summaryrefslogtreecommitdiff
path: root/kernel
diff options
context:
space:
mode:
Diffstat (limited to 'kernel')
-rw-r--r--kernel/Makefile2
-rw-r--r--kernel/bpf/bpf_iter.c58
-rw-r--r--kernel/bpf/core.c2
-rw-r--r--kernel/bpf/map_iter.c37
-rw-r--r--kernel/bpf/syscall.c2
-rw-r--r--kernel/bpf/verifier.c2
-rw-r--r--kernel/dma/Kconfig5
-rw-r--r--kernel/dma/debug.c55
-rw-r--r--kernel/irq/manage.c7
-rw-r--r--kernel/irq/pm.c8
-rw-r--r--kernel/module.c60
-rw-r--r--kernel/sys_ni.c1
-rw-r--r--kernel/sysctl_binary.c171
-rw-r--r--kernel/time/Kconfig9
-rw-r--r--kernel/time/alarmtimer.c2
-rw-r--r--kernel/time/posix-cpu-timers.c216
-rw-r--r--kernel/time/sched_clock.c2
-rw-r--r--kernel/time/timekeeping.c4
-rw-r--r--kernel/time/timekeeping_internal.h11
-rw-r--r--kernel/time/timer.c1
-rw-r--r--kernel/time/vsyscall.c41
-rw-r--r--kernel/trace/bpf_trace.c2
22 files changed, 372 insertions, 326 deletions
diff --git a/kernel/Makefile b/kernel/Makefile
index b3da548691c9..9a20016d4900 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -5,7 +5,7 @@
obj-y = fork.o exec_domain.o panic.o \
cpu.o exit.o softirq.o resource.o \
- sysctl.o sysctl_binary.o capability.o ptrace.o user.o \
+ sysctl.o capability.o ptrace.o user.o \
signal.o sys.o umh.o workqueue.o pid.o task_work.o \
extable.o params.o \
kthread.o sys_ni.o nsproxy.o \
diff --git a/kernel/bpf/bpf_iter.c b/kernel/bpf/bpf_iter.c
index 363b9cafc2d8..b6715964b685 100644
--- a/kernel/bpf/bpf_iter.c
+++ b/kernel/bpf/bpf_iter.c
@@ -338,8 +338,8 @@ static void bpf_iter_link_release(struct bpf_link *link)
struct bpf_iter_link *iter_link =
container_of(link, struct bpf_iter_link, link);
- if (iter_link->aux.map)
- bpf_map_put_with_uref(iter_link->aux.map);
+ if (iter_link->tinfo->reg_info->detach_target)
+ iter_link->tinfo->reg_info->detach_target(&iter_link->aux);
}
static void bpf_iter_link_dealloc(struct bpf_link *link)
@@ -390,15 +390,35 @@ bool bpf_link_is_iter(struct bpf_link *link)
int bpf_iter_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
{
+ union bpf_iter_link_info __user *ulinfo;
struct bpf_link_primer link_primer;
struct bpf_iter_target_info *tinfo;
- struct bpf_iter_aux_info aux = {};
+ union bpf_iter_link_info linfo;
struct bpf_iter_link *link;
- u32 prog_btf_id, target_fd;
+ u32 prog_btf_id, linfo_len;
bool existed = false;
- struct bpf_map *map;
int err;
+ if (attr->link_create.target_fd || attr->link_create.flags)
+ return -EINVAL;
+
+ memset(&linfo, 0, sizeof(union bpf_iter_link_info));
+
+ ulinfo = u64_to_user_ptr(attr->link_create.iter_info);
+ linfo_len = attr->link_create.iter_info_len;
+ if (!ulinfo ^ !linfo_len)
+ return -EINVAL;
+
+ if (ulinfo) {
+ err = bpf_check_uarg_tail_zero(ulinfo, sizeof(linfo),
+ linfo_len);
+ if (err)
+ return err;
+ linfo_len = min_t(u32, linfo_len, sizeof(linfo));
+ if (copy_from_user(&linfo, ulinfo, linfo_len))
+ return -EFAULT;
+ }
+
prog_btf_id = prog->aux->attach_btf_id;
mutex_lock(&targets_mutex);
list_for_each_entry(tinfo, &targets, list) {
@@ -411,13 +431,6 @@ int bpf_iter_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
if (!existed)
return -ENOENT;
- /* Make sure user supplied flags are target expected. */
- target_fd = attr->link_create.target_fd;
- if (attr->link_create.flags != tinfo->reg_info->req_linfo)
- return -EINVAL;
- if (!attr->link_create.flags && target_fd)
- return -EINVAL;
-
link = kzalloc(sizeof(*link), GFP_USER | __GFP_NOWARN);
if (!link)
return -ENOMEM;
@@ -431,28 +444,15 @@ int bpf_iter_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
return err;
}
- if (tinfo->reg_info->req_linfo == BPF_ITER_LINK_MAP_FD) {
- map = bpf_map_get_with_uref(target_fd);
- if (IS_ERR(map)) {
- err = PTR_ERR(map);
- goto cleanup_link;
- }
-
- aux.map = map;
- err = tinfo->reg_info->check_target(prog, &aux);
+ if (tinfo->reg_info->attach_target) {
+ err = tinfo->reg_info->attach_target(prog, &linfo, &link->aux);
if (err) {
- bpf_map_put_with_uref(map);
- goto cleanup_link;
+ bpf_link_cleanup(&link_primer);
+ return err;
}
-
- link->aux.map = map;
}
return bpf_link_settle(&link_primer);
-
-cleanup_link:
- bpf_link_cleanup(&link_primer);
- return err;
}
static void init_seq_meta(struct bpf_iter_priv_data *priv_data,
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index bde93344164d..ed0b3578867c 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -1966,7 +1966,7 @@ void bpf_prog_array_delete_safe(struct bpf_prog_array *array,
* @index: the index of the program to replace
*
* Skips over dummy programs, by not counting them, when calculating
- * the the position of the program to replace.
+ * the position of the program to replace.
*
* Return:
* * 0 - Success
diff --git a/kernel/bpf/map_iter.c b/kernel/bpf/map_iter.c
index fbe1f557cb88..af86048e5afd 100644
--- a/kernel/bpf/map_iter.c
+++ b/kernel/bpf/map_iter.c
@@ -98,12 +98,21 @@ static struct bpf_iter_reg bpf_map_reg_info = {
.seq_info = &bpf_map_seq_info,
};
-static int bpf_iter_check_map(struct bpf_prog *prog,
- struct bpf_iter_aux_info *aux)
+static int bpf_iter_attach_map(struct bpf_prog *prog,
+ union bpf_iter_link_info *linfo,
+ struct bpf_iter_aux_info *aux)
{
u32 key_acc_size, value_acc_size, key_size, value_size;
- struct bpf_map *map = aux->map;
+ struct bpf_map *map;
bool is_percpu = false;
+ int err = -EINVAL;
+
+ if (!linfo->map.map_fd)
+ return -EBADF;
+
+ map = bpf_map_get_with_uref(linfo->map.map_fd);
+ if (IS_ERR(map))
+ return PTR_ERR(map);
if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH ||
map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH ||
@@ -112,7 +121,7 @@ static int bpf_iter_check_map(struct bpf_prog *prog,
else if (map->map_type != BPF_MAP_TYPE_HASH &&
map->map_type != BPF_MAP_TYPE_LRU_HASH &&
map->map_type != BPF_MAP_TYPE_ARRAY)
- return -EINVAL;
+ goto put_map;
key_acc_size = prog->aux->max_rdonly_access;
value_acc_size = prog->aux->max_rdwr_access;
@@ -122,10 +131,22 @@ static int bpf_iter_check_map(struct bpf_prog *prog,
else
value_size = round_up(map->value_size, 8) * num_possible_cpus();
- if (key_acc_size > key_size || value_acc_size > value_size)
- return -EACCES;
+ if (key_acc_size > key_size || value_acc_size > value_size) {
+ err = -EACCES;
+ goto put_map;
+ }
+ aux->map = map;
return 0;
+
+put_map:
+ bpf_map_put_with_uref(map);
+ return err;
+}
+
+static void bpf_iter_detach_map(struct bpf_iter_aux_info *aux)
+{
+ bpf_map_put_with_uref(aux->map);
}
DEFINE_BPF_ITER_FUNC(bpf_map_elem, struct bpf_iter_meta *meta,
@@ -133,8 +154,8 @@ DEFINE_BPF_ITER_FUNC(bpf_map_elem, struct bpf_iter_meta *meta,
static const struct bpf_iter_reg bpf_map_elem_reg_info = {
.target = "bpf_map_elem",
- .check_target = bpf_iter_check_map,
- .req_linfo = BPF_ITER_LINK_MAP_FD,
+ .attach_target = bpf_iter_attach_map,
+ .detach_target = bpf_iter_detach_map,
.ctx_arg_info_size = 2,
.ctx_arg_info = {
{ offsetof(struct bpf_iter__bpf_map_elem, key),
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 2f343ce15747..86299a292214 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -3883,7 +3883,7 @@ static int tracing_bpf_link_attach(const union bpf_attr *attr, struct bpf_prog *
return -EINVAL;
}
-#define BPF_LINK_CREATE_LAST_FIELD link_create.flags
+#define BPF_LINK_CREATE_LAST_FIELD link_create.iter_info_len
static int link_create(union bpf_attr *attr)
{
enum bpf_prog_type ptype;
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index b6ccfce3bf4c..ef938f17b944 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -8294,7 +8294,7 @@ static bool stacksafe(struct bpf_func_state *old,
if (old->stack[spi].slot_type[i % BPF_REG_SIZE] !=
cur->stack[spi].slot_type[i % BPF_REG_SIZE])
/* Ex: old explored (safe) state has STACK_SPILL in
- * this stack slot, but current has has STACK_MISC ->
+ * this stack slot, but current has STACK_MISC ->
* this verifier states are not equivalent,
* return false to continue verification of this path
*/
diff --git a/kernel/dma/Kconfig b/kernel/dma/Kconfig
index f4770fcfa62b..5732b2b3ef17 100644
--- a/kernel/dma/Kconfig
+++ b/kernel/dma/Kconfig
@@ -186,11 +186,6 @@ config DMA_API_DEBUG
drivers like double-freeing of DMA mappings or freeing mappings that
were never allocated.
- This also attempts to catch cases where a page owned by DMA is
- accessed by the cpu in a way that could cause data corruption. For
- example, this enables cow_user_page() to check that the source page is
- not undergoing DMA.
-
This option causes a performance degradation. Use only if you want to
debug device drivers and dma interactions.
diff --git a/kernel/dma/debug.c b/kernel/dma/debug.c
index f7f807fb6ebb..8e9f7b301c6d 100644
--- a/kernel/dma/debug.c
+++ b/kernel/dma/debug.c
@@ -448,9 +448,6 @@ void debug_dma_dump_mappings(struct device *dev)
* dma_active_cacheline entry to track per event. dma_map_sg(), on the
* other hand, consumes a single dma_debug_entry, but inserts 'nents'
* entries into the tree.
- *
- * At any time debug_dma_assert_idle() can be called to trigger a
- * warning if any cachelines in the given page are in the active set.
*/
static RADIX_TREE(dma_active_cacheline, GFP_NOWAIT);
static DEFINE_SPINLOCK(radix_lock);
@@ -497,10 +494,7 @@ static void active_cacheline_inc_overlap(phys_addr_t cln)
overlap = active_cacheline_set_overlap(cln, ++overlap);
/* If we overflowed the overlap counter then we're potentially
- * leaking dma-mappings. Otherwise, if maps and unmaps are
- * balanced then this overflow may cause false negatives in
- * debug_dma_assert_idle() as the cacheline may be marked idle
- * prematurely.
+ * leaking dma-mappings.
*/
WARN_ONCE(overlap > ACTIVE_CACHELINE_MAX_OVERLAP,
pr_fmt("exceeded %d overlapping mappings of cacheline %pa\n"),
@@ -555,53 +549,6 @@ static void active_cacheline_remove(struct dma_debug_entry *entry)
spin_unlock_irqrestore(&radix_lock, flags);
}
-/**
- * debug_dma_assert_idle() - assert that a page is not undergoing dma
- * @page: page to lookup in the dma_active_cacheline tree
- *
- * Place a call to this routine in cases where the cpu touching the page
- * before the dma completes (page is dma_unmapped) will lead to data
- * corruption.
- */
-void debug_dma_assert_idle(struct page *page)
-{
- static struct dma_debug_entry *ents[CACHELINES_PER_PAGE];
- struct dma_debug_entry *entry = NULL;
- void **results = (void **) &ents;
- unsigned int nents, i;
- unsigned long flags;
- phys_addr_t cln;
-
- if (dma_debug_disabled())
- return;
-
- if (!page)
- return;
-
- cln = (phys_addr_t) page_to_pfn(page) << CACHELINE_PER_PAGE_SHIFT;
- spin_lock_irqsave(&radix_lock, flags);
- nents = radix_tree_gang_lookup(&dma_active_cacheline, results, cln,
- CACHELINES_PER_PAGE);
- for (i = 0; i < nents; i++) {
- phys_addr_t ent_cln = to_cacheline_number(ents[i]);
-
- if (ent_cln == cln) {
- entry = ents[i];
- break;
- } else if (ent_cln >= cln + CACHELINES_PER_PAGE)
- break;
- }
- spin_unlock_irqrestore(&radix_lock, flags);
-
- if (!entry)
- return;
-
- cln = to_cacheline_number(entry);
- err_printk(entry->dev, entry,
- "cpu touching an active dma mapped cacheline [cln=%pa]\n",
- &cln);
-}
-
/*
* Wrapper function for adding an entry to the hash.
* This function takes care of locking itself.
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index d55ba625d426..52ac5391dcc6 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -2731,8 +2731,10 @@ int irq_set_irqchip_state(unsigned int irq, enum irqchip_irq_state which,
do {
chip = irq_data_get_irq_chip(data);
- if (WARN_ON_ONCE(!chip))
- return -ENODEV;
+ if (WARN_ON_ONCE(!chip)) {
+ err = -ENODEV;
+ goto out_unlock;
+ }
if (chip->irq_set_irqchip_state)
break;
#ifdef CONFIG_IRQ_DOMAIN_HIERARCHY
@@ -2745,6 +2747,7 @@ int irq_set_irqchip_state(unsigned int irq, enum irqchip_irq_state which,
if (data)
err = chip->irq_set_irqchip_state(data, which, val);
+out_unlock:
irq_put_desc_busunlock(desc, flags);
return err;
}
diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c
index 8f557fa1f4fe..c6c7e187ae74 100644
--- a/kernel/irq/pm.c
+++ b/kernel/irq/pm.c
@@ -185,14 +185,18 @@ void rearm_wake_irq(unsigned int irq)
unsigned long flags;
struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, IRQ_GET_DESC_CHECK_GLOBAL);
- if (!desc || !(desc->istate & IRQS_SUSPENDED) ||
- !irqd_is_wakeup_set(&desc->irq_data))
+ if (!desc)
return;
+ if (!(desc->istate & IRQS_SUSPENDED) ||
+ !irqd_is_wakeup_set(&desc->irq_data))
+ goto unlock;
+
desc->istate &= ~IRQS_SUSPENDED;
irqd_set(&desc->irq_data, IRQD_WAKEUP_ARMED);
__enable_irq(desc);
+unlock:
irq_put_desc_busunlock(desc, flags);
}
diff --git a/kernel/module.c b/kernel/module.c
index 8fa2600bde6a..1c5cff34d9f2 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -422,7 +422,7 @@ static bool each_symbol_in_section(const struct symsearch *arr,
}
/* Returns true as soon as fn returns true, otherwise false. */
-bool each_symbol_section(bool (*fn)(const struct symsearch *arr,
+static bool each_symbol_section(bool (*fn)(const struct symsearch *arr,
struct module *owner,
void *data),
void *data)
@@ -484,7 +484,6 @@ bool each_symbol_section(bool (*fn)(const struct symsearch *arr,
}
return false;
}
-EXPORT_SYMBOL_GPL(each_symbol_section);
struct find_symbol_arg {
/* Input */
@@ -496,6 +495,7 @@ struct find_symbol_arg {
struct module *owner;
const s32 *crc;
const struct kernel_symbol *sym;
+ enum mod_license license;
};
static bool check_exported_symbol(const struct symsearch *syms,
@@ -505,9 +505,9 @@ static bool check_exported_symbol(const struct symsearch *syms,
struct find_symbol_arg *fsa = data;
if (!fsa->gplok) {
- if (syms->licence == GPL_ONLY)
+ if (syms->license == GPL_ONLY)
return false;
- if (syms->licence == WILL_BE_GPL_ONLY && fsa->warn) {
+ if (syms->license == WILL_BE_GPL_ONLY && fsa->warn) {
pr_warn("Symbol %s is being used by a non-GPL module, "
"which will not be allowed in the future\n",
fsa->name);
@@ -529,6 +529,7 @@ static bool check_exported_symbol(const struct symsearch *syms,
fsa->owner = owner;
fsa->crc = symversion(syms->crcs, symnum);
fsa->sym = &syms->start[symnum];
+ fsa->license = syms->license;
return true;
}
@@ -585,9 +586,10 @@ static bool find_exported_symbol_in_section(const struct symsearch *syms,
/* Find an exported symbol and return it, along with, (optional) crc and
* (optional) module which owns it. Needs preempt disabled or module_mutex. */
-const struct kernel_symbol *find_symbol(const char *name,
+static const struct kernel_symbol *find_symbol(const char *name,
struct module **owner,
const s32 **crc,
+ enum mod_license *license,
bool gplok,
bool warn)
{
@@ -602,13 +604,14 @@ const struct kernel_symbol *find_symbol(const char *name,
*owner = fsa.owner;
if (crc)
*crc = fsa.crc;
+ if (license)
+ *license = fsa.license;
return fsa.sym;
}
pr_debug("Failed to find symbol %s\n", name);
return NULL;
}
-EXPORT_SYMBOL_GPL(find_symbol);
/*
* Search for module by name: must hold module_mutex (or preempt disabled
@@ -869,7 +872,7 @@ static int add_module_usage(struct module *a, struct module *b)
}
/* Module a uses b: caller needs module_mutex() */
-int ref_module(struct module *a, struct module *b)
+static int ref_module(struct module *a, struct module *b)
{
int err;
@@ -888,7 +891,6 @@ int ref_module(struct module *a, struct module *b)
}
return 0;
}
-EXPORT_SYMBOL_GPL(ref_module);
/* Clear the unload stuff of the module. */
static void module_unload_free(struct module *mod)
@@ -1077,7 +1079,7 @@ void __symbol_put(const char *symbol)
struct module *owner;
preempt_disable();
- if (!find_symbol(symbol, &owner, NULL, true, false))
+ if (!find_symbol(symbol, &owner, NULL, NULL, true, false))
BUG();
module_put(owner);
preempt_enable();
@@ -1169,11 +1171,10 @@ static inline void module_unload_free(struct module *mod)
{
}
-int ref_module(struct module *a, struct module *b)
+static int ref_module(struct module *a, struct module *b)
{
return strong_try_module_get(b);
}
-EXPORT_SYMBOL_GPL(ref_module);
static inline int module_unload_init(struct module *mod)
{
@@ -1356,7 +1357,7 @@ static inline int check_modstruct_version(const struct load_info *info,
* locking is necessary -- use preempt_disable() to placate lockdep.
*/
preempt_disable();
- if (!find_symbol("module_layout", NULL, &crc, true, false)) {
+ if (!find_symbol("module_layout", NULL, &crc, NULL, true, false)) {
preempt_enable();
BUG();
}
@@ -1430,6 +1431,24 @@ static int verify_namespace_is_imported(const struct load_info *info,
return 0;
}
+static bool inherit_taint(struct module *mod, struct module *owner)
+{
+ if (!owner || !test_bit(TAINT_PROPRIETARY_MODULE, &owner->taints))
+ return true;
+
+ if (mod->using_gplonly_symbols) {
+ pr_err("%s: module using GPL-only symbols uses symbols from proprietary module %s.\n",
+ mod->name, owner->name);
+ return false;
+ }
+
+ if (!test_bit(TAINT_PROPRIETARY_MODULE, &mod->taints)) {
+ pr_warn("%s: module uses symbols from proprietary module %s, inheriting taint.\n",
+ mod->name, owner->name);
+ set_bit(TAINT_PROPRIETARY_MODULE, &mod->taints);
+ }
+ return true;
+}
/* Resolve a symbol for this module. I.e. if we find one, record usage. */
static const struct kernel_symbol *resolve_symbol(struct module *mod,
@@ -1440,6 +1459,7 @@ static const struct kernel_symbol *resolve_symbol(struct module *mod,
struct module *owner;
const struct kernel_symbol *sym;
const s32 *crc;
+ enum mod_license license;
int err;
/*
@@ -1449,11 +1469,19 @@ static const struct kernel_symbol *resolve_symbol(struct module *mod,
*/
sched_annotate_sleep();
mutex_lock(&module_mutex);
- sym = find_symbol(name, &owner, &crc,
+ sym = find_symbol(name, &owner, &crc, &license,
!(mod->taints & (1 << TAINT_PROPRIETARY_MODULE)), true);
if (!sym)
goto unlock;
+ if (license == GPL_ONLY)
+ mod->using_gplonly_symbols = true;
+
+ if (!inherit_taint(mod, owner)) {
+ sym = NULL;
+ goto getname;
+ }
+
if (!check_version(info, name, mod, crc)) {
sym = ERR_PTR(-EINVAL);
goto getname;
@@ -2236,7 +2264,7 @@ void *__symbol_get(const char *symbol)
const struct kernel_symbol *sym;
preempt_disable();
- sym = find_symbol(symbol, &owner, NULL, true, true);
+ sym = find_symbol(symbol, &owner, NULL, NULL, true, true);
if (sym && strong_try_module_get(owner))
sym = NULL;
preempt_enable();
@@ -2272,7 +2300,7 @@ static int verify_exported_symbols(struct module *mod)
for (i = 0; i < ARRAY_SIZE(arr); i++) {
for (s = arr[i].sym; s < arr[i].sym + arr[i].num; s++) {
if (find_symbol(kernel_symbol_name(s), &owner, NULL,
- true, false)) {
+ NULL, true, false)) {
pr_err("%s: exports duplicate symbol %s"
" (owned by %s)\n",
mod->name, kernel_symbol_name(s),
@@ -4489,7 +4517,6 @@ struct module *__module_address(unsigned long addr)
}
return mod;
}
-EXPORT_SYMBOL_GPL(__module_address);
/*
* is_module_text_address - is this address inside module code?
@@ -4528,7 +4555,6 @@ struct module *__module_text_address(unsigned long addr)
}
return mod;
}
-EXPORT_SYMBOL_GPL(__module_text_address);
/* Don't grab lock, we're oopsing. */
void print_modules(void)
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 3b69a560a7ac..4d59775ea79c 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -364,7 +364,6 @@ COND_SYSCALL(socketcall);
COND_SYSCALL_COMPAT(socketcall);
/* compat syscalls for arm64, x86, ... */
-COND_SYSCALL_COMPAT(sysctl);
COND_SYSCALL_COMPAT(fanotify_mark);
/* x86 */
diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c
deleted file mode 100644
index 7d550cc76a3b..000000000000
--- a/kernel/sysctl_binary.c
+++ /dev/null
@@ -1,171 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-#include <linux/stat.h>
-#include <linux/sysctl.h>
-#include "../fs/xfs/xfs_sysctl.h"
-#include <linux/sunrpc/debug.h>
-#include <linux/string.h>
-#include <linux/syscalls.h>
-#include <linux/namei.h>
-#include <linux/mount.h>
-#include <linux/fs.h>
-#include <linux/nsproxy.h>
-#include <linux/pid_namespace.h>
-#include <linux/file.h>
-#include <linux/ctype.h>
-#include <linux/netdevice.h>
-#include <linux/kernel.h>
-#include <linux/uuid.h>
-#include <linux/slab.h>
-#include <linux/compat.h>
-
-static ssize_t binary_sysctl(const int *name, int nlen,
- void __user *oldval, size_t oldlen, void __user *newval, size_t newlen)
-{
- return -ENOSYS;
-}
-
-static void deprecated_sysctl_warning(const int *name, int nlen)
-{
- int i;
-
- /*
- * CTL_KERN/KERN_VERSION is used by older glibc and cannot
- * ever go away.
- */
- if (nlen >= 2 && name[0] == CTL_KERN && name[1] == KERN_VERSION)
- return;
-
- if (printk_ratelimit()) {
- printk(KERN_INFO
- "warning: process `%s' used the deprecated sysctl "
- "system call with ", current->comm);
- for (i = 0; i < nlen; i++)
- printk(KERN_CONT "%d.", name[i]);
- printk(KERN_CONT "\n");
- }
- return;
-}
-
-#define WARN_ONCE_HASH_BITS 8
-#define WARN_ONCE_HASH_SIZE (1<<WARN_ONCE_HASH_BITS)
-
-static DECLARE_BITMAP(warn_once_bitmap, WARN_ONCE_HASH_SIZE);
-
-#define FNV32_OFFSET 2166136261U
-#define FNV32_PRIME 0x01000193
-
-/*
- * Print each legacy sysctl (approximately) only once.
- * To avoid making the tables non-const use a external
- * hash-table instead.
- * Worst case hash collision: 6, but very rarely.
- * NOTE! We don't use the SMP-safe bit tests. We simply
- * don't care enough.
- */
-static void warn_on_bintable(const int *name, int nlen)
-{
- int i;
- u32 hash = FNV32_OFFSET;
-
- for (i = 0; i < nlen; i++)
- hash = (hash ^ name[i]) * FNV32_PRIME;
- hash %= WARN_ONCE_HASH_SIZE;
- if (__test_and_set_bit(hash, warn_once_bitmap))
- return;
- deprecated_sysctl_warning(name, nlen);
-}
-
-static ssize_t do_sysctl(int __user *args_name, int nlen,
- void __user *oldval, size_t oldlen, void __user *newval, size_t newlen)
-{
- int name[CTL_MAXNAME];
- int i;
-
- /* Check args->nlen. */
- if (nlen < 0 || nlen > CTL_MAXNAME)
- return -ENOTDIR;
- /* Read in the sysctl name for simplicity */
- for (i = 0; i < nlen; i++)
- if (get_user(name[i], args_name + i))
- return -EFAULT;
-
- warn_on_bintable(name, nlen);
-
- return binary_sysctl(name, nlen, oldval, oldlen, newval, newlen);
-}
-
-SYSCALL_DEFINE1(sysctl, struct __sysctl_args __user *, args)
-{
- struct __sysctl_args tmp;
- size_t oldlen = 0;
- ssize_t result;
-
- if (copy_from_user(&tmp, args, sizeof(tmp)))
- return -EFAULT;
-
- if (tmp.oldval && !tmp.oldlenp)
- return -EFAULT;
-
- if (tmp.oldlenp && get_user(oldlen, tmp.oldlenp))
- return -EFAULT;
-
- result = do_sysctl(tmp.name, tmp.nlen, tmp.oldval, oldlen,
- tmp.newval, tmp.newlen);
-
- if (result >= 0) {
- oldlen = result;
- result = 0;
- }
-
- if (tmp.oldlenp && put_user(oldlen, tmp.oldlenp))
- return -EFAULT;
-
- return result;
-}
-
-
-#ifdef CONFIG_COMPAT
-
-struct compat_sysctl_args {
- compat_uptr_t name;
- int nlen;
- compat_uptr_t oldval;
- compat_uptr_t oldlenp;
- compat_uptr_t newval;
- compat_size_t newlen;
- compat_ulong_t __unused[4];
-};
-
-COMPAT_SYSCALL_DEFINE1(sysctl, struct compat_sysctl_args __user *, args)
-{
- struct compat_sysctl_args tmp;
- compat_size_t __user *compat_oldlenp;
- size_t oldlen = 0;
- ssize_t result;
-
- if (copy_from_user(&tmp, args, sizeof(tmp)))
- return -EFAULT;
-
- if (tmp.oldval && !tmp.oldlenp)
- return -EFAULT;
-
- compat_oldlenp = compat_ptr(tmp.oldlenp);
- if (compat_oldlenp && get_user(oldlen, compat_oldlenp))
- return -EFAULT;
-
- result = do_sysctl(compat_ptr(tmp.name), tmp.nlen,
- compat_ptr(tmp.oldval), oldlen,
- compat_ptr(tmp.newval), tmp.newlen);
-
- if (result >= 0) {
- oldlen = result;
- result = 0;
- }
-
- if (compat_oldlenp && put_user(oldlen, compat_oldlenp))
- return -EFAULT;
-
- return result;
-}
-
-#endif /* CONFIG_COMPAT */
diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig
index fcc42353f125..a09b1d61df6a 100644
--- a/kernel/time/Kconfig
+++ b/kernel/time/Kconfig
@@ -52,6 +52,15 @@ config GENERIC_CLOCKEVENTS_MIN_ADJUST
config GENERIC_CMOS_UPDATE
bool
+# Select to handle posix CPU timers from task_work
+# and not from the timer interrupt context
+config HAVE_POSIX_CPU_TIMERS_TASK_WORK
+ bool
+
+config POSIX_CPU_TIMERS_TASK_WORK
+ bool
+ default y if POSIX_TIMERS && HAVE_POSIX_CPU_TIMERS_TASK_WORK
+
if GENERIC_CLOCKEVENTS
menu "Timers subsystem"
diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c
index 2ffb466af77e..ca223a89530a 100644
--- a/kernel/time/alarmtimer.c
+++ b/kernel/time/alarmtimer.c
@@ -192,7 +192,7 @@ static void alarmtimer_dequeue(struct alarm_base *base, struct alarm *alarm)
* When a alarm timer fires, this runs through the timerqueue to
* see which alarms expired, and runs those. If there are more alarm
* timers queued for the future, we set the hrtimer to fire when
- * when the next future alarm timer expires.
+ * the next future alarm timer expires.
*/
static enum hrtimer_restart alarmtimer_fired(struct hrtimer *timer)
{
diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c
index 165117996ea0..a71758e34e45 100644
--- a/kernel/time/posix-cpu-timers.c
+++ b/kernel/time/posix-cpu-timers.c
@@ -377,6 +377,7 @@ static int posix_cpu_clock_get(const clockid_t clock, struct timespec64 *tp)
*/
static int posix_cpu_timer_create(struct k_itimer *new_timer)
{
+ static struct lock_class_key posix_cpu_timers_key;
struct pid *pid;
rcu_read_lock();
@@ -386,6 +387,17 @@ static int posix_cpu_timer_create(struct k_itimer *new_timer)
return -EINVAL;
}
+ /*
+ * If posix timer expiry is handled in task work context then
+ * timer::it_lock can be taken without disabling interrupts as all
+ * other locking happens in task context. This requires a seperate
+ * lock class key otherwise regular posix timer expiry would record
+ * the lock class being taken in interrupt context and generate a
+ * false positive warning.
+ */
+ if (IS_ENABLED(CONFIG_POSIX_CPU_TIMERS_TASK_WORK))
+ lockdep_set_class(&new_timer->it_lock, &posix_cpu_timers_key);
+
new_timer->kclock = &clock_posix_cpu;
timerqueue_init(&new_timer->it.cpu.node);
new_timer->it.cpu.pid = get_pid(pid);
@@ -1080,43 +1092,163 @@ static inline bool fastpath_timer_check(struct task_struct *tsk)
return false;
}
+static void handle_posix_cpu_timers(struct task_struct *tsk);
+
+#ifdef CONFIG_POSIX_CPU_TIMERS_TASK_WORK
+static void posix_cpu_timers_work(struct callback_head *work)
+{
+ handle_posix_cpu_timers(current);
+}
+
/*
- * This is called from the timer interrupt handler. The irq handler has
- * already updated our counts. We need to check if any timers fire now.
- * Interrupts are disabled.
+ * Initialize posix CPU timers task work in init task. Out of line to
+ * keep the callback static and to avoid header recursion hell.
*/
-void run_posix_cpu_timers(void)
+void __init posix_cputimers_init_work(void)
{
- struct task_struct *tsk = current;
- struct k_itimer *timer, *next;
- unsigned long flags;
- LIST_HEAD(firing);
+ init_task_work(&current->posix_cputimers_work.work,
+ posix_cpu_timers_work);
+}
- lockdep_assert_irqs_disabled();
+/*
+ * Note: All operations on tsk->posix_cputimer_work.scheduled happen either
+ * in hard interrupt context or in task context with interrupts
+ * disabled. Aside of that the writer/reader interaction is always in the
+ * context of the current task, which means they are strict per CPU.
+ */
+static inline bool posix_cpu_timers_work_scheduled(struct task_struct *tsk)
+{
+ return tsk->posix_cputimers_work.scheduled;
+}
- /*
- * The fast path checks that there are no expired thread or thread
- * group timers. If that's so, just return.
- */
- if (!fastpath_timer_check(tsk))
+static inline void __run_posix_cpu_timers(struct task_struct *tsk)
+{
+ if (WARN_ON_ONCE(tsk->posix_cputimers_work.scheduled))
return;
- lockdep_posixtimer_enter();
- if (!lock_task_sighand(tsk, &flags)) {
- lockdep_posixtimer_exit();
- return;
+ /* Schedule task work to actually expire the timers */
+ tsk->posix_cputimers_work.scheduled = true;
+ task_work_add(tsk, &tsk->posix_cputimers_work.work, TWA_RESUME);
+}
+
+static inline bool posix_cpu_timers_enable_work(struct task_struct *tsk,
+ unsigned long start)
+{
+ bool ret = true;
+
+ /*
+ * On !RT kernels interrupts are disabled while collecting expired
+ * timers, so no tick can happen and the fast path check can be
+ * reenabled without further checks.
+ */
+ if (!IS_ENABLED(CONFIG_PREEMPT_RT)) {
+ tsk->posix_cputimers_work.scheduled = false;
+ return true;
}
+
/*
- * Here we take off tsk->signal->cpu_timers[N] and
- * tsk->cpu_timers[N] all the timers that are firing, and
- * put them on the firing list.
+ * On RT enabled kernels ticks can happen while the expired timers
+ * are collected under sighand lock. But any tick which observes
+ * the CPUTIMERS_WORK_SCHEDULED bit set, does not run the fastpath
+ * checks. So reenabling the tick work has do be done carefully:
+ *
+ * Disable interrupts and run the fast path check if jiffies have
+ * advanced since the collecting of expired timers started. If
+ * jiffies have not advanced or the fast path check did not find
+ * newly expired timers, reenable the fast path check in the timer
+ * interrupt. If there are newly expired timers, return false and
+ * let the collection loop repeat.
*/
- check_thread_timers(tsk, &firing);
+ local_irq_disable();
+ if (start != jiffies && fastpath_timer_check(tsk))
+ ret = false;
+ else
+ tsk->posix_cputimers_work.scheduled = false;
+ local_irq_enable();
+
+ return ret;
+}
+#else /* CONFIG_POSIX_CPU_TIMERS_TASK_WORK */
+static inline void __run_posix_cpu_timers(struct task_struct *tsk)
+{
+ lockdep_posixtimer_enter();
+ handle_posix_cpu_timers(tsk);
+ lockdep_posixtimer_exit();
+}
+
+static inline bool posix_cpu_timers_work_scheduled(struct task_struct *tsk)
+{
+ return false;
+}
+
+static inline bool posix_cpu_timers_enable_work(struct task_struct *tsk,
+ unsigned long start)
+{
+ return true;
+}
+#endif /* CONFIG_POSIX_CPU_TIMERS_TASK_WORK */
+
+static void handle_posix_cpu_timers(struct task_struct *tsk)
+{
+ struct k_itimer *timer, *next;
+ unsigned long flags, start;
+ LIST_HEAD(firing);
+
+ if (!lock_task_sighand(tsk, &flags))
+ return;
- check_process_timers(tsk, &firing);
+ do {
+ /*
+ * On RT locking sighand lock does not disable interrupts,
+ * so this needs to be careful vs. ticks. Store the current
+ * jiffies value.
+ */
+ start = READ_ONCE(jiffies);
+ barrier();
+
+ /*
+ * Here we take off tsk->signal->cpu_timers[N] and
+ * tsk->cpu_timers[N] all the timers that are firing, and
+ * put them on the firing list.
+ */
+ check_thread_timers(tsk, &firing);
+
+ check_process_timers(tsk, &firing);
+
+ /*
+ * The above timer checks have updated the exipry cache and
+ * because nothing can have queued or modified timers after
+ * sighand lock was taken above it is guaranteed to be
+ * consistent. So the next timer interrupt fastpath check
+ * will find valid data.
+ *
+ * If timer expiry runs in the timer interrupt context then
+ * the loop is not relevant as timers will be directly
+ * expired in interrupt context. The stub function below
+ * returns always true which allows the compiler to
+ * optimize the loop out.
+ *
+ * If timer expiry is deferred to task work context then
+ * the following rules apply:
+ *
+ * - On !RT kernels no tick can have happened on this CPU
+ * after sighand lock was acquired because interrupts are
+ * disabled. So reenabling task work before dropping
+ * sighand lock and reenabling interrupts is race free.
+ *
+ * - On RT kernels ticks might have happened but the tick
+ * work ignored posix CPU timer handling because the
+ * CPUTIMERS_WORK_SCHEDULED bit is set. Reenabling work
+ * must be done very carefully including a check whether
+ * ticks have happened since the start of the timer
+ * expiry checks. posix_cpu_timers_enable_work() takes
+ * care of that and eventually lets the expiry checks
+ * run again.
+ */
+ } while (!posix_cpu_timers_enable_work(tsk, start));
/*
- * We must release these locks before taking any timer's lock.
+ * We must release sighand lock before taking any timer's lock.
* There is a potential race with timer deletion here, as the
* siglock now protects our private firing list. We have set
* the firing flag in each timer, so that a deletion attempt
@@ -1134,6 +1266,13 @@ void run_posix_cpu_timers(void)
list_for_each_entry_safe(timer, next, &firing, it.cpu.elist) {
int cpu_firing;
+ /*
+ * spin_lock() is sufficient here even independent of the
+ * expiry context. If expiry happens in hard interrupt
+ * context it's obvious. For task work context it's safe
+ * because all other operations on timer::it_lock happen in
+ * task context (syscall or exit).
+ */
spin_lock(&timer->it_lock);
list_del_init(&timer->it.cpu.elist);
cpu_firing = timer->it.cpu.firing;
@@ -1147,7 +1286,34 @@ void run_posix_cpu_timers(void)
cpu_timer_fire(timer);
spin_unlock(&timer->it_lock);
}
- lockdep_posixtimer_exit();
+}
+
+/*
+ * This is called from the timer interrupt handler. The irq handler has
+ * already updated our counts. We need to check if any timers fire now.
+ * Interrupts are disabled.
+ */
+void run_posix_cpu_timers(void)
+{
+ struct task_struct *tsk = current;
+
+ lockdep_assert_irqs_disabled();
+
+ /*
+ * If the actual expiry is deferred to task work context and the
+ * work is already scheduled there is no point to do anything here.
+ */
+ if (posix_cpu_timers_work_scheduled(tsk))
+ return;
+
+ /*
+ * The fast path checks that there are no expired thread or thread
+ * group timers. If that's so, just return.
+ */
+ if (!fastpath_timer_check(tsk))
+ return;
+
+ __run_posix_cpu_timers(tsk);
}
/*
diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c
index 0deaf4b79fb4..1c03eec6ca9b 100644
--- a/kernel/time/sched_clock.c
+++ b/kernel/time/sched_clock.c
@@ -229,7 +229,7 @@ void __init generic_sched_clock_init(void)
{
/*
* If no sched_clock() function has been provided at that point,
- * make it the final one one.
+ * make it the final one.
*/
if (cd.actual_read_sched_clock == jiffy_sched_clock_read)
sched_clock_register(jiffy_sched_clock_read, BITS_PER_LONG, HZ);
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 406306b33452..4c47f388a83f 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -39,7 +39,7 @@ enum timekeeping_adv_mode {
TK_ADV_FREQ
};
-static DEFINE_RAW_SPINLOCK(timekeeper_lock);
+DEFINE_RAW_SPINLOCK(timekeeper_lock);
/*
* The most important data for readout fits into a single 64 byte
@@ -2004,7 +2004,7 @@ static inline unsigned int accumulate_nsecs_to_secs(struct timekeeper *tk)
* logarithmic_accumulation - shifted accumulation of cycles
*
* This functions accumulates a shifted interval of cycles into
- * into a shifted interval nanoseconds. Allows for O(log) accumulation
+ * a shifted interval nanoseconds. Allows for O(log) accumulation
* loop.
*
* Returns the unconsumed cycles.
diff --git a/kernel/time/timekeeping_internal.h b/kernel/time/timekeeping_internal.h
index bcbb52db2256..4ca2787d1642 100644
--- a/kernel/time/timekeeping_internal.h
+++ b/kernel/time/timekeeping_internal.h
@@ -1,12 +1,14 @@
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _TIMEKEEPING_INTERNAL_H
#define _TIMEKEEPING_INTERNAL_H
-/*
- * timekeeping debug functions
- */
+
#include <linux/clocksource.h>
+#include <linux/spinlock.h>
#include <linux/time.h>
+/*
+ * timekeeping debug functions
+ */
#ifdef CONFIG_DEBUG_FS
extern void tk_debug_account_sleep_time(const struct timespec64 *t);
#else
@@ -31,4 +33,7 @@ static inline u64 clocksource_delta(u64 now, u64 last, u64 mask)
}
#endif
+/* Semi public for serialization of non timekeeper VDSO updates. */
+extern raw_spinlock_t timekeeper_lock;
+
#endif /* _TIMEKEEPING_INTERNAL_H */
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index ae5029f984a8..a16764b0116e 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -2017,6 +2017,7 @@ static void __init init_timer_cpus(void)
void __init init_timers(void)
{
init_timer_cpus();
+ posix_cputimers_init_work();
open_softirq(TIMER_SOFTIRQ, run_timer_softirq);
}
diff --git a/kernel/time/vsyscall.c b/kernel/time/vsyscall.c
index 54ce6eb2ca36..88e6b8ed6ca5 100644
--- a/kernel/time/vsyscall.c
+++ b/kernel/time/vsyscall.c
@@ -13,6 +13,8 @@
#include <vdso/helpers.h>
#include <vdso/vsyscall.h>
+#include "timekeeping_internal.h"
+
static inline void update_vdso_data(struct vdso_data *vdata,
struct timekeeper *tk)
{
@@ -127,3 +129,42 @@ void update_vsyscall_tz(void)
__arch_sync_vdso_data(vdata);
}
+
+/**
+ * vdso_update_begin - Start of a VDSO update section
+ *
+ * Allows architecture code to safely update the architecture specific VDSO
+ * data. Disables interrupts, acquires timekeeper lock to serialize against
+ * concurrent updates from timekeeping and invalidates the VDSO data
+ * sequence counter to prevent concurrent readers from accessing
+ * inconsistent data.
+ *
+ * Returns: Saved interrupt flags which need to be handed in to
+ * vdso_update_end().
+ */
+unsigned long vdso_update_begin(void)
+{
+ struct vdso_data *vdata = __arch_get_k_vdso_data();
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&timekeeper_lock, flags);
+ vdso_write_begin(vdata);
+ return flags;
+}
+
+/**
+ * vdso_update_end - End of a VDSO update section
+ * @flags: Interrupt flags as returned from vdso_update_begin()
+ *
+ * Pairs with vdso_update_begin(). Marks vdso data consistent, invokes data
+ * synchronization if the architecture requires it, drops timekeeper lock
+ * and restores interrupt flags.
+ */
+void vdso_update_end(unsigned long flags)
+{
+ struct vdso_data *vdata = __arch_get_k_vdso_data();
+
+ vdso_write_end(vdata);
+ __arch_sync_vdso_data(vdata);
+ raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
+}
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index cb91ef902cc4..a8d4f253ed77 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -383,7 +383,7 @@ static DEFINE_RAW_SPINLOCK(trace_printk_lock);
#define BPF_TRACE_PRINTK_SIZE 1024
-static inline __printf(1, 0) int bpf_do_trace_printk(const char *fmt, ...)
+static __printf(1, 0) int bpf_do_trace_printk(const char *fmt, ...)
{
static char buf[BPF_TRACE_PRINTK_SIZE];
unsigned long flags;