#include <linux/math64.h>
#include <asm/uaccess.h>
#include <linux/kernel_stat.h>
+#include <trace/events/timer.h>
/*
* Called after updating RLIMIT_CPU to set timer expiration if necessary.
*/
void update_rlimit_cpu(unsigned long rlim_new)
{
- cputime_t cputime;
+ cputime_t cputime = secs_to_cputime(rlim_new);
+ struct signal_struct *const sig = current->signal;
- cputime = secs_to_cputime(rlim_new);
- if (cputime_eq(current->signal->it_prof_expires, cputime_zero) ||
- cputime_lt(current->signal->it_prof_expires, cputime)) {
+ if (cputime_eq(sig->it[CPUCLOCK_PROF].expires, cputime_zero) ||
+ cputime_gt(sig->it[CPUCLOCK_PROF].expires, cputime)) {
spin_lock_irq(¤t->sighand->siglock);
set_process_cpu_timer(current, CPUCLOCK_PROF, &cputime, NULL);
spin_unlock_irq(¤t->sighand->siglock);
cpu->cpu = virt_ticks(p);
break;
case CPUCLOCK_SCHED:
- cpu->sched = p->se.sum_exec_runtime + task_delta_exec(p);
+ cpu->sched = task_sched_runtime(p);
break;
}
return 0;
rcu_read_unlock();
}
+static void update_gt_cputime(struct task_cputime *a, struct task_cputime *b)
+{
+ if (cputime_gt(b->utime, a->utime))
+ a->utime = b->utime;
+
+ if (cputime_gt(b->stime, a->stime))
+ a->stime = b->stime;
+
+ if (b->sum_exec_runtime > a->sum_exec_runtime)
+ a->sum_exec_runtime = b->sum_exec_runtime;
+}
+
+void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times)
+{
+ struct thread_group_cputimer *cputimer = &tsk->signal->cputimer;
+ struct task_cputime sum;
+ unsigned long flags;
+
+ spin_lock_irqsave(&cputimer->lock, flags);
+ if (!cputimer->running) {
+ cputimer->running = 1;
+ /*
+ * The POSIX timer interface allows for absolute time expiry
+ * values through the TIMER_ABSTIME flag, therefore we have
+ * to synchronize the timer to the clock every time we start
+ * it.
+ */
+ thread_group_cputime(tsk, &sum);
+ update_gt_cputime(&cputimer->cputime, &sum);
+ }
+ *times = cputimer->cputime;
+ spin_unlock_irqrestore(&cputimer->lock, flags);
+}
+
/*
* Sample a process (thread group) clock for the given group_leader task.
* Must be called with tasklist_lock held for reading.
{
struct task_cputime cputime;
- thread_group_cputime(p, &cputime);
switch (CPUCLOCK_WHICH(which_clock)) {
default:
return -EINVAL;
case CPUCLOCK_PROF:
+ thread_group_cputime(p, &cputime);
cpu->cpu = cputime_add(cputime.utime, cputime.stime);
break;
case CPUCLOCK_VIRT:
+ thread_group_cputime(p, &cputime);
cpu->cpu = cputime.utime;
break;
case CPUCLOCK_SCHED:
- cpu->sched = cputime.sum_exec_runtime + task_delta_exec(p);
+ cpu->sched = thread_group_sched_runtime(p);
break;
}
return 0;
/*
* Validate the clockid_t for a new CPU-clock timer, and initialize the timer.
- * This is called from sys_timer_create with the new timer already locked.
+ * This is called from sys_timer_create() and do_cpu_nanosleep() with the
+ * new timer already all-zeros initialized.
*/
int posix_cpu_timer_create(struct k_itimer *new_timer)
{
return -EINVAL;
INIT_LIST_HEAD(&new_timer->it.cpu.entry);
- new_timer->it.cpu.incr.sched = 0;
- new_timer->it.cpu.expires.sched = 0;
read_lock(&tasklist_lock);
if (CPUCLOCK_PERTHREAD(new_timer->it_clock)) {
}
void posix_cpu_timers_exit_group(struct task_struct *tsk)
{
- struct task_cputime cputime;
+ struct signal_struct *const sig = tsk->signal;
- thread_group_cputime(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)
now);
}
-/*
- * Enable the process wide cpu timer accounting.
- *
- * serialized using ->sighand->siglock
- */
-static void start_process_timers(struct task_struct *tsk)
+static inline int expires_gt(cputime_t expires, cputime_t new_exp)
{
- tsk->signal->cputimer.running = 1;
- barrier();
+ return cputime_eq(expires, cputime_zero) ||
+ cputime_gt(expires, new_exp);
}
-/*
- * Release the process wide timer accounting -- timer stops ticking when
- * nobody cares about it.
- *
- * serialized using ->sighand->siglock
- */
-static void stop_process_timers(struct task_struct *tsk)
+static inline int expires_le(cputime_t expires, cputime_t new_exp)
{
- tsk->signal->cputimer.running = 0;
- barrier();
+ return !cputime_eq(expires, cputime_zero) &&
+ cputime_le(expires, new_exp);
}
-
/*
* Insert the timer on the appropriate list before any timers that
* expire later. This must be called with the tasklist_lock held
BUG_ON(!irqs_disabled());
spin_lock(&p->sighand->siglock);
- if (!CPUCLOCK_PERTHREAD(timer->it_clock))
- start_process_timers(p);
-
listpos = head;
if (CPUCLOCK_WHICH(timer->it_clock) == CPUCLOCK_SCHED) {
list_for_each_entry(next, head, entry) {
*/
if (CPUCLOCK_PERTHREAD(timer->it_clock)) {
+ union cpu_time_count *exp = &nt->expires;
+
switch (CPUCLOCK_WHICH(timer->it_clock)) {
default:
BUG();
case CPUCLOCK_PROF:
- if (cputime_eq(p->cputime_expires.prof_exp,
- cputime_zero) ||
- cputime_gt(p->cputime_expires.prof_exp,
- nt->expires.cpu))
- p->cputime_expires.prof_exp =
- nt->expires.cpu;
+ if (expires_gt(p->cputime_expires.prof_exp,
+ exp->cpu))
+ p->cputime_expires.prof_exp = exp->cpu;
break;
case CPUCLOCK_VIRT:
- if (cputime_eq(p->cputime_expires.virt_exp,
- cputime_zero) ||
- cputime_gt(p->cputime_expires.virt_exp,
- nt->expires.cpu))
- p->cputime_expires.virt_exp =
- nt->expires.cpu;
+ if (expires_gt(p->cputime_expires.virt_exp,
+ exp->cpu))
+ p->cputime_expires.virt_exp = exp->cpu;
break;
case CPUCLOCK_SCHED:
if (p->cputime_expires.sched_exp == 0 ||
- p->cputime_expires.sched_exp >
- nt->expires.sched)
+ p->cputime_expires.sched_exp > exp->sched)
p->cputime_expires.sched_exp =
- nt->expires.sched;
+ exp->sched;
break;
}
} else {
+ struct signal_struct *const sig = p->signal;
+ union cpu_time_count *exp = &timer->it.cpu.expires;
+
/*
* For a process timer, set the cached expiration time.
*/
default:
BUG();
case CPUCLOCK_VIRT:
- if (!cputime_eq(p->signal->it_virt_expires,
- cputime_zero) &&
- cputime_lt(p->signal->it_virt_expires,
- timer->it.cpu.expires.cpu))
+ if (expires_le(sig->it[CPUCLOCK_VIRT].expires,
+ exp->cpu))
break;
- p->signal->cputime_expires.virt_exp =
- timer->it.cpu.expires.cpu;
+ sig->cputime_expires.virt_exp = exp->cpu;
break;
case CPUCLOCK_PROF:
- if (!cputime_eq(p->signal->it_prof_expires,
- cputime_zero) &&
- cputime_lt(p->signal->it_prof_expires,
- timer->it.cpu.expires.cpu))
+ if (expires_le(sig->it[CPUCLOCK_PROF].expires,
+ exp->cpu))
break;
- i = p->signal->rlim[RLIMIT_CPU].rlim_cur;
+ i = sig->rlim[RLIMIT_CPU].rlim_cur;
if (i != RLIM_INFINITY &&
- i <= cputime_to_secs(timer->it.cpu.expires.cpu))
+ i <= cputime_to_secs(exp->cpu))
break;
- p->signal->cputime_expires.prof_exp =
- timer->it.cpu.expires.cpu;
+ sig->cputime_expires.prof_exp = exp->cpu;
break;
case CPUCLOCK_SCHED:
- p->signal->cputime_expires.sched_exp =
- timer->it.cpu.expires.sched;
+ sig->cputime_expires.sched_exp = exp->sched;
break;
}
}
}
/*
+ * Sample a process (thread group) timer for the given group_leader task.
+ * Must be called with tasklist_lock held for reading.
+ */
+static int cpu_timer_sample_group(const clockid_t which_clock,
+ struct task_struct *p,
+ union cpu_time_count *cpu)
+{
+ struct task_cputime cputime;
+
+ thread_group_cputimer(p, &cputime);
+ switch (CPUCLOCK_WHICH(which_clock)) {
+ default:
+ return -EINVAL;
+ case CPUCLOCK_PROF:
+ cpu->cpu = cputime_add(cputime.utime, cputime.stime);
+ break;
+ case CPUCLOCK_VIRT:
+ cpu->cpu = cputime.utime;
+ break;
+ case CPUCLOCK_SCHED:
+ cpu->sched = cputime.sum_exec_runtime + task_delta_exec(p);
+ break;
+ }
+ return 0;
+}
+
+/*
* Guts of sys_timer_settime for CPU timers.
* This is called with the timer locked and interrupts disabled.
* If we return TIMER_RETRY, it's necessary to release the timer's lock
if (CPUCLOCK_PERTHREAD(timer->it_clock)) {
cpu_clock_sample(timer->it_clock, p, &val);
} else {
- cpu_clock_sample_group(timer->it_clock, p, &val);
+ cpu_timer_sample_group(timer->it_clock, p, &val);
}
if (old) {
read_unlock(&tasklist_lock);
goto dead;
} else {
- cpu_clock_sample_group(timer->it_clock, p, &now);
+ cpu_timer_sample_group(timer->it_clock, p, &now);
clear_dead = (unlikely(p->exit_state) &&
thread_group_empty(p));
}
}
}
+static void stop_process_timers(struct task_struct *tsk)
+{
+ struct thread_group_cputimer *cputimer = &tsk->signal->cputimer;
+ unsigned long flags;
+
+ if (!cputimer->running)
+ return;
+
+ spin_lock_irqsave(&cputimer->lock, flags);
+ cputimer->running = 0;
+ spin_unlock_irqrestore(&cputimer->lock, flags);
+}
+
+static u32 onecputick;
+
+static void check_cpu_itimer(struct task_struct *tsk, struct cpu_itimer *it,
+ cputime_t *expires, cputime_t cur_time, int signo)
+{
+ if (cputime_eq(it->expires, cputime_zero))
+ return;
+
+ if (cputime_ge(cur_time, it->expires)) {
+ if (!cputime_eq(it->incr, cputime_zero)) {
+ it->expires = cputime_add(it->expires, it->incr);
+ it->error += it->incr_error;
+ if (it->error >= onecputick) {
+ it->expires = cputime_sub(it->expires,
+ cputime_one_jiffy);
+ it->error -= onecputick;
+ }
+ } else {
+ it->expires = cputime_zero;
+ }
+
+ trace_itimer_expire(signo == SIGPROF ?
+ ITIMER_PROF : ITIMER_VIRTUAL,
+ tsk->signal->leader_pid, cur_time);
+ __group_send_sig_info(signo, SEND_SIG_PRIV, tsk);
+ }
+
+ if (!cputime_eq(it->expires, cputime_zero) &&
+ (cputime_eq(*expires, cputime_zero) ||
+ cputime_lt(it->expires, *expires))) {
+ *expires = it->expires;
+ }
+}
+
/*
* Check for any per-thread CPU timers that have fired and move them
* off the tsk->*_timers list onto the firing list. Per-thread timers
* Don't sample the current process CPU clocks if there are no timers.
*/
if (list_empty(&timers[CPUCLOCK_PROF]) &&
- cputime_eq(sig->it_prof_expires, cputime_zero) &&
+ cputime_eq(sig->it[CPUCLOCK_PROF].expires, cputime_zero) &&
sig->rlim[RLIMIT_CPU].rlim_cur == RLIM_INFINITY &&
list_empty(&timers[CPUCLOCK_VIRT]) &&
- cputime_eq(sig->it_virt_expires, cputime_zero) &&
+ cputime_eq(sig->it[CPUCLOCK_VIRT].expires, cputime_zero) &&
list_empty(&timers[CPUCLOCK_SCHED])) {
stop_process_timers(tsk);
return;
/*
* Check for the special case process timers.
*/
- if (!cputime_eq(sig->it_prof_expires, cputime_zero)) {
- if (cputime_ge(ptime, sig->it_prof_expires)) {
- /* ITIMER_PROF fires and reloads. */
- sig->it_prof_expires = sig->it_prof_incr;
- if (!cputime_eq(sig->it_prof_expires, cputime_zero)) {
- sig->it_prof_expires = cputime_add(
- sig->it_prof_expires, ptime);
- }
- __group_send_sig_info(SIGPROF, SEND_SIG_PRIV, tsk);
- }
- if (!cputime_eq(sig->it_prof_expires, cputime_zero) &&
- (cputime_eq(prof_expires, cputime_zero) ||
- cputime_lt(sig->it_prof_expires, prof_expires))) {
- prof_expires = sig->it_prof_expires;
- }
- }
- if (!cputime_eq(sig->it_virt_expires, cputime_zero)) {
- if (cputime_ge(utime, sig->it_virt_expires)) {
- /* ITIMER_VIRTUAL fires and reloads. */
- sig->it_virt_expires = sig->it_virt_incr;
- if (!cputime_eq(sig->it_virt_expires, cputime_zero)) {
- sig->it_virt_expires = cputime_add(
- sig->it_virt_expires, utime);
- }
- __group_send_sig_info(SIGVTALRM, SEND_SIG_PRIV, tsk);
- }
- if (!cputime_eq(sig->it_virt_expires, cputime_zero) &&
- (cputime_eq(virt_expires, cputime_zero) ||
- cputime_lt(sig->it_virt_expires, virt_expires))) {
- virt_expires = sig->it_virt_expires;
- }
- }
+ check_cpu_itimer(tsk, &sig->it[CPUCLOCK_PROF], &prof_expires, ptime,
+ SIGPROF);
+ check_cpu_itimer(tsk, &sig->it[CPUCLOCK_VIRT], &virt_expires, utime,
+ SIGVTALRM);
+
if (sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY) {
unsigned long psecs = cputime_to_secs(ptime);
cputime_t x;
clear_dead_task(timer, now);
goto out_unlock;
}
- cpu_clock_sample_group(timer->it_clock, p, &now);
+ cpu_timer_sample_group(timer->it_clock, p, &now);
bump_cpu_timer(timer, now);
/* Leave the tasklist_lock locked for the call below. */
}
if (task_cputime_expired(&group_sample, &sig->cputime_expires))
return 1;
}
- return 0;
+
+ return sig->rlim[RLIMIT_CPU].rlim_cur != RLIM_INFINITY;
}
/*
* timer call will interfere.
*/
list_for_each_entry_safe(timer, next, &firing, it.cpu.entry) {
- int firing;
+ int cpu_firing;
+
spin_lock(&timer->it_lock);
list_del_init(&timer->it.cpu.entry);
- firing = timer->it.cpu.firing;
+ cpu_firing = timer->it.cpu.firing;
timer->it.cpu.firing = 0;
/*
* The firing flag is -1 if we collided with a reset
* of the timer, which already reported this
* almost-firing as an overrun. So don't generate an event.
*/
- if (likely(firing >= 0)) {
+ if (likely(cpu_firing >= 0))
cpu_timer_fire(timer);
- }
spin_unlock(&timer->it_lock);
}
}
/*
- * Sample a process (thread group) timer for the given group_leader task.
- * Must be called with tasklist_lock held for reading.
- */
-static int cpu_timer_sample_group(const clockid_t which_clock,
- struct task_struct *p,
- union cpu_time_count *cpu)
-{
- struct task_cputime cputime;
-
- thread_group_cputimer(p, &cputime);
- switch (CPUCLOCK_WHICH(which_clock)) {
- default:
- return -EINVAL;
- case CPUCLOCK_PROF:
- cpu->cpu = cputime_add(cputime.utime, cputime.stime);
- break;
- case CPUCLOCK_VIRT:
- cpu->cpu = cputime.utime;
- break;
- case CPUCLOCK_SCHED:
- cpu->sched = cputime.sum_exec_runtime + task_delta_exec(p);
- break;
- }
- return 0;
-}
-
-/*
* Set one of the process-wide special case CPU timers.
* The tsk->sighand->siglock must be held by the caller.
* The *newval argument is relative and we update it to be absolute, *oldval
struct list_head *head;
BUG_ON(clock_idx == CPUCLOCK_SCHED);
- start_process_timers(tsk);
cpu_timer_sample_group(clock_idx, tsk, &now);
if (oldval) {
if (!cputime_eq(*oldval, cputime_zero)) {
if (cputime_le(*oldval, now.cpu)) {
/* Just about to fire. */
- *oldval = jiffies_to_cputime(1);
+ *oldval = cputime_one_jiffy;
} else {
*oldval = cputime_sub(*oldval, now.cpu);
}
.nsleep = thread_cpu_nsleep,
.nsleep_restart = thread_cpu_nsleep_restart,
};
+ struct timespec ts;
register_posix_clock(CLOCK_PROCESS_CPUTIME_ID, &process);
register_posix_clock(CLOCK_THREAD_CPUTIME_ID, &thread);
+ cputime_to_timespec(cputime_one_jiffy, &ts);
+ onecputick = ts.tv_nsec;
+ WARN_ON(ts.tv_sec != 0);
+
return 0;
}
__initcall(init_posix_cpu_timers);