linux-kernel - Re: [PATCH 2/4] sched/uclamp: Remove rq max aggregation

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]

Date:   Mon, 11 Dec 2023 00:08:09 +0000
From:   Qais Yousef <qyousef@...alina.io>
To:     Ingo Molnar <mingo@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        "Rafael J. Wysocki" <rafael@...nel.org>,
        Viresh Kumar <viresh.kumar@...aro.org>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>
Cc:     linux-kernel@...r.kernel.org, linux-pm@...r.kernel.org,
        Lukasz Luba <lukasz.luba@....com>, Wei Wang <wvw@...gle.com>,
        Rick Yiu <rickyiu@...gle.com>,
        Chung-Kai Mei <chungkai@...gle.com>,
        Hongyan Xia <hongyan.xia2@....com>
Subject: Re: [PATCH 2/4] sched/uclamp: Remove rq max aggregation

On 12/08/23 01:52, Qais Yousef wrote:

> +/* Request freq update on context switch if necessary */
> +static inline void uclamp_context_switch(struct rq *rq)
>  {
> -	return static_branch_likely(&sched_uclamp_used);
> +	unsigned long uclamp_min;
> +	unsigned long uclamp_max;
> +	unsigned long util;
> +
> +	/* Only RT and FAIR tasks are aware of uclamp */
> +	if (!rt_policy(current->policy) && !fair_policy(current->policy))
> +		return;

We have a dependency on min_granularity_ns (or base_slice_ns) here that
I forgot to add before posting.

If our base_slice_ns is smaller than dvfs_update_delay, then tasks won't run
long enough for the hardware to apply their performance hints before they get
context switched out.

Beside the new proposed sched_runtime being able to request a smaller slice; in
practice default base_slice_ns is okay-ish.

> +
> +	uclamp_min = uclamp_eff_value(current, UCLAMP_MIN);
> +	uclamp_max = uclamp_eff_value(current, UCLAMP_MAX);
> +	util = rq->cfs.avg.util_avg;
> +
> +	if (uclamp_min > util || uclamp_max < util)
> +		cpufreq_update_util(rq, SCHED_CPUFREQ_PERF_HINTS);
>  }