Bug 98903

Summary:	[AMDGPU] System hang with latest CS:GO update
Product:	DRI	Reporter:	Ernst Sjöstrand <ernstp>
Component:	DRM/AMDgpu	Assignee:	Default DRI bug account <dri-devel>
Status:	RESOLVED WORKSFORME	QA Contact:
Severity:	normal
Priority:	medium
Version:	DRI git
Hardware:	Other
OS:	All
Whiteboard:
i915 platform:		i915 features:

Description Ernst Sjöstrand 2016-11-29 15:48:19 UTC

Running the latest CS:GO update caused this hang. Before todays update it was rock solid.
Running under gnome-shell.

But I'm not running the latest kernel code, let me try to update...
Fiji card.
Mesa: 13.1~git161128183000.7c277ea~y~padoka0
LLVM: 4.0~svn287552-0~y~padoka0
libdrm: 2.4.73+git1611281831.a831583~y~padoka0

Nov 29 16:37:50 mammut kernel: [ 1088.547557] INFO: task Compositor:2862 blocked for more than 120 seconds.
Nov 29 16:37:50 mammut kernel: [ 1088.547563]       Not tainted 4.9.0-rc4+ #80
Nov 29 16:37:50 mammut kernel: [ 1088.547565] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Nov 29 16:37:50 mammut kernel: [ 1088.547567] Compositor      D    0  2862      1 0x00000000
Nov 29 16:37:50 mammut kernel: [ 1088.547571]  ffff980e057a8800 0000000000000000 ffff980e7f4192c0 ffff980e341e6c00
Nov 29 16:37:50 mammut kernel: [ 1088.547575]  ffffffffbec0e540 ffffaad5c2f3f940 ffffffffbe65b8d4 0000000000000000
Nov 29 16:37:50 mammut kernel: [ 1088.547578]  ffffaad5c2f3f920 00ffffffbe221517 ffff980e16bdf110 ffff980e341e6c00
Nov 29 16:37:50 mammut kernel: [ 1088.547581] Call Trace:
Nov 29 16:37:50 mammut kernel: [ 1088.547588]  [<ffffffffbe65b8d4>] ? __schedule+0x1a4/0x680
Nov 29 16:37:50 mammut kernel: [ 1088.547590]  [<ffffffffbe65bde6>] schedule+0x36/0x80
Nov 29 16:37:50 mammut kernel: [ 1088.547622]  [<ffffffffc03631c4>] amd_sched_entity_push_job+0xc4/0x110 [amdgpu]
Nov 29 16:37:50 mammut kernel: [ 1088.547625]  [<ffffffffbdec74f0>] ? wake_atomic_t_function+0x60/0x60
Nov 29 16:37:50 mammut kernel: [ 1088.547651]  [<ffffffffc0363c58>] amdgpu_job_submit+0x88/0xd0 [amdgpu]
Nov 29 16:37:50 mammut kernel: [ 1088.547672]  [<ffffffffc0325b1a>] amdgpu_vm_bo_split_mapping+0x50a/0x7b0 [amdgpu]
Nov 29 16:37:50 mammut kernel: [ 1088.547693]  [<ffffffffc0324f70>] ? amdgpu_gem_prime_export+0x60/0x60 [amdgpu]
Nov 29 16:37:50 mammut kernel: [ 1088.547712]  [<ffffffffc03271d0>] amdgpu_vm_bo_update+0x140/0x2f0 [amdgpu]
Nov 29 16:37:50 mammut kernel: [ 1088.547731]  [<ffffffffc0314c64>] amdgpu_gem_va_update_vm+0x194/0x1b0 [amdgpu]
Nov 29 16:37:50 mammut kernel: [ 1088.547735]  [<ffffffffbe226d01>] ? interval_tree_iter_next+0x51/0x70
Nov 29 16:37:50 mammut kernel: [ 1088.547754]  [<ffffffffc0315d5f>] amdgpu_gem_va_ioctl+0x21f/0x2e0 [amdgpu]
Nov 29 16:37:50 mammut kernel: [ 1088.547770]  [<ffffffffc0205f8b>] drm_ioctl+0x21b/0x4c0 [drm]
Nov 29 16:37:50 mammut kernel: [ 1088.547789]  [<ffffffffc0315b40>] ? amdgpu_gem_metadata_ioctl+0x1d0/0x1d0 [amdgpu]
Nov 29 16:37:50 mammut kernel: [ 1088.547804]  [<ffffffffc02fb04f>] amdgpu_drm_ioctl+0x4f/0x90 [amdgpu]
Nov 29 16:37:50 mammut kernel: [ 1088.547807]  [<ffffffffbe040c03>] do_vfs_ioctl+0xa3/0x600
Nov 29 16:37:50 mammut kernel: [ 1088.547810]  [<ffffffffbde6a516>] ? __do_page_fault+0x266/0x4e0
Nov 29 16:37:50 mammut kernel: [ 1088.547813]  [<ffffffffbe0411d9>] SyS_ioctl+0x79/0x90
Nov 29 16:37:50 mammut kernel: [ 1088.547816]  [<ffffffffbe660b3b>] entry_SYSCALL_64_fastpath+0x1e/0xad
Nov 29 16:39:01 mammut CRON[18393]: (root) CMD (  [ -x /usr/lib/php5/sessionclean ] && /usr/lib/php5/sessionclean)
Nov 29 16:39:51 mammut kernel: [ 1209.379338] INFO: task Compositor:2862 blocked for more than 120 seconds.
Nov 29 16:39:51 mammut kernel: [ 1209.379344]       Not tainted 4.9.0-rc4+ #80
Nov 29 16:39:51 mammut kernel: [ 1209.379345] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Nov 29 16:39:51 mammut kernel: [ 1209.379348] Compositor      D    0  2862      1 0x00000000
Nov 29 16:39:51 mammut kernel: [ 1209.379352]  ffff980e057a8800 0000000000000000 ffff980e7f4192c0 ffff980e341e6c00
Nov 29 16:39:51 mammut kernel: [ 1209.379355]  ffffffffbec0e540 ffffaad5c2f3f940 ffffffffbe65b8d4 0000000000000000
Nov 29 16:39:51 mammut kernel: [ 1209.379358]  ffffaad5c2f3f920 00ffffffbe221517 ffff980e16bdf110 ffff980e341e6c00
Nov 29 16:39:51 mammut kernel: [ 1209.379361] Call Trace:
Nov 29 16:39:51 mammut kernel: [ 1209.379369]  [<ffffffffbe65b8d4>] ? __schedule+0x1a4/0x680
Nov 29 16:39:51 mammut kernel: [ 1209.379371]  [<ffffffffbe65bde6>] schedule+0x36/0x80
Nov 29 16:39:51 mammut kernel: [ 1209.379404]  [<ffffffffc03631c4>] amd_sched_entity_push_job+0xc4/0x110 [amdgpu]
Nov 29 16:39:51 mammut kernel: [ 1209.379407]  [<ffffffffbdec74f0>] ? wake_atomic_t_function+0x60/0x60
Nov 29 16:39:51 mammut kernel: [ 1209.379431]  [<ffffffffc0363c58>] amdgpu_job_submit+0x88/0xd0 [amdgpu]
Nov 29 16:39:51 mammut kernel: [ 1209.379453]  [<ffffffffc0325b1a>] amdgpu_vm_bo_split_mapping+0x50a/0x7b0 [amdgpu]
Nov 29 16:39:51 mammut kernel: [ 1209.379472]  [<ffffffffc0324f70>] ? amdgpu_gem_prime_export+0x60/0x60 [amdgpu]
Nov 29 16:39:51 mammut kernel: [ 1209.379490]  [<ffffffffc03271d0>] amdgpu_vm_bo_update+0x140/0x2f0 [amdgpu]
Nov 29 16:39:51 mammut kernel: [ 1209.379509]  [<ffffffffc0314c64>] amdgpu_gem_va_update_vm+0x194/0x1b0 [amdgpu]
Nov 29 16:39:51 mammut kernel: [ 1209.379513]  [<ffffffffbe226d01>] ? interval_tree_iter_next+0x51/0x70
Nov 29 16:39:51 mammut kernel: [ 1209.379531]  [<ffffffffc0315d5f>] amdgpu_gem_va_ioctl+0x21f/0x2e0 [amdgpu]
Nov 29 16:39:51 mammut kernel: [ 1209.379547]  [<ffffffffc0205f8b>] drm_ioctl+0x21b/0x4c0 [drm]
Nov 29 16:39:51 mammut kernel: [ 1209.379565]  [<ffffffffc0315b40>] ? amdgpu_gem_metadata_ioctl+0x1d0/0x1d0 [amdgpu]
Nov 29 16:39:51 mammut kernel: [ 1209.379580]  [<ffffffffc02fb04f>] amdgpu_drm_ioctl+0x4f/0x90 [amdgpu]
Nov 29 16:39:51 mammut kernel: [ 1209.379583]  [<ffffffffbe040c03>] do_vfs_ioctl+0xa3/0x600
Nov 29 16:39:51 mammut kernel: [ 1209.379586]  [<ffffffffbde6a516>] ? __do_page_fault+0x266/0x4e0
Nov 29 16:39:51 mammut kernel: [ 1209.379589]  [<ffffffffbe0411d9>] SyS_ioctl+0x79/0x90
Nov 29 16:39:51 mammut kernel: [ 1209.379592]  [<ffffffffbe660b3b>] entry_SYSCALL_64_fastpath+0x1e/0xad
Nov 29 16:41:16 mammut kernel: [ 1294.482836] sysrq: SysRq : Emergency Sync
Nov 29 16:41:16 mammut kernel: [ 1294.484975] Emergency Sync complete
Nov 29 16:41:17 mammut kernel: [ 1296.178828] sysrq: SysRq : Emergency Remount R/O

Comment 1 Ernst Sjöstrand 2016-12-14 10:58:03 UTC

Can't reproduce anymore.

Comment 2 Ernst Sjöstrand 2017-01-05 10:03:00 UTC

This was probably an old libxcb in the steam runtime, like: https://github.com/ValveSoftware/steam-for-linux/issues/4638

FYI.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.