linux-kernel - Re: [PATCH] sched/preempt: Bypass setting same mode for sched_dynamic

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: <20240620123338.GX31592@noisy.programming.kicks-ass.net>
Date: Thu, 20 Jun 2024 14:33:38 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Jinjie Ruan <ruanjinjie@...wei.com>
Cc: mingo@...hat.com, juri.lelli@...hat.com, vincent.guittot@...aro.org,
	dietmar.eggemann@....com, rostedt@...dmis.org, bsegall@...gle.com,
	mgorman@...e.de, bristot@...hat.com, vschneid@...hat.com,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] sched/preempt: Bypass setting same mode for
 sched_dynamic_update()

On Thu, Jun 20, 2024 at 08:10:20PM +0800, Jinjie Ruan wrote:
> If the preempt mode to set is same with current preempt mode, there is no
> need to update all the `cond_resched`, `might_resched`, `preempt_schedule`,
> `preempt_schedule_notrace` and `irqentry_exit_cond_resched` state.

IIRC this actively breaks things, also this is a super slow path, nobody
should care.

> Signed-off-by: Jinjie Ruan <ruanjinjie@...wei.com>
> ---
>  kernel/sched/core.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index bcf2c4cc0522..eb409901c64c 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -8777,6 +8777,9 @@ static bool klp_override;
>  
>  static void __sched_dynamic_update(int mode)
>  {
> +	if (mode == preempt_dynamic_mode)
> +		return;
> +
>  	/*
>  	 * Avoid {NONE,VOLUNTARY} -> FULL transitions from ever ending up in
>  	 * the ZERO state, which is invalid.
> -- 
> 2.34.1
>