[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <eb8cd1eb-1cd0-4eed-a87b-1edd3e26a732@nvidia.com>
Date: Mon, 17 Mar 2025 23:45:36 +0100
From: Joel Fernandes <joelagnelf@...dia.com>
To: Tejun Heo <tj@...nel.org>
Cc: linux-kernel@...r.kernel.org, David Vernet <void@...ifault.com>,
Andrea Righi <arighi@...dia.com>, Changwoo Min <changwoo@...lia.com>,
Ingo Molnar <mingo@...hat.com>, Peter Zijlstra <peterz@...radead.org>,
Juri Lelli <juri.lelli@...hat.com>,
Vincent Guittot <vincent.guittot@...aro.org>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Steven Rostedt <rostedt@...dmis.org>, Ben Segall <bsegall@...gle.com>,
Mel Gorman <mgorman@...e.de>, Valentin Schneider <vschneid@...hat.com>
Subject: Re: [PATCH RFC] sched_ext: Choose prev_cpu if idle and cache affine
without WF_SYNC
On 3/17/2025 11:25 PM, Tejun Heo wrote:
> Hello,
>
> On Mon, Mar 17, 2025 at 11:07:36PM +0100, Joel Fernandes wrote:
> ...
>>> I think it's preferring idle core a
>>> bit too much - it probably doesn't make sense to cross the NUMA boundary if
>>> there is an idle CPU in this node, at least.
>>
>> Yes, that is a bit extreme. I believe that is what my patch is fixing. If
>> previous CPU and current CPU share cache, we prefer busy cores with free
>> previous idle SMT, otherwise go looking for fully idle cores. But it sounds like
>> from Peter's reply that is not necessarily a good thing to do in case 'fast
>> numa'. So I guess there is no good answer (?) or way of doing it.
>
> Yeah, recent AMD CPUs can be configured into NUMA mode where each chiplet is
> reported as a node and they do behave like one as each has its own memory
> connection but the distances among them are significantly closer than
> traditional multi-socket. That said, I think the implementation is still a
> bit too happy to jump the boundary. What Andrea is suggesting seems
> reasonable?
>
Thanks for sharing the AMD configurations. Yeah, agreed on Andrea's suggestion.
- Joel
Powered by blists - more mailing lists