linux-ext4 - Re: [PATCH v4 1/2] dax: dax_layout_busy

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: <20180827160903.GF4002@quack2.suse.cz>
Date:   Mon, 27 Aug 2018 18:09:03 +0200
From:   Jan Kara <jack@...e.cz>
To:     "Theodore Y. Ts'o" <tytso@....edu>
Cc:     sandeen@...hat.com, Ross Zwisler <ross.zwisler@...ux.intel.com>,
        Jan Kara <jack@...e.cz>,
        Dan Williams <dan.j.williams@...el.com>,
        Dave Chinner <david@...morbit.com>,
        "Darrick J. Wong" <darrick.wong@...cle.com>,
        Christoph Hellwig <hch@....de>, linux-nvdimm@...ts.01.org,
        Jeff Moyer <jmoyer@...hat.com>, linux-ext4@...r.kernel.org,
        Lukas Czerner <lczerner@...hat.com>
Subject: Re: [PATCH v4 1/2] dax: dax_layout_busy_page() warn on !exceptional

On Mon 13-08-18 12:12:52, Jan Kara wrote:
> On Fri 10-08-18 22:10:53, Theodore Y. Ts'o wrote:
> > On Fri, Aug 10, 2018 at 04:33:49PM -0400, Theodore Y. Ts'o wrote:
> > > I just kicked off a DAX test ("gce-xfstests -c dax -g auto") with
> > > CONFIG_KASAN disabled, and I expect it shouldn't show up anything
> > > concerning.  So assuming nothing surprising pops up, yes it should be
> > > merged at the next merge window.
> > 
> > ... and here are the results.  The first is 4.17, and the second is
> > the ext4 git tree:
> > 
> > ext4/dax: 488 tests, 4 failures, 97 skipped, 2647 seconds
> >   Failures: ext4/033 generic/344 generic/491 generic/503
> > 
> > ext4/dax: 488 tests, 3 failures, 97 skipped, 2637 seconds
> >   Failures: generic/081 generic/344 generic/388
> > 
> > The generic/388 failure is a known flake (shutdown stress test).
> > 
> > The generic/081 regression appears to be a device-mapper issue:
> > 
> > generic/081		[22:06:33][   15.079661] run fstests generic/081 at 2018-08-10 22:06:33
> > [   15.795745] device-mapper: ioctl: can't change device type (old=4 vs new=1) after initial table load.
> > [failed, exit status 1] [22:06:36]- output mismatch (see /results/ext4/results-dax/generic/081.out.bad)
> >     --- tests/generic/081.out	2018-08-09 18:00:42.000000000 -0400
> >     +++ /results/ext4/results-dax/generic/081.out.bad	2018-08-10 22:06:36.440005460 -0400
> >     @@ -1,2 +1,4 @@
> >      QA output created by 081
> >      Silence is golden
> >     +Failed to create snapshot
> >     +(see /results/ext4/results-dax/generic/081.full for details)
> >     ...
> >     (Run 'diff -u tests/generic/081.out /results/ext4/results-dax/generic/081.out.bad'  to see the entire diff)
> 
> I'll see if this reproduces for me. Doesn't seem to be related to the DAX
> patches you caary though.

This is caused by commit dbc626597c39b "dm: prevent DAX mounts if not
supported". I've started discussion with relevant developers how to fix
this.

								Honza
-- 
Jan Kara <jack@...e.com>
SUSE Labs, CR