[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20231215041027.GX2194132@linux.vnet.ibm.com>
Date: Fri, 15 Dec 2023 09:40:27 +0530
From: Srikar Dronamraju <srikar@...ux.vnet.ibm.com>
To: Ingo Molnar <mingo@...nel.org>, Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...hat.com>, 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>,
Daniel Bristot de Oliveira <bristot@...hat.com>,
Valentin Schneider <vschneid@...hat.com>
Cc: LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] sched/fair: Enable group_asym_packing in
find_idlest_group
* Srikar Dronamraju <srikar@...ux.vnet.ibm.com> [2023-10-18 21:20:35]:
Hi Ingo, Peter,
> Current scheduler code doesn't handle SD_ASYM_PACKING in the
> find_idlest_cpu path. On few architectures, like Powerpc, cache is at a
> core. Moving threads across cores may end up in cache misses.
>
> While asym_packing can be enabled above SMT level, enabling Asym packing
> across cores could result in poorer performance due to cache misses.
> However if the initial task placement via find_idlest_cpu does take
> Asym_packing into consideration, then scheduler can avoid asym_packing
> migrations. This will result in lesser migrations and better packing and
> better overall performance.
>
> Signed-off-by: Srikar Dronamraju <srikar@...ux.vnet.ibm.com>
I haven't heard any comments or seen any reviews on this patch.
I have verified that it does still apply cleanly on v6.7-rc5 based
tip/master
Do let me know your thoughts on the same.
Also let me know if you want me to repost the same.
--
Thanks and Regards
Srikar Dronamraju
Powered by blists - more mailing lists