linux-kernel - Re: [PATCH 3/4] sched/eevdf: O(1) fastpath for task selection

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: <20231107104207.GT8262@noisy.programming.kicks-ass.net>
Date:   Tue, 7 Nov 2023 11:42:07 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     Abel Wu <wuyun.abel@...edance.com>
Cc:     Ingo Molnar <mingo@...nel.org>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Valentin Schneider <valentin.schneider@....com>,
        Barry Song <21cnbao@...il.com>,
        Benjamin Segall <bsegall@...gle.com>,
        Chen Yu <yu.c.chen@...el.com>,
        Daniel Jordan <daniel.m.jordan@...cle.com>,
        "Gautham R . Shenoy" <gautham.shenoy@....com>,
        Joel Fernandes <joel@...lfernandes.org>,
        K Prateek Nayak <kprateek.nayak@....com>,
        Mike Galbraith <efault@....de>,
        Qais Yousef <qyousef@...alina.io>,
        Tim Chen <tim.c.chen@...ux.intel.com>,
        Yicong Yang <yangyicong@...wei.com>,
        Youssef Esmat <youssefesmat@...omium.org>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/4] sched/eevdf: O(1) fastpath for task selection

On Tue, Nov 07, 2023 at 06:12:49PM +0800, Abel Wu wrote:

> > @@ -904,7 +905,7 @@ static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq)
> >   	 * in this cfs_rq, saving some cycles.
> >   	 */
> >   	if (cfs_rq->nr_running == 1)
> > -		return curr && curr->on_rq ? curr : __node_2_se(node);
> > +		return curr && curr->on_rq ? curr : se;
> 
> Maybe we can reduce memory footprint on curr by:
> 
> 		return se ? se : curr;

Irrespective, I think that logic makes more sense. If we know we have
but one task and the tree has a task, it must be that task, otherwise,
current must be it. 

Anyway, I was still staring at the previous patch, flipping the tree
around like that is clever. Yes I suppose that ought to work just fine.