linux-kernel - Re: [PATCH 3.10] cgroup: break infinite loop in __css

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: <20141202193538.GO10918@htj.dyndns.org>
Date:	Tue, 2 Dec 2014 14:35:38 -0500
From:	Tejun Heo <tj@...nel.org>
To:	Roman Gushchin <klamm@...dex-team.ru>
Cc:	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Li Zefan <lizefan@...wei.com>
Subject: Re: [PATCH 3.10] cgroup: break infinite loop in __css_tryget()

On Tue, Dec 02, 2014 at 09:40:45PM +0300, Roman Gushchin wrote:
> Hi, Tejun!
> 
> 02.12.2014, 19:56, "Tejun Heo" <tj@...nel.org>:
> > Hello, Roman.
> >
> > On Fri, Nov 28, 2014 at 07:47:54PM +0300, Roman Gushchin wrote:
> >>  If cgroup_destroy_locked() sets the css refcount to a negative value,
> >>  we get an infinite loop in __css_tryget().
> >>
> >>  In this case css_refcnt() returns modified (non-negative value), so
> >>  both (t == v) and (t < 0) conditions are always false.
> >
> > I don't follow.  The count is biased and modified by unbiasing iff the
> > value is negative.  Here, @v is the unbiased value and @t is the
> > verbatim value.  If @v is different from @t due to unbiasing, @t must
> > be negative satisfying the second condition and returning NULL, no?
> 
> Yep. I missed that we compare t with 0 (not v).

Hmm... what's up with the rcu stall message then?

-- 
tejun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/