From 7c8903f6373f9abecf060bad53ca36bc4ac037f2 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 14 Feb 2006 13:53:03 -0800 Subject: [PATCH] jbd: revert checkpoint list changes This patch reverts commit f93ea411b73594f7d144855fd34278bcf34a9afc: [PATCH] jbd: split checkpoint lists This broke journal_flush() for OCFS2, which is its method of being sure that metadata is sent to disk for another node. And two related commits 8d3c7fce2d20ecc3264c8d8c91ae3beacdeaed1b and 43c3e6f5abdf6acac9b90c86bf03f995bf7d3d92 with the subjects: [PATCH] jbd: log_do_checkpoint fix [PATCH] jbd: remove_transaction fix These seem to be incremental bugfixes on the original patch and as such are no longer needed. Signed-off-by: Mark Fasheh Cc: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/jbd.h | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jbd.h b/include/linux/jbd.h index 0fe4aa891ddc..41ee79962bb2 100644 --- a/include/linux/jbd.h +++ b/include/linux/jbd.h @@ -497,12 +497,6 @@ struct transaction_s */ struct journal_head *t_checkpoint_list; - /* - * Doubly-linked circular list of all buffers submitted for IO while - * checkpointing. [j_list_lock] - */ - struct journal_head *t_checkpoint_io_list; - /* * Doubly-linked circular list of temporary buffers currently undergoing * IO in the log [j_list_lock] @@ -852,7 +846,7 @@ extern void journal_commit_transaction(journal_t *); /* Checkpoint list management */ int __journal_clean_checkpoint_list(journal_t *journal); -int __journal_remove_checkpoint(struct journal_head *); +void __journal_remove_checkpoint(struct journal_head *); void __journal_insert_checkpoint(struct journal_head *, transaction_t *); /* Buffer IO */ -- cgit v1.2.3 From 5ac5f9d1ce8492163dbde5d357dc5d03becf7e36 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 14 Feb 2006 13:53:04 -0800 Subject: [PATCH] NLM: Fix the NLM_GRANTED callback checks If 2 threads attached to the same process are blocking on different locks on different files (maybe even on different servers) but have the same lock arguments (i.e. same offset+length - actually quite common, since most processes try to lock the entire file) then the first GRANTED call that wakes one up will also wake the other. Currently when the NLM_GRANTED callback comes in, lockd walks the list of blocked locks in search of a match to the lock that the NLM server has granted. Although it checks the lock pid, start and end, it fails to check the filehandle and the server address. By checking the filehandle and server IP address, we ensure that this only happens if the locks truly are referencing the same file. Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/lockd/clntlock.c | 27 +++++++++++++++++---------- fs/lockd/svc4proc.c | 2 +- fs/lockd/svcproc.c | 2 +- include/linux/lockd/lockd.h | 6 +++--- 4 files changed, 22 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index 3eaf6e701087..da6354baa0b8 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -111,9 +111,10 @@ long nlmclnt_block(struct nlm_rqst *req, long timeout) /* * The server lockd has called us back to tell us the lock was granted */ -u32 -nlmclnt_grant(struct nlm_lock *lock) +u32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *lock) { + const struct file_lock *fl = &lock->fl; + const struct nfs_fh *fh = &lock->fh; struct nlm_wait *block; u32 res = nlm_lck_denied; @@ -122,14 +123,20 @@ nlmclnt_grant(struct nlm_lock *lock) * Warning: must not use cookie to match it! */ list_for_each_entry(block, &nlm_blocked, b_list) { - if (nlm_compare_locks(block->b_lock, &lock->fl)) { - /* Alright, we found a lock. Set the return status - * and wake up the caller - */ - block->b_status = NLM_LCK_GRANTED; - wake_up(&block->b_wait); - res = nlm_granted; - } + struct file_lock *fl_blocked = block->b_lock; + + if (!nlm_compare_locks(fl_blocked, fl)) + continue; + if (!nlm_cmp_addr(&block->b_host->h_addr, addr)) + continue; + if (nfs_compare_fh(NFS_FH(fl_blocked->fl_file->f_dentry->d_inode) ,fh) != 0) + continue; + /* Alright, we found a lock. Set the return status + * and wake up the caller + */ + block->b_status = NLM_LCK_GRANTED; + wake_up(&block->b_wait); + res = nlm_granted; } return res; } diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index 4063095d849e..b10f913aa06a 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -228,7 +228,7 @@ nlm4svc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp, resp->cookie = argp->cookie; dprintk("lockd: GRANTED called\n"); - resp->status = nlmclnt_grant(&argp->lock); + resp->status = nlmclnt_grant(&rqstp->rq_addr, &argp->lock); dprintk("lockd: GRANTED status %d\n", ntohl(resp->status)); return rpc_success; } diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 3bc437e0cf5b..35681d9cf1fc 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -256,7 +256,7 @@ nlmsvc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp, resp->cookie = argp->cookie; dprintk("lockd: GRANTED called\n"); - resp->status = nlmclnt_grant(&argp->lock); + resp->status = nlmclnt_grant(&rqstp->rq_addr, &argp->lock); dprintk("lockd: GRANTED status %d\n", ntohl(resp->status)); return rpc_success; } diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h index 920766cea79c..ef21ed296039 100644 --- a/include/linux/lockd/lockd.h +++ b/include/linux/lockd/lockd.h @@ -149,7 +149,7 @@ struct nlm_rqst * nlmclnt_alloc_call(void); int nlmclnt_prepare_block(struct nlm_rqst *req, struct nlm_host *host, struct file_lock *fl); void nlmclnt_finish_block(struct nlm_rqst *req); long nlmclnt_block(struct nlm_rqst *req, long timeout); -u32 nlmclnt_grant(struct nlm_lock *); +u32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *); void nlmclnt_recovery(struct nlm_host *, u32); int nlmclnt_reclaim(struct nlm_host *, struct file_lock *); int nlmclnt_setgrantargs(struct nlm_rqst *, struct nlm_lock *); @@ -204,7 +204,7 @@ nlmsvc_file_inode(struct nlm_file *file) * Compare two host addresses (needs modifying for ipv6) */ static __inline__ int -nlm_cmp_addr(struct sockaddr_in *sin1, struct sockaddr_in *sin2) +nlm_cmp_addr(const struct sockaddr_in *sin1, const struct sockaddr_in *sin2) { return sin1->sin_addr.s_addr == sin2->sin_addr.s_addr; } @@ -214,7 +214,7 @@ nlm_cmp_addr(struct sockaddr_in *sin1, struct sockaddr_in *sin2) * When the second lock is of type F_UNLCK, this acts like a wildcard. */ static __inline__ int -nlm_compare_locks(struct file_lock *fl1, struct file_lock *fl2) +nlm_compare_locks(const struct file_lock *fl1, const struct file_lock *fl2) { return fl1->fl_pid == fl2->fl_pid && fl1->fl_start == fl2->fl_start -- cgit v1.2.3 From d6077cb80cde4506720f9165eba99ee07438513f Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Tue, 14 Feb 2006 13:53:10 -0800 Subject: [PATCH] sched: revert "filter affine wakeups" Revert commit d7102e95b7b9c00277562c29aad421d2d521c5f6: [PATCH] sched: filter affine wakeups Apparently caused more than 10% performance regression for aim7 benchmark. The setup in use is 16-cpu HP rx8620, 64Gb of memory and 12 MSA1000s with 144 disks. Each disk is 72Gb with a single ext3 filesystem (courtesy of HP, who supplied benchmark results). The problem is, for aim7, the wake-up pattern is random, but it still needs load balancing action in the wake-up path to achieve best performance. With the above commit, lack of load balancing hurts that workload. However, for workloads like database transaction processing, the requirement is exactly opposite. In the wake up path, best performance is achieved with absolutely zero load balancing. We simply wake up the process on the CPU that it was previously run. Worst performance is obtained when we do load balancing at wake up. There isn't an easy way to auto detect the workload characteristics. Ingo's earlier patch that detects idle CPU and decide whether to load balance or not doesn't perform with aim7 either since all CPUs are busy (it causes even bigger perf. regression). Revert commit d7102e95b7b9c00277562c29aad421d2d521c5f6, which causes more than 10% performance regression with aim7. Signed-off-by: Ken Chen Acked-by: Ingo Molnar Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 5 +---- kernel/sched.c | 10 +--------- 2 files changed, 2 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 9c1da0269a18..b6f51e3a38ec 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -697,11 +697,8 @@ struct task_struct { int lock_depth; /* BKL lock depth */ -#if defined(CONFIG_SMP) - int last_waker_cpu; /* CPU that last woke this task up */ -#if defined(__ARCH_WANT_UNLOCKED_CTXSW) +#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) int oncpu; -#endif #endif int prio, static_prio; struct list_head run_list; diff --git a/kernel/sched.c b/kernel/sched.c index 87d93be336a1..66d957227de9 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1204,9 +1204,6 @@ static int try_to_wake_up(task_t *p, unsigned int state, int sync) } } - if (p->last_waker_cpu != this_cpu) - goto out_set_cpu; - if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed))) goto out_set_cpu; @@ -1277,8 +1274,6 @@ out_set_cpu: cpu = task_cpu(p); } - p->last_waker_cpu = this_cpu; - out_activate: #endif /* CONFIG_SMP */ if (old_state == TASK_UNINTERRUPTIBLE) { @@ -1360,12 +1355,9 @@ void fastcall sched_fork(task_t *p, int clone_flags) #ifdef CONFIG_SCHEDSTATS memset(&p->sched_info, 0, sizeof(p->sched_info)); #endif -#if defined(CONFIG_SMP) - p->last_waker_cpu = cpu; -#if defined(__ARCH_WANT_UNLOCKED_CTXSW) +#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) p->oncpu = 0; #endif -#endif #ifdef CONFIG_PREEMPT /* Want to start with kernel preemption disabled. */ task_thread_info(p)->preempt_count = 1; -- cgit v1.2.3 From ee68cea2c26b7a8222f9020f54d22c6067011e8b Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 15 Feb 2006 01:34:23 -0800 Subject: [NETFILTER]: Fix xfrm lookup after SNAT To find out if a packet needs to be handled by IPsec after SNAT, packets are currently rerouted in POST_ROUTING and a new xfrm lookup is done. This breaks SNAT of non-unicast packets to non-local addresses because the packet is routed as incoming packet and no neighbour entry is bound to the dst_entry. In general, it seems to be a bad idea to replace the dst_entry after the packet was already sent to the output routine because its state might not match what's expected. This patch changes the xfrm lookup in POST_ROUTING to re-use the original dst_entry without routing the packet again. This means no policy routing can be used for transport mode transforms (which keep the original route) when packets are SNATed to match the policy, but it looks like the best we can do for now. Signed-off-by: Patrick McHardy Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4.h | 2 +- net/ipv4/netfilter.c | 41 ++++++++++++++++++++++++++++++++++ net/ipv4/netfilter/ip_nat_standalone.c | 6 ++--- 3 files changed, 45 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h index fdc4a9527343..43c09d790b83 100644 --- a/include/linux/netfilter_ipv4.h +++ b/include/linux/netfilter_ipv4.h @@ -79,7 +79,7 @@ enum nf_ip_hook_priorities { #ifdef __KERNEL__ extern int ip_route_me_harder(struct sk_buff **pskb); - +extern int ip_xfrm_me_harder(struct sk_buff **pskb); #endif /*__KERNEL__*/ #endif /*__LINUX_IP_NETFILTER_H*/ diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index 52a3d7c57907..ed42cdc57cd9 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -78,6 +78,47 @@ int ip_route_me_harder(struct sk_buff **pskb) } EXPORT_SYMBOL(ip_route_me_harder); +#ifdef CONFIG_XFRM +int ip_xfrm_me_harder(struct sk_buff **pskb) +{ + struct flowi fl; + unsigned int hh_len; + struct dst_entry *dst; + + if (IPCB(*pskb)->flags & IPSKB_XFRM_TRANSFORMED) + return 0; + if (xfrm_decode_session(*pskb, &fl, AF_INET) < 0) + return -1; + + dst = (*pskb)->dst; + if (dst->xfrm) + dst = ((struct xfrm_dst *)dst)->route; + dst_hold(dst); + + if (xfrm_lookup(&dst, &fl, (*pskb)->sk, 0) < 0) + return -1; + + dst_release((*pskb)->dst); + (*pskb)->dst = dst; + + /* Change in oif may mean change in hh_len. */ + hh_len = (*pskb)->dst->dev->hard_header_len; + if (skb_headroom(*pskb) < hh_len) { + struct sk_buff *nskb; + + nskb = skb_realloc_headroom(*pskb, hh_len); + if (!nskb) + return -1; + if ((*pskb)->sk) + skb_set_owner_w(nskb, (*pskb)->sk); + kfree_skb(*pskb); + *pskb = nskb; + } + return 0; +} +EXPORT_SYMBOL(ip_xfrm_me_harder); +#endif + void (*ip_nat_decode_session)(struct sk_buff *, struct flowi *); EXPORT_SYMBOL(ip_nat_decode_session); diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c index 92c54999a19d..7c3f7d380240 100644 --- a/net/ipv4/netfilter/ip_nat_standalone.c +++ b/net/ipv4/netfilter/ip_nat_standalone.c @@ -235,19 +235,19 @@ ip_nat_out(unsigned int hooknum, return NF_ACCEPT; ret = ip_nat_fn(hooknum, pskb, in, out, okfn); +#ifdef CONFIG_XFRM if (ret != NF_DROP && ret != NF_STOLEN && (ct = ip_conntrack_get(*pskb, &ctinfo)) != NULL) { enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); if (ct->tuplehash[dir].tuple.src.ip != ct->tuplehash[!dir].tuple.dst.ip -#ifdef CONFIG_XFRM || ct->tuplehash[dir].tuple.src.u.all != ct->tuplehash[!dir].tuple.dst.u.all -#endif ) - return ip_route_me_harder(pskb) == 0 ? ret : NF_DROP; + return ip_xfrm_me_harder(pskb) == 0 ? ret : NF_DROP; } +#endif return ret; } -- cgit v1.2.3 From 5ecfbae093f0c37311e89b29bfc0c9d586eace87 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 15 Feb 2006 22:50:10 +0300 Subject: [PATCH] fix zap_thread's ptrace related problems 1. The tracee can go from ptrace_stop() to do_signal_stop() after __ptrace_unlink(p). 2. It is unsafe to __ptrace_unlink(p) while p->parent may wait for tasklist_lock in ptrace_detach(). Signed-off-by: Oleg Nesterov Cc: Roland McGrath Cc: Ingo Molnar Cc: Christoph Hellwig Cc: Eric W. Biederman Signed-off-by: Linus Torvalds --- fs/exec.c | 2 +- include/linux/ptrace.h | 1 + kernel/ptrace.c | 25 +++++++++++++++---------- 3 files changed, 17 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/fs/exec.c b/fs/exec.c index 055378d2513e..0e1c95074d42 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1403,7 +1403,7 @@ static void zap_threads (struct mm_struct *mm) do_each_thread(g,p) { if (mm == p->mm && p != tsk && p->ptrace && p->parent->mm == mm) { - __ptrace_unlink(p); + __ptrace_detach(p, 0); } } while_each_thread(g,p); write_unlock_irq(&tasklist_lock); diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h index 9d5cd106b344..0d36750fc0f1 100644 --- a/include/linux/ptrace.h +++ b/include/linux/ptrace.h @@ -84,6 +84,7 @@ extern int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __us extern int ptrace_writedata(struct task_struct *tsk, char __user *src, unsigned long dst, int len); extern int ptrace_attach(struct task_struct *tsk); extern int ptrace_detach(struct task_struct *, unsigned int); +extern void __ptrace_detach(struct task_struct *, unsigned int); extern void ptrace_disable(struct task_struct *); extern int ptrace_check_attach(struct task_struct *task, int kill); extern int ptrace_request(struct task_struct *child, long request, long addr, long data); diff --git a/kernel/ptrace.c b/kernel/ptrace.c index d2cf144d0af5..d95a72c9279d 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -72,8 +72,8 @@ void ptrace_untrace(task_t *child) */ void __ptrace_unlink(task_t *child) { - if (!child->ptrace) - BUG(); + BUG_ON(!child->ptrace); + child->ptrace = 0; if (!list_empty(&child->ptrace_list)) { list_del_init(&child->ptrace_list); @@ -184,22 +184,27 @@ bad: return retval; } +void __ptrace_detach(struct task_struct *child, unsigned int data) +{ + child->exit_code = data; + /* .. re-parent .. */ + __ptrace_unlink(child); + /* .. and wake it up. */ + if (child->exit_state != EXIT_ZOMBIE) + wake_up_process(child); +} + int ptrace_detach(struct task_struct *child, unsigned int data) { if (!valid_signal(data)) - return -EIO; + return -EIO; /* Architecture-specific hardware disable .. */ ptrace_disable(child); - /* .. re-parent .. */ - child->exit_code = data; - write_lock_irq(&tasklist_lock); - __ptrace_unlink(child); - /* .. and wake it up. */ - if (child->exit_state != EXIT_ZOMBIE) - wake_up_process(child); + if (child->ptrace) + __ptrace_detach(child, data); write_unlock_irq(&tasklist_lock); return 0; -- cgit v1.2.3 From 48d5cad87c3a4998d0bda16ccfb5c60dfe4de5fb Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 15 Feb 2006 15:10:22 -0800 Subject: [XFRM]: Fix SNAT-related crash in xfrm4_output_finish When a packet matching an IPsec policy is SNATed so it doesn't match any policy anymore it looses its xfrm bundle, which makes xfrm4_output_finish crash because of a NULL pointer dereference. This patch directs these packets to the original output path instead. Since the packets have already passed the POST_ROUTING hook, but need to start at the beginning of the original output path which includes another POST_ROUTING invocation, a flag is added to the IPCB to indicate that the packet was rerouted and doesn't need to pass the POST_ROUTING hook again. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter.h | 19 +++++++++++++++---- include/net/ip.h | 1 + include/net/xfrm.h | 1 - net/ipv4/ip_gre.c | 3 ++- net/ipv4/ip_output.c | 16 ++++++++++------ net/ipv4/ipip.c | 3 ++- net/ipv4/xfrm4_output.c | 13 ++++++++++--- 7 files changed, 40 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 4cf6088625c1..3ca3d9ee78a9 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -184,8 +184,11 @@ static inline int nf_hook_thresh(int pf, unsigned int hook, struct sk_buff **pskb, struct net_device *indev, struct net_device *outdev, - int (*okfn)(struct sk_buff *), int thresh) + int (*okfn)(struct sk_buff *), int thresh, + int cond) { + if (!cond) + return 1; #ifndef CONFIG_NETFILTER_DEBUG if (list_empty(&nf_hooks[pf][hook])) return 1; @@ -197,7 +200,7 @@ static inline int nf_hook(int pf, unsigned int hook, struct sk_buff **pskb, struct net_device *indev, struct net_device *outdev, int (*okfn)(struct sk_buff *)) { - return nf_hook_thresh(pf, hook, pskb, indev, outdev, okfn, INT_MIN); + return nf_hook_thresh(pf, hook, pskb, indev, outdev, okfn, INT_MIN, 1); } /* Activate hook; either okfn or kfree_skb called, unless a hook @@ -224,7 +227,13 @@ static inline int nf_hook(int pf, unsigned int hook, struct sk_buff **pskb, #define NF_HOOK_THRESH(pf, hook, skb, indev, outdev, okfn, thresh) \ ({int __ret; \ -if ((__ret=nf_hook_thresh(pf, hook, &(skb), indev, outdev, okfn, thresh)) == 1)\ +if ((__ret=nf_hook_thresh(pf, hook, &(skb), indev, outdev, okfn, thresh, 1)) == 1)\ + __ret = (okfn)(skb); \ +__ret;}) + +#define NF_HOOK_COND(pf, hook, skb, indev, outdev, okfn, cond) \ +({int __ret; \ +if ((__ret=nf_hook_thresh(pf, hook, &(skb), indev, outdev, okfn, INT_MIN, cond)) == 1)\ __ret = (okfn)(skb); \ __ret;}) @@ -295,11 +304,13 @@ extern struct proc_dir_entry *proc_net_netfilter; #else /* !CONFIG_NETFILTER */ #define NF_HOOK(pf, hook, skb, indev, outdev, okfn) (okfn)(skb) +#define NF_HOOK_COND(pf, hook, skb, indev, outdev, okfn, cond) (okfn)(skb) static inline int nf_hook_thresh(int pf, unsigned int hook, struct sk_buff **pskb, struct net_device *indev, struct net_device *outdev, - int (*okfn)(struct sk_buff *), int thresh) + int (*okfn)(struct sk_buff *), int thresh, + int cond) { return okfn(*pskb); } diff --git a/include/net/ip.h b/include/net/ip.h index 8de0697b364c..fab3d5b3ab1c 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -41,6 +41,7 @@ struct inet_skb_parm #define IPSKB_XFRM_TUNNEL_SIZE 2 #define IPSKB_XFRM_TRANSFORMED 4 #define IPSKB_FRAG_COMPLETE 8 +#define IPSKB_REROUTED 16 }; struct ipcm_cookie diff --git a/include/net/xfrm.h b/include/net/xfrm.h index d09ca0e7d139..d6111a2f0a23 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -866,7 +866,6 @@ extern int xfrm_state_mtu(struct xfrm_state *x, int mtu); extern int xfrm_init_state(struct xfrm_state *x); extern int xfrm4_rcv(struct sk_buff *skb); extern int xfrm4_output(struct sk_buff *skb); -extern int xfrm4_output_finish(struct sk_buff *skb); extern int xfrm4_tunnel_register(struct xfrm_tunnel *handler); extern int xfrm4_tunnel_deregister(struct xfrm_tunnel *handler); extern int xfrm6_rcv_spi(struct sk_buff **pskb, u32 spi); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index abe23923e4e7..9981dcd68f11 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -830,7 +830,8 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) skb->h.raw = skb->nh.raw; skb->nh.raw = skb_push(skb, gre_hlen); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); - IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE|IPSKB_XFRM_TRANSFORMED); + IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED | + IPSKB_REROUTED); dst_release(skb->dst); skb->dst = &rt->u.dst; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 3324fbfe528a..57d290d89ec2 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -207,8 +207,10 @@ static inline int ip_finish_output(struct sk_buff *skb) { #if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM) /* Policy lookup after SNAT yielded a new policy */ - if (skb->dst->xfrm != NULL) - return xfrm4_output_finish(skb); + if (skb->dst->xfrm != NULL) { + IPCB(skb)->flags |= IPSKB_REROUTED; + return dst_output(skb); + } #endif if (skb->len > dst_mtu(skb->dst) && !(skb_shinfo(skb)->ufo_size || skb_shinfo(skb)->tso_size)) @@ -271,8 +273,9 @@ int ip_mc_output(struct sk_buff *skb) newskb->dev, ip_dev_loopback_xmit); } - return NF_HOOK(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dev, - ip_finish_output); + return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dev, + ip_finish_output, + !(IPCB(skb)->flags & IPSKB_REROUTED)); } int ip_output(struct sk_buff *skb) @@ -284,8 +287,9 @@ int ip_output(struct sk_buff *skb) skb->dev = dev; skb->protocol = htons(ETH_P_IP); - return NF_HOOK(PF_INET, NF_IP_POST_ROUTING, skb, NULL, dev, - ip_finish_output); + return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, dev, + ip_finish_output, + !(IPCB(skb)->flags & IPSKB_REROUTED)); } int ip_queue_xmit(struct sk_buff *skb, int ipfragok) diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index e5cbe72c6b80..03d13742a4b8 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -622,7 +622,8 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) skb->h.raw = skb->nh.raw; skb->nh.raw = skb_push(skb, sizeof(struct iphdr)); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); - IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE|IPSKB_XFRM_TRANSFORMED); + IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED | + IPSKB_REROUTED); dst_release(skb->dst); skb->dst = &rt->u.dst; diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c index d4df0ddd424b..32ad229b4fed 100644 --- a/net/ipv4/xfrm4_output.c +++ b/net/ipv4/xfrm4_output.c @@ -152,10 +152,16 @@ error_nolock: goto out_exit; } -int xfrm4_output_finish(struct sk_buff *skb) +static int xfrm4_output_finish(struct sk_buff *skb) { int err; +#ifdef CONFIG_NETFILTER + if (!skb->dst->xfrm) { + IPCB(skb)->flags |= IPSKB_REROUTED; + return dst_output(skb); + } +#endif while (likely((err = xfrm4_output_one(skb)) == 0)) { nf_reset(skb); @@ -178,6 +184,7 @@ int xfrm4_output_finish(struct sk_buff *skb) int xfrm4_output(struct sk_buff *skb) { - return NF_HOOK(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dst->dev, - xfrm4_output_finish); + return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dst->dev, + xfrm4_output_finish, + !(IPCB(skb)->flags & IPSKB_REROUTED)); } -- cgit v1.2.3 From 9c92d3486434e7310cb288587953e2dae4a79701 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 15 Feb 2006 15:18:19 -0800 Subject: [NETFILTER]: Don't invoke okfn in CONFIG_NETFILTER=n variant of nf_hook() nf_hook() is supposed to call the netfilter hook and return control of the packet back to the caller in case it may pass, the okfn is only used for queueing. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 3ca3d9ee78a9..468896939843 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -318,7 +318,7 @@ static inline int nf_hook(int pf, unsigned int hook, struct sk_buff **pskb, struct net_device *indev, struct net_device *outdev, int (*okfn)(struct sk_buff *)) { - return okfn(*pskb); + return 1; } static inline void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb) {} struct flowi; -- cgit v1.2.3 From b2ee9dbfad14ba8e34a589d552ddc67300a26bec Mon Sep 17 00:00:00 2001 From: Roman Zippel Date: Wed, 15 Feb 2006 15:17:40 -0800 Subject: [PATCH] hrtimer: fix multiple macro argument expansion For two macros the arguments were expanded twice, change them to inline functions to avoid it. Signed-off-by: Roman Zippel Acked-by: Ingo Molnar Acked-by: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ktime.h | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ktime.h b/include/linux/ktime.h index 6aca67a569a2..f3dec45ef874 100644 --- a/include/linux/ktime.h +++ b/include/linux/ktime.h @@ -96,10 +96,16 @@ static inline ktime_t ktime_set(const long secs, const unsigned long nsecs) ({ (ktime_t){ .tv64 = (kt).tv64 + (nsval) }; }) /* convert a timespec to ktime_t format: */ -#define timespec_to_ktime(ts) ktime_set((ts).tv_sec, (ts).tv_nsec) +static inline ktime_t timespec_to_ktime(struct timespec ts) +{ + return ktime_set(ts.tv_sec, ts.tv_nsec); +} /* convert a timeval to ktime_t format: */ -#define timeval_to_ktime(tv) ktime_set((tv).tv_sec, (tv).tv_usec * 1000) +static inline ktime_t timeval_to_ktime(struct timeval tv) +{ + return ktime_set(tv.tv_sec, tv.tv_usec * NSEC_PER_USEC); +} /* Map the ktime_t to timespec conversion to ns_to_timespec function */ #define ktime_to_timespec(kt) ns_to_timespec((kt).tv64) -- cgit v1.2.3 From a62eaf151d9cb478d127cfbc2e93c498869785b0 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 16 Feb 2006 23:41:58 +0100 Subject: [PATCH] x86_64: Add boot option to disable randomized mappings and cleanup AMD SimNow!'s JIT doesn't like them at all in the guest. For distribution installation it's easiest if it's a boot time option. Also I moved the variable to a more appropiate place and make it independent from sysctl And marked __read_mostly which it is. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 3 +++ arch/i386/kernel/cpu/transmeta.c | 1 + include/linux/kernel.h | 6 ------ include/linux/mm.h | 2 ++ kernel/sysctl.c | 2 -- mm/memory.c | 10 ++++++++++ 6 files changed, 16 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index ac75b57edf2e..b874771385cd 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1638,6 +1638,9 @@ running once the system is up. Format: ,,,,,[,[,[,]]] + norandmaps Don't use address space randomization + Equivalent to echo 0 > /proc/sys/kernel/randomize_va_space + ______________________________________________________________________ Changelog: diff --git a/arch/i386/kernel/cpu/transmeta.c b/arch/i386/kernel/cpu/transmeta.c index bdbeb77f4e22..7214c9b577ab 100644 --- a/arch/i386/kernel/cpu/transmeta.c +++ b/arch/i386/kernel/cpu/transmeta.c @@ -1,4 +1,5 @@ #include +#include #include #include #include diff --git a/include/linux/kernel.h b/include/linux/kernel.h index b49affa0ac5a..3b507bf05d09 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -326,12 +326,6 @@ struct sysinfo { /* Force a compilation error if condition is true */ #define BUILD_BUG_ON(condition) ((void)sizeof(char[1 - 2*!!(condition)])) -#ifdef CONFIG_SYSCTL -extern int randomize_va_space; -#else -#define randomize_va_space 1 -#endif - /* Trap pasters of __FUNCTION__ at compile-time */ #define __FUNCTION__ (__func__) diff --git a/include/linux/mm.h b/include/linux/mm.h index 75e9f0724997..26e1663a5cbe 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1051,5 +1051,7 @@ int shrink_slab(unsigned long scanned, gfp_t gfp_mask, void drop_pagecache(void); void drop_slab(void); +extern int randomize_va_space; + #endif /* __KERNEL__ */ #endif /* _LINUX_MM_H */ diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 71dd6f62efec..7654d55c47f5 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -126,8 +126,6 @@ extern int sysctl_hz_timer; extern int acct_parm[]; #endif -int randomize_va_space = 1; - static int parse_table(int __user *, int, void __user *, size_t __user *, void __user *, size_t, ctl_table *, void **); static int proc_doutsstring(ctl_table *table, int write, struct file *filp, diff --git a/mm/memory.c b/mm/memory.c index 2bee1f21aa8a..9abc6008544b 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -82,6 +82,16 @@ EXPORT_SYMBOL(num_physpages); EXPORT_SYMBOL(high_memory); EXPORT_SYMBOL(vmalloc_earlyreserve); +int randomize_va_space __read_mostly = 1; + +static int __init disable_randmaps(char *s) +{ + randomize_va_space = 0; + return 0; +} +__setup("norandmaps", disable_randmaps); + + /* * If a p?d_bad entry is found while walking page tables, report * the error, before resetting entry to p?d_none. Usually (but -- cgit v1.2.3 From 726c14bf499e91e7ede4f1728830aba05c675061 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Fri, 17 Feb 2006 10:30:23 +1100 Subject: [PATCH] Provide an interface for getting the current tick length This provides an interface for arch code to find out how many nanoseconds are going to be added on to xtime by the next call to do_timer. The value returned is a fixed-point number in 52.12 format in nanoseconds. The reason for this format is that it gives the full precision that the timekeeping code is using internally. The motivation for this is to fix a problem that has arisen on 32-bit powerpc in that the value returned by do_gettimeofday drifts apart from xtime if NTP is being used. PowerPC is now using a lockless do_gettimeofday based on reading the timebase register and performing some simple arithmetic. (This method of getting the time is also exported to userspace via the VDSO.) However, the factor and offset it uses were calculated based on the nominal tick length and weren't being adjusted when NTP varied the tick length. Note that 64-bit powerpc has had the lockless do_gettimeofday for a long time now. It also had an extremely hairy routine that got called from the 32-bit compat routine for adjtimex, which adjusted the factor and offset according to what it thought the timekeeping code was going to do. Not only was this only called if a 32-bit task did adjtimex (i.e. not if a 64-bit task did adjtimex), it was also duplicating computations from kernel/timer.c and it wasn't clear that it was (still) correct. The simple solution is to ask the timekeeping code how long the current jiffy will be on each timer interrupt, after calling do_timer. If this jiffy will be a different length from the last one, we then need to compute new values for the factor and offset used in the lockless do_gettimeofday. In this way we can keep xtime and do_gettimeofday in sync, even when NTP is varying the tick length. Note that when adjtimex varies the tick length, it almost always introduces the variation from the next tick on. The only case I could see where adjtimex would vary the length of the current tick is when an old-style adjtime adjustment is being cancelled. (It's not clear to me why the adjustment has to be cancelled immediately rather than from the next tick on.) Thus I don't see any real need for a hook in adjtimex; the rare case of an old-style adjustment being cancelled can be fixed up at the next tick. Signed-off-by: Paul Mackerras Acked-by: john stultz Signed-off-by: Linus Torvalds --- include/linux/timex.h | 3 +++ kernel/timer.c | 39 ++++++++++++++++++++++++++++++++++----- 2 files changed, 37 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/timex.h b/include/linux/timex.h index 04a4a8cb4ed3..b7ca1204e42a 100644 --- a/include/linux/timex.h +++ b/include/linux/timex.h @@ -345,6 +345,9 @@ time_interpolator_reset(void) #endif /* !CONFIG_TIME_INTERPOLATION */ +/* Returns how long ticks are at present, in ns / 2^(SHIFT_SCALE-10). */ +extern u64 current_tick_length(void); + #endif /* KERNEL */ #endif /* LINUX_TIMEX_H */ diff --git a/kernel/timer.c b/kernel/timer.c index b9dad3994676..fe3a9a9f8328 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -717,12 +717,16 @@ static void second_overflow(void) #endif } -/* in the NTP reference this is called "hardclock()" */ -static void update_wall_time_one_tick(void) +/* + * Returns how many microseconds we need to add to xtime this tick + * in doing an adjustment requested with adjtime. + */ +static long adjtime_adjustment(void) { - long time_adjust_step, delta_nsec; + long time_adjust_step; - if ((time_adjust_step = time_adjust) != 0 ) { + time_adjust_step = time_adjust; + if (time_adjust_step) { /* * We are doing an adjtime thing. Prepare time_adjust_step to * be within bounds. Note that a positive time_adjust means we @@ -733,10 +737,19 @@ static void update_wall_time_one_tick(void) */ time_adjust_step = min(time_adjust_step, (long)tickadj); time_adjust_step = max(time_adjust_step, (long)-tickadj); + } + return time_adjust_step; +} +/* in the NTP reference this is called "hardclock()" */ +static void update_wall_time_one_tick(void) +{ + long time_adjust_step, delta_nsec; + + time_adjust_step = adjtime_adjustment(); + if (time_adjust_step) /* Reduce by this step the amount of time left */ time_adjust -= time_adjust_step; - } delta_nsec = tick_nsec + time_adjust_step * 1000; /* * Advance the phase, once it gets to one microsecond, then @@ -758,6 +771,22 @@ static void update_wall_time_one_tick(void) } } +/* + * Return how long ticks are at the moment, that is, how much time + * update_wall_time_one_tick will add to xtime next time we call it + * (assuming no calls to do_adjtimex in the meantime). + * The return value is in fixed-point nanoseconds with SHIFT_SCALE-10 + * bits to the right of the binary point. + * This function has no side-effects. + */ +u64 current_tick_length(void) +{ + long delta_nsec; + + delta_nsec = tick_nsec + adjtime_adjustment() * 1000; + return ((u64) delta_nsec << (SHIFT_SCALE - 10)) + time_adj; +} + /* * Using a loop looks inefficient, but "ticks" is * usually just one (we shouldn't be losing ticks, -- cgit v1.2.3