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
next prev parent reply other threads:[~2019-11-21 2:44 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-21 2:44 [PATCH 0/6] sched/nohz: Make the rest of kcpustat vtime aware v3 Frederic Weisbecker
2019-11-21 2:44 ` Frederic Weisbecker [this message]
2019-11-21 16:48 ` [tip: sched/core] sched/cputime: Support other fields on kcpustat_field() tip-bot2 for Frederic Weisbecker
2019-11-21 2:44 ` [PATCH 2/6] sched/vtime: Bring up complete kcpustat accessor Frederic Weisbecker
2019-11-21 16:48 ` [tip: sched/core] " tip-bot2 for Frederic Weisbecker
2019-12-28 20:56 ` [PATCH 2/6] " Chris Wilson
2019-12-30 1:08 ` Frederic Weisbecker
2019-12-30 9:04 ` [PATCH] sched/vtime: Prevent unstable evaluation of WARN(vtime->state) Chris Wilson
2019-12-30 17:39 ` Frederic Weisbecker
2019-11-21 2:44 ` [PATCH 3/6] procfs: Use all-in-one vtime aware kcpustat accessor Frederic Weisbecker
2019-11-21 16:48 ` [tip: sched/core] " tip-bot2 for Frederic Weisbecker
2019-12-08 15:57 ` [PATCH 3/6] " Paul Orlyk
2019-12-09 15:50 ` Frederic Weisbecker
2019-11-21 2:44 ` [PATCH 4/6] cpufreq: Use vtime aware kcpustat accessors for user time Frederic Weisbecker
2019-11-21 16:48 ` [tip: sched/core] " tip-bot2 for Frederic Weisbecker
2019-11-21 2:44 ` [PATCH 5/6] leds: Use all-in-one vtime aware kcpustat accessor Frederic Weisbecker
2019-11-21 6:58 ` Ingo Molnar
2019-11-21 14:14 ` Frederic Weisbecker
2019-11-21 16:49 ` Ingo Molnar
2019-11-21 16:48 ` [tip: sched/core] " tip-bot2 for Frederic Weisbecker
2019-11-21 2:44 ` [PATCH 6/6] rackmeter: Use " Frederic Weisbecker
2019-11-21 16:48 ` [tip: sched/core] " tip-bot2 for Frederic Weisbecker
-- strict thread matches above, loose matches on Subject: below --
2019-11-19 23:22 [PATCH 0/6] sched/nohz: Make the rest of kcpustat vtime aware v2 Frederic Weisbecker
2019-11-19 23:22 ` [PATCH 1/6] sched/cputime: Support other fields on kcpustat_field() Frederic Weisbecker
2019-11-20 11:51 ` Ingo Molnar
2019-11-20 21:04 ` Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20191121024430.19938-2-frederic@kernel.org \
--to=frederic@kernel.org \
--cc=benh@kernel.crashing.org \
--cc=jacek.anaszewski@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=pavel@ucw.cz \
--cc=peterz@infradead.org \
--cc=riel@surriel.com \
--cc=rjw@rjwysocki.net \
--cc=tglx@linutronix.de \
--cc=viresh.kumar@linaro.org \
--cc=wanpengli@tencent.com \
--cc=yauheni.kaliuta@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).