)ntohl(nla_get_be32(cda[CTA_TIMEOUT])) * HZ; - __nf_ct_set_timeout(ct, timeout); - rcu_read_lock(); if (cda[CTA_HELP]) { char *helpname = NULL; @@ -2319,6 +2321,9 @@ ctnetlink_create_conntrack(struct net *net, /* we must add conntrack extensions before confirmation. */ ct->status |= IPS_CONFIRMED; + timeout = (u64)ntohl(nla_get_be32(cda[CTA_TIMEOUT])) * HZ; + __nf_ct_set_timeout(ct, timeout); + if (cda[CTA_STATUS]) { err = ctnetlink_change_status(ct, cda); if (err < 0) -- cgit v1.2.3 From eb6fba7555a812c07aa984fb9e8e9b151a65ca16 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 17 Apr 2023 08:53:46 -0700 Subject: net: skbuff: hide wifi_acked when CONFIG_WIRELESS not set Datacenter kernel builds will very likely not include WIRELESS, so let them shave 2 bits off the skb by hiding the wifi fields. Reviewed-by: Florian Fainelli Reviewed-by: Eric Dumazet Signed-off-by: Jakub Kicinski Acked-by: Johannes Berg Signed-off-by: David S. Miller --- include/linux/skbuff.h | 11 +++++++++++ include/net/sock.h | 2 +- net/core/skbuff.c | 2 ++ net/socket.c | 2 ++ 4 files changed, 16 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index a823ec3aa326..513f03b23a73 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -953,8 +953,10 @@ struct sk_buff { __u8 l4_hash:1; __u8 sw_hash:1; +#ifdef CONFIG_WIRELESS __u8 wifi_acked_valid:1; __u8 wifi_acked:1; +#endif __u8 no_fcs:1; /* Indicates the inner headers are valid in the skbuff. */ __u8 encapsulation:1; @@ -1187,6 +1189,15 @@ static inline unsigned int skb_napi_id(const struct sk_buff *skb) #endif } +static inline bool skb_wifi_acked_valid(const struct sk_buff *skb) +{ +#ifdef CONFIG_WIRELESS + return skb->wifi_acked_valid; +#else + return 0; +#endif +} + /** * skb_unref - decrement the skb's reference count * @skb: buffer diff --git a/include/net/sock.h b/include/net/sock.h index 5edf0038867c..8b7ed7167243 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2697,7 +2697,7 @@ sock_recv_timestamp(struct msghdr *msg, struct sock *sk, struct sk_buff *skb) else sock_write_timestamp(sk, kt); - if (sock_flag(sk, SOCK_WIFI_STATUS) && skb->wifi_acked_valid) + if (sock_flag(sk, SOCK_WIFI_STATUS) && skb_wifi_acked_valid(skb)) __sock_recv_wifi_status(msg, sk, skb); } diff --git a/net/core/skbuff.c b/net/core/skbuff.c index ef81452759be..768f9d04911f 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -5189,6 +5189,7 @@ void skb_tstamp_tx(struct sk_buff *orig_skb, } EXPORT_SYMBOL_GPL(skb_tstamp_tx); +#ifdef CONFIG_WIRELESS void skb_complete_wifi_ack(struct sk_buff *skb, bool acked) { struct sock *sk = skb->sk; @@ -5214,6 +5215,7 @@ void skb_complete_wifi_ack(struct sk_buff *skb, bool acked) kfree_skb(skb); } EXPORT_SYMBOL_GPL(skb_complete_wifi_ack); +#endif /* CONFIG_WIRELESS */ /** * skb_partial_csum_set - set up and verify partial csum values for packet diff --git a/net/socket.c b/net/socket.c index 73e493da4589..a7b4b37d86df 100644 --- a/net/socket.c +++ b/net/socket.c @@ -957,6 +957,7 @@ void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk, } EXPORT_SYMBOL_GPL(__sock_recv_timestamp); +#ifdef CONFIG_WIRELESS void __sock_recv_wifi_status(struct msghdr *msg, struct sock *sk, struct sk_buff *skb) { @@ -972,6 +973,7 @@ void __sock_recv_wifi_status(struct msghdr *msg, struct sock *sk, put_cmsg(msg, SOL_SOCKET, SCM_WIFI_STATUS, sizeof(ack), &ack); } EXPORT_SYMBOL_GPL(__sock_recv_wifi_status); +#endif static inline void sock_recv_drops(struct msghdr *msg, struct sock *sk, struct sk_buff *skb) -- cgit v1.2.3 From c24831a13ba2e472f874483525084da2f2feb890 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 17 Apr 2023 08:53:47 -0700 Subject: net: skbuff: hide csum_not_inet when CONFIG_IP_SCTP not set SCTP is not universally deployed, allow hiding its bit from the skb. Reviewed-by: Florian Fainelli Reviewed-by: Eric Dumazet Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- include/linux/skbuff.h | 14 ++++++++++++++ net/core/dev.c | 3 +-- net/sched/act_csum.c | 3 +-- 3 files changed, 16 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 513f03b23a73..98d6b48f4dcf 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -983,7 +983,9 @@ struct sk_buff { __u8 decrypted:1; #endif __u8 slow_gro:1; +#if IS_ENABLED(CONFIG_IP_SCTP) __u8 csum_not_inet:1; +#endif #ifdef CONFIG_NET_SCHED __u16 tc_index; /* traffic control index */ @@ -5060,7 +5062,19 @@ static inline void skb_reset_redirect(struct sk_buff *skb) static inline bool skb_csum_is_sctp(struct sk_buff *skb) { +#if IS_ENABLED(CONFIG_IP_SCTP) return skb->csum_not_inet; +#else + return 0; +#endif +} + +static inline void skb_reset_csum_not_inet(struct sk_buff *skb) +{ + skb->ip_summed = CHECKSUM_NONE; +#if IS_ENABLED(CONFIG_IP_SCTP) + skb->csum_not_inet = 0; +#endif } static inline void skb_set_kcov_handle(struct sk_buff *skb, diff --git a/net/core/dev.c b/net/core/dev.c index 8aea68275172..3fc4dba71f9d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3315,8 +3315,7 @@ int skb_crc32c_csum_help(struct sk_buff *skb) skb->len - start, ~(__u32)0, crc32c_csum_stub)); *(__le32 *)(skb->data + offset) = crc32c_csum; - skb->ip_summed = CHECKSUM_NONE; - skb->csum_not_inet = 0; + skb_reset_csum_not_inet(skb); out: return ret; } diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c index 95e9304024b7..8ed285023a40 100644 --- a/net/sched/act_csum.c +++ b/net/sched/act_csum.c @@ -376,8 +376,7 @@ static int tcf_csum_sctp(struct sk_buff *skb, unsigned int ihl, sctph->checksum = sctp_compute_cksum(skb, skb_network_offset(skb) + ihl); - skb->ip_summed = CHECKSUM_NONE; - skb->csum_not_inet = 0; + skb_reset_csum_not_inet(skb); return 1; } -- cgit v1.2.3 From 8e4c62c7d980eaf0f64c1c0ef0c80f5685af0fb6 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 17 Apr 2023 08:28:05 -0700 Subject: page_pool: add DMA_ATTR_WEAK_ORDERING on all mappings Commit c519fe9a4f0d ("bnxt: add dma mapping attributes") added DMA_ATTR_WEAK_ORDERING to DMA attrs on bnxt. It has since spread to a few more drivers (possibly as a copy'n'paste). DMA_ATTR_WEAK_ORDERING only seems to matter on Sparc and PowerPC/cell, the rarity of these platforms is likely why we never bothered adding the attribute in the page pool, even though it should be safe to add. To make the page pool migration in drivers which set this flag less of a risk (of regressing the precious sparc database workloads or whatever needed this) let's add DMA_ATTR_WEAK_ORDERING on all page pool DMA mappings. We could make this a driver opt-in but frankly I don't think it's worth complicating the API. I can't think of a reason why device accesses to packet memory would have to be ordered. Acked-by: Ilias Apalodimas Acked-by: Somnath Kotur Reviewed-by: Alexander Lobakin Acked-by: Jesper Dangaard Brouer Link: https://lore.kernel.org/r/20230417152805.331865-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/core/page_pool.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 2f6bf422ed30..97f20f7ff4fc 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -316,7 +316,8 @@ static bool page_pool_dma_map(struct page_pool *pool, struct page *page) */ dma = dma_map_page_attrs(pool->p.dev, page, 0, (PAGE_SIZE << pool->p.order), - pool->p.dma_dir, DMA_ATTR_SKIP_CPU_SYNC); + pool->p.dma_dir, DMA_ATTR_SKIP_CPU_SYNC | + DMA_ATTR_WEAK_ORDERING); if (dma_mapping_error(pool->p.dev, dma)) return false; @@ -484,7 +485,7 @@ void page_pool_release_page(struct page_pool *pool, struct page *page) /* When page is unmapped, it cannot be returned to our pool */ dma_unmap_page_attrs(pool->p.dev, dma, PAGE_SIZE << pool->p.order, pool->p.dma_dir, - DMA_ATTR_SKIP_CPU_SYNC); + DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING); page_pool_set_dma_addr(page, 0); skip_dma_unmap: page_pool_clear_pp_info(page); -- cgit v1.2.3 From 691d0b782066a6eeeecbfceb7910a8f6184e6105 Mon Sep 17 00:00:00 2001 From: Dai Ngo Date: Tue, 18 Apr 2023 13:19:02 -0700 Subject: SUNRPC: remove the maximum number of retries in call_bind_status Currently call_bind_status places a hard limit of 3 to the number of retries on EACCES error. This limit was done to prevent NLM unlock requests from being hang forever when the server keeps returning garbage. However this change causes problem for cases when NLM service takes longer than 9 seconds to register with the port mapper after a restart. This patch removes this hard coded limit and let the RPC handles the retry based on the standard hard/soft task semantics. Fixes: 0b760113a3a1 ("NLM: Don't hang forever on NLM unlock requests") Reported-by: Helen Chao Tested-by: Helen Chao Signed-off-by: Dai Ngo Reviewed-by: Jeff Layton Signed-off-by: Anna Schumaker --- include/linux/sunrpc/sched.h | 3 +-- net/sunrpc/clnt.c | 3 --- net/sunrpc/sched.c | 1 - 3 files changed, 1 insertion(+), 6 deletions(-) (limited to 'net') diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index b8ca3ecaf8d7..8ada7dc802d3 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -90,8 +90,7 @@ struct rpc_task { #endif unsigned char tk_priority : 2,/* Task priority */ tk_garb_retry : 2, - tk_cred_retry : 2, - tk_rebind_retry : 2; + tk_cred_retry : 2; }; typedef void (*rpc_action)(struct rpc_task *); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index fd7e1c630493..d2ee56634308 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -2050,9 +2050,6 @@ call_bind_status(struct rpc_task *task) status = -EOPNOTSUPP; break; } - if (task->tk_rebind_retry == 0) - break; - task->tk_rebind_retry--; rpc_delay(task, 3*HZ); goto retry_timeout; case -ENOBUFS: diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index be587a308e05..c8321de341ee 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -817,7 +817,6 @@ rpc_init_task_statistics(struct rpc_task *task) /* Initialize retry counters */ task->tk_garb_retry = 2; task->tk_cred_retry = 2; - task->tk_rebind_retry = 2; /* starting timestamp */ task->tk_start = ktime_get(); -- cgit v1.2.3 From 3b3009ea8abb713b022d94fba95ec270cf6e7eae Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 17 Apr 2023 10:32:26 -0400 Subject: net/handshake: Create a NETLINK service for handling handshake requests When a kernel consumer needs a transport layer security session, it first needs a handshake to negotiate and establish a session. This negotiation can be done in user space via one of the several existing library implementations, or it can be done in the kernel. No in-kernel handshake implementations yet exist. In their absence, we add a netlink service that can: a. Notify a user space daemon that a handshake is needed. b. Once notified, the daemon calls the kernel back via this netlink service to get the handshake parameters, including an open socket on which to establish the session. c. Once the handshake is complete, the daemon reports the session status and other information via a second netlink operation. This operation marks that it is safe for the kernel to use the open socket and the security session established there. The notification service uses a multicast group. Each handshake mechanism (eg, tlshd) adopts its own group number so that the handshake services are completely independent of one another. The kernel can then tell via netlink_has_listeners() whether a handshake service is active and prepared to handle a handshake request. A new netlink operation, ACCEPT, acts like accept(2) in that it instantiates a file descriptor in the user space daemon's fd table. If this operation is successful, the reply carries the fd number, which can be treated as an open and ready file descriptor. While user space is performing the handshake, the kernel keeps its muddy paws off the open socket. A second new netlink operation, DONE, indicates that the user space daemon is finished with the socket and it is safe for the kernel to use again. The operation also indicates whether a session was established successfully. Signed-off-by: Chuck Lever Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/handshake.yaml | 122 +++++++++++ MAINTAINERS | 9 + include/trace/events/handshake.h | 159 ++++++++++++++ include/uapi/linux/handshake.h | 71 ++++++ net/Kconfig | 5 + net/Makefile | 1 + net/handshake/Makefile | 11 + net/handshake/genl.c | 57 +++++ net/handshake/genl.h | 23 ++ net/handshake/handshake.h | 82 +++++++ net/handshake/netlink.c | 312 ++++++++++++++++++++++++++ net/handshake/request.c | 339 +++++++++++++++++++++++++++++ net/handshake/trace.c | 20 ++ 13 files changed, 1211 insertions(+) create mode 100644 Documentation/netlink/specs/handshake.yaml create mode 100644 include/trace/events/handshake.h create mode 100644 include/uapi/linux/handshake.h create mode 100644 net/handshake/Makefile create mode 100644 net/handshake/genl.c create mode 100644 net/handshake/genl.h create mode 100644 net/handshake/handshake.h create mode 100644 net/handshake/netlink.c create mode 100644 net/handshake/request.c create mode 100644 net/handshake/trace.c (limited to 'net') diff --git a/Documentation/netlink/specs/handshake.yaml b/Documentation/netlink/specs/handshake.yaml new file mode 100644 index 000000000000..0333d92b1438 --- /dev/null +++ b/Documentation/netlink/specs/handshake.yaml @@ -0,0 +1,122 @@ +# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) +# +# Author: Chuck Lever +# +# Copyright (c) 2023, Oracle and/or its affiliates. +# + +name: handshake + +protocol: genetlink + +doc: Netlink protocol to request a transport layer security handshake. + +definitions: + - + type: enum + name: handler-class + value-start: 0 + entries: [ none, max ] + - + type: enum + name: msg-type + value-start: 0 + entries: [ unspec, clienthello, serverhello ] + - + type: enum + name: auth + value-start: 0 + entries: [ unspec, unauth, psk, x509 ] + +attribute-sets: + - + name: x509 + attributes: + - + name: cert + type: u32 + - + name: privkey + type: u32 + - + name: accept + attributes: + - + name: sockfd + type: u32 + - + name: handler-class + type: u32 + enum: handler-class + - + name: message-type + type: u32 + enum: msg-type + - + name: timeout + type: u32 + - + name: auth-mode + type: u32 + enum: auth + - + name: peer-identity + type: u32 + multi-attr: true + - + name: certificate + type: nest + nested-attributes: x509 + multi-attr: true + - + name: done + attributes: + - + name: status + type: u32 + - + name: sockfd + type: u32 + - + name: remote-auth + type: u32 + multi-attr: true + +operations: + list: + - + name: ready + doc: Notify handlers that a new handshake request is waiting + notify: accept + - + name: accept + doc: Handler retrieves next queued handshake request + attribute-set: accept + flags: [ admin-perm ] + do: + request: + attributes: + - handler-class + reply: + attributes: + - sockfd + - message-type + - timeout + - auth-mode + - peer-identity + - certificate + - + name: done + doc: Handler reports handshake completion + attribute-set: done + do: + request: + attributes: + - status + - sockfd + - remote-auth + +mcast-groups: + list: + - + name: none diff --git a/MAINTAINERS b/MAINTAINERS index 4fc57dfd5fd0..cdc7748d15b8 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8947,6 +8947,15 @@ Q: http://patchwork.linuxtv.org/project/linux-media/list/ T: git git://linuxtv.org/anttip/media_tree.git F: drivers/media/usb/hackrf/ +HANDSHAKE UPCALL FOR TRANSPORT LAYER SECURITY +M: Chuck Lever +L: kernel-tls-handshake@lists.linux.dev +L: netdev@vger.kernel.org +S: Maintained +F: Documentation/netlink/specs/handshake.yaml +F: include/trace/events/handshake.h +F: net/handshake/ + HANTRO VPU CODEC DRIVER M: Ezequiel Garcia M: Philipp Zabel diff --git a/include/trace/events/handshake.h b/include/trace/events/handshake.h new file mode 100644 index 000000000000..8dadcab5f12a --- /dev/null +++ b/include/trace/events/handshake.h @@ -0,0 +1,159 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM handshake + +#if !defined(_TRACE_HANDSHAKE_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_HANDSHAKE_H + +#include +#include + +DECLARE_EVENT_CLASS(handshake_event_class, + TP_PROTO( + const struct net *net, + const struct handshake_req *req, + const struct sock *sk + ), + TP_ARGS(net, req, sk), + TP_STRUCT__entry( + __field(const void *, req) + __field(const void *, sk) + __field(unsigned int, netns_ino) + ), + TP_fast_assign( + __entry->req = req; + __entry->sk = sk; + __entry->netns_ino = net->ns.inum; + ), + TP_printk("req=%p sk=%p", + __entry->req, __entry->sk + ) +); +#define DEFINE_HANDSHAKE_EVENT(name) \ + DEFINE_EVENT(handshake_event_class, name, \ + TP_PROTO( \ + const struct net *net, \ + const struct handshake_req *req, \ + const struct sock *sk \ + ), \ + TP_ARGS(net, req, sk)) + +DECLARE_EVENT_CLASS(handshake_fd_class, + TP_PROTO( + const struct net *net, + const struct handshake_req *req, + const struct sock *sk, + int fd + ), + TP_ARGS(net, req, sk, fd), + TP_STRUCT__entry( + __field(const void *, req) + __field(const void *, sk) + __field(int, fd) + __field(unsigned int, netns_ino) + ), + TP_fast_assign( + __entry->req = req; + __entry->sk = req->hr_sk; + __entry->fd = fd; + __entry->netns_ino = net->ns.inum; + ), + TP_printk("req=%p sk=%p fd=%d", + __entry->req, __entry->sk, __entry->fd + ) +); +#define DEFINE_HANDSHAKE_FD_EVENT(name) \ + DEFINE_EVENT(handshake_fd_class, name, \ + TP_PROTO( \ + const struct net *net, \ + const struct handshake_req *req, \ + const struct sock *sk, \ + int fd \ + ), \ + TP_ARGS(net, req, sk, fd)) + +DECLARE_EVENT_CLASS(handshake_error_class, + TP_PROTO( + const struct net *net, + const struct handshake_req *req, + const struct sock *sk, + int err + ), + TP_ARGS(net, req, sk, err), + TP_STRUCT__entry( + __field(const void *, req) + __field(const void *, sk) + __field(int, err) + __field(unsigned int, netns_ino) + ), + TP_fast_assign( + __entry->req = req; + __entry->sk = sk; + __entry->err = err; + __entry->netns_ino = net->ns.inum; + ), + TP_printk("req=%p sk=%p err=%d", + __entry->req, __entry->sk, __entry->err + ) +); +#define DEFINE_HANDSHAKE_ERROR(name) \ + DEFINE_EVENT(handshake_error_class, name, \ + TP_PROTO( \ + const struct net *net, \ + const struct handshake_req *req, \ + const struct sock *sk, \ + int err \ + ), \ + TP_ARGS(net, req, sk, err)) + + +/* + * Request lifetime events + */ + +DEFINE_HANDSHAKE_EVENT(handshake_submit); +DEFINE_HANDSHAKE_ERROR(handshake_submit_err); +DEFINE_HANDSHAKE_EVENT(handshake_cancel); +DEFINE_HANDSHAKE_EVENT(handshake_cancel_none); +DEFINE_HANDSHAKE_EVENT(handshake_cancel_busy); +DEFINE_HANDSHAKE_EVENT(handshake_destruct); + + +TRACE_EVENT(handshake_complete, + TP_PROTO( + const struct net *net, + const struct handshake_req *req, + const struct sock *sk, + int status + ), + TP_ARGS(net, req, sk, status), + TP_STRUCT__entry( + __field(const void *, req) + __field(const void *, sk) + __field(int, status) + __field(unsigned int, netns_ino) + ), + TP_fast_assign( + __entry->req = req; + __entry->sk = sk; + __entry->status = status; + __entry->netns_ino = net->ns.inum; + ), + TP_printk("req=%p sk=%p status=%d", + __entry->req, __entry->sk, __entry->status + ) +); + +/* + * Netlink events + */ + +DEFINE_HANDSHAKE_ERROR(handshake_notify_err); +DEFINE_HANDSHAKE_FD_EVENT(handshake_cmd_accept); +DEFINE_HANDSHAKE_ERROR(handshake_cmd_accept_err); +DEFINE_HANDSHAKE_FD_EVENT(handshake_cmd_done); +DEFINE_HANDSHAKE_ERROR(handshake_cmd_done_err); + +#endif /* _TRACE_HANDSHAKE_H */ + +#include diff --git a/include/uapi/linux/handshake.h b/include/uapi/linux/handshake.h new file mode 100644 index 000000000000..7f66ff489b87 --- /dev/null +++ b/include/uapi/linux/handshake.h @@ -0,0 +1,71 @@ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) */ +/* Do not edit directly, auto-generated from: */ +/* Documentation/netlink/specs/handshake.yaml */ +/* YNL-GEN uapi header */ + +#ifndef _UAPI_LINUX_HANDSHAKE_H +#define _UAPI_LINUX_HANDSHAKE_H + +#define HANDSHAKE_FAMILY_NAME "handshake" +#define HANDSHAKE_FAMILY_VERSION 1 + +enum handshake_handler_class { + HANDSHAKE_HANDLER_CLASS_NONE, + HANDSHAKE_HANDLER_CLASS_MAX, +}; + +enum handshake_msg_type { + HANDSHAKE_MSG_TYPE_UNSPEC, + HANDSHAKE_MSG_TYPE_CLIENTHELLO, + HANDSHAKE_MSG_TYPE_SERVERHELLO, +}; + +enum handshake_auth { + HANDSHAKE_AUTH_UNSPEC, + HANDSHAKE_AUTH_UNAUTH, + HANDSHAKE_AUTH_PSK, + HANDSHAKE_AUTH_X509, +}; + +enum { + HANDSHAKE_A_X509_CERT = 1, + HANDSHAKE_A_X509_PRIVKEY, + + __HANDSHAKE_A_X509_MAX, + HANDSHAKE_A_X509_MAX = (__HANDSHAKE_A_X509_MAX - 1) +}; + +enum { + HANDSHAKE_A_ACCEPT_SOCKFD = 1, + HANDSHAKE_A_ACCEPT_HANDLER_CLASS, + HANDSHAKE_A_ACCEPT_MESSAGE_TYPE, + HANDSHAKE_A_ACCEPT_TIMEOUT, + HANDSHAKE_A_ACCEPT_AUTH_MODE, + HANDSHAKE_A_ACCEPT_PEER_IDENTITY, + HANDSHAKE_A_ACCEPT_CERTIFICATE, + + __HANDSHAKE_A_ACCEPT_MAX, + HANDSHAKE_A_ACCEPT_MAX = (__HANDSHAKE_A_ACCEPT_MAX - 1) +}; + +enum { + HANDSHAKE_A_DONE_STATUS = 1, + HANDSHAKE_A_DONE_SOCKFD, + HANDSHAKE_A_DONE_REMOTE_AUTH, + + __HANDSHAKE_A_DONE_MAX, + HANDSHAKE_A_DONE_MAX = (__HANDSHAKE_A_DONE_MAX - 1) +}; + +enum { + HANDSHAKE_CMD_READY = 1, + HANDSHAKE_CMD_ACCEPT, + HANDSHAKE_CMD_DONE, + + __HANDSHAKE_CMD_MAX, + HANDSHAKE_CMD_MAX = (__HANDSHAKE_CMD_MAX - 1) +}; + +#define HANDSHAKE_MCGRP_NONE "none" + +#endif /* _UAPI_LINUX_HANDSHAKE_H */ diff --git a/net/Kconfig b/net/Kconfig index f806722bccf4..4b800706cc76 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -68,6 +68,11 @@ source "net/iucv/Kconfig" source "net/smc/Kconfig" source "net/xdp/Kconfig" +config NET_HANDSHAKE + bool + depends on SUNRPC || NVME_TARGET_TCP || NVME_TCP + default y + config INET bool "TCP/IP networking" help diff --git a/net/Makefile b/net/Makefile index 87592009366f..4c4dc535453d 100644 --- a/net/Makefile +++ b/net/Makefile @@ -79,3 +79,4 @@ obj-$(CONFIG_NET_NCSI) += ncsi/ obj-$(CONFIG_XDP_SOCKETS) += xdp/ obj-$(CONFIG_MPTCP) += mptcp/ obj-$(CONFIG_MCTP) += mctp/ +obj-$(CONFIG_NET_HANDSHAKE) += handshake/ diff --git a/net/handshake/Makefile b/net/handshake/Makefile new file mode 100644 index 000000000000..d38736de45da --- /dev/null +++ b/net/handshake/Makefile @@ -0,0 +1,11 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +# Makefile for the Generic HANDSHAKE service +# +# Author: Chuck Lever +# +# Copyright (c) 2023, Oracle and/or its affiliates. +# + +obj-y += handshake.o +handshake-y := genl.o netlink.o request.o trace.o diff --git a/net/handshake/genl.c b/net/handshake/genl.c new file mode 100644 index 000000000000..652f37d19bd6 --- /dev/null +++ b/net/handshake/genl.c @@ -0,0 +1,57 @@ +// SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) +/* Do not edit directly, auto-generated from: */ +/* Documentation/netlink/specs/handshake.yaml */ +/* YNL-GEN kernel source */ + +#include +#include + +#include "genl.h" + +#include + +/* HANDSHAKE_CMD_ACCEPT - do */ +static const struct nla_policy handshake_accept_nl_policy[HANDSHAKE_A_ACCEPT_HANDLER_CLASS + 1] = { + [HANDSHAKE_A_ACCEPT_HANDLER_CLASS] = NLA_POLICY_MAX(NLA_U32, 1), +}; + +/* HANDSHAKE_CMD_DONE - do */ +static const struct nla_policy handshake_done_nl_policy[HANDSHAKE_A_DONE_REMOTE_AUTH + 1] = { + [HANDSHAKE_A_DONE_STATUS] = { .type = NLA_U32, }, + [HANDSHAKE_A_DONE_SOCKFD] = { .type = NLA_U32, }, + [HANDSHAKE_A_DONE_REMOTE_AUTH] = { .type = NLA_U32, }, +}; + +/* Ops table for handshake */ +static const struct genl_split_ops handshake_nl_ops[] = { + { + .cmd = HANDSHAKE_CMD_ACCEPT, + .doit = handshake_nl_accept_doit, + .policy = handshake_accept_nl_policy, + .maxattr = HANDSHAKE_A_ACCEPT_HANDLER_CLASS, + .flags = GENL_ADMIN_PERM | GENL_CMD_CAP_DO, + }, + { + .cmd = HANDSHAKE_CMD_DONE, + .doit = handshake_nl_done_doit, + .policy = handshake_done_nl_policy, + .maxattr = HANDSHAKE_A_DONE_REMOTE_AUTH, + .flags = GENL_CMD_CAP_DO, + }, +}; + +static const struct genl_multicast_group handshake_nl_mcgrps[] = { + [HANDSHAKE_NLGRP_NONE] = { "none", }, +}; + +struct genl_family handshake_nl_family __ro_after_init = { + .name = HANDSHAKE_FAMILY_NAME, + .version = HANDSHAKE_FAMILY_VERSION, + .netnsok = true, + .parallel_ops = true, + .module = THIS_MODULE, + .split_ops = handshake_nl_ops, + .n_split_ops = ARRAY_SIZE(handshake_nl_ops), + .mcgrps = handshake_nl_mcgrps, + .n_mcgrps = ARRAY_SIZE(handshake_nl_mcgrps), +}; diff --git a/net/handshake/genl.h b/net/handshake/genl.h new file mode 100644 index 000000000000..a1eb7ccccc7f --- /dev/null +++ b/net/handshake/genl.h @@ -0,0 +1,23 @@ +/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) */ +/* Do not edit directly, auto-generated from: */ +/* Documentation/netlink/specs/handshake.yaml */ +/* YNL-GEN kernel header */ + +#ifndef _LINUX_HANDSHAKE_GEN_H +#define _LINUX_HANDSHAKE_GEN_H + +#include +#include + +#include + +int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info); +int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info); + +enum { + HANDSHAKE_NLGRP_NONE, +}; + +extern struct genl_family handshake_nl_family; + +#endif /* _LINUX_HANDSHAKE_GEN_H */ diff --git a/net/handshake/handshake.h b/net/handshake/handshake.h new file mode 100644 index 000000000000..52568dbe24f1 --- /dev/null +++ b/net/handshake/handshake.h @@ -0,0 +1,82 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Generic netlink handshake service + * + * Author: Chuck Lever + * + * Copyright (c) 2023, Oracle and/or its affiliates. + */ + +#ifndef _INTERNAL_HANDSHAKE_H +#define _INTERNAL_HANDSHAKE_H + +/* Per-net namespace context */ +struct handshake_net { + spinlock_t hn_lock; /* protects next 3 fields */ + int hn_pending; + int hn_pending_max; + struct list_head hn_requests; + + unsigned long hn_flags; +}; + +enum hn_flags_bits { + HANDSHAKE_F_NET_DRAINING, +}; + +struct handshake_proto; + +/* One handshake request */ +struct handshake_req { + struct list_head hr_list; + struct rhash_head hr_rhash; + unsigned long hr_flags; + const struct handshake_proto *hr_proto; + struct sock *hr_sk; + void (*hr_odestruct)(struct sock *sk); + + /* Always the last field */ + char hr_priv[]; +}; + +enum hr_flags_bits { + HANDSHAKE_F_REQ_COMPLETED, +}; + +/* Invariants for all handshake requests for one transport layer + * security protocol + */ +struct handshake_proto { + int hp_handler_class; + size_t hp_privsize; + + int (*hp_accept)(struct handshake_req *req, + struct genl_info *info, int fd); + void (*hp_done)(struct handshake_req *req, + unsigned int status, + struct genl_info *info); + void (*hp_destroy)(struct handshake_req *req); +}; + +/* netlink.c */ +int handshake_genl_notify(struct net *net, const struct handshake_proto *proto, + gfp_t flags); +struct nlmsghdr *handshake_genl_put(struct sk_buff *msg, + struct genl_info *info); +struct handshake_net *handshake_pernet(struct net *net); + +/* request.c */ +struct handshake_req *handshake_req_alloc(const struct handshake_proto *proto, + gfp_t flags); +int handshake_req_hash_init(void); +void handshake_req_hash_destroy(void); +void *handshake_req_private(struct handshake_req *req); +struct handshake_req *handshake_req_hash_lookup(struct sock *sk); +struct handshake_req *handshake_req_next(struct handshake_net *hn, int class); +int handshake_req_submit(struct socket *sock, struct handshake_req *req, + gfp_t flags); +void handshake_complete(struct handshake_req *req, unsigned int status, + struct genl_info *info); +bool handshake_req_cancel(struct sock *sk); + +#endif /* _INTERNAL_HANDSHAKE_H */ diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c new file mode 100644 index 000000000000..7264cac04047 --- /dev/null +++ b/net/handshake/netlink.c @@ -0,0 +1,312 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Generic netlink handshake service + * + * Author: Chuck Lever + * + * Copyright (c) 2023, Oracle and/or its affiliates. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include +#include "handshake.h" +#include "genl.h" + +#include + +/** + * handshake_genl_notify - Notify handlers that a request is waiting + * @net: target network namespace + * @proto: handshake protocol + * @flags: memory allocation control flags + * + * Returns zero on success or a negative errno if notification failed. + */ +int handshake_genl_notify(struct net *net, const struct handshake_proto *proto, + gfp_t flags) +{ + struct sk_buff *msg; + void *hdr; + + if (!genl_has_listeners(&handshake_nl_family, net, + proto->hp_handler_class)) + return -ESRCH; + + msg = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + hdr = genlmsg_put(msg, 0, 0, &handshake_nl_family, 0, + HANDSHAKE_CMD_READY); + if (!hdr) + goto out_free; + + if (nla_put_u32(msg, HANDSHAKE_A_ACCEPT_HANDLER_CLASS, + proto->hp_handler_class) < 0) { + genlmsg_cancel(msg, hdr); + goto out_free; + } + + genlmsg_end(msg, hdr); + return genlmsg_multicast_netns(&handshake_nl_family, net, msg, + 0, proto->hp_handler_class, flags); + +out_free: + nlmsg_free(msg); + return -EMSGSIZE; +} + +/** + * handshake_genl_put - Create a generic netlink message header + * @msg: buffer in which to create the header + * @info: generic netlink message context + * + * Returns a ready-to-use header, or NULL. + */ +struct nlmsghdr *handshake_genl_put(struct sk_buff *msg, + struct genl_info *info) +{ + return genlmsg_put(msg, info->snd_portid, info->snd_seq, + &handshake_nl_family, 0, info->genlhdr->cmd); +} +EXPORT_SYMBOL(handshake_genl_put); + +/* + * dup() a kernel socket for use as a user space file descriptor + * in the current process. The kernel socket must have an + * instatiated struct file. + * + * Implicit argument: "current()" + */ +static int handshake_dup(struct socket *sock) +{ + struct file *file; + int newfd; + + if (!sock->file) + return -EBADF; + + file = get_file(sock->file); + newfd = get_unused_fd_flags(O_CLOEXEC); + if (newfd < 0) { + fput(file); + return newfd; + } + + fd_install(newfd, file); + return newfd; +} + +int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = sock_net(skb->sk); + struct handshake_net *hn = handshake_pernet(net); + struct handshake_req *req = NULL; + struct socket *sock; + int class, fd, err; + + err = -EOPNOTSUPP; + if (!hn) + goto out_status; + + err = -EINVAL; + if (GENL_REQ_ATTR_CHECK(info, HANDSHAKE_A_ACCEPT_HANDLER_CLASS)) + goto out_status; + class = nla_get_u32(info->attrs[HANDSHAKE_A_ACCEPT_HANDLER_CLASS]); + + err = -EAGAIN; + req = handshake_req_next(hn, class); + if (!req) + goto out_status; + + sock = req->hr_sk->sk_socket; + fd = handshake_dup(sock); + if (fd < 0) { + err = fd; + goto out_complete; + } + err = req->hr_proto->hp_accept(req, info, fd); + if (err) + goto out_complete; + + trace_handshake_cmd_accept(net, req, req->hr_sk, fd); + return 0; + +out_complete: + handshake_complete(req, -EIO, NULL); + fput(sock->file); +out_status: + trace_handshake_cmd_accept_err(net, req, NULL, err); + return err; +} + +int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = sock_net(skb->sk); + struct socket *sock = NULL; + struct handshake_req *req; + int fd, status, err; + + if (GENL_REQ_ATTR_CHECK(info, HANDSHAKE_A_DONE_SOCKFD)) + return -EINVAL; + fd = nla_get_u32(info->attrs[HANDSHAKE_A_DONE_SOCKFD]); + + err = 0; + sock = sockfd_lookup(fd, &err); + if (err) { + err = -EBADF; + goto out_status; + } + + req = handshake_req_hash_lookup(sock->sk); + if (!req) { + err = -EBUSY; + fput(sock->file); + goto out_status; + } + + trace_handshake_cmd_done(net, req, sock->sk, fd); + + status = -EIO; + if (info->attrs[HANDSHAKE_A_DONE_STATUS]) + status = nla_get_u32(info->attrs[HANDSHAKE_A_DONE_STATUS]); + + handshake_complete(req, status, info); + fput(sock->file); + return 0; + +out_status: + trace_handshake_cmd_done_err(net, req, sock->sk, err); + return err; +} + +static unsigned int handshake_net_id; + +static int __net_init handshake_net_init(struct net *net) +{ + struct handshake_net *hn = net_generic(net, handshake_net_id); + unsigned long tmp; + struct sysinfo si; + + /* + * Arbitrary limit to prevent handshakes that do not make + * progress from clogging up the system. The cap scales up + * with the amount of physical memory on the system. + */ + si_meminfo(&si); + tmp = si.totalram / (25 * si.mem_unit); + hn->hn_pending_max = clamp(tmp, 3UL, 50UL); + + spin_lock_init(&hn->hn_lock); + hn->hn_pending = 0; + hn->hn_flags = 0; + INIT_LIST_HEAD(&hn->hn_requests); + return 0; +} + +static void __net_exit handshake_net_exit(struct net *net) +{ + struct handshake_net *hn = net_generic(net, handshake_net_id); + struct handshake_req *req; + LIST_HEAD(requests); + + /* + * Drain the net's pending list. Requests that have been + * accepted and are in progress will be destroyed when + * the socket is closed. + */ + spin_lock(&hn->hn_lock); + set_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags); + list_splice_init(&requests, &hn->hn_requests); + spin_unlock(&hn->hn_lock); + + while (!list_empty(&requests)) { + req = list_first_entry(&requests, struct handshake_req, hr_list); + list_del(&req->hr_list); + + /* + * Requests on this list have not yet been + * accepted, so they do not have an fd to put. + */ + + handshake_complete(req, -ETIMEDOUT, NULL); + } +} + +static struct pernet_operations __net_initdata handshake_genl_net_ops = { + .init = handshake_net_init, + .exit = handshake_net_exit, + .id = &handshake_net_id, + .size = sizeof(struct handshake_net), +}; + +/** + * handshake_pernet - Get the handshake private per-net structure + * @net: network namespace + * + * Returns a pointer to the net's private per-net structure for the + * handshake module, or NULL if handshake_init() failed. + */ +struct handshake_net *handshake_pernet(struct net *net) +{ + return handshake_net_id ? + net_generic(net, handshake_net_id) : NULL; +} + +static int __init handshake_init(void) +{ + int ret; + + ret = handshake_req_hash_init(); + if (ret) { + pr_warn("handshake: hash initialization failed (%d)\n", ret); + return ret; + } + + ret = genl_register_family(&handshake_nl_family); + if (ret) { + pr_warn("handshake: netlink registration failed (%d)\n", ret); + handshake_req_hash_destroy(); + return ret; + } + + /* + * ORDER: register_pernet_subsys must be done last. + * + * If initialization does not make it past pernet_subsys + * registration, then handshake_net_id will remain 0. That + * shunts the handshake consumer API to return ENOTSUPP + * to prevent it from dereferencing something that hasn't + * been allocated. + */ + ret = register_pernet_subsys(&handshake_genl_net_ops); + if (ret) { + pr_warn("handshake: pernet registration failed (%d)\n", ret); + genl_unregister_family(&handshake_nl_family); + handshake_req_hash_destroy(); + } + + return ret; +} + +static void __exit handshake_exit(void) +{ + unregister_pernet_subsys(&handshake_genl_net_ops); + handshake_net_id = 0; + + handshake_req_hash_destroy(); + genl_unregister_family(&handshake_nl_family); +} + +module_init(handshake_init); +module_exit(handshake_exit); diff --git a/net/handshake/request.c b/net/handshake/request.c new file mode 100644 index 000000000000..d5b2bc6de057 --- /dev/null +++ b/net/handshake/request.c @@ -0,0 +1,339 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Handshake request lifetime events + * + * Author: Chuck Lever + * + * Copyright (c) 2023, Oracle and/or its affiliates. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include +#include "handshake.h" + +#include + +/* + * We need both a handshake_req -> sock mapping, and a sock -> + * handshake_req mapping. Both are one-to-one. + * + * To avoid adding another pointer field to struct sock, net/handshake + * maintains a hash table, indexed by the memory address of @sock, to + * find the struct handshake_req outstanding for that socket. The + * reverse direction uses a simple pointer field in the handshake_req + * struct. + */ + +static struct rhashtable handshake_rhashtbl ____cacheline_aligned_in_smp; + +static const struct rhashtable_params handshake_rhash_params = { + .key_len = sizeof_field(struct handshake_req, hr_sk), + .key_offset = offsetof(struct handshake_req, hr_sk), + .head_offset = offsetof(struct handshake_req, hr_rhash), + .automatic_shrinking = true, +}; + +int handshake_req_hash_init(void) +{ + return rhashtable_init(&handshake_rhashtbl, &handshake_rhash_params); +} + +void handshake_req_hash_destroy(void) +{ + rhashtable_destroy(&handshake_rhashtbl); +} + +struct handshake_req *handshake_req_hash_lookup(struct sock *sk) +{ + return rhashtable_lookup_fast(&handshake_rhashtbl, &sk, + handshake_rhash_params); +} + +static bool handshake_req_hash_add(struct handshake_req *req) +{ + int ret; + + ret = rhashtable_lookup_insert_fast(&handshake_rhashtbl, + &req->hr_rhash, + handshake_rhash_params); + return ret == 0; +} + +static void handshake_req_destroy(struct handshake_req *req) +{ + if (req->hr_proto->hp_destroy) + req->hr_proto->hp_destroy(req); + rhashtable_remove_fast(&handshake_rhashtbl, &req->hr_rhash, + handshake_rhash_params); + kfree(req); +} + +static void handshake_sk_destruct(struct sock *sk) +{ + void (*sk_destruct)(struct sock *sk); + struct handshake_req *req; + + req = handshake_req_hash_lookup(sk); + if (!req) + return; + + trace_handshake_destruct(sock_net(sk), req, sk); + sk_destruct = req->hr_odestruct; + handshake_req_destroy(req); + if (sk_destruct) + sk_destruct(sk); +} + +/** + * handshake_req_alloc - Allocate a handshake request + * @proto: security protocol + * @flags: memory allocation flags + * + * Returns an initialized handshake_req or NULL. + */ +struct handshake_req *handshake_req_alloc(const struct handshake_proto *proto, + gfp_t flags) +{ + struct handshake_req *req; + + if (!proto) + return NULL; + if (proto->hp_handler_class <= HANDSHAKE_HANDLER_CLASS_NONE) + return NULL; + if (proto->hp_handler_class >= HANDSHAKE_HANDLER_CLASS_MAX) + return NULL; + if (!proto->hp_accept || !proto->hp_done) + return NULL; + + req = kzalloc(struct_size(req, hr_priv, proto->hp_privsize), flags); + if (!req) + return NULL; + + INIT_LIST_HEAD(&req->hr_list); + req->hr_proto = proto; + return req; +} +EXPORT_SYMBOL(handshake_req_alloc); + +/** + * handshake_req_private - Get per-handshake private data + * @req: handshake arguments + * + */ +void *handshake_req_private(struct handshake_req *req) +{ + return (void *)&req->hr_priv; +} +EXPORT_SYMBOL(handshake_req_private); + +static bool __add_pending_locked(struct handshake_net *hn, + struct handshake_req *req) +{ + if (WARN_ON_ONCE(!list_empty(&req->hr_list))) + return false; + hn->hn_pending++; + list_add_tail(&req->hr_list, &hn->hn_requests); + return true; +} + +static void __remove_pending_locked(struct handshake_net *hn, + struct handshake_req *req) +{ + hn->hn_pending--; + list_del_init(&req->hr_list); +} + +/* + * Returns %true if the request was found on @net's pending list, + * otherwise %false. + * + * If @req was on a pending list, it has not yet been accepted. + */ +static bool remove_pending(struct handshake_net *hn, struct handshake_req *req) +{ + bool ret = false; + + spin_lock(&hn->hn_lock); + if (!list_empty(&req->hr_list)) { + __remove_pending_locked(hn, req); + ret = true; + } + spin_unlock(&hn->hn_lock); + + return ret; +} + +struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) +{ + struct handshake_req *req, *pos; + + req = NULL; + spin_lock(&hn->hn_lock); + list_for_each_entry(pos, &hn->hn_requests, hr_list) { + if (pos->hr_proto->hp_handler_class != class) + continue; + __remove_pending_locked(hn, pos); + req = pos; + break; + } + spin_unlock(&hn->hn_lock); + + return req; +} + +/** + * handshake_req_submit - Submit a handshake request + * @sock: open socket on which to perform the handshake + * @req: handshake arguments + * @flags: memory allocation flags + * + * Return values: + * %0: Request queued + * %-EINVAL: Invalid argument + * %-EBUSY: A handshake is already under way for this socket + * %-ESRCH: No handshake agent is available + * %-EAGAIN: Too many pending handshake requests + * %-ENOMEM: Failed to allocate memory + * %-EMSGSIZE: Failed to construct notification message + * %-EOPNOTSUPP: Handshake module not initialized + * + * A zero return value from handshake_req_submit() means that + * exactly one subsequent completion callback is guaranteed. + * + * A negative return value from handshake_req_submit() means that + * no completion callback will be done and that @req has been + * destroyed. + */ +int handshake_req_submit(struct socket *sock, struct handshake_req *req, + gfp_t flags) +{ + struct handshake_net *hn; + struct net *net; + int ret; + + if (!sock || !req || !sock->file) { + kfree(req); + return -EINVAL; + } + + req->hr_sk = sock->sk; + if (!req->hr_sk) { + kfree(req); + return -EINVAL; + } + req->hr_odestruct = req->hr_sk->sk_destruct; + req->hr_sk->sk_destruct = handshake_sk_destruct; + + ret = -EOPNOTSUPP; + net = sock_net(req->hr_sk); + hn = handshake_pernet(net); + if (!hn) + goto out_err; + + ret = -EAGAIN; + if (READ_ONCE(hn->hn_pending) >= hn->hn_pending_max) + goto out_err; + + spin_lock(&hn->hn_lock); + ret = -EOPNOTSUPP; + if (test_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags)) + goto out_unlock; + ret = -EBUSY; + if (!handshake_req_hash_add(req)) + goto out_unlock; + if (!__add_pending_locked(hn, req)) + goto out_unlock; + spin_unlock(&hn->hn_lock); + + ret = handshake_genl_notify(net, req->hr_proto, flags); + if (ret) { + trace_handshake_notify_err(net, req, req->hr_sk, ret); + if (remove_pending(hn, req)) + goto out_err; + } + + /* Prevent socket release while a handshake request is pending */ + sock_hold(req->hr_sk); + + trace_handshake_submit(net, req, req->hr_sk); + return 0; + +out_unlock: + spin_unlock(&hn->hn_lock); +out_err: + trace_handshake_submit_err(net, req, req->hr_sk, ret); + handshake_req_destroy(req); + return ret; +} +EXPORT_SYMBOL(handshake_req_submit); + +void handshake_complete(struct handshake_req *req, unsigned int status, + struct genl_info *info) +{ + struct sock *sk = req->hr_sk; + struct net *net = sock_net(sk); + + if (!test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED, &req->hr_flags)) { + trace_handshake_complete(net, req, sk, status); + req->hr_proto->hp_done(req, status, info); + + /* Handshake request is no longer pending */ + sock_put(sk); + } +} + +/** + * handshake_req_cancel - Cancel an in-progress handshake + * @sk: socket on which there is an ongoing handshake + * + * Request cancellation races with request completion. To determine + * who won, callers examine the return value from this function. + * + * Return values: + * %true - Uncompleted handshake request was canceled + * %false - Handshake request already completed or not found + */ +bool handshake_req_cancel(struct sock *sk) +{ + struct handshake_req *req; + struct handshake_net *hn; + struct net *net; + + net = sock_net(sk); + req = handshake_req_hash_lookup(sk); + if (!req) { + trace_handshake_cancel_none(net, req, sk); + return false; + } + + hn = handshake_pernet(net); + if (hn && remove_pending(hn, req)) { + /* Request hadn't been accepted */ + goto out_true; + } + if (test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED, &req->hr_flags)) { + /* Request already completed */ + trace_handshake_cancel_busy(net, req, sk); + return false; + } + +out_true: + trace_handshake_cancel(net, req, sk); + + /* Handshake request is no longer pending */ + sock_put(sk); + return true; +} +EXPORT_SYMBOL(handshake_req_cancel); diff --git a/net/handshake/trace.c b/net/handshake/trace.c new file mode 100644 index 000000000000..1c4d8e27e17a --- /dev/null +++ b/net/handshake/trace.c @@ -0,0 +1,20 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Trace points for transport security layer handshakes. + * + * Author: Chuck Lever + * + * Copyright (c) 2023, Oracle and/or its affiliates. + */ + +#include + +#include +#include +#include + +#include "handshake.h" + +#define CREATE_TRACE_POINTS + +#include -- cgit v1.2.3 From 2fd5532044a89d2403b543520b4902e196f7d165 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 17 Apr 2023 10:32:33 -0400 Subject: net/handshake: Add a kernel API for requesting a TLSv1.3 handshake To enable kernel consumers of TLS to request a TLS handshake, add support to net/handshake/ to request a handshake upcall. This patch also acts as a template for adding handshake upcall support for other kernel transport layer security providers. Signed-off-by: Chuck Lever Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/handshake.yaml | 4 +- Documentation/networking/index.rst | 1 + Documentation/networking/tls-handshake.rst | 217 +++++++++++++++ MAINTAINERS | 2 + include/net/handshake.h | 43 +++ include/uapi/linux/handshake.h | 2 + net/handshake/Makefile | 2 +- net/handshake/genl.c | 3 +- net/handshake/genl.h | 1 + net/handshake/tlshd.c | 417 +++++++++++++++++++++++++++++ 10 files changed, 689 insertions(+), 3 deletions(-) create mode 100644 Documentation/networking/tls-handshake.rst create mode 100644 include/net/handshake.h create mode 100644 net/handshake/tlshd.c (limited to 'net') diff --git a/Documentation/netlink/specs/handshake.yaml b/Documentation/netlink/specs/handshake.yaml index 0333d92b1438..614f1a585511 100644 --- a/Documentation/netlink/specs/handshake.yaml +++ b/Documentation/netlink/specs/handshake.yaml @@ -16,7 +16,7 @@ definitions: type: enum name: handler-class value-start: 0 - entries: [ none, max ] + entries: [ none, tlshd, max ] - type: enum name: msg-type @@ -120,3 +120,5 @@ mcast-groups: list: - name: none + - + name: tlshd diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst index 24bb256d6d53..a164ff074356 100644 --- a/Documentation/networking/index.rst +++ b/Documentation/networking/index.rst @@ -36,6 +36,7 @@ Contents: scaling tls tls-offload + tls-handshake nfc 6lowpan 6pack diff --git a/Documentation/networking/tls-handshake.rst b/Documentation/networking/tls-handshake.rst new file mode 100644 index 000000000000..a2817a88e905 --- /dev/null +++ b/Documentation/networking/tls-handshake.rst @@ -0,0 +1,217 @@ +.. SPDX-License-Identifier: GPL-2.0 + +======================= +In-Kernel TLS Handshake +======================= + +Overview +======== + +Transport Layer Security (TLS) is a Upper Layer Protocol (ULP) that runs +over TCP. TLS provides end-to-end data integrity and confidentiality in +addition to peer authentication. + +The kernel's kTLS implementation handles the TLS record subprotocol, but +does not handle the TLS handshake subprotocol which is used to establish +a TLS session. Kernel consumers can use the API described here to +request TLS session establishment. + +There are several possible ways to provide a handshake service in the +kernel. The API described here is designed to hide the details of those +implementations so that in-kernel TLS consumers do not need to be +aware of how the handshake gets done. + + +User handshake agent +==================== + +As of this writing, there is no TLS handshake implementation in the +Linux kernel. To provide a handshake service, a handshake agent +(typically in user space) is started in each network namespace where a +kernel consumer might require a TLS handshake. Handshake agents listen +for events sent from the kernel that indicate a handshake request is +waiting. + +An open socket is passed to a handshake agent via a netlink operation, +which creates a socket descriptor in the agent's file descriptor table. +If the handshake completes successfully, the handshake agent promotes +the socket to use the TLS ULP and sets the session information using the +SOL_TLS socket options. The handshake agent returns the socket to the +kernel via a second netlink operation. + + +Kernel Handshake API +==================== + +A kernel TLS consumer initiates a client-side TLS handshake on an open +socket by invoking one of the tls_client_hello() functions. First, it +fills in a structure that contains the parameters of the request: + +.. code-block:: c + + struct tls_handshake_args { + struct socket *ta_sock; + tls_done_func_t ta_done; + void *ta_data; + unsigned int ta_timeout_ms; + key_serial_t ta_keyring; + key_serial_t ta_my_cert; + key_serial_t ta_my_privkey; + unsigned int ta_num_peerids; + key_serial_t ta_my_peerids[5]; + }; + +The @ta_sock field references an open and connected socket. The consumer +must hold a reference on the socket to prevent it from being destroyed +while the handshake is in progress. The consumer must also have +instantiated a struct file in sock->file. + + +@ta_done contains a callback function that is invoked when the handshake +has completed. Further explanation of this function is in the "Handshake +Completion" sesction below. + +The consumer can fill in the @ta_timeout_ms field to force the servicing +handshake agent to exit after a number of milliseconds. This enables the +socket to be fully closed once both the kernel and the handshake agent +have closed their endpoints. + +Authentication material such as x.509 certificates, private certificate +keys, and pre-shared keys are provided to the handshake agent in keys +that are instantiated by the consumer before making the handshake +request. The consumer can provide a private keyring that is linked into +the handshake agent's process keyring in the @ta_keyring field to prevent +access of those keys by other subsystems. + +To request an x.509-authenticated TLS session, the consumer fills in +the @ta_my_cert and @ta_my_privkey fields with the serial numbers of +keys containing an x.509 certificate and the private key for that +certificate. Then, it invokes this function: + +.. code-block:: c + + ret = tls_client_hello_x509(args, gfp_flags); + +The function returns zero when the handshake request is under way. A +zero return guarantees the callback function @ta_done will be invoked +for this socket. The function returns a negative errno if the handshake +could not be started. A negative errno guarantees the callback function +@ta_done will not be invoked on this socket. + + +To initiate a client-side TLS handshake with a pre-shared key, use: + +.. code-block:: c + + ret = tls_client_hello_psk(args, gfp_flags); + +However, in this case, the consumer fills in the @ta_my_peerids array +with serial numbers of keys containing the peer identities it wishes +to offer, and the @ta_num_peerids field with the number of array +entries it has filled in. The other fields are filled in as above. + + +To initiate an anonymous client-side TLS handshake use: + +.. code-block:: c + + ret = tls_client_hello_anon(args, gfp_flags); + +The handshake agent presents no peer identity information to the remote +during this type of handshake. Only server authentication (ie the client +verifies the server's identity) is performed during the handshake. Thus +the established session uses encryption only. + + +Consumers that are in-kernel servers use: + +.. code-block:: c + + ret = tls_server_hello_x509(args, gfp_flags); + +or + +.. code-block:: c + + ret = tls_server_hello_psk(args, gfp_flags); + +The argument structure is filled in as above. + + +If the consumer needs to cancel the handshake request, say, due to a ^C +or other exigent event, the consumer can invoke: + +.. code-block:: c + + bool tls_handshake_cancel(sock); + +This function returns true if the handshake request associated with +@sock has been canceled. The consumer's handshake completion callback +will not be invoked. If this function returns false, then the consumer's +completion callback has already been invoked. + + +Handshake Completion +==================== + +When the handshake agent has completed processing, it notifies the +kernel that the socket may be used by the consumer again. At this point, +the consumer's handshake completion callback, provided in the @ta_done +field in the tls_handshake_args structure, is invoked. + +The synopsis of this function is: + +.. code-block:: c + + typedef void (*tls_done_func_t)(void *data, int status, + key_serial_t peerid); + +The consumer provides a cookie in the @ta_data field of the +tls_handshake_args structure that is returned in the @data parameter of +this callback. The consumer uses the cookie to match the callback to the +thread waiting for the handshake to complete. + +The success status of the handshake is returned via the @status +parameter: + ++------------+----------------------------------------------+ +| status | meaning | ++============+==============================================+ +| 0 | TLS session established successfully | ++------------+----------------------------------------------+ +| -EACCESS | Remote peer rejected the handshake or | +| | authentication failed | ++------------+----------------------------------------------+ +| -ENOMEM | Temporary resource allocation failure | ++------------+----------------------------------------------+ +| -EINVAL | Consumer provided an invalid argument | ++------------+----------------------------------------------+ +| -ENOKEY | Missing authentication material | ++------------+----------------------------------------------+ +| -EIO | An unexpected fault occurred | ++------------+----------------------------------------------+ + +The @peerid parameter contains the serial number of a key containing the +remote peer's identity or the value TLS_NO_PEERID if the session is not +authenticated. + +A best practice is to close and destroy the socket immediately if the +handshake failed. + + +Other considerations +-------------------- + +While a handshake is under way, the kernel consumer must alter the +socket's sk_data_ready callback function to ignore all incoming data. +Once the handshake completion callback function has been invoked, normal +receive operation can be resumed. + +Once a TLS session is established, the consumer must provide a buffer +for and then examine the control message (CMSG) that is part of every +subsequent sock_recvmsg(). Each control message indicates whether the +received message data is TLS record data or session metadata. + +See tls.rst for details on how a kTLS consumer recognizes incoming +(decrypted) application data, alerts, and handshake packets once the +socket has been promoted to use the TLS ULP. diff --git a/MAINTAINERS b/MAINTAINERS index cdc7748d15b8..04ebde8ccb75 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8953,6 +8953,8 @@ L: kernel-tls-handshake@lists.linux.dev L: netdev@vger.kernel.org S: Maintained F: Documentation/netlink/specs/handshake.yaml +F: Documentation/networking/tls-handshake.rst +F: include/net/handshake.h F: include/trace/events/handshake.h F: net/handshake/ diff --git a/include/net/handshake.h b/include/net/handshake.h new file mode 100644 index 000000000000..3352b1ab43b3 --- /dev/null +++ b/include/net/handshake.h @@ -0,0 +1,43 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Generic netlink HANDSHAKE service. + * + * Author: Chuck Lever + * + * Copyright (c) 2023, Oracle and/or its affiliates. + */ + +#ifndef _NET_HANDSHAKE_H +#define _NET_HANDSHAKE_H + +enum { + TLS_NO_KEYRING = 0, + TLS_NO_PEERID = 0, + TLS_NO_CERT = 0, + TLS_NO_PRIVKEY = 0, +}; + +typedef void (*tls_done_func_t)(void *data, int status, + key_serial_t peerid); + +struct tls_handshake_args { + struct socket *ta_sock; + tls_done_func_t ta_done; + void *ta_data; + unsigned int ta_timeout_ms; + key_serial_t ta_keyring; + key_serial_t ta_my_cert; + key_serial_t ta_my_privkey; + unsigned int ta_num_peerids; + key_serial_t ta_my_peerids[5]; +}; + +int tls_client_hello_anon(const struct tls_handshake_args *args, gfp_t flags); +int tls_client_hello_x509(const struct tls_handshake_args *args, gfp_t flags); +int tls_client_hello_psk(const struct tls_handshake_args *args, gfp_t flags); +int tls_server_hello_x509(const struct tls_handshake_args *args, gfp_t flags); +int tls_server_hello_psk(const struct tls_handshake_args *args, gfp_t flags); + +bool tls_handshake_cancel(struct sock *sk); + +#endif /* _NET_HANDSHAKE_H */ diff --git a/include/uapi/linux/handshake.h b/include/uapi/linux/handshake.h index 7f66ff489b87..1de4d0b95325 100644 --- a/include/uapi/linux/handshake.h +++ b/include/uapi/linux/handshake.h @@ -11,6 +11,7 @@ enum handshake_handler_class { HANDSHAKE_HANDLER_CLASS_NONE, + HANDSHAKE_HANDLER_CLASS_TLSHD, HANDSHAKE_HANDLER_CLASS_MAX, }; @@ -67,5 +68,6 @@ enum { }; #define HANDSHAKE_MCGRP_NONE "none" +#define HANDSHAKE_MCGRP_TLSHD "tlshd" #endif /* _UAPI_LINUX_HANDSHAKE_H */ diff --git a/net/handshake/Makefile b/net/handshake/Makefile index d38736de45da..a089f7e3df24 100644 --- a/net/handshake/Makefile +++ b/net/handshake/Makefile @@ -8,4 +8,4 @@ # obj-y += handshake.o -handshake-y := genl.o netlink.o request.o trace.o +handshake-y := genl.o netlink.o request.o tlshd.o trace.o diff --git a/net/handshake/genl.c b/net/handshake/genl.c index 652f37d19bd6..9f29efb1493e 100644 --- a/net/handshake/genl.c +++ b/net/handshake/genl.c @@ -12,7 +12,7 @@ /* HANDSHAKE_CMD_ACCEPT - do */ static const struct nla_policy handshake_accept_nl_policy[HANDSHAKE_A_ACCEPT_HANDLER_CLASS + 1] = { - [HANDSHAKE_A_ACCEPT_HANDLER_CLASS] = NLA_POLICY_MAX(NLA_U32, 1), + [HANDSHAKE_A_ACCEPT_HANDLER_CLASS] = NLA_POLICY_MAX(NLA_U32, 2), }; /* HANDSHAKE_CMD_DONE - do */ @@ -42,6 +42,7 @@ static const struct genl_split_ops handshake_nl_ops[] = { static const struct genl_multicast_group handshake_nl_mcgrps[] = { [HANDSHAKE_NLGRP_NONE] = { "none", }, + [HANDSHAKE_NLGRP_TLSHD] = { "tlshd", }, }; struct genl_family handshake_nl_family __ro_after_init = { diff --git a/net/handshake/genl.h b/net/handshake/genl.h index a1eb7ccccc7f..2c1f1aa6a02a 100644 --- a/net/handshake/genl.h +++ b/net/handshake/genl.h @@ -16,6 +16,7 @@ int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info); enum { HANDSHAKE_NLGRP_NONE, + HANDSHAKE_NLGRP_TLSHD, }; extern struct genl_family handshake_nl_family; diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c new file mode 100644 index 000000000000..1b8353296060 --- /dev/null +++ b/net/handshake/tlshd.c @@ -0,0 +1,417 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Establish a TLS session for a kernel socket consumer + * using the tlshd user space handler. + * + * Author: Chuck Lever + * + * Copyright (c) 2021-2023, Oracle and/or its affiliates. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include +#include +#include "handshake.h" + +struct tls_handshake_req { + void (*th_consumer_done)(void *data, int status, + key_serial_t peerid); + void *th_consumer_data; + + int th_type; + unsigned int th_timeout_ms; + int th_auth_mode; + key_serial_t th_keyring; + key_serial_t th_certificate; + key_serial_t th_privkey; + + unsigned int th_num_peerids; + key_serial_t th_peerid[5]; +}; + +static struct tls_handshake_req * +tls_handshake_req_init(struct handshake_req *req, + const struct tls_handshake_args *args) +{ + struct tls_handshake_req *treq = handshake_req_private(req); + + treq->th_timeout_ms = args->ta_timeout_ms; + treq->th_consumer_done = args->ta_done; + treq->th_consumer_data = args->ta_data; + treq->th_keyring = args->ta_keyring; + treq->th_num_peerids = 0; + treq->th_certificate = TLS_NO_CERT; + treq->th_privkey = TLS_NO_PRIVKEY; + return treq; +} + +static void tls_handshake_remote_peerids(struct tls_handshake_req *treq, + struct genl_info *info) +{ + struct nlattr *head = nlmsg_attrdata(info->nlhdr, GENL_HDRLEN); + int rem, len = nlmsg_attrlen(info->nlhdr, GENL_HDRLEN); + struct nlattr *nla; + unsigned int i; + + i = 0; + nla_for_each_attr(nla, head, len, rem) { + if (nla_type(nla) == HANDSHAKE_A_DONE_REMOTE_AUTH) + i++; + } + if (!i) + return; + treq->th_num_peerids = min_t(unsigned int, i, + ARRAY_SIZE(treq->th_peerid)); + + i = 0; + nla_for_each_attr(nla, head, len, rem) { + if (nla_type(nla) == HANDSHAKE_A_DONE_REMOTE_AUTH) + treq->th_peerid[i++] = nla_get_u32(nla); + if (i >= treq->th_num_peerids) + break; + } +} + +/** + * tls_handshake_done - callback to handle a CMD_DONE request + * @req: socket on which the handshake was performed + * @status: session status code + * @info: full results of session establishment + * + */ +static void tls_handshake_done(struct handshake_req *req, + unsigned int status, struct genl_info *info) +{ + struct tls_handshake_req *treq = handshake_req_private(req); + + treq->th_peerid[0] = TLS_NO_PEERID; + if (info) + tls_handshake_remote_peerids(treq, info); + + treq->th_consumer_done(treq->th_consumer_data, -status, + treq->th_peerid[0]); +} + +#if IS_ENABLED(CONFIG_KEYS) +static int tls_handshake_private_keyring(struct tls_handshake_req *treq) +{ + key_ref_t process_keyring_ref, keyring_ref; + int ret; + + if (treq->th_keyring == TLS_NO_KEYRING) + return 0; + + process_keyring_ref = lookup_user_key(KEY_SPEC_PROCESS_KEYRING, + KEY_LOOKUP_CREATE, + KEY_NEED_WRITE); + if (IS_ERR(process_keyring_ref)) { + ret = PTR_ERR(process_keyring_ref); + goto out; + } + + keyring_ref = lookup_user_key(treq->th_keyring, KEY_LOOKUP_CREATE, + KEY_NEED_LINK); + if (IS_ERR(keyring_ref)) { + ret = PTR_ERR(keyring_ref); + goto out_put_key; + } + + ret = key_link(key_ref_to_ptr(process_keyring_ref), + key_ref_to_ptr(keyring_ref)); + + key_ref_put(keyring_ref); +out_put_key: + key_ref_put(process_keyring_ref); +out: + return ret; +} +#else +static int tls_handshake_private_keyring(struct tls_handshake_req *treq) +{ + return 0; +} +#endif + +static int tls_handshake_put_peer_identity(struct sk_buff *msg, + struct tls_handshake_req *treq) +{ + unsigned int i; + + for (i = 0; i < treq->th_num_peerids; i++) + if (nla_put_u32(msg, HANDSHAKE_A_ACCEPT_PEER_IDENTITY, + treq->th_peerid[i]) < 0) + return -EMSGSIZE; + return 0; +} + +static int tls_handshake_put_certificate(struct sk_buff *msg, + struct tls_handshake_req *treq) +{ + struct nlattr *entry_attr; + + if (treq->th_certificate == TLS_NO_CERT && + treq->th_privkey == TLS_NO_PRIVKEY) + return 0; + + entry_attr = nla_nest_start(msg, HANDSHAKE_A_ACCEPT_CERTIFICATE); + if (!entry_attr) + return -EMSGSIZE; + + if (nla_put_u32(msg, HANDSHAKE_A_X509_CERT, + treq->th_certificate) || + nla_put_u32(msg, HANDSHAKE_A_X509_PRIVKEY, + treq->th_privkey)) { + nla_nest_cancel(msg, entry_attr); + return -EMSGSIZE; + } + + nla_nest_end(msg, entry_attr); + return 0; +} + +/** + * tls_handshake_accept - callback to construct a CMD_ACCEPT response + * @req: handshake parameters to return + * @info: generic netlink message context + * @fd: file descriptor to be returned + * + * Returns zero on success, or a negative errno on failure. + */ +static int tls_handshake_accept(struct handshake_req *req, + struct genl_info *info, int fd) +{ + struct tls_handshake_req *treq = handshake_req_private(req); + struct nlmsghdr *hdr; + struct sk_buff *msg; + int ret; + + ret = tls_handshake_private_keyring(treq); + if (ret < 0) + goto out; + + ret = -ENOMEM; + msg = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + goto out; + hdr = handshake_genl_put(msg, info); + if (!hdr) + goto out_cancel; + + ret = -EMSGSIZE; + ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_SOCKFD, fd); + if (ret < 0) + goto out_cancel; + ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_MESSAGE_TYPE, treq->th_type); + if (ret < 0) + goto out_cancel; + if (treq->th_timeout_ms) { + ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_TIMEOUT, treq->th_timeout_ms); + if (ret < 0) + goto out_cancel; + } + + ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_AUTH_MODE, + treq->th_auth_mode); + if (ret < 0) + goto out_cancel; + switch (treq->th_auth_mode) { + case HANDSHAKE_AUTH_PSK: + ret = tls_handshake_put_peer_identity(msg, treq); + if (ret < 0) + goto out_cancel; + break; + case HANDSHAKE_AUTH_X509: + ret = tls_handshake_put_certificate(msg, treq); + if (ret < 0) + goto out_cancel; + break; + } + + genlmsg_end(msg, hdr); + return genlmsg_reply(msg, info); + +out_cancel: + genlmsg_cancel(msg, hdr); +out: + return ret; +} + +static const struct handshake_proto tls_handshake_proto = { + .hp_handler_class = HANDSHAKE_HANDLER_CLASS_TLSHD, + .hp_privsize = sizeof(struct tls_handshake_req), + + .hp_accept = tls_handshake_accept, + .hp_done = tls_handshake_done, +}; + +/** + * tls_client_hello_anon - request an anonymous TLS handshake on a socket + * @args: socket and handshake parameters for this request + * @flags: memory allocation control flags + * + * Return values: + * %0: Handshake request enqueue; ->done will be called when complete + * %-ESRCH: No user agent is available + * %-ENOMEM: Memory allocation failed + */ +int tls_client_hello_anon(const struct tls_handshake_args *args, gfp_t flags) +{ + struct tls_handshake_req *treq; + struct handshake_req *req; + + req = handshake_req_alloc(&tls_handshake_proto, flags); + if (!req) + return -ENOMEM; + treq = tls_handshake_req_init(req, args); + treq->th_type = HANDSHAKE_MSG_TYPE_CLIENTHELLO; + treq->th_auth_mode = HANDSHAKE_AUTH_UNAUTH; + + return handshake_req_submit(args->ta_sock, req, flags); +} +EXPORT_SYMBOL(tls_client_hello_anon); + +/** + * tls_client_hello_x509 - request an x.509-based TLS handshake on a socket + * @args: socket and handshake parameters for this request + * @flags: memory allocation control flags + * + * Return values: + * %0: Handshake request enqueue; ->done will be called when complete + * %-ESRCH: No user agent is available + * %-ENOMEM: Memory allocation failed + */ +int tls_client_hello_x509(const struct tls_handshake_args *args, gfp_t flags) +{ + struct tls_handshake_req *treq; + struct handshake_req *req; + + req = handshake_req_alloc(&tls_handshake_proto, flags); + if (!req) + return -ENOMEM; + treq = tls_handshake_req_init(req, args); + treq->th_type = HANDSHAKE_MSG_TYPE_CLIENTHELLO; + treq->th_auth_mode = HANDSHAKE_AUTH_X509; + treq->th_certificate = args->ta_my_cert; + treq->th_privkey = args->ta_my_privkey; + + return handshake_req_submit(args->ta_sock, req, flags); +} +EXPORT_SYMBOL(tls_client_hello_x509); + +/** + * tls_client_hello_psk - request a PSK-based TLS handshake on a socket + * @args: socket and handshake parameters for this request + * @flags: memory allocation control flags + * + * Return values: + * %0: Handshake request enqueue; ->done will be called when complete + * %-EINVAL: Wrong number of local peer IDs + * %-ESRCH: No user agent is available + * %-ENOMEM: Memory allocation failed + */ +int tls_client_hello_psk(const struct tls_handshake_args *args, gfp_t flags) +{ + struct tls_handshake_req *treq; + struct handshake_req *req; + unsigned int i; + + if (!args->ta_num_peerids || + args->ta_num_peerids > ARRAY_SIZE(treq->th_peerid)) + return -EINVAL; + + req = handshake_req_alloc(&tls_handshake_proto, flags); + if (!req) + return -ENOMEM; + treq = tls_handshake_req_init(req, args); + treq->th_type = HANDSHAKE_MSG_TYPE_CLIENTHELLO; + treq->th_auth_mode = HANDSHAKE_AUTH_PSK; + treq->th_num_peerids = args->ta_num_peerids; + for (i = 0; i < args->ta_num_peerids; i++) + treq->th_peerid[i] = args->ta_my_peerids[i]; + + return handshake_req_submit(args->ta_sock, req, flags); +} +EXPORT_SYMBOL(tls_client_hello_psk); + +/** + * tls_server_hello_x509 - request a server TLS handshake on a socket + * @args: socket and handshake parameters for this request + * @flags: memory allocation control flags + * + * Return values: + * %0: Handshake request enqueue; ->done will be called when complete + * %-ESRCH: No user agent is available + * %-ENOMEM: Memory allocation failed + */ +int tls_server_hello_x509(const struct tls_handshake_args *args, gfp_t flags) +{ + struct tls_handshake_req *treq; + struct handshake_req *req; + + req = handshake_req_alloc(&tls_handshake_proto, flags); + if (!req) + return -ENOMEM; + treq = tls_handshake_req_init(req, args); + treq->th_type = HANDSHAKE_MSG_TYPE_SERVERHELLO; + treq->th_auth_mode = HANDSHAKE_AUTH_X509; + treq->th_certificate = args->ta_my_cert; + treq->th_privkey = args->ta_my_privkey; + + return handshake_req_submit(args->ta_sock, req, flags); +} +EXPORT_SYMBOL(tls_server_hello_x509); + +/** + * tls_server_hello_psk - request a server TLS handshake on a socket + * @args: socket and handshake parameters for this request + * @flags: memory allocation control flags + * + * Return values: + * %0: Handshake request enqueue; ->done will be called when complete + * %-ESRCH: No user agent is available + * %-ENOMEM: Memory allocation failed + */ +int tls_server_hello_psk(const struct tls_handshake_args *args, gfp_t flags) +{ + struct tls_handshake_req *treq; + struct handshake_req *req; + + req = handshake_req_alloc(&tls_handshake_proto, flags); + if (!req) + return -ENOMEM; + treq = tls_handshake_req_init(req, args); + treq->th_type = HANDSHAKE_MSG_TYPE_SERVERHELLO; + treq->th_auth_mode = HANDSHAKE_AUTH_PSK; + treq->th_num_peerids = 1; + treq->th_peerid[0] = args->ta_my_peerids[0]; + + return handshake_req_submit(args->ta_sock, req, flags); +} +EXPORT_SYMBOL(tls_server_hello_psk); + +/** + * tls_handshake_cancel - cancel a pending handshake + * @sk: socket on which there is an ongoing handshake + * + * Request cancellation races with request completion. To determine + * who won, callers examine the return value from this function. + * + * Return values: + * %true - Uncompleted handshake request was canceled + * %false - Handshake request already completed or not found + */ +bool tls_handshake_cancel(struct sock *sk) +{ + return handshake_req_cancel(sk); +} +EXPORT_SYMBOL(tls_handshake_cancel); -- cgit v1.2.3 From 88232ec1ec5ecf4aa5de439cff3d5e2b7adcac93 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 17 Apr 2023 10:32:39 -0400 Subject: net/handshake: Add Kunit tests for the handshake consumer API These verify the API contracts and help exercise lifetime rules for consumer sockets and handshake_req structures. One way to run these tests: ./tools/testing/kunit/kunit.py run --kunitconfig ./net/handshake/.kunitconfig Signed-off-by: Chuck Lever Signed-off-by: Jakub Kicinski --- net/Kconfig | 15 ++ net/handshake/.kunitconfig | 11 + net/handshake/Makefile | 2 + net/handshake/handshake-test.c | 523 +++++++++++++++++++++++++++++++++++++++++ net/handshake/handshake.h | 5 + net/handshake/netlink.c | 7 + net/handshake/request.c | 5 + net/handshake/tlshd.c | 1 + 8 files changed, 569 insertions(+) create mode 100644 net/handshake/.kunitconfig create mode 100644 net/handshake/handshake-test.c (limited to 'net') diff --git a/net/Kconfig b/net/Kconfig index 4b800706cc76..7d39c1773eb4 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -73,6 +73,21 @@ config NET_HANDSHAKE depends on SUNRPC || NVME_TARGET_TCP || NVME_TCP default y +config NET_HANDSHAKE_KUNIT_TEST + tristate "KUnit tests for the handshake upcall mechanism" if !KUNIT_ALL_TESTS + default KUNIT_ALL_TESTS + depends on KUNIT + help + This builds the KUnit tests for the handshake upcall mechanism. + + KUnit tests run during boot and output the results to the debug + log in TAP format (https://testanything.org/). Only useful for + kernel devs running KUnit test harness and are not for inclusion + into a production build. + + For more information on KUnit and unit tests in general, refer + to the KUnit documentation in Documentation/dev-tools/kunit/. + config INET bool "TCP/IP networking" help diff --git a/net/handshake/.kunitconfig b/net/handshake/.kunitconfig new file mode 100644 index 000000000000..5c48cf4abca2 --- /dev/null +++ b/net/handshake/.kunitconfig @@ -0,0 +1,11 @@ +CONFIG_KUNIT=y +CONFIG_UBSAN=y +CONFIG_STACKTRACE=y +CONFIG_NET=y +CONFIG_NETWORK_FILESYSTEMS=y +CONFIG_INET=y +CONFIG_MULTIUSER=y +CONFIG_NFS_FS=y +CONFIG_SUNRPC=y +CONFIG_NET_HANDSHAKE=y +CONFIG_NET_HANDSHAKE_KUNIT_TEST=y diff --git a/net/handshake/Makefile b/net/handshake/Makefile index a089f7e3df24..247d73c6ff6e 100644 --- a/net/handshake/Makefile +++ b/net/handshake/Makefile @@ -9,3 +9,5 @@ obj-y += handshake.o handshake-y := genl.o netlink.o request.o tlshd.o trace.o + +obj-$(CONFIG_NET_HANDSHAKE_KUNIT_TEST) += handshake-test.o diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c new file mode 100644 index 000000000000..e6adc5dec11a --- /dev/null +++ b/net/handshake/handshake-test.c @@ -0,0 +1,523 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2023 Oracle and/or its affiliates. + * + * KUnit test of the handshake upcall mechanism. + */ + +#include +#include + +#include + +#include +#include +#include + +#include +#include "handshake.h" + +MODULE_IMPORT_NS(EXPORTED_FOR_KUNIT_TESTING); + +static int test_accept_func(struct handshake_req *req, struct genl_info *info, + int fd) +{ + return 0; +} + +static void test_done_func(struct handshake_req *req, unsigned int status, + struct genl_info *info) +{ +} + +struct handshake_req_alloc_test_param { + const char *desc; + struct handshake_proto *proto; + gfp_t gfp; + bool expect_success; +}; + +static struct handshake_proto handshake_req_alloc_proto_2 = { + .hp_handler_class = HANDSHAKE_HANDLER_CLASS_NONE, +}; + +static struct handshake_proto handshake_req_alloc_proto_3 = { + .hp_handler_class = HANDSHAKE_HANDLER_CLASS_MAX, +}; + +static struct handshake_proto handshake_req_alloc_proto_4 = { + .hp_handler_class = HANDSHAKE_HANDLER_CLASS_TLSHD, +}; + +static struct handshake_proto handshake_req_alloc_proto_5 = { + .hp_handler_class = HANDSHAKE_HANDLER_CLASS_TLSHD, + .hp_accept = test_accept_func, +}; + +static struct handshake_proto handshake_req_alloc_proto_6 = { + .hp_handler_class = HANDSHAKE_HANDLER_CLASS_TLSHD, + .hp_privsize = UINT_MAX, + .hp_accept = test_accept_func, + .hp_done = test_done_func, +}; + +static struct handshake_proto handshake_req_alloc_proto_good = { + .hp_handler_class = HANDSHAKE_HANDLER_CLASS_TLSHD, + .hp_accept = test_accept_func, + .hp_done = test_done_func, +}; + +static const +struct handshake_req_alloc_test_param handshake_req_alloc_params[] = { + { + .desc = "handshake_req_alloc NULL proto", + .proto = NULL, + .gfp = GFP_KERNEL, + .expect_success = false, + }, + { + .desc = "handshake_req_alloc CLASS_NONE", + .proto = &handshake_req_alloc_proto_2, + .gfp = GFP_KERNEL, + .expect_success = false, + }, + { + .desc = "handshake_req_alloc CLASS_MAX", + .proto = &handshake_req_alloc_proto_3, + .gfp = GFP_KERNEL, + .expect_success = false, + }, + { + .desc = "handshake_req_alloc no callbacks", + .proto = &handshake_req_alloc_proto_4, + .gfp = GFP_KERNEL, + .expect_success = false, + }, + { + .desc = "handshake_req_alloc no done callback", + .proto = &handshake_req_alloc_proto_5, + .gfp = GFP_KERNEL, + .expect_success = false, + }, + { + .desc = "handshake_req_alloc excessive privsize", + .proto = &handshake_req_alloc_proto_6, + .gfp = GFP_KERNEL, + .expect_success = false, + }, + { + .desc = "handshake_req_alloc all good", + .proto = &handshake_req_alloc_proto_good, + .gfp = GFP_KERNEL, + .expect_success = true, + }, +}; + +static void +handshake_req_alloc_get_desc(const struct handshake_req_alloc_test_param *param, + char *desc) +{ + strscpy(desc, param->desc, KUNIT_PARAM_DESC_SIZE); +} + +/* Creates the function handshake_req_alloc_gen_params */ +KUNIT_ARRAY_PARAM(handshake_req_alloc, handshake_req_alloc_params, + handshake_req_alloc_get_desc); + +static void handshake_req_alloc_case(struct kunit *test) +{ + const struct handshake_req_alloc_test_param *param = test->param_value; + struct handshake_req *result; + + /* Arrange */ + + /* Act */ + result = handshake_req_alloc(param->proto, param->gfp); + + /* Assert */ + if (param->expect_success) + KUNIT_EXPECT_NOT_NULL(test, result); + else + KUNIT_EXPECT_NULL(test, result); + + kfree(result); +} + +static void handshake_req_submit_test1(struct kunit *test) +{ + struct socket *sock; + int err, result; + + /* Arrange */ + err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, + &sock, 1); + KUNIT_ASSERT_EQ(test, err, 0); + + /* Act */ + result = handshake_req_submit(sock, NULL, GFP_KERNEL); + + /* Assert */ + KUNIT_EXPECT_EQ(test, result, -EINVAL); + + sock_release(sock); +} + +static void handshake_req_submit_test2(struct kunit *test) +{ + struct handshake_req *req; + int result; + + /* Arrange */ + req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req); + + /* Act */ + result = handshake_req_submit(NULL, req, GFP_KERNEL); + + /* Assert */ + KUNIT_EXPECT_EQ(test, result, -EINVAL); + + /* handshake_req_submit() destroys @req on error */ +} + +static void handshake_req_submit_test3(struct kunit *test) +{ + struct handshake_req *req; + struct socket *sock; + int err, result; + + /* Arrange */ + req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req); + + err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, + &sock, 1); + KUNIT_ASSERT_EQ(test, err, 0); + sock->file = NULL; + + /* Act */ + result = handshake_req_submit(sock, req, GFP_KERNEL); + + /* Assert */ + KUNIT_EXPECT_EQ(test, result, -EINVAL); + + /* handshake_req_submit() destroys @req on error */ + sock_release(sock); +} + +static void handshake_req_submit_test4(struct kunit *test) +{ + struct handshake_req *req, *result; + struct socket *sock; + int err; + + /* Arrange */ + req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req); + + err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, + &sock, 1); + KUNIT_ASSERT_EQ(test, err, 0); + sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + KUNIT_ASSERT_NOT_NULL(test, sock->sk); + + err = handshake_req_submit(sock, req, GFP_KERNEL); + KUNIT_ASSERT_EQ(test, err, 0); + + /* Act */ + result = handshake_req_hash_lookup(sock->sk); + + /* Assert */ + KUNIT_EXPECT_NOT_NULL(test, result); + KUNIT_EXPECT_PTR_EQ(test, req, result); + + handshake_req_cancel(sock->sk); + sock_release(sock); +} + +static void handshake_req_submit_test5(struct kunit *test) +{ + struct handshake_req *req; + struct handshake_net *hn; + struct socket *sock; + struct net *net; + int saved, err; + + /* Arrange */ + req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req); + + err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, + &sock, 1); + KUNIT_ASSERT_EQ(test, err, 0); + sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + KUNIT_ASSERT_NOT_NULL(test, sock->sk); + + net = sock_net(sock->sk); + hn = handshake_pernet(net); + KUNIT_ASSERT_NOT_NULL(test, hn); + + saved = hn->hn_pending; + hn->hn_pending = hn->hn_pending_max + 1; + + /* Act */ + err = handshake_req_submit(sock, req, GFP_KERNEL); + + /* Assert */ + KUNIT_EXPECT_EQ(test, err, -EAGAIN); + + sock_release(sock); + hn->hn_pending = saved; +} + +static void handshake_req_submit_test6(struct kunit *test) +{ + struct handshake_req *req1, *req2; + struct socket *sock; + int err; + + /* Arrange */ + req1 = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req1); + req2 = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req2); + + err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, + &sock, 1); + KUNIT_ASSERT_EQ(test, err, 0); + sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + KUNIT_ASSERT_NOT_NULL(test, sock->sk); + + /* Act */ + err = handshake_req_submit(sock, req1, GFP_KERNEL); + KUNIT_ASSERT_EQ(test, err, 0); + err = handshake_req_submit(sock, req2, GFP_KERNEL); + + /* Assert */ + KUNIT_EXPECT_EQ(test, err, -EBUSY); + + handshake_req_cancel(sock->sk); + sock_release(sock); +} + +static void handshake_req_cancel_test1(struct kunit *test) +{ + struct handshake_req *req; + struct socket *sock; + bool result; + int err; + + /* Arrange */ + req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req); + + err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, + &sock, 1); + KUNIT_ASSERT_EQ(test, err, 0); + + sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + + err = handshake_req_submit(sock, req, GFP_KERNEL); + KUNIT_ASSERT_EQ(test, err, 0); + + /* NB: handshake_req hasn't been accepted */ + + /* Act */ + result = handshake_req_cancel(sock->sk); + + /* Assert */ + KUNIT_EXPECT_TRUE(test, result); + + sock_release(sock); +} + +static void handshake_req_cancel_test2(struct kunit *test) +{ + struct handshake_req *req, *next; + struct handshake_net *hn; + struct socket *sock; + struct net *net; + bool result; + int err; + + /* Arrange */ + req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req); + + err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, + &sock, 1); + KUNIT_ASSERT_EQ(test, err, 0); + + sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + + err = handshake_req_submit(sock, req, GFP_KERNEL); + KUNIT_ASSERT_EQ(test, err, 0); + + net = sock_net(sock->sk); + hn = handshake_pernet(net); + KUNIT_ASSERT_NOT_NULL(test, hn); + + /* Pretend to accept this request */ + next = handshake_req_next(hn, HANDSHAKE_HANDLER_CLASS_TLSHD); + KUNIT_ASSERT_PTR_EQ(test, req, next); + + /* Act */ + result = handshake_req_cancel(sock->sk); + + /* Assert */ + KUNIT_EXPECT_TRUE(test, result); + + sock_release(sock); +} + +static void handshake_req_cancel_test3(struct kunit *test) +{ + struct handshake_req *req, *next; + struct handshake_net *hn; + struct socket *sock; + struct net *net; + bool result; + int err; + + /* Arrange */ + req = handshake_req_alloc(&handshake_req_alloc_proto_good, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req); + + err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, + &sock, 1); + KUNIT_ASSERT_EQ(test, err, 0); + + sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + + err = handshake_req_submit(sock, req, GFP_KERNEL); + KUNIT_ASSERT_EQ(test, err, 0); + + net = sock_net(sock->sk); + hn = handshake_pernet(net); + KUNIT_ASSERT_NOT_NULL(test, hn); + + /* Pretend to accept this request */ + next = handshake_req_next(hn, HANDSHAKE_HANDLER_CLASS_TLSHD); + KUNIT_ASSERT_PTR_EQ(test, req, next); + + /* Pretend to complete this request */ + handshake_complete(next, -ETIMEDOUT, NULL); + + /* Act */ + result = handshake_req_cancel(sock->sk); + + /* Assert */ + KUNIT_EXPECT_FALSE(test, result); + + sock_release(sock); +} + +static struct handshake_req *handshake_req_destroy_test; + +static void test_destroy_func(struct handshake_req *req) +{ + handshake_req_destroy_test = req; +} + +static struct handshake_proto handshake_req_alloc_proto_destroy = { + .hp_handler_class = HANDSHAKE_HANDLER_CLASS_TLSHD, + .hp_accept = test_accept_func, + .hp_done = test_done_func, + .hp_destroy = test_destroy_func, +}; + +static void handshake_req_destroy_test1(struct kunit *test) +{ + struct handshake_req *req; + struct socket *sock; + int err; + + /* Arrange */ + handshake_req_destroy_test = NULL; + + req = handshake_req_alloc(&handshake_req_alloc_proto_destroy, GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, req); + + err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, + &sock, 1); + KUNIT_ASSERT_EQ(test, err, 0); + + sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + + err = handshake_req_submit(sock, req, GFP_KERNEL); + KUNIT_ASSERT_EQ(test, err, 0); + + handshake_req_cancel(sock->sk); + + /* Act */ + sock_release(sock); + + /* Assert */ + KUNIT_EXPECT_PTR_EQ(test, handshake_req_destroy_test, req); +} + +static struct kunit_case handshake_api_test_cases[] = { + { + .name = "req_alloc API fuzzing", + .run_case = handshake_req_alloc_case, + .generate_params = handshake_req_alloc_gen_params, + }, + { + .name = "req_submit NULL req arg", + .run_case = handshake_req_submit_test1, + }, + { + .name = "req_submit NULL sock arg", + .run_case = handshake_req_submit_test2, + }, + { + .name = "req_submit NULL sock->file", + .run_case = handshake_req_submit_test3, + }, + { + .name = "req_lookup works", + .run_case = handshake_req_submit_test4, + }, + { + .name = "req_submit max pending", + .run_case = handshake_req_submit_test5, + }, + { + .name = "req_submit multiple", + .run_case = handshake_req_submit_test6, + }, + { + .name = "req_cancel before accept", + .run_case = handshake_req_cancel_test1, + }, + { + .name = "req_cancel after accept", + .run_case = handshake_req_cancel_test2, + }, + { + .name = "req_cancel after done", + .run_case = handshake_req_cancel_test3, + }, + { + .name = "req_destroy works", + .run_case = handshake_req_destroy_test1, + }, + {} +}; + +static struct kunit_suite handshake_api_suite = { + .name = "Handshake API tests", + .test_cases = handshake_api_test_cases, +}; + +kunit_test_suites(&handshake_api_suite); + +MODULE_DESCRIPTION("Test handshake upcall API functions"); +MODULE_LICENSE("GPL"); diff --git a/net/handshake/handshake.h b/net/handshake/handshake.h index 52568dbe24f1..4dac965c99df 100644 --- a/net/handshake/handshake.h +++ b/net/handshake/handshake.h @@ -49,6 +49,7 @@ enum hr_flags_bits { struct handshake_proto { int hp_handler_class; size_t hp_privsize; + unsigned long hp_flags; int (*hp_accept)(struct handshake_req *req, struct genl_info *info, int fd); @@ -58,6 +59,10 @@ struct handshake_proto { void (*hp_destroy)(struct handshake_req *req); }; +enum hp_flags_bits { + HANDSHAKE_F_PROTO_NOTIFY, +}; + /* netlink.c */ int handshake_genl_notify(struct net *net, const struct handshake_proto *proto, gfp_t flags); diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 7264cac04047..8ea0ff993f9f 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -18,6 +18,8 @@ #include #include +#include + #include #include "handshake.h" #include "genl.h" @@ -38,6 +40,10 @@ int handshake_genl_notify(struct net *net, const struct handshake_proto *proto, struct sk_buff *msg; void *hdr; + /* Disable notifications during unit testing */ + if (!test_bit(HANDSHAKE_F_PROTO_NOTIFY, &proto->hp_flags)) + return 0; + if (!genl_has_listeners(&handshake_nl_family, net, proto->hp_handler_class)) return -ESRCH; @@ -262,6 +268,7 @@ struct handshake_net *handshake_pernet(struct net *net) return handshake_net_id ? net_generic(net, handshake_net_id) : NULL; } +EXPORT_SYMBOL_IF_KUNIT(handshake_pernet); static int __init handshake_init(void) { diff --git a/net/handshake/request.c b/net/handshake/request.c index d5b2bc6de057..94d5cef3e048 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -20,6 +20,8 @@ #include #include +#include + #include #include "handshake.h" @@ -60,6 +62,7 @@ struct handshake_req *handshake_req_hash_lookup(struct sock *sk) return rhashtable_lookup_fast(&handshake_rhashtbl, &sk, handshake_rhash_params); } +EXPORT_SYMBOL_IF_KUNIT(handshake_req_hash_lookup); static bool handshake_req_hash_add(struct handshake_req *req) { @@ -192,6 +195,7 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) return req; } +EXPORT_SYMBOL_IF_KUNIT(handshake_req_next); /** * handshake_req_submit - Submit a handshake request @@ -293,6 +297,7 @@ void handshake_complete(struct handshake_req *req, unsigned int status, sock_put(sk); } } +EXPORT_SYMBOL_IF_KUNIT(handshake_complete); /** * handshake_req_cancel - Cancel an in-progress handshake diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c index 1b8353296060..fcbeb63b4eb1 100644 --- a/net/handshake/tlshd.c +++ b/net/handshake/tlshd.c @@ -249,6 +249,7 @@ out: static const struct handshake_proto tls_handshake_proto = { .hp_handler_class = HANDSHAKE_HANDLER_CLASS_TLSHD, .hp_privsize = sizeof(struct tls_handshake_req), + .hp_flags = BIT(HANDSHAKE_F_PROTO_NOTIFY), .hp_accept = tls_handshake_accept, .hp_done = tls_handshake_done, -- cgit v1.2.3 From 927cdea5d2095287ddd5246e5aa68eb5d68db2be Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 18 Apr 2023 18:59:02 +0300 Subject: net: bridge: switchdev: don't notify FDB entries with "master dynamic" There is a structural problem in switchdev, where the flag bits in struct switchdev_notifier_fdb_info (added_by_user, is_local etc) only represent a simplified / denatured view of what's in struct net_bridge_fdb_entry :: flags (BR_FDB_ADDED_BY_USER, BR_FDB_LOCAL etc). Each time we want to pass more information about struct net_bridge_fdb_entry :: flags to struct switchdev_notifier_fdb_info (here, BR_FDB_STATIC), we find that FDB entries were already notified to switchdev with no regard to this flag, and thus, switchdev drivers had no indication whether the notified entries were static or not. For example, this command: ip link add br0 type bridge && ip link set swp0 master br0 bridge fdb add dev swp0 00:01:02:03:04:05 master dynamic has never worked as intended with switchdev. It causes a struct net_bridge_fdb_entry to be passed to br_switchdev_fdb_notify() which has a single flag set: BR_FDB_ADDED_BY_USER. This is further passed to the switchdev notifier chain, where interested drivers have no choice but to assume this is a static (does not age) and sticky (does not migrate) FDB entry. So currently, all drivers offload it to hardware as such, as can be seen below ("offload" is set). bridge fdb get 00:01:02:03:04:05 dev swp0 master 00:01:02:03:04:05 dev swp0 offload master br0 The software FDB entry expires $ageing_time centiseconds after the kernel last sees a packet with this MAC SA, and the bridge notifies its deletion as well, so it eventually disappears from hardware too. This is a problem, because it is actually desirable to start offloading "master dynamic" FDB entries correctly - they should expire $ageing_time centiseconds after the *hardware* port last sees a packet with this MAC SA - and this is how the current incorrect behavior was discovered. With an offloaded data plane, it can be expected that software only sees exception path packets, so an otherwise active dynamic FDB entry would be aged out by software sooner than it should. With the change in place, these FDB entries are no longer offloaded: bridge fdb get 00:01:02:03:04:05 dev swp0 master 00:01:02:03:04:05 dev swp0 master br0 and this also constitutes a better way (assuming a backport to stable kernels) for user space to determine whether the kernel has the capability of doing something sane with these or not. As opposed to "master dynamic" FDB entries, on the current behavior of which no one currently depends on (which can be deduced from the lack of kselftests), Ido Schimmel explains that entries with the "extern_learn" flag (BR_FDB_ADDED_BY_EXT_LEARN) should still be notified to switchdev, since the spectrum driver listens to them (and this is kind of okay, because although they are treated identically to "static", they are expected to not age, and to roam). Fixes: 6b26b51b1d13 ("net: bridge: Add support for notifying devices about FDB add/del") Link: https://lore.kernel.org/netdev/20230327115206.jk5q5l753aoelwus@skbuf/ Signed-off-by: Vladimir Oltean Reviewed-by: Jesse Brandeburg Reviewed-by: Ido Schimmel Tested-by: Ido Schimmel Link: https://lore.kernel.org/r/20230418155902.898627-1-vladimir.oltean@nxp.com Signed-off-by: Paolo Abeni --- net/bridge/br_switchdev.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'net') diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c index de18e9c1d7a7..ba95c4d74a60 100644 --- a/net/bridge/br_switchdev.c +++ b/net/bridge/br_switchdev.c @@ -148,6 +148,17 @@ br_switchdev_fdb_notify(struct net_bridge *br, if (test_bit(BR_FDB_LOCKED, &fdb->flags)) return; + /* Entries with these flags were created using ndm_state == NUD_REACHABLE, + * ndm_flags == NTF_MASTER( | NTF_STICKY), ext_flags == 0 by something + * equivalent to 'bridge fdb add ... master dynamic (sticky)'. + * Drivers don't know how to deal with these, so don't notify them to + * avoid confusing them. + */ + if (test_bit(BR_FDB_ADDED_BY_USER, &fdb->flags) && + !test_bit(BR_FDB_STATIC, &fdb->flags) && + !test_bit(BR_FDB_ADDED_BY_EXT_LEARN, &fdb->flags)) + return; + br_switchdev_fdb_populate(br, &item, fdb, NULL); switch (type) { -- cgit v1.2.3 From eb74bfcfa99415f1a65ce021b8ac2256b7ed50bc Mon Sep 17 00:00:00 2001 From: Yingsha Xu Date: Wed, 19 Apr 2023 18:45:47 +0800 Subject: wifi: mac80211: remove return value check of debugfs_create_dir() Smatch complains that: debugfs_hw_add() warn: 'statsd' is an error pointer or valid Debugfs checks are generally not supposed to be checked for errors and it is not necessary here. Just delete the dead code. Signed-off-by: Yingsha Xu Reviewed-by: Dongliang Mu Link: https://lore.kernel.org/r/20230419104548.30124-1-ysxu@hust.edu.cn Signed-off-by: Johannes Berg --- net/mac80211/debugfs.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'net') diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c index dfb9f55e2685..207f772bd8ce 100644 --- a/net/mac80211/debugfs.c +++ b/net/mac80211/debugfs.c @@ -673,10 +673,6 @@ void debugfs_hw_add(struct ieee80211_local *local) statsd = debugfs_create_dir("statistics", phyd); - /* if the dir failed, don't put all the other things into the root! */ - if (!statsd) - return; - #ifdef CONFIG_MAC80211_DEBUG_COUNTERS DEBUGFS_STATS_ADD(dot11TransmittedFragmentCount); DEBUGFS_STATS_ADD(dot11MulticastTransmittedFrameCount); -- cgit v1.2.3 From dd64b232deb8d48812a2ea739d1fedaeaffb59ed Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Apr 2023 11:20:06 -0700 Subject: page_pool: unlink from napi during destroy Jesper points out that we must prevent recycling into cache after page_pool_destroy() is called, because page_pool_destroy() is not synchronized with recycling (some pages may still be outstanding when destroy() gets called). I assumed this will not happen because NAPI can't be scheduled if its page pool is being destroyed. But I missed the fact that NAPI may get reused. For instance when user changes ring configuration driver may allocate a new page pool, stop NAPI, swap, start NAPI, and then destroy the old pool. The NAPI is running so old page pool will think it can recycle to the cache, but the consumer at that point is the destroy() path, not NAPI. To avoid extra synchronization let the drivers do "unlinking" during the "swap" stage while NAPI is indeed disabled. Fixes: 8c48eea3adf3 ("page_pool: allow caching from safely localized NAPI") Reported-by: Jesper Dangaard Brouer Link: https://lore.kernel.org/all/e8df2654-6a5b-3c92-489d-2fe5e444135f@redhat.com/ Acked-by: Jesper Dangaard Brouer Link: https://lore.kernel.org/r/20230419182006.719923-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- include/net/page_pool.h | 5 +++++ net/core/page_pool.c | 18 +++++++++++++++++- 2 files changed, 22 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/page_pool.h b/include/net/page_pool.h index 91b808dade82..c8ec2f34722b 100644 --- a/include/net/page_pool.h +++ b/include/net/page_pool.h @@ -247,6 +247,7 @@ struct page_pool *page_pool_create(const struct page_pool_params *params); struct xdp_mem_info; #ifdef CONFIG_PAGE_POOL +void page_pool_unlink_napi(struct page_pool *pool); void page_pool_destroy(struct page_pool *pool); void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *), struct xdp_mem_info *mem); @@ -254,6 +255,10 @@ void page_pool_release_page(struct page_pool *pool, struct page *page); void page_pool_put_page_bulk(struct page_pool *pool, void **data, int count); #else +static inline void page_pool_unlink_napi(struct page_pool *pool) +{ +} + static inline void page_pool_destroy(struct page_pool *pool) { } diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 97f20f7ff4fc..e212e9d7edcb 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -839,6 +839,21 @@ void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *), pool->xdp_mem_id = mem->id; } +void page_pool_unlink_napi(struct page_pool *pool) +{ + if (!pool->p.napi) + return; + + /* To avoid races with recycling and additional barriers make sure + * pool and NAPI are unlinked when NAPI is disabled. + */ + WARN_ON(!test_bit(NAPI_STATE_SCHED, &pool->p.napi->state) || + READ_ONCE(pool->p.napi->list_owner) != -1); + + WRITE_ONCE(pool->p.napi, NULL); +} +EXPORT_SYMBOL(page_pool_unlink_napi); + void page_pool_destroy(struct page_pool *pool) { if (!pool) @@ -847,6 +862,7 @@ void page_pool_destroy(struct page_pool *pool) if (!page_pool_put(pool)) return; + page_pool_unlink_napi(pool); page_pool_free_frag(pool); if (!page_pool_release(pool)) @@ -900,7 +916,7 @@ bool page_pool_return_skb_page(struct page *page, bool napi_safe) * in the same context as the consumer would run, so there's * no possible race. */ - napi = pp->p.napi; + napi = READ_ONCE(pp->p.napi); allow_direct = napi_safe && napi && READ_ONCE(napi->list_owner) == smp_processor_id(); -- cgit v1.2.3 From 8fa66e4a1bdd41d55d7842928e60a40fed65715d Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 19 Apr 2023 19:00:05 -0700 Subject: net: skbuff: update and rename __kfree_skb_defer() __kfree_skb_defer() uses the old naming where "defer" meant slab bulk free/alloc APIs. In the meantime we also made __kfree_skb_defer() feed the per-NAPI skb cache, which implies bulk APIs. So take away the 'defer' and add 'napi'. While at it add a drop reason. This only matters on the tx_action path, if the skb has a frag_list. But getting rid of a SKB_DROP_REASON_NOT_SPECIFIED seems like a net benefit so why not. Reviewed-by: Alexander Lobakin Link: https://lore.kernel.org/r/20230420020005.815854-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 2 +- net/core/dev.c | 3 ++- net/core/gro.c | 2 +- net/core/skbuff.c | 4 ++-- 4 files changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index db5973559042..03aa7ed076f0 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3253,7 +3253,7 @@ static inline struct sk_buff *napi_alloc_skb(struct napi_struct *napi, void napi_consume_skb(struct sk_buff *skb, int budget); void napi_skb_free_stolen_head(struct sk_buff *skb); -void __kfree_skb_defer(struct sk_buff *skb); +void __napi_kfree_skb(struct sk_buff *skb, enum skb_drop_reason reason); /** * __dev_alloc_pages - allocate page for network Rx diff --git a/net/core/dev.c b/net/core/dev.c index 3fc4dba71f9d..1551aabac343 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5040,7 +5040,8 @@ static __latent_entropy void net_tx_action(struct softirq_action *h) if (skb->fclone != SKB_FCLONE_UNAVAILABLE) __kfree_skb(skb); else - __kfree_skb_defer(skb); + __napi_kfree_skb(skb, + get_kfree_skb_cb(skb)->reason); } } diff --git a/net/core/gro.c b/net/core/gro.c index a606705a0859..2d84165cb4f1 100644 --- a/net/core/gro.c +++ b/net/core/gro.c @@ -633,7 +633,7 @@ static gro_result_t napi_skb_finish(struct napi_struct *napi, else if (skb->fclone != SKB_FCLONE_UNAVAILABLE) __kfree_skb(skb); else - __kfree_skb_defer(skb); + __napi_kfree_skb(skb, SKB_CONSUMED); break; case GRO_HELD: diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 768f9d04911f..8764653bede7 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1226,9 +1226,9 @@ static void napi_skb_cache_put(struct sk_buff *skb) } } -void __kfree_skb_defer(struct sk_buff *skb) +void __napi_kfree_skb(struct sk_buff *skb, enum skb_drop_reason reason) { - skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED, true); + skb_release_all(skb, reason, true); napi_skb_cache_put(skb); } -- cgit v1.2.3 From 38e124086282715e3b9b3d193d9308bb3499b46c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 15 May 2022 18:16:54 -0400 Subject: kill the last remaining user of proc_ns_fget() lookups by descriptor are better off closer to syscall surface... Reviewed-by: Christian Brauner Signed-off-by: Al Viro --- fs/nsfs.c | 18 ------------------ include/linux/proc_ns.h | 1 - net/core/net_namespace.c | 23 +++++++++++------------ 3 files changed, 11 insertions(+), 31 deletions(-) (limited to 'net') diff --git a/fs/nsfs.c b/fs/nsfs.c index f8df60b3b901..f602a96a1afe 100644 --- a/fs/nsfs.c +++ b/fs/nsfs.c @@ -235,24 +235,6 @@ bool proc_ns_file(const struct file *file) return file->f_op == &ns_file_operations; } -struct file *proc_ns_fget(int fd) -{ - struct file *file; - - file = fget(fd); - if (!file) - return ERR_PTR(-EBADF); - - if (file->f_op != &ns_file_operations) - goto out_invalid; - - return file; - -out_invalid: - fput(file); - return ERR_PTR(-EINVAL); -} - /** * ns_match() - Returns true if current namespace matches dev/ino provided. * @ns: current namespace diff --git a/include/linux/proc_ns.h b/include/linux/proc_ns.h index 75807ecef880..49539bc416ce 100644 --- a/include/linux/proc_ns.h +++ b/include/linux/proc_ns.h @@ -72,7 +72,6 @@ static inline int ns_alloc_inum(struct ns_common *ns) #define ns_free_inum(ns) proc_free_inum((ns)->inum) -extern struct file *proc_ns_fget(int fd); #define get_proc_ns(inode) ((struct ns_common *)(inode)->i_private) extern int ns_get_path(struct path *path, struct task_struct *task, const struct proc_ns_operations *ns_ops); diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 7b69cf882b8e..3e3598cd49f2 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -676,21 +677,19 @@ EXPORT_SYMBOL_GPL(get_net_ns); struct net *get_net_ns_by_fd(int fd) { - struct file *file; - struct ns_common *ns; - struct net *net; + struct fd f = fdget(fd); + struct net *net = ERR_PTR(-EINVAL); - file = proc_ns_fget(fd); - if (IS_ERR(file)) - return ERR_CAST(file); + if (!f.file) + return ERR_PTR(-EBADF); - ns = get_proc_ns(file_inode(file)); - if (ns->ops == &netns_operations) - net = get_net(container_of(ns, struct net, ns)); - else - net = ERR_PTR(-EINVAL); + if (proc_ns_file(f.file)) { + struct ns_common *ns = get_proc_ns(file_inode(f.file)); + if (ns->ops == &netns_operations) + net = get_net(container_of(ns, struct net, ns)); + } + fdput(f); - fput(file); return net; } EXPORT_SYMBOL_GPL(get_net_ns_by_fd); -- cgit v1.2.3 From 35b288d6e3d467f6f7bf466fee6184b275999f14 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Tue, 18 Apr 2023 14:14:55 +0300 Subject: net: ethtool: mm: sanitize some UAPI configurations The verify-enabled boolean (ETHTOOL_A_MM_VERIFY_ENABLED) was intended to be a sub-setting of tx-enabled (ETHTOOL_A_MM_TX_ENABLED). IOW, MAC Merge TX can be enabled with or without verification, but verification with TX disabled makes no sense. The pmac-enabled boolean (ETHTOOL_A_MM_PMAC_ENABLED) was intended to be a global toggle from an API perspective, whereas tx-enabled just handles the TX direction. IOW, the pMAC can be enabled with or without TX, but it doesn't make sense to enable TX if the pMAC is not enabled. Add two checks which sanitize and reject these invalid cases. Signed-off-by: Vladimir Oltean Reviewed-by: Simon Horman Signed-off-by: Jakub Kicinski --- net/ethtool/mm.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'net') diff --git a/net/ethtool/mm.c b/net/ethtool/mm.c index e00d7d5cea7e..4058a557b5a4 100644 --- a/net/ethtool/mm.c +++ b/net/ethtool/mm.c @@ -214,6 +214,16 @@ static int ethnl_set_mm(struct ethnl_req_info *req_info, struct genl_info *info) return -ERANGE; } + if (cfg.verify_enabled && !cfg.tx_enabled) { + NL_SET_ERR_MSG(extack, "Verification requires TX enabled"); + return -EINVAL; + } + + if (cfg.tx_enabled && !cfg.pmac_enabled) { + NL_SET_ERR_MSG(extack, "TX enabled requires pMAC enabled"); + return -EINVAL; + } + ret = dev->ethtool_ops->set_mm(dev, &cfg, extack); return ret < 0 ? ret : 1; } -- cgit v1.2.3 From 7ab75456be144a354fbb3df1516d82fc24d3d67d Mon Sep 17 00:00:00 2001 From: Mahesh Bandewar Date: Tue, 18 Apr 2023 18:32:38 -0700 Subject: ipv6: add icmpv6_error_anycast_as_unicast for ICMPv6 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ICMPv6 error packets are not sent to the anycast destinations and this prevents things like traceroute from working. So create a setting similar to ECHO when dealing with Anycast sources (icmpv6_echo_ignore_anycast). Signed-off-by: Mahesh Bandewar Reviewed-by: David Ahern Reviewed-by: Maciej Żenczykowski Link: https://lore.kernel.org/r/20230419013238.2691167-1-maheshb@google.com Signed-off-by: Jakub Kicinski --- Documentation/networking/ip-sysctl.rst | 7 +++++++ include/net/netns/ipv6.h | 1 + net/ipv6/af_inet6.c | 1 + net/ipv6/icmp.c | 15 +++++++++++++-- 4 files changed, 22 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index 58a78a316697..6ec06a33688a 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -2721,6 +2721,13 @@ echo_ignore_anycast - BOOLEAN Default: 0 +error_anycast_as_unicast - BOOLEAN + If set to 1, then the kernel will respond with ICMP Errors + resulting from requests sent to it over the IPv6 protocol destined + to anycast address essentially treating anycast as unicast. + + Default: 0 + xfrm6_gc_thresh - INTEGER (Obsolete since linux-4.14) The threshold at which we will start garbage collecting for IPv6 diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index b4af4837d80b..3cceb3e9320b 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -55,6 +55,7 @@ struct netns_sysctl_ipv6 { u64 ioam6_id_wide; bool skip_notify_on_dev_down; u8 fib_notify_on_flag_change; + u8 icmpv6_error_anycast_as_unicast; }; struct netns_ipv6 { diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index e1b679a590c9..2bbf13216a3d 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -952,6 +952,7 @@ static int __net_init inet6_net_init(struct net *net) net->ipv6.sysctl.icmpv6_echo_ignore_all = 0; net->ipv6.sysctl.icmpv6_echo_ignore_multicast = 0; net->ipv6.sysctl.icmpv6_echo_ignore_anycast = 0; + net->ipv6.sysctl.icmpv6_error_anycast_as_unicast = 0; /* By default, rate limit error messages. * Except for pmtu discovery, it would break it. diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 1f53f2a74480..9edf1f45b1ed 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -362,9 +362,10 @@ static struct dst_entry *icmpv6_route_lookup(struct net *net, /* * We won't send icmp if the destination is known - * anycast. + * anycast unless we need to treat anycast as unicast. */ - if (ipv6_anycast_destination(dst, &fl6->daddr)) { + if (!READ_ONCE(net->ipv6.sysctl.icmpv6_error_anycast_as_unicast) && + ipv6_anycast_destination(dst, &fl6->daddr)) { net_dbg_ratelimited("icmp6_send: acast source\n"); dst_release(dst); return ERR_PTR(-EINVAL); @@ -1195,6 +1196,15 @@ static struct ctl_table ipv6_icmp_table_template[] = { .mode = 0644, .proc_handler = proc_do_large_bitmap, }, + { + .procname = "error_anycast_as_unicast", + .data = &init_net.ipv6.sysctl.icmpv6_error_anycast_as_unicast, + .maxlen = sizeof(u8), + .mode = 0644, + .proc_handler = proc_dou8vec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, { }, }; @@ -1212,6 +1222,7 @@ struct ctl_table * __net_init ipv6_icmp_sysctl_init(struct net *net) table[2].data = &net->ipv6.sysctl.icmpv6_echo_ignore_multicast; table[3].data = &net->ipv6.sysctl.icmpv6_echo_ignore_anycast; table[4].data = &net->ipv6.sysctl.icmpv6_ratemask_ptr; + table[5].data = &net->ipv6.sysctl.icmpv6_error_anycast_as_unicast; } return table; } -- cgit v1.2.3 From 071c0fc6fb919dcf29c676a842dda08a674877d7 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 19 Apr 2023 14:52:53 +0200 Subject: net: extend drop reasons for multiple subsystems Extend drop reasons to make them usable by subsystems other than core by reserving the high 16 bits for a new subsystem ID, of which 0 of course is used for the existing reasons immediately. To still be able to have string reasons, restructure that code a bit to make the loopup under RCU, the only user of this (right now) is drop_monitor. Link: https://lore.kernel.org/netdev/00659771ed54353f92027702c5bbb84702da62ce.camel@sipsolutions.net Signed-off-by: Johannes Berg Signed-off-by: Jakub Kicinski --- include/net/dropreason-core.h | 14 +++++++--- include/net/dropreason.h | 31 +++++++++++++++++++++++ net/core/drop_monitor.c | 33 +++++++++++++++++------- net/core/skbuff.c | 59 ++++++++++++++++++++++++++++++++++++++++--- 4 files changed, 121 insertions(+), 16 deletions(-) create mode 100644 include/net/dropreason.h (limited to 'net') diff --git a/include/net/dropreason-core.h b/include/net/dropreason-core.h index ade6d5b9186c..a2b953b57689 100644 --- a/include/net/dropreason-core.h +++ b/include/net/dropreason-core.h @@ -340,12 +340,20 @@ enum skb_drop_reason { */ SKB_DROP_REASON_IPV6_NDISC_NS_OTHERHOST, /** - * @SKB_DROP_REASON_MAX: the maximum of drop reason, which shouldn't be - * used as a real 'reason' + * @SKB_DROP_REASON_MAX: the maximum of core drop reasons, which + * shouldn't be used as a real 'reason' - only for tracing code gen */ SKB_DROP_REASON_MAX, + + /** + * @SKB_DROP_REASON_SUBSYS_MASK: subsystem mask in drop reasons, + * see &enum skb_drop_reason_subsys + */ + SKB_DROP_REASON_SUBSYS_MASK = 0xffff0000, }; +#define SKB_DROP_REASON_SUBSYS_SHIFT 16 + #define SKB_DR_INIT(name, reason) \ enum skb_drop_reason name = SKB_DROP_REASON_##reason #define SKB_DR(name) \ @@ -359,6 +367,4 @@ enum skb_drop_reason { SKB_DR_SET(name, reason); \ } while (0) -extern const char * const drop_reasons[]; - #endif diff --git a/include/net/dropreason.h b/include/net/dropreason.h new file mode 100644 index 000000000000..f0f2378dbed0 --- /dev/null +++ b/include/net/dropreason.h @@ -0,0 +1,31 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ + +#ifndef _LINUX_DROPREASON_H +#define _LINUX_DROPREASON_H +#include + +/** + * enum skb_drop_reason_subsys - subsystem tag for (extended) drop reasons + */ +enum skb_drop_reason_subsys { + /** @SKB_DROP_REASON_SUBSYS_CORE: core drop reasons defined above */ + SKB_DROP_REASON_SUBSYS_CORE, + + /** @SKB_DROP_REASON_SUBSYS_NUM: number of subsystems defined */ + SKB_DROP_REASON_SUBSYS_NUM +}; + +struct drop_reason_list { + const char * const *reasons; + size_t n_reasons; +}; + +/* Note: due to dynamic registrations, access must be under RCU */ +extern const struct drop_reason_list __rcu * +drop_reasons_by_subsys[SKB_DROP_REASON_SUBSYS_NUM]; + +void drop_reasons_register_subsys(enum skb_drop_reason_subsys subsys, + const struct drop_reason_list *list); +void drop_reasons_unregister_subsys(enum skb_drop_reason_subsys subsys); + +#endif diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c index 5a782d1d8fd3..aff31cd944c2 100644 --- a/net/core/drop_monitor.c +++ b/net/core/drop_monitor.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include @@ -29,6 +30,7 @@ #include #include #include +#include #include #include @@ -504,8 +506,6 @@ static void net_dm_packet_trace_kfree_skb_hit(void *ignore, if (!nskb) return; - if (unlikely(reason >= SKB_DROP_REASON_MAX || reason <= 0)) - reason = SKB_DROP_REASON_NOT_SPECIFIED; cb = NET_DM_SKB_CB(nskb); cb->reason = reason; cb->pc = location; @@ -552,9 +552,9 @@ static size_t net_dm_in_port_size(void) } #define NET_DM_MAX_SYMBOL_LEN 40 +#define NET_DM_MAX_REASON_LEN 50 -static size_t net_dm_packet_report_size(size_t payload_len, - enum skb_drop_reason reason) +static size_t net_dm_packet_report_size(size_t payload_len) { size_t size; @@ -576,7 +576,7 @@ static size_t net_dm_packet_report_size(size_t payload_len, /* NET_DM_ATTR_PROTO */ nla_total_size(sizeof(u16)) + /* NET_DM_ATTR_REASON */ - nla_total_size(strlen(drop_reasons[reason]) + 1) + + nla_total_size(NET_DM_MAX_REASON_LEN + 1) + /* NET_DM_ATTR_PAYLOAD */ nla_total_size(payload_len); } @@ -610,6 +610,8 @@ static int net_dm_packet_report_fill(struct sk_buff *msg, struct sk_buff *skb, size_t payload_len) { struct net_dm_skb_cb *cb = NET_DM_SKB_CB(skb); + const struct drop_reason_list *list = NULL; + unsigned int subsys, subsys_reason; char buf[NET_DM_MAX_SYMBOL_LEN]; struct nlattr *attr; void *hdr; @@ -627,9 +629,24 @@ static int net_dm_packet_report_fill(struct sk_buff *msg, struct sk_buff *skb, NET_DM_ATTR_PAD)) goto nla_put_failure; + rcu_read_lock(); + subsys = u32_get_bits(cb->reason, SKB_DROP_REASON_SUBSYS_MASK); + if (subsys < SKB_DROP_REASON_SUBSYS_NUM) + list = rcu_dereference(drop_reasons_by_subsys[subsys]); + subsys_reason = cb->reason & ~SKB_DROP_REASON_SUBSYS_MASK; + if (!list || + subsys_reason >= list->n_reasons || + !list->reasons[subsys_reason] || + strlen(list->reasons[subsys_reason]) > NET_DM_MAX_REASON_LEN) { + list = rcu_dereference(drop_reasons_by_subsys[SKB_DROP_REASON_SUBSYS_CORE]); + subsys_reason = SKB_DROP_REASON_NOT_SPECIFIED; + } if (nla_put_string(msg, NET_DM_ATTR_REASON, - drop_reasons[cb->reason])) + list->reasons[subsys_reason])) { + rcu_read_unlock(); goto nla_put_failure; + } + rcu_read_unlock(); snprintf(buf, sizeof(buf), "%pS", cb->pc); if (nla_put_string(msg, NET_DM_ATTR_SYMBOL, buf)) @@ -687,9 +704,7 @@ static void net_dm_packet_report(struct sk_buff *skb) if (net_dm_trunc_len) payload_len = min_t(size_t, net_dm_trunc_len, payload_len); - msg = nlmsg_new(net_dm_packet_report_size(payload_len, - NET_DM_SKB_CB(skb)->reason), - GFP_KERNEL); + msg = nlmsg_new(net_dm_packet_report_size(payload_len), GFP_KERNEL); if (!msg) goto out; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 8764653bede7..0d998806b377 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -58,6 +58,7 @@ #include #include #include +#include #include #include #include @@ -72,6 +73,7 @@ #include #include #include +#include #include #include @@ -122,11 +124,59 @@ EXPORT_SYMBOL(sysctl_max_skb_frags); #undef FN #define FN(reason) [SKB_DROP_REASON_##reason] = #reason, -const char * const drop_reasons[] = { +static const char * const drop_reasons[] = { [SKB_CONSUMED] = "CONSUMED", DEFINE_DROP_REASON(FN, FN) }; -EXPORT_SYMBOL(drop_reasons); + +static const struct drop_reason_list drop_reasons_core = { + .reasons = drop_reasons, + .n_reasons = ARRAY_SIZE(drop_reasons), +}; + +const struct drop_reason_list __rcu * +drop_reasons_by_subsys[SKB_DROP_REASON_SUBSYS_NUM] = { + [SKB_DROP_REASON_SUBSYS_CORE] = RCU_INITIALIZER(&drop_reasons_core), +}; +EXPORT_SYMBOL(drop_reasons_by_subsys); + +/** + * drop_reasons_register_subsys - register another drop reason subsystem + * @subsys: the subsystem to register, must not be the core + * @list: the list of drop reasons within the subsystem, must point to + * a statically initialized list + */ +void drop_reasons_register_subsys(enum skb_drop_reason_subsys subsys, + const struct drop_reason_list *list) +{ + if (WARN(subsys <= SKB_DROP_REASON_SUBSYS_CORE || + subsys >= ARRAY_SIZE(drop_reasons_by_subsys), + "invalid subsystem %d\n", subsys)) + return; + + /* must point to statically allocated memory, so INIT is OK */ + RCU_INIT_POINTER(drop_reasons_by_subsys[subsys], list); +} +EXPORT_SYMBOL_GPL(drop_reasons_register_subsys); + +/** + * drop_reasons_unregister_subsys - unregister a drop reason subsystem + * @subsys: the subsystem to remove, must not be the core + * + * Note: This will synchronize_rcu() to ensure no users when it returns. + */ +void drop_reasons_unregister_subsys(enum skb_drop_reason_subsys subsys) +{ + if (WARN(subsys <= SKB_DROP_REASON_SUBSYS_CORE || + subsys >= ARRAY_SIZE(drop_reasons_by_subsys), + "invalid subsystem %d\n", subsys)) + return; + + RCU_INIT_POINTER(drop_reasons_by_subsys[subsys], NULL); + + synchronize_rcu(); +} +EXPORT_SYMBOL_GPL(drop_reasons_unregister_subsys); /** * skb_panic - private function for out-of-line support @@ -986,7 +1036,10 @@ bool __kfree_skb_reason(struct sk_buff *skb, enum skb_drop_reason reason) if (unlikely(!skb_unref(skb))) return false; - DEBUG_NET_WARN_ON_ONCE(reason <= 0 || reason >= SKB_DROP_REASON_MAX); + DEBUG_NET_WARN_ON_ONCE(reason == SKB_NOT_DROPPED_YET || + u32_get_bits(reason, + SKB_DROP_REASON_SUBSYS_MASK) >= + SKB_DROP_REASON_SUBSYS_NUM); if (reason == SKB_CONSUMED) trace_consume_skb(skb, __builtin_return_address(0)); -- cgit v1.2.3 From baa951a1c1771810f3a378a95fc93e81953027d5 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 19 Apr 2023 14:52:54 +0200 Subject: mac80211: use the new drop reasons infrastructure It can be really hard to analyse or debug why packets are going missing in mac80211, so add the needed infrastructure to use use the new per-subsystem drop reasons. We actually use two drop reason subsystems here because of the different handling of frames that are dropped but still go to monitor for old versions of hostapd, and those that are just completely unusable (e.g. crypto failed.) Annotate a few reasons here just to illustrate this, we'll need to go through and annotate more of them later. Signed-off-by: Johannes Berg Signed-off-by: Jakub Kicinski --- include/net/dropreason.h | 12 ++++++++++ net/mac80211/drop.h | 56 ++++++++++++++++++++++++++++++++++++++++++++++ net/mac80211/ieee80211_i.h | 8 +------ net/mac80211/main.c | 31 +++++++++++++++++++++++++ net/mac80211/rx.c | 55 +++++++++++++++++++++------------------------ net/mac80211/wpa.c | 24 ++++++++++---------- 6 files changed, 138 insertions(+), 48 deletions(-) create mode 100644 net/mac80211/drop.h (limited to 'net') diff --git a/include/net/dropreason.h b/include/net/dropreason.h index f0f2378dbed0..685fb37df8e8 100644 --- a/include/net/dropreason.h +++ b/include/net/dropreason.h @@ -11,6 +11,18 @@ enum skb_drop_reason_subsys { /** @SKB_DROP_REASON_SUBSYS_CORE: core drop reasons defined above */ SKB_DROP_REASON_SUBSYS_CORE, + /** + * @SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE: mac80211 drop reasons + * for unusable frames, see net/mac80211/drop.h + */ + SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE, + + /** + * @SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR: mac80211 drop reasons + * for frames still going to monitor, see net/mac80211/drop.h + */ + SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR, + /** @SKB_DROP_REASON_SUBSYS_NUM: number of subsystems defined */ SKB_DROP_REASON_SUBSYS_NUM }; diff --git a/net/mac80211/drop.h b/net/mac80211/drop.h new file mode 100644 index 000000000000..49dc809cab29 --- /dev/null +++ b/net/mac80211/drop.h @@ -0,0 +1,56 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * mac80211 drop reason list + * + * Copyright (C) 2023 Intel Corporation + */ + +#ifndef MAC80211_DROP_H +#define MAC80211_DROP_H +#include + +typedef unsigned int __bitwise ieee80211_rx_result; + +#define MAC80211_DROP_REASONS_MONITOR(R) \ + R(RX_DROP_M_UNEXPECTED_4ADDR_FRAME) \ + R(RX_DROP_M_BAD_BCN_KEYIDX) \ + R(RX_DROP_M_BAD_MGMT_KEYIDX) \ +/* this line for the trailing \ - add before this */ + +#define MAC80211_DROP_REASONS_UNUSABLE(R) \ + R(RX_DROP_U_MIC_FAIL) \ + R(RX_DROP_U_REPLAY) \ + R(RX_DROP_U_BAD_MMIE) \ +/* this line for the trailing \ - add before this */ + +/* having two enums allows for checking ieee80211_rx_result use with sparse */ +enum ___mac80211_drop_reason { +/* if we get to the end of handlers with RX_CONTINUE this will be the reason */ + ___RX_CONTINUE = SKB_CONSUMED, + +/* this never gets used as an argument to kfree_skb_reason() */ + ___RX_QUEUED = SKB_NOT_DROPPED_YET, + +#define ENUM(x) ___ ## x, + ___RX_DROP_MONITOR = SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR << + SKB_DROP_REASON_SUBSYS_SHIFT, + MAC80211_DROP_REASONS_MONITOR(ENUM) + + ___RX_DROP_UNUSABLE = SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE << + SKB_DROP_REASON_SUBSYS_SHIFT, + MAC80211_DROP_REASONS_UNUSABLE(ENUM) +#undef ENUM +}; + +enum mac80211_drop_reason { + RX_CONTINUE = (__force ieee80211_rx_result)___RX_CONTINUE, + RX_QUEUED = (__force ieee80211_rx_result)___RX_QUEUED, + RX_DROP_MONITOR = (__force ieee80211_rx_result)___RX_DROP_MONITOR, + RX_DROP_UNUSABLE = (__force ieee80211_rx_result)___RX_DROP_UNUSABLE, +#define DEF(x) x = (__force ieee80211_rx_result)___ ## x, + MAC80211_DROP_REASONS_MONITOR(DEF) + MAC80211_DROP_REASONS_UNUSABLE(DEF) +#undef DEF +}; + +#endif /* MAC80211_DROP_H */ diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 9b7e184430b8..a0a7839cb961 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -33,6 +33,7 @@ #include "key.h" #include "sta_info.h" #include "debug.h" +#include "drop.h" extern const struct cfg80211_ops mac80211_config_ops; @@ -170,13 +171,6 @@ struct ieee80211_tx_data { unsigned int flags; }; - -typedef unsigned __bitwise ieee80211_rx_result; -#define RX_CONTINUE ((__force ieee80211_rx_result) 0u) -#define RX_DROP_UNUSABLE ((__force ieee80211_rx_result) 1u) -#define RX_DROP_MONITOR ((__force ieee80211_rx_result) 2u) -#define RX_QUEUED ((__force ieee80211_rx_result) 3u) - /** * enum ieee80211_packet_rx_flags - packet RX flags * @IEEE80211_RX_AMSDU: a-MSDU packet diff --git a/net/mac80211/main.c b/net/mac80211/main.c index ddf2b7811c55..55cdfaef0f5d 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include @@ -1542,6 +1543,28 @@ void ieee80211_free_hw(struct ieee80211_hw *hw) } EXPORT_SYMBOL(ieee80211_free_hw); +static const char * const drop_reasons_monitor[] = { +#define V(x) #x, + [0] = "RX_DROP_MONITOR", + MAC80211_DROP_REASONS_MONITOR(V) +}; + +static struct drop_reason_list drop_reason_list_monitor = { + .reasons = drop_reasons_monitor, + .n_reasons = ARRAY_SIZE(drop_reasons_monitor), +}; + +static const char * const drop_reasons_unusable[] = { + [0] = "RX_DROP_UNUSABLE", + MAC80211_DROP_REASONS_UNUSABLE(V) +#undef V +}; + +static struct drop_reason_list drop_reason_list_unusable = { + .reasons = drop_reasons_unusable, + .n_reasons = ARRAY_SIZE(drop_reasons_unusable), +}; + static int __init ieee80211_init(void) { struct sk_buff *skb; @@ -1559,6 +1582,11 @@ static int __init ieee80211_init(void) if (ret) goto err_netdev; + drop_reasons_register_subsys(SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR, + &drop_reason_list_monitor); + drop_reasons_register_subsys(SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE, + &drop_reason_list_unusable); + return 0; err_netdev: rc80211_minstrel_exit(); @@ -1574,6 +1602,9 @@ static void __exit ieee80211_exit(void) ieee80211_iface_exit(); + drop_reasons_unregister_subsys(SKB_DROP_REASON_SUBSYS_MAC80211_MONITOR); + drop_reasons_unregister_subsys(SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE); + rcu_barrier(); } diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index db3451f5f2fb..58222c077898 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1826,7 +1826,7 @@ ieee80211_rx_h_sta_process(struct ieee80211_rx_data *rx) cfg80211_rx_unexpected_4addr_frame( rx->sdata->dev, sta->sta.addr, GFP_ATOMIC); - return RX_DROP_MONITOR; + return RX_DROP_M_UNEXPECTED_4ADDR_FRAME; } /* * Update counter and free packet here to avoid @@ -1961,7 +1961,7 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) cfg80211_rx_unprot_mlme_mgmt(rx->sdata->dev, skb->data, skb->len); - return RX_DROP_MONITOR; /* unexpected BIP keyidx */ + return RX_DROP_M_BAD_BCN_KEYIDX; } rx->key = ieee80211_rx_get_bigtk(rx, mmie_keyidx); @@ -1975,7 +1975,7 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) if (mmie_keyidx < NUM_DEFAULT_KEYS || mmie_keyidx >= NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS) - return RX_DROP_MONITOR; /* unexpected BIP keyidx */ + return RX_DROP_M_BAD_MGMT_KEYIDX; /* unexpected BIP keyidx */ if (rx->link_sta) { if (ieee80211_is_group_privacy_action(skb) && test_sta_flag(rx->sta, WLAN_STA_MFP)) @@ -3960,7 +3960,8 @@ ieee80211_rx_h_mgmt(struct ieee80211_rx_data *rx) } static void ieee80211_rx_cooked_monitor(struct ieee80211_rx_data *rx, - struct ieee80211_rate *rate) + struct ieee80211_rate *rate, + ieee80211_rx_result reason) { struct ieee80211_sub_if_data *sdata; struct ieee80211_local *local = rx->local; @@ -4024,42 +4025,38 @@ static void ieee80211_rx_cooked_monitor(struct ieee80211_rx_data *rx, } out_free_skb: - dev_kfree_skb(skb); + kfree_skb_reason(skb, (__force u32)reason); } static void ieee80211_rx_handlers_result(struct ieee80211_rx_data *rx, ieee80211_rx_result res) { - switch (res) { - case RX_DROP_MONITOR: - I802_DEBUG_INC(rx->sdata->local->rx_handlers_drop); - if (rx->sta) - rx->link_sta->rx_stats.dropped++; - fallthrough; - case RX_CONTINUE: { - struct ieee80211_rate *rate = NULL; - struct ieee80211_supported_band *sband; - struct ieee80211_rx_status *status; - - status = IEEE80211_SKB_RXCB((rx->skb)); + struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(rx->skb); + struct ieee80211_supported_band *sband; + struct ieee80211_rate *rate = NULL; - sband = rx->local->hw.wiphy->bands[status->band]; - if (status->encoding == RX_ENC_LEGACY) - rate = &sband->bitrates[status->rate_idx]; + if (res == RX_QUEUED) { + I802_DEBUG_INC(rx->sdata->local->rx_handlers_queued); + return; + } - ieee80211_rx_cooked_monitor(rx, rate); - break; - } - case RX_DROP_UNUSABLE: + if (res != RX_CONTINUE) { I802_DEBUG_INC(rx->sdata->local->rx_handlers_drop); if (rx->sta) rx->link_sta->rx_stats.dropped++; - dev_kfree_skb(rx->skb); - break; - case RX_QUEUED: - I802_DEBUG_INC(rx->sdata->local->rx_handlers_queued); - break; } + + if (u32_get_bits((__force u32)res, SKB_DROP_REASON_SUBSYS_MASK) == + SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE) { + kfree_skb_reason(rx->skb, (__force u32)res); + return; + } + + sband = rx->local->hw.wiphy->bands[status->band]; + if (status->encoding == RX_ENC_LEGACY) + rate = &sband->bitrates[status->rate_idx]; + + ieee80211_rx_cooked_monitor(rx, rate, res); } static void ieee80211_rx_handlers(struct ieee80211_rx_data *rx, diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c index 20f742b5503b..4133496da378 100644 --- a/net/mac80211/wpa.c +++ b/net/mac80211/wpa.c @@ -550,7 +550,7 @@ ieee80211_crypto_ccmp_decrypt(struct ieee80211_rx_data *rx, if (res < 0 || (!res && !(status->flag & RX_FLAG_ALLOW_SAME_PN))) { key->u.ccmp.replays++; - return RX_DROP_UNUSABLE; + return RX_DROP_U_REPLAY; } if (!(status->flag & RX_FLAG_DECRYPTED)) { @@ -564,7 +564,7 @@ ieee80211_crypto_ccmp_decrypt(struct ieee80211_rx_data *rx, skb->data + hdrlen + IEEE80211_CCMP_HDR_LEN, data_len, skb->data + skb->len - mic_len)) - return RX_DROP_UNUSABLE; + return RX_DROP_U_MIC_FAIL; } memcpy(key->u.ccmp.rx_pn[queue], pn, IEEE80211_CCMP_PN_LEN); @@ -746,7 +746,7 @@ ieee80211_crypto_gcmp_decrypt(struct ieee80211_rx_data *rx) if (res < 0 || (!res && !(status->flag & RX_FLAG_ALLOW_SAME_PN))) { key->u.gcmp.replays++; - return RX_DROP_UNUSABLE; + return RX_DROP_U_REPLAY; } if (!(status->flag & RX_FLAG_DECRYPTED)) { @@ -761,7 +761,7 @@ ieee80211_crypto_gcmp_decrypt(struct ieee80211_rx_data *rx) data_len, skb->data + skb->len - IEEE80211_GCMP_MIC_LEN)) - return RX_DROP_UNUSABLE; + return RX_DROP_U_MIC_FAIL; } memcpy(key->u.gcmp.rx_pn[queue], pn, IEEE80211_GCMP_PN_LEN); @@ -930,13 +930,13 @@ ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx) (skb->data + skb->len - sizeof(*mmie)); if (mmie->element_id != WLAN_EID_MMIE || mmie->length != sizeof(*mmie) - 2) - return RX_DROP_UNUSABLE; /* Invalid MMIE */ + return RX_DROP_U_BAD_MMIE; /* Invalid MMIE */ bip_ipn_swap(ipn, mmie->sequence_number); if (memcmp(ipn, key->u.aes_cmac.rx_pn, 6) <= 0) { key->u.aes_cmac.replays++; - return RX_DROP_UNUSABLE; + return RX_DROP_U_REPLAY; } if (!(status->flag & RX_FLAG_DECRYPTED)) { @@ -946,7 +946,7 @@ ieee80211_crypto_aes_cmac_decrypt(struct ieee80211_rx_data *rx) skb->data + 24, skb->len - 24, mic); if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) { key->u.aes_cmac.icverrors++; - return RX_DROP_UNUSABLE; + return RX_DROP_U_MIC_FAIL; } } @@ -986,7 +986,7 @@ ieee80211_crypto_aes_cmac_256_decrypt(struct ieee80211_rx_data *rx) if (memcmp(ipn, key->u.aes_cmac.rx_pn, 6) <= 0) { key->u.aes_cmac.replays++; - return RX_DROP_UNUSABLE; + return RX_DROP_U_REPLAY; } if (!(status->flag & RX_FLAG_DECRYPTED)) { @@ -996,7 +996,7 @@ ieee80211_crypto_aes_cmac_256_decrypt(struct ieee80211_rx_data *rx) skb->data + 24, skb->len - 24, mic); if (crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) { key->u.aes_cmac.icverrors++; - return RX_DROP_UNUSABLE; + return RX_DROP_U_MIC_FAIL; } } @@ -1079,13 +1079,13 @@ ieee80211_crypto_aes_gmac_decrypt(struct ieee80211_rx_data *rx) (skb->data + skb->len - sizeof(*mmie)); if (mmie->element_id != WLAN_EID_MMIE || mmie->length != sizeof(*mmie) - 2) - return RX_DROP_UNUSABLE; /* Invalid MMIE */ + return RX_DROP_U_BAD_MMIE; /* Invalid MMIE */ bip_ipn_swap(ipn, mmie->sequence_number); if (memcmp(ipn, key->u.aes_gmac.rx_pn, 6) <= 0) { key->u.aes_gmac.replays++; - return RX_DROP_UNUSABLE; + return RX_DROP_U_REPLAY; } if (!(status->flag & RX_FLAG_DECRYPTED)) { @@ -1104,7 +1104,7 @@ ieee80211_crypto_aes_gmac_decrypt(struct ieee80211_rx_data *rx) crypto_memneq(mic, mmie->mic, sizeof(mmie->mic))) { key->u.aes_gmac.icverrors++; kfree(mic); - return RX_DROP_UNUSABLE; + return RX_DROP_U_MIC_FAIL; } kfree(mic); } -- cgit v1.2.3 From 94b95dfaa814f565d92f5a65f0ff12a483095522 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Wed, 19 Apr 2023 15:19:07 +0300 Subject: xfrm: release all offloaded policy memory Failure to add offloaded policy will cause to the following error once user will try to reload driver. Unregister_netdevice: waiting for eth3 to become free. Usage count = 2 This was caused by xfrm_dev_policy_add() which increments reference to net_device. That reference was supposed to be decremented in xfrm_dev_policy_free(). However the latter wasn't called. unregister_netdevice: waiting for eth3 to become free. Usage count = 2 leaked reference. xfrm_dev_policy_add+0xff/0x3d0 xfrm_policy_construct+0x352/0x420 xfrm_add_policy+0x179/0x320 xfrm_user_rcv_msg+0x1d2/0x3d0 netlink_rcv_skb+0xe0/0x210 xfrm_netlink_rcv+0x45/0x50 netlink_unicast+0x346/0x490 netlink_sendmsg+0x3b0/0x6c0 sock_sendmsg+0x73/0xc0 sock_write_iter+0x13b/0x1f0 vfs_write+0x528/0x5d0 ksys_write+0x120/0x150 do_syscall_64+0x3d/0x90 entry_SYSCALL_64_after_hwframe+0x46/0xb0 Fixes: 919e43fad516 ("xfrm: add an interface to offload policy") Signed-off-by: Leon Romanovsky Reviewed-by: Simon Horman Reviewed-by: Eric Dumazet Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 103af2b3e986..af8fbcbfbe69 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1978,6 +1978,7 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, if (err) { xfrm_dev_policy_delete(xp); + xfrm_dev_policy_free(xp); security_xfrm_policy_free(xp->security); kfree(xp); return err; -- cgit v1.2.3 From ec8f32ad9a65a8cbb465b69e154aaec9d2fe45c4 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Wed, 19 Apr 2023 15:19:08 +0300 Subject: xfrm: Fix leak of dev tracker At the stage of direction checks, the netdev reference tracker is already initialized, but released with wrong *_put() call. Fixes: 919e43fad516 ("xfrm: add an interface to offload policy") Signed-off-by: Leon Romanovsky Reviewed-by: Simon Horman Reviewed-by: Eric Dumazet Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c index 95f1436bf6a2..e2ca50bfca24 100644 --- a/net/xfrm/xfrm_device.c +++ b/net/xfrm/xfrm_device.c @@ -378,7 +378,7 @@ int xfrm_dev_policy_add(struct net *net, struct xfrm_policy *xp, break; default: xdo->dev = NULL; - dev_put(dev); + netdev_put(dev, &xdo->dev_tracker); NL_SET_ERR_MSG(extack, "Unrecognized offload direction"); return -EINVAL; } -- cgit v1.2.3 From add7370a398930077c6bc257ef5016b040d476eb Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 19 Apr 2023 11:16:28 -0400 Subject: sctp: delete the nested flexible array params This patch deletes the flexible-array params[] from the structure sctp_inithdr, sctp_addiphdr and sctp_reconf_chunk to avoid some sparse warnings: # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/ net/sctp/input.c: note: in included file (through include/net/sctp/structs.h, include/net/sctp/sctp.h): ./include/linux/sctp.h:278:29: warning: nested flexible array ./include/linux/sctp.h:675:30: warning: nested flexible array This warning is reported if a structure having a flexible array member is included by other structures. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/linux/sctp.h | 6 +++--- include/net/sctp/sctp.h | 8 ++++---- net/sctp/input.c | 2 +- net/sctp/sm_make_chunk.c | 18 +++++++++--------- net/sctp/sm_statefuns.c | 2 +- net/sctp/stream.c | 2 +- 6 files changed, 19 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/include/linux/sctp.h b/include/linux/sctp.h index 358dc08e0831..0ff36a2737a3 100644 --- a/include/linux/sctp.h +++ b/include/linux/sctp.h @@ -270,7 +270,7 @@ struct sctp_inithdr { __be16 num_outbound_streams; __be16 num_inbound_streams; __be32 initial_tsn; - __u8 params[]; + /* __u8 params[]; */ }; struct sctp_init_chunk { @@ -667,7 +667,7 @@ struct sctp_addip_param { struct sctp_addiphdr { __be32 serial; - __u8 params[]; + /* __u8 params[]; */ }; struct sctp_addip_chunk { @@ -742,7 +742,7 @@ struct sctp_infox { struct sctp_reconf_chunk { struct sctp_chunkhdr chunk_hdr; - __u8 params[]; + /* __u8 params[]; */ }; struct sctp_strreset_outreq { diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index c335dd01a597..74fae532b944 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -425,11 +425,11 @@ static inline bool sctp_chunk_pending(const struct sctp_chunk *chunk) * the chunk length to indicate when to stop. Make sure * there is room for a param header too. */ -#define sctp_walk_params(pos, chunk, member)\ -_sctp_walk_params((pos), (chunk), ntohs((chunk)->chunk_hdr.length), member) +#define sctp_walk_params(pos, chunk)\ +_sctp_walk_params((pos), (chunk), ntohs((chunk)->chunk_hdr.length)) -#define _sctp_walk_params(pos, chunk, end, member)\ -for (pos.v = chunk->member;\ +#define _sctp_walk_params(pos, chunk, end)\ +for (pos.v = (u8 *)(chunk + 1);\ (pos.v + offsetof(struct sctp_paramhdr, length) + sizeof(pos.p->length) <=\ (void *)chunk + end) &&\ pos.v <= (void *)chunk + end - ntohs(pos.p->length) &&\ diff --git a/net/sctp/input.c b/net/sctp/input.c index 127bf28a6033..2613c4d74b16 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -1150,7 +1150,7 @@ static struct sctp_association *__sctp_rcv_init_lookup(struct net *net, init = (struct sctp_init_chunk *)skb->data; /* Walk the parameters looking for embedded addresses. */ - sctp_walk_params(params, init, init_hdr.params) { + sctp_walk_params(params, init) { /* Note: Ignoring hostname addresses. */ af = sctp_get_af_specific(param_type2af(params.p->type)); diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index c8f4ec5d5f98..4dbbbc2a7742 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -2306,7 +2306,7 @@ int sctp_verify_init(struct net *net, const struct sctp_endpoint *ep, ntohl(peer_init->init_hdr.a_rwnd) < SCTP_DEFAULT_MINWINDOW) return sctp_process_inv_mandatory(asoc, chunk, errp); - sctp_walk_params(param, peer_init, init_hdr.params) { + sctp_walk_params(param, peer_init) { if (param.p->type == SCTP_PARAM_STATE_COOKIE) has_cookie = true; } @@ -2329,7 +2329,7 @@ int sctp_verify_init(struct net *net, const struct sctp_endpoint *ep, chunk, errp); /* Verify all the variable length parameters */ - sctp_walk_params(param, peer_init, init_hdr.params) { + sctp_walk_params(param, peer_init) { result = sctp_verify_param(net, ep, asoc, param, cid, chunk, errp); switch (result) { @@ -2381,7 +2381,7 @@ int sctp_process_init(struct sctp_association *asoc, struct sctp_chunk *chunk, src_match = 1; /* Process the initialization parameters. */ - sctp_walk_params(param, peer_init, init_hdr.params) { + sctp_walk_params(param, peer_init) { if (!src_match && (param.p->type == SCTP_PARAM_IPV4_ADDRESS || param.p->type == SCTP_PARAM_IPV6_ADDRESS)) { @@ -3202,7 +3202,7 @@ bool sctp_verify_asconf(const struct sctp_association *asoc, union sctp_params param; addip = (struct sctp_addip_chunk *)chunk->chunk_hdr; - sctp_walk_params(param, addip, addip_hdr.params) { + sctp_walk_params(param, addip) { size_t length = ntohs(param.p->length); *errp = param.p; @@ -3215,14 +3215,14 @@ bool sctp_verify_asconf(const struct sctp_association *asoc, /* ensure there is only one addr param and it's in the * beginning of addip_hdr params, or we reject it. */ - if (param.v != addip->addip_hdr.params) + if (param.v != (addip + 1)) return false; addr_param_seen = true; break; case SCTP_PARAM_IPV6_ADDRESS: if (length != sizeof(struct sctp_ipv6addr_param)) return false; - if (param.v != addip->addip_hdr.params) + if (param.v != (addip + 1)) return false; addr_param_seen = true; break; @@ -3302,7 +3302,7 @@ struct sctp_chunk *sctp_process_asconf(struct sctp_association *asoc, goto done; /* Process the TLVs contained within the ASCONF chunk. */ - sctp_walk_params(param, addip, addip_hdr.params) { + sctp_walk_params(param, addip) { /* Skip preceeding address parameters. */ if (param.p->type == SCTP_PARAM_IPV4_ADDRESS || param.p->type == SCTP_PARAM_IPV6_ADDRESS) @@ -3636,7 +3636,7 @@ static struct sctp_chunk *sctp_make_reconf(const struct sctp_association *asoc, return NULL; reconf = (struct sctp_reconf_chunk *)retval->chunk_hdr; - retval->param_hdr.v = reconf->params; + retval->param_hdr.v = (u8 *)(reconf + 1); return retval; } @@ -3878,7 +3878,7 @@ bool sctp_verify_reconf(const struct sctp_association *asoc, __u16 cnt = 0; hdr = (struct sctp_reconf_chunk *)chunk->chunk_hdr; - sctp_walk_params(param, hdr, params) { + sctp_walk_params(param, hdr) { __u16 length = ntohs(param.p->length); *errp = param.p; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index ce5426171206..39d416e7f795 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -4142,7 +4142,7 @@ enum sctp_disposition sctp_sf_do_reconf(struct net *net, (void *)err_param, commands); hdr = (struct sctp_reconf_chunk *)chunk->chunk_hdr; - sctp_walk_params(param, hdr, params) { + sctp_walk_params(param, hdr) { struct sctp_chunk *reply = NULL; struct sctp_ulpevent *ev = NULL; diff --git a/net/sctp/stream.c b/net/sctp/stream.c index ee6514af830f..c241cc552e8d 100644 --- a/net/sctp/stream.c +++ b/net/sctp/stream.c @@ -491,7 +491,7 @@ static struct sctp_paramhdr *sctp_chunk_lookup_strreset_param( return NULL; hdr = (struct sctp_reconf_chunk *)chunk->chunk_hdr; - sctp_walk_params(param, hdr, params) { + sctp_walk_params(param, hdr) { /* sctp_strreset_tsnreq is actually the basic structure * of all stream reconf params, so it's safe to use it * to access request_seq. -- cgit v1.2.3 From 73175a042955e531ec355a8708585befa67a22db Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 19 Apr 2023 11:16:29 -0400 Subject: sctp: delete the nested flexible array skip This patch deletes the flexible-array skip[] from the structure sctp_ifwdtsn/fwdtsn_hdr to avoid some sparse warnings: # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/ net/sctp/stream_interleave.c: note: in included file (through include/net/sctp/structs.h, include/net/sctp/sctp.h): ./include/linux/sctp.h:611:32: warning: nested flexible array ./include/linux/sctp.h:628:33: warning: nested flexible array Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/linux/sctp.h | 4 ++-- include/net/sctp/sctp.h | 4 ++-- net/sctp/stream_interleave.c | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/linux/sctp.h b/include/linux/sctp.h index 0ff36a2737a3..9815b801fec0 100644 --- a/include/linux/sctp.h +++ b/include/linux/sctp.h @@ -603,7 +603,7 @@ struct sctp_fwdtsn_skip { struct sctp_fwdtsn_hdr { __be32 new_cum_tsn; - struct sctp_fwdtsn_skip skip[]; + /* struct sctp_fwdtsn_skip skip[]; */ }; struct sctp_fwdtsn_chunk { @@ -620,7 +620,7 @@ struct sctp_ifwdtsn_skip { struct sctp_ifwdtsn_hdr { __be32 new_cum_tsn; - struct sctp_ifwdtsn_skip skip[]; + /* struct sctp_ifwdtsn_skip skip[]; */ }; struct sctp_ifwdtsn_chunk { diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 74fae532b944..2a67100b2a17 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -452,8 +452,8 @@ for (err = (struct sctp_errhdr *)((void *)chunk_hdr + \ _sctp_walk_fwdtsn((pos), (chunk), ntohs((chunk)->chunk_hdr->length) - sizeof(struct sctp_fwdtsn_chunk)) #define _sctp_walk_fwdtsn(pos, chunk, end)\ -for (pos = chunk->subh.fwdtsn_hdr->skip;\ - (void *)pos <= (void *)chunk->subh.fwdtsn_hdr->skip + end - sizeof(struct sctp_fwdtsn_skip);\ +for (pos = (void *)(chunk->subh.fwdtsn_hdr + 1);\ + (void *)pos <= (void *)(chunk->subh.fwdtsn_hdr + 1) + end - sizeof(struct sctp_fwdtsn_skip);\ pos++) /* External references. */ diff --git a/net/sctp/stream_interleave.c b/net/sctp/stream_interleave.c index b046b11200c9..840f24045ae2 100644 --- a/net/sctp/stream_interleave.c +++ b/net/sctp/stream_interleave.c @@ -1153,8 +1153,8 @@ static void sctp_generate_iftsn(struct sctp_outq *q, __u32 ctsn) } #define _sctp_walk_ifwdtsn(pos, chunk, end) \ - for (pos = chunk->subh.ifwdtsn_hdr->skip; \ - (void *)pos <= (void *)chunk->subh.ifwdtsn_hdr->skip + (end) - \ + for (pos = (void *)(chunk->subh.ifwdtsn_hdr + 1); \ + (void *)pos <= (void *)(chunk->subh.ifwdtsn_hdr + 1) + (end) - \ sizeof(struct sctp_ifwdtsn_skip); pos++) #define sctp_walk_ifwdtsn(pos, ch) \ -- cgit v1.2.3 From 9789c1c6619e0a5eccfc31abe49b1ce5ca3cd11f Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 19 Apr 2023 11:16:30 -0400 Subject: sctp: delete the nested flexible array variable This patch deletes the flexible-array variable[] from the structure sctp_sackhdr and sctp_errhdr to avoid some sparse warnings: # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/ net/sctp/sm_statefuns.c: note: in included file (through include/net/sctp/structs.h, include/net/sctp/sctp.h): ./include/linux/sctp.h:451:28: warning: nested flexible array ./include/linux/sctp.h:393:29: warning: nested flexible array Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/linux/sctp.h | 4 ++-- net/sctp/outqueue.c | 11 +++++++---- net/sctp/sm_sideeffect.c | 3 +-- net/sctp/sm_statefuns.c | 2 +- 4 files changed, 11 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/include/linux/sctp.h b/include/linux/sctp.h index 9815b801fec0..01a0eb7e9fa1 100644 --- a/include/linux/sctp.h +++ b/include/linux/sctp.h @@ -385,7 +385,7 @@ struct sctp_sackhdr { __be32 a_rwnd; __be16 num_gap_ack_blocks; __be16 num_dup_tsns; - union sctp_sack_variable variable[]; + /* union sctp_sack_variable variable[]; */ }; struct sctp_sack_chunk { @@ -443,7 +443,7 @@ struct sctp_shutdown_chunk { struct sctp_errhdr { __be16 cause; __be16 length; - __u8 variable[]; + /* __u8 variable[]; */ }; struct sctp_operr_chunk { diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index 20831079fb09..0dc6b8ab9963 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -1231,7 +1231,7 @@ static void sctp_sack_update_unack_data(struct sctp_association *assoc, unack_data = assoc->next_tsn - assoc->ctsn_ack_point - 1; - frags = sack->variable; + frags = (union sctp_sack_variable *)(sack + 1); for (i = 0; i < ntohs(sack->num_gap_ack_blocks); i++) { unack_data -= ((ntohs(frags[i].gab.end) - ntohs(frags[i].gab.start) + 1)); @@ -1252,7 +1252,6 @@ int sctp_outq_sack(struct sctp_outq *q, struct sctp_chunk *chunk) struct sctp_transport *transport; struct sctp_chunk *tchunk = NULL; struct list_head *lchunk, *transport_list, *temp; - union sctp_sack_variable *frags = sack->variable; __u32 sack_ctsn, ctsn, tsn; __u32 highest_tsn, highest_new_tsn; __u32 sack_a_rwnd; @@ -1313,8 +1312,12 @@ int sctp_outq_sack(struct sctp_outq *q, struct sctp_chunk *chunk) /* Get the highest TSN in the sack. */ highest_tsn = sack_ctsn; - if (gap_ack_blocks) + if (gap_ack_blocks) { + union sctp_sack_variable *frags = + (union sctp_sack_variable *)(sack + 1); + highest_tsn += ntohs(frags[gap_ack_blocks - 1].gab.end); + } if (TSN_lt(asoc->highest_sacked, highest_tsn)) asoc->highest_sacked = highest_tsn; @@ -1789,7 +1792,7 @@ static int sctp_acked(struct sctp_sackhdr *sack, __u32 tsn) * Block are assumed to have been received correctly. */ - frags = sack->variable; + frags = (union sctp_sack_variable *)(sack + 1); blocks = ntohs(sack->num_gap_ack_blocks); tsn_offset = tsn - ctsn; for (i = 0; i < blocks; ++i) { diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 463c4a58d2c3..7fbeb99d8d32 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -984,8 +984,7 @@ static void sctp_cmd_process_operr(struct sctp_cmd_seq *cmds, { struct sctp_chunkhdr *unk_chunk_hdr; - unk_chunk_hdr = (struct sctp_chunkhdr *) - err_hdr->variable; + unk_chunk_hdr = (struct sctp_chunkhdr *)(err_hdr + 1); switch (unk_chunk_hdr->type) { /* ADDIP 4.1 A9) If the peer responds to an ASCONF with * an ERROR chunk reporting that it did not recognized diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 39d416e7f795..8d0cfd689b20 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -1337,7 +1337,7 @@ static int sctp_sf_send_restart_abort(struct net *net, union sctp_addr *ssa, * throughout the code today. */ errhdr = (struct sctp_errhdr *)buffer; - addrparm = (union sctp_addr_param *)errhdr->variable; + addrparm = (union sctp_addr_param *)(errhdr + 1); /* Copy into a parm format. */ len = af->to_addr_param(ssa, addrparm); -- cgit v1.2.3 From f97278ff346a5f11c68d0ac499999d5ad06a6db2 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 19 Apr 2023 11:16:31 -0400 Subject: sctp: delete the nested flexible array peer_init This patch deletes the flexible-array peer_init[] from the structure sctp_cookie to avoid some sparse warnings: # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/ net/sctp/sm_make_chunk.c: note: in included file (through include/net/sctp/sctp.h): ./include/net/sctp/structs.h:1588:28: warning: nested flexible array ./include/net/sctp/structs.h:343:28: warning: nested flexible array Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 2 +- net/sctp/associola.c | 5 +++-- net/sctp/sm_make_chunk.c | 4 ++-- net/sctp/sm_statefuns.c | 8 +++----- 4 files changed, 9 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 070c9458fff4..5c72d1864dd6 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -332,7 +332,7 @@ struct sctp_cookie { * the association TCB is re-constructed from the cookie. */ __u32 raw_addr_list_len; - struct sctp_init_chunk peer_init[]; + /* struct sctp_init_chunk peer_init[]; */ }; diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 63ba5551c13f..796529167e8d 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -1597,9 +1597,10 @@ int sctp_assoc_set_bind_addr_from_cookie(struct sctp_association *asoc, struct sctp_cookie *cookie, gfp_t gfp) { - int var_size2 = ntohs(cookie->peer_init->chunk_hdr.length); + struct sctp_init_chunk *peer_init = (struct sctp_init_chunk *)(cookie + 1); + int var_size2 = ntohs(peer_init->chunk_hdr.length); int var_size3 = cookie->raw_addr_list_len; - __u8 *raw = (__u8 *)cookie->peer_init + var_size2; + __u8 *raw = (__u8 *)peer_init + var_size2; return sctp_raw_to_bind_addrs(&asoc->base.bind_addr, raw, var_size3, asoc->ep->base.bind_addr.port, gfp); diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 4dbbbc2a7742..08527d882e56 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1707,11 +1707,11 @@ static struct sctp_cookie_param *sctp_pack_cookie( ktime_get_real()); /* Copy the peer's init packet. */ - memcpy(&cookie->c.peer_init[0], init_chunk->chunk_hdr, + memcpy(cookie + 1, init_chunk->chunk_hdr, ntohs(init_chunk->chunk_hdr->length)); /* Copy the raw local address list of the association. */ - memcpy((__u8 *)&cookie->c.peer_init[0] + + memcpy((__u8 *)(cookie + 1) + ntohs(init_chunk->chunk_hdr->length), raw_addrs, addrs_len); if (sctp_sk(ep->base.sk)->hmac) { diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 8d0cfd689b20..7b8eb735fa88 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -794,8 +794,7 @@ enum sctp_disposition sctp_sf_do_5_1D_ce(struct net *net, /* This is a brand-new association, so these are not yet side * effects--it is safe to run them here. */ - peer_init = &chunk->subh.cookie_hdr->c.peer_init[0]; - + peer_init = (struct sctp_init_chunk *)(chunk->subh.cookie_hdr + 1); if (!sctp_process_init(new_asoc, chunk, &chunk->subh.cookie_hdr->c.peer_addr, peer_init, GFP_ATOMIC)) @@ -1869,8 +1868,7 @@ static enum sctp_disposition sctp_sf_do_dupcook_a( /* new_asoc is a brand-new association, so these are not yet * side effects--it is safe to run them here. */ - peer_init = &chunk->subh.cookie_hdr->c.peer_init[0]; - + peer_init = (struct sctp_init_chunk *)(chunk->subh.cookie_hdr + 1); if (!sctp_process_init(new_asoc, chunk, sctp_source(chunk), peer_init, GFP_ATOMIC)) goto nomem; @@ -1990,7 +1988,7 @@ static enum sctp_disposition sctp_sf_do_dupcook_b( /* new_asoc is a brand-new association, so these are not yet * side effects--it is safe to run them here. */ - peer_init = &chunk->subh.cookie_hdr->c.peer_init[0]; + peer_init = (struct sctp_init_chunk *)(chunk->subh.cookie_hdr + 1); if (!sctp_process_init(new_asoc, chunk, sctp_source(chunk), peer_init, GFP_ATOMIC)) goto nomem; -- cgit v1.2.3 From 2ab399a931dddacdf7202cd4b49a5187154623d1 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 19 Apr 2023 11:16:32 -0400 Subject: sctp: delete the nested flexible array hmac This patch deletes the flexible-array hmac[] from the structure sctp_authhdr to avoid some sparse warnings: # make C=2 CF="-Wflexible-array-nested" M=./net/sctp/ net/sctp/auth.c: note: in included file (through include/net/sctp/structs.h, include/net/sctp/sctp.h): ./include/linux/sctp.h:735:29: warning: nested flexible array Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/linux/sctp.h | 2 +- net/sctp/auth.c | 2 +- net/sctp/sm_statefuns.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/linux/sctp.h b/include/linux/sctp.h index 01a0eb7e9fa1..d182e8c41985 100644 --- a/include/linux/sctp.h +++ b/include/linux/sctp.h @@ -727,7 +727,7 @@ struct sctp_addip_chunk { struct sctp_authhdr { __be16 shkey_id; __be16 hmac_id; - __u8 hmac[]; + /* __u8 hmac[]; */ }; struct sctp_auth_chunk { diff --git a/net/sctp/auth.c b/net/sctp/auth.c index 34964145514e..c58fffc86a0c 100644 --- a/net/sctp/auth.c +++ b/net/sctp/auth.c @@ -738,7 +738,7 @@ void sctp_auth_calculate_hmac(const struct sctp_association *asoc, tfm = asoc->ep->auth_hmacs[hmac_id]; - digest = auth->auth_hdr.hmac; + digest = (u8 *)(&auth->auth_hdr + 1); if (crypto_shash_setkey(tfm, &asoc_key->data[0], asoc_key->len)) goto free; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 7b8eb735fa88..97f1155a2045 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -4391,7 +4391,7 @@ static enum sctp_ierror sctp_sf_authenticate( * 3. Compute the new digest * 4. Compare saved and new digests. */ - digest = auth_hdr->hmac; + digest = (u8 *)(auth_hdr + 1); skb_pull(chunk->skb, sig_len); save_digest = kmemdup(digest, sig_len, GFP_ATOMIC); -- cgit v1.2.3 From abff3e5e2935f5cc34feb559b352179b13eaa066 Mon Sep 17 00:00:00 2001 From: Emeel Hakim Date: Wed, 19 Apr 2023 17:21:22 +0300 Subject: vlan: Add MACsec offload operations for VLAN interface Add support for MACsec offload operations for VLAN driver to allow offloading MACsec when VLAN's real device supports Macsec offload by forwarding the offload request to it. Signed-off-by: Emeel Hakim Reviewed-by: Subbaraya Sundeep Signed-off-by: David S. Miller --- net/8021q/vlan_dev.c | 242 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 242 insertions(+) (limited to 'net') diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 5920544e93e8..870e4935d6e6 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -26,6 +26,7 @@ #include #include #include +#include #include "vlan.h" #include "vlanproc.h" @@ -572,6 +573,9 @@ static int vlan_dev_init(struct net_device *dev) NETIF_F_HIGHDMA | NETIF_F_SCTP_CRC | NETIF_F_ALL_FCOE; + if (real_dev->vlan_features & NETIF_F_HW_MACSEC) + dev->hw_features |= NETIF_F_HW_MACSEC; + dev->features |= dev->hw_features | NETIF_F_LLTX; netif_inherit_tso_max(dev, real_dev); if (dev->features & NETIF_F_VLAN_FEATURES) @@ -803,6 +807,241 @@ static int vlan_dev_fill_forward_path(struct net_device_path_ctx *ctx, return 0; } +#if IS_ENABLED(CONFIG_MACSEC) + +static const struct macsec_ops *vlan_get_macsec_ops(const struct macsec_context *ctx) +{ + return vlan_dev_priv(ctx->netdev)->real_dev->macsec_ops; +} + +static int vlan_macsec_offload(int (* const func)(struct macsec_context *), + struct macsec_context *ctx) +{ + if (unlikely(!func)) + return 0; + + return (*func)(ctx); +} + +static int vlan_macsec_dev_open(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_dev_open, ctx); +} + +static int vlan_macsec_dev_stop(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_dev_stop, ctx); +} + +static int vlan_macsec_add_secy(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_add_secy, ctx); +} + +static int vlan_macsec_upd_secy(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_upd_secy, ctx); +} + +static int vlan_macsec_del_secy(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_del_secy, ctx); +} + +static int vlan_macsec_add_rxsc(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_add_rxsc, ctx); +} + +static int vlan_macsec_upd_rxsc(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_upd_rxsc, ctx); +} + +static int vlan_macsec_del_rxsc(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_del_rxsc, ctx); +} + +static int vlan_macsec_add_rxsa(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_add_rxsa, ctx); +} + +static int vlan_macsec_upd_rxsa(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_upd_rxsa, ctx); +} + +static int vlan_macsec_del_rxsa(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_del_rxsa, ctx); +} + +static int vlan_macsec_add_txsa(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_add_txsa, ctx); +} + +static int vlan_macsec_upd_txsa(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_upd_txsa, ctx); +} + +static int vlan_macsec_del_txsa(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_del_txsa, ctx); +} + +static int vlan_macsec_get_dev_stats(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_get_dev_stats, ctx); +} + +static int vlan_macsec_get_tx_sc_stats(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_get_tx_sc_stats, ctx); +} + +static int vlan_macsec_get_tx_sa_stats(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_get_tx_sa_stats, ctx); +} + +static int vlan_macsec_get_rx_sc_stats(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_get_rx_sc_stats, ctx); +} + +static int vlan_macsec_get_rx_sa_stats(struct macsec_context *ctx) +{ + const struct macsec_ops *ops = vlan_get_macsec_ops(ctx); + + if (!ops) + return -EOPNOTSUPP; + + return vlan_macsec_offload(ops->mdo_get_rx_sa_stats, ctx); +} + +static const struct macsec_ops macsec_offload_ops = { + /* Device wide */ + .mdo_dev_open = vlan_macsec_dev_open, + .mdo_dev_stop = vlan_macsec_dev_stop, + /* SecY */ + .mdo_add_secy = vlan_macsec_add_secy, + .mdo_upd_secy = vlan_macsec_upd_secy, + .mdo_del_secy = vlan_macsec_del_secy, + /* Security channels */ + .mdo_add_rxsc = vlan_macsec_add_rxsc, + .mdo_upd_rxsc = vlan_macsec_upd_rxsc, + .mdo_del_rxsc = vlan_macsec_del_rxsc, + /* Security associations */ + .mdo_add_rxsa = vlan_macsec_add_rxsa, + .mdo_upd_rxsa = vlan_macsec_upd_rxsa, + .mdo_del_rxsa = vlan_macsec_del_rxsa, + .mdo_add_txsa = vlan_macsec_add_txsa, + .mdo_upd_txsa = vlan_macsec_upd_txsa, + .mdo_del_txsa = vlan_macsec_del_txsa, + /* Statistics */ + .mdo_get_dev_stats = vlan_macsec_get_dev_stats, + .mdo_get_tx_sc_stats = vlan_macsec_get_tx_sc_stats, + .mdo_get_tx_sa_stats = vlan_macsec_get_tx_sa_stats, + .mdo_get_rx_sc_stats = vlan_macsec_get_rx_sc_stats, + .mdo_get_rx_sa_stats = vlan_macsec_get_rx_sa_stats, +}; + +#endif + static const struct ethtool_ops vlan_ethtool_ops = { .get_link_ksettings = vlan_ethtool_get_link_ksettings, .get_drvinfo = vlan_ethtool_get_drvinfo, @@ -869,6 +1108,9 @@ void vlan_setup(struct net_device *dev) dev->priv_destructor = vlan_dev_free; dev->ethtool_ops = &vlan_ethtool_ops; +#if IS_ENABLED(CONFIG_MACSEC) + dev->macsec_ops = &macsec_offload_ops; +#endif dev->min_mtu = 0; dev->max_mtu = ETH_MAX_MTU; -- cgit v1.2.3 From 013a7ce81dd8a5bba31f802053829d91f9a43a8f Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 19 Apr 2023 18:34:52 +0300 Subject: bridge: Reorder neighbor suppression check when flooding The bridge does not flood ARP / NS packets for which a reply was sent to bridge ports that have neighbor suppression enabled. Subsequent patches are going to add per-{Port, VLAN} neighbor suppression, which is going to make it more expensive to check whether neighbor suppression is enabled since a VLAN lookup will be required. Therefore, instead of unnecessarily performing this lookup for every packet, only perform it for ARP / NS packets for which a reply was sent. Signed-off-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_forward.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 02bb620d3b8d..0fe133fa214c 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -224,8 +224,8 @@ void br_flood(struct net_bridge *br, struct sk_buff *skb, /* Do not flood to ports that enable proxy ARP */ if (p->flags & BR_PROXYARP) continue; - if ((p->flags & (BR_PROXYARP_WIFI | BR_NEIGH_SUPPRESS)) && - BR_INPUT_SKB_CB(skb)->proxyarp_replied) + if (BR_INPUT_SKB_CB(skb)->proxyarp_replied && + (p->flags & (BR_PROXYARP_WIFI | BR_NEIGH_SUPPRESS))) continue; prev = maybe_deliver(prev, p, skb, local_orig); -- cgit v1.2.3 From e408336a693e0fd74ba1b959627594b80f1761c3 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 19 Apr 2023 18:34:53 +0300 Subject: bridge: Pass VLAN ID to br_flood() Subsequent patches are going to add per-{Port, VLAN} neighbor suppression, which will require br_flood() to potentially suppress ARP / NS packets on a per-{Port, VLAN} basis. As a preparation, pass the VLAN ID of the packet as another argument to br_flood(). Signed-off-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_device.c | 8 ++++---- net/bridge/br_forward.c | 3 ++- net/bridge/br_input.c | 2 +- net/bridge/br_private.h | 3 ++- 4 files changed, 9 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index df47c876230e..8eca8a5c80c6 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -80,10 +80,10 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) dest = eth_hdr(skb)->h_dest; if (is_broadcast_ether_addr(dest)) { - br_flood(br, skb, BR_PKT_BROADCAST, false, true); + br_flood(br, skb, BR_PKT_BROADCAST, false, true, vid); } else if (is_multicast_ether_addr(dest)) { if (unlikely(netpoll_tx_running(dev))) { - br_flood(br, skb, BR_PKT_MULTICAST, false, true); + br_flood(br, skb, BR_PKT_MULTICAST, false, true, vid); goto out; } if (br_multicast_rcv(&brmctx, &pmctx_null, vlan, skb, vid)) { @@ -96,11 +96,11 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) br_multicast_querier_exists(brmctx, eth_hdr(skb), mdst)) br_multicast_flood(mdst, skb, brmctx, false, true); else - br_flood(br, skb, BR_PKT_MULTICAST, false, true); + br_flood(br, skb, BR_PKT_MULTICAST, false, true, vid); } else if ((dst = br_fdb_find_rcu(br, dest, vid)) != NULL) { br_forward(dst->dst, skb, false, true); } else { - br_flood(br, skb, BR_PKT_UNICAST, false, true); + br_flood(br, skb, BR_PKT_UNICAST, false, true, vid); } out: rcu_read_unlock(); diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 0fe133fa214c..94a8d757ae4e 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -197,7 +197,8 @@ out: /* called under rcu_read_lock */ void br_flood(struct net_bridge *br, struct sk_buff *skb, - enum br_pkt_type pkt_type, bool local_rcv, bool local_orig) + enum br_pkt_type pkt_type, bool local_rcv, bool local_orig, + u16 vid) { struct net_bridge_port *prev = NULL; struct net_bridge_port *p; diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 3027e8f6be15..fc17b9fd93e6 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -207,7 +207,7 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb br_forward(dst->dst, skb, local_rcv, false); } else { if (!mcast_hit) - br_flood(br, skb, pkt_type, local_rcv, false); + br_flood(br, skb, pkt_type, local_rcv, false, vid); else br_multicast_flood(mdst, skb, brmctx, local_rcv, false); } diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 7264fd40f82f..1ff4d64ab584 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -849,7 +849,8 @@ void br_forward(const struct net_bridge_port *to, struct sk_buff *skb, bool local_rcv, bool local_orig); int br_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb); void br_flood(struct net_bridge *br, struct sk_buff *skb, - enum br_pkt_type pkt_type, bool local_rcv, bool local_orig); + enum br_pkt_type pkt_type, bool local_rcv, bool local_orig, + u16 vid); /* return true if both source port and dest port are isolated */ static inline bool br_skb_isolated(const struct net_bridge_port *to, -- cgit v1.2.3 From a714e3ec230892039b5d5ae6902b58bb084a15c1 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 19 Apr 2023 18:34:54 +0300 Subject: bridge: Add internal flags for per-{Port, VLAN} neighbor suppression Add two internal flags that will be used to enable / disable per-{Port, VLAN} neighbor suppression: 1. 'BR_NEIGH_VLAN_SUPPRESS': A per-port flag used to indicate that per-{Port, VLAN} neighbor suppression is enabled on the bridge port. When set, 'BR_NEIGH_SUPPRESS' has no effect. 2. 'BR_VLFLAG_NEIGH_SUPPRESS_ENABLED': A per-VLAN flag used to indicate that neighbor suppression is enabled on the given VLAN. Signed-off-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/if_bridge.h | 1 + net/bridge/br_private.h | 1 + 2 files changed, 2 insertions(+) (limited to 'net') diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h index 1668ac4d7adc..3ff96ae31bf6 100644 --- a/include/linux/if_bridge.h +++ b/include/linux/if_bridge.h @@ -60,6 +60,7 @@ struct br_ip_list { #define BR_TX_FWD_OFFLOAD BIT(20) #define BR_PORT_LOCKED BIT(21) #define BR_PORT_MAB BIT(22) +#define BR_NEIGH_VLAN_SUPPRESS BIT(23) #define BR_DEFAULT_AGEING_TIME (300 * HZ) diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 1ff4d64ab584..b17fc821ecc8 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -178,6 +178,7 @@ enum { BR_VLFLAG_ADDED_BY_SWITCHDEV = BIT(1), BR_VLFLAG_MCAST_ENABLED = BIT(2), BR_VLFLAG_GLOBAL_MCAST_ENABLED = BIT(3), + BR_VLFLAG_NEIGH_SUPPRESS_ENABLED = BIT(4), }; /** -- cgit v1.2.3 From 6be42ed0a5f4b26e77b51cd10a09b41d53f97570 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 19 Apr 2023 18:34:55 +0300 Subject: bridge: Take per-{Port, VLAN} neighbor suppression into account The bridge driver gates the neighbor suppression code behind an internal per-bridge flag called 'BROPT_NEIGH_SUPPRESS_ENABLED'. The flag is set when at least one bridge port has neighbor suppression enabled. As a preparation for per-{Port, VLAN} neighbor suppression, make sure the global flag is also set if per-{Port, VLAN} neighbor suppression is enabled. That is, when the 'BR_NEIGH_VLAN_SUPPRESS' flag is set on at least one bridge port. Signed-off-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_arp_nd_proxy.c | 2 +- net/bridge/br_if.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/bridge/br_arp_nd_proxy.c b/net/bridge/br_arp_nd_proxy.c index b45c00c01dea..016a25a9e444 100644 --- a/net/bridge/br_arp_nd_proxy.c +++ b/net/bridge/br_arp_nd_proxy.c @@ -30,7 +30,7 @@ void br_recalculate_neigh_suppress_enabled(struct net_bridge *br) bool neigh_suppress = false; list_for_each_entry(p, &br->port_list, list) { - if (p->flags & BR_NEIGH_SUPPRESS) { + if (p->flags & (BR_NEIGH_SUPPRESS | BR_NEIGH_VLAN_SUPPRESS)) { neigh_suppress = true; break; } diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 24f01ff113f0..3f04b40f6056 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -759,7 +759,7 @@ void br_port_flags_change(struct net_bridge_port *p, unsigned long mask) if (mask & BR_AUTO_MASK) nbp_update_port_count(br); - if (mask & BR_NEIGH_SUPPRESS) + if (mask & (BR_NEIGH_SUPPRESS | BR_NEIGH_VLAN_SUPPRESS)) br_recalculate_neigh_suppress_enabled(br); } -- cgit v1.2.3 From 3aca683e0654a2f4ff6996341cba98a5396ff050 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 19 Apr 2023 18:34:56 +0300 Subject: bridge: Encapsulate data path neighbor suppression logic Currently, there are various places in the bridge data path that check whether neighbor suppression is enabled on a given bridge port. As a preparation for per-{Port, VLAN} neighbor suppression, encapsulate this logic in a function and pass the VLAN ID of the packet as an argument. Signed-off-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_arp_nd_proxy.c | 15 ++++++++++----- net/bridge/br_forward.c | 3 ++- net/bridge/br_private.h | 1 + 3 files changed, 13 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/bridge/br_arp_nd_proxy.c b/net/bridge/br_arp_nd_proxy.c index 016a25a9e444..16c3a1c5d0ae 100644 --- a/net/bridge/br_arp_nd_proxy.c +++ b/net/bridge/br_arp_nd_proxy.c @@ -158,7 +158,7 @@ void br_do_proxy_suppress_arp(struct sk_buff *skb, struct net_bridge *br, return; if (br_opt_get(br, BROPT_NEIGH_SUPPRESS_ENABLED)) { - if (p && (p->flags & BR_NEIGH_SUPPRESS)) + if (br_is_neigh_suppress_enabled(p, vid)) return; if (parp->ar_op != htons(ARPOP_RREQUEST) && parp->ar_op != htons(ARPOP_RREPLY) && @@ -202,8 +202,8 @@ void br_do_proxy_suppress_arp(struct sk_buff *skb, struct net_bridge *br, bool replied = false; if ((p && (p->flags & BR_PROXYARP)) || - (f->dst && (f->dst->flags & (BR_PROXYARP_WIFI | - BR_NEIGH_SUPPRESS)))) { + (f->dst && (f->dst->flags & BR_PROXYARP_WIFI)) || + br_is_neigh_suppress_enabled(f->dst, vid)) { if (!vid) br_arp_send(br, p, skb->dev, sip, tip, sha, n->ha, sha, 0, 0); @@ -407,7 +407,7 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br, BR_INPUT_SKB_CB(skb)->proxyarp_replied = 0; - if (p && (p->flags & BR_NEIGH_SUPPRESS)) + if (br_is_neigh_suppress_enabled(p, vid)) return; if (msg->icmph.icmp6_type == NDISC_NEIGHBOUR_ADVERTISEMENT && @@ -461,7 +461,7 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br, if (f) { bool replied = false; - if (f->dst && (f->dst->flags & BR_NEIGH_SUPPRESS)) { + if (br_is_neigh_suppress_enabled(f->dst, vid)) { if (vid != 0) br_nd_send(br, p, skb, n, skb->vlan_proto, @@ -483,3 +483,8 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br, } } #endif + +bool br_is_neigh_suppress_enabled(const struct net_bridge_port *p, u16 vid) +{ + return p && (p->flags & BR_NEIGH_SUPPRESS); +} diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 94a8d757ae4e..57744704ff69 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -226,7 +226,8 @@ void br_flood(struct net_bridge *br, struct sk_buff *skb, if (p->flags & BR_PROXYARP) continue; if (BR_INPUT_SKB_CB(skb)->proxyarp_replied && - (p->flags & (BR_PROXYARP_WIFI | BR_NEIGH_SUPPRESS))) + ((p->flags & BR_PROXYARP_WIFI) || + br_is_neigh_suppress_enabled(p, vid))) continue; prev = maybe_deliver(prev, p, skb, local_orig); diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index b17fc821ecc8..2119729ded2b 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -2220,4 +2220,5 @@ void br_do_proxy_suppress_arp(struct sk_buff *skb, struct net_bridge *br, void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br, u16 vid, struct net_bridge_port *p, struct nd_msg *msg); struct nd_msg *br_is_nd_neigh_msg(struct sk_buff *skb, struct nd_msg *m); +bool br_is_neigh_suppress_enabled(const struct net_bridge_port *p, u16 vid); #endif -- cgit v1.2.3 From 412614b1457af278b00152f3804af5481c6c55ac Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 19 Apr 2023 18:34:57 +0300 Subject: bridge: Add per-{Port, VLAN} neighbor suppression data path support When the bridge is not VLAN-aware (i.e., VLAN ID is 0), determine if neighbor suppression is enabled on a given bridge port solely based on the existing 'BR_NEIGH_SUPPRESS' flag. Otherwise, if the bridge is VLAN-aware, first check if per-{Port, VLAN} neighbor suppression is enabled on the given bridge port using the 'BR_NEIGH_VLAN_SUPPRESS' flag. If so, look up the VLAN and check whether it has neighbor suppression enabled based on the per-VLAN 'BR_VLFLAG_NEIGH_SUPPRESS_ENABLED' flag. If the bridge is VLAN-aware, but the bridge port does not have per-{Port, VLAN} neighbor suppression enabled, then fallback to determine neighbor suppression based on the 'BR_NEIGH_SUPPRESS' flag. Signed-off-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_arp_nd_proxy.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bridge/br_arp_nd_proxy.c b/net/bridge/br_arp_nd_proxy.c index 16c3a1c5d0ae..c7869a286df4 100644 --- a/net/bridge/br_arp_nd_proxy.c +++ b/net/bridge/br_arp_nd_proxy.c @@ -486,5 +486,21 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br, bool br_is_neigh_suppress_enabled(const struct net_bridge_port *p, u16 vid) { - return p && (p->flags & BR_NEIGH_SUPPRESS); + if (!p) + return false; + + if (!vid) + return !!(p->flags & BR_NEIGH_SUPPRESS); + + if (p->flags & BR_NEIGH_VLAN_SUPPRESS) { + struct net_bridge_vlan_group *vg = nbp_vlan_group_rcu(p); + struct net_bridge_vlan *v; + + v = br_vlan_find(vg, vid); + if (!v) + return false; + return !!(v->priv_flags & BR_VLFLAG_NEIGH_SUPPRESS_ENABLED); + } else { + return !!(p->flags & BR_NEIGH_SUPPRESS); + } } -- cgit v1.2.3 From 83f6d600796c65ab34b08dbddb5795099dfda4d1 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 19 Apr 2023 18:34:58 +0300 Subject: bridge: vlan: Allow setting VLAN neighbor suppression state Add a new VLAN attribute that allows user space to set the neighbor suppression state of the port VLAN. Example: # bridge -d -j -p vlan show dev swp1 vid 10 | jq '.[]["vlans"][]["neigh_suppress"]' false # bridge vlan set vid 10 dev swp1 neigh_suppress on # bridge -d -j -p vlan show dev swp1 vid 10 | jq '.[]["vlans"][]["neigh_suppress"]' true # bridge vlan set vid 10 dev swp1 neigh_suppress off # bridge -d -j -p vlan show dev swp1 vid 10 | jq '.[]["vlans"][]["neigh_suppress"]' false # bridge vlan set vid 10 dev br0 neigh_suppress on Error: bridge: Can't set neigh_suppress for non-port vlans. Signed-off-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/uapi/linux/if_bridge.h | 1 + net/bridge/br_vlan.c | 1 + net/bridge/br_vlan_options.c | 20 +++++++++++++++++++- 3 files changed, 21 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h index c9d624f528c5..f95326fce6bb 100644 --- a/include/uapi/linux/if_bridge.h +++ b/include/uapi/linux/if_bridge.h @@ -525,6 +525,7 @@ enum { BRIDGE_VLANDB_ENTRY_MCAST_ROUTER, BRIDGE_VLANDB_ENTRY_MCAST_N_GROUPS, BRIDGE_VLANDB_ENTRY_MCAST_MAX_GROUPS, + BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS, __BRIDGE_VLANDB_ENTRY_MAX, }; #define BRIDGE_VLANDB_ENTRY_MAX (__BRIDGE_VLANDB_ENTRY_MAX - 1) diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c index 8a3dbc09ba38..15f44d026e75 100644 --- a/net/bridge/br_vlan.c +++ b/net/bridge/br_vlan.c @@ -2134,6 +2134,7 @@ static const struct nla_policy br_vlan_db_policy[BRIDGE_VLANDB_ENTRY_MAX + 1] = [BRIDGE_VLANDB_ENTRY_MCAST_ROUTER] = { .type = NLA_U8 }, [BRIDGE_VLANDB_ENTRY_MCAST_N_GROUPS] = { .type = NLA_REJECT }, [BRIDGE_VLANDB_ENTRY_MCAST_MAX_GROUPS] = { .type = NLA_U32 }, + [BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS] = NLA_POLICY_MAX(NLA_U8, 1), }; static int br_vlan_rtm_process_one(struct net_device *dev, diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c index e378c2f3a9e2..8fa89b04ee94 100644 --- a/net/bridge/br_vlan_options.c +++ b/net/bridge/br_vlan_options.c @@ -52,7 +52,9 @@ bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v, const struct net_bridge_port *p) { if (nla_put_u8(skb, BRIDGE_VLANDB_ENTRY_STATE, br_vlan_get_state(v)) || - !__vlan_tun_put(skb, v)) + !__vlan_tun_put(skb, v) || + nla_put_u8(skb, BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS, + !!(v->priv_flags & BR_VLFLAG_NEIGH_SUPPRESS_ENABLED))) return false; #ifdef CONFIG_BRIDGE_IGMP_SNOOPING @@ -80,6 +82,7 @@ size_t br_vlan_opts_nl_size(void) + nla_total_size(sizeof(u32)) /* BRIDGE_VLANDB_ENTRY_MCAST_N_GROUPS */ + nla_total_size(sizeof(u32)) /* BRIDGE_VLANDB_ENTRY_MCAST_MAX_GROUPS */ #endif + + nla_total_size(sizeof(u8)) /* BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS */ + 0; } @@ -239,6 +242,21 @@ static int br_vlan_process_one_opts(const struct net_bridge *br, } #endif + if (tb[BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS]) { + bool enabled = v->priv_flags & BR_VLFLAG_NEIGH_SUPPRESS_ENABLED; + bool val = nla_get_u8(tb[BRIDGE_VLANDB_ENTRY_NEIGH_SUPPRESS]); + + if (!p) { + NL_SET_ERR_MSG_MOD(extack, "Can't set neigh_suppress for non-port vlans"); + return -EINVAL; + } + + if (val != enabled) { + v->priv_flags ^= BR_VLFLAG_NEIGH_SUPPRESS_ENABLED; + *changed = true; + } + } + return 0; } -- cgit v1.2.3 From 160656d7201d861a1f2a0bf279a765e8cda2317a Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 19 Apr 2023 18:34:59 +0300 Subject: bridge: Allow setting per-{Port, VLAN} neighbor suppression state Add a new bridge port attribute that allows user space to enable per-{Port, VLAN} neighbor suppression. Example: # bridge -d -j -p link show dev swp1 | jq '.[]["neigh_vlan_suppress"]' false # bridge link set dev swp1 neigh_vlan_suppress on # bridge -d -j -p link show dev swp1 | jq '.[]["neigh_vlan_suppress"]' true # bridge link set dev swp1 neigh_vlan_suppress off # bridge -d -j -p link show dev swp1 | jq '.[]["neigh_vlan_suppress"]' false Signed-off-by: Ido Schimmel Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/uapi/linux/if_link.h | 1 + net/bridge/br_netlink.c | 8 +++++++- net/core/rtnetlink.c | 2 +- 3 files changed, 9 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 8d679688efe0..4ac1000b0ef2 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -569,6 +569,7 @@ enum { IFLA_BRPORT_MAB, IFLA_BRPORT_MCAST_N_GROUPS, IFLA_BRPORT_MCAST_MAX_GROUPS, + IFLA_BRPORT_NEIGH_VLAN_SUPPRESS, __IFLA_BRPORT_MAX }; #define IFLA_BRPORT_MAX (__IFLA_BRPORT_MAX - 1) diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index fefb1c0e248b..05c5863d2e20 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -189,6 +189,7 @@ static inline size_t br_port_info_size(void) + nla_total_size(1) /* IFLA_BRPORT_ISOLATED */ + nla_total_size(1) /* IFLA_BRPORT_LOCKED */ + nla_total_size(1) /* IFLA_BRPORT_MAB */ + + nla_total_size(1) /* IFLA_BRPORT_NEIGH_VLAN_SUPPRESS */ + nla_total_size(sizeof(struct ifla_bridge_id)) /* IFLA_BRPORT_ROOT_ID */ + nla_total_size(sizeof(struct ifla_bridge_id)) /* IFLA_BRPORT_BRIDGE_ID */ + nla_total_size(sizeof(u16)) /* IFLA_BRPORT_DESIGNATED_PORT */ @@ -278,7 +279,9 @@ static int br_port_fill_attrs(struct sk_buff *skb, !!(p->flags & BR_MRP_LOST_IN_CONT)) || nla_put_u8(skb, IFLA_BRPORT_ISOLATED, !!(p->flags & BR_ISOLATED)) || nla_put_u8(skb, IFLA_BRPORT_LOCKED, !!(p->flags & BR_PORT_LOCKED)) || - nla_put_u8(skb, IFLA_BRPORT_MAB, !!(p->flags & BR_PORT_MAB))) + nla_put_u8(skb, IFLA_BRPORT_MAB, !!(p->flags & BR_PORT_MAB)) || + nla_put_u8(skb, IFLA_BRPORT_NEIGH_VLAN_SUPPRESS, + !!(p->flags & BR_NEIGH_VLAN_SUPPRESS))) return -EMSGSIZE; timerval = br_timer_value(&p->message_age_timer); @@ -891,6 +894,7 @@ static const struct nla_policy br_port_policy[IFLA_BRPORT_MAX + 1] = { [IFLA_BRPORT_MCAST_EHT_HOSTS_LIMIT] = { .type = NLA_U32 }, [IFLA_BRPORT_MCAST_N_GROUPS] = { .type = NLA_REJECT }, [IFLA_BRPORT_MCAST_MAX_GROUPS] = { .type = NLA_U32 }, + [IFLA_BRPORT_NEIGH_VLAN_SUPPRESS] = NLA_POLICY_MAX(NLA_U8, 1), }; /* Change the state of the port and notify spanning tree */ @@ -957,6 +961,8 @@ static int br_setport(struct net_bridge_port *p, struct nlattr *tb[], br_set_port_flag(p, tb, IFLA_BRPORT_ISOLATED, BR_ISOLATED); br_set_port_flag(p, tb, IFLA_BRPORT_LOCKED, BR_PORT_LOCKED); br_set_port_flag(p, tb, IFLA_BRPORT_MAB, BR_PORT_MAB); + br_set_port_flag(p, tb, IFLA_BRPORT_NEIGH_VLAN_SUPPRESS, + BR_NEIGH_VLAN_SUPPRESS); if ((p->flags & BR_PORT_MAB) && (!(p->flags & BR_PORT_LOCKED) || !(p->flags & BR_LEARNING))) { diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index e844d75220fb..653901a1bf75 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -61,7 +61,7 @@ #include "dev.h" #define RTNL_MAX_TYPE 50 -#define RTNL_SLAVE_MAX_TYPE 42 +#define RTNL_SLAVE_MAX_TYPE 43 struct rtnl_link { rtnl_doit_func doit; -- cgit v1.2.3 From dfc39d4026fb2432363c0f77543c4cf3adca4c7b Mon Sep 17 00:00:00 2001 From: Jianfeng Tan Date: Wed, 19 Apr 2023 15:24:16 +0800 Subject: net/packet: support mergeable feature of virtio Packet sockets, like tap, can be used as the backend for kernel vhost. In packet sockets, virtio net header size is currently hardcoded to be the size of struct virtio_net_hdr, which is 10 bytes; however, it is not always the case: some virtio features, such as mrg_rxbuf, need virtio net header to be 12-byte long. Mergeable buffers, as a virtio feature, is worthy of supporting: packets that are larger than one-mbuf size will be dropped in vhost worker's handle_rx if mrg_rxbuf feature is not used, but large packets cannot be avoided and increasing mbuf's size is not economical. With this virtio feature enabled by virtio-user, packet sockets with hardcoded 10-byte virtio net header will parse mac head incorrectly in packet_snd by taking the last two bytes of virtio net header as part of mac header. This incorrect mac header parsing will cause packet to be dropped due to invalid ether head checking in later under-layer device packet receiving. By adding extra field vnet_hdr_sz with utilizing holes in struct packet_sock to record currently used virtio net header size and supporting extra sockopt PACKET_VNET_HDR_SZ to set specified vnet_hdr_sz, packet sockets can know the exact length of virtio net header that virtio user gives. In packet_snd, tpacket_snd and packet_recvmsg, instead of using hardcoded virtio net header size, it can get the exact vnet_hdr_sz from corresponding packet_sock, and parse mac header correctly based on this information to avoid the packets being mistakenly dropped. Signed-off-by: Jianfeng Tan Co-developed-by: Anqi Shen Signed-off-by: Anqi Shen Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- include/uapi/linux/if_packet.h | 1 + net/packet/af_packet.c | 95 +++++++++++++++++++++++++----------------- net/packet/diag.c | 2 +- net/packet/internal.h | 2 +- 4 files changed, 60 insertions(+), 40 deletions(-) (limited to 'net') diff --git a/include/uapi/linux/if_packet.h b/include/uapi/linux/if_packet.h index 78c981d6a9d4..9efc42382fdb 100644 --- a/include/uapi/linux/if_packet.h +++ b/include/uapi/linux/if_packet.h @@ -59,6 +59,7 @@ struct sockaddr_ll { #define PACKET_ROLLOVER_STATS 21 #define PACKET_FANOUT_DATA 22 #define PACKET_IGNORE_OUTGOING 23 +#define PACKET_VNET_HDR_SZ 24 #define PACKET_FANOUT_HASH 0 #define PACKET_FANOUT_LB 1 diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 568f8d76e3c1..6080c0db0814 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2090,18 +2090,18 @@ static unsigned int run_filter(struct sk_buff *skb, } static int packet_rcv_vnet(struct msghdr *msg, const struct sk_buff *skb, - size_t *len) + size_t *len, int vnet_hdr_sz) { - struct virtio_net_hdr vnet_hdr; + struct virtio_net_hdr_mrg_rxbuf vnet_hdr = { .num_buffers = 0 }; - if (*len < sizeof(vnet_hdr)) + if (*len < vnet_hdr_sz) return -EINVAL; - *len -= sizeof(vnet_hdr); + *len -= vnet_hdr_sz; - if (virtio_net_hdr_from_skb(skb, &vnet_hdr, vio_le(), true, 0)) + if (virtio_net_hdr_from_skb(skb, (struct virtio_net_hdr *)&vnet_hdr, vio_le(), true, 0)) return -EINVAL; - return memcpy_to_msg(msg, (void *)&vnet_hdr, sizeof(vnet_hdr)); + return memcpy_to_msg(msg, (void *)&vnet_hdr, vnet_hdr_sz); } /* @@ -2250,7 +2250,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, __u32 ts_status; bool is_drop_n_account = false; unsigned int slot_id = 0; - bool do_vnet = false; + int vnet_hdr_sz = 0; /* struct tpacket{2,3}_hdr is aligned to a multiple of TPACKET_ALIGNMENT. * We may add members to them until current aligned size without forcing @@ -2308,10 +2308,9 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, netoff = TPACKET_ALIGN(po->tp_hdrlen + (maclen < 16 ? 16 : maclen)) + po->tp_reserve; - if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) { - netoff += sizeof(struct virtio_net_hdr); - do_vnet = true; - } + vnet_hdr_sz = READ_ONCE(po->vnet_hdr_sz); + if (vnet_hdr_sz) + netoff += vnet_hdr_sz; macoff = netoff - maclen; } if (netoff > USHRT_MAX) { @@ -2337,7 +2336,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, snaplen = po->rx_ring.frame_size - macoff; if ((int)snaplen < 0) { snaplen = 0; - do_vnet = false; + vnet_hdr_sz = 0; } } } else if (unlikely(macoff + snaplen > @@ -2351,7 +2350,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, if (unlikely((int)snaplen < 0)) { snaplen = 0; macoff = GET_PBDQC_FROM_RB(&po->rx_ring)->max_frame_len; - do_vnet = false; + vnet_hdr_sz = 0; } } spin_lock(&sk->sk_receive_queue.lock); @@ -2367,7 +2366,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, __set_bit(slot_id, po->rx_ring.rx_owner_map); } - if (do_vnet && + if (vnet_hdr_sz && virtio_net_hdr_from_skb(skb, h.raw + macoff - sizeof(struct virtio_net_hdr), vio_le(), true, 0)) { @@ -2551,16 +2550,26 @@ static int __packet_snd_vnet_parse(struct virtio_net_hdr *vnet_hdr, size_t len) } static int packet_snd_vnet_parse(struct msghdr *msg, size_t *len, - struct virtio_net_hdr *vnet_hdr) + struct virtio_net_hdr *vnet_hdr, int vnet_hdr_sz) { - if (*len < sizeof(*vnet_hdr)) + int ret; + + if (*len < vnet_hdr_sz) return -EINVAL; - *len -= sizeof(*vnet_hdr); + *len -= vnet_hdr_sz; if (!copy_from_iter_full(vnet_hdr, sizeof(*vnet_hdr), &msg->msg_iter)) return -EFAULT; - return __packet_snd_vnet_parse(vnet_hdr, *len); + ret = __packet_snd_vnet_parse(vnet_hdr, *len); + if (ret) + return ret; + + /* move iter to point to the start of mac header */ + if (vnet_hdr_sz != sizeof(struct virtio_net_hdr)) + iov_iter_advance(&msg->msg_iter, vnet_hdr_sz - sizeof(struct virtio_net_hdr)); + + return 0; } static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb, @@ -2722,6 +2731,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg) void *ph; DECLARE_SOCKADDR(struct sockaddr_ll *, saddr, msg->msg_name); bool need_wait = !(msg->msg_flags & MSG_DONTWAIT); + int vnet_hdr_sz = READ_ONCE(po->vnet_hdr_sz); unsigned char *addr = NULL; int tp_len, size_max; void *data; @@ -2779,8 +2789,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg) size_max = po->tx_ring.frame_size - (po->tp_hdrlen - sizeof(struct sockaddr_ll)); - if ((size_max > dev->mtu + reserve + VLAN_HLEN) && - !packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) + if ((size_max > dev->mtu + reserve + VLAN_HLEN) && !vnet_hdr_sz) size_max = dev->mtu + reserve + VLAN_HLEN; reinit_completion(&po->skb_completion); @@ -2809,10 +2818,10 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg) status = TP_STATUS_SEND_REQUEST; hlen = LL_RESERVED_SPACE(dev); tlen = dev->needed_tailroom; - if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) { + if (vnet_hdr_sz) { vnet_hdr = data; - data += sizeof(*vnet_hdr); - tp_len -= sizeof(*vnet_hdr); + data += vnet_hdr_sz; + tp_len -= vnet_hdr_sz; if (tp_len < 0 || __packet_snd_vnet_parse(vnet_hdr, tp_len)) { tp_len = -EINVAL; @@ -2837,7 +2846,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg) addr, hlen, copylen, &sockc); if (likely(tp_len >= 0) && tp_len > dev->mtu + reserve && - !packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR) && + !vnet_hdr_sz && !packet_extra_vlan_len_allowed(dev, skb)) tp_len = -EMSGSIZE; @@ -2856,7 +2865,7 @@ tpacket_error: } } - if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) { + if (vnet_hdr_sz) { if (virtio_net_hdr_to_skb(skb, vnet_hdr, vio_le())) { tp_len = -EINVAL; goto tpacket_error; @@ -2946,7 +2955,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) struct virtio_net_hdr vnet_hdr = { 0 }; int offset = 0; struct packet_sock *po = pkt_sk(sk); - bool has_vnet_hdr = false; + int vnet_hdr_sz = READ_ONCE(po->vnet_hdr_sz); int hlen, tlen, linear; int extra_len = 0; @@ -2990,11 +2999,10 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) if (sock->type == SOCK_RAW) reserve = dev->hard_header_len; - if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) { - err = packet_snd_vnet_parse(msg, &len, &vnet_hdr); + if (vnet_hdr_sz) { + err = packet_snd_vnet_parse(msg, &len, &vnet_hdr, vnet_hdr_sz); if (err) goto out_unlock; - has_vnet_hdr = true; } if (unlikely(sock_flag(sk, SOCK_NOFCS))) { @@ -3064,11 +3072,11 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) packet_parse_headers(skb, sock); - if (has_vnet_hdr) { + if (vnet_hdr_sz) { err = virtio_net_hdr_to_skb(skb, &vnet_hdr, vio_le()); if (err) goto out_free; - len += sizeof(vnet_hdr); + len += vnet_hdr_sz; virtio_net_hdr_set_proto(skb, &vnet_hdr); } @@ -3408,7 +3416,7 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, struct sock *sk = sock->sk; struct sk_buff *skb; int copied, err; - int vnet_hdr_len = 0; + int vnet_hdr_len = READ_ONCE(pkt_sk(sk)->vnet_hdr_sz); unsigned int origlen = 0; err = -EINVAL; @@ -3449,11 +3457,10 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, packet_rcv_try_clear_pressure(pkt_sk(sk)); - if (packet_sock_flag(pkt_sk(sk), PACKET_SOCK_HAS_VNET_HDR)) { - err = packet_rcv_vnet(msg, skb, &len); + if (vnet_hdr_len) { + err = packet_rcv_vnet(msg, skb, &len, vnet_hdr_len); if (err) goto out_free; - vnet_hdr_len = sizeof(struct virtio_net_hdr); } /* You lose any data beyond the buffer you gave. If it worries @@ -3915,8 +3922,9 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval, return 0; } case PACKET_VNET_HDR: + case PACKET_VNET_HDR_SZ: { - int val; + int val, hdr_len; if (sock->type != SOCK_RAW) return -EINVAL; @@ -3925,11 +3933,19 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval, if (copy_from_sockptr(&val, optval, sizeof(val))) return -EFAULT; + if (optname == PACKET_VNET_HDR_SZ) { + if (val && val != sizeof(struct virtio_net_hdr) && + val != sizeof(struct virtio_net_hdr_mrg_rxbuf)) + return -EINVAL; + hdr_len = val; + } else { + hdr_len = val ? sizeof(struct virtio_net_hdr) : 0; + } lock_sock(sk); if (po->rx_ring.pg_vec || po->tx_ring.pg_vec) { ret = -EBUSY; } else { - packet_sock_flag_set(po, PACKET_SOCK_HAS_VNET_HDR, val); + WRITE_ONCE(po->vnet_hdr_sz, hdr_len); ret = 0; } release_sock(sk); @@ -4062,7 +4078,10 @@ static int packet_getsockopt(struct socket *sock, int level, int optname, val = packet_sock_flag(po, PACKET_SOCK_ORIGDEV); break; case PACKET_VNET_HDR: - val = packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR); + val = !!READ_ONCE(po->vnet_hdr_sz); + break; + case PACKET_VNET_HDR_SZ: + val = READ_ONCE(po->vnet_hdr_sz); break; case PACKET_VERSION: val = po->tp_version; diff --git a/net/packet/diag.c b/net/packet/diag.c index de4ced5cf3e8..d0c4eda4cdc6 100644 --- a/net/packet/diag.c +++ b/net/packet/diag.c @@ -27,7 +27,7 @@ static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb) pinfo.pdi_flags |= PDI_AUXDATA; if (packet_sock_flag(po, PACKET_SOCK_ORIGDEV)) pinfo.pdi_flags |= PDI_ORIGDEV; - if (packet_sock_flag(po, PACKET_SOCK_HAS_VNET_HDR)) + if (READ_ONCE(po->vnet_hdr_sz)) pinfo.pdi_flags |= PDI_VNETHDR; if (packet_sock_flag(po, PACKET_SOCK_TP_LOSS)) pinfo.pdi_flags |= PDI_LOSS; diff --git a/net/packet/internal.h b/net/packet/internal.h index 27930f69f368..63f4865202c1 100644 --- a/net/packet/internal.h +++ b/net/packet/internal.h @@ -118,6 +118,7 @@ struct packet_sock { struct mutex pg_vec_lock; unsigned long flags; int ifindex; /* bound device */ + u8 vnet_hdr_sz; __be16 num; struct packet_rollover *rollover; struct packet_mclist *mclist; @@ -139,7 +140,6 @@ enum packet_sock_flags { PACKET_SOCK_AUXDATA, PACKET_SOCK_TX_HAS_OFF, PACKET_SOCK_TP_LOSS, - PACKET_SOCK_HAS_VNET_HDR, PACKET_SOCK_RUNNING, PACKET_SOCK_PRESSURE, PACKET_SOCK_QDISC_BYPASS, -- cgit v1.2.3 From 84601d6ee68ae820dec97450934797046d62db4b Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 21 Apr 2023 19:02:54 +0200 Subject: bpf: add bpf_link support for BPF_NETFILTER programs Add bpf_link support skeleton. To keep this reviewable, no bpf program can be invoked yet, if a program is attached only a c-stub is called and not the actual bpf program. Defaults to 'y' if both netfilter and bpf syscall are enabled in kconfig. Uapi example usage: union bpf_attr attr = { }; attr.link_create.prog_fd = progfd; attr.link_create.attach_type = 0; /* unused */ attr.link_create.netfilter.pf = PF_INET; attr.link_create.netfilter.hooknum = NF_INET_LOCAL_IN; attr.link_create.netfilter.priority = -128; err = bpf(BPF_LINK_CREATE, &attr, sizeof(attr)); ... this would attach progfd to ipv4:input hook. Such hook gets removed automatically if the calling program exits. BPF_NETFILTER program invocation is added in followup change. NF_HOOK_OP_BPF enum will eventually be read from nfnetlink_hook, it allows to tell userspace which program is attached at the given hook when user runs 'nft hook list' command rather than just the priority and not-very-helpful 'this hook runs a bpf prog but I can't tell which one'. Will also be used to disallow registration of two bpf programs with same priority in a followup patch. v4: arm32 cmpxchg only supports 32bit operand s/prio/priority/ v3: restrict prog attachment to ip/ip6 for now, lets lift restrictions if more use cases pop up (arptables, ebtables, netdev ingress/egress etc). Signed-off-by: Florian Westphal Link: https://lore.kernel.org/r/20230421170300.24115-2-fw@strlen.de Signed-off-by: Alexei Starovoitov --- include/linux/netfilter.h | 1 + include/net/netfilter/nf_bpf_link.h | 10 +++ include/uapi/linux/bpf.h | 14 ++++ kernel/bpf/syscall.c | 6 ++ net/netfilter/Kconfig | 3 + net/netfilter/Makefile | 1 + net/netfilter/nf_bpf_link.c | 159 ++++++++++++++++++++++++++++++++++++ 7 files changed, 194 insertions(+) create mode 100644 include/net/netfilter/nf_bpf_link.h create mode 100644 net/netfilter/nf_bpf_link.c (limited to 'net') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index c8e03bcaecaa..0762444e3767 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -80,6 +80,7 @@ typedef unsigned int nf_hookfn(void *priv, enum nf_hook_ops_type { NF_HOOK_OP_UNDEFINED, NF_HOOK_OP_NF_TABLES, + NF_HOOK_OP_BPF, }; struct nf_hook_ops { diff --git a/include/net/netfilter/nf_bpf_link.h b/include/net/netfilter/nf_bpf_link.h new file mode 100644 index 000000000000..eeaeaf3d15de --- /dev/null +++ b/include/net/netfilter/nf_bpf_link.h @@ -0,0 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#if IS_ENABLED(CONFIG_NETFILTER_BPF_LINK) +int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog); +#else +static inline int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) +{ + return -EOPNOTSUPP; +} +#endif diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 4b20a7269bee..1bb11a6ee667 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -986,6 +986,7 @@ enum bpf_prog_type { BPF_PROG_TYPE_LSM, BPF_PROG_TYPE_SK_LOOKUP, BPF_PROG_TYPE_SYSCALL, /* a program that can execute syscalls */ + BPF_PROG_TYPE_NETFILTER, }; enum bpf_attach_type { @@ -1050,6 +1051,7 @@ enum bpf_link_type { BPF_LINK_TYPE_PERF_EVENT = 7, BPF_LINK_TYPE_KPROBE_MULTI = 8, BPF_LINK_TYPE_STRUCT_OPS = 9, + BPF_LINK_TYPE_NETFILTER = 10, MAX_BPF_LINK_TYPE, }; @@ -1560,6 +1562,12 @@ union bpf_attr { */ __u64 cookie; } tracing; + struct { + __u32 pf; + __u32 hooknum; + __s32 priority; + __u32 flags; + } netfilter; }; } link_create; @@ -6410,6 +6418,12 @@ struct bpf_link_info { struct { __u32 map_id; } struct_ops; + struct { + __u32 pf; + __u32 hooknum; + __s32 priority; + __u32 flags; + } netfilter; }; } __attribute__((aligned(8))); diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index bcf1a1920ddd..14f39c1e573e 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -35,6 +35,7 @@ #include #include #include +#include #define IS_FD_ARRAY(map) ((map)->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || \ (map)->map_type == BPF_MAP_TYPE_CGROUP_ARRAY || \ @@ -2462,6 +2463,7 @@ static bool is_net_admin_prog_type(enum bpf_prog_type prog_type) case BPF_PROG_TYPE_CGROUP_SYSCTL: case BPF_PROG_TYPE_SOCK_OPS: case BPF_PROG_TYPE_EXT: /* extends any prog */ + case BPF_PROG_TYPE_NETFILTER: return true; case BPF_PROG_TYPE_CGROUP_SKB: /* always unpriv */ @@ -4588,6 +4590,7 @@ static int link_create(union bpf_attr *attr, bpfptr_t uattr) switch (prog->type) { case BPF_PROG_TYPE_EXT: + case BPF_PROG_TYPE_NETFILTER: break; case BPF_PROG_TYPE_PERF_EVENT: case BPF_PROG_TYPE_TRACEPOINT: @@ -4654,6 +4657,9 @@ static int link_create(union bpf_attr *attr, bpfptr_t uattr) case BPF_PROG_TYPE_XDP: ret = bpf_xdp_link_attach(attr, prog); break; + case BPF_PROG_TYPE_NETFILTER: + ret = bpf_nf_link_attach(attr, prog); + break; #endif case BPF_PROG_TYPE_PERF_EVENT: case BPF_PROG_TYPE_TRACEPOINT: diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index d0bf630482c1..441d1f134110 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -30,6 +30,9 @@ config NETFILTER_FAMILY_BRIDGE config NETFILTER_FAMILY_ARP bool +config NETFILTER_BPF_LINK + def_bool BPF_SYSCALL + config NETFILTER_NETLINK_HOOK tristate "Netfilter base hook dump support" depends on NETFILTER_ADVANCED diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 5ffef1cd6143..d4958e7e7631 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -22,6 +22,7 @@ nf_conntrack-$(CONFIG_DEBUG_INFO_BTF) += nf_conntrack_bpf.o endif obj-$(CONFIG_NETFILTER) = netfilter.o +obj-$(CONFIG_NETFILTER_BPF_LINK) += nf_bpf_link.o obj-$(CONFIG_NETFILTER_NETLINK) += nfnetlink.o obj-$(CONFIG_NETFILTER_NETLINK_ACCT) += nfnetlink_acct.o diff --git a/net/netfilter/nf_bpf_link.c b/net/netfilter/nf_bpf_link.c new file mode 100644 index 000000000000..efa4f3390742 --- /dev/null +++ b/net/netfilter/nf_bpf_link.c @@ -0,0 +1,159 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include + +#include +#include + +static unsigned int nf_hook_run_bpf(void *bpf_prog, struct sk_buff *skb, + const struct nf_hook_state *s) +{ + return NF_ACCEPT; +} + +struct bpf_nf_link { + struct bpf_link link; + struct nf_hook_ops hook_ops; + struct net *net; + u32 dead; +}; + +static void bpf_nf_link_release(struct bpf_link *link) +{ + struct bpf_nf_link *nf_link = container_of(link, struct bpf_nf_link, link); + + if (nf_link->dead) + return; + + /* prevent hook-not-found warning splat from netfilter core when + * .detach was already called + */ + if (!cmpxchg(&nf_link->dead, 0, 1)) + nf_unregister_net_hook(nf_link->net, &nf_link->hook_ops); +} + +static void bpf_nf_link_dealloc(struct bpf_link *link) +{ + struct bpf_nf_link *nf_link = container_of(link, struct bpf_nf_link, link); + + kfree(nf_link); +} + +static int bpf_nf_link_detach(struct bpf_link *link) +{ + bpf_nf_link_release(link); + return 0; +} + +static void bpf_nf_link_show_info(const struct bpf_link *link, + struct seq_file *seq) +{ + struct bpf_nf_link *nf_link = container_of(link, struct bpf_nf_link, link); + + seq_printf(seq, "pf:\t%u\thooknum:\t%u\tprio:\t%d\n", + nf_link->hook_ops.pf, nf_link->hook_ops.hooknum, + nf_link->hook_ops.priority); +} + +static int bpf_nf_link_fill_link_info(const struct bpf_link *link, + struct bpf_link_info *info) +{ + struct bpf_nf_link *nf_link = container_of(link, struct bpf_nf_link, link); + + info->netfilter.pf = nf_link->hook_ops.pf; + info->netfilter.hooknum = nf_link->hook_ops.hooknum; + info->netfilter.priority = nf_link->hook_ops.priority; + info->netfilter.flags = 0; + + return 0; +} + +static int bpf_nf_link_update(struct bpf_link *link, struct bpf_prog *new_prog, + struct bpf_prog *old_prog) +{ + return -EOPNOTSUPP; +} + +static const struct bpf_link_ops bpf_nf_link_lops = { + .release = bpf_nf_link_release, + .dealloc = bpf_nf_link_dealloc, + .detach = bpf_nf_link_detach, + .show_fdinfo = bpf_nf_link_show_info, + .fill_link_info = bpf_nf_link_fill_link_info, + .update_prog = bpf_nf_link_update, +}; + +static int bpf_nf_check_pf_and_hooks(const union bpf_attr *attr) +{ + switch (attr->link_create.netfilter.pf) { + case NFPROTO_IPV4: + case NFPROTO_IPV6: + if (attr->link_create.netfilter.hooknum >= NF_INET_NUMHOOKS) + return -EPROTO; + break; + default: + return -EAFNOSUPPORT; + } + + if (attr->link_create.netfilter.flags) + return -EOPNOTSUPP; + + /* make sure conntrack confirm is always last. + * + * In the future, if userspace can e.g. request defrag, then + * "defrag_requested && prio before NF_IP_PRI_CONNTRACK_DEFRAG" + * should fail. + */ + switch (attr->link_create.netfilter.priority) { + case NF_IP_PRI_FIRST: return -ERANGE; /* sabotage_in and other warts */ + case NF_IP_PRI_LAST: return -ERANGE; /* e.g. conntrack confirm */ + } + + return 0; +} + +int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) +{ + struct net *net = current->nsproxy->net_ns; + struct bpf_link_primer link_primer; + struct bpf_nf_link *link; + int err; + + if (attr->link_create.flags) + return -EINVAL; + + err = bpf_nf_check_pf_and_hooks(attr); + if (err) + return err; + + link = kzalloc(sizeof(*link), GFP_USER); + if (!link) + return -ENOMEM; + + bpf_link_init(&link->link, BPF_LINK_TYPE_NETFILTER, &bpf_nf_link_lops, prog); + + link->hook_ops.hook = nf_hook_run_bpf; + link->hook_ops.hook_ops_type = NF_HOOK_OP_BPF; + link->hook_ops.priv = prog; + + link->hook_ops.pf = attr->link_create.netfilter.pf; + link->hook_ops.priority = attr->link_create.netfilter.priority; + link->hook_ops.hooknum = attr->link_create.netfilter.hooknum; + + link->net = net; + link->dead = false; + + err = bpf_link_prime(&link->link, &link_primer); + if (err) { + kfree(link); + return err; + } + + err = nf_register_net_hook(net, &link->hook_ops); + if (err) { + bpf_link_cleanup(&link_primer); + return err; + } + + return bpf_link_settle(&link_primer); +} -- cgit v1.2.3 From fd9c663b9ad67dedfc9a3fd3429ddd3e83782b4d Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 21 Apr 2023 19:02:55 +0200 Subject: bpf: minimal support for programs hooked into netfilter framework This adds minimal support for BPF_PROG_TYPE_NETFILTER bpf programs that will be invoked via the NF_HOOK() points in the ip stack. Invocation incurs an indirect call. This is not a necessity: Its possible to add 'DEFINE_BPF_DISPATCHER(nf_progs)' and handle the program invocation with the same method already done for xdp progs. This isn't done here to keep the size of this chunk down. Verifier restricts verdicts to either DROP or ACCEPT. Signed-off-by: Florian Westphal Link: https://lore.kernel.org/r/20230421170300.24115-3-fw@strlen.de Signed-off-by: Alexei Starovoitov --- include/linux/bpf_types.h | 4 +++ include/net/netfilter/nf_bpf_link.h | 5 +++ kernel/bpf/btf.c | 6 ++++ kernel/bpf/verifier.c | 3 ++ net/core/filter.c | 1 + net/netfilter/nf_bpf_link.c | 70 ++++++++++++++++++++++++++++++++++++- 6 files changed, 88 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/linux/bpf_types.h b/include/linux/bpf_types.h index d4ee3ccd3753..39a999abb0ce 100644 --- a/include/linux/bpf_types.h +++ b/include/linux/bpf_types.h @@ -79,6 +79,10 @@ BPF_PROG_TYPE(BPF_PROG_TYPE_LSM, lsm, #endif BPF_PROG_TYPE(BPF_PROG_TYPE_SYSCALL, bpf_syscall, void *, void *) +#ifdef CONFIG_NETFILTER +BPF_PROG_TYPE(BPF_PROG_TYPE_NETFILTER, netfilter, + struct bpf_nf_ctx, struct bpf_nf_ctx) +#endif BPF_MAP_TYPE(BPF_MAP_TYPE_ARRAY, array_map_ops) BPF_MAP_TYPE(BPF_MAP_TYPE_PERCPU_ARRAY, percpu_array_map_ops) diff --git a/include/net/netfilter/nf_bpf_link.h b/include/net/netfilter/nf_bpf_link.h index eeaeaf3d15de..6c984b0ea838 100644 --- a/include/net/netfilter/nf_bpf_link.h +++ b/include/net/netfilter/nf_bpf_link.h @@ -1,5 +1,10 @@ /* SPDX-License-Identifier: GPL-2.0 */ +struct bpf_nf_ctx { + const struct nf_hook_state *state; + struct sk_buff *skb; +}; + #if IS_ENABLED(CONFIG_NETFILTER_BPF_LINK) int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog); #else diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index 7db4ec125fbd..6b682b8e4b50 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -25,6 +25,9 @@ #include #include #include + +#include + #include #include "../tools/lib/bpf/relo_core.h" @@ -212,6 +215,7 @@ enum btf_kfunc_hook { BTF_KFUNC_HOOK_SK_SKB, BTF_KFUNC_HOOK_SOCKET_FILTER, BTF_KFUNC_HOOK_LWT, + BTF_KFUNC_HOOK_NETFILTER, BTF_KFUNC_HOOK_MAX, }; @@ -7802,6 +7806,8 @@ static int bpf_prog_type_to_kfunc_hook(enum bpf_prog_type prog_type) case BPF_PROG_TYPE_LWT_XMIT: case BPF_PROG_TYPE_LWT_SEG6LOCAL: return BTF_KFUNC_HOOK_LWT; + case BPF_PROG_TYPE_NETFILTER: + return BTF_KFUNC_HOOK_NETFILTER; default: return BTF_KFUNC_HOOK_MAX; } diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 1e05355facdc..fc7281d39e46 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -13816,6 +13816,9 @@ static int check_return_code(struct bpf_verifier_env *env) } break; + case BPF_PROG_TYPE_NETFILTER: + range = tnum_range(NF_DROP, NF_ACCEPT); + break; case BPF_PROG_TYPE_EXT: /* freplace program can return anything as its return value * depends on the to-be-replaced kernel func or bpf program. diff --git a/net/core/filter.c b/net/core/filter.c index 44fb997434ad..d9ce04ca22ce 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -11717,6 +11717,7 @@ static int __init bpf_kfunc_init(void) ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_IN, &bpf_kfunc_set_skb); ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_XMIT, &bpf_kfunc_set_skb); ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_LWT_SEG6LOCAL, &bpf_kfunc_set_skb); + ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_NETFILTER, &bpf_kfunc_set_skb); return ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_XDP, &bpf_kfunc_set_xdp); } late_initcall(bpf_kfunc_init); diff --git a/net/netfilter/nf_bpf_link.c b/net/netfilter/nf_bpf_link.c index efa4f3390742..49cfc5215386 100644 --- a/net/netfilter/nf_bpf_link.c +++ b/net/netfilter/nf_bpf_link.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 #include +#include #include #include @@ -8,7 +9,13 @@ static unsigned int nf_hook_run_bpf(void *bpf_prog, struct sk_buff *skb, const struct nf_hook_state *s) { - return NF_ACCEPT; + const struct bpf_prog *prog = bpf_prog; + struct bpf_nf_ctx ctx = { + .state = s, + .skb = skb, + }; + + return bpf_prog_run(prog, &ctx); } struct bpf_nf_link { @@ -157,3 +164,64 @@ int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) return bpf_link_settle(&link_primer); } + +const struct bpf_prog_ops netfilter_prog_ops = { +}; + +static bool nf_ptr_to_btf_id(struct bpf_insn_access_aux *info, const char *name) +{ + struct btf *btf; + s32 type_id; + + btf = bpf_get_btf_vmlinux(); + if (IS_ERR_OR_NULL(btf)) + return false; + + type_id = btf_find_by_name_kind(btf, name, BTF_KIND_STRUCT); + if (WARN_ON_ONCE(type_id < 0)) + return false; + + info->btf = btf; + info->btf_id = type_id; + info->reg_type = PTR_TO_BTF_ID | PTR_TRUSTED; + return true; +} + +static bool nf_is_valid_access(int off, int size, enum bpf_access_type type, + const struct bpf_prog *prog, + struct bpf_insn_access_aux *info) +{ + if (off < 0 || off >= sizeof(struct bpf_nf_ctx)) + return false; + + if (type == BPF_WRITE) + return false; + + switch (off) { + case bpf_ctx_range(struct bpf_nf_ctx, skb): + if (size != sizeof_field(struct bpf_nf_ctx, skb)) + return false; + + return nf_ptr_to_btf_id(info, "sk_buff"); + case bpf_ctx_range(struct bpf_nf_ctx, state): + if (size != sizeof_field(struct bpf_nf_ctx, state)) + return false; + + return nf_ptr_to_btf_id(info, "nf_hook_state"); + default: + return false; + } + + return false; +} + +static const struct bpf_func_proto * +bpf_nf_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) +{ + return bpf_base_func_proto(func_id); +} + +const struct bpf_verifier_ops netfilter_verifier_ops = { + .is_valid_access = nf_is_valid_access, + .get_func_proto = bpf_nf_func_proto, +}; -- cgit v1.2.3 From 506a74db7e019a277e987fa65654bdd953859d5b Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 21 Apr 2023 19:02:56 +0200 Subject: netfilter: nfnetlink hook: dump bpf prog id This allows userspace ("nft list hooks") to show which bpf program is attached to which hook. Without this, user only knows bpf prog is attached at prio x, y, z at INPUT and FORWARD, but can't tell which program is where. v4: kdoc fixups (Simon Horman) Link: https://lore.kernel.org/bpf/ZEELzpNCnYJuZyod@corigine.com/ Signed-off-by: Florian Westphal Link: https://lore.kernel.org/r/20230421170300.24115-4-fw@strlen.de Signed-off-by: Alexei Starovoitov --- include/uapi/linux/netfilter/nfnetlink_hook.h | 24 +++++++- net/netfilter/nfnetlink_hook.c | 81 ++++++++++++++++++++++----- 2 files changed, 89 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/include/uapi/linux/netfilter/nfnetlink_hook.h b/include/uapi/linux/netfilter/nfnetlink_hook.h index bbcd285b22e1..84a561a74b98 100644 --- a/include/uapi/linux/netfilter/nfnetlink_hook.h +++ b/include/uapi/linux/netfilter/nfnetlink_hook.h @@ -32,8 +32,12 @@ enum nfnl_hook_attributes { /** * enum nfnl_hook_chain_info_attributes - chain description * - * NFNLA_HOOK_INFO_DESC: nft chain and table name (enum nft_table_attributes) (NLA_NESTED) - * NFNLA_HOOK_INFO_TYPE: chain type (enum nfnl_hook_chaintype) (NLA_U32) + * @NFNLA_HOOK_INFO_DESC: nft chain and table name (NLA_NESTED) + * @NFNLA_HOOK_INFO_TYPE: chain type (enum nfnl_hook_chaintype) (NLA_U32) + * + * NFNLA_HOOK_INFO_DESC depends on NFNLA_HOOK_INFO_TYPE value: + * NFNL_HOOK_TYPE_NFTABLES: enum nft_table_attributes + * NFNL_HOOK_TYPE_BPF: enum nfnl_hook_bpf_attributes */ enum nfnl_hook_chain_info_attributes { NFNLA_HOOK_INFO_UNSPEC, @@ -55,10 +59,24 @@ enum nfnl_hook_chain_desc_attributes { /** * enum nfnl_hook_chaintype - chain type * - * @NFNL_HOOK_TYPE_NFTABLES nf_tables base chain + * @NFNL_HOOK_TYPE_NFTABLES: nf_tables base chain + * @NFNL_HOOK_TYPE_BPF: bpf program */ enum nfnl_hook_chaintype { NFNL_HOOK_TYPE_NFTABLES = 0x1, + NFNL_HOOK_TYPE_BPF, +}; + +/** + * enum nfnl_hook_bpf_attributes - bpf prog description + * + * @NFNLA_HOOK_BPF_ID: bpf program id (NLA_U32) + */ +enum nfnl_hook_bpf_attributes { + NFNLA_HOOK_BPF_UNSPEC, + NFNLA_HOOK_BPF_ID, + __NFNLA_HOOK_BPF_MAX, }; +#define NFNLA_HOOK_BPF_MAX (__NFNLA_HOOK_BPF_MAX - 1) #endif /* _NFNL_HOOK_H */ diff --git a/net/netfilter/nfnetlink_hook.c b/net/netfilter/nfnetlink_hook.c index 8120aadf6a0f..ade8ee1988b1 100644 --- a/net/netfilter/nfnetlink_hook.c +++ b/net/netfilter/nfnetlink_hook.c @@ -5,6 +5,7 @@ * Author: Florian Westphal */ +#include #include #include #include @@ -57,35 +58,76 @@ struct nfnl_dump_hook_data { u8 hook; }; +static struct nlattr *nfnl_start_info_type(struct sk_buff *nlskb, enum nfnl_hook_chaintype t) +{ + struct nlattr *nest = nla_nest_start(nlskb, NFNLA_HOOK_CHAIN_INFO); + int ret; + + if (!nest) + return NULL; + + ret = nla_put_be32(nlskb, NFNLA_HOOK_INFO_TYPE, htonl(t)); + if (ret == 0) + return nest; + + nla_nest_cancel(nlskb, nest); + return NULL; +} + +static int nfnl_hook_put_bpf_prog_info(struct sk_buff *nlskb, + const struct nfnl_dump_hook_data *ctx, + unsigned int seq, + const struct bpf_prog *prog) +{ + struct nlattr *nest, *nest2; + int ret; + + if (!IS_ENABLED(CONFIG_NETFILTER_BPF_LINK)) + return 0; + + if (WARN_ON_ONCE(!prog)) + return 0; + + nest = nfnl_start_info_type(nlskb, NFNL_HOOK_TYPE_BPF); + if (!nest) + return -EMSGSIZE; + + nest2 = nla_nest_start(nlskb, NFNLA_HOOK_INFO_DESC); + if (!nest2) + goto cancel_nest; + + ret = nla_put_be32(nlskb, NFNLA_HOOK_BPF_ID, htonl(prog->aux->id)); + if (ret) + goto cancel_nest; + + nla_nest_end(nlskb, nest2); + nla_nest_end(nlskb, nest); + return 0; + +cancel_nest: + nla_nest_cancel(nlskb, nest); + return -EMSGSIZE; +} + static int nfnl_hook_put_nft_chain_info(struct sk_buff *nlskb, const struct nfnl_dump_hook_data *ctx, unsigned int seq, - const struct nf_hook_ops *ops) + struct nft_chain *chain) { struct net *net = sock_net(nlskb->sk); struct nlattr *nest, *nest2; - struct nft_chain *chain; int ret = 0; - if (ops->hook_ops_type != NF_HOOK_OP_NF_TABLES) - return 0; - - chain = ops->priv; if (WARN_ON_ONCE(!chain)) return 0; if (!nft_is_active(net, chain)) return 0; - nest = nla_nest_start(nlskb, NFNLA_HOOK_CHAIN_INFO); + nest = nfnl_start_info_type(nlskb, NFNL_HOOK_TYPE_NFTABLES); if (!nest) return -EMSGSIZE; - ret = nla_put_be32(nlskb, NFNLA_HOOK_INFO_TYPE, - htonl(NFNL_HOOK_TYPE_NFTABLES)); - if (ret) - goto cancel_nest; - nest2 = nla_nest_start(nlskb, NFNLA_HOOK_INFO_DESC); if (!nest2) goto cancel_nest; @@ -171,7 +213,20 @@ static int nfnl_hook_dump_one(struct sk_buff *nlskb, if (ret) goto nla_put_failure; - ret = nfnl_hook_put_nft_chain_info(nlskb, ctx, seq, ops); + switch (ops->hook_ops_type) { + case NF_HOOK_OP_NF_TABLES: + ret = nfnl_hook_put_nft_chain_info(nlskb, ctx, seq, ops->priv); + break; + case NF_HOOK_OP_BPF: + ret = nfnl_hook_put_bpf_prog_info(nlskb, ctx, seq, ops->priv); + break; + case NF_HOOK_OP_UNDEFINED: + break; + default: + WARN_ON_ONCE(1); + break; + } + if (ret) goto nla_put_failure; -- cgit v1.2.3 From 0bdc6da88f5bac0f3f272cb6f545c2cc70e8c66a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 21 Apr 2023 19:02:57 +0200 Subject: netfilter: disallow bpf hook attachment at same priority This is just to avoid ordering issues between multiple bpf programs, this could be removed later in case it turns out to be too cautious. bpf prog could still be shared with non-bpf hook, otherwise we'd have to make conntrack hook registration fail just because a bpf program has same priority. Signed-off-by: Florian Westphal Link: https://lore.kernel.org/r/20230421170300.24115-5-fw@strlen.de Signed-off-by: Alexei Starovoitov --- net/netfilter/core.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'net') diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 358220b58521..f0783e42108b 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -119,6 +119,18 @@ nf_hook_entries_grow(const struct nf_hook_entries *old, for (i = 0; i < old_entries; i++) { if (orig_ops[i] != &dummy_ops) alloc_entries++; + + /* Restrict BPF hook type to force a unique priority, not + * shared at attach time. + * + * This is mainly to avoid ordering issues between two + * different bpf programs, this doesn't prevent a normal + * hook at same priority as a bpf one (we don't want to + * prevent defrag, conntrack, iptables etc from attaching). + */ + if (reg->priority == orig_ops[i]->priority && + reg->hook_ops_type == NF_HOOK_OP_BPF) + return ERR_PTR(-EBUSY); } } -- cgit v1.2.3 From 2b99ef22e0d237e08bfc437e7d051f78f352aeb2 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 21 Apr 2023 19:02:59 +0200 Subject: bpf: add test_run support for netfilter program type add glue code so a bpf program can be run using userspace-provided netfilter state and packet/skb. Default is to use ipv4:output hook point, but this can be overridden by userspace. Userspace provided netfilter state is restricted, only hook and protocol families can be overridden and only to ipv4/ipv6. Signed-off-by: Florian Westphal Link: https://lore.kernel.org/r/20230421170300.24115-7-fw@strlen.de Signed-off-by: Alexei Starovoitov --- include/linux/bpf.h | 3 + net/bpf/test_run.c | 158 ++++++++++++++++++++++++++++++++++++++++++++ net/netfilter/nf_bpf_link.c | 1 + 3 files changed, 162 insertions(+) (limited to 'net') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 18b592fde896..e53ceee1df37 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -2264,6 +2264,9 @@ int bpf_prog_test_run_raw_tp(struct bpf_prog *prog, int bpf_prog_test_run_sk_lookup(struct bpf_prog *prog, const union bpf_attr *kattr, union bpf_attr __user *uattr); +int bpf_prog_test_run_nf(struct bpf_prog *prog, + const union bpf_attr *kattr, + union bpf_attr __user *uattr); bool btf_ctx_access(int off, int size, enum bpf_access_type type, const struct bpf_prog *prog, struct bpf_insn_access_aux *info); diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index f170e8a17974..e79e3a415ca9 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -19,7 +19,9 @@ #include #include #include +#include #include +#include #define CREATE_TRACE_POINTS #include @@ -1691,6 +1693,162 @@ out: return err; } +static int verify_and_copy_hook_state(struct nf_hook_state *state, + const struct nf_hook_state *user, + struct net_device *dev) +{ + if (user->in || user->out) + return -EINVAL; + + if (user->net || user->sk || user->okfn) + return -EINVAL; + + switch (user->pf) { + case NFPROTO_IPV4: + case NFPROTO_IPV6: + switch (state->hook) { + case NF_INET_PRE_ROUTING: + state->in = dev; + break; + case NF_INET_LOCAL_IN: + state->in = dev; + break; + case NF_INET_FORWARD: + state->in = dev; + state->out = dev; + break; + case NF_INET_LOCAL_OUT: + state->out = dev; + break; + case NF_INET_POST_ROUTING: + state->out = dev; + break; + } + + break; + default: + return -EINVAL; + } + + state->pf = user->pf; + state->hook = user->hook; + + return 0; +} + +static __be16 nfproto_eth(int nfproto) +{ + switch (nfproto) { + case NFPROTO_IPV4: + return htons(ETH_P_IP); + case NFPROTO_IPV6: + break; + } + + return htons(ETH_P_IPV6); +} + +int bpf_prog_test_run_nf(struct bpf_prog *prog, + const union bpf_attr *kattr, + union bpf_attr __user *uattr) +{ + struct net *net = current->nsproxy->net_ns; + struct net_device *dev = net->loopback_dev; + struct nf_hook_state *user_ctx, hook_state = { + .pf = NFPROTO_IPV4, + .hook = NF_INET_LOCAL_OUT, + }; + u32 size = kattr->test.data_size_in; + u32 repeat = kattr->test.repeat; + struct bpf_nf_ctx ctx = { + .state = &hook_state, + }; + struct sk_buff *skb = NULL; + u32 retval, duration; + void *data; + int ret; + + if (kattr->test.flags || kattr->test.cpu || kattr->test.batch_size) + return -EINVAL; + + if (size < sizeof(struct iphdr)) + return -EINVAL; + + data = bpf_test_init(kattr, kattr->test.data_size_in, size, + NET_SKB_PAD + NET_IP_ALIGN, + SKB_DATA_ALIGN(sizeof(struct skb_shared_info))); + if (IS_ERR(data)) + return PTR_ERR(data); + + if (!repeat) + repeat = 1; + + user_ctx = bpf_ctx_init(kattr, sizeof(struct nf_hook_state)); + if (IS_ERR(user_ctx)) { + kfree(data); + return PTR_ERR(user_ctx); + } + + if (user_ctx) { + ret = verify_and_copy_hook_state(&hook_state, user_ctx, dev); + if (ret) + goto out; + } + + skb = slab_build_skb(data); + if (!skb) { + ret = -ENOMEM; + goto out; + } + + data = NULL; /* data released via kfree_skb */ + + skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN); + __skb_put(skb, size); + + ret = -EINVAL; + + if (hook_state.hook != NF_INET_LOCAL_OUT) { + if (size < ETH_HLEN + sizeof(struct iphdr)) + goto out; + + skb->protocol = eth_type_trans(skb, dev); + switch (skb->protocol) { + case htons(ETH_P_IP): + if (hook_state.pf == NFPROTO_IPV4) + break; + goto out; + case htons(ETH_P_IPV6): + if (size < ETH_HLEN + sizeof(struct ipv6hdr)) + goto out; + if (hook_state.pf == NFPROTO_IPV6) + break; + goto out; + default: + ret = -EPROTO; + goto out; + } + + skb_reset_network_header(skb); + } else { + skb->protocol = nfproto_eth(hook_state.pf); + } + + ctx.skb = skb; + + ret = bpf_test_run(prog, &ctx, repeat, &retval, &duration, false); + if (ret) + goto out; + + ret = bpf_test_finish(kattr, uattr, NULL, NULL, 0, retval, duration); + +out: + kfree(user_ctx); + kfree_skb(skb); + kfree(data); + return ret; +} + static const struct btf_kfunc_id_set bpf_prog_test_kfunc_set = { .owner = THIS_MODULE, .set = &test_sk_check_kfunc_ids, diff --git a/net/netfilter/nf_bpf_link.c b/net/netfilter/nf_bpf_link.c index 49cfc5215386..c36da56d756f 100644 --- a/net/netfilter/nf_bpf_link.c +++ b/net/netfilter/nf_bpf_link.c @@ -166,6 +166,7 @@ int bpf_nf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog) } const struct bpf_prog_ops netfilter_prog_ops = { + .test_run = bpf_prog_test_run_nf, }; static bool nf_ptr_to_btf_id(struct bpf_insn_access_aux *info, const char *name) -- cgit v1.2.3 From e38fbfa972eb0944ce4387caf23d166c1df4b114 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 11 Apr 2023 16:29:45 +0200 Subject: netfilter: nf_tables: merge nft_rules_old structure and end of ruleblob marker In order to free the rules in a chain via call_rcu, the rule array used to stash a rcu_head and space for a pointer at the end of the rule array. When the current nft_rule_dp blob format got added in 2c865a8a28a1 ("netfilter: nf_tables: add rule blob layout"), this results in a double-trailer: size (unsigned long) struct nft_rule_dp struct nft_expr ... struct nft_rule_dp struct nft_expr ... struct nft_rule_dp (is_last=1) // Trailer The trailer, struct nft_rule_dp (is_last=1), is not accounted for in size, so it can be located via start_addr + size. Because the rcu_head is stored after 'start+size' as well this means the is_last trailer is *aliased* to the rcu_head (struct nft_rules_old). This is harmless, because at this time the nft_do_chain function never evaluates/accesses the trailer, it only checks the address boundary: for (; rule < last_rule; rule = nft_rule_next(rule)) { ... But this way the last_rule address has to be stashed in the jump structure to restore it after returning from a chain. nft_do_chain stack usage has become way too big, so put it on a diet. Without this patch is impossible to use for (; !rule->is_last; rule = nft_rule_next(rule)) { ... because on free, the needed update of the rcu_head will clobber the nft_rule_dp is_last bit. Furthermore, also stash the chain pointer in the trailer, this allows to recover the original chain structure from nf_tables_trace infra without a need to place them in the jump struct. After this patch it is trivial to diet the jump stack structure, done in the next two patches. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 55 +++++++++++++++++++++---------------------- 1 file changed, 27 insertions(+), 28 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index e48ab8dfb541..79848a27e640 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -2110,38 +2110,41 @@ static void nft_chain_release_hook(struct nft_chain_hook *hook) module_put(hook->type->owner); } -struct nft_rules_old { +struct nft_rule_dp_last { + struct nft_rule_dp end; /* end of nft_rule_blob marker */ struct rcu_head h; struct nft_rule_blob *blob; + const struct nft_chain *chain; /* for tracing */ }; -static void nft_last_rule(struct nft_rule_blob *blob, const void *ptr) +static void nft_last_rule(const struct nft_chain *chain, const void *ptr) { - struct nft_rule_dp *prule; + struct nft_rule_dp_last *lrule; + + BUILD_BUG_ON(offsetof(struct nft_rule_dp_last, end) != 0); - prule = (struct nft_rule_dp *)ptr; - prule->is_last = 1; + lrule = (struct nft_rule_dp_last *)ptr; + lrule->end.is_last = 1; + lrule->chain = chain; /* blob size does not include the trailer rule */ } -static struct nft_rule_blob *nf_tables_chain_alloc_rules(unsigned int size) +static struct nft_rule_blob *nf_tables_chain_alloc_rules(const struct nft_chain *chain, + unsigned int size) { struct nft_rule_blob *blob; - /* size must include room for the last rule */ - if (size < offsetof(struct nft_rule_dp, data)) - return NULL; - - size += sizeof(struct nft_rule_blob) + sizeof(struct nft_rules_old); if (size > INT_MAX) return NULL; + size += sizeof(struct nft_rule_blob) + sizeof(struct nft_rule_dp_last); + blob = kvmalloc(size, GFP_KERNEL_ACCOUNT); if (!blob) return NULL; blob->size = 0; - nft_last_rule(blob, blob->data); + nft_last_rule(chain, blob->data); return blob; } @@ -2220,7 +2223,6 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask, struct nft_rule_blob *blob; struct nft_trans *trans; struct nft_chain *chain; - unsigned int data_size; int err; if (table->use == UINT_MAX) @@ -2308,8 +2310,7 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask, chain->udlen = nla_len(nla[NFTA_CHAIN_USERDATA]); } - data_size = offsetof(struct nft_rule_dp, data); /* last rule */ - blob = nf_tables_chain_alloc_rules(data_size); + blob = nf_tables_chain_alloc_rules(chain, 0); if (!blob) { err = -ENOMEM; goto err_destroy_chain; @@ -8817,9 +8818,8 @@ static int nf_tables_commit_chain_prepare(struct net *net, struct nft_chain *cha return -ENOMEM; } } - data_size += offsetof(struct nft_rule_dp, data); /* last rule */ - chain->blob_next = nf_tables_chain_alloc_rules(data_size); + chain->blob_next = nf_tables_chain_alloc_rules(chain, data_size); if (!chain->blob_next) return -ENOMEM; @@ -8864,12 +8864,11 @@ static int nf_tables_commit_chain_prepare(struct net *net, struct nft_chain *cha chain->blob_next->size += (unsigned long)(data - (void *)prule); } - prule = (struct nft_rule_dp *)data; - data += offsetof(struct nft_rule_dp, data); if (WARN_ON_ONCE(data > data_boundary)) return -ENOMEM; - nft_last_rule(chain->blob_next, prule); + prule = (struct nft_rule_dp *)data; + nft_last_rule(chain, prule); return 0; } @@ -8890,22 +8889,22 @@ static void nf_tables_commit_chain_prepare_cancel(struct net *net) } } -static void __nf_tables_commit_chain_free_rules_old(struct rcu_head *h) +static void __nf_tables_commit_chain_free_rules(struct rcu_head *h) { - struct nft_rules_old *o = container_of(h, struct nft_rules_old, h); + struct nft_rule_dp_last *l = container_of(h, struct nft_rule_dp_last, h); - kvfree(o->blob); + kvfree(l->blob); } static void nf_tables_commit_chain_free_rules_old(struct nft_rule_blob *blob) { - struct nft_rules_old *old; + struct nft_rule_dp_last *last; - /* rcu_head is after end marker */ - old = (void *)blob + sizeof(*blob) + blob->size; - old->blob = blob; + /* last rule trailer is after end marker */ + last = (void *)blob + sizeof(*blob) + blob->size; + last->blob = blob; - call_rcu(&old->h, __nf_tables_commit_chain_free_rules_old); + call_rcu(&last->h, __nf_tables_commit_chain_free_rules); } static void nf_tables_commit_chain(struct net *net, struct nft_chain *chain) -- cgit v1.2.3 From d4d89e6546e0d1ac09cb9dd353f0cb31c8a8deb1 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 11 Apr 2023 16:29:46 +0200 Subject: netfilter: nf_tables: don't store address of last rule on jump Walk the rule headers until the trailer one (last_bit flag set) instead of stopping at last_rule address. This avoids the need to store the address when jumping to another chain. This cuts size of jumpstack array by one third, on 64bit from 384 to 256 bytes. Still, stack usage is still quite large: scripts/stackusage: nf_tables_core.c:258 nft_do_chain 496 static Next patch will also remove chain pointer. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_core.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index 6ecd0ba2e546..ec3bab751092 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -205,7 +205,6 @@ static noinline void nft_update_chain_stats(const struct nft_chain *chain, struct nft_jumpstack { const struct nft_chain *chain; const struct nft_rule_dp *rule; - const struct nft_rule_dp *last_rule; }; static void expr_call_ops_eval(const struct nft_expr *expr, @@ -259,9 +258,9 @@ unsigned int nft_do_chain(struct nft_pktinfo *pkt, void *priv) { const struct nft_chain *chain = priv, *basechain = chain; - const struct nft_rule_dp *rule, *last_rule; const struct net *net = nft_net(pkt); const struct nft_expr *expr, *last; + const struct nft_rule_dp *rule; struct nft_regs regs = {}; unsigned int stackptr = 0; struct nft_jumpstack jumpstack[NFT_JUMP_STACK_SIZE]; @@ -279,10 +278,9 @@ do_chain: blob = rcu_dereference(chain->blob_gen_0); rule = (struct nft_rule_dp *)blob->data; - last_rule = (void *)blob->data + blob->size; next_rule: regs.verdict.code = NFT_CONTINUE; - for (; rule < last_rule; rule = nft_rule_next(rule)) { + for (; !rule->is_last ; rule = nft_rule_next(rule)) { nft_rule_dp_for_each_expr(expr, last, rule) { if (expr->ops == &nft_cmp_fast_ops) nft_cmp_fast_eval(expr, ®s); @@ -327,7 +325,6 @@ next_rule: return NF_DROP; jumpstack[stackptr].chain = chain; jumpstack[stackptr].rule = nft_rule_next(rule); - jumpstack[stackptr].last_rule = last_rule; stackptr++; fallthrough; case NFT_GOTO: @@ -344,7 +341,6 @@ next_rule: stackptr--; chain = jumpstack[stackptr].chain; rule = jumpstack[stackptr].rule; - last_rule = jumpstack[stackptr].last_rule; goto next_rule; } -- cgit v1.2.3 From 63e9bbbcca60333490e13744ae736d8f988e4950 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 11 Apr 2023 16:29:47 +0200 Subject: netfilter: nf_tables: don't store chain address on jump Now that the rule trailer/end marker and the rcu head reside in the same structure, we no longer need to save/restore the chain pointer when performing/returning from a jump. We can simply let the trace infra walk the evaluated rule until it hits the end marker and then fetch the chain pointer from there. When the rule is NULL (policy tracing), then chain and basechain pointers were already identical, so just use the basechain. This cuts size of jumpstack in half, from 256 to 128 bytes in 64bit, scripts/stackusage says: nf_tables_core.c:251 nft_do_chain 328 static Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 14 ++++++++++++-- net/netfilter/nf_tables_api.c | 7 ------- net/netfilter/nf_tables_core.c | 21 ++++++--------------- net/netfilter/nf_tables_trace.c | 30 ++++++++++++++++++++++++++---- 4 files changed, 44 insertions(+), 28 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 1b8e305bb54a..f476fd030626 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -1046,6 +1046,18 @@ struct nft_rule_dp { __attribute__((aligned(__alignof__(struct nft_expr)))); }; +struct nft_rule_dp_last { + struct nft_rule_dp end; /* end of nft_rule_blob marker */ + struct rcu_head h; /* call_rcu head */ + struct nft_rule_blob *blob; /* ptr to free via call_rcu */ + const struct nft_chain *chain; /* for nftables tracing */ +}; + +static inline const struct nft_rule_dp *nft_rule_next(const struct nft_rule_dp *rule) +{ + return (void *)rule + sizeof(*rule) + rule->dlen; +} + struct nft_rule_blob { unsigned long size; unsigned char data[] @@ -1396,7 +1408,6 @@ void nft_unregister_flowtable_type(struct nf_flowtable_type *type); * @packet_dumped: packet headers sent in a previous traceinfo message * @pkt: pktinfo currently processed * @basechain: base chain currently processed - * @chain: chain currently processed * @rule: rule that was evaluated * @verdict: verdict given by rule */ @@ -1408,7 +1419,6 @@ struct nft_traceinfo { u32 skbid; const struct nft_pktinfo *pkt; const struct nft_base_chain *basechain; - const struct nft_chain *chain; const struct nft_rule_dp *rule; const struct nft_verdict *verdict; }; diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 79848a27e640..0e1c86bb51a2 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -2110,13 +2110,6 @@ static void nft_chain_release_hook(struct nft_chain_hook *hook) module_put(hook->type->owner); } -struct nft_rule_dp_last { - struct nft_rule_dp end; /* end of nft_rule_blob marker */ - struct rcu_head h; - struct nft_rule_blob *blob; - const struct nft_chain *chain; /* for tracing */ -}; - static void nft_last_rule(const struct nft_chain *chain, const void *ptr) { struct nft_rule_dp_last *lrule; diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index ec3bab751092..89c05b64c2a2 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -42,13 +42,11 @@ static inline void nf_skip_indirect_calls_enable(void) { } #endif static noinline void __nft_trace_packet(struct nft_traceinfo *info, - const struct nft_chain *chain, enum nft_trace_types type) { if (!info->trace || !info->nf_trace) return; - info->chain = chain; info->type = type; nft_trace_notify(info); @@ -56,14 +54,13 @@ static noinline void __nft_trace_packet(struct nft_traceinfo *info, static inline void nft_trace_packet(const struct nft_pktinfo *pkt, struct nft_traceinfo *info, - const struct nft_chain *chain, const struct nft_rule_dp *rule, enum nft_trace_types type) { if (static_branch_unlikely(&nft_trace_enabled)) { info->nf_trace = pkt->skb->nf_trace; info->rule = rule; - __nft_trace_packet(info, chain, type); + __nft_trace_packet(info, type); } } @@ -111,7 +108,6 @@ static void nft_cmp16_fast_eval(const struct nft_expr *expr, } static noinline void __nft_trace_verdict(struct nft_traceinfo *info, - const struct nft_chain *chain, const struct nft_regs *regs) { enum nft_trace_types type; @@ -133,17 +129,16 @@ static noinline void __nft_trace_verdict(struct nft_traceinfo *info, break; } - __nft_trace_packet(info, chain, type); + __nft_trace_packet(info, type); } static inline void nft_trace_verdict(struct nft_traceinfo *info, - const struct nft_chain *chain, const struct nft_rule_dp *rule, const struct nft_regs *regs) { if (static_branch_unlikely(&nft_trace_enabled)) { info->rule = rule; - __nft_trace_verdict(info, chain, regs); + __nft_trace_verdict(info, regs); } } @@ -203,7 +198,6 @@ static noinline void nft_update_chain_stats(const struct nft_chain *chain, } struct nft_jumpstack { - const struct nft_chain *chain; const struct nft_rule_dp *rule; }; @@ -247,7 +241,6 @@ indirect_call: #define nft_rule_expr_first(rule) (struct nft_expr *)&rule->data[0] #define nft_rule_expr_next(expr) ((void *)expr) + expr->ops->size #define nft_rule_expr_last(rule) (struct nft_expr *)&rule->data[rule->dlen] -#define nft_rule_next(rule) (void *)rule + sizeof(*rule) + rule->dlen #define nft_rule_dp_for_each_expr(expr, last, rule) \ for ((expr) = nft_rule_expr_first(rule), (last) = nft_rule_expr_last(rule); \ @@ -302,14 +295,14 @@ next_rule: nft_trace_copy_nftrace(pkt, &info); continue; case NFT_CONTINUE: - nft_trace_packet(pkt, &info, chain, rule, + nft_trace_packet(pkt, &info, rule, NFT_TRACETYPE_RULE); continue; } break; } - nft_trace_verdict(&info, chain, rule, ®s); + nft_trace_verdict(&info, rule, ®s); switch (regs.verdict.code & NF_VERDICT_MASK) { case NF_ACCEPT: @@ -323,7 +316,6 @@ next_rule: case NFT_JUMP: if (WARN_ON_ONCE(stackptr >= NFT_JUMP_STACK_SIZE)) return NF_DROP; - jumpstack[stackptr].chain = chain; jumpstack[stackptr].rule = nft_rule_next(rule); stackptr++; fallthrough; @@ -339,12 +331,11 @@ next_rule: if (stackptr > 0) { stackptr--; - chain = jumpstack[stackptr].chain; rule = jumpstack[stackptr].rule; goto next_rule; } - nft_trace_packet(pkt, &info, basechain, NULL, NFT_TRACETYPE_POLICY); + nft_trace_packet(pkt, &info, NULL, NFT_TRACETYPE_POLICY); if (static_branch_unlikely(&nft_counters_enabled)) nft_update_chain_stats(basechain, pkt); diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c index 1163ba9c1401..3d9b83d84a84 100644 --- a/net/netfilter/nf_tables_trace.c +++ b/net/netfilter/nf_tables_trace.c @@ -164,9 +164,29 @@ static bool nft_trace_have_verdict_chain(struct nft_traceinfo *info) return true; } +static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *info) +{ + const struct nft_rule_dp *rule = info->rule; + const struct nft_rule_dp_last *last; + + if (!rule) + return &info->basechain->chain; + + while (!rule->is_last) + rule = nft_rule_next(rule); + + last = (const struct nft_rule_dp_last *)rule; + + if (WARN_ON_ONCE(!last->chain)) + return &info->basechain->chain; + + return last->chain; +} + void nft_trace_notify(struct nft_traceinfo *info) { const struct nft_pktinfo *pkt = info->pkt; + const struct nft_chain *chain; struct nlmsghdr *nlh; struct sk_buff *skb; unsigned int size; @@ -176,9 +196,11 @@ void nft_trace_notify(struct nft_traceinfo *info) if (!nfnetlink_has_listeners(nft_net(pkt), NFNLGRP_NFTRACE)) return; + chain = nft_trace_get_chain(info); + size = nlmsg_total_size(sizeof(struct nfgenmsg)) + - nla_total_size(strlen(info->chain->table->name)) + - nla_total_size(strlen(info->chain->name)) + + nla_total_size(strlen(chain->table->name)) + + nla_total_size(strlen(chain->name)) + nla_total_size_64bit(sizeof(__be64)) + /* rule handle */ nla_total_size(sizeof(__be32)) + /* trace type */ nla_total_size(0) + /* VERDICT, nested */ @@ -217,10 +239,10 @@ void nft_trace_notify(struct nft_traceinfo *info) if (nla_put_u32(skb, NFTA_TRACE_ID, info->skbid)) goto nla_put_failure; - if (nla_put_string(skb, NFTA_TRACE_CHAIN, info->chain->name)) + if (nla_put_string(skb, NFTA_TRACE_CHAIN, chain->name)) goto nla_put_failure; - if (nla_put_string(skb, NFTA_TRACE_TABLE, info->chain->table->name)) + if (nla_put_string(skb, NFTA_TRACE_TABLE, chain->table->name)) goto nla_put_failure; if (nf_trace_fill_rule_info(skb, info)) -- cgit v1.2.3 From 9a32e9850686599ed194ccdceb6cd3dd56b2d9b9 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 13 Apr 2023 17:13:19 +0200 Subject: netfilter: nf_tables: don't write table validation state without mutex The ->cleanup callback needs to be removed, this doesn't work anymore as the transaction mutex is already released in the ->abort function. Just do it after a successful validation pass, this either happens from commit or abort phases where transaction mutex is held. Fixes: f102d66b335a ("netfilter: nf_tables: use dedicated mutex to guard transactions") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nfnetlink.h | 1 - net/netfilter/nf_tables_api.c | 8 ++------ net/netfilter/nfnetlink.c | 2 -- 3 files changed, 2 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index 241e005f290a..e9a9ab34a7cc 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -45,7 +45,6 @@ struct nfnetlink_subsystem { int (*commit)(struct net *net, struct sk_buff *skb); int (*abort)(struct net *net, struct sk_buff *skb, enum nfnl_abort_action action); - void (*cleanup)(struct net *net); bool (*valid_genid)(struct net *net, u32 genid); }; diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 0e1c86bb51a2..21eb273d2740 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -8639,6 +8639,8 @@ static int nf_tables_validate(struct net *net) if (nft_table_validate(net, table) < 0) return -EAGAIN; } + + nft_validate_state_update(net, NFT_VALIDATE_SKIP); break; } @@ -9578,11 +9580,6 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action) return 0; } -static void nf_tables_cleanup(struct net *net) -{ - nft_validate_state_update(net, NFT_VALIDATE_SKIP); -} - static int nf_tables_abort(struct net *net, struct sk_buff *skb, enum nfnl_abort_action action) { @@ -9616,7 +9613,6 @@ static const struct nfnetlink_subsystem nf_tables_subsys = { .cb = nf_tables_cb, .commit = nf_tables_commit, .abort = nf_tables_abort, - .cleanup = nf_tables_cleanup, .valid_genid = nf_tables_valid_genid, .owner = THIS_MODULE, }; diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 81c7737c803a..ae7146475d17 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -590,8 +590,6 @@ done: goto replay_abort; } } - if (ss->cleanup) - ss->cleanup(net); nfnl_err_deliver(&err_list, oskb); kfree_skb(skb); -- cgit v1.2.3 From 00c320f9b75560628e840bef027a27c746706759 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 13 Apr 2023 17:13:20 +0200 Subject: netfilter: nf_tables: make validation state per table We only need to validate tables that saw changes in the current transaction. The existing code revalidates all tables, but this isn't needed as cross-table jumps are not allowed (chains have table scope). Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 3 ++- net/netfilter/nf_tables_api.c | 38 ++++++++++++++++++-------------------- 2 files changed, 20 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index f476fd030626..ec347d9cff9e 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -1209,6 +1209,7 @@ unsigned int nft_do_chain(struct nft_pktinfo *pkt, void *priv); * @genmask: generation mask * @afinfo: address family info * @name: name of the table + * @validate_state: internal, set when transaction adds jumps */ struct nft_table { struct list_head list; @@ -1227,6 +1228,7 @@ struct nft_table { char *name; u16 udlen; u8 *udata; + u8 validate_state; }; static inline bool nft_table_has_owner(const struct nft_table *table) @@ -1698,7 +1700,6 @@ struct nftables_pernet { struct mutex commit_mutex; u64 table_handle; unsigned int base_seq; - u8 validate_state; }; extern unsigned int nf_tables_net_id; diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 21eb273d2740..44ebc5f9598e 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -102,11 +102,9 @@ static const u8 nft2audit_op[NFT_MSG_MAX] = { // enum nf_tables_msg_types [NFT_MSG_DELFLOWTABLE] = AUDIT_NFT_OP_FLOWTABLE_UNREGISTER, }; -static void nft_validate_state_update(struct net *net, u8 new_validate_state) +static void nft_validate_state_update(struct nft_table *table, u8 new_validate_state) { - struct nftables_pernet *nft_net = nft_pernet(net); - - switch (nft_net->validate_state) { + switch (table->validate_state) { case NFT_VALIDATE_SKIP: WARN_ON_ONCE(new_validate_state == NFT_VALIDATE_DO); break; @@ -117,7 +115,7 @@ static void nft_validate_state_update(struct net *net, u8 new_validate_state) return; } - nft_net->validate_state = new_validate_state; + table->validate_state = new_validate_state; } static void nf_tables_trans_destroy_work(struct work_struct *w); static DECLARE_WORK(trans_destroy_work, nf_tables_trans_destroy_work); @@ -1224,6 +1222,7 @@ static int nf_tables_newtable(struct sk_buff *skb, const struct nfnl_info *info, if (table == NULL) goto err_kzalloc; + table->validate_state = NFT_VALIDATE_SKIP; table->name = nla_strdup(attr, GFP_KERNEL_ACCOUNT); if (table->name == NULL) goto err_strdup; @@ -3660,7 +3659,7 @@ static int nf_tables_newrule(struct sk_buff *skb, const struct nfnl_info *info, } if (expr_info[i].ops->validate) - nft_validate_state_update(net, NFT_VALIDATE_NEED); + nft_validate_state_update(table, NFT_VALIDATE_NEED); expr_info[i].ops = NULL; expr = nft_expr_next(expr); @@ -3710,7 +3709,7 @@ static int nf_tables_newrule(struct sk_buff *skb, const struct nfnl_info *info, if (flow) nft_trans_flow_rule(trans) = flow; - if (nft_net->validate_state == NFT_VALIDATE_DO) + if (table->validate_state == NFT_VALIDATE_DO) return nft_table_validate(net, table); return 0; @@ -6312,7 +6311,7 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set, if (desc.type == NFT_DATA_VERDICT && (elem.data.val.verdict.code == NFT_GOTO || elem.data.val.verdict.code == NFT_JUMP)) - nft_validate_state_update(ctx->net, + nft_validate_state_update(ctx->table, NFT_VALIDATE_NEED); } @@ -6437,7 +6436,6 @@ static int nf_tables_newsetelem(struct sk_buff *skb, const struct nfnl_info *info, const struct nlattr * const nla[]) { - struct nftables_pernet *nft_net = nft_pernet(info->net); struct netlink_ext_ack *extack = info->extack; u8 genmask = nft_genmask_next(info->net); u8 family = info->nfmsg->nfgen_family; @@ -6476,7 +6474,7 @@ static int nf_tables_newsetelem(struct sk_buff *skb, } } - if (nft_net->validate_state == NFT_VALIDATE_DO) + if (table->validate_state == NFT_VALIDATE_DO) return nft_table_validate(net, table); return 0; @@ -8628,19 +8626,20 @@ static int nf_tables_validate(struct net *net) struct nftables_pernet *nft_net = nft_pernet(net); struct nft_table *table; - switch (nft_net->validate_state) { - case NFT_VALIDATE_SKIP: - break; - case NFT_VALIDATE_NEED: - nft_validate_state_update(net, NFT_VALIDATE_DO); - fallthrough; - case NFT_VALIDATE_DO: - list_for_each_entry(table, &nft_net->tables, list) { + list_for_each_entry(table, &nft_net->tables, list) { + switch (table->validate_state) { + case NFT_VALIDATE_SKIP: + continue; + case NFT_VALIDATE_NEED: + nft_validate_state_update(table, NFT_VALIDATE_DO); + fallthrough; + case NFT_VALIDATE_DO: if (nft_table_validate(net, table) < 0) return -EAGAIN; + + nft_validate_state_update(table, NFT_VALIDATE_SKIP); } - nft_validate_state_update(net, NFT_VALIDATE_SKIP); break; } @@ -10355,7 +10354,6 @@ static int __net_init nf_tables_init_net(struct net *net) INIT_LIST_HEAD(&nft_net->notify_list); mutex_init(&nft_net->commit_mutex); nft_net->base_seq = 1; - nft_net->validate_state = NFT_VALIDATE_SKIP; return 0; } -- cgit v1.2.3 From 2a1d6abd7ebeafc45f2fb5fabfd53b8b3beaa170 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 14 Apr 2023 15:01:31 +0200 Subject: netfilter: nf_tables: remove unneeded conditional This helper is inlined, so keep it as small as possible. If the static key is true, there is only a very small chance that info->trace is false: 1. tracing was enabled at this very moment, the static key was updated to active right after nft_do_table was called. 2. tracing was disabled at this very moment. trace->info is already false, the static key is about to be patched to false soon. In both cases, no event will be sent because info->trace is false (checked in noinline slowpath). info->nf_trace is irrelevant. The nf_trace update is redunant in this case, but this will only happen for short duration, when static key flips. text data bss dec hex filename old: 2980 192 32 3204 c84 nf_tables_core.o new: 2964 192 32 3188 c74i nf_tables_core.o Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_core.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index 89c05b64c2a2..bed855638050 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -67,10 +67,8 @@ static inline void nft_trace_packet(const struct nft_pktinfo *pkt, static inline void nft_trace_copy_nftrace(const struct nft_pktinfo *pkt, struct nft_traceinfo *info) { - if (static_branch_unlikely(&nft_trace_enabled)) { - if (info->trace) - info->nf_trace = pkt->skb->nf_trace; - } + if (static_branch_unlikely(&nft_trace_enabled)) + info->nf_trace = pkt->skb->nf_trace; } static void nft_bitwise_fast_eval(const struct nft_expr *expr, -- cgit v1.2.3 From 698bb828a6c20c86e30b307175be1827c071ce23 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 14 Apr 2023 15:01:32 +0200 Subject: netfilter: nf_tables: do not store pktinfo in traceinfo structure pass it as argument. No change in object size. stack usage decreases by 8 byte: nf_tables_core.c:254 nft_do_chain 320 static Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 5 ++--- net/netfilter/nf_tables_core.c | 21 ++++++++++++--------- net/netfilter/nf_tables_trace.c | 5 ++--- 3 files changed, 16 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index ec347d9cff9e..cab351928cd2 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -1408,7 +1408,6 @@ void nft_unregister_flowtable_type(struct nf_flowtable_type *type); * @type: event type (enum nft_trace_types) * @skbid: hash of skb to be used as trace id * @packet_dumped: packet headers sent in a previous traceinfo message - * @pkt: pktinfo currently processed * @basechain: base chain currently processed * @rule: rule that was evaluated * @verdict: verdict given by rule @@ -1419,7 +1418,6 @@ struct nft_traceinfo { bool packet_dumped; enum nft_trace_types type:8; u32 skbid; - const struct nft_pktinfo *pkt; const struct nft_base_chain *basechain; const struct nft_rule_dp *rule; const struct nft_verdict *verdict; @@ -1429,7 +1427,8 @@ void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt, const struct nft_verdict *verdict, const struct nft_chain *basechain); -void nft_trace_notify(struct nft_traceinfo *info); +void nft_trace_notify(const struct nft_pktinfo *pkt, + struct nft_traceinfo *info); #define MODULE_ALIAS_NFT_CHAIN(family, name) \ MODULE_ALIAS("nft-chain-" __stringify(family) "-" name) diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index bed855638050..776eb2b9f632 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -41,7 +41,8 @@ static inline bool nf_skip_indirect_calls(void) { return false; } static inline void nf_skip_indirect_calls_enable(void) { } #endif -static noinline void __nft_trace_packet(struct nft_traceinfo *info, +static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt, + struct nft_traceinfo *info, enum nft_trace_types type) { if (!info->trace || !info->nf_trace) @@ -49,7 +50,7 @@ static noinline void __nft_trace_packet(struct nft_traceinfo *info, info->type = type; - nft_trace_notify(info); + nft_trace_notify(pkt, info); } static inline void nft_trace_packet(const struct nft_pktinfo *pkt, @@ -60,7 +61,7 @@ static inline void nft_trace_packet(const struct nft_pktinfo *pkt, if (static_branch_unlikely(&nft_trace_enabled)) { info->nf_trace = pkt->skb->nf_trace; info->rule = rule; - __nft_trace_packet(info, type); + __nft_trace_packet(pkt, info, type); } } @@ -105,7 +106,8 @@ static void nft_cmp16_fast_eval(const struct nft_expr *expr, regs->verdict.code = NFT_BREAK; } -static noinline void __nft_trace_verdict(struct nft_traceinfo *info, +static noinline void __nft_trace_verdict(const struct nft_pktinfo *pkt, + struct nft_traceinfo *info, const struct nft_regs *regs) { enum nft_trace_types type; @@ -123,20 +125,21 @@ static noinline void __nft_trace_verdict(struct nft_traceinfo *info, type = NFT_TRACETYPE_RULE; if (info->trace) - info->nf_trace = info->pkt->skb->nf_trace; + info->nf_trace = pkt->skb->nf_trace; break; } - __nft_trace_packet(info, type); + __nft_trace_packet(pkt, info, type); } -static inline void nft_trace_verdict(struct nft_traceinfo *info, +static inline void nft_trace_verdict(const struct nft_pktinfo *pkt, + struct nft_traceinfo *info, const struct nft_rule_dp *rule, const struct nft_regs *regs) { if (static_branch_unlikely(&nft_trace_enabled)) { info->rule = rule; - __nft_trace_verdict(info, regs); + __nft_trace_verdict(pkt, info, regs); } } @@ -300,7 +303,7 @@ next_rule: break; } - nft_trace_verdict(&info, rule, ®s); + nft_trace_verdict(pkt, &info, rule, ®s); switch (regs.verdict.code & NF_VERDICT_MASK) { case NF_ACCEPT: diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c index 3d9b83d84a84..0a0dcf2587fd 100644 --- a/net/netfilter/nf_tables_trace.c +++ b/net/netfilter/nf_tables_trace.c @@ -183,9 +183,9 @@ static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *i return last->chain; } -void nft_trace_notify(struct nft_traceinfo *info) +void nft_trace_notify(const struct nft_pktinfo *pkt, + struct nft_traceinfo *info) { - const struct nft_pktinfo *pkt = info->pkt; const struct nft_chain *chain; struct nlmsghdr *nlh; struct sk_buff *skb; @@ -305,7 +305,6 @@ void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt, info->trace = true; info->nf_trace = pkt->skb->nf_trace; info->packet_dumped = false; - info->pkt = pkt; info->verdict = verdict; net_get_random_once(&trace_key, sizeof(trace_key)); -- cgit v1.2.3 From 0a202145d5f9277dd24474aca8245731d030f29e Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 14 Apr 2023 15:01:33 +0200 Subject: netfilter: nf_tables: do not store verdict in traceinfo structure Just pass it as argument to nft_trace_notify. Stack is reduced by 8 bytes: nf_tables_core.c:256 nft_do_chain 312 static Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 4 +--- net/netfilter/nf_tables_core.c | 14 ++++++++------ net/netfilter/nf_tables_trace.c | 21 +++++++++++---------- 3 files changed, 20 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index cab351928cd2..693469ecfa54 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -1410,7 +1410,6 @@ void nft_unregister_flowtable_type(struct nf_flowtable_type *type); * @packet_dumped: packet headers sent in a previous traceinfo message * @basechain: base chain currently processed * @rule: rule that was evaluated - * @verdict: verdict given by rule */ struct nft_traceinfo { bool trace; @@ -1420,14 +1419,13 @@ struct nft_traceinfo { u32 skbid; const struct nft_base_chain *basechain; const struct nft_rule_dp *rule; - const struct nft_verdict *verdict; }; void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt, - const struct nft_verdict *verdict, const struct nft_chain *basechain); void nft_trace_notify(const struct nft_pktinfo *pkt, + const struct nft_verdict *verdict, struct nft_traceinfo *info); #define MODULE_ALIAS_NFT_CHAIN(family, name) \ diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index 776eb2b9f632..6debe8b2623f 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -42,6 +42,7 @@ static inline void nf_skip_indirect_calls_enable(void) { } #endif static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt, + const struct nft_verdict *verdict, struct nft_traceinfo *info, enum nft_trace_types type) { @@ -50,10 +51,11 @@ static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt, info->type = type; - nft_trace_notify(pkt, info); + nft_trace_notify(pkt, verdict, info); } static inline void nft_trace_packet(const struct nft_pktinfo *pkt, + struct nft_verdict *verdict, struct nft_traceinfo *info, const struct nft_rule_dp *rule, enum nft_trace_types type) @@ -61,7 +63,7 @@ static inline void nft_trace_packet(const struct nft_pktinfo *pkt, if (static_branch_unlikely(&nft_trace_enabled)) { info->nf_trace = pkt->skb->nf_trace; info->rule = rule; - __nft_trace_packet(pkt, info, type); + __nft_trace_packet(pkt, verdict, info, type); } } @@ -129,7 +131,7 @@ static noinline void __nft_trace_verdict(const struct nft_pktinfo *pkt, break; } - __nft_trace_packet(pkt, info, type); + __nft_trace_packet(pkt, ®s->verdict, info, type); } static inline void nft_trace_verdict(const struct nft_pktinfo *pkt, @@ -264,7 +266,7 @@ nft_do_chain(struct nft_pktinfo *pkt, void *priv) info.trace = false; if (static_branch_unlikely(&nft_trace_enabled)) - nft_trace_init(&info, pkt, ®s.verdict, basechain); + nft_trace_init(&info, pkt, basechain); do_chain: if (genbit) blob = rcu_dereference(chain->blob_gen_1); @@ -296,7 +298,7 @@ next_rule: nft_trace_copy_nftrace(pkt, &info); continue; case NFT_CONTINUE: - nft_trace_packet(pkt, &info, rule, + nft_trace_packet(pkt, ®s.verdict, &info, rule, NFT_TRACETYPE_RULE); continue; } @@ -336,7 +338,7 @@ next_rule: goto next_rule; } - nft_trace_packet(pkt, &info, NULL, NFT_TRACETYPE_POLICY); + nft_trace_packet(pkt, ®s.verdict, &info, NULL, NFT_TRACETYPE_POLICY); if (static_branch_unlikely(&nft_counters_enabled)) nft_update_chain_stats(basechain, pkt); diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c index 0a0dcf2587fd..e635104a42be 100644 --- a/net/netfilter/nf_tables_trace.c +++ b/net/netfilter/nf_tables_trace.c @@ -124,6 +124,7 @@ static int nf_trace_fill_pkt_info(struct sk_buff *nlskb, } static int nf_trace_fill_rule_info(struct sk_buff *nlskb, + const struct nft_verdict *verdict, const struct nft_traceinfo *info) { if (!info->rule || info->rule->is_last) @@ -135,7 +136,7 @@ static int nf_trace_fill_rule_info(struct sk_buff *nlskb, * Since no rule matched, the ->rule pointer is invalid. */ if (info->type == NFT_TRACETYPE_RETURN && - info->verdict->code == NFT_CONTINUE) + verdict->code == NFT_CONTINUE) return 0; return nla_put_be64(nlskb, NFTA_TRACE_RULE_HANDLE, @@ -143,7 +144,8 @@ static int nf_trace_fill_rule_info(struct sk_buff *nlskb, NFTA_TRACE_PAD); } -static bool nft_trace_have_verdict_chain(struct nft_traceinfo *info) +static bool nft_trace_have_verdict_chain(const struct nft_verdict *verdict, + struct nft_traceinfo *info) { switch (info->type) { case NFT_TRACETYPE_RETURN: @@ -153,7 +155,7 @@ static bool nft_trace_have_verdict_chain(struct nft_traceinfo *info) return false; } - switch (info->verdict->code) { + switch (verdict->code) { case NFT_JUMP: case NFT_GOTO: break; @@ -184,6 +186,7 @@ static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *i } void nft_trace_notify(const struct nft_pktinfo *pkt, + const struct nft_verdict *verdict, struct nft_traceinfo *info) { const struct nft_chain *chain; @@ -217,8 +220,8 @@ void nft_trace_notify(const struct nft_pktinfo *pkt, nla_total_size(sizeof(u32)) + /* nfproto */ nla_total_size(sizeof(u32)); /* policy */ - if (nft_trace_have_verdict_chain(info)) - size += nla_total_size(strlen(info->verdict->chain->name)); /* jump target */ + if (nft_trace_have_verdict_chain(verdict, info)) + size += nla_total_size(strlen(verdict->chain->name)); /* jump target */ skb = nlmsg_new(size, GFP_ATOMIC); if (!skb) @@ -245,7 +248,7 @@ void nft_trace_notify(const struct nft_pktinfo *pkt, if (nla_put_string(skb, NFTA_TRACE_TABLE, chain->table->name)) goto nla_put_failure; - if (nf_trace_fill_rule_info(skb, info)) + if (nf_trace_fill_rule_info(skb, verdict, info)) goto nla_put_failure; switch (info->type) { @@ -254,11 +257,11 @@ void nft_trace_notify(const struct nft_pktinfo *pkt, break; case NFT_TRACETYPE_RETURN: case NFT_TRACETYPE_RULE: - if (nft_verdict_dump(skb, NFTA_TRACE_VERDICT, info->verdict)) + if (nft_verdict_dump(skb, NFTA_TRACE_VERDICT, verdict)) goto nla_put_failure; /* pkt->skb undefined iff NF_STOLEN, disable dump */ - if (info->verdict->code == NF_STOLEN) + if (verdict->code == NF_STOLEN) info->packet_dumped = true; else mark = pkt->skb->mark; @@ -295,7 +298,6 @@ void nft_trace_notify(const struct nft_pktinfo *pkt, } void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt, - const struct nft_verdict *verdict, const struct nft_chain *chain) { static siphash_key_t trace_key __read_mostly; @@ -305,7 +307,6 @@ void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt, info->trace = true; info->nf_trace = pkt->skb->nf_trace; info->packet_dumped = false; - info->verdict = verdict; net_get_random_once(&trace_key, sizeof(trace_key)); -- cgit v1.2.3 From 46df417544f4f7fa3438caada0dc0e534a694343 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 14 Apr 2023 15:01:34 +0200 Subject: netfilter: nf_tables: do not store rule in traceinfo structure pass it as argument instead. This reduces size of traceinfo to 16 bytes. Total stack usage: nf_tables_core.c:252 nft_do_chain 304 static While its possible to also pass basechain as argument, doing so increases nft_do_chaininfo function size. Unlike pktinfo/verdict/rule the basechain info isn't used in the expression evaluation path. gcc places it on the stack, which results in extra push/pop when it gets passed to the trace helpers as argument rather than as part of the traceinfo structure. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 3 +-- net/netfilter/nf_tables_core.c | 15 +++++++-------- net/netfilter/nf_tables_trace.c | 14 ++++++++------ 3 files changed, 16 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 693469ecfa54..58a4d217faaf 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -1409,7 +1409,6 @@ void nft_unregister_flowtable_type(struct nf_flowtable_type *type); * @skbid: hash of skb to be used as trace id * @packet_dumped: packet headers sent in a previous traceinfo message * @basechain: base chain currently processed - * @rule: rule that was evaluated */ struct nft_traceinfo { bool trace; @@ -1418,7 +1417,6 @@ struct nft_traceinfo { enum nft_trace_types type:8; u32 skbid; const struct nft_base_chain *basechain; - const struct nft_rule_dp *rule; }; void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt, @@ -1426,6 +1424,7 @@ void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt, void nft_trace_notify(const struct nft_pktinfo *pkt, const struct nft_verdict *verdict, + const struct nft_rule_dp *rule, struct nft_traceinfo *info); #define MODULE_ALIAS_NFT_CHAIN(family, name) \ diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index 6debe8b2623f..4d0ce12221f6 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -43,6 +43,7 @@ static inline void nf_skip_indirect_calls_enable(void) { } static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt, const struct nft_verdict *verdict, + const struct nft_rule_dp *rule, struct nft_traceinfo *info, enum nft_trace_types type) { @@ -51,7 +52,7 @@ static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt, info->type = type; - nft_trace_notify(pkt, verdict, info); + nft_trace_notify(pkt, verdict, rule, info); } static inline void nft_trace_packet(const struct nft_pktinfo *pkt, @@ -62,8 +63,7 @@ static inline void nft_trace_packet(const struct nft_pktinfo *pkt, { if (static_branch_unlikely(&nft_trace_enabled)) { info->nf_trace = pkt->skb->nf_trace; - info->rule = rule; - __nft_trace_packet(pkt, verdict, info, type); + __nft_trace_packet(pkt, verdict, rule, info, type); } } @@ -110,6 +110,7 @@ static void nft_cmp16_fast_eval(const struct nft_expr *expr, static noinline void __nft_trace_verdict(const struct nft_pktinfo *pkt, struct nft_traceinfo *info, + const struct nft_rule_dp *rule, const struct nft_regs *regs) { enum nft_trace_types type; @@ -131,7 +132,7 @@ static noinline void __nft_trace_verdict(const struct nft_pktinfo *pkt, break; } - __nft_trace_packet(pkt, ®s->verdict, info, type); + __nft_trace_packet(pkt, ®s->verdict, rule, info, type); } static inline void nft_trace_verdict(const struct nft_pktinfo *pkt, @@ -139,10 +140,8 @@ static inline void nft_trace_verdict(const struct nft_pktinfo *pkt, const struct nft_rule_dp *rule, const struct nft_regs *regs) { - if (static_branch_unlikely(&nft_trace_enabled)) { - info->rule = rule; - __nft_trace_verdict(pkt, info, regs); - } + if (static_branch_unlikely(&nft_trace_enabled)) + __nft_trace_verdict(pkt, info, rule, regs); } static bool nft_payload_fast_eval(const struct nft_expr *expr, diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c index e635104a42be..6d41c0bd3d78 100644 --- a/net/netfilter/nf_tables_trace.c +++ b/net/netfilter/nf_tables_trace.c @@ -125,9 +125,10 @@ static int nf_trace_fill_pkt_info(struct sk_buff *nlskb, static int nf_trace_fill_rule_info(struct sk_buff *nlskb, const struct nft_verdict *verdict, + const struct nft_rule_dp *rule, const struct nft_traceinfo *info) { - if (!info->rule || info->rule->is_last) + if (!rule || rule->is_last) return 0; /* a continue verdict with ->type == RETURN means that this is @@ -140,7 +141,7 @@ static int nf_trace_fill_rule_info(struct sk_buff *nlskb, return 0; return nla_put_be64(nlskb, NFTA_TRACE_RULE_HANDLE, - cpu_to_be64(info->rule->handle), + cpu_to_be64(rule->handle), NFTA_TRACE_PAD); } @@ -166,9 +167,9 @@ static bool nft_trace_have_verdict_chain(const struct nft_verdict *verdict, return true; } -static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *info) +static const struct nft_chain *nft_trace_get_chain(const struct nft_rule_dp *rule, + const struct nft_traceinfo *info) { - const struct nft_rule_dp *rule = info->rule; const struct nft_rule_dp_last *last; if (!rule) @@ -187,6 +188,7 @@ static const struct nft_chain *nft_trace_get_chain(const struct nft_traceinfo *i void nft_trace_notify(const struct nft_pktinfo *pkt, const struct nft_verdict *verdict, + const struct nft_rule_dp *rule, struct nft_traceinfo *info) { const struct nft_chain *chain; @@ -199,7 +201,7 @@ void nft_trace_notify(const struct nft_pktinfo *pkt, if (!nfnetlink_has_listeners(nft_net(pkt), NFNLGRP_NFTRACE)) return; - chain = nft_trace_get_chain(info); + chain = nft_trace_get_chain(rule, info); size = nlmsg_total_size(sizeof(struct nfgenmsg)) + nla_total_size(strlen(chain->table->name)) + @@ -248,7 +250,7 @@ void nft_trace_notify(const struct nft_pktinfo *pkt, if (nla_put_string(skb, NFTA_TRACE_TABLE, chain->table->name)) goto nla_put_failure; - if (nf_trace_fill_rule_info(skb, verdict, info)) + if (nf_trace_fill_rule_info(skb, verdict, rule, info)) goto nla_put_failure; switch (info->type) { -- cgit v1.2.3 From e3478c68f6704638d08f437cbc552ca5970c151a Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Mon, 17 Apr 2023 17:10:45 +0200 Subject: ipvs: Update width of source for ip_vs_sync_conn_options In ip_vs_sync_conn_v0() copy is made to struct ip_vs_sync_conn_options. That structure looks like this: struct ip_vs_sync_conn_options { struct ip_vs_seq in_seq; struct ip_vs_seq out_seq; }; The source of the copy is the in_seq field of struct ip_vs_conn. Whose type is struct ip_vs_seq. Thus we can see that the source - is not as wide as the amount of data copied, which is the width of struct ip_vs_sync_conn_option. The copy is safe because the next field in is another struct ip_vs_seq. Make use of struct_group() to annotate this. Flagged by gcc-13 as: In file included from ./include/linux/string.h:254, from ./include/linux/bitmap.h:11, from ./include/linux/cpumask.h:12, from ./arch/x86/include/asm/paravirt.h:17, from ./arch/x86/include/asm/cpuid.h:62, from ./arch/x86/include/asm/processor.h:19, from ./arch/x86/include/asm/timex.h:5, from ./include/linux/timex.h:67, from ./include/linux/time32.h:13, from ./include/linux/time.h:60, from ./include/linux/stat.h:19, from ./include/linux/module.h:13, from net/netfilter/ipvs/ip_vs_sync.c:38: In function 'fortify_memcpy_chk', inlined from 'ip_vs_sync_conn_v0' at net/netfilter/ipvs/ip_vs_sync.c:606:3: ./include/linux/fortify-string.h:529:25: error: call to '__read_overflow2_field' declared with attribute warning: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Werror=attribute-warning] 529 | __read_overflow2_field(q_size_field, size); | Compile tested only. Signed-off-by: Simon Horman Reviewed-by: Horatiu Vultur Signed-off-by: Pablo Neira Ayuso --- include/net/ip_vs.h | 6 ++++-- net/netfilter/ipvs/ip_vs_sync.c | 2 +- 2 files changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index 6d71a5ff52df..e20f1f92066d 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -630,8 +630,10 @@ struct ip_vs_conn { */ struct ip_vs_app *app; /* bound ip_vs_app object */ void *app_data; /* Application private data */ - struct ip_vs_seq in_seq; /* incoming seq. struct */ - struct ip_vs_seq out_seq; /* outgoing seq. struct */ + struct_group(sync_conn_opt, + struct ip_vs_seq in_seq; /* incoming seq. struct */ + struct ip_vs_seq out_seq; /* outgoing seq. struct */ + ); const struct ip_vs_pe *pe; char *pe_data; diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 4963fec815da..d4fe7bb4f853 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -603,7 +603,7 @@ static void ip_vs_sync_conn_v0(struct netns_ipvs *ipvs, struct ip_vs_conn *cp, if (cp->flags & IP_VS_CONN_F_SEQ_MASK) { struct ip_vs_sync_conn_options *opt = (struct ip_vs_sync_conn_options *)&s[1]; - memcpy(opt, &cp->in_seq, sizeof(*opt)); + memcpy(opt, &cp->sync_conn_opt, sizeof(*opt)); } m->nr_conns++; -- cgit v1.2.3 From 280654932e341bf2860503af49bd5cb8e2b78522 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Mon, 17 Apr 2023 17:10:46 +0200 Subject: ipvs: Consistently use array_size() in ip_vs_conn_init() Consistently use array_size() to calculate the size of ip_vs_conn_tab in bytes. Flagged by Coccinelle: WARNING: array_size is already used (line 1498) to compute the same size No functional change intended. Compile tested only. Signed-off-by: Simon Horman Reviewed-by: Horatiu Vultur Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipvs/ip_vs_conn.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 13534e02346c..928e64653837 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -1481,6 +1481,7 @@ void __net_exit ip_vs_conn_net_cleanup(struct netns_ipvs *ipvs) int __init ip_vs_conn_init(void) { + size_t tab_array_size; int idx; /* Compute size and mask */ @@ -1494,8 +1495,9 @@ int __init ip_vs_conn_init(void) /* * Allocate the connection hash table and initialize its list heads */ - ip_vs_conn_tab = vmalloc(array_size(ip_vs_conn_tab_size, - sizeof(*ip_vs_conn_tab))); + tab_array_size = array_size(ip_vs_conn_tab_size, + sizeof(*ip_vs_conn_tab)); + ip_vs_conn_tab = vmalloc(tab_array_size); if (!ip_vs_conn_tab) return -ENOMEM; @@ -1508,10 +1510,8 @@ int __init ip_vs_conn_init(void) return -ENOMEM; } - pr_info("Connection hash table configured " - "(size=%d, memory=%ldKbytes)\n", - ip_vs_conn_tab_size, - (long)(ip_vs_conn_tab_size*sizeof(*ip_vs_conn_tab))/1024); + pr_info("Connection hash table configured (size=%d, memory=%zdKbytes)\n", + ip_vs_conn_tab_size, tab_array_size / 1024); IP_VS_DBG(0, "Each connection entry needs %zd bytes at least\n", sizeof(struct ip_vs_conn)); -- cgit v1.2.3 From 210ffe4a74caead4d6790747d32b63c6152c70b7 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Mon, 17 Apr 2023 17:10:47 +0200 Subject: ipvs: Remove {Enter,Leave}Function Remove EnterFunction and LeaveFunction. These debugging macros seem well past their use-by date. And seem to have little value these days. Removing them allows some trivial cleanup of some exit paths for some functions. These are also included in this patch. There is likely scope for further cleanup of both debugging and unwind paths. But let's leave that for another day. Only intended to change debug output, and only when CONFIG_IP_VS_DEBUG is enabled. Compile tested only. Signed-off-by: Simon Horman Reviewed-by: Horatiu Vultur Signed-off-by: Pablo Neira Ayuso --- include/net/ip_vs.h | 20 ------------- net/netfilter/ipvs/ip_vs_core.c | 8 ------ net/netfilter/ipvs/ip_vs_ctl.c | 26 +---------------- net/netfilter/ipvs/ip_vs_sync.c | 5 ---- net/netfilter/ipvs/ip_vs_xmit.c | 62 ++++++----------------------------------- 5 files changed, 9 insertions(+), 112 deletions(-) (limited to 'net') diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index e20f1f92066d..a3adc246ee31 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -265,26 +265,6 @@ static inline const char *ip_vs_dbg_addr(int af, char *buf, size_t buf_len, pr_err(msg, ##__VA_ARGS__); \ } while (0) -#ifdef CONFIG_IP_VS_DEBUG -#define EnterFunction(level) \ - do { \ - if (level <= ip_vs_get_debug_level()) \ - printk(KERN_DEBUG \ - pr_fmt("Enter: %s, %s line %i\n"), \ - __func__, __FILE__, __LINE__); \ - } while (0) -#define LeaveFunction(level) \ - do { \ - if (level <= ip_vs_get_debug_level()) \ - printk(KERN_DEBUG \ - pr_fmt("Leave: %s, %s line %i\n"), \ - __func__, __FILE__, __LINE__); \ - } while (0) -#else -#define EnterFunction(level) do {} while (0) -#define LeaveFunction(level) do {} while (0) -#endif - /* The port number of FTP service (in network order). */ #define FTPPORT cpu_to_be16(21) #define FTPDATA cpu_to_be16(20) diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 2fcc26507d69..cb83ca506c5c 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1140,7 +1140,6 @@ struct ip_vs_conn *ip_vs_new_conn_out(struct ip_vs_service *svc, __be16 vport; unsigned int flags; - EnterFunction(12); vaddr = &svc->addr; vport = svc->port; daddr = &iph->saddr; @@ -1208,7 +1207,6 @@ struct ip_vs_conn *ip_vs_new_conn_out(struct ip_vs_service *svc, IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport), IP_VS_DBG_ADDR(cp->af, &cp->daddr), ntohs(cp->dport), cp->flags, refcount_read(&cp->refcnt)); - LeaveFunction(12); return cp; } @@ -1316,13 +1314,11 @@ after_nat: ip_vs_update_conntrack(skb, cp, 0); ip_vs_conn_put(cp); - LeaveFunction(11); return NF_ACCEPT; drop: ip_vs_conn_put(cp); kfree_skb(skb); - LeaveFunction(11); return NF_STOLEN; } @@ -1341,8 +1337,6 @@ ip_vs_out_hook(void *priv, struct sk_buff *skb, const struct nf_hook_state *stat int af = state->pf; struct sock *sk; - EnterFunction(11); - /* Already marked as IPVS request or reply? */ if (skb->ipvs_property) return NF_ACCEPT; @@ -2365,7 +2359,6 @@ static void __net_exit __ip_vs_dev_cleanup_batch(struct list_head *net_list) struct netns_ipvs *ipvs; struct net *net; - EnterFunction(2); list_for_each_entry(net, net_list, exit_list) { ipvs = net_ipvs(net); ip_vs_unregister_hooks(ipvs, AF_INET); @@ -2374,7 +2367,6 @@ static void __net_exit __ip_vs_dev_cleanup_batch(struct list_head *net_list) smp_wmb(); ip_vs_sync_net_cleanup(ipvs); } - LeaveFunction(2); } static struct pernet_operations ipvs_core_ops = { diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 2a5ed71c82c3..62606fb44d02 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -1061,8 +1061,6 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) unsigned int atype; int ret; - EnterFunction(2); - #ifdef CONFIG_IP_VS_IPV6 if (udest->af == AF_INET6) { atype = ipv6_addr_type(&udest->addr.in6); @@ -1111,7 +1109,6 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) spin_lock_init(&dest->dst_lock); __ip_vs_update_dest(svc, dest, udest, 1); - LeaveFunction(2); return 0; err_stats: @@ -1134,8 +1131,6 @@ ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) __be16 dport = udest->port; int ret; - EnterFunction(2); - if (udest->weight < 0) { pr_err("%s(): server weight less than zero\n", __func__); return -ERANGE; @@ -1183,7 +1178,7 @@ ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) ret = ip_vs_start_estimator(svc->ipvs, &dest->stats); if (ret < 0) - goto err; + return ret; __ip_vs_update_dest(svc, dest, udest, 1); } else { /* @@ -1192,9 +1187,6 @@ ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) ret = ip_vs_new_dest(svc, udest); } -err: - LeaveFunction(2); - return ret; } @@ -1209,8 +1201,6 @@ ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) union nf_inet_addr daddr; __be16 dport = udest->port; - EnterFunction(2); - if (udest->weight < 0) { pr_err("%s(): server weight less than zero\n", __func__); return -ERANGE; @@ -1242,7 +1232,6 @@ ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) } __ip_vs_update_dest(svc, dest, udest, 0); - LeaveFunction(2); return 0; } @@ -1317,8 +1306,6 @@ ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) struct ip_vs_dest *dest; __be16 dport = udest->port; - EnterFunction(2); - /* We use function that requires RCU lock */ rcu_read_lock(); dest = ip_vs_lookup_dest(svc, udest->af, &udest->addr, dport); @@ -1339,8 +1326,6 @@ ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) */ __ip_vs_del_dest(svc->ipvs, dest, false); - LeaveFunction(2); - return 0; } @@ -1746,7 +1731,6 @@ void ip_vs_service_nets_cleanup(struct list_head *net_list) struct netns_ipvs *ipvs; struct net *net; - EnterFunction(2); /* Check for "full" addressed entries */ mutex_lock(&__ip_vs_mutex); list_for_each_entry(net, net_list, exit_list) { @@ -1754,7 +1738,6 @@ void ip_vs_service_nets_cleanup(struct list_head *net_list) ip_vs_flush(ipvs, true); } mutex_unlock(&__ip_vs_mutex); - LeaveFunction(2); } /* Put all references for device (dst_cache) */ @@ -1792,7 +1775,6 @@ static int ip_vs_dst_event(struct notifier_block *this, unsigned long event, if (event != NETDEV_DOWN || !ipvs) return NOTIFY_DONE; IP_VS_DBG(3, "%s() dev=%s\n", __func__, dev->name); - EnterFunction(2); mutex_lock(&__ip_vs_mutex); for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { hlist_for_each_entry(svc, &ip_vs_svc_table[idx], s_list) { @@ -1821,7 +1803,6 @@ static int ip_vs_dst_event(struct notifier_block *this, unsigned long event, } spin_unlock_bh(&ipvs->dest_trash_lock); mutex_unlock(&__ip_vs_mutex); - LeaveFunction(2); return NOTIFY_DONE; } @@ -4537,8 +4518,6 @@ int __init ip_vs_control_init(void) int idx; int ret; - EnterFunction(2); - /* Initialize svc_table, ip_vs_svc_fwm_table */ for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { INIT_HLIST_HEAD(&ip_vs_svc_table[idx]); @@ -4551,15 +4530,12 @@ int __init ip_vs_control_init(void) if (ret < 0) return ret; - LeaveFunction(2); return 0; } void ip_vs_control_cleanup(void) { - EnterFunction(2); unregister_netdevice_notifier(&ip_vs_dst_notifier); /* relying on common rcu_barrier() in ip_vs_cleanup() */ - LeaveFunction(2); } diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index d4fe7bb4f853..264f2f87a437 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -1582,13 +1582,11 @@ ip_vs_send_async(struct socket *sock, const char *buffer, const size_t length) struct kvec iov; int len; - EnterFunction(7); iov.iov_base = (void *)buffer; iov.iov_len = length; len = kernel_sendmsg(sock, &msg, &iov, 1, (size_t)(length)); - LeaveFunction(7); return len; } @@ -1614,15 +1612,12 @@ ip_vs_receive(struct socket *sock, char *buffer, const size_t buflen) struct kvec iov = {buffer, buflen}; int len; - EnterFunction(7); - /* Receive a packet */ iov_iter_kvec(&msg.msg_iter, ITER_DEST, &iov, 1, buflen); len = sock_recvmsg(sock, &msg, MSG_DONTWAIT); if (len < 0) return len; - LeaveFunction(7); return len; } diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 99c349c0d968..feb1d7fcb09f 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -706,8 +706,6 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, { struct iphdr *iph = ip_hdr(skb); - EnterFunction(10); - if (__ip_vs_get_out_rt(cp->ipvs, cp->af, skb, NULL, iph->daddr, IP_VS_RT_MODE_NON_LOCAL, NULL, ipvsh) < 0) goto tx_error; @@ -719,12 +717,10 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0); - LeaveFunction(10); return NF_STOLEN; tx_error: kfree_skb(skb); - LeaveFunction(10); return NF_STOLEN; } @@ -735,8 +731,6 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, { struct ipv6hdr *iph = ipv6_hdr(skb); - EnterFunction(10); - if (__ip_vs_get_out_rt_v6(cp->ipvs, cp->af, skb, NULL, &iph->daddr, NULL, ipvsh, 0, IP_VS_RT_MODE_NON_LOCAL) < 0) @@ -747,12 +741,10 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0); - LeaveFunction(10); return NF_STOLEN; tx_error: kfree_skb(skb); - LeaveFunction(10); return NF_STOLEN; } #endif @@ -768,8 +760,6 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, struct rtable *rt; /* Route to the other host */ int local, rc, was_input; - EnterFunction(10); - /* check if it is a connection of no-client-port */ if (unlikely(cp->flags & IP_VS_CONN_F_NO_CPORT)) { __be16 _pt, *p; @@ -839,12 +829,10 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local); - LeaveFunction(10); return rc; tx_error: kfree_skb(skb); - LeaveFunction(10); return NF_STOLEN; } @@ -856,8 +844,6 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, struct rt6_info *rt; /* Route to the other host */ int local, rc; - EnterFunction(10); - /* check if it is a connection of no-client-port */ if (unlikely(cp->flags & IP_VS_CONN_F_NO_CPORT && !ipvsh->fragoffs)) { __be16 _pt, *p; @@ -927,11 +913,9 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local); - LeaveFunction(10); return rc; tx_error: - LeaveFunction(10); kfree_skb(skb); return NF_STOLEN; } @@ -1149,8 +1133,6 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, int tun_type, gso_type; int tun_flags; - EnterFunction(10); - local = __ip_vs_get_out_rt(ipvs, cp->af, skb, cp->dest, cp->daddr.ip, IP_VS_RT_MODE_LOCAL | IP_VS_RT_MODE_NON_LOCAL | @@ -1199,7 +1181,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, &next_protocol, NULL, &dsfield, &ttl, dfp); if (IS_ERR(skb)) - goto tx_error; + return NF_STOLEN; gso_type = __tun_gso_type_mask(AF_INET, cp->af); if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) { @@ -1267,14 +1249,10 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, else if (ret == NF_DROP) kfree_skb(skb); - LeaveFunction(10); - return NF_STOLEN; tx_error: - if (!IS_ERR(skb)) - kfree_skb(skb); - LeaveFunction(10); + kfree_skb(skb); return NF_STOLEN; } @@ -1298,8 +1276,6 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, int tun_type, gso_type; int tun_flags; - EnterFunction(10); - local = __ip_vs_get_out_rt_v6(ipvs, cp->af, skb, cp->dest, &cp->daddr.in6, &saddr, ipvsh, 1, @@ -1347,7 +1323,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, &next_protocol, &payload_len, &dsfield, &ttl, NULL); if (IS_ERR(skb)) - goto tx_error; + return NF_STOLEN; gso_type = __tun_gso_type_mask(AF_INET6, cp->af); if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GUE) { @@ -1414,14 +1390,10 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, else if (ret == NF_DROP) kfree_skb(skb); - LeaveFunction(10); - return NF_STOLEN; tx_error: - if (!IS_ERR(skb)) - kfree_skb(skb); - LeaveFunction(10); + kfree_skb(skb); return NF_STOLEN; } #endif @@ -1437,8 +1409,6 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, { int local; - EnterFunction(10); - local = __ip_vs_get_out_rt(cp->ipvs, cp->af, skb, cp->dest, cp->daddr.ip, IP_VS_RT_MODE_LOCAL | IP_VS_RT_MODE_NON_LOCAL | @@ -1455,12 +1425,10 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0); - LeaveFunction(10); return NF_STOLEN; tx_error: kfree_skb(skb); - LeaveFunction(10); return NF_STOLEN; } @@ -1471,8 +1439,6 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, { int local; - EnterFunction(10); - local = __ip_vs_get_out_rt_v6(cp->ipvs, cp->af, skb, cp->dest, &cp->daddr.in6, NULL, ipvsh, 0, @@ -1489,12 +1455,10 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0); - LeaveFunction(10); return NF_STOLEN; tx_error: kfree_skb(skb); - LeaveFunction(10); return NF_STOLEN; } #endif @@ -1514,8 +1478,6 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, int local; int rt_mode, was_input; - EnterFunction(10); - /* The ICMP packet for VS/TUN, VS/DR and LOCALNODE will be forwarded directly here, because there is no need to translate address/port back */ @@ -1526,7 +1488,7 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, rc = NF_ACCEPT; /* do not touch skb anymore */ atomic_inc(&cp->in_pkts); - goto out; + return rc; } /* @@ -1582,14 +1544,11 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, /* Another hack: avoid icmp_send in ip_fragment */ skb->ignore_df = 1; - rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local); - goto out; + return ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local); tx_error: kfree_skb(skb); rc = NF_STOLEN; - out: - LeaveFunction(10); return rc; } @@ -1604,8 +1563,6 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, int local; int rt_mode; - EnterFunction(10); - /* The ICMP packet for VS/TUN, VS/DR and LOCALNODE will be forwarded directly here, because there is no need to translate address/port back */ @@ -1616,7 +1573,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, rc = NF_ACCEPT; /* do not touch skb anymore */ atomic_inc(&cp->in_pkts); - goto out; + return rc; } /* @@ -1671,14 +1628,11 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, /* Another hack: avoid icmp_send in ip_fragment */ skb->ignore_df = 1; - rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local); - goto out; + return ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local); tx_error: kfree_skb(skb); rc = NF_STOLEN; -out: - LeaveFunction(10); return rc; } #endif -- cgit v1.2.3 From c3c060adc0249355411a93e61888051e6902b8a1 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 21 Apr 2023 00:34:28 +0200 Subject: netfilter: nf_tables: extended netlink error reporting for netdevice Flowtable and netdev chains are bound to one or several netdevice, extend netlink error reporting to specify the the netdevice that triggers the error. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 38 ++++++++++++++++++++++++-------------- 1 file changed, 24 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 44ebc5f9598e..0fd4e28fbb60 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1954,7 +1954,8 @@ static struct nft_hook *nft_hook_list_find(struct list_head *hook_list, static int nf_tables_parse_netdev_hooks(struct net *net, const struct nlattr *attr, - struct list_head *hook_list) + struct list_head *hook_list, + struct netlink_ext_ack *extack) { struct nft_hook *hook, *next; const struct nlattr *tmp; @@ -1968,10 +1969,12 @@ static int nf_tables_parse_netdev_hooks(struct net *net, hook = nft_netdev_hook_alloc(net, tmp); if (IS_ERR(hook)) { + NL_SET_BAD_ATTR(extack, tmp); err = PTR_ERR(hook); goto err_hook; } if (nft_hook_list_find(hook_list, hook)) { + NL_SET_BAD_ATTR(extack, tmp); kfree(hook); err = -EEXIST; goto err_hook; @@ -2004,20 +2007,23 @@ struct nft_chain_hook { static int nft_chain_parse_netdev(struct net *net, struct nlattr *tb[], - struct list_head *hook_list) + struct list_head *hook_list, + struct netlink_ext_ack *extack) { struct nft_hook *hook; int err; if (tb[NFTA_HOOK_DEV]) { hook = nft_netdev_hook_alloc(net, tb[NFTA_HOOK_DEV]); - if (IS_ERR(hook)) + if (IS_ERR(hook)) { + NL_SET_BAD_ATTR(extack, tb[NFTA_HOOK_DEV]); return PTR_ERR(hook); + } list_add_tail(&hook->list, hook_list); } else if (tb[NFTA_HOOK_DEVS]) { err = nf_tables_parse_netdev_hooks(net, tb[NFTA_HOOK_DEVS], - hook_list); + hook_list, extack); if (err < 0) return err; @@ -2085,7 +2091,7 @@ static int nft_chain_parse_hook(struct net *net, INIT_LIST_HEAD(&hook->list); if (nft_base_chain_netdev(family, hook->num)) { - err = nft_chain_parse_netdev(net, ha, &hook->list); + err = nft_chain_parse_netdev(net, ha, &hook->list, extack); if (err < 0) { module_put(type->owner); return err; @@ -7560,7 +7566,8 @@ static const struct nla_policy nft_flowtable_hook_policy[NFTA_FLOWTABLE_HOOK_MAX static int nft_flowtable_parse_hook(const struct nft_ctx *ctx, const struct nlattr *attr, struct nft_flowtable_hook *flowtable_hook, - struct nft_flowtable *flowtable, bool add) + struct nft_flowtable *flowtable, + struct netlink_ext_ack *extack, bool add) { struct nlattr *tb[NFTA_FLOWTABLE_HOOK_MAX + 1]; struct nft_hook *hook; @@ -7607,7 +7614,8 @@ static int nft_flowtable_parse_hook(const struct nft_ctx *ctx, if (tb[NFTA_FLOWTABLE_HOOK_DEVS]) { err = nf_tables_parse_netdev_hooks(ctx->net, tb[NFTA_FLOWTABLE_HOOK_DEVS], - &flowtable_hook->list); + &flowtable_hook->list, + extack); if (err < 0) return err; } @@ -7750,7 +7758,8 @@ static void nft_flowtable_hooks_destroy(struct list_head *hook_list) } static int nft_flowtable_update(struct nft_ctx *ctx, const struct nlmsghdr *nlh, - struct nft_flowtable *flowtable) + struct nft_flowtable *flowtable, + struct netlink_ext_ack *extack) { const struct nlattr * const *nla = ctx->nla; struct nft_flowtable_hook flowtable_hook; @@ -7761,7 +7770,7 @@ static int nft_flowtable_update(struct nft_ctx *ctx, const struct nlmsghdr *nlh, int err; err = nft_flowtable_parse_hook(ctx, nla[NFTA_FLOWTABLE_HOOK], - &flowtable_hook, flowtable, false); + &flowtable_hook, flowtable, extack, false); if (err < 0) return err; @@ -7866,7 +7875,7 @@ static int nf_tables_newflowtable(struct sk_buff *skb, nft_ctx_init(&ctx, net, skb, info->nlh, family, table, NULL, nla); - return nft_flowtable_update(&ctx, info->nlh, flowtable); + return nft_flowtable_update(&ctx, info->nlh, flowtable, extack); } nft_ctx_init(&ctx, net, skb, info->nlh, family, table, NULL, nla); @@ -7907,7 +7916,7 @@ static int nf_tables_newflowtable(struct sk_buff *skb, goto err3; err = nft_flowtable_parse_hook(&ctx, nla[NFTA_FLOWTABLE_HOOK], - &flowtable_hook, flowtable, true); + &flowtable_hook, flowtable, extack, true); if (err < 0) goto err4; @@ -7959,7 +7968,8 @@ static void nft_flowtable_hook_release(struct nft_flowtable_hook *flowtable_hook } static int nft_delflowtable_hook(struct nft_ctx *ctx, - struct nft_flowtable *flowtable) + struct nft_flowtable *flowtable, + struct netlink_ext_ack *extack) { const struct nlattr * const *nla = ctx->nla; struct nft_flowtable_hook flowtable_hook; @@ -7969,7 +7979,7 @@ static int nft_delflowtable_hook(struct nft_ctx *ctx, int err; err = nft_flowtable_parse_hook(ctx, nla[NFTA_FLOWTABLE_HOOK], - &flowtable_hook, flowtable, false); + &flowtable_hook, flowtable, extack, false); if (err < 0) return err; @@ -8051,7 +8061,7 @@ static int nf_tables_delflowtable(struct sk_buff *skb, nft_ctx_init(&ctx, net, skb, info->nlh, family, table, NULL, nla); if (nla[NFTA_FLOWTABLE_HOOK]) - return nft_delflowtable_hook(&ctx, flowtable); + return nft_delflowtable_hook(&ctx, flowtable, extack); if (flowtable->use > 0) { NL_SET_BAD_ATTR(extack, attr); -- cgit v1.2.3 From 28339b21a365a8cb9edc6956cbfd679f569daa81 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 21 Apr 2023 00:34:29 +0200 Subject: netfilter: nf_tables: do not send complete notification of deletions In most cases, table, name and handle is sufficient for userspace to identify an object that has been deleted. Skipping unneeded fields in the netlink attributes in the message saves bandwidth (ie. less chances of hitting ENOBUFS). Rules are an exception: the existing userspace monitor code relies on the rule definition. This exception can be removed by implementing a rule cache in userspace, this is already supported by the tracing infrastructure. Regarding flowtables, incremental deletion of devices is possible. Skipping a full notification allows userspace to differentiate between flowtable removal and incremental removal of devices. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 70 +++++++++++++++++++++++++++++++------------ 1 file changed, 51 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 0fd4e28fbb60..3f95fb24a0bc 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -819,12 +819,20 @@ static int nf_tables_fill_table_info(struct sk_buff *skb, struct net *net, goto nla_put_failure; if (nla_put_string(skb, NFTA_TABLE_NAME, table->name) || - nla_put_be32(skb, NFTA_TABLE_FLAGS, - htonl(table->flags & NFT_TABLE_F_MASK)) || nla_put_be32(skb, NFTA_TABLE_USE, htonl(table->use)) || nla_put_be64(skb, NFTA_TABLE_HANDLE, cpu_to_be64(table->handle), NFTA_TABLE_PAD)) goto nla_put_failure; + + if (event == NFT_MSG_DELTABLE) { + nlmsg_end(skb, nlh); + return 0; + } + + if (nla_put_be32(skb, NFTA_TABLE_FLAGS, + htonl(table->flags & NFT_TABLE_F_MASK))) + goto nla_put_failure; + if (nft_table_has_owner(table) && nla_put_be32(skb, NFTA_TABLE_OWNER, htonl(table->nlpid))) goto nla_put_failure; @@ -1626,13 +1634,16 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net, if (!nlh) goto nla_put_failure; - if (nla_put_string(skb, NFTA_CHAIN_TABLE, table->name)) - goto nla_put_failure; - if (nla_put_be64(skb, NFTA_CHAIN_HANDLE, cpu_to_be64(chain->handle), + if (nla_put_string(skb, NFTA_CHAIN_TABLE, table->name) || + nla_put_string(skb, NFTA_CHAIN_NAME, chain->name) || + nla_put_be64(skb, NFTA_CHAIN_HANDLE, cpu_to_be64(chain->handle), NFTA_CHAIN_PAD)) goto nla_put_failure; - if (nla_put_string(skb, NFTA_CHAIN_NAME, chain->name)) - goto nla_put_failure; + + if (event == NFT_MSG_DELCHAIN) { + nlmsg_end(skb, nlh); + return 0; + } if (nft_is_base_chain(chain)) { const struct nft_base_chain *basechain = nft_base_chain(chain); @@ -4150,6 +4161,12 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, if (nla_put_be64(skb, NFTA_SET_HANDLE, cpu_to_be64(set->handle), NFTA_SET_PAD)) goto nla_put_failure; + + if (event == NFT_MSG_DELSET) { + nlmsg_end(skb, nlh); + return 0; + } + if (set->flags != 0) if (nla_put_be32(skb, NFTA_SET_FLAGS, htonl(set->flags))) goto nla_put_failure; @@ -7154,13 +7171,20 @@ static int nf_tables_fill_obj_info(struct sk_buff *skb, struct net *net, if (nla_put_string(skb, NFTA_OBJ_TABLE, table->name) || nla_put_string(skb, NFTA_OBJ_NAME, obj->key.name) || - nla_put_be32(skb, NFTA_OBJ_TYPE, htonl(obj->ops->type->type)) || - nla_put_be32(skb, NFTA_OBJ_USE, htonl(obj->use)) || - nft_object_dump(skb, NFTA_OBJ_DATA, obj, reset) || nla_put_be64(skb, NFTA_OBJ_HANDLE, cpu_to_be64(obj->handle), NFTA_OBJ_PAD)) goto nla_put_failure; + if (event == NFT_MSG_DELOBJ) { + nlmsg_end(skb, nlh); + return 0; + } + + if (nla_put_be32(skb, NFTA_OBJ_TYPE, htonl(obj->ops->type->type)) || + nla_put_be32(skb, NFTA_OBJ_USE, htonl(obj->use)) || + nft_object_dump(skb, NFTA_OBJ_DATA, obj, reset)) + goto nla_put_failure; + if (obj->udata && nla_put(skb, NFTA_OBJ_USERDATA, obj->udlen, obj->udata)) goto nla_put_failure; @@ -8089,9 +8113,16 @@ static int nf_tables_fill_flowtable_info(struct sk_buff *skb, struct net *net, if (nla_put_string(skb, NFTA_FLOWTABLE_TABLE, flowtable->table->name) || nla_put_string(skb, NFTA_FLOWTABLE_NAME, flowtable->name) || - nla_put_be32(skb, NFTA_FLOWTABLE_USE, htonl(flowtable->use)) || nla_put_be64(skb, NFTA_FLOWTABLE_HANDLE, cpu_to_be64(flowtable->handle), - NFTA_FLOWTABLE_PAD) || + NFTA_FLOWTABLE_PAD)) + goto nla_put_failure; + + if (event == NFT_MSG_DELFLOWTABLE && !hook_list) { + nlmsg_end(skb, nlh); + return 0; + } + + if (nla_put_be32(skb, NFTA_FLOWTABLE_USE, htonl(flowtable->use)) || nla_put_be32(skb, NFTA_FLOWTABLE_FLAGS, htonl(flowtable->data.flags))) goto nla_put_failure; @@ -8106,6 +8137,9 @@ static int nf_tables_fill_flowtable_info(struct sk_buff *skb, struct net *net, if (!nest_devs) goto nla_put_failure; + if (!hook_list) + hook_list = &flowtable->hook_list; + list_for_each_entry_rcu(hook, hook_list, list) { if (nla_put_string(skb, NFTA_DEVICE_NAME, hook->ops.dev->name)) goto nla_put_failure; @@ -8162,8 +8196,7 @@ static int nf_tables_dump_flowtable(struct sk_buff *skb, NFT_MSG_NEWFLOWTABLE, NLM_F_MULTI | NLM_F_APPEND, table->family, - flowtable, - &flowtable->hook_list) < 0) + flowtable, NULL) < 0) goto done; nl_dump_check_consistent(cb, nlmsg_hdr(skb)); @@ -8258,7 +8291,7 @@ static int nf_tables_getflowtable(struct sk_buff *skb, err = nf_tables_fill_flowtable_info(skb2, net, NETLINK_CB(skb).portid, info->nlh->nlmsg_seq, NFT_MSG_NEWFLOWTABLE, 0, family, - flowtable, &flowtable->hook_list); + flowtable, NULL); if (err < 0) goto err_fill_flowtable_info; @@ -8271,8 +8304,7 @@ err_fill_flowtable_info: static void nf_tables_flowtable_notify(struct nft_ctx *ctx, struct nft_flowtable *flowtable, - struct list_head *hook_list, - int event) + struct list_head *hook_list, int event) { struct nftables_pernet *nft_net = nft_pernet(ctx->net); struct sk_buff *skb; @@ -9333,7 +9365,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) nft_clear(net, nft_trans_flowtable(trans)); nf_tables_flowtable_notify(&trans->ctx, nft_trans_flowtable(trans), - &nft_trans_flowtable(trans)->hook_list, + NULL, NFT_MSG_NEWFLOWTABLE); } nft_trans_destroy(trans); @@ -9351,7 +9383,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) list_del_rcu(&nft_trans_flowtable(trans)->list); nf_tables_flowtable_notify(&trans->ctx, nft_trans_flowtable(trans), - &nft_trans_flowtable(trans)->hook_list, + NULL, trans->msg_type); nft_unregister_flowtable_net_hooks(net, &nft_trans_flowtable(trans)->hook_list); -- cgit v1.2.3 From cdc32546632354305afdcf399a5431138a31c9e0 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 21 Apr 2023 00:34:30 +0200 Subject: netfilter: nf_tables: rename function to destroy hook list Rename nft_flowtable_hooks_destroy() by nft_hooks_destroy() to prepare for netdev chain device updates. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 3f95fb24a0bc..77975b4b0fdc 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -7771,7 +7771,7 @@ err_unregister_net_hooks: return err; } -static void nft_flowtable_hooks_destroy(struct list_head *hook_list) +static void nft_hooks_destroy(struct list_head *hook_list) { struct nft_hook *hook, *next; @@ -7952,7 +7952,7 @@ static int nf_tables_newflowtable(struct sk_buff *skb, &flowtable->hook_list, flowtable); if (err < 0) { - nft_flowtable_hooks_destroy(&flowtable->hook_list); + nft_hooks_destroy(&flowtable->hook_list); goto err4; } @@ -8789,7 +8789,7 @@ static void nft_commit_release(struct nft_trans *trans) case NFT_MSG_DELFLOWTABLE: case NFT_MSG_DESTROYFLOWTABLE: if (nft_trans_flowtable_update(trans)) - nft_flowtable_hooks_destroy(&nft_trans_flowtable_hooks(trans)); + nft_hooks_destroy(&nft_trans_flowtable_hooks(trans)); else nf_tables_flowtable_destroy(nft_trans_flowtable(trans)); break; @@ -9440,7 +9440,7 @@ static void nf_tables_abort_release(struct nft_trans *trans) break; case NFT_MSG_NEWFLOWTABLE: if (nft_trans_flowtable_update(trans)) - nft_flowtable_hooks_destroy(&nft_trans_flowtable_hooks(trans)); + nft_hooks_destroy(&nft_trans_flowtable_hooks(trans)); else nf_tables_flowtable_destroy(nft_trans_flowtable(trans)); break; -- cgit v1.2.3 From b9703ed44ffbfba85c103b9de01886a225e14b38 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 21 Apr 2023 00:34:31 +0200 Subject: netfilter: nf_tables: support for adding new devices to an existing netdev chain This patch allows users to add devices to an existing netdev chain. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 6 ++ net/netfilter/nf_tables_api.c | 217 ++++++++++++++++++++++++-------------- 2 files changed, 142 insertions(+), 81 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 58a4d217faaf..3ed21d2d5659 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -1609,6 +1609,8 @@ struct nft_trans_chain { struct nft_stats __percpu *stats; u8 policy; u32 chain_id; + struct nft_base_chain *basechain; + struct list_head hook_list; }; #define nft_trans_chain_update(trans) \ @@ -1621,6 +1623,10 @@ struct nft_trans_chain { (((struct nft_trans_chain *)trans->data)->policy) #define nft_trans_chain_id(trans) \ (((struct nft_trans_chain *)trans->data)->chain_id) +#define nft_trans_basechain(trans) \ + (((struct nft_trans_chain *)trans->data)->basechain) +#define nft_trans_chain_hooks(trans) \ + (((struct nft_trans_chain *)trans->data)->hook_list) struct nft_trans_table { bool update; diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 77975b4b0fdc..f8d8cace0c7d 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1582,7 +1582,8 @@ nla_put_failure: } static int nft_dump_basechain_hook(struct sk_buff *skb, int family, - const struct nft_base_chain *basechain) + const struct nft_base_chain *basechain, + const struct list_head *hook_list) { const struct nf_hook_ops *ops = &basechain->ops; struct nft_hook *hook, *first = NULL; @@ -1599,7 +1600,11 @@ static int nft_dump_basechain_hook(struct sk_buff *skb, int family, if (nft_base_chain_netdev(family, ops->hooknum)) { nest_devs = nla_nest_start_noflag(skb, NFTA_HOOK_DEVS); - list_for_each_entry(hook, &basechain->hook_list, list) { + + if (!hook_list) + hook_list = &basechain->hook_list; + + list_for_each_entry(hook, hook_list, list) { if (!first) first = hook; @@ -1624,7 +1629,8 @@ nla_put_failure: static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net, u32 portid, u32 seq, int event, u32 flags, int family, const struct nft_table *table, - const struct nft_chain *chain) + const struct nft_chain *chain, + const struct list_head *hook_list) { struct nlmsghdr *nlh; @@ -1649,7 +1655,7 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net, const struct nft_base_chain *basechain = nft_base_chain(chain); struct nft_stats __percpu *stats; - if (nft_dump_basechain_hook(skb, family, basechain)) + if (nft_dump_basechain_hook(skb, family, basechain, hook_list)) goto nla_put_failure; if (nla_put_be32(skb, NFTA_CHAIN_POLICY, @@ -1684,7 +1690,8 @@ nla_put_failure: return -1; } -static void nf_tables_chain_notify(const struct nft_ctx *ctx, int event) +static void nf_tables_chain_notify(const struct nft_ctx *ctx, int event, + const struct list_head *hook_list) { struct nftables_pernet *nft_net; struct sk_buff *skb; @@ -1704,7 +1711,7 @@ static void nf_tables_chain_notify(const struct nft_ctx *ctx, int event) err = nf_tables_fill_chain_info(skb, ctx->net, ctx->portid, ctx->seq, event, flags, ctx->family, ctx->table, - ctx->chain); + ctx->chain, hook_list); if (err < 0) { kfree_skb(skb); goto err; @@ -1750,7 +1757,7 @@ static int nf_tables_dump_chains(struct sk_buff *skb, NFT_MSG_NEWCHAIN, NLM_F_MULTI, table->family, table, - chain) < 0) + chain, NULL) < 0) goto done; nl_dump_check_consistent(cb, nlmsg_hdr(skb)); @@ -1804,7 +1811,7 @@ static int nf_tables_getchain(struct sk_buff *skb, const struct nfnl_info *info, err = nf_tables_fill_chain_info(skb2, net, NETLINK_CB(skb).portid, info->nlh->nlmsg_seq, NFT_MSG_NEWCHAIN, - 0, family, table, chain); + 0, family, table, chain, NULL); if (err < 0) goto err_fill_chain_info; @@ -2048,9 +2055,10 @@ static int nft_chain_parse_netdev(struct net *net, } static int nft_chain_parse_hook(struct net *net, + struct nft_base_chain *basechain, const struct nlattr * const nla[], struct nft_chain_hook *hook, u8 family, - struct netlink_ext_ack *extack, bool autoload) + struct netlink_ext_ack *extack) { struct nftables_pernet *nft_net = nft_pernet(net); struct nlattr *ha[NFTA_HOOK_MAX + 1]; @@ -2066,31 +2074,46 @@ static int nft_chain_parse_hook(struct net *net, if (err < 0) return err; - if (ha[NFTA_HOOK_HOOKNUM] == NULL || - ha[NFTA_HOOK_PRIORITY] == NULL) - return -EINVAL; + if (!basechain) { + if (!ha[NFTA_HOOK_HOOKNUM] || + !ha[NFTA_HOOK_PRIORITY]) + return -EINVAL; - hook->num = ntohl(nla_get_be32(ha[NFTA_HOOK_HOOKNUM])); - hook->priority = ntohl(nla_get_be32(ha[NFTA_HOOK_PRIORITY])); + hook->num = ntohl(nla_get_be32(ha[NFTA_HOOK_HOOKNUM])); + hook->priority = ntohl(nla_get_be32(ha[NFTA_HOOK_PRIORITY])); - type = __nft_chain_type_get(family, NFT_CHAIN_T_DEFAULT); - if (!type) - return -EOPNOTSUPP; + type = __nft_chain_type_get(family, NFT_CHAIN_T_DEFAULT); + if (!type) + return -EOPNOTSUPP; - if (nla[NFTA_CHAIN_TYPE]) { - type = nf_tables_chain_type_lookup(net, nla[NFTA_CHAIN_TYPE], - family, autoload); - if (IS_ERR(type)) { - NL_SET_BAD_ATTR(extack, nla[NFTA_CHAIN_TYPE]); - return PTR_ERR(type); + if (nla[NFTA_CHAIN_TYPE]) { + type = nf_tables_chain_type_lookup(net, nla[NFTA_CHAIN_TYPE], + family, true); + if (IS_ERR(type)) { + NL_SET_BAD_ATTR(extack, nla[NFTA_CHAIN_TYPE]); + return PTR_ERR(type); + } } - } - if (hook->num >= NFT_MAX_HOOKS || !(type->hook_mask & (1 << hook->num))) - return -EOPNOTSUPP; + if (hook->num >= NFT_MAX_HOOKS || !(type->hook_mask & (1 << hook->num))) + return -EOPNOTSUPP; - if (type->type == NFT_CHAIN_T_NAT && - hook->priority <= NF_IP_PRI_CONNTRACK) - return -EOPNOTSUPP; + if (type->type == NFT_CHAIN_T_NAT && + hook->priority <= NF_IP_PRI_CONNTRACK) + return -EOPNOTSUPP; + } else { + if (ha[NFTA_HOOK_HOOKNUM]) { + hook->num = ntohl(nla_get_be32(ha[NFTA_HOOK_HOOKNUM])); + if (hook->num != basechain->ops.hooknum) + return -EOPNOTSUPP; + } + if (ha[NFTA_HOOK_PRIORITY]) { + hook->priority = ntohl(nla_get_be32(ha[NFTA_HOOK_PRIORITY])); + if (hook->priority != basechain->ops.priority) + return -EOPNOTSUPP; + } + + type = basechain->type; + } if (!try_module_get(type->owner)) { if (nla[NFTA_CHAIN_TYPE]) @@ -2184,12 +2207,8 @@ static int nft_basechain_init(struct nft_base_chain *basechain, u8 family, list_splice_init(&hook->list, &basechain->hook_list); list_for_each_entry(h, &basechain->hook_list, list) nft_basechain_hook_init(&h->ops, family, hook, chain); - - basechain->ops.hooknum = hook->num; - basechain->ops.priority = hook->priority; - } else { - nft_basechain_hook_init(&basechain->ops, family, hook, chain); } + nft_basechain_hook_init(&basechain->ops, family, hook, chain); chain->flags |= NFT_CHAIN_BASE | flags; basechain->policy = NF_ACCEPT; @@ -2239,13 +2258,13 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask, if (nla[NFTA_CHAIN_HOOK]) { struct nft_stats __percpu *stats = NULL; - struct nft_chain_hook hook; + struct nft_chain_hook hook = {}; if (flags & NFT_CHAIN_BINDING) return -EOPNOTSUPP; - err = nft_chain_parse_hook(net, nla, &hook, family, extack, - true); + err = nft_chain_parse_hook(net, NULL, nla, &hook, family, + extack); if (err < 0) return err; @@ -2359,65 +2378,57 @@ err_destroy_chain: return err; } -static bool nft_hook_list_equal(struct list_head *hook_list1, - struct list_head *hook_list2) -{ - struct nft_hook *hook; - int n = 0, m = 0; - - n = 0; - list_for_each_entry(hook, hook_list2, list) { - if (!nft_hook_list_find(hook_list1, hook)) - return false; - - n++; - } - list_for_each_entry(hook, hook_list1, list) - m++; - - return n == m; -} - static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy, u32 flags, const struct nlattr *attr, struct netlink_ext_ack *extack) { const struct nlattr * const *nla = ctx->nla; + struct nft_base_chain *basechain = NULL; struct nft_table *table = ctx->table; struct nft_chain *chain = ctx->chain; - struct nft_base_chain *basechain; + struct nft_chain_hook hook = {}; struct nft_stats *stats = NULL; - struct nft_chain_hook hook; + struct nft_hook *h, *next; struct nf_hook_ops *ops; struct nft_trans *trans; + bool unregister = false; int err; if (chain->flags ^ flags) return -EOPNOTSUPP; + INIT_LIST_HEAD(&hook.list); + if (nla[NFTA_CHAIN_HOOK]) { if (!nft_is_base_chain(chain)) { NL_SET_BAD_ATTR(extack, attr); return -EEXIST; } - err = nft_chain_parse_hook(ctx->net, nla, &hook, ctx->family, - extack, false); + + basechain = nft_base_chain(chain); + err = nft_chain_parse_hook(ctx->net, basechain, nla, &hook, + ctx->family, extack); if (err < 0) return err; - basechain = nft_base_chain(chain); if (basechain->type != hook.type) { nft_chain_release_hook(&hook); NL_SET_BAD_ATTR(extack, attr); return -EEXIST; } - if (nft_base_chain_netdev(ctx->family, hook.num)) { - if (!nft_hook_list_equal(&basechain->hook_list, - &hook.list)) { - nft_chain_release_hook(&hook); - NL_SET_BAD_ATTR(extack, attr); - return -EEXIST; + if (nft_base_chain_netdev(ctx->family, basechain->ops.hooknum)) { + list_for_each_entry_safe(h, next, &hook.list, list) { + h->ops.pf = basechain->ops.pf; + h->ops.hooknum = basechain->ops.hooknum; + h->ops.priority = basechain->ops.priority; + h->ops.priv = basechain->ops.priv; + h->ops.hook = basechain->ops.hook; + + if (nft_hook_list_find(&basechain->hook_list, h)) { + list_del(&h->list); + kfree(h); + } } } else { ops = &basechain->ops; @@ -2428,7 +2439,6 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy, return -EEXIST; } } - nft_chain_release_hook(&hook); } if (nla[NFTA_CHAIN_HANDLE] && @@ -2439,24 +2449,43 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy, nla[NFTA_CHAIN_NAME], genmask); if (!IS_ERR(chain2)) { NL_SET_BAD_ATTR(extack, nla[NFTA_CHAIN_NAME]); - return -EEXIST; + err = -EEXIST; + goto err_hooks; } } if (nla[NFTA_CHAIN_COUNTERS]) { - if (!nft_is_base_chain(chain)) - return -EOPNOTSUPP; + if (!nft_is_base_chain(chain)) { + err = -EOPNOTSUPP; + goto err_hooks; + } stats = nft_stats_alloc(nla[NFTA_CHAIN_COUNTERS]); - if (IS_ERR(stats)) - return PTR_ERR(stats); + if (IS_ERR(stats)) { + err = PTR_ERR(stats); + goto err_hooks; + } } + if (!(table->flags & NFT_TABLE_F_DORMANT) && + nft_is_base_chain(chain) && + !list_empty(&hook.list)) { + basechain = nft_base_chain(chain); + ops = &basechain->ops; + + if (nft_base_chain_netdev(table->family, basechain->ops.hooknum)) { + err = nft_netdev_register_hooks(ctx->net, &hook.list); + if (err < 0) + goto err_hooks; + } + } + + unregister = true; err = -ENOMEM; trans = nft_trans_alloc(ctx, NFT_MSG_NEWCHAIN, sizeof(struct nft_trans_chain)); if (trans == NULL) - goto err; + goto err_trans; nft_trans_chain_stats(trans) = stats; nft_trans_chain_update(trans) = true; @@ -2475,7 +2504,7 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy, err = -ENOMEM; name = nla_strdup(nla[NFTA_CHAIN_NAME], GFP_KERNEL_ACCOUNT); if (!name) - goto err; + goto err_trans; err = -EEXIST; list_for_each_entry(tmp, &nft_net->commit_list, list) { @@ -2486,18 +2515,35 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy, strcmp(name, nft_trans_chain_name(tmp)) == 0) { NL_SET_BAD_ATTR(extack, nla[NFTA_CHAIN_NAME]); kfree(name); - goto err; + goto err_trans; } } nft_trans_chain_name(trans) = name; } + + nft_trans_basechain(trans) = basechain; + INIT_LIST_HEAD(&nft_trans_chain_hooks(trans)); + list_splice(&hook.list, &nft_trans_chain_hooks(trans)); + nft_trans_commit_list_add_tail(ctx->net, trans); return 0; -err: + +err_trans: free_percpu(stats); kfree(trans); +err_hooks: + if (nla[NFTA_CHAIN_HOOK]) { + list_for_each_entry_safe(h, next, &hook.list, list) { + if (unregister) + nf_unregister_net_hook(ctx->net, &h->ops); + list_del(&h->list); + kfree_rcu(h, rcu); + } + module_put(hook.type->owner); + } + return err; } @@ -9244,19 +9290,22 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) case NFT_MSG_NEWCHAIN: if (nft_trans_chain_update(trans)) { nft_chain_commit_update(trans); - nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN); + nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN, + &nft_trans_chain_hooks(trans)); + list_splice(&nft_trans_chain_hooks(trans), + &nft_trans_basechain(trans)->hook_list); /* trans destroyed after rcu grace period */ } else { nft_chain_commit_drop_policy(trans); nft_clear(net, trans->ctx.chain); - nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN); + nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN, NULL); nft_trans_destroy(trans); } break; case NFT_MSG_DELCHAIN: case NFT_MSG_DESTROYCHAIN: nft_chain_del(trans->ctx.chain); - nf_tables_chain_notify(&trans->ctx, trans->msg_type); + nf_tables_chain_notify(&trans->ctx, trans->msg_type, NULL); nf_tables_unregister_hook(trans->ctx.net, trans->ctx.table, trans->ctx.chain); @@ -9423,7 +9472,10 @@ static void nf_tables_abort_release(struct nft_trans *trans) nf_tables_table_destroy(&trans->ctx); break; case NFT_MSG_NEWCHAIN: - nf_tables_chain_destroy(&trans->ctx); + if (nft_trans_chain_update(trans)) + nft_hooks_destroy(&nft_trans_chain_hooks(trans)); + else + nf_tables_chain_destroy(&trans->ctx); break; case NFT_MSG_NEWRULE: nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans)); @@ -9486,6 +9538,9 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action) break; case NFT_MSG_NEWCHAIN: if (nft_trans_chain_update(trans)) { + nft_netdev_unregister_hooks(net, + &nft_trans_chain_hooks(trans), + true); free_percpu(nft_trans_chain_stats(trans)); kfree(nft_trans_chain_name(trans)); nft_trans_destroy(trans); -- cgit v1.2.3 From 7d937b107108f918906fc867ea070cc3d0cbcd32 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 21 Apr 2023 00:34:32 +0200 Subject: netfilter: nf_tables: support for deleting devices in an existing netdev chain This patch allows for deleting devices in an existing netdev chain. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 99 ++++++++++++++++++++++++++++++++++++++----- 1 file changed, 88 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index f8d8cace0c7d..c55ac3d57c51 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1646,7 +1646,7 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net, NFTA_CHAIN_PAD)) goto nla_put_failure; - if (event == NFT_MSG_DELCHAIN) { + if (event == NFT_MSG_DELCHAIN && !hook_list) { nlmsg_end(skb, nlh); return 0; } @@ -2667,6 +2667,59 @@ static int nf_tables_newchain(struct sk_buff *skb, const struct nfnl_info *info, return nf_tables_addchain(&ctx, family, genmask, policy, flags, extack); } +static int nft_delchain_hook(struct nft_ctx *ctx, struct nft_chain *chain, + struct netlink_ext_ack *extack) +{ + const struct nlattr * const *nla = ctx->nla; + struct nft_chain_hook chain_hook = {}; + struct nft_base_chain *basechain; + struct nft_hook *this, *hook; + LIST_HEAD(chain_del_list); + struct nft_trans *trans; + int err; + + if (!nft_is_base_chain(chain)) + return -EOPNOTSUPP; + + basechain = nft_base_chain(chain); + err = nft_chain_parse_hook(ctx->net, basechain, nla, &chain_hook, + ctx->family, extack); + if (err < 0) + return err; + + list_for_each_entry(this, &chain_hook.list, list) { + hook = nft_hook_list_find(&basechain->hook_list, this); + if (!hook) { + err = -ENOENT; + goto err_chain_del_hook; + } + list_move(&hook->list, &chain_del_list); + } + + trans = nft_trans_alloc(ctx, NFT_MSG_DELCHAIN, + sizeof(struct nft_trans_chain)); + if (!trans) { + err = -ENOMEM; + goto err_chain_del_hook; + } + + nft_trans_basechain(trans) = basechain; + nft_trans_chain_update(trans) = true; + INIT_LIST_HEAD(&nft_trans_chain_hooks(trans)); + list_splice(&chain_del_list, &nft_trans_chain_hooks(trans)); + nft_chain_release_hook(&chain_hook); + + nft_trans_commit_list_add_tail(ctx->net, trans); + + return 0; + +err_chain_del_hook: + list_splice(&chain_del_list, &basechain->hook_list); + nft_chain_release_hook(&chain_hook); + + return err; +} + static int nf_tables_delchain(struct sk_buff *skb, const struct nfnl_info *info, const struct nlattr * const nla[]) { @@ -2707,12 +2760,19 @@ static int nf_tables_delchain(struct sk_buff *skb, const struct nfnl_info *info, return PTR_ERR(chain); } + nft_ctx_init(&ctx, net, skb, info->nlh, family, table, chain, nla); + + if (nla[NFTA_CHAIN_HOOK]) { + if (chain->flags & NFT_CHAIN_HW_OFFLOAD) + return -EOPNOTSUPP; + + return nft_delchain_hook(&ctx, chain, extack); + } + if (info->nlh->nlmsg_flags & NLM_F_NONREC && chain->use > 0) return -EBUSY; - nft_ctx_init(&ctx, net, skb, info->nlh, family, table, chain, nla); - use = chain->use; list_for_each_entry(rule, &chain->rules, list) { if (!nft_is_active_next(net, rule)) @@ -8812,7 +8872,10 @@ static void nft_commit_release(struct nft_trans *trans) break; case NFT_MSG_DELCHAIN: case NFT_MSG_DESTROYCHAIN: - nf_tables_chain_destroy(&trans->ctx); + if (nft_trans_chain_update(trans)) + nft_hooks_destroy(&nft_trans_chain_hooks(trans)); + else + nf_tables_chain_destroy(&trans->ctx); break; case NFT_MSG_DELRULE: case NFT_MSG_DESTROYRULE: @@ -9304,11 +9367,20 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) break; case NFT_MSG_DELCHAIN: case NFT_MSG_DESTROYCHAIN: - nft_chain_del(trans->ctx.chain); - nf_tables_chain_notify(&trans->ctx, trans->msg_type, NULL); - nf_tables_unregister_hook(trans->ctx.net, - trans->ctx.table, - trans->ctx.chain); + if (nft_trans_chain_update(trans)) { + nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN, + &nft_trans_chain_hooks(trans)); + nft_netdev_unregister_hooks(net, + &nft_trans_chain_hooks(trans), + true); + } else { + nft_chain_del(trans->ctx.chain); + nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN, + NULL); + nf_tables_unregister_hook(trans->ctx.net, + trans->ctx.table, + trans->ctx.chain); + } break; case NFT_MSG_NEWRULE: nft_clear(trans->ctx.net, nft_trans_rule(trans)); @@ -9558,8 +9630,13 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action) break; case NFT_MSG_DELCHAIN: case NFT_MSG_DESTROYCHAIN: - trans->ctx.table->use++; - nft_clear(trans->ctx.net, trans->ctx.chain); + if (nft_trans_chain_update(trans)) { + list_splice(&nft_trans_chain_hooks(trans), + &nft_trans_basechain(trans)->hook_list); + } else { + trans->ctx.table->use++; + nft_clear(trans->ctx.net, trans->ctx.chain); + } nft_trans_destroy(trans); break; case NFT_MSG_NEWRULE: -- cgit v1.2.3 From 207296f1a03bfead0110ffc4f192f242100ce4ff Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 21 Apr 2023 00:34:33 +0200 Subject: netfilter: nf_tables: allow to create netdev chain without device Relax netdev chain creation to allow for loading the ruleset, then adding/deleting devices at a later stage. Hardware offload does not support for this feature yet. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index c55ac3d57c51..09542951656c 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -2023,10 +2023,9 @@ struct nft_chain_hook { struct list_head list; }; -static int nft_chain_parse_netdev(struct net *net, - struct nlattr *tb[], +static int nft_chain_parse_netdev(struct net *net, struct nlattr *tb[], struct list_head *hook_list, - struct netlink_ext_ack *extack) + struct netlink_ext_ack *extack, u32 flags) { struct nft_hook *hook; int err; @@ -2045,12 +2044,12 @@ static int nft_chain_parse_netdev(struct net *net, if (err < 0) return err; - if (list_empty(hook_list)) - return -EINVAL; - } else { - return -EINVAL; } + if (flags & NFT_CHAIN_HW_OFFLOAD && + list_empty(hook_list)) + return -EINVAL; + return 0; } @@ -2058,7 +2057,7 @@ static int nft_chain_parse_hook(struct net *net, struct nft_base_chain *basechain, const struct nlattr * const nla[], struct nft_chain_hook *hook, u8 family, - struct netlink_ext_ack *extack) + u32 flags, struct netlink_ext_ack *extack) { struct nftables_pernet *nft_net = nft_pernet(net); struct nlattr *ha[NFTA_HOOK_MAX + 1]; @@ -2125,7 +2124,7 @@ static int nft_chain_parse_hook(struct net *net, INIT_LIST_HEAD(&hook->list); if (nft_base_chain_netdev(family, hook->num)) { - err = nft_chain_parse_netdev(net, ha, &hook->list, extack); + err = nft_chain_parse_netdev(net, ha, &hook->list, extack, flags); if (err < 0) { module_put(type->owner); return err; @@ -2263,7 +2262,7 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask, if (flags & NFT_CHAIN_BINDING) return -EOPNOTSUPP; - err = nft_chain_parse_hook(net, NULL, nla, &hook, family, + err = nft_chain_parse_hook(net, NULL, nla, &hook, family, flags, extack); if (err < 0) return err; @@ -2407,7 +2406,7 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy, basechain = nft_base_chain(chain); err = nft_chain_parse_hook(ctx->net, basechain, nla, &hook, - ctx->family, extack); + ctx->family, flags, extack); if (err < 0) return err; @@ -2683,7 +2682,7 @@ static int nft_delchain_hook(struct nft_ctx *ctx, struct nft_chain *chain, basechain = nft_base_chain(chain); err = nft_chain_parse_hook(ctx->net, basechain, nla, &chain_hook, - ctx->family, extack); + ctx->family, chain->flags, extack); if (err < 0) return err; -- cgit v1.2.3 From 7041101ff6c3073fd8f2e99920f535b111c929cb Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Thu, 20 Apr 2023 16:59:46 +0200 Subject: net/sched: sch_fq: fix integer overflow of "credit" if sch_fq is configured with "initial quantum" having values greater than INT_MAX, the first assignment of "credit" does signed integer overflow to a very negative value. In this situation, the syzkaller script provided by Cristoph triggers the CPU soft-lockup warning even with few sockets. It's not an infinite loop, but "credit" wasn't probably meant to be minus 2Gb for each new flow. Capping "initial quantum" to INT_MAX proved to fix the issue. v2: validation of "initial quantum" is done in fq_policy, instead of open coding in fq_change() _ suggested by Jakub Kicinski Reported-by: Christoph Paasch Link: https://github.com/multipath-tcp/mptcp_net-next/issues/377 Fixes: afe4fd062416 ("pkt_sched: fq: Fair Queue packet scheduler") Reviewed-by: Eric Dumazet Signed-off-by: Davide Caratti Link: https://lore.kernel.org/r/7b3a3c7e36d03068707a021760a194a8eb5ad41a.1682002300.git.dcaratti@redhat.com Signed-off-by: Jakub Kicinski --- net/sched/sch_fq.c | 6 +++++- .../selftests/tc-testing/tc-tests/qdiscs/fq.json | 22 ++++++++++++++++++++++ 2 files changed, 27 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index 48d14fb90ba0..f59a2cb2c803 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -779,13 +779,17 @@ static int fq_resize(struct Qdisc *sch, u32 log) return 0; } +static struct netlink_range_validation iq_range = { + .max = INT_MAX, +}; + static const struct nla_policy fq_policy[TCA_FQ_MAX + 1] = { [TCA_FQ_UNSPEC] = { .strict_start_type = TCA_FQ_TIMER_SLACK }, [TCA_FQ_PLIMIT] = { .type = NLA_U32 }, [TCA_FQ_FLOW_PLIMIT] = { .type = NLA_U32 }, [TCA_FQ_QUANTUM] = { .type = NLA_U32 }, - [TCA_FQ_INITIAL_QUANTUM] = { .type = NLA_U32 }, + [TCA_FQ_INITIAL_QUANTUM] = NLA_POLICY_FULL_RANGE(NLA_U32, &iq_range), [TCA_FQ_RATE_ENABLE] = { .type = NLA_U32 }, [TCA_FQ_FLOW_DEFAULT_RATE] = { .type = NLA_U32 }, [TCA_FQ_FLOW_MAX_RATE] = { .type = NLA_U32 }, diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json index 8acb904d1419..3593fb8f79ad 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq.json @@ -114,6 +114,28 @@ "$IP link del dev $DUMMY type dummy" ] }, + { + "id": "10f7", + "name": "Create FQ with invalid initial_quantum setting", + "category": [ + "qdisc", + "fq" + ], + "plugins": { + "requires": "nsPlugin" + }, + "setup": [ + "$IP link add dev $DUMMY type dummy || /bin/true" + ], + "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root fq initial_quantum 0x80000000", + "expExitCode": "2", + "verifyCmd": "$TC qdisc show dev $DUMMY", + "matchPattern": "qdisc fq 1: root.*initial_quantum 2048Mb", + "matchCount": "0", + "teardown": [ + "$IP link del dev $DUMMY type dummy" + ] + }, { "id": "9398", "name": "Create FQ with maxrate setting", -- cgit v1.2.3 From 6aa445e39693bff9c98b12f960e66b4e18c7378b Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 20 Apr 2023 19:37:23 +0200 Subject: net/handshake: Fix section mismatch in handshake_exit If CONFIG_NET_NS=n (e.g. m68k/defconfig): WARNING: modpost: vmlinux.o: section mismatch in reference: handshake_exit (section: .exit.text) -> handshake_genl_net_ops (section: .init.data) ERROR: modpost: Section mismatches detected. Fix this by dropping the __net_initdata tag from handshake_genl_net_ops. Fixes: 3b3009ea8abb713b ("net/handshake: Create a NETLINK service for handling handshake requests") Reported-by: noreply@ellerman.id.au Closes: http://kisskb.ellerman.id.au/kisskb/buildresult/14912987 Signed-off-by: Geert Uytterhoeven Reviewed-by: Chuck Lever Link: https://lore.kernel.org/r/20230420173723.3773434-1-geert@linux-m68k.org Signed-off-by: Jakub Kicinski --- net/handshake/netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 8ea0ff993f9f..35c9c445e0b8 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -249,7 +249,7 @@ static void __net_exit handshake_net_exit(struct net *net) } } -static struct pernet_operations __net_initdata handshake_genl_net_ops = { +static struct pernet_operations handshake_genl_net_ops = { .init = handshake_net_init, .exit = handshake_net_exit, .id = &handshake_net_id, -- cgit v1.2.3 From 2cc8a008d62f3c04eeb7ec6fe59e542802bb8df3 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Thu, 20 Apr 2023 20:36:33 +0200 Subject: net/sched: cls_api: Initialize miss_cookie_node when action miss is not used Function tcf_exts_init_ex() sets exts->miss_cookie_node ptr only when use_action_miss is true so it assumes in other case that the field is set to NULL by the caller. If not then the field contains garbage and subsequent tcf_exts_destroy() call results in a crash. Ensure that the field .miss_cookie_node pointer is NULL when use_action_miss parameter is false to avoid this potential scenario. Fixes: 80cd22c35c90 ("net/sched: cls_api: Support hardware miss to tc action") Signed-off-by: Ivan Vecera Reviewed-by: Pedro Tammela Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230420183634.1139391-1-ivecera@redhat.com Signed-off-by: Jakub Kicinski --- net/sched/cls_api.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 35785a36c802..3c3629c9e7b6 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -3211,6 +3211,7 @@ int tcf_exts_init_ex(struct tcf_exts *exts, struct net *net, int action, #ifdef CONFIG_NET_CLS_ACT exts->type = 0; exts->nr_actions = 0; + exts->miss_cookie_node = NULL; /* Note: we do not own yet a reference on net. * This reference might be taken later from tcf_exts_get_net(). */ -- cgit v1.2.3 From 418a73074da9182f571e467eaded03ea501f3281 Mon Sep 17 00:00:00 2001 From: Maxime Bizon Date: Thu, 20 Apr 2023 20:25:08 +0200 Subject: net: dst: fix missing initialization of rt_uncached xfrm_alloc_dst() followed by xfrm4_dst_destroy(), without a xfrm4_fill_dst() call in between, causes the following BUG: BUG: spinlock bad magic on CPU#0, fbxhostapd/732 lock: 0x890b7668, .magic: 890b7668, .owner: /-1, .owner_cpu: 0 CPU: 0 PID: 732 Comm: fbxhostapd Not tainted 6.3.0-rc6-next-20230414-00613-ge8de66369925-dirty #9 Hardware name: Marvell Kirkwood (Flattened Device Tree) unwind_backtrace from show_stack+0x10/0x14 show_stack from dump_stack_lvl+0x28/0x30 dump_stack_lvl from do_raw_spin_lock+0x20/0x80 do_raw_spin_lock from rt_del_uncached_list+0x30/0x64 rt_del_uncached_list from xfrm4_dst_destroy+0x3c/0xbc xfrm4_dst_destroy from dst_destroy+0x5c/0xb0 dst_destroy from rcu_process_callbacks+0xc4/0xec rcu_process_callbacks from __do_softirq+0xb4/0x22c __do_softirq from call_with_stack+0x1c/0x24 call_with_stack from do_softirq+0x60/0x6c do_softirq from __local_bh_enable_ip+0xa0/0xcc Patch "net: dst: Prevent false sharing vs. dst_entry:: __refcnt" moved rt_uncached and rt_uncached_list fields from rtable struct to dst struct, so they are more zeroed by memset_after(xdst, 0, u.dst) in xfrm_alloc_dst(). Note that rt_uncached (list_head) was never properly initialized at alloc time, but xfrm[46]_dst_destroy() is written in such a way that it was not an issue thanks to the memset: if (xdst->u.rt.dst.rt_uncached_list) rt_del_uncached_list(&xdst->u.rt); The route code does it the other way around: rt_uncached_list is assumed to be valid IIF rt_uncached list_head is not empty: void rt_del_uncached_list(struct rtable *rt) { if (!list_empty(&rt->dst.rt_uncached)) { struct uncached_list *ul = rt->dst.rt_uncached_list; spin_lock_bh(&ul->lock); list_del_init(&rt->dst.rt_uncached); spin_unlock_bh(&ul->lock); } } This patch adds mandatory rt_uncached list_head initialization in generic dst_init(), and adapt xfrm[46]_dst_destroy logic to match the rest of the code. Fixes: d288a162dd1c ("net: dst: Prevent false sharing vs. dst_entry:: __refcnt") Reported-by: kernel test robot Link: https://lore.kernel.org/oe-lkp/202304162125.18b7bcdd-oliver.sang@intel.com Reviewed-by: David Ahern Reviewed-by: Eric Dumazet CC: Leon Romanovsky Signed-off-by: Maxime Bizon Link: https://lore.kernel.org/r/20230420182508.2417582-1-mbizon@freebox.fr Signed-off-by: Jakub Kicinski --- net/core/dst.c | 1 + net/ipv4/route.c | 4 ---- net/ipv4/xfrm4_policy.c | 4 +--- net/ipv6/route.c | 1 - net/ipv6/xfrm6_policy.c | 4 +--- 5 files changed, 3 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/core/dst.c b/net/core/dst.c index 3247e84045ca..79d9306ad1ee 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -67,6 +67,7 @@ void dst_init(struct dst_entry *dst, struct dst_ops *ops, #endif dst->lwtstate = NULL; rcuref_init(&dst->__rcuref, initial_ref); + INIT_LIST_HEAD(&dst->rt_uncached); dst->__use = 0; dst->lastuse = jiffies; dst->flags = flags; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 2a3d14d95ada..98d7e6ba7493 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1644,7 +1644,6 @@ struct rtable *rt_dst_alloc(struct net_device *dev, rt->rt_uses_gateway = 0; rt->rt_gw_family = 0; rt->rt_gw4 = 0; - INIT_LIST_HEAD(&rt->dst.rt_uncached); rt->dst.output = ip_output; if (flags & RTCF_LOCAL) @@ -1675,7 +1674,6 @@ struct rtable *rt_dst_clone(struct net_device *dev, struct rtable *rt) new_rt->rt_gw4 = rt->rt_gw4; else if (rt->rt_gw_family == AF_INET6) new_rt->rt_gw6 = rt->rt_gw6; - INIT_LIST_HEAD(&new_rt->dst.rt_uncached); new_rt->dst.input = rt->dst.input; new_rt->dst.output = rt->dst.output; @@ -2858,8 +2856,6 @@ struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_or rt->rt_gw4 = ort->rt_gw4; else if (rt->rt_gw_family == AF_INET6) rt->rt_gw6 = ort->rt_gw6; - - INIT_LIST_HEAD(&rt->dst.rt_uncached); } dst_release(dst_orig); diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 47861c8b7340..9403bbaf1b61 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -91,7 +91,6 @@ static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev, xdst->u.rt.rt_gw6 = rt->rt_gw6; xdst->u.rt.rt_pmtu = rt->rt_pmtu; xdst->u.rt.rt_mtu_locked = rt->rt_mtu_locked; - INIT_LIST_HEAD(&xdst->u.rt.dst.rt_uncached); rt_add_uncached_list(&xdst->u.rt); return 0; @@ -121,8 +120,7 @@ static void xfrm4_dst_destroy(struct dst_entry *dst) struct xfrm_dst *xdst = (struct xfrm_dst *)dst; dst_destroy_metrics_generic(dst); - if (xdst->u.rt.dst.rt_uncached_list) - rt_del_uncached_list(&xdst->u.rt); + rt_del_uncached_list(&xdst->u.rt); xfrm_dst_destroy(xdst); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 35085fc0cf15..e3aec46bd466 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -334,7 +334,6 @@ static const struct rt6_info ip6_blk_hole_entry_template = { static void rt6_info_init(struct rt6_info *rt) { memset_after(rt, 0, dst); - INIT_LIST_HEAD(&rt->dst.rt_uncached); } /* allocate dst with ip6_dst_ops */ diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 2b493f8d0091..eecc5e59da17 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -89,7 +89,6 @@ static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev, xdst->u.rt6.rt6i_gateway = rt->rt6i_gateway; xdst->u.rt6.rt6i_dst = rt->rt6i_dst; xdst->u.rt6.rt6i_src = rt->rt6i_src; - INIT_LIST_HEAD(&xdst->u.rt6.dst.rt_uncached); rt6_uncached_list_add(&xdst->u.rt6); return 0; @@ -121,8 +120,7 @@ static void xfrm6_dst_destroy(struct dst_entry *dst) if (likely(xdst->u.rt6.rt6i_idev)) in6_dev_put(xdst->u.rt6.rt6i_idev); dst_destroy_metrics_generic(dst); - if (xdst->u.rt6.dst.rt_uncached_list) - rt6_uncached_list_del(&xdst->u.rt6); + rt6_uncached_list_del(&xdst->u.rt6); xfrm_dst_destroy(xdst); } -- cgit v1.2.3 From 99e5acae193e369b71217efe6f1dad42f3f18815 Mon Sep 17 00:00:00 2001 From: Ziyang Xuan Date: Thu, 20 Apr 2023 20:40:35 +0800 Subject: ipv4: Fix potential uninit variable access bug in __ip_make_skb() Like commit ea30388baebc ("ipv6: Fix an uninit variable access bug in __ip6_make_skb()"). icmphdr does not in skb linear region under the scenario of SOCK_RAW socket. Access icmp_hdr(skb)->type directly will trigger the uninit variable access bug. Use a local variable icmp_type to carry the correct value in different scenarios. Fixes: 96793b482540 ("[IPV4]: Add ICMPMsgStats MIB (RFC 4293)") Reviewed-by: Willem de Bruijn Signed-off-by: Ziyang Xuan Signed-off-by: David S. Miller --- net/ipv4/ip_output.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 4e4e308c3230..89bd7872c629 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1570,9 +1570,19 @@ struct sk_buff *__ip_make_skb(struct sock *sk, cork->dst = NULL; skb_dst_set(skb, &rt->dst); - if (iph->protocol == IPPROTO_ICMP) - icmp_out_count(net, ((struct icmphdr *) - skb_transport_header(skb))->type); + if (iph->protocol == IPPROTO_ICMP) { + u8 icmp_type; + + /* For such sockets, transhdrlen is zero when do ip_append_data(), + * so icmphdr does not in skb linear region and can not get icmp_type + * by icmp_hdr(skb)->type. + */ + if (sk->sk_type == SOCK_RAW && !inet_sk(sk)->hdrincl) + icmp_type = fl4->fl4_icmp_type; + else + icmp_type = icmp_hdr(skb)->type; + icmp_out_count(net, icmp_type); + } ip_cork_release(cork); out: -- cgit v1.2.3 From e0416e7d33361d2ad0bf9f007428346579ac854a Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 21 Apr 2023 23:03:46 +0100 Subject: rxrpc: Fix potential race in error handling in afs_make_call() If the rxrpc call set up by afs_make_call() receives an error whilst it is transmitting the request, there's the possibility that it may get to the point the rxrpc call is ended (after the error_kill_call label) just as the call is queued for async processing. This could manifest itself as call->rxcall being seen as NULL in afs_deliver_to_call() when it tries to lock the call. Fix this by splitting rxrpc_kernel_end_call() into a function to shut down an rxrpc call and a function to release the caller's reference and calling the latter only when we get to afs_put_call(). Reported-by: Jeffrey Altman Signed-off-by: David Howells Tested-by: kafs-testing+fedora36_64checkkafs-build-306@auristor.com cc: Marc Dionne cc: "David S. Miller" cc: Eric Dumazet cc: Jakub Kicinski cc: Paolo Abeni cc: linux-afs@lists.infradead.org cc: netdev@vger.kernel.org Signed-off-by: David S. Miller --- Documentation/networking/rxrpc.rst | 17 ++++++++++++----- fs/afs/rxrpc.c | 9 ++++----- include/net/af_rxrpc.h | 3 ++- net/rxrpc/af_rxrpc.c | 37 +++++++++++++++++++++++++------------ net/rxrpc/rxperf.c | 3 ++- 5 files changed, 45 insertions(+), 24 deletions(-) (limited to 'net') diff --git a/Documentation/networking/rxrpc.rst b/Documentation/networking/rxrpc.rst index ec1323d92c96..e807e18ba32a 100644 --- a/Documentation/networking/rxrpc.rst +++ b/Documentation/networking/rxrpc.rst @@ -848,14 +848,21 @@ The kernel interface functions are as follows: returned. The caller now holds a reference on this and it must be properly ended. - (#) End a client call:: + (#) Shut down a client call:: - void rxrpc_kernel_end_call(struct socket *sock, + void rxrpc_kernel_shutdown_call(struct socket *sock, + struct rxrpc_call *call); + + This is used to shut down a previously begun call. The user_call_ID is + expunged from AF_RXRPC's knowledge and will not be seen again in + association with the specified call. + + (#) Release the ref on a client call:: + + void rxrpc_kernel_put_call(struct socket *sock, struct rxrpc_call *call); - This is used to end a previously begun call. The user_call_ID is expunged - from AF_RXRPC's knowledge and will not be seen again in association with - the specified call. + This is used to release the caller's ref on an rxrpc call. (#) Send data through a call:: diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index 7817e2b860e5..e08b850c3e6d 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -179,7 +179,8 @@ void afs_put_call(struct afs_call *call) ASSERT(call->type->name != NULL); if (call->rxcall) { - rxrpc_kernel_end_call(net->socket, call->rxcall); + rxrpc_kernel_shutdown_call(net->socket, call->rxcall); + rxrpc_kernel_put_call(net->socket, call->rxcall); call->rxcall = NULL; } if (call->type->destructor) @@ -420,10 +421,8 @@ error_kill_call: * The call, however, might be queued on afs_async_calls and we need to * make sure we don't get any more notifications that might requeue it. */ - if (call->rxcall) { - rxrpc_kernel_end_call(call->net->socket, call->rxcall); - call->rxcall = NULL; - } + if (call->rxcall) + rxrpc_kernel_shutdown_call(call->net->socket, call->rxcall); if (call->async) { if (cancel_work_sync(&call->async_work)) afs_put_call(call); diff --git a/include/net/af_rxrpc.h b/include/net/af_rxrpc.h index ba717eac0229..01a35e113ab9 100644 --- a/include/net/af_rxrpc.h +++ b/include/net/af_rxrpc.h @@ -57,7 +57,8 @@ int rxrpc_kernel_recv_data(struct socket *, struct rxrpc_call *, struct iov_iter *, size_t *, bool, u32 *, u16 *); bool rxrpc_kernel_abort_call(struct socket *, struct rxrpc_call *, u32, int, enum rxrpc_abort_reason); -void rxrpc_kernel_end_call(struct socket *, struct rxrpc_call *); +void rxrpc_kernel_shutdown_call(struct socket *sock, struct rxrpc_call *call); +void rxrpc_kernel_put_call(struct socket *sock, struct rxrpc_call *call); void rxrpc_kernel_get_peer(struct socket *, struct rxrpc_call *, struct sockaddr_rxrpc *); bool rxrpc_kernel_get_srtt(struct socket *, struct rxrpc_call *, u32 *); diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 102f5cbff91a..c32b164206f9 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -342,31 +342,44 @@ static void rxrpc_dummy_notify_rx(struct sock *sk, struct rxrpc_call *rxcall, } /** - * rxrpc_kernel_end_call - Allow a kernel service to end a call it was using + * rxrpc_kernel_shutdown_call - Allow a kernel service to shut down a call it was using * @sock: The socket the call is on * @call: The call to end * - * Allow a kernel service to end a call it was using. The call must be + * Allow a kernel service to shut down a call it was using. The call must be * complete before this is called (the call should be aborted if necessary). */ -void rxrpc_kernel_end_call(struct socket *sock, struct rxrpc_call *call) +void rxrpc_kernel_shutdown_call(struct socket *sock, struct rxrpc_call *call) { _enter("%d{%d}", call->debug_id, refcount_read(&call->ref)); mutex_lock(&call->user_mutex); - rxrpc_release_call(rxrpc_sk(sock->sk), call); - - /* Make sure we're not going to call back into a kernel service */ - if (call->notify_rx) { - spin_lock(&call->notify_lock); - call->notify_rx = rxrpc_dummy_notify_rx; - spin_unlock(&call->notify_lock); + if (!test_bit(RXRPC_CALL_RELEASED, &call->flags)) { + rxrpc_release_call(rxrpc_sk(sock->sk), call); + + /* Make sure we're not going to call back into a kernel service */ + if (call->notify_rx) { + spin_lock(&call->notify_lock); + call->notify_rx = rxrpc_dummy_notify_rx; + spin_unlock(&call->notify_lock); + } } - mutex_unlock(&call->user_mutex); +} +EXPORT_SYMBOL(rxrpc_kernel_shutdown_call); + +/** + * rxrpc_kernel_put_call - Release a reference to a call + * @sock: The socket the call is on + * @call: The call to put + * + * Drop the application's ref on an rxrpc call. + */ +void rxrpc_kernel_put_call(struct socket *sock, struct rxrpc_call *call) +{ rxrpc_put_call(call, rxrpc_call_put_kernel); } -EXPORT_SYMBOL(rxrpc_kernel_end_call); +EXPORT_SYMBOL(rxrpc_kernel_put_call); /** * rxrpc_kernel_check_life - Check to see whether a call is still alive diff --git a/net/rxrpc/rxperf.c b/net/rxrpc/rxperf.c index 4a2e90015ca7..085e7892d310 100644 --- a/net/rxrpc/rxperf.c +++ b/net/rxrpc/rxperf.c @@ -342,7 +342,8 @@ static void rxperf_deliver_to_call(struct work_struct *work) call_complete: rxperf_set_call_complete(call, ret, remote_abort); /* The call may have been requeued */ - rxrpc_kernel_end_call(rxperf_socket, call->rxcall); + rxrpc_kernel_shutdown_call(rxperf_socket, call->rxcall); + rxrpc_kernel_put_call(rxperf_socket, call->rxcall); cancel_work(&call->work); kfree(call); } -- cgit v1.2.3 From e8e1ce8454c9cc8ad2e4422bef346428e52455e3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Apr 2023 09:43:53 +0000 Subject: net: add debugging checks in skb_attempt_defer_free() Make sure skbs that are stored in softnet_data.defer_list do not have a dst attached. Also make sure the the skb was orphaned. Link: https://lore.kernel.org/netdev/CANn89iJuEVe72bPmEftyEJHLzzN=QNR2yueFjTxYXCEpS5S8HQ@mail.gmail.com/T/ Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/skbuff.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 0d998806b377..bd815a00d2af 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6881,6 +6881,9 @@ nodefer: __kfree_skb(skb); return; } + DEBUG_NET_WARN_ON_ONCE(skb_dst(skb)); + DEBUG_NET_WARN_ON_ONCE(skb->destructor); + sd = &per_cpu(softnet_data, cpu); defer_max = READ_ONCE(sysctl_skb_defer_max); if (READ_ONCE(sd->defer_count) >= defer_max) -- cgit v1.2.3 From 931e93bdf8ca71cef1f8759c43bc2c5385392b8b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Apr 2023 09:43:54 +0000 Subject: net: do not provide hard irq safety for sd->defer_lock kfree_skb() can be called from hard irq handlers, but skb_attempt_defer_free() is meant to be used from process or BH contexts, and skb_defer_free_flush() is meant to be called from BH contexts. Not having to mask hard irq can save some cycles. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 4 ++-- net/core/skbuff.c | 5 ++--- 2 files changed, 4 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 1551aabac343..d15568f5a44f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6632,11 +6632,11 @@ static void skb_defer_free_flush(struct softnet_data *sd) if (!READ_ONCE(sd->defer_list)) return; - spin_lock_irq(&sd->defer_lock); + spin_lock(&sd->defer_lock); skb = sd->defer_list; sd->defer_list = NULL; sd->defer_count = 0; - spin_unlock_irq(&sd->defer_lock); + spin_unlock(&sd->defer_lock); while (skb != NULL) { next = skb->next; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index bd815a00d2af..304a966164d8 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6870,7 +6870,6 @@ void skb_attempt_defer_free(struct sk_buff *skb) { int cpu = skb->alloc_cpu; struct softnet_data *sd; - unsigned long flags; unsigned int defer_max; bool kick; @@ -6889,7 +6888,7 @@ nodefer: __kfree_skb(skb); if (READ_ONCE(sd->defer_count) >= defer_max) goto nodefer; - spin_lock_irqsave(&sd->defer_lock, flags); + spin_lock_bh(&sd->defer_lock); /* Send an IPI every time queue reaches half capacity. */ kick = sd->defer_count == (defer_max >> 1); /* Paired with the READ_ONCE() few lines above */ @@ -6898,7 +6897,7 @@ nodefer: __kfree_skb(skb); skb->next = sd->defer_list; /* Paired with READ_ONCE() in skb_defer_free_flush() */ WRITE_ONCE(sd->defer_list, skb); - spin_unlock_irqrestore(&sd->defer_lock, flags); + spin_unlock_bh(&sd->defer_lock); /* Make sure to trigger NET_RX_SOFTIRQ on the remote CPU * if we are unlucky enough (this seems very unlikely). -- cgit v1.2.3 From e6f50edfef046cf8ad541b4d5972bf38fdcdec39 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Apr 2023 09:43:55 +0000 Subject: net: move skb_defer_free_flush() up We plan using skb_defer_free_flush() from napi_threaded_poll() in the following patch. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 42 +++++++++++++++++++++--------------------- 1 file changed, 21 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index d15568f5a44f..81ded215731b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6598,6 +6598,27 @@ static int napi_thread_wait(struct napi_struct *napi) return -1; } +static void skb_defer_free_flush(struct softnet_data *sd) +{ + struct sk_buff *skb, *next; + + /* Paired with WRITE_ONCE() in skb_attempt_defer_free() */ + if (!READ_ONCE(sd->defer_list)) + return; + + spin_lock(&sd->defer_lock); + skb = sd->defer_list; + sd->defer_list = NULL; + sd->defer_count = 0; + spin_unlock(&sd->defer_lock); + + while (skb != NULL) { + next = skb->next; + napi_consume_skb(skb, 1); + skb = next; + } +} + static int napi_threaded_poll(void *data) { struct napi_struct *napi = data; @@ -6624,27 +6645,6 @@ static int napi_threaded_poll(void *data) return 0; } -static void skb_defer_free_flush(struct softnet_data *sd) -{ - struct sk_buff *skb, *next; - - /* Paired with WRITE_ONCE() in skb_attempt_defer_free() */ - if (!READ_ONCE(sd->defer_list)) - return; - - spin_lock(&sd->defer_lock); - skb = sd->defer_list; - sd->defer_list = NULL; - sd->defer_count = 0; - spin_unlock(&sd->defer_lock); - - while (skb != NULL) { - next = skb->next; - napi_consume_skb(skb, 1); - skb = next; - } -} - static __latent_entropy void net_rx_action(struct softirq_action *h) { struct softnet_data *sd = this_cpu_ptr(&softnet_data); -- cgit v1.2.3 From a1aaee7f8f79d1b0595e24f8c3caed24630d6cb6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Apr 2023 09:43:56 +0000 Subject: net: make napi_threaded_poll() aware of sd->defer_list If we call skb_defer_free_flush() from napi_threaded_poll(), we can avoid to raise IPI from skb_attempt_defer_free() when the list becomes too big. This allows napi_threaded_poll() to rely less on softirqs, and lowers latency caused by a too big list. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 81ded215731b..7d9ec23f97c6 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6622,6 +6622,7 @@ static void skb_defer_free_flush(struct softnet_data *sd) static int napi_threaded_poll(void *data) { struct napi_struct *napi = data; + struct softnet_data *sd; void *have; while (!napi_thread_wait(napi)) { @@ -6629,11 +6630,13 @@ static int napi_threaded_poll(void *data) bool repoll = false; local_bh_disable(); + sd = this_cpu_ptr(&softnet_data); have = netpoll_poll_lock(napi); __napi_poll(napi, &repoll); netpoll_poll_unlock(have); + skb_defer_free_flush(sd); local_bh_enable(); if (!repoll) -- cgit v1.2.3 From 87eff2ec57b6d68d294013d8dd21e839a1175e3a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 21 Apr 2023 09:43:57 +0000 Subject: net: optimize napi_threaded_poll() vs RPS/RFS We use napi_threaded_poll() in order to reduce our softirq dependency. We can add a followup of 821eba962d95 ("net: optimize napi_schedule_rps()") to further remove the need of firing NET_RX_SOFTIRQ whenever RPS/RFS are used. Signed-off-by: Eric Dumazet Acked-by: Paolo Abeni Signed-off-by: David S. Miller --- include/linux/netdevice.h | 3 +++ net/core/dev.c | 12 ++++++++++-- 2 files changed, 13 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index a6a3e9457d6c..08fbd4622ccf 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3194,7 +3194,10 @@ struct softnet_data { #ifdef CONFIG_RPS struct softnet_data *rps_ipi_list; #endif + bool in_net_rx_action; + bool in_napi_threaded_poll; + #ifdef CONFIG_NET_FLOW_LIMIT struct sd_flow_limit __rcu *flow_limit; #endif diff --git a/net/core/dev.c b/net/core/dev.c index 7d9ec23f97c6..735096d42c1d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4603,10 +4603,10 @@ static void napi_schedule_rps(struct softnet_data *sd) sd->rps_ipi_next = mysd->rps_ipi_list; mysd->rps_ipi_list = sd; - /* If not called from net_rx_action() + /* If not called from net_rx_action() or napi_threaded_poll() * we have to raise NET_RX_SOFTIRQ. */ - if (!mysd->in_net_rx_action) + if (!mysd->in_net_rx_action && !mysd->in_napi_threaded_poll) __raise_softirq_irqoff(NET_RX_SOFTIRQ); return; } @@ -6631,11 +6631,19 @@ static int napi_threaded_poll(void *data) local_bh_disable(); sd = this_cpu_ptr(&softnet_data); + sd->in_napi_threaded_poll = true; have = netpoll_poll_lock(napi); __napi_poll(napi, &repoll); netpoll_poll_unlock(have); + sd->in_napi_threaded_poll = false; + barrier(); + + if (sd_has_rps_ipi_waiting(sd)) { + local_irq_disable(); + net_rps_action_and_irq_enable(sd); + } skb_defer_free_flush(sd); local_bh_enable(); -- cgit v1.2.3 From 788352191c85326beb74f518ba761ed38b82f533 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Fri, 21 Apr 2023 14:08:53 +0100 Subject: rxrpc: Replace fake flex-array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Zero-length arrays as fake flexible arrays are deprecated and we are moving towards adopting C99 flexible-array members instead. Transform zero-length array into flexible-array member in struct rxrpc_ackpacket. Address the following warnings found with GCC-13 and -fstrict-flex-arrays=3 enabled: net/rxrpc/call_event.c:149:38: warning: array subscript i is outside array bounds of ‘uint8_t[0]’ {aka ‘unsigned char[]’} [-Warray-bounds=] This helps with the ongoing efforts to tighten the FORTIFY_SOURCE routines on memcpy() and help us make progress towards globally enabling -fstrict-flex-arrays=3 [1]. Link: https://github.com/KSPP/linux/issues/21 Link: https://github.com/KSPP/linux/issues/263 Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1] Signed-off-by: Gustavo A. R. Silva Signed-off-by: David Howells Reviewed-by: Simon Horman Reviewed-by: Kees Cook Reviewed-by: Jeffrey Altman cc: Marc Dionne cc: "David S. Miller" cc: Eric Dumazet cc: Jakub Kicinski cc: Paolo Abeni cc: linux-afs@lists.infradead.org cc: netdev@vger.kernel.org cc: linux-hardening@vger.kernel.org Link: https://lore.kernel.org/r/ZAZT11n4q5bBttW0@work/ Signed-off-by: David S. Miller --- net/rxrpc/protocol.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/rxrpc/protocol.h b/net/rxrpc/protocol.h index 6760cb99c6d6..e8ee4af43ca8 100644 --- a/net/rxrpc/protocol.h +++ b/net/rxrpc/protocol.h @@ -126,7 +126,7 @@ struct rxrpc_ackpacket { uint8_t nAcks; /* number of ACKs */ #define RXRPC_MAXACKS 255 - uint8_t acks[0]; /* list of ACK/NAKs */ + uint8_t acks[]; /* list of ACK/NAKs */ #define RXRPC_ACK_TYPE_NACK 0 #define RXRPC_ACK_TYPE_ACK 1 -- cgit v1.2.3 From fadfc57cc8047080a123b16f288b7138524fb1e2 Mon Sep 17 00:00:00 2001 From: Marc Dionne Date: Fri, 21 Apr 2023 17:16:17 +0100 Subject: rxrpc: Fix error when reading rxrpc tokens When converting from ASSERTCMP to WARN_ON, the tested condition must be inverted, which was missed for this case. This would cause an EIO error when trying to read an rxrpc token, for instance when trying to display tokens with AuriStor's "tokens" command. Fixes: 84924aac08a4 ("rxrpc: Fix checker warning") Signed-off-by: Marc Dionne Signed-off-by: David Howells cc: "David S. Miller" cc: Eric Dumazet cc: Jakub Kicinski cc: Paolo Abeni cc: linux-afs@lists.infradead.org cc: netdev@vger.kernel.org Signed-off-by: David S. Miller --- net/rxrpc/key.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/rxrpc/key.c b/net/rxrpc/key.c index 8d53aded09c4..33e8302a79e3 100644 --- a/net/rxrpc/key.c +++ b/net/rxrpc/key.c @@ -680,7 +680,7 @@ static long rxrpc_read(const struct key *key, return -ENOPKG; } - if (WARN_ON((unsigned long)xdr - (unsigned long)oldxdr == + if (WARN_ON((unsigned long)xdr - (unsigned long)oldxdr != toksize)) return -EIO; } -- cgit v1.2.3 From 1f5020acb33f926030f62563c86dffca35c7b701 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 21 Apr 2023 01:55:54 +0300 Subject: net: vlan: introduce skb_vlan_eth_hdr() Similar to skb_eth_hdr() introduced in commit 96cc4b69581d ("macvlan: do not assume mac_header is set in macvlan_broadcast()"), let's introduce a skb_vlan_eth_hdr() helper which can be used in TX-only code paths to get to the VLAN header based on skb->data rather than based on the skb_mac_header(skb). We also consolidate the drivers that dereference skb->data to go through this helper. Signed-off-by: Vladimir Oltean Reviewed-by: Eric Dumazet Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 3 +-- drivers/net/ethernet/emulex/benet/be_main.c | 2 +- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 2 +- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 2 +- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 2 +- drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c | 2 +- drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c | 4 ++-- drivers/net/ethernet/sfc/tx_tso.c | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 7 ++----- drivers/staging/gdm724x/gdm_lte.c | 4 ++-- include/linux/if_vlan.h | 12 ++++++++++-- net/batman-adv/soft-interface.c | 2 +- 12 files changed, 24 insertions(+), 20 deletions(-) (limited to 'net') diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 12083b9679b5..6ea5521074d3 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -1935,8 +1935,7 @@ u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb, /* Skip VLAN tag if present */ if (ether_type == ETH_P_8021Q) { - struct vlan_ethhdr *vhdr = - (struct vlan_ethhdr *)skb->data; + struct vlan_ethhdr *vhdr = skb_vlan_eth_hdr(skb); ether_type = ntohs(vhdr->h_vlan_encapsulated_proto); } diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index aed1b622f51f..7e408bcc88de 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -1124,7 +1124,7 @@ static struct sk_buff *be_lancer_xmit_workarounds(struct be_adapter *adapter, struct be_wrb_params *wrb_params) { - struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data; + struct vlan_ethhdr *veh = skb_vlan_eth_hdr(skb); unsigned int eth_hdr_len; struct iphdr *ip; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index 5caea154362f..7356ad965487 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -1532,7 +1532,7 @@ static int hns3_handle_vtags(struct hns3_enet_ring *tx_ring, if (unlikely(rc < 0)) return rc; - vhdr = (struct vlan_ethhdr *)skb->data; + vhdr = skb_vlan_eth_hdr(skb); vhdr->h_vlan_TCI |= cpu_to_be16((skb->priority << VLAN_PRIO_SHIFT) & VLAN_PRIO_MASK); diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c index c8c2cbaa0ede..8b8bf4880faa 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c @@ -3063,7 +3063,7 @@ static inline int i40e_tx_prepare_vlan_flags(struct sk_buff *skb, rc = skb_cow_head(skb, 0); if (rc < 0) return rc; - vhdr = (struct vlan_ethhdr *)skb->data; + vhdr = skb_vlan_eth_hdr(skb); vhdr->h_vlan_TCI = htons(tx_flags >> I40E_TX_FLAGS_VLAN_SHIFT); } else { diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index f2604fc05991..e961ef4bbf4d 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -8798,7 +8798,7 @@ netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, if (skb_cow_head(skb, 0)) goto out_drop; - vhdr = (struct vlan_ethhdr *)skb->data; + vhdr = skb_vlan_eth_hdr(skb); vhdr->h_vlan_TCI = htons(tx_flags >> IXGBE_TX_FLAGS_VLAN_SHIFT); } else { diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c index 59d0dd862fd1..1d1e183d3a8b 100644 --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c @@ -1854,7 +1854,7 @@ netxen_tso_check(struct net_device *netdev, if (protocol == cpu_to_be16(ETH_P_8021Q)) { - vh = (struct vlan_ethhdr *)skb->data; + vh = skb_vlan_eth_hdr(skb); protocol = vh->h_vlan_encapsulated_proto; flags = FLAGS_VLAN_TAGGED; diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c index 92930a055cbc..41894d154013 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c @@ -318,7 +318,7 @@ static void qlcnic_send_filter(struct qlcnic_adapter *adapter, if (adapter->flags & QLCNIC_VLAN_FILTERING) { if (protocol == ETH_P_8021Q) { - vh = (struct vlan_ethhdr *)skb->data; + vh = skb_vlan_eth_hdr(skb); vlan_id = ntohs(vh->h_vlan_TCI); } else if (skb_vlan_tag_present(skb)) { vlan_id = skb_vlan_tag_get(skb); @@ -468,7 +468,7 @@ static int qlcnic_tx_pkt(struct qlcnic_adapter *adapter, u32 producer = tx_ring->producer; if (protocol == ETH_P_8021Q) { - vh = (struct vlan_ethhdr *)skb->data; + vh = skb_vlan_eth_hdr(skb); flags = QLCNIC_FLAGS_VLAN_TAGGED; vlan_tci = ntohs(vh->h_vlan_TCI); protocol = ntohs(vh->h_vlan_encapsulated_proto); diff --git a/drivers/net/ethernet/sfc/tx_tso.c b/drivers/net/ethernet/sfc/tx_tso.c index 898e5c61d908..d381d8164f07 100644 --- a/drivers/net/ethernet/sfc/tx_tso.c +++ b/drivers/net/ethernet/sfc/tx_tso.c @@ -147,7 +147,7 @@ static __be16 efx_tso_check_protocol(struct sk_buff *skb) EFX_WARN_ON_ONCE_PARANOID(((struct ethhdr *)skb->data)->h_proto != protocol); if (protocol == htons(ETH_P_8021Q)) { - struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data; + struct vlan_ethhdr *veh = skb_vlan_eth_hdr(skb); protocol = veh->h_vlan_encapsulated_proto; } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index f116e4ae293b..0fca81507a77 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -4569,13 +4569,10 @@ dma_map_err: static void stmmac_rx_vlan(struct net_device *dev, struct sk_buff *skb) { - struct vlan_ethhdr *veth; - __be16 vlan_proto; + struct vlan_ethhdr *veth = skb_vlan_eth_hdr(skb); + __be16 vlan_proto = veth->h_vlan_proto; u16 vlanid; - veth = (struct vlan_ethhdr *)skb->data; - vlan_proto = veth->h_vlan_proto; - if ((vlan_proto == htons(ETH_P_8021Q) && dev->features & NETIF_F_HW_VLAN_CTAG_RX) || (vlan_proto == htons(ETH_P_8021AD) && diff --git a/drivers/staging/gdm724x/gdm_lte.c b/drivers/staging/gdm724x/gdm_lte.c index 671ee8843c88..5703a9ddb6d0 100644 --- a/drivers/staging/gdm724x/gdm_lte.c +++ b/drivers/staging/gdm724x/gdm_lte.c @@ -349,7 +349,7 @@ static s32 gdm_lte_tx_nic_type(struct net_device *dev, struct sk_buff *skb) /* Get ethernet protocol */ eth = (struct ethhdr *)skb->data; if (ntohs(eth->h_proto) == ETH_P_8021Q) { - vlan_eth = (struct vlan_ethhdr *)skb->data; + vlan_eth = skb_vlan_eth_hdr(skb); mac_proto = ntohs(vlan_eth->h_vlan_encapsulated_proto); network_data = skb->data + VLAN_ETH_HLEN; nic_type |= NIC_TYPE_F_VLAN; @@ -435,7 +435,7 @@ static netdev_tx_t gdm_lte_tx(struct sk_buff *skb, struct net_device *dev) * driver based on the NIC mac */ if (nic_type & NIC_TYPE_F_VLAN) { - struct vlan_ethhdr *vlan_eth = (struct vlan_ethhdr *)skb->data; + struct vlan_ethhdr *vlan_eth = skb_vlan_eth_hdr(skb); nic->vlan_id = ntohs(vlan_eth->h_vlan_TCI) & VLAN_VID_MASK; data_buf = skb->data + (VLAN_ETH_HLEN - ETH_HLEN); diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 90b76d63c11c..3698f2b391cd 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -62,6 +62,14 @@ static inline struct vlan_ethhdr *vlan_eth_hdr(const struct sk_buff *skb) return (struct vlan_ethhdr *)skb_mac_header(skb); } +/* Prefer this version in TX path, instead of + * skb_reset_mac_header() + vlan_eth_hdr() + */ +static inline struct vlan_ethhdr *skb_vlan_eth_hdr(const struct sk_buff *skb) +{ + return (struct vlan_ethhdr *)skb->data; +} + #define VLAN_PRIO_MASK 0xe000 /* Priority Code Point */ #define VLAN_PRIO_SHIFT 13 #define VLAN_CFI_MASK 0x1000 /* Canonical Format Indicator / Drop Eligible Indicator */ @@ -529,7 +537,7 @@ static inline void __vlan_hwaccel_put_tag(struct sk_buff *skb, */ static inline int __vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci) { - struct vlan_ethhdr *veth = (struct vlan_ethhdr *)skb->data; + struct vlan_ethhdr *veth = skb_vlan_eth_hdr(skb); if (!eth_type_vlan(veth->h_vlan_proto)) return -EINVAL; @@ -713,7 +721,7 @@ static inline bool skb_vlan_tagged_multi(struct sk_buff *skb) if (unlikely(!pskb_may_pull(skb, VLAN_ETH_HLEN))) return false; - veh = (struct vlan_ethhdr *)skb->data; + veh = skb_vlan_eth_hdr(skb); protocol = veh->h_vlan_encapsulated_proto; } diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index 125f4628687c..d3fdf82282af 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -439,7 +439,7 @@ void batadv_interface_rx(struct net_device *soft_iface, if (!pskb_may_pull(skb, VLAN_ETH_HLEN)) goto dropped; - vhdr = (struct vlan_ethhdr *)skb->data; + vhdr = skb_vlan_eth_hdr(skb); /* drop batman-in-batman packets to prevent loops */ if (vhdr->h_vlan_encapsulated_proto != htons(ETH_P_BATMAN)) -- cgit v1.2.3 From eabb1494c9f20362ae53a9991481a1523be4f4b7 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 21 Apr 2023 01:55:56 +0300 Subject: net: dsa: tag_ocelot: do not rely on skb_mac_header() for VLAN xmit skb_mac_header() will no longer be available in the TX path when reverting commit 6d1ccff62780 ("net: reset mac header in dev_start_xmit()"). As preparation for that, let's use skb_vlan_eth_hdr() to get to the VLAN header instead, which assumes it's located at skb->data (assumption which holds true here). Signed-off-by: Vladimir Oltean Reviewed-by: Eric Dumazet Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/tag_ocelot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/dsa/tag_ocelot.c b/net/dsa/tag_ocelot.c index 28ebecafdd24..73ee09de1a3a 100644 --- a/net/dsa/tag_ocelot.c +++ b/net/dsa/tag_ocelot.c @@ -26,7 +26,7 @@ static void ocelot_xmit_get_vlan_info(struct sk_buff *skb, struct dsa_port *dp, return; } - hdr = (struct vlan_ethhdr *)skb_mac_header(skb); + hdr = skb_vlan_eth_hdr(skb); br_vlan_get_proto(br, &proto); if (ntohs(hdr->h_vlan_proto) == proto) { -- cgit v1.2.3 From 499b2491d550677b824b828ff431e4ef4d1d3b9d Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 21 Apr 2023 01:55:57 +0300 Subject: net: dsa: tag_ksz: do not rely on skb_mac_header() in TX paths skb_mac_header() will no longer be available in the TX path when reverting commit 6d1ccff62780 ("net: reset mac header in dev_start_xmit()"). As preparation for that, let's use skb_eth_hdr() to get to the Ethernet header's MAC DA instead, helper which assumes this header is located at skb->data (assumption which holds true here). Signed-off-by: Vladimir Oltean Reviewed-by: Eric Dumazet Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/tag_ksz.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/dsa/tag_ksz.c b/net/dsa/tag_ksz.c index 0eb1c7784c3d..ea100bd25939 100644 --- a/net/dsa/tag_ksz.c +++ b/net/dsa/tag_ksz.c @@ -120,18 +120,18 @@ static struct sk_buff *ksz_common_rcv(struct sk_buff *skb, static struct sk_buff *ksz8795_xmit(struct sk_buff *skb, struct net_device *dev) { struct dsa_port *dp = dsa_slave_to_port(dev); + struct ethhdr *hdr; u8 *tag; - u8 *addr; if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb)) return NULL; /* Tag encoding */ tag = skb_put(skb, KSZ_INGRESS_TAG_LEN); - addr = skb_mac_header(skb); + hdr = skb_eth_hdr(skb); *tag = 1 << dp->index; - if (is_link_local_ether_addr(addr)) + if (is_link_local_ether_addr(hdr->h_dest)) *tag |= KSZ8795_TAIL_TAG_OVERRIDE; return skb; @@ -273,8 +273,8 @@ static struct sk_buff *ksz9477_xmit(struct sk_buff *skb, u16 queue_mapping = skb_get_queue_mapping(skb); u8 prio = netdev_txq_to_tc(dev, queue_mapping); struct dsa_port *dp = dsa_slave_to_port(dev); + struct ethhdr *hdr; __be16 *tag; - u8 *addr; u16 val; if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb)) @@ -284,13 +284,13 @@ static struct sk_buff *ksz9477_xmit(struct sk_buff *skb, ksz_xmit_timestamp(dp, skb); tag = skb_put(skb, KSZ9477_INGRESS_TAG_LEN); - addr = skb_mac_header(skb); + hdr = skb_eth_hdr(skb); val = BIT(dp->index); val |= FIELD_PREP(KSZ9477_TAIL_TAG_PRIO, prio); - if (is_link_local_ether_addr(addr)) + if (is_link_local_ether_addr(hdr->h_dest)) val |= KSZ9477_TAIL_TAG_OVERRIDE; *tag = cpu_to_be16(val); @@ -337,7 +337,7 @@ static struct sk_buff *ksz9893_xmit(struct sk_buff *skb, u16 queue_mapping = skb_get_queue_mapping(skb); u8 prio = netdev_txq_to_tc(dev, queue_mapping); struct dsa_port *dp = dsa_slave_to_port(dev); - u8 *addr; + struct ethhdr *hdr; u8 *tag; if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb)) @@ -347,13 +347,13 @@ static struct sk_buff *ksz9893_xmit(struct sk_buff *skb, ksz_xmit_timestamp(dp, skb); tag = skb_put(skb, KSZ_INGRESS_TAG_LEN); - addr = skb_mac_header(skb); + hdr = skb_eth_hdr(skb); *tag = BIT(dp->index); *tag |= FIELD_PREP(KSZ9893_TAIL_TAG_PRIO, prio); - if (is_link_local_ether_addr(addr)) + if (is_link_local_ether_addr(hdr->h_dest)) *tag |= KSZ9893_TAIL_TAG_OVERRIDE; return ksz_defer_xmit(dp, skb); -- cgit v1.2.3 From f9346f00b5af0b32503dcd74072392d9592163ef Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 21 Apr 2023 01:55:58 +0300 Subject: net: dsa: tag_sja1105: don't rely on skb_mac_header() in TX paths skb_mac_header() will no longer be available in the TX path when reverting commit 6d1ccff62780 ("net: reset mac header in dev_start_xmit()"). As preparation for that, let's use skb_vlan_eth_hdr() to get to the VLAN header instead, which assumes it's located at skb->data (assumption which holds true here). Signed-off-by: Vladimir Oltean Reviewed-by: Eric Dumazet Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/tag_sja1105.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c index 1c2ceba4771b..a7ca97b7ac9e 100644 --- a/net/dsa/tag_sja1105.c +++ b/net/dsa/tag_sja1105.c @@ -256,7 +256,7 @@ static struct sk_buff *sja1105_pvid_tag_control_pkt(struct dsa_port *dp, return NULL; } - hdr = (struct vlan_ethhdr *)skb_mac_header(skb); + hdr = skb_vlan_eth_hdr(skb); /* If skb is already VLAN-tagged, leave that VLAN ID in place */ if (hdr->h_vlan_proto == xmit_tpid) -- cgit v1.2.3 From b5653b157e55a25ec76506e8cd31353e0ed4944c Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 21 Apr 2023 01:55:59 +0300 Subject: net: dsa: tag_sja1105: replace skb_mac_header() with vlan_eth_hdr() This is a cosmetic patch which consolidates the code to use the helper function offered by if_vlan.h. Signed-off-by: Vladimir Oltean Reviewed-by: Eric Dumazet Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/tag_sja1105.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c index a7ca97b7ac9e..a5f3b73da417 100644 --- a/net/dsa/tag_sja1105.c +++ b/net/dsa/tag_sja1105.c @@ -516,7 +516,7 @@ static bool sja1110_skb_has_inband_control_extension(const struct sk_buff *skb) static void sja1105_vlan_rcv(struct sk_buff *skb, int *source_port, int *switch_id, int *vbid, u16 *vid) { - struct vlan_ethhdr *hdr = (struct vlan_ethhdr *)skb_mac_header(skb); + struct vlan_ethhdr *hdr = vlan_eth_hdr(skb); u16 vlan_tci; if (skb_vlan_tag_present(skb)) -- cgit v1.2.3 From f0a9d563064c3040a2d7b32317b37a6bcb1099b2 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 21 Apr 2023 01:56:00 +0300 Subject: net: dsa: update TX path comments to not mention skb_mac_header() Once commit 6d1ccff62780 ("net: reset mac header in dev_start_xmit()") will be reverted, it will no longer be true that skb->data points at skb_mac_header(skb) - since the skb->mac_header will not be set - so stop saying that, and just say that it points to the MAC header. I've reviewed vlan_insert_tag() and it does not *actually* depend on skb_mac_header(), so reword that to avoid the confusion. Signed-off-by: Vladimir Oltean Reviewed-by: Eric Dumazet Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/tag.h | 2 +- net/dsa/tag_8021q.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/dsa/tag.h b/net/dsa/tag.h index 7cfbca824f1c..32d12f4a9d73 100644 --- a/net/dsa/tag.h +++ b/net/dsa/tag.h @@ -229,7 +229,7 @@ static inline void *dsa_etype_header_pos_rx(struct sk_buff *skb) return skb->data - 2; } -/* On TX, skb->data points to skb_mac_header(skb), which means that EtherType +/* On TX, skb->data points to the MAC header, which means that EtherType * header taggers start exactly where the EtherType is (the EtherType is * treated as part of the DSA header). */ diff --git a/net/dsa/tag_8021q.c b/net/dsa/tag_8021q.c index 5ee9ef00954e..cbdfc392f7e0 100644 --- a/net/dsa/tag_8021q.c +++ b/net/dsa/tag_8021q.c @@ -461,8 +461,8 @@ EXPORT_SYMBOL_GPL(dsa_tag_8021q_unregister); struct sk_buff *dsa_8021q_xmit(struct sk_buff *skb, struct net_device *netdev, u16 tpid, u16 tci) { - /* skb->data points at skb_mac_header, which - * is fine for vlan_insert_tag. + /* skb->data points at the MAC header, which is fine + * for vlan_insert_tag(). */ return vlan_insert_tag(skb, htons(tpid), tci); } -- cgit v1.2.3 From 0bcf2e4aca6c29a07555b713f2fb461dc38d5977 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 21 Apr 2023 01:56:01 +0300 Subject: net: dsa: tag_ocelot: call only the relevant portion of __skb_vlan_pop() on TX ocelot_xmit_get_vlan_info() calls __skb_vlan_pop() as the most appropriate helper I could find which strips away a VLAN header. That's all I need it to do, but __skb_vlan_pop() has more logic, which will become incompatible with the future revert of commit 6d1ccff62780 ("net: reset mac header in dev_start_xmit()"). Namely, it performs a sanity check on skb_mac_header(), which will stop being set after the above revert, so it will return an error instead of removing the VLAN tag. ocelot_xmit_get_vlan_info() gets called in 2 circumstances: (1) the port is under a VLAN-aware bridge and the bridge sends VLAN-tagged packets (2) the port is under a VLAN-aware bridge and somebody else (an 8021q upper) sends VLAN-tagged packets (using a VID that isn't in the bridge vlan tables) In case (1), there is actually no bug to defend against, because br_dev_xmit() calls skb_reset_mac_header() and things continue to work. However, in case (2), illustrated using the commands below, it can be seen that our intervention is needed, since __skb_vlan_pop() complains: $ ip link add br0 type bridge vlan_filtering 1 && ip link set br0 up $ ip link set $eth master br0 && ip link set $eth up $ ip link add link $eth name $eth.100 type vlan id 100 && ip link set $eth.100 up $ ip addr add 192.168.100.1/24 dev $eth.100 I could fend off the checks in __skb_vlan_pop() with some skb_mac_header_was_set() calls, but seeing how few callers of __skb_vlan_pop() there are from TX paths, that seems rather unproductive. As an alternative solution, extract the bare minimum logic to strip a VLAN header, and move it to a new helper named vlan_remove_tag(), close to the definition of vlan_insert_tag(). Document it appropriately and make ocelot_xmit_get_vlan_info() call this smaller helper instead. Seeing that it doesn't appear illegal to test skb->protocol in the TX path, I guess it would be a good for vlan_remove_tag() to also absorb the vlan_set_encap_proto() function call. Signed-off-by: Vladimir Oltean Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/if_vlan.h | 21 +++++++++++++++++++++ net/core/skbuff.c | 8 +------- net/dsa/tag_ocelot.c | 2 +- 3 files changed, 23 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 3698f2b391cd..0f40f379d75c 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -685,6 +685,27 @@ static inline void vlan_set_encap_proto(struct sk_buff *skb, skb->protocol = htons(ETH_P_802_2); } +/** + * vlan_remove_tag - remove outer VLAN tag from payload + * @skb: skbuff to remove tag from + * @vlan_tci: buffer to store value + * + * Expects the skb to contain a VLAN tag in the payload, and to have skb->data + * pointing at the MAC header. + * + * Returns a new pointer to skb->data, or NULL on failure to pull. + */ +static inline void *vlan_remove_tag(struct sk_buff *skb, u16 *vlan_tci) +{ + struct vlan_hdr *vhdr = (struct vlan_hdr *)(skb->data + ETH_HLEN); + + *vlan_tci = ntohs(vhdr->h_vlan_TCI); + + memmove(skb->data + VLAN_HLEN, skb->data, 2 * ETH_ALEN); + vlan_set_encap_proto(skb, vhdr); + return __skb_pull(skb, VLAN_HLEN); +} + /** * skb_vlan_tagged - check if skb is vlan tagged. * @skb: skbuff to query diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 304a966164d8..c7c141f6fc14 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -5996,7 +5996,6 @@ EXPORT_SYMBOL(skb_ensure_writable); */ int __skb_vlan_pop(struct sk_buff *skb, u16 *vlan_tci) { - struct vlan_hdr *vhdr; int offset = skb->data - skb_mac_header(skb); int err; @@ -6012,13 +6011,8 @@ int __skb_vlan_pop(struct sk_buff *skb, u16 *vlan_tci) skb_postpull_rcsum(skb, skb->data + (2 * ETH_ALEN), VLAN_HLEN); - vhdr = (struct vlan_hdr *)(skb->data + ETH_HLEN); - *vlan_tci = ntohs(vhdr->h_vlan_TCI); - - memmove(skb->data + VLAN_HLEN, skb->data, 2 * ETH_ALEN); - __skb_pull(skb, VLAN_HLEN); + vlan_remove_tag(skb, vlan_tci); - vlan_set_encap_proto(skb, vhdr); skb->mac_header += VLAN_HLEN; if (skb_network_offset(skb) < ETH_HLEN) diff --git a/net/dsa/tag_ocelot.c b/net/dsa/tag_ocelot.c index 73ee09de1a3a..20bf7074d5a6 100644 --- a/net/dsa/tag_ocelot.c +++ b/net/dsa/tag_ocelot.c @@ -30,7 +30,7 @@ static void ocelot_xmit_get_vlan_info(struct sk_buff *skb, struct dsa_port *dp, br_vlan_get_proto(br, &proto); if (ntohs(hdr->h_vlan_proto) == proto) { - __skb_vlan_pop(skb, &tci); + vlan_remove_tag(skb, &tci); *vlan_tci = tci; } else { rcu_read_lock(); -- cgit v1.2.3 From 2f0f9465ad9fa9c93f30009184c10da0f504f313 Mon Sep 17 00:00:00 2001 From: Yajun Deng Date: Fri, 21 Apr 2023 16:26:06 +0800 Subject: net: sched: Print msecs when transmit queue time out The kernel will print several warnings in a short period of time when it stalls. Like this: First warning: [ 7100.097547] ------------[ cut here ]------------ [ 7100.097550] NETDEV WATCHDOG: eno2 (xxx): transmit queue 8 timed out [ 7100.097571] WARNING: CPU: 8 PID: 0 at net/sched/sch_generic.c:467 dev_watchdog+0x260/0x270 ... Second warning: [ 7147.756952] rcu: INFO: rcu_preempt self-detected stall on CPU [ 7147.756958] rcu: 24-....: (59999 ticks this GP) idle=546/1/0x400000000000000 softirq=367 3137/3673146 fqs=13844 [ 7147.756960] (t=60001 jiffies g=4322709 q=133381) [ 7147.756962] NMI backtrace for cpu 24 ... We calculate that the transmit queue start stall should occur before 7095s according to watchdog_timeo, the rcu start stall at 7087s. These two times are close together, it is difficult to confirm which happened first. To let users know the exact time the stall started, print msecs when the transmit queue time out. Signed-off-by: Yajun Deng Signed-off-by: David S. Miller --- net/sched/sch_generic.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index a9aadc4e6858..37e41f972f69 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -502,7 +502,7 @@ static void dev_watchdog(struct timer_list *t) if (netif_device_present(dev) && netif_running(dev) && netif_carrier_ok(dev)) { - int some_queue_timedout = 0; + unsigned int timedout_ms = 0; unsigned int i; unsigned long trans_start; @@ -514,16 +514,16 @@ static void dev_watchdog(struct timer_list *t) if (netif_xmit_stopped(txq) && time_after(jiffies, (trans_start + dev->watchdog_timeo))) { - some_queue_timedout = 1; + timedout_ms = jiffies_to_msecs(jiffies - trans_start); atomic_long_inc(&txq->trans_timeout); break; } } - if (unlikely(some_queue_timedout)) { + if (unlikely(timedout_ms)) { trace_net_dev_xmit_timeout(dev, i); - WARN_ONCE(1, KERN_INFO "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out\n", - dev->name, netdev_drivername(dev), i); + WARN_ONCE(1, "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out %u ms\n", + dev->name, netdev_drivername(dev), i, timedout_ms); netif_freeze_queues(dev); dev->netdev_ops->ndo_tx_timeout(dev, i); netif_unfreeze_queues(dev); -- cgit v1.2.3 From 5036034572b79daa6d6600338e8e8229e2a44b09 Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Fri, 21 Apr 2023 18:25:13 -0300 Subject: net/sched: act_pedit: use NLA_POLICY for parsing 'ex' keys Transform two checks in the 'ex' key parsing into netlink policies removing extra if checks. Signed-off-by: Pedro Tammela Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/act_pedit.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 4559a1507ea5..45201f75e896 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -30,8 +30,9 @@ static const struct nla_policy pedit_policy[TCA_PEDIT_MAX + 1] = { }; static const struct nla_policy pedit_key_ex_policy[TCA_PEDIT_KEY_EX_MAX + 1] = { - [TCA_PEDIT_KEY_EX_HTYPE] = { .type = NLA_U16 }, - [TCA_PEDIT_KEY_EX_CMD] = { .type = NLA_U16 }, + [TCA_PEDIT_KEY_EX_HTYPE] = + NLA_POLICY_MAX(NLA_U16, TCA_PEDIT_HDR_TYPE_MAX), + [TCA_PEDIT_KEY_EX_CMD] = NLA_POLICY_MAX(NLA_U16, TCA_PEDIT_CMD_MAX), }; static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla, @@ -81,12 +82,6 @@ static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla, k->htype = nla_get_u16(tb[TCA_PEDIT_KEY_EX_HTYPE]); k->cmd = nla_get_u16(tb[TCA_PEDIT_KEY_EX_CMD]); - if (k->htype > TCA_PEDIT_HDR_TYPE_MAX || - k->cmd > TCA_PEDIT_CMD_MAX) { - err = -EINVAL; - goto err_out; - } - k++; } -- cgit v1.2.3 From 0c83c5210e18763ec54252b990d270748c99b01d Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Fri, 21 Apr 2023 18:25:14 -0300 Subject: net/sched: act_pedit: use extack in 'ex' parsing errors We have extack available when parsing 'ex' keys, so pass it to tcf_pedit_keys_ex_parse and add more detailed error messages. Reviewed-by: Simon Horman Signed-off-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/act_pedit.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 45201f75e896..24976cd4e4a2 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -36,7 +36,7 @@ static const struct nla_policy pedit_key_ex_policy[TCA_PEDIT_KEY_EX_MAX + 1] = { }; static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla, - u8 n) + u8 n, struct netlink_ext_ack *extack) { struct tcf_pedit_key_ex *keys_ex; struct tcf_pedit_key_ex *k; @@ -57,12 +57,14 @@ static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla, struct nlattr *tb[TCA_PEDIT_KEY_EX_MAX + 1]; if (!n) { + NL_SET_ERR_MSG_MOD(extack, "Can't parse more extended keys than requested"); err = -EINVAL; goto err_out; } n--; if (nla_type(ka) != TCA_PEDIT_KEY_EX) { + NL_SET_ERR_MSG_ATTR(extack, ka, "Unknown attribute, expected extended key"); err = -EINVAL; goto err_out; } @@ -73,8 +75,14 @@ static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla, if (err) goto err_out; - if (!tb[TCA_PEDIT_KEY_EX_HTYPE] || - !tb[TCA_PEDIT_KEY_EX_CMD]) { + if (NL_REQ_ATTR_CHECK(extack, nla, tb, TCA_PEDIT_KEY_EX_HTYPE)) { + NL_SET_ERR_MSG(extack, "Missing required attribute"); + err = -EINVAL; + goto err_out; + } + + if (NL_REQ_ATTR_CHECK(extack, nla, tb, TCA_PEDIT_KEY_EX_CMD)) { + NL_SET_ERR_MSG(extack, "Missing required attribute"); err = -EINVAL; goto err_out; } @@ -86,6 +94,7 @@ static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla, } if (n) { + NL_SET_ERR_MSG_MOD(extack, "Not enough extended keys to parse"); err = -EINVAL; goto err_out; } @@ -217,7 +226,7 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, } nparms->tcfp_keys_ex = - tcf_pedit_keys_ex_parse(tb[TCA_PEDIT_KEYS_EX], parm->nkeys); + tcf_pedit_keys_ex_parse(tb[TCA_PEDIT_KEYS_EX], parm->nkeys, extack); if (IS_ERR(nparms->tcfp_keys_ex)) { ret = PTR_ERR(nparms->tcfp_keys_ex); goto out_free; -- cgit v1.2.3 From e1201bc781c28766720e78a5e099ffa568be4d74 Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Fri, 21 Apr 2023 18:25:15 -0300 Subject: net/sched: act_pedit: check static offsets a priori Static key offsets should always be on 32 bit boundaries. Validate them on create/update time for static offsets and move the datapath validation for runtime offsets only. iproute2 already errors out if a given offset and data size cannot be packed to a 32 bit boundary. This change will make sure users which create/update pedit instances directly via netlink also error out, instead of finding out when packets are traversing. Reviewed-by: Jamal Hadi Salim Reviewed-by: Simon Horman Signed-off-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/act_pedit.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 24976cd4e4a2..cc4dfb01c6c7 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -251,8 +251,16 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, memcpy(nparms->tcfp_keys, parm->keys, ksize); for (i = 0; i < nparms->tcfp_nkeys; ++i) { + u32 offmask = nparms->tcfp_keys[i].offmask; u32 cur = nparms->tcfp_keys[i].off; + /* The AT option can be added to static offsets in the datapath */ + if (!offmask && cur % 4) { + NL_SET_ERR_MSG_MOD(extack, "Offsets must be on 32bit boundaries"); + ret = -EINVAL; + goto put_chain; + } + /* sanitize the shift value for any later use */ nparms->tcfp_keys[i].shift = min_t(size_t, BITS_PER_TYPE(int) - 1, @@ -261,7 +269,7 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, /* The AT option can read a single byte, we can bound the actual * value with uchar max. */ - cur += (0xff & nparms->tcfp_keys[i].offmask) >> nparms->tcfp_keys[i].shift; + cur += (0xff & offmask) >> nparms->tcfp_keys[i].shift; /* Each key touches 4 bytes starting from the computed offset */ nparms->tcfp_off_max_hint = @@ -411,12 +419,12 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, sizeof(_d), &_d); if (!d) goto bad; - offset += (*d & tkey->offmask) >> tkey->shift; - } - if (offset % 4) { - pr_info("tc action pedit offset must be on 32 bit boundaries\n"); - goto bad; + offset += (*d & tkey->offmask) >> tkey->shift; + if (offset % 4) { + pr_info("tc action pedit offset must be on 32 bit boundaries\n"); + goto bad; + } } if (!offset_valid(skb, hoffset + offset)) { -- cgit v1.2.3 From 577140180ba28d0d37bc898c7bd6702c83aa106f Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Fri, 21 Apr 2023 18:25:16 -0300 Subject: net/sched: act_pedit: remove extra check for key type The netlink parsing already validates the key 'htype'. Remove the datapath check as it's redundant. Reviewed-by: Jamal Hadi Salim Reviewed-by: Simon Horman Signed-off-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/act_pedit.c | 29 +++++++---------------------- 1 file changed, 7 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index cc4dfb01c6c7..2fec4473d800 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -325,37 +325,28 @@ static bool offset_valid(struct sk_buff *skb, int offset) return true; } -static int pedit_skb_hdr_offset(struct sk_buff *skb, - enum pedit_header_type htype, int *hoffset) +static void pedit_skb_hdr_offset(struct sk_buff *skb, + enum pedit_header_type htype, int *hoffset) { - int ret = -EINVAL; - + /* 'htype' is validated in the netlink parsing */ switch (htype) { case TCA_PEDIT_KEY_EX_HDR_TYPE_ETH: - if (skb_mac_header_was_set(skb)) { + if (skb_mac_header_was_set(skb)) *hoffset = skb_mac_offset(skb); - ret = 0; - } break; case TCA_PEDIT_KEY_EX_HDR_TYPE_NETWORK: case TCA_PEDIT_KEY_EX_HDR_TYPE_IP4: case TCA_PEDIT_KEY_EX_HDR_TYPE_IP6: *hoffset = skb_network_offset(skb); - ret = 0; break; case TCA_PEDIT_KEY_EX_HDR_TYPE_TCP: case TCA_PEDIT_KEY_EX_HDR_TYPE_UDP: - if (skb_transport_header_was_set(skb)) { + if (skb_transport_header_was_set(skb)) *hoffset = skb_transport_offset(skb); - ret = 0; - } break; default: - ret = -EINVAL; break; } - - return ret; } TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, @@ -388,10 +379,9 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, for (i = parms->tcfp_nkeys; i > 0; i--, tkey++) { int offset = tkey->off; + int hoffset = 0; u32 *ptr, hdata; - int hoffset; u32 val; - int rc; if (tkey_ex) { htype = tkey_ex->htype; @@ -400,12 +390,7 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, tkey_ex++; } - rc = pedit_skb_hdr_offset(skb, htype, &hoffset); - if (rc) { - pr_info("tc action pedit bad header type specified (0x%x)\n", - htype); - goto bad; - } + pedit_skb_hdr_offset(skb, htype, &hoffset); if (tkey->offmask) { u8 *d, _d; -- cgit v1.2.3 From e3c9673e2f6e1b3aa4bb87c570336e10f364c28a Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Fri, 21 Apr 2023 18:25:17 -0300 Subject: net/sched: act_pedit: rate limit datapath messages Unbounded info messages in the pedit datapath can flood the printk ring buffer quite easily depending on the action created. As these messages are informational, usually printing some, not all, is enough to bring attention to the real issue. Reviewed-by: Jamal Hadi Salim Reviewed-by: Simon Horman Signed-off-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/act_pedit.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 2fec4473d800..fb93d4c1faca 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -396,8 +396,8 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, u8 *d, _d; if (!offset_valid(skb, hoffset + tkey->at)) { - pr_info("tc action pedit 'at' offset %d out of bounds\n", - hoffset + tkey->at); + pr_info_ratelimited("tc action pedit 'at' offset %d out of bounds\n", + hoffset + tkey->at); goto bad; } d = skb_header_pointer(skb, hoffset + tkey->at, @@ -407,14 +407,13 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, offset += (*d & tkey->offmask) >> tkey->shift; if (offset % 4) { - pr_info("tc action pedit offset must be on 32 bit boundaries\n"); + pr_info_ratelimited("tc action pedit offset must be on 32 bit boundaries\n"); goto bad; } } if (!offset_valid(skb, hoffset + offset)) { - pr_info("tc action pedit offset %d out of bounds\n", - hoffset + offset); + pr_info_ratelimited("tc action pedit offset %d out of bounds\n", hoffset + offset); goto bad; } @@ -431,8 +430,7 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, val = (*ptr + tkey->val) & ~tkey->mask; break; default: - pr_info("tc action pedit bad command (%d)\n", - cmd); + pr_info_ratelimited("tc action pedit bad command (%d)\n", cmd); goto bad; } -- cgit v1.2.3 From 807cfded92b0a2e8be9c078c693075790c7c4131 Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Sat, 22 Apr 2023 12:56:09 -0300 Subject: net/sched: sch_htb: use extack on errors messages Some error messages are still being printed to dmesg. Since extack is available, provide error messages there. Reviewed-by: Simon Horman Acked-by: Jamal Hadi Salim Signed-off-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/sch_htb.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 92f2975b6a82..8aef7dd9fb88 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1786,7 +1786,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, goto failure; err = nla_parse_nested_deprecated(tb, TCA_HTB_MAX, opt, htb_policy, - NULL); + extack); if (err < 0) goto failure; @@ -1858,7 +1858,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, /* check maximal depth */ if (parent && parent->parent && parent->parent->level < 2) { - pr_err("htb: tree is too deep\n"); + NL_SET_ERR_MSG_MOD(extack, "tree is too deep"); goto failure; } err = -ENOBUFS; @@ -1917,8 +1917,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, }; err = htb_offload(dev, &offload_opt); if (err) { - pr_err("htb: TC_HTB_LEAF_ALLOC_QUEUE failed with err = %d\n", - err); + NL_SET_ERR_MSG_WEAK(extack, + "Failed to offload TC_HTB_LEAF_ALLOC_QUEUE"); goto err_kill_estimator; } dev_queue = netdev_get_tx_queue(dev, offload_opt.qid); @@ -1937,8 +1937,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, }; err = htb_offload(dev, &offload_opt); if (err) { - pr_err("htb: TC_HTB_LEAF_TO_INNER failed with err = %d\n", - err); + NL_SET_ERR_MSG_WEAK(extack, + "Failed to offload TC_HTB_LEAF_TO_INNER"); htb_graft_helper(dev_queue, old_q); goto err_kill_estimator; } @@ -2067,8 +2067,9 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, qdisc_put(parent_qdisc); if (warn) - pr_warn("HTB: quantum of class %X is %s. Consider r2q change.\n", - cl->common.classid, (warn == -1 ? "small" : "big")); + NL_SET_ERR_MSG_FMT_MOD(extack, + "quantum of class %X is %s. Consider r2q change.", + cl->common.classid, (warn == -1 ? "small" : "big")); qdisc_class_hash_grow(sch, &q->clhash); -- cgit v1.2.3 From c69a9b023f65b73068a17f2f4eb6a1b6e257f5bf Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Sat, 22 Apr 2023 12:56:10 -0300 Subject: net/sched: sch_qfq: use extack on errors messages Some error messages are still being printed to dmesg. Since extack is available, provide error messages there. Reviewed-by: Simon Horman Acked-by: Jamal Hadi Salim Signed-off-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/sch_qfq.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 02098a02943e..abcc48087831 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -402,8 +402,8 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, int err; int delta_w; - if (tca[TCA_OPTIONS] == NULL) { - pr_notice("qfq: no options\n"); + if (NL_REQ_ATTR_CHECK(extack, NULL, tca, TCA_OPTIONS)) { + NL_SET_ERR_MSG_MOD(extack, "missing options"); return -EINVAL; } @@ -442,8 +442,9 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, delta_w = weight - (cl ? cl->agg->class_weight : 0); if (q->wsum + delta_w > QFQ_MAX_WSUM) { - pr_notice("qfq: total weight out of range (%d + %u)\n", - delta_w, q->wsum); + NL_SET_ERR_MSG_FMT_MOD(extack, + "total weight out of range (%d + %u)\n", + delta_w, q->wsum); return -EINVAL; } -- cgit v1.2.3 From 25369891fcef373540f8b4e0b3bccf77a04490d5 Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Sat, 22 Apr 2023 12:56:11 -0300 Subject: net/sched: sch_qfq: refactor parsing of netlink parameters Two parameters can be transformed into netlink policies and validated while parsing the netlink message. Reviewed-by: Simon Horman Acked-by: Jamal Hadi Salim Signed-off-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/sch_qfq.c | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index abcc48087831..dfd9a99e6257 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -113,6 +113,7 @@ #define QFQ_MTU_SHIFT 16 /* to support TSO/GSO */ #define QFQ_MIN_LMAX 512 /* see qfq_slot_insert */ +#define QFQ_MAX_LMAX (1UL << QFQ_MTU_SHIFT) #define QFQ_MAX_AGG_CLASSES 8 /* max num classes per aggregate allowed */ @@ -214,9 +215,14 @@ static struct qfq_class *qfq_find_class(struct Qdisc *sch, u32 classid) return container_of(clc, struct qfq_class, common); } +static struct netlink_range_validation lmax_range = { + .min = QFQ_MIN_LMAX, + .max = QFQ_MAX_LMAX, +}; + static const struct nla_policy qfq_policy[TCA_QFQ_MAX + 1] = { - [TCA_QFQ_WEIGHT] = { .type = NLA_U32 }, - [TCA_QFQ_LMAX] = { .type = NLA_U32 }, + [TCA_QFQ_WEIGHT] = NLA_POLICY_RANGE(NLA_U32, 1, QFQ_MAX_WEIGHT), + [TCA_QFQ_LMAX] = NLA_POLICY_FULL_RANGE(NLA_U32, &lmax_range), }; /* @@ -408,17 +414,13 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, } err = nla_parse_nested_deprecated(tb, TCA_QFQ_MAX, tca[TCA_OPTIONS], - qfq_policy, NULL); + qfq_policy, extack); if (err < 0) return err; - if (tb[TCA_QFQ_WEIGHT]) { + if (tb[TCA_QFQ_WEIGHT]) weight = nla_get_u32(tb[TCA_QFQ_WEIGHT]); - if (!weight || weight > (1UL << QFQ_MAX_WSHIFT)) { - pr_notice("qfq: invalid weight %u\n", weight); - return -EINVAL; - } - } else + else weight = 1; if (tb[TCA_QFQ_LMAX]) @@ -426,11 +428,6 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, else lmax = psched_mtu(qdisc_dev(sch)); - if (lmax < QFQ_MIN_LMAX || lmax > (1UL << QFQ_MTU_SHIFT)) { - pr_notice("qfq: invalid max length %u\n", lmax); - return -EINVAL; - } - inv_w = ONE_FP / weight; weight = ONE_FP / inv_w; -- cgit v1.2.3 From ae051b044ebf5938940741c25a8989e07a5d7e88 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 27 Feb 2023 15:50:58 -0800 Subject: Bluetooth: hci_sync: Fix smatch warning This fixes the following new warning: net/bluetooth/hci_sync.c:2403 hci_pause_addr_resolution() warn: missing error code? 'err' Reported-by: kernel test robot Reported-by: Dan Carpenter Link: https://lore.kernel.org/r/202302251952.xryXOegd-lkp@intel.com/ Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 632be1267288..b64af017ab99 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -2403,7 +2403,7 @@ static int hci_pause_addr_resolution(struct hci_dev *hdev) /* Return if address resolution is disabled and RPA is not used. */ if (!err && scan_use_rpa(hdev)) - return err; + return 0; hci_resume_advertising_sync(hdev); return err; -- cgit v1.2.3 From f5d13029a41d6504831ab99da0a9a0332fa7ea19 Mon Sep 17 00:00:00 2001 From: Archie Pusaka Date: Mon, 6 Mar 2023 17:07:07 +0800 Subject: Bluetooth: hci_sync: Don't wait peer's reply when powering off Currently, when we initiate disconnection, we will wait for the peer's reply unless when we are suspending, where we fire and forget the disconnect request. A similar case is when adapter is powering off. However, we still wait for the peer's reply in this case. Therefore, if the peer is unresponsive, the command will time out and the power off sequence will fail, causing "bluetooth powered on by itself" to users. This patch makes the host doesn't wait for the peer's reply when the disconnection reason is powering off. Signed-off-by: Archie Pusaka Reviewed-by: Abhishek Pandit-Subedi Reviewed-by: Simon Horman Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index b64af017ab99..eb5fd2fe1dc2 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -5108,10 +5108,12 @@ static int hci_disconnect_sync(struct hci_dev *hdev, struct hci_conn *conn, cp.handle = cpu_to_le16(conn->handle); cp.reason = reason; - /* Wait for HCI_EV_DISCONN_COMPLETE not HCI_EV_CMD_STATUS when not - * suspending. + /* Wait for HCI_EV_DISCONN_COMPLETE, not HCI_EV_CMD_STATUS, when the + * reason is anything but HCI_ERROR_REMOTE_POWER_OFF. This reason is + * used when suspending or powering off, where we don't want to wait + * for the peer's response. */ - if (!hdev->suspended) + if (reason != HCI_ERROR_REMOTE_POWER_OFF) return __hci_cmd_sync_status_sk(hdev, HCI_OP_DISCONNECT, sizeof(cp), &cp, HCI_EV_DISCONN_COMPLETE, -- cgit v1.2.3 From 493ea69970e2a4e4102fdf0be6d262456f564c3a Mon Sep 17 00:00:00 2001 From: Brian Gix Date: Fri, 24 Feb 2023 09:56:32 -0800 Subject: Bluetooth: Convert MSFT filter HCI cmd to hci_sync The msft_set_filter_enable() command was using the deprecated hci_request mechanism rather than hci_sync. This caused the warning error: hci0: HCI_REQ-0xfcf0 Signed-off-by: Brian Gix Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/msft.c | 36 +++++++++++------------------------- 1 file changed, 11 insertions(+), 25 deletions(-) (limited to 'net') diff --git a/net/bluetooth/msft.c b/net/bluetooth/msft.c index bee6a4c656be..bf5cee48916c 100644 --- a/net/bluetooth/msft.c +++ b/net/bluetooth/msft.c @@ -743,17 +743,12 @@ __u64 msft_get_features(struct hci_dev *hdev) } static void msft_le_set_advertisement_filter_enable_cb(struct hci_dev *hdev, - u8 status, u16 opcode, - struct sk_buff *skb) + void *user_data, + u8 status) { - struct msft_cp_le_set_advertisement_filter_enable *cp; - struct msft_rp_le_set_advertisement_filter_enable *rp; + struct msft_cp_le_set_advertisement_filter_enable *cp = user_data; struct msft_data *msft = hdev->msft_data; - rp = (struct msft_rp_le_set_advertisement_filter_enable *)skb->data; - if (skb->len < sizeof(*rp)) - return; - /* Error 0x0C would be returned if the filter enabled status is * already set to whatever we were trying to set. * Although the default state should be disabled, some controller set @@ -766,7 +761,6 @@ static void msft_le_set_advertisement_filter_enable_cb(struct hci_dev *hdev, hci_dev_lock(hdev); - cp = hci_sent_cmd_data(hdev, hdev->msft_opcode); msft->filter_enabled = cp->enable; if (status == 0x0C) @@ -804,31 +798,23 @@ int msft_remove_monitor(struct hci_dev *hdev, struct adv_monitor *monitor) return msft_remove_monitor_sync(hdev, monitor); } -void msft_req_add_set_filter_enable(struct hci_request *req, bool enable) -{ - struct hci_dev *hdev = req->hdev; - struct msft_cp_le_set_advertisement_filter_enable cp; - - cp.sub_opcode = MSFT_OP_LE_SET_ADVERTISEMENT_FILTER_ENABLE; - cp.enable = enable; - - hci_req_add(req, hdev->msft_opcode, sizeof(cp), &cp); -} - int msft_set_filter_enable(struct hci_dev *hdev, bool enable) { - struct hci_request req; + struct msft_cp_le_set_advertisement_filter_enable cp; struct msft_data *msft = hdev->msft_data; int err; if (!msft) return -EOPNOTSUPP; - hci_req_init(&req, hdev); - msft_req_add_set_filter_enable(&req, enable); - err = hci_req_run_skb(&req, msft_le_set_advertisement_filter_enable_cb); + cp.sub_opcode = MSFT_OP_LE_SET_ADVERTISEMENT_FILTER_ENABLE; + cp.enable = enable; + err = __hci_cmd_sync_status(hdev, hdev->msft_opcode, sizeof(cp), &cp, + HCI_CMD_TIMEOUT); + + msft_le_set_advertisement_filter_enable_cb(hdev, &cp, err); - return err; + return 0; } bool msft_curve_validity(struct hci_dev *hdev) -- cgit v1.2.3 From 9f5ae8fcae1227e9bae3d973fcbef8319a1b25c7 Mon Sep 17 00:00:00 2001 From: Inga Stotland Date: Wed, 8 Mar 2023 10:32:35 -0800 Subject: Bluetooth: hci_sync: Remove duplicate statement This removes the following duplicate statement in hci_le_ext_directed_advertising_sync(): cp.own_addr_type = own_addr_type; Signed-off-by: Inga Stotland Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index eb5fd2fe1dc2..31231f0e4a28 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -5855,7 +5855,6 @@ static int hci_le_ext_directed_advertising_sync(struct hci_dev *hdev, memset(&cp, 0, sizeof(cp)); cp.evt_properties = cpu_to_le16(LE_LEGACY_ADV_DIRECT_IND); - cp.own_addr_type = own_addr_type; cp.channel_map = hdev->le_adv_channel_map; cp.tx_power = HCI_TX_POWER_INVALID; cp.primary_phy = HCI_ADV_PHY_1M; -- cgit v1.2.3 From b8b23001b8025a61f0979578884a74faa825023e Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Wed, 8 Mar 2023 16:16:31 -0800 Subject: Bluetooth: L2CAP: Delay identity address updates This delays the identity address updates to give time for userspace to process the new address otherwise there is a risk that userspace creates a duplicated device if the MGMT event is delayed for some reason. Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/l2cap.h | 2 +- net/bluetooth/l2cap_core.c | 7 +++---- net/bluetooth/smp.c | 9 ++++++++- 3 files changed, 12 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h index 2f766e3437ce..cf393e72d6ed 100644 --- a/include/net/bluetooth/l2cap.h +++ b/include/net/bluetooth/l2cap.h @@ -694,7 +694,7 @@ struct l2cap_conn { struct sk_buff_head pending_rx; struct work_struct pending_rx_work; - struct work_struct id_addr_update_work; + struct delayed_work id_addr_timer; __u8 disc_reason; diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 55a7226233f9..5cc95fd17f7d 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -745,7 +745,7 @@ EXPORT_SYMBOL_GPL(l2cap_chan_list); static void l2cap_conn_update_id_addr(struct work_struct *work) { struct l2cap_conn *conn = container_of(work, struct l2cap_conn, - id_addr_update_work); + id_addr_timer.work); struct hci_conn *hcon = conn->hcon; struct l2cap_chan *chan; @@ -1907,8 +1907,7 @@ static void l2cap_conn_del(struct hci_conn *hcon, int err) if (work_pending(&conn->pending_rx_work)) cancel_work_sync(&conn->pending_rx_work); - if (work_pending(&conn->id_addr_update_work)) - cancel_work_sync(&conn->id_addr_update_work); + cancel_delayed_work_sync(&conn->id_addr_timer); l2cap_unregister_all_users(conn); @@ -7874,7 +7873,7 @@ static struct l2cap_conn *l2cap_conn_add(struct hci_conn *hcon) skb_queue_head_init(&conn->pending_rx); INIT_WORK(&conn->pending_rx_work, process_pending_rx); - INIT_WORK(&conn->id_addr_update_work, l2cap_conn_update_id_addr); + INIT_DELAYED_WORK(&conn->id_addr_timer, l2cap_conn_update_id_addr); conn->disc_reason = HCI_ERROR_REMOTE_USER_TERM; diff --git a/net/bluetooth/smp.c b/net/bluetooth/smp.c index 70663229b3cc..f1a9fc0012f0 100644 --- a/net/bluetooth/smp.c +++ b/net/bluetooth/smp.c @@ -58,6 +58,8 @@ #define SMP_TIMEOUT msecs_to_jiffies(30000) +#define ID_ADDR_TIMEOUT msecs_to_jiffies(200) + #define AUTH_REQ_MASK(dev) (hci_dev_test_flag(dev, HCI_SC_ENABLED) ? \ 0x3f : 0x07) #define KEY_DIST_MASK 0x07 @@ -1067,7 +1069,12 @@ static void smp_notify_keys(struct l2cap_conn *conn) if (hcon->type == LE_LINK) { bacpy(&hcon->dst, &smp->remote_irk->bdaddr); hcon->dst_type = smp->remote_irk->addr_type; - queue_work(hdev->workqueue, &conn->id_addr_update_work); + /* Use a short delay to make sure the new address is + * propagated _before_ the channels. + */ + queue_delayed_work(hdev->workqueue, + &conn->id_addr_timer, + ID_ADDR_TIMEOUT); } } -- cgit v1.2.3 From 8194f1ef5a815aea815a91daf2c721eab2674f1f Mon Sep 17 00:00:00 2001 From: Vasily Khoruzhick Date: Tue, 7 Mar 2023 23:17:30 +0100 Subject: Bluetooth: Add new quirk for broken local ext features page 2 Some adapters (e.g. RTL8723CS) advertise that they have more than 2 pages for local ext features, but they don't support any features declared in these pages. RTL8723CS reports max_page = 2 and declares support for sync train and secure connection, but it responds with either garbage or with error in status on corresponding commands. Signed-off-by: Vasily Khoruzhick Signed-off-by: Bastian Germann Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci.h | 7 +++++++ net/bluetooth/hci_event.c | 9 +++++++-- 2 files changed, 14 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 400f8a7d0c3f..997107bfc0b1 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -294,6 +294,13 @@ enum { * during the hdev->setup vendor callback. */ HCI_QUIRK_BROKEN_MWS_TRANSPORT_CONFIG, + + /* When this quirk is set, max_page for local extended features + * is set to 1, even if controller reports higher number. Some + * controllers (e.g. RTL8723CS) report more pages, but they + * don't actually support features declared there. + */ + HCI_QUIRK_BROKEN_LOCAL_EXT_FEATURES_PAGE_2, }; /* HCI device flags */ diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index e87c928c9e17..51f13518dba9 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -886,8 +886,13 @@ static u8 hci_cc_read_local_ext_features(struct hci_dev *hdev, void *data, if (rp->status) return rp->status; - if (hdev->max_page < rp->max_page) - hdev->max_page = rp->max_page; + if (hdev->max_page < rp->max_page) { + if (test_bit(HCI_QUIRK_BROKEN_LOCAL_EXT_FEATURES_PAGE_2, + &hdev->quirks)) + bt_dev_warn(hdev, "broken local ext features page 2"); + else + hdev->max_page = rp->max_page; + } if (rp->page < HCI_MAX_PAGES) memcpy(hdev->features[rp->page], rp->features, 8); -- cgit v1.2.3 From 9695ef876fd122cb7bbc04a4a93b8727d2e36bda Mon Sep 17 00:00:00 2001 From: Abhishek Pandit-Subedi Date: Thu, 30 Mar 2023 09:58:23 -0700 Subject: Bluetooth: Add support for hci devcoredump Add devcoredump APIs to hci core so that drivers only have to provide the dump skbs instead of managing the synchronization and timeouts. The devcoredump APIs should be used in the following manner: - hci_devcoredump_init is called to allocate the dump. - hci_devcoredump_append is called to append any skbs with dump data OR hci_devcoredump_append_pattern is called to insert a pattern. - hci_devcoredump_complete is called when all dump packets have been sent OR hci_devcoredump_abort is called to indicate an error and cancel an ongoing dump collection. The high level APIs just prepare some skbs with the appropriate data and queue it for the dump to process. Packets part of the crashdump can be intercepted in the driver in interrupt context and forwarded directly to the devcoredump APIs. Internally, there are 5 states for the dump: idle, active, complete, abort and timeout. A devcoredump will only be in active state after it has been initialized. Once active, it accepts data to be appended, patterns to be inserted (i.e. memset) and a completion event or an abort event to generate a devcoredump. The timeout is initialized at the same time the dump is initialized (defaulting to 10s) and will be cleared either when the timeout occurs or the dump is complete or aborted. Signed-off-by: Abhishek Pandit-Subedi Signed-off-by: Manish Mandlik Reviewed-by: Abhishek Pandit-Subedi Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/coredump.h | 116 +++++++++ include/net/bluetooth/hci_core.h | 14 + net/bluetooth/Makefile | 2 + net/bluetooth/coredump.c | 535 +++++++++++++++++++++++++++++++++++++++ net/bluetooth/hci_core.c | 1 + net/bluetooth/hci_sync.c | 2 + 6 files changed, 670 insertions(+) create mode 100644 include/net/bluetooth/coredump.h create mode 100644 net/bluetooth/coredump.c (limited to 'net') diff --git a/include/net/bluetooth/coredump.h b/include/net/bluetooth/coredump.h new file mode 100644 index 000000000000..72f51b587a04 --- /dev/null +++ b/include/net/bluetooth/coredump.h @@ -0,0 +1,116 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2022 Google Corporation + */ + +#ifndef __COREDUMP_H +#define __COREDUMP_H + +#define DEVCOREDUMP_TIMEOUT msecs_to_jiffies(10000) /* 10 sec */ + +typedef void (*coredump_t)(struct hci_dev *hdev); +typedef void (*dmp_hdr_t)(struct hci_dev *hdev, struct sk_buff *skb); +typedef void (*notify_change_t)(struct hci_dev *hdev, int state); + +/* struct hci_devcoredump - Devcoredump state + * + * @supported: Indicates if FW dump collection is supported by driver + * @state: Current state of dump collection + * @timeout: Indicates a timeout for collecting the devcoredump + * + * @alloc_size: Total size of the dump + * @head: Start of the dump + * @tail: Pointer to current end of dump + * @end: head + alloc_size for easy comparisons + * + * @dump_q: Dump queue for state machine to process + * @dump_rx: Devcoredump state machine work + * @dump_timeout: Devcoredump timeout work + * + * @coredump: Called from the driver's .coredump() function. + * @dmp_hdr: Create a dump header to identify controller/fw/driver info + * @notify_change: Notify driver when devcoredump state has changed + */ +struct hci_devcoredump { + bool supported; + + enum devcoredump_state { + HCI_DEVCOREDUMP_IDLE, + HCI_DEVCOREDUMP_ACTIVE, + HCI_DEVCOREDUMP_DONE, + HCI_DEVCOREDUMP_ABORT, + HCI_DEVCOREDUMP_TIMEOUT, + } state; + + unsigned long timeout; + + size_t alloc_size; + char *head; + char *tail; + char *end; + + struct sk_buff_head dump_q; + struct work_struct dump_rx; + struct delayed_work dump_timeout; + + coredump_t coredump; + dmp_hdr_t dmp_hdr; + notify_change_t notify_change; +}; + +#ifdef CONFIG_DEV_COREDUMP + +void hci_devcd_reset(struct hci_dev *hdev); +void hci_devcd_rx(struct work_struct *work); +void hci_devcd_timeout(struct work_struct *work); + +int hci_devcd_register(struct hci_dev *hdev, coredump_t coredump, + dmp_hdr_t dmp_hdr, notify_change_t notify_change); +int hci_devcd_init(struct hci_dev *hdev, u32 dump_size); +int hci_devcd_append(struct hci_dev *hdev, struct sk_buff *skb); +int hci_devcd_append_pattern(struct hci_dev *hdev, u8 pattern, u32 len); +int hci_devcd_complete(struct hci_dev *hdev); +int hci_devcd_abort(struct hci_dev *hdev); + +#else + +static inline void hci_devcd_reset(struct hci_dev *hdev) {} +static inline void hci_devcd_rx(struct work_struct *work) {} +static inline void hci_devcd_timeout(struct work_struct *work) {} + +static inline int hci_devcd_register(struct hci_dev *hdev, coredump_t coredump, + dmp_hdr_t dmp_hdr, + notify_change_t notify_change) +{ + return -EOPNOTSUPP; +} + +static inline int hci_devcd_init(struct hci_dev *hdev, u32 dump_size) +{ + return -EOPNOTSUPP; +} + +static inline int hci_devcd_append(struct hci_dev *hdev, struct sk_buff *skb) +{ + return -EOPNOTSUPP; +} + +static inline int hci_devcd_append_pattern(struct hci_dev *hdev, + u8 pattern, u32 len) +{ + return -EOPNOTSUPP; +} + +static inline int hci_devcd_complete(struct hci_dev *hdev) +{ + return -EOPNOTSUPP; +} + +static inline int hci_devcd_abort(struct hci_dev *hdev) +{ + return -EOPNOTSUPP; +} + +#endif /* CONFIG_DEV_COREDUMP */ + +#endif /* __COREDUMP_H */ diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index e22e45fbe8db..a461335beacf 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -32,6 +32,7 @@ #include #include #include +#include /* HCI priority */ #define HCI_PRIO_MAX 7 @@ -590,6 +591,10 @@ struct hci_dev { const char *fw_info; struct dentry *debugfs; +#ifdef CONFIG_DEV_COREDUMP + struct hci_devcoredump dump; +#endif + struct device dev; struct rfkill *rfkill; @@ -1497,6 +1502,15 @@ static inline void hci_set_aosp_capable(struct hci_dev *hdev) #endif } +static inline void hci_devcd_setup(struct hci_dev *hdev) +{ +#ifdef CONFIG_DEV_COREDUMP + INIT_WORK(&hdev->dump.dump_rx, hci_devcd_rx); + INIT_DELAYED_WORK(&hdev->dump.dump_timeout, hci_devcd_timeout); + skb_queue_head_init(&hdev->dump.dump_q); +#endif +} + int hci_dev_open(__u16 dev); int hci_dev_close(__u16 dev); int hci_dev_do_close(struct hci_dev *hdev); diff --git a/net/bluetooth/Makefile b/net/bluetooth/Makefile index 0e7b7db42750..141ac1fda0bf 100644 --- a/net/bluetooth/Makefile +++ b/net/bluetooth/Makefile @@ -17,6 +17,8 @@ bluetooth-y := af_bluetooth.o hci_core.o hci_conn.o hci_event.o mgmt.o \ ecdh_helper.o hci_request.o mgmt_util.o mgmt_config.o hci_codec.o \ eir.o hci_sync.o +bluetooth-$(CONFIG_DEV_COREDUMP) += coredump.o + bluetooth-$(CONFIG_BT_BREDR) += sco.o bluetooth-$(CONFIG_BT_LE) += iso.o bluetooth-$(CONFIG_BT_HS) += a2mp.o amp.o diff --git a/net/bluetooth/coredump.c b/net/bluetooth/coredump.c new file mode 100644 index 000000000000..08fa98505454 --- /dev/null +++ b/net/bluetooth/coredump.c @@ -0,0 +1,535 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2023 Google Corporation + */ + +#include + +#include +#include + +enum hci_devcoredump_pkt_type { + HCI_DEVCOREDUMP_PKT_INIT, + HCI_DEVCOREDUMP_PKT_SKB, + HCI_DEVCOREDUMP_PKT_PATTERN, + HCI_DEVCOREDUMP_PKT_COMPLETE, + HCI_DEVCOREDUMP_PKT_ABORT, +}; + +struct hci_devcoredump_skb_cb { + u16 pkt_type; +}; + +struct hci_devcoredump_skb_pattern { + u8 pattern; + u32 len; +} __packed; + +#define hci_dmp_cb(skb) ((struct hci_devcoredump_skb_cb *)((skb)->cb)) + +#define DBG_UNEXPECTED_STATE() \ + bt_dev_dbg(hdev, \ + "Unexpected packet (%d) for state (%d). ", \ + hci_dmp_cb(skb)->pkt_type, hdev->dump.state) + +#define MAX_DEVCOREDUMP_HDR_SIZE 512 /* bytes */ + +static int hci_devcd_update_hdr_state(char *buf, size_t size, int state) +{ + int len = 0; + + if (!buf) + return 0; + + len = scnprintf(buf, size, "Bluetooth devcoredump\nState: %d\n", state); + + return len + 1; /* scnprintf adds \0 at the end upon state rewrite */ +} + +/* Call with hci_dev_lock only. */ +static int hci_devcd_update_state(struct hci_dev *hdev, int state) +{ + bt_dev_dbg(hdev, "Updating devcoredump state from %d to %d.", + hdev->dump.state, state); + + hdev->dump.state = state; + + return hci_devcd_update_hdr_state(hdev->dump.head, + hdev->dump.alloc_size, state); +} + +static int hci_devcd_mkheader(struct hci_dev *hdev, struct sk_buff *skb) +{ + char dump_start[] = "--- Start dump ---\n"; + char hdr[80]; + int hdr_len; + + hdr_len = hci_devcd_update_hdr_state(hdr, sizeof(hdr), + HCI_DEVCOREDUMP_IDLE); + skb_put_data(skb, hdr, hdr_len); + + if (hdev->dump.dmp_hdr) + hdev->dump.dmp_hdr(hdev, skb); + + skb_put_data(skb, dump_start, strlen(dump_start)); + + return skb->len; +} + +/* Do not call with hci_dev_lock since this calls driver code. */ +static void hci_devcd_notify(struct hci_dev *hdev, int state) +{ + if (hdev->dump.notify_change) + hdev->dump.notify_change(hdev, state); +} + +/* Call with hci_dev_lock only. */ +void hci_devcd_reset(struct hci_dev *hdev) +{ + hdev->dump.head = NULL; + hdev->dump.tail = NULL; + hdev->dump.alloc_size = 0; + + hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_IDLE); + + cancel_delayed_work(&hdev->dump.dump_timeout); + skb_queue_purge(&hdev->dump.dump_q); +} + +/* Call with hci_dev_lock only. */ +static void hci_devcd_free(struct hci_dev *hdev) +{ + if (hdev->dump.head) + vfree(hdev->dump.head); + + hci_devcd_reset(hdev); +} + +/* Call with hci_dev_lock only. */ +static int hci_devcd_alloc(struct hci_dev *hdev, u32 size) +{ + hdev->dump.head = vmalloc(size); + if (!hdev->dump.head) + return -ENOMEM; + + hdev->dump.alloc_size = size; + hdev->dump.tail = hdev->dump.head; + hdev->dump.end = hdev->dump.head + size; + + hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_IDLE); + + return 0; +} + +/* Call with hci_dev_lock only. */ +static bool hci_devcd_copy(struct hci_dev *hdev, char *buf, u32 size) +{ + if (hdev->dump.tail + size > hdev->dump.end) + return false; + + memcpy(hdev->dump.tail, buf, size); + hdev->dump.tail += size; + + return true; +} + +/* Call with hci_dev_lock only. */ +static bool hci_devcd_memset(struct hci_dev *hdev, u8 pattern, u32 len) +{ + if (hdev->dump.tail + len > hdev->dump.end) + return false; + + memset(hdev->dump.tail, pattern, len); + hdev->dump.tail += len; + + return true; +} + +/* Call with hci_dev_lock only. */ +static int hci_devcd_prepare(struct hci_dev *hdev, u32 dump_size) +{ + struct sk_buff *skb; + int dump_hdr_size; + int err = 0; + + skb = alloc_skb(MAX_DEVCOREDUMP_HDR_SIZE, GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + dump_hdr_size = hci_devcd_mkheader(hdev, skb); + + if (hci_devcd_alloc(hdev, dump_hdr_size + dump_size)) { + err = -ENOMEM; + goto hdr_free; + } + + /* Insert the device header */ + if (!hci_devcd_copy(hdev, skb->data, skb->len)) { + bt_dev_err(hdev, "Failed to insert header"); + hci_devcd_free(hdev); + + err = -ENOMEM; + goto hdr_free; + } + +hdr_free: + kfree_skb(skb); + + return err; +} + +static void hci_devcd_handle_pkt_init(struct hci_dev *hdev, struct sk_buff *skb) +{ + u32 *dump_size; + + if (hdev->dump.state != HCI_DEVCOREDUMP_IDLE) { + DBG_UNEXPECTED_STATE(); + return; + } + + if (skb->len != sizeof(*dump_size)) { + bt_dev_dbg(hdev, "Invalid dump init pkt"); + return; + } + + dump_size = skb_pull_data(skb, sizeof(*dump_size)); + if (!*dump_size) { + bt_dev_err(hdev, "Zero size dump init pkt"); + return; + } + + if (hci_devcd_prepare(hdev, *dump_size)) { + bt_dev_err(hdev, "Failed to prepare for dump"); + return; + } + + hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_ACTIVE); + queue_delayed_work(hdev->workqueue, &hdev->dump.dump_timeout, + hdev->dump.timeout); +} + +static void hci_devcd_handle_pkt_skb(struct hci_dev *hdev, struct sk_buff *skb) +{ + if (hdev->dump.state != HCI_DEVCOREDUMP_ACTIVE) { + DBG_UNEXPECTED_STATE(); + return; + } + + if (!hci_devcd_copy(hdev, skb->data, skb->len)) + bt_dev_dbg(hdev, "Failed to insert skb"); +} + +static void hci_devcd_handle_pkt_pattern(struct hci_dev *hdev, + struct sk_buff *skb) +{ + struct hci_devcoredump_skb_pattern *pattern; + + if (hdev->dump.state != HCI_DEVCOREDUMP_ACTIVE) { + DBG_UNEXPECTED_STATE(); + return; + } + + if (skb->len != sizeof(*pattern)) { + bt_dev_dbg(hdev, "Invalid pattern skb"); + return; + } + + pattern = skb_pull_data(skb, sizeof(*pattern)); + + if (!hci_devcd_memset(hdev, pattern->pattern, pattern->len)) + bt_dev_dbg(hdev, "Failed to set pattern"); +} + +static void hci_devcd_handle_pkt_complete(struct hci_dev *hdev, + struct sk_buff *skb) +{ + u32 dump_size; + + if (hdev->dump.state != HCI_DEVCOREDUMP_ACTIVE) { + DBG_UNEXPECTED_STATE(); + return; + } + + hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_DONE); + dump_size = hdev->dump.tail - hdev->dump.head; + + bt_dev_dbg(hdev, "complete with size %u (expect %zu)", dump_size, + hdev->dump.alloc_size); + + dev_coredumpv(&hdev->dev, hdev->dump.head, dump_size, GFP_KERNEL); +} + +static void hci_devcd_handle_pkt_abort(struct hci_dev *hdev, + struct sk_buff *skb) +{ + u32 dump_size; + + if (hdev->dump.state != HCI_DEVCOREDUMP_ACTIVE) { + DBG_UNEXPECTED_STATE(); + return; + } + + hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_ABORT); + dump_size = hdev->dump.tail - hdev->dump.head; + + bt_dev_dbg(hdev, "aborted with size %u (expect %zu)", dump_size, + hdev->dump.alloc_size); + + /* Emit a devcoredump with the available data */ + dev_coredumpv(&hdev->dev, hdev->dump.head, dump_size, GFP_KERNEL); +} + +/* Bluetooth devcoredump state machine. + * + * Devcoredump states: + * + * HCI_DEVCOREDUMP_IDLE: The default state. + * + * HCI_DEVCOREDUMP_ACTIVE: A devcoredump will be in this state once it has + * been initialized using hci_devcd_init(). Once active, the driver + * can append data using hci_devcd_append() or insert a pattern + * using hci_devcd_append_pattern(). + * + * HCI_DEVCOREDUMP_DONE: Once the dump collection is complete, the drive + * can signal the completion using hci_devcd_complete(). A + * devcoredump is generated indicating the completion event and + * then the state machine is reset to the default state. + * + * HCI_DEVCOREDUMP_ABORT: The driver can cancel ongoing dump collection in + * case of any error using hci_devcd_abort(). A devcoredump is + * still generated with the available data indicating the abort + * event and then the state machine is reset to the default state. + * + * HCI_DEVCOREDUMP_TIMEOUT: A timeout timer for HCI_DEVCOREDUMP_TIMEOUT sec + * is started during devcoredump initialization. Once the timeout + * occurs, the driver is notified, a devcoredump is generated with + * the available data indicating the timeout event and then the + * state machine is reset to the default state. + * + * The driver must register using hci_devcd_register() before using the hci + * devcoredump APIs. + */ +void hci_devcd_rx(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, dump.dump_rx); + struct sk_buff *skb; + int start_state; + + while ((skb = skb_dequeue(&hdev->dump.dump_q))) { + /* Return if timeout occurs. The timeout handler function + * hci_devcd_timeout() will report the available dump data. + */ + if (hdev->dump.state == HCI_DEVCOREDUMP_TIMEOUT) { + kfree_skb(skb); + return; + } + + hci_dev_lock(hdev); + start_state = hdev->dump.state; + + switch (hci_dmp_cb(skb)->pkt_type) { + case HCI_DEVCOREDUMP_PKT_INIT: + hci_devcd_handle_pkt_init(hdev, skb); + break; + + case HCI_DEVCOREDUMP_PKT_SKB: + hci_devcd_handle_pkt_skb(hdev, skb); + break; + + case HCI_DEVCOREDUMP_PKT_PATTERN: + hci_devcd_handle_pkt_pattern(hdev, skb); + break; + + case HCI_DEVCOREDUMP_PKT_COMPLETE: + hci_devcd_handle_pkt_complete(hdev, skb); + break; + + case HCI_DEVCOREDUMP_PKT_ABORT: + hci_devcd_handle_pkt_abort(hdev, skb); + break; + + default: + bt_dev_dbg(hdev, "Unknown packet (%d) for state (%d). ", + hci_dmp_cb(skb)->pkt_type, hdev->dump.state); + break; + } + + hci_dev_unlock(hdev); + kfree_skb(skb); + + /* Notify the driver about any state changes before resetting + * the state machine + */ + if (start_state != hdev->dump.state) + hci_devcd_notify(hdev, hdev->dump.state); + + /* Reset the state machine if the devcoredump is complete */ + hci_dev_lock(hdev); + if (hdev->dump.state == HCI_DEVCOREDUMP_DONE || + hdev->dump.state == HCI_DEVCOREDUMP_ABORT) + hci_devcd_reset(hdev); + hci_dev_unlock(hdev); + } +} +EXPORT_SYMBOL(hci_devcd_rx); + +void hci_devcd_timeout(struct work_struct *work) +{ + struct hci_dev *hdev = container_of(work, struct hci_dev, + dump.dump_timeout.work); + u32 dump_size; + + hci_devcd_notify(hdev, HCI_DEVCOREDUMP_TIMEOUT); + + hci_dev_lock(hdev); + + cancel_work(&hdev->dump.dump_rx); + + hci_devcd_update_state(hdev, HCI_DEVCOREDUMP_TIMEOUT); + + dump_size = hdev->dump.tail - hdev->dump.head; + bt_dev_dbg(hdev, "timeout with size %u (expect %zu)", dump_size, + hdev->dump.alloc_size); + + /* Emit a devcoredump with the available data */ + dev_coredumpv(&hdev->dev, hdev->dump.head, dump_size, GFP_KERNEL); + + hci_devcd_reset(hdev); + + hci_dev_unlock(hdev); +} +EXPORT_SYMBOL(hci_devcd_timeout); + +int hci_devcd_register(struct hci_dev *hdev, coredump_t coredump, + dmp_hdr_t dmp_hdr, notify_change_t notify_change) +{ + /* Driver must implement coredump() and dmp_hdr() functions for + * bluetooth devcoredump. The coredump() should trigger a coredump + * event on the controller when the device's coredump sysfs entry is + * written to. The dmp_hdr() should create a dump header to identify + * the controller/fw/driver info. + */ + if (!coredump || !dmp_hdr) + return -EINVAL; + + hci_dev_lock(hdev); + hdev->dump.coredump = coredump; + hdev->dump.dmp_hdr = dmp_hdr; + hdev->dump.notify_change = notify_change; + hdev->dump.supported = true; + hdev->dump.timeout = DEVCOREDUMP_TIMEOUT; + hci_dev_unlock(hdev); + + return 0; +} +EXPORT_SYMBOL(hci_devcd_register); + +static inline bool hci_devcd_enabled(struct hci_dev *hdev) +{ + return hdev->dump.supported; +} + +int hci_devcd_init(struct hci_dev *hdev, u32 dump_size) +{ + struct sk_buff *skb; + + if (!hci_devcd_enabled(hdev)) + return -EOPNOTSUPP; + + skb = alloc_skb(sizeof(dump_size), GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_INIT; + skb_put_data(skb, &dump_size, sizeof(dump_size)); + + skb_queue_tail(&hdev->dump.dump_q, skb); + queue_work(hdev->workqueue, &hdev->dump.dump_rx); + + return 0; +} +EXPORT_SYMBOL(hci_devcd_init); + +int hci_devcd_append(struct hci_dev *hdev, struct sk_buff *skb) +{ + if (!skb) + return -ENOMEM; + + if (!hci_devcd_enabled(hdev)) { + kfree_skb(skb); + return -EOPNOTSUPP; + } + + hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_SKB; + + skb_queue_tail(&hdev->dump.dump_q, skb); + queue_work(hdev->workqueue, &hdev->dump.dump_rx); + + return 0; +} +EXPORT_SYMBOL(hci_devcd_append); + +int hci_devcd_append_pattern(struct hci_dev *hdev, u8 pattern, u32 len) +{ + struct hci_devcoredump_skb_pattern p; + struct sk_buff *skb; + + if (!hci_devcd_enabled(hdev)) + return -EOPNOTSUPP; + + skb = alloc_skb(sizeof(p), GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + p.pattern = pattern; + p.len = len; + + hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_PATTERN; + skb_put_data(skb, &p, sizeof(p)); + + skb_queue_tail(&hdev->dump.dump_q, skb); + queue_work(hdev->workqueue, &hdev->dump.dump_rx); + + return 0; +} +EXPORT_SYMBOL(hci_devcd_append_pattern); + +int hci_devcd_complete(struct hci_dev *hdev) +{ + struct sk_buff *skb; + + if (!hci_devcd_enabled(hdev)) + return -EOPNOTSUPP; + + skb = alloc_skb(0, GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_COMPLETE; + + skb_queue_tail(&hdev->dump.dump_q, skb); + queue_work(hdev->workqueue, &hdev->dump.dump_rx); + + return 0; +} +EXPORT_SYMBOL(hci_devcd_complete); + +int hci_devcd_abort(struct hci_dev *hdev) +{ + struct sk_buff *skb; + + if (!hci_devcd_enabled(hdev)) + return -EOPNOTSUPP; + + skb = alloc_skb(0, GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_ABORT; + + skb_queue_tail(&hdev->dump.dump_q, skb); + queue_work(hdev->workqueue, &hdev->dump.dump_rx); + + return 0; +} +EXPORT_SYMBOL(hci_devcd_abort); diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 334e308451f5..393b317ae68f 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -2544,6 +2544,7 @@ struct hci_dev *hci_alloc_dev_priv(int sizeof_priv) INIT_DELAYED_WORK(&hdev->cmd_timer, hci_cmd_timeout); INIT_DELAYED_WORK(&hdev->ncmd_timer, hci_ncmd_timeout); + hci_devcd_setup(hdev); hci_request_setup(hdev); hci_init_sysfs(hdev); diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 31231f0e4a28..f21497ebc814 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -4727,6 +4727,8 @@ int hci_dev_open_sync(struct hci_dev *hdev) goto done; } + hci_devcd_reset(hdev); + set_bit(HCI_RUNNING, &hdev->flags); hci_sock_dev_event(hdev, HCI_DEV_OPEN); -- cgit v1.2.3 From 0fe8c8d071343fa9278980ce4b6f8e6ea24a2ed1 Mon Sep 17 00:00:00 2001 From: Iulia Tanasescu Date: Fri, 31 Mar 2023 18:38:01 +0300 Subject: Bluetooth: Split bt_iso_qos into dedicated structures Split bt_iso_qos into dedicated unicast and broadcast structures and add additional broadcast parameters. Fixes: eca0ae4aea66 ("Bluetooth: Add initial implementation of BIS connections") Signed-off-by: Iulia Tanasescu Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/bluetooth.h | 43 +++++++--- include/net/bluetooth/hci_core.h | 9 ++- net/bluetooth/hci_conn.c | 162 ++++++++++++++++++++------------------ net/bluetooth/hci_event.c | 33 ++++---- net/bluetooth/iso.c | 125 +++++++++++++++++++++++------ 5 files changed, 237 insertions(+), 135 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h index bcc5a4cd2c17..1b4230cd42a3 100644 --- a/include/net/bluetooth/bluetooth.h +++ b/include/net/bluetooth/bluetooth.h @@ -1,6 +1,7 @@ /* BlueZ - Bluetooth protocol stack for Linux Copyright (C) 2000-2001 Qualcomm Incorporated + Copyright 2023 NXP Written 2000,2001 by Maxim Krasnyansky @@ -171,23 +172,39 @@ struct bt_iso_io_qos { __u8 rtn; }; -struct bt_iso_qos { - union { - __u8 cig; - __u8 big; - }; - union { - __u8 cis; - __u8 bis; - }; - union { - __u8 sca; - __u8 sync_interval; - }; +struct bt_iso_ucast_qos { + __u8 cig; + __u8 cis; + __u8 sca; + __u8 packing; + __u8 framing; + struct bt_iso_io_qos in; + struct bt_iso_io_qos out; +}; + +struct bt_iso_bcast_qos { + __u8 big; + __u8 bis; + __u8 sync_interval; __u8 packing; __u8 framing; struct bt_iso_io_qos in; struct bt_iso_io_qos out; + __u8 encryption; + __u8 bcode[16]; + __u8 options; + __u16 skip; + __u16 sync_timeout; + __u8 sync_cte_type; + __u8 mse; + __u16 timeout; +}; + +struct bt_iso_qos { + union { + struct bt_iso_ucast_qos ucast; + struct bt_iso_bcast_qos bcast; + }; }; #define BT_ISO_PHY_1M 0x01 diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index a461335beacf..f11689284112 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1,6 +1,7 @@ /* BlueZ - Bluetooth protocol stack for Linux Copyright (c) 2000-2001, 2010, Code Aurora Forum. All rights reserved. + Copyright 2023 NXP Written 2000,2001 by Maxim Krasnyansky @@ -1096,7 +1097,7 @@ static inline struct hci_conn *hci_conn_hash_lookup_bis(struct hci_dev *hdev, if (bacmp(&c->dst, ba) || c->type != ISO_LINK) continue; - if (c->iso_qos.big == big && c->iso_qos.bis == bis) { + if (c->iso_qos.bcast.big == big && c->iso_qos.bcast.bis == bis) { rcu_read_unlock(); return c; } @@ -1205,7 +1206,7 @@ static inline struct hci_conn *hci_conn_hash_lookup_cig(struct hci_dev *hdev, if (c->type != ISO_LINK) continue; - if (handle == c->iso_qos.cig) { + if (handle == c->iso_qos.ucast.cig) { rcu_read_unlock(); return c; } @@ -1228,7 +1229,7 @@ static inline struct hci_conn *hci_conn_hash_lookup_big(struct hci_dev *hdev, if (bacmp(&c->dst, BDADDR_ANY) || c->type != ISO_LINK) continue; - if (handle == c->iso_qos.big) { + if (handle == c->iso_qos.bcast.big) { rcu_read_unlock(); return c; } @@ -1337,7 +1338,7 @@ struct hci_conn *hci_connect_bis(struct hci_dev *hdev, bdaddr_t *dst, __u8 dst_type, struct bt_iso_qos *qos, __u8 data_len, __u8 *data); int hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst, __u8 dst_type, - __u8 sid); + __u8 sid, struct bt_iso_qos *qos); int hci_le_big_create_sync(struct hci_dev *hdev, struct bt_iso_qos *qos, __u16 sync_handle, __u8 num_bis, __u8 bis[]); int hci_conn_check_link_mode(struct hci_conn *conn); diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 8455ba141ee6..5672b4924572 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1,6 +1,7 @@ /* BlueZ - Bluetooth protocol stack for Linux Copyright (c) 2000-2001, 2010, Code Aurora Forum. All rights reserved. + Copyright 2023 NXP Written 2000,2001 by Maxim Krasnyansky @@ -795,8 +796,8 @@ static void bis_list(struct hci_conn *conn, void *data) if (bacmp(&conn->dst, BDADDR_ANY)) return; - if (d->big != conn->iso_qos.big || d->bis == BT_ISO_QOS_BIS_UNSET || - d->bis != conn->iso_qos.bis) + if (d->big != conn->iso_qos.bcast.big || d->bis == BT_ISO_QOS_BIS_UNSET || + d->bis != conn->iso_qos.bcast.bis) return; d->count++; @@ -916,10 +917,10 @@ static void bis_cleanup(struct hci_conn *conn) if (!test_and_clear_bit(HCI_CONN_PER_ADV, &conn->flags)) return; - hci_le_terminate_big(hdev, conn->iso_qos.big, - conn->iso_qos.bis); + hci_le_terminate_big(hdev, conn->iso_qos.bcast.big, + conn->iso_qos.bcast.bis); } else { - hci_le_big_terminate(hdev, conn->iso_qos.big, + hci_le_big_terminate(hdev, conn->iso_qos.bcast.big, conn->sync_handle); } } @@ -959,7 +960,7 @@ static void cis_cleanup(struct hci_conn *conn) struct iso_list_data d; memset(&d, 0, sizeof(d)); - d.cig = conn->iso_qos.cig; + d.cig = conn->iso_qos.ucast.cig; /* Check if ISO connection is a CIS and remove CIG if there are * no other connections using it. @@ -968,7 +969,7 @@ static void cis_cleanup(struct hci_conn *conn) if (d.count) return; - hci_le_remove_cig(hdev, conn->iso_qos.cig); + hci_le_remove_cig(hdev, conn->iso_qos.ucast.cig); } struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, @@ -1411,7 +1412,7 @@ static int qos_set_big(struct hci_dev *hdev, struct bt_iso_qos *qos) struct iso_list_data data; /* Allocate a BIG if not set */ - if (qos->big == BT_ISO_QOS_BIG_UNSET) { + if (qos->bcast.big == BT_ISO_QOS_BIG_UNSET) { for (data.big = 0x00; data.big < 0xef; data.big++) { data.count = 0; data.bis = 0xff; @@ -1426,7 +1427,7 @@ static int qos_set_big(struct hci_dev *hdev, struct bt_iso_qos *qos) return -EADDRNOTAVAIL; /* Update BIG */ - qos->big = data.big; + qos->bcast.big = data.big; } return 0; @@ -1437,7 +1438,7 @@ static int qos_set_bis(struct hci_dev *hdev, struct bt_iso_qos *qos) struct iso_list_data data; /* Allocate BIS if not set */ - if (qos->bis == BT_ISO_QOS_BIS_UNSET) { + if (qos->bcast.bis == BT_ISO_QOS_BIS_UNSET) { /* Find an unused adv set to advertise BIS, skip instance 0x00 * since it is reserved as general purpose set. */ @@ -1455,7 +1456,7 @@ static int qos_set_bis(struct hci_dev *hdev, struct bt_iso_qos *qos) return -EADDRNOTAVAIL; /* Update BIS */ - qos->bis = data.bis; + qos->bcast.bis = data.bis; } return 0; @@ -1484,8 +1485,8 @@ static struct hci_conn *hci_add_bis(struct hci_dev *hdev, bdaddr_t *dst, if (err) return ERR_PTR(err); - data.big = qos->big; - data.bis = qos->bis; + data.big = qos->bcast.big; + data.bis = qos->bcast.bis; data.count = 0; /* Check if there is already a matching BIG/BIS */ @@ -1493,7 +1494,7 @@ static struct hci_conn *hci_add_bis(struct hci_dev *hdev, bdaddr_t *dst, if (data.count) return ERR_PTR(-EADDRINUSE); - conn = hci_conn_hash_lookup_bis(hdev, dst, qos->big, qos->bis); + conn = hci_conn_hash_lookup_bis(hdev, dst, qos->bcast.big, qos->bcast.bis); if (conn) return ERR_PTR(-EADDRINUSE); @@ -1648,13 +1649,13 @@ static void cis_add(struct iso_list_data *d, struct bt_iso_qos *qos) { struct hci_cis_params *cis = &d->pdu.cis[d->pdu.cp.num_cis]; - cis->cis_id = qos->cis; - cis->c_sdu = cpu_to_le16(qos->out.sdu); - cis->p_sdu = cpu_to_le16(qos->in.sdu); - cis->c_phy = qos->out.phy ? qos->out.phy : qos->in.phy; - cis->p_phy = qos->in.phy ? qos->in.phy : qos->out.phy; - cis->c_rtn = qos->out.rtn; - cis->p_rtn = qos->in.rtn; + cis->cis_id = qos->ucast.cis; + cis->c_sdu = cpu_to_le16(qos->ucast.out.sdu); + cis->p_sdu = cpu_to_le16(qos->ucast.in.sdu); + cis->c_phy = qos->ucast.out.phy ? qos->ucast.out.phy : qos->ucast.in.phy; + cis->p_phy = qos->ucast.in.phy ? qos->ucast.in.phy : qos->ucast.out.phy; + cis->c_rtn = qos->ucast.out.rtn; + cis->p_rtn = qos->ucast.in.rtn; d->pdu.cp.num_cis++; } @@ -1667,8 +1668,8 @@ static void cis_list(struct hci_conn *conn, void *data) if (!bacmp(&conn->dst, BDADDR_ANY)) return; - if (d->cig != conn->iso_qos.cig || d->cis == BT_ISO_QOS_CIS_UNSET || - d->cis != conn->iso_qos.cis) + if (d->cig != conn->iso_qos.ucast.cig || d->cis == BT_ISO_QOS_CIS_UNSET || + d->cis != conn->iso_qos.ucast.cis) return; d->count++; @@ -1687,17 +1688,18 @@ static int hci_le_create_big(struct hci_conn *conn, struct bt_iso_qos *qos) memset(&cp, 0, sizeof(cp)); - cp.handle = qos->big; - cp.adv_handle = qos->bis; + cp.handle = qos->bcast.big; + cp.adv_handle = qos->bcast.bis; cp.num_bis = 0x01; - hci_cpu_to_le24(qos->out.interval, cp.bis.sdu_interval); - cp.bis.sdu = cpu_to_le16(qos->out.sdu); - cp.bis.latency = cpu_to_le16(qos->out.latency); - cp.bis.rtn = qos->out.rtn; - cp.bis.phy = qos->out.phy; - cp.bis.packing = qos->packing; - cp.bis.framing = qos->framing; - cp.bis.encryption = 0x00; + hci_cpu_to_le24(qos->bcast.out.interval, cp.bis.sdu_interval); + cp.bis.sdu = cpu_to_le16(qos->bcast.out.sdu); + cp.bis.latency = cpu_to_le16(qos->bcast.out.latency); + cp.bis.rtn = qos->bcast.out.rtn; + cp.bis.phy = qos->bcast.out.phy; + cp.bis.packing = qos->bcast.packing; + cp.bis.framing = qos->bcast.framing; + cp.bis.encryption = qos->bcast.encryption; + memcpy(cp.bis.bcode, qos->bcast.bcode, sizeof(cp.bis.bcode)); memset(&cp.bis.bcode, 0, sizeof(cp.bis.bcode)); return hci_send_cmd(hdev, HCI_OP_LE_CREATE_BIG, sizeof(cp), &cp); @@ -1711,7 +1713,7 @@ static bool hci_le_set_cig_params(struct hci_conn *conn, struct bt_iso_qos *qos) memset(&data, 0, sizeof(data)); /* Allocate a CIG if not set */ - if (qos->cig == BT_ISO_QOS_CIG_UNSET) { + if (qos->ucast.cig == BT_ISO_QOS_CIG_UNSET) { for (data.cig = 0x00; data.cig < 0xff; data.cig++) { data.count = 0; data.cis = 0xff; @@ -1731,22 +1733,22 @@ static bool hci_le_set_cig_params(struct hci_conn *conn, struct bt_iso_qos *qos) return false; /* Update CIG */ - qos->cig = data.cig; + qos->ucast.cig = data.cig; } - data.pdu.cp.cig_id = qos->cig; - hci_cpu_to_le24(qos->out.interval, data.pdu.cp.c_interval); - hci_cpu_to_le24(qos->in.interval, data.pdu.cp.p_interval); - data.pdu.cp.sca = qos->sca; - data.pdu.cp.packing = qos->packing; - data.pdu.cp.framing = qos->framing; - data.pdu.cp.c_latency = cpu_to_le16(qos->out.latency); - data.pdu.cp.p_latency = cpu_to_le16(qos->in.latency); + data.pdu.cp.cig_id = qos->ucast.cig; + hci_cpu_to_le24(qos->ucast.out.interval, data.pdu.cp.c_interval); + hci_cpu_to_le24(qos->ucast.in.interval, data.pdu.cp.p_interval); + data.pdu.cp.sca = qos->ucast.sca; + data.pdu.cp.packing = qos->ucast.packing; + data.pdu.cp.framing = qos->ucast.framing; + data.pdu.cp.c_latency = cpu_to_le16(qos->ucast.out.latency); + data.pdu.cp.p_latency = cpu_to_le16(qos->ucast.in.latency); - if (qos->cis != BT_ISO_QOS_CIS_UNSET) { + if (qos->ucast.cis != BT_ISO_QOS_CIS_UNSET) { data.count = 0; - data.cig = qos->cig; - data.cis = qos->cis; + data.cig = qos->ucast.cig; + data.cis = qos->ucast.cis; hci_conn_hash_list_state(hdev, cis_list, ISO_LINK, BT_BOUND, &data); @@ -1757,7 +1759,7 @@ static bool hci_le_set_cig_params(struct hci_conn *conn, struct bt_iso_qos *qos) } /* Reprogram all CIS(s) with the same CIG */ - for (data.cig = qos->cig, data.cis = 0x00; data.cis < 0x11; + for (data.cig = qos->ucast.cig, data.cis = 0x00; data.cis < 0x11; data.cis++) { data.count = 0; @@ -1767,14 +1769,14 @@ static bool hci_le_set_cig_params(struct hci_conn *conn, struct bt_iso_qos *qos) continue; /* Allocate a CIS if not set */ - if (qos->cis == BT_ISO_QOS_CIS_UNSET) { + if (qos->ucast.cis == BT_ISO_QOS_CIS_UNSET) { /* Update CIS */ - qos->cis = data.cis; + qos->ucast.cis = data.cis; cis_add(&data, qos); } } - if (qos->cis == BT_ISO_QOS_CIS_UNSET || !data.pdu.cp.num_cis) + if (qos->ucast.cis == BT_ISO_QOS_CIS_UNSET || !data.pdu.cp.num_cis) return false; if (hci_send_cmd(hdev, HCI_OP_LE_SET_CIG_PARAMS, @@ -1809,32 +1811,32 @@ struct hci_conn *hci_bind_cis(struct hci_dev *hdev, bdaddr_t *dst, return cis; /* Update LINK PHYs according to QoS preference */ - cis->le_tx_phy = qos->out.phy; - cis->le_rx_phy = qos->in.phy; + cis->le_tx_phy = qos->ucast.out.phy; + cis->le_rx_phy = qos->ucast.in.phy; /* If output interval is not set use the input interval as it cannot be * 0x000000. */ - if (!qos->out.interval) - qos->out.interval = qos->in.interval; + if (!qos->ucast.out.interval) + qos->ucast.out.interval = qos->ucast.in.interval; /* If input interval is not set use the output interval as it cannot be * 0x000000. */ - if (!qos->in.interval) - qos->in.interval = qos->out.interval; + if (!qos->ucast.in.interval) + qos->ucast.in.interval = qos->ucast.out.interval; /* If output latency is not set use the input latency as it cannot be * 0x0000. */ - if (!qos->out.latency) - qos->out.latency = qos->in.latency; + if (!qos->ucast.out.latency) + qos->ucast.out.latency = qos->ucast.in.latency; /* If input latency is not set use the output latency as it cannot be * 0x0000. */ - if (!qos->in.latency) - qos->in.latency = qos->out.latency; + if (!qos->ucast.in.latency) + qos->ucast.in.latency = qos->ucast.out.latency; if (!hci_le_set_cig_params(cis, qos)) { hci_conn_drop(cis); @@ -1854,7 +1856,7 @@ bool hci_iso_setup_path(struct hci_conn *conn) memset(&cmd, 0, sizeof(cmd)); - if (conn->iso_qos.out.sdu) { + if (conn->iso_qos.ucast.out.sdu) { cmd.handle = cpu_to_le16(conn->handle); cmd.direction = 0x00; /* Input (Host to Controller) */ cmd.path = 0x00; /* HCI path if enabled */ @@ -1865,7 +1867,7 @@ bool hci_iso_setup_path(struct hci_conn *conn) return false; } - if (conn->iso_qos.in.sdu) { + if (conn->iso_qos.ucast.in.sdu) { cmd.handle = cpu_to_le16(conn->handle); cmd.direction = 0x01; /* Output (Controller to Host) */ cmd.path = 0x00; /* HCI path if enabled */ @@ -1892,7 +1894,7 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data) cmd.cis[0].acl_handle = cpu_to_le16(conn->link->handle); cmd.cis[0].cis_handle = cpu_to_le16(conn->handle); cmd.cp.num_cis++; - cig = conn->iso_qos.cig; + cig = conn->iso_qos.ucast.cig; hci_dev_lock(hdev); @@ -1902,7 +1904,7 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data) struct hci_cis *cis = &cmd.cis[cmd.cp.num_cis]; if (conn == data || conn->type != ISO_LINK || - conn->state == BT_CONNECTED || conn->iso_qos.cig != cig) + conn->state == BT_CONNECTED || conn->iso_qos.ucast.cig != cig) continue; /* Check if all CIS(s) belonging to a CIG are ready */ @@ -2002,8 +2004,8 @@ static void hci_bind_bis(struct hci_conn *conn, struct bt_iso_qos *qos) { /* Update LINK PHYs according to QoS preference */ - conn->le_tx_phy = qos->out.phy; - conn->le_tx_phy = qos->out.phy; + conn->le_tx_phy = qos->bcast.out.phy; + conn->le_tx_phy = qos->bcast.out.phy; conn->iso_qos = *qos; conn->state = BT_BOUND; } @@ -2016,16 +2018,16 @@ static int create_big_sync(struct hci_dev *hdev, void *data) u32 flags = 0; int err; - if (qos->out.phy == 0x02) + if (qos->bcast.out.phy == 0x02) flags |= MGMT_ADV_FLAG_SEC_2M; /* Align intervals */ - interval = qos->out.interval / 1250; + interval = qos->bcast.out.interval / 1250; - if (qos->bis) - sync_interval = qos->sync_interval * 1600; + if (qos->bcast.bis) + sync_interval = qos->bcast.sync_interval * 1600; - err = hci_start_per_adv_sync(hdev, qos->bis, conn->le_per_adv_data_len, + err = hci_start_per_adv_sync(hdev, qos->bcast.bis, conn->le_per_adv_data_len, conn->le_per_adv_data, flags, interval, interval, sync_interval); if (err) @@ -2062,7 +2064,7 @@ static int create_pa_sync(struct hci_dev *hdev, void *data) } int hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst, __u8 dst_type, - __u8 sid) + __u8 sid, struct bt_iso_qos *qos) { struct hci_cp_le_pa_create_sync *cp; @@ -2075,9 +2077,13 @@ int hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst, __u8 dst_type, return -ENOMEM; } + cp->options = qos->bcast.options; cp->sid = sid; cp->addr_type = dst_type; bacpy(&cp->addr, dst); + cp->skip = cpu_to_le16(qos->bcast.skip); + cp->sync_timeout = cpu_to_le16(qos->bcast.sync_timeout); + cp->sync_cte_type = qos->bcast.sync_cte_type; /* Queue start pa_create_sync and scan */ return hci_cmd_sync_queue(hdev, create_pa_sync, cp, create_pa_complete); @@ -2100,8 +2106,12 @@ int hci_le_big_create_sync(struct hci_dev *hdev, struct bt_iso_qos *qos, return err; memset(&pdu, 0, sizeof(pdu)); - pdu.cp.handle = qos->big; + pdu.cp.handle = qos->bcast.big; pdu.cp.sync_handle = cpu_to_le16(sync_handle); + pdu.cp.encryption = qos->bcast.encryption; + memcpy(pdu.cp.bcode, qos->bcast.bcode, sizeof(pdu.cp.bcode)); + pdu.cp.mse = qos->bcast.mse; + pdu.cp.timeout = cpu_to_le16(qos->bcast.timeout); pdu.cp.num_bis = num_bis; memcpy(pdu.bis, bis, num_bis); @@ -2151,7 +2161,7 @@ struct hci_conn *hci_connect_bis(struct hci_dev *hdev, bdaddr_t *dst, return ERR_PTR(err); } - hci_iso_qos_setup(hdev, conn, &qos->out, + hci_iso_qos_setup(hdev, conn, &qos->bcast.out, conn->le_tx_phy ? conn->le_tx_phy : hdev->le_tx_def_phys); @@ -2177,9 +2187,9 @@ struct hci_conn *hci_connect_cis(struct hci_dev *hdev, bdaddr_t *dst, if (IS_ERR(le)) return le; - hci_iso_qos_setup(hdev, le, &qos->out, + hci_iso_qos_setup(hdev, le, &qos->ucast.out, le->le_tx_phy ? le->le_tx_phy : hdev->le_tx_def_phys); - hci_iso_qos_setup(hdev, le, &qos->in, + hci_iso_qos_setup(hdev, le, &qos->ucast.in, le->le_rx_phy ? le->le_rx_phy : hdev->le_rx_def_phys); cis = hci_bind_cis(hdev, dst, dst_type, qos); diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 51f13518dba9..0e0a93cc1218 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -1,6 +1,7 @@ /* BlueZ - Bluetooth protocol stack for Linux Copyright (c) 2000-2001, 2010, Code Aurora Forum. All rights reserved. + Copyright 2023 NXP Written 2000,2001 by Maxim Krasnyansky @@ -3833,7 +3834,7 @@ static u8 hci_cc_le_set_cig_params(struct hci_dev *hdev, void *data, rcu_read_lock(); list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) { - if (conn->type != ISO_LINK || conn->iso_qos.cig != rp->cig_id || + if (conn->type != ISO_LINK || conn->iso_qos.ucast.cig != rp->cig_id || conn->state == BT_CONNECTED) continue; @@ -3890,7 +3891,7 @@ static u8 hci_cc_le_setup_iso_path(struct hci_dev *hdev, void *data, /* Input (Host to Controller) */ case 0x00: /* Only confirm connection if output only */ - if (conn->iso_qos.out.sdu && !conn->iso_qos.in.sdu) + if (conn->iso_qos.ucast.out.sdu && !conn->iso_qos.ucast.in.sdu) hci_connect_cfm(conn, rp->status); break; /* Output (Controller to Host) */ @@ -6818,15 +6819,15 @@ static void hci_le_cis_estabilished_evt(struct hci_dev *hdev, void *data, memset(&interval, 0, sizeof(interval)); memcpy(&interval, ev->c_latency, sizeof(ev->c_latency)); - conn->iso_qos.in.interval = le32_to_cpu(interval); + conn->iso_qos.ucast.in.interval = le32_to_cpu(interval); memcpy(&interval, ev->p_latency, sizeof(ev->p_latency)); - conn->iso_qos.out.interval = le32_to_cpu(interval); - conn->iso_qos.in.latency = le16_to_cpu(ev->interval); - conn->iso_qos.out.latency = le16_to_cpu(ev->interval); - conn->iso_qos.in.sdu = le16_to_cpu(ev->c_mtu); - conn->iso_qos.out.sdu = le16_to_cpu(ev->p_mtu); - conn->iso_qos.in.phy = ev->c_phy; - conn->iso_qos.out.phy = ev->p_phy; + conn->iso_qos.ucast.out.interval = le32_to_cpu(interval); + conn->iso_qos.ucast.in.latency = le16_to_cpu(ev->interval); + conn->iso_qos.ucast.out.latency = le16_to_cpu(ev->interval); + conn->iso_qos.ucast.in.sdu = le16_to_cpu(ev->c_mtu); + conn->iso_qos.ucast.out.sdu = le16_to_cpu(ev->p_mtu); + conn->iso_qos.ucast.in.phy = ev->c_phy; + conn->iso_qos.ucast.out.phy = ev->p_phy; } if (!ev->status) { @@ -6900,8 +6901,8 @@ static void hci_le_cis_req_evt(struct hci_dev *hdev, void *data, cis->handle = cis_handle; } - cis->iso_qos.cig = ev->cig_id; - cis->iso_qos.cis = ev->cis_id; + cis->iso_qos.ucast.cig = ev->cig_id; + cis->iso_qos.ucast.cis = ev->cis_id; if (!(flags & HCI_PROTO_DEFER)) { hci_le_accept_cis(hdev, ev->cis_handle); @@ -6988,13 +6989,13 @@ static void hci_le_big_sync_established_evt(struct hci_dev *hdev, void *data, bis->handle = handle; } - bis->iso_qos.big = ev->handle; + bis->iso_qos.bcast.big = ev->handle; memset(&interval, 0, sizeof(interval)); memcpy(&interval, ev->latency, sizeof(ev->latency)); - bis->iso_qos.in.interval = le32_to_cpu(interval); + bis->iso_qos.bcast.in.interval = le32_to_cpu(interval); /* Convert ISO Interval (1.25 ms slots) to latency (ms) */ - bis->iso_qos.in.latency = le16_to_cpu(ev->interval) * 125 / 100; - bis->iso_qos.in.sdu = le16_to_cpu(ev->max_pdu); + bis->iso_qos.bcast.in.latency = le16_to_cpu(ev->interval) * 125 / 100; + bis->iso_qos.bcast.in.sdu = le16_to_cpu(ev->max_pdu); hci_iso_setup_path(bis); } diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 8d136a730163..74117df03a3f 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -3,6 +3,7 @@ * BlueZ - Bluetooth protocol stack for Linux * * Copyright (C) 2022 Intel Corporation + * Copyright 2023 NXP */ #include @@ -59,11 +60,17 @@ struct iso_pinfo { __u16 sync_handle; __u32 flags; struct bt_iso_qos qos; + bool qos_user_set; __u8 base_len; __u8 base[BASE_MAX_LENGTH]; struct iso_conn *conn; }; +static struct bt_iso_qos default_qos; + +static bool check_ucast_qos(struct bt_iso_qos *qos); +static bool check_bcast_qos(struct bt_iso_qos *qos); + /* ---- ISO timers ---- */ #define ISO_CONN_TIMEOUT (HZ * 40) #define ISO_DISCONN_TIMEOUT (HZ * 2) @@ -264,8 +271,15 @@ static int iso_connect_bis(struct sock *sk) goto unlock; } + /* Fail if user set invalid QoS */ + if (iso_pi(sk)->qos_user_set && !check_bcast_qos(&iso_pi(sk)->qos)) { + iso_pi(sk)->qos = default_qos; + err = -EINVAL; + goto unlock; + } + /* Fail if out PHYs are marked as disabled */ - if (!iso_pi(sk)->qos.out.phy) { + if (!iso_pi(sk)->qos.bcast.out.phy) { err = -EINVAL; goto unlock; } @@ -336,8 +350,15 @@ static int iso_connect_cis(struct sock *sk) goto unlock; } + /* Fail if user set invalid QoS */ + if (iso_pi(sk)->qos_user_set && !check_ucast_qos(&iso_pi(sk)->qos)) { + iso_pi(sk)->qos = default_qos; + err = -EINVAL; + goto unlock; + } + /* Fail if either PHYs are marked as disabled */ - if (!iso_pi(sk)->qos.in.phy && !iso_pi(sk)->qos.out.phy) { + if (!iso_pi(sk)->qos.ucast.in.phy && !iso_pi(sk)->qos.ucast.out.phy) { err = -EINVAL; goto unlock; } @@ -417,7 +438,7 @@ static int iso_send_frame(struct sock *sk, struct sk_buff *skb) BT_DBG("sk %p len %d", sk, skb->len); - if (skb->len > qos->out.sdu) + if (skb->len > qos->ucast.out.sdu) return -EMSGSIZE; len = skb->len; @@ -680,13 +701,23 @@ static struct proto iso_proto = { } static struct bt_iso_qos default_qos = { - .cig = BT_ISO_QOS_CIG_UNSET, - .cis = BT_ISO_QOS_CIS_UNSET, - .sca = 0x00, - .packing = 0x00, - .framing = 0x00, - .in = DEFAULT_IO_QOS, - .out = DEFAULT_IO_QOS, + .bcast = { + .big = BT_ISO_QOS_BIG_UNSET, + .bis = BT_ISO_QOS_BIS_UNSET, + .sync_interval = 0x00, + .packing = 0x00, + .framing = 0x00, + .in = DEFAULT_IO_QOS, + .out = DEFAULT_IO_QOS, + .encryption = 0x00, + .bcode = {0x00}, + .options = 0x00, + .skip = 0x0000, + .sync_timeout = 0x4000, + .sync_cte_type = 0x00, + .mse = 0x00, + .timeout = 0x4000, + }, }; static struct sock *iso_sock_alloc(struct net *net, struct socket *sock, @@ -893,9 +924,15 @@ static int iso_listen_bis(struct sock *sk) if (!hdev) return -EHOSTUNREACH; + /* Fail if user set invalid QoS */ + if (iso_pi(sk)->qos_user_set && !check_bcast_qos(&iso_pi(sk)->qos)) { + iso_pi(sk)->qos = default_qos; + return -EINVAL; + } + err = hci_pa_create_sync(hdev, &iso_pi(sk)->dst, le_addr_type(iso_pi(sk)->dst_type), - iso_pi(sk)->bc_sid); + iso_pi(sk)->bc_sid, &iso_pi(sk)->qos); hci_dev_put(hdev); @@ -1154,21 +1191,62 @@ static bool check_io_qos(struct bt_iso_io_qos *qos) return true; } -static bool check_qos(struct bt_iso_qos *qos) +static bool check_ucast_qos(struct bt_iso_qos *qos) { - if (qos->sca > 0x07) + if (qos->ucast.sca > 0x07) return false; - if (qos->packing > 0x01) + if (qos->ucast.packing > 0x01) return false; - if (qos->framing > 0x01) + if (qos->ucast.framing > 0x01) return false; - if (!check_io_qos(&qos->in)) + if (!check_io_qos(&qos->ucast.in)) return false; - if (!check_io_qos(&qos->out)) + if (!check_io_qos(&qos->ucast.out)) + return false; + + return true; +} + +static bool check_bcast_qos(struct bt_iso_qos *qos) +{ + if (qos->bcast.sync_interval > 0x07) + return false; + + if (qos->bcast.packing > 0x01) + return false; + + if (qos->bcast.framing > 0x01) + return false; + + if (!check_io_qos(&qos->bcast.in)) + return false; + + if (!check_io_qos(&qos->bcast.out)) + return false; + + if (qos->bcast.encryption > 0x01) + return false; + + if (qos->bcast.options > 0x07) + return false; + + if (qos->bcast.skip > 0x01f3) + return false; + + if (qos->bcast.sync_timeout < 0x000a || qos->bcast.sync_timeout > 0x4000) + return false; + + if (qos->bcast.sync_cte_type > 0x1f) + return false; + + if (qos->bcast.mse > 0x1f) + return false; + + if (qos->bcast.timeout < 0x000a || qos->bcast.timeout > 0x4000) return false; return true; @@ -1179,7 +1257,7 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname, { struct sock *sk = sock->sk; int len, err = 0; - struct bt_iso_qos qos; + struct bt_iso_qos qos = default_qos; u32 opt; BT_DBG("sk %p", sk); @@ -1212,24 +1290,19 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname, } len = min_t(unsigned int, sizeof(qos), optlen); - if (len != sizeof(qos)) { - err = -EINVAL; - break; - } - - memset(&qos, 0, sizeof(qos)); if (copy_from_sockptr(&qos, optval, len)) { err = -EFAULT; break; } - if (!check_qos(&qos)) { + if (len == sizeof(qos.ucast) && !check_ucast_qos(&qos)) { err = -EINVAL; break; } iso_pi(sk)->qos = qos; + iso_pi(sk)->qos_user_set = true; break; @@ -1419,7 +1492,7 @@ static bool iso_match_big(struct sock *sk, void *data) { struct hci_evt_le_big_sync_estabilished *ev = data; - return ev->handle == iso_pi(sk)->qos.big; + return ev->handle == iso_pi(sk)->qos.bcast.big; } static void iso_conn_ready(struct iso_conn *conn) -- cgit v1.2.3 From 288c90224eec55d13e786844b7954ef060752089 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 19 Dec 2022 13:37:02 -0800 Subject: Bluetooth: Enable all supported LE PHY by default This enables 2M and Coded PHY by default if they are marked as supported in the LE features bits. Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 4 ++++ net/bluetooth/hci_sync.c | 28 ++++++++++++++++++++++++---- net/bluetooth/mgmt.c | 4 ++-- 3 files changed, 30 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index f11689284112..827e67159523 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1683,9 +1683,13 @@ void hci_conn_del_sysfs(struct hci_conn *conn); #define scan_1m(dev) (((dev)->le_tx_def_phys & HCI_LE_SET_PHY_1M) || \ ((dev)->le_rx_def_phys & HCI_LE_SET_PHY_1M)) +#define le_2m_capable(dev) (((dev)->le_features[1] & HCI_LE_PHY_2M)) + #define scan_2m(dev) (((dev)->le_tx_def_phys & HCI_LE_SET_PHY_2M) || \ ((dev)->le_rx_def_phys & HCI_LE_SET_PHY_2M)) +#define le_coded_capable(dev) (((dev)->le_features[1] & HCI_LE_PHY_CODED)) + #define scan_coded(dev) (((dev)->le_tx_def_phys & HCI_LE_SET_PHY_CODED) || \ ((dev)->le_rx_def_phys & HCI_LE_SET_PHY_CODED)) diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index f21497ebc814..00017f75cd41 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -4414,18 +4414,38 @@ static int hci_le_set_write_def_data_len_sync(struct hci_dev *hdev) sizeof(cp), &cp, HCI_CMD_TIMEOUT); } -/* Set Default PHY parameters if command is supported */ +/* Set Default PHY parameters if command is supported, enables all supported + * PHYs according to the LE Features bits. + */ static int hci_le_set_default_phy_sync(struct hci_dev *hdev) { struct hci_cp_le_set_default_phy cp; - if (!(hdev->commands[35] & 0x20)) + if (!(hdev->commands[35] & 0x20)) { + /* If the command is not supported it means only 1M PHY is + * supported. + */ + hdev->le_tx_def_phys = HCI_LE_SET_PHY_1M; + hdev->le_rx_def_phys = HCI_LE_SET_PHY_1M; return 0; + } memset(&cp, 0, sizeof(cp)); cp.all_phys = 0x00; - cp.tx_phys = hdev->le_tx_def_phys; - cp.rx_phys = hdev->le_rx_def_phys; + cp.tx_phys = HCI_LE_SET_PHY_1M; + cp.rx_phys = HCI_LE_SET_PHY_1M; + + /* Enables 2M PHY if supported */ + if (le_2m_capable(hdev)) { + cp.tx_phys |= HCI_LE_SET_PHY_2M; + cp.rx_phys |= HCI_LE_SET_PHY_2M; + } + + /* Enables Coded PHY if supported */ + if (le_coded_capable(hdev)) { + cp.tx_phys |= HCI_LE_SET_PHY_CODED; + cp.rx_phys |= HCI_LE_SET_PHY_CODED; + } return __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_DEFAULT_PHY, sizeof(cp), &cp, HCI_CMD_TIMEOUT); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 249dc6777fb4..5f8c144c84b8 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -8393,10 +8393,10 @@ static u32 get_supported_adv_flags(struct hci_dev *hdev) flags |= MGMT_ADV_FLAG_HW_OFFLOAD; flags |= MGMT_ADV_FLAG_CAN_SET_TX_POWER; - if (hdev->le_features[1] & HCI_LE_PHY_2M) + if (le_2m_capable(hdev)) flags |= MGMT_ADV_FLAG_SEC_2M; - if (hdev->le_features[1] & HCI_LE_PHY_CODED) + if (le_coded_capable(hdev)) flags |= MGMT_ADV_FLAG_SEC_CODED; } -- cgit v1.2.3 From 3c690a0d64f5964977050a91b6bba8de09c13d94 Mon Sep 17 00:00:00 2001 From: Lanzhe Li Date: Sun, 9 Apr 2023 21:02:29 +0800 Subject: Bluetooth: fix inconsistent indenting Fixed a wrong indentation before "return".This line uses a 7 space indent instead of a tab. Signed-off-by: Lanzhe Li Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_debugfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_debugfs.c b/net/bluetooth/hci_debugfs.c index b7f682922a16..ec0df2f9188e 100644 --- a/net/bluetooth/hci_debugfs.c +++ b/net/bluetooth/hci_debugfs.c @@ -189,7 +189,7 @@ static int uuids_show(struct seq_file *f, void *p) } hci_dev_unlock(hdev); - return 0; + return 0; } DEFINE_SHOW_ATTRIBUTE(uuids); -- cgit v1.2.3 From 9e3c2ea67a4c24f7a7517e863e3adc89c4e3b66d Mon Sep 17 00:00:00 2001 From: Iulia Tanasescu Date: Tue, 11 Apr 2023 10:41:35 +0300 Subject: Bluetooth: hci_conn: remove extra line in hci_le_big_create_sync Remove extra line setting the broadcast code parameter of the hci_cp_le_create_big struct to 0. The broadcast code is copied from the QoS struct. Signed-off-by: Iulia Tanasescu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 5672b4924572..01e0b7255174 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1700,7 +1700,6 @@ static int hci_le_create_big(struct hci_conn *conn, struct bt_iso_qos *qos) cp.bis.framing = qos->bcast.framing; cp.bis.encryption = qos->bcast.encryption; memcpy(cp.bis.bcode, qos->bcast.bcode, sizeof(cp.bis.bcode)); - memset(&cp.bis.bcode, 0, sizeof(cp.bis.bcode)); return hci_send_cmd(hdev, HCI_OP_LE_CREATE_BIG, sizeof(cp), &cp); } -- cgit v1.2.3 From 06149746e7203d5ffe2d6faf9799ee36203aa8b8 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Tue, 11 Apr 2023 16:02:22 -0700 Subject: Bluetooth: hci_conn: Add support for linking multiple hcon Since it is required for some configurations to have multiple CIS with the same peer which is now covered by iso-tester in the following test cases: ISO AC 6(i) - Success ISO AC 7(i) - Success ISO AC 8(i) - Success ISO AC 9(i) - Success ISO AC 11(i) - Success Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 14 +++- net/bluetooth/hci_conn.c | 155 ++++++++++++++++++++++++++++----------- net/bluetooth/hci_event.c | 92 +++++++++++------------ net/bluetooth/iso.c | 8 +- 4 files changed, 172 insertions(+), 97 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 827e67159523..4fe1e71cb9d8 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -770,7 +770,10 @@ struct hci_conn { void *iso_data; struct amp_mgr *amp_mgr; - struct hci_conn *link; + struct list_head link_list; + struct hci_conn *parent; + struct hci_link *link; + struct bt_codec codec; void (*connect_cfm_cb) (struct hci_conn *conn, u8 status); @@ -780,6 +783,11 @@ struct hci_conn { void (*cleanup)(struct hci_conn *conn); }; +struct hci_link { + struct list_head list; + struct hci_conn *conn; +}; + struct hci_chan { struct list_head list; __u16 handle; @@ -1383,12 +1391,14 @@ static inline void hci_conn_put(struct hci_conn *conn) put_device(&conn->dev); } -static inline void hci_conn_hold(struct hci_conn *conn) +static inline struct hci_conn *hci_conn_hold(struct hci_conn *conn) { BT_DBG("hcon %p orig refcnt %d", conn, atomic_read(&conn->refcnt)); atomic_inc(&conn->refcnt); cancel_delayed_work(&conn->disc_work); + + return conn; } static inline void hci_conn_drop(struct hci_conn *conn) diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 01e0b7255174..d8466abbb36a 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -330,8 +330,11 @@ static void hci_add_sco(struct hci_conn *conn, __u16 handle) static bool find_next_esco_param(struct hci_conn *conn, const struct sco_param *esco_param, int size) { + if (!conn->parent) + return false; + for (; conn->attempt <= size; conn->attempt++) { - if (lmp_esco_2m_capable(conn->link) || + if (lmp_esco_2m_capable(conn->parent) || (esco_param[conn->attempt - 1].pkt_type & ESCO_2EV3)) break; BT_DBG("hcon %p skipped attempt %d, eSCO 2M not supported", @@ -461,7 +464,7 @@ static int hci_enhanced_setup_sync(struct hci_dev *hdev, void *data) break; case BT_CODEC_CVSD: - if (lmp_esco_capable(conn->link)) { + if (conn->parent && lmp_esco_capable(conn->parent)) { if (!find_next_esco_param(conn, esco_param_cvsd, ARRAY_SIZE(esco_param_cvsd))) return -EINVAL; @@ -531,7 +534,7 @@ static bool hci_setup_sync_conn(struct hci_conn *conn, __u16 handle) param = &esco_param_msbc[conn->attempt - 1]; break; case SCO_AIRMODE_CVSD: - if (lmp_esco_capable(conn->link)) { + if (conn->parent && lmp_esco_capable(conn->parent)) { if (!find_next_esco_param(conn, esco_param_cvsd, ARRAY_SIZE(esco_param_cvsd))) return false; @@ -637,21 +640,22 @@ void hci_le_start_enc(struct hci_conn *conn, __le16 ediv, __le64 rand, /* Device _must_ be locked */ void hci_sco_setup(struct hci_conn *conn, __u8 status) { - struct hci_conn *sco = conn->link; + struct hci_link *link; - if (!sco) + link = list_first_entry_or_null(&conn->link_list, struct hci_link, list); + if (!link || !link->conn) return; BT_DBG("hcon %p", conn); if (!status) { if (lmp_esco_capable(conn->hdev)) - hci_setup_sync(sco, conn->handle); + hci_setup_sync(link->conn, conn->handle); else - hci_add_sco(sco, conn->handle); + hci_add_sco(link->conn, conn->handle); } else { - hci_connect_cfm(sco, status); - hci_conn_del(sco); + hci_connect_cfm(link->conn, status); + hci_conn_del(link->conn); } } @@ -1042,6 +1046,7 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, skb_queue_head_init(&conn->data_q); INIT_LIST_HEAD(&conn->chan_list); + INIT_LIST_HEAD(&conn->link_list); INIT_DELAYED_WORK(&conn->disc_work, hci_conn_timeout); INIT_DELAYED_WORK(&conn->auto_accept_work, hci_conn_auto_accept); @@ -1069,15 +1074,39 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, return conn; } -static bool hci_conn_unlink(struct hci_conn *conn) +static void hci_conn_unlink(struct hci_conn *conn) { + struct hci_dev *hdev = conn->hdev; + + bt_dev_dbg(hdev, "hcon %p", conn); + + if (!conn->parent) { + struct hci_link *link, *t; + + list_for_each_entry_safe(link, t, &conn->link_list, list) + hci_conn_unlink(link->conn); + + return; + } + if (!conn->link) - return false; + return; + + hci_conn_put(conn->parent); + conn->parent = NULL; - conn->link->link = NULL; + list_del_rcu(&conn->link->list); + synchronize_rcu(); + + kfree(conn->link); conn->link = NULL; - return true; + /* Due to race, SCO connection might be not established + * yet at this point. Delete it now, otherwise it is + * possible for it to be stuck and can't be deleted. + */ + if (conn->handle == HCI_CONN_HANDLE_UNSET) + hci_conn_del(conn); } int hci_conn_del(struct hci_conn *conn) @@ -1091,18 +1120,7 @@ int hci_conn_del(struct hci_conn *conn) cancel_delayed_work_sync(&conn->idle_work); if (conn->type == ACL_LINK) { - struct hci_conn *link = conn->link; - - if (link) { - hci_conn_unlink(conn); - /* Due to race, SCO connection might be not established - * yet at this point. Delete it now, otherwise it is - * possible for it to be stuck and can't be deleted. - */ - if (link->handle == HCI_CONN_HANDLE_UNSET) - hci_conn_del(link); - } - + hci_conn_unlink(conn); /* Unacked frames */ hdev->acl_cnt += conn->sent; } else if (conn->type == LE_LINK) { @@ -1113,7 +1131,7 @@ int hci_conn_del(struct hci_conn *conn) else hdev->acl_cnt += conn->sent; } else { - struct hci_conn *acl = conn->link; + struct hci_conn *acl = conn->parent; if (acl) { hci_conn_unlink(conn); @@ -1600,11 +1618,40 @@ struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst, return acl; } +static struct hci_link *hci_conn_link(struct hci_conn *parent, + struct hci_conn *conn) +{ + struct hci_dev *hdev = parent->hdev; + struct hci_link *link; + + bt_dev_dbg(hdev, "parent %p hcon %p", parent, conn); + + if (conn->link) + return conn->link; + + if (conn->parent) + return NULL; + + link = kzalloc(sizeof(*link), GFP_KERNEL); + if (!link) + return NULL; + + link->conn = hci_conn_hold(conn); + conn->link = link; + conn->parent = hci_conn_get(parent); + + /* Use list_add_tail_rcu append to the list */ + list_add_tail_rcu(&link->list, &parent->link_list); + + return link; +} + struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, bdaddr_t *dst, __u16 setting, struct bt_codec *codec) { struct hci_conn *acl; struct hci_conn *sco; + struct hci_link *link; acl = hci_connect_acl(hdev, dst, BT_SECURITY_LOW, HCI_AT_NO_BONDING, CONN_REASON_SCO_CONNECT); @@ -1620,10 +1667,12 @@ struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, bdaddr_t *dst, } } - acl->link = sco; - sco->link = acl; - - hci_conn_hold(sco); + link = hci_conn_link(acl, sco); + if (!link) { + hci_conn_drop(acl); + hci_conn_drop(sco); + return NULL; + } sco->setting = setting; sco->codec = *codec; @@ -1890,7 +1939,7 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data) u8 cig; memset(&cmd, 0, sizeof(cmd)); - cmd.cis[0].acl_handle = cpu_to_le16(conn->link->handle); + cmd.cis[0].acl_handle = cpu_to_le16(conn->parent->handle); cmd.cis[0].cis_handle = cpu_to_le16(conn->handle); cmd.cp.num_cis++; cig = conn->iso_qos.ucast.cig; @@ -1903,11 +1952,12 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data) struct hci_cis *cis = &cmd.cis[cmd.cp.num_cis]; if (conn == data || conn->type != ISO_LINK || - conn->state == BT_CONNECTED || conn->iso_qos.ucast.cig != cig) + conn->state == BT_CONNECTED || + conn->iso_qos.ucast.cig != cig) continue; /* Check if all CIS(s) belonging to a CIG are ready */ - if (!conn->link || conn->link->state != BT_CONNECTED || + if (!conn->parent || conn->parent->state != BT_CONNECTED || conn->state != BT_CONNECT) { cmd.cp.num_cis = 0; break; @@ -1924,7 +1974,7 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data) * command have been generated, the Controller shall return the * error code Command Disallowed (0x0C). */ - cis->acl_handle = cpu_to_le16(conn->link->handle); + cis->acl_handle = cpu_to_le16(conn->parent->handle); cis->cis_handle = cpu_to_le16(conn->handle); cmd.cp.num_cis++; } @@ -1943,15 +1993,33 @@ static int hci_create_cis_sync(struct hci_dev *hdev, void *data) int hci_le_create_cis(struct hci_conn *conn) { struct hci_conn *cis; + struct hci_link *link, *t; struct hci_dev *hdev = conn->hdev; int err; + bt_dev_dbg(hdev, "hcon %p", conn); + switch (conn->type) { case LE_LINK: - if (!conn->link || conn->state != BT_CONNECTED) + if (conn->state != BT_CONNECTED || list_empty(&conn->link_list)) return -EINVAL; - cis = conn->link; - break; + + cis = NULL; + + /* hci_conn_link uses list_add_tail_rcu so the list is in + * the same order as the connections are requested. + */ + list_for_each_entry_safe(link, t, &conn->link_list, list) { + if (link->conn->state == BT_BOUND) { + err = hci_le_create_cis(link->conn); + if (err) + return err; + + cis = link->conn; + } + } + + return cis ? 0 : -EINVAL; case ISO_LINK: cis = conn; break; @@ -2172,6 +2240,7 @@ struct hci_conn *hci_connect_cis(struct hci_dev *hdev, bdaddr_t *dst, { struct hci_conn *le; struct hci_conn *cis; + struct hci_link *link; if (hci_dev_test_flag(hdev, HCI_ADVERTISING)) le = hci_connect_le(hdev, dst, dst_type, false, @@ -2197,16 +2266,18 @@ struct hci_conn *hci_connect_cis(struct hci_dev *hdev, bdaddr_t *dst, return cis; } - le->link = cis; - cis->link = le; - - hci_conn_hold(cis); + link = hci_conn_link(le, cis); + if (!link) { + hci_conn_drop(le); + hci_conn_drop(cis); + return NULL; + } /* If LE is already connected and CIS handle is already set proceed to * Create CIS immediately. */ if (le->state == BT_CONNECTED && cis->handle != HCI_CONN_HANDLE_UNSET) - hci_le_create_cis(le); + hci_le_create_cis(cis); return cis; } diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 0e0a93cc1218..d00ef6e3fc45 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -2345,7 +2345,8 @@ static void hci_cs_create_conn(struct hci_dev *hdev, __u8 status) static void hci_cs_add_sco(struct hci_dev *hdev, __u8 status) { struct hci_cp_add_sco *cp; - struct hci_conn *acl, *sco; + struct hci_conn *acl; + struct hci_link *link; __u16 handle; bt_dev_dbg(hdev, "status 0x%2.2x", status); @@ -2365,12 +2366,13 @@ static void hci_cs_add_sco(struct hci_dev *hdev, __u8 status) acl = hci_conn_hash_lookup_handle(hdev, handle); if (acl) { - sco = acl->link; - if (sco) { - sco->state = BT_CLOSED; + link = list_first_entry_or_null(&acl->link_list, + struct hci_link, list); + if (link && link->conn) { + link->conn->state = BT_CLOSED; - hci_connect_cfm(sco, status); - hci_conn_del(sco); + hci_connect_cfm(link->conn, status); + hci_conn_del(link->conn); } } @@ -2637,74 +2639,61 @@ static void hci_cs_read_remote_ext_features(struct hci_dev *hdev, __u8 status) hci_dev_unlock(hdev); } -static void hci_cs_setup_sync_conn(struct hci_dev *hdev, __u8 status) +static void hci_setup_sync_conn_status(struct hci_dev *hdev, __u16 handle, + __u8 status) { - struct hci_cp_setup_sync_conn *cp; - struct hci_conn *acl, *sco; - __u16 handle; - - bt_dev_dbg(hdev, "status 0x%2.2x", status); - - if (!status) - return; - - cp = hci_sent_cmd_data(hdev, HCI_OP_SETUP_SYNC_CONN); - if (!cp) - return; + struct hci_conn *acl; + struct hci_link *link; - handle = __le16_to_cpu(cp->handle); - - bt_dev_dbg(hdev, "handle 0x%4.4x", handle); + bt_dev_dbg(hdev, "handle 0x%4.4x status 0x%2.2x", handle, status); hci_dev_lock(hdev); acl = hci_conn_hash_lookup_handle(hdev, handle); if (acl) { - sco = acl->link; - if (sco) { - sco->state = BT_CLOSED; + link = list_first_entry_or_null(&acl->link_list, + struct hci_link, list); + if (link && link->conn) { + link->conn->state = BT_CLOSED; - hci_connect_cfm(sco, status); - hci_conn_del(sco); + hci_connect_cfm(link->conn, status); + hci_conn_del(link->conn); } } hci_dev_unlock(hdev); } -static void hci_cs_enhanced_setup_sync_conn(struct hci_dev *hdev, __u8 status) +static void hci_cs_setup_sync_conn(struct hci_dev *hdev, __u8 status) { - struct hci_cp_enhanced_setup_sync_conn *cp; - struct hci_conn *acl, *sco; - __u16 handle; + struct hci_cp_setup_sync_conn *cp; bt_dev_dbg(hdev, "status 0x%2.2x", status); if (!status) return; - cp = hci_sent_cmd_data(hdev, HCI_OP_ENHANCED_SETUP_SYNC_CONN); + cp = hci_sent_cmd_data(hdev, HCI_OP_SETUP_SYNC_CONN); if (!cp) return; - handle = __le16_to_cpu(cp->handle); + hci_setup_sync_conn_status(hdev, __le16_to_cpu(cp->handle), status); +} - bt_dev_dbg(hdev, "handle 0x%4.4x", handle); +static void hci_cs_enhanced_setup_sync_conn(struct hci_dev *hdev, __u8 status) +{ + struct hci_cp_enhanced_setup_sync_conn *cp; - hci_dev_lock(hdev); + bt_dev_dbg(hdev, "status 0x%2.2x", status); - acl = hci_conn_hash_lookup_handle(hdev, handle); - if (acl) { - sco = acl->link; - if (sco) { - sco->state = BT_CLOSED; + if (!status) + return; - hci_connect_cfm(sco, status); - hci_conn_del(sco); - } - } + cp = hci_sent_cmd_data(hdev, HCI_OP_ENHANCED_SETUP_SYNC_CONN); + if (!cp) + return; - hci_dev_unlock(hdev); + hci_setup_sync_conn_status(hdev, __le16_to_cpu(cp->handle), status); } static void hci_cs_sniff_mode(struct hci_dev *hdev, __u8 status) @@ -3834,19 +3823,20 @@ static u8 hci_cc_le_set_cig_params(struct hci_dev *hdev, void *data, rcu_read_lock(); list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) { - if (conn->type != ISO_LINK || conn->iso_qos.ucast.cig != rp->cig_id || + if (conn->type != ISO_LINK || + conn->iso_qos.ucast.cig != rp->cig_id || conn->state == BT_CONNECTED) continue; conn->handle = __le16_to_cpu(rp->handle[i++]); - bt_dev_dbg(hdev, "%p handle 0x%4.4x link %p", conn, - conn->handle, conn->link); + bt_dev_dbg(hdev, "%p handle 0x%4.4x parent %p", conn, + conn->handle, conn->parent); /* Create CIS if LE is already connected */ - if (conn->link && conn->link->state == BT_CONNECTED) { + if (conn->parent && conn->parent->state == BT_CONNECTED) { rcu_read_unlock(); - hci_le_create_cis(conn->link); + hci_le_create_cis(conn); rcu_read_lock(); } @@ -5031,7 +5021,7 @@ static void hci_sync_conn_complete_evt(struct hci_dev *hdev, void *data, if (conn->out) { conn->pkt_type = (hdev->esco_type & SCO_ESCO_MASK) | (hdev->esco_type & EDR_ESCO_MASK); - if (hci_setup_sync(conn, conn->link->handle)) + if (hci_setup_sync(conn, conn->parent->handle)) goto unlock; } fallthrough; diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 74117df03a3f..34d55a85d8f6 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -1657,8 +1657,12 @@ static void iso_connect_cfm(struct hci_conn *hcon, __u8 status) /* Check if LE link has failed */ if (status) { - if (hcon->link) - iso_conn_del(hcon->link, bt_to_errno(status)); + struct hci_link *link, *t; + + list_for_each_entry_safe(link, t, &hcon->link_list, + list) + iso_conn_del(link->conn, bt_to_errno(status)); + return; } -- cgit v1.2.3 From c14516faede33c2c31da45cf950d55dbff42962e Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Tue, 11 Apr 2023 16:14:25 -0700 Subject: Bluetooth: hci_conn: Fix not matching by CIS ID This fixes only matching CIS by address which prevents creating new hcon if upper layer is requesting a specific CIS ID. Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 12 +++++++++++- net/bluetooth/hci_conn.c | 3 ++- 2 files changed, 13 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 4fe1e71cb9d8..a6c8aee2f256 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1180,7 +1180,9 @@ static inline struct hci_conn *hci_conn_hash_lookup_le(struct hci_dev *hdev, static inline struct hci_conn *hci_conn_hash_lookup_cis(struct hci_dev *hdev, bdaddr_t *ba, - __u8 ba_type) + __u8 ba_type, + __u8 cig, + __u8 id) { struct hci_conn_hash *h = &hdev->conn_hash; struct hci_conn *c; @@ -1191,6 +1193,14 @@ static inline struct hci_conn *hci_conn_hash_lookup_cis(struct hci_dev *hdev, if (c->type != ISO_LINK) continue; + /* Match CIG ID if set */ + if (cig != BT_ISO_QOS_CIG_UNSET && cig != c->iso_qos.ucast.cig) + continue; + + /* Match CIS ID if set */ + if (id != BT_ISO_QOS_CIS_UNSET && id != c->iso_qos.ucast.cis) + continue; + if (ba_type == c->dst_type && !bacmp(&c->dst, ba)) { rcu_read_unlock(); return c; diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index d8466abbb36a..c215e983e287 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1841,7 +1841,8 @@ struct hci_conn *hci_bind_cis(struct hci_dev *hdev, bdaddr_t *dst, { struct hci_conn *cis; - cis = hci_conn_hash_lookup_cis(hdev, dst, dst_type); + cis = hci_conn_hash_lookup_cis(hdev, dst, dst_type, qos->ucast.cig, + qos->ucast.cis); if (!cis) { cis = hci_conn_add(hdev, ISO_LINK, dst, HCI_ROLE_MASTER); if (!cis) -- cgit v1.2.3 From c09b80be6ffc338634b2f5f8cfa12b6843410834 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Wed, 12 Apr 2023 17:45:51 -0700 Subject: Bluetooth: hci_conn: Fix not waiting for HCI_EVT_LE_CIS_ESTABLISHED When submitting HCI_OP_LE_CREATE_CIS the code shall wait for HCI_EVT_LE_CIS_ESTABLISHED thus enforcing the serialization of HCI_OP_LE_CREATE_CIS as the Core spec does not allow to send them in parallel: BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E page 2566: If the Host issues this command before all the HCI_LE_CIS_Established events from the previous use of the command have been generated, the Controller shall return the error code Command Disallowed (0x0C). Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_sync.h | 2 ++ net/bluetooth/hci_conn.c | 58 +---------------------------------- net/bluetooth/hci_sync.c | 65 ++++++++++++++++++++++++++++++++++++++++ 3 files changed, 68 insertions(+), 57 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h index 17f5a4c32f36..f61b249787fc 100644 --- a/include/net/bluetooth/hci_sync.h +++ b/include/net/bluetooth/hci_sync.h @@ -122,6 +122,8 @@ int hci_abort_conn_sync(struct hci_dev *hdev, struct hci_conn *conn, u8 reason); int hci_le_create_conn_sync(struct hci_dev *hdev, struct hci_conn *conn); +int hci_le_create_cis_sync(struct hci_dev *hdev, struct hci_conn *conn); + int hci_le_remove_cig_sync(struct hci_dev *hdev, u8 handle); int hci_le_terminate_big_sync(struct hci_dev *hdev, u8 handle, u8 reason); diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index c215e983e287..640b951bf40a 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1932,63 +1932,7 @@ bool hci_iso_setup_path(struct hci_conn *conn) static int hci_create_cis_sync(struct hci_dev *hdev, void *data) { - struct { - struct hci_cp_le_create_cis cp; - struct hci_cis cis[0x1f]; - } cmd; - struct hci_conn *conn = data; - u8 cig; - - memset(&cmd, 0, sizeof(cmd)); - cmd.cis[0].acl_handle = cpu_to_le16(conn->parent->handle); - cmd.cis[0].cis_handle = cpu_to_le16(conn->handle); - cmd.cp.num_cis++; - cig = conn->iso_qos.ucast.cig; - - hci_dev_lock(hdev); - - rcu_read_lock(); - - list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) { - struct hci_cis *cis = &cmd.cis[cmd.cp.num_cis]; - - if (conn == data || conn->type != ISO_LINK || - conn->state == BT_CONNECTED || - conn->iso_qos.ucast.cig != cig) - continue; - - /* Check if all CIS(s) belonging to a CIG are ready */ - if (!conn->parent || conn->parent->state != BT_CONNECTED || - conn->state != BT_CONNECT) { - cmd.cp.num_cis = 0; - break; - } - - /* Group all CIS with state BT_CONNECT since the spec don't - * allow to send them individually: - * - * BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E - * page 2566: - * - * If the Host issues this command before all the - * HCI_LE_CIS_Established events from the previous use of the - * command have been generated, the Controller shall return the - * error code Command Disallowed (0x0C). - */ - cis->acl_handle = cpu_to_le16(conn->parent->handle); - cis->cis_handle = cpu_to_le16(conn->handle); - cmd.cp.num_cis++; - } - - rcu_read_unlock(); - - hci_dev_unlock(hdev); - - if (!cmd.cp.num_cis) - return 0; - - return hci_send_cmd(hdev, HCI_OP_LE_CREATE_CIS, sizeof(cmd.cp) + - sizeof(cmd.cis[0]) * cmd.cp.num_cis, &cmd); + return hci_le_create_cis_sync(hdev, data); } int hci_le_create_cis(struct hci_conn *conn) diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 00017f75cd41..6f060d00a70a 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -6137,6 +6137,71 @@ done: return err; } +int hci_le_create_cis_sync(struct hci_dev *hdev, struct hci_conn *conn) +{ + struct { + struct hci_cp_le_create_cis cp; + struct hci_cis cis[0x1f]; + } cmd; + u8 cig; + struct hci_conn *hcon = conn; + + memset(&cmd, 0, sizeof(cmd)); + cmd.cis[0].acl_handle = cpu_to_le16(conn->parent->handle); + cmd.cis[0].cis_handle = cpu_to_le16(conn->handle); + cmd.cp.num_cis++; + cig = conn->iso_qos.ucast.cig; + + hci_dev_lock(hdev); + + rcu_read_lock(); + + list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) { + struct hci_cis *cis = &cmd.cis[cmd.cp.num_cis]; + + if (conn == hcon || conn->type != ISO_LINK || + conn->state == BT_CONNECTED || + conn->iso_qos.ucast.cig != cig) + continue; + + /* Check if all CIS(s) belonging to a CIG are ready */ + if (!conn->parent || conn->parent->state != BT_CONNECTED || + conn->state != BT_CONNECT) { + cmd.cp.num_cis = 0; + break; + } + + /* Group all CIS with state BT_CONNECT since the spec don't + * allow to send them individually: + * + * BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E + * page 2566: + * + * If the Host issues this command before all the + * HCI_LE_CIS_Established events from the previous use of the + * command have been generated, the Controller shall return the + * error code Command Disallowed (0x0C). + */ + cis->acl_handle = cpu_to_le16(conn->parent->handle); + cis->cis_handle = cpu_to_le16(conn->handle); + cmd.cp.num_cis++; + } + + rcu_read_unlock(); + + hci_dev_unlock(hdev); + + if (!cmd.cp.num_cis) + return 0; + + /* Wait for HCI_LE_CIS_Established */ + return __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_CREATE_CIS, + sizeof(cmd.cp) + sizeof(cmd.cis[0]) * + cmd.cp.num_cis, &cmd, + HCI_EVT_LE_CIS_ESTABLISHED, + conn->conn_timeout, NULL); +} + int hci_le_remove_cig_sync(struct hci_dev *hdev, u8 handle) { struct hci_cp_le_remove_cig cp; -- cgit v1.2.3 From 91b6d02ddcd113352bdd895990b252065c596de7 Mon Sep 17 00:00:00 2001 From: Raul Cheleguini Date: Thu, 23 Mar 2023 10:45:39 -0300 Subject: Bluetooth: Add new quirk for broken set random RPA timeout for ATS2851 The ATS2851 based controller advertises support for command "LE Set Random Private Address Timeout" but does not actually implement it, impeding the controller initialization. Add the quirk HCI_QUIRK_BROKEN_SET_RPA_TIMEOUT to unblock the controller initialization. < HCI Command: LE Set Resolvable Private... (0x08|0x002e) plen 2 Timeout: 900 seconds > HCI Event: Command Status (0x0f) plen 4 LE Set Resolvable Private Address Timeout (0x08|0x002e) ncmd 1 Status: Unknown HCI Command (0x01) Co-developed-by: imoc Signed-off-by: imoc Signed-off-by: Raul Cheleguini Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 1 + include/net/bluetooth/hci.h | 8 ++++++++ net/bluetooth/hci_sync.c | 6 +++++- 3 files changed, 14 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 2303b0a66323..3aa189b1986d 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -4149,6 +4149,7 @@ static int btusb_probe(struct usb_interface *intf, /* Support is advertised, but not implemented */ set_bit(HCI_QUIRK_BROKEN_ERR_DATA_REPORTING, &hdev->quirks); set_bit(HCI_QUIRK_BROKEN_READ_TRANSMIT_POWER, &hdev->quirks); + set_bit(HCI_QUIRK_BROKEN_SET_RPA_TIMEOUT, &hdev->quirks); set_bit(HCI_QUIRK_BROKEN_EXT_SCAN, &hdev->quirks); } diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 997107bfc0b1..07df96c47ef4 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -301,6 +301,14 @@ enum { * don't actually support features declared there. */ HCI_QUIRK_BROKEN_LOCAL_EXT_FEATURES_PAGE_2, + + /* + * When this quirk is set, the HCI_OP_LE_SET_RPA_TIMEOUT command is + * skipped during initialization. This is required for the Actions + * Semiconductor ATS2851 based controllers, which erroneously claims + * to support it. + */ + HCI_QUIRK_BROKEN_SET_RPA_TIMEOUT, }; /* HCI device flags */ diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 6f060d00a70a..771aaa808967 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -4093,7 +4093,8 @@ static int hci_le_set_rpa_timeout_sync(struct hci_dev *hdev) { __le16 timeout = cpu_to_le16(hdev->rpa_timeout); - if (!(hdev->commands[35] & 0x04)) + if (!(hdev->commands[35] & 0x04) || + test_bit(HCI_QUIRK_BROKEN_SET_RPA_TIMEOUT, &hdev->quirks)) return 0; return __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_RPA_TIMEOUT, @@ -4553,6 +4554,9 @@ static const struct { "HCI Set Event Filter command not supported."), HCI_QUIRK_BROKEN(ENHANCED_SETUP_SYNC_CONN, "HCI Enhanced Setup Synchronous Connection command is " + "advertised, but not supported."), + HCI_QUIRK_BROKEN(SET_RPA_TIMEOUT, + "HCI LE Set Random Private Address Timeout command is " "advertised, but not supported.") }; -- cgit v1.2.3 From 000c2fa2c144c499c881a101819cf1936a1f7cf2 Mon Sep 17 00:00:00 2001 From: Ruihan Li Date: Sun, 16 Apr 2023 16:02:51 +0800 Subject: bluetooth: Add cmd validity checks at the start of hci_sock_ioctl() Previously, channel open messages were always sent to monitors on the first ioctl() call for unbound HCI sockets, even if the command and arguments were completely invalid. This can leave an exploitable hole with the abuse of invalid ioctl calls. This commit hardens the ioctl processing logic by first checking if the command is valid, and immediately returning with an ENOIOCTLCMD error code if it is not. This ensures that ioctl calls with invalid commands are free of side effects, and increases the difficulty of further exploitation by forcing exploitation to find a way to pass a valid command first. Signed-off-by: Ruihan Li Co-developed-by: Marcel Holtmann Signed-off-by: Marcel Holtmann Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sock.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) (limited to 'net') diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 06581223238c..8d8950101cee 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -987,6 +987,34 @@ static int hci_sock_ioctl(struct socket *sock, unsigned int cmd, BT_DBG("cmd %x arg %lx", cmd, arg); + /* Make sure the cmd is valid before doing anything */ + switch (cmd) { + case HCIGETDEVLIST: + case HCIGETDEVINFO: + case HCIGETCONNLIST: + case HCIDEVUP: + case HCIDEVDOWN: + case HCIDEVRESET: + case HCIDEVRESTAT: + case HCISETSCAN: + case HCISETAUTH: + case HCISETENCRYPT: + case HCISETPTYPE: + case HCISETLINKPOL: + case HCISETLINKMODE: + case HCISETACLMTU: + case HCISETSCOMTU: + case HCIINQUIRY: + case HCISETRAW: + case HCIGETCONNINFO: + case HCIGETAUTHINFO: + case HCIBLOCKADDR: + case HCIUNBLOCKADDR: + break; + default: + return -ENOIOCTLCMD; + } + lock_sock(sk); if (hci_pi(sk)->channel != HCI_CHANNEL_RAW) { -- cgit v1.2.3 From 25e97f7b1866e6b8503be349eeea44bb52d661ce Mon Sep 17 00:00:00 2001 From: Min Li Date: Mon, 17 Apr 2023 10:27:54 +0800 Subject: Bluetooth: L2CAP: fix "bad unlock balance" in l2cap_disconnect_rsp conn->chan_lock isn't acquired before l2cap_get_chan_by_scid, if l2cap_get_chan_by_scid returns NULL, then 'bad unlock balance' is triggered. Reported-by: syzbot+9519d6b5b79cf7787cf3@syzkaller.appspotmail.com Link: https://lore.kernel.org/all/000000000000894f5f05f95e9f4d@google.com/ Signed-off-by: Min Li Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 5cc95fd17f7d..376b523c7b26 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -4693,7 +4693,6 @@ static inline int l2cap_disconnect_rsp(struct l2cap_conn *conn, chan = l2cap_get_chan_by_scid(conn, scid); if (!chan) { - mutex_unlock(&conn->chan_lock); return 0; } -- cgit v1.2.3 From 25c150ac103a4ebeed0319994c742a90634ddf18 Mon Sep 17 00:00:00 2001 From: Ruihan Li Date: Sun, 16 Apr 2023 16:14:04 +0800 Subject: bluetooth: Perform careful capability checks in hci_sock_ioctl() Previously, capability was checked using capable(), which verified that the caller of the ioctl system call had the required capability. In addition, the result of the check would be stored in the HCI_SOCK_TRUSTED flag, making it persistent for the socket. However, malicious programs can abuse this approach by deliberately sharing an HCI socket with a privileged task. The HCI socket will be marked as trusted when the privileged task occasionally makes an ioctl call. This problem can be solved by using sk_capable() to check capability, which ensures that not only the current task but also the socket opener has the specified capability, thus reducing the risk of privilege escalation through the previously identified vulnerability. Cc: stable@vger.kernel.org Fixes: f81f5b2db869 ("Bluetooth: Send control open and close messages for HCI raw sockets") Signed-off-by: Ruihan Li Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sock.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 8d8950101cee..1d249d839819 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -1031,7 +1031,14 @@ static int hci_sock_ioctl(struct socket *sock, unsigned int cmd, if (hci_sock_gen_cookie(sk)) { struct sk_buff *skb; - if (capable(CAP_NET_ADMIN)) + /* Perform careful checks before setting the HCI_SOCK_TRUSTED + * flag. Make sure that not only the current task but also + * the socket opener has the required capability, since + * privileged programs can be tricked into making ioctl calls + * on HCI sockets, and the socket should not be marked as + * trusted simply because the ioctl caller is privileged. + */ + if (sk_capable(sk, CAP_NET_ADMIN)) hci_sock_set_flag(sk, HCI_SOCK_TRUSTED); /* Send event to monitor */ -- cgit v1.2.3 From 0ab905c3be8c2be368821dd2bd1f8d126064d993 Mon Sep 17 00:00:00 2001 From: Zijun Hu Date: Mon, 17 Apr 2023 17:39:59 +0800 Subject: Bluetooth: Devcoredump: Fix storing u32 without specifying byte order issue API hci_devcd_init() stores its u32 type parameter @dump_size into skb, but it does not specify which byte order is used to store the integer, let us take little endian to store and parse the integer. Fixes: f5cc609d09d4 ("Bluetooth: Add support for hci devcoredump") Signed-off-by: Zijun Hu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/coredump.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/bluetooth/coredump.c b/net/bluetooth/coredump.c index 08fa98505454..d2d2624ec708 100644 --- a/net/bluetooth/coredump.c +++ b/net/bluetooth/coredump.c @@ -5,6 +5,7 @@ #include +#include #include #include @@ -180,25 +181,25 @@ hdr_free: static void hci_devcd_handle_pkt_init(struct hci_dev *hdev, struct sk_buff *skb) { - u32 *dump_size; + u32 dump_size; if (hdev->dump.state != HCI_DEVCOREDUMP_IDLE) { DBG_UNEXPECTED_STATE(); return; } - if (skb->len != sizeof(*dump_size)) { + if (skb->len != sizeof(dump_size)) { bt_dev_dbg(hdev, "Invalid dump init pkt"); return; } - dump_size = skb_pull_data(skb, sizeof(*dump_size)); - if (!*dump_size) { + dump_size = get_unaligned_le32(skb_pull_data(skb, 4)); + if (!dump_size) { bt_dev_err(hdev, "Zero size dump init pkt"); return; } - if (hci_devcd_prepare(hdev, *dump_size)) { + if (hci_devcd_prepare(hdev, dump_size)) { bt_dev_err(hdev, "Failed to prepare for dump"); return; } @@ -441,7 +442,7 @@ int hci_devcd_init(struct hci_dev *hdev, u32 dump_size) return -ENOMEM; hci_dmp_cb(skb)->pkt_type = HCI_DEVCOREDUMP_PKT_INIT; - skb_put_data(skb, &dump_size, sizeof(dump_size)); + put_unaligned_le32(dump_size, skb_put(skb, 4)); skb_queue_tail(&hdev->dump.dump_q, skb); queue_work(hdev->workqueue, &hdev->dump.dump_rx); -- cgit v1.2.3 From f419863588217f76eaf754e1dfce21ea7fcb026d Mon Sep 17 00:00:00 2001 From: Archie Pusaka Date: Thu, 20 Apr 2023 20:23:36 +0800 Subject: Bluetooth: Cancel sync command before suspend and power off Some of the sync commands might take a long time to complete, e.g. LE Create Connection when the peer device isn't responding might take 20 seconds before it times out. If suspend command is issued during this time, it will need to wait for completion since both commands are using the same sync lock. This patch cancel any running sync commands before attempting to suspend or adapter power off. Signed-off-by: Archie Pusaka Reviewed-by: Ying Hsu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_core.c | 3 +++ net/bluetooth/mgmt.c | 4 ++++ 2 files changed, 7 insertions(+) (limited to 'net') diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 393b317ae68f..a856b1051d35 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -2803,6 +2803,9 @@ int hci_suspend_dev(struct hci_dev *hdev) if (mgmt_powering_down(hdev)) return 0; + /* Cancel potentially blocking sync operation before suspend */ + __hci_cmd_sync_cancel(hdev, -EHOSTDOWN); + hci_req_sync_lock(hdev); ret = hci_suspend_sync(hdev); hci_req_sync_unlock(hdev); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 5f8c144c84b8..13c745876b39 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -1399,6 +1399,10 @@ static int set_powered(struct sock *sk, struct hci_dev *hdev, void *data, goto failed; } + /* Cancel potentially blocking sync operation before power off */ + if (cp->val == 0x00) + __hci_cmd_sync_cancel(hdev, -EHOSTDOWN); + err = hci_cmd_sync_queue(hdev, set_powered_sync, cmd, mgmt_set_powered_complete); -- cgit v1.2.3 From d883a4669a1def6d121ccf5e64ad28260d1c9531 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Fri, 21 Apr 2023 11:37:55 -0700 Subject: Bluetooth: hci_sync: Only allow hci_cmd_sync_queue if running This makes sure hci_cmd_sync_queue only queue new work if HCI_RUNNING has been set otherwise there is a risk of commands being sent while turning off. Because hci_cmd_sync_queue can no longer queue work while HCI_RUNNING is not set it cannot be used to power on adapters so instead hci_cmd_sync_submit is introduced which bypass the HCI_RUNNING check, so it behaves like the old implementation. Link: https://lore.kernel.org/all/CAB4PzUpDMvdc8j2MdeSAy1KkAE-D3woprCwAdYWeOc-3v3c9Sw@mail.gmail.com/ Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_sync.h | 2 ++ net/bluetooth/hci_sync.c | 25 +++++++++++++++++++++++-- net/bluetooth/mgmt.c | 12 ++++++++---- 3 files changed, 33 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h index f61b249787fc..2495be4d8b82 100644 --- a/include/net/bluetooth/hci_sync.h +++ b/include/net/bluetooth/hci_sync.h @@ -41,6 +41,8 @@ void hci_cmd_sync_clear(struct hci_dev *hdev); void hci_cmd_sync_cancel(struct hci_dev *hdev, int err); void __hci_cmd_sync_cancel(struct hci_dev *hdev, int err); +int hci_cmd_sync_submit(struct hci_dev *hdev, hci_cmd_sync_work_func_t func, + void *data, hci_cmd_sync_work_destroy_t destroy); int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func, void *data, hci_cmd_sync_work_destroy_t destroy); diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 771aaa808967..647a8ce54062 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -684,8 +684,12 @@ void hci_cmd_sync_cancel(struct hci_dev *hdev, int err) } EXPORT_SYMBOL(hci_cmd_sync_cancel); -int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func, - void *data, hci_cmd_sync_work_destroy_t destroy) +/* Submit HCI command to be run in as cmd_sync_work: + * + * - hdev must _not_ be unregistered + */ +int hci_cmd_sync_submit(struct hci_dev *hdev, hci_cmd_sync_work_func_t func, + void *data, hci_cmd_sync_work_destroy_t destroy) { struct hci_cmd_sync_work_entry *entry; @@ -708,6 +712,23 @@ int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func, return 0; } +EXPORT_SYMBOL(hci_cmd_sync_submit); + +/* Queue HCI command: + * + * - hdev must be running + */ +int hci_cmd_sync_queue(struct hci_dev *hdev, hci_cmd_sync_work_func_t func, + void *data, hci_cmd_sync_work_destroy_t destroy) +{ + /* Only queue command if hdev is running which means it had been opened + * and is either on init phase or is already up. + */ + if (!test_bit(HCI_RUNNING, &hdev->flags)) + return -ENETDOWN; + + return hci_cmd_sync_submit(hdev, func, data, destroy); +} EXPORT_SYMBOL(hci_cmd_sync_queue); int hci_update_eir_sync(struct hci_dev *hdev) diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 13c745876b39..f7b2d0971f24 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -1400,11 +1400,15 @@ static int set_powered(struct sock *sk, struct hci_dev *hdev, void *data, } /* Cancel potentially blocking sync operation before power off */ - if (cp->val == 0x00) + if (cp->val == 0x00) { __hci_cmd_sync_cancel(hdev, -EHOSTDOWN); - - err = hci_cmd_sync_queue(hdev, set_powered_sync, cmd, - mgmt_set_powered_complete); + err = hci_cmd_sync_queue(hdev, set_powered_sync, cmd, + mgmt_set_powered_complete); + } else { + /* Use hci_cmd_sync_submit since hdev might not be running */ + err = hci_cmd_sync_submit(hdev, set_powered_sync, cmd, + mgmt_set_powered_complete); + } if (err < 0) mgmt_pending_remove(cmd); -- cgit v1.2.3 From d913d32cc2707e9cd24fe6fa6d7d470e9c728980 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 21 Apr 2023 11:52:55 -0700 Subject: netlink: Use copy_to_user() for optval in netlink_getsockopt(). Brad Spencer provided a detailed report [0] that when calling getsockopt() for AF_NETLINK, some SOL_NETLINK options set only 1 byte even though such options require at least sizeof(int) as length. The options return a flag value that fits into 1 byte, but such behaviour confuses users who do not initialise the variable before calling getsockopt() and do not strictly check the returned value as char. Currently, netlink_getsockopt() uses put_user() to copy data to optlen and optval, but put_user() casts the data based on the pointer, char *optval. As a result, only 1 byte is set to optval. To avoid this behaviour, we need to use copy_to_user() or cast optval for put_user(). Note that this changes the behaviour on big-endian systems, but we document that the size of optval is int in the man page. $ man 7 netlink ... Socket options To set or get a netlink socket option, call getsockopt(2) to read or setsockopt(2) to write the option with the option level argument set to SOL_NETLINK. Unless otherwise noted, optval is a pointer to an int. Fixes: 9a4595bc7e67 ("[NETLINK]: Add set/getsockopt options to support more than 32 groups") Fixes: be0c22a46cfb ("netlink: add NETLINK_BROADCAST_ERROR socket option") Fixes: 38938bfe3489 ("netlink: add NETLINK_NO_ENOBUFS socket flag") Fixes: 0a6a3a23ea6e ("netlink: add NETLINK_CAP_ACK socket option") Fixes: 2d4bc93368f5 ("netlink: extended ACK reporting") Fixes: 89d35528d17d ("netlink: Add new socket option to enable strict checking on dumps") Reported-by: Brad Spencer Link: https://lore.kernel.org/netdev/ZD7VkNWFfp22kTDt@datsun.rim.net/ Signed-off-by: Kuniyuki Iwashima Reviewed-by: Johannes Berg Link: https://lore.kernel.org/r/20230421185255.94606-1-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/netlink/af_netlink.c | 75 +++++++++++++++--------------------------------- 1 file changed, 23 insertions(+), 52 deletions(-) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index f365dfdd672d..9b6eb28e6e94 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1742,7 +1742,8 @@ static int netlink_getsockopt(struct socket *sock, int level, int optname, { struct sock *sk = sock->sk; struct netlink_sock *nlk = nlk_sk(sk); - int len, val, err; + unsigned int flag; + int len, val; if (level != SOL_NETLINK) return -ENOPROTOOPT; @@ -1754,39 +1755,17 @@ static int netlink_getsockopt(struct socket *sock, int level, int optname, switch (optname) { case NETLINK_PKTINFO: - if (len < sizeof(int)) - return -EINVAL; - len = sizeof(int); - val = nlk->flags & NETLINK_F_RECV_PKTINFO ? 1 : 0; - if (put_user(len, optlen) || - put_user(val, optval)) - return -EFAULT; - err = 0; + flag = NETLINK_F_RECV_PKTINFO; break; case NETLINK_BROADCAST_ERROR: - if (len < sizeof(int)) - return -EINVAL; - len = sizeof(int); - val = nlk->flags & NETLINK_F_BROADCAST_SEND_ERROR ? 1 : 0; - if (put_user(len, optlen) || - put_user(val, optval)) - return -EFAULT; - err = 0; + flag = NETLINK_F_BROADCAST_SEND_ERROR; break; case NETLINK_NO_ENOBUFS: - if (len < sizeof(int)) - return -EINVAL; - len = sizeof(int); - val = nlk->flags & NETLINK_F_RECV_NO_ENOBUFS ? 1 : 0; - if (put_user(len, optlen) || - put_user(val, optval)) - return -EFAULT; - err = 0; + flag = NETLINK_F_RECV_NO_ENOBUFS; break; case NETLINK_LIST_MEMBERSHIPS: { - int pos, idx, shift; + int pos, idx, shift, err = 0; - err = 0; netlink_lock_table(); for (pos = 0; pos * 8 < nlk->ngroups; pos += sizeof(u32)) { if (len - pos < sizeof(u32)) @@ -1803,40 +1782,32 @@ static int netlink_getsockopt(struct socket *sock, int level, int optname, if (put_user(ALIGN(nlk->ngroups / 8, sizeof(u32)), optlen)) err = -EFAULT; netlink_unlock_table(); - break; + return err; } case NETLINK_CAP_ACK: - if (len < sizeof(int)) - return -EINVAL; - len = sizeof(int); - val = nlk->flags & NETLINK_F_CAP_ACK ? 1 : 0; - if (put_user(len, optlen) || - put_user(val, optval)) - return -EFAULT; - err = 0; + flag = NETLINK_F_CAP_ACK; break; case NETLINK_EXT_ACK: - if (len < sizeof(int)) - return -EINVAL; - len = sizeof(int); - val = nlk->flags & NETLINK_F_EXT_ACK ? 1 : 0; - if (put_user(len, optlen) || put_user(val, optval)) - return -EFAULT; - err = 0; + flag = NETLINK_F_EXT_ACK; break; case NETLINK_GET_STRICT_CHK: - if (len < sizeof(int)) - return -EINVAL; - len = sizeof(int); - val = nlk->flags & NETLINK_F_STRICT_CHK ? 1 : 0; - if (put_user(len, optlen) || put_user(val, optval)) - return -EFAULT; - err = 0; + flag = NETLINK_F_STRICT_CHK; break; default: - err = -ENOPROTOOPT; + return -ENOPROTOOPT; } - return err; + + if (len < sizeof(int)) + return -EINVAL; + + len = sizeof(int); + val = nlk->flags & flag ? 1 : 0; + + if (put_user(len, optlen) || + copy_to_user(optval, &val, len)) + return -EFAULT; + + return 0; } static void netlink_cmsg_recv_pktinfo(struct msghdr *msg, struct sk_buff *skb) -- cgit v1.2.3 From 00d0f31a1ec8eb3e4d692bdf2455daf0fb865bb3 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 20 Apr 2023 16:33:02 -0700 Subject: net: ethtool: coalesce: try to make user settings stick twice SET_COALESCE may change operation mode and parameters in one call. Changing operation mode may cause the driver to reset the parameter values to what is a reasonable default for new operation mode. Since driver does not know which parameters come from user and which are echoed back from ->get, driver may ignore the parameters when switching operation modes. This used to be inevitable for ioctl() but in netlink we know which parameters are actually specified by the user. We could inform which parameters were set by the user but this would lead to a lot of code duplication in the drivers. Instead try to call the drivers twice if both mode and params are changed. The set method already checks if any params need updating so in case the driver did the right thing the first time around - there will be no second call to it's ->set method (only an extra call to ->get()). For mlx5 for example before this patch we'd see: # ethtool -C eth0 adaptive-rx on adaptive-tx on # ethtool -C eth0 adaptive-rx off adaptive-tx off \ tx-usecs 123 rx-usecs 123 Adaptive RX: off TX: off rx-usecs: 3 rx-frames: 32 tx-usecs: 16 tx-frames: 32 [...] After the change: # ethtool -C eth0 adaptive-rx on adaptive-tx on # ethtool -C eth0 adaptive-rx off adaptive-tx off \ tx-usecs 123 rx-usecs 123 Adaptive RX: off TX: off rx-usecs: 123 rx-frames: 32 tx-usecs: 123 tx-frames: 32 [...] This only works for netlink, so it's a small discrepancy between netlink and ioctl(). Since we anticipate most users to move to netlink I believe it's worth making their lives easier. Link: https://lore.kernel.org/r/20230420233302.944382-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/networking/ethtool-netlink.rst | 4 +++ net/ethtool/coalesce.c | 54 ++++++++++++++++++++++------ 2 files changed, 47 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst index cd0973d4ba01..2540c70952ff 100644 --- a/Documentation/networking/ethtool-netlink.rst +++ b/Documentation/networking/ethtool-netlink.rst @@ -1099,6 +1099,10 @@ such that the corresponding bit in ``ethtool_ops::supported_coalesce_params`` is not set), regardless of their values. Driver may impose additional constraints on coalescing parameters and their values. +Compared to requests issued via the ``ioctl()`` netlink version of this request +will try harder to make sure that values specified by the user have been applied +and may call the driver twice. + PAUSE_GET ========= diff --git a/net/ethtool/coalesce.c b/net/ethtool/coalesce.c index 443e7e642c96..01a59ce211c8 100644 --- a/net/ethtool/coalesce.c +++ b/net/ethtool/coalesce.c @@ -254,13 +254,14 @@ ethnl_set_coalesce_validate(struct ethnl_req_info *req_info, } static int -ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info) +__ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info, + bool *dual_change) { struct kernel_ethtool_coalesce kernel_coalesce = {}; struct net_device *dev = req_info->dev; struct ethtool_coalesce coalesce = {}; + bool mod_mode = false, mod = false; struct nlattr **tb = info->attrs; - bool mod = false; int ret; ret = dev->ethtool_ops->get_coalesce(dev, &coalesce, &kernel_coalesce, @@ -268,6 +269,7 @@ ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info) if (ret < 0) return ret; + /* Update values */ ethnl_update_u32(&coalesce.rx_coalesce_usecs, tb[ETHTOOL_A_COALESCE_RX_USECS], &mod); ethnl_update_u32(&coalesce.rx_max_coalesced_frames, @@ -286,10 +288,6 @@ ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info) tb[ETHTOOL_A_COALESCE_TX_MAX_FRAMES_IRQ], &mod); ethnl_update_u32(&coalesce.stats_block_coalesce_usecs, tb[ETHTOOL_A_COALESCE_STATS_BLOCK_USECS], &mod); - ethnl_update_bool32(&coalesce.use_adaptive_rx_coalesce, - tb[ETHTOOL_A_COALESCE_USE_ADAPTIVE_RX], &mod); - ethnl_update_bool32(&coalesce.use_adaptive_tx_coalesce, - tb[ETHTOOL_A_COALESCE_USE_ADAPTIVE_TX], &mod); ethnl_update_u32(&coalesce.pkt_rate_low, tb[ETHTOOL_A_COALESCE_PKT_RATE_LOW], &mod); ethnl_update_u32(&coalesce.rx_coalesce_usecs_low, @@ -312,17 +310,25 @@ ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info) tb[ETHTOOL_A_COALESCE_TX_MAX_FRAMES_HIGH], &mod); ethnl_update_u32(&coalesce.rate_sample_interval, tb[ETHTOOL_A_COALESCE_RATE_SAMPLE_INTERVAL], &mod); - ethnl_update_u8(&kernel_coalesce.use_cqe_mode_tx, - tb[ETHTOOL_A_COALESCE_USE_CQE_MODE_TX], &mod); - ethnl_update_u8(&kernel_coalesce.use_cqe_mode_rx, - tb[ETHTOOL_A_COALESCE_USE_CQE_MODE_RX], &mod); ethnl_update_u32(&kernel_coalesce.tx_aggr_max_bytes, tb[ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES], &mod); ethnl_update_u32(&kernel_coalesce.tx_aggr_max_frames, tb[ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES], &mod); ethnl_update_u32(&kernel_coalesce.tx_aggr_time_usecs, tb[ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS], &mod); - if (!mod) + + /* Update operation modes */ + ethnl_update_bool32(&coalesce.use_adaptive_rx_coalesce, + tb[ETHTOOL_A_COALESCE_USE_ADAPTIVE_RX], &mod_mode); + ethnl_update_bool32(&coalesce.use_adaptive_tx_coalesce, + tb[ETHTOOL_A_COALESCE_USE_ADAPTIVE_TX], &mod_mode); + ethnl_update_u8(&kernel_coalesce.use_cqe_mode_tx, + tb[ETHTOOL_A_COALESCE_USE_CQE_MODE_TX], &mod_mode); + ethnl_update_u8(&kernel_coalesce.use_cqe_mode_rx, + tb[ETHTOOL_A_COALESCE_USE_CQE_MODE_RX], &mod_mode); + + *dual_change = mod && mod_mode; + if (!mod && !mod_mode) return 0; ret = dev->ethtool_ops->set_coalesce(dev, &coalesce, &kernel_coalesce, @@ -330,6 +336,32 @@ ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info) return ret < 0 ? ret : 1; } +static int +ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info) +{ + bool dual_change; + int err, ret; + + /* SET_COALESCE may change operation mode and parameters in one call. + * Changing operation mode may cause the driver to reset the parameter + * values, and therefore ignore user input (driver does not know which + * parameters come from user and which are echoed back from ->get). + * To not complicate the drivers if user tries to change both the mode + * and parameters at once - call the driver twice. + */ + err = __ethnl_set_coalesce(req_info, info, &dual_change); + if (err < 0) + return err; + ret = err; + + if (ret && dual_change) { + err = __ethnl_set_coalesce(req_info, info, &dual_change); + if (err < 0) + return err; + } + return ret; +} + const struct ethnl_request_ops ethnl_coalesce_request_ops = { .request_cmd = ETHTOOL_MSG_COALESCE_GET, .reply_cmd = ETHTOOL_MSG_COALESCE_GET_REPLY, -- cgit v1.2.3 From 5fc46f94219d1d103ffb5f0832be9da674d85a73 Mon Sep 17 00:00:00 2001 From: Martin Willi Date: Tue, 25 Apr 2023 09:46:18 +0200 Subject: Revert "Fix XFRM-I support for nested ESP tunnels" This reverts commit b0355dbbf13c0052931dd14c38c789efed64d3de. The reverted commit clears the secpath on packets received via xfrm interfaces to support nested IPsec tunnels. This breaks Netfilter policy matching using xt_policy in the FORWARD chain, as the secpath is missing during forwarding. Additionally, Benedict Wong reports that it breaks Transport-in-Tunnel mode. Fix this regression by reverting the commit until we have a better approach for nested IPsec tunnels. Fixes: b0355dbbf13c ("Fix XFRM-I support for nested ESP tunnels") Link: https://lore.kernel.org/netdev/20230412085615.124791-1-martin@strongswan.org/ Signed-off-by: Martin Willi Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_interface_core.c | 54 ++++-------------------------------------- net/xfrm/xfrm_policy.c | 3 --- 2 files changed, 4 insertions(+), 53 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_interface_core.c b/net/xfrm/xfrm_interface_core.c index 35279c220bd7..1f99dc469027 100644 --- a/net/xfrm/xfrm_interface_core.c +++ b/net/xfrm/xfrm_interface_core.c @@ -310,52 +310,6 @@ static void xfrmi_scrub_packet(struct sk_buff *skb, bool xnet) skb->mark = 0; } -static int xfrmi_input(struct sk_buff *skb, int nexthdr, __be32 spi, - int encap_type, unsigned short family) -{ - struct sec_path *sp; - - sp = skb_sec_path(skb); - if (sp && (sp->len || sp->olen) && - !xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family)) - goto discard; - - XFRM_SPI_SKB_CB(skb)->family = family; - if (family == AF_INET) { - XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr); - XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = NULL; - } else { - XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct ipv6hdr, daddr); - XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6 = NULL; - } - - return xfrm_input(skb, nexthdr, spi, encap_type); -discard: - kfree_skb(skb); - return 0; -} - -static int xfrmi4_rcv(struct sk_buff *skb) -{ - return xfrmi_input(skb, ip_hdr(skb)->protocol, 0, 0, AF_INET); -} - -static int xfrmi6_rcv(struct sk_buff *skb) -{ - return xfrmi_input(skb, skb_network_header(skb)[IP6CB(skb)->nhoff], - 0, 0, AF_INET6); -} - -static int xfrmi4_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) -{ - return xfrmi_input(skb, nexthdr, spi, encap_type, AF_INET); -} - -static int xfrmi6_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) -{ - return xfrmi_input(skb, nexthdr, spi, encap_type, AF_INET6); -} - static int xfrmi_rcv_cb(struct sk_buff *skb, int err) { const struct xfrm_mode *inner_mode; @@ -991,8 +945,8 @@ static struct pernet_operations xfrmi_net_ops = { }; static struct xfrm6_protocol xfrmi_esp6_protocol __read_mostly = { - .handler = xfrmi6_rcv, - .input_handler = xfrmi6_input, + .handler = xfrm6_rcv, + .input_handler = xfrm_input, .cb_handler = xfrmi_rcv_cb, .err_handler = xfrmi6_err, .priority = 10, @@ -1042,8 +996,8 @@ static struct xfrm6_tunnel xfrmi_ip6ip_handler __read_mostly = { #endif static struct xfrm4_protocol xfrmi_esp4_protocol __read_mostly = { - .handler = xfrmi4_rcv, - .input_handler = xfrmi4_input, + .handler = xfrm4_rcv, + .input_handler = xfrm_input, .cb_handler = xfrmi_rcv_cb, .err_handler = xfrmi4_err, .priority = 10, diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 62be042f2ebc..21a3a1cd3d6d 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -3739,9 +3739,6 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, goto reject; } - if (if_id) - secpath_reset(skb); - xfrm_pols_put(pols, npols); return 1; } -- cgit v1.2.3 From 50749f2dd6854a41830996ad302aef2ffaf011d8 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 24 Apr 2023 15:20:22 -0700 Subject: tcp/udp: Fix memleaks of sk and zerocopy skbs with TX timestamp. syzkaller reported [0] memory leaks of an UDP socket and ZEROCOPY skbs. We can reproduce the problem with these sequences: sk = socket(AF_INET, SOCK_DGRAM, 0) sk.setsockopt(SOL_SOCKET, SO_TIMESTAMPING, SOF_TIMESTAMPING_TX_SOFTWARE) sk.setsockopt(SOL_SOCKET, SO_ZEROCOPY, 1) sk.sendto(b'', MSG_ZEROCOPY, ('127.0.0.1', 53)) sk.close() sendmsg() calls msg_zerocopy_alloc(), which allocates a skb, sets skb->cb->ubuf.refcnt to 1, and calls sock_hold(). Here, struct ubuf_info_msgzc indirectly holds a refcnt of the socket. When the skb is sent, __skb_tstamp_tx() clones it and puts the clone into the socket's error queue with the TX timestamp. When the original skb is received locally, skb_copy_ubufs() calls skb_unclone(), and pskb_expand_head() increments skb->cb->ubuf.refcnt. This additional count is decremented while freeing the skb, but struct ubuf_info_msgzc still has a refcnt, so __msg_zerocopy_callback() is not called. The last refcnt is not released unless we retrieve the TX timestamped skb by recvmsg(). Since we clear the error queue in inet_sock_destruct() after the socket's refcnt reaches 0, there is a circular dependency. If we close() the socket holding such skbs, we never call sock_put() and leak the count, sk, and skb. TCP has the same problem, and commit e0c8bccd40fc ("net: stream: purge sk_error_queue in sk_stream_kill_queues()") tried to fix it by calling skb_queue_purge() during close(). However, there is a small chance that skb queued in a qdisc or device could be put into the error queue after the skb_queue_purge() call. In __skb_tstamp_tx(), the cloned skb should not have a reference to the ubuf to remove the circular dependency, but skb_clone() does not call skb_copy_ubufs() for zerocopy skb. So, we need to call skb_orphan_frags_rx() for the cloned skb to call skb_copy_ubufs(). [0]: BUG: memory leak unreferenced object 0xffff88800c6d2d00 (size 1152): comm "syz-executor392", pid 264, jiffies 4294785440 (age 13.044s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 cd af e8 81 00 00 00 00 ................ 02 00 07 40 00 00 00 00 00 00 00 00 00 00 00 00 ...@............ backtrace: [<0000000055636812>] sk_prot_alloc+0x64/0x2a0 net/core/sock.c:2024 [<0000000054d77b7a>] sk_alloc+0x3b/0x800 net/core/sock.c:2083 [<0000000066f3c7e0>] inet_create net/ipv4/af_inet.c:319 [inline] [<0000000066f3c7e0>] inet_create+0x31e/0xe40 net/ipv4/af_inet.c:245 [<000000009b83af97>] __sock_create+0x2ab/0x550 net/socket.c:1515 [<00000000b9b11231>] sock_create net/socket.c:1566 [inline] [<00000000b9b11231>] __sys_socket_create net/socket.c:1603 [inline] [<00000000b9b11231>] __sys_socket_create net/socket.c:1588 [inline] [<00000000b9b11231>] __sys_socket+0x138/0x250 net/socket.c:1636 [<000000004fb45142>] __do_sys_socket net/socket.c:1649 [inline] [<000000004fb45142>] __se_sys_socket net/socket.c:1647 [inline] [<000000004fb45142>] __x64_sys_socket+0x73/0xb0 net/socket.c:1647 [<0000000066999e0e>] do_syscall_x64 arch/x86/entry/common.c:50 [inline] [<0000000066999e0e>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80 [<0000000017f238c1>] entry_SYSCALL_64_after_hwframe+0x63/0xcd BUG: memory leak unreferenced object 0xffff888017633a00 (size 240): comm "syz-executor392", pid 264, jiffies 4294785440 (age 13.044s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 2d 6d 0c 80 88 ff ff .........-m..... backtrace: [<000000002b1c4368>] __alloc_skb+0x229/0x320 net/core/skbuff.c:497 [<00000000143579a6>] alloc_skb include/linux/skbuff.h:1265 [inline] [<00000000143579a6>] sock_omalloc+0xaa/0x190 net/core/sock.c:2596 [<00000000be626478>] msg_zerocopy_alloc net/core/skbuff.c:1294 [inline] [<00000000be626478>] msg_zerocopy_realloc+0x1ce/0x7f0 net/core/skbuff.c:1370 [<00000000cbfc9870>] __ip_append_data+0x2adf/0x3b30 net/ipv4/ip_output.c:1037 [<0000000089869146>] ip_make_skb+0x26c/0x2e0 net/ipv4/ip_output.c:1652 [<00000000098015c2>] udp_sendmsg+0x1bac/0x2390 net/ipv4/udp.c:1253 [<0000000045e0e95e>] inet_sendmsg+0x10a/0x150 net/ipv4/af_inet.c:819 [<000000008d31bfde>] sock_sendmsg_nosec net/socket.c:714 [inline] [<000000008d31bfde>] sock_sendmsg+0x141/0x190 net/socket.c:734 [<0000000021e21aa4>] __sys_sendto+0x243/0x360 net/socket.c:2117 [<00000000ac0af00c>] __do_sys_sendto net/socket.c:2129 [inline] [<00000000ac0af00c>] __se_sys_sendto net/socket.c:2125 [inline] [<00000000ac0af00c>] __x64_sys_sendto+0xe1/0x1c0 net/socket.c:2125 [<0000000066999e0e>] do_syscall_x64 arch/x86/entry/common.c:50 [inline] [<0000000066999e0e>] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80 [<0000000017f238c1>] entry_SYSCALL_64_after_hwframe+0x63/0xcd Fixes: f214f915e7db ("tcp: enable MSG_ZEROCOPY") Fixes: b5947e5d1e71 ("udp: msg_zerocopy") Reported-by: syzbot Signed-off-by: Kuniyuki Iwashima Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/core/skbuff.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 4c0879798eb8..2f9bb98170ab 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -5162,6 +5162,9 @@ void __skb_tstamp_tx(struct sk_buff *orig_skb, skb = alloc_skb(0, GFP_ATOMIC); } else { skb = skb_clone(orig_skb, GFP_ATOMIC); + + if (skb_orphan_frags_rx(skb, GFP_ATOMIC)) + return; } if (!skb) return; -- cgit v1.2.3 From cf64b9bce95095b80f4589e4f54572cc5d8c1538 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 8 Mar 2023 17:51:00 +1100 Subject: SUNRPC: return proper error from get_expiry() The get_expiry() function currently returns a timestamp, and uses the special return value of 0 to indicate an error. Unfortunately this causes a problem when 0 is the correct return value. On a system with no RTC it is possible that the boot time will be seen to be "3". When exportfs probes to see if a particular filesystem supports NFS export it tries to cache information with an expiry time of "3". The intention is for this to be "long in the past". Even with no RTC it will not be far in the future (at most a second or two) so this is harmless. But if the boot time happens to have been calculated to be "3", then get_expiry will fail incorrectly as it converts the number to "seconds since bootime" - 0. To avoid this problem we change get_expiry() to report the error quite separately from the expiry time. The error is now the return value. The expiry time is reported through a by-reference parameter. Reported-by: Jerry Zhang Tested-by: Jerry Zhang Signed-off-by: NeilBrown Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/export.c | 13 ++++++------- fs/nfsd/nfs4idmap.c | 8 ++++---- include/linux/sunrpc/cache.h | 15 ++++++++------- net/sunrpc/auth_gss/svcauth_gss.c | 12 ++++++------ net/sunrpc/svcauth_unix.c | 12 ++++++------ 5 files changed, 30 insertions(+), 30 deletions(-) (limited to 'net') diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 668c7527b17e..6da74aebe1fb 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -123,11 +123,11 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen) /* OK, we seem to have a valid key */ key.h.flags = 0; - key.h.expiry_time = get_expiry(&mesg); - if (key.h.expiry_time == 0) + err = get_expiry(&mesg, &key.h.expiry_time); + if (err) goto out; - key.ek_client = dom; + key.ek_client = dom; key.ek_fsidtype = fsidtype; memcpy(key.ek_fsid, buf, len); @@ -610,9 +610,8 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) exp.ex_devid_map = NULL; /* expiry */ - err = -EINVAL; - exp.h.expiry_time = get_expiry(&mesg); - if (exp.h.expiry_time == 0) + err = get_expiry(&mesg, &exp.h.expiry_time); + if (err) goto out3; /* flags */ @@ -624,7 +623,7 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) if (err || an_int < 0) goto out3; exp.ex_flags= an_int; - + /* anon uid */ err = get_int(&mesg, &an_int); if (err) diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c index 5e9809aff37e..7a806ac13e31 100644 --- a/fs/nfsd/nfs4idmap.c +++ b/fs/nfsd/nfs4idmap.c @@ -240,8 +240,8 @@ idtoname_parse(struct cache_detail *cd, char *buf, int buflen) goto out; /* expiry */ - ent.h.expiry_time = get_expiry(&buf); - if (ent.h.expiry_time == 0) + error = get_expiry(&buf, &ent.h.expiry_time); + if (error) goto out; error = -ENOMEM; @@ -408,8 +408,8 @@ nametoid_parse(struct cache_detail *cd, char *buf, int buflen) memcpy(ent.name, buf1, sizeof(ent.name)); /* expiry */ - ent.h.expiry_time = get_expiry(&buf); - if (ent.h.expiry_time == 0) + error = get_expiry(&buf, &ent.h.expiry_time); + if (error) goto out; /* ID */ diff --git a/include/linux/sunrpc/cache.h b/include/linux/sunrpc/cache.h index ec5a555df96f..518bd28f5ab8 100644 --- a/include/linux/sunrpc/cache.h +++ b/include/linux/sunrpc/cache.h @@ -300,17 +300,18 @@ static inline int get_time(char **bpp, time64_t *time) return 0; } -static inline time64_t get_expiry(char **bpp) +static inline int get_expiry(char **bpp, time64_t *rvp) { - time64_t rv; + int error; struct timespec64 boot; - if (get_time(bpp, &rv)) - return 0; - if (rv < 0) - return 0; + error = get_time(bpp, rvp); + if (error) + return error; + getboottime64(&boot); - return rv - boot.tv_sec; + (*rvp) -= boot.tv_sec; + return 0; } #endif /* _LINUX_SUNRPC_CACHE_H_ */ diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c index 9c843974bb48..c4a566737085 100644 --- a/net/sunrpc/auth_gss/svcauth_gss.c +++ b/net/sunrpc/auth_gss/svcauth_gss.c @@ -257,11 +257,11 @@ static int rsi_parse(struct cache_detail *cd, rsii.h.flags = 0; /* expiry */ - expiry = get_expiry(&mesg); - status = -EINVAL; - if (expiry == 0) + status = get_expiry(&mesg, &expiry); + if (status) goto out; + status = -EINVAL; /* major/minor */ len = qword_get(&mesg, buf, mlen); if (len <= 0) @@ -483,11 +483,11 @@ static int rsc_parse(struct cache_detail *cd, rsci.h.flags = 0; /* expiry */ - expiry = get_expiry(&mesg); - status = -EINVAL; - if (expiry == 0) + status = get_expiry(&mesg, &expiry); + if (status) goto out; + status = -EINVAL; rscp = rsc_lookup(cd, &rsci); if (!rscp) goto out; diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c index 4246363cb095..4485088ce27b 100644 --- a/net/sunrpc/svcauth_unix.c +++ b/net/sunrpc/svcauth_unix.c @@ -225,9 +225,9 @@ static int ip_map_parse(struct cache_detail *cd, return -EINVAL; } - expiry = get_expiry(&mesg); - if (expiry ==0) - return -EINVAL; + err = get_expiry(&mesg, &expiry); + if (err) + return err; /* domainname, or empty for NEGATIVE */ len = qword_get(&mesg, buf, mlen); @@ -506,9 +506,9 @@ static int unix_gid_parse(struct cache_detail *cd, uid = make_kuid(current_user_ns(), id); ug.uid = uid; - expiry = get_expiry(&mesg); - if (expiry == 0) - return -EINVAL; + err = get_expiry(&mesg, &expiry); + if (err) + return err; rv = get_int(&mesg, &gids); if (rv || gids < 0 || gids > 8192) -- cgit v1.2.3 From 376bcd9b37632cf191711a68aa25ab42f0048c2e Mon Sep 17 00:00:00 2001 From: Luis Chamberlain Date: Sat, 11 Mar 2023 15:39:40 -0800 Subject: sunrpc: simplify two-level sysctl registration for svcrdma_parm_table There is no need to declare two tables to just create directories, this can be easily be done with a prefix path with register_sysctl(). Simplify this registration. Signed-off-by: Luis Chamberlain Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- net/sunrpc/xprtrdma/svc_rdma.c | 21 ++------------------- 1 file changed, 2 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/sunrpc/xprtrdma/svc_rdma.c b/net/sunrpc/xprtrdma/svc_rdma.c index 5bc20e9d09cd..f0d5eeed4c88 100644 --- a/net/sunrpc/xprtrdma/svc_rdma.c +++ b/net/sunrpc/xprtrdma/svc_rdma.c @@ -212,24 +212,6 @@ static struct ctl_table svcrdma_parm_table[] = { { }, }; -static struct ctl_table svcrdma_table[] = { - { - .procname = "svc_rdma", - .mode = 0555, - .child = svcrdma_parm_table - }, - { }, -}; - -static struct ctl_table svcrdma_root_table[] = { - { - .procname = "sunrpc", - .mode = 0555, - .child = svcrdma_table - }, - { }, -}; - static void svc_rdma_proc_cleanup(void) { if (!svcrdma_table_header) @@ -263,7 +245,8 @@ static int svc_rdma_proc_init(void) if (rc) goto out_err; - svcrdma_table_header = register_sysctl_table(svcrdma_root_table); + svcrdma_table_header = register_sysctl("sunrpc/svc_rdma", + svcrdma_parm_table); return 0; out_err: -- cgit v1.2.3 From 0f5162480bd25bd97b91c9153db7afbd89698804 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 17 Mar 2023 17:09:20 -0400 Subject: NFSD: Watch for rq_pages bounds checking errors in nfsd_splice_actor() There have been several bugs over the years where the NFSD splice actor has attempted to write outside the rq_pages array. This is a "should never happen" condition, but if for some reason the pipe splice actor should attempt to walk past the end of rq_pages, it needs to terminate the READ operation to prevent corruption of the pointer addresses in the fields just beyond the array. A server crash is thus prevented. Since the code is not behaving, the READ operation returns -EIO to the client. None of the READ payload data can be trusted if the splice actor isn't operating as expected. Suggested-by: Jeff Layton Signed-off-by: Chuck Lever Reviewed-by: Jeff Layton --- fs/nfsd/vfs.c | 6 +++++- include/linux/sunrpc/svc.h | 2 +- include/trace/events/sunrpc.h | 25 +++++++++++++++++++++++++ net/sunrpc/svc.c | 15 ++++++++++++++- 4 files changed, 45 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 5783209f17fc..10aa68ca82ef 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -930,6 +930,9 @@ nfsd_open_verified(struct svc_rqst *rqstp, struct svc_fh *fhp, int may_flags, * Grab and keep cached pages associated with a file in the svc_rqst * so that they can be passed to the network sendmsg/sendpage routines * directly. They will be released after the sending has completed. + * + * Return values: Number of bytes consumed, or -EIO if there are no + * remaining pages in rqstp->rq_pages. */ static int nfsd_splice_actor(struct pipe_inode_info *pipe, struct pipe_buffer *buf, @@ -948,7 +951,8 @@ nfsd_splice_actor(struct pipe_inode_info *pipe, struct pipe_buffer *buf, */ if (page == *(rqstp->rq_next_page - 1)) continue; - svc_rqst_replace_page(rqstp, page); + if (unlikely(!svc_rqst_replace_page(rqstp, page))) + return -EIO; } if (rqstp->rq_res.page_len == 0) // first call rqstp->rq_res.page_base = offset % PAGE_SIZE; diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 877891536c2f..f5af055280ff 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -422,7 +422,7 @@ struct svc_serv *svc_create(struct svc_program *, unsigned int, int (*threadfn)(void *data)); struct svc_rqst *svc_rqst_alloc(struct svc_serv *serv, struct svc_pool *pool, int node); -void svc_rqst_replace_page(struct svc_rqst *rqstp, +bool svc_rqst_replace_page(struct svc_rqst *rqstp, struct page *page); void svc_rqst_free(struct svc_rqst *); void svc_exit_thread(struct svc_rqst *); diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h index 3ca54536f8f7..5a3bb42e1f50 100644 --- a/include/trace/events/sunrpc.h +++ b/include/trace/events/sunrpc.h @@ -1790,6 +1790,31 @@ DEFINE_EVENT(svc_rqst_status, svc_send, TP_PROTO(const struct svc_rqst *rqst, int status), TP_ARGS(rqst, status)); +TRACE_EVENT(svc_replace_page_err, + TP_PROTO(const struct svc_rqst *rqst), + + TP_ARGS(rqst), + TP_STRUCT__entry( + SVC_RQST_ENDPOINT_FIELDS(rqst) + + __field(const void *, begin) + __field(const void *, respages) + __field(const void *, nextpage) + ), + + TP_fast_assign( + SVC_RQST_ENDPOINT_ASSIGNMENTS(rqst); + + __entry->begin = rqst->rq_pages; + __entry->respages = rqst->rq_respages; + __entry->nextpage = rqst->rq_next_page; + ), + + TP_printk(SVC_RQST_ENDPOINT_FORMAT " begin=%p respages=%p nextpage=%p", + SVC_RQST_ENDPOINT_VARARGS, + __entry->begin, __entry->respages, __entry->nextpage) +); + TRACE_EVENT(svc_stats_latency, TP_PROTO( const struct svc_rqst *rqst diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index fea7ce8fba14..633aa1eb476b 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -842,9 +842,21 @@ EXPORT_SYMBOL_GPL(svc_set_num_threads); * * When replacing a page in rq_pages, batch the release of the * replaced pages to avoid hammering the page allocator. + * + * Return values: + * %true: page replaced + * %false: array bounds checking failed */ -void svc_rqst_replace_page(struct svc_rqst *rqstp, struct page *page) +bool svc_rqst_replace_page(struct svc_rqst *rqstp, struct page *page) { + struct page **begin = rqstp->rq_pages; + struct page **end = &rqstp->rq_pages[RPCSVC_MAXPAGES]; + + if (unlikely(rqstp->rq_next_page < begin || rqstp->rq_next_page > end)) { + trace_svc_replace_page_err(rqstp); + return false; + } + if (*rqstp->rq_next_page) { if (!pagevec_space(&rqstp->rq_pvec)) __pagevec_release(&rqstp->rq_pvec); @@ -853,6 +865,7 @@ void svc_rqst_replace_page(struct svc_rqst *rqstp, struct page *page) get_page(page); *(rqstp->rq_next_page++) = page; + return true; } EXPORT_SYMBOL_GPL(svc_rqst_replace_page); -- cgit v1.2.3 From ae0d77708aae219a9264a74188d5c1b1a5754da6 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 7 Mar 2023 09:14:46 -0500 Subject: SUNRPC: Ensure server-side sockets have a sock->file The TLS handshake upcall mechanism requires a non-NULL sock->file on the socket it hands to user space. svc_sock_free() already releases sock->file properly if one exists. Signed-off-by: Chuck Lever --- net/sunrpc/svcsock.c | 25 ++++++++++++++++++------- 1 file changed, 18 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 03a4f5615086..302a14dd7882 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1293,26 +1293,37 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, struct socket *sock, int flags) { + struct file *filp = NULL; struct svc_sock *svsk; struct sock *inet; int pmap_register = !(flags & SVC_SOCK_ANONYMOUS); - int err = 0; svsk = kzalloc(sizeof(*svsk), GFP_KERNEL); if (!svsk) return ERR_PTR(-ENOMEM); + if (!sock->file) { + filp = sock_alloc_file(sock, O_NONBLOCK, NULL); + if (IS_ERR(filp)) { + kfree(svsk); + return ERR_CAST(filp); + } + } + inet = sock->sk; - /* Register socket with portmapper */ - if (pmap_register) + if (pmap_register) { + int err; + err = svc_register(serv, sock_net(sock->sk), inet->sk_family, inet->sk_protocol, ntohs(inet_sk(inet)->inet_sport)); - - if (err < 0) { - kfree(svsk); - return ERR_PTR(err); + if (err < 0) { + if (filp) + fput(filp); + kfree(svsk); + return ERR_PTR(err); + } } svsk->sk_sock = sock; -- cgit v1.2.3 From 55fcc7d9159de886296626e47db2c81f8578c7e1 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 3 Apr 2023 13:53:07 -0400 Subject: SUNRPC: Ignore return value of ->xpo_sendto Clean up: All callers of svc_process() ignore its return value, so svc_process() can safely be converted to return void. Ditto for svc_send(). The return value of ->xpo_sendto() is now used only as part of a trace event. Signed-off-by: Chuck Lever --- include/linux/sunrpc/svc.h | 2 +- include/linux/sunrpc/svcsock.h | 2 +- net/sunrpc/svc.c | 13 +++++++------ net/sunrpc/svc_xprt.c | 21 +++++++++------------ 4 files changed, 18 insertions(+), 20 deletions(-) (limited to 'net') diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index f5af055280ff..2d31121fc2e6 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -430,7 +430,7 @@ struct svc_serv * svc_create_pooled(struct svc_program *, unsigned int, int (*threadfn)(void *data)); int svc_set_num_threads(struct svc_serv *, struct svc_pool *, int); int svc_pool_stats_open(struct svc_serv *serv, struct file *file); -int svc_process(struct svc_rqst *); +void svc_process(struct svc_rqst *rqstp); int bc_svc_process(struct svc_serv *, struct rpc_rqst *, struct svc_rqst *); int svc_register(const struct svc_serv *, struct net *, const int, diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h index bcc555c7ae9c..dd73fa174af5 100644 --- a/include/linux/sunrpc/svcsock.h +++ b/include/linux/sunrpc/svcsock.h @@ -56,7 +56,7 @@ static inline u32 svc_sock_final_rec(struct svc_sock *svsk) */ void svc_close_net(struct svc_serv *, struct net *); int svc_recv(struct svc_rqst *, long); -int svc_send(struct svc_rqst *); +void svc_send(struct svc_rqst *rqstp); void svc_drop(struct svc_rqst *); void svc_sock_update_bufs(struct svc_serv *serv); bool svc_alien_sock(struct net *net, int fd); diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 633aa1eb476b..0aa8892fad63 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -1444,11 +1444,12 @@ err_system_err: goto sendit; } -/* - * Process the RPC request. +/** + * svc_process - Execute one RPC transaction + * @rqstp: RPC transaction context + * */ -int -svc_process(struct svc_rqst *rqstp) +void svc_process(struct svc_rqst *rqstp) { struct kvec *resv = &rqstp->rq_res.head[0]; __be32 *p; @@ -1484,7 +1485,8 @@ svc_process(struct svc_rqst *rqstp) if (!svc_process_common(rqstp)) goto out_drop; - return svc_send(rqstp); + svc_send(rqstp); + return; out_baddir: svc_printk(rqstp, "bad direction 0x%08x, dropping request\n", @@ -1492,7 +1494,6 @@ out_baddir: rqstp->rq_server->sv_stats->rpcbadfmt++; out_drop: svc_drop(rqstp); - return 0; } EXPORT_SYMBOL_GPL(svc_process); diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index ba629297da4e..36c79b718323 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -909,18 +909,20 @@ void svc_drop(struct svc_rqst *rqstp) } EXPORT_SYMBOL_GPL(svc_drop); -/* - * Return reply to client. +/** + * svc_send - Return reply to client + * @rqstp: RPC transaction context + * */ -int svc_send(struct svc_rqst *rqstp) +void svc_send(struct svc_rqst *rqstp) { struct svc_xprt *xprt; - int len = -EFAULT; struct xdr_buf *xb; + int status; xprt = rqstp->rq_xprt; if (!xprt) - goto out; + return; /* calculate over-all length */ xb = &rqstp->rq_res; @@ -930,15 +932,10 @@ int svc_send(struct svc_rqst *rqstp) trace_svc_xdr_sendto(rqstp->rq_xid, xb); trace_svc_stats_latency(rqstp); - len = xprt->xpt_ops->xpo_sendto(rqstp); + status = xprt->xpt_ops->xpo_sendto(rqstp); - trace_svc_send(rqstp, len); + trace_svc_send(rqstp, status); svc_xprt_release(rqstp); - - if (len == -ECONNREFUSED || len == -ENOTCONN || len == -EAGAIN) - len = 0; -out: - return len; } /* -- cgit v1.2.3 From b20cb39def085723868972182fb58fa906839a4f Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 14 Apr 2023 20:17:56 -0400 Subject: SUNRPC: Relocate svc_free_res_pages() Clean-up: There doesn't seem to be a reason why this function is stuck in a header. One thing it prevents is the convenient addition of tracing. Moving it to a source file also makes the rq_respages clean-up logic easier to find. Reviewed-by: Calum Mackay Signed-off-by: Chuck Lever --- include/linux/sunrpc/svc.h | 12 +----------- net/sunrpc/svc.c | 19 +++++++++++++++++++ net/sunrpc/svc_xprt.c | 2 +- 3 files changed, 21 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 2d31121fc2e6..762d7231e574 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -309,17 +309,6 @@ static inline struct sockaddr *svc_daddr(const struct svc_rqst *rqst) return (struct sockaddr *) &rqst->rq_daddr; } -static inline void svc_free_res_pages(struct svc_rqst *rqstp) -{ - while (rqstp->rq_next_page != rqstp->rq_respages) { - struct page **pp = --rqstp->rq_next_page; - if (*pp) { - put_page(*pp); - *pp = NULL; - } - } -} - struct svc_deferred_req { u32 prot; /* protocol (UDP or TCP) */ struct svc_xprt *xprt; @@ -424,6 +413,7 @@ struct svc_rqst *svc_rqst_alloc(struct svc_serv *serv, struct svc_pool *pool, int node); bool svc_rqst_replace_page(struct svc_rqst *rqstp, struct page *page); +void svc_rqst_release_pages(struct svc_rqst *rqstp); void svc_rqst_free(struct svc_rqst *); void svc_exit_thread(struct svc_rqst *); struct svc_serv * svc_create_pooled(struct svc_program *, unsigned int, diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 0aa8892fad63..0fc70cc405b2 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -869,6 +869,25 @@ bool svc_rqst_replace_page(struct svc_rqst *rqstp, struct page *page) } EXPORT_SYMBOL_GPL(svc_rqst_replace_page); +/** + * svc_rqst_release_pages - Release Reply buffer pages + * @rqstp: RPC transaction context + * + * Release response pages that might still be in flight after + * svc_send, and any spliced filesystem-owned pages. + */ +void svc_rqst_release_pages(struct svc_rqst *rqstp) +{ + while (rqstp->rq_next_page != rqstp->rq_respages) { + struct page **pp = --rqstp->rq_next_page; + + if (*pp) { + put_page(*pp); + *pp = NULL; + } + } +} + /* * Called from a server thread as it's exiting. Caller must hold the "service * mutex" for the service. diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 36c79b718323..533e08c4f319 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -542,7 +542,7 @@ static void svc_xprt_release(struct svc_rqst *rqstp) rqstp->rq_deferred = NULL; pagevec_release(&rqstp->rq_pvec); - svc_free_res_pages(rqstp); + svc_rqst_release_pages(rqstp); rqstp->rq_res.page_len = 0; rqstp->rq_res.page_base = 0; -- cgit v1.2.3 From 647a2a6428f2cd01e53079ac16e17fdeff229e68 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 14 Apr 2023 20:18:02 -0400 Subject: SUNRPC: Convert svc_xprt_release() to the release_pages() API Instead of invoking put_page() one-at-a-time, pass the "response" portion of rq_pages directly to release_pages() to reduce the number of times each nfsd thread invokes a page allocator API. Since svc_xprt_release() is not invoked while a client is waiting for an RPC Reply, this is not expected to directly impact mean request latencies on a lightly or moderately loaded server. However as workload intensity increases, I expect somewhat better scalability: the same number of server threads should be able to handle more work. Reviewed-by: Calum Mackay Signed-off-by: Chuck Lever --- net/sunrpc/svc.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 0fc70cc405b2..b982f802f2a0 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -878,13 +878,12 @@ EXPORT_SYMBOL_GPL(svc_rqst_replace_page); */ void svc_rqst_release_pages(struct svc_rqst *rqstp) { - while (rqstp->rq_next_page != rqstp->rq_respages) { - struct page **pp = --rqstp->rq_next_page; + int i, count = rqstp->rq_next_page - rqstp->rq_respages; - if (*pp) { - put_page(*pp); - *pp = NULL; - } + if (count) { + release_pages(rqstp->rq_respages, count); + for (i = 0; i < count; i++) + rqstp->rq_respages[i] = NULL; } } -- cgit v1.2.3 From 2b5fdc0f5caa505afe34d608e2eefadadf2ee67a Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 25 Apr 2023 13:56:35 +0100 Subject: rxrpc: Fix potential data race in rxrpc_wait_to_be_connected() Inside the loop in rxrpc_wait_to_be_connected() it checks call->error to see if it should exit the loop without first checking the call state. This is probably safe as if call->error is set, the call is dead anyway, but we should probably wait for the call state to have been set to completion first, lest it cause surprise on the way out. Fix this by only accessing call->error if the call is complete. We don't actually need to access the error inside the loop as we'll do that after. This caused the following report: BUG: KCSAN: data-race in rxrpc_send_data / rxrpc_set_call_completion write to 0xffff888159cf3c50 of 4 bytes by task 25673 on cpu 1: rxrpc_set_call_completion+0x71/0x1c0 net/rxrpc/call_state.c:22 rxrpc_send_data_packet+0xba9/0x1650 net/rxrpc/output.c:479 rxrpc_transmit_one+0x1e/0x130 net/rxrpc/output.c:714 rxrpc_decant_prepared_tx net/rxrpc/call_event.c:326 [inline] rxrpc_transmit_some_data+0x496/0x600 net/rxrpc/call_event.c:350 rxrpc_input_call_event+0x564/0x1220 net/rxrpc/call_event.c:464 rxrpc_io_thread+0x307/0x1d80 net/rxrpc/io_thread.c:461 kthread+0x1ac/0x1e0 kernel/kthread.c:376 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:308 read to 0xffff888159cf3c50 of 4 bytes by task 25672 on cpu 0: rxrpc_send_data+0x29e/0x1950 net/rxrpc/sendmsg.c:296 rxrpc_do_sendmsg+0xb7a/0xc20 net/rxrpc/sendmsg.c:726 rxrpc_sendmsg+0x413/0x520 net/rxrpc/af_rxrpc.c:565 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg net/socket.c:747 [inline] ____sys_sendmsg+0x375/0x4c0 net/socket.c:2501 ___sys_sendmsg net/socket.c:2555 [inline] __sys_sendmmsg+0x263/0x500 net/socket.c:2641 __do_sys_sendmmsg net/socket.c:2670 [inline] __se_sys_sendmmsg net/socket.c:2667 [inline] __x64_sys_sendmmsg+0x57/0x60 net/socket.c:2667 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd value changed: 0x00000000 -> 0xffffffea Fixes: 9d35d880e0e4 ("rxrpc: Move client call connection to the I/O thread") Reported-by: syzbot+ebc945fdb4acd72cba78@syzkaller.appspotmail.com Link: https://lore.kernel.org/r/000000000000e7c6d205fa10a3cd@google.com/ Signed-off-by: David Howells cc: Marc Dionne cc: Dmitry Vyukov cc: "David S. Miller" cc: Eric Dumazet cc: Jakub Kicinski cc: Paolo Abeni cc: linux-afs@lists.infradead.org cc: linux-fsdevel@vger.kernel.org cc: netdev@vger.kernel.org Link: https://lore.kernel.org/r/508133.1682427395@warthog.procyon.org.uk Signed-off-by: Paolo Abeni --- net/rxrpc/sendmsg.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index da49fcf1c456..6caa47d352ed 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -50,15 +50,11 @@ static int rxrpc_wait_to_be_connected(struct rxrpc_call *call, long *timeo) _enter("%d", call->debug_id); if (rxrpc_call_state(call) != RXRPC_CALL_CLIENT_AWAIT_CONN) - return call->error; + goto no_wait; add_wait_queue_exclusive(&call->waitq, &myself); for (;;) { - ret = call->error; - if (ret < 0) - break; - switch (call->interruptibility) { case RXRPC_INTERRUPTIBLE: case RXRPC_PREINTERRUPTIBLE: @@ -69,10 +65,9 @@ static int rxrpc_wait_to_be_connected(struct rxrpc_call *call, long *timeo) set_current_state(TASK_UNINTERRUPTIBLE); break; } - if (rxrpc_call_state(call) != RXRPC_CALL_CLIENT_AWAIT_CONN) { - ret = call->error; + + if (rxrpc_call_state(call) != RXRPC_CALL_CLIENT_AWAIT_CONN) break; - } if ((call->interruptibility == RXRPC_INTERRUPTIBLE || call->interruptibility == RXRPC_PREINTERRUPTIBLE) && signal_pending(current)) { @@ -85,6 +80,7 @@ static int rxrpc_wait_to_be_connected(struct rxrpc_call *call, long *timeo) remove_wait_queue(&call->waitq, &myself); __set_current_state(TASK_RUNNING); +no_wait: if (ret == 0 && rxrpc_call_is_complete(call)) ret = call->error; -- cgit v1.2.3 From 32eff6bacec2cb574677c15378169a9fa30043ef Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Tue, 25 Apr 2023 16:06:04 +0200 Subject: net/sched: flower: Fix wrong handle assignment during filter change Commit 08a0063df3ae ("net/sched: flower: Move filter handle initialization earlier") moved filter handle initialization but an assignment of the handle to fnew->handle is done regardless of fold value. This is wrong because if fold != NULL (so fold->handle == handle) no new handle is allocated and passed handle is assigned to fnew->handle. Then if any subsequent action in fl_change() fails then the handle value is removed from IDR that is incorrect as we will have still valid old filter instance with handle that is not present in IDR. Fix this issue by moving the assignment so it is done only when passed fold == NULL. Prior the patch: [root@machine tc-testing]# ./tdc.py -d enp1s0f0np0 -e 14be Test 14be: Concurrently replace same range of 100k flower filters from 10 tc instances exit: 123 exit: 0 RTNETLINK answers: Invalid argument We have an error talking to the kernel Command failed tmp/replace_6:1885 All test results: 1..1 not ok 1 14be - Concurrently replace same range of 100k flower filters from 10 tc instances Command exited with 123, expected 0 RTNETLINK answers: Invalid argument We have an error talking to the kernel Command failed tmp/replace_6:1885 After the patch: [root@machine tc-testing]# ./tdc.py -d enp1s0f0np0 -e 14be Test 14be: Concurrently replace same range of 100k flower filters from 10 tc instances All test results: 1..1 ok 1 14be - Concurrently replace same range of 100k flower filters from 10 tc instances Fixes: 08a0063df3ae ("net/sched: flower: Move filter handle initialization earlier") Signed-off-by: Ivan Vecera Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230425140604.169881-1-ivecera@redhat.com Signed-off-by: Paolo Abeni --- net/sched/cls_flower.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index cc49256d5318..6ab6aadc07b8 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -2231,8 +2231,8 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, kfree(fnew); goto errout_tb; } + fnew->handle = handle; } - fnew->handle = handle; err = tcf_exts_init_ex(&fnew->exts, net, TCA_FLOWER_ACT, 0, tp, handle, !tc_skip_hw(fnew->flags)); -- cgit v1.2.3 From 1b483d9f5805c7e3d628d4995e97f4311fcb82eb Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Tue, 25 Apr 2023 11:47:25 -0300 Subject: net/sched: act_pedit: free pedit keys on bail from offset check Ido Schimmel reports a memleak on a syzkaller instance: BUG: memory leak unreferenced object 0xffff88803d45e400 (size 1024): comm "syz-executor292", pid 563, jiffies 4295025223 (age 51.781s) hex dump (first 32 bytes): 28 bd 70 00 fb db df 25 02 00 14 1f ff 02 00 02 (.p....%........ 00 32 00 00 1f 00 00 00 ac 14 14 3e 08 00 07 00 .2.........>.... backtrace: [] kmemleak_alloc_recursive include/linux/kmemleak.h:42 [inline] [] slab_post_alloc_hook mm/slab.h:772 [inline] [] slab_alloc_node mm/slub.c:3452 [inline] [] __kmem_cache_alloc_node+0x25c/0x320 mm/slub.c:3491 [] __do_kmalloc_node mm/slab_common.c:966 [inline] [] __kmalloc+0x59/0x1a0 mm/slab_common.c:980 [] kmalloc include/linux/slab.h:584 [inline] [] tcf_pedit_init+0x793/0x1ae0 net/sched/act_pedit.c:245 [] tcf_action_init_1+0x453/0x6e0 net/sched/act_api.c:1394 [] tcf_action_init+0x5a8/0x950 net/sched/act_api.c:1459 [] tcf_action_add+0x118/0x4e0 net/sched/act_api.c:1985 [] tc_ctl_action+0x377/0x490 net/sched/act_api.c:2044 [] rtnetlink_rcv_msg+0x46d/0xd70 net/core/rtnetlink.c:6395 [] netlink_rcv_skb+0x185/0x490 net/netlink/af_netlink.c:2575 [] rtnetlink_rcv+0x26/0x30 net/core/rtnetlink.c:6413 [] netlink_unicast_kernel net/netlink/af_netlink.c:1339 [inline] [] netlink_unicast+0x5be/0x8a0 net/netlink/af_netlink.c:1365 [] netlink_sendmsg+0x9af/0xed0 net/netlink/af_netlink.c:1942 [] sock_sendmsg_nosec net/socket.c:724 [inline] [] sock_sendmsg net/socket.c:747 [inline] [] ____sys_sendmsg+0x3ef/0xaa0 net/socket.c:2503 [] ___sys_sendmsg+0x122/0x1c0 net/socket.c:2557 [] __sys_sendmsg+0x11f/0x200 net/socket.c:2586 [] __do_sys_sendmsg net/socket.c:2595 [inline] [] __se_sys_sendmsg net/socket.c:2593 [inline] [] __x64_sys_sendmsg+0x80/0xc0 net/socket.c:2593 The recently added static offset check missed a free to the key buffer when bailing out on error. Fixes: e1201bc781c2 ("net/sched: act_pedit: check static offsets a priori") Reported-by: Ido Schimmel Signed-off-by: Pedro Tammela Reviewed-by: Ido Schimmel Tested-by: Ido Schimmel Link: https://lore.kernel.org/r/20230425144725.669262-1-pctammela@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/act_pedit.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index fb93d4c1faca..fc945c7e4123 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -258,7 +258,7 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, if (!offmask && cur % 4) { NL_SET_ERR_MSG_MOD(extack, "Offsets must be on 32bit boundaries"); ret = -EINVAL; - goto put_chain; + goto out_free_keys; } /* sanitize the shift value for any later use */ @@ -291,6 +291,8 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, return ret; +out_free_keys: + kfree(nparms->tcfp_keys); put_chain: if (goto_ch) tcf_chain_put_by_act(goto_ch); -- cgit v1.2.3 From 6a0cdf56bfc9a1791ddd6955d60f3678523e599c Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 14 Apr 2023 20:18:15 -0400 Subject: SUNRPC: Be even lazier about releasing pages A single RPC transaction that touches only a couple of pages means rq_pvec will not be even close to full in svc_xpt_release(). This is a common case. Instead, just leave the pages in rq_pvec until it is completely full. This improves the efficiency of the batch release mechanism on workloads that involve small RPC messages. The rq_pvec is also fully emptied just before thread exit. Reviewed-by: Calum Mackay Signed-off-by: Chuck Lever --- net/sunrpc/svc.c | 3 +++ net/sunrpc/svc_xprt.c | 3 --- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index b982f802f2a0..26367cf4c17a 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -649,6 +649,8 @@ svc_rqst_alloc(struct svc_serv *serv, struct svc_pool *pool, int node) if (!rqstp) return rqstp; + pagevec_init(&rqstp->rq_pvec); + __set_bit(RQ_BUSY, &rqstp->rq_flags); rqstp->rq_server = serv; rqstp->rq_pool = pool; @@ -894,6 +896,7 @@ void svc_rqst_release_pages(struct svc_rqst *rqstp) void svc_rqst_free(struct svc_rqst *rqstp) { + pagevec_release(&rqstp->rq_pvec); svc_release_buffer(rqstp); if (rqstp->rq_scratch_page) put_page(rqstp->rq_scratch_page); diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 533e08c4f319..e3952b690f54 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -541,7 +541,6 @@ static void svc_xprt_release(struct svc_rqst *rqstp) kfree(rqstp->rq_deferred); rqstp->rq_deferred = NULL; - pagevec_release(&rqstp->rq_pvec); svc_rqst_release_pages(rqstp); rqstp->rq_res.page_len = 0; rqstp->rq_res.page_base = 0; @@ -667,8 +666,6 @@ static int svc_alloc_arg(struct svc_rqst *rqstp) struct xdr_buf *arg = &rqstp->rq_arg; unsigned long pages, filled, ret; - pagevec_init(&rqstp->rq_pvec); - pages = (serv->sv_max_mesg + 2 * PAGE_SIZE) >> PAGE_SHIFT; if (pages > RPCSVC_MAXPAGES) { pr_warn_once("svc: warning: pages=%lu > RPCSVC_MAXPAGES=%lu\n", -- cgit v1.2.3 From 5e052dda121e2870dd87181783da4a95d7d2927b Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 17 Apr 2023 09:42:14 -0400 Subject: SUNRPC: Recognize control messages in server-side TCP socket code To support kTLS, the server-side TCP socket receive path needs to watch for CMSGs. Acked-by: Jakub Kicinski Signed-off-by: Chuck Lever --- include/net/tls.h | 2 ++ net/sunrpc/svcsock.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++-- 2 files changed, 48 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/tls.h b/include/net/tls.h index 154949c7b0c8..6056ce5a2aa5 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -69,6 +69,8 @@ extern const struct tls_cipher_size_desc tls_cipher_size_desc[]; #define TLS_CRYPTO_INFO_READY(info) ((info)->cipher_type) +#define TLS_RECORD_TYPE_ALERT 0x15 +#define TLS_RECORD_TYPE_HANDSHAKE 0x16 #define TLS_RECORD_TYPE_DATA 0x17 #define TLS_AAD_SPACE_SIZE 13 diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 302a14dd7882..c5b74f523fc4 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -43,6 +43,7 @@ #include #include #include +#include #include #include #include @@ -216,6 +217,49 @@ static int svc_one_sock_name(struct svc_sock *svsk, char *buf, int remaining) return len; } +static int +svc_tcp_sock_process_cmsg(struct svc_sock *svsk, struct msghdr *msg, + struct cmsghdr *cmsg, int ret) +{ + if (cmsg->cmsg_level == SOL_TLS && + cmsg->cmsg_type == TLS_GET_RECORD_TYPE) { + u8 content_type = *((u8 *)CMSG_DATA(cmsg)); + + switch (content_type) { + case TLS_RECORD_TYPE_DATA: + /* TLS sets EOR at the end of each application data + * record, even though there might be more frames + * waiting to be decrypted. + */ + msg->msg_flags &= ~MSG_EOR; + break; + case TLS_RECORD_TYPE_ALERT: + ret = -ENOTCONN; + break; + default: + ret = -EAGAIN; + } + } + return ret; +} + +static int +svc_tcp_sock_recv_cmsg(struct svc_sock *svsk, struct msghdr *msg) +{ + union { + struct cmsghdr cmsg; + u8 buf[CMSG_SPACE(sizeof(u8))]; + } u; + int ret; + + msg->msg_control = &u; + msg->msg_controllen = sizeof(u); + ret = sock_recvmsg(svsk->sk_sock, msg, MSG_DONTWAIT); + if (unlikely(msg->msg_controllen != sizeof(u))) + ret = svc_tcp_sock_process_cmsg(svsk, msg, &u.cmsg, ret); + return ret; +} + #if ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE static void svc_flush_bvec(const struct bio_vec *bvec, size_t size, size_t seek) { @@ -263,7 +307,7 @@ static ssize_t svc_tcp_read_msg(struct svc_rqst *rqstp, size_t buflen, iov_iter_advance(&msg.msg_iter, seek); buflen -= seek; } - len = sock_recvmsg(svsk->sk_sock, &msg, MSG_DONTWAIT); + len = svc_tcp_sock_recv_cmsg(svsk, &msg); if (len > 0) svc_flush_bvec(bvec, len, seek); @@ -877,7 +921,7 @@ static ssize_t svc_tcp_read_marker(struct svc_sock *svsk, iov.iov_base = ((char *)&svsk->sk_marker) + svsk->sk_tcplen; iov.iov_len = want; iov_iter_kvec(&msg.msg_iter, ITER_DEST, &iov, 1, want); - len = sock_recvmsg(svsk->sk_sock, &msg, MSG_DONTWAIT); + len = svc_tcp_sock_recv_cmsg(svsk, &msg); if (len < 0) return len; svsk->sk_tcplen += len; -- cgit v1.2.3 From 695bc1f32c6bc806218b322096b23dfd601e59ca Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 17 Apr 2023 22:15:28 -0400 Subject: SUNRPC: Clear rq_xid when receiving a new RPC Call This is an eye-catcher for tracepoints that record the XID: it means svc_rqst() has not received a full RPC Call with an XID yet. Signed-off-by: Chuck Lever --- net/sunrpc/svc_xprt.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index e3952b690f54..3b9708b39e35 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -701,6 +701,8 @@ static int svc_alloc_arg(struct svc_rqst *rqstp) arg->page_len = (pages-2)*PAGE_SIZE; arg->len = (pages-1)*PAGE_SIZE; arg->tail[0].iov_len = 0; + + rqstp->rq_xid = xdr_zero; return 0; } -- cgit v1.2.3 From b3cbf98e2fdf3cb147a95161560cd25987284330 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 20 Apr 2023 13:56:24 -0400 Subject: SUNRPC: Support TLS handshake in the server-side TCP socket code This patch adds opportunitistic RPC-with-TLS to the Linux in-kernel NFS server. If the client requests RPC-with-TLS and the user space handshake agent is running, the server will set up a TLS session. There are no policy settings yet. For example, the server cannot yet require the use of RPC-with-TLS to access its data. Signed-off-by: Chuck Lever --- include/linux/sunrpc/svc_xprt.h | 5 +- include/linux/sunrpc/svcsock.h | 2 + include/trace/events/sunrpc.h | 16 ++++++- net/sunrpc/svc_xprt.c | 5 +- net/sunrpc/svcauth_unix.c | 11 ++++- net/sunrpc/svcsock.c | 101 ++++++++++++++++++++++++++++++++++++++-- 6 files changed, 132 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index 775368802762..867479204840 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -27,7 +27,7 @@ struct svc_xprt_ops { void (*xpo_detach)(struct svc_xprt *); void (*xpo_free)(struct svc_xprt *); void (*xpo_kill_temp_xprt)(struct svc_xprt *); - void (*xpo_start_tls)(struct svc_xprt *); + void (*xpo_handshake)(struct svc_xprt *xprt); }; struct svc_xprt_class { @@ -70,6 +70,9 @@ struct svc_xprt { #define XPT_LOCAL 12 /* connection from loopback interface */ #define XPT_KILL_TEMP 13 /* call xpo_kill_temp_xprt before closing */ #define XPT_CONG_CTRL 14 /* has congestion control */ +#define XPT_HANDSHAKE 15 /* xprt requests a handshake */ +#define XPT_TLS_SESSION 16 /* transport-layer security established */ +#define XPT_PEER_AUTH 17 /* peer has been authenticated */ struct svc_serv *xpt_server; /* service for transport */ atomic_t xpt_reserved; /* space on outq that is rsvd */ diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h index dd73fa174af5..d16ae621782c 100644 --- a/include/linux/sunrpc/svcsock.h +++ b/include/linux/sunrpc/svcsock.h @@ -38,6 +38,8 @@ struct svc_sock { /* Number of queued send requests */ atomic_t sk_sendqlen; + struct completion sk_handshake_done; + struct page * sk_pages[RPCSVC_MAXPAGES]; /* received data */ }; diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h index 5a3bb42e1f50..31bc7025cb44 100644 --- a/include/trace/events/sunrpc.h +++ b/include/trace/events/sunrpc.h @@ -1857,7 +1857,10 @@ TRACE_EVENT(svc_stats_latency, { BIT(XPT_CACHE_AUTH), "CACHE_AUTH" }, \ { BIT(XPT_LOCAL), "LOCAL" }, \ { BIT(XPT_KILL_TEMP), "KILL_TEMP" }, \ - { BIT(XPT_CONG_CTRL), "CONG_CTRL" }) + { BIT(XPT_CONG_CTRL), "CONG_CTRL" }, \ + { BIT(XPT_HANDSHAKE), "HANDSHAKE" }, \ + { BIT(XPT_TLS_SESSION), "TLS_SESSION" }, \ + { BIT(XPT_PEER_AUTH), "PEER_AUTH" }) TRACE_EVENT(svc_xprt_create_err, TP_PROTO( @@ -1990,6 +1993,17 @@ DEFINE_SVC_XPRT_EVENT(close); DEFINE_SVC_XPRT_EVENT(detach); DEFINE_SVC_XPRT_EVENT(free); +#define DEFINE_SVC_TLS_EVENT(name) \ + DEFINE_EVENT(svc_xprt_event, svc_tls_##name, \ + TP_PROTO(const struct svc_xprt *xprt), \ + TP_ARGS(xprt)) + +DEFINE_SVC_TLS_EVENT(start); +DEFINE_SVC_TLS_EVENT(upcall); +DEFINE_SVC_TLS_EVENT(unavailable); +DEFINE_SVC_TLS_EVENT(not_started); +DEFINE_SVC_TLS_EVENT(timed_out); + TRACE_EVENT(svc_xprt_accept, TP_PROTO( const struct svc_xprt *xprt, diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 3b9708b39e35..84e5d7d31481 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -427,7 +427,7 @@ static bool svc_xprt_ready(struct svc_xprt *xprt) if (xpt_flags & BIT(XPT_BUSY)) return false; - if (xpt_flags & (BIT(XPT_CONN) | BIT(XPT_CLOSE))) + if (xpt_flags & (BIT(XPT_CONN) | BIT(XPT_CLOSE) | BIT(XPT_HANDSHAKE))) return true; if (xpt_flags & (BIT(XPT_DATA) | BIT(XPT_DEFERRED))) { if (xprt->xpt_ops->xpo_has_wspace(xprt) && @@ -828,6 +828,9 @@ static int svc_handle_xprt(struct svc_rqst *rqstp, struct svc_xprt *xprt) module_put(xprt->xpt_class->xcl_owner); } svc_xprt_received(xprt); + } else if (test_bit(XPT_HANDSHAKE, &xprt->xpt_flags)) { + xprt->xpt_ops->xpo_handshake(xprt); + svc_xprt_received(xprt); } else if (svc_xprt_reserve_slot(rqstp, xprt)) { /* XPT_DATA|XPT_DEFERRED case: */ dprintk("svc: server %p, pool %u, transport %p, inuse=%d\n", diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c index 4485088ce27b..174783f804fa 100644 --- a/net/sunrpc/svcauth_unix.c +++ b/net/sunrpc/svcauth_unix.c @@ -17,8 +17,9 @@ #include #include #include -#define RPCDBG_FACILITY RPCDBG_AUTH +#include +#define RPCDBG_FACILITY RPCDBG_AUTH #include "netns.h" @@ -832,6 +833,7 @@ svcauth_tls_accept(struct svc_rqst *rqstp) { struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct svc_cred *cred = &rqstp->rq_cred; + struct svc_xprt *xprt = rqstp->rq_xprt; u32 flavor, len; void *body; __be32 *p; @@ -865,14 +867,19 @@ svcauth_tls_accept(struct svc_rqst *rqstp) if (cred->cr_group_info == NULL) return SVC_CLOSE; - if (rqstp->rq_xprt->xpt_ops->xpo_start_tls) { + if (xprt->xpt_ops->xpo_handshake) { p = xdr_reserve_space(&rqstp->rq_res_stream, XDR_UNIT * 2 + 8); if (!p) return SVC_CLOSE; + trace_svc_tls_start(xprt); *p++ = rpc_auth_null; *p++ = cpu_to_be32(8); memcpy(p, "STARTTLS", 8); + + set_bit(XPT_HANDSHAKE, &xprt->xpt_flags); + svc_xprt_enqueue(xprt); } else { + trace_svc_tls_unavailable(xprt); if (xdr_stream_encode_opaque_auth(&rqstp->rq_res_stream, RPC_AUTH_NULL, NULL, 0) < 0) return SVC_CLOSE; diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index c5b74f523fc4..a51c9b989d58 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -44,9 +44,11 @@ #include #include #include +#include #include #include #include +#include #include #include @@ -64,6 +66,12 @@ #define RPCDBG_FACILITY RPCDBG_SVCXPRT +/* To-do: to avoid tying up an nfsd thread while waiting for a + * handshake request, the request could instead be deferred. + */ +enum { + SVC_HANDSHAKE_TO = 5U * HZ +}; static struct svc_sock *svc_setup_socket(struct svc_serv *, struct socket *, int flags); @@ -359,6 +367,8 @@ static void svc_data_ready(struct sock *sk) rmb(); svsk->sk_odata(sk); trace_svcsock_data_ready(&svsk->sk_xprt, 0); + if (test_bit(XPT_HANDSHAKE, &svsk->sk_xprt.xpt_flags)) + return; if (!test_and_set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags)) svc_xprt_enqueue(&svsk->sk_xprt); } @@ -396,6 +406,88 @@ static void svc_tcp_kill_temp_xprt(struct svc_xprt *xprt) sock_no_linger(svsk->sk_sock->sk); } +/** + * svc_tcp_handshake_done - Handshake completion handler + * @data: address of xprt to wake + * @status: status of handshake + * @peerid: serial number of key containing the remote peer's identity + * + * If a security policy is specified as an export option, we don't + * have a specific export here to check. So we set a "TLS session + * is present" flag on the xprt and let an upper layer enforce local + * security policy. + */ +static void svc_tcp_handshake_done(void *data, int status, key_serial_t peerid) +{ + struct svc_xprt *xprt = data; + struct svc_sock *svsk = container_of(xprt, struct svc_sock, sk_xprt); + + if (!status) { + if (peerid != TLS_NO_PEERID) + set_bit(XPT_PEER_AUTH, &xprt->xpt_flags); + set_bit(XPT_TLS_SESSION, &xprt->xpt_flags); + } + clear_bit(XPT_HANDSHAKE, &xprt->xpt_flags); + complete_all(&svsk->sk_handshake_done); +} + +/** + * svc_tcp_handshake - Perform a transport-layer security handshake + * @xprt: connected transport endpoint + * + */ +static void svc_tcp_handshake(struct svc_xprt *xprt) +{ + struct svc_sock *svsk = container_of(xprt, struct svc_sock, sk_xprt); + struct sock *sk = svsk->sk_sock->sk; + struct tls_handshake_args args = { + .ta_sock = svsk->sk_sock, + .ta_done = svc_tcp_handshake_done, + .ta_data = xprt, + }; + int ret; + + trace_svc_tls_upcall(xprt); + + clear_bit(XPT_TLS_SESSION, &xprt->xpt_flags); + init_completion(&svsk->sk_handshake_done); + + ret = tls_server_hello_x509(&args, GFP_KERNEL); + if (ret) { + trace_svc_tls_not_started(xprt); + goto out_failed; + } + + ret = wait_for_completion_interruptible_timeout(&svsk->sk_handshake_done, + SVC_HANDSHAKE_TO); + if (ret <= 0) { + if (tls_handshake_cancel(sk)) { + trace_svc_tls_timed_out(xprt); + goto out_close; + } + } + + if (!test_bit(XPT_TLS_SESSION, &xprt->xpt_flags)) { + trace_svc_tls_unavailable(xprt); + goto out_close; + } + + /* Mark the transport ready in case the remote sent RPC + * traffic before the kernel received the handshake + * completion downcall. + */ + set_bit(XPT_DATA, &xprt->xpt_flags); + svc_xprt_enqueue(xprt); + return; + +out_close: + set_bit(XPT_CLOSE, &xprt->xpt_flags); +out_failed: + clear_bit(XPT_HANDSHAKE, &xprt->xpt_flags); + set_bit(XPT_DATA, &xprt->xpt_flags); + svc_xprt_enqueue(xprt); +} + /* * See net/ipv6/ip_sockglue.c : ip_cmsg_recv_pktinfo */ @@ -1257,6 +1349,7 @@ static const struct svc_xprt_ops svc_tcp_ops = { .xpo_has_wspace = svc_tcp_has_wspace, .xpo_accept = svc_tcp_accept, .xpo_kill_temp_xprt = svc_tcp_kill_temp_xprt, + .xpo_handshake = svc_tcp_handshake, }; static struct svc_xprt_class svc_tcp_class = { @@ -1580,10 +1673,12 @@ static void svc_tcp_sock_detach(struct svc_xprt *xprt) static void svc_sock_free(struct svc_xprt *xprt) { struct svc_sock *svsk = container_of(xprt, struct svc_sock, sk_xprt); + struct socket *sock = svsk->sk_sock; - if (svsk->sk_sock->file) - sockfd_put(svsk->sk_sock); + tls_handshake_cancel(sock->sk); + if (sock->file) + sockfd_put(sock); else - sock_release(svsk->sk_sock); + sock_release(sock); kfree(svsk); } -- cgit v1.2.3 From 6f75cd166a5a3c0bc50441faa8b8304f60522fdd Mon Sep 17 00:00:00 2001 From: Cosmo Chou Date: Wed, 26 Apr 2023 16:13:50 +0800 Subject: net/ncsi: clear Tx enable mode when handling a Config required AEN ncsi_channel_is_tx() determines whether a given channel should be used for Tx or not. However, when reconfiguring the channel by handling a Configuration Required AEN, there is a misjudgment that the channel Tx has already been enabled, which results in the Enable Channel Network Tx command not being sent. Clear the channel Tx enable flag before reconfiguring the channel to avoid the misjudgment. Fixes: 8d951a75d022 ("net/ncsi: Configure multi-package, multi-channel modes with failover") Signed-off-by: Cosmo Chou Signed-off-by: David S. Miller --- net/ncsi/ncsi-aen.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/ncsi/ncsi-aen.c b/net/ncsi/ncsi-aen.c index b635c194f0a8..62fb1031763d 100644 --- a/net/ncsi/ncsi-aen.c +++ b/net/ncsi/ncsi-aen.c @@ -165,6 +165,7 @@ static int ncsi_aen_handler_cr(struct ncsi_dev_priv *ndp, nc->state = NCSI_CHANNEL_INACTIVE; list_add_tail_rcu(&nc->link, &ndp->channel_queue); spin_unlock_irqrestore(&ndp->lock, flags); + nc->modes[NCSI_MODE_TX_ENABLE].enable = 0; return ncsi_process_next_channel(ndp); } -- cgit v1.2.3 From 7e692df3933628d974acb9f5b334d2b3e885e2a6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 28 Apr 2023 04:32:31 +0000 Subject: tcp: fix skb_copy_ubufs() vs BIG TCP David Ahern reported crashes in skb_copy_ubufs() caused by TCP tx zerocopy using hugepages, and skb length bigger than ~68 KB. skb_copy_ubufs() assumed it could copy all payload using up to MAX_SKB_FRAGS order-0 pages. This assumption broke when BIG TCP was able to put up to 512 KB per skb. We did not hit this bug at Google because we use CONFIG_MAX_SKB_FRAGS=45 and limit gso_max_size to 180000. A solution is to use higher order pages if needed. v2: add missing __GFP_COMP, or we leak memory. Fixes: 7c4e983c4f3c ("net: allow gso_max_size to exceed 65536") Reported-by: David Ahern Link: https://lore.kernel.org/netdev/c70000f6-baa4-4a05-46d0-4b3e0dc1ccc8@gmail.com/T/ Signed-off-by: Eric Dumazet Cc: Xin Long Cc: Willem de Bruijn Cc: Coco Li Signed-off-by: David S. Miller --- net/core/skbuff.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 2112146092bf..26a586007d8b 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1758,7 +1758,7 @@ int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask) { int num_frags = skb_shinfo(skb)->nr_frags; struct page *page, *head = NULL; - int i, new_frags; + int i, order, psize, new_frags; u32 d_off; if (skb_shared(skb) || skb_unclone(skb, gfp_mask)) @@ -1767,9 +1767,17 @@ int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask) if (!num_frags) goto release; - new_frags = (__skb_pagelen(skb) + PAGE_SIZE - 1) >> PAGE_SHIFT; + /* We might have to allocate high order pages, so compute what minimum + * page order is needed. + */ + order = 0; + while ((PAGE_SIZE << order) * MAX_SKB_FRAGS < __skb_pagelen(skb)) + order++; + psize = (PAGE_SIZE << order); + + new_frags = (__skb_pagelen(skb) + psize - 1) >> (PAGE_SHIFT + order); for (i = 0; i < new_frags; i++) { - page = alloc_page(gfp_mask); + page = alloc_pages(gfp_mask | __GFP_COMP, order); if (!page) { while (head) { struct page *next = (struct page *)page_private(head); @@ -1796,11 +1804,11 @@ int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask) vaddr = kmap_atomic(p); while (done < p_len) { - if (d_off == PAGE_SIZE) { + if (d_off == psize) { d_off = 0; page = (struct page *)page_private(page); } - copy = min_t(u32, PAGE_SIZE - d_off, p_len - done); + copy = min_t(u32, psize - d_off, p_len - done); memcpy(page_address(page) + d_off, vaddr + p_off + done, copy); done += copy; @@ -1816,7 +1824,7 @@ int skb_copy_ubufs(struct sk_buff *skb, gfp_t gfp_mask) /* skb frags point to kernel buffers */ for (i = 0; i < new_frags - 1; i++) { - __skb_fill_page_desc(skb, i, head, 0, PAGE_SIZE); + __skb_fill_page_desc(skb, i, head, 0, psize); head = (struct page *)page_private(head); } __skb_fill_page_desc(skb, new_frags - 1, head, 0, d_off); -- cgit v1.2.3 From da94a7781fc3c92e7df7832bc2746f4d39bc624e Mon Sep 17 00:00:00 2001 From: Vlad Buslov Date: Wed, 26 Apr 2023 14:31:11 +0200 Subject: net/sched: cls_api: remove block_cb from driver_list before freeing Error handler of tcf_block_bind() frees the whole bo->cb_list on error. However, by that time the flow_block_cb instances are already in the driver list because driver ndo_setup_tc() callback is called before that up the call chain in tcf_block_offload_cmd(). This leaves dangling pointers to freed objects in the list and causes use-after-free[0]. Fix it by also removing flow_block_cb instances from driver_list before deallocating them. [0]: [ 279.868433] ================================================================== [ 279.869964] BUG: KASAN: slab-use-after-free in flow_block_cb_setup_simple+0x631/0x7c0 [ 279.871527] Read of size 8 at addr ffff888147e2bf20 by task tc/2963 [ 279.873151] CPU: 6 PID: 2963 Comm: tc Not tainted 6.3.0-rc6+ #4 [ 279.874273] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 [ 279.876295] Call Trace: [ 279.876882] [ 279.877413] dump_stack_lvl+0x33/0x50 [ 279.878198] print_report+0xc2/0x610 [ 279.878987] ? flow_block_cb_setup_simple+0x631/0x7c0 [ 279.879994] kasan_report+0xae/0xe0 [ 279.880750] ? flow_block_cb_setup_simple+0x631/0x7c0 [ 279.881744] ? mlx5e_tc_reoffload_flows_work+0x240/0x240 [mlx5_core] [ 279.883047] flow_block_cb_setup_simple+0x631/0x7c0 [ 279.884027] tcf_block_offload_cmd.isra.0+0x189/0x2d0 [ 279.885037] ? tcf_block_setup+0x6b0/0x6b0 [ 279.885901] ? mutex_lock+0x7d/0xd0 [ 279.886669] ? __mutex_unlock_slowpath.constprop.0+0x2d0/0x2d0 [ 279.887844] ? ingress_init+0x1c0/0x1c0 [sch_ingress] [ 279.888846] tcf_block_get_ext+0x61c/0x1200 [ 279.889711] ingress_init+0x112/0x1c0 [sch_ingress] [ 279.890682] ? clsact_init+0x2b0/0x2b0 [sch_ingress] [ 279.891701] qdisc_create+0x401/0xea0 [ 279.892485] ? qdisc_tree_reduce_backlog+0x470/0x470 [ 279.893473] tc_modify_qdisc+0x6f7/0x16d0 [ 279.894344] ? tc_get_qdisc+0xac0/0xac0 [ 279.895213] ? mutex_lock+0x7d/0xd0 [ 279.896005] ? __mutex_lock_slowpath+0x10/0x10 [ 279.896910] rtnetlink_rcv_msg+0x5fe/0x9d0 [ 279.897770] ? rtnl_calcit.isra.0+0x2b0/0x2b0 [ 279.898672] ? __sys_sendmsg+0xb5/0x140 [ 279.899494] ? do_syscall_64+0x3d/0x90 [ 279.900302] ? entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 279.901337] ? kasan_save_stack+0x2e/0x40 [ 279.902177] ? kasan_save_stack+0x1e/0x40 [ 279.903058] ? kasan_set_track+0x21/0x30 [ 279.903913] ? kasan_save_free_info+0x2a/0x40 [ 279.904836] ? ____kasan_slab_free+0x11a/0x1b0 [ 279.905741] ? kmem_cache_free+0x179/0x400 [ 279.906599] netlink_rcv_skb+0x12c/0x360 [ 279.907450] ? rtnl_calcit.isra.0+0x2b0/0x2b0 [ 279.908360] ? netlink_ack+0x1550/0x1550 [ 279.909192] ? rhashtable_walk_peek+0x170/0x170 [ 279.910135] ? kmem_cache_alloc_node+0x1af/0x390 [ 279.911086] ? _copy_from_iter+0x3d6/0xc70 [ 279.912031] netlink_unicast+0x553/0x790 [ 279.912864] ? netlink_attachskb+0x6a0/0x6a0 [ 279.913763] ? netlink_recvmsg+0x416/0xb50 [ 279.914627] netlink_sendmsg+0x7a1/0xcb0 [ 279.915473] ? netlink_unicast+0x790/0x790 [ 279.916334] ? iovec_from_user.part.0+0x4d/0x220 [ 279.917293] ? netlink_unicast+0x790/0x790 [ 279.918159] sock_sendmsg+0xc5/0x190 [ 279.918938] ____sys_sendmsg+0x535/0x6b0 [ 279.919813] ? import_iovec+0x7/0x10 [ 279.920601] ? kernel_sendmsg+0x30/0x30 [ 279.921423] ? __copy_msghdr+0x3c0/0x3c0 [ 279.922254] ? import_iovec+0x7/0x10 [ 279.923041] ___sys_sendmsg+0xeb/0x170 [ 279.923854] ? copy_msghdr_from_user+0x110/0x110 [ 279.924797] ? ___sys_recvmsg+0xd9/0x130 [ 279.925630] ? __perf_event_task_sched_in+0x183/0x470 [ 279.926656] ? ___sys_sendmsg+0x170/0x170 [ 279.927529] ? ctx_sched_in+0x530/0x530 [ 279.928369] ? update_curr+0x283/0x4f0 [ 279.929185] ? perf_event_update_userpage+0x570/0x570 [ 279.930201] ? __fget_light+0x57/0x520 [ 279.931023] ? __switch_to+0x53d/0xe70 [ 279.931846] ? sockfd_lookup_light+0x1a/0x140 [ 279.932761] __sys_sendmsg+0xb5/0x140 [ 279.933560] ? __sys_sendmsg_sock+0x20/0x20 [ 279.934436] ? fpregs_assert_state_consistent+0x1d/0xa0 [ 279.935490] do_syscall_64+0x3d/0x90 [ 279.936300] entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 279.937311] RIP: 0033:0x7f21c814f887 [ 279.938085] Code: 0a 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b9 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10 [ 279.941448] RSP: 002b:00007fff11efd478 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [ 279.942964] RAX: ffffffffffffffda RBX: 0000000064401979 RCX: 00007f21c814f887 [ 279.944337] RDX: 0000000000000000 RSI: 00007fff11efd4e0 RDI: 0000000000000003 [ 279.945660] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000 [ 279.947003] R10: 00007f21c8008708 R11: 0000000000000246 R12: 0000000000000001 [ 279.948345] R13: 0000000000409980 R14: 000000000047e538 R15: 0000000000485400 [ 279.949690] [ 279.950706] Allocated by task 2960: [ 279.951471] kasan_save_stack+0x1e/0x40 [ 279.952338] kasan_set_track+0x21/0x30 [ 279.953165] __kasan_kmalloc+0x77/0x90 [ 279.954006] flow_block_cb_setup_simple+0x3dd/0x7c0 [ 279.955001] tcf_block_offload_cmd.isra.0+0x189/0x2d0 [ 279.956020] tcf_block_get_ext+0x61c/0x1200 [ 279.956881] ingress_init+0x112/0x1c0 [sch_ingress] [ 279.957873] qdisc_create+0x401/0xea0 [ 279.958656] tc_modify_qdisc+0x6f7/0x16d0 [ 279.959506] rtnetlink_rcv_msg+0x5fe/0x9d0 [ 279.960392] netlink_rcv_skb+0x12c/0x360 [ 279.961216] netlink_unicast+0x553/0x790 [ 279.962044] netlink_sendmsg+0x7a1/0xcb0 [ 279.962906] sock_sendmsg+0xc5/0x190 [ 279.963702] ____sys_sendmsg+0x535/0x6b0 [ 279.964534] ___sys_sendmsg+0xeb/0x170 [ 279.965343] __sys_sendmsg+0xb5/0x140 [ 279.966132] do_syscall_64+0x3d/0x90 [ 279.966908] entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 279.968407] Freed by task 2960: [ 279.969114] kasan_save_stack+0x1e/0x40 [ 279.969929] kasan_set_track+0x21/0x30 [ 279.970729] kasan_save_free_info+0x2a/0x40 [ 279.971603] ____kasan_slab_free+0x11a/0x1b0 [ 279.972483] __kmem_cache_free+0x14d/0x280 [ 279.973337] tcf_block_setup+0x29d/0x6b0 [ 279.974173] tcf_block_offload_cmd.isra.0+0x226/0x2d0 [ 279.975186] tcf_block_get_ext+0x61c/0x1200 [ 279.976080] ingress_init+0x112/0x1c0 [sch_ingress] [ 279.977065] qdisc_create+0x401/0xea0 [ 279.977857] tc_modify_qdisc+0x6f7/0x16d0 [ 279.978695] rtnetlink_rcv_msg+0x5fe/0x9d0 [ 279.979562] netlink_rcv_skb+0x12c/0x360 [ 279.980388] netlink_unicast+0x553/0x790 [ 279.981214] netlink_sendmsg+0x7a1/0xcb0 [ 279.982043] sock_sendmsg+0xc5/0x190 [ 279.982827] ____sys_sendmsg+0x535/0x6b0 [ 279.983703] ___sys_sendmsg+0xeb/0x170 [ 279.984510] __sys_sendmsg+0xb5/0x140 [ 279.985298] do_syscall_64+0x3d/0x90 [ 279.986076] entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 279.987532] The buggy address belongs to the object at ffff888147e2bf00 which belongs to the cache kmalloc-192 of size 192 [ 279.989747] The buggy address is located 32 bytes inside of freed 192-byte region [ffff888147e2bf00, ffff888147e2bfc0) [ 279.992367] The buggy address belongs to the physical page: [ 279.993430] page:00000000550f405c refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x147e2a [ 279.995182] head:00000000550f405c order:1 entire_mapcount:0 nr_pages_mapped:0 pincount:0 [ 279.996713] anon flags: 0x200000000010200(slab|head|node=0|zone=2) [ 279.997878] raw: 0200000000010200 ffff888100042a00 0000000000000000 dead000000000001 [ 279.999384] raw: 0000000000000000 0000000000200020 00000001ffffffff 0000000000000000 [ 280.000894] page dumped because: kasan: bad access detected [ 280.002386] Memory state around the buggy address: [ 280.003338] ffff888147e2be00: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ 280.004781] ffff888147e2be80: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc [ 280.006224] >ffff888147e2bf00: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ 280.007700] ^ [ 280.008592] ffff888147e2bf80: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc [ 280.010035] ffff888147e2c000: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ 280.011564] ================================================================== Fixes: 59094b1e5094 ("net: sched: use flow block API") Signed-off-by: Vlad Buslov Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/cls_api.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 3c3629c9e7b6..2621550bfddc 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1589,6 +1589,7 @@ static int tcf_block_bind(struct tcf_block *block, err_unroll: list_for_each_entry_safe(block_cb, next, &bo->cb_list, list) { + list_del(&block_cb->driver_list); if (i-- > 0) { list_del(&block_cb->list); tcf_block_playback_offloads(block, block_cb->cb, -- cgit v1.2.3 From c88f8d5cd95fd039cff95d682b8e71100c001df0 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Wed, 26 Apr 2023 23:00:06 -0700 Subject: sit: update dev->needed_headroom in ipip6_tunnel_bind_dev() When a tunnel device is bound with the underlying device, its dev->needed_headroom needs to be updated properly. IPv4 tunnels already do the same in ip_tunnel_bind_dev(). Otherwise we may not have enough header room for skb, especially after commit b17f709a2401 ("gue: TX support for using remote checksum offload option"). Fixes: 32b8a8e59c9c ("sit: add IPv4 over IPv4 support") Reported-by: Palash Oswal Link: https://lore.kernel.org/netdev/CAGyP=7fDcSPKu6nttbGwt7RXzE3uyYxLjCSE97J64pRxJP8jPA@mail.gmail.com/ Cc: Kuniyuki Iwashima Cc: Eric Dumazet Signed-off-by: Cong Wang Reviewed-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Signed-off-by: David S. Miller --- net/ipv6/sit.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 063560e2cb1a..cc24cefdb85c 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -1095,12 +1095,13 @@ tx_err: static void ipip6_tunnel_bind_dev(struct net_device *dev) { + struct ip_tunnel *tunnel = netdev_priv(dev); + int t_hlen = tunnel->hlen + sizeof(struct iphdr); struct net_device *tdev = NULL; - struct ip_tunnel *tunnel; + int hlen = LL_MAX_HEADER; const struct iphdr *iph; struct flowi4 fl4; - tunnel = netdev_priv(dev); iph = &tunnel->parms.iph; if (iph->daddr) { @@ -1123,14 +1124,15 @@ static void ipip6_tunnel_bind_dev(struct net_device *dev) tdev = __dev_get_by_index(tunnel->net, tunnel->parms.link); if (tdev && !netif_is_l3_master(tdev)) { - int t_hlen = tunnel->hlen + sizeof(struct iphdr); int mtu; mtu = tdev->mtu - t_hlen; if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; WRITE_ONCE(dev->mtu, mtu); + hlen = tdev->hard_header_len + tdev->needed_headroom; } + dev->needed_headroom = t_hlen + hlen; } static void ipip6_tunnel_update(struct ip_tunnel *t, struct ip_tunnel_parm *p, -- cgit v1.2.3 From dc6456e938e938d64ffb6383a286b2ac9790a37f Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Thu, 27 Apr 2023 11:21:59 +0200 Subject: net: ipv6: fix skb hash for some RST packets The skb hash comes from sk->sk_txhash when using TCP, except for some IPv6 RST packets. This is because in tcp_v6_send_reset when not in TIME_WAIT the hash is taken from sk->sk_hash, while it should come from sk->sk_txhash as those two hashes are not computed the same way. Packetdrill script to test the above, 0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3 +0 fcntl(3, F_SETFL, O_RDWR|O_NONBLOCK) = 0 +0 connect(3, ..., ...) = -1 EINPROGRESS (Operation now in progress) +0 > (flowlabel 0x1) S 0:0(0) <...> // Wrong ack seq, trigger a rst. +0 < S. 0:0(0) ack 0 win 4000 // Check the flowlabel matches prior one from SYN. +0 > (flowlabel 0x1) R 0:0(0) <...> Fixes: 9258b8b1be2e ("ipv6: tcp: send consistent autoflowlabel in RST packets") Signed-off-by: Antoine Tenart Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/tcp_ipv6.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 244cf86c4cbb..7132eb213a7a 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1065,7 +1065,7 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb) if (np->repflow) label = ip6_flowlabel(ipv6h); priority = sk->sk_priority; - txhash = sk->sk_hash; + txhash = sk->sk_txhash; } if (sk->sk_state == TCP_TIME_WAIT) { label = cpu_to_be32(inet_twsk(sk)->tw_flowlabel); -- cgit v1.2.3 From 526f28bd0fbdc699cda31426928802650c1528e5 Mon Sep 17 00:00:00 2001 From: Victor Nogueira Date: Wed, 26 Apr 2023 15:19:40 +0000 Subject: net/sched: act_mirred: Add carrier check There are cases where the device is adminstratively UP, but operationally down. For example, we have a physical device (Nvidia ConnectX-6 Dx, 25Gbps) who's cable was pulled out, here is its ip link output: 5: ens2f1: mtu 1500 qdisc mq state DOWN mode DEFAULT group default qlen 1000 link/ether b8:ce:f6:4b:68:35 brd ff:ff:ff:ff:ff:ff altname enp179s0f1np1 As you can see, it's administratively UP but operationally down. In this case, sending a packet to this port caused a nasty kernel hang (so nasty that we were unable to capture it). Aborting a transmit based on operational status (in addition to administrative status) fixes the issue. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Acked-by: Jamal Hadi Salim Signed-off-by: Victor Nogueira v1->v2: Add fixes tag v2->v3: Remove blank line between tags + add change log, suggested by Leon Signed-off-by: David S. Miller --- net/sched/act_mirred.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index ec43764e92e7..0a711c184c29 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -264,7 +264,7 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, goto out; } - if (unlikely(!(dev->flags & IFF_UP))) { + if (unlikely(!(dev->flags & IFF_UP)) || !netif_carrier_ok(dev)) { net_notice_ratelimited("tc mirred to Houston: device %s is down\n", dev->name); goto out; -- cgit v1.2.3 From 0d098d83c5d9e107b2df7f5e11f81492f56d2fe7 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 28 Apr 2023 21:27:54 +0100 Subject: rxrpc: Fix hard call timeout units The hard call timeout is specified in the RXRPC_SET_CALL_TIMEOUT cmsg in seconds, so fix the point at which sendmsg() applies it to the call to convert to jiffies from seconds, not milliseconds. Fixes: a158bdd3247b ("rxrpc: Fix timeout of a call that hasn't yet been granted a channel") Signed-off-by: David Howells cc: Marc Dionne cc: "David S. Miller" cc: Eric Dumazet cc: Jakub Kicinski cc: Paolo Abeni cc: linux-afs@lists.infradead.org cc: netdev@vger.kernel.org cc: linux-kernel@vger.kernel.org Signed-off-by: David S. Miller --- net/rxrpc/sendmsg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index 6caa47d352ed..7498a77b5d39 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -699,7 +699,7 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) fallthrough; case 1: if (p.call.timeouts.hard > 0) { - j = msecs_to_jiffies(p.call.timeouts.hard); + j = p.call.timeouts.hard * HZ; now = jiffies; j += now; WRITE_ONCE(call->expect_term_by, j); -- cgit v1.2.3 From 0eb362d254814ce04848730bf32e75b8ee1a4d6c Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 28 Apr 2023 21:27:55 +0100 Subject: rxrpc: Make it so that a waiting process can be aborted When sendmsg() creates an rxrpc call, it queues it to wait for a connection and channel to be assigned and then waits before it can start shovelling data as the encrypted DATA packet content includes a summary of the connection parameters. However, sendmsg() may get interrupted before a connection gets assigned and further sendmsg() calls will fail with EBUSY until an assignment is made. Fix this so that the call can at least be aborted without failing on EBUSY. We have to be careful here as sendmsg() mustn't be allowed to start the call timer if the call doesn't yet have a connection assigned as an oops may follow shortly thereafter. Fixes: 540b1c48c37a ("rxrpc: Fix deadlock between call creation and sendmsg/recvmsg") Reported-by: Marc Dionne Signed-off-by: David Howells cc: "David S. Miller" cc: Eric Dumazet cc: Jakub Kicinski cc: Paolo Abeni cc: linux-afs@lists.infradead.org cc: netdev@vger.kernel.org cc: linux-kernel@vger.kernel.org Signed-off-by: David S. Miller --- net/rxrpc/sendmsg.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index 7498a77b5d39..c1b074c17b33 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -656,10 +656,13 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) goto out_put_unlock; } else { switch (rxrpc_call_state(call)) { - case RXRPC_CALL_UNINITIALISED: case RXRPC_CALL_CLIENT_AWAIT_CONN: - case RXRPC_CALL_SERVER_PREALLOC: case RXRPC_CALL_SERVER_SECURING: + if (p.command == RXRPC_CMD_SEND_ABORT) + break; + fallthrough; + case RXRPC_CALL_UNINITIALISED: + case RXRPC_CALL_SERVER_PREALLOC: rxrpc_put_call(call, rxrpc_call_put_sendmsg); ret = -EBUSY; goto error_release_sock; -- cgit v1.2.3 From db099c625b13a74d462521a46d98a8ce5b53af5d Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 28 Apr 2023 21:27:56 +0100 Subject: rxrpc: Fix timeout of a call that hasn't yet been granted a channel afs_make_call() calls rxrpc_kernel_begin_call() to begin a call (which may get stalled in the background waiting for a connection to become available); it then calls rxrpc_kernel_set_max_life() to set the timeouts - but that starts the call timer so the call timer might then expire before we get a connection assigned - leading to the following oops if the call stalled: BUG: kernel NULL pointer dereference, address: 0000000000000000 ... CPU: 1 PID: 5111 Comm: krxrpcio/0 Not tainted 6.3.0-rc7-build3+ #701 RIP: 0010:rxrpc_alloc_txbuf+0xc0/0x157 ... Call Trace: rxrpc_send_ACK+0x50/0x13b rxrpc_input_call_event+0x16a/0x67d rxrpc_io_thread+0x1b6/0x45f ? _raw_spin_unlock_irqrestore+0x1f/0x35 ? rxrpc_input_packet+0x519/0x519 kthread+0xe7/0xef ? kthread_complete_and_exit+0x1b/0x1b ret_from_fork+0x22/0x30 Fix this by noting the timeouts in struct rxrpc_call when the call is created. The timer will be started when the first packet is transmitted. It shouldn't be possible to trigger this directly from userspace through AF_RXRPC as sendmsg() will return EBUSY if the call is in the waiting-for-conn state if it dropped out of the wait due to a signal. Fixes: 9d35d880e0e4 ("rxrpc: Move client call connection to the I/O thread") Reported-by: Marc Dionne Signed-off-by: David Howells cc: "David S. Miller" cc: Eric Dumazet cc: Jakub Kicinski cc: Paolo Abeni cc: linux-afs@lists.infradead.org cc: netdev@vger.kernel.org cc: linux-kernel@vger.kernel.org Signed-off-by: David S. Miller --- fs/afs/afs.h | 4 ++-- fs/afs/internal.h | 2 +- fs/afs/rxrpc.c | 8 +++----- include/net/af_rxrpc.h | 21 +++++++++++---------- net/rxrpc/af_rxrpc.c | 3 +++ net/rxrpc/ar-internal.h | 1 + net/rxrpc/call_object.c | 9 ++++++++- net/rxrpc/sendmsg.c | 1 + 8 files changed, 30 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/fs/afs/afs.h b/fs/afs/afs.h index 432cb4b23961..81815724db6c 100644 --- a/fs/afs/afs.h +++ b/fs/afs/afs.h @@ -19,8 +19,8 @@ #define AFSPATHMAX 1024 /* Maximum length of a pathname plus NUL */ #define AFSOPAQUEMAX 1024 /* Maximum length of an opaque field */ -#define AFS_VL_MAX_LIFESPAN (120 * HZ) -#define AFS_PROBE_MAX_LIFESPAN (30 * HZ) +#define AFS_VL_MAX_LIFESPAN 120 +#define AFS_PROBE_MAX_LIFESPAN 30 typedef u64 afs_volid_t; typedef u64 afs_vnodeid_t; diff --git a/fs/afs/internal.h b/fs/afs/internal.h index ad8523d0d038..68ae91d21b57 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -128,7 +128,7 @@ struct afs_call { spinlock_t state_lock; int error; /* error code */ u32 abort_code; /* Remote abort ID or 0 */ - unsigned int max_lifespan; /* Maximum lifespan to set if not 0 */ + unsigned int max_lifespan; /* Maximum lifespan in secs to set if not 0 */ unsigned request_size; /* size of request data */ unsigned reply_max; /* maximum size of reply */ unsigned count2; /* count used in unmarshalling */ diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index e08b850c3e6d..ed1644e7683f 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -335,7 +335,9 @@ void afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, gfp_t gfp) /* create a call */ rxcall = rxrpc_kernel_begin_call(call->net->socket, srx, call->key, (unsigned long)call, - tx_total_len, gfp, + tx_total_len, + call->max_lifespan, + gfp, (call->async ? afs_wake_up_async_call : afs_wake_up_call_waiter), @@ -350,10 +352,6 @@ void afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, gfp_t gfp) } call->rxcall = rxcall; - - if (call->max_lifespan) - rxrpc_kernel_set_max_life(call->net->socket, rxcall, - call->max_lifespan); call->issue_time = ktime_get_real(); /* send the request */ diff --git a/include/net/af_rxrpc.h b/include/net/af_rxrpc.h index 01a35e113ab9..5531dd08061e 100644 --- a/include/net/af_rxrpc.h +++ b/include/net/af_rxrpc.h @@ -40,16 +40,17 @@ typedef void (*rxrpc_user_attach_call_t)(struct rxrpc_call *, unsigned long); void rxrpc_kernel_new_call_notification(struct socket *, rxrpc_notify_new_call_t, rxrpc_discard_new_call_t); -struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *, - struct sockaddr_rxrpc *, - struct key *, - unsigned long, - s64, - gfp_t, - rxrpc_notify_rx_t, - bool, - enum rxrpc_interruptibility, - unsigned int); +struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, + struct sockaddr_rxrpc *srx, + struct key *key, + unsigned long user_call_ID, + s64 tx_total_len, + u32 hard_timeout, + gfp_t gfp, + rxrpc_notify_rx_t notify_rx, + bool upgrade, + enum rxrpc_interruptibility interruptibility, + unsigned int debug_id); int rxrpc_kernel_send_data(struct socket *, struct rxrpc_call *, struct msghdr *, size_t, rxrpc_notify_end_tx_t); diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index c32b164206f9..31f738d65f1c 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -265,6 +265,7 @@ static int rxrpc_listen(struct socket *sock, int backlog) * @key: The security context to use (defaults to socket setting) * @user_call_ID: The ID to use * @tx_total_len: Total length of data to transmit during the call (or -1) + * @hard_timeout: The maximum lifespan of the call in sec * @gfp: The allocation constraints * @notify_rx: Where to send notifications instead of socket queue * @upgrade: Request service upgrade for call @@ -283,6 +284,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, struct key *key, unsigned long user_call_ID, s64 tx_total_len, + u32 hard_timeout, gfp_t gfp, rxrpc_notify_rx_t notify_rx, bool upgrade, @@ -313,6 +315,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, p.tx_total_len = tx_total_len; p.interruptibility = interruptibility; p.kernel = true; + p.timeouts.hard = hard_timeout; memset(&cp, 0, sizeof(cp)); cp.local = rx->local; diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 67b0a894162d..5d44dc08f66d 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -616,6 +616,7 @@ struct rxrpc_call { unsigned long expect_term_by; /* When we expect call termination by */ u32 next_rx_timo; /* Timeout for next Rx packet (jif) */ u32 next_req_timo; /* Timeout for next Rx request packet (jif) */ + u32 hard_timo; /* Maximum lifetime or 0 (jif) */ struct timer_list timer; /* Combined event timer */ struct work_struct destroyer; /* In-process-context destroyer */ rxrpc_notify_rx_t notify_rx; /* kernel service Rx notification function */ diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index e9f1f49d18c2..fecbc73054bc 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -226,6 +226,13 @@ static struct rxrpc_call *rxrpc_alloc_client_call(struct rxrpc_sock *rx, if (cp->exclusive) __set_bit(RXRPC_CALL_EXCLUSIVE, &call->flags); + if (p->timeouts.normal) + call->next_rx_timo = min(msecs_to_jiffies(p->timeouts.normal), 1UL); + if (p->timeouts.idle) + call->next_req_timo = min(msecs_to_jiffies(p->timeouts.idle), 1UL); + if (p->timeouts.hard) + call->hard_timo = p->timeouts.hard * HZ; + ret = rxrpc_init_client_call_security(call); if (ret < 0) { rxrpc_prefail_call(call, RXRPC_CALL_LOCAL_ERROR, ret); @@ -257,7 +264,7 @@ void rxrpc_start_call_timer(struct rxrpc_call *call) call->keepalive_at = j; call->expect_rx_by = j; call->expect_req_by = j; - call->expect_term_by = j; + call->expect_term_by = j + call->hard_timo; call->timer.expires = now; } diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index c1b074c17b33..8e0b94714e84 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -651,6 +651,7 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) if (IS_ERR(call)) return PTR_ERR(call); /* ... and we have the call lock. */ + p.call.nr_timeouts = 0; ret = 0; if (rxrpc_call_is_complete(call)) goto out_put_unlock; -- cgit v1.2.3 From af97b7dfb0d4636d58f2341346fffce30c6c2259 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Mon, 1 May 2023 16:04:08 +0200 Subject: SUNRPC: Avoid relying on crypto API to derive CBC-CTS output IV Scott reports SUNRPC self-test failures regarding the output IV on arm64 when using the SIMD accelerated implementation of AES in CBC mode with ciphertext stealing ("cts(cbc(aes))" in crypto API speak). These failures are the result of the fact that, while RFC 3962 does specify what the output IV should be and includes test vectors for it, the general concept of an output IV is poorly defined, and generally, not specified by the various algorithms implemented by the crypto API. Only algorithms that support transparent chaining (e.g., CBC mode on a block boundary) have requirements on the output IV, but ciphertext stealing (CTS) is fundamentally about how to encapsulate CBC in a way where the length of the entire message may not be an integral multiple of the cipher block size, and the concept of an output IV does not exist here because it has no defined purpose past the end of the message. The generic CTS template takes advantage of this chaining capability of the CBC implementations, and as a result, happens to return an output IV, simply because it passes its IV buffer directly to the encapsulated CBC implementation, which operates on full blocks only, and always returns an IV. This output IV happens to match how RFC 3962 defines it, even though the CTS template itself does not contain any output IV logic whatsoever, and, for this reason, lacks any test vectors that exercise this accidental output IV generation. The arm64 SIMD implementation of cts(cbc(aes)) does not use the generic CTS template at all, but instead, implements the CBC mode and ciphertext stealing directly, and therefore does not encapsule a CBC implementation that returns an output IV in the same way. The arm64 SIMD implementation complies with the specification and passes all internal tests, but when invoked by the SUNRPC code, fails to produce the expected output IV and causes its selftests to fail. Given that the output IV is defined as the penultimate block (where the final block may smaller than the block size), we can quite easily derive it in the caller by copying the appropriate slice of ciphertext after encryption. Cc: Trond Myklebust Cc: Anna Schumaker Cc: Chuck Lever Cc: Jeff Layton Reported-by: Scott Mayhew Tested-by: Scott Mayhew Signed-off-by: Ard Biesheuvel Signed-off-by: Chuck Lever --- net/sunrpc/auth_gss/gss_krb5_crypto.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'net') diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c index 212c5d57465a..9734e1d9f991 100644 --- a/net/sunrpc/auth_gss/gss_krb5_crypto.c +++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c @@ -639,6 +639,16 @@ gss_krb5_cts_crypt(struct crypto_sync_skcipher *cipher, struct xdr_buf *buf, ret = write_bytes_to_xdr_buf(buf, offset, data, len); +#if IS_ENABLED(CONFIG_KUNIT) + /* + * CBC-CTS does not define an output IV but RFC 3962 defines it as the + * penultimate block of ciphertext, so copy that into the IV buffer + * before returning. + */ + if (encrypt) + memcpy(iv, data, crypto_sync_skcipher_ivsize(cipher)); +#endif + out: kfree(data); return ret; -- cgit v1.2.3 From 8509f62b0b07ae8d6dec5aa9613ab1b250ff632f Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 25 Apr 2023 16:50:32 +0200 Subject: netfilter: nf_tables: hit ENOENT on unexisting chain/flowtable update with missing attributes If user does not specify hook number and priority, then assume this is a chain/flowtable update. Therefore, report ENOENT which provides a better hint than EINVAL. Set on extended netlink error report to refer to the chain name. Fixes: 5b6743fb2c2a ("netfilter: nf_tables: skip flowtable hooknum and priority on device updates") Fixes: 5efe72698a97 ("netfilter: nf_tables: support for adding new devices to an existing netdev chain") Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 29 +++++++++++++++++------------ 1 file changed, 17 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 09542951656c..8b6c61a2196c 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -2075,8 +2075,10 @@ static int nft_chain_parse_hook(struct net *net, if (!basechain) { if (!ha[NFTA_HOOK_HOOKNUM] || - !ha[NFTA_HOOK_PRIORITY]) - return -EINVAL; + !ha[NFTA_HOOK_PRIORITY]) { + NL_SET_BAD_ATTR(extack, nla[NFTA_CHAIN_NAME]); + return -ENOENT; + } hook->num = ntohl(nla_get_be32(ha[NFTA_HOOK_HOOKNUM])); hook->priority = ntohl(nla_get_be32(ha[NFTA_HOOK_PRIORITY])); @@ -7693,7 +7695,7 @@ static const struct nla_policy nft_flowtable_hook_policy[NFTA_FLOWTABLE_HOOK_MAX }; static int nft_flowtable_parse_hook(const struct nft_ctx *ctx, - const struct nlattr *attr, + const struct nlattr * const nla[], struct nft_flowtable_hook *flowtable_hook, struct nft_flowtable *flowtable, struct netlink_ext_ack *extack, bool add) @@ -7705,15 +7707,18 @@ static int nft_flowtable_parse_hook(const struct nft_ctx *ctx, INIT_LIST_HEAD(&flowtable_hook->list); - err = nla_parse_nested_deprecated(tb, NFTA_FLOWTABLE_HOOK_MAX, attr, + err = nla_parse_nested_deprecated(tb, NFTA_FLOWTABLE_HOOK_MAX, + nla[NFTA_FLOWTABLE_HOOK], nft_flowtable_hook_policy, NULL); if (err < 0) return err; if (add) { if (!tb[NFTA_FLOWTABLE_HOOK_NUM] || - !tb[NFTA_FLOWTABLE_HOOK_PRIORITY]) - return -EINVAL; + !tb[NFTA_FLOWTABLE_HOOK_PRIORITY]) { + NL_SET_BAD_ATTR(extack, nla[NFTA_FLOWTABLE_NAME]); + return -ENOENT; + } hooknum = ntohl(nla_get_be32(tb[NFTA_FLOWTABLE_HOOK_NUM])); if (hooknum != NF_NETDEV_INGRESS) @@ -7898,8 +7903,8 @@ static int nft_flowtable_update(struct nft_ctx *ctx, const struct nlmsghdr *nlh, u32 flags; int err; - err = nft_flowtable_parse_hook(ctx, nla[NFTA_FLOWTABLE_HOOK], - &flowtable_hook, flowtable, extack, false); + err = nft_flowtable_parse_hook(ctx, nla, &flowtable_hook, flowtable, + extack, false); if (err < 0) return err; @@ -8044,8 +8049,8 @@ static int nf_tables_newflowtable(struct sk_buff *skb, if (err < 0) goto err3; - err = nft_flowtable_parse_hook(&ctx, nla[NFTA_FLOWTABLE_HOOK], - &flowtable_hook, flowtable, extack, true); + err = nft_flowtable_parse_hook(&ctx, nla, &flowtable_hook, flowtable, + extack, true); if (err < 0) goto err4; @@ -8107,8 +8112,8 @@ static int nft_delflowtable_hook(struct nft_ctx *ctx, struct nft_trans *trans; int err; - err = nft_flowtable_parse_hook(ctx, nla[NFTA_FLOWTABLE_HOOK], - &flowtable_hook, flowtable, extack, false); + err = nft_flowtable_parse_hook(ctx, nla, &flowtable_hook, flowtable, + extack, false); if (err < 0) return err; -- cgit v1.2.3 From c1592a89942e9678f7d9c8030efa777c0d57edab Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 2 May 2023 10:25:24 +0200 Subject: netfilter: nf_tables: deactivate anonymous set from preparation phase Toggle deleted anonymous sets as inactive in the next generation, so users cannot perform any update on it. Clear the generation bitmask in case the transaction is aborted. The following KASAN splat shows a set element deletion for a bound anonymous set that has been already removed in the same transaction. [ 64.921510] ================================================================== [ 64.923123] BUG: KASAN: wild-memory-access in nf_tables_commit+0xa24/0x1490 [nf_tables] [ 64.924745] Write of size 8 at addr dead000000000122 by task test/890 [ 64.927903] CPU: 3 PID: 890 Comm: test Not tainted 6.3.0+ #253 [ 64.931120] Call Trace: [ 64.932699] [ 64.934292] dump_stack_lvl+0x33/0x50 [ 64.935908] ? nf_tables_commit+0xa24/0x1490 [nf_tables] [ 64.937551] kasan_report+0xda/0x120 [ 64.939186] ? nf_tables_commit+0xa24/0x1490 [nf_tables] [ 64.940814] nf_tables_commit+0xa24/0x1490 [nf_tables] [ 64.942452] ? __kasan_slab_alloc+0x2d/0x60 [ 64.944070] ? nf_tables_setelem_notify+0x190/0x190 [nf_tables] [ 64.945710] ? kasan_set_track+0x21/0x30 [ 64.947323] nfnetlink_rcv_batch+0x709/0xd90 [nfnetlink] [ 64.948898] ? nfnetlink_rcv_msg+0x480/0x480 [nfnetlink] Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 1 + net/netfilter/nf_tables_api.c | 12 ++++++++++++ net/netfilter/nft_dynset.c | 2 +- net/netfilter/nft_lookup.c | 2 +- net/netfilter/nft_objref.c | 2 +- 5 files changed, 16 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 3ed21d2d5659..2e24ea1d744c 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -619,6 +619,7 @@ struct nft_set_binding { }; enum nft_trans_phase; +void nf_tables_activate_set(const struct nft_ctx *ctx, struct nft_set *set); void nf_tables_deactivate_set(const struct nft_ctx *ctx, struct nft_set *set, struct nft_set_binding *binding, enum nft_trans_phase phase); diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 8b6c61a2196c..59fb8320ab4d 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -5127,12 +5127,24 @@ static void nf_tables_unbind_set(const struct nft_ctx *ctx, struct nft_set *set, } } +void nf_tables_activate_set(const struct nft_ctx *ctx, struct nft_set *set) +{ + if (nft_set_is_anonymous(set)) + nft_clear(ctx->net, set); + + set->use++; +} +EXPORT_SYMBOL_GPL(nf_tables_activate_set); + void nf_tables_deactivate_set(const struct nft_ctx *ctx, struct nft_set *set, struct nft_set_binding *binding, enum nft_trans_phase phase) { switch (phase) { case NFT_TRANS_PREPARE: + if (nft_set_is_anonymous(set)) + nft_deactivate_next(ctx->net, set); + set->use--; return; case NFT_TRANS_ABORT: diff --git a/net/netfilter/nft_dynset.c b/net/netfilter/nft_dynset.c index 274579b1696e..bd19c7aec92e 100644 --- a/net/netfilter/nft_dynset.c +++ b/net/netfilter/nft_dynset.c @@ -342,7 +342,7 @@ static void nft_dynset_activate(const struct nft_ctx *ctx, { struct nft_dynset *priv = nft_expr_priv(expr); - priv->set->use++; + nf_tables_activate_set(ctx, priv->set); } static void nft_dynset_destroy(const struct nft_ctx *ctx, diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index cecf8ab90e58..03ef4fdaa460 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -167,7 +167,7 @@ static void nft_lookup_activate(const struct nft_ctx *ctx, { struct nft_lookup *priv = nft_expr_priv(expr); - priv->set->use++; + nf_tables_activate_set(ctx, priv->set); } static void nft_lookup_destroy(const struct nft_ctx *ctx, diff --git a/net/netfilter/nft_objref.c b/net/netfilter/nft_objref.c index cb37169608ba..a48dd5b5d45b 100644 --- a/net/netfilter/nft_objref.c +++ b/net/netfilter/nft_objref.c @@ -185,7 +185,7 @@ static void nft_objref_map_activate(const struct nft_ctx *ctx, { struct nft_objref_map *priv = nft_expr_priv(expr); - priv->set->use++; + nf_tables_activate_set(ctx, priv->set); } static void nft_objref_map_destroy(const struct nft_ctx *ctx, -- cgit v1.2.3 From 9ad685dbfe7e856bbf17a7177b64676d324d6ed7 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 2 May 2023 15:20:50 +0300 Subject: ethtool: Fix uninitialized number of lanes It is not possible to set the number of lanes when setting link modes using the legacy IOCTL ethtool interface. Since 'struct ethtool_link_ksettings' is not initialized in this path, drivers receive an uninitialized number of lanes in 'struct ethtool_link_ksettings::lanes'. When this information is later queried from drivers, it results in the ethtool code making decisions based on uninitialized memory, leading to the following KMSAN splat [1]. In practice, this most likely only happens with the tun driver that simply returns whatever it got in the set operation. As far as I can tell, this uninitialized memory is not leaked to user space thanks to the 'ethtool_ops->cap_link_lanes_supported' check in linkmodes_prepare_data(). Fix by initializing the structure in the IOCTL path. Did not find any more call sites that pass an uninitialized structure when calling 'ethtool_ops::set_link_ksettings()'. [1] BUG: KMSAN: uninit-value in ethnl_update_linkmodes net/ethtool/linkmodes.c:273 [inline] BUG: KMSAN: uninit-value in ethnl_set_linkmodes+0x190b/0x19d0 net/ethtool/linkmodes.c:333 ethnl_update_linkmodes net/ethtool/linkmodes.c:273 [inline] ethnl_set_linkmodes+0x190b/0x19d0 net/ethtool/linkmodes.c:333 ethnl_default_set_doit+0x88d/0xde0 net/ethtool/netlink.c:640 genl_family_rcv_msg_doit net/netlink/genetlink.c:968 [inline] genl_family_rcv_msg net/netlink/genetlink.c:1048 [inline] genl_rcv_msg+0x141a/0x14c0 net/netlink/genetlink.c:1065 netlink_rcv_skb+0x3f8/0x750 net/netlink/af_netlink.c:2577 genl_rcv+0x40/0x60 net/netlink/genetlink.c:1076 netlink_unicast_kernel net/netlink/af_netlink.c:1339 [inline] netlink_unicast+0xf41/0x1270 net/netlink/af_netlink.c:1365 netlink_sendmsg+0x127d/0x1430 net/netlink/af_netlink.c:1942 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg net/socket.c:747 [inline] ____sys_sendmsg+0xa24/0xe40 net/socket.c:2501 ___sys_sendmsg+0x2a1/0x3f0 net/socket.c:2555 __sys_sendmsg net/socket.c:2584 [inline] __do_sys_sendmsg net/socket.c:2593 [inline] __se_sys_sendmsg net/socket.c:2591 [inline] __x64_sys_sendmsg+0x36b/0x540 net/socket.c:2591 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Uninit was stored to memory at: tun_get_link_ksettings+0x37/0x60 drivers/net/tun.c:3544 __ethtool_get_link_ksettings+0x17b/0x260 net/ethtool/ioctl.c:441 ethnl_set_linkmodes+0xee/0x19d0 net/ethtool/linkmodes.c:327 ethnl_default_set_doit+0x88d/0xde0 net/ethtool/netlink.c:640 genl_family_rcv_msg_doit net/netlink/genetlink.c:968 [inline] genl_family_rcv_msg net/netlink/genetlink.c:1048 [inline] genl_rcv_msg+0x141a/0x14c0 net/netlink/genetlink.c:1065 netlink_rcv_skb+0x3f8/0x750 net/netlink/af_netlink.c:2577 genl_rcv+0x40/0x60 net/netlink/genetlink.c:1076 netlink_unicast_kernel net/netlink/af_netlink.c:1339 [inline] netlink_unicast+0xf41/0x1270 net/netlink/af_netlink.c:1365 netlink_sendmsg+0x127d/0x1430 net/netlink/af_netlink.c:1942 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg net/socket.c:747 [inline] ____sys_sendmsg+0xa24/0xe40 net/socket.c:2501 ___sys_sendmsg+0x2a1/0x3f0 net/socket.c:2555 __sys_sendmsg net/socket.c:2584 [inline] __do_sys_sendmsg net/socket.c:2593 [inline] __se_sys_sendmsg net/socket.c:2591 [inline] __x64_sys_sendmsg+0x36b/0x540 net/socket.c:2591 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Uninit was stored to memory at: tun_set_link_ksettings+0x37/0x60 drivers/net/tun.c:3553 ethtool_set_link_ksettings+0x600/0x690 net/ethtool/ioctl.c:609 __dev_ethtool net/ethtool/ioctl.c:3024 [inline] dev_ethtool+0x1db9/0x2a70 net/ethtool/ioctl.c:3078 dev_ioctl+0xb07/0x1270 net/core/dev_ioctl.c:524 sock_do_ioctl+0x295/0x540 net/socket.c:1213 sock_ioctl+0x729/0xd90 net/socket.c:1316 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:870 [inline] __se_sys_ioctl+0x222/0x400 fs/ioctl.c:856 __x64_sys_ioctl+0x96/0xe0 fs/ioctl.c:856 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Local variable link_ksettings created at: ethtool_set_link_ksettings+0x54/0x690 net/ethtool/ioctl.c:577 __dev_ethtool net/ethtool/ioctl.c:3024 [inline] dev_ethtool+0x1db9/0x2a70 net/ethtool/ioctl.c:3078 Fixes: 012ce4dd3102 ("ethtool: Extend link modes settings uAPI with lanes") Reported-and-tested-by: syzbot+ef6edd9f1baaa54d6235@syzkaller.appspotmail.com Link: https://lore.kernel.org/netdev/0000000000004bb41105fa70f361@google.com/ Reviewed-by: Danielle Ratson Signed-off-by: Ido Schimmel Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- net/ethtool/ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c index 59adc4e6e9ee..6bb778e10461 100644 --- a/net/ethtool/ioctl.c +++ b/net/ethtool/ioctl.c @@ -574,8 +574,8 @@ static int ethtool_get_link_ksettings(struct net_device *dev, static int ethtool_set_link_ksettings(struct net_device *dev, void __user *useraddr) { + struct ethtool_link_ksettings link_ksettings = {}; int err; - struct ethtool_link_ksettings link_ksettings; ASSERT_RTNL(); -- cgit v1.2.3 From 6a341729fb31b4c5df9f74f24b4b1c98410c9b87 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 1 May 2023 13:28:57 -0700 Subject: af_packet: Don't send zero-byte data in packet_sendmsg_spkt(). syzkaller reported a warning below [0]. We can reproduce it by sending 0-byte data from the (AF_PACKET, SOCK_PACKET) socket via some devices whose dev->hard_header_len is 0. struct sockaddr_pkt addr = { .spkt_family = AF_PACKET, .spkt_device = "tun0", }; int fd; fd = socket(AF_PACKET, SOCK_PACKET, 0); sendto(fd, NULL, 0, 0, (struct sockaddr *)&addr, sizeof(addr)); We have a similar fix for the (AF_PACKET, SOCK_RAW) socket as commit dc633700f00f ("net/af_packet: check len when min_header_len equals to 0"). Let's add the same test for the SOCK_PACKET socket. [0]: skb_assert_len WARNING: CPU: 1 PID: 19945 at include/linux/skbuff.h:2552 skb_assert_len include/linux/skbuff.h:2552 [inline] WARNING: CPU: 1 PID: 19945 at include/linux/skbuff.h:2552 __dev_queue_xmit+0x1f26/0x31d0 net/core/dev.c:4159 Modules linked in: CPU: 1 PID: 19945 Comm: syz-executor.0 Not tainted 6.3.0-rc7-02330-gca6270c12e20 #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 RIP: 0010:skb_assert_len include/linux/skbuff.h:2552 [inline] RIP: 0010:__dev_queue_xmit+0x1f26/0x31d0 net/core/dev.c:4159 Code: 89 de e8 1d a2 85 fd 84 db 75 21 e8 64 a9 85 fd 48 c7 c6 80 2a 1f 86 48 c7 c7 c0 06 1f 86 c6 05 23 cf 27 04 01 e8 fa ee 56 fd <0f> 0b e8 43 a9 85 fd 0f b6 1d 0f cf 27 04 31 ff 89 de e8 e3 a1 85 RSP: 0018:ffff8880217af6e0 EFLAGS: 00010282 RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffc90001133000 RDX: 0000000000040000 RSI: ffffffff81186922 RDI: 0000000000000001 RBP: ffff8880217af8b0 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000001 R12: ffff888030045640 R13: ffff8880300456b0 R14: ffff888030045650 R15: ffff888030045718 FS: 00007fc5864da640(0000) GS:ffff88806cd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020005740 CR3: 000000003f856003 CR4: 0000000000770ee0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: dev_queue_xmit include/linux/netdevice.h:3085 [inline] packet_sendmsg_spkt+0xc4b/0x1230 net/packet/af_packet.c:2066 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg+0x1b4/0x200 net/socket.c:747 ____sys_sendmsg+0x331/0x970 net/socket.c:2503 ___sys_sendmsg+0x11d/0x1c0 net/socket.c:2557 __sys_sendmmsg+0x18c/0x430 net/socket.c:2643 __do_sys_sendmmsg net/socket.c:2672 [inline] __se_sys_sendmmsg net/socket.c:2669 [inline] __x64_sys_sendmmsg+0x9c/0x100 net/socket.c:2669 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3c/0x90 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x72/0xdc RIP: 0033:0x7fc58791de5d Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 73 9f 1b 00 f7 d8 64 89 01 48 RSP: 002b:00007fc5864d9cc8 EFLAGS: 00000246 ORIG_RAX: 0000000000000133 RAX: ffffffffffffffda RBX: 00000000004bbf80 RCX: 00007fc58791de5d RDX: 0000000000000001 RSI: 0000000020005740 RDI: 0000000000000004 RBP: 00000000004bbf80 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 000000000000000b R14: 00007fc58797e530 R15: 0000000000000000 ---[ end trace 0000000000000000 ]--- skb len=0 headroom=16 headlen=0 tailroom=304 mac=(16,0) net=(16,-1) trans=-1 shinfo(txflags=0 nr_frags=0 gso(size=0 type=0 segs=0)) csum(0x0 ip_summed=0 complete_sw=0 valid=0 level=0) hash(0x0 sw=0 l4=0) proto=0x0000 pkttype=0 iif=0 dev name=sit0 feat=0x00000006401d7869 sk family=17 type=10 proto=0 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot Signed-off-by: Kuniyuki Iwashima Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/packet/af_packet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 6080c0db0814..640d94e34635 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2033,7 +2033,7 @@ retry: goto retry; } - if (!dev_validate_header(dev, skb->data, len)) { + if (!dev_validate_header(dev, skb->data, len) || !skb->len) { err = -EINVAL; goto out_unlock; } -- cgit v1.2.3 From f057b63bc11d86a98176de31b437e46789f44d8f Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 3 May 2023 12:00:18 +0200 Subject: netfilter: nf_tables: fix ct untracked match breakage "ct untracked" no longer works properly due to erroneous NFT_BREAK. We have to check ctinfo enum first. Fixes: d9e789147605 ("netfilter: nf_tables: avoid retpoline overhead for some ct expression calls") Reported-by: Rvfg Link: https://marc.info/?l=netfilter&m=168294996212038&w=2 Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_ct_fast.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_ct_fast.c b/net/netfilter/nft_ct_fast.c index 89983b0613fa..e684c8a91848 100644 --- a/net/netfilter/nft_ct_fast.c +++ b/net/netfilter/nft_ct_fast.c @@ -15,10 +15,6 @@ void nft_ct_get_fast_eval(const struct nft_expr *expr, unsigned int state; ct = nf_ct_get(pkt->skb, &ctinfo); - if (!ct) { - regs->verdict.code = NFT_BREAK; - return; - } switch (priv->key) { case NFT_CT_STATE: @@ -30,6 +26,16 @@ void nft_ct_get_fast_eval(const struct nft_expr *expr, state = NF_CT_STATE_INVALID_BIT; *dest = state; return; + default: + break; + } + + if (!ct) { + regs->verdict.code = NFT_BREAK; + return; + } + + switch (priv->key) { case NFT_CT_DIRECTION: nft_reg_store8(dest, CTINFO2DIR(ctinfo)); return; -- cgit v1.2.3 From 29cd2927fb914cc53b5ba4f67d2b74695c994ba4 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 2 May 2023 14:59:10 -0400 Subject: SUNRPC: Fix encoding of accepted but unsuccessful RPC replies Jiri Slaby says: > I bisected to this ... as it breaks nfs3-only servers in 6.3. > I.e. /etc/nfs.conf containing: > [nfsd] > vers4=no > > The client sees: > mount("10.0.2.15:/tmp", "/mnt", "nfs", 0, "vers=4.2,addr=10.0.2.15,clientad"...) = -1 EIO (Input/output error) > write(2, "mount.nfs: mount system call fai"..., 45 > mount.nfs: mount system call failed for /mnt > > And the kernel says: > nfs4_discover_server_trunking unhandled error -5. Exiting with error EIO Reported-by: Jiri Slaby Link: https://bugzilla.suse.com/show_bug.cgi?id=1210995 Fixes: 4bcf0343e8a6 ("SUNRPC: Set rq_accept_statp inside ->accept methods") Tested-by: Jiri Slaby Signed-off-by: Chuck Lever --- net/sunrpc/svc.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 26367cf4c17a..43e32129a583 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -1416,7 +1416,7 @@ err_bad_rpc: /* Only RPCv2 supported */ xdr_stream_encode_u32(xdr, RPC_VERSION); xdr_stream_encode_u32(xdr, RPC_VERSION); - goto sendit; + return 1; /* don't wrap */ err_bad_auth: dprintk("svc: authentication failed (%d)\n", @@ -1432,7 +1432,7 @@ err_bad_auth: err_bad_prog: dprintk("svc: unknown program %d\n", rqstp->rq_prog); serv->sv_stats->rpcbadfmt++; - xdr_stream_encode_u32(xdr, RPC_PROG_UNAVAIL); + *rqstp->rq_accept_statp = rpc_prog_unavail; goto sendit; err_bad_vers: @@ -1440,7 +1440,12 @@ err_bad_vers: rqstp->rq_vers, rqstp->rq_prog, progp->pg_name); serv->sv_stats->rpcbadfmt++; - xdr_stream_encode_u32(xdr, RPC_PROG_MISMATCH); + *rqstp->rq_accept_statp = rpc_prog_mismatch; + + /* + * svc_authenticate() has already added the verifier and + * advanced the stream just past rq_accept_statp. + */ xdr_stream_encode_u32(xdr, process.mismatch.lovers); xdr_stream_encode_u32(xdr, process.mismatch.hivers); goto sendit; @@ -1449,19 +1454,19 @@ err_bad_proc: svc_printk(rqstp, "unknown procedure (%d)\n", rqstp->rq_proc); serv->sv_stats->rpcbadfmt++; - xdr_stream_encode_u32(xdr, RPC_PROC_UNAVAIL); + *rqstp->rq_accept_statp = rpc_proc_unavail; goto sendit; err_garbage_args: svc_printk(rqstp, "failed to decode RPC header\n"); serv->sv_stats->rpcbadfmt++; - xdr_stream_encode_u32(xdr, RPC_GARBAGE_ARGS); + *rqstp->rq_accept_statp = rpc_garbage_args; goto sendit; err_system_err: serv->sv_stats->rpcbadfmt++; - xdr_stream_encode_u32(xdr, RPC_SYSTEM_ERR); + *rqstp->rq_accept_statp = rpc_system_err; goto sendit; } -- cgit v1.2.3 From d7385ba137711ea71527a605cac162610a621de8 Mon Sep 17 00:00:00 2001 From: Jason Andryuk Date: Thu, 4 May 2023 15:47:23 -0400 Subject: 9p: Remove INET dependency 9pfs can run over assorted transports, so it doesn't have an INET dependency. Drop it and remove the includes of linux/inet.h. NET_9P_FD/trans_fd.o builds without INET or UNIX and is usable over plain file descriptors. However, tcp and unix functionality is still built and would generate runtime failures if used. Add imply INET and UNIX to NET_9P_FD, so functionality is enabled by default but can still be explicitly disabled. This allows configuring 9pfs over Xen with INET and UNIX disabled. Signed-off-by: Jason Andryuk Signed-off-by: David S. Miller --- fs/9p/Kconfig | 2 +- fs/9p/vfs_addr.c | 1 - fs/9p/vfs_dentry.c | 1 - fs/9p/vfs_dir.c | 1 - fs/9p/vfs_file.c | 1 - fs/9p/vfs_inode.c | 1 - fs/9p/vfs_inode_dotl.c | 1 - fs/9p/vfs_super.c | 1 - net/9p/Kconfig | 2 ++ 9 files changed, 3 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/fs/9p/Kconfig b/fs/9p/Kconfig index d7bc93447c85..0c63df574ee7 100644 --- a/fs/9p/Kconfig +++ b/fs/9p/Kconfig @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only config 9P_FS tristate "Plan 9 Resource Sharing Support (9P2000)" - depends on INET && NET_9P + depends on NET_9P select NETFS_SUPPORT help If you say Y here, you will get experimental support for diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 6f46d7e4c750..425956eb9fde 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -12,7 +12,6 @@ #include #include #include -#include #include #include #include diff --git a/fs/9p/vfs_dentry.c b/fs/9p/vfs_dentry.c index 65fa2df5e49b..f16f73581634 100644 --- a/fs/9p/vfs_dentry.c +++ b/fs/9p/vfs_dentry.c @@ -13,7 +13,6 @@ #include #include #include -#include #include #include #include diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c index 3d74b04fe0de..52bf87934650 100644 --- a/fs/9p/vfs_dir.c +++ b/fs/9p/vfs_dir.c @@ -13,7 +13,6 @@ #include #include #include -#include #include #include #include diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index 44c15eb2b908..367a851eaa82 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -14,7 +14,6 @@ #include #include #include -#include #include #include #include diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 1d523bec0a94..502ac74e4959 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -15,7 +15,6 @@ #include #include #include -#include #include #include #include diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index 331ed60d8fcb..a7da49906d99 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -13,7 +13,6 @@ #include #include #include -#include #include #include #include diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index 266c4693e20c..10449994a972 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -12,7 +12,6 @@ #include #include #include -#include #include #include #include diff --git a/net/9p/Kconfig b/net/9p/Kconfig index deabbd376cb1..00ebce9e5a65 100644 --- a/net/9p/Kconfig +++ b/net/9p/Kconfig @@ -17,6 +17,8 @@ if NET_9P config NET_9P_FD default NET_9P + imply INET + imply UNIX tristate "9P FD Transport" help This builds support for transports over TCP, Unix sockets and -- cgit v1.2.3 From dd4f6bbfa646f258e5bcdfac57a5c413d687f588 Mon Sep 17 00:00:00 2001 From: Vlad Buslov Date: Thu, 4 May 2023 20:16:14 +0200 Subject: net/sched: flower: fix filter idr initialization The cited commit moved idr initialization too early in fl_change() which allows concurrent users to access the filter that is still being initialized and is in inconsistent state, which, in turn, can cause NULL pointer dereference [0]. Since there is no obvious way to fix the ordering without reverting the whole cited commit, alternative approach taken to first insert NULL pointer into idr in order to allocate the handle but still cause fl_get() to return NULL and prevent concurrent users from seeing the filter while providing miss-to-action infrastructure with valid handle id early in fl_change(). [ 152.434728] general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] SMP KASAN [ 152.436163] KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] [ 152.437269] CPU: 4 PID: 3877 Comm: tc Not tainted 6.3.0-rc4+ #5 [ 152.438110] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 [ 152.439644] RIP: 0010:fl_dump_key+0x8b/0x1d10 [cls_flower] [ 152.440461] Code: 01 f2 02 f2 c7 40 08 04 f2 04 f2 c7 40 0c 04 f3 f3 f3 65 48 8b 04 25 28 00 00 00 48 89 84 24 00 01 00 00 48 89 c8 48 c1 e8 03 <0f> b6 04 10 84 c0 74 08 3c 03 0f 8e 98 19 00 00 8b 13 85 d2 74 57 [ 152.442885] RSP: 0018:ffff88817a28f158 EFLAGS: 00010246 [ 152.443851] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ 152.444826] RDX: dffffc0000000000 RSI: ffffffff8500ae80 RDI: ffff88810a987900 [ 152.445791] RBP: ffff888179d88240 R08: ffff888179d8845c R09: ffff888179d88240 [ 152.446780] R10: ffffed102f451e48 R11: 00000000fffffff2 R12: ffff88810a987900 [ 152.447741] R13: ffffffff8500ae80 R14: ffff88810a987900 R15: ffff888149b3c738 [ 152.448756] FS: 00007f5eb2a34800(0000) GS:ffff88881ec00000(0000) knlGS:0000000000000000 [ 152.449888] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 152.450685] CR2: 000000000046ad19 CR3: 000000010b0bd006 CR4: 0000000000370ea0 [ 152.451641] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 152.452628] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 152.453588] Call Trace: [ 152.454032] [ 152.454447] ? netlink_sendmsg+0x7a1/0xcb0 [ 152.455109] ? sock_sendmsg+0xc5/0x190 [ 152.455689] ? ____sys_sendmsg+0x535/0x6b0 [ 152.456320] ? ___sys_sendmsg+0xeb/0x170 [ 152.456916] ? do_syscall_64+0x3d/0x90 [ 152.457529] ? entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 152.458321] ? ___sys_sendmsg+0xeb/0x170 [ 152.458958] ? __sys_sendmsg+0xb5/0x140 [ 152.459564] ? do_syscall_64+0x3d/0x90 [ 152.460122] ? entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 152.460852] ? fl_dump_key_options.part.0+0xea0/0xea0 [cls_flower] [ 152.461710] ? _raw_spin_lock+0x7a/0xd0 [ 152.462299] ? _raw_read_lock_irq+0x30/0x30 [ 152.462924] ? nla_put+0x15e/0x1c0 [ 152.463480] fl_dump+0x228/0x650 [cls_flower] [ 152.464112] ? fl_tmplt_dump+0x210/0x210 [cls_flower] [ 152.464854] ? __kmem_cache_alloc_node+0x1a7/0x330 [ 152.465592] ? nla_put+0x15e/0x1c0 [ 152.466160] tcf_fill_node+0x515/0x9a0 [ 152.466766] ? tc_setup_offload_action+0xf0/0xf0 [ 152.467463] ? __alloc_skb+0x13c/0x2a0 [ 152.468067] ? __build_skb_around+0x330/0x330 [ 152.468814] ? fl_get+0x107/0x1a0 [cls_flower] [ 152.469503] tc_del_tfilter+0x718/0x1330 [ 152.470115] ? is_bpf_text_address+0xa/0x20 [ 152.470765] ? tc_ctl_chain+0xee0/0xee0 [ 152.471335] ? __kernel_text_address+0xe/0x30 [ 152.471948] ? unwind_get_return_address+0x56/0xa0 [ 152.472639] ? __thaw_task+0x150/0x150 [ 152.473218] ? arch_stack_walk+0x98/0xf0 [ 152.473839] ? __stack_depot_save+0x35/0x4c0 [ 152.474501] ? stack_trace_save+0x91/0xc0 [ 152.475119] ? security_capable+0x51/0x90 [ 152.475741] rtnetlink_rcv_msg+0x2c1/0x9d0 [ 152.476387] ? rtnl_calcit.isra.0+0x2b0/0x2b0 [ 152.477042] ? __sys_sendmsg+0xb5/0x140 [ 152.477664] ? do_syscall_64+0x3d/0x90 [ 152.478255] ? entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 152.479010] ? __stack_depot_save+0x35/0x4c0 [ 152.479679] ? __stack_depot_save+0x35/0x4c0 [ 152.480346] netlink_rcv_skb+0x12c/0x360 [ 152.480929] ? rtnl_calcit.isra.0+0x2b0/0x2b0 [ 152.481517] ? do_syscall_64+0x3d/0x90 [ 152.482061] ? netlink_ack+0x1550/0x1550 [ 152.482612] ? rhashtable_walk_peek+0x170/0x170 [ 152.483262] ? kmem_cache_alloc_node+0x1af/0x390 [ 152.483875] ? _copy_from_iter+0x3d6/0xc70 [ 152.484528] netlink_unicast+0x553/0x790 [ 152.485168] ? netlink_attachskb+0x6a0/0x6a0 [ 152.485848] ? unwind_next_frame+0x11cc/0x1a10 [ 152.486538] ? arch_stack_walk+0x61/0xf0 [ 152.487169] netlink_sendmsg+0x7a1/0xcb0 [ 152.487799] ? netlink_unicast+0x790/0x790 [ 152.488355] ? iovec_from_user.part.0+0x4d/0x220 [ 152.488990] ? _raw_spin_lock+0x7a/0xd0 [ 152.489598] ? netlink_unicast+0x790/0x790 [ 152.490236] sock_sendmsg+0xc5/0x190 [ 152.490796] ____sys_sendmsg+0x535/0x6b0 [ 152.491394] ? import_iovec+0x7/0x10 [ 152.491964] ? kernel_sendmsg+0x30/0x30 [ 152.492561] ? __copy_msghdr+0x3c0/0x3c0 [ 152.493160] ? do_syscall_64+0x3d/0x90 [ 152.493706] ___sys_sendmsg+0xeb/0x170 [ 152.494283] ? may_open_dev+0xd0/0xd0 [ 152.494858] ? copy_msghdr_from_user+0x110/0x110 [ 152.495541] ? __handle_mm_fault+0x2678/0x4ad0 [ 152.496205] ? copy_page_range+0x2360/0x2360 [ 152.496862] ? __fget_light+0x57/0x520 [ 152.497449] ? mas_find+0x1c0/0x1c0 [ 152.498026] ? sockfd_lookup_light+0x1a/0x140 [ 152.498703] __sys_sendmsg+0xb5/0x140 [ 152.499306] ? __sys_sendmsg_sock+0x20/0x20 [ 152.499951] ? do_user_addr_fault+0x369/0xd80 [ 152.500595] do_syscall_64+0x3d/0x90 [ 152.501185] entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 152.501917] RIP: 0033:0x7f5eb294f887 [ 152.502494] Code: 0a 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b9 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10 [ 152.505008] RSP: 002b:00007ffd2c708f78 EFLAGS: 00000246 ORIG_RAX: 000000000000002e [ 152.506152] RAX: ffffffffffffffda RBX: 00000000642d9472 RCX: 00007f5eb294f887 [ 152.507134] RDX: 0000000000000000 RSI: 00007ffd2c708fe0 RDI: 0000000000000003 [ 152.508113] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000 [ 152.509119] R10: 00007f5eb2808708 R11: 0000000000000246 R12: 0000000000000001 [ 152.510068] R13: 0000000000000000 R14: 00007ffd2c70d1b8 R15: 0000000000485400 [ 152.511031] [ 152.511444] Modules linked in: cls_flower sch_ingress openvswitch nsh mlx5_vdpa vringh vhost_iotlb vdpa mlx5_ib mlx5_core rpcrdma rdma_ucm ib_iser libiscsi scsi_transport_iscsi ib_umad rdma_cm ib_ipoib iw_cm ib_cm ib_uverbs ib_core xt_conntrack xt_MASQUERADE nf_conntrack_netlink nfnetlink xt_addrtype iptable_nat nf_nat br_netfilter overlay zram zsmalloc fuse [last unloaded: mlx5_core] [ 152.515720] ---[ end trace 0000000000000000 ]--- Fixes: 08a0063df3ae ("net/sched: flower: Move filter handle initialization earlier") Signed-off-by: Vlad Buslov Reviewed-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/cls_flower.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 6ab6aadc07b8..4dc3a9007f30 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -2210,10 +2210,10 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, spin_lock(&tp->lock); if (!handle) { handle = 1; - err = idr_alloc_u32(&head->handle_idr, fnew, &handle, + err = idr_alloc_u32(&head->handle_idr, NULL, &handle, INT_MAX, GFP_ATOMIC); } else { - err = idr_alloc_u32(&head->handle_idr, fnew, &handle, + err = idr_alloc_u32(&head->handle_idr, NULL, &handle, handle, GFP_ATOMIC); /* Filter with specified handle was concurrently @@ -2378,7 +2378,7 @@ static void fl_walk(struct tcf_proto *tp, struct tcf_walker *arg, rcu_read_lock(); idr_for_each_entry_continue_ul(&head->handle_idr, f, tmp, id) { /* don't return filters that are being deleted */ - if (!refcount_inc_not_zero(&f->refcnt)) + if (!f || !refcount_inc_not_zero(&f->refcnt)) continue; rcu_read_unlock(); -- cgit v1.2.3 From 5110f3ff6d3c986df9575c8da86630578b7f0846 Mon Sep 17 00:00:00 2001 From: Vlad Buslov Date: Thu, 4 May 2023 20:16:15 +0200 Subject: Revert "net/sched: flower: Fix wrong handle assignment during filter change" This reverts commit 32eff6bacec2cb574677c15378169a9fa30043ef. Superseded by the following commit in this series. Signed-off-by: Vlad Buslov Signed-off-by: David S. Miller --- net/sched/cls_flower.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 4dc3a9007f30..ac4f344c52e0 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -2231,8 +2231,8 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, kfree(fnew); goto errout_tb; } - fnew->handle = handle; } + fnew->handle = handle; err = tcf_exts_init_ex(&fnew->exts, net, TCA_FLOWER_ACT, 0, tp, handle, !tc_skip_hw(fnew->flags)); -- cgit v1.2.3 From fd741f0d9f702c193b2b44225c004f8c5d5be163 Mon Sep 17 00:00:00 2001 From: Vlad Buslov Date: Thu, 4 May 2023 20:16:16 +0200 Subject: net/sched: flower: fix error handler on replace When replacing a filter (i.e. 'fold' pointer is not NULL) the insertion of new filter to idr is postponed until later in code since handle is already provided by the user. However, the error handling code in fl_change() always assumes that the new filter had been inserted into idr. If error handler is reached when replacing existing filter it may remove it from idr therefore making it unreachable for delete or dump afterwards. Fix the issue by verifying that 'fold' argument wasn't provided by caller before calling idr_remove(). Fixes: 08a0063df3ae ("net/sched: flower: Move filter handle initialization earlier") Signed-off-by: Vlad Buslov Reviewed-by: Pedro Tammela Signed-off-by: David S. Miller --- net/sched/cls_flower.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index ac4f344c52e0..9dbc43388e57 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -2339,7 +2339,8 @@ errout_hw: errout_mask: fl_mask_put(head, fnew->mask); errout_idr: - idr_remove(&head->handle_idr, fnew->handle); + if (!fold) + idr_remove(&head->handle_idr, fnew->handle); __fl_put(fnew); errout_tb: kfree(tb); -- cgit v1.2.3 From 319050d4302ee9b216fb8cea847a9c4427628a98 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 4 May 2023 13:43:33 -0400 Subject: SUNRPC: Fix error handling in svc_setup_socket() Dan points out that sock_alloc_file() releases @sock on error, but so do all of svc_setup_socket's callers, resulting in a double- release if sock_alloc_file() returns an error. Rather than allocating a struct file for all new sockets, allocate one only for sockets created during a TCP accept. For the moment, those are the only ones that will ever be used with RPC-with-TLS. Reported-by: Dan Carpenter Fixes: ae0d77708aae ("SUNRPC: Ensure server-side sockets have a sock->file") Signed-off-by: Chuck Lever --- net/sunrpc/svcsock.c | 16 ++++------------ 1 file changed, 4 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index a51c9b989d58..9989194446a5 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -895,6 +895,9 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt) trace_svcsock_accept_err(xprt, serv->sv_name, err); return NULL; } + if (IS_ERR(sock_alloc_file(newsock, O_NONBLOCK, NULL))) + return NULL; + set_bit(XPT_CONN, &svsk->sk_xprt.xpt_flags); err = kernel_getpeername(newsock, sin); @@ -935,7 +938,7 @@ static struct svc_xprt *svc_tcp_accept(struct svc_xprt *xprt) return &newsvsk->sk_xprt; failed: - sock_release(newsock); + sockfd_put(newsock); return NULL; } @@ -1430,7 +1433,6 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, struct socket *sock, int flags) { - struct file *filp = NULL; struct svc_sock *svsk; struct sock *inet; int pmap_register = !(flags & SVC_SOCK_ANONYMOUS); @@ -1439,14 +1441,6 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, if (!svsk) return ERR_PTR(-ENOMEM); - if (!sock->file) { - filp = sock_alloc_file(sock, O_NONBLOCK, NULL); - if (IS_ERR(filp)) { - kfree(svsk); - return ERR_CAST(filp); - } - } - inet = sock->sk; if (pmap_register) { @@ -1456,8 +1450,6 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, inet->sk_protocol, ntohs(inet_sk(inet)->inet_sport)); if (err < 0) { - if (filp) - fput(filp); kfree(svsk); return ERR_PTR(err); } -- cgit v1.2.3 From 424f8416bb39936df6365442d651ee729b283460 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 5 May 2023 17:06:18 +0000 Subject: net: skb_partial_csum_set() fix against transport header magic value skb->transport_header uses the special 0xFFFF value to mark if the transport header was set or not. We must prevent callers to accidentaly set skb->transport_header to 0xFFFF. Note that only fuzzers can possibly do this today. syzbot reported: WARNING: CPU: 0 PID: 2340 at include/linux/skbuff.h:2847 skb_transport_offset include/linux/skbuff.h:2956 [inline] WARNING: CPU: 0 PID: 2340 at include/linux/skbuff.h:2847 virtio_net_hdr_to_skb+0xbcc/0x10c0 include/linux/virtio_net.h:103 Modules linked in: CPU: 0 PID: 2340 Comm: syz-executor.0 Not tainted 6.3.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/14/2023 RIP: 0010:skb_transport_header include/linux/skbuff.h:2847 [inline] RIP: 0010:skb_transport_offset include/linux/skbuff.h:2956 [inline] RIP: 0010:virtio_net_hdr_to_skb+0xbcc/0x10c0 include/linux/virtio_net.h:103 Code: 41 39 df 0f 82 c3 04 00 00 48 8b 7c 24 10 44 89 e6 e8 08 6e 59 ff 48 85 c0 74 54 e8 ce 36 7e fc e9 37 f8 ff ff e8 c4 36 7e fc <0f> 0b e9 93 f8 ff ff 44 89 f7 44 89 e6 e8 32 38 7e fc 45 39 e6 0f RSP: 0018:ffffc90004497880 EFLAGS: 00010293 RAX: ffffffff84fea55c RBX: 000000000000ffff RCX: ffff888120be2100 RDX: 0000000000000000 RSI: 000000000000ffff RDI: 000000000000ffff RBP: ffffc90004497990 R08: ffffffff84fe9de5 R09: 0000000000000034 R10: ffffea00048ebd80 R11: 0000000000000034 R12: ffff88811dc2d9c8 R13: dffffc0000000000 R14: ffff88811dc2d9ae R15: 1ffff11023b85b35 FS: 00007f9211a59700(0000) GS:ffff8881f6c00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000200002c0 CR3: 00000001215a5000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: packet_snd net/packet/af_packet.c:3076 [inline] packet_sendmsg+0x4590/0x61a0 net/packet/af_packet.c:3115 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg net/socket.c:747 [inline] __sys_sendto+0x472/0x630 net/socket.c:2144 __do_sys_sendto net/socket.c:2156 [inline] __se_sys_sendto net/socket.c:2152 [inline] __x64_sys_sendto+0xe5/0x100 net/socket.c:2152 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x2f/0x50 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0033:0x7f9210c8c169 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 f1 19 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f9211a59168 EFLAGS: 00000246 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 00007f9210dabf80 RCX: 00007f9210c8c169 RDX: 000000000000ffed RSI: 00000000200000c0 RDI: 0000000000000003 RBP: 00007f9210ce7ca1 R08: 0000000020000540 R09: 0000000000000014 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 00007ffe135d65cf R14: 00007f9211a59300 R15: 0000000000022000 Fixes: 66e4c8d95008 ("net: warn if transport header was not set") Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Willem de Bruijn Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/core/skbuff.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 26a586007d8b..515ec5cdc79c 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -5298,7 +5298,7 @@ bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off) u32 csum_end = (u32)start + (u32)off + sizeof(__sum16); u32 csum_start = skb_headroom(skb) + (u32)start; - if (unlikely(csum_start > U16_MAX || csum_end > skb_headlen(skb))) { + if (unlikely(csum_start >= U16_MAX || csum_end > skb_headlen(skb))) { net_warn_ratelimited("bad partial csum: csum=%u/%u headroom=%u headlen=%u\n", start, off, skb_headroom(skb), skb_headlen(skb)); return false; @@ -5306,7 +5306,7 @@ bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off) skb->ip_summed = CHECKSUM_PARTIAL; skb->csum_start = csum_start; skb->csum_offset = off; - skb_set_transport_header(skb, start); + skb->transport_header = csum_start; return true; } EXPORT_SYMBOL_GPL(skb_partial_csum_set); -- cgit v1.2.3 From 3d776e31c841ba2f69895d2255a49320bec7cea6 Mon Sep 17 00:00:00 2001 From: Tobias Brunner Date: Tue, 9 May 2023 10:59:58 +0200 Subject: xfrm: Reject optional tunnel/BEET mode templates in outbound policies xfrm_state_find() uses `encap_family` of the current template with the passed local and remote addresses to find a matching state. If an optional tunnel or BEET mode template is skipped in a mixed-family scenario, there could be a mismatch causing an out-of-bounds read as the addresses were not replaced to match the family of the next template. While there are theoretical use cases for optional templates in outbound policies, the only practical one is to skip IPComp states in inbound policies if uncompressed packets are received that are handled by an implicitly created IPIP state instead. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Tobias Brunner Acked-by: Herbert Xu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index af8fbcbfbe69..6794b9dea27a 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1768,7 +1768,7 @@ static void copy_templates(struct xfrm_policy *xp, struct xfrm_user_tmpl *ut, } static int validate_tmpl(int nr, struct xfrm_user_tmpl *ut, u16 family, - struct netlink_ext_ack *extack) + int dir, struct netlink_ext_ack *extack) { u16 prev_family; int i; @@ -1794,6 +1794,10 @@ static int validate_tmpl(int nr, struct xfrm_user_tmpl *ut, u16 family, switch (ut[i].mode) { case XFRM_MODE_TUNNEL: case XFRM_MODE_BEET: + if (ut[i].optional && dir == XFRM_POLICY_OUT) { + NL_SET_ERR_MSG(extack, "Mode in optional template not allowed in outbound policy"); + return -EINVAL; + } break; default: if (ut[i].family != prev_family) { @@ -1831,7 +1835,7 @@ static int validate_tmpl(int nr, struct xfrm_user_tmpl *ut, u16 family, } static int copy_from_user_tmpl(struct xfrm_policy *pol, struct nlattr **attrs, - struct netlink_ext_ack *extack) + int dir, struct netlink_ext_ack *extack) { struct nlattr *rt = attrs[XFRMA_TMPL]; @@ -1842,7 +1846,7 @@ static int copy_from_user_tmpl(struct xfrm_policy *pol, struct nlattr **attrs, int nr = nla_len(rt) / sizeof(*utmpl); int err; - err = validate_tmpl(nr, utmpl, pol->family, extack); + err = validate_tmpl(nr, utmpl, pol->family, dir, extack); if (err) return err; @@ -1919,7 +1923,7 @@ static struct xfrm_policy *xfrm_policy_construct(struct net *net, if (err) goto error; - if (!(err = copy_from_user_tmpl(xp, attrs, extack))) + if (!(err = copy_from_user_tmpl(xp, attrs, p->dir, extack))) err = copy_from_user_sec_ctx(xp, attrs); if (err) goto error; @@ -3498,7 +3502,7 @@ static struct xfrm_policy *xfrm_compile_policy(struct sock *sk, int opt, return NULL; nr = ((len - sizeof(*p)) / sizeof(*ut)); - if (validate_tmpl(nr, ut, p->sel.family, NULL)) + if (validate_tmpl(nr, ut, p->sel.family, p->dir, NULL)) return NULL; if (p->dir > XFRM_POLICY_OUT) -- cgit v1.2.3 From cf3128a7aca55b2eefb68281d44749c683bdc96f Mon Sep 17 00:00:00 2001 From: Tobias Brunner Date: Tue, 9 May 2023 11:00:06 +0200 Subject: af_key: Reject optional tunnel/BEET mode templates in outbound policies xfrm_state_find() uses `encap_family` of the current template with the passed local and remote addresses to find a matching state. If an optional tunnel or BEET mode template is skipped in a mixed-family scenario, there could be a mismatch causing an out-of-bounds read as the addresses were not replaced to match the family of the next template. While there are theoretical use cases for optional templates in outbound policies, the only practical one is to skip IPComp states in inbound policies if uncompressed packets are received that are handled by an implicitly created IPIP state instead. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Tobias Brunner Acked-by: Herbert Xu Signed-off-by: Steffen Klassert --- net/key/af_key.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/key/af_key.c b/net/key/af_key.c index a815f5ab4c49..31ab12fd720a 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -1940,7 +1940,8 @@ static u32 gen_reqid(struct net *net) } static int -parse_ipsecrequest(struct xfrm_policy *xp, struct sadb_x_ipsecrequest *rq) +parse_ipsecrequest(struct xfrm_policy *xp, struct sadb_x_policy *pol, + struct sadb_x_ipsecrequest *rq) { struct net *net = xp_net(xp); struct xfrm_tmpl *t = xp->xfrm_vec + xp->xfrm_nr; @@ -1958,9 +1959,12 @@ parse_ipsecrequest(struct xfrm_policy *xp, struct sadb_x_ipsecrequest *rq) if ((mode = pfkey_mode_to_xfrm(rq->sadb_x_ipsecrequest_mode)) < 0) return -EINVAL; t->mode = mode; - if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_USE) + if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_USE) { + if ((mode == XFRM_MODE_TUNNEL || mode == XFRM_MODE_BEET) && + pol->sadb_x_policy_dir == IPSEC_DIR_OUTBOUND) + return -EINVAL; t->optional = 1; - else if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_UNIQUE) { + } else if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_UNIQUE) { t->reqid = rq->sadb_x_ipsecrequest_reqid; if (t->reqid > IPSEC_MANUAL_REQID_MAX) t->reqid = 0; @@ -2002,7 +2006,7 @@ parse_ipsecrequests(struct xfrm_policy *xp, struct sadb_x_policy *pol) rq->sadb_x_ipsecrequest_len < sizeof(*rq)) return -EINVAL; - if ((err = parse_ipsecrequest(xp, rq)) < 0) + if ((err = parse_ipsecrequest(xp, pol, rq)) < 0) return err; len -= rq->sadb_x_ipsecrequest_len; rq = (void*)((u8*)rq + rq->sadb_x_ipsecrequest_len); -- cgit v1.2.3 From 8680407b6f8f5fba59e8f1d63c869abc280f04df Mon Sep 17 00:00:00 2001 From: Benedict Wong Date: Wed, 10 May 2023 01:14:14 +0000 Subject: xfrm: Check if_id in inbound policy/secpath match This change ensures that if configured in the policy, the if_id set in the policy and secpath states match during the inbound policy check. Without this, there is potential for ambiguity where entries in the secpath differing by only the if_id could be mismatched. Notably, this is checked in the outbound direction when resolving templates to SAs, but not on the inbound path when matching SAs and policies. Test: Tested against Android kernel unit tests & CTS Signed-off-by: Benedict Wong Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 21a3a1cd3d6d..6d15788b5123 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -3312,7 +3312,7 @@ xfrm_secpath_reject(int idx, struct sk_buff *skb, const struct flowi *fl) static inline int xfrm_state_ok(const struct xfrm_tmpl *tmpl, const struct xfrm_state *x, - unsigned short family) + unsigned short family, u32 if_id) { if (xfrm_state_kern(x)) return tmpl->optional && !xfrm_state_addr_cmp(tmpl, x, tmpl->encap_family); @@ -3323,7 +3323,8 @@ xfrm_state_ok(const struct xfrm_tmpl *tmpl, const struct xfrm_state *x, (tmpl->allalgs || (tmpl->aalgos & (1<props.aalgo)) || !(xfrm_id_proto_match(tmpl->id.proto, IPSEC_PROTO_ANY))) && !(x->props.mode != XFRM_MODE_TRANSPORT && - xfrm_state_addr_cmp(tmpl, x, family)); + xfrm_state_addr_cmp(tmpl, x, family)) && + (if_id == 0 || if_id == x->if_id); } /* @@ -3335,7 +3336,7 @@ xfrm_state_ok(const struct xfrm_tmpl *tmpl, const struct xfrm_state *x, */ static inline int xfrm_policy_ok(const struct xfrm_tmpl *tmpl, const struct sec_path *sp, int start, - unsigned short family) + unsigned short family, u32 if_id) { int idx = start; @@ -3345,7 +3346,7 @@ xfrm_policy_ok(const struct xfrm_tmpl *tmpl, const struct sec_path *sp, int star } else start = -1; for (; idx < sp->len; idx++) { - if (xfrm_state_ok(tmpl, sp->xvec[idx], family)) + if (xfrm_state_ok(tmpl, sp->xvec[idx], family, if_id)) return ++idx; if (sp->xvec[idx]->props.mode != XFRM_MODE_TRANSPORT) { if (start == -1) @@ -3724,7 +3725,7 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb, * are implied between each two transformations. */ for (i = xfrm_nr-1, k = 0; i >= 0; i--) { - k = xfrm_policy_ok(tpp[i], sp, k, family); + k = xfrm_policy_ok(tpp[i], sp, k, family, if_id); if (k < 0) { if (k < -1) /* "-2 - errored_index" returned */ -- cgit v1.2.3 From dc1c9fd4a8bbe1e06add9053010b652449bfe411 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 4 May 2023 14:20:21 +0200 Subject: netfilter: nf_tables: always release netdev hooks from notifier This reverts "netfilter: nf_tables: skip netdev events generated on netns removal". The problem is that when a veth device is released, the veth release callback will also queue the peer netns device for removal. Its possible that the peer netns is also slated for removal. In this case, the device memory is already released before the pre_exit hook of the peer netns runs: BUG: KASAN: slab-use-after-free in nf_hook_entry_head+0x1b8/0x1d0 Read of size 8 at addr ffff88812c0124f0 by task kworker/u8:1/45 Workqueue: netns cleanup_net Call Trace: nf_hook_entry_head+0x1b8/0x1d0 __nf_unregister_net_hook+0x76/0x510 nft_netdev_unregister_hooks+0xa0/0x220 __nft_release_hook+0x184/0x490 nf_tables_pre_exit_net+0x12f/0x1b0 .. Order is: 1. First netns is released, veth_dellink() queues peer netns device for removal 2. peer netns is queued for removal 3. peer netns device is released, unreg event is triggered 4. unreg event is ignored because netns is going down 5. pre_exit hook calls nft_netdev_unregister_hooks but device memory might be free'd already. Fixes: 68a3765c659f ("netfilter: nf_tables: skip netdev events generated on netns removal") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_chain_filter.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_chain_filter.c b/net/netfilter/nft_chain_filter.c index c3563f0be269..680fe557686e 100644 --- a/net/netfilter/nft_chain_filter.c +++ b/net/netfilter/nft_chain_filter.c @@ -344,6 +344,12 @@ static void nft_netdev_event(unsigned long event, struct net_device *dev, return; } + /* UNREGISTER events are also happening on netns exit. + * + * Although nf_tables core releases all tables/chains, only this event + * handler provides guarantee that hook->ops.dev is still accessible, + * so we cannot skip exiting net namespaces. + */ __nft_release_basechain(ctx); } @@ -362,9 +368,6 @@ static int nf_tables_netdev_event(struct notifier_block *this, event != NETDEV_CHANGENAME) return NOTIFY_DONE; - if (!check_net(ctx.net)) - return NOTIFY_DONE; - nft_net = nft_pernet(ctx.net); mutex_lock(&nft_net->commit_mutex); list_for_each_entry(table, &nft_net->tables, list) { -- cgit v1.2.3 From e72eeab542dbf4f544e389e64fa13b82a1b6d003 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 4 May 2023 14:55:02 +0200 Subject: netfilter: conntrack: fix possible bug_on with enable_hooks=1 I received a bug report (no reproducer so far) where we trip over 712 rcu_read_lock(); 713 ct_hook = rcu_dereference(nf_ct_hook); 714 BUG_ON(ct_hook == NULL); // here In nf_conntrack_destroy(). First turn this BUG_ON into a WARN. I think it was triggered via enable_hooks=1 flag. When this flag is turned on, the conntrack hooks are registered before nf_ct_hook pointer gets assigned. This opens a short window where packets enter the conntrack machinery, can have skb->_nfct set up and a subsequent kfree_skb might occur before nf_ct_hook is set. Call nf_conntrack_init_end() to set nf_ct_hook before we register the pernet ops. Fixes: ba3fbe663635 ("netfilter: nf_conntrack: provide modparam to always register conntrack hooks") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/core.c | 6 ++++-- net/netfilter/nf_conntrack_standalone.c | 3 ++- 2 files changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/netfilter/core.c b/net/netfilter/core.c index f0783e42108b..5f76ae86a656 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -711,9 +711,11 @@ void nf_conntrack_destroy(struct nf_conntrack *nfct) rcu_read_lock(); ct_hook = rcu_dereference(nf_ct_hook); - BUG_ON(ct_hook == NULL); - ct_hook->destroy(nfct); + if (ct_hook) + ct_hook->destroy(nfct); rcu_read_unlock(); + + WARN_ON(!ct_hook); } EXPORT_SYMBOL(nf_conntrack_destroy); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 57f6724c99a7..169e16fc2bce 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -1218,11 +1218,12 @@ static int __init nf_conntrack_standalone_init(void) nf_conntrack_htable_size_user = nf_conntrack_htable_size; #endif + nf_conntrack_init_end(); + ret = register_pernet_subsys(&nf_conntrack_net_ops); if (ret < 0) goto out_pernet; - nf_conntrack_init_end(); return 0; out_pernet: -- cgit v1.2.3 From a939d14919b799e6fff8a9c80296ca229ba2f8a4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 9 May 2023 16:56:34 +0000 Subject: netlink: annotate accesses to nlk->cb_running Both netlink_recvmsg() and netlink_native_seq_show() read nlk->cb_running locklessly. Use READ_ONCE() there. Add corresponding WRITE_ONCE() to netlink_dump() and __netlink_dump_start() syzbot reported: BUG: KCSAN: data-race in __netlink_dump_start / netlink_recvmsg write to 0xffff88813ea4db59 of 1 bytes by task 28219 on cpu 0: __netlink_dump_start+0x3af/0x4d0 net/netlink/af_netlink.c:2399 netlink_dump_start include/linux/netlink.h:308 [inline] rtnetlink_rcv_msg+0x70f/0x8c0 net/core/rtnetlink.c:6130 netlink_rcv_skb+0x126/0x220 net/netlink/af_netlink.c:2577 rtnetlink_rcv+0x1c/0x20 net/core/rtnetlink.c:6192 netlink_unicast_kernel net/netlink/af_netlink.c:1339 [inline] netlink_unicast+0x56f/0x640 net/netlink/af_netlink.c:1365 netlink_sendmsg+0x665/0x770 net/netlink/af_netlink.c:1942 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg net/socket.c:747 [inline] sock_write_iter+0x1aa/0x230 net/socket.c:1138 call_write_iter include/linux/fs.h:1851 [inline] new_sync_write fs/read_write.c:491 [inline] vfs_write+0x463/0x760 fs/read_write.c:584 ksys_write+0xeb/0x1a0 fs/read_write.c:637 __do_sys_write fs/read_write.c:649 [inline] __se_sys_write fs/read_write.c:646 [inline] __x64_sys_write+0x42/0x50 fs/read_write.c:646 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd read to 0xffff88813ea4db59 of 1 bytes by task 28222 on cpu 1: netlink_recvmsg+0x3b4/0x730 net/netlink/af_netlink.c:2022 sock_recvmsg_nosec+0x4c/0x80 net/socket.c:1017 ____sys_recvmsg+0x2db/0x310 net/socket.c:2718 ___sys_recvmsg net/socket.c:2762 [inline] do_recvmmsg+0x2e5/0x710 net/socket.c:2856 __sys_recvmmsg net/socket.c:2935 [inline] __do_sys_recvmmsg net/socket.c:2958 [inline] __se_sys_recvmmsg net/socket.c:2951 [inline] __x64_sys_recvmmsg+0xe2/0x160 net/socket.c:2951 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd value changed: 0x00 -> 0x01 Fixes: 16b304f3404f ("netlink: Eliminate kmalloc in netlink dump operation.") Reported-by: syzbot Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 7ef8b9a1e30c..c87804112d0c 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1990,7 +1990,7 @@ static int netlink_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, skb_free_datagram(sk, skb); - if (nlk->cb_running && + if (READ_ONCE(nlk->cb_running) && atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf / 2) { ret = netlink_dump(sk); if (ret) { @@ -2302,7 +2302,7 @@ static int netlink_dump(struct sock *sk) if (cb->done) cb->done(cb); - nlk->cb_running = false; + WRITE_ONCE(nlk->cb_running, false); module = cb->module; skb = cb->skb; mutex_unlock(nlk->cb_mutex); @@ -2365,7 +2365,7 @@ int __netlink_dump_start(struct sock *ssk, struct sk_buff *skb, goto error_put; } - nlk->cb_running = true; + WRITE_ONCE(nlk->cb_running, true); nlk->dump_done_errno = INT_MAX; mutex_unlock(nlk->cb_mutex); @@ -2703,7 +2703,7 @@ static int netlink_native_seq_show(struct seq_file *seq, void *v) nlk->groups ? (u32)nlk->groups[0] : 0, sk_rmem_alloc_get(s), sk_wmem_alloc_get(s), - nlk->cb_running, + READ_ONCE(nlk->cb_running), refcount_read(&s->sk_refcnt), atomic_read(&s->sk_drops), sock_i_ino(s) -- cgit v1.2.3 From e05a5f510f26607616fecdd4ac136310c8bea56b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 9 May 2023 16:35:53 +0000 Subject: net: annotate sk->sk_err write from do_recvmmsg() do_recvmmsg() can write to sk->sk_err from multiple threads. As said before, many other points reading or writing sk_err need annotations. Fixes: 34b88a68f26a ("net: Fix use after free in the recvmmsg exit path") Signed-off-by: Eric Dumazet Reported-by: syzbot Reviewed-by: Kuniyuki Iwashima Signed-off-by: David S. Miller --- net/socket.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/socket.c b/net/socket.c index a7b4b37d86df..b7e01d0fe082 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2911,7 +2911,7 @@ static int do_recvmmsg(int fd, struct mmsghdr __user *mmsg, * error to return on the next call or if the * app asks about it using getsockopt(SO_ERROR). */ - sock->sk->sk_err = -err; + WRITE_ONCE(sock->sk->sk_err, -err); } out_put: fput_light(sock->file, fput_needed); -- cgit v1.2.3 From d0ac89f6f9879fae316c155de77b5173b3e2c9c9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 9 May 2023 18:29:48 +0000 Subject: net: deal with most data-races in sk_wait_event() __condition is evaluated twice in sk_wait_event() macro. First invocation is lockless, and reads can race with writes, as spotted by syzbot. BUG: KCSAN: data-race in sk_stream_wait_connect / tcp_disconnect write to 0xffff88812d83d6a0 of 4 bytes by task 9065 on cpu 1: tcp_disconnect+0x2cd/0xdb0 inet_shutdown+0x19e/0x1f0 net/ipv4/af_inet.c:911 __sys_shutdown_sock net/socket.c:2343 [inline] __sys_shutdown net/socket.c:2355 [inline] __do_sys_shutdown net/socket.c:2363 [inline] __se_sys_shutdown+0xf8/0x140 net/socket.c:2361 __x64_sys_shutdown+0x31/0x40 net/socket.c:2361 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd read to 0xffff88812d83d6a0 of 4 bytes by task 9040 on cpu 0: sk_stream_wait_connect+0x1de/0x3a0 net/core/stream.c:75 tcp_sendmsg_locked+0x2e4/0x2120 net/ipv4/tcp.c:1266 tcp_sendmsg+0x30/0x50 net/ipv4/tcp.c:1484 inet6_sendmsg+0x63/0x80 net/ipv6/af_inet6.c:651 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg net/socket.c:747 [inline] __sys_sendto+0x246/0x300 net/socket.c:2142 __do_sys_sendto net/socket.c:2154 [inline] __se_sys_sendto net/socket.c:2150 [inline] __x64_sys_sendto+0x78/0x90 net/socket.c:2150 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd value changed: 0x00000000 -> 0x00000068 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/stream.c | 12 ++++++------ net/ipv4/tcp_bpf.c | 2 +- net/llc/af_llc.c | 8 +++++--- net/smc/smc_close.c | 4 ++-- net/smc/smc_rx.c | 4 ++-- net/smc/smc_tx.c | 4 ++-- net/tipc/socket.c | 4 ++-- net/tls/tls_main.c | 3 ++- 8 files changed, 22 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/net/core/stream.c b/net/core/stream.c index 434446ab14c5..f5c4e47df165 100644 --- a/net/core/stream.c +++ b/net/core/stream.c @@ -73,8 +73,8 @@ int sk_stream_wait_connect(struct sock *sk, long *timeo_p) add_wait_queue(sk_sleep(sk), &wait); sk->sk_write_pending++; done = sk_wait_event(sk, timeo_p, - !sk->sk_err && - !((1 << sk->sk_state) & + !READ_ONCE(sk->sk_err) && + !((1 << READ_ONCE(sk->sk_state)) & ~(TCPF_ESTABLISHED | TCPF_CLOSE_WAIT)), &wait); remove_wait_queue(sk_sleep(sk), &wait); sk->sk_write_pending--; @@ -87,9 +87,9 @@ EXPORT_SYMBOL(sk_stream_wait_connect); * sk_stream_closing - Return 1 if we still have things to send in our buffers. * @sk: socket to verify */ -static inline int sk_stream_closing(struct sock *sk) +static int sk_stream_closing(const struct sock *sk) { - return (1 << sk->sk_state) & + return (1 << READ_ONCE(sk->sk_state)) & (TCPF_FIN_WAIT1 | TCPF_CLOSING | TCPF_LAST_ACK); } @@ -142,8 +142,8 @@ int sk_stream_wait_memory(struct sock *sk, long *timeo_p) set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); sk->sk_write_pending++; - sk_wait_event(sk, ¤t_timeo, sk->sk_err || - (sk->sk_shutdown & SEND_SHUTDOWN) || + sk_wait_event(sk, ¤t_timeo, READ_ONCE(sk->sk_err) || + (READ_ONCE(sk->sk_shutdown) & SEND_SHUTDOWN) || (sk_stream_memory_free(sk) && !vm_wait), &wait); sk->sk_write_pending--; diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index ebf917511937..2e9547467edb 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -168,7 +168,7 @@ static int tcp_msg_wait_data(struct sock *sk, struct sk_psock *psock, sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); ret = sk_wait_event(sk, &timeo, !list_empty(&psock->ingress_msg) || - !skb_queue_empty(&sk->sk_receive_queue), &wait); + !skb_queue_empty_lockless(&sk->sk_receive_queue), &wait); sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk); remove_wait_queue(sk_sleep(sk), &wait); return ret; diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index da7fe94bea2e..9ffbc667be6c 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -583,7 +583,8 @@ static int llc_ui_wait_for_disc(struct sock *sk, long timeout) add_wait_queue(sk_sleep(sk), &wait); while (1) { - if (sk_wait_event(sk, &timeout, sk->sk_state == TCP_CLOSE, &wait)) + if (sk_wait_event(sk, &timeout, + READ_ONCE(sk->sk_state) == TCP_CLOSE, &wait)) break; rc = -ERESTARTSYS; if (signal_pending(current)) @@ -603,7 +604,8 @@ static bool llc_ui_wait_for_conn(struct sock *sk, long timeout) add_wait_queue(sk_sleep(sk), &wait); while (1) { - if (sk_wait_event(sk, &timeout, sk->sk_state != TCP_SYN_SENT, &wait)) + if (sk_wait_event(sk, &timeout, + READ_ONCE(sk->sk_state) != TCP_SYN_SENT, &wait)) break; if (signal_pending(current) || !timeout) break; @@ -622,7 +624,7 @@ static int llc_ui_wait_for_busy_core(struct sock *sk, long timeout) while (1) { rc = 0; if (sk_wait_event(sk, &timeout, - (sk->sk_shutdown & RCV_SHUTDOWN) || + (READ_ONCE(sk->sk_shutdown) & RCV_SHUTDOWN) || (!llc_data_accept_state(llc->state) && !llc->remote_busy_flag && !llc->p_flag), &wait)) diff --git a/net/smc/smc_close.c b/net/smc/smc_close.c index 31db7438857c..dbdf03e8aa5b 100644 --- a/net/smc/smc_close.c +++ b/net/smc/smc_close.c @@ -67,8 +67,8 @@ static void smc_close_stream_wait(struct smc_sock *smc, long timeout) rc = sk_wait_event(sk, &timeout, !smc_tx_prepared_sends(&smc->conn) || - sk->sk_err == ECONNABORTED || - sk->sk_err == ECONNRESET || + READ_ONCE(sk->sk_err) == ECONNABORTED || + READ_ONCE(sk->sk_err) == ECONNRESET || smc->conn.killed, &wait); if (rc) diff --git a/net/smc/smc_rx.c b/net/smc/smc_rx.c index 4380d32f5a5f..9a2f3638d161 100644 --- a/net/smc/smc_rx.c +++ b/net/smc/smc_rx.c @@ -267,9 +267,9 @@ int smc_rx_wait(struct smc_sock *smc, long *timeo, sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); add_wait_queue(sk_sleep(sk), &wait); rc = sk_wait_event(sk, timeo, - sk->sk_err || + READ_ONCE(sk->sk_err) || cflags->peer_conn_abort || - sk->sk_shutdown & RCV_SHUTDOWN || + READ_ONCE(sk->sk_shutdown) & RCV_SHUTDOWN || conn->killed || fcrit(conn), &wait); diff --git a/net/smc/smc_tx.c b/net/smc/smc_tx.c index f4b6a71ac488..45128443f1f1 100644 --- a/net/smc/smc_tx.c +++ b/net/smc/smc_tx.c @@ -113,8 +113,8 @@ static int smc_tx_wait(struct smc_sock *smc, int flags) break; /* at least 1 byte of free & no urgent data */ set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); sk_wait_event(sk, &timeo, - sk->sk_err || - (sk->sk_shutdown & SEND_SHUTDOWN) || + READ_ONCE(sk->sk_err) || + (READ_ONCE(sk->sk_shutdown) & SEND_SHUTDOWN) || smc_cdc_rxed_any_close(conn) || (atomic_read(&conn->sndbuf_space) && !conn->urg_tx_pend), diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 37edfe10f8c6..dd73d71c02a9 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -314,9 +314,9 @@ static void tsk_rej_rx_queue(struct sock *sk, int error) tipc_sk_respond(sk, skb, error); } -static bool tipc_sk_connected(struct sock *sk) +static bool tipc_sk_connected(const struct sock *sk) { - return sk->sk_state == TIPC_ESTABLISHED; + return READ_ONCE(sk->sk_state) == TIPC_ESTABLISHED; } /* tipc_sk_type_connectionless - check if the socket is datagram socket diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c index b32c112984dd..f2e7302a4d96 100644 --- a/net/tls/tls_main.c +++ b/net/tls/tls_main.c @@ -111,7 +111,8 @@ int wait_on_pending_writer(struct sock *sk, long *timeo) break; } - if (sk_wait_event(sk, timeo, !sk->sk_write_pending, &wait)) + if (sk_wait_event(sk, timeo, + !READ_ONCE(sk->sk_write_pending), &wait)) break; } remove_wait_queue(sk_sleep(sk), &wait); -- cgit v1.2.3 From 4063384ef762cc5946fc7a3f89879e76c6ec51e2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 9 May 2023 13:18:57 +0000 Subject: net: add vlan_get_protocol_and_depth() helper MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Before blamed commit, pskb_may_pull() was used instead of skb_header_pointer() in __vlan_get_protocol() and friends. Few callers depended on skb->head being populated with MAC header, syzbot caught one of them (skb_mac_gso_segment()) Add vlan_get_protocol_and_depth() to make the intent clearer and use it where sensible. This is a more generic fix than commit e9d3f80935b6 ("net/af_packet: make sure to pull mac header") which was dealing with a similar issue. kernel BUG at include/linux/skbuff.h:2655 ! invalid opcode: 0000 [#1] SMP KASAN CPU: 0 PID: 1441 Comm: syz-executor199 Not tainted 6.1.24-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/14/2023 RIP: 0010:__skb_pull include/linux/skbuff.h:2655 [inline] RIP: 0010:skb_mac_gso_segment+0x68f/0x6a0 net/core/gro.c:136 Code: fd 48 8b 5c 24 10 44 89 6b 70 48 c7 c7 c0 ae 0d 86 44 89 e6 e8 a1 91 d0 00 48 c7 c7 00 af 0d 86 48 89 de 31 d2 e8 d1 4a e9 ff <0f> 0b 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 41 RSP: 0018:ffffc90001bd7520 EFLAGS: 00010286 RAX: ffffffff8469736a RBX: ffff88810f31dac0 RCX: ffff888115a18b00 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: ffffc90001bd75e8 R08: ffffffff84697183 R09: fffff5200037adf9 R10: 0000000000000000 R11: dffffc0000000001 R12: 0000000000000012 R13: 000000000000fee5 R14: 0000000000005865 R15: 000000000000fed7 FS: 000055555633f300(0000) GS:ffff8881f6a00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020000000 CR3: 0000000116fea000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: [] __skb_gso_segment+0x32d/0x4c0 net/core/dev.c:3419 [] skb_gso_segment include/linux/netdevice.h:4819 [inline] [] validate_xmit_skb+0x3aa/0xee0 net/core/dev.c:3725 [] __dev_queue_xmit+0x1332/0x3300 net/core/dev.c:4313 [] dev_queue_xmit+0x17/0x20 include/linux/netdevice.h:3029 [] packet_snd net/packet/af_packet.c:3111 [inline] [] packet_sendmsg+0x49d2/0x6470 net/packet/af_packet.c:3142 [] sock_sendmsg_nosec net/socket.c:716 [inline] [] sock_sendmsg net/socket.c:736 [inline] [] __sys_sendto+0x472/0x5f0 net/socket.c:2139 [] __do_sys_sendto net/socket.c:2151 [inline] [] __se_sys_sendto net/socket.c:2147 [inline] [] __x64_sys_sendto+0xe5/0x100 net/socket.c:2147 [] do_syscall_x64 arch/x86/entry/common.c:50 [inline] [] do_syscall_64+0x2f/0x50 arch/x86/entry/common.c:80 [] entry_SYSCALL_64_after_hwframe+0x63/0xcd Fixes: 469aceddfa3e ("vlan: consolidate VLAN parsing code and limit max parsing depth") Reported-by: syzbot Signed-off-by: Eric Dumazet Cc: Toke Høiland-Jørgensen Cc: Willem de Bruijn Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/tap.c | 4 ++-- include/linux/if_vlan.h | 17 +++++++++++++++++ net/bridge/br_forward.c | 2 +- net/core/dev.c | 2 +- net/packet/af_packet.c | 6 ++---- 5 files changed, 23 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/drivers/net/tap.c b/drivers/net/tap.c index ce993cc75bf3..d30d730ed5a7 100644 --- a/drivers/net/tap.c +++ b/drivers/net/tap.c @@ -742,7 +742,7 @@ static ssize_t tap_get_user(struct tap_queue *q, void *msg_control, /* Move network header to the right position for VLAN tagged packets */ if (eth_type_vlan(skb->protocol) && - __vlan_get_protocol(skb, skb->protocol, &depth) != 0) + vlan_get_protocol_and_depth(skb, skb->protocol, &depth) != 0) skb_set_network_header(skb, depth); /* copy skb_ubuf_info for callback when skb has no error */ @@ -1197,7 +1197,7 @@ static int tap_get_user_xdp(struct tap_queue *q, struct xdp_buff *xdp) /* Move network header to the right position for VLAN tagged packets */ if (eth_type_vlan(skb->protocol) && - __vlan_get_protocol(skb, skb->protocol, &depth) != 0) + vlan_get_protocol_and_depth(skb, skb->protocol, &depth) != 0) skb_set_network_header(skb, depth); rcu_read_lock(); diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 0f40f379d75c..6ba71957851e 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -637,6 +637,23 @@ static inline __be16 vlan_get_protocol(const struct sk_buff *skb) return __vlan_get_protocol(skb, skb->protocol, NULL); } +/* This version of __vlan_get_protocol() also pulls mac header in skb->head */ +static inline __be16 vlan_get_protocol_and_depth(struct sk_buff *skb, + __be16 type, int *depth) +{ + int maclen; + + type = __vlan_get_protocol(skb, type, &maclen); + + if (type) { + if (!pskb_may_pull(skb, maclen)) + type = 0; + else if (depth) + *depth = maclen; + } + return type; +} + /* A getter for the SKB protocol field which will handle VLAN tags consistently * whether VLAN acceleration is enabled or not. */ diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 57744704ff69..84d6dd5e5b1a 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -42,7 +42,7 @@ int br_dev_queue_push_xmit(struct net *net, struct sock *sk, struct sk_buff *skb eth_type_vlan(skb->protocol)) { int depth; - if (!__vlan_get_protocol(skb, skb->protocol, &depth)) + if (!vlan_get_protocol_and_depth(skb, skb->protocol, &depth)) goto drop; skb_set_network_header(skb, depth); diff --git a/net/core/dev.c b/net/core/dev.c index 735096d42c1d..b3c13e041935 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3335,7 +3335,7 @@ __be16 skb_network_protocol(struct sk_buff *skb, int *depth) type = eth->h_proto; } - return __vlan_get_protocol(skb, type, depth); + return vlan_get_protocol_and_depth(skb, type, depth); } /* openvswitch calls this on rx path, so we need a different check. diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 640d94e34635..94c6a1ffa459 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1934,10 +1934,8 @@ static void packet_parse_headers(struct sk_buff *skb, struct socket *sock) /* Move network header to the right position for VLAN tagged packets */ if (likely(skb->dev->type == ARPHRD_ETHER) && eth_type_vlan(skb->protocol) && - __vlan_get_protocol(skb, skb->protocol, &depth) != 0) { - if (pskb_may_pull(skb, depth)) - skb_set_network_header(skb, depth); - } + vlan_get_protocol_and_depth(skb, skb->protocol, &depth) != 0) + skb_set_network_header(skb, depth); skb_probe_transport_header(skb); } -- cgit v1.2.3 From e14cadfd80d76f01bfaa1a8d745b1db19b57d6be Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 9 May 2023 20:36:56 +0000 Subject: tcp: add annotations around sk->sk_shutdown accesses Now sk->sk_shutdown is no longer a bitfield, we can add standard READ_ONCE()/WRITE_ONCE() annotations to silence KCSAN reports like the following: BUG: KCSAN: data-race in tcp_disconnect / tcp_poll write to 0xffff88814588582c of 1 bytes by task 3404 on cpu 1: tcp_disconnect+0x4d6/0xdb0 net/ipv4/tcp.c:3121 __inet_stream_connect+0x5dd/0x6e0 net/ipv4/af_inet.c:715 inet_stream_connect+0x48/0x70 net/ipv4/af_inet.c:727 __sys_connect_file net/socket.c:2001 [inline] __sys_connect+0x19b/0x1b0 net/socket.c:2018 __do_sys_connect net/socket.c:2028 [inline] __se_sys_connect net/socket.c:2025 [inline] __x64_sys_connect+0x41/0x50 net/socket.c:2025 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd read to 0xffff88814588582c of 1 bytes by task 3374 on cpu 0: tcp_poll+0x2e6/0x7d0 net/ipv4/tcp.c:562 sock_poll+0x253/0x270 net/socket.c:1383 vfs_poll include/linux/poll.h:88 [inline] io_poll_check_events io_uring/poll.c:281 [inline] io_poll_task_func+0x15a/0x820 io_uring/poll.c:333 handle_tw_list io_uring/io_uring.c:1184 [inline] tctx_task_work+0x1fe/0x4d0 io_uring/io_uring.c:1246 task_work_run+0x123/0x160 kernel/task_work.c:179 get_signal+0xe64/0xff0 kernel/signal.c:2635 arch_do_signal_or_restart+0x89/0x2a0 arch/x86/kernel/signal.c:306 exit_to_user_mode_loop+0x6f/0xe0 kernel/entry/common.c:168 exit_to_user_mode_prepare+0x6c/0xb0 kernel/entry/common.c:204 __syscall_exit_to_user_mode_work kernel/entry/common.c:286 [inline] syscall_exit_to_user_mode+0x26/0x140 kernel/entry/common.c:297 do_syscall_64+0x4d/0xc0 arch/x86/entry/common.c:86 entry_SYSCALL_64_after_hwframe+0x63/0xcd value changed: 0x03 -> 0x00 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/af_inet.c | 2 +- net/ipv4/tcp.c | 14 ++++++++------ net/ipv4/tcp_input.c | 4 ++-- 3 files changed, 11 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 940062e08f57..c4aab3aacbd8 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -894,7 +894,7 @@ int inet_shutdown(struct socket *sock, int how) EPOLLHUP, even on eg. unconnected UDP sockets -- RR */ fallthrough; default: - sk->sk_shutdown |= how; + WRITE_ONCE(sk->sk_shutdown, sk->sk_shutdown | how); if (sk->sk_prot->shutdown) sk->sk_prot->shutdown(sk, how); break; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 20db115c38c4..4d6392c16b7a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -498,6 +498,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) __poll_t mask; struct sock *sk = sock->sk; const struct tcp_sock *tp = tcp_sk(sk); + u8 shutdown; int state; sock_poll_wait(file, sock, wait); @@ -540,9 +541,10 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) * NOTE. Check for TCP_CLOSE is added. The goal is to prevent * blocking on fresh not-connected or disconnected socket. --ANK */ - if (sk->sk_shutdown == SHUTDOWN_MASK || state == TCP_CLOSE) + shutdown = READ_ONCE(sk->sk_shutdown); + if (shutdown == SHUTDOWN_MASK || state == TCP_CLOSE) mask |= EPOLLHUP; - if (sk->sk_shutdown & RCV_SHUTDOWN) + if (shutdown & RCV_SHUTDOWN) mask |= EPOLLIN | EPOLLRDNORM | EPOLLRDHUP; /* Connected or passive Fast Open socket? */ @@ -559,7 +561,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) if (tcp_stream_is_readable(sk, target)) mask |= EPOLLIN | EPOLLRDNORM; - if (!(sk->sk_shutdown & SEND_SHUTDOWN)) { + if (!(shutdown & SEND_SHUTDOWN)) { if (__sk_stream_is_writeable(sk, 1)) { mask |= EPOLLOUT | EPOLLWRNORM; } else { /* send SIGIO later */ @@ -2867,7 +2869,7 @@ void __tcp_close(struct sock *sk, long timeout) int data_was_unread = 0; int state; - sk->sk_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(sk->sk_shutdown, SHUTDOWN_MASK); if (sk->sk_state == TCP_LISTEN) { tcp_set_state(sk, TCP_CLOSE); @@ -3119,7 +3121,7 @@ int tcp_disconnect(struct sock *sk, int flags) inet_bhash2_reset_saddr(sk); - sk->sk_shutdown = 0; + WRITE_ONCE(sk->sk_shutdown, 0); sock_reset_flag(sk, SOCK_DONE); tp->srtt_us = 0; tp->mdev_us = jiffies_to_usecs(TCP_TIMEOUT_INIT); @@ -4649,7 +4651,7 @@ void tcp_done(struct sock *sk) if (req) reqsk_fastopen_remove(sk, req, false); - sk->sk_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(sk->sk_shutdown, SHUTDOWN_MASK); if (!sock_flag(sk, SOCK_DEAD)) sk->sk_state_change(sk); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index a057330d6f59..61b6710f337a 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4362,7 +4362,7 @@ void tcp_fin(struct sock *sk) inet_csk_schedule_ack(sk); - sk->sk_shutdown |= RCV_SHUTDOWN; + WRITE_ONCE(sk->sk_shutdown, sk->sk_shutdown | RCV_SHUTDOWN); sock_set_flag(sk, SOCK_DONE); switch (sk->sk_state) { @@ -6599,7 +6599,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) break; tcp_set_state(sk, TCP_FIN_WAIT2); - sk->sk_shutdown |= SEND_SHUTDOWN; + WRITE_ONCE(sk->sk_shutdown, sk->sk_shutdown | SEND_SHUTDOWN); sk_dst_confirm(sk); -- cgit v1.2.3 From 5bca1d081f44c9443e61841842ce4e9179d327b6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 9 May 2023 17:31:31 +0000 Subject: net: datagram: fix data-races in datagram_poll() datagram_poll() runs locklessly, we should add READ_ONCE() annotations while reading sk->sk_err, sk->sk_shutdown and sk->sk_state. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20230509173131.3263780-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/datagram.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/datagram.c b/net/core/datagram.c index 5662dff3d381..176eb5834746 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -807,18 +807,21 @@ __poll_t datagram_poll(struct file *file, struct socket *sock, { struct sock *sk = sock->sk; __poll_t mask; + u8 shutdown; sock_poll_wait(file, sock, wait); mask = 0; /* exceptional events? */ - if (sk->sk_err || !skb_queue_empty_lockless(&sk->sk_error_queue)) + if (READ_ONCE(sk->sk_err) || + !skb_queue_empty_lockless(&sk->sk_error_queue)) mask |= EPOLLERR | (sock_flag(sk, SOCK_SELECT_ERR_QUEUE) ? EPOLLPRI : 0); - if (sk->sk_shutdown & RCV_SHUTDOWN) + shutdown = READ_ONCE(sk->sk_shutdown); + if (shutdown & RCV_SHUTDOWN) mask |= EPOLLRDHUP | EPOLLIN | EPOLLRDNORM; - if (sk->sk_shutdown == SHUTDOWN_MASK) + if (shutdown == SHUTDOWN_MASK) mask |= EPOLLHUP; /* readable? */ @@ -827,10 +830,12 @@ __poll_t datagram_poll(struct file *file, struct socket *sock, /* Connection-based need to check for termination and startup */ if (connection_based(sk)) { - if (sk->sk_state == TCP_CLOSE) + int state = READ_ONCE(sk->sk_state); + + if (state == TCP_CLOSE) mask |= EPOLLHUP; /* connection hasn't started yet? */ - if (sk->sk_state == TCP_SYN_SENT) + if (state == TCP_SYN_SENT) return mask; } -- cgit v1.2.3 From 679ed006d416ea0cecfe24a99d365d1dea69c683 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 9 May 2023 17:34:55 -0700 Subject: af_unix: Fix a data race of sk->sk_receive_queue->qlen. KCSAN found a data race of sk->sk_receive_queue->qlen where recvmsg() updates qlen under the queue lock and sendmsg() checks qlen under unix_state_sock(), not the queue lock, so the reader side needs READ_ONCE(). BUG: KCSAN: data-race in __skb_try_recv_from_queue / unix_wait_for_peer write (marked) to 0xffff888019fe7c68 of 4 bytes by task 49792 on cpu 0: __skb_unlink include/linux/skbuff.h:2347 [inline] __skb_try_recv_from_queue+0x3de/0x470 net/core/datagram.c:197 __skb_try_recv_datagram+0xf7/0x390 net/core/datagram.c:263 __unix_dgram_recvmsg+0x109/0x8a0 net/unix/af_unix.c:2452 unix_dgram_recvmsg+0x94/0xa0 net/unix/af_unix.c:2549 sock_recvmsg_nosec net/socket.c:1019 [inline] ____sys_recvmsg+0x3a3/0x3b0 net/socket.c:2720 ___sys_recvmsg+0xc8/0x150 net/socket.c:2764 do_recvmmsg+0x182/0x560 net/socket.c:2858 __sys_recvmmsg net/socket.c:2937 [inline] __do_sys_recvmmsg net/socket.c:2960 [inline] __se_sys_recvmmsg net/socket.c:2953 [inline] __x64_sys_recvmmsg+0x153/0x170 net/socket.c:2953 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3b/0x90 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x72/0xdc read to 0xffff888019fe7c68 of 4 bytes by task 49793 on cpu 1: skb_queue_len include/linux/skbuff.h:2127 [inline] unix_recvq_full net/unix/af_unix.c:229 [inline] unix_wait_for_peer+0x154/0x1a0 net/unix/af_unix.c:1445 unix_dgram_sendmsg+0x13bc/0x14b0 net/unix/af_unix.c:2048 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg+0x148/0x160 net/socket.c:747 ____sys_sendmsg+0x20e/0x620 net/socket.c:2503 ___sys_sendmsg+0xc6/0x140 net/socket.c:2557 __sys_sendmmsg+0x11d/0x370 net/socket.c:2643 __do_sys_sendmmsg net/socket.c:2672 [inline] __se_sys_sendmmsg net/socket.c:2669 [inline] __x64_sys_sendmmsg+0x58/0x70 net/socket.c:2669 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3b/0x90 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x72/0xdc value changed: 0x0000000b -> 0x00000001 Reported by Kernel Concurrency Sanitizer on: CPU: 1 PID: 49793 Comm: syz-executor.0 Not tainted 6.3.0-rc7-02330-gca6270c12e20 #2 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Reviewed-by: Michal Kubiak Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index fb31e8a4409e..08102e728b15 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1442,7 +1442,7 @@ static long unix_wait_for_peer(struct sock *other, long timeo) sched = !sock_flag(other, SOCK_DEAD) && !(other->sk_shutdown & RCV_SHUTDOWN) && - unix_recvq_full(other); + unix_recvq_full_lockless(other); unix_state_unlock(other); -- cgit v1.2.3 From e1d09c2c2f5793474556b60f83900e088d0d366d Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 9 May 2023 17:34:56 -0700 Subject: af_unix: Fix data races around sk->sk_shutdown. KCSAN found a data race around sk->sk_shutdown where unix_release_sock() and unix_shutdown() update it under unix_state_lock(), OTOH unix_poll() and unix_dgram_poll() read it locklessly. We need to annotate the writes and reads with WRITE_ONCE() and READ_ONCE(). BUG: KCSAN: data-race in unix_poll / unix_release_sock write to 0xffff88800d0f8aec of 1 bytes by task 264 on cpu 0: unix_release_sock+0x75c/0x910 net/unix/af_unix.c:631 unix_release+0x59/0x80 net/unix/af_unix.c:1042 __sock_release+0x7d/0x170 net/socket.c:653 sock_close+0x19/0x30 net/socket.c:1397 __fput+0x179/0x5e0 fs/file_table.c:321 ____fput+0x15/0x20 fs/file_table.c:349 task_work_run+0x116/0x1a0 kernel/task_work.c:179 resume_user_mode_work include/linux/resume_user_mode.h:49 [inline] exit_to_user_mode_loop kernel/entry/common.c:171 [inline] exit_to_user_mode_prepare+0x174/0x180 kernel/entry/common.c:204 __syscall_exit_to_user_mode_work kernel/entry/common.c:286 [inline] syscall_exit_to_user_mode+0x1a/0x30 kernel/entry/common.c:297 do_syscall_64+0x4b/0x90 arch/x86/entry/common.c:86 entry_SYSCALL_64_after_hwframe+0x72/0xdc read to 0xffff88800d0f8aec of 1 bytes by task 222 on cpu 1: unix_poll+0xa3/0x2a0 net/unix/af_unix.c:3170 sock_poll+0xcf/0x2b0 net/socket.c:1385 vfs_poll include/linux/poll.h:88 [inline] ep_item_poll.isra.0+0x78/0xc0 fs/eventpoll.c:855 ep_send_events fs/eventpoll.c:1694 [inline] ep_poll fs/eventpoll.c:1823 [inline] do_epoll_wait+0x6c4/0xea0 fs/eventpoll.c:2258 __do_sys_epoll_wait fs/eventpoll.c:2270 [inline] __se_sys_epoll_wait fs/eventpoll.c:2265 [inline] __x64_sys_epoll_wait+0xcc/0x190 fs/eventpoll.c:2265 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x3b/0x90 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x72/0xdc value changed: 0x00 -> 0x03 Reported by Kernel Concurrency Sanitizer on: CPU: 1 PID: 222 Comm: dbus-broker Not tainted 6.3.0-rc7-02330-gca6270c12e20 #2 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 Fixes: 3c73419c09a5 ("af_unix: fix 'poll for write'/ connected DGRAM sockets") Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Reviewed-by: Michal Kubiak Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 08102e728b15..cc695c9f09ec 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -603,7 +603,7 @@ static void unix_release_sock(struct sock *sk, int embrion) /* Clear state */ unix_state_lock(sk); sock_orphan(sk); - sk->sk_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(sk->sk_shutdown, SHUTDOWN_MASK); path = u->path; u->path.dentry = NULL; u->path.mnt = NULL; @@ -628,7 +628,7 @@ static void unix_release_sock(struct sock *sk, int embrion) if (sk->sk_type == SOCK_STREAM || sk->sk_type == SOCK_SEQPACKET) { unix_state_lock(skpair); /* No more writes */ - skpair->sk_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(skpair->sk_shutdown, SHUTDOWN_MASK); if (!skb_queue_empty(&sk->sk_receive_queue) || embrion) WRITE_ONCE(skpair->sk_err, ECONNRESET); unix_state_unlock(skpair); @@ -3008,7 +3008,7 @@ static int unix_shutdown(struct socket *sock, int mode) ++mode; unix_state_lock(sk); - sk->sk_shutdown |= mode; + WRITE_ONCE(sk->sk_shutdown, sk->sk_shutdown | mode); other = unix_peer(sk); if (other) sock_hold(other); @@ -3028,7 +3028,7 @@ static int unix_shutdown(struct socket *sock, int mode) if (mode&SEND_SHUTDOWN) peer_mode |= RCV_SHUTDOWN; unix_state_lock(other); - other->sk_shutdown |= peer_mode; + WRITE_ONCE(other->sk_shutdown, other->sk_shutdown | peer_mode); unix_state_unlock(other); other->sk_state_change(other); if (peer_mode == SHUTDOWN_MASK) @@ -3160,16 +3160,18 @@ static __poll_t unix_poll(struct file *file, struct socket *sock, poll_table *wa { struct sock *sk = sock->sk; __poll_t mask; + u8 shutdown; sock_poll_wait(file, sock, wait); mask = 0; + shutdown = READ_ONCE(sk->sk_shutdown); /* exceptional events? */ if (READ_ONCE(sk->sk_err)) mask |= EPOLLERR; - if (sk->sk_shutdown == SHUTDOWN_MASK) + if (shutdown == SHUTDOWN_MASK) mask |= EPOLLHUP; - if (sk->sk_shutdown & RCV_SHUTDOWN) + if (shutdown & RCV_SHUTDOWN) mask |= EPOLLRDHUP | EPOLLIN | EPOLLRDNORM; /* readable? */ @@ -3203,9 +3205,11 @@ static __poll_t unix_dgram_poll(struct file *file, struct socket *sock, struct sock *sk = sock->sk, *other; unsigned int writable; __poll_t mask; + u8 shutdown; sock_poll_wait(file, sock, wait); mask = 0; + shutdown = READ_ONCE(sk->sk_shutdown); /* exceptional events? */ if (READ_ONCE(sk->sk_err) || @@ -3213,9 +3217,9 @@ static __poll_t unix_dgram_poll(struct file *file, struct socket *sock, mask |= EPOLLERR | (sock_flag(sk, SOCK_SELECT_ERR_QUEUE) ? EPOLLPRI : 0); - if (sk->sk_shutdown & RCV_SHUTDOWN) + if (shutdown & RCV_SHUTDOWN) mask |= EPOLLRDHUP | EPOLLIN | EPOLLRDNORM; - if (sk->sk_shutdown == SHUTDOWN_MASK) + if (shutdown == SHUTDOWN_MASK) mask |= EPOLLHUP; /* readable? */ -- cgit v1.2.3 From e93c9378e33f68b61ea9318580d841caa22fb9ea Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 10 May 2023 16:46:21 +0200 Subject: devlink: change per-devlink netdev notifier to static one The commit 565b4824c39f ("devlink: change port event netdev notifier from per-net to global") changed original per-net notifier to be per-devlink instance. That fixed the issue of non-receiving events of netdev uninit if that moved to a different namespace. That worked fine in -net tree. However, later on when commit ee75f1fc44dd ("net/mlx5e: Create separate devlink instance for ethernet auxiliary device") and commit 72ed5d5624af ("net/mlx5: Suspend auxiliary devices only in case of PCI device suspend") were merged, a deadlock was introduced when removing a namespace with devlink instance with another nested instance. Here there is the bad flow example resulting in deadlock with mlx5: net_cleanup_work -> cleanup_net (takes down_read(&pernet_ops_rwsem) -> devlink_pernet_pre_exit() -> devlink_reload() -> mlx5_devlink_reload_down() -> mlx5_unload_one_devl_locked() -> mlx5_detach_device() -> del_adev() -> mlx5e_remove() -> mlx5e_destroy_devlink() -> devlink_free() -> unregister_netdevice_notifier() (takes down_write(&pernet_ops_rwsem) Steps to reproduce: $ modprobe mlx5_core $ ip netns add ns1 $ devlink dev reload pci/0000:08:00.0 netns ns1 $ ip netns del ns1 Resolve this by converting the notifier from per-devlink instance to a static one registered during init phase and leaving it registered forever. Use this notifier for all devlink port instances created later on. Note what a tree needs this fix only in case all of the cited fixes commits are present. Reported-by: Moshe Shemesh Fixes: 565b4824c39f ("devlink: change port event netdev notifier from per-net to global") Fixes: ee75f1fc44dd ("net/mlx5e: Create separate devlink instance for ethernet auxiliary device") Fixes: 72ed5d5624af ("net/mlx5: Suspend auxiliary devices only in case of PCI device suspend") Signed-off-by: Jiri Pirko Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230510144621.932017-1-jiri@resnulli.us Signed-off-by: Jakub Kicinski --- net/devlink/core.c | 16 +++++++--------- net/devlink/devl_internal.h | 1 - net/devlink/leftover.c | 5 ++--- 3 files changed, 9 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/devlink/core.c b/net/devlink/core.c index 777b091ef74d..0e58eee44bdb 100644 --- a/net/devlink/core.c +++ b/net/devlink/core.c @@ -204,11 +204,6 @@ struct devlink *devlink_alloc_ns(const struct devlink_ops *ops, if (ret < 0) goto err_xa_alloc; - devlink->netdevice_nb.notifier_call = devlink_port_netdevice_event; - ret = register_netdevice_notifier(&devlink->netdevice_nb); - if (ret) - goto err_register_netdevice_notifier; - devlink->dev = dev; devlink->ops = ops; xa_init_flags(&devlink->ports, XA_FLAGS_ALLOC); @@ -233,8 +228,6 @@ struct devlink *devlink_alloc_ns(const struct devlink_ops *ops, return devlink; -err_register_netdevice_notifier: - xa_erase(&devlinks, devlink->index); err_xa_alloc: kfree(devlink); return NULL; @@ -266,8 +259,6 @@ void devlink_free(struct devlink *devlink) xa_destroy(&devlink->params); xa_destroy(&devlink->ports); - WARN_ON_ONCE(unregister_netdevice_notifier(&devlink->netdevice_nb)); - xa_erase(&devlinks, devlink->index); devlink_put(devlink); @@ -303,6 +294,10 @@ static struct pernet_operations devlink_pernet_ops __net_initdata = { .pre_exit = devlink_pernet_pre_exit, }; +static struct notifier_block devlink_port_netdevice_nb __net_initdata = { + .notifier_call = devlink_port_netdevice_event, +}; + static int __init devlink_init(void) { int err; @@ -311,6 +306,9 @@ static int __init devlink_init(void) if (err) goto out; err = register_pernet_subsys(&devlink_pernet_ops); + if (err) + goto out; + err = register_netdevice_notifier(&devlink_port_netdevice_nb); out: WARN_ON(err); diff --git a/net/devlink/devl_internal.h b/net/devlink/devl_internal.h index e133f423294a..62921b2eb0d3 100644 --- a/net/devlink/devl_internal.h +++ b/net/devlink/devl_internal.h @@ -50,7 +50,6 @@ struct devlink { u8 reload_failed:1; refcount_t refcount; struct rcu_work rwork; - struct notifier_block netdevice_nb; char priv[] __aligned(NETDEV_ALIGN); }; diff --git a/net/devlink/leftover.c b/net/devlink/leftover.c index dffca2f9bfa7..cd0254968076 100644 --- a/net/devlink/leftover.c +++ b/net/devlink/leftover.c @@ -7073,10 +7073,9 @@ int devlink_port_netdevice_event(struct notifier_block *nb, struct devlink_port *devlink_port = netdev->devlink_port; struct devlink *devlink; - devlink = container_of(nb, struct devlink, netdevice_nb); - - if (!devlink_port || devlink_port->devlink != devlink) + if (!devlink_port) return NOTIFY_OK; + devlink = devlink_port->devlink; switch (event) { case NETDEV_POST_INIT: -- cgit v1.2.3 From ef1148d4487438a3408d6face2a8360d91b4af70 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 10 May 2023 15:46:46 +0000 Subject: ipv6: remove nexthop_fib6_nh_bh() After blamed commit, nexthop_fib6_nh_bh() and nexthop_fib6_nh() are the same. Delete nexthop_fib6_nh_bh(), and convert /proc/net/ipv6_route to standard rcu to avoid this splat: [ 5723.180080] WARNING: suspicious RCU usage [ 5723.180083] ----------------------------- [ 5723.180084] include/net/nexthop.h:516 suspicious rcu_dereference_check() usage! [ 5723.180086] other info that might help us debug this: [ 5723.180087] rcu_scheduler_active = 2, debug_locks = 1 [ 5723.180089] 2 locks held by cat/55856: [ 5723.180091] #0: ffff9440a582afa8 (&p->lock){+.+.}-{3:3}, at: seq_read_iter (fs/seq_file.c:188) [ 5723.180100] #1: ffffffffaac07040 (rcu_read_lock_bh){....}-{1:2}, at: rcu_lock_acquire (include/linux/rcupdate.h:326) [ 5723.180109] stack backtrace: [ 5723.180111] CPU: 14 PID: 55856 Comm: cat Tainted: G S I 6.3.0-dbx-DEV #528 [ 5723.180115] Call Trace: [ 5723.180117] [ 5723.180119] dump_stack_lvl (lib/dump_stack.c:107) [ 5723.180124] dump_stack (lib/dump_stack.c:114) [ 5723.180126] lockdep_rcu_suspicious (include/linux/context_tracking.h:122) [ 5723.180132] ipv6_route_seq_show (include/net/nexthop.h:?) [ 5723.180135] ? ipv6_route_seq_next (net/ipv6/ip6_fib.c:2605) [ 5723.180140] seq_read_iter (fs/seq_file.c:272) [ 5723.180145] seq_read (fs/seq_file.c:163) [ 5723.180151] proc_reg_read (fs/proc/inode.c:316 fs/proc/inode.c:328) [ 5723.180155] vfs_read (fs/read_write.c:468) [ 5723.180160] ? up_read (kernel/locking/rwsem.c:1617) [ 5723.180164] ksys_read (fs/read_write.c:613) [ 5723.180168] __x64_sys_read (fs/read_write.c:621) [ 5723.180170] do_syscall_64 (arch/x86/entry/common.c:?) [ 5723.180174] entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120) [ 5723.180177] RIP: 0033:0x7fa455677d2a Fixes: 09eed1192cec ("neighbour: switch to standard rcu, instead of rcu_bh") Reported-by: syzbot Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20230510154646.370659-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/nexthop.h | 23 ----------------------- net/ipv6/ip6_fib.c | 16 ++++++++-------- 2 files changed, 8 insertions(+), 31 deletions(-) (limited to 'net') diff --git a/include/net/nexthop.h b/include/net/nexthop.h index 9fa291a04621..2b12725de9c0 100644 --- a/include/net/nexthop.h +++ b/include/net/nexthop.h @@ -497,29 +497,6 @@ static inline struct fib6_nh *nexthop_fib6_nh(struct nexthop *nh) return NULL; } -/* Variant of nexthop_fib6_nh(). - * Caller should either hold rcu_read_lock(), or RTNL. - */ -static inline struct fib6_nh *nexthop_fib6_nh_bh(struct nexthop *nh) -{ - struct nh_info *nhi; - - if (nh->is_group) { - struct nh_group *nh_grp; - - nh_grp = rcu_dereference_rtnl(nh->nh_grp); - nh = nexthop_mpath_select(nh_grp, 0); - if (!nh) - return NULL; - } - - nhi = rcu_dereference_rtnl(nh->nh_info); - if (nhi->family == AF_INET6) - return &nhi->fib6_nh; - - return NULL; -} - static inline struct net_device *fib6_info_nh_dev(struct fib6_info *f6i) { struct fib6_nh *fib6_nh; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 2438da5ff6da..bac768d36cc1 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -2491,7 +2491,7 @@ static int ipv6_route_native_seq_show(struct seq_file *seq, void *v) const struct net_device *dev; if (rt->nh) - fib6_nh = nexthop_fib6_nh_bh(rt->nh); + fib6_nh = nexthop_fib6_nh(rt->nh); seq_printf(seq, "%pi6 %02x ", &rt->fib6_dst.addr, rt->fib6_dst.plen); @@ -2556,14 +2556,14 @@ static struct fib6_table *ipv6_route_seq_next_table(struct fib6_table *tbl, if (tbl) { h = (tbl->tb6_id & (FIB6_TABLE_HASHSZ - 1)) + 1; - node = rcu_dereference_bh(hlist_next_rcu(&tbl->tb6_hlist)); + node = rcu_dereference(hlist_next_rcu(&tbl->tb6_hlist)); } else { h = 0; node = NULL; } while (!node && h < FIB6_TABLE_HASHSZ) { - node = rcu_dereference_bh( + node = rcu_dereference( hlist_first_rcu(&net->ipv6.fib_table_hash[h++])); } return hlist_entry_safe(node, struct fib6_table, tb6_hlist); @@ -2593,7 +2593,7 @@ static void *ipv6_route_seq_next(struct seq_file *seq, void *v, loff_t *pos) if (!v) goto iter_table; - n = rcu_dereference_bh(((struct fib6_info *)v)->fib6_next); + n = rcu_dereference(((struct fib6_info *)v)->fib6_next); if (n) return n; @@ -2619,12 +2619,12 @@ iter_table: } static void *ipv6_route_seq_start(struct seq_file *seq, loff_t *pos) - __acquires(RCU_BH) + __acquires(RCU) { struct net *net = seq_file_net(seq); struct ipv6_route_iter *iter = seq->private; - rcu_read_lock_bh(); + rcu_read_lock(); iter->tbl = ipv6_route_seq_next_table(NULL, net); iter->skip = *pos; @@ -2645,7 +2645,7 @@ static bool ipv6_route_iter_active(struct ipv6_route_iter *iter) } static void ipv6_route_native_seq_stop(struct seq_file *seq, void *v) - __releases(RCU_BH) + __releases(RCU) { struct net *net = seq_file_net(seq); struct ipv6_route_iter *iter = seq->private; @@ -2653,7 +2653,7 @@ static void ipv6_route_native_seq_stop(struct seq_file *seq, void *v) if (ipv6_route_iter_active(iter)) fib6_walker_unlink(net, &iter->w); - rcu_read_unlock_bh(); + rcu_read_unlock(); } #if IS_BUILTIN(CONFIG_IPV6) && defined(CONFIG_BPF_SYSCALL) -- cgit v1.2.3 From 6d4486efe9c69626cab423456169e250a5cd3af5 Mon Sep 17 00:00:00 2001 From: Zhuang Shengen Date: Thu, 11 May 2023 19:34:30 +0800 Subject: vsock: avoid to close connected socket after the timeout MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When client and server establish a connection through vsock, the client send a request to the server to initiate the connection, then start a timer to wait for the server's response. When the server's RESPONSE message arrives, the timer also times out and exits. The server's RESPONSE message is processed first, and the connection is established. However, the client's timer also times out, the original processing logic of the client is to directly set the state of this vsock to CLOSE and return ETIMEDOUT. It will not notify the server when the port is released, causing the server port remain. when client's vsock_connect timeout，it should check sk state is ESTABLISHED or not. if sk state is ESTABLISHED, it means the connection is established, the client should not set the sk state to CLOSE Note: I encountered this issue on kernel-4.18, which can be fixed by this patch. Then I checked the latest code in the community and found similar issue. Fixes: d021c344051a ("VSOCK: Introduce VM Sockets") Signed-off-by: Zhuang Shengen Reviewed-by: Stefano Garzarella Signed-off-by: David S. Miller --- net/vmw_vsock/af_vsock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 413407bb646c..efb8a0937a13 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -1462,7 +1462,7 @@ static int vsock_connect(struct socket *sock, struct sockaddr *addr, vsock_transport_cancel_pkt(vsk); vsock_remove_connected(vsk); goto out_wait; - } else if (timeout == 0) { + } else if ((sk->sk_state != TCP_ESTABLISHED) && (timeout == 0)) { err = -ETIMEDOUT; sk->sk_state = TCP_CLOSE; sock->state = SS_UNCONNECTED; -- cgit v1.2.3 From 1e306ec49a1f206fd2cc89a42fac6e6f592a8cc1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 11 May 2023 11:47:49 +0000 Subject: tcp: fix possible sk_priority leak in tcp_v4_send_reset() When tcp_v4_send_reset() is called with @sk == NULL, we do not change ctl_sk->sk_priority, which could have been set from a prior invocation. Change tcp_v4_send_reset() to set sk_priority and sk_mark fields before calling ip_send_unicast_reply(). This means tcp_v4_send_reset() and tcp_v4_send_ack() no longer have to clear ctl_sk->sk_mark after their call to ip_send_unicast_reply(). Fixes: f6c0f5d209fa ("tcp: honor SO_PRIORITY in TIME_WAIT state") Signed-off-by: Eric Dumazet Cc: Antoine Tenart Signed-off-by: David S. Miller --- net/ipv4/tcp_ipv4.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 39bda2b1066e..06d2573685ca 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -829,6 +829,9 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) inet_twsk(sk)->tw_priority : sk->sk_priority; transmit_time = tcp_transmit_time(sk); xfrm_sk_clone_policy(ctl_sk, sk); + } else { + ctl_sk->sk_mark = 0; + ctl_sk->sk_priority = 0; } ip_send_unicast_reply(ctl_sk, skb, &TCP_SKB_CB(skb)->header.h4.opt, @@ -836,7 +839,6 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) &arg, arg.iov[0].iov_len, transmit_time); - ctl_sk->sk_mark = 0; xfrm_sk_free_policy(ctl_sk); sock_net_set(ctl_sk, &init_net); __TCP_INC_STATS(net, TCP_MIB_OUTSEGS); @@ -935,7 +937,6 @@ static void tcp_v4_send_ack(const struct sock *sk, &arg, arg.iov[0].iov_len, transmit_time); - ctl_sk->sk_mark = 0; sock_net_set(ctl_sk, &init_net); __TCP_INC_STATS(net, TCP_MIB_OUTSEGS); local_bh_enable(); -- cgit v1.2.3 From d80fc101d2eb9b3188c228d61223890aeea480a4 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Thu, 11 May 2023 19:22:11 -0400 Subject: erspan: get the proto with the md version for collect_md In commit 20704bd1633d ("erspan: build the header with the right proto according to erspan_ver"), it gets the proto with t->parms.erspan_ver, but t->parms.erspan_ver is not used by collect_md branch, and instead it should get the proto with md->version for collect_md. Thanks to Kevin for pointing this out. Fixes: 20704bd1633d ("erspan: build the header with the right proto according to erspan_ver") Fixes: 94d7d8f29287 ("ip6_gre: add erspan v2 support") Reported-by: Kevin Traynor Signed-off-by: Xin Long Reviewed-by: Simon Horman Reviewed-by: William Tu Signed-off-by: David S. Miller --- net/ipv6/ip6_gre.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index a4ecfc9d2593..da80974ad23a 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -1015,12 +1015,14 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb, ntohl(tun_id), ntohl(md->u.index), truncate, false); + proto = htons(ETH_P_ERSPAN); } else if (md->version == 2) { erspan_build_header_v2(skb, ntohl(tun_id), md->u.md2.dir, get_hwid(&md->u.md2), truncate, false); + proto = htons(ETH_P_ERSPAN2); } else { goto tx_err; } @@ -1043,24 +1045,25 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb, break; } - if (t->parms.erspan_ver == 1) + if (t->parms.erspan_ver == 1) { erspan_build_header(skb, ntohl(t->parms.o_key), t->parms.index, truncate, false); - else if (t->parms.erspan_ver == 2) + proto = htons(ETH_P_ERSPAN); + } else if (t->parms.erspan_ver == 2) { erspan_build_header_v2(skb, ntohl(t->parms.o_key), t->parms.dir, t->parms.hwid, truncate, false); - else + proto = htons(ETH_P_ERSPAN2); + } else { goto tx_err; + } fl6.daddr = t->parms.raddr; } /* Push GRE header. */ - proto = (t->parms.erspan_ver == 1) ? htons(ETH_P_ERSPAN) - : htons(ETH_P_ERSPAN2); gre_build_header(skb, 8, TUNNEL_SEQ, proto, 0, htonl(atomic_fetch_inc(&t->o_seqno))); /* TooBig packet may have updated dst->dev's mtu */ -- cgit v1.2.3 From eb8d3a2c809abd73ab0a060fe971d6b9019aa3c1 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 9 May 2023 09:41:49 +1000 Subject: SUNRPC: double free xprt_ctxt while still in use When an RPC request is deferred, the rq_xprt_ctxt pointer is moved out of the svc_rqst into the svc_deferred_req. When the deferred request is revisited, the pointer is copied into the new svc_rqst - and also remains in the svc_deferred_req. In the (rare?) case that the request is deferred a second time, the old svc_deferred_req is reused - it still has all the correct content. However in that case the rq_xprt_ctxt pointer is NOT cleared so that when xpo_release_xprt is called, the ctxt is freed (UDP) or possible added to a free list (RDMA). When the deferred request is revisited for a second time, it will reference this ctxt which may be invalid, and the free the object a second time which is likely to oops. So change svc_defer() to *always* clear rq_xprt_ctxt, and assert that the value is now stored in the svc_deferred_req. Fixes: 773f91b2cf3f ("SUNRPC: Fix NFSD's request deferral on RDMA transports") Signed-off-by: NeilBrown Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- net/sunrpc/svc_xprt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 84e5d7d31481..5fd94f6bdc75 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -1223,13 +1223,14 @@ static struct cache_deferred_req *svc_defer(struct cache_req *req) dr->daddr = rqstp->rq_daddr; dr->argslen = rqstp->rq_arg.len >> 2; dr->xprt_ctxt = rqstp->rq_xprt_ctxt; - rqstp->rq_xprt_ctxt = NULL; /* back up head to the start of the buffer and copy */ skip = rqstp->rq_arg.len - rqstp->rq_arg.head[0].iov_len; memcpy(dr->args, rqstp->rq_arg.head[0].iov_base - skip, dr->argslen << 2); } + WARN_ON_ONCE(rqstp->rq_xprt_ctxt != dr->xprt_ctxt); + rqstp->rq_xprt_ctxt = NULL; trace_svc_defer(rqstp); svc_xprt_get(rqstp->rq_xprt); dr->xprt = rqstp->rq_xprt; -- cgit v1.2.3 From 948f072ada23e0a504c5e4d7d71d4c83bd0785ec Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 9 May 2023 09:42:47 +1000 Subject: SUNRPC: always free ctxt when freeing deferred request Since the ->xprt_ctxt pointer was added to svc_deferred_req, it has not been sufficient to use kfree() to free a deferred request. We may need to free the ctxt as well. As freeing the ctxt is all that ->xpo_release_rqst() does, we repurpose it to explicit do that even when the ctxt is not stored in an rqst. So we now have ->xpo_release_ctxt() which is given an xprt and a ctxt, which may have been taken either from an rqst or from a dreq. The caller is now responsible for clearing that pointer after the call to ->xpo_release_ctxt. We also clear dr->xprt_ctxt when the ctxt is moved into a new rqst when revisiting a deferred request. This ensures there is only one pointer to the ctxt, so the risk of double freeing in future is reduced. The new code in svc_xprt_release which releases both the ctxt and any rq_deferred depends on this. Fixes: 773f91b2cf3f ("SUNRPC: Fix NFSD's request deferral on RDMA transports") Signed-off-by: NeilBrown Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- include/linux/sunrpc/svc_rdma.h | 2 +- include/linux/sunrpc/svc_xprt.h | 2 +- net/sunrpc/svc_xprt.c | 23 +++++++++++++++++------ net/sunrpc/svcsock.c | 30 ++++++++++++++++-------------- net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 11 +++++------ net/sunrpc/xprtrdma/svc_rdma_transport.c | 2 +- 6 files changed, 41 insertions(+), 29 deletions(-) (limited to 'net') diff --git a/include/linux/sunrpc/svc_rdma.h b/include/linux/sunrpc/svc_rdma.h index 24aa159d29a7..fbc4bd423b35 100644 --- a/include/linux/sunrpc/svc_rdma.h +++ b/include/linux/sunrpc/svc_rdma.h @@ -176,7 +176,7 @@ extern struct svc_rdma_recv_ctxt * extern void svc_rdma_recv_ctxt_put(struct svcxprt_rdma *rdma, struct svc_rdma_recv_ctxt *ctxt); extern void svc_rdma_flush_recv_queues(struct svcxprt_rdma *rdma); -extern void svc_rdma_release_rqst(struct svc_rqst *rqstp); +extern void svc_rdma_release_ctxt(struct svc_xprt *xprt, void *ctxt); extern int svc_rdma_recvfrom(struct svc_rqst *); /* svc_rdma_rw.c */ diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index 867479204840..a6b12631db21 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -23,7 +23,7 @@ struct svc_xprt_ops { int (*xpo_sendto)(struct svc_rqst *); int (*xpo_result_payload)(struct svc_rqst *, unsigned int, unsigned int); - void (*xpo_release_rqst)(struct svc_rqst *); + void (*xpo_release_ctxt)(struct svc_xprt *xprt, void *ctxt); void (*xpo_detach)(struct svc_xprt *); void (*xpo_free)(struct svc_xprt *); void (*xpo_kill_temp_xprt)(struct svc_xprt *); diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 5fd94f6bdc75..13a14897bc17 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -532,13 +532,23 @@ void svc_reserve(struct svc_rqst *rqstp, int space) } EXPORT_SYMBOL_GPL(svc_reserve); +static void free_deferred(struct svc_xprt *xprt, struct svc_deferred_req *dr) +{ + if (!dr) + return; + + xprt->xpt_ops->xpo_release_ctxt(xprt, dr->xprt_ctxt); + kfree(dr); +} + static void svc_xprt_release(struct svc_rqst *rqstp) { struct svc_xprt *xprt = rqstp->rq_xprt; - xprt->xpt_ops->xpo_release_rqst(rqstp); + xprt->xpt_ops->xpo_release_ctxt(xprt, rqstp->rq_xprt_ctxt); + rqstp->rq_xprt_ctxt = NULL; - kfree(rqstp->rq_deferred); + free_deferred(xprt, rqstp->rq_deferred); rqstp->rq_deferred = NULL; svc_rqst_release_pages(rqstp); @@ -1054,7 +1064,7 @@ static void svc_delete_xprt(struct svc_xprt *xprt) spin_unlock_bh(&serv->sv_lock); while ((dr = svc_deferred_dequeue(xprt)) != NULL) - kfree(dr); + free_deferred(xprt, dr); call_xpt_users(xprt); svc_xprt_put(xprt); @@ -1176,8 +1186,8 @@ static void svc_revisit(struct cache_deferred_req *dreq, int too_many) if (too_many || test_bit(XPT_DEAD, &xprt->xpt_flags)) { spin_unlock(&xprt->xpt_lock); trace_svc_defer_drop(dr); + free_deferred(xprt, dr); svc_xprt_put(xprt); - kfree(dr); return; } dr->xprt = NULL; @@ -1222,14 +1232,13 @@ static struct cache_deferred_req *svc_defer(struct cache_req *req) dr->addrlen = rqstp->rq_addrlen; dr->daddr = rqstp->rq_daddr; dr->argslen = rqstp->rq_arg.len >> 2; - dr->xprt_ctxt = rqstp->rq_xprt_ctxt; /* back up head to the start of the buffer and copy */ skip = rqstp->rq_arg.len - rqstp->rq_arg.head[0].iov_len; memcpy(dr->args, rqstp->rq_arg.head[0].iov_base - skip, dr->argslen << 2); } - WARN_ON_ONCE(rqstp->rq_xprt_ctxt != dr->xprt_ctxt); + dr->xprt_ctxt = rqstp->rq_xprt_ctxt; rqstp->rq_xprt_ctxt = NULL; trace_svc_defer(rqstp); svc_xprt_get(rqstp->rq_xprt); @@ -1263,6 +1272,8 @@ static noinline int svc_deferred_recv(struct svc_rqst *rqstp) rqstp->rq_daddr = dr->daddr; rqstp->rq_respages = rqstp->rq_pages; rqstp->rq_xprt_ctxt = dr->xprt_ctxt; + + dr->xprt_ctxt = NULL; svc_xprt_received(rqstp->rq_xprt); return dr->argslen << 2; } diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 9989194446a5..63fe7a338992 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -121,27 +121,27 @@ static void svc_reclassify_socket(struct socket *sock) #endif /** - * svc_tcp_release_rqst - Release transport-related resources - * @rqstp: request structure with resources to be released + * svc_tcp_release_ctxt - Release transport-related resources + * @xprt: the transport which owned the context + * @ctxt: the context from rqstp->rq_xprt_ctxt or dr->xprt_ctxt * */ -static void svc_tcp_release_rqst(struct svc_rqst *rqstp) +static void svc_tcp_release_ctxt(struct svc_xprt *xprt, void *ctxt) { } /** - * svc_udp_release_rqst - Release transport-related resources - * @rqstp: request structure with resources to be released + * svc_udp_release_ctxt - Release transport-related resources + * @xprt: the transport which owned the context + * @ctxt: the context from rqstp->rq_xprt_ctxt or dr->xprt_ctxt * */ -static void svc_udp_release_rqst(struct svc_rqst *rqstp) +static void svc_udp_release_ctxt(struct svc_xprt *xprt, void *ctxt) { - struct sk_buff *skb = rqstp->rq_xprt_ctxt; + struct sk_buff *skb = ctxt; - if (skb) { - rqstp->rq_xprt_ctxt = NULL; + if (skb) consume_skb(skb); - } } union svc_pktinfo_u { @@ -696,7 +696,8 @@ static int svc_udp_sendto(struct svc_rqst *rqstp) unsigned int sent; int err; - svc_udp_release_rqst(rqstp); + svc_udp_release_ctxt(xprt, rqstp->rq_xprt_ctxt); + rqstp->rq_xprt_ctxt = NULL; svc_set_cmsg_data(rqstp, cmh); @@ -768,7 +769,7 @@ static const struct svc_xprt_ops svc_udp_ops = { .xpo_recvfrom = svc_udp_recvfrom, .xpo_sendto = svc_udp_sendto, .xpo_result_payload = svc_sock_result_payload, - .xpo_release_rqst = svc_udp_release_rqst, + .xpo_release_ctxt = svc_udp_release_ctxt, .xpo_detach = svc_sock_detach, .xpo_free = svc_sock_free, .xpo_has_wspace = svc_udp_has_wspace, @@ -1301,7 +1302,8 @@ static int svc_tcp_sendto(struct svc_rqst *rqstp) unsigned int sent; int err; - svc_tcp_release_rqst(rqstp); + svc_tcp_release_ctxt(xprt, rqstp->rq_xprt_ctxt); + rqstp->rq_xprt_ctxt = NULL; atomic_inc(&svsk->sk_sendqlen); mutex_lock(&xprt->xpt_mutex); @@ -1346,7 +1348,7 @@ static const struct svc_xprt_ops svc_tcp_ops = { .xpo_recvfrom = svc_tcp_recvfrom, .xpo_sendto = svc_tcp_sendto, .xpo_result_payload = svc_sock_result_payload, - .xpo_release_rqst = svc_tcp_release_rqst, + .xpo_release_ctxt = svc_tcp_release_ctxt, .xpo_detach = svc_tcp_sock_detach, .xpo_free = svc_sock_free, .xpo_has_wspace = svc_tcp_has_wspace, diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c index 1c658fa43063..a22fe7587fa6 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c +++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c @@ -239,21 +239,20 @@ void svc_rdma_recv_ctxt_put(struct svcxprt_rdma *rdma, } /** - * svc_rdma_release_rqst - Release transport-specific per-rqst resources - * @rqstp: svc_rqst being released + * svc_rdma_release_ctxt - Release transport-specific per-rqst resources + * @xprt: the transport which owned the context + * @vctxt: the context from rqstp->rq_xprt_ctxt or dr->xprt_ctxt * * Ensure that the recv_ctxt is released whether or not a Reply * was sent. For example, the client could close the connection, * or svc_process could drop an RPC, before the Reply is sent. */ -void svc_rdma_release_rqst(struct svc_rqst *rqstp) +void svc_rdma_release_ctxt(struct svc_xprt *xprt, void *vctxt) { - struct svc_rdma_recv_ctxt *ctxt = rqstp->rq_xprt_ctxt; - struct svc_xprt *xprt = rqstp->rq_xprt; + struct svc_rdma_recv_ctxt *ctxt = vctxt; struct svcxprt_rdma *rdma = container_of(xprt, struct svcxprt_rdma, sc_xprt); - rqstp->rq_xprt_ctxt = NULL; if (ctxt) svc_rdma_recv_ctxt_put(rdma, ctxt); } diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c index 416b298f74dd..ca04f7a6a085 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_transport.c +++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c @@ -80,7 +80,7 @@ static const struct svc_xprt_ops svc_rdma_ops = { .xpo_recvfrom = svc_rdma_recvfrom, .xpo_sendto = svc_rdma_sendto, .xpo_result_payload = svc_rdma_result_payload, - .xpo_release_rqst = svc_rdma_release_rqst, + .xpo_release_ctxt = svc_rdma_release_ctxt, .xpo_detach = svc_rdma_detach, .xpo_free = svc_rdma_free, .xpo_has_wspace = svc_rdma_has_wspace, -- cgit v1.2.3 From 07a27305938559fb35f7a46fb90a5e37728bdee6 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Sun, 14 May 2023 15:51:48 -0400 Subject: SUNRPC: Fix trace_svc_register() call site The trace event recorded incorrect values for the registered family, protocol, and port because the arguments are in the wrong order. Fixes: b4af59328c25 ("SUNRPC: Trace server-side rpcbind registration events") Signed-off-by: Chuck Lever --- net/sunrpc/svc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 43e32129a583..79967b6925bd 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -1052,7 +1052,7 @@ static int __svc_register(struct net *net, const char *progname, #endif } - trace_svc_register(progname, version, protocol, port, family, error); + trace_svc_register(progname, version, family, protocol, port, error); return error; } -- cgit v1.2.3 From c83b49383b595be50647f0c764a48c78b5f3c4f8 Mon Sep 17 00:00:00 2001 From: Dong Chenchen Date: Thu, 11 May 2023 20:54:40 +0800 Subject: net: nsh: Use correct mac_offset to unwind gso skb in nsh_gso_segment() As the call trace shows, skb_panic was caused by wrong skb->mac_header in nsh_gso_segment(): invalid opcode: 0000 [#1] PREEMPT SMP KASAN PTI CPU: 3 PID: 2737 Comm: syz Not tainted 6.3.0-next-20230505 #1 RIP: 0010:skb_panic+0xda/0xe0 call Trace: skb_push+0x91/0xa0 nsh_gso_segment+0x4f3/0x570 skb_mac_gso_segment+0x19e/0x270 __skb_gso_segment+0x1e8/0x3c0 validate_xmit_skb+0x452/0x890 validate_xmit_skb_list+0x99/0xd0 sch_direct_xmit+0x294/0x7c0 __dev_queue_xmit+0x16f0/0x1d70 packet_xmit+0x185/0x210 packet_snd+0xc15/0x1170 packet_sendmsg+0x7b/0xa0 sock_sendmsg+0x14f/0x160 The root cause is: nsh_gso_segment() use skb->network_header - nhoff to reset mac_header in skb_gso_error_unwind() if inner-layer protocol gso fails. However, skb->network_header may be reset by inner-layer protocol gso function e.g. mpls_gso_segment. skb->mac_header reset by the inaccurate network_header will be larger than skb headroom. nsh_gso_segment nhoff = skb->network_header - skb->mac_header; __skb_pull(skb,nsh_len) skb_mac_gso_segment mpls_gso_segment skb_reset_network_header(skb);//skb->network_header+=nsh_len return -EINVAL; skb_gso_error_unwind skb_push(skb, nsh_len); skb->mac_header = skb->network_header - nhoff; // skb->mac_header > skb->headroom, cause skb_push panic Use correct mac_offset to restore mac_header and get rid of nhoff. Fixes: c411ed854584 ("nsh: add GSO support") Reported-by: syzbot+632b5d9964208bfef8c0@syzkaller.appspotmail.com Suggested-by: Eric Dumazet Signed-off-by: Dong Chenchen Signed-off-by: David S. Miller --- net/nsh/nsh.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/nsh/nsh.c b/net/nsh/nsh.c index e9ca007718b7..0f23e5e8e03e 100644 --- a/net/nsh/nsh.c +++ b/net/nsh/nsh.c @@ -77,13 +77,12 @@ static struct sk_buff *nsh_gso_segment(struct sk_buff *skb, netdev_features_t features) { struct sk_buff *segs = ERR_PTR(-EINVAL); + u16 mac_offset = skb->mac_header; unsigned int nsh_len, mac_len; __be16 proto; - int nhoff; skb_reset_network_header(skb); - nhoff = skb->network_header - skb->mac_header; mac_len = skb->mac_len; if (unlikely(!pskb_may_pull(skb, NSH_BASE_HDR_LEN))) @@ -108,15 +107,14 @@ static struct sk_buff *nsh_gso_segment(struct sk_buff *skb, segs = skb_mac_gso_segment(skb, features); if (IS_ERR_OR_NULL(segs)) { skb_gso_error_unwind(skb, htons(ETH_P_NSH), nsh_len, - skb->network_header - nhoff, - mac_len); + mac_offset, mac_len); goto out; } for (skb = segs; skb; skb = skb->next) { skb->protocol = htons(ETH_P_NSH); __skb_push(skb, nsh_len); - skb_set_mac_header(skb, -nhoff); + skb->mac_header = mac_offset; skb->network_header = skb->mac_header + mac_len; skb->mac_len = mac_len; } -- cgit v1.2.3 From 3ae6d66b605be604644d4bb5708a7ffd9cf1abe8 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sun, 14 May 2023 15:52:27 -0400 Subject: tipc: add tipc_bearer_min_mtu to calculate min mtu As different media may requires different min mtu, and even the same media with different net family requires different min mtu, add tipc_bearer_min_mtu() to calculate min mtu accordingly. This API will be used to check the new mtu when doing the link mtu negotiation in the next patch. Signed-off-by: Xin Long Acked-by: Jon Maloy Signed-off-by: David S. Miller --- net/tipc/bearer.c | 13 +++++++++++++ net/tipc/bearer.h | 3 +++ net/tipc/udp_media.c | 5 +++-- 3 files changed, 19 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 35cac7733fd3..0e9a29e1536b 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -541,6 +541,19 @@ int tipc_bearer_mtu(struct net *net, u32 bearer_id) return mtu; } +int tipc_bearer_min_mtu(struct net *net, u32 bearer_id) +{ + int mtu = TIPC_MIN_BEARER_MTU; + struct tipc_bearer *b; + + rcu_read_lock(); + b = bearer_get(net, bearer_id); + if (b) + mtu += b->encap_hlen; + rcu_read_unlock(); + return mtu; +} + /* tipc_bearer_xmit_skb - sends buffer to destination over bearer */ void tipc_bearer_xmit_skb(struct net *net, u32 bearer_id, diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index 490ad6e5f7a3..bd0cc5c287ef 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -146,6 +146,7 @@ struct tipc_media { * @identity: array index of this bearer within TIPC bearer array * @disc: ptr to link setup request * @net_plane: network plane ('A' through 'H') currently associated with bearer + * @encap_hlen: encap headers length * @up: bearer up flag (bit 0) * @refcnt: tipc_bearer reference counter * @@ -170,6 +171,7 @@ struct tipc_bearer { u32 identity; struct tipc_discoverer *disc; char net_plane; + u16 encap_hlen; unsigned long up; refcount_t refcnt; }; @@ -232,6 +234,7 @@ int tipc_bearer_setup(void); void tipc_bearer_cleanup(void); void tipc_bearer_stop(struct net *net); int tipc_bearer_mtu(struct net *net, u32 bearer_id); +int tipc_bearer_min_mtu(struct net *net, u32 bearer_id); bool tipc_bearer_bcast_support(struct net *net, u32 bearer_id); void tipc_bearer_xmit_skb(struct net *net, u32 bearer_id, struct sk_buff *skb, diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c index c2bb818704c8..0a85244fd618 100644 --- a/net/tipc/udp_media.c +++ b/net/tipc/udp_media.c @@ -738,8 +738,8 @@ static int tipc_udp_enable(struct net *net, struct tipc_bearer *b, udp_conf.local_ip.s_addr = local.ipv4.s_addr; udp_conf.use_udp_checksums = false; ub->ifindex = dev->ifindex; - if (tipc_mtu_bad(dev, sizeof(struct iphdr) + - sizeof(struct udphdr))) { + b->encap_hlen = sizeof(struct iphdr) + sizeof(struct udphdr); + if (tipc_mtu_bad(dev, b->encap_hlen)) { err = -EINVAL; goto err; } @@ -760,6 +760,7 @@ static int tipc_udp_enable(struct net *net, struct tipc_bearer *b, else udp_conf.local_ip6 = local.ipv6; ub->ifindex = dev->ifindex; + b->encap_hlen = sizeof(struct ipv6hdr) + sizeof(struct udphdr); b->mtu = 1280; #endif } else { -- cgit v1.2.3 From 56077b56cd3fb78e1c8619e29581ba25a5c55e86 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sun, 14 May 2023 15:52:28 -0400 Subject: tipc: do not update mtu if msg_max is too small in mtu negotiation When doing link mtu negotiation, a malicious peer may send Activate msg with a very small mtu, e.g. 4 in Shuang's testing, without checking for the minimum mtu, l->mtu will be set to 4 in tipc_link_proto_rcv(), then n->links[bearer_id].mtu is set to 4294967228, which is a overflow of '4 - INT_H_SIZE - EMSG_OVERHEAD' in tipc_link_mss(). With tipc_link.mtu = 4, tipc_link_xmit() kept printing the warning: tipc: Too large msg, purging xmit list 1 5 0 40 4! tipc: Too large msg, purging xmit list 1 15 0 60 4! And with tipc_link_entry.mtu 4294967228, a huge skb was allocated in named_distribute(), and when purging it in tipc_link_xmit(), a crash was even caused: general protection fault, probably for non-canonical address 0x2100001011000dd: 0000 [#1] PREEMPT SMP PTI CPU: 0 PID: 0 Comm: swapper/0 Kdump: loaded Not tainted 6.3.0.neta #19 RIP: 0010:kfree_skb_list_reason+0x7e/0x1f0 Call Trace: skb_release_data+0xf9/0x1d0 kfree_skb_reason+0x40/0x100 tipc_link_xmit+0x57a/0x740 [tipc] tipc_node_xmit+0x16c/0x5c0 [tipc] tipc_named_node_up+0x27f/0x2c0 [tipc] tipc_node_write_unlock+0x149/0x170 [tipc] tipc_rcv+0x608/0x740 [tipc] tipc_udp_recv+0xdc/0x1f0 [tipc] udp_queue_rcv_one_skb+0x33e/0x620 udp_unicast_rcv_skb.isra.72+0x75/0x90 __udp4_lib_rcv+0x56d/0xc20 ip_protocol_deliver_rcu+0x100/0x2d0 This patch fixes it by checking the new mtu against tipc_bearer_min_mtu(), and not updating mtu if it is too small. Fixes: ed193ece2649 ("tipc: simplify link mtu negotiation") Reported-by: Shuang Li Signed-off-by: Xin Long Acked-by: Jon Maloy Signed-off-by: David S. Miller --- net/tipc/link.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/tipc/link.c b/net/tipc/link.c index b3ce24823f50..2eff1c7949cb 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -2200,7 +2200,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, struct tipc_msg *hdr = buf_msg(skb); struct tipc_gap_ack_blks *ga = NULL; bool reply = msg_probe(hdr), retransmitted = false; - u32 dlen = msg_data_sz(hdr), glen = 0; + u32 dlen = msg_data_sz(hdr), glen = 0, msg_max; u16 peers_snd_nxt = msg_next_sent(hdr); u16 peers_tol = msg_link_tolerance(hdr); u16 peers_prio = msg_linkprio(hdr); @@ -2239,6 +2239,9 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, switch (mtyp) { case RESET_MSG: case ACTIVATE_MSG: + msg_max = msg_max_pkt(hdr); + if (msg_max < tipc_bearer_min_mtu(l->net, l->bearer_id)) + break; /* Complete own link name with peer's interface name */ if_name = strrchr(l->name, ':') + 1; if (sizeof(l->name) - (if_name - l->name) <= TIPC_MAX_IF_NAME) @@ -2283,8 +2286,8 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, l->peer_session = msg_session(hdr); l->in_session = true; l->peer_bearer_id = msg_bearer_id(hdr); - if (l->mtu > msg_max_pkt(hdr)) - l->mtu = msg_max_pkt(hdr); + if (l->mtu > msg_max) + l->mtu = msg_max; break; case STATE_MSG: -- cgit v1.2.3 From 35a089b5d793d2bfd2cc7cfa6104545184de2ce7 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sun, 14 May 2023 15:52:29 -0400 Subject: tipc: check the bearer min mtu properly when setting it by netlink Checking the bearer min mtu with tipc_udp_mtu_bad() only works for IPv4 UDP bearer, and IPv6 UDP bearer has a different value for the min mtu. This patch checks with encap_hlen + TIPC_MIN_BEARER_MTU for min mtu, which works for both IPv4 and IPv6 UDP bearer. Note that tipc_udp_mtu_bad() is still used to check media min mtu in __tipc_nl_media_set(), as m->mtu currently is only used by the IPv4 UDP bearer as its default mtu value. Fixes: 682cd3cf946b ("tipc: confgiure and apply UDP bearer MTU on running links") Signed-off-by: Xin Long Acked-by: Jon Maloy Signed-off-by: David S. Miller --- net/tipc/bearer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 0e9a29e1536b..53881406e200 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -1151,8 +1151,8 @@ int __tipc_nl_bearer_set(struct sk_buff *skb, struct genl_info *info) return -EINVAL; } #ifdef CONFIG_TIPC_MEDIA_UDP - if (tipc_udp_mtu_bad(nla_get_u32 - (props[TIPC_NLA_PROP_MTU]))) { + if (nla_get_u32(props[TIPC_NLA_PROP_MTU]) < + b->encap_hlen + TIPC_MIN_BEARER_MTU) { NL_SET_ERR_MSG(info->extack, "MTU value is out-of-range"); return -EINVAL; -- cgit v1.2.3 From db2773d65b02aed319a93efdfb958087771d4e19 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Thu, 6 Apr 2023 13:08:45 +0200 Subject: can: isotp: recvmsg(): allow MSG_CMSG_COMPAT flag The control message provided by isotp support MSG_CMSG_COMPAT but blocked recvmsg() syscalls that have set this flag, i.e. on 32bit user space on 64 bit kernels. Link: https://github.com/hartkopp/can-isotp/issues/59 Cc: Oleksij Rempel Suggested-by: Marc Kleine-Budde Signed-off-by: Oliver Hartkopp Fixes: 42bf50a1795a ("can: isotp: support MSG_TRUNC flag when reading from socket") Link: https://lore.kernel.org/20230505110308.81087-2-mkl@pengutronix.de Cc: stable@vger.kernel.org Signed-off-by: Marc Kleine-Budde --- net/can/isotp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/can/isotp.c b/net/can/isotp.c index a750259cb79c..84f9aba02901 100644 --- a/net/can/isotp.c +++ b/net/can/isotp.c @@ -1139,7 +1139,7 @@ static int isotp_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, struct isotp_sock *so = isotp_sk(sk); int ret = 0; - if (flags & ~(MSG_DONTWAIT | MSG_TRUNC | MSG_PEEK)) + if (flags & ~(MSG_DONTWAIT | MSG_TRUNC | MSG_PEEK | MSG_CMSG_COMPAT)) return -EINVAL; if (!so->bound) -- cgit v1.2.3 From 1db080cbdbab28752bbb1c86d64daf96253a5da1 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Thu, 6 Apr 2023 13:08:45 +0200 Subject: can: j1939: recvmsg(): allow MSG_CMSG_COMPAT flag The control message provided by J1939 support MSG_CMSG_COMPAT but blocked recvmsg() syscalls that have set this flag, i.e. on 32bit user space on 64 bit kernels. Link: https://github.com/hartkopp/can-isotp/issues/59 Cc: Oleksij Rempel Suggested-by: Marc Kleine-Budde Signed-off-by: Oliver Hartkopp Tested-by: Oleksij Rempel Acked-by: Oleksij Rempel Fixes: 9d71dd0c7009 ("can: add support of SAE J1939 protocol") Link: https://lore.kernel.org/20230505110308.81087-3-mkl@pengutronix.de Cc: stable@vger.kernel.org Signed-off-by: Marc Kleine-Budde --- net/can/j1939/socket.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/can/j1939/socket.c b/net/can/j1939/socket.c index 7e90f9e61d9b..1790469b2580 100644 --- a/net/can/j1939/socket.c +++ b/net/can/j1939/socket.c @@ -798,7 +798,7 @@ static int j1939_sk_recvmsg(struct socket *sock, struct msghdr *msg, struct j1939_sk_buff_cb *skcb; int ret = 0; - if (flags & ~(MSG_DONTWAIT | MSG_ERRQUEUE)) + if (flags & ~(MSG_DONTWAIT | MSG_ERRQUEUE | MSG_CMSG_COMPAT)) return -EINVAL; if (flags & MSG_ERRQUEUE) -- cgit v1.2.3 From 1b6b4ed01493b7ea2205ab83c49198f7d13ca9d2 Mon Sep 17 00:00:00 2001 From: Ilan Peer Date: Mon, 24 Apr 2023 10:32:24 +0300 Subject: wifi: cfg80211: Drop entries with invalid BSSIDs in RNR Ignore AP information for entries that include an invalid BSSID in the TBTT information field, e.g., all zeros BSSIDs. Fixes: c8cb5b854b40 ("nl80211/cfg80211: support 6 GHz scanning") Signed-off-by: Ilan Peer Signed-off-by: Gregory Greenman Link: https://lore.kernel.org/r/20230424103224.5e65d04d1448.Ic10c8577ae4a85272c407106c9d0a2ecb5372743@changeid Signed-off-by: Johannes Berg --- net/wireless/scan.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/wireless/scan.c b/net/wireless/scan.c index a1382255fab3..c501db7bbdb3 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -5,7 +5,7 @@ * Copyright 2008 Johannes Berg * Copyright 2013-2014 Intel Mobile Communications GmbH * Copyright 2016 Intel Deutschland GmbH - * Copyright (C) 2018-2022 Intel Corporation + * Copyright (C) 2018-2023 Intel Corporation */ #include #include @@ -540,6 +540,10 @@ static int cfg80211_parse_ap_info(struct cfg80211_colocated_ap *entry, /* skip the TBTT offset */ pos++; + /* ignore entries with invalid BSSID */ + if (!is_valid_ether_addr(pos)) + return -EINVAL; + memcpy(entry->bssid, pos, ETH_ALEN); pos += ETH_ALEN; -- cgit v1.2.3 From ef6e1997da63ad0ac3fe33153fec9524c9ae56c9 Mon Sep 17 00:00:00 2001 From: Mirsad Goran Todorovac Date: Tue, 25 Apr 2023 18:40:08 +0200 Subject: wifi: mac80211: fortify the spinlock against deadlock by interrupt In the function ieee80211_tx_dequeue() there is a particular locking sequence: begin: spin_lock(&local->queue_stop_reason_lock); q_stopped = local->queue_stop_reasons[q]; spin_unlock(&local->queue_stop_reason_lock); However small the chance (increased by ftracetest), an asynchronous interrupt can occur in between of spin_lock() and spin_unlock(), and the interrupt routine will attempt to lock the same &local->queue_stop_reason_lock again. This will cause a costly reset of the CPU and the wifi device or an altogether hang in the single CPU and single core scenario. The only remaining spin_lock(&local->queue_stop_reason_lock) that did not disable interrupts was patched, which should prevent any deadlocks on the same CPU/core and the same wifi device. This is the probable trace of the deadlock: kernel: ================================ kernel: WARNING: inconsistent lock state kernel: 6.3.0-rc6-mt-20230401-00001-gf86822a1170f #4 Tainted: G W kernel: -------------------------------- kernel: inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage. kernel: kworker/5:0/25656 [HC0[0]:SC0[0]:HE1:SE1] takes: kernel: ffff9d6190779478 (&local->queue_stop_reason_lock){+.?.}-{2:2}, at: return_to_handler+0x0/0x40 kernel: {IN-SOFTIRQ-W} state was registered at: kernel: lock_acquire+0xc7/0x2d0 kernel: _raw_spin_lock+0x36/0x50 kernel: ieee80211_tx_dequeue+0xb4/0x1330 [mac80211] kernel: iwl_mvm_mac_itxq_xmit+0xae/0x210 [iwlmvm] kernel: iwl_mvm_mac_wake_tx_queue+0x2d/0xd0 [iwlmvm] kernel: ieee80211_queue_skb+0x450/0x730 [mac80211] kernel: __ieee80211_xmit_fast.constprop.66+0x834/0xa50 [mac80211] kernel: __ieee80211_subif_start_xmit+0x217/0x530 [mac80211] kernel: ieee80211_subif_start_xmit+0x60/0x580 [mac80211] kernel: dev_hard_start_xmit+0xb5/0x260 kernel: __dev_queue_xmit+0xdbe/0x1200 kernel: neigh_resolve_output+0x166/0x260 kernel: ip_finish_output2+0x216/0xb80 kernel: __ip_finish_output+0x2a4/0x4d0 kernel: ip_finish_output+0x2d/0xd0 kernel: ip_output+0x82/0x2b0 kernel: ip_local_out+0xec/0x110 kernel: igmpv3_sendpack+0x5c/0x90 kernel: igmp_ifc_timer_expire+0x26e/0x4e0 kernel: call_timer_fn+0xa5/0x230 kernel: run_timer_softirq+0x27f/0x550 kernel: __do_softirq+0xb4/0x3a4 kernel: irq_exit_rcu+0x9b/0xc0 kernel: sysvec_apic_timer_interrupt+0x80/0xa0 kernel: asm_sysvec_apic_timer_interrupt+0x1f/0x30 kernel: _raw_spin_unlock_irqrestore+0x3f/0x70 kernel: free_to_partial_list+0x3d6/0x590 kernel: __slab_free+0x1b7/0x310 kernel: kmem_cache_free+0x52d/0x550 kernel: putname+0x5d/0x70 kernel: do_sys_openat2+0x1d7/0x310 kernel: do_sys_open+0x51/0x80 kernel: __x64_sys_openat+0x24/0x30 kernel: do_syscall_64+0x5c/0x90 kernel: entry_SYSCALL_64_after_hwframe+0x72/0xdc kernel: irq event stamp: 5120729 kernel: hardirqs last enabled at (5120729): [] trace_graph_return+0xd6/0x120 kernel: hardirqs last disabled at (5120728): [] trace_graph_return+0xf0/0x120 kernel: softirqs last enabled at (5069900): [] return_to_handler+0x0/0x40 kernel: softirqs last disabled at (5067555): [] return_to_handler+0x0/0x40 kernel: other info that might help us debug this: kernel: Possible unsafe locking scenario: kernel: CPU0 kernel: ---- kernel: lock(&local->queue_stop_reason_lock); kernel: kernel: lock(&local->queue_stop_reason_lock); kernel: *** DEADLOCK *** kernel: 8 locks held by kworker/5:0/25656: kernel: #0: ffff9d618009d138 ((wq_completion)events_freezable){+.+.}-{0:0}, at: process_one_work+0x1ca/0x530 kernel: #1: ffffb1ef4637fe68 ((work_completion)(&local->restart_work)){+.+.}-{0:0}, at: process_one_work+0x1ce/0x530 kernel: #2: ffffffff9f166548 (rtnl_mutex){+.+.}-{3:3}, at: return_to_handler+0x0/0x40 kernel: #3: ffff9d6190778728 (&rdev->wiphy.mtx){+.+.}-{3:3}, at: return_to_handler+0x0/0x40 kernel: #4: ffff9d619077b480 (&mvm->mutex){+.+.}-{3:3}, at: return_to_handler+0x0/0x40 kernel: #5: ffff9d61907bacd8 (&trans_pcie->mutex){+.+.}-{3:3}, at: return_to_handler+0x0/0x40 kernel: #6: ffffffff9ef9cda0 (rcu_read_lock){....}-{1:2}, at: iwl_mvm_queue_state_change+0x59/0x3a0 [iwlmvm] kernel: #7: ffffffff9ef9cda0 (rcu_read_lock){....}-{1:2}, at: iwl_mvm_mac_itxq_xmit+0x42/0x210 [iwlmvm] kernel: stack backtrace: kernel: CPU: 5 PID: 25656 Comm: kworker/5:0 Tainted: G W 6.3.0-rc6-mt-20230401-00001-gf86822a1170f #4 kernel: Hardware name: LENOVO 82H8/LNVNB161216, BIOS GGCN51WW 11/16/2022 kernel: Workqueue: events_freezable ieee80211_restart_work [mac80211] kernel: Call Trace: kernel: kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: dump_stack_lvl+0x5f/0xa0 kernel: dump_stack+0x14/0x20 kernel: print_usage_bug.part.46+0x208/0x2a0 kernel: mark_lock.part.47+0x605/0x630 kernel: ? sched_clock+0xd/0x20 kernel: ? trace_clock_local+0x14/0x30 kernel: ? __rb_reserve_next+0x5f/0x490 kernel: ? _raw_spin_lock+0x1b/0x50 kernel: __lock_acquire+0x464/0x1990 kernel: ? mark_held_locks+0x4e/0x80 kernel: lock_acquire+0xc7/0x2d0 kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: ? ftrace_return_to_handler+0x8b/0x100 kernel: ? preempt_count_add+0x4/0x70 kernel: _raw_spin_lock+0x36/0x50 kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: ieee80211_tx_dequeue+0xb4/0x1330 [mac80211] kernel: ? prepare_ftrace_return+0xc5/0x190 kernel: ? ftrace_graph_func+0x16/0x20 kernel: ? 0xffffffffc02ab0b1 kernel: ? lock_acquire+0xc7/0x2d0 kernel: ? iwl_mvm_mac_itxq_xmit+0x42/0x210 [iwlmvm] kernel: ? ieee80211_tx_dequeue+0x9/0x1330 [mac80211] kernel: ? __rcu_read_lock+0x4/0x40 kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: iwl_mvm_mac_itxq_xmit+0xae/0x210 [iwlmvm] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: iwl_mvm_queue_state_change+0x311/0x3a0 [iwlmvm] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: iwl_mvm_wake_sw_queue+0x17/0x20 [iwlmvm] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: iwl_txq_gen2_unmap+0x1c9/0x1f0 [iwlwifi] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: iwl_txq_gen2_free+0x55/0x130 [iwlwifi] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: iwl_txq_gen2_tx_free+0x63/0x80 [iwlwifi] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: _iwl_trans_pcie_gen2_stop_device+0x3f3/0x5b0 [iwlwifi] kernel: ? _iwl_trans_pcie_gen2_stop_device+0x9/0x5b0 [iwlwifi] kernel: ? mutex_lock_nested+0x4/0x30 kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: iwl_trans_pcie_gen2_stop_device+0x5f/0x90 [iwlwifi] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: iwl_mvm_stop_device+0x78/0xd0 [iwlmvm] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: __iwl_mvm_mac_start+0x114/0x210 [iwlmvm] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: iwl_mvm_mac_start+0x76/0x150 [iwlmvm] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: drv_start+0x79/0x180 [mac80211] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: ieee80211_reconfig+0x1523/0x1ce0 [mac80211] kernel: ? synchronize_net+0x4/0x50 kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: ieee80211_restart_work+0x108/0x170 [mac80211] kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: process_one_work+0x250/0x530 kernel: ? ftrace_regs_caller_end+0x66/0x66 kernel: worker_thread+0x48/0x3a0 kernel: ? __pfx_worker_thread+0x10/0x10 kernel: kthread+0x10f/0x140 kernel: ? __pfx_kthread+0x10/0x10 kernel: ret_from_fork+0x29/0x50 kernel: Fixes: 4444bc2116ae ("wifi: mac80211: Proper mark iTXQs for resumption") Link: https://lore.kernel.org/all/1f58a0d1-d2b9-d851-73c3-93fcc607501c@alu.unizg.hr/ Reported-by: Mirsad Goran Todorovac Cc: Gregory Greenman Cc: Johannes Berg Link: https://lore.kernel.org/all/cdc80531-f25f-6f9d-b15f-25e16130b53a@alu.unizg.hr/ Cc: David S. Miller Cc: Eric Dumazet Cc: Jakub Kicinski Cc: Paolo Abeni Cc: Leon Romanovsky Cc: Alexander Wetzel Signed-off-by: Mirsad Goran Todorovac Reviewed-by: Leon Romanovsky Reviewed-by: tag, or it goes automatically? Link: https://lore.kernel.org/r/20230425164005.25272-1-mirsad.todorovac@alu.unizg.hr Signed-off-by: Johannes Berg --- net/mac80211/tx.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 1a3327407552..0d9fbc8458fd 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -3791,6 +3791,7 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, ieee80211_tx_result r; struct ieee80211_vif *vif = txq->vif; int q = vif->hw_queue[txq->ac]; + unsigned long flags; bool q_stopped; WARN_ON_ONCE(softirq_count() == 0); @@ -3799,9 +3800,9 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, return NULL; begin: - spin_lock(&local->queue_stop_reason_lock); + spin_lock_irqsave(&local->queue_stop_reason_lock, flags); q_stopped = local->queue_stop_reasons[q]; - spin_unlock(&local->queue_stop_reason_lock); + spin_unlock_irqrestore(&local->queue_stop_reason_lock, flags); if (unlikely(q_stopped)) { /* mark for waking later */ -- cgit v1.2.3 From 13ad2b1eeacd48ec0f31f55964e6dc7dfc2c0299 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Mon, 24 Apr 2023 19:42:04 +0200 Subject: wifi: mac80211: Fix puncturing bitmap handling in __ieee80211_csa_finalize() 'changed' can be OR'ed with BSS_CHANGED_EHT_PUNCTURING which is larger than an u32. So, turn 'changed' into an u64 and update ieee80211_set_after_csa_beacon() accordingly. In the commit in Fixes, only ieee80211_start_ap() was updated. Fixes: 2cc25e4b2a04 ("wifi: mac80211: configure puncturing bitmap") Signed-off-by: Christophe JAILLET Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/e84a3f80fe536787f7a2c7180507efc36cd14f95.1682358088.git.christophe.jaillet@wanadoo.fr Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 7317e4a5d1ff..c5e5f783f137 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -3589,7 +3589,7 @@ void ieee80211_channel_switch_disconnect(struct ieee80211_vif *vif, bool block_t EXPORT_SYMBOL(ieee80211_channel_switch_disconnect); static int ieee80211_set_after_csa_beacon(struct ieee80211_sub_if_data *sdata, - u32 *changed) + u64 *changed) { int err; @@ -3632,7 +3632,7 @@ static int ieee80211_set_after_csa_beacon(struct ieee80211_sub_if_data *sdata, static int __ieee80211_csa_finalize(struct ieee80211_sub_if_data *sdata) { struct ieee80211_local *local = sdata->local; - u32 changed = 0; + u64 changed = 0; int err; sdata_assert_lock(sdata); -- cgit v1.2.3 From 248e4776514bf70236e6b1a54c65aa5324c8b1eb Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 4 May 2023 16:45:01 +0300 Subject: wifi: mac80211: fix min center freq offset tracing We need to set the correct trace variable, otherwise we're overwriting something else instead and the right one that we print later is not initialized. Fixes: b6011960f392 ("mac80211: handle channel frequency offset") Signed-off-by: Johannes Berg Signed-off-by: Gregory Greenman Link: https://lore.kernel.org/r/20230504134511.828474-2-gregory.greenman@intel.com Signed-off-by: Johannes Berg --- net/mac80211/trace.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h index de5d69f21306..db0d0132c58c 100644 --- a/net/mac80211/trace.h +++ b/net/mac80211/trace.h @@ -67,7 +67,7 @@ __entry->min_freq_offset = (c)->chan ? (c)->chan->freq_offset : 0; \ __entry->min_chan_width = (c)->width; \ __entry->min_center_freq1 = (c)->center_freq1; \ - __entry->freq1_offset = (c)->freq1_offset; \ + __entry->min_freq1_offset = (c)->freq1_offset; \ __entry->min_center_freq2 = (c)->center_freq2; #define MIN_CHANDEF_PR_FMT " min_control:%d.%03d MHz min_width:%d min_center: %d.%03d/%d MHz" #define MIN_CHANDEF_PR_ARG __entry->min_control_freq, __entry->min_freq_offset, \ -- cgit v1.2.3 From a23d7f5b2fbda114de60c4b53311e052281d7533 Mon Sep 17 00:00:00 2001 From: Michael Lee Date: Thu, 4 May 2023 16:04:41 +0800 Subject: wifi: mac80211: Abort running color change when stopping the AP When stopping the AP, there might be a color change in progress. It should be deactivated here, or the driver might later finalize a color change on a stopped AP. Fixes: 5f9404abdf2a (mac80211: add support for BSS color change) Signed-off-by: Michael Lee Link: https://lore.kernel.org/r/20230504080441.22958-1-michael-cy.lee@mediatek.com Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index c5e5f783f137..86b2036d73ff 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1578,9 +1578,10 @@ static int ieee80211_stop_ap(struct wiphy *wiphy, struct net_device *dev, sdata_dereference(link->u.ap.unsol_bcast_probe_resp, sdata); - /* abort any running channel switch */ + /* abort any running channel switch or color change */ mutex_lock(&local->mtx); link_conf->csa_active = false; + link_conf->color_change_active = false; if (link->csa_block_tx) { ieee80211_wake_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); -- cgit v1.2.3 From 860e1b43da94551cd1e73adc36b3c64cc3e5dc01 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 4 May 2023 16:45:02 +0300 Subject: wifi: mac80211: simplify chanctx allocation There's no need to call ieee80211_recalc_chanctx_min_def() since it cannot and won't call the driver anyway; just use _ieee80211_recalc_chanctx_min_def() instead. Signed-off-by: Johannes Berg Signed-off-by: Gregory Greenman Link: https://lore.kernel.org/r/20230504134511.828474-3-gregory.greenman@intel.com Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index dbc34fbe7c8f..d23d1a7b4cc3 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -638,7 +638,7 @@ ieee80211_alloc_chanctx(struct ieee80211_local *local, ctx->conf.rx_chains_dynamic = 1; ctx->mode = mode; ctx->conf.radar_enabled = false; - ieee80211_recalc_chanctx_min_def(local, ctx); + _ieee80211_recalc_chanctx_min_def(local, ctx); return ctx; } -- cgit v1.2.3 From b72a455a2409fd94d6d9b4eb51d659a88213243b Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 4 May 2023 16:45:03 +0300 Subject: wifi: mac80211: consider reserved chanctx for mindef When a chanctx is reserved for a new vif and we recalculate the minimal definition for it, we need to consider the new interface it's being reserved for before we assign it, so it can be used directly with the correct min channel width. Fix the code to - optionally - consider that, and use that option just before doing the reassignment. Also, when considering channel context reservations, we should only consider the one link we're currently working with. Change the boolean argument to a link pointer to do that. Signed-off-by: Johannes Berg Signed-off-by: Gregory Greenman Link: https://lore.kernel.org/r/20230504134511.828474-4-gregory.greenman@intel.com Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 72 ++++++++++++++++++++++++++++------------------ net/mac80211/ieee80211_i.h | 3 +- net/mac80211/util.c | 2 +- 3 files changed, 47 insertions(+), 30 deletions(-) (limited to 'net') diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index d23d1a7b4cc3..1b182cf9d661 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -258,7 +258,8 @@ ieee80211_get_max_required_bw(struct ieee80211_sub_if_data *sdata, static enum nl80211_chan_width ieee80211_get_chanctx_vif_max_required_bw(struct ieee80211_sub_if_data *sdata, - struct ieee80211_chanctx_conf *conf) + struct ieee80211_chanctx *ctx, + struct ieee80211_link_data *rsvd_for) { enum nl80211_chan_width max_bw = NL80211_CHAN_WIDTH_20_NOHT; struct ieee80211_vif *vif = &sdata->vif; @@ -267,13 +268,14 @@ ieee80211_get_chanctx_vif_max_required_bw(struct ieee80211_sub_if_data *sdata, rcu_read_lock(); for (link_id = 0; link_id < ARRAY_SIZE(sdata->link); link_id++) { enum nl80211_chan_width width = NL80211_CHAN_WIDTH_20_NOHT; - struct ieee80211_bss_conf *link_conf = - rcu_dereference(sdata->vif.link_conf[link_id]); + struct ieee80211_link_data *link = + rcu_dereference(sdata->link[link_id]); - if (!link_conf) + if (!link) continue; - if (rcu_access_pointer(link_conf->chanctx_conf) != conf) + if (link != rsvd_for && + rcu_access_pointer(link->conf->chanctx_conf) != &ctx->conf) continue; switch (vif->type) { @@ -287,7 +289,7 @@ ieee80211_get_chanctx_vif_max_required_bw(struct ieee80211_sub_if_data *sdata, * point, so take the width from the chandef, but * account also for TDLS peers */ - width = max(link_conf->chandef.width, + width = max(link->conf->chandef.width, ieee80211_get_max_required_bw(sdata, link_id)); break; case NL80211_IFTYPE_P2P_DEVICE: @@ -296,7 +298,7 @@ ieee80211_get_chanctx_vif_max_required_bw(struct ieee80211_sub_if_data *sdata, case NL80211_IFTYPE_ADHOC: case NL80211_IFTYPE_MESH_POINT: case NL80211_IFTYPE_OCB: - width = link_conf->chandef.width; + width = link->conf->chandef.width; break; case NL80211_IFTYPE_WDS: case NL80211_IFTYPE_UNSPECIFIED: @@ -316,7 +318,8 @@ ieee80211_get_chanctx_vif_max_required_bw(struct ieee80211_sub_if_data *sdata, static enum nl80211_chan_width ieee80211_get_chanctx_max_required_bw(struct ieee80211_local *local, - struct ieee80211_chanctx_conf *conf) + struct ieee80211_chanctx *ctx, + struct ieee80211_link_data *rsvd_for) { struct ieee80211_sub_if_data *sdata; enum nl80211_chan_width max_bw = NL80211_CHAN_WIDTH_20_NOHT; @@ -328,7 +331,8 @@ ieee80211_get_chanctx_max_required_bw(struct ieee80211_local *local, if (!ieee80211_sdata_running(sdata)) continue; - width = ieee80211_get_chanctx_vif_max_required_bw(sdata, conf); + width = ieee80211_get_chanctx_vif_max_required_bw(sdata, ctx, + rsvd_for); max_bw = max(max_bw, width); } @@ -336,8 +340,8 @@ ieee80211_get_chanctx_max_required_bw(struct ieee80211_local *local, /* use the configured bandwidth in case of monitor interface */ sdata = rcu_dereference(local->monitor_sdata); if (sdata && - rcu_access_pointer(sdata->vif.bss_conf.chanctx_conf) == conf) - max_bw = max(max_bw, conf->def.width); + rcu_access_pointer(sdata->vif.bss_conf.chanctx_conf) == &ctx->conf) + max_bw = max(max_bw, ctx->conf.def.width); rcu_read_unlock(); @@ -349,8 +353,10 @@ ieee80211_get_chanctx_max_required_bw(struct ieee80211_local *local, * the max of min required widths of all the interfaces bound to this * channel context. */ -static u32 _ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, - struct ieee80211_chanctx *ctx) +static u32 +_ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, + struct ieee80211_chanctx *ctx, + struct ieee80211_link_data *rsvd_for) { enum nl80211_chan_width max_bw; struct cfg80211_chan_def min_def; @@ -370,7 +376,7 @@ static u32 _ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, return 0; } - max_bw = ieee80211_get_chanctx_max_required_bw(local, &ctx->conf); + max_bw = ieee80211_get_chanctx_max_required_bw(local, ctx, rsvd_for); /* downgrade chandef up to max_bw */ min_def = ctx->conf.def; @@ -448,9 +454,10 @@ static void ieee80211_chan_bw_change(struct ieee80211_local *local, * channel context. */ void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, - struct ieee80211_chanctx *ctx) + struct ieee80211_chanctx *ctx, + struct ieee80211_link_data *rsvd_for) { - u32 changed = _ieee80211_recalc_chanctx_min_def(local, ctx); + u32 changed = _ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for); if (!changed) return; @@ -464,10 +471,11 @@ void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, ieee80211_chan_bw_change(local, ctx, false); } -static void ieee80211_change_chanctx(struct ieee80211_local *local, - struct ieee80211_chanctx *ctx, - struct ieee80211_chanctx *old_ctx, - const struct cfg80211_chan_def *chandef) +static void _ieee80211_change_chanctx(struct ieee80211_local *local, + struct ieee80211_chanctx *ctx, + struct ieee80211_chanctx *old_ctx, + const struct cfg80211_chan_def *chandef, + struct ieee80211_link_data *rsvd_for) { u32 changed; @@ -492,7 +500,7 @@ static void ieee80211_change_chanctx(struct ieee80211_local *local, ieee80211_chan_bw_change(local, old_ctx, true); if (cfg80211_chandef_identical(&ctx->conf.def, chandef)) { - ieee80211_recalc_chanctx_min_def(local, ctx); + ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for); return; } @@ -502,7 +510,7 @@ static void ieee80211_change_chanctx(struct ieee80211_local *local, /* check if min chanctx also changed */ changed = IEEE80211_CHANCTX_CHANGE_WIDTH | - _ieee80211_recalc_chanctx_min_def(local, ctx); + _ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for); drv_change_chanctx(local, ctx, changed); if (!local->use_chanctx) { @@ -514,6 +522,14 @@ static void ieee80211_change_chanctx(struct ieee80211_local *local, ieee80211_chan_bw_change(local, old_ctx, false); } +static void ieee80211_change_chanctx(struct ieee80211_local *local, + struct ieee80211_chanctx *ctx, + struct ieee80211_chanctx *old_ctx, + const struct cfg80211_chan_def *chandef) +{ + _ieee80211_change_chanctx(local, ctx, old_ctx, chandef, NULL); +} + static struct ieee80211_chanctx * ieee80211_find_chanctx(struct ieee80211_local *local, const struct cfg80211_chan_def *chandef, @@ -638,7 +654,7 @@ ieee80211_alloc_chanctx(struct ieee80211_local *local, ctx->conf.rx_chains_dynamic = 1; ctx->mode = mode; ctx->conf.radar_enabled = false; - _ieee80211_recalc_chanctx_min_def(local, ctx); + _ieee80211_recalc_chanctx_min_def(local, ctx, NULL); return ctx; } @@ -873,12 +889,12 @@ out: ieee80211_recalc_chanctx_chantype(local, curr_ctx); ieee80211_recalc_smps_chanctx(local, curr_ctx); ieee80211_recalc_radar_chanctx(local, curr_ctx); - ieee80211_recalc_chanctx_min_def(local, curr_ctx); + ieee80211_recalc_chanctx_min_def(local, curr_ctx, NULL); } if (new_ctx && ieee80211_chanctx_num_assigned(local, new_ctx) > 0) { ieee80211_recalc_txpower(sdata, false); - ieee80211_recalc_chanctx_min_def(local, new_ctx); + ieee80211_recalc_chanctx_min_def(local, new_ctx, NULL); } if (sdata->vif.type != NL80211_IFTYPE_P2P_DEVICE && @@ -1270,7 +1286,7 @@ ieee80211_link_use_reserved_reassign(struct ieee80211_link_data *link) ieee80211_link_update_chandef(link, &link->reserved_chandef); - ieee80211_change_chanctx(local, new_ctx, old_ctx, chandef); + _ieee80211_change_chanctx(local, new_ctx, old_ctx, chandef, link); vif_chsw[0].vif = &sdata->vif; vif_chsw[0].old_ctx = &old_ctx->conf; @@ -1300,7 +1316,7 @@ ieee80211_link_use_reserved_reassign(struct ieee80211_link_data *link) if (ieee80211_chanctx_refcount(local, old_ctx) == 0) ieee80211_free_chanctx(local, old_ctx); - ieee80211_recalc_chanctx_min_def(local, new_ctx); + ieee80211_recalc_chanctx_min_def(local, new_ctx, NULL); ieee80211_recalc_smps_chanctx(local, new_ctx); ieee80211_recalc_radar_chanctx(local, new_ctx); @@ -1665,7 +1681,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local) ieee80211_recalc_chanctx_chantype(local, ctx); ieee80211_recalc_smps_chanctx(local, ctx); ieee80211_recalc_radar_chanctx(local, ctx); - ieee80211_recalc_chanctx_min_def(local, ctx); + ieee80211_recalc_chanctx_min_def(local, ctx, NULL); list_for_each_entry_safe(link, link_tmp, &ctx->reserved_links, reserved_chanctx_list) { diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index a0a7839cb961..b0372e76f373 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2537,7 +2537,8 @@ int ieee80211_chanctx_refcount(struct ieee80211_local *local, void ieee80211_recalc_smps_chanctx(struct ieee80211_local *local, struct ieee80211_chanctx *chanctx); void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local, - struct ieee80211_chanctx *ctx); + struct ieee80211_chanctx *ctx, + struct ieee80211_link_data *rsvd_for); bool ieee80211_is_radar_required(struct ieee80211_local *local); void ieee80211_dfs_cac_timer(unsigned long data); diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 1527d6aafc14..4bf76150925d 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -3015,7 +3015,7 @@ void ieee80211_recalc_min_chandef(struct ieee80211_sub_if_data *sdata, chanctx = container_of(chanctx_conf, struct ieee80211_chanctx, conf); - ieee80211_recalc_chanctx_min_def(local, chanctx); + ieee80211_recalc_chanctx_min_def(local, chanctx, NULL); } unlock: mutex_unlock(&local->chanctx_mtx); -- cgit v1.2.3 From 04312de4ced4b152749614e8179f3978a20a992f Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 4 May 2023 16:45:04 +0300 Subject: wifi: mac80211: recalc chanctx mindef before assigning When we allocate a new channel context, or find an existing one that is compatible, we currently assign it to a link before its mindef is updated. This leads to strange situations, especially in link switching where you switch to an 80 MHz link and expect it to be active immediately, but the mindef is still configured to 20 MHz while assigning. Also, it's strange that the chandef passed to the assign method's argument is wider than the one in the context. Fix this by calculating the mindef with the new link considered before calling the driver. In particular, this fixes an iwlwifi problem during link switch where the firmware would assert because the (link) station that was added for the AP is configured to transmit at a bandwidth that's wider than the channel context that it's configured on. Signed-off-by: Johannes Berg Signed-off-by: Gregory Greenman Link: https://lore.kernel.org/r/20230504134511.828474-5-gregory.greenman@intel.com Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index 1b182cf9d661..77c90ed8f5d7 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -871,6 +871,9 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, } if (new_ctx) { + /* recalc considering the link we'll use it for now */ + ieee80211_recalc_chanctx_min_def(local, new_ctx, link); + ret = drv_assign_vif_chanctx(local, sdata, link->conf, new_ctx); if (ret) goto out; -- cgit v1.2.3 From d6352dae0903fe8beae4c007dc320e9e9f1fed45 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 15 May 2023 19:29:25 +0300 Subject: devlink: Fix crash with CONFIG_NET_NS=n '__net_initdata' becomes a no-op with CONFIG_NET_NS=y, but when this option is disabled it becomes '__initdata', which means the data can be freed after the initialization phase. This annotation is obviously incorrect for the devlink net device notifier block which is still registered after the initialization phase [1]. Fix this crash by removing the '__net_initdata' annotation. [1] general protection fault, probably for non-canonical address 0xcccccccccccccccc: 0000 [#1] PREEMPT SMP CPU: 3 PID: 117 Comm: (udev-worker) Not tainted 6.4.0-rc1-custom-gdf0acdc59b09 #64 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-1.fc37 04/01/2014 RIP: 0010:notifier_call_chain+0x58/0xc0 [...] Call Trace: dev_set_mac_address+0x85/0x120 dev_set_mac_address_user+0x30/0x50 do_setlink+0x219/0x1270 rtnl_setlink+0xf7/0x1a0 rtnetlink_rcv_msg+0x142/0x390 netlink_rcv_skb+0x58/0x100 netlink_unicast+0x188/0x270 netlink_sendmsg+0x214/0x470 __sys_sendto+0x12f/0x1a0 __x64_sys_sendto+0x24/0x30 do_syscall_64+0x38/0x80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Fixes: e93c9378e33f ("devlink: change per-devlink netdev notifier to static one") Reported-by: Marek Szyprowski Closes: https://lore.kernel.org/netdev/600ddf9e-589a-2aa0-7b69-a438f833ca10@samsung.com/ Tested-by: Marek Szyprowski Signed-off-by: Ido Schimmel Reviewed-by: Jiri Pirko Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230515162925.1144416-1-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- net/devlink/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/devlink/core.c b/net/devlink/core.c index 0e58eee44bdb..c23ebabadc52 100644 --- a/net/devlink/core.c +++ b/net/devlink/core.c @@ -294,7 +294,7 @@ static struct pernet_operations devlink_pernet_ops __net_initdata = { .pre_exit = devlink_pernet_pre_exit, }; -static struct notifier_block devlink_port_netdevice_nb __net_initdata = { +static struct notifier_block devlink_port_netdevice_nb = { .notifier_call = devlink_port_netdevice_event, }; -- cgit v1.2.3 From dacab578c7c6cd06c50c89dfa36b0e0f10decd4e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 16 May 2023 14:23:42 +0000 Subject: vlan: fix a potential uninit-value in vlan_dev_hard_start_xmit() syzbot triggered the following splat [1], sending an empty message through pppoe_sendmsg(). When VLAN_FLAG_REORDER_HDR flag is set, vlan_dev_hard_header() does not push extra bytes for the VLAN header, because vlan is offloaded. Unfortunately vlan_dev_hard_start_xmit() first reads veth->h_vlan_proto before testing (vlan->flags & VLAN_FLAG_REORDER_HDR). We need to swap the two conditions. [1] BUG: KMSAN: uninit-value in vlan_dev_hard_start_xmit+0x171/0x7f0 net/8021q/vlan_dev.c:111 vlan_dev_hard_start_xmit+0x171/0x7f0 net/8021q/vlan_dev.c:111 __netdev_start_xmit include/linux/netdevice.h:4883 [inline] netdev_start_xmit include/linux/netdevice.h:4897 [inline] xmit_one net/core/dev.c:3580 [inline] dev_hard_start_xmit+0x253/0xa20 net/core/dev.c:3596 __dev_queue_xmit+0x3c7f/0x5ac0 net/core/dev.c:4246 dev_queue_xmit include/linux/netdevice.h:3053 [inline] pppoe_sendmsg+0xa93/0xb80 drivers/net/ppp/pppoe.c:900 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg net/socket.c:747 [inline] ____sys_sendmsg+0xa24/0xe40 net/socket.c:2501 ___sys_sendmsg+0x2a1/0x3f0 net/socket.c:2555 __sys_sendmmsg+0x411/0xa50 net/socket.c:2641 __do_sys_sendmmsg net/socket.c:2670 [inline] __se_sys_sendmmsg net/socket.c:2667 [inline] __x64_sys_sendmmsg+0xbc/0x120 net/socket.c:2667 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Uninit was created at: slab_post_alloc_hook+0x12d/0xb60 mm/slab.h:774 slab_alloc_node mm/slub.c:3452 [inline] kmem_cache_alloc_node+0x543/0xab0 mm/slub.c:3497 kmalloc_reserve+0x148/0x470 net/core/skbuff.c:520 __alloc_skb+0x3a7/0x850 net/core/skbuff.c:606 alloc_skb include/linux/skbuff.h:1277 [inline] sock_wmalloc+0xfe/0x1a0 net/core/sock.c:2583 pppoe_sendmsg+0x3af/0xb80 drivers/net/ppp/pppoe.c:867 sock_sendmsg_nosec net/socket.c:724 [inline] sock_sendmsg net/socket.c:747 [inline] ____sys_sendmsg+0xa24/0xe40 net/socket.c:2501 ___sys_sendmsg+0x2a1/0x3f0 net/socket.c:2555 __sys_sendmmsg+0x411/0xa50 net/socket.c:2641 __do_sys_sendmmsg net/socket.c:2670 [inline] __se_sys_sendmmsg net/socket.c:2667 [inline] __x64_sys_sendmmsg+0xbc/0x120 net/socket.c:2667 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd CPU: 0 PID: 29770 Comm: syz-executor.0 Not tainted 6.3.0-rc6-syzkaller-gc478e5b17829 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/30/2023 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/8021q/vlan_dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 870e4935d6e6..b90781b9ece6 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -109,8 +109,8 @@ static netdev_tx_t vlan_dev_hard_start_xmit(struct sk_buff *skb, * NOTE: THIS ASSUMES DIX ETHERNET, SPECIFICALLY NOT SUPPORTING * OTHER THINGS LIKE FDDI/TokenRing/802.3 SNAPs... */ - if (veth->h_vlan_proto != vlan->vlan_proto || - vlan->flags & VLAN_FLAG_REORDER_HDR) { + if (vlan->flags & VLAN_FLAG_REORDER_HDR || + veth->h_vlan_proto != vlan->vlan_proto) { u16 vlan_tci; vlan_tci = vlan->vlan_id; vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb->priority); -- cgit v1.2.3 From 224a876e37543eee111bf9b6aa4935080e619335 Mon Sep 17 00:00:00 2001 From: Tom Rix Date: Sun, 14 May 2023 10:00:10 -0400 Subject: netfilter: conntrack: define variables exp_nat_nla_policy and any_addr with CONFIG_NF_NAT MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit gcc with W=1 and ! CONFIG_NF_NAT net/netfilter/nf_conntrack_netlink.c:3463:32: error: ‘exp_nat_nla_policy’ defined but not used [-Werror=unused-const-variable=] 3463 | static const struct nla_policy exp_nat_nla_policy[CTA_EXPECT_NAT_MAX+1] = { | ^~~~~~~~~~~~~~~~~~ net/netfilter/nf_conntrack_netlink.c:2979:33: error: ‘any_addr’ defined but not used [-Werror=unused-const-variable=] 2979 | static const union nf_inet_addr any_addr; | ^~~~~~~~ These variables use is controlled by CONFIG_NF_NAT, so should their definitions. Signed-off-by: Tom Rix Reviewed-by: Simon Horman Signed-off-by: Florian Westphal --- net/netfilter/nf_conntrack_netlink.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index d40544cd61a6..69c8c8c7e9b8 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -2976,7 +2976,9 @@ nla_put_failure: return -1; } +#if IS_ENABLED(CONFIG_NF_NAT) static const union nf_inet_addr any_addr; +#endif static __be32 nf_expect_get_id(const struct nf_conntrack_expect *exp) { @@ -3460,10 +3462,12 @@ ctnetlink_change_expect(struct nf_conntrack_expect *x, return 0; } +#if IS_ENABLED(CONFIG_NF_NAT) static const struct nla_policy exp_nat_nla_policy[CTA_EXPECT_NAT_MAX+1] = { [CTA_EXPECT_NAT_DIR] = { .type = NLA_U32 }, [CTA_EXPECT_NAT_TUPLE] = { .type = NLA_NESTED }, }; +#endif static int ctnetlink_parse_expect_nat(const struct nlattr *attr, -- cgit v1.2.3 From e3c361b8acd636f5fe80c02849ca175201edf10c Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 11 May 2023 14:15:15 +0200 Subject: netfilter: nf_tables: fix nft_trans type confusion nft_trans_FOO objects all share a common nft_trans base structure, but trailing fields depend on the real object size. Access is only safe after trans->msg_type check. Check for rule type first. Found by code inspection. Fixes: 1a94e38d254b ("netfilter: nf_tables: add NFTA_RULE_ID attribute") Signed-off-by: Florian Westphal --- net/netfilter/nf_tables_api.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 59fb8320ab4d..dc5675962de4 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -3865,12 +3865,10 @@ static struct nft_rule *nft_rule_lookup_byid(const struct net *net, struct nft_trans *trans; list_for_each_entry(trans, &nft_net->commit_list, list) { - struct nft_rule *rule = nft_trans_rule(trans); - if (trans->msg_type == NFT_MSG_NEWRULE && trans->ctx.chain == chain && id == nft_trans_rule_id(trans)) - return rule; + return nft_trans_rule(trans); } return ERR_PTR(-ENOENT); } -- cgit v1.2.3 From 61ae320a29b0540c16931816299eb86bf2b66c08 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 11 May 2023 22:39:30 +0200 Subject: netfilter: nft_set_rbtree: fix null deref on element insertion There is no guarantee that rb_prev() will not return NULL in nft_rbtree_gc_elem(): general protection fault, probably for non-canonical address 0xdffffc0000000003: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000018-0x000000000000001f] nft_add_set_elem+0x14b0/0x2990 nf_tables_newsetelem+0x528/0xb30 Furthermore, there is a possible use-after-free while iterating, 'node' can be free'd so we need to cache the next value to use. Fixes: c9e6978e2725 ("netfilter: nft_set_rbtree: Switch to node list walk for overlap detection") Signed-off-by: Florian Westphal --- net/netfilter/nft_set_rbtree.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c index 19ea4d3c3553..2f114aa10f1a 100644 --- a/net/netfilter/nft_set_rbtree.c +++ b/net/netfilter/nft_set_rbtree.c @@ -221,7 +221,7 @@ static int nft_rbtree_gc_elem(const struct nft_set *__set, { struct nft_set *set = (struct nft_set *)__set; struct rb_node *prev = rb_prev(&rbe->node); - struct nft_rbtree_elem *rbe_prev; + struct nft_rbtree_elem *rbe_prev = NULL; struct nft_set_gc_batch *gcb; gcb = nft_set_gc_batch_check(set, NULL, GFP_ATOMIC); @@ -229,17 +229,21 @@ static int nft_rbtree_gc_elem(const struct nft_set *__set, return -ENOMEM; /* search for expired end interval coming before this element. */ - do { + while (prev) { rbe_prev = rb_entry(prev, struct nft_rbtree_elem, node); if (nft_rbtree_interval_end(rbe_prev)) break; prev = rb_prev(prev); - } while (prev != NULL); + } + + if (rbe_prev) { + rb_erase(&rbe_prev->node, &priv->root); + atomic_dec(&set->nelems); + } - rb_erase(&rbe_prev->node, &priv->root); rb_erase(&rbe->node, &priv->root); - atomic_sub(2, &set->nelems); + atomic_dec(&set->nelems); nft_set_gc_batch_add(gcb, rbe); nft_set_gc_batch_complete(gcb); @@ -268,7 +272,7 @@ static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set, struct nft_set_ext **ext) { struct nft_rbtree_elem *rbe, *rbe_le = NULL, *rbe_ge = NULL; - struct rb_node *node, *parent, **p, *first = NULL; + struct rb_node *node, *next, *parent, **p, *first = NULL; struct nft_rbtree *priv = nft_set_priv(set); u8 genmask = nft_genmask_next(net); int d, err; @@ -307,7 +311,9 @@ static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set, * Values stored in the tree are in reversed order, starting from * highest to lowest value. */ - for (node = first; node != NULL; node = rb_next(node)) { + for (node = first; node != NULL; node = next) { + next = rb_next(node); + rbe = rb_entry(node, struct nft_rbtree_elem, node); if (!nft_set_elem_active(&rbe->ext, genmask)) -- cgit v1.2.3 From fb1b7be9b16c1f4626969ba4e95a97da2a452b41 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 16 May 2023 21:45:34 +0200 Subject: atm: hide unused procfs functions When CONFIG_PROC_FS is disabled, the function declarations for some procfs functions are hidden, but the definitions are still build, as shown by this compiler warning: net/atm/resources.c:403:7: error: no previous prototype for 'atm_dev_seq_start' [-Werror=missing-prototypes] net/atm/resources.c:409:6: error: no previous prototype for 'atm_dev_seq_stop' [-Werror=missing-prototypes] net/atm/resources.c:414:7: error: no previous prototype for 'atm_dev_seq_next' [-Werror=missing-prototypes] Add another #ifdef to leave these out of the build. Signed-off-by: Arnd Bergmann Link: https://lore.kernel.org/r/20230516194625.549249-2-arnd@kernel.org Signed-off-by: Jakub Kicinski --- net/atm/resources.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/atm/resources.c b/net/atm/resources.c index 2b2d33eeaf20..995d29e7fb13 100644 --- a/net/atm/resources.c +++ b/net/atm/resources.c @@ -400,6 +400,7 @@ done: return error; } +#ifdef CONFIG_PROC_FS void *atm_dev_seq_start(struct seq_file *seq, loff_t *pos) { mutex_lock(&atm_dev_mutex); @@ -415,3 +416,4 @@ void *atm_dev_seq_next(struct seq_file *seq, void *v, loff_t *pos) { return seq_list_next(v, &atm_devs, pos); } +#endif -- cgit v1.2.3 From 89dcd87ce534a3a7f267cfd58505803006f51301 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 16 May 2023 21:45:35 +0200 Subject: bridge: always declare tunnel functions When CONFIG_BRIDGE_VLAN_FILTERING is disabled, two functions are still defined but have no prototype or caller. This causes a W=1 warning for the missing prototypes: net/bridge/br_netlink_tunnel.c:29:6: error: no previous prototype for 'vlan_tunid_inrange' [-Werror=missing-prototypes] net/bridge/br_netlink_tunnel.c:199:5: error: no previous prototype for 'br_vlan_tunnel_info' [-Werror=missing-prototypes] The functions are already contitional on CONFIG_BRIDGE_VLAN_FILTERING, and I coulnd't easily figure out the right set of #ifdefs, so just move the declarations out of the #ifdef to avoid the warning, at a small cost in code size over a more elaborate fix. Fixes: 188c67dd1906 ("net: bridge: vlan options: add support for tunnel id dumping") Fixes: 569da0822808 ("net: bridge: vlan options: add support for tunnel mapping set/del") Signed-off-by: Arnd Bergmann Acked-by: Nikolay Aleksandrov Link: https://lore.kernel.org/r/20230516194625.549249-3-arnd@kernel.org Signed-off-by: Jakub Kicinski --- net/bridge/br_private_tunnel.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/bridge/br_private_tunnel.h b/net/bridge/br_private_tunnel.h index 2b053289f016..efb096025151 100644 --- a/net/bridge/br_private_tunnel.h +++ b/net/bridge/br_private_tunnel.h @@ -27,6 +27,10 @@ int br_process_vlan_tunnel_info(const struct net_bridge *br, int br_get_vlan_tunnel_info_size(struct net_bridge_vlan_group *vg); int br_fill_vlan_tunnel_info(struct sk_buff *skb, struct net_bridge_vlan_group *vg); +bool vlan_tunid_inrange(const struct net_bridge_vlan *v_curr, + const struct net_bridge_vlan *v_last); +int br_vlan_tunnel_info(const struct net_bridge_port *p, int cmd, + u16 vid, u32 tun_id, bool *changed); #ifdef CONFIG_BRIDGE_VLAN_FILTERING /* br_vlan_tunnel.c */ @@ -43,10 +47,6 @@ void br_handle_ingress_vlan_tunnel(struct sk_buff *skb, struct net_bridge_vlan_group *vg); int br_handle_egress_vlan_tunnel(struct sk_buff *skb, struct net_bridge_vlan *vlan); -bool vlan_tunid_inrange(const struct net_bridge_vlan *v_curr, - const struct net_bridge_vlan *v_last); -int br_vlan_tunnel_info(const struct net_bridge_port *p, int cmd, - u16 vid, u32 tun_id, bool *changed); #else static inline int vlan_tunnel_init(struct net_bridge_vlan_group *vg) { -- cgit v1.2.3 From b3a03b540e3cf62a255213d084d76d71c02793d5 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 16 May 2023 18:50:36 -0700 Subject: tls: rx: device: fix checking decryption status skb->len covers the entire skb, including the frag_list. In fact we're guaranteed that rxm->full_len <= skb->len, so since the change under Fixes we were not checking decrypt status of any skb but the first. Note that the skb_pagelen() added here may feel a bit costly, but it's removed by subsequent fixes, anyway. Reported-by: Tariq Toukan Fixes: 86b259f6f888 ("tls: rx: device: bound the frag walk") Tested-by: Shai Amiram Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/tls/tls_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c index a7cc4f9faac2..3b87c7b04ac8 100644 --- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -1012,7 +1012,7 @@ int tls_device_decrypted(struct sock *sk, struct tls_context *tls_ctx) struct sk_buff *skb_iter; int left; - left = rxm->full_len - skb->len; + left = rxm->full_len + rxm->offset - skb_pagelen(skb); /* Check if all the data is decrypted already */ skb_iter = skb_shinfo(skb)->frag_list; while (skb_iter && left > 0) { -- cgit v1.2.3 From 210620ae44a83f25220450bbfcc22e6fe986b25f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 16 May 2023 18:50:37 -0700 Subject: tls: rx: strp: set the skb->len of detached / CoW'ed skbs alloc_skb_with_frags() fills in page frag sizes but does not set skb->len and skb->data_len. Set those correctly otherwise device offload will most likely generate an empty skb and hit the BUG() at the end of __skb_nsg(). Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser") Tested-by: Shai Amiram Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/tls/tls_strp.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/tls/tls_strp.c b/net/tls/tls_strp.c index 955ac3e0bf4d..24016c865e00 100644 --- a/net/tls/tls_strp.c +++ b/net/tls/tls_strp.c @@ -56,6 +56,8 @@ static struct sk_buff *tls_strp_msg_make_copy(struct tls_strparser *strp) offset += skb_frag_size(frag); } + skb->len = strp->stm.full_len; + skb->data_len = strp->stm.full_len; skb_copy_header(skb, strp->anchor); rxm = strp_msg(skb); rxm->offset = 0; -- cgit v1.2.3 From 14c4be92ebb3e36e392aa9dd8f314038a9f96f3c Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 16 May 2023 18:50:38 -0700 Subject: tls: rx: strp: force mixed decrypted records into copy mode If a record is partially decrypted we'll have to CoW it, anyway, so go into copy mode and allocate a writable skb right away. This will make subsequent fix simpler because we won't have to teach tls_strp_msg_make_copy() how to copy skbs while preserving decrypt status. Tested-by: Shai Amiram Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- include/linux/skbuff.h | 10 ++++++++++ net/tls/tls_strp.c | 16 +++++++++++----- 2 files changed, 21 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 738776ab8838..0b40417457cd 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1587,6 +1587,16 @@ static inline void skb_copy_hash(struct sk_buff *to, const struct sk_buff *from) to->l4_hash = from->l4_hash; }; +static inline int skb_cmp_decrypted(const struct sk_buff *skb1, + const struct sk_buff *skb2) +{ +#ifdef CONFIG_TLS_DEVICE + return skb2->decrypted - skb1->decrypted; +#else + return 0; +#endif +} + static inline void skb_copy_decrypted(struct sk_buff *to, const struct sk_buff *from) { diff --git a/net/tls/tls_strp.c b/net/tls/tls_strp.c index 24016c865e00..2b6fa9855999 100644 --- a/net/tls/tls_strp.c +++ b/net/tls/tls_strp.c @@ -317,15 +317,19 @@ static int tls_strp_read_copy(struct tls_strparser *strp, bool qshort) return 0; } -static bool tls_strp_check_no_dup(struct tls_strparser *strp) +static bool tls_strp_check_queue_ok(struct tls_strparser *strp) { unsigned int len = strp->stm.offset + strp->stm.full_len; - struct sk_buff *skb; + struct sk_buff *first, *skb; u32 seq; - skb = skb_shinfo(strp->anchor)->frag_list; - seq = TCP_SKB_CB(skb)->seq; + first = skb_shinfo(strp->anchor)->frag_list; + skb = first; + seq = TCP_SKB_CB(first)->seq; + /* Make sure there's no duplicate data in the queue, + * and the decrypted status matches. + */ while (skb->len < len) { seq += skb->len; len -= skb->len; @@ -333,6 +337,8 @@ static bool tls_strp_check_no_dup(struct tls_strparser *strp) if (TCP_SKB_CB(skb)->seq != seq) return false; + if (skb_cmp_decrypted(first, skb)) + return false; } return true; @@ -413,7 +419,7 @@ static int tls_strp_read_sock(struct tls_strparser *strp) return tls_strp_read_copy(strp, true); } - if (!tls_strp_check_no_dup(strp)) + if (!tls_strp_check_queue_ok(strp)) return tls_strp_read_copy(strp, false); strp->msg_ready = 1; -- cgit v1.2.3 From 8b0c0dc9fbbd01e58a573a41c38885f9e4c17696 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 16 May 2023 18:50:39 -0700 Subject: tls: rx: strp: fix determining record length in copy mode We call tls_rx_msg_size(skb) before doing skb->len += chunk. So the tls_rx_msg_size() code will see old skb->len, most likely leading to an over-read. Worst case we will over read an entire record, next iteration will try to trim the skb but may end up turning frag len negative or discarding the subsequent record (since we already told TCP we've read it during previous read but now we'll trim it out of the skb). Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser") Tested-by: Shai Amiram Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/tls/tls_strp.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/tls/tls_strp.c b/net/tls/tls_strp.c index 2b6fa9855999..e2e48217e7ac 100644 --- a/net/tls/tls_strp.c +++ b/net/tls/tls_strp.c @@ -210,19 +210,28 @@ static int tls_strp_copyin(read_descriptor_t *desc, struct sk_buff *in_skb, skb_frag_size(frag), chunk)); - sz = tls_rx_msg_size(strp, strp->anchor); + skb->len += chunk; + skb->data_len += chunk; + skb_frag_size_add(frag, chunk); + + sz = tls_rx_msg_size(strp, skb); if (sz < 0) { desc->error = sz; return 0; } /* We may have over-read, sz == 0 is guaranteed under-read */ - if (sz > 0) - chunk = min_t(size_t, chunk, sz - skb->len); + if (unlikely(sz && sz < skb->len)) { + int over = skb->len - sz; + + WARN_ON_ONCE(over > chunk); + skb->len -= over; + skb->data_len -= over; + skb_frag_size_add(frag, -over); + + chunk -= over; + } - skb->len += chunk; - skb->data_len += chunk; - skb_frag_size_add(frag, chunk); frag++; len -= chunk; offset += chunk; -- cgit v1.2.3 From c1c607b1e5d5477d82ca6a86a05a4f10907b33ee Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 16 May 2023 18:50:40 -0700 Subject: tls: rx: strp: factor out copying skb data We'll need to copy input skbs individually in the next patch. Factor that code out (without assuming we're copying a full record). Tested-by: Shai Amiram Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/tls/tls_strp.c | 33 +++++++++++++++++++++++---------- 1 file changed, 23 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/tls/tls_strp.c b/net/tls/tls_strp.c index e2e48217e7ac..61fbf84baf9e 100644 --- a/net/tls/tls_strp.c +++ b/net/tls/tls_strp.c @@ -34,31 +34,44 @@ static void tls_strp_anchor_free(struct tls_strparser *strp) strp->anchor = NULL; } -/* Create a new skb with the contents of input copied to its page frags */ -static struct sk_buff *tls_strp_msg_make_copy(struct tls_strparser *strp) +static struct sk_buff * +tls_strp_skb_copy(struct tls_strparser *strp, struct sk_buff *in_skb, + int offset, int len) { - struct strp_msg *rxm; struct sk_buff *skb; - int i, err, offset; + int i, err; - skb = alloc_skb_with_frags(0, strp->stm.full_len, TLS_PAGE_ORDER, + skb = alloc_skb_with_frags(0, len, TLS_PAGE_ORDER, &err, strp->sk->sk_allocation); if (!skb) return NULL; - offset = strp->stm.offset; for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; - WARN_ON_ONCE(skb_copy_bits(strp->anchor, offset, + WARN_ON_ONCE(skb_copy_bits(in_skb, offset, skb_frag_address(frag), skb_frag_size(frag))); offset += skb_frag_size(frag); } - skb->len = strp->stm.full_len; - skb->data_len = strp->stm.full_len; - skb_copy_header(skb, strp->anchor); + skb->len = len; + skb->data_len = len; + skb_copy_header(skb, in_skb); + return skb; +} + +/* Create a new skb with the contents of input copied to its page frags */ +static struct sk_buff *tls_strp_msg_make_copy(struct tls_strparser *strp) +{ + struct strp_msg *rxm; + struct sk_buff *skb; + + skb = tls_strp_skb_copy(strp, strp->anchor, strp->stm.offset, + strp->stm.full_len); + if (!skb) + return NULL; + rxm = strp_msg(skb); rxm->offset = 0; return skb; -- cgit v1.2.3 From eca9bfafee3a0487e59c59201ae14c7594ba940a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 16 May 2023 18:50:41 -0700 Subject: tls: rx: strp: preserve decryption status of skbs when needed When receive buffer is small we try to copy out the data from TCP into a skb maintained by TLS to prevent connection from stalling. Unfortunately if a single record is made up of a mix of decrypted and non-decrypted skbs combining them into a single skb leads to loss of decryption status, resulting in decryption errors or data corruption. Similarly when trying to use TCP receive queue directly we need to make sure that all the skbs within the record have the same status. If we don't the mixed status will be detected correctly but we'll CoW the anchor, again collapsing it into a single paged skb without decrypted status preserved. So the "fixup" code will not know which parts of skb to re-encrypt. Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser") Tested-by: Shai Amiram Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- include/net/tls.h | 1 + net/tls/tls.h | 5 +++ net/tls/tls_device.c | 22 ++++------ net/tls/tls_strp.c | 117 +++++++++++++++++++++++++++++++++++++++++++-------- 4 files changed, 114 insertions(+), 31 deletions(-) (limited to 'net') diff --git a/include/net/tls.h b/include/net/tls.h index 6056ce5a2aa5..596595c4b1af 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -126,6 +126,7 @@ struct tls_strparser { u32 mark : 8; u32 stopped : 1; u32 copy_mode : 1; + u32 mixed_decrypted : 1; u32 msg_ready : 1; struct strp_msg stm; diff --git a/net/tls/tls.h b/net/tls/tls.h index 804c3880d028..0672acab2773 100644 --- a/net/tls/tls.h +++ b/net/tls/tls.h @@ -167,6 +167,11 @@ static inline bool tls_strp_msg_ready(struct tls_sw_context_rx *ctx) return ctx->strp.msg_ready; } +static inline bool tls_strp_msg_mixed_decrypted(struct tls_sw_context_rx *ctx) +{ + return ctx->strp.mixed_decrypted; +} + #ifdef CONFIG_TLS_DEVICE int tls_device_init(void); void tls_device_cleanup(void); diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c index 3b87c7b04ac8..bf69c9d6d06c 100644 --- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -1007,20 +1007,14 @@ int tls_device_decrypted(struct sock *sk, struct tls_context *tls_ctx) struct tls_sw_context_rx *sw_ctx = tls_sw_ctx_rx(tls_ctx); struct sk_buff *skb = tls_strp_msg(sw_ctx); struct strp_msg *rxm = strp_msg(skb); - int is_decrypted = skb->decrypted; - int is_encrypted = !is_decrypted; - struct sk_buff *skb_iter; - int left; - - left = rxm->full_len + rxm->offset - skb_pagelen(skb); - /* Check if all the data is decrypted already */ - skb_iter = skb_shinfo(skb)->frag_list; - while (skb_iter && left > 0) { - is_decrypted &= skb_iter->decrypted; - is_encrypted &= !skb_iter->decrypted; - - left -= skb_iter->len; - skb_iter = skb_iter->next; + int is_decrypted, is_encrypted; + + if (!tls_strp_msg_mixed_decrypted(sw_ctx)) { + is_decrypted = skb->decrypted; + is_encrypted = !is_decrypted; + } else { + is_decrypted = 0; + is_encrypted = 0; } trace_tls_device_decrypted(sk, tcp_sk(sk)->copied_seq - rxm->full_len, diff --git a/net/tls/tls_strp.c b/net/tls/tls_strp.c index 61fbf84baf9e..da95abbb7ea3 100644 --- a/net/tls/tls_strp.c +++ b/net/tls/tls_strp.c @@ -29,7 +29,8 @@ static void tls_strp_anchor_free(struct tls_strparser *strp) struct skb_shared_info *shinfo = skb_shinfo(strp->anchor); DEBUG_NET_WARN_ON_ONCE(atomic_read(&shinfo->dataref) != 1); - shinfo->frag_list = NULL; + if (!strp->copy_mode) + shinfo->frag_list = NULL; consume_skb(strp->anchor); strp->anchor = NULL; } @@ -195,22 +196,22 @@ static void tls_strp_flush_anchor_copy(struct tls_strparser *strp) for (i = 0; i < shinfo->nr_frags; i++) __skb_frag_unref(&shinfo->frags[i], false); shinfo->nr_frags = 0; + if (strp->copy_mode) { + kfree_skb_list(shinfo->frag_list); + shinfo->frag_list = NULL; + } strp->copy_mode = 0; + strp->mixed_decrypted = 0; } -static int tls_strp_copyin(read_descriptor_t *desc, struct sk_buff *in_skb, - unsigned int offset, size_t in_len) +static int tls_strp_copyin_frag(struct tls_strparser *strp, struct sk_buff *skb, + struct sk_buff *in_skb, unsigned int offset, + size_t in_len) { - struct tls_strparser *strp = (struct tls_strparser *)desc->arg.data; - struct sk_buff *skb; - skb_frag_t *frag; size_t len, chunk; + skb_frag_t *frag; int sz; - if (strp->msg_ready) - return 0; - - skb = strp->anchor; frag = &skb_shinfo(skb)->frags[skb->len / PAGE_SIZE]; len = in_len; @@ -228,10 +229,8 @@ static int tls_strp_copyin(read_descriptor_t *desc, struct sk_buff *in_skb, skb_frag_size_add(frag, chunk); sz = tls_rx_msg_size(strp, skb); - if (sz < 0) { - desc->error = sz; - return 0; - } + if (sz < 0) + return sz; /* We may have over-read, sz == 0 is guaranteed under-read */ if (unlikely(sz && sz < skb->len)) { @@ -271,15 +270,99 @@ static int tls_strp_copyin(read_descriptor_t *desc, struct sk_buff *in_skb, offset += chunk; } - if (strp->stm.full_len == skb->len) { +read_done: + return in_len - len; +} + +static int tls_strp_copyin_skb(struct tls_strparser *strp, struct sk_buff *skb, + struct sk_buff *in_skb, unsigned int offset, + size_t in_len) +{ + struct sk_buff *nskb, *first, *last; + struct skb_shared_info *shinfo; + size_t chunk; + int sz; + + if (strp->stm.full_len) + chunk = strp->stm.full_len - skb->len; + else + chunk = TLS_MAX_PAYLOAD_SIZE + PAGE_SIZE; + chunk = min(chunk, in_len); + + nskb = tls_strp_skb_copy(strp, in_skb, offset, chunk); + if (!nskb) + return -ENOMEM; + + shinfo = skb_shinfo(skb); + if (!shinfo->frag_list) { + shinfo->frag_list = nskb; + nskb->prev = nskb; + } else { + first = shinfo->frag_list; + last = first->prev; + last->next = nskb; + first->prev = nskb; + } + + skb->len += chunk; + skb->data_len += chunk; + + if (!strp->stm.full_len) { + sz = tls_rx_msg_size(strp, skb); + if (sz < 0) + return sz; + + /* We may have over-read, sz == 0 is guaranteed under-read */ + if (unlikely(sz && sz < skb->len)) { + int over = skb->len - sz; + + WARN_ON_ONCE(over > chunk); + skb->len -= over; + skb->data_len -= over; + __pskb_trim(nskb, nskb->len - over); + + chunk -= over; + } + + strp->stm.full_len = sz; + } + + return chunk; +} + +static int tls_strp_copyin(read_descriptor_t *desc, struct sk_buff *in_skb, + unsigned int offset, size_t in_len) +{ + struct tls_strparser *strp = (struct tls_strparser *)desc->arg.data; + struct sk_buff *skb; + int ret; + + if (strp->msg_ready) + return 0; + + skb = strp->anchor; + if (!skb->len) + skb_copy_decrypted(skb, in_skb); + else + strp->mixed_decrypted |= !!skb_cmp_decrypted(skb, in_skb); + + if (IS_ENABLED(CONFIG_TLS_DEVICE) && strp->mixed_decrypted) + ret = tls_strp_copyin_skb(strp, skb, in_skb, offset, in_len); + else + ret = tls_strp_copyin_frag(strp, skb, in_skb, offset, in_len); + if (ret < 0) { + desc->error = ret; + ret = 0; + } + + if (strp->stm.full_len && strp->stm.full_len == skb->len) { desc->count = 0; strp->msg_ready = 1; tls_rx_msg_ready(strp); } -read_done: - return in_len - len; + return ret; } static int tls_strp_read_copyin(struct tls_strparser *strp) -- cgit v1.2.3 From 74836ec828fe17b63f2006fdbf53311d691396bf Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 16 May 2023 18:50:42 -0700 Subject: tls: rx: strp: don't use GFP_KERNEL in softirq context When receive buffer is small, or the TCP rx queue looks too complicated to bother using it directly - we allocate a new skb and copy data into it. We already use sk->sk_allocation... but nothing actually sets it to GFP_ATOMIC on the ->sk_data_ready() path. Users of HW offload are far more likely to experience problems due to scheduling while atomic. "Copy mode" is very rarely triggered with SW crypto. Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser") Tested-by: Shai Amiram Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/tls/tls_sw.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 635b8bf6b937..6e6a7c37d685 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -2304,10 +2304,14 @@ static void tls_data_ready(struct sock *sk) struct tls_context *tls_ctx = tls_get_ctx(sk); struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx); struct sk_psock *psock; + gfp_t alloc_save; trace_sk_data_ready(sk); + alloc_save = sk->sk_allocation; + sk->sk_allocation = GFP_ATOMIC; tls_strp_data_ready(&ctx->strp); + sk->sk_allocation = alloc_save; psock = sk_psock_get(sk); if (psock) { -- cgit v1.2.3 From 35112271672ae98f45df7875244a4e33aa215e31 Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Thu, 18 May 2023 13:14:55 +0800 Subject: net/smc: Reset connection when trying to use SMCRv2 fails. We found a crash when using SMCRv2 with 2 Mellanox ConnectX-4. It can be reproduced by: - smc_run nginx - smc_run wrk -t 32 -c 500 -d 30 http://: BUG: kernel NULL pointer dereference, address: 0000000000000014 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 8000000108713067 P4D 8000000108713067 PUD 151127067 PMD 0 Oops: 0000 [#1] PREEMPT SMP PTI CPU: 4 PID: 2441 Comm: kworker/4:249 Kdump: loaded Tainted: G W E 6.4.0-rc1+ #42 Workqueue: smc_hs_wq smc_listen_work [smc] RIP: 0010:smc_clc_send_confirm_accept+0x284/0x580 [smc] RSP: 0018:ffffb8294b2d7c78 EFLAGS: 00010a06 RAX: ffff8f1873238880 RBX: ffffb8294b2d7dc8 RCX: 0000000000000000 RDX: 00000000000000b4 RSI: 0000000000000001 RDI: 0000000000b40c00 RBP: ffffb8294b2d7db8 R08: ffff8f1815c5860c R09: 0000000000000000 R10: 0000000000000400 R11: 0000000000000000 R12: ffff8f1846f56180 R13: ffff8f1815c5860c R14: 0000000000000001 R15: 0000000000000001 FS: 0000000000000000(0000) GS:ffff8f1aefd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000014 CR3: 00000001027a0001 CR4: 00000000003706e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: ? mlx5_ib_map_mr_sg+0xa1/0xd0 [mlx5_ib] ? smcr_buf_map_link+0x24b/0x290 [smc] ? __smc_buf_create+0x4ee/0x9b0 [smc] smc_clc_send_accept+0x4c/0xb0 [smc] smc_listen_work+0x346/0x650 [smc] ? __schedule+0x279/0x820 process_one_work+0x1e5/0x3f0 worker_thread+0x4d/0x2f0 ? __pfx_worker_thread+0x10/0x10 kthread+0xe5/0x120 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x2c/0x50 During the CLC handshake, server sequentially tries available SMCRv2 and SMCRv1 devices in smc_listen_work(). If an SMCRv2 device is found. SMCv2 based link group and link will be assigned to the connection. Then assumed that some buffer assignment errors happen later in the CLC handshake, such as RMB registration failure, server will give up SMCRv2 and try SMCRv1 device instead. But the resources assigned to the connection won't be reset. When server tries SMCRv1 device, the connection creation process will be executed again. Since conn->lnk has been assigned when trying SMCRv2, it will not be set to the correct SMCRv1 link in smcr_lgr_conn_assign_link(). So in such situation, conn->lgr points to correct SMCRv1 link group but conn->lnk points to the SMCRv2 link mistakenly. Then in smc_clc_send_confirm_accept(), conn->rmb_desc->mr[link->link_idx] will be accessed. Since the link->link_idx is not correct, the related MR may not have been initialized, so crash happens. | Try SMCRv2 device first | |-> conn->lgr: assign existed SMCRv2 link group; | |-> conn->link: assign existed SMCRv2 link (link_idx may be 1 in SMC_LGR_SYMMETRIC); | |-> sndbuf & RMB creation fails, quit; | | Try SMCRv1 device then | |-> conn->lgr: create SMCRv1 link group and assign; | |-> conn->link: keep SMCRv2 link mistakenly; | |-> sndbuf & RMB creation succeed, only RMB->mr[link_idx = 0] | initialized. | | Then smc_clc_send_confirm_accept() accesses | conn->rmb_desc->mr[conn->link->link_idx, which is 1], then crash. v This patch tries to fix this by cleaning conn->lnk before assigning link. In addition, it is better to reset the connection and clean the resources assigned if trying SMCRv2 failed in buffer creation or registration. Fixes: e49300a6bf62 ("net/smc: add listen processing for SMC-Rv2") Link: https://lore.kernel.org/r/20220523055056.2078994-1-liuyacan@corp.netease.com/ Signed-off-by: Wen Gu Reviewed-by: Tony Lu Signed-off-by: David S. Miller --- net/smc/af_smc.c | 9 +++++++-- net/smc/smc_core.c | 1 + 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 50c38b624f77..538e9c6ec8c9 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -2000,8 +2000,10 @@ static int smc_listen_rdma_init(struct smc_sock *new_smc, return rc; /* create send buffer and rmb */ - if (smc_buf_create(new_smc, false)) + if (smc_buf_create(new_smc, false)) { + smc_conn_abort(new_smc, ini->first_contact_local); return SMC_CLC_DECL_MEM; + } return 0; } @@ -2217,8 +2219,11 @@ static void smc_find_rdma_v2_device_serv(struct smc_sock *new_smc, smcr_version = ini->smcr_version; ini->smcr_version = SMC_V2; rc = smc_listen_rdma_init(new_smc, ini); - if (!rc) + if (!rc) { rc = smc_listen_rdma_reg(new_smc, ini->first_contact_local); + if (rc) + smc_conn_abort(new_smc, ini->first_contact_local); + } if (!rc) return; ini->smcr_version = smcr_version; diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 454356771cda..3f465faf2b68 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -127,6 +127,7 @@ static int smcr_lgr_conn_assign_link(struct smc_connection *conn, bool first) int i, j; /* do link balancing */ + conn->lnk = NULL; /* reset conn->lnk first */ for (i = 0; i < SMC_LINKS_PER_LGR_MAX; i++) { struct smc_link *lnk = &conn->lgr->lnk[i]; -- cgit v1.2.3 From d180891fba995bd54e25b089b1ec98d134873586 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 10 May 2023 12:28:00 -0400 Subject: SUNRPC: Don't change task->tk_status after the call to rpc_exit_task Some calls to rpc_exit_task() may deliberately change the value of task->tk_status, for instance because it gets checked by the RPC call's rpc_release() callback. That makes it wrong to reset the value to task->tk_rpc_status. In particular this causes a bug where the rpc_call_done() callback tries to fail over a set of pNFS/flexfiles writes to a different IP address, but the reset of task->tk_status causes nfs_commit_release_pages() to immediately mark the file as having a fatal error. Fixes: 39494194f93b ("SUNRPC: Fix races with rpc_killall_tasks()") Cc: stable@vger.kernel.org # 6.1.x Signed-off-by: Trond Myklebust Signed-off-by: Anna Schumaker --- net/sunrpc/sched.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index c8321de341ee..6debf4fd42d4 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -927,11 +927,10 @@ static void __rpc_execute(struct rpc_task *task) */ do_action = task->tk_action; /* Tasks with an RPC error status should exit */ - if (do_action != rpc_exit_task && + if (do_action && do_action != rpc_exit_task && (status = READ_ONCE(task->tk_rpc_status)) != 0) { task->tk_status = status; - if (do_action != NULL) - do_action = rpc_exit_task; + do_action = rpc_exit_task; } /* Callbacks override all actions */ if (task->tk_callback) { -- cgit v1.2.3 From ca1fd42e7dbfcb34890ffbf1f2f4b356776dab6f Mon Sep 17 00:00:00 2001 From: Ruihan Li Date: Wed, 3 May 2023 21:39:34 +0800 Subject: Bluetooth: Fix potential double free caused by hci_conn_unlink The hci_conn_unlink function is being called by hci_conn_del, which means it should not call hci_conn_del with the input parameter conn again. If it does, conn may have already been released when hci_conn_unlink returns, leading to potential UAF and double-free issues. This patch resolves the problem by modifying hci_conn_unlink to release only conn's child links when necessary, but never release conn itself. Reported-by: syzbot+690b90b14f14f43f4688@syzkaller.appspotmail.com Closes: https://lore.kernel.org/linux-bluetooth/000000000000484a8205faafe216@google.com/ Fixes: 06149746e720 ("Bluetooth: hci_conn: Add support for linking multiple hcon") Signed-off-by: Ruihan Li Signed-off-by: Luiz Augusto von Dentz Reported-by: syzbot+690b90b14f14f43f4688@syzkaller.appspotmail.com Reported-by: Luiz Augusto von Dentz Reported-by: syzbot+8bb72f86fc823817bc5d@syzkaller.appspotmail.com --- net/bluetooth/hci_conn.c | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 640b951bf40a..70e1655a9df6 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1083,8 +1083,18 @@ static void hci_conn_unlink(struct hci_conn *conn) if (!conn->parent) { struct hci_link *link, *t; - list_for_each_entry_safe(link, t, &conn->link_list, list) - hci_conn_unlink(link->conn); + list_for_each_entry_safe(link, t, &conn->link_list, list) { + struct hci_conn *child = link->conn; + + hci_conn_unlink(child); + + /* Due to race, SCO connection might be not established + * yet at this point. Delete it now, otherwise it is + * possible for it to be stuck and can't be deleted. + */ + if (child->handle == HCI_CONN_HANDLE_UNSET) + hci_conn_del(child); + } return; } @@ -1100,13 +1110,6 @@ static void hci_conn_unlink(struct hci_conn *conn) kfree(conn->link); conn->link = NULL; - - /* Due to race, SCO connection might be not established - * yet at this point. Delete it now, otherwise it is - * possible for it to be stuck and can't be deleted. - */ - if (conn->handle == HCI_CONN_HANDLE_UNSET) - hci_conn_del(conn); } int hci_conn_del(struct hci_conn *conn) -- cgit v1.2.3 From 2910431ab0e500dfc5df12299bb15eef0f30b43e Mon Sep 17 00:00:00 2001 From: Ruihan Li Date: Wed, 3 May 2023 21:39:35 +0800 Subject: Bluetooth: Refcnt drop must be placed last in hci_conn_unlink If hci_conn_put(conn->parent) reduces conn->parent's reference count to zero, it can immediately deallocate conn->parent. At the same time, conn->link->list has its head in conn->parent, causing use-after-free problems in the latter list_del_rcu(&conn->link->list). This problem can be easily solved by reordering the two operations, i.e., first performing the list removal with list_del_rcu and then decreasing the refcnt with hci_conn_put. Reported-by: Luiz Augusto von Dentz Closes: https://lore.kernel.org/linux-bluetooth/CABBYNZ+1kce8_RJrLNOXd_8=Mdpb=2bx4Nto-hFORk=qiOkoCg@mail.gmail.com/ Fixes: 06149746e720 ("Bluetooth: hci_conn: Add support for linking multiple hcon") Signed-off-by: Ruihan Li Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 70e1655a9df6..44d0643fc681 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1102,12 +1102,12 @@ static void hci_conn_unlink(struct hci_conn *conn) if (!conn->link) return; - hci_conn_put(conn->parent); - conn->parent = NULL; - list_del_rcu(&conn->link->list); synchronize_rcu(); + hci_conn_put(conn->parent); + conn->parent = NULL; + kfree(conn->link); conn->link = NULL; } -- cgit v1.2.3 From a2ac591cb4d83e1f2d4b4adb3c14b2c79764650a Mon Sep 17 00:00:00 2001 From: Ruihan Li Date: Wed, 3 May 2023 21:39:36 +0800 Subject: Bluetooth: Fix UAF in hci_conn_hash_flush again Commit 06149746e720 ("Bluetooth: hci_conn: Add support for linking multiple hcon") reintroduced a previously fixed bug [1] ("KASAN: slab-use-after-free Read in hci_conn_hash_flush"). This bug was originally fixed by commit 5dc7d23e167e ("Bluetooth: hci_conn: Fix possible UAF"). The hci_conn_unlink function was added to avoid invalidating the link traversal caused by successive hci_conn_del operations releasing extra connections. However, currently hci_conn_unlink itself also releases extra connections, resulted in the reintroduced bug. This patch follows a more robust solution for cleaning up all connections, by repeatedly removing the first connection until there are none left. This approach does not rely on the inner workings of hci_conn_del and ensures proper cleanup of all connections. Meanwhile, we need to make sure that hci_conn_del never fails. Indeed it doesn't, as it now always returns zero. To make this a bit clearer, this patch also changes its return type to void. Reported-by: syzbot+8bb72f86fc823817bc5d@syzkaller.appspotmail.com Closes: https://lore.kernel.org/linux-bluetooth/000000000000aa920505f60d25ad@google.com/ Fixes: 06149746e720 ("Bluetooth: hci_conn: Add support for linking multiple hcon") Signed-off-by: Ruihan Li Co-developed-by: Luiz Augusto von Dentz Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 2 +- net/bluetooth/hci_conn.c | 33 ++++++++++++++++++++++----------- 2 files changed, 23 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index a6c8aee2f256..8baf34639939 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1327,7 +1327,7 @@ int hci_le_create_cis(struct hci_conn *conn); struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, u8 role); -int hci_conn_del(struct hci_conn *conn); +void hci_conn_del(struct hci_conn *conn); void hci_conn_hash_flush(struct hci_dev *hdev); void hci_conn_check_pending(struct hci_dev *hdev); diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 44d0643fc681..ce588359b290 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1088,6 +1088,14 @@ static void hci_conn_unlink(struct hci_conn *conn) hci_conn_unlink(child); + /* If hdev is down it means + * hci_dev_close_sync/hci_conn_hash_flush is in progress + * and links don't need to be cleanup as all connections + * would be cleanup. + */ + if (!test_bit(HCI_UP, &hdev->flags)) + continue; + /* Due to race, SCO connection might be not established * yet at this point. Delete it now, otherwise it is * possible for it to be stuck and can't be deleted. @@ -1112,7 +1120,7 @@ static void hci_conn_unlink(struct hci_conn *conn) conn->link = NULL; } -int hci_conn_del(struct hci_conn *conn) +void hci_conn_del(struct hci_conn *conn) { struct hci_dev *hdev = conn->hdev; @@ -1163,8 +1171,6 @@ int hci_conn_del(struct hci_conn *conn) * rest of hci_conn_del. */ hci_conn_cleanup(conn); - - return 0; } struct hci_dev *hci_get_route(bdaddr_t *dst, bdaddr_t *src, uint8_t src_type) @@ -2465,22 +2471,27 @@ timer: /* Drop all connection on the device */ void hci_conn_hash_flush(struct hci_dev *hdev) { - struct hci_conn_hash *h = &hdev->conn_hash; - struct hci_conn *c, *n; + struct list_head *head = &hdev->conn_hash.list; + struct hci_conn *conn; BT_DBG("hdev %s", hdev->name); - list_for_each_entry_safe(c, n, &h->list, list) { - c->state = BT_CLOSED; - - hci_disconn_cfm(c, HCI_ERROR_LOCAL_HOST_TERM); + /* We should not traverse the list here, because hci_conn_del + * can remove extra links, which may cause the list traversal + * to hit items that have already been released. + */ + while ((conn = list_first_entry_or_null(head, + struct hci_conn, + list)) != NULL) { + conn->state = BT_CLOSED; + hci_disconn_cfm(conn, HCI_ERROR_LOCAL_HOST_TERM); /* Unlink before deleting otherwise it is possible that * hci_conn_del removes the link which may cause the list to * contain items already freed. */ - hci_conn_unlink(c); - hci_conn_del(c); + hci_conn_unlink(conn); + hci_conn_del(conn); } } -- cgit v1.2.3 From a2904d2825536aa896a149a9174d11b0958e7095 Mon Sep 17 00:00:00 2001 From: Ruihan Li Date: Wed, 3 May 2023 21:39:37 +0800 Subject: Bluetooth: Unlink CISes when LE disconnects in hci_conn_del Currently, hci_conn_del calls hci_conn_unlink for BR/EDR, (e)SCO, and CIS connections, i.e., everything except LE connections. However, if (e)SCO connections are unlinked when BR/EDR disconnects, CIS connections should also be unlinked when LE disconnects. In terms of disconnection behavior, CIS and (e)SCO connections are not too different. One peculiarity of CIS is that when CIS connections are disconnected, the CIS handle isn't deleted, as per [BLUETOOTH CORE SPECIFICATION Version 5.4 | Vol 4, Part E] 7.1.6 Disconnect command: All SCO, eSCO, and CIS connections on a physical link should be disconnected before the ACL connection on the same physical connection is disconnected. If it does not, they will be implicitly disconnected as part of the ACL disconnection. ... Note: As specified in Section 7.7.5, on the Central, the handle for a CIS remains valid even after disconnection and, therefore, the Host can recreate a disconnected CIS at a later point in time using the same connection handle. Since hci_conn_link invokes both hci_conn_get and hci_conn_hold, hci_conn_unlink should perform both hci_conn_put and hci_conn_drop as well. However, currently it performs only hci_conn_put. This patch makes hci_conn_unlink call hci_conn_drop as well, which simplifies the logic in hci_conn_del a bit and may benefit future users of hci_conn_unlink. But it is noted that this change additionally implies that hci_conn_unlink can queue disc_work on conn itself, with the following call stack: hci_conn_unlink(conn) [conn->parent == NULL] -> hci_conn_unlink(child) [child->parent == conn] -> hci_conn_drop(child->parent) -> queue_delayed_work(&conn->disc_work) Queued disc_work after hci_conn_del can be spurious, so during the process of hci_conn_del, it is necessary to make the call to cancel_delayed_work(&conn->disc_work) after invoking hci_conn_unlink. Signed-off-by: Ruihan Li Co-developed-by: Luiz Augusto von Dentz Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 21 ++++++--------------- 1 file changed, 6 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index ce588359b290..f75ef12f18f7 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1100,7 +1100,9 @@ static void hci_conn_unlink(struct hci_conn *conn) * yet at this point. Delete it now, otherwise it is * possible for it to be stuck and can't be deleted. */ - if (child->handle == HCI_CONN_HANDLE_UNSET) + if ((child->type == SCO_LINK || + child->type == ESCO_LINK) && + child->handle == HCI_CONN_HANDLE_UNSET) hci_conn_del(child); } @@ -1113,6 +1115,7 @@ static void hci_conn_unlink(struct hci_conn *conn) list_del_rcu(&conn->link->list); synchronize_rcu(); + hci_conn_drop(conn->parent); hci_conn_put(conn->parent); conn->parent = NULL; @@ -1126,12 +1129,13 @@ void hci_conn_del(struct hci_conn *conn) BT_DBG("%s hcon %p handle %d", hdev->name, conn, conn->handle); + hci_conn_unlink(conn); + cancel_delayed_work_sync(&conn->disc_work); cancel_delayed_work_sync(&conn->auto_accept_work); cancel_delayed_work_sync(&conn->idle_work); if (conn->type == ACL_LINK) { - hci_conn_unlink(conn); /* Unacked frames */ hdev->acl_cnt += conn->sent; } else if (conn->type == LE_LINK) { @@ -1142,13 +1146,6 @@ void hci_conn_del(struct hci_conn *conn) else hdev->acl_cnt += conn->sent; } else { - struct hci_conn *acl = conn->parent; - - if (acl) { - hci_conn_unlink(conn); - hci_conn_drop(acl); - } - /* Unacked ISO frames */ if (conn->type == ISO_LINK) { if (hdev->iso_pkts) @@ -2485,12 +2482,6 @@ void hci_conn_hash_flush(struct hci_dev *hdev) list)) != NULL) { conn->state = BT_CLOSED; hci_disconn_cfm(conn, HCI_ERROR_LOCAL_HOST_TERM); - - /* Unlink before deleting otherwise it is possible that - * hci_conn_del removes the link which may cause the list to - * contain items already freed. - */ - hci_conn_unlink(conn); hci_conn_del(conn); } } -- cgit v1.2.3 From 6ca328e985cd995dfd1d5de44046e6074f853fbb Mon Sep 17 00:00:00 2001 From: Xin Long Date: Thu, 18 May 2023 16:03:00 -0400 Subject: sctp: fix an issue that plpmtu can never go to complete state When doing plpmtu probe, the probe size is growing every time when it receives the ACK during the Search state until the probe fails. When the failure occurs, pl.probe_high is set and it goes to the Complete state. However, if the link pmtu is huge, like 65535 in loopback_dev, the probe eventually keeps using SCTP_MAX_PLPMTU as the probe size and never fails. Because of that, pl.probe_high can not be set, and the plpmtu probe can never go to the Complete state. Fix it by setting pl.probe_high to SCTP_MAX_PLPMTU when the probe size grows to SCTP_MAX_PLPMTU in sctp_transport_pl_recv(). Also, not allow the probe size greater than SCTP_MAX_PLPMTU in the Complete state. Fixes: b87641aff9e7 ("sctp: do state transition when a probe succeeds on HB ACK recv path") Signed-off-by: Xin Long Signed-off-by: David S. Miller --- net/sctp/transport.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sctp/transport.c b/net/sctp/transport.c index 2f66a2006517..2abe45af98e7 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -324,9 +324,12 @@ bool sctp_transport_pl_recv(struct sctp_transport *t) t->pl.probe_size += SCTP_PL_BIG_STEP; } else if (t->pl.state == SCTP_PL_SEARCH) { if (!t->pl.probe_high) { - t->pl.probe_size = min(t->pl.probe_size + SCTP_PL_BIG_STEP, - SCTP_MAX_PLPMTU); - return false; + if (t->pl.probe_size < SCTP_MAX_PLPMTU) { + t->pl.probe_size = min(t->pl.probe_size + SCTP_PL_BIG_STEP, + SCTP_MAX_PLPMTU); + return false; + } + t->pl.probe_high = SCTP_MAX_PLPMTU; } t->pl.probe_size += SCTP_PL_MIN_STEP; if (t->pl.probe_size >= t->pl.probe_high) { @@ -341,7 +344,7 @@ bool sctp_transport_pl_recv(struct sctp_transport *t) } else if (t->pl.state == SCTP_PL_COMPLETE) { /* Raise probe_size again after 30 * interval in Search Complete */ t->pl.state = SCTP_PL_SEARCH; /* Search Complete -> Search */ - t->pl.probe_size += SCTP_PL_MIN_STEP; + t->pl.probe_size = min(t->pl.probe_size + SCTP_PL_MIN_STEP, SCTP_MAX_PLPMTU); } return t->pl.state == SCTP_PL_COMPLETE; -- cgit v1.2.3 From b21c7ba6d9a5532add3827a3b49f49cbc0cb9779 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 19 May 2023 13:12:50 -0400 Subject: net/handshake: Squelch allocation warning during Kunit test The "handshake_req_alloc excessive privsize" kunit test is intended to check what happens when the maximum privsize is exceeded. The WARN_ON_ONCE_GFP at mm/page_alloc.c:4744 can be disabled safely for this test. Reported-by: Linux Kernel Functional Testing Fixes: 88232ec1ec5e ("net/handshake: Add Kunit tests for the handshake consumer API") Signed-off-by: Chuck Lever Link: https://lore.kernel.org/r/168451636052.47152.9600443326570457947.stgit@oracle-102.nfsv4bat.org Signed-off-by: Jakub Kicinski --- net/handshake/handshake-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c index e6adc5dec11a..6193e46ee6d9 100644 --- a/net/handshake/handshake-test.c +++ b/net/handshake/handshake-test.c @@ -102,7 +102,7 @@ struct handshake_req_alloc_test_param handshake_req_alloc_params[] = { { .desc = "handshake_req_alloc excessive privsize", .proto = &handshake_req_alloc_proto_6, - .gfp = GFP_KERNEL, + .gfp = GFP_KERNEL | __GFP_NOWARN, .expect_success = false, }, { -- cgit v1.2.3 From 18c40a1cc1d990c51381ef48cd93fdb31d5cd903 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 19 May 2023 13:08:24 -0400 Subject: net/handshake: Fix sock->file allocation sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); ^^^^ ^^^^ sock_alloc_file() calls release_sock() on error but the left hand side of the assignment dereferences "sock". This isn't the bug and I didn't report this earlier because there is an assert that it doesn't fail. net/handshake/handshake-test.c:221 handshake_req_submit_test4() error: dereferencing freed memory 'sock' net/handshake/handshake-test.c:233 handshake_req_submit_test4() warn: 'req' was already freed. net/handshake/handshake-test.c:254 handshake_req_submit_test5() error: dereferencing freed memory 'sock' net/handshake/handshake-test.c:290 handshake_req_submit_test6() error: dereferencing freed memory 'sock' net/handshake/handshake-test.c:321 handshake_req_cancel_test1() error: dereferencing freed memory 'sock' net/handshake/handshake-test.c:355 handshake_req_cancel_test2() error: dereferencing freed memory 'sock' net/handshake/handshake-test.c:367 handshake_req_cancel_test2() warn: 'req' was already freed. net/handshake/handshake-test.c:395 handshake_req_cancel_test3() error: dereferencing freed memory 'sock' net/handshake/handshake-test.c:407 handshake_req_cancel_test3() warn: 'req' was already freed. net/handshake/handshake-test.c:451 handshake_req_destroy_test1() error: dereferencing freed memory 'sock' net/handshake/handshake-test.c:463 handshake_req_destroy_test1() warn: 'req' was already freed. Reported-by: Dan Carpenter Fixes: 88232ec1ec5e ("net/handshake: Add Kunit tests for the handshake consumer API") Signed-off-by: Chuck Lever Link: https://lore.kernel.org/r/168451609436.45209.15407022385441542980.stgit@oracle-102.nfsv4bat.org Signed-off-by: Jakub Kicinski --- net/handshake/handshake-test.c | 42 ++++++++++++++++++++++++++++-------------- 1 file changed, 28 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c index 6193e46ee6d9..6d37bab35c8f 100644 --- a/net/handshake/handshake-test.c +++ b/net/handshake/handshake-test.c @@ -209,6 +209,7 @@ static void handshake_req_submit_test4(struct kunit *test) { struct handshake_req *req, *result; struct socket *sock; + struct file *filp; int err; /* Arrange */ @@ -218,9 +219,10 @@ static void handshake_req_submit_test4(struct kunit *test) err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock, 1); KUNIT_ASSERT_EQ(test, err, 0); - sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + filp = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); KUNIT_ASSERT_NOT_NULL(test, sock->sk); + sock->file = filp; err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); @@ -241,6 +243,7 @@ static void handshake_req_submit_test5(struct kunit *test) struct handshake_req *req; struct handshake_net *hn; struct socket *sock; + struct file *filp; struct net *net; int saved, err; @@ -251,9 +254,10 @@ static void handshake_req_submit_test5(struct kunit *test) err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock, 1); KUNIT_ASSERT_EQ(test, err, 0); - sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + filp = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); KUNIT_ASSERT_NOT_NULL(test, sock->sk); + sock->file = filp; net = sock_net(sock->sk); hn = handshake_pernet(net); @@ -276,6 +280,7 @@ static void handshake_req_submit_test6(struct kunit *test) { struct handshake_req *req1, *req2; struct socket *sock; + struct file *filp; int err; /* Arrange */ @@ -287,9 +292,10 @@ static void handshake_req_submit_test6(struct kunit *test) err = __sock_create(&init_net, PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock, 1); KUNIT_ASSERT_EQ(test, err, 0); - sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + filp = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); KUNIT_ASSERT_NOT_NULL(test, sock->sk); + sock->file = filp; /* Act */ err = handshake_req_submit(sock, req1, GFP_KERNEL); @@ -307,6 +313,7 @@ static void handshake_req_cancel_test1(struct kunit *test) { struct handshake_req *req; struct socket *sock; + struct file *filp; bool result; int err; @@ -318,8 +325,9 @@ static void handshake_req_cancel_test1(struct kunit *test) &sock, 1); KUNIT_ASSERT_EQ(test, err, 0); - sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + filp = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); + sock->file = filp; err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); @@ -340,6 +348,7 @@ static void handshake_req_cancel_test2(struct kunit *test) struct handshake_req *req, *next; struct handshake_net *hn; struct socket *sock; + struct file *filp; struct net *net; bool result; int err; @@ -352,8 +361,9 @@ static void handshake_req_cancel_test2(struct kunit *test) &sock, 1); KUNIT_ASSERT_EQ(test, err, 0); - sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + filp = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); + sock->file = filp; err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); @@ -380,6 +390,7 @@ static void handshake_req_cancel_test3(struct kunit *test) struct handshake_req *req, *next; struct handshake_net *hn; struct socket *sock; + struct file *filp; struct net *net; bool result; int err; @@ -392,8 +403,9 @@ static void handshake_req_cancel_test3(struct kunit *test) &sock, 1); KUNIT_ASSERT_EQ(test, err, 0); - sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + filp = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); + sock->file = filp; err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); @@ -436,6 +448,7 @@ static void handshake_req_destroy_test1(struct kunit *test) { struct handshake_req *req; struct socket *sock; + struct file *filp; int err; /* Arrange */ @@ -448,8 +461,9 @@ static void handshake_req_destroy_test1(struct kunit *test) &sock, 1); KUNIT_ASSERT_EQ(test, err, 0); - sock->file = sock_alloc_file(sock, O_NONBLOCK, NULL); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, sock->file); + filp = sock_alloc_file(sock, O_NONBLOCK, NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); + sock->file = filp; err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); -- cgit v1.2.3 From 3632679d9e4f879f49949bb5b050e0de553e4739 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Mon, 22 May 2023 14:08:20 +0200 Subject: ipv{4,6}/raw: fix output xfrm lookup wrt protocol With a raw socket bound to IPPROTO_RAW (ie with hdrincl enabled), the protocol field of the flow structure, build by raw_sendmsg() / rawv6_sendmsg()), is set to IPPROTO_RAW. This breaks the ipsec policy lookup when some policies are defined with a protocol in the selector. For ipv6, the sin6_port field from 'struct sockaddr_in6' could be used to specify the protocol. Just accept all values for IPPROTO_RAW socket. For ipv4, the sin_port field of 'struct sockaddr_in' could not be used without breaking backward compatibility (the value of this field was never checked). Let's add a new kind of control message, so that the userland could specify which protocol is used. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") CC: stable@vger.kernel.org Signed-off-by: Nicolas Dichtel Link: https://lore.kernel.org/r/20230522120820.1319391-1-nicolas.dichtel@6wind.com Signed-off-by: Paolo Abeni --- include/net/ip.h | 2 ++ include/uapi/linux/in.h | 1 + net/ipv4/ip_sockglue.c | 12 +++++++++++- net/ipv4/raw.c | 5 ++++- net/ipv6/raw.c | 3 ++- 5 files changed, 20 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/net/ip.h b/include/net/ip.h index c3fffaa92d6e..acec504c469a 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -76,6 +76,7 @@ struct ipcm_cookie { __be32 addr; int oif; struct ip_options_rcu *opt; + __u8 protocol; __u8 ttl; __s16 tos; char priority; @@ -96,6 +97,7 @@ static inline void ipcm_init_sk(struct ipcm_cookie *ipcm, ipcm->sockc.tsflags = inet->sk.sk_tsflags; ipcm->oif = READ_ONCE(inet->sk.sk_bound_dev_if); ipcm->addr = inet->inet_saddr; + ipcm->protocol = inet->inet_num; } #define IPCB(skb) ((struct inet_skb_parm*)((skb)->cb)) diff --git a/include/uapi/linux/in.h b/include/uapi/linux/in.h index 4b7f2df66b99..e682ab628dfa 100644 --- a/include/uapi/linux/in.h +++ b/include/uapi/linux/in.h @@ -163,6 +163,7 @@ struct in_addr { #define IP_MULTICAST_ALL 49 #define IP_UNICAST_IF 50 #define IP_LOCAL_PORT_RANGE 51 +#define IP_PROTOCOL 52 #define MCAST_EXCLUDE 0 #define MCAST_INCLUDE 1 diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index b511ff0adc0a..8e97d8d4cc9d 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -317,7 +317,14 @@ int ip_cmsg_send(struct sock *sk, struct msghdr *msg, struct ipcm_cookie *ipc, ipc->tos = val; ipc->priority = rt_tos2priority(ipc->tos); break; - + case IP_PROTOCOL: + if (cmsg->cmsg_len != CMSG_LEN(sizeof(int))) + return -EINVAL; + val = *(int *)CMSG_DATA(cmsg); + if (val < 1 || val > 255) + return -EINVAL; + ipc->protocol = val; + break; default: return -EINVAL; } @@ -1761,6 +1768,9 @@ int do_ip_getsockopt(struct sock *sk, int level, int optname, case IP_LOCAL_PORT_RANGE: val = inet->local_port_range.hi << 16 | inet->local_port_range.lo; break; + case IP_PROTOCOL: + val = inet_sk(sk)->inet_num; + break; default: sockopt_release_sock(sk); return -ENOPROTOOPT; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index ff712bf2a98d..eadf1c9ef7e4 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -532,6 +532,9 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) } ipcm_init_sk(&ipc, inet); + /* Keep backward compat */ + if (hdrincl) + ipc.protocol = IPPROTO_RAW; if (msg->msg_controllen) { err = ip_cmsg_send(sk, msg, &ipc, false); @@ -599,7 +602,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) flowi4_init_output(&fl4, ipc.oif, ipc.sockc.mark, tos, RT_SCOPE_UNIVERSE, - hdrincl ? IPPROTO_RAW : sk->sk_protocol, + hdrincl ? ipc.protocol : sk->sk_protocol, inet_sk_flowi_flags(sk) | (hdrincl ? FLOWI_FLAG_KNOWN_NH : 0), daddr, saddr, 0, 0, sk->sk_uid); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 7d0adb612bdd..44ee7a2e72ac 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -793,7 +793,8 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) if (!proto) proto = inet->inet_num; - else if (proto != inet->inet_num) + else if (proto != inet->inet_num && + inet->inet_num != IPPROTO_RAW) return -EINVAL; if (proto > 255) -- cgit v1.2.3 From 78fa0d61d97a728d306b0c23d353c0e340756437 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Mon, 22 May 2023 19:56:05 -0700 Subject: bpf, sockmap: Pass skb ownership through read_skb The read_skb hook calls consume_skb() now, but this means that if the recv_actor program wants to use the skb it needs to inc the ref cnt so that the consume_skb() doesn't kfree the sk_buff. This is problematic because in some error cases under memory pressure we may need to linearize the sk_buff from sk_psock_skb_ingress_enqueue(). Then we get this, skb_linearize() __pskb_pull_tail() pskb_expand_head() BUG_ON(skb_shared(skb)) Because we incremented users refcnt from sk_psock_verdict_recv() we hit the bug on with refcnt > 1 and trip it. To fix lets simply pass ownership of the sk_buff through the skb_read call. Then we can drop the consume from read_skb handlers and assume the verdict recv does any required kfree. Bug found while testing in our CI which runs in VMs that hit memory constraints rather regularly. William tested TCP read_skb handlers. [ 106.536188] ------------[ cut here ]------------ [ 106.536197] kernel BUG at net/core/skbuff.c:1693! [ 106.536479] invalid opcode: 0000 [#1] PREEMPT SMP PTI [ 106.536726] CPU: 3 PID: 1495 Comm: curl Not tainted 5.19.0-rc5 #1 [ 106.537023] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ArchLinux 1.16.0-1 04/01/2014 [ 106.537467] RIP: 0010:pskb_expand_head+0x269/0x330 [ 106.538585] RSP: 0018:ffffc90000138b68 EFLAGS: 00010202 [ 106.538839] RAX: 000000000000003f RBX: ffff8881048940e8 RCX: 0000000000000a20 [ 106.539186] RDX: 0000000000000002 RSI: 0000000000000000 RDI: ffff8881048940e8 [ 106.539529] RBP: ffffc90000138be8 R08: 00000000e161fd1a R09: 0000000000000000 [ 106.539877] R10: 0000000000000018 R11: 0000000000000000 R12: ffff8881048940e8 [ 106.540222] R13: 0000000000000003 R14: 0000000000000000 R15: ffff8881048940e8 [ 106.540568] FS: 00007f277dde9f00(0000) GS:ffff88813bd80000(0000) knlGS:0000000000000000 [ 106.540954] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 106.541227] CR2: 00007f277eeede64 CR3: 000000000ad3e000 CR4: 00000000000006e0 [ 106.541569] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 106.541915] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 106.542255] Call Trace: [ 106.542383] [ 106.542487] __pskb_pull_tail+0x4b/0x3e0 [ 106.542681] skb_ensure_writable+0x85/0xa0 [ 106.542882] sk_skb_pull_data+0x18/0x20 [ 106.543084] bpf_prog_b517a65a242018b0_bpf_skskb_http_verdict+0x3a9/0x4aa9 [ 106.543536] ? migrate_disable+0x66/0x80 [ 106.543871] sk_psock_verdict_recv+0xe2/0x310 [ 106.544258] ? sk_psock_write_space+0x1f0/0x1f0 [ 106.544561] tcp_read_skb+0x7b/0x120 [ 106.544740] tcp_data_queue+0x904/0xee0 [ 106.544931] tcp_rcv_established+0x212/0x7c0 [ 106.545142] tcp_v4_do_rcv+0x174/0x2a0 [ 106.545326] tcp_v4_rcv+0xe70/0xf60 [ 106.545500] ip_protocol_deliver_rcu+0x48/0x290 [ 106.545744] ip_local_deliver_finish+0xa7/0x150 Fixes: 04919bed948dc ("tcp: Introduce tcp_read_skb()") Reported-by: William Findlay Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Tested-by: William Findlay Reviewed-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20230523025618.113937-2-john.fastabend@gmail.com --- net/core/skmsg.c | 2 -- net/ipv4/tcp.c | 1 - net/ipv4/udp.c | 7 ++----- net/unix/af_unix.c | 7 ++----- net/vmw_vsock/virtio_transport_common.c | 5 +---- 5 files changed, 5 insertions(+), 17 deletions(-) (limited to 'net') diff --git a/net/core/skmsg.c b/net/core/skmsg.c index f81883759d38..4a3dc8d27295 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -1183,8 +1183,6 @@ static int sk_psock_verdict_recv(struct sock *sk, struct sk_buff *skb) int ret = __SK_DROP; int len = skb->len; - skb_get(skb); - rcu_read_lock(); psock = sk_psock(sk); if (unlikely(!psock)) { diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4d6392c16b7a..e914e3446377 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1773,7 +1773,6 @@ int tcp_read_skb(struct sock *sk, skb_read_actor_t recv_actor) WARN_ON_ONCE(!skb_set_owner_sk_safe(skb, sk)); tcp_flags = TCP_SKB_CB(skb)->tcp_flags; used = recv_actor(sk, skb); - consume_skb(skb); if (used < 0) { if (!copied) copied = used; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index aa32afd871ee..9482def1f310 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1818,7 +1818,7 @@ EXPORT_SYMBOL(__skb_recv_udp); int udp_read_skb(struct sock *sk, skb_read_actor_t recv_actor) { struct sk_buff *skb; - int err, copied; + int err; try_again: skb = skb_recv_udp(sk, MSG_DONTWAIT, &err); @@ -1837,10 +1837,7 @@ try_again: } WARN_ON_ONCE(!skb_set_owner_sk_safe(skb, sk)); - copied = recv_actor(sk, skb); - kfree_skb(skb); - - return copied; + return recv_actor(sk, skb); } EXPORT_SYMBOL(udp_read_skb); diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index cc695c9f09ec..e7728b57a8c7 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2553,7 +2553,7 @@ static int unix_read_skb(struct sock *sk, skb_read_actor_t recv_actor) { struct unix_sock *u = unix_sk(sk); struct sk_buff *skb; - int err, copied; + int err; mutex_lock(&u->iolock); skb = skb_recv_datagram(sk, MSG_DONTWAIT, &err); @@ -2561,10 +2561,7 @@ static int unix_read_skb(struct sock *sk, skb_read_actor_t recv_actor) if (!skb) return err; - copied = recv_actor(sk, skb); - kfree_skb(skb); - - return copied; + return recv_actor(sk, skb); } /* diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index e4878551f140..b769fc258931 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -1441,7 +1441,6 @@ int virtio_transport_read_skb(struct vsock_sock *vsk, skb_read_actor_t recv_acto struct sock *sk = sk_vsock(vsk); struct sk_buff *skb; int off = 0; - int copied; int err; spin_lock_bh(&vvs->rx_lock); @@ -1454,9 +1453,7 @@ int virtio_transport_read_skb(struct vsock_sock *vsk, skb_read_actor_t recv_acto if (!skb) return err; - copied = recv_actor(sk, skb); - kfree_skb(skb); - return copied; + return recv_actor(sk, skb); } EXPORT_SYMBOL_GPL(virtio_transport_read_skb); -- cgit v1.2.3 From 29173d07f79883ac94f5570294f98af3d4287382 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Mon, 22 May 2023 19:56:06 -0700 Subject: bpf, sockmap: Convert schedule_work into delayed_work Sk_buffs are fed into sockmap verdict programs either from a strparser (when the user might want to decide how framing of skb is done by attaching another parser program) or directly through tcp_read_sock. The tcp_read_sock is the preferred method for performance when the BPF logic is a stream parser. The flow for Cilium's common use case with a stream parser is, tcp_read_sock() sk_psock_verdict_recv ret = bpf_prog_run_pin_on_cpu() sk_psock_verdict_apply(sock, skb, ret) // if system is under memory pressure or app is slow we may // need to queue skb. Do this queuing through ingress_skb and // then kick timer to wake up handler skb_queue_tail(ingress_skb, skb) schedule_work(work); The work queue is wired up to sk_psock_backlog(). This will then walk the ingress_skb skb list that holds our sk_buffs that could not be handled, but should be OK to run at some later point. However, its possible that the workqueue doing this work still hits an error when sending the skb. When this happens the skbuff is requeued on a temporary 'state' struct kept with the workqueue. This is necessary because its possible to partially send an skbuff before hitting an error and we need to know how and where to restart when the workqueue runs next. Now for the trouble, we don't rekick the workqueue. This can cause a stall where the skbuff we just cached on the state variable might never be sent. This happens when its the last packet in a flow and no further packets come along that would cause the system to kick the workqueue from that side. To fix we could do simple schedule_work(), but while under memory pressure it makes sense to back off some instead of continue to retry repeatedly. So instead to fix convert schedule_work to schedule_delayed_work and add backoff logic to reschedule from backlog queue on errors. Its not obvious though what a good backoff is so use '1'. To test we observed some flakes whil running NGINX compliance test with sockmap we attributed these failed test to this bug and subsequent issue. >From on list discussion. This commit bec217197b41("skmsg: Schedule psock work if the cached skb exists on the psock") was intended to address similar race, but had a couple cases it missed. Most obvious it only accounted for receiving traffic on the local socket so if redirecting into another socket we could still get an sk_buff stuck here. Next it missed the case where copied=0 in the recv() handler and then we wouldn't kick the scheduler. Also its sub-optimal to require userspace to kick the internal mechanisms of sockmap to wake it up and copy data to user. It results in an extra syscall and requires the app to actual handle the EAGAIN correctly. Fixes: 04919bed948dc ("tcp: Introduce tcp_read_skb()") Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Tested-by: William Findlay Reviewed-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20230523025618.113937-3-john.fastabend@gmail.com --- include/linux/skmsg.h | 2 +- net/core/skmsg.c | 21 ++++++++++++++------- net/core/sock_map.c | 3 ++- 3 files changed, 17 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h index 84f787416a54..904ff9a32ad6 100644 --- a/include/linux/skmsg.h +++ b/include/linux/skmsg.h @@ -105,7 +105,7 @@ struct sk_psock { struct proto *sk_proto; struct mutex work_mutex; struct sk_psock_work_state work_state; - struct work_struct work; + struct delayed_work work; struct rcu_work rwork; }; diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 4a3dc8d27295..0a9ee2acac0b 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -482,7 +482,7 @@ int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg, } out: if (psock->work_state.skb && copied > 0) - schedule_work(&psock->work); + schedule_delayed_work(&psock->work, 0); return copied; } EXPORT_SYMBOL_GPL(sk_msg_recvmsg); @@ -640,7 +640,8 @@ static void sk_psock_skb_state(struct sk_psock *psock, static void sk_psock_backlog(struct work_struct *work) { - struct sk_psock *psock = container_of(work, struct sk_psock, work); + struct delayed_work *dwork = to_delayed_work(work); + struct sk_psock *psock = container_of(dwork, struct sk_psock, work); struct sk_psock_work_state *state = &psock->work_state; struct sk_buff *skb = NULL; bool ingress; @@ -680,6 +681,12 @@ start: if (ret == -EAGAIN) { sk_psock_skb_state(psock, state, skb, len, off); + + /* Delay slightly to prioritize any + * other work that might be here. + */ + if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) + schedule_delayed_work(&psock->work, 1); goto end; } /* Hard errors break pipe and stop xmit. */ @@ -734,7 +741,7 @@ struct sk_psock *sk_psock_init(struct sock *sk, int node) INIT_LIST_HEAD(&psock->link); spin_lock_init(&psock->link_lock); - INIT_WORK(&psock->work, sk_psock_backlog); + INIT_DELAYED_WORK(&psock->work, sk_psock_backlog); mutex_init(&psock->work_mutex); INIT_LIST_HEAD(&psock->ingress_msg); spin_lock_init(&psock->ingress_lock); @@ -823,7 +830,7 @@ static void sk_psock_destroy(struct work_struct *work) sk_psock_done_strp(psock); - cancel_work_sync(&psock->work); + cancel_delayed_work_sync(&psock->work); mutex_destroy(&psock->work_mutex); psock_progs_drop(&psock->progs); @@ -938,7 +945,7 @@ static int sk_psock_skb_redirect(struct sk_psock *from, struct sk_buff *skb) } skb_queue_tail(&psock_other->ingress_skb, skb); - schedule_work(&psock_other->work); + schedule_delayed_work(&psock_other->work, 0); spin_unlock_bh(&psock_other->ingress_lock); return 0; } @@ -1018,7 +1025,7 @@ static int sk_psock_verdict_apply(struct sk_psock *psock, struct sk_buff *skb, spin_lock_bh(&psock->ingress_lock); if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) { skb_queue_tail(&psock->ingress_skb, skb); - schedule_work(&psock->work); + schedule_delayed_work(&psock->work, 0); err = 0; } spin_unlock_bh(&psock->ingress_lock); @@ -1049,7 +1056,7 @@ static void sk_psock_write_space(struct sock *sk) psock = sk_psock(sk); if (likely(psock)) { if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) - schedule_work(&psock->work); + schedule_delayed_work(&psock->work, 0); write_space = psock->saved_write_space; } rcu_read_unlock(); diff --git a/net/core/sock_map.c b/net/core/sock_map.c index 7c189c2e2fbf..00afb66cd095 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -1644,9 +1644,10 @@ void sock_map_close(struct sock *sk, long timeout) rcu_read_unlock(); sk_psock_stop(psock); release_sock(sk); - cancel_work_sync(&psock->work); + cancel_delayed_work_sync(&psock->work); sk_psock_put(sk, psock); } + /* Make sure we do not recurse. This is a bug. * Leak the socket instead of crashing on a stack overflow. */ -- cgit v1.2.3 From bce22552f92ea7c577f49839b8e8f7d29afaf880 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Mon, 22 May 2023 19:56:07 -0700 Subject: bpf, sockmap: Reschedule is now done through backlog Now that the backlog manages the reschedule() logic correctly we can drop the partial fix to reschedule from recvmsg hook. Rescheduling on recvmsg hook was added to address a corner case where we still had data in the backlog state but had nothing to kick it and reschedule the backlog worker to run and finish copying data out of the state. This had a couple limitations, first it required user space to kick it introducing an unnecessary EBUSY and retry. Second it only handled the ingress case and egress redirects would still be hung. With the correct fix, pushing the reschedule logic down to where the enomem error occurs we can drop this fix. Fixes: bec217197b412 ("skmsg: Schedule psock work if the cached skb exists on the psock") Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Reviewed-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20230523025618.113937-4-john.fastabend@gmail.com --- net/core/skmsg.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net') diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 0a9ee2acac0b..76ff15f8bb06 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -481,8 +481,6 @@ int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg, msg_rx = sk_psock_peek_msg(psock); } out: - if (psock->work_state.skb && copied > 0) - schedule_delayed_work(&psock->work, 0); return copied; } EXPORT_SYMBOL_GPL(sk_msg_recvmsg); -- cgit v1.2.3 From 405df89dd52cbcd69a3cd7d9a10d64de38f854b2 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Mon, 22 May 2023 19:56:08 -0700 Subject: bpf, sockmap: Improved check for empty queue We noticed some rare sk_buffs were stepping past the queue when system was under memory pressure. The general theory is to skip enqueueing sk_buffs when its not necessary which is the normal case with a system that is properly provisioned for the task, no memory pressure and enough cpu assigned. But, if we can't allocate memory due to an ENOMEM error when enqueueing the sk_buff into the sockmap receive queue we push it onto a delayed workqueue to retry later. When a new sk_buff is received we then check if that queue is empty. However, there is a problem with simply checking the queue length. When a sk_buff is being processed from the ingress queue but not yet on the sockmap msg receive queue its possible to also recv a sk_buff through normal path. It will check the ingress queue which is zero and then skip ahead of the pkt being processed. Previously we used sock lock from both contexts which made the problem harder to hit, but not impossible. To fix instead of popping the skb from the queue entirely we peek the skb from the queue and do the copy there. This ensures checks to the queue length are non-zero while skb is being processed. Then finally when the entire skb has been copied to user space queue or another socket we pop it off the queue. This way the queue length check allows bypassing the queue only after the list has been completely processed. To reproduce issue we run NGINX compliance test with sockmap running and observe some flakes in our testing that we attributed to this issue. Fixes: 04919bed948dc ("tcp: Introduce tcp_read_skb()") Suggested-by: Jakub Sitnicki Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Tested-by: William Findlay Reviewed-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20230523025618.113937-5-john.fastabend@gmail.com --- include/linux/skmsg.h | 1 - net/core/skmsg.c | 32 ++++++++------------------------ 2 files changed, 8 insertions(+), 25 deletions(-) (limited to 'net') diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h index 904ff9a32ad6..054d7911bfc9 100644 --- a/include/linux/skmsg.h +++ b/include/linux/skmsg.h @@ -71,7 +71,6 @@ struct sk_psock_link { }; struct sk_psock_work_state { - struct sk_buff *skb; u32 len; u32 off; }; diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 76ff15f8bb06..bcd45a99a3db 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -622,16 +622,12 @@ static int sk_psock_handle_skb(struct sk_psock *psock, struct sk_buff *skb, static void sk_psock_skb_state(struct sk_psock *psock, struct sk_psock_work_state *state, - struct sk_buff *skb, int len, int off) { spin_lock_bh(&psock->ingress_lock); if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) { - state->skb = skb; state->len = len; state->off = off; - } else { - sock_drop(psock->sk, skb); } spin_unlock_bh(&psock->ingress_lock); } @@ -642,23 +638,17 @@ static void sk_psock_backlog(struct work_struct *work) struct sk_psock *psock = container_of(dwork, struct sk_psock, work); struct sk_psock_work_state *state = &psock->work_state; struct sk_buff *skb = NULL; + u32 len = 0, off = 0; bool ingress; - u32 len, off; int ret; mutex_lock(&psock->work_mutex); - if (unlikely(state->skb)) { - spin_lock_bh(&psock->ingress_lock); - skb = state->skb; + if (unlikely(state->len)) { len = state->len; off = state->off; - state->skb = NULL; - spin_unlock_bh(&psock->ingress_lock); } - if (skb) - goto start; - while ((skb = skb_dequeue(&psock->ingress_skb))) { + while ((skb = skb_peek(&psock->ingress_skb))) { len = skb->len; off = 0; if (skb_bpf_strparser(skb)) { @@ -667,7 +657,6 @@ static void sk_psock_backlog(struct work_struct *work) off = stm->offset; len = stm->full_len; } -start: ingress = skb_bpf_ingress(skb); skb_bpf_redirect_clear(skb); do { @@ -677,8 +666,7 @@ start: len, ingress); if (ret <= 0) { if (ret == -EAGAIN) { - sk_psock_skb_state(psock, state, skb, - len, off); + sk_psock_skb_state(psock, state, len, off); /* Delay slightly to prioritize any * other work that might be here. @@ -690,15 +678,16 @@ start: /* Hard errors break pipe and stop xmit. */ sk_psock_report_error(psock, ret ? -ret : EPIPE); sk_psock_clear_state(psock, SK_PSOCK_TX_ENABLED); - sock_drop(psock->sk, skb); goto end; } off += ret; len -= ret; } while (len); - if (!ingress) + skb = skb_dequeue(&psock->ingress_skb); + if (!ingress) { kfree_skb(skb); + } } end: mutex_unlock(&psock->work_mutex); @@ -791,11 +780,6 @@ static void __sk_psock_zap_ingress(struct sk_psock *psock) skb_bpf_redirect_clear(skb); sock_drop(psock->sk, skb); } - kfree_skb(psock->work_state.skb); - /* We null the skb here to ensure that calls to sk_psock_backlog - * do not pick up the free'd skb. - */ - psock->work_state.skb = NULL; __sk_psock_purge_ingress_msg(psock); } @@ -814,7 +798,6 @@ void sk_psock_stop(struct sk_psock *psock) spin_lock_bh(&psock->ingress_lock); sk_psock_clear_state(psock, SK_PSOCK_TX_ENABLED); sk_psock_cork_free(psock); - __sk_psock_zap_ingress(psock); spin_unlock_bh(&psock->ingress_lock); } @@ -829,6 +812,7 @@ static void sk_psock_destroy(struct work_struct *work) sk_psock_done_strp(psock); cancel_delayed_work_sync(&psock->work); + __sk_psock_zap_ingress(psock); mutex_destroy(&psock->work_mutex); psock_progs_drop(&psock->progs); -- cgit v1.2.3 From 901546fd8f9ca4b5c481ce00928ab425ce9aacc0 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Mon, 22 May 2023 19:56:09 -0700 Subject: bpf, sockmap: Handle fin correctly The sockmap code is returning EAGAIN after a FIN packet is received and no more data is on the receive queue. Correct behavior is to return 0 to the user and the user can then close the socket. The EAGAIN causes many apps to retry which masks the problem. Eventually the socket is evicted from the sockmap because its released from sockmap sock free handling. The issue creates a delay and can cause some errors on application side. To fix this check on sk_msg_recvmsg side if length is zero and FIN flag is set then set return to zero. A selftest will be added to check this condition. Fixes: 04919bed948dc ("tcp: Introduce tcp_read_skb()") Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Tested-by: William Findlay Reviewed-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20230523025618.113937-6-john.fastabend@gmail.com --- net/ipv4/tcp_bpf.c | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) (limited to 'net') diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index 2e9547467edb..73c13642d47f 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -174,6 +174,24 @@ static int tcp_msg_wait_data(struct sock *sk, struct sk_psock *psock, return ret; } +static bool is_next_msg_fin(struct sk_psock *psock) +{ + struct scatterlist *sge; + struct sk_msg *msg_rx; + int i; + + msg_rx = sk_psock_peek_msg(psock); + i = msg_rx->sg.start; + sge = sk_msg_elem(msg_rx, i); + if (!sge->length) { + struct sk_buff *skb = msg_rx->skb; + + if (skb && TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN) + return true; + } + return false; +} + static int tcp_bpf_recvmsg_parser(struct sock *sk, struct msghdr *msg, size_t len, @@ -196,6 +214,19 @@ static int tcp_bpf_recvmsg_parser(struct sock *sk, lock_sock(sk); msg_bytes_ready: copied = sk_msg_recvmsg(sk, psock, msg, len, flags); + /* The typical case for EFAULT is the socket was gracefully + * shutdown with a FIN pkt. So check here the other case is + * some error on copy_page_to_iter which would be unexpected. + * On fin return correct return code to zero. + */ + if (copied == -EFAULT) { + bool is_fin = is_next_msg_fin(psock); + + if (is_fin) { + copied = 0; + goto out; + } + } if (!copied) { long timeo; int data; -- cgit v1.2.3 From ea444185a6bf7da4dd0df1598ee953e4f7174858 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Mon, 22 May 2023 19:56:10 -0700 Subject: bpf, sockmap: TCP data stall on recv before accept A common mechanism to put a TCP socket into the sockmap is to hook the BPF_SOCK_OPS_{ACTIVE_PASSIVE}_ESTABLISHED_CB event with a BPF program that can map the socket info to the correct BPF verdict parser. When the user adds the socket to the map the psock is created and the new ops are assigned to ensure the verdict program will 'see' the sk_buffs as they arrive. Part of this process hooks the sk_data_ready op with a BPF specific handler to wake up the BPF verdict program when data is ready to read. The logic is simple enough (posted here for easy reading) static void sk_psock_verdict_data_ready(struct sock *sk) { struct socket *sock = sk->sk_socket; if (unlikely(!sock || !sock->ops || !sock->ops->read_skb)) return; sock->ops->read_skb(sk, sk_psock_verdict_recv); } The oversight here is sk->sk_socket is not assigned until the application accepts() the new socket. However, its entirely ok for the peer application to do a connect() followed immediately by sends. The socket on the receiver is sitting on the backlog queue of the listening socket until its accepted and the data is queued up. If the peer never accepts the socket or is slow it will eventually hit data limits and rate limit the session. But, important for BPF sockmap hooks when this data is received TCP stack does the sk_data_ready() call but the read_skb() for this data is never called because sk_socket is missing. The data sits on the sk_receive_queue. Then once the socket is accepted if we never receive more data from the peer there will be no further sk_data_ready calls and all the data is still on the sk_receive_queue(). Then user calls recvmsg after accept() and for TCP sockets in sockmap we use the tcp_bpf_recvmsg_parser() handler. The handler checks for data in the sk_msg ingress queue expecting that the BPF program has already run from the sk_data_ready hook and enqueued the data as needed. So we are stuck. To fix do an unlikely check in recvmsg handler for data on the sk_receive_queue and if it exists wake up data_ready. We have the sock locked in both read_skb and recvmsg so should avoid having multiple runners. Fixes: 04919bed948dc ("tcp: Introduce tcp_read_skb()") Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Reviewed-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20230523025618.113937-7-john.fastabend@gmail.com --- net/ipv4/tcp_bpf.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'net') diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index 73c13642d47f..01dd76be1a58 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -212,6 +212,26 @@ static int tcp_bpf_recvmsg_parser(struct sock *sk, return tcp_recvmsg(sk, msg, len, flags, addr_len); lock_sock(sk); + + /* We may have received data on the sk_receive_queue pre-accept and + * then we can not use read_skb in this context because we haven't + * assigned a sk_socket yet so have no link to the ops. The work-around + * is to check the sk_receive_queue and in these cases read skbs off + * queue again. The read_skb hook is not running at this point because + * of lock_sock so we avoid having multiple runners in read_skb. + */ + if (unlikely(!skb_queue_empty(&sk->sk_receive_queue))) { + tcp_data_ready(sk); + /* This handles the ENOMEM errors if we both receive data + * pre accept and are already under memory pressure. At least + * let user know to retry. + */ + if (unlikely(!skb_queue_empty(&sk->sk_receive_queue))) { + copied = -EAGAIN; + goto out; + } + } + msg_bytes_ready: copied = sk_msg_recvmsg(sk, psock, msg, len, flags); /* The typical case for EFAULT is the socket was gracefully -- cgit v1.2.3 From 6df7f764cd3cf5a03a4a47b23be47e57e41fcd85 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Mon, 22 May 2023 19:56:11 -0700 Subject: bpf, sockmap: Wake up polling after data copy When TCP stack has data ready to read sk_data_ready() is called. Sockmap overwrites this with its own handler to call into BPF verdict program. But, the original TCP socket had sock_def_readable that would additionally wake up any user space waiters with sk_wake_async(). Sockmap saved the callback when the socket was created so call the saved data ready callback and then we can wake up any epoll() logic waiting on the read. Note we call on 'copied >= 0' to account for returning 0 when a FIN is received because we need to wake up user for this as well so they can do the recvmsg() -> 0 and detect the shutdown. Fixes: 04919bed948dc ("tcp: Introduce tcp_read_skb()") Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Reviewed-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20230523025618.113937-8-john.fastabend@gmail.com --- net/core/skmsg.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/skmsg.c b/net/core/skmsg.c index bcd45a99a3db..08be5f409fb8 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -1199,12 +1199,21 @@ out: static void sk_psock_verdict_data_ready(struct sock *sk) { struct socket *sock = sk->sk_socket; + int copied; trace_sk_data_ready(sk); if (unlikely(!sock || !sock->ops || !sock->ops->read_skb)) return; - sock->ops->read_skb(sk, sk_psock_verdict_recv); + copied = sock->ops->read_skb(sk, sk_psock_verdict_recv); + if (copied >= 0) { + struct sk_psock *psock; + + rcu_read_lock(); + psock = sk_psock(sk); + psock->saved_data_ready(sk); + rcu_read_unlock(); + } } void sk_psock_start_verdict(struct sock *sk, struct sk_psock *psock) -- cgit v1.2.3 From e5c6de5fa025882babf89cecbed80acf49b987fa Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Mon, 22 May 2023 19:56:12 -0700 Subject: bpf, sockmap: Incorrectly handling copied_seq The read_skb() logic is incrementing the tcp->copied_seq which is used for among other things calculating how many outstanding bytes can be read by the application. This results in application errors, if the application does an ioctl(FIONREAD) we return zero because this is calculated from the copied_seq value. To fix this we move tcp->copied_seq accounting into the recv handler so that we update these when the recvmsg() hook is called and data is in fact copied into user buffers. This gives an accurate FIONREAD value as expected and improves ACK handling. Before we were calling the tcp_rcv_space_adjust() which would update 'number of bytes copied to user in last RTT' which is wrong for programs returning SK_PASS. The bytes are only copied to the user when recvmsg is handled. Doing the fix for recvmsg is straightforward, but fixing redirect and SK_DROP pkts is a bit tricker. Build a tcp_psock_eat() helper and then call this from skmsg handlers. This fixes another issue where a broken socket with a BPF program doing a resubmit could hang the receiver. This happened because although read_skb() consumed the skb through sock_drop() it did not update the copied_seq. Now if a single reccv socket is redirecting to many sockets (for example for lb) the receiver sk will be hung even though we might expect it to continue. The hang comes from not updating the copied_seq numbers and memory pressure resulting from that. We have a slight layer problem of calling tcp_eat_skb even if its not a TCP socket. To fix we could refactor and create per type receiver handlers. I decided this is more work than we want in the fix and we already have some small tweaks depending on caller that use the helper skb_bpf_strparser(). So we extend that a bit and always set the strparser bit when it is in use and then we can gate the seq_copied updates on this. Fixes: 04919bed948dc ("tcp: Introduce tcp_read_skb()") Signed-off-by: John Fastabend Signed-off-by: Daniel Borkmann Reviewed-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20230523025618.113937-9-john.fastabend@gmail.com --- include/net/tcp.h | 10 ++++++++++ net/core/skmsg.c | 15 +++++++-------- net/ipv4/tcp.c | 10 +--------- net/ipv4/tcp_bpf.c | 28 +++++++++++++++++++++++++++- 4 files changed, 45 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 04a31643cda3..18a038d16434 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1470,6 +1470,8 @@ static inline void tcp_adjust_rcv_ssthresh(struct sock *sk) } void tcp_cleanup_rbuf(struct sock *sk, int copied); +void __tcp_cleanup_rbuf(struct sock *sk, int copied); + /* We provision sk_rcvbuf around 200% of sk_rcvlowat. * If 87.5 % (7/8) of the space has been consumed, we want to override @@ -2326,6 +2328,14 @@ int tcp_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore); void tcp_bpf_clone(const struct sock *sk, struct sock *newsk); #endif /* CONFIG_BPF_SYSCALL */ +#ifdef CONFIG_INET +void tcp_eat_skb(struct sock *sk, struct sk_buff *skb); +#else +static inline void tcp_eat_skb(struct sock *sk, struct sk_buff *skb) +{ +} +#endif + int tcp_bpf_sendmsg_redir(struct sock *sk, bool ingress, struct sk_msg *msg, u32 bytes, int flags); #endif /* CONFIG_NET_SOCK_MSG */ diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 08be5f409fb8..a9060e1f0e43 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -979,10 +979,8 @@ static int sk_psock_verdict_apply(struct sk_psock *psock, struct sk_buff *skb, err = -EIO; sk_other = psock->sk; if (sock_flag(sk_other, SOCK_DEAD) || - !sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) { - skb_bpf_redirect_clear(skb); + !sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED)) goto out_free; - } skb_bpf_set_ingress(skb); @@ -1011,18 +1009,19 @@ static int sk_psock_verdict_apply(struct sk_psock *psock, struct sk_buff *skb, err = 0; } spin_unlock_bh(&psock->ingress_lock); - if (err < 0) { - skb_bpf_redirect_clear(skb); + if (err < 0) goto out_free; - } } break; case __SK_REDIRECT: + tcp_eat_skb(psock->sk, skb); err = sk_psock_skb_redirect(psock, skb); break; case __SK_DROP: default: out_free: + skb_bpf_redirect_clear(skb); + tcp_eat_skb(psock->sk, skb); sock_drop(psock->sk, skb); } @@ -1067,8 +1066,7 @@ static void sk_psock_strp_read(struct strparser *strp, struct sk_buff *skb) skb_dst_drop(skb); skb_bpf_redirect_clear(skb); ret = bpf_prog_run_pin_on_cpu(prog, skb); - if (ret == SK_PASS) - skb_bpf_set_strparser(skb); + skb_bpf_set_strparser(skb); ret = sk_psock_map_verd(ret, skb_bpf_redirect_fetch(skb)); skb->sk = NULL; } @@ -1176,6 +1174,7 @@ static int sk_psock_verdict_recv(struct sock *sk, struct sk_buff *skb) psock = sk_psock(sk); if (unlikely(!psock)) { len = 0; + tcp_eat_skb(sk, skb); sock_drop(sk, skb); goto out; } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index e914e3446377..a60f6f4e7cd9 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1571,7 +1571,7 @@ static int tcp_peek_sndq(struct sock *sk, struct msghdr *msg, int len) * calculation of whether or not we must ACK for the sake of * a window update. */ -static void __tcp_cleanup_rbuf(struct sock *sk, int copied) +void __tcp_cleanup_rbuf(struct sock *sk, int copied) { struct tcp_sock *tp = tcp_sk(sk); bool time_to_ack = false; @@ -1786,14 +1786,6 @@ int tcp_read_skb(struct sock *sk, skb_read_actor_t recv_actor) break; } } - WRITE_ONCE(tp->copied_seq, seq); - - tcp_rcv_space_adjust(sk); - - /* Clean up data we have read: This will do ACK frames. */ - if (copied > 0) - __tcp_cleanup_rbuf(sk, copied); - return copied; } EXPORT_SYMBOL(tcp_read_skb); diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index 01dd76be1a58..5f93918c063c 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -11,6 +11,24 @@ #include #include +void tcp_eat_skb(struct sock *sk, struct sk_buff *skb) +{ + struct tcp_sock *tcp; + int copied; + + if (!skb || !skb->len || !sk_is_tcp(sk)) + return; + + if (skb_bpf_strparser(skb)) + return; + + tcp = tcp_sk(sk); + copied = tcp->copied_seq + skb->len; + WRITE_ONCE(tcp->copied_seq, copied); + tcp_rcv_space_adjust(sk); + __tcp_cleanup_rbuf(sk, skb->len); +} + static int bpf_tcp_ingress(struct sock *sk, struct sk_psock *psock, struct sk_msg *msg, u32 apply_bytes, int flags) { @@ -198,8 +216,10 @@ static int tcp_bpf_recvmsg_parser(struct sock *sk, int flags, int *addr_len) { + struct tcp_sock *tcp = tcp_sk(sk); + u32 seq = tcp->copied_seq; struct sk_psock *psock; - int copied; + int copied = 0; if (unlikely(flags & MSG_ERRQUEUE)) return inet_recv_error(sk, msg, len, addr_len); @@ -244,9 +264,11 @@ msg_bytes_ready: if (is_fin) { copied = 0; + seq++; goto out; } } + seq += copied; if (!copied) { long timeo; int data; @@ -284,6 +306,10 @@ msg_bytes_ready: copied = -EAGAIN; } out: + WRITE_ONCE(tcp->copied_seq, seq); + tcp_rcv_space_adjust(sk); + if (copied > 0) + __tcp_cleanup_rbuf(sk, copied); release_sock(sk); sk_psock_put(sk, psock); return copied; -- cgit v1.2.3 From 368d3cb406cdd074d1df2ad9ec06d1bfcb664882 Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Mon, 22 May 2023 11:17:14 +0800 Subject: page_pool: fix inconsistency for page_pool_ring_[un]lock() page_pool_ring_[un]lock() use in_softirq() to decide which spin lock variant to use, and when they are called in the context with in_softirq() being false, spin_lock_bh() is called in page_pool_ring_lock() while spin_unlock() is called in page_pool_ring_unlock(), because spin_lock_bh() has disabled the softirq in page_pool_ring_lock(), which causes inconsistency for spin lock pair calling. This patch fixes it by returning in_softirq state from page_pool_producer_lock(), and use it to decide which spin lock variant to use in page_pool_producer_unlock(). As pool->ring has both producer and consumer lock, so rename it to page_pool_producer_[un]lock() to reflect the actual usage. Also move them to page_pool.c as they are only used there, and remove the 'inline' as the compiler may have better idea to do inlining or not. Fixes: 7886244736a4 ("net: page_pool: Add bulk support for ptr_ring") Signed-off-by: Yunsheng Lin Acked-by: Jesper Dangaard Brouer Acked-by: Ilias Apalodimas Link: https://lore.kernel.org/r/20230522031714.5089-1-linyunsheng@huawei.com Signed-off-by: Jakub Kicinski --- include/net/page_pool.h | 18 ------------------ net/core/page_pool.c | 28 ++++++++++++++++++++++++++-- 2 files changed, 26 insertions(+), 20 deletions(-) (limited to 'net') diff --git a/include/net/page_pool.h b/include/net/page_pool.h index c8ec2f34722b..126f9e294389 100644 --- a/include/net/page_pool.h +++ b/include/net/page_pool.h @@ -399,22 +399,4 @@ static inline void page_pool_nid_changed(struct page_pool *pool, int new_nid) page_pool_update_nid(pool, new_nid); } -static inline void page_pool_ring_lock(struct page_pool *pool) - __acquires(&pool->ring.producer_lock) -{ - if (in_softirq()) - spin_lock(&pool->ring.producer_lock); - else - spin_lock_bh(&pool->ring.producer_lock); -} - -static inline void page_pool_ring_unlock(struct page_pool *pool) - __releases(&pool->ring.producer_lock) -{ - if (in_softirq()) - spin_unlock(&pool->ring.producer_lock); - else - spin_unlock_bh(&pool->ring.producer_lock); -} - #endif /* _NET_PAGE_POOL_H */ diff --git a/net/core/page_pool.c b/net/core/page_pool.c index e212e9d7edcb..a3e12a61d456 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -134,6 +134,29 @@ EXPORT_SYMBOL(page_pool_ethtool_stats_get); #define recycle_stat_add(pool, __stat, val) #endif +static bool page_pool_producer_lock(struct page_pool *pool) + __acquires(&pool->ring.producer_lock) +{ + bool in_softirq = in_softirq(); + + if (in_softirq) + spin_lock(&pool->ring.producer_lock); + else + spin_lock_bh(&pool->ring.producer_lock); + + return in_softirq; +} + +static void page_pool_producer_unlock(struct page_pool *pool, + bool in_softirq) + __releases(&pool->ring.producer_lock) +{ + if (in_softirq) + spin_unlock(&pool->ring.producer_lock); + else + spin_unlock_bh(&pool->ring.producer_lock); +} + static int page_pool_init(struct page_pool *pool, const struct page_pool_params *params) { @@ -617,6 +640,7 @@ void page_pool_put_page_bulk(struct page_pool *pool, void **data, int count) { int i, bulk_len = 0; + bool in_softirq; for (i = 0; i < count; i++) { struct page *page = virt_to_head_page(data[i]); @@ -635,7 +659,7 @@ void page_pool_put_page_bulk(struct page_pool *pool, void **data, return; /* Bulk producer into ptr_ring page_pool cache */ - page_pool_ring_lock(pool); + in_softirq = page_pool_producer_lock(pool); for (i = 0; i < bulk_len; i++) { if (__ptr_ring_produce(&pool->ring, data[i])) { /* ring full */ @@ -644,7 +668,7 @@ void page_pool_put_page_bulk(struct page_pool *pool, void **data, } } recycle_stat_add(pool, ring, i); - page_pool_ring_unlock(pool); + page_pool_producer_unlock(pool, in_softirq); /* Hopefully all pages was return into ptr_ring */ if (likely(i == bulk_len)) -- cgit v1.2.3 From 8a02fb71d7192ff1a9a47c9d937624966c6e09af Mon Sep 17 00:00:00 2001 From: Pratyush Yadav Date: Mon, 22 May 2023 17:30:20 +0200 Subject: net: fix skb leak in __skb_tstamp_tx() Commit 50749f2dd685 ("tcp/udp: Fix memleaks of sk and zerocopy skbs with TX timestamp.") added a call to skb_orphan_frags_rx() to fix leaks with zerocopy skbs. But it ended up adding a leak of its own. When skb_orphan_frags_rx() fails, the function just returns, leaking the skb it just cloned. Free it before returning. This bug was discovered and resolved using Coverity Static Analysis Security Testing (SAST) by Synopsys, Inc. Fixes: 50749f2dd685 ("tcp/udp: Fix memleaks of sk and zerocopy skbs with TX timestamp.") Signed-off-by: Pratyush Yadav Reviewed-by: Kuniyuki Iwashima Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/r/20230522153020.32422-1-ptyadav@amazon.de Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 515ec5cdc79c..cea28d30abb5 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -5224,8 +5224,10 @@ void __skb_tstamp_tx(struct sk_buff *orig_skb, } else { skb = skb_clone(orig_skb, GFP_ATOMIC); - if (skb_orphan_frags_rx(skb, GFP_ATOMIC)) + if (skb_orphan_frags_rx(skb, GFP_ATOMIC)) { + kfree_skb(skb); return; + } } if (!skb) return; -- cgit v1.2.3 From 878ecb0897f4737a4c9401f3523fd49589025671 Mon Sep 17 00:00:00 2001 From: Gavrilov Ilia Date: Tue, 23 May 2023 08:29:44 +0000 Subject: ipv6: Fix out-of-bounds access in ipv6_find_tlv() optlen is fetched without checking whether there is more than one byte to parse. It can lead to out-of-bounds access. Found by InfoTeCS on behalf of Linux Verification Center (linuxtesting.org) with SVACE. Fixes: c61a40432509 ("[IPV6]: Find option offset by type.") Signed-off-by: Gavrilov Ilia Reviewed-by: Jiri Pirko Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/exthdrs_core.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c index da46c4284676..49e31e4ae7b7 100644 --- a/net/ipv6/exthdrs_core.c +++ b/net/ipv6/exthdrs_core.c @@ -143,6 +143,8 @@ int ipv6_find_tlv(const struct sk_buff *skb, int offset, int type) optlen = 1; break; default: + if (len < 2) + goto bad; optlen = nh[offset + 1] + 2; if (optlen > len) goto bad; -- cgit v1.2.3 From a095326e2c0f33743ce8e887d5b90edf3f36cced Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 May 2023 11:47:09 -0400 Subject: net/handshake: Remove unneeded check from handshake_dup() handshake_req_submit() now verifies that the socket has a file. Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Reviewed-by: Simon Horman Signed-off-by: Chuck Lever Signed-off-by: Jakub Kicinski --- net/handshake/netlink.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'net') diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 35c9c445e0b8..7ec8a76c3c8a 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -99,9 +99,6 @@ static int handshake_dup(struct socket *sock) struct file *file; int newfd; - if (!sock->file) - return -EBADF; - file = get_file(sock->file); newfd = get_unused_fd_flags(O_CLOEXEC); if (newfd < 0) { -- cgit v1.2.3 From 7ea9c1ec66bc099b0bfba961a8a46dfe25d7d8e4 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 May 2023 11:47:40 -0400 Subject: net/handshake: Fix handshake_dup() ref counting If get_unused_fd_flags() fails, we ended up calling fput(sock->file) twice. Reported-by: Dan Carpenter Suggested-by: Paolo Abeni Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Signed-off-by: Jakub Kicinski --- net/handshake/netlink.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 7ec8a76c3c8a..f5dc170689d9 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -139,15 +139,16 @@ int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info) goto out_complete; } err = req->hr_proto->hp_accept(req, info, fd); - if (err) + if (err) { + fput(sock->file); goto out_complete; + } trace_handshake_cmd_accept(net, req, req->hr_sk, fd); return 0; out_complete: handshake_complete(req, -EIO, NULL); - fput(sock->file); out_status: trace_handshake_cmd_accept_err(net, req, NULL, err); return err; -- cgit v1.2.3 From 7afc6d0a107ffbd448c96eb2458b9e64a5af7860 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 May 2023 11:48:13 -0400 Subject: net/handshake: Fix uninitialized local variable trace_handshake_cmd_done_err() simply records the pointer in @req, so initializing it to NULL is sufficient and safe. Reported-by: Dan Carpenter Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Reviewed-by: Simon Horman Signed-off-by: Chuck Lever Signed-off-by: Jakub Kicinski --- net/handshake/netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index f5dc170689d9..16a4bde648ba 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -157,8 +157,8 @@ out_status: int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info) { struct net *net = sock_net(skb->sk); + struct handshake_req *req = NULL; struct socket *sock = NULL; - struct handshake_req *req; int fd, status, err; if (GENL_REQ_ATTR_CHECK(info, HANDSHAKE_A_DONE_SOCKFD)) -- cgit v1.2.3 From fc490880e39d86c65ab2bcbd357af1950fa55e48 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 May 2023 11:48:45 -0400 Subject: net/handshake: handshake_genl_notify() shouldn't ignore @flags Reported-by: Dan Carpenter Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Reviewed-by: Simon Horman Signed-off-by: Chuck Lever Signed-off-by: Jakub Kicinski --- net/handshake/netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 16a4bde648ba..1086653e1fad 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -48,7 +48,7 @@ int handshake_genl_notify(struct net *net, const struct handshake_proto *proto, proto->hp_handler_class)) return -ESRCH; - msg = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL); + msg = genlmsg_new(GENLMSG_DEFAULT_SIZE, flags); if (!msg) return -ENOMEM; -- cgit v1.2.3 From 1ce77c998f0415d7d9d91cb9bd7665e25c8f75f1 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 May 2023 11:49:17 -0400 Subject: net/handshake: Unpin sock->file if a handshake is cancelled If user space never calls DONE, sock->file's reference count remains elevated. Enable sock->file to be freed eventually in this case. Reported-by: Jakub Kacinski Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Signed-off-by: Jakub Kicinski --- net/handshake/handshake.h | 1 + net/handshake/request.c | 4 ++++ 2 files changed, 5 insertions(+) (limited to 'net') diff --git a/net/handshake/handshake.h b/net/handshake/handshake.h index 4dac965c99df..8aeaadca844f 100644 --- a/net/handshake/handshake.h +++ b/net/handshake/handshake.h @@ -31,6 +31,7 @@ struct handshake_req { struct list_head hr_list; struct rhash_head hr_rhash; unsigned long hr_flags; + struct file *hr_file; const struct handshake_proto *hr_proto; struct sock *hr_sk; void (*hr_odestruct)(struct sock *sk); diff --git a/net/handshake/request.c b/net/handshake/request.c index 94d5cef3e048..d78d41abb3d9 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -239,6 +239,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, } req->hr_odestruct = req->hr_sk->sk_destruct; req->hr_sk->sk_destruct = handshake_sk_destruct; + req->hr_file = sock->file; ret = -EOPNOTSUPP; net = sock_net(req->hr_sk); @@ -334,6 +335,9 @@ bool handshake_req_cancel(struct sock *sk) return false; } + /* Request accepted and waiting for DONE */ + fput(req->hr_file); + out_true: trace_handshake_cancel(net, req, sk); -- cgit v1.2.3 From 26fb5480a27d34975cc2b680b77af189620dd740 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 11 May 2023 11:49:50 -0400 Subject: net/handshake: Enable the SNI extension to work properly Enable the upper layer protocol to specify the SNI peername. This avoids the need for tlshd to use a DNS lookup, which can return a hostname that doesn't match the incoming certificate's SubjectName. Fixes: 2fd5532044a8 ("net/handshake: Add a kernel API for requesting a TLSv1.3 handshake") Reviewed-by: Simon Horman Signed-off-by: Chuck Lever Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/handshake.yaml | 4 ++++ Documentation/networking/tls-handshake.rst | 5 +++++ include/net/handshake.h | 1 + include/uapi/linux/handshake.h | 1 + net/handshake/tlshd.c | 8 ++++++++ 5 files changed, 19 insertions(+) (limited to 'net') diff --git a/Documentation/netlink/specs/handshake.yaml b/Documentation/netlink/specs/handshake.yaml index 614f1a585511..6d89e30f5fd5 100644 --- a/Documentation/netlink/specs/handshake.yaml +++ b/Documentation/netlink/specs/handshake.yaml @@ -68,6 +68,9 @@ attribute-sets: type: nest nested-attributes: x509 multi-attr: true + - + name: peername + type: string - name: done attributes: @@ -105,6 +108,7 @@ operations: - auth-mode - peer-identity - certificate + - peername - name: done doc: Handler reports handshake completion diff --git a/Documentation/networking/tls-handshake.rst b/Documentation/networking/tls-handshake.rst index a2817a88e905..6f5ea1646a47 100644 --- a/Documentation/networking/tls-handshake.rst +++ b/Documentation/networking/tls-handshake.rst @@ -53,6 +53,7 @@ fills in a structure that contains the parameters of the request: struct socket *ta_sock; tls_done_func_t ta_done; void *ta_data; + const char *ta_peername; unsigned int ta_timeout_ms; key_serial_t ta_keyring; key_serial_t ta_my_cert; @@ -71,6 +72,10 @@ instantiated a struct file in sock->file. has completed. Further explanation of this function is in the "Handshake Completion" sesction below. +The consumer can provide a NUL-terminated hostname in the @ta_peername +field that is sent as part of ClientHello. If no peername is provided, +the DNS hostname associated with the server's IP address is used instead. + The consumer can fill in the @ta_timeout_ms field to force the servicing handshake agent to exit after a number of milliseconds. This enables the socket to be fully closed once both the kernel and the handshake agent diff --git a/include/net/handshake.h b/include/net/handshake.h index 3352b1ab43b3..2e26e436e85f 100644 --- a/include/net/handshake.h +++ b/include/net/handshake.h @@ -24,6 +24,7 @@ struct tls_handshake_args { struct socket *ta_sock; tls_done_func_t ta_done; void *ta_data; + const char *ta_peername; unsigned int ta_timeout_ms; key_serial_t ta_keyring; key_serial_t ta_my_cert; diff --git a/include/uapi/linux/handshake.h b/include/uapi/linux/handshake.h index 1de4d0b95325..3d7ea58778c9 100644 --- a/include/uapi/linux/handshake.h +++ b/include/uapi/linux/handshake.h @@ -44,6 +44,7 @@ enum { HANDSHAKE_A_ACCEPT_AUTH_MODE, HANDSHAKE_A_ACCEPT_PEER_IDENTITY, HANDSHAKE_A_ACCEPT_CERTIFICATE, + HANDSHAKE_A_ACCEPT_PEERNAME, __HANDSHAKE_A_ACCEPT_MAX, HANDSHAKE_A_ACCEPT_MAX = (__HANDSHAKE_A_ACCEPT_MAX - 1) diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c index fcbeb63b4eb1..b735f5cced2f 100644 --- a/net/handshake/tlshd.c +++ b/net/handshake/tlshd.c @@ -31,6 +31,7 @@ struct tls_handshake_req { int th_type; unsigned int th_timeout_ms; int th_auth_mode; + const char *th_peername; key_serial_t th_keyring; key_serial_t th_certificate; key_serial_t th_privkey; @@ -48,6 +49,7 @@ tls_handshake_req_init(struct handshake_req *req, treq->th_timeout_ms = args->ta_timeout_ms; treq->th_consumer_done = args->ta_done; treq->th_consumer_data = args->ta_data; + treq->th_peername = args->ta_peername; treq->th_keyring = args->ta_keyring; treq->th_num_peerids = 0; treq->th_certificate = TLS_NO_CERT; @@ -214,6 +216,12 @@ static int tls_handshake_accept(struct handshake_req *req, ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_MESSAGE_TYPE, treq->th_type); if (ret < 0) goto out_cancel; + if (treq->th_peername) { + ret = nla_put_string(msg, HANDSHAKE_A_ACCEPT_PEERNAME, + treq->th_peername); + if (ret < 0) + goto out_cancel; + } if (treq->th_timeout_ms) { ret = nla_put_u32(msg, HANDSHAKE_A_ACCEPT_TIMEOUT, treq->th_timeout_ms); if (ret < 0) -- cgit v1.2.3 From ad42a35bdfc6d3c0fc4cb4027d7b2757ce665665 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 23 May 2023 09:33:05 -0700 Subject: udplite: Fix NULL pointer dereference in __sk_mem_raise_allocated(). syzbot reported [0] a null-ptr-deref in sk_get_rmem0() while using IPPROTO_UDPLITE (0x88): 14:25:52 executing program 1: r0 = socket$inet6(0xa, 0x80002, 0x88) We had a similar report [1] for probably sk_memory_allocated_add() in __sk_mem_raise_allocated(), and commit c915fe13cbaa ("udplite: fix NULL pointer dereference") fixed it by setting .memory_allocated for udplite_prot and udplitev6_prot. To fix the variant, we need to set either .sysctl_wmem_offset or .sysctl_rmem. Now UDP and UDPLITE share the same value for .memory_allocated, so we use the same .sysctl_wmem_offset for UDP and UDPLITE. [0]: general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] CPU: 0 PID: 6829 Comm: syz-executor.1 Not tainted 6.4.0-rc2-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/28/2023 RIP: 0010:sk_get_rmem0 include/net/sock.h:2907 [inline] RIP: 0010:__sk_mem_raise_allocated+0x806/0x17a0 net/core/sock.c:3006 Code: c1 ea 03 80 3c 02 00 0f 85 23 0f 00 00 48 8b 44 24 08 48 8b 98 38 01 00 00 48 b8 00 00 00 00 00 fc ff df 48 89 da 48 c1 ea 03 <0f> b6 14 02 48 89 d8 83 e0 07 83 c0 03 38 d0 0f 8d 6f 0a 00 00 8b RSP: 0018:ffffc90005d7f450 EFLAGS: 00010246 RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffc90004d92000 RDX: 0000000000000000 RSI: ffffffff88066482 RDI: ffffffff8e2ccbb8 RBP: ffff8880173f7000 R08: 0000000000000005 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000030000 R13: 0000000000000001 R14: 0000000000000340 R15: 0000000000000001 FS: 0000000000000000(0000) GS:ffff8880b9800000(0063) knlGS:00000000f7f1cb40 CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033 CR2: 000000002e82f000 CR3: 0000000034ff0000 CR4: 00000000003506f0 Call Trace: __sk_mem_schedule+0x6c/0xe0 net/core/sock.c:3077 udp_rmem_schedule net/ipv4/udp.c:1539 [inline] __udp_enqueue_schedule_skb+0x776/0xb30 net/ipv4/udp.c:1581 __udpv6_queue_rcv_skb net/ipv6/udp.c:666 [inline] udpv6_queue_rcv_one_skb+0xc39/0x16c0 net/ipv6/udp.c:775 udpv6_queue_rcv_skb+0x194/0xa10 net/ipv6/udp.c:793 __udp6_lib_mcast_deliver net/ipv6/udp.c:906 [inline] __udp6_lib_rcv+0x1bda/0x2bd0 net/ipv6/udp.c:1013 ip6_protocol_deliver_rcu+0x2e7/0x1250 net/ipv6/ip6_input.c:437 ip6_input_finish+0x150/0x2f0 net/ipv6/ip6_input.c:482 NF_HOOK include/linux/netfilter.h:303 [inline] NF_HOOK include/linux/netfilter.h:297 [inline] ip6_input+0xa0/0xd0 net/ipv6/ip6_input.c:491 ip6_mc_input+0x40b/0xf50 net/ipv6/ip6_input.c:585 dst_input include/net/dst.h:468 [inline] ip6_rcv_finish net/ipv6/ip6_input.c:79 [inline] NF_HOOK include/linux/netfilter.h:303 [inline] NF_HOOK include/linux/netfilter.h:297 [inline] ipv6_rcv+0x250/0x380 net/ipv6/ip6_input.c:309 __netif_receive_skb_one_core+0x114/0x180 net/core/dev.c:5491 __netif_receive_skb+0x1f/0x1c0 net/core/dev.c:5605 netif_receive_skb_internal net/core/dev.c:5691 [inline] netif_receive_skb+0x133/0x7a0 net/core/dev.c:5750 tun_rx_batched+0x4b3/0x7a0 drivers/net/tun.c:1553 tun_get_user+0x2452/0x39c0 drivers/net/tun.c:1989 tun_chr_write_iter+0xdf/0x200 drivers/net/tun.c:2035 call_write_iter include/linux/fs.h:1868 [inline] new_sync_write fs/read_write.c:491 [inline] vfs_write+0x945/0xd50 fs/read_write.c:584 ksys_write+0x12b/0x250 fs/read_write.c:637 do_syscall_32_irqs_on arch/x86/entry/common.c:112 [inline] __do_fast_syscall_32+0x65/0xf0 arch/x86/entry/common.c:178 do_fast_syscall_32+0x33/0x70 arch/x86/entry/common.c:203 entry_SYSENTER_compat_after_hwframe+0x70/0x82 RIP: 0023:0xf7f21579 Code: b8 01 10 06 03 74 b4 01 10 07 03 74 b0 01 10 08 03 74 d8 01 00 00 00 00 00 00 00 00 00 00 00 00 00 51 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 8d b4 26 00 00 00 00 8d b4 26 00 00 00 00 RSP: 002b:00000000f7f1c590 EFLAGS: 00000282 ORIG_RAX: 0000000000000004 RAX: ffffffffffffffda RBX: 00000000000000c8 RCX: 0000000020000040 RDX: 0000000000000083 RSI: 00000000f734e000 RDI: 0000000000000000 RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000296 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 Modules linked in: Link: https://lore.kernel.org/netdev/CANaxB-yCk8hhP68L4Q2nFOJht8sqgXGGQO2AftpHs0u1xyGG5A@mail.gmail.com/ [1] Fixes: 850cbaddb52d ("udp: use it's own memory accounting schema") Reported-by: syzbot+444ca0907e96f7c5e48b@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=444ca0907e96f7c5e48b Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20230523163305.66466-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni --- net/ipv4/udplite.c | 2 ++ net/ipv6/udplite.c | 2 ++ 2 files changed, 4 insertions(+) (limited to 'net') diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c index e0c9cc39b81e..56d94d23b9e0 100644 --- a/net/ipv4/udplite.c +++ b/net/ipv4/udplite.c @@ -64,6 +64,8 @@ struct proto udplite_prot = { .per_cpu_fw_alloc = &udp_memory_per_cpu_fw_alloc, .sysctl_mem = sysctl_udp_mem, + .sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_udp_wmem_min), + .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min), .obj_size = sizeof(struct udp_sock), .h.udp_table = &udplite_table, }; diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index 67eaf3ca14ce..3bab0cc13697 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -60,6 +60,8 @@ struct proto udplitev6_prot = { .per_cpu_fw_alloc = &udp_memory_per_cpu_fw_alloc, .sysctl_mem = sysctl_udp_mem, + .sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_udp_wmem_min), + .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min), .obj_size = sizeof(struct udp6_sock), .h.udp_table = &udplite_table, }; -- cgit v1.2.3