From: K Prateek Nayak <kprateek.nayak@amd.com>
To: Peter Zijlstra <peterz@infradead.org>, Mario Roy <marioeroy@gmail.com>
Cc: Chris Mason <clm@meta.com>,
Joseph Salisbury <joseph.salisbury@oracle.com>,
Adam Li <adamli@os.amperecomputing.com>,
"Hazem Mohamed Abuelfotoh" <abuehaze@amazon.com>,
Josh Don <joshdon@google.com>, <mingo@redhat.com>,
<juri.lelli@redhat.com>, <vincent.guittot@linaro.org>,
<dietmar.eggemann@arm.com>, <rostedt@goodmis.org>,
<bsegall@google.com>, <mgorman@suse.de>, <vschneid@redhat.com>,
<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 4/4] sched/fair: Proportional newidle balance
Date: Fri, 23 Jan 2026 17:54:30 +0530 [thread overview]
Message-ID: <52b5b07d-89ee-4266-a229-ce9a8c6f8cb1@amd.com> (raw)
In-Reply-To: <20260123110306.GA217302@noisy.programming.kicks-ass.net>
Hello Peter,
On 1/23/2026 4:33 PM, Peter Zijlstra wrote:
> On Fri, Jan 23, 2026 at 11:50:46AM +0100, Peter Zijlstra wrote:
>> On Sun, Jan 18, 2026 at 03:46:22PM -0500, Mario Roy wrote:
>>> The patch "Proportional newidle balance" introduced a regression
>>> with Linux 6.12.65 and 6.18.5. There is noticeable regression with
>>> easyWave testing. [1]
>>>
>>> The CPU is AMD Threadripper 9960X CPU (24/48). I followed the source
>>> to install easyWave [2]. That is fetching the two tar.gz archives.
>>
>> What is the actual configuration of that chip? Is it like 3*8 or 4*6
>> (CCX wise). A quick google couldn't find me the answer :/
>
> Obviously I found it right after sending this. It's a 4x6 config.
> Meaning it needs newidle to balance between those 4 domains.
>
> Pratheek -- are you guys still considering that SIS_NODE thing? That
> worked really well for workstation chips, but there were some issues on
> Epyc or so.
SIS_NODE was really turned out to be a trade-off between search
time vs search opportunity, especially when the system was heavily
overloaded.
Let me rebase those old patches and give it a spin over the weekend
on a couple of those large machines (128C/256T and 192C/384T per
socket) to see the damage. I'll update here by Tuesday or post out
a series if I see the situation having changed on the recent
kernels - some benchmarks had a completely different bottleneck
there when we looked closer last.
>
>>> #!/bin/bash
>>> # CXXFLAGS="-O3 $CXXFLAGS" ./configure
>>> # make -j8
>>>
>>> trap 'rm -f *.ssh *.idx *.log *.sshmax *.time' EXIT
>>>
>>> OMP_NUM_THREADS=48 ./src/easywave \
>>> -grid examples/e2Asean.grd -source examples/BengkuluSept2007.flt \
>>> -time 1200
>>>
>>>
>>> Before results with CachyOS 6.12.63-2 and 6.18.3-2 kernels.
I'll go look at the benchmark too to see if I can reproduce on my end
and get some stats for these too. Thanks for bringing it to my notice.
--
Thanks and Regards,
Prateek
next prev parent reply other threads:[~2026-01-23 12:24 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-07 16:06 [PATCH 0/4] sched: The newidle balance regression Peter Zijlstra
2025-11-07 16:06 ` [PATCH 1/4] sched/fair: Revert max_newidle_lb_cost bump Peter Zijlstra
2025-11-14 12:19 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2025-11-17 16:23 ` tip-bot2 for Peter Zijlstra
2025-11-07 16:06 ` [PATCH 2/4] sched/fair: Small cleanup to sched_balance_newidle() Peter Zijlstra
2025-11-10 13:55 ` Dietmar Eggemann
2025-11-10 14:04 ` Peter Zijlstra
2025-11-12 14:37 ` Shrikanth Hegde
2025-11-12 14:42 ` Peter Zijlstra
2025-11-12 15:08 ` Peter Zijlstra
2025-11-12 15:28 ` Shrikanth Hegde
2025-11-14 9:49 ` Peter Zijlstra
2025-11-14 10:22 ` Vincent Guittot
2025-11-14 11:05 ` Peter Zijlstra
2025-11-14 13:11 ` Vincent Guittot
2025-11-14 12:19 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2025-11-17 16:23 ` tip-bot2 for Peter Zijlstra
2025-11-07 16:06 ` [PATCH 3/4] sched/fair: Small cleanup to update_newidle_cost() Peter Zijlstra
2025-11-14 12:19 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2025-11-17 16:23 ` tip-bot2 for Peter Zijlstra
2025-11-07 16:06 ` [PATCH 4/4] sched/fair: Proportional newidle balance Peter Zijlstra
2025-11-10 13:55 ` Dietmar Eggemann
2025-11-11 9:07 ` Adam Li
2025-11-11 9:20 ` Peter Zijlstra
2025-11-12 12:04 ` Adam Li
2025-11-12 13:41 ` Peter Zijlstra
2025-11-12 15:42 ` Shrikanth Hegde
2025-11-14 9:35 ` Peter Zijlstra
2025-11-14 12:18 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2025-11-17 16:23 ` tip-bot2 for Peter Zijlstra
2026-01-18 20:46 ` [PATCH 4/4] " Mario Roy
2026-01-23 10:50 ` Peter Zijlstra
2026-01-23 11:03 ` Peter Zijlstra
2026-01-23 12:24 ` K Prateek Nayak [this message]
2026-01-28 4:08 ` K Prateek Nayak
2026-01-27 4:15 ` Mario Roy
2026-01-27 10:40 ` Peter Zijlstra
2026-01-27 15:17 ` Peter Zijlstra
2026-01-30 1:44 ` Mario Roy
2026-01-30 4:14 ` Mario Roy
2026-02-24 9:13 ` [tip: sched/core] sched/fair: More complex proportional " tip-bot2 for Peter Zijlstra
2026-01-25 12:22 ` [PATCH 4/4] sched/fair: Proportional " Mohamed Abuelfotoh, Hazem
2026-01-27 8:44 ` Peter Zijlstra
2026-01-28 15:48 ` Mohamed Abuelfotoh, Hazem
2026-01-29 9:19 ` Peter Zijlstra
2026-01-29 9:24 ` Peter Zijlstra
2026-01-30 16:12 ` Mohamed Abuelfotoh, Hazem
2026-01-30 13:16 ` Mohamed Abuelfotoh, Hazem
2026-02-02 10:51 ` Peter Zijlstra
2026-02-02 11:07 ` Mohamed Abuelfotoh, Hazem
2026-02-04 12:45 ` Mohamed Abuelfotoh, Hazem
2026-02-04 13:27 ` Peter Zijlstra
2026-02-04 13:59 ` Mohamed Abuelfotoh, Hazem
2026-02-04 14:05 ` Peter Zijlstra
2026-02-04 22:48 ` Mohamed Abuelfotoh, Hazem
2026-01-27 8:50 ` Peter Zijlstra
2026-01-27 9:13 ` Peter Zijlstra
2026-01-28 16:24 ` Mohamed Abuelfotoh, Hazem
2026-01-28 16:03 ` Mohamed Abuelfotoh, Hazem
2026-04-29 8:51 ` Qing Wang
2026-05-06 2:44 ` [PATCH] sched/fair: Replace random newidle_balance with Bresenham accumulator Qing Wang
2025-11-10 19:47 ` [PATCH 0/4] sched: The newidle balance regression Chris Mason
2025-11-11 19:08 ` Josh Don
2025-11-12 21:59 ` Chris Mason
2025-11-14 9:37 ` Peter Zijlstra
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=52b5b07d-89ee-4266-a229-ce9a8c6f8cb1@amd.com \
--to=kprateek.nayak@amd.com \
--cc=abuehaze@amazon.com \
--cc=adamli@os.amperecomputing.com \
--cc=bsegall@google.com \
--cc=clm@meta.com \
--cc=dietmar.eggemann@arm.com \
--cc=joseph.salisbury@oracle.com \
--cc=joshdon@google.com \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=marioeroy@gmail.com \
--cc=mgorman@suse.de \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=vincent.guittot@linaro.org \
--cc=vschneid@redhat.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