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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Fri, 11 Aug 2023 19:33:54 -0700
From:   Shakeel Butt <shakeelb@...gle.com>
To:     Ivan Babrou <ivan@...udflare.com>
Cc:     Waiman Long <longman@...hat.com>, cgroups@...r.kernel.org,
        Linux MM <linux-mm@...ck.org>,
        kernel-team <kernel-team@...udflare.com>,
        Johannes Weiner <hannes@...xchg.org>,
        Michal Hocko <mhocko@...nel.org>,
        Roman Gushchin <roman.gushchin@...ux.dev>,
        Muchun Song <muchun.song@...ux.dev>,
        Andrew Morton <akpm@...ux-foundation.org>,
        linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: Expensive memory.stat + cpu.stat reads

Hi Ivan,

(sorry for late response as I was away)

On Fri, Aug 11, 2023 at 3:35 PM Ivan Babrou <ivan@...udflare.com> wrote:
[...]
> > > I spent some time looking into this and I think I landed on a fix:
> > >
> > > * https://github.com/bobrik/linux/commit/50b627811d54
> > >
> > > I'm not 100% sure if it's the right fix for the issue, but it reduces
> > > the runtime significantly.

In your patch, can you try to replace mem_cgroup_flush_stats() with
mem_cgroup_flush_stats_ratelimited() instead of cgroup_rstat_flush().
I wanted to see if you observe any stale stats issues.

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ