[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20241125134115.GB38837@noisy.programming.kicks-ass.net>
Date: Mon, 25 Nov 2024 14:41:15 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: "Sapkal, Swapnil" <swapnil.sapkal@....com>
Cc: mingo@...hat.com, acme@...nel.org, namhyung@...nel.org,
irogers@...gle.com, james.clark@....com, ravi.bangoria@....com,
yu.c.chen@...el.com, mark.rutland@....com,
alexander.shishkin@...ux.intel.com, jolsa@...nel.org,
rostedt@...dmis.org, vincent.guittot@...aro.org,
adrian.hunter@...el.com, kan.liang@...ux.intel.com,
gautham.shenoy@....com, kprateek.nayak@....com,
juri.lelli@...hat.com, yangjihong@...edance.com, void@...ifault.com,
tj@...nel.org, vineethr@...ux.ibm.com, linux-kernel@...r.kernel.org,
linux-perf-users@...r.kernel.org, santosh.shukla@....com,
ananth.narayan@....com, sandipan.das@....com,
James Clark <james.clark@...aro.org>
Subject: Re: [PATCH v2 1/6] sched/stats: Print domain name in /proc/schedstat
On Mon, Nov 25, 2024 at 06:01:16PM +0530, Sapkal, Swapnil wrote:
> Hello Peter,
>
> On 11/22/2024 4:42 PM, Peter Zijlstra wrote:
> > On Fri, Nov 22, 2024 at 08:44:47AM +0000, Swapnil Sapkal wrote:
> >
> > > schedstat version has not been updated since this change merely adds
> > > additional information to the domain name field and does not add a new
> > > field altogether.
> >
> > So I don't care much either way, but if an existing tool is trying to
> > parse the domain number, it might now get confused by the extra
> > characters. I know of no such tool, just being pedantic.
>
> I will change the schedstat version so that it will not break any existing
> tools.
>
> However the domain name is under CONFIG_SCHED_DEBUG and thus incrementing
> the schedstats version is not sufficient when this config option is
> disabled. Is it okay for you to move domain name out of CONFIG_SCHED_DEBUG?
I suppose so..
> Previously, I sent couple of patches to change the schedstat layout for
> * Correctly accounting the different types of imbalances [1]
> * Accurately computing the task hot migrations [2] (This was your patch)
> I would like to include these two patches that better helps justify the
> version change. Is that okay by you?
Yeah, sure.
Powered by blists - more mailing lists