linux-ext4 - Re: [PATCH 0/2] bigalloc + delalloc reservation leak fixes

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: <20180510002929.3zizzsmpaac3ppvd@localhost.localdomain>
Date:   Wed, 9 May 2018 20:29:29 -0400
From:   Eric Whitney <enwlinux@...il.com>
To:     Liu Bo <bo.liu@...ux.alibaba.com>
Cc:     "Theodore Y. Ts'o" <tytso@....edu>, linux-ext4@...r.kernel.org,
        Eric Whitney <enwlinux@...il.com>
Subject: Re: [PATCH 0/2] bigalloc + delalloc reservation leak fixes

* Liu Bo <bo.liu@...ux.alibaba.com>:
> On Mon, May 07, 2018 at 10:52:16AM -0400, Theodore Y. Ts'o wrote:
> > On Mon, May 07, 2018 at 10:18:10PM +0800, Liu Bo wrote:
> > > These two fixes are to address the reservation leak problem we've
> > > found when using bigalloc with delalloc.
> > > 
> > > A specific fstests case is available at
> > > https://patchwork.kernel.org/patch/10384221/
> > > 
> > > Liu Bo (2):
> > >   Ext4: bigalloc: fix overreservation on quota accounting
> > >   Ext4: bigalloc: do not reserve space for delalloc extents if there is
> > >     an available cluster
> > 
> > Hi Liu,
> > 
> > Eric has been working in this area as well, and has been working on a
> > series of patches.  (This had been discussed on the weekly ext4
> > teleconference calls; if this is something you would be interested in
> > joining, please contact me via e-mail privately.)
> 
> Glad to know others are looking into the same issue.
> 
> Will contact you about the time of the meeting (most time I'm base in
> the U.S. so that timezone is not a problem for me).
> 
> > 
> > My understanding is that his patch series is a bit more complex, since
> > he was working on a number of cleanups at the same time.  Eric, could
> > you send out your patch series, even if they aren't fully polished, so
> > we can see how much overlap there is between your work and Liu Bo's?
> >
> 
> Cool, I'm happy to review Eric's patches.
> 
> My two patches addressed the problem found by the posted fstests
> cases, maybe we can test Eric's patches against the testcase as a
> start.
> 
> thanks,
> -liubo


Hi Liu, Ted:

Sorry for the late reply - I've been offline with some other matters.

I've been working on a bug in my changes, so I'll post an RFC with them in a
day or so.  It'll contain the bug, but it's time to get some feedback on my
direction anyway.  I've been working on a general solution to bugzilla #151491,
and convinced myself along the way that a rewrite of bigalloc reserved cluster
accounting was required.  (I'd actually be happy to learn that I'm wrong about
that.)  I think the code covers the cases Liu was addressing with his patches,
as some of the changes are similar.

Thanks,
Eric