From: Valentin Schneider <valentin.schneider@arm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: "chengjian (D)" <cj.chengjian@huawei.com>,
mingo@kernel.org, linux-kernel@vger.kernel.org,
chenwandun@huawei.com, xiexiuqi@huawei.com, liwei391@huawei.com,
huawei.libin@huawei.com, bobo.shaobowang@huawei.com,
juri.lelli@redhat.com, vincent.guittot@linaro.org
Subject: Re: [PATCH] sched/fair: Optimize select_idle_cpu
Date: Fri, 13 Dec 2019 12:20:03 +0000 [thread overview]
Message-ID: <0a0744f4-24cf-f8b6-cc91-f63847560675@arm.com> (raw)
In-Reply-To: <20191213120913.GB2844@hirez.programming.kicks-ass.net>
On 13/12/2019 12:09, Peter Zijlstra wrote:
>> Like you said the gains here would probably be small - the highest SMT
>> count I'm aware of is SMT8 (POWER9). Still, if we end up with both
>> select_idle_core() and select_idle_cpu() using that pattern, it would make
>> sense IMO to align select_idle_smt() with those.
>
> The cpumask_and() operation added would also have cost. I really don't
> see that paying off.
>
> The other sites have the problem that we combine an iteration limit with
> affinity constraints. This loop doesn't do that and therefore doesn't
> suffer the problem.
>
select_idle_core() doesn't really have an iteration limit, right? That
being said, yeah, the cpumask_and() for e.g. SMT2 systems would be
mostly wasteful.
prev parent reply other threads:[~2019-12-13 20:36 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-12 14:41 [PATCH] sched/fair: Optimize select_idle_cpu Cheng Jian
2019-12-12 14:56 ` chengjian (D)
2019-12-12 15:04 ` Peter Zijlstra
2019-12-13 1:51 ` chengjian (D)
2019-12-13 8:10 ` Peter Zijlstra
2019-12-12 15:24 ` Peter Zijlstra
2019-12-12 15:44 ` Valentin Schneider
2019-12-13 9:47 ` chengjian (D)
2019-12-13 9:57 ` chengjian (D)
2019-12-13 11:28 ` Valentin Schneider
2019-12-13 12:09 ` Peter Zijlstra
2019-12-13 12:20 ` Valentin Schneider [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=0a0744f4-24cf-f8b6-cc91-f63847560675@arm.com \
--to=valentin.schneider@arm.com \
--cc=bobo.shaobowang@huawei.com \
--cc=chenwandun@huawei.com \
--cc=cj.chengjian@huawei.com \
--cc=huawei.libin@huawei.com \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=liwei391@huawei.com \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=vincent.guittot@linaro.org \
--cc=xiexiuqi@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).