[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <163766531523.11128.10280253346273305725.tip-bot2@tip-bot2>
Date: Tue, 23 Nov 2021 11:01:55 -0000
From: "tip-bot2 for Andrey Ryabinin" <tip-bot2@...utronix.de>
To: linux-tip-commits@...r.kernel.org
Cc: Andrey Ryabinin <arbn@...dex-team.com>,
"Peter Zijlstra (Intel)" <peterz@...radead.org>,
Daniel Jordan <daniel.m.jordan@...cle.com>,
Tejun Heo <tj@...nel.org>, x86@...nel.org,
linux-kernel@...r.kernel.org
Subject: [tip: sched/core] sched/cpuacct: Make user/system times in
cpuacct.stat more precise
The following commit has been merged into the sched/core branch of tip:
Commit-ID: 8c92606ab81086db00cbb73347d124b4eb169b7e
Gitweb: https://git.kernel.org/tip/8c92606ab81086db00cbb73347d124b4eb169b7e
Author: Andrey Ryabinin <arbn@...dex-team.com>
AuthorDate: Mon, 15 Nov 2021 19:46:07 +03:00
Committer: Peter Zijlstra <peterz@...radead.org>
CommitterDate: Tue, 23 Nov 2021 09:55:22 +01:00
sched/cpuacct: Make user/system times in cpuacct.stat more precise
cpuacct.stat shows user time based on raw random precision tick
based counters. Use cputime_addjust() to scale these values against the
total runtime accounted by the scheduler, like we already do
for user/system times in /proc/<pid>/stat.
Signed-off-by: Andrey Ryabinin <arbn@...dex-team.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
Reviewed-by: Daniel Jordan <daniel.m.jordan@...cle.com>
Acked-by: Tejun Heo <tj@...nel.org>
Link: https://lore.kernel.org/r/20211115164607.23784-4-arbn@yandex-team.com
---
kernel/sched/cpuacct.c | 25 +++++++++++++++----------
1 file changed, 15 insertions(+), 10 deletions(-)
diff --git a/kernel/sched/cpuacct.c b/kernel/sched/cpuacct.c
index 9de7dd5..3d06c5e 100644
--- a/kernel/sched/cpuacct.c
+++ b/kernel/sched/cpuacct.c
@@ -261,25 +261,30 @@ static int cpuacct_all_seq_show(struct seq_file *m, void *V)
static int cpuacct_stats_show(struct seq_file *sf, void *v)
{
struct cpuacct *ca = css_ca(seq_css(sf));
- s64 val[CPUACCT_STAT_NSTATS];
+ struct task_cputime cputime;
+ u64 val[CPUACCT_STAT_NSTATS];
int cpu;
int stat;
- memset(val, 0, sizeof(val));
+ memset(&cputime, 0, sizeof(cputime));
for_each_possible_cpu(cpu) {
u64 *cpustat = per_cpu_ptr(ca->cpustat, cpu)->cpustat;
- val[CPUACCT_STAT_USER] += cpustat[CPUTIME_USER];
- val[CPUACCT_STAT_USER] += cpustat[CPUTIME_NICE];
- val[CPUACCT_STAT_SYSTEM] += cpustat[CPUTIME_SYSTEM];
- val[CPUACCT_STAT_SYSTEM] += cpustat[CPUTIME_IRQ];
- val[CPUACCT_STAT_SYSTEM] += cpustat[CPUTIME_SOFTIRQ];
+ cputime.utime += cpustat[CPUTIME_USER];
+ cputime.utime += cpustat[CPUTIME_NICE];
+ cputime.stime += cpustat[CPUTIME_SYSTEM];
+ cputime.stime += cpustat[CPUTIME_IRQ];
+ cputime.stime += cpustat[CPUTIME_SOFTIRQ];
+
+ cputime.sum_exec_runtime += *per_cpu_ptr(ca->cpuusage, cpu);
}
+ cputime_adjust(&cputime, &seq_css(sf)->cgroup->prev_cputime,
+ &val[CPUACCT_STAT_USER], &val[CPUACCT_STAT_SYSTEM]);
+
for (stat = 0; stat < CPUACCT_STAT_NSTATS; stat++) {
- seq_printf(sf, "%s %lld\n",
- cpuacct_stat_desc[stat],
- (long long)nsec_to_clock_t(val[stat]));
+ seq_printf(sf, "%s %llu\n", cpuacct_stat_desc[stat],
+ nsec_to_clock_t(val[stat]));
}
return 0;
Powered by blists - more mailing lists