[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <aBsGXCKX8-2_Cn9x@gmail.com>
Date: Wed, 7 May 2025 09:06:04 +0200
From: Ingo Molnar <mingo@...nel.org>
To: Yafang Shao <laoar.shao@...il.com>,
Peter Zijlstra <peterz@...radead.org>
Cc: ardb@...nel.org, arnd@...db.de, bp@...en8.de, dwmw@...zon.co.uk,
hpa@...or.com, linux-kernel@...r.kernel.org,
michal.lkml@...kovi.net, peterz@...radead.org, tglx@...utronix.de,
torvalds@...ux-foundation.org, vkuznets@...hat.com,
yamada.masahiro@...ionext.com
Subject: Re: [PATCH 13/15] x86/kconfig/64: Enable popular scheduler, cgroups
and namespaces options in the defconfig
* Yafang Shao <laoar.shao@...il.com> wrote:
> Hello Mingo,
>
> > +CONFIG_VIRT_CPU_ACCOUNTING_GEN=y
> > +CONFIG_IRQ_TIME_ACCOUNTING=y
>
> Enabling CONFIG_IRQ_TIME_ACCOUNTING=y can lead to user-visible behavioral
> changes. For more context, please refer to the related discussion here:
> https://lore.kernel.org/all/20241222024734.63894-1-laoar.shao@gmail.com/ .
Yeah. I actually agree with your series. It (re-)includes IRQ/softirq
time in task CPU usage statistics even under IRQ_TIME_ACCOUNTING=y,
while still keeping the finegrained IRQ/softirq statistics as well,
correct?
The Kconfig option is also arguably rather misleading:
config IRQ_TIME_ACCOUNTING
bool "Fine granularity task level IRQ time accounting"
depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE
help
Select this option to enable fine granularity task irq time
accounting. This is done by reading a timestamp on each
transitions between softirq and hardirq state, so there can be a
small performance impact.
It only warns about a small performance impact, but doesn't warn that
CPU accounting is changed in an incompatible fashion that surprises
tooling...
But I think we should probably treat this as a bug, not as lack of
documentation. Peter, do you concur?
> If we decide to enable it by default, we should clearly document this
> behavior change. Below is the patch I wrote earlier but haven’t sent
> out for review yet.
Note that it's not enabled by default - this patch is just about the
x86 defconfig.
Thanks,
Ingo
Powered by blists - more mailing lists