Created attachment 127567 [details] kern.log Bug description: =========================================== gem_busy sub-test extended-parallel-render causes GPU hang this a test regression, please see the following info WW43 setup (pass) ============= kernel commit 15dfed2b90e84e7c277f81842fc3f19355293061 igt commit 54f8a3f drm commit a44c9c3 cairo commit db8a7f1 WW44 setup (fail) ========== kernel commit 17dc529acb9a6a4328b419048e32df586b90646b igt commit 93437cb drm commit 9e24d0c cairo commit db8a7f1 test output ================================================= IGT-Version: 1.16-g93437cb (x86_64) (Linux: 4.9.0-rc1-nightly+ x86_64) (gem_busy:2214) igt-core-DEBUG: Test requirement passed: !igt_run_in_simulation() (gem_busy:2214) drmtest-DEBUG: Test requirement passed: !(fd<0) (gem_busy:2214) drmtest-DEBUG: Test requirement passed: drmSetMaster(fd) == 0 (gem_busy:2214) igt-aux-CRITICAL: Test assertion failure function sig_abort, file igt_aux.c:401: (gem_busy:2214) igt-aux-CRITICAL: Failed assertion: !"GPU hung" Stack trace: #0 [__igt_fail_assert+0x101] #1 [sig_abort+0x3a] #2 [killpg+0x40] #3 [has_extended_busy_ioctl+0x166] #4 [__real_main596+0x2b3] #5 [main+0x23] #6 [__libc_start_main+0xf0] #7 [_start+0x29] #8 [<unknown>+0x29] Test gem_busy failed. **** DEBUG **** (gem_busy:2214) igt-core-DEBUG: Test requirement passed: !igt_run_in_simulation() (gem_busy:2214) drmtest-DEBUG: Test requirement passed: !(fd<0) (gem_busy:2214) drmtest-DEBUG: Test requirement passed: drmSetMaster(fd) == 0 (gem_busy:2214) igt-aux-CRITICAL: Test assertion failure function sig_abort, file igt_aux.c:401: (gem_busy:2214) igt-aux-CRITICAL: Failed assertion: !"GPU hung" **** END **** Subtest extended-parallel-render: FAIL (gem_busy:2214) igt-aux-CRITICAL: Test assertion failure function sig_abort, file igt_aux.c:401: (gem_busy:2214) igt-aux-CRITICAL: Failed assertion: !"GPU hung" Stack trace: #0 [__igt_fail_assert+0x101] #1 [sig_abort+0x3a] #2 [killpg+0x40] #3 [has_extended_busy_ioctl+0x166] #4 [__real_main596+0xf53] #5 [main+0x23] #6 [__libc_start_main+0xf0] #7 [_start+0x29] #8 [<unknown>+0x29] Test gem_busy failed. **** DEBUG **** (gem_busy:2214) igt-aux-CRITICAL: Test assertion failure function sig_abort, file igt_aux.c:401: (gem_busy:2214) igt-aux-CRITICAL: Failed assertion: !"GPU hung" **** END **** (gem_busy:2214) ioctl-wrappers-DEBUG: Test requirement passed: has_ban_period (gem_busy:2214) igt-gt-DEBUG: Test requirement passed: has_gpu_reset(fd) (gem_busy:2214) ioctl-wrappers-CRITICAL: Test assertion failure function gem_execbuf, file ioctl_wrappers.c:604: (gem_busy:2214) ioctl-wrappers-CRITICAL: Failed assertion: __gem_execbuf(fd, execbuf) == 0 (gem_busy:2214) ioctl-wrappers-CRITICAL: error: -5 != 0 Stack trace: #0 [__igt_fail_assert+0x101] #1 [gem_execbuf+0x44] #2 [has_extended_busy_ioctl+0x151] #3 [__real_main596+0xd04] #4 [main+0x23] #5 [__libc_start_main+0xf0] #6 [_start+0x29] #7 [<unknown>+0x29] Test gem_busy failed. **** DEBUG **** (gem_busy:2214) ioctl-wrappers-DEBUG: Test requirement passed: has_ban_period (gem_busy:2214) igt-gt-DEBUG: Test requirement passed: has_gpu_reset(fd) (gem_busy:2214) ioctl-wrappers-CRITICAL: Test assertion failure function gem_execbuf, file ioctl_wrappers.c:604: (gem_busy:2214) ioctl-wrappers-CRITICAL: Failed assertion: __gem_execbuf(fd, execbuf) == 0 (gem_busy:2214) ioctl-wrappers-CRITICAL: error: -5 != 0 **** END **** (gem_busy:2214) igt-core-DEBUG: Exiting with status code 99 relevant dmesg output ================================================= Oct 27 13:06:14 SNB-1 kernel: [ 166.813633] [drm] GPU HANG: ecode 6:0:0xe77ffeff, in gem_busy [1272], reason: Hang on render ring, action: reset Oct 27 13:06:14 SNB-1 kernel: [ 166.813634] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. Oct 27 13:06:14 SNB-1 kernel: [ 166.813635] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel Oct 27 13:06:14 SNB-1 kernel: [ 166.813636] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. Oct 27 13:06:14 SNB-1 kernel: [ 166.813636] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. Oct 27 13:06:14 SNB-1 kernel: [ 166.813637] [drm] GPU crash dump saved to /sys/class/drm/card0/error Software information ============================================ Kernel version : 4.9.0-rc1-nightly+ Linux distribution : Ubuntu 16.04.1 LTS Architecture : 64-bit Gfx stack code : 4283756084 xf86-video-intel version : 2.99.917 Xorg-Xserver version : 1.18.99.901 (1.19.0 RC 1) DRM version : 2.4.71 VAAPI version : Intel i965 driver for Intel(R) Sandybridge Desktop - 1.7.3.pre1 (1.7.2-140-g852cea1) Cairo version : 1.15.2 Intel GPU Tools version : Tag [intel-gpu-tools-1.16-96-g93437cb] / Commit [93437cb] Kernel driver in use : i915 Bios revision : 4.6 doesn't has this firmware Hardware information ============================================ Platform : SNB Motherboard model : OptiPlex990 Motherboard type : 06D7TR Mini Tower Motherboard manufacturer : DellInc. CPU family : Core i7 CPU information : Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz GPU Card : Intel Corporation 2nd Generation Core Processor Family Integrated Graphics Controller (rev 09) (prog-if 00 [VGA controller]) Memory ram : 16 GB Maximum memory ram allowed : 32 GB CPU thread : 8 CPU core : 4 Signature : Type 0, Family 6, Model 42, Stepping 7 Hard drive capacity : 111GiB (120GB) Kernel parameters ============================================ quiet drm.debug=0xe resume=/dev/sda3 Attachments ================================================== kern.log gpu_hang.log gpu_crash.log
Created attachment 127568 [details] gpu_hang
Created attachment 127569 [details] gpu_crash
Can you please make sure your kernel is uptodate and allows /dev/kmsg access?
(In reply to Chris Wilson from comment #3) > Can you please make sure your kernel is uptodate and allows /dev/kmsg access? I tried with the latest kernel and test is still failing. Added kmsg
Created attachment 127572 [details] kmsg
14,1280,255241527,-;[IGT] gem_busy: executing 7,1281,255241602,-;[drm:i915_gem_open [i915]] 7,1282,255241959,-;[drm:i915_gem_open [i915]] 6,1283,268864341,-;[drm] GPU HANG: ecode 6:0:0xe77ffeff, in gem_busy [1280], reason: Hang on render ring, action: reset 6,1284,268864342,-;[drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. 6,1285,268864342,-;[drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel 6,1286,268864342,-;[drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. 6,1287,268864343,-;[drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. 6,1288,268864343,-;[drm] GPU crash dump saved to /sys/class/drm/card0/error 7,1289,268864371,-;[drm:i915_reset_and_wakeup [i915]] resetting chip 5,1290,268864378,-;drm/i915: Resetting chip after gpu hang 7,1291,268864397,-;[drm:i915_gem_reset [i915]] resetting render ring to restart from tail of request 0x22 7,1292,268864419,-;[drm:intel_print_rc6_info [i915]] Enabling RC6 states: RC6 on RC6p off RC6pp off 7,1293,268868451,-;[drm:intel_guc_setup [i915]] GuC fw status: path (null), fetch NONE, load NONE 7,1294,281854085,-;[drm:i915_reset_and_wakeup [i915]] resetting chip 5,1295,281854091,-;drm/i915: Resetting chip after gpu hang 7,1296,281854111,-;[drm:i915_gem_reset [i915]] resetting render ring to restart from tail of request 0x23 14,1304,297808259,-;[IGT] gem_busy: exiting, ret=99 Either we are not getting all the [IGT] breadcrumbs or that dies before we even start the test.
Highest due to regression without workaround
Lupita could you re-test with latest configuration
(In reply to ricardo.vega from comment #8) > Lupita could you re-test with latest configuration This issue does not occur using the next configuration: Software information ============================================ Kernel version : 4.9.0-rc7drm-intel-qa-ww49-commit-477cc06+ Linux distribution : Ubuntu 16.04.1 LTS Architecture : 64-bit Gfx stack code : 4152731784 xf86-video-intel version : 2.99.917 Xorg-Xserver version : 1.18.3 DRM version : 2.4.73 Cairo version : 1.15.2 Intel GPU Tools version : Tag [intel-gpu-tools-1.16-166-gd072258] / Commit [d072258] Kernel driver in use : i915 Bios revision : 4.6 Kernel parameters ============================================ quiet drm.debug=0xe resume=/dev/sd result ============================================ IGT-Version: 1.16-gd072258 (x86_64) (Linux: 4.9.0-rc7drm-intel-qa-ww49-commit-477cc06+ x86_64) Subtest extended-parallel-render: SUCCESS (0.040s)
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.