[PATCH 1/6] sched/cputime: Support other fields on kcpustat_field()

From: Frederic Weisbecker <frederic@kernel.org>
To: Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Frederic Weisbecker <frederic@kernel.org>,
	Jacek Anaszewski <jacek.anaszewski@gmail.com>,
	Wanpeng Li <wanpengli@tencent.com>,
	"Rafael J . Wysocki" <rjw@rjwysocki.net>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Rik van Riel <riel@surriel.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Yauheni Kaliuta <yauheni.kaliuta@redhat.com>,
	Viresh Kumar <viresh.kumar@linaro.org>,
	Pavel Machek <pavel@ucw.cz>
Subject: [PATCH 1/6] sched/cputime: Support other fields on kcpustat_field()
Date: Thu, 21 Nov 2019 03:44:25 +0100	[thread overview]
Message-ID: <20191121024430.19938-2-frederic@kernel.org> (raw)
In-Reply-To: <20191121024430.19938-1-frederic@kernel.org>

Provide support for user, nice, guest and guest_nice fields through
kcpustat_field().

Whether we account the delta to a nice or not nice field is decided on
top of the nice value snapshot taken at the time we call kcpustat_field().
If the nice value of the task has been changed since the last vtime
update, we may have inacurrate distribution of the nice VS unnice
cputime.

However this is considered as a minor issue compared to the proper fix
that would involve interrupting the target on nice updates, which is
undesired on nohz_full CPUs.

Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Cc: Yauheni Kaliuta <yauheni.kaliuta@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Rik van Riel <riel@surriel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Wanpeng Li <wanpengli@tencent.com>
Cc: Ingo Molnar <mingo@kernel.org>
---
 kernel/sched/cputime.c | 54 +++++++++++++++++++++++++++++++++---------
 1 file changed, 43 insertions(+), 11 deletions(-)

diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index e0cd20693ef5..27b5406222fc 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -912,11 +912,21 @@ void task_cputime(struct task_struct *t, u64 *utime, u64 *stime)
 	} while (read_seqcount_retry(&vtime->seqcount, seq));
 }
 
+static u64 kcpustat_user_vtime(struct vtime *vtime)
+{
+	if (vtime->state == VTIME_USER)
+		return vtime->utime + vtime_delta(vtime);
+	else if (vtime->state == VTIME_GUEST)
+		return vtime->gtime + vtime_delta(vtime);
+	return 0;
+}
+
 static int kcpustat_field_vtime(u64 *cpustat,
-				struct vtime *vtime,
+				struct task_struct *tsk,
 				enum cpu_usage_stat usage,
 				int cpu, u64 *val)
 {
+	struct vtime *vtime = &tsk->vtime;
 	unsigned int seq;
 	int err;
 
@@ -946,9 +956,37 @@ static int kcpustat_field_vtime(u64 *cpustat,
 
 		*val = cpustat[usage];
 
-		if (vtime->state == VTIME_SYS)
-			*val += vtime->stime + vtime_delta(vtime);
-
+		/*
+		 * Nice VS unnice cputime accounting may be inaccurate if
+		 * the nice value has changed since the last vtime update.
+		 * But proper fix would involve interrupting target on nice
+		 * updates which is a no go on nohz_full (although the scheduler
+		 * may still interrupt the target if rescheduling is needed...)
+		 */
+		switch (usage) {
+		case CPUTIME_SYSTEM:
+			if (vtime->state == VTIME_SYS)
+				*val += vtime->stime + vtime_delta(vtime);
+			break;
+		case CPUTIME_USER:
+			if (task_nice(tsk) <= 0)
+				*val += kcpustat_user_vtime(vtime);
+			break;
+		case CPUTIME_NICE:
+			if (task_nice(tsk) > 0)
+				*val += kcpustat_user_vtime(vtime);
+			break;
+		case CPUTIME_GUEST:
+			if (vtime->state == VTIME_GUEST && task_nice(tsk) <= 0)
+				*val += vtime->gtime + vtime_delta(vtime);
+			break;
+		case CPUTIME_GUEST_NICE:
+			if (vtime->state == VTIME_GUEST && task_nice(tsk) > 0)
+				*val += vtime->gtime + vtime_delta(vtime);
+			break;
+		default:
+			break;
+		}
 	} while (read_seqcount_retry(&vtime->seqcount, seq));
 
 	return 0;
@@ -965,15 +1003,10 @@ u64 kcpustat_field(struct kernel_cpustat *kcpustat,
 	if (!vtime_accounting_enabled_cpu(cpu))
 		return cpustat[usage];
 
-	/* Only support sys vtime for now */
-	if (usage != CPUTIME_SYSTEM)
-		return cpustat[usage];
-
 	rq = cpu_rq(cpu);
 
 	for (;;) {
 		struct task_struct *curr;
-		struct vtime *vtime;
 
 		rcu_read_lock();
 		curr = rcu_dereference(rq->curr);
@@ -982,8 +1015,7 @@ u64 kcpustat_field(struct kernel_cpustat *kcpustat,
 			return cpustat[usage];
 		}
 
-		vtime = &curr->vtime;
-		err = kcpustat_field_vtime(cpustat, vtime, usage, cpu, &val);
+		err = kcpustat_field_vtime(cpustat, curr, usage, cpu, &val);
 		rcu_read_unlock();
 
 		if (!err)
-- 
2.23.0

