linux-kernel - Re: [PATCH 2/4] lib/percpu-refcount: introduce 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: <20180919030909.GC20560@ming.t460p>
Date:   Wed, 19 Sep 2018 11:09:10 +0800
From:   Ming Lei <ming.lei@...hat.com>
To:     Tejun Heo <tj@...nel.org>
Cc:     linux-kernel@...r.kernel.org,
        Jianchao Wang <jianchao.w.wang@...cle.com>,
        Kent Overstreet <kent.overstreet@...il.com>,
        linux-block@...r.kernel.org
Subject: Re: [PATCH 2/4] lib/percpu-refcount: introduce percpu_ref_resurge()

On Tue, Sep 18, 2018 at 05:50:21AM -0700, Tejun Heo wrote:
> On Tue, Sep 18, 2018 at 06:19:44PM +0800, Ming Lei wrote:
> > Now percpu_ref_reinit() can only be done on one percpu refcounter when it
> > drops zero. And the limit shouldn't be so strict, and it is quite
> > straightforward that we can do it when the refcount doesn't drop zero
> > because it is at atomic mode.
> > 
> > This patch introduces percpu_ref_resurge() in which the above limit is
> > relaxed, so we may avoid extra change[1] for NVMe timeout's requirement.
> 
> For now,
> 
> Nacked-by: Tejun Heo <tj@...nel.org>
> 
> Please see the original discussion thread.

Your comment in that thread supposes that synchronize_rcu() is used for
avoiding race with .release().

But this patchset doesn't use that approach at all for avoiding this
race.

Thanks,
Ming