lkml.org 
[lkml]   [2013]   [May]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH 1/8] posix-cpu-timers: don't account cpu timer after stopped thread runtime accounting
Date
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>

When tsk->signal->cputimer->running is 1, signal->cputimer (i.e. per process
timer account) and tsk->sum_sched_runtime (i.e. per thread timer account)
increase at the same pace because update_curr() increases both accounting.

However, there is one exception. When thread exiting, __exit_signal() turns
over task's sum_shced_runtime to sig->sum_sched_runtime, but it doesn't stop
signal->cputimer accounting.

This inconsistency makes POSIX timer wake up too early. This patch fixes it.

Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Olivier Langlois <olivier@trillion01.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
---
kernel/sched/stats.h | 33 ++++++++++++++++++++++++++++++---
1 files changed, 30 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/stats.h b/kernel/sched/stats.h
index 2ef90a5..8a0567f 100644
--- a/kernel/sched/stats.h
+++ b/kernel/sched/stats.h
@@ -162,6 +162,33 @@ sched_info_switch(struct task_struct *prev, struct task_struct *next)
*/

/**
+ * cputimer_running - return true if cputimer is running
+ *
+ * @tsk: Pointer to target task.
+ */
+static inline bool cputimer_running(struct task_struct *tsk)
+
+{
+ struct thread_group_cputimer *cputimer = &tsk->signal->cputimer;
+
+ if (!cputimer->running)
+ return false;
+
+ /*
+ * After turning over se.sum_exec_runtime's time accounting to
+ * sig->sum_sched_runtime in __exit_signal(), a per-thread cputime
+ * of the thread is going to be lost. We must not account exec_runtime
+ * here too because we need to keep consistent cputime between
+ * per-thread and per-process. Otherwise, the inconsistency is
+ * observable when single thread program run.
+ */
+ if (unlikely(!tsk->sighand))
+ return false;
+
+ return true;
+}
+
+/**
* account_group_user_time - Maintain utime for a thread group.
*
* @tsk: Pointer to task structure.
@@ -176,7 +203,7 @@ static inline void account_group_user_time(struct task_struct *tsk,
{
struct thread_group_cputimer *cputimer = &tsk->signal->cputimer;

- if (!cputimer->running)
+ if (!cputimer_running(tsk))
return;

raw_spin_lock(&cputimer->lock);
@@ -199,7 +226,7 @@ static inline void account_group_system_time(struct task_struct *tsk,
{
struct thread_group_cputimer *cputimer = &tsk->signal->cputimer;

- if (!cputimer->running)
+ if (!cputimer_running(tsk))
return;

raw_spin_lock(&cputimer->lock);
@@ -222,7 +249,7 @@ static inline void account_group_exec_runtime(struct task_struct *tsk,
{
struct thread_group_cputimer *cputimer = &tsk->signal->cputimer;

- if (!cputimer->running)
+ if (!cputimer_running(tsk))
return;

raw_spin_lock(&cputimer->lock);
--
1.7.1


\
 
 \ /
  Last update: 2013-05-26 23:41    [W:0.218 / U:0.052 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site