Created attachment 144626 [details] GPU crash dump dmesg output: [ 4337.121156] i915 0000:00:02.0: GPU HANG: ecode 9:0:0x00000000, hang on rcs0 [....] [ 4337.121168] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0 [ 4337.122893] [drm:gen8_reset_engines [i915]] *ERROR* rcs0: reset request timeout [ 4337.122918] i915 0000:00:02.0: Resetting chip for hang on rcs0 [ 4337.124650] [drm:gen8_reset_engines [i915]] *ERROR* rcs0: reset request timeout [ 4337.126370] [drm:gen8_reset_engines [i915]] *ERROR* rcs0: reset request timeout Kernel version: 5.1.14 OS: Arch Linux Mobo: ASRock Z170M Extreme4
rcs: START: 0x0000d000 HEAD: 0x00e01470 [0x00000000] head = 0x00001470, wraps = 7 TAIL: 0x00001470 [0x00000000, 0x00000000] CTL: 0x00003001 len=16384, enabled MODE: 0x00000000 HWS: 0xffffe000 ACTHD: 0x00000000 00e01470 at ring: 0x00000000 IPEIR: 0x00000000 IPEHR: 0x02800000 ELSP[0]: pid 688, ban score 0, seqno 3:0002b24b!, prio 2, emitted -1119ms, start 0000d000, head 000013d0, tail 00001470 ELSP[1]: pid 611, ban score 0, seqno 2:0002b24c, prio 0, emitted -1114ms, start 00009000, head 00003b60, tail 00003bf8 The CS reached the end of the ring, but the GPU failed to switch contexts. Then the GPU failed to respond to reset requests, further indicating it had suffered a terminal shock. Immediate suspicion would be on powergating, for the similarity with say bug 110450.
Or it could just be a userspace fault that leaves the GPU barely functional up to the point it fails completely.
A new one: [40476.867483] i915 0000:00:02.0: GPU HANG: ecode 9:1:0xfffffffe, in ffmpeg [30494], hang on rcs0, vcs0 [40476.867484] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. [40476.867484] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel [40476.867485] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. [40476.867485] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. [40476.867485] [drm] GPU crash dump saved to /sys/class/drm/card0/error [40476.868493] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0, vcs0 [40476.868560] i915 0000:00:02.0: Resetting vcs0 for hang on rcs0, vcs0 ffmpeg command: ffmpeg -vaapi_device /dev/dri/renderD128 -i 'http://****' -c:v hevc_vaapi -b:v 840K -vf 'format=nv12,hwupload' -c:a copy -y <output_file.mkv> vaapi driver: iHD_drv_video.so System: Machine: Device: desktop Mobo: ASRock model: Z170M Extreme4 serial: M80-68001700529 UEFI [Legacy]: American Megatrends v: P7.20 date: 12/13/2016 CPU: Dual core Intel Core i3-6100 (-MT-MCP-) cache: 3072 KB clock speeds: max: 3700 MHz 1: 800 MHz 2: 3223 MHz 3: 2982 MHz 4: 3049 MHz Graphics: Card: Intel HD Graphics 530 Display Server: X.Org 1.20.5 drivers: intel (unloaded: modesetting) Resolution: 1920x1080@60.00hz OpenGL: renderer: Mesa DRI Intel HD Graphics 530 (Skylake GT2) version: 4.5 Mesa 19.1.2
Created attachment 144812 [details] i915 crash dump crash dump attached.
Ah, libva, userspace hang leaving the GPU barely functional is no longer a surprise.
Closing this issue as NOTOURBUG. Rik, can you please create an issue for libva? ttps://github.com/intel/libva/issues
(In reply to Lakshmi from comment #6) > Closing this issue as NOTOURBUG. > Rik, can you please create an issue for libva? > ttps://github.com/intel/libva/issues No, but thanks for your support.
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.