linux-kernel - [RFC] sched: idle: Proposal to remove idle=poll

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] [day] [month] [year] [list]

Message-ID: <jv5aqz342nsq5s2ducloqcwt55naxl3iafdaz33ynp3yzea2pc@kycn6jegsf2w>
Date: Sun, 7 Sep 2025 19:18:01 -0400
From: Aaron Tomlin <atomlin@...mlin.com>
To: Mel Gorman <mgorman@...hsingularity.net>, 
	Peter Zijlstra <peterz@...radead.org>
Cc: tglx@...utronix.de, mingo@...hat.com, bp@...en8.de, 
	dave.hansen@...ux.intel.com, x86@...nel.org, juri.lelli@...hat.com, 
	vincent.guittot@...aro.org, hpa@...or.com, oleg@...hat.com, dietmar.eggemann@....com, 
	rostedt@...dmis.org, bsegall@...gle.com, mgorman@...e.de, vschneid@...hat.com, 
	linux-kernel@...r.kernel.org, atomlin@...mlin.com
Subject: [RFC] sched: idle: Proposal to remove idle=poll

On Sun, Aug 31, 2025 at 06:29:18PM -0400, Aaron Tomlin wrote:
> Yes, I can confirm the PM QoS subsystem allows one to set constraints on a
> per-CPU basis or for the entire system. One can specify a maximum allowed
> latency constraint. As per function pm_qos_resume_latency_us_store() a
> value of "n/a" will prevent the specified CPU from entering the shallowest
> CPU idle-state (namely "C1") given its exit latency constraint.
> Indeed using "idle=poll" to prevent a CPU from entering idle C-states is
> problematic given its crude, all-or-nothing approach.
> 
> Thank you for the suggestion.

Mel, Peter,

I wanted to ask given the availability of the PM QoS subsystem on a Linux
kernel that has Kconfig option CONFIG_PM enabled, does this feature make
'idle=poll' redundant? Furthermore, if I understand correctly, for
platforms that require a generic idle polling mechanism
(CONFIG_GENERIC_IDLE_POLL_SETUP) boot-time parameter 'nohlt' can still be
used, no?

I was thinking about submitting a RFC patch to remove idle=poll.


Kind regards,

-- 
Aaron Tomlin