linux-kernel - Re: [PATCH -v2 12/17] sched,rt: Use cpumask_any*

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: <20201006155527.w6jck2rgk64t45wm@e107158-lin.cambridge.arm.com>
Date:   Tue, 6 Oct 2020 16:55:27 +0100
From:   Qais Yousef <qais.yousef@....com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     tglx@...utronix.de, mingo@...nel.org, linux-kernel@...r.kernel.org,
        bigeasy@...utronix.de, swood@...hat.com,
        valentin.schneider@....com, juri.lelli@...hat.com,
        vincent.guittot@...aro.org, dietmar.eggemann@....com,
        rostedt@...dmis.org, bsegall@...gle.com, mgorman@...e.de,
        bristot@...hat.com, vincent.donnefort@....com, tj@...nel.org
Subject: Re: [PATCH -v2 12/17] sched,rt: Use cpumask_any*_distribute()

On 10/05/20 16:57, Peter Zijlstra wrote:
> Replace a bunch of cpumask_any*() instances with
> cpumask_any*_distribute(), by injecting this little bit of random in
> cpu selection, we reduce the chance two competing balance operations
> working off the same lowest_mask pick the same CPU.
> 
> Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
> ---
>  include/linux/cpumask.h |    6 ++++++
>  kernel/sched/cpupri.c   |    4 ++--
>  kernel/sched/deadline.c |    2 +-
>  kernel/sched/rt.c       |    6 +++---
>  lib/cpumask.c           |   18 ++++++++++++++++++
>  5 files changed, 30 insertions(+), 6 deletions(-)
> 

[...]

> --- a/kernel/sched/rt.c
> +++ b/kernel/sched/rt.c
> @@ -1752,8 +1752,8 @@ static int find_lowest_rq(struct task_st
>  				return this_cpu;
>  			}
>  
> -			best_cpu = cpumask_first_and(lowest_mask,
> -						     sched_domain_span(sd));
> +			best_cpu = cpumask_any_and_distribute(lowest_mask,
> +							      sched_domain_span(sd));

I guess I should have done this 6 months ago and just got done with it :)

	20200414150556.10920-1-qais.yousef@....com

>  			if (best_cpu < nr_cpu_ids) {
>  				rcu_read_unlock();
>  				return best_cpu;
> @@ -1770,7 +1770,7 @@ static int find_lowest_rq(struct task_st
>  	if (this_cpu != -1)
>  		return this_cpu;
>  
> -	cpu = cpumask_any(lowest_mask);
> +	cpu = cpumask_any_distribute(lowest_mask);
>  	if (cpu < nr_cpu_ids)
>  		return cpu;
>  
> --- a/lib/cpumask.c
> +++ b/lib/cpumask.c
> @@ -267,3 +267,21 @@ int cpumask_any_and_distribute(const str
>  	return next;
>  }
>  EXPORT_SYMBOL(cpumask_any_and_distribute);
> +
> +int cpumask_any_distribute(const struct cpumask *srcp)
> +{
> +	int next, prev;
> +
> +	/* NOTE: our first selection will skip 0. */
> +	prev = __this_cpu_read(distribute_cpu_mask_prev);

We had a discussion then that __this_cpu*() variant assumes preemption being
disabled and it's safer to use this_cpu*() variant instead. Still holds true
here?

Thanks

--
Qais Yousef

> +
> +	next = cpumask_next(prev, srcp);
> +	if (next >= nr_cpu_ids)
> +		next = cpumask_first(srcp);
> +
> +	if (next < nr_cpu_ids)
> +		__this_cpu_write(distribute_cpu_mask_prev, next);
> +
> +	return next;
> +}
> +EXPORT_SYMBOL(cpumask_any_distribute);
> 
>