From: Peter Zijlstra <peterz@infradead.org>
To: Valentin Schneider <vschneid@redhat.com>
Cc: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>,
Juri Lelli <juri.lelli@redhat.com>,
Vincent Guittot <vincent.guittot@linaro.org>,
Ricardo Neri <ricardo.neri@intel.com>,
"Ravi V. Shankar" <ravi.v.shankar@intel.com>,
Ben Segall <bsegall@google.com>,
Daniel Bristot de Oliveira <bristot@redhat.com>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Len Brown <len.brown@intel.com>, Mel Gorman <mgorman@suse.de>,
"Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>,
Steven Rostedt <rostedt@goodmis.org>,
Tim Chen <tim.c.chen@linux.intel.com>,
Ionela Voinescu <ionela.voinescu@arm.com>,
x86@kernel.org, linux-kernel@vger.kernel.org,
"Tim C . Chen" <tim.c.chen@intel.com>
Subject: Re: [PATCH v3 06/10] sched/fair: Use the prefer_sibling flag of the current sched domain
Date: Fri, 10 Feb 2023 17:53:03 +0100 [thread overview]
Message-ID: <Y+Z2b/OtZDk9cT53@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <xhsmhmt5lr2nz.mognet@vschneid.remote.csb>
On Fri, Feb 10, 2023 at 02:54:56PM +0000, Valentin Schneider wrote:
> So something like have SD_PREFER_SIBLING affect the SD it's on (and not
> its parent), but remove it from the lowest non-degenerated topology level?
So I was rather confused about the whole moving it between levels things
this morning -- conceptually, prefer siblings says you want to try
sibling domains before filling up your current domain. Now, balancing
between siblings happens one level up, hence looking at child->flags
makes perfect sense.
But looking at the current domain and still calling it prefer sibling
makes absolutely no sense what so ever.
In that confusion I think I also got the polarity wrong, I thought you
wanted to kill prefer_sibling for the assymetric SMT cases, instead you
want to force enable it as long as there is one SMT child around.
Whichever way around it we do it, I'm thinking perhaps some renaming
might be in order to clarify things.
How about adding a flag SD_SPREAD_TASKS, which is the effective toggle
of the behaviour, but have it be set by children with SD_PREFER_SIBLING
or something.
OTOH, there's also
if (busiest->group_weight == 1 || sds->prefer_sibling) {
which explicitly also takes the group-of-one (the !child case) into
account, but that's not consistently done.
sds->prefer_sibling = !child || child->flags & SD_PREFER_SIBLING;
seems an interesting option, except perhaps ASYM_CPUCAPACITY -- I
forget, can CPUs of different capacity be in the same leaf domain? With
big.LITTLE I think not, they had their own cache domains and so you get
at least MC domains per capacity, but DynamiQ might have totally wrecked
that party.
> (+ add it to the first NUMA level to keep things as they are, even if TBF I
> find relying on it for NUMA balancing a bit odd).
Arguably it ought to perhaps be one of those node_reclaim_distance
things. The thing is that NUMA-1 is often fairly quick, esp. these days
where it's basically on die numa.
next prev parent reply other threads:[~2023-02-10 16:53 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-07 4:58 [PATCH v3 00/10] sched/fair: Avoid unnecessary migrations within SMT domains Ricardo Neri
2023-02-07 4:58 ` [PATCH v3 01/10] sched/fair: Generalize asym_packing logic for SMT cores Ricardo Neri
2023-02-07 4:58 ` [PATCH v3 02/10] sched/fair: Move is_core_idle() out of CONFIG_NUMA Ricardo Neri
2023-02-07 4:58 ` [PATCH v3 03/10] sched/fair: Only do asym_packing load balancing from fully idle SMT cores Ricardo Neri
2023-02-07 4:58 ` [PATCH v3 04/10] sched/fair: Let low-priority cores help high-priority busy " Ricardo Neri
2023-02-08 7:56 ` Vincent Guittot
2023-02-09 11:53 ` Peter Zijlstra
2023-02-10 0:43 ` Ricardo Neri
2023-02-10 8:41 ` Peter Zijlstra
2023-02-10 13:05 ` Ricardo Neri
2023-02-10 1:52 ` Ricardo Neri
2023-02-13 13:40 ` Dietmar Eggemann
2023-02-13 23:23 ` Ricardo Neri
2023-03-10 0:51 ` Tim Chen
2023-03-14 23:54 ` Ricardo Neri
2023-02-07 4:58 ` [PATCH v3 05/10] sched/fair: Keep a fully_busy SMT sched group as busiest Ricardo Neri
2023-02-07 4:58 ` [PATCH v3 06/10] sched/fair: Use the prefer_sibling flag of the current sched domain Ricardo Neri
2023-02-08 7:48 ` Vincent Guittot
2023-02-10 13:24 ` Ricardo Neri
2023-02-09 13:17 ` Chen Yu
2023-02-09 20:00 ` Chen, Tim C
2023-02-09 23:05 ` Tim Chen
2023-02-10 3:16 ` Ricardo Neri
2023-02-10 6:55 ` Chen Yu
2023-02-10 10:08 ` Peter Zijlstra
2023-02-10 14:54 ` Valentin Schneider
2023-02-10 16:53 ` Peter Zijlstra [this message]
2023-02-10 17:12 ` Valentin Schneider
2023-02-10 18:31 ` Ricardo Neri
2023-02-13 12:17 ` Dietmar Eggemann
2023-02-14 6:43 ` Ricardo Neri
2023-02-16 5:21 ` Ricardo Neri
2023-02-16 12:16 ` Peter Zijlstra
2023-02-17 1:41 ` Ricardo Neri
2023-02-23 10:09 ` Dietmar Eggemann
2023-02-24 12:29 ` Ricardo Neri
[not found] ` <tencent_60A804570F09C0CFE0495D5B984941123A05@qq.com>
2023-02-20 9:45 ` Valentin Schneider
[not found] ` <tencent_6C38D389245FD03C6E1312999FEDD394F606@qq.com>
2023-02-21 18:15 ` Valentin Schneider
2023-02-07 4:58 ` [PATCH v3 07/10] sched/fair: Do not even the number of busy CPUs via asym_packing Ricardo Neri
2023-02-13 12:44 ` Dietmar Eggemann
2023-02-13 19:47 ` Ricardo Neri
2023-02-07 4:58 ` [PATCH v3 08/10] sched/topology: Remove SHARED_CHILD from ASYM_PACKING Ricardo Neri
2023-03-03 11:29 ` Ionela Voinescu
2023-03-05 19:08 ` Ricardo Neri
2023-03-06 13:10 ` Ionela Voinescu
2023-03-06 18:17 ` Ricardo Neri
2023-02-07 4:58 ` [PATCH v3 09/10] x86/sched: Remove SD_ASYM_PACKING from the SMT domain flags Ricardo Neri
2023-02-07 4:58 ` [PATCH v3 10/10] x86/sched/itmt: Give all SMT siblings of a core the same priority Ricardo Neri
2023-02-09 8:07 ` [PATCH v3 00/10] sched/fair: Avoid unnecessary migrations within SMT domains Zhang, Rui
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=Y+Z2b/OtZDk9cT53@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=bristot@redhat.com \
--cc=bsegall@google.com \
--cc=dietmar.eggemann@arm.com \
--cc=ionela.voinescu@arm.com \
--cc=juri.lelli@redhat.com \
--cc=len.brown@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=rafael.j.wysocki@intel.com \
--cc=ravi.v.shankar@intel.com \
--cc=ricardo.neri-calderon@linux.intel.com \
--cc=ricardo.neri@intel.com \
--cc=rostedt@goodmis.org \
--cc=srinivas.pandruvada@linux.intel.com \
--cc=tim.c.chen@intel.com \
--cc=tim.c.chen@linux.intel.com \
--cc=vincent.guittot@linaro.org \
--cc=vschneid@redhat.com \
--cc=x86@kernel.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
Powered by JetHome