netdev - Re: [PATCH 07/16] smp: optimize smp_call_function_many

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: <YtXQGMijINmEoiVi@worktop.programming.kicks-ass.net>
Date:   Mon, 18 Jul 2022 23:26:48 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Yury Norov <yury.norov@...il.com>
Cc:     linux-kernel@...r.kernel.org,
        Alexander Lobakin <alexandr.lobakin@...el.com>,
        Alexei Starovoitov <ast@...nel.org>,
        Alexey Klimov <aklimov@...hat.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Andrii Nakryiko <andrii@...nel.org>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        Ben Segall <bsegall@...gle.com>,
        Christoph Lameter <cl@...ux.com>,
        Dan Williams <dan.j.williams@...el.com>,
        Daniel Borkmann <daniel@...earbox.net>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        Dennis Zhou <dennis@...nel.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Eric Dumazet <edumazet@...gle.com>,
        Frederic Weisbecker <fweisbec@...il.com>,
        Guenter Roeck <linux@...ck-us.net>,
        Ingo Molnar <mingo@...hat.com>,
        Isabella Basso <isabbasso@...eup.net>,
        John Fastabend <john.fastabend@...il.com>,
        Josh Poimboeuf <jpoimboe@...nel.org>,
        Juergen Gross <jgross@...e.com>,
        Juri Lelli <juri.lelli@...hat.com>,
        KP Singh <kpsingh@...nel.org>,
        Kees Cook <keescook@...omium.org>,
        Martin KaFai Lau <kafai@...com>,
        Mel Gorman <mgorman@...e.de>, Miroslav Benes <mbenes@...e.cz>,
        Nathan Chancellor <nathan@...nel.org>,
        "Paul E . McKenney" <paulmck@...nel.org>,
        Randy Dunlap <rdunlap@...radead.org>,
        Rasmus Villemoes <linux@...musvillemoes.dk>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        Song Liu <songliubraving@...com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Tejun Heo <tj@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Valentin Schneider <vschneid@...hat.com>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Vlastimil Babka <vbabka@...e.cz>, Yonghong Song <yhs@...com>,
        linux-mm@...ck.org, netdev@...r.kernel.org, bpf@...r.kernel.org
Subject: Re: [PATCH 07/16] smp: optimize smp_call_function_many_cond()

On Mon, Jul 18, 2022 at 12:28:35PM -0700, Yury Norov wrote:

> diff --git a/kernel/smp.c b/kernel/smp.c
> index dd215f439426..7ed2b9b12f74 100644
> --- a/kernel/smp.c
> +++ b/kernel/smp.c
> @@ -880,6 +880,28 @@ EXPORT_SYMBOL_GPL(smp_call_function_any);
>  #define SCF_WAIT	(1U << 0)
>  #define SCF_RUN_LOCAL	(1U << 1)
>  
> +/* Check if we need remote execution, i.e., any CPU excluding this one. */
> +static inline bool __need_remote_exec(const struct cpumask *mask, unsigned int this_cpu)
> +{
> +	unsigned int cpu;
> +
> +	switch (num_online_cpus()) {
> +	case 0:
> +		return false;
> +	case 1:
> +		return cpu_online(this_cpu) ? false : true;
> +	default:
> +		if (mask == cpu_online_mask)
> +			return true;
> +	}
> +
> +	cpu = cpumask_first_and(mask, cpu_online_mask);
> +	if (cpu == this_cpu)
> +		cpu = cpumask_next_and(cpu, mask, cpu_online_mask);
> +
> +	return cpu < nr_cpu_ids;
> +}
> +
>  static void smp_call_function_many_cond(const struct cpumask *mask,
>  					smp_call_func_t func, void *info,
>  					unsigned int scf_flags,
> @@ -916,12 +938,7 @@ static void smp_call_function_many_cond(const struct cpumask *mask,
>  	if ((scf_flags & SCF_RUN_LOCAL) && cpumask_test_cpu(this_cpu, mask))
>  		run_local = true;
>  
> -	/* Check if we need remote execution, i.e., any CPU excluding this one. */
> -	cpu = cpumask_first_and(mask, cpu_online_mask);
> -	if (cpu == this_cpu)
> -		cpu = cpumask_next_and(cpu, mask, cpu_online_mask);
> -	if (cpu < nr_cpu_ids)
> -		run_remote = true;
> +	run_remote = __need_remote_exec(mask, this_cpu);
>  
>  	if (run_remote) {
>  		cfd = this_cpu_ptr(&cfd_data);

This is more complex code for, very little to no gain. Why ?!