[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5447e3b5-db58-1ace-9974-6dfc250b4a1e@oracle.com>
Date: Mon, 14 Jan 2019 12:04:19 -0500
From: Steven Sistare <steven.sistare@...cle.com>
To: Rick Lindsley <ricklind@...ux.vnet.ibm.com>, mingo@...hat.com,
peterz@...radead.org
Cc: subhra.mazumdar@...cle.com, dhaval.giani@...cle.com,
daniel.m.jordan@...cle.com, pavel.tatashin@...rosoft.com,
matt@...eblueprint.co.uk, umgwanakikbuti@...il.com,
riel@...hat.com, jbacik@...com, juri.lelli@...hat.com,
valentin.schneider@....com, vincent.guittot@...aro.org,
quentin.perret@....com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4 10/10] sched/fair: Provide idle search schedstats
On 12/24/2018 7:25 AM, Rick Lindsley wrote:
> On 12/06/2018 01:28 PM, Steve Sistare wrote:
>> Add schedstats to measure the effectiveness of searching for idle CPUs
>> and stealing tasks. This is a temporary patch intended for use during
>> development only. SCHEDSTAT_VERSION is bumped to 16, and the following
>> fields are added to the per-CPU statistics of /proc/schedstat:
>
> Why only temporary? It would seem that this might be useful information in the future too, for other developers with either new code or new use cases.
>
> Rick
Yes, they could be useful. I suspect it would take a while to reach consensus
on new field definition, and whether their usefulness outweighs the downside
of breaking out of tree tools that assume version 15.
- Steve
Powered by blists - more mailing lists