[RFC git pull] timer fixes

From: Ingo Molnar
Date: Wed Feb 11 2009 - 09:40:28 EST

Linus,

[RFC]

Please consider pulling the latest timers-fixes-for-linus git tree from:

git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git timers-fixes-for-linus

They reason for the RFC is these commits:

signal: re-add dead task accumulation stats.
timers: split process wide cpu clocks/timers
timers: split process wide cpu clocks/timers, fix
timers: fix TIMER_ABSTIME for process wide cpu timers

They fix a reported 3% regression, but the fix got a bit large. The plain revert of:

490dea4: itimers: remove the per-cpu-ish-ness

Would be quite large in itself too, and would likely regress on larger boxes due to
its n^2 nature. My preferred solution would be to go with Peter's fixes in this
tree.

Thanks,

Ingo

------------------>
Ingo Molnar (1):
timers: split process wide cpu clocks/timers, remove spurious warning

Pavel Emelyanov (2):
x86: fix hpet timer reinit for x86_64
x86: clean up hpet timer reinit

Peter Zijlstra (4):
signal: re-add dead task accumulation stats.
timers: split process wide cpu clocks/timers
timers: split process wide cpu clocks/timers, fix
timers: fix TIMER_ABSTIME for process wide cpu timers

Suresh Siddha (1):
sched: fix nohz load balancer on cpu offline


arch/x86/kernel/hpet.c | 12 ++++-
include/linux/init_task.h | 11 ++--
include/linux/sched.h | 64 +++++++++++++------------
kernel/exit.c | 3 +
kernel/fork.c | 3 +-
kernel/itimer.c | 4 +-
kernel/posix-cpu-timers.c | 117 +++++++++++++++++++++++++++++++++++++++++++--
kernel/sched.c | 17 ++++--
kernel/sched_stats.h | 45 ++++++++++-------
kernel/signal.c | 8 ++--
10 files changed, 209 insertions(+), 75 deletions(-)

diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 64d5ad0..388254f 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -897,7 +897,7 @@ static unsigned long hpet_rtc_flags;
static int hpet_prev_update_sec;
static struct rtc_time hpet_alarm_time;
static unsigned long hpet_pie_count;
-static unsigned long hpet_t1_cmp;
+static u32 hpet_t1_cmp;
static unsigned long hpet_default_delta;
static unsigned long hpet_pie_delta;
static unsigned long hpet_pie_limit;
@@ -905,6 +905,14 @@ static unsigned long hpet_pie_limit;
static rtc_irq_handler irq_handler;

/*
+ * Check that the hpet counter c1 is ahead of the c2
+ */
+static inline int hpet_cnt_ahead(u32 c1, u32 c2)
+{
+ return (s32)(c2 - c1) < 0;
+}
+
+/*
* Registers a IRQ handler.
*/
int hpet_register_irq_handler(rtc_irq_handler handler)
@@ -1075,7 +1083,7 @@ static void hpet_rtc_timer_reinit(void)
hpet_t1_cmp += delta;
hpet_writel(hpet_t1_cmp, HPET_T1_CMP);
lost_ints++;
- } while ((long)(hpet_readl(HPET_COUNTER) - hpet_t1_cmp) > 0);
+ } while (!hpet_cnt_ahead(hpet_t1_cmp, hpet_readl(HPET_COUNTER)));

if (lost_ints) {
if (hpet_rtc_flags & RTC_PIE)
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index ea0ea1a..e752d97 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -48,12 +48,11 @@ extern struct fs_struct init_fs;
.posix_timers = LIST_HEAD_INIT(sig.posix_timers), \
.cpu_timers = INIT_CPU_TIMERS(sig.cpu_timers), \
.rlim = INIT_RLIMITS, \
- .cputime = { .totals = { \
- .utime = cputime_zero, \
- .stime = cputime_zero, \
- .sum_exec_runtime = 0, \
- .lock = __SPIN_LOCK_UNLOCKED(sig.cputime.totals.lock), \
- }, }, \
+ .cputimer = { \
+ .cputime = INIT_CPUTIME, \
+ .running = 0, \
+ .lock = __SPIN_LOCK_UNLOCKED(sig.cputimer.lock), \
+ }, \
}

extern struct nsproxy init_nsproxy;
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2127e95..8981e52 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -443,7 +443,6 @@ struct pacct_struct {
* @utime: time spent in user mode, in &cputime_t units
* @stime: time spent in kernel mode, in &cputime_t units
* @sum_exec_runtime: total time spent on the CPU, in nanoseconds
- * @lock: lock for fields in this struct
*
* This structure groups together three kinds of CPU time that are
* tracked for threads and thread groups. Most things considering
@@ -454,23 +453,33 @@ struct task_cputime {
cputime_t utime;
cputime_t stime;
unsigned long long sum_exec_runtime;
- spinlock_t lock;
};
/* Alternate field names when used to cache expirations. */
#define prof_exp stime
#define virt_exp utime
#define sched_exp sum_exec_runtime

+#define INIT_CPUTIME \
+ (struct task_cputime) { \
+ .utime = cputime_zero, \
+ .stime = cputime_zero, \
+ .sum_exec_runtime = 0, \
+ }
+
/**
- * struct thread_group_cputime - thread group interval timer counts
- * @totals: thread group interval timers; substructure for
- * uniprocessor kernel, per-cpu for SMP kernel.
+ * struct thread_group_cputimer - thread group interval timer counts
+ * @cputime: thread group interval timers.
+ * @running: non-zero when there are timers running and
+ * @cputime receives updates.
+ * @lock: lock for fields in this struct.
*
* This structure contains the version of task_cputime, above, that is
- * used for thread group CPU clock calculations.
+ * used for thread group CPU timer calculations.
*/
-struct thread_group_cputime {
- struct task_cputime totals;
+struct thread_group_cputimer {
+ struct task_cputime cputime;
+ int running;
+ spinlock_t lock;
};

/*
@@ -519,10 +528,10 @@ struct signal_struct {
cputime_t it_prof_incr, it_virt_incr;

/*
- * Thread group totals for process CPU clocks.
- * See thread_group_cputime(), et al, for details.
+ * Thread group totals for process CPU timers.
+ * See thread_group_cputimer(), et al, for details.
*/
- struct thread_group_cputime cputime;
+ struct thread_group_cputimer cputimer;

/* Earliest-expiration cache. */
struct task_cputime cputime_expires;
@@ -559,7 +568,7 @@ struct signal_struct {
* Live threads maintain their own counters and add to these
* in __exit_signal, except for the group leader.
*/
- cputime_t cutime, cstime;
+ cputime_t utime, stime, cutime, cstime;
cputime_t gtime;
cputime_t cgtime;
unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw;
@@ -568,6 +577,14 @@ struct signal_struct {
struct task_io_accounting ioac;

/*
+ * Cumulative ns of schedule CPU time fo dead threads in the
+ * group, not including a zombie group leader, (This only differs
+ * from jiffies_to_ns(utime + stime) if sched_clock uses something
+ * other than jiffies.)
+ */
+ unsigned long long sum_sched_runtime;
+
+ /*
* We don't bother to synchronize most readers of this at all,
* because there is no reader checking a limit that actually needs
* to get both rlim_cur and rlim_max atomically, and either one
@@ -2183,27 +2200,14 @@ static inline int spin_needbreak(spinlock_t *lock)
/*
* Thread group CPU time accounting.
*/
-
-static inline
-void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times)
-{
- struct task_cputime *totals = &tsk->signal->cputime.totals;
- unsigned long flags;
-
- spin_lock_irqsave(&totals->lock, flags);
- *times = *totals;
- spin_unlock_irqrestore(&totals->lock, flags);
-}
+void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times);
+void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times);

static inline void thread_group_cputime_init(struct signal_struct *sig)
{
- sig->cputime.totals = (struct task_cputime){
- .utime = cputime_zero,
- .stime = cputime_zero,
- .sum_exec_runtime = 0,
- };
-
- spin_lock_init(&sig->cputime.totals.lock);
+ sig->cputimer.cputime = INIT_CPUTIME;
+ spin_lock_init(&sig->cputimer.lock);
+ sig->cputimer.running = 0;
}

static inline void thread_group_cputime_free(struct signal_struct *sig)
diff --git a/kernel/exit.c b/kernel/exit.c
index f80dec3..efd30cc 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -118,6 +118,8 @@ static void __exit_signal(struct task_struct *tsk)
* We won't ever get here for the group leader, since it
* will have been the last reference on the signal_struct.
*/
+ sig->utime = cputime_add(sig->utime, task_utime(tsk));
+ sig->stime = cputime_add(sig->stime, task_stime(tsk));
sig->gtime = cputime_add(sig->gtime, task_gtime(tsk));
sig->min_flt += tsk->min_flt;
sig->maj_flt += tsk->maj_flt;
@@ -126,6 +128,7 @@ static void __exit_signal(struct task_struct *tsk)
sig->inblock += task_io_get_inblock(tsk);
sig->oublock += task_io_get_oublock(tsk);
task_io_accounting_add(&sig->ioac, &tsk->ioac);
+ sig->sum_sched_runtime += tsk->se.sum_exec_runtime;
sig = NULL; /* Marker for below. */
}

diff --git a/kernel/fork.c b/kernel/fork.c
index 242a706..e8e854a 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -851,13 +851,14 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
sig->tty_old_pgrp = NULL;
sig->tty = NULL;

- sig->cutime = sig->cstime = cputime_zero;
+ sig->utime = sig->stime = sig->cutime = sig->cstime = cputime_zero;
sig->gtime = cputime_zero;
sig->cgtime = cputime_zero;
sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0;
sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0;
sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0;
task_io_accounting_init(&sig->ioac);
+ sig->sum_sched_runtime = 0;
taskstats_tgid_init(sig);

task_lock(current->group_leader);
diff --git a/kernel/itimer.c b/kernel/itimer.c
index 6a5fe93..58762f7 100644
--- a/kernel/itimer.c
+++ b/kernel/itimer.c
@@ -62,7 +62,7 @@ int do_getitimer(int which, struct itimerval *value)
struct task_cputime cputime;
cputime_t utime;

- thread_group_cputime(tsk, &cputime);
+ thread_group_cputimer(tsk, &cputime);
utime = cputime.utime;
if (cputime_le(cval, utime)) { /* about to fire */
cval = jiffies_to_cputime(1);
@@ -82,7 +82,7 @@ int do_getitimer(int which, struct itimerval *value)
struct task_cputime times;
cputime_t ptime;

- thread_group_cputime(tsk, &times);
+ thread_group_cputimer(tsk, &times);
ptime = cputime_add(times.utime, times.stime);
if (cputime_le(cval, ptime)) { /* about to fire */
cval = jiffies_to_cputime(1);
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index fa07da9..2313a4c 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -230,6 +230,71 @@ static int cpu_clock_sample(const clockid_t which_clock, struct task_struct *p,
return 0;
}

+void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times)
+{
+ struct sighand_struct *sighand;
+ struct signal_struct *sig;
+ struct task_struct *t;
+
+ *times = INIT_CPUTIME;
+
+ rcu_read_lock();
+ sighand = rcu_dereference(tsk->sighand);
+ if (!sighand)
+ goto out;
+
+ sig = tsk->signal;
+
+ t = tsk;
+ do {
+ times->utime = cputime_add(times->utime, t->utime);
+ times->stime = cputime_add(times->stime, t->stime);
+ times->sum_exec_runtime += t->se.sum_exec_runtime;
+
+ t = next_thread(t);
+ } while (t != tsk);
+
+ times->utime = cputime_add(times->utime, sig->utime);
+ times->stime = cputime_add(times->stime, sig->stime);
+ times->sum_exec_runtime += sig->sum_sched_runtime;
+out:
+ 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.
@@ -457,7 +522,7 @@ void posix_cpu_timers_exit_group(struct task_struct *tsk)
{
struct task_cputime cputime;

- thread_group_cputime(tsk, &cputime);
+ thread_group_cputimer(tsk, &cputime);
cleanup_timers(tsk->signal->cpu_timers,
cputime.utime, cputime.stime, cputime.sum_exec_runtime);
}
@@ -964,6 +1029,19 @@ static void check_thread_timers(struct task_struct *tsk,
}
}

+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);
+}
+
/*
* 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
@@ -987,13 +1065,15 @@ static void check_process_timers(struct task_struct *tsk,
sig->rlim[RLIMIT_CPU].rlim_cur == RLIM_INFINITY &&
list_empty(&timers[CPUCLOCK_VIRT]) &&
cputime_eq(sig->it_virt_expires, cputime_zero) &&
- list_empty(&timers[CPUCLOCK_SCHED]))
+ list_empty(&timers[CPUCLOCK_SCHED])) {
+ stop_process_timers(tsk);
return;
+ }

/*
* Collect the current process totals.
*/
- thread_group_cputime(tsk, &cputime);
+ thread_group_cputimer(tsk, &cputime);
utime = cputime.utime;
ptime = cputime_add(utime, cputime.stime);
sum_sched_runtime = cputime.sum_exec_runtime;
@@ -1259,7 +1339,7 @@ static inline int fastpath_timer_check(struct task_struct *tsk)
if (!task_cputime_zero(&sig->cputime_expires)) {
struct task_cputime group_sample;

- thread_group_cputime(tsk, &group_sample);
+ thread_group_cputimer(tsk, &group_sample);
if (task_cputime_expired(&group_sample, &sig->cputime_expires))
return 1;
}
@@ -1329,6 +1409,33 @@ void run_posix_cpu_timers(struct task_struct *tsk)
}

/*
+ * 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
@@ -1341,7 +1448,7 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx,
struct list_head *head;

BUG_ON(clock_idx == CPUCLOCK_SCHED);
- cpu_clock_sample_group(clock_idx, tsk, &now);
+ cpu_timer_sample_group(clock_idx, tsk, &now);

if (oldval) {
if (!cputime_eq(*oldval, cputime_zero)) {
diff --git a/kernel/sched.c b/kernel/sched.c
index 242d0d4..e1fc67d 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3890,19 +3890,24 @@ int select_nohz_load_balancer(int stop_tick)
int cpu = smp_processor_id();

if (stop_tick) {
- cpumask_set_cpu(cpu, nohz.cpu_mask);
cpu_rq(cpu)->in_nohz_recently = 1;

- /*
- * If we are going offline and still the leader, give up!
- */
- if (!cpu_active(cpu) &&
- atomic_read(&nohz.load_balancer) == cpu) {
+ if (!cpu_active(cpu)) {
+ if (atomic_read(&nohz.load_balancer) != cpu)
+ return 0;
+
+ /*
+ * If we are going offline and still the leader,
+ * give up!
+ */
if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
BUG();
+
return 0;
}

+ cpumask_set_cpu(cpu, nohz.cpu_mask);
+
/* time for ilb owner also to sleep */
if (cpumask_weight(nohz.cpu_mask) == num_online_cpus()) {
if (atomic_read(&nohz.load_balancer) == cpu)
diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h
index 8ab0cef..a8f93dd 100644
--- a/kernel/sched_stats.h
+++ b/kernel/sched_stats.h
@@ -296,19 +296,21 @@ sched_info_switch(struct task_struct *prev, struct task_struct *next)
static inline void account_group_user_time(struct task_struct *tsk,
cputime_t cputime)
{
- struct task_cputime *times;
- struct signal_struct *sig;
+ struct thread_group_cputimer *cputimer;

/* tsk == current, ensure it is safe to use ->signal */
if (unlikely(tsk->exit_state))
return;

- sig = tsk->signal;
- times = &sig->cputime.totals;
+ cputimer = &tsk->signal->cputimer;

- spin_lock(&times->lock);
- times->utime = cputime_add(times->utime, cputime);
- spin_unlock(&times->lock);
+ if (!cputimer->running)
+ return;
+
+ spin_lock(&cputimer->lock);
+ cputimer->cputime.utime =
+ cputime_add(cputimer->cputime.utime, cputime);
+ spin_unlock(&cputimer->lock);
}

/**
@@ -324,19 +326,21 @@ static inline void account_group_user_time(struct task_struct *tsk,
static inline void account_group_system_time(struct task_struct *tsk,
cputime_t cputime)
{
- struct task_cputime *times;
- struct signal_struct *sig;
+ struct thread_group_cputimer *cputimer;

/* tsk == current, ensure it is safe to use ->signal */
if (unlikely(tsk->exit_state))
return;

- sig = tsk->signal;
- times = &sig->cputime.totals;
+ cputimer = &tsk->signal->cputimer;
+
+ if (!cputimer->running)
+ return;

- spin_lock(&times->lock);
- times->stime = cputime_add(times->stime, cputime);
- spin_unlock(&times->lock);
+ spin_lock(&cputimer->lock);
+ cputimer->cputime.stime =
+ cputime_add(cputimer->cputime.stime, cputime);
+ spin_unlock(&cputimer->lock);
}

/**
@@ -352,7 +356,7 @@ static inline void account_group_system_time(struct task_struct *tsk,
static inline void account_group_exec_runtime(struct task_struct *tsk,
unsigned long long ns)
{
- struct task_cputime *times;
+ struct thread_group_cputimer *cputimer;
struct signal_struct *sig;

sig = tsk->signal;
@@ -361,9 +365,12 @@ static inline void account_group_exec_runtime(struct task_struct *tsk,
if (unlikely(!sig))
return;

- times = &sig->cputime.totals;
+ cputimer = &sig->cputimer;
+
+ if (!cputimer->running)
+ return;

- spin_lock(&times->lock);
- times->sum_exec_runtime += ns;
- spin_unlock(&times->lock);
+ spin_lock(&cputimer->lock);
+ cputimer->cputime.sum_exec_runtime += ns;
+ spin_unlock(&cputimer->lock);
}
diff --git a/kernel/signal.c b/kernel/signal.c
index b6b3676..2a74fe8 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1367,7 +1367,6 @@ int do_notify_parent(struct task_struct *tsk, int sig)
struct siginfo info;
unsigned long flags;
struct sighand_struct *psig;
- struct task_cputime cputime;
int ret = sig;

BUG_ON(sig == -1);
@@ -1397,9 +1396,10 @@ int do_notify_parent(struct task_struct *tsk, int sig)
info.si_uid = __task_cred(tsk)->uid;
rcu_read_unlock();

- thread_group_cputime(tsk, &cputime);
- info.si_utime = cputime_to_jiffies(cputime.utime);
- info.si_stime = cputime_to_jiffies(cputime.stime);
+ info.si_utime = cputime_to_clock_t(cputime_add(tsk->utime,
+ tsk->signal->utime));
+ info.si_stime = cputime_to_clock_t(cputime_add(tsk->stime,
+ tsk->signal->stime));

info.si_status = tsk->exit_code & 0x7f;
if (tsk->exit_code & 0x80)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/