From: Peter Zijlstra <peterz@infradead.org>
To: Mel Gorman <mgorman@techsingularity.net>
Cc: Vincent Guittot <vincent.guittot@linaro.org>,
Ingo Molnar <mingo@kernel.org>, Phil Auld <pauld@redhat.com>,
Valentin Schneider <valentin.schneider@arm.com>,
Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
Quentin Perret <quentin.perret@arm.com>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Morten Rasmussen <Morten.Rasmussen@arm.com>,
Hillf Danton <hdanton@sina.com>, Parth Shah <parth@linux.ibm.com>,
Rik van Riel <riel@surriel.com>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] sched, fair: Allow a small degree of load imbalance between SD_NUMA domains v2
Date: Tue, 7 Jan 2020 12:22:55 +0100 [thread overview]
Message-ID: <20200107112255.GV2827@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20200107095655.GF3466@techsingularity.net>
On Tue, Jan 07, 2020 at 09:56:55AM +0000, Mel Gorman wrote:
> Much more importantly, doing what you suggest allows an imbalance
> of more CPUs than are backed by a single LLC. On high-end AMD EPYC 2
> machines, busiest->group_weight scaled by imbalance_pct spans multiple L3
> caches. That is going to have side-effects. While I also do not account
> for the LLC group_weight, it's unlikely the cut-off I used would be
> smaller than an LLC cache on a large machine as the cache.
>
> These two points are why I didn't take the group weight into account.
>
> Now if you want, I can do what you suggest anyway as long as you are happy
> that the child domain weight is also taken into account and to bound the
> largest possible allowed imbalance to deal with the case of a node having
> multiple small LLC caches. That means that some machines will be using the
> size of the node and some machines will use the size of an LLC. It's less
> predictable overall as some machines will be "special" relative to others
> making it harder to reproduce certain problems locally but it would take
> imbalance_pct into account in a way that you're happy with.
>
> Also bear in mind that whether LLC is accounted for or not, the final
> result should be halved similar to the other imbalance calculations to
> avoid over or under load balancing.
> + /* Consider allowing a small imbalance between NUMA groups */
> + if (env->sd->flags & SD_NUMA) {
> + struct sched_domain *child = env->sd->child;
This assumes sd-child exists, which should be true for NUMA domains I
suppose.
> + unsigned int imbalance_adj;
> +
> + /*
> + * Calculate an acceptable degree of imbalance based
> + * on imbalance_adj. However, do not allow a greater
> + * imbalance than the child domains weight to avoid
> + * a case where the allowed imbalance spans multiple
> + * LLCs.
> + */
That comment is a wee misleading, @child is not an LLC per se. This
could be the NUMA distance 2 domain, in which case @child is the NUMA
distance 1 group.
That said, even then it probably makes sense to ensure you don't idle a
whole smaller distance group.
> + imbalance_adj = busiest->group_weight * (env->sd->imbalance_pct - 100) / 100;
> + imbalance_adj = min(imbalance_adj, child->span_weight);
> + imbalance_adj >>= 1;
> +
> + /*
> + * Ignore small imbalances when the busiest group has
> + * low utilisation.
> + */
> + if (busiest->sum_nr_running < imbalance_adj)
> + env->imbalance = 0;
> + }
> +
> return;
> }
>
next prev parent reply other threads:[~2020-01-07 11:23 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-20 8:42 [PATCH] sched, fair: Allow a small degree of load imbalance between SD_NUMA domains v2 Mel Gorman
2019-12-20 12:40 ` Valentin Schneider
2019-12-20 14:22 ` Mel Gorman
2019-12-20 15:32 ` Valentin Schneider
2019-12-21 11:25 ` Mel Gorman
2019-12-22 12:00 ` Srikar Dronamraju
2019-12-23 13:31 ` Vincent Guittot
2019-12-23 13:41 ` Vincent Guittot
2020-01-03 14:31 ` Mel Gorman
2020-01-06 13:55 ` Vincent Guittot
2020-01-06 14:52 ` Mel Gorman
2020-01-07 8:38 ` Vincent Guittot
2020-01-07 9:56 ` Mel Gorman
2020-01-07 11:17 ` Vincent Guittot
2020-01-07 11:56 ` Mel Gorman
2020-01-07 16:00 ` Vincent Guittot
2020-01-07 20:24 ` Mel Gorman
2020-01-08 8:25 ` Vincent Guittot
2020-01-08 8:49 ` Mel Gorman
2020-01-08 13:18 ` Peter Zijlstra
2020-01-08 14:03 ` Mel Gorman
2020-01-08 16:46 ` Vincent Guittot
2020-01-08 18:03 ` Mel Gorman
2020-01-07 11:22 ` Peter Zijlstra [this message]
2020-01-07 11:42 ` Mel Gorman
2020-01-07 12:29 ` Peter Zijlstra
2020-01-07 12:28 ` Peter Zijlstra
2020-01-07 19:26 ` Phil Auld
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=20200107112255.GV2827@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=Morten.Rasmussen@arm.com \
--cc=dietmar.eggemann@arm.com \
--cc=hdanton@sina.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@techsingularity.net \
--cc=mingo@kernel.org \
--cc=parth@linux.ibm.com \
--cc=pauld@redhat.com \
--cc=quentin.perret@arm.com \
--cc=riel@surriel.com \
--cc=srikar@linux.vnet.ibm.com \
--cc=valentin.schneider@arm.com \
--cc=vincent.guittot@linaro.org \
/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).