[Bug 110848] Everything using GPU gets stuck after running+killing parallel Media loads (after running 3D benchmarks)

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=110848

--- Comment #7 from Eero Tamminen <eero.t.tamminen at intel.com> ---
I've now verified that every time this freeze happens, these are stuck:
-----------------------------------------------
Name:   khugepaged
State:  D (disk sleep)
Pid:    38
[<0>] kthread+0x113/0x130
[<0>] ret_from_fork+0x35/0x40
--
Name:   kworker/u8:2+i915
State:  D (disk sleep)
Pid:    7911
[<0>] __i915_gem_free_work+0x5f/0x90 [i915]
[<0>] process_one_work+0x1e9/0x410
[<0>] worker_thread+0x2d/0x3d0
[<0>] kthread+0x113/0x130
[<0>] ret_from_fork+0x35/0x40
-----------------------------------------------

in addition to the stuck transcoding tasks:
-----------------------------------------------
...
Name:   sample_multi_tr
State:  D (disk sleep)
Pid:    20118
[<0>] __fput+0xae/0x200
[<0>] task_work_run+0x84/0xa0
[<0>] do_exit+0x308/0xba0
[<0>] do_group_exit+0x33/0xa0
[<0>] get_signal+0x121/0x910
[<0>] do_signal+0x30/0x6b0
[<0>] exit_to_usermode_loop+0x96/0xb0
[<0>] do_syscall_64+0x103/0x130
[<0>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
--
Name:   sample_multi_tr
State:  D (disk sleep)
Pid:    20221
[<0>] chrdev_open+0xa3/0x1b0
[<0>] do_dentry_open+0x1c4/0x380
[<0>] path_openat+0x564/0x11f0
[<0>] path_openat+0x564/0x11f0
[<0>] do_filp_open+0x9b/0x110
[<0>] do_sys_open+0x1bd/0x260
[<0>] do_syscall_64+0x4f/0x130
[<0>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
...
-----------------------------------------------

Above is from last evening drm-tip git build.  For some reason on ClearLinux
(comment 5) ffmpeg git build gets stuck, and on Ubuntu 18.04 (this comment)
it's the MSDK git build test app.  In both setups, kernel & Media stack are
using same binaries.  Main difference between these setups are:
- HW is supposed to be identical, but they are separate machines
- Weston vs. X server running in background (which aren't used in Media tests)
- optimizations in ClearLinux libc

I think the differences in freezing app are just due to some differences in
timings.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are on the CC list for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20190611/014c70e9/attachment-0001.html>