sched: Introduce task_times() to replace task_{u,s}time() pair

Functions task_{u,s}time() are called in pair in almost all
cases.  However task_stime() is implemented to call task_utime()
from its inside, so such paired calls run task_utime() twice.

It means we do heavy divisions (div_u64 + do_div) twice to get
utime and stime which can be obtained at same time by one set
of divisions.

This patch introduces a function task_times(*tsk, *utime,
*stime) to retrieve utime and stime at once in better, optimized
way.

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Stanislaw Gruszka <sgruszka@redhat.com>
Cc: Spencer Candland <spencer@bluehost.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Americo Wang <xiyou.wangcong@gmail.com>
LKML-Reference: <4B0E16AE.906@jp.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

fs/proc/array.c

@@ -535,8 +535,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns,
 	if (!whole) {
 		min_flt = task->min_flt;
 		maj_flt = task->maj_flt;
-		utime = task_utime(task);
-		stime = task_stime(task);
+		task_times(task, &utime, &stime);
 		gtime = task_gtime(task);
 	}
 

include/linux/sched.h

@@ -1723,6 +1723,7 @@ static inline void put_task_struct(struct task_struct *t)
 extern cputime_t task_utime(struct task_struct *p);
 extern cputime_t task_stime(struct task_struct *p);
 extern cputime_t task_gtime(struct task_struct *p);
+extern void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st);
 
 /*
  * Per process flags

kernel/exit.c

@@ -91,6 +91,8 @@ static void __exit_signal(struct task_struct *tsk)
 	if (atomic_dec_and_test(&sig->count))
 		posix_cpu_timers_exit_group(tsk);
 	else {
+		cputime_t utime, stime;
+
 		/*
 		 * If there is any task waiting for the group exit
 		 * then notify it:
@@ -110,8 +112,9 @@ 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));
+		task_times(tsk, &utime, &stime);
+		sig->utime = cputime_add(sig->utime, utime);
+		sig->stime = cputime_add(sig->stime, stime);
 		sig->gtime = cputime_add(sig->gtime, task_gtime(tsk));
 		sig->min_flt += tsk->min_flt;
 		sig->maj_flt += tsk->maj_flt;

kernel/sched.c

@@ -5191,6 +5191,14 @@ cputime_t task_stime(struct task_struct *p)
 {
 	return p->stime;
 }
+
+void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
+{
+	if (ut)
+		*ut = task_utime(p);
+	if (st)
+		*st = task_stime(p);
+}
 #else
 
 #ifndef nsecs_to_cputime
@@ -5198,41 +5206,48 @@ cputime_t task_stime(struct task_struct *p)
 	msecs_to_cputime(div_u64((__nsecs), NSEC_PER_MSEC))
 #endif
 
-cputime_t task_utime(struct task_struct *p)
+void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
 {
-	cputime_t utime = p->utime, total = utime + p->stime;
-	u64 temp;
+	cputime_t rtime, utime = p->utime, total = utime + p->stime;
 
 	/*
 	 * Use CFS's precise accounting:
 	 */
-	temp = (u64)nsecs_to_cputime(p->se.sum_exec_runtime);
+	rtime = nsecs_to_cputime(p->se.sum_exec_runtime);
 
 	if (total) {
-		temp *= utime;
-		do_div(temp, total);
-	}
-	utime = (cputime_t)temp;
+		u64 temp;
 
+		temp = (u64)(rtime * utime);
+		do_div(temp, total);
+		utime = (cputime_t)temp;
+	} else
+		utime = rtime;
+
+	/*
+	 * Compare with previous values, to keep monotonicity:
+	 */
 	p->prev_utime = max(p->prev_utime, utime);
-	return p->prev_utime;
+	p->prev_stime = max(p->prev_stime, rtime - p->prev_utime);
+
+	if (ut)
+		*ut = p->prev_utime;
+	if (st)
+		*st = p->prev_stime;
+}
+
+cputime_t task_utime(struct task_struct *p)
+{
+	cputime_t utime;
+	task_times(p, &utime, NULL);
+	return utime;
 }
 
 cputime_t task_stime(struct task_struct *p)
 {
 	cputime_t stime;
-
-	/*
-	 * Use CFS's precise accounting. (we subtract utime from
-	 * the total, to make sure the total observed by userspace
-	 * grows monotonically - apps rely on that):
-	 */
-	stime = nsecs_to_cputime(p->se.sum_exec_runtime) - task_utime(p);
-
-	if (stime >= 0)
-		p->prev_stime = max(p->prev_stime, stime);
-
-	return p->prev_stime;
+	task_times(p, NULL, &stime);
+	return stime;
 }
 #endif
 

kernel/sys.c

@@ -1346,8 +1346,7 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
 	utime = stime = cputime_zero;
 
 	if (who == RUSAGE_THREAD) {
-		utime = task_utime(current);
-		stime = task_stime(current);
+		task_times(current, &utime, &stime);
 		accumulate_thread_rusage(p, r);
 		maxrss = p->signal->maxrss;
 		goto out;