[Bug 196291] New: amdgpu: Freeze because of syscall not returning

[Bug 196291] New: amdgpu: Freeze because of syscall not returning

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=196291

            Bug ID: 196291
           Summary: amdgpu: Freeze because of syscall not returning
           Product: Drivers
           Version: 2.5
    Kernel Version: 4.11.8
          Hardware: x86-64
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: Video(DRI - non Intel)
          Assignee: drivers_video-dri@kernel-bugs.osdl.org
          Reporter: tobi291019@gmail.com
        Regression: No

Created attachment 257397
  --> https://bugzilla.kernel.org/attachment.cgi?id=257397&action=edit
dmesg with lockup warning at the end

An amdgpu syscall, called by plasmashell, appears to deadlock randomly and
freeze X.org completely. Several graphics processes, plasmashell and X.org are
left stuck in D-State. Everything else continues to operate correctly,
including audio, networking, etc..

The issue seems to appear more frequently whilst running games, although I am
unable to find any particular pattern to it.

Running Arch Linux with a custom compiled linux-zen kernel (with ACS override
patches) and ZFS, although as far as I can tell those are not related to the
issue, Mesa 17.1.4 with Radeon RX 480. The issue has been around for a while
and I sadly do not remember when it first occured, but definitely the entire
4.11.x lineup is affected and I am fairly sure 4.10.x was as well. The issue is
way too rare though for me to bisect the exact cause however.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 196291] amdgpu: Freeze because of syscall not returning

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=196291

Christian König (christian.koenig@amd.com) changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |christian.koenig@amd.com

--- Comment #1 from Christian König (christian.koenig@amd.com) ---
Please provide the output of "cat /sys/kernel/debug/dri/0/amdgpu_fence_info"
when this happens.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 196291] amdgpu: Freeze because of syscall not returning

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=196291

--- Comment #2 from Tobias Auerochs (tobi291019@gmail.com) ---
Created attachment 257449
  --> https://bugzilla.kernel.org/attachment.cgi?id=257449&action=edit
/sys/kernel/debug/dri/0/amdgpu_fence_info after being frozen for a few minutes

Got the freeze again randomly, attached the output from
/sys/kernel/debug/dri/0/amdgpu_fence_info.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 196291] amdgpu: Freeze because of syscall not returning

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=196291

--- Comment #3 from Christian König (christian.koenig@amd.com) ---
That isn't related to any system call. The problem is simply that the hardware
has crashed and some task is trying to push new commands to it, waiting for
previous commands to end (which never happens).

That is most likely a problem on the user space driver side and not related to
the kernel at all.

Please open a bug report on FDO for this.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 196291] amdgpu: Freeze because of syscall not returning

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=196291

--- Comment #4 from Tobias Auerochs (tobi291019@gmail.com) ---
Submitted on freedesktop.org bugzilla:
https://bugs.freedesktop.org/show_bug.cgi?id=101746

-- 
You are receiving this mail because:
You are watching the assignee of the bug.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 196291] amdgpu: Freeze because of syscall not returning

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=196291

Tobias Auerochs (tobi291019@gmail.com) changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |INVALID

--- Comment #5 from Tobias Auerochs (tobi291019@gmail.com) ---
Well, after encountering a possibly unrelated (reproducible) issue, causing the
exact same symptons and a GPU reset (in debugfs) seems to recover correctly
from that, I think this issue really just runs down to GPU resets not being
issued automatically on the kernel side yet.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel