linux-kernel - Re: [PATCH] percpu-refcount: relax limit on percpu_ref

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: <20180911134404.GF1100574@devbig004.ftw2.facebook.com>
Date:   Tue, 11 Sep 2018 06:44:04 -0700
From:   Tejun Heo <tj@...nel.org>
To:     "jianchao.wang" <jianchao.w.wang@...cle.com>
Cc:     Ming Lei <ming.lei@...hat.com>, linux-kernel@...r.kernel.org,
        Kent Overstreet <kent.overstreet@...il.com>,
        linux-block@...r.kernel.org
Subject: Re: [PATCH] percpu-refcount: relax limit on percpu_ref_reinit()

Hello,

On Tue, Sep 11, 2018 at 12:40:36PM +0800, jianchao.wang wrote:
> (long)(sum (every cpu's refcounter)) + atomic refcounter
> 
> and when switch to atomic mode, there could be value left in atomic refcounter.
> then the unsigned long percpu refcounter cound be decreased from 0.

It can always do that.  Imagine a task getting a ref on one cpu,
migrating and putting it on another.  The only thing which matters is
that the sum of everything adds upto what it should be.

Thanks.

-- 
tejun