tmakin/ack-tegra
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'linus' into tracing/core

Browse Source 
Conflicts:
	kernel/trace/trace_events_filter.c

We use the tracing/core version.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
This commit is contained in:
Ingo Molnar
2009-08-11 14:19:09 +02:00
parent fb82ad7198 85dfd81dc5
commit 89034bc2c7
1029 changed files with 32122 additions and 16291 deletions
Download Patch File Download Diff File Expand all files Collapse all files
kernel/cgroup.c
+110 -41
View File
@@ -47,6 +47,7 @@
#include <linux/hash.h>
#include <linux/namei.h>
#include <linux/smp_lock.h>
#include <linux/pid_namespace.h>
#include <asm/atomic.h>
@@ -734,16 +735,28 @@ static void cgroup_d_remove_dir(struct dentry *dentry)
* reference to css->refcnt. In general, this refcnt is expected to goes down
* to zero, soon.
*
* CGRP_WAIT_ON_RMDIR flag is modified under cgroup's inode->i_mutex;
* CGRP_WAIT_ON_RMDIR flag is set under cgroup's inode->i_mutex;
*/
DECLARE_WAIT_QUEUE_HEAD(cgroup_rmdir_waitq);
static void cgroup_wakeup_rmdir_waiters(const struct cgroup *cgrp)
static void cgroup_wakeup_rmdir_waiter(struct cgroup *cgrp)
{
if (unlikely(test_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags)))
if (unlikely(test_and_clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags)))
wake_up_all(&cgroup_rmdir_waitq);
}
void cgroup_exclude_rmdir(struct cgroup_subsys_state *css)
{
css_get(css);
}
void cgroup_release_and_wakeup_rmdir(struct cgroup_subsys_state *css)
{
cgroup_wakeup_rmdir_waiter(css->cgroup);
css_put(css);
}
static int rebind_subsystems(struct cgroupfs_root *root,
unsigned long final_bits)
{
@@ -960,6 +973,7 @@ static void init_cgroup_housekeeping(struct cgroup *cgrp)
INIT_LIST_HEAD(&cgrp->children);
INIT_LIST_HEAD(&cgrp->css_sets);
INIT_LIST_HEAD(&cgrp->release_list);
INIT_LIST_HEAD(&cgrp->pids_list);
init_rwsem(&cgrp->pids_mutex);
}
static void init_cgroup_root(struct cgroupfs_root *root)
@@ -1357,7 +1371,7 @@ int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk)
* wake up rmdir() waiter. the rmdir should fail since the cgroup
* is no longer empty.
*/
cgroup_wakeup_rmdir_waiters(cgrp);
cgroup_wakeup_rmdir_waiter(cgrp);
return 0;
}
@@ -2201,12 +2215,30 @@ err:
return ret;
}
/*
* Cache pids for all threads in the same pid namespace that are
* opening the same "tasks" file.
*/
struct cgroup_pids {
/* The node in cgrp->pids_list */
struct list_head list;
/* The cgroup those pids belong to */
struct cgroup *cgrp;
/* The namepsace those pids belong to */
struct pid_namespace *ns;
/* Array of process ids in the cgroup */
pid_t *tasks_pids;
/* How many files are using the this tasks_pids array */
int use_count;
/* Length of the current tasks_pids array */
int length;
};
static int cmppid(const void *a, const void *b)
{
return *(pid_t *)a - *(pid_t *)b;
}
/*
* seq_file methods for the "tasks" file. The seq_file position is the
* next pid to display; the seq_file iterator is a pointer to the pid
@@ -2221,45 +2253,47 @@ static void *cgroup_tasks_start(struct seq_file *s, loff_t *pos)
* after a seek to the start). Use a binary-search to find the
* next pid to display, if any
*/
struct cgroup *cgrp = s->private;
struct cgroup_pids *cp = s->private;
struct cgroup *cgrp = cp->cgrp;
int index = 0, pid = *pos;
int *iter;
down_read(&cgrp->pids_mutex);
if (pid) {
int end = cgrp->pids_length;
int end = cp->length;
while (index < end) {
int mid = (index + end) / 2;
if (cgrp->tasks_pids[mid] == pid) {
if (cp->tasks_pids[mid] == pid) {
index = mid;
break;
} else if (cgrp->tasks_pids[mid] <= pid)
} else if (cp->tasks_pids[mid] <= pid)
index = mid + 1;
else
end = mid;
}
}
/* If we're off the end of the array, we're done */
if (index >= cgrp->pids_length)
if (index >= cp->length)
return NULL;
/* Update the abstract position to be the actual pid that we found */
iter = cgrp->tasks_pids + index;
iter = cp->tasks_pids + index;
*pos = *iter;
return iter;
}
static void cgroup_tasks_stop(struct seq_file *s, void *v)
{
struct cgroup *cgrp = s->private;
struct cgroup_pids *cp = s->private;
struct cgroup *cgrp = cp->cgrp;
up_read(&cgrp->pids_mutex);
}
static void *cgroup_tasks_next(struct seq_file *s, void *v, loff_t *pos)
{
struct cgroup *cgrp = s->private;
struct cgroup_pids *cp = s->private;
int *p = v;
int *end = cgrp->tasks_pids + cgrp->pids_length;
int *end = cp->tasks_pids + cp->length;
/*
* Advance to the next pid in the array. If this goes off the
@@ -2286,26 +2320,33 @@ static struct seq_operations cgroup_tasks_seq_operations = {
.show = cgroup_tasks_show,
};
static void release_cgroup_pid_array(struct cgroup *cgrp)
static void release_cgroup_pid_array(struct cgroup_pids *cp)
{
struct cgroup *cgrp = cp->cgrp;
down_write(&cgrp->pids_mutex);
BUG_ON(!cgrp->pids_use_count);
if (!--cgrp->pids_use_count) {
kfree(cgrp->tasks_pids);
cgrp->tasks_pids = NULL;
cgrp->pids_length = 0;
BUG_ON(!cp->use_count);
if (!--cp->use_count) {
list_del(&cp->list);
put_pid_ns(cp->ns);
kfree(cp->tasks_pids);
kfree(cp);
}
up_write(&cgrp->pids_mutex);
}
static int cgroup_tasks_release(struct inode *inode, struct file *file)
{
struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent);
struct seq_file *seq;
struct cgroup_pids *cp;
if (!(file->f_mode & FMODE_READ))
return 0;
release_cgroup_pid_array(cgrp);
seq = file->private_data;
cp = seq->private;
release_cgroup_pid_array(cp);
return seq_release(inode, file);
}
@@ -2324,6 +2365,8 @@ static struct file_operations cgroup_tasks_operations = {
static int cgroup_tasks_open(struct inode *unused, struct file *file)
{
struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent);
struct pid_namespace *ns = current->nsproxy->pid_ns;
struct cgroup_pids *cp;
pid_t *pidarray;
int npids;
int retval;
@@ -2350,20 +2393,37 @@ static int cgroup_tasks_open(struct inode *unused, struct file *file)
* array if necessary
*/
down_write(&cgrp->pids_mutex);
kfree(cgrp->tasks_pids);
cgrp->tasks_pids = pidarray;
cgrp->pids_length = npids;
cgrp->pids_use_count++;
list_for_each_entry(cp, &cgrp->pids_list, list) {
if (ns == cp->ns)
goto found;
}
cp = kzalloc(sizeof(*cp), GFP_KERNEL);
if (!cp) {
up_write(&cgrp->pids_mutex);
kfree(pidarray);
return -ENOMEM;
}
cp->cgrp = cgrp;
cp->ns = ns;
get_pid_ns(ns);
list_add(&cp->list, &cgrp->pids_list);
found:
kfree(cp->tasks_pids);
cp->tasks_pids = pidarray;
cp->length = npids;
cp->use_count++;
up_write(&cgrp->pids_mutex);
file->f_op = &cgroup_tasks_operations;
retval = seq_open(file, &cgroup_tasks_seq_operations);
if (retval) {
release_cgroup_pid_array(cgrp);
release_cgroup_pid_array(cp);
return retval;
}
((struct seq_file *)file->private_data)->private = cgrp;
((struct seq_file *)file->private_data)->private = cp;
return 0;
}
@@ -2695,34 +2755,43 @@ again:
}
mutex_unlock(&cgroup_mutex);
/*
* In general, subsystem has no css->refcnt after pre_destroy(). But
* in racy cases, subsystem may have to get css->refcnt after
* pre_destroy() and it makes rmdir return with -EBUSY. This sometimes
* make rmdir return -EBUSY too often. To avoid that, we use waitqueue
* for cgroup's rmdir. CGRP_WAIT_ON_RMDIR is for synchronizing rmdir
* and subsystem's reference count handling. Please see css_get/put
* and css_tryget() and cgroup_wakeup_rmdir_waiter() implementation.
*/
set_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags);
/*
* Call pre_destroy handlers of subsys. Notify subsystems
* that rmdir() request comes.
*/
ret = cgroup_call_pre_destroy(cgrp);
if (ret)
if (ret) {
clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags);
return ret;
}
mutex_lock(&cgroup_mutex);
parent = cgrp->parent;
if (atomic_read(&cgrp->count) || !list_empty(&cgrp->children)) {
clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags);
mutex_unlock(&cgroup_mutex);
return -EBUSY;
}
/*
* css_put/get is provided for subsys to grab refcnt to css. In typical
* case, subsystem has no reference after pre_destroy(). But, under
* hierarchy management, some *temporal* refcnt can be hold.
* To avoid returning -EBUSY to a user, waitqueue is used. If subsys
* is really busy, it should return -EBUSY at pre_destroy(). wake_up
* is called when css_put() is called and refcnt goes down to 0.
*/
set_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags);
prepare_to_wait(&cgroup_rmdir_waitq, &wait, TASK_INTERRUPTIBLE);
if (!cgroup_clear_css_refs(cgrp)) {
mutex_unlock(&cgroup_mutex);
schedule();
/*
* Because someone may call cgroup_wakeup_rmdir_waiter() before
* prepare_to_wait(), we need to check this flag.
*/
if (test_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags))
schedule();
finish_wait(&cgroup_rmdir_waitq, &wait);
clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags);
if (signal_pending(current))
@@ -3294,7 +3363,7 @@ void __css_put(struct cgroup_subsys_state *css)
set_bit(CGRP_RELEASABLE, &cgrp->flags);
check_for_release(cgrp);
}
cgroup_wakeup_rmdir_waiters(cgrp);
cgroup_wakeup_rmdir_waiter(cgrp);
}
rcu_read_unlock();
}
kernel/fork.c
+13 -17
View File
@@ -426,6 +426,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p)
init_rwsem(&mm->mmap_sem);
INIT_LIST_HEAD(&mm->mmlist);
mm->flags = (current->mm) ? current->mm->flags : default_dump_filter;
mm->oom_adj = (current->mm) ? current->mm->oom_adj : 0;
mm->core_state = NULL;
mm->nr_ptes = 0;
set_mm_counter(mm, file_rss, 0);
@@ -567,18 +568,18 @@ void mm_release(struct task_struct *tsk, struct mm_struct *mm)
* the value intact in a core dump, and to save the unnecessary
* trouble otherwise. Userland only wants this done for a sys_exit.
*/
if (tsk->clear_child_tid
&& !(tsk->flags & PF_SIGNALED)
&& atomic_read(&mm->mm_users) > 1) {
u32 __user * tidptr = tsk->clear_child_tid;
if (tsk->clear_child_tid) {
if (!(tsk->flags & PF_SIGNALED) &&
atomic_read(&mm->mm_users) > 1) {
/*
* We don't check the error code - if userspace has
* not set up a proper pointer then tough luck.
*/
put_user(0, tsk->clear_child_tid);
sys_futex(tsk->clear_child_tid, FUTEX_WAKE,
1, NULL, NULL, 0);
}
tsk->clear_child_tid = NULL;
/*
* We don't check the error code - if userspace has
* not set up a proper pointer then tough luck.
*/
put_user(0, tidptr);
sys_futex(tidptr, FUTEX_WAKE, 1, NULL, NULL, 0);
}
}
@@ -1268,6 +1269,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
write_unlock_irq(&tasklist_lock);
proc_fork_connector(p);
cgroup_post_fork(p);
perf_counter_fork(p);
return p;
bad_fork_free_pid:
@@ -1407,12 +1409,6 @@ long do_fork(unsigned long clone_flags,
if (clone_flags & CLONE_VFORK) {
p->vfork_done = &vfork;
init_completion(&vfork);
} else if (!(clone_flags & CLONE_VM)) {
/*
* vfork will do an exec which will call
* set_task_comm()
*/
perf_counter_fork(p);
}
audit_finish_fork(p);
kernel/freezer.c
+7
View File
@@ -44,12 +44,19 @@ void refrigerator(void)
recalc_sigpending(); /* We sent fake signal, clean it up */
spin_unlock_irq(&current->sighand->siglock);
/* prevent accounting of that task to load */
current->flags |= PF_FREEZING;
for (;;) {
set_current_state(TASK_UNINTERRUPTIBLE);
if (!frozen(current))
break;
schedule();
}
/* Remove the accounting blocker */
current->flags &= ~PF_FREEZING;
pr_debug("%s left refrigerator\n", current->comm);
__set_current_state(save);
}
kernel/irq/internals.h
+1 -2
View File
@@ -42,8 +42,7 @@ static inline void unregister_handler_proc(unsigned int irq,
extern int irq_select_affinity_usr(unsigned int irq);
extern void
irq_set_thread_affinity(struct irq_desc *desc, const struct cpumask *cpumask);
extern void irq_set_thread_affinity(struct irq_desc *desc);
/*
* Debugging printout:
kernel/irq/manage.c
+49 -6
View File
@@ -80,14 +80,22 @@ int irq_can_set_affinity(unsigned int irq)
return 1;
}
void
irq_set_thread_affinity(struct irq_desc *desc, const struct cpumask *cpumask)
/**
* irq_set_thread_affinity - Notify irq threads to adjust affinity
* @desc: irq descriptor which has affitnity changed
*
* We just set IRQTF_AFFINITY and delegate the affinity setting
* to the interrupt thread itself. We can not call
* set_cpus_allowed_ptr() here as we hold desc->lock and this
* code can be called from hard interrupt context.
*/
void irq_set_thread_affinity(struct irq_desc *desc)
{
struct irqaction *action = desc->action;
while (action) {
if (action->thread)
set_cpus_allowed_ptr(action->thread, cpumask);
set_bit(IRQTF_AFFINITY, &action->thread_flags);
action = action->next;
}
}
@@ -112,7 +120,7 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask)
if (desc->status & IRQ_MOVE_PCNTXT) {
if (!desc->chip->set_affinity(irq, cpumask)) {
cpumask_copy(desc->affinity, cpumask);
irq_set_thread_affinity(desc, cpumask);
irq_set_thread_affinity(desc);
}
}
else {
@@ -122,7 +130,7 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask)
#else
if (!desc->chip->set_affinity(irq, cpumask)) {
cpumask_copy(desc->affinity, cpumask);
irq_set_thread_affinity(desc, cpumask);
irq_set_thread_affinity(desc);
}
#endif
desc->status |= IRQ_AFFINITY_SET;
@@ -176,7 +184,7 @@ int irq_select_affinity_usr(unsigned int irq)
spin_lock_irqsave(&desc->lock, flags);
ret = setup_affinity(irq, desc);
if (!ret)
irq_set_thread_affinity(desc, desc->affinity);
irq_set_thread_affinity(desc);
spin_unlock_irqrestore(&desc->lock, flags);
return ret;
@@ -443,6 +451,39 @@ static int irq_wait_for_interrupt(struct irqaction *action)
return -1;
}
#ifdef CONFIG_SMP
/*
* Check whether we need to change the affinity of the interrupt thread.
*/
static void
irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action)
{
cpumask_var_t mask;
if (!test_and_clear_bit(IRQTF_AFFINITY, &action->thread_flags))
return;
/*
* In case we are out of memory we set IRQTF_AFFINITY again and
* try again next time
*/
if (!alloc_cpumask_var(&mask, GFP_KERNEL)) {
set_bit(IRQTF_AFFINITY, &action->thread_flags);
return;
}
spin_lock_irq(&desc->lock);
cpumask_copy(mask, desc->affinity);
spin_unlock_irq(&desc->lock);
set_cpus_allowed_ptr(current, mask);
free_cpumask_var(mask);
}
#else
static inline void
irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action) { }
#endif
/*
* Interrupt handler thread
*/
@@ -458,6 +499,8 @@ static int irq_thread(void *data)
while (!irq_wait_for_interrupt(action)) {
irq_thread_check_affinity(desc, action);
atomic_inc(&desc->threads_active);
spin_lock_irq(&desc->lock);
kernel/irq/migration.c
+1 -1
View File
@@ -45,7 +45,7 @@ void move_masked_irq(int irq)
< nr_cpu_ids))
if (!desc->chip->set_affinity(irq, desc->pending_mask)) {
cpumask_copy(desc->affinity, desc->pending_mask);
irq_set_thread_affinity(desc, desc->pending_mask);
irq_set_thread_affinity(desc);
}
cpumask_clear(desc->pending_mask);
kernel/irq/numa_migrate.c
+2 -2
View File
@@ -107,8 +107,8 @@ out_unlock:
struct irq_desc *move_irq_desc(struct irq_desc *desc, int node)
{
/* those all static, do move them */
if (desc->irq < NR_IRQS_LEGACY)
/* those static or target node is -1, do not move them */
if (desc->irq < NR_IRQS_LEGACY || node == -1)
return desc;
if (desc->node != node)
kernel/kexec.c
+1 -1
View File
@@ -1228,7 +1228,7 @@ static int __init parse_crashkernel_mem(char *cmdline,
} while (*cur++ == ',');
if (*crash_size > 0) {
while (*cur != ' ' && *cur != '@')
while (*cur && *cur != ' ' && *cur != '@')
cur++;
if (*cur == '@') {
cur++;
kernel/kprobes.c
+1 -1
View File
@@ -686,7 +686,7 @@ int __kprobes register_kprobe(struct kprobe *p)
p->addr = addr;
preempt_disable();
if (!__kernel_text_address((unsigned long) p->addr) ||
if (!kernel_text_address((unsigned long) p->addr) ||
in_kprobes_functions((unsigned long) p->addr)) {
preempt_enable();
return -EINVAL;
kernel/kthread.c
+6 -4
View File
@@ -180,10 +180,12 @@ EXPORT_SYMBOL(kthread_bind);
* @k: thread created by kthread_create().
*
* Sets kthread_should_stop() for @k to return true, wakes it, and
* waits for it to exit. Your threadfn() must not call do_exit()
* itself if you use this function! This can also be called after
* kthread_create() instead of calling wake_up_process(): the thread
* will exit without calling threadfn().
* waits for it to exit. This can also be called after kthread_create()
* instead of calling wake_up_process(): the thread will exit without
* calling threadfn().
*
* If threadfn() may call do_exit() itself, the caller must ensure
* task_struct can't go away.
*
* Returns the result of threadfn(), or %-EINTR if wake_up_process()
* was never called.
kernel/lockdep_proc.c
+2 -1
View File
@@ -758,7 +758,8 @@ static int __init lockdep_proc_init(void)
&proc_lockdep_stats_operations);
#ifdef CONFIG_LOCK_STAT
proc_create("lock_stat", S_IRUSR, NULL, &proc_lock_stat_operations);
proc_create("lock_stat", S_IRUSR | S_IWUSR, NULL,
&proc_lock_stat_operations);
#endif
return 0;
kernel/module.c
+2 -1
View File
@@ -1068,7 +1068,8 @@ static inline int check_modstruct_version(Elf_Shdr *sechdrs,
{
const unsigned long *crc;
if (!find_symbol("module_layout", NULL, &crc, true, false))
if (!find_symbol(MODULE_SYMBOL_PREFIX "module_layout", NULL,
&crc, true, false))
BUG();
return check_version(sechdrs, versindex, "module_layout", mod, crc);
}
kernel/panic.c
+1
View File
@@ -301,6 +301,7 @@ int oops_may_print(void)
*/
void oops_enter(void)
{
tracing_off();
/* can't trust the integrity of the kernel anymore: */
debug_locks_off();
do_oops_enter_exit();
kernel/perf_counter.c
+330 -224
View File
@@ -42,6 +42,7 @@ static int perf_overcommit __read_mostly = 1;
static atomic_t nr_counters __read_mostly;
static atomic_t nr_mmap_counters __read_mostly;
static atomic_t nr_comm_counters __read_mostly;
static atomic_t nr_task_counters __read_mostly;
/*
* perf counter paranoia level:
@@ -146,6 +147,28 @@ static void put_ctx(struct perf_counter_context *ctx)
}
}
static void unclone_ctx(struct perf_counter_context *ctx)
{
if (ctx->parent_ctx) {
put_ctx(ctx->parent_ctx);
ctx->parent_ctx = NULL;
}
}
/*
* If we inherit counters we want to return the parent counter id
* to userspace.
*/
static u64 primary_counter_id(struct perf_counter *counter)
{
u64 id = counter->id;
if (counter->parent)
id = counter->parent->id;
return id;
}
/*
* Get the perf_counter_context for a task and lock it.
* This has to cope with with the fact that until it is locked,
@@ -1081,7 +1104,7 @@ static void perf_counter_sync_stat(struct perf_counter_context *ctx,
__perf_counter_sync_stat(counter, next_counter);
counter = list_next_entry(counter, event_entry);
next_counter = list_next_entry(counter, event_entry);
next_counter = list_next_entry(next_counter, event_entry);
}
}
@@ -1288,7 +1311,6 @@ static void perf_counter_cpu_sched_in(struct perf_cpu_context *cpuctx, int cpu)
#define MAX_INTERRUPTS (~0ULL)
static void perf_log_throttle(struct perf_counter *counter, int enable);
static void perf_log_period(struct perf_counter *counter, u64 period);
static void perf_adjust_period(struct perf_counter *counter, u64 events)
{
@@ -1307,8 +1329,6 @@ static void perf_adjust_period(struct perf_counter *counter, u64 events)
if (!sample_period)
sample_period = 1;
perf_log_period(counter, sample_period);
hwc->sample_period = sample_period;
}
@@ -1463,10 +1483,8 @@ static void perf_counter_enable_on_exec(struct task_struct *task)
/*
* Unclone this context if we enabled any counter.
*/
if (enabled && ctx->parent_ctx) {
put_ctx(ctx->parent_ctx);
ctx->parent_ctx = NULL;
}
if (enabled)
unclone_ctx(ctx);
spin_unlock(&ctx->lock);
@@ -1526,7 +1544,6 @@ __perf_counter_init_context(struct perf_counter_context *ctx,
static struct perf_counter_context *find_get_context(pid_t pid, int cpu)
{
struct perf_counter_context *parent_ctx;
struct perf_counter_context *ctx;
struct perf_cpu_context *cpuctx;
struct task_struct *task;
@@ -1586,11 +1603,7 @@ static struct perf_counter_context *find_get_context(pid_t pid, int cpu)
retry:
ctx = perf_lock_task_context(task, &flags);
if (ctx) {
parent_ctx = ctx->parent_ctx;
if (parent_ctx) {
put_ctx(parent_ctx);
ctx->parent_ctx = NULL; /* no longer a clone */
}
unclone_ctx(ctx);
spin_unlock_irqrestore(&ctx->lock, flags);
}
@@ -1642,6 +1655,8 @@ static void free_counter(struct perf_counter *counter)
atomic_dec(&nr_mmap_counters);
if (counter->attr.comm)
atomic_dec(&nr_comm_counters);
if (counter->attr.task)
atomic_dec(&nr_task_counters);
}
if (counter->destroy)
@@ -1676,6 +1691,18 @@ static int perf_release(struct inode *inode, struct file *file)
return 0;
}
static u64 perf_counter_read_tree(struct perf_counter *counter)
{
struct perf_counter *child;
u64 total = 0;
total += perf_counter_read(counter);
list_for_each_entry(child, &counter->child_list, child_list)
total += perf_counter_read(child);
return total;
}
/*
* Read the performance counter - simple non blocking version for now
*/
@@ -1695,7 +1722,7 @@ perf_read_hw(struct perf_counter *counter, char __user *buf, size_t count)
WARN_ON_ONCE(counter->ctx->parent_ctx);
mutex_lock(&counter->child_mutex);
values[0] = perf_counter_read(counter);
values[0] = perf_counter_read_tree(counter);
n = 1;
if (counter->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
values[n++] = counter->total_time_enabled +
@@ -1704,7 +1731,7 @@ perf_read_hw(struct perf_counter *counter, char __user *buf, size_t count)
values[n++] = counter->total_time_running +
atomic64_read(&counter->child_total_time_running);
if (counter->attr.read_format & PERF_FORMAT_ID)
values[n++] = counter->id;
values[n++] = primary_counter_id(counter);
mutex_unlock(&counter->child_mutex);
if (count < n * sizeof(u64))
@@ -1811,8 +1838,6 @@ static int perf_counter_period(struct perf_counter *counter, u64 __user *arg)
counter->attr.sample_freq = value;
} else {
perf_log_period(counter, value);
counter->attr.sample_period = value;
counter->hw.sample_period = value;
}
@@ -2661,10 +2686,14 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
if (sample_type & PERF_SAMPLE_ID)
header.size += sizeof(u64);
if (sample_type & PERF_SAMPLE_STREAM_ID)
header.size += sizeof(u64);
if (sample_type & PERF_SAMPLE_CPU) {
header.size += sizeof(cpu_entry);
cpu_entry.cpu = raw_smp_processor_id();
cpu_entry.reserved = 0;
}
if (sample_type & PERF_SAMPLE_PERIOD)
@@ -2685,6 +2714,18 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
header.size += sizeof(u64);
}
if (sample_type & PERF_SAMPLE_RAW) {
int size = sizeof(u32);
if (data->raw)
size += data->raw->size;
else
size += sizeof(u32);
WARN_ON_ONCE(size & (sizeof(u64)-1));
header.size += size;
}
ret = perf_output_begin(&handle, counter, header.size, nmi, 1);
if (ret)
return;
@@ -2703,7 +2744,13 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
if (sample_type & PERF_SAMPLE_ADDR)
perf_output_put(&handle, data->addr);
if (sample_type & PERF_SAMPLE_ID)
if (sample_type & PERF_SAMPLE_ID) {
u64 id = primary_counter_id(counter);
perf_output_put(&handle, id);
}
if (sample_type & PERF_SAMPLE_STREAM_ID)
perf_output_put(&handle, counter->id);
if (sample_type & PERF_SAMPLE_CPU)
@@ -2726,7 +2773,7 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
if (sub != counter)
sub->pmu->read(sub);
group_entry.id = sub->id;
group_entry.id = primary_counter_id(sub);
group_entry.counter = atomic64_read(&sub->count);
perf_output_put(&handle, group_entry);
@@ -2742,6 +2789,22 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
}
}
if (sample_type & PERF_SAMPLE_RAW) {
if (data->raw) {
perf_output_put(&handle, data->raw->size);
perf_output_copy(&handle, data->raw->data, data->raw->size);
} else {
struct {
u32 size;
u32 data;
} raw = {
.size = sizeof(u32),
.data = 0,
};
perf_output_put(&handle, raw);
}
}
perf_output_end(&handle);
}
@@ -2786,15 +2849,8 @@ perf_counter_read_event(struct perf_counter *counter,
}
if (counter->attr.read_format & PERF_FORMAT_ID) {
u64 id;
event.header.size += sizeof(u64);
if (counter->parent)
id = counter->parent->id;
else
id = counter->id;
event.format[i++] = id;
event.format[i++] = primary_counter_id(counter);
}
ret = perf_output_begin(&handle, counter, event.header.size, 0, 0);
@@ -2806,48 +2862,56 @@ perf_counter_read_event(struct perf_counter *counter,
}
/*
* fork tracking
* task tracking -- fork/exit
*
* enabled by: attr.comm | attr.mmap | attr.task
*/
struct perf_fork_event {
struct task_struct *task;
struct perf_task_event {
struct task_struct *task;
struct perf_counter_context *task_ctx;
struct {
struct perf_event_header header;
u32 pid;
u32 ppid;
u32 tid;
u32 ptid;
} event;
};
static void perf_counter_fork_output(struct perf_counter *counter,
struct perf_fork_event *fork_event)
static void perf_counter_task_output(struct perf_counter *counter,
struct perf_task_event *task_event)
{
struct perf_output_handle handle;
int size = fork_event->event.header.size;
struct task_struct *task = fork_event->task;
int size = task_event->event.header.size;
struct task_struct *task = task_event->task;
int ret = perf_output_begin(&handle, counter, size, 0, 0);
if (ret)
return;
fork_event->event.pid = perf_counter_pid(counter, task);
fork_event->event.ppid = perf_counter_pid(counter, task->real_parent);
task_event->event.pid = perf_counter_pid(counter, task);
task_event->event.ppid = perf_counter_pid(counter, task->real_parent);
perf_output_put(&handle, fork_event->event);
task_event->event.tid = perf_counter_tid(counter, task);
task_event->event.ptid = perf_counter_tid(counter, task->real_parent);
perf_output_put(&handle, task_event->event);
perf_output_end(&handle);
}
static int perf_counter_fork_match(struct perf_counter *counter)
static int perf_counter_task_match(struct perf_counter *counter)
{
if (counter->attr.comm || counter->attr.mmap)
if (counter->attr.comm || counter->attr.mmap || counter->attr.task)
return 1;
return 0;
}
static void perf_counter_fork_ctx(struct perf_counter_context *ctx,
struct perf_fork_event *fork_event)
static void perf_counter_task_ctx(struct perf_counter_context *ctx,
struct perf_task_event *task_event)
{
struct perf_counter *counter;
@@ -2856,51 +2920,62 @@ static void perf_counter_fork_ctx(struct perf_counter_context *ctx,
rcu_read_lock();
list_for_each_entry_rcu(counter, &ctx->event_list, event_entry) {
if (perf_counter_fork_match(counter))
perf_counter_fork_output(counter, fork_event);
if (perf_counter_task_match(counter))
perf_counter_task_output(counter, task_event);
}
rcu_read_unlock();
}
static void perf_counter_fork_event(struct perf_fork_event *fork_event)
static void perf_counter_task_event(struct perf_task_event *task_event)
{
struct perf_cpu_context *cpuctx;
struct perf_counter_context *ctx;
struct perf_counter_context *ctx = task_event->task_ctx;
cpuctx = &get_cpu_var(perf_cpu_context);
perf_counter_fork_ctx(&cpuctx->ctx, fork_event);
perf_counter_task_ctx(&cpuctx->ctx, task_event);
put_cpu_var(perf_cpu_context);
rcu_read_lock();
/*
* doesn't really matter which of the child contexts the
* events ends up in.
*/
ctx = rcu_dereference(current->perf_counter_ctxp);
if (!ctx)
ctx = rcu_dereference(task_event->task->perf_counter_ctxp);
if (ctx)
perf_counter_fork_ctx(ctx, fork_event);
perf_counter_task_ctx(ctx, task_event);
rcu_read_unlock();
}
static void perf_counter_task(struct task_struct *task,
struct perf_counter_context *task_ctx,
int new)
{
struct perf_task_event task_event;
if (!atomic_read(&nr_comm_counters) &&
!atomic_read(&nr_mmap_counters) &&
!atomic_read(&nr_task_counters))
return;
task_event = (struct perf_task_event){
.task = task,
.task_ctx = task_ctx,
.event = {
.header = {
.type = new ? PERF_EVENT_FORK : PERF_EVENT_EXIT,
.misc = 0,
.size = sizeof(task_event.event),
},
/* .pid */
/* .ppid */
/* .tid */
/* .ptid */
},
};
perf_counter_task_event(&task_event);
}
void perf_counter_fork(struct task_struct *task)
{
struct perf_fork_event fork_event;
if (!atomic_read(&nr_comm_counters) &&
!atomic_read(&nr_mmap_counters))
return;
fork_event = (struct perf_fork_event){
.task = task,
.event = {
.header = {
.type = PERF_EVENT_FORK,
.size = sizeof(fork_event.event),
},
},
};
perf_counter_fork_event(&fork_event);
perf_counter_task(task, NULL, 1);
}
/*
@@ -2968,8 +3043,10 @@ static void perf_counter_comm_event(struct perf_comm_event *comm_event)
struct perf_cpu_context *cpuctx;
struct perf_counter_context *ctx;
unsigned int size;
char *comm = comm_event->task->comm;
char comm[TASK_COMM_LEN];
memset(comm, 0, sizeof(comm));
strncpy(comm, comm_event->task->comm, sizeof(comm));
size = ALIGN(strlen(comm)+1, sizeof(u64));
comm_event->comm = comm;
@@ -3004,8 +3081,16 @@ void perf_counter_comm(struct task_struct *task)
comm_event = (struct perf_comm_event){
.task = task,
/* .comm */
/* .comm_size */
.event = {
.header = { .type = PERF_EVENT_COMM, },
.header = {
.type = PERF_EVENT_COMM,
.misc = 0,
/* .size */
},
/* .pid */
/* .tid */
},
};
@@ -3088,8 +3173,15 @@ static void perf_counter_mmap_event(struct perf_mmap_event *mmap_event)
char *buf = NULL;
const char *name;
memset(tmp, 0, sizeof(tmp));
if (file) {
buf = kzalloc(PATH_MAX, GFP_KERNEL);
/*
* d_path works from the end of the buffer backwards, so we
* need to add enough zero bytes after the string to handle
* the 64bit alignment we do later.
*/
buf = kzalloc(PATH_MAX + sizeof(u64), GFP_KERNEL);
if (!buf) {
name = strncpy(tmp, "//enomem", sizeof(tmp));
goto got_name;
@@ -3100,9 +3192,11 @@ static void perf_counter_mmap_event(struct perf_mmap_event *mmap_event)
goto got_name;
}
} else {
name = arch_vma_name(mmap_event->vma);
if (name)
if (arch_vma_name(mmap_event->vma)) {
name = strncpy(tmp, arch_vma_name(mmap_event->vma),
sizeof(tmp));
goto got_name;
}
if (!vma->vm_mm) {
name = strncpy(tmp, "[vdso]", sizeof(tmp));
@@ -3147,8 +3241,16 @@ void __perf_counter_mmap(struct vm_area_struct *vma)
mmap_event = (struct perf_mmap_event){
.vma = vma,
/* .file_name */
/* .file_size */
.event = {
.header = { .type = PERF_EVENT_MMAP, },
.header = {
.type = PERF_EVENT_MMAP,
.misc = 0,
/* .size */
},
/* .pid */
/* .tid */
.start = vma->vm_start,
.len = vma->vm_end - vma->vm_start,
.pgoff = vma->vm_pgoff,
@@ -3158,49 +3260,6 @@ void __perf_counter_mmap(struct vm_area_struct *vma)
perf_counter_mmap_event(&mmap_event);
}
/*
* Log sample_period changes so that analyzing tools can re-normalize the
* event flow.
*/
struct freq_event {
struct perf_event_header header;
u64 time;
u64 id;
u64 period;
};
static void perf_log_period(struct perf_counter *counter, u64 period)
{
struct perf_output_handle handle;
struct freq_event event;
int ret;
if (counter->hw.sample_period == period)
return;
if (counter->attr.sample_type & PERF_SAMPLE_PERIOD)
return;
event = (struct freq_event) {
.header = {
.type = PERF_EVENT_PERIOD,
.misc = 0,
.size = sizeof(event),
},
.time = sched_clock(),
.id = counter->id,
.period = period,
};
ret = perf_output_begin(&handle, counter, sizeof(event), 1, 0);
if (ret)
return;
perf_output_put(&handle, event);
perf_output_end(&handle);
}
/*
* IRQ throttle logging
*/
@@ -3214,16 +3273,21 @@ static void perf_log_throttle(struct perf_counter *counter, int enable)
struct perf_event_header header;
u64 time;
u64 id;
u64 stream_id;
} throttle_event = {
.header = {
.type = PERF_EVENT_THROTTLE + 1,
.type = PERF_EVENT_THROTTLE,
.misc = 0,
.size = sizeof(throttle_event),
},
.time = sched_clock(),
.id = counter->id,
.time = sched_clock(),
.id = primary_counter_id(counter),
.stream_id = counter->id,
};
if (enable)
throttle_event.header.type = PERF_EVENT_UNTHROTTLE;
ret = perf_output_begin(&handle, counter, sizeof(throttle_event), 1, 0);
if (ret)
return;
@@ -3300,87 +3364,81 @@ int perf_counter_overflow(struct perf_counter *counter, int nmi,
* Generic software counter infrastructure
*/
static void perf_swcounter_update(struct perf_counter *counter)
/*
* We directly increment counter->count and keep a second value in
* counter->hw.period_left to count intervals. This period counter
* is kept in the range [-sample_period, 0] so that we can use the
* sign as trigger.
*/
static u64 perf_swcounter_set_period(struct perf_counter *counter)
{
struct hw_perf_counter *hwc = &counter->hw;
u64 prev, now;
s64 delta;
u64 period = hwc->last_period;
u64 nr, offset;
s64 old, val;
hwc->last_period = hwc->sample_period;
again:
prev = atomic64_read(&hwc->prev_count);
now = atomic64_read(&hwc->count);
if (atomic64_cmpxchg(&hwc->prev_count, prev, now) != prev)
old = val = atomic64_read(&hwc->period_left);
if (val < 0)
return 0;
nr = div64_u64(period + val, period);
offset = nr * period;
val -= offset;
if (atomic64_cmpxchg(&hwc->period_left, old, val) != old)
goto again;
delta = now - prev;
atomic64_add(delta, &counter->count);
atomic64_sub(delta, &hwc->period_left);
}
static void perf_swcounter_set_period(struct perf_counter *counter)
{
struct hw_perf_counter *hwc = &counter->hw;
s64 left = atomic64_read(&hwc->period_left);
s64 period = hwc->sample_period;
if (unlikely(left <= -period)) {
left = period;
atomic64_set(&hwc->period_left, left);
hwc->last_period = period;
}
if (unlikely(left <= 0)) {
left += period;
atomic64_add(period, &hwc->period_left);
hwc->last_period = period;
}
atomic64_set(&hwc->prev_count, -left);
atomic64_set(&hwc->count, -left);
}
static enum hrtimer_restart perf_swcounter_hrtimer(struct hrtimer *hrtimer)
{
enum hrtimer_restart ret = HRTIMER_RESTART;
struct perf_sample_data data;
struct perf_counter *counter;
u64 period;
counter = container_of(hrtimer, struct perf_counter, hw.hrtimer);
counter->pmu->read(counter);
data.addr = 0;
data.regs = get_irq_regs();
/*
* In case we exclude kernel IPs or are somehow not in interrupt
* context, provide the next best thing, the user IP.
*/
if ((counter->attr.exclude_kernel || !data.regs) &&
!counter->attr.exclude_user)
data.regs = task_pt_regs(current);
if (data.regs) {
if (perf_counter_overflow(counter, 0, &data))
ret = HRTIMER_NORESTART;
}
period = max_t(u64, 10000, counter->hw.sample_period);
hrtimer_forward_now(hrtimer, ns_to_ktime(period));
return ret;
return nr;
}
static void perf_swcounter_overflow(struct perf_counter *counter,
int nmi, struct perf_sample_data *data)
{
data->period = counter->hw.last_period;
struct hw_perf_counter *hwc = &counter->hw;
u64 overflow;
perf_swcounter_update(counter);
perf_swcounter_set_period(counter);
if (perf_counter_overflow(counter, nmi, data))
/* soft-disable the counter */
;
data->period = counter->hw.last_period;
overflow = perf_swcounter_set_period(counter);
if (hwc->interrupts == MAX_INTERRUPTS)
return;
for (; overflow; overflow--) {
if (perf_counter_overflow(counter, nmi, data)) {
/*
* We inhibit the overflow from happening when
* hwc->interrupts == MAX_INTERRUPTS.
*/
break;
}
}
}
static void perf_swcounter_unthrottle(struct perf_counter *counter)
{
/*
* Nothing to do, we already reset hwc->interrupts.
*/
}
static void perf_swcounter_add(struct perf_counter *counter, u64 nr,
int nmi, struct perf_sample_data *data)
{
struct hw_perf_counter *hwc = &counter->hw;
atomic64_add(nr, &counter->count);
if (!hwc->sample_period)
return;
if (!data->regs)
return;
if (!atomic64_add_negative(nr, &hwc->period_left))
perf_swcounter_overflow(counter, nmi, data);
}
static int perf_swcounter_is_counting(struct perf_counter *counter)
@@ -3444,15 +3502,6 @@ static int perf_swcounter_match(struct perf_counter *counter,
return 1;
}
static void perf_swcounter_add(struct perf_counter *counter, u64 nr,
int nmi, struct perf_sample_data *data)
{
int neg = atomic64_add_negative(nr, &counter->hw.count);
if (counter->hw.sample_period && !neg && data->regs)
perf_swcounter_overflow(counter, nmi, data);
}
static void perf_swcounter_ctx_event(struct perf_counter_context *ctx,
enum perf_type_id type,
u32 event, u64 nr, int nmi,
@@ -3531,26 +3580,65 @@ void __perf_swcounter_event(u32 event, u64 nr, int nmi,
static void perf_swcounter_read(struct perf_counter *counter)
{
perf_swcounter_update(counter);
}
static int perf_swcounter_enable(struct perf_counter *counter)
{
perf_swcounter_set_period(counter);
struct hw_perf_counter *hwc = &counter->hw;
if (hwc->sample_period) {
hwc->last_period = hwc->sample_period;
perf_swcounter_set_period(counter);
}
return 0;
}
static void perf_swcounter_disable(struct perf_counter *counter)
{
perf_swcounter_update(counter);
}
static const struct pmu perf_ops_generic = {
.enable = perf_swcounter_enable,
.disable = perf_swcounter_disable,
.read = perf_swcounter_read,
.unthrottle = perf_swcounter_unthrottle,
};
/*
* hrtimer based swcounter callback
*/
static enum hrtimer_restart perf_swcounter_hrtimer(struct hrtimer *hrtimer)
{
enum hrtimer_restart ret = HRTIMER_RESTART;
struct perf_sample_data data;
struct perf_counter *counter;
u64 period;
counter = container_of(hrtimer, struct perf_counter, hw.hrtimer);
counter->pmu->read(counter);
data.addr = 0;
data.regs = get_irq_regs();
/*
* In case we exclude kernel IPs or are somehow not in interrupt
* context, provide the next best thing, the user IP.
*/
if ((counter->attr.exclude_kernel || !data.regs) &&
!counter->attr.exclude_user)
data.regs = task_pt_regs(current);
if (data.regs) {
if (perf_counter_overflow(counter, 0, &data))
ret = HRTIMER_NORESTART;
}
period = max_t(u64, 10000, counter->hw.sample_period);
hrtimer_forward_now(hrtimer, ns_to_ktime(period));
return ret;
}
/*
* Software counter: cpu wall time clock
*/
@@ -3668,17 +3756,24 @@ static const struct pmu perf_ops_task_clock = {
};
#ifdef CONFIG_EVENT_PROFILE
void perf_tpcounter_event(int event_id)
void perf_tpcounter_event(int event_id, u64 addr, u64 count, void *record,
int entry_size)
{
struct perf_raw_record raw = {
.size = entry_size,
.data = record,
};
struct perf_sample_data data = {
.regs = get_irq_regs();
.addr = 0,
.regs = get_irq_regs(),
.addr = addr,
.raw = &raw,
};
if (!data.regs)
data.regs = task_pt_regs(current);
do_perf_swcounter_event(PERF_TYPE_TRACEPOINT, event_id, 1, 1, &data);
do_perf_swcounter_event(PERF_TYPE_TRACEPOINT, event_id, count, 1, &data);
}
EXPORT_SYMBOL_GPL(perf_tpcounter_event);
@@ -3687,16 +3782,20 @@ extern void ftrace_profile_disable(int);
static void tp_perf_counter_destroy(struct perf_counter *counter)
{
ftrace_profile_disable(perf_event_id(&counter->attr));
ftrace_profile_disable(counter->attr.config);
}
static const struct pmu *tp_perf_counter_init(struct perf_counter *counter)
{
int event_id = perf_event_id(&counter->attr);
int ret;
/*
* Raw tracepoint data is a severe data leak, only allow root to
* have these.
*/
if ((counter->attr.sample_type & PERF_SAMPLE_RAW) &&
!capable(CAP_SYS_ADMIN))
return ERR_PTR(-EPERM);
ret = ftrace_profile_enable(event_id);
if (ret)
if (ftrace_profile_enable(counter->attr.config))
return NULL;
counter->destroy = tp_perf_counter_destroy;
@@ -3874,6 +3973,8 @@ done:
atomic_inc(&nr_mmap_counters);
if (counter->attr.comm)
atomic_inc(&nr_comm_counters);
if (counter->attr.task)
atomic_inc(&nr_task_counters);
}
return counter;
@@ -4235,8 +4336,10 @@ void perf_counter_exit_task(struct task_struct *child)
struct perf_counter_context *child_ctx;
unsigned long flags;
if (likely(!child->perf_counter_ctxp))
if (likely(!child->perf_counter_ctxp)) {
perf_counter_task(child, NULL, 0);
return;
}
local_irq_save(flags);
/*
@@ -4255,17 +4358,20 @@ void perf_counter_exit_task(struct task_struct *child)
*/
spin_lock(&child_ctx->lock);
child->perf_counter_ctxp = NULL;
if (child_ctx->parent_ctx) {
/*
* This context is a clone; unclone it so it can't get
* swapped to another process while we're removing all
* the counters from it.
*/
put_ctx(child_ctx->parent_ctx);
child_ctx->parent_ctx = NULL;
}
spin_unlock(&child_ctx->lock);
local_irq_restore(flags);
/*
* If this context is a clone; unclone it so it can't get
* swapped to another process while we're removing all
* the counters from it.
*/
unclone_ctx(child_ctx);
spin_unlock_irqrestore(&child_ctx->lock, flags);
/*
* Report the task dead after unscheduling the counters so that we
* won't get any samples after PERF_EVENT_EXIT. We can however still
* get a few PERF_EVENT_READ events.
*/
perf_counter_task(child, child_ctx, 0);
/*
* We can recurse on the same lock type through:
kernel/posix-cpu-timers.c
+4 -3
View File
@@ -521,11 +521,12 @@ void posix_cpu_timers_exit(struct task_struct *tsk)
}
void posix_cpu_timers_exit_group(struct task_struct *tsk)
{
struct task_cputime cputime;
struct signal_struct *const sig = tsk->signal;
thread_group_cputimer(tsk, &cputime);
cleanup_timers(tsk->signal->cpu_timers,
cputime.utime, cputime.stime, cputime.sum_exec_runtime);
cputime_add(tsk->utime, sig->utime),
cputime_add(tsk->stime, sig->stime),
tsk->se.sum_exec_runtime + sig->sum_sched_runtime);
}
static void clear_dead_task(struct k_itimer *timer, union cpu_time_count now)
kernel/posix-timers.c
+7
View File
@@ -202,6 +202,12 @@ static int no_timer_create(struct k_itimer *new_timer)
return -EOPNOTSUPP;
}
static int no_nsleep(const clockid_t which_clock, int flags,
struct timespec *tsave, struct timespec __user *rmtp)
{
return -EOPNOTSUPP;
}
/*
* Return nonzero if we know a priori this clockid_t value is bogus.
*/
@@ -254,6 +260,7 @@ static __init int init_posix_timers(void)
.clock_get = posix_get_monotonic_raw,
.clock_set = do_posix_clock_nosettime,
.timer_create = no_timer_create,
.nsleep = no_nsleep,
};
register_posix_clock(CLOCK_REALTIME, &clock_realtime);
kernel/profile.c
+3 -2
View File
@@ -117,11 +117,12 @@ int __ref profile_init(void)
cpumask_copy(prof_cpu_mask, cpu_possible_mask);
prof_buffer = kzalloc(buffer_bytes, GFP_KERNEL);
prof_buffer = kzalloc(buffer_bytes, GFP_KERNEL|__GFP_NOWARN);
if (prof_buffer)
return 0;
prof_buffer = alloc_pages_exact(buffer_bytes, GFP_KERNEL|__GFP_ZERO);
prof_buffer = alloc_pages_exact(buffer_bytes,
GFP_KERNEL|__GFP_ZERO|__GFP_NOWARN);
if (prof_buffer)
return 0;
kernel/rtmutex.c
+1 -3
View File
@@ -1039,16 +1039,14 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock,
if (!rt_mutex_owner(lock) || try_to_steal_lock(lock, task)) {
/* We got the lock for task. */
debug_rt_mutex_lock(lock);
rt_mutex_set_owner(lock, task, 0);
spin_unlock(&lock->wait_lock);
rt_mutex_deadlock_account_lock(lock, task);
return 1;
}
ret = task_blocks_on_rt_mutex(lock, waiter, task, detect_deadlock);
if (ret && !waiter->task) {
/*
* Reset the return value. We might have
kernel/sched.c
+2 -2
View File
@@ -7289,6 +7289,7 @@ static void migrate_dead_tasks(unsigned int dead_cpu)
static void calc_global_load_remove(struct rq *rq)
{
atomic_long_sub(rq->calc_load_active, &calc_load_tasks);
rq->calc_load_active = 0;
}
#endif /* CONFIG_HOTPLUG_CPU */
@@ -7515,6 +7516,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
task_rq_unlock(rq, &flags);
get_task_struct(p);
cpu_rq(cpu)->migration_thread = p;
rq->calc_load_update = calc_load_update;
break;
case CPU_ONLINE:
@@ -7525,8 +7527,6 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
/* Update our root-domain */
rq = cpu_rq(cpu);
spin_lock_irqsave(&rq->lock, flags);
rq->calc_load_update = calc_load_update;
rq->calc_load_active = 0;
if (rq->rd) {
BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
kernel/sched_cpupri.c
+14 -1
View File
@@ -81,8 +81,21 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p,
if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids)
continue;
if (lowest_mask)
if (lowest_mask) {
cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask);
/*
* We have to ensure that we have at least one bit
* still set in the array, since the map could have
* been concurrently emptied between the first and
* second reads of vec->mask. If we hit this
* condition, simply act as though we never hit this
* priority level and continue on.
*/
if (cpumask_any(lowest_mask) >= nr_cpu_ids)
continue;
}
return 1;
}
kernel/sched_fair.c
+27 -15
View File
@@ -266,6 +266,12 @@ static inline u64 min_vruntime(u64 min_vruntime, u64 vruntime)
return min_vruntime;
}
static inline int entity_before(struct sched_entity *a,
struct sched_entity *b)
{
return (s64)(a->vruntime - b->vruntime) < 0;
}
static inline s64 entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
return se->vruntime - cfs_rq->min_vruntime;
@@ -605,9 +611,13 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se)
static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
#ifdef CONFIG_SCHEDSTATS
struct task_struct *tsk = NULL;
if (entity_is_task(se))
tsk = task_of(se);
if (se->sleep_start) {
u64 delta = rq_of(cfs_rq)->clock - se->sleep_start;
struct task_struct *tsk = task_of(se);
if ((s64)delta < 0)
delta = 0;
@@ -618,11 +628,11 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
se->sleep_start = 0;
se->sum_sleep_runtime += delta;
account_scheduler_latency(tsk, delta >> 10, 1);
if (tsk)
account_scheduler_latency(tsk, delta >> 10, 1);
}
if (se->block_start) {
u64 delta = rq_of(cfs_rq)->clock - se->block_start;
struct task_struct *tsk = task_of(se);
if ((s64)delta < 0)
delta = 0;
@@ -633,17 +643,19 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
se->block_start = 0;
se->sum_sleep_runtime += delta;
/*
* Blocking time is in units of nanosecs, so shift by 20 to
* get a milliseconds-range estimation of the amount of
* time that the task spent sleeping:
*/
if (unlikely(prof_on == SLEEP_PROFILING)) {
profile_hits(SLEEP_PROFILING, (void *)get_wchan(tsk),
delta >> 20);
if (tsk) {
/*
* Blocking time is in units of nanosecs, so shift by
* 20 to get a milliseconds-range estimation of the
* amount of time that the task spent sleeping:
*/
if (unlikely(prof_on == SLEEP_PROFILING)) {
profile_hits(SLEEP_PROFILING,
(void *)get_wchan(tsk),
delta >> 20);
}
account_scheduler_latency(tsk, delta >> 10, 0);
}
account_scheduler_latency(tsk, delta >> 10, 0);
}
#endif
}
@@ -1017,7 +1029,7 @@ static void yield_task_fair(struct rq *rq)
/*
* Already in the rightmost position?
*/
if (unlikely(!rightmost || rightmost->vruntime < se->vruntime))
if (unlikely(!rightmost || entity_before(rightmost, se)))
return;
/*
@@ -1713,7 +1725,7 @@ static void task_new_fair(struct rq *rq, struct task_struct *p)
/* 'curr' will be NULL if the child belongs to a different group */
if (sysctl_sched_child_runs_first && this_cpu == task_cpu(p) &&
curr && curr->vruntime < se->vruntime) {
curr && entity_before(curr, se)) {
/*
* Upon rescheduling, sched_class::put_prev_task() will place
* 'current' within the tree based on its new key value.
kernel/signal.c
+14 -11
View File
@@ -2454,11 +2454,9 @@ do_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, unsigned long s
stack_t oss;
int error;
if (uoss) {
oss.ss_sp = (void __user *) current->sas_ss_sp;
oss.ss_size = current->sas_ss_size;
oss.ss_flags = sas_ss_flags(sp);
}
oss.ss_sp = (void __user *) current->sas_ss_sp;
oss.ss_size = current->sas_ss_size;
oss.ss_flags = sas_ss_flags(sp);
if (uss) {
void __user *ss_sp;
@@ -2466,10 +2464,12 @@ do_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, unsigned long s
int ss_flags;
error = -EFAULT;
if (!access_ok(VERIFY_READ, uss, sizeof(*uss))
|| __get_user(ss_sp, &uss->ss_sp)
|| __get_user(ss_flags, &uss->ss_flags)
|| __get_user(ss_size, &uss->ss_size))
if (!access_ok(VERIFY_READ, uss, sizeof(*uss)))
goto out;
error = __get_user(ss_sp, &uss->ss_sp) |
__get_user(ss_flags, &uss->ss_flags) |
__get_user(ss_size, &uss->ss_size);
if (error)
goto out;
error = -EPERM;
@@ -2501,13 +2501,16 @@ do_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, unsigned long s
current->sas_ss_size = ss_size;
}
error = 0;
if (uoss) {
error = -EFAULT;
if (copy_to_user(uoss, &oss, sizeof(oss)))
if (!access_ok(VERIFY_WRITE, uoss, sizeof(*uoss)))
goto out;
error = __put_user(oss.ss_sp, &uoss->ss_sp) |
__put_user(oss.ss_size, &uoss->ss_size) |
__put_user(oss.ss_flags, &uoss->ss_flags);
}
error = 0;
out:
return error;
}
kernel/smp.c
+1 -1
View File
@@ -57,7 +57,7 @@ hotplug_cfd(struct notifier_block *nfb, unsigned long action, void *hcpu)
return NOTIFY_BAD;
break;
#ifdef CONFIG_CPU_HOTPLUG
#ifdef CONFIG_HOTPLUG_CPU
case CPU_UP_CANCELED:
case CPU_UP_CANCELED_FROZEN:
kernel/softirq.c
+63 -1
View File
@@ -345,7 +345,9 @@ void open_softirq(int nr, void (*action)(struct softirq_action *))
softirq_vec[nr].action = action;
}
/* Tasklets */
/*
* Tasklets
*/
struct tasklet_head
{
struct tasklet_struct *head;
@@ -493,6 +495,66 @@ void tasklet_kill(struct tasklet_struct *t)
EXPORT_SYMBOL(tasklet_kill);
/*
* tasklet_hrtimer
*/
/*
* The trampoline is called when the hrtimer expires. If this is
* called from the hrtimer interrupt then we schedule the tasklet as
* the timer callback function expects to run in softirq context. If
* it's called in softirq context anyway (i.e. high resolution timers
* disabled) then the hrtimer callback is called right away.
*/
static enum hrtimer_restart __hrtimer_tasklet_trampoline(struct hrtimer *timer)
{
struct tasklet_hrtimer *ttimer =
container_of(timer, struct tasklet_hrtimer, timer);
if (hrtimer_is_hres_active(timer)) {
tasklet_hi_schedule(&ttimer->tasklet);
return HRTIMER_NORESTART;
}
return ttimer->function(timer);
}
/*
* Helper function which calls the hrtimer callback from
* tasklet/softirq context
*/
static void __tasklet_hrtimer_trampoline(unsigned long data)
{
struct tasklet_hrtimer *ttimer = (void *)data;
enum hrtimer_restart restart;
restart = ttimer->function(&ttimer->timer);
if (restart != HRTIMER_NORESTART)
hrtimer_restart(&ttimer->timer);
}
/**
* tasklet_hrtimer_init - Init a tasklet/hrtimer combo for softirq callbacks
* @ttimer: tasklet_hrtimer which is initialized
* @function: hrtimer callback funtion which gets called from softirq context
* @which_clock: clock id (CLOCK_MONOTONIC/CLOCK_REALTIME)
* @mode: hrtimer mode (HRTIMER_MODE_ABS/HRTIMER_MODE_REL)
*/
void tasklet_hrtimer_init(struct tasklet_hrtimer *ttimer,
enum hrtimer_restart (*function)(struct hrtimer *),
clockid_t which_clock, enum hrtimer_mode mode)
{
hrtimer_init(&ttimer->timer, which_clock, mode);
ttimer->timer.function = __hrtimer_tasklet_trampoline;
tasklet_init(&ttimer->tasklet, __tasklet_hrtimer_trampoline,
(unsigned long)ttimer);
ttimer->function = function;
}
EXPORT_SYMBOL_GPL(tasklet_hrtimer_init);
/*
* Remote softirq bits
*/
DEFINE_PER_CPU(struct list_head [NR_SOFTIRQS], softirq_work_list);
EXPORT_PER_CPU_SYMBOL(softirq_work_list);
kernel/time/clocksource.c
+1 -1
View File
@@ -513,7 +513,7 @@ static ssize_t sysfs_override_clocksource(struct sys_device *dev,
* Check to make sure we don't switch to a non-highres capable
* clocksource if the tick code is in oneshot mode (highres or nohz)
*/
if (tick_oneshot_mode_active() &&
if (tick_oneshot_mode_active() && ovr &&
!(ovr->flags & CLOCK_SOURCE_VALID_FOR_HRES)) {
printk(KERN_WARNING "%s clocksource is not HRT compatible. "
"Cannot switch while in HRT/NOHZ mode\n", ovr->name);
kernel/timer.c
+1 -1
View File
@@ -714,7 +714,7 @@ int mod_timer(struct timer_list *timer, unsigned long expires)
* networking code - if the timer is re-modified
* to be the same thing then just return:
*/
if (timer->expires == expires && timer_pending(timer))
if (timer_pending(timer) && timer->expires == expires)
return 1;
return __mod_timer(timer, expires, false, TIMER_NOT_PINNED);
kernel/trace/ftrace.c
+14 -5
View File
@@ -1617,7 +1617,7 @@ ftrace_regex_open(struct inode *inode, struct file *file, int enable)
mutex_lock(&ftrace_regex_lock);
if ((file->f_mode & FMODE_WRITE) &&
!(file->f_flags & O_APPEND))
(file->f_flags & O_TRUNC))
ftrace_filter_reset(enable);
if (file->f_mode & FMODE_READ) {
@@ -2527,7 +2527,7 @@ ftrace_graph_open(struct inode *inode, struct file *file)
mutex_lock(&graph_lock);
if ((file->f_mode & FMODE_WRITE) &&
!(file->f_flags & O_APPEND)) {
(file->f_flags & O_TRUNC)) {
ftrace_graph_count = 0;
memset(ftrace_graph_funcs, 0, sizeof(ftrace_graph_funcs));
}
@@ -2545,6 +2545,14 @@ ftrace_graph_open(struct inode *inode, struct file *file)
return ret;
}
static int
ftrace_graph_release(struct inode *inode, struct file *file)
{
if (file->f_mode & FMODE_READ)
seq_release(inode, file);
return 0;
}
static int
ftrace_set_func(unsigned long *array, int *idx, char *buffer)
{
@@ -2674,9 +2682,10 @@ ftrace_graph_write(struct file *file, const char __user *ubuf,
}
static const struct file_operations ftrace_graph_fops = {
.open = ftrace_graph_open,
.read = seq_read,
.write = ftrace_graph_write,
.open = ftrace_graph_open,
.read = seq_read,
.write = ftrace_graph_write,
.release = ftrace_graph_release,
};
#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
kernel/trace/ring_buffer.c
+7 -8
View File
@@ -1170,6 +1170,7 @@ ring_buffer_free(struct ring_buffer *buffer)
put_online_cpus();
kfree(buffer->buffers);
free_cpumask_var(buffer->cpumask);
kfree(buffer);
@@ -2379,7 +2380,7 @@ void ring_buffer_discard_commit(struct ring_buffer *buffer,
*/
RB_WARN_ON(buffer, !local_read(&cpu_buffer->committing));
if (!rb_try_to_discard(cpu_buffer, event))
if (rb_try_to_discard(cpu_buffer, event))
goto out;
/*
@@ -2990,7 +2991,6 @@ rb_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts)
* the box. Return the padding, and we will release
* the current locks, and try again.
*/
rb_advance_reader(cpu_buffer);
return event;
case RINGBUF_TYPE_TIME_EXTEND:
@@ -3093,7 +3093,7 @@ static inline int rb_ok_to_lock(void)
* buffer too. A one time deal is all you get from reading
* the ring buffer from an NMI.
*/
if (likely(!in_nmi() && !oops_in_progress))
if (likely(!in_nmi()))
return 1;
tracing_off_permanent();
@@ -3126,6 +3126,8 @@ ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts)
if (dolock)
spin_lock(&cpu_buffer->reader_lock);
event = rb_buffer_peek(buffer, cpu, ts);
if (event && event->type_len == RINGBUF_TYPE_PADDING)
rb_advance_reader(cpu_buffer);
if (dolock)
spin_unlock(&cpu_buffer->reader_lock);
local_irq_restore(flags);
@@ -3197,12 +3199,9 @@ ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts)
spin_lock(&cpu_buffer->reader_lock);
event = rb_buffer_peek(buffer, cpu, ts);
if (!event)
goto out_unlock;
if (event)
rb_advance_reader(cpu_buffer);
rb_advance_reader(cpu_buffer);
out_unlock:
if (dolock)
spin_unlock(&cpu_buffer->reader_lock);
local_irq_restore(flags);
kernel/trace/trace.c
+9 -4
View File
@@ -848,6 +848,7 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags,
((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) |
(need_resched() ? TRACE_FLAG_NEED_RESCHED : 0);
}
EXPORT_SYMBOL_GPL(tracing_generic_entry_update);
struct ring_buffer_event *trace_buffer_lock_reserve(struct trace_array *tr,
int type,
@@ -1857,7 +1858,7 @@ static int tracing_open(struct inode *inode, struct file *file)
/* If this file was open for write, then erase contents */
if ((file->f_mode & FMODE_WRITE) &&
!(file->f_flags & O_APPEND)) {
(file->f_flags & O_TRUNC)) {
long cpu = (long) inode->i_private;
if (cpu == TRACE_PIPE_ALL_CPU)
@@ -2911,7 +2912,8 @@ tracing_fill_pipe_page(size_t rem, struct trace_iterator *iter)
break;
}
trace_consume(iter);
if (ret != TRACE_TYPE_NO_CONSUME)
trace_consume(iter);
rem -= count;
if (!find_next_entry_inc(iter)) {
rem = 0;
@@ -4056,8 +4058,11 @@ static void __ftrace_dump(bool disable_tracing)
iter.pos = -1;
if (find_next_entry_inc(&iter) != NULL) {
print_trace_line(&iter);
trace_consume(&iter);
int ret;
ret = print_trace_line(&iter);
if (ret != TRACE_TYPE_NO_CONSUME)
trace_consume(&iter);
}
trace_printk_seq(&iter.seq);
kernel/trace/trace.h
-4
View File
@@ -438,10 +438,6 @@ struct trace_entry *tracing_get_trace_entry(struct trace_array *tr,
struct trace_entry *trace_find_next_entry(struct trace_iterator *iter,
int *ent_cpu, u64 *ent_ts);
void tracing_generic_entry_update(struct trace_entry *entry,
unsigned long flags,
int pc);
void default_wait_pipe(struct trace_iterator *iter);
void poll_wait_pipe(struct trace_iterator *iter);
kernel/trace/trace_event_profile.c
+1 -1
View File
@@ -14,7 +14,7 @@ int ftrace_profile_enable(int event_id)
mutex_lock(&event_mutex);
list_for_each_entry(event, &ftrace_events, list) {
if (event->id == event_id) {
if (event->id == event_id && event->profile_enable) {
ret = event->profile_enable(event);
break;
}
kernel/trace/trace_events.c
+2 -2
View File
@@ -378,7 +378,7 @@ ftrace_event_seq_open(struct inode *inode, struct file *file)
const struct seq_operations *seq_ops;
if ((file->f_mode & FMODE_WRITE) &&
!(file->f_flags & O_APPEND))
(file->f_flags & O_TRUNC))
ftrace_clear_events();
seq_ops = inode->i_private;
@@ -945,7 +945,7 @@ event_create_dir(struct ftrace_event_call *call, struct dentry *d_events,
entry = trace_create_file("enable", 0644, call->dir, call,
enable);
if (call->id)
if (call->id && call->profile_enable)
entry = trace_create_file("id", 0444, call->dir, call,
id);
kernel/trace/trace_functions_graph.c
+9 -2
View File
@@ -927,9 +927,16 @@ print_graph_function(struct trace_iterator *iter)
switch (entry->type) {
case TRACE_GRAPH_ENT: {
struct ftrace_graph_ent_entry *field;
/*
* print_graph_entry() may consume the current event,
* thus @field may become invalid, so we need to save it.
* sizeof(struct ftrace_graph_ent_entry) is very small,
* it can be safely saved at the stack.
*/
struct ftrace_graph_ent_entry *field, saved;
trace_assign_type(field, entry);
return print_graph_entry(field, s, iter);
saved = *field;
return print_graph_entry(&saved, s, iter);
}
case TRACE_GRAPH_RET: {
struct ftrace_graph_ret_entry *field;
kernel/trace/trace_printk.c
+1 -1
View File
@@ -176,7 +176,7 @@ static int t_show(struct seq_file *m, void *v)
const char *str = *fmt;
int i;
seq_printf(m, "0x%lx : \"", (unsigned long)fmt);
seq_printf(m, "0x%lx : \"", *(unsigned long *)fmt);
/*
* Tabs and new lines need to be converted.
kernel/trace/trace_stack.c
+2 -5
View File
@@ -294,17 +294,14 @@ static const struct seq_operations stack_trace_seq_ops = {
static int stack_trace_open(struct inode *inode, struct file *file)
{
int ret;
ret = seq_open(file, &stack_trace_seq_ops);
return ret;
return seq_open(file, &stack_trace_seq_ops);
}
static const struct file_operations stack_trace_fops = {
.open = stack_trace_open,
.read = seq_read,
.llseek = seq_lseek,
.release = seq_release,
};
int
kernel/trace/trace_stat.c
+22 -12
View File
@@ -76,7 +76,7 @@ static struct rb_node *release_next(struct tracer_stat *ts,
}
}
static void reset_stat_session(struct stat_session *session)
static void __reset_stat_session(struct stat_session *session)
{
struct rb_node *node = session->stat_root.rb_node;
@@ -86,10 +86,17 @@ static void reset_stat_session(struct stat_session *session)
session->stat_root = RB_ROOT;
}
static void reset_stat_session(struct stat_session *session)
{
mutex_lock(&session->stat_mutex);
__reset_stat_session(session);
mutex_unlock(&session->stat_mutex);
}
static void destroy_session(struct stat_session *session)
{
debugfs_remove(session->file);
reset_stat_session(session);
__reset_stat_session(session);
mutex_destroy(&session->stat_mutex);
kfree(session);
}
@@ -153,7 +160,7 @@ static int stat_seq_init(struct stat_session *session)
int i;
mutex_lock(&session->stat_mutex);
reset_stat_session(session);
__reset_stat_session(session);
if (!ts->stat_cmp)
ts->stat_cmp = dummy_cmp;
@@ -186,7 +193,7 @@ exit:
return ret;
exit_free_rbtree:
reset_stat_session(session);
__reset_stat_session(session);
mutex_unlock(&session->stat_mutex);
return ret;
}
@@ -253,16 +260,21 @@ static const struct seq_operations trace_stat_seq_ops = {
static int tracing_stat_open(struct inode *inode, struct file *file)
{
int ret;
struct seq_file *m;
struct stat_session *session = inode->i_private;
ret = stat_seq_init(session);
if (ret)
return ret;
ret = seq_open(file, &trace_stat_seq_ops);
if (!ret) {
struct seq_file *m = file->private_data;
m->private = session;
ret = stat_seq_init(session);
if (ret) {
reset_stat_session(session);
return ret;
}
m = file->private_data;
m->private = session;
return ret;
}
@@ -273,11 +285,9 @@ static int tracing_stat_release(struct inode *i, struct file *f)
{
struct stat_session *session = i->i_private;
mutex_lock(&session->stat_mutex);
reset_stat_session(session);
mutex_unlock(&session->stat_mutex);
return 0;
return seq_release(i, f);
}
static const struct file_operations tracing_stat_fops = {