Bug 104484 - [drm] GPU HANG: ecode 9:0:0x86dfdff9, in csgo_linux64 [14845], reason: Hang on rcs0, action: reset
Summary: [drm] GPU HANG: ecode 9:0:0x86dfdff9, in csgo_linux64 [14845], reason: Hang o...
Status: RESOLVED INVALID
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i965 (show other bugs)
Version: git
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Intel 3D Bugs Mailing List
QA Contact: Intel 3D Bugs Mailing List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2018-01-04 02:57 UTC by Robert
Modified: 2018-01-04 09:23 UTC (History)
0 users

See Also:
i915 platform:
i915 features:

Attachments
CSGO Crash Dump (112.94 KB, application/zip)
2018-01-04 02:57 UTC, Robert 		Details
System Info (1.88 KB, text/plain)
2018-01-04 02:57 UTC, Robert 		Details
Console stdout (11.16 MB, text/plain)
2018-01-04 02:58 UTC, Robert 		Details
View All

Description Robert 2018-01-04 02:57:08 UTC 
Created attachment 136539 [details]
CSGO Crash Dump

CSGO Crash while playing.  The crash does not appear to be happening very often at all.  I will look out for future occurances.

Note, I just recently got the GuC firmware to start loading due to a bug in Ubuntu that is being tracked here:
https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/1728238

My /etc/default/grub is using the following to load the GuC:
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash i915.enable_guc_loading=1 i915.enable_guc_submission=1 i915.enable_rc6=1 i915.enable_fbc=1"

dmesg output at CSGO crash:

[44374.990060] [drm] GPU HANG: ecode 9:0:0x86dfdff9, in csgo_linux64 [14845], reason: Hang on rcs0, action: reset
[44374.990104] drm/i915: Resetting chip after gpu hang
[44374.990510] [drm] RC6 on
[44374.993570] [drm] GuC submission enabled (firmware i915/kbl_guc_ver9_14.bin [version 9.14])
[44382.725582] drm/i915: Resetting chip after gpu hang
[44382.725703] [drm] RC6 on
[44382.729514] [drm] GuC submission enabled (firmware i915/kbl_guc_ver9_14.bin [version 9.14])
[44390.725632] drm/i915: Resetting chip after gpu hang
[44390.725767] [drm] RC6 on
[44390.730716] [drm] GuC submission enabled (firmware i915/kbl_guc_ver9_14.bin [version 9.14])
[44398.725661] drm/i915: Resetting chip after gpu hang
[44398.725835] [drm] RC6 on
[44398.728494] [drm] GuC submission enabled (firmware i915/kbl_guc_ver9_14.bin [version 9.14])
[44400.649134] [UFW BLOCK] IN=wlp58s0 OUT= MAC=01:00:5e:00:00:01:60:38:e0:7e:aa:ff:08:00 SRC=192.168.1.1 DST=224.0.0.1 LEN=28 TOS=0x00 PREC=0xC0 TTL=1 ID=15931 PROTO=2 
[44406.725704] drm/i915: Resetting chip after gpu hang
[44406.725844] [drm] RC6 on
[44406.735753] [drm] GuC submission enabled (firmware i915/kbl_guc_ver9_14.bin [version 9.14])
[44407.086988] show_signal_msg: 79 callbacks suppressed
[44407.086990] csgo_linux64[14890]: segfault at 0 ip 00007f39fba5cb3e sp 00007f39a22ee000 error 6 in libtier0_client.so[7f39fba4a000+29000]

Mesa:
Version: 1:17.4~git171229194900.ab95724~a~padoka0
Comment 1 Robert 2018-01-04 02:57:57 UTC 
Created attachment 136540 [details]
System Info
Comment 2 Robert 2018-01-04 02:58:49 UTC 
Created attachment 136541 [details]
Console stdout
Comment 4 Chris Wilson 2018-01-04 09:23:49 UTC 
There is nothing mesa can do about potential bugs or performance regressions in the guc fw. Please reproduce without that loaded and reopen.

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.