linux-kernel - Re: [PATCH] autofs4 - fix get_next_positive

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: <1337757721.2225.8.camel@perseus.themaw.net>
Date:	Wed, 23 May 2012 15:22:01 +0800
From:	Ian Kent <ikent@...hat.com>
To:	Linus Torvalds <torvalds@...ux-foundation.org>
Cc:	Kernel Mailing List <linux-kernel@...r.kernel.org>,
	linux-fsdevel <linux-fsdevel@...r.kernel.org>,
	autofs mailing list <autofs@...ux.kernel.org>
Subject: Re: [PATCH] autofs4 - fix get_next_positive_subdir()

On Tue, 2012-05-22 at 19:53 -0700, Linus Torvalds wrote:
> On Tue, May 22, 2012 at 7:49 PM, Ian Kent <ikent@...hat.com> wrote:
> >
> > The locking for the list traversal in get_next_positive_subdir() is
> > wrong, so fix it.
> 
> As an explanation, this kind of thing is totally useless. It doesn't
> actually give any information at all. It's like saying "change
> locking"
> 
> What happened, and why? Why is the new nested spinlock ok and won't
> deadlock against other nested users? Wazzup?

It's good that you questioned this Linus.

Looking again at dput() I think the traversal still isn't quite right.

For a start the test for d_count 0 or positive and hashed can never be
true since the point of the change was to take the d_lock of the
d_subdirs dentry for the traversal.

I'll need to work on this some more, thanks.
Ian

--
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/