Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1631543

Re: [PATCH] sched/cputime: Fix ksoftirqd cputime accounting regression

From Rik van Riel <riel@redhat.com>
Newsgroups linux.kernel
Subject Re: [PATCH] sched/cputime: Fix ksoftirqd cputime accounting regression
Date 2017-04-26 17:50 +0200
Message-ID <tAxmy-4e9-19@gated-at.bofh.it> (permalink)
References <tA9tU-5Lr-9@gated-at.bofh.it>
Organization Red Hat, Inc

Show all headers | View raw


On Tue, 2017-04-25 at 16:10 +0200, Frederic Weisbecker wrote:
> irq_time_read() returns the irqtime minus the ksoftirqd time. This
> is necessary because irq_time_read() is used to substract the IRQ
> time
> from the sum_exec_runtime of a task. If we were to include the
> softirq
> time of ksoftirqd, this task would substract its own CPU time
> everytime
> it updates ksoftirqd->sum_exec_runtime which would therefore never
> progress.
> 
> But this behaviour got broken by commit a499a5a14db:
> 	("sched/cputime: Increment kcpustat directly on irqtime
> account")
> which now includes ksoftirqd softirq time in the time returned by
> irq_time_read().

Ahhh, after a look at account_system_time, it becomes
clear why that was happening!

Reviewed-by: Rik van Riel <riel@redhat.com>

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH] sched/cputime: Fix ksoftirqd cputime accounting regression Frederic Weisbecker <fweisbec@gmail.com> - 2017-04-25 16:20 +0200
  Re: [PATCH] sched/cputime: Fix ksoftirqd cputime accounting  regression Jesper Dangaard Brouer <brouer@redhat.com> - 2017-04-26 16:40 +0200
  Re: [PATCH] sched/cputime: Fix ksoftirqd cputime accounting  regression Rik van Riel <riel@redhat.com> - 2017-04-26 17:50 +0200
  [tip:sched/urgent] sched/cputime: Fix ksoftirqd cputime accounting  regression tip-bot for Frederic Weisbecker <tipbot@zytor.com> - 2017-04-27 16:30 +0200

csiph-web