[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAP_z_Ci=E3Bbb3sncDfktugNrzUK74nESEM-qLBPubyx+W4HeA@mail.gmail.com>
Date: Mon, 8 Sep 2025 15:26:10 -0700
From: Blake Jones <blakejones@...gle.com>
To: Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>
Cc: Juri Lelli <juri.lelli@...hat.com>, Vincent Guittot <vincent.guittot@...aro.org>,
Madadi Vineeth Reddy <vineethr@...ux.ibm.com>, Dietmar Eggemann <dietmar.eggemann@....com>,
Steven Rostedt <rostedt@...dmis.org>, Ben Segall <bsegall@...gle.com>, Josh Don <joshdon@...gle.com>,
Mel Gorman <mgorman@...e.de>, Valentin Schneider <vschneid@...hat.com>,
LKML <linux-kernel@...r.kernel.org>
Subject: [Ping] Re: [PATCH v2] Reorder some fields in struct rq.
Hi Josh,
On Wed, Aug 20, 2025 at 5:11 PM Josh Don <joshdon@...gle.com> wrote:
> On Wed, Jul 30, 2025 at 1:56 PM Blake Jones <blakejones@...gle.com> wrote:
> > This colocates some hot fields in "struct rq" to be on the same cache line
> > as others that are often accessed at the same time or in similar ways.
>
> Thanks for the analysis and this patch.
>
> I was going to suggest ____cacheline_aligned_in_smp, but it'll behave
> nearly identical in practice so it doesn't matter (would save 64 bytes
> on a 128 byte cacheline UP system).
Thanks for your comments. I suspect there aren't a lot of
128-byte-cacheline UP systems out there, so I'm not going to worry
about this.
> Peter, any thoughts on this patch?
Echoing this - any thoughts from the maintainers?
Blake
Powered by blists - more mailing lists