[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#762306: lockup when accessing cifs-mounted file on 3.17.0-rc1 and later when remote samba server is restarted



On Sat, 2014-09-27 at 13:36 +0930, Arthur Marsh wrote:
> I tried each of the -o vers=2.0 and the -o vers=3.0 options to the mount 
> -t cifs command with kernel 3.17.0-rc6 and also received a general 
> protection fault.
> 
> Arthur.
> 
> Steve French wrote on 27/09/14 02:02:
> > Am curious whether it also fails with vers=3.0 on mount
> >
> > On Fri, Sep 26, 2014 at 9:32 AM, Steve French <smfrench@gmail.com> wrote:
> >> cifs mount (default mount options) or mount with SMB2 or SMB3
> >> (vers=2.02 or vers=3.0)?
> >>
> >> On Fri, Sep 26, 2014 at 9:15 AM, Arthur Marsh
> >> <arthur.marsh@internode.on.net> wrote:
> >>> Arthur Marsh wrote on 16/09/14 03:17:
> >>>>
> >>>> Arthur Marsh wrote, on 14/09/14 21:15:
> >>>>>
> >>>>> On 3.16.0 kernels, whether compiled by myself or the stock Debian
> >>>>> kernel, if I have program accessing a file on a CIFS-mounted file system
> >>>>> (specifically playing an audio file), and restart the samba server on
> >>>>> the remote machine, I might get a brief pause, but no major problems.
> >>>>>
> >>>>> On 3.17.0-rc1 and later I get a complete machine lock-up.
> >>>>>
> >>>>> Illustration:
> >>>>>
> >>>>> https://pbs.twimg.com/media/BxfPkdgCcAAlpUo.jpg:large
> >>>>>
> >>>>> Has anyone else experienced similar?
> >>>>>
> >>>>> I'm happy to supply any further details and try test fixes.
> >>>>>
> >>>>> Arthur.
> >>>>
> >>>>
> >>>> I'm still getting the problem with kernel 3.17.0-rc5
> >>>>
> >>>> Has anyone been able to reproduce the problem?
> >>>>
> >>>> Arthur.
> >>>>
> >>>
> >>> I was off sick this past week but completed the bisect:
> >>>
> >>> git bisect good
> >>> 66386c08be5d1a2eefc1f7ab8c008561b6c811e5 is the first bad commit
> >>> commit 66386c08be5d1a2eefc1f7ab8c008561b6c811e5
> >>> Author: Pavel Shilovsky <pshilovsky@samba.org>
> >>> Date:   Fri Jun 20 15:48:40 2014 +0400
> >>>
> >>>      CIFS: Separate filling pages from iovec write
> >>>
> >>>      Reviewed-by: Shirish Pargaonkar <spargaonkar@suse.com>
> >>>      Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org>
> >>>      Signed-off-by: Steve French <smfrench@gmail.com>
> >>>
> >>> :040000 040000 1be960992f3d75b4998082fda3f3c2fca213b9e5
> >>> 318c46c63c983fe898480e1b88e97b818af1e502 M      fs
> >>>
> >>>
> >>> git bisect log:
> >>>
> >>> git bisect start
> >>> # good: [19583ca584d6f574384e17fe7613dfaeadcdc4a6] Linux 3.16
> >>> git bisect good 19583ca584d6f574384e17fe7613dfaeadcdc4a6
> >>> # bad: [7d1311b93e58ed55f3a31cc8f94c4b8fe988a2b9] Linux 3.17-rc1
> >>> git bisect bad 7d1311b93e58ed55f3a31cc8f94c4b8fe988a2b9
> >>> # good: [ae045e2455429c418a418a3376301a9e5753a0a8] Merge
> >>> git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
> >>> git bisect good ae045e2455429c418a418a3376301a9e5753a0a8
> >>> # good: [44c916d58b9ef1f2c4aec2def57fa8289c716a60] Merge tag
> >>> 'cleanup-for-3.17' of
> >>> git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
> >>> git bisect good 44c916d58b9ef1f2c4aec2def57fa8289c716a60
> >>> # skip: [023f78b02c729070116fa3a7ebd4107a032d3f5c] Merge branch 'for-next'
> >>> of git://git.samba.org/sfrench/cifs-2.6
> >>> git bisect skip 023f78b02c729070116fa3a7ebd4107a032d3f5c
> >>> # skip: [908790fa3b779d37365e6b28e3aa0f6e833020c3] dcache: d_splice_alias
> >>> mustn't create directory aliases
> >>> git bisect skip 908790fa3b779d37365e6b28e3aa0f6e833020c3
> >>> # bad: [58d08e3b2c2033354b91467da33deffa06360c28] Merge tag 'for-linus' of
> >>> git://git.kernel.org/pub/scm/linux/kernel/git/olof/chrome-platform
> >>> git bisect bad 58d08e3b2c2033354b91467da33deffa06360c28
> >>> # good: [27d438c56009e5ae632de36fe70985d1aab5e344] Revert "drm/i915: Enable
> >>> PSR by default."
> >>> git bisect good 27d438c56009e5ae632de36fe70985d1aab5e344
> >>> # good: [ee34fb97a96ceac3334705ebab8b541ca291699f] ARM: dts: exynos5420:
> >>> remove disp_pd
> >>> git bisect good ee34fb97a96ceac3334705ebab8b541ca291699f
> >>> # good: [d1e458fe671baf1e60afafc88bda090202a412f1] svcrdma: remove
> >>> rdma_create_qp() failure recovery logic
> >>> git bisect good d1e458fe671baf1e60afafc88bda090202a412f1
> >>> # good: [95484b57265caa671a57efed06e322d56461774b] drm/nouveau/ltc:
> >>> s/ltcg/ltc/ + cleanup
> >>> git bisect good 95484b57265caa671a57efed06e322d56461774b
> >>> # good: [63b12bdb0d21aca527996fb2c547387bfd3e14b8] Merge branch
> >>> 'signal-cleanup' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/misc
> >>> git bisect good 63b12bdb0d21aca527996fb2c547387bfd3e14b8
> >>> # bad: [96784de59fb35077c2bb33c39328992b836d87d3] Merge branch 'stable-3.17'
> >>> of git://git.infradead.org/users/pcmoore/selinux
> >>> git bisect bad 96784de59fb35077c2bb33c39328992b836d87d3
> >>> # bad: [25f402598d2c8f0808d93715ad33e43b265c1604] CIFS: Fix rsize usage in
> >>> user read
> >>> git bisect bad 25f402598d2c8f0808d93715ad33e43b265c1604
> >>> # good: [90ac1387c2dfcd9b4bd302fce03b9ddff73d0093] CIFS: Separate pages
> >>> initialization from writepages
> >>> git bisect good 90ac1387c2dfcd9b4bd302fce03b9ddff73d0093
> >>> # skip: [6ec0b01b2691d1465bb7219e031e8bf38ccd9397] CIFS: Fix wsize usage in
> >>> iovec write
> >>> git bisect skip 6ec0b01b2691d1465bb7219e031e8bf38ccd9397
> >>> # bad: [66386c08be5d1a2eefc1f7ab8c008561b6c811e5] CIFS: Separate filling
> >>> pages from iovec write
> >>> git bisect bad 66386c08be5d1a2eefc1f7ab8c008561b6c811e5
> >>> # good: [66231a47965c551d3056d5104f8b06688065748c] CIFS: Fix wsize usage in
> >>> writepages
> >>> git bisect good 66231a47965c551d3056d5104f8b06688065748c
> >>> # good: [7f6c50086a6f5bc0fee46548afc836070a439313] CIFS: Fix
> >>> cifs_writev_requeue when wsize changes
> >>> git bisect good 7f6c50086a6f5bc0fee46548afc836070a439313
> >>> # first bad commit: [66386c08be5d1a2eefc1f7ab8c008561b6c811e5] CIFS:
> >>> Separate filling pages from iovec write
> >>>
> >>>
> >>>
> >>>
> >>
> >>
> >>
> >> --
> >> Thanks,
> >>
> >> Steve
> >
> >
> >
> 

Sorry, easier for me at this address, so just cc'ed my gmail address.

Regards,

Shirish


Reply to: