linux-kernel - Re: [PATCH] sched: Skip useless sched_balance_running acquisition if load balance is not due

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <b38301cb-6d8a-40c5-a84b-c59711ea1548@intel.com>
Date: Wed, 4 Jun 2025 12:26:23 +0800
From: "Chen, Yu C" <yu.c.chen@...el.com>
To: Tim Chen <tim.c.chen@...ux.intel.com>
CC: Vincent Guittot <vincent.guittot@...aro.org>, Doug Nelson
	<doug.nelson@...el.com>, Mohini Narkhede <mohini.narkhede@...el.com>,
	<linux-kernel@...r.kernel.org>, Peter Zijlstra <peterz@...radead.org>, "Ingo
 Molnar" <mingo@...nel.org>
Subject: Re: [PATCH] sched: Skip useless sched_balance_running acquisition if
 load balance is not due

On 4/16/2025 11:58 AM, Tim Chen wrote:
> At load balance time, balance of last level cache domains and
> above needs to be serialized. The scheduler checks the atomic var
> sched_balance_running first and then see if time is due for a load
> balance. This is an expensive operation as multiple CPUs can attempt
> sched_balance_running acquisition at the same time.
> 
> On a 2 socket Granite Rapid systems enabling sub-numa cluster and
> running OLTP workloads, 7.6% of cpu cycles are spent on cmpxchg of
> sched_balance_running.  Most of the time, a balance attempt is aborted
> immediately after acquiring sched_balance_running as load balance time
> is not due.
> 
> Instead, check balance due time first before acquiring
> sched_balance_running. This skips many useless acquisitions
> of sched_balance_running and knocks the 7.6% CPU overhead on
> sched_balance_domain() down to 0.05%.  Throughput of the OLTP workload
> improved by 11%.
> 
> Signed-off-by: Tim Chen <tim.c.chen@...ux.intel.com>
> Reported-by: Mohini Narkhede <mohini.narkhede@...el.com>
> Tested-by: Mohini Narkhede <mohini.narkhede@...el.com>
> ---

This change is straightforward(simple enough) to mitigate the
costly atomic operation, it looks good from my understanding, so:

Reviewed-by: Chen Yu yu.c.chen@...el.com


thanks,
Chenyu