Bug 99788 - [SKL][BAT] kms_pipe_crc_basic/read-crc-pipe-b-frame-sequence fail on CI
Summary: [SKL][BAT] kms_pipe_crc_basic/read-crc-pipe-b-frame-sequence fail on CI
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: DRI git
Hardware: Other All
: highest critical
Assignee: Mika Kahola
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard: ReadyForDev
Keywords:
Depends on:
Blocks:
 
Reported: 2017-02-13 08:40 UTC by Jani Saarinen
Modified: 2017-06-16 18:12 UTC (History)
3 users (show)

See Also:
i915 platform: SKL
i915 features: display/Other


Attachments
dmesg during run (6.69 MB, text/plain)
2017-02-13 08:40 UTC, Jani Saarinen
no flags Details
Reset GPU before running the tests (735 bytes, patch)
2017-04-28 13:07 UTC, Mika Kahola
no flags Details | Splinter Review

Description Jani Saarinen 2017-02-13 08:40:47 UTC
Created attachment 129549 [details]
dmesg during run

On Patchwork run 
kms_pipe_crc_basic/read-crc-pipe-b-frame-sequence fails on CI
pass       -> FAIL       (fi-skl-6770hq)

Result:
IGT-Version: 1.17-g3e65f19 (x86_64) (Linux: 4.10.0-rc7-CI-Patchwork_3776+ x86_64)
read-crc-pipe-B-frame-sequence: Testing connector DP-2 using pipe B
read-crc-pipe-B-frame-sequence: Testing connector DP-4 using pipe B
Stack trace:
  #0 [__igt_fail_assert+0x101]
  #1 [test_read_crc.constprop.3+0x48c]
  #2 [<unknown>+0x48c]
  #3 [<unknown>+0x48c]
Subtest read-crc-pipe-B-frame-sequence: FAIL (1.155s)
Err	
(kms_pipe_crc_basic:8766) CRITICAL: Test assertion failure function test_read_crc_for_output, file kms_pipe_crc_basic.c:168:
(kms_pipe_crc_basic:8766) CRITICAL: Failed assertion: crcs[j].frame + 1 == crcs[j + 1].frame
(kms_pipe_crc_basic:8766) CRITICAL: error: 20240 != 20241
Subtest read-crc-pipe-B-frame-sequence failed.
**** DEBUG ****
(kms_pipe_crc_basic:8766) DEBUG: Test requirement passed: !(pipe >= data->display.n_pipes)
(kms_pipe_crc_basic:8766) INFO: read-crc-pipe-B-frame-sequence: Testing connector DP-2 using pipe B
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-2: set_pipe(B)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-2: Selecting pipe B
(kms_pipe_crc_basic:8766) DEBUG: Clearing the fb with color (0.00,1.00,0.00)
(kms_pipe_crc_basic:8766) igt-fb-DEBUG: igt_create_fb_with_bo_size(width=1920, height=1200, format=0x34325258, tiling=0x0, size=0)
(kms_pipe_crc_basic:8766) igt-fb-DEBUG: igt_create_fb_with_bo_size(handle=1, pitch=7680)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: B.0: plane_set_fb(68)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: commit {
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     DP-2: Selecting pipe B
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     SetCrtc pipe A, disabling
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     SetPlane pipe A, plane 1, disabling
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     SetCursor pipe A, disabling
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     DP-2: SetCrtc pipe B, fb 68, src (0, 0), mode 1920x1200
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     SetPlane pipe B, plane 1, disabling
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     SetCursor pipe B, disabling
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     SetCrtc pipe C, disabling
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     SetPlane pipe C, plane 1, disabling
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     SetCursor pipe C, disabling
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: }
(kms_pipe_crc_basic:8766) igt-debugfs-DEBUG: Using generic frame CRC ABI
(kms_pipe_crc_basic:8766) DEBUG: CRC for this fb: 92b15e48 00000000 00000000 00000000 00000000
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: B.0: plane_set_fb(0)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-2: set_pipe(any)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-2: set_pipe(B)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-2: Selecting pipe B
(kms_pipe_crc_basic:8766) DEBUG: Clearing the fb with color (0.00,1.00,1.00)
(kms_pipe_crc_basic:8766) igt-fb-DEBUG: igt_create_fb_with_bo_size(width=1920, height=1200, format=0x34325258, tiling=0x0, size=0)
(kms_pipe_crc_basic:8766) igt-fb-DEBUG: igt_create_fb_with_bo_size(handle=1, pitch=7680)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: B.0: plane_set_fb(68)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: commit {
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     DP-2: Selecting pipe B
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     DP-2: SetCrtc pipe B, fb 68, src (0, 0), mode 1920x1200
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: }
(kms_pipe_crc_basic:8766) igt-debugfs-DEBUG: Using generic frame CRC ABI
(kms_pipe_crc_basic:8766) DEBUG: CRC for this fb: 52112cfc 00000000 00000000 00000000 00000000
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: B.0: plane_set_fb(0)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-2: set_pipe(any)
(kms_pipe_crc_basic:8766) INFO: read-crc-pipe-B-frame-sequence: Testing connector DP-4 using pipe B
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-4: set_pipe(B)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-4: Selecting pipe B
(kms_pipe_crc_basic:8766) DEBUG: Clearing the fb with color (0.00,1.00,0.00)
(kms_pipe_crc_basic:8766) igt-fb-DEBUG: igt_create_fb_with_bo_size(width=3840, height=2160, format=0x34325258, tiling=0x0, size=0)
(kms_pipe_crc_basic:8766) igt-fb-DEBUG: igt_create_fb_with_bo_size(handle=1, pitch=15360)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: B.0: plane_set_fb(68)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: commit {
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     DP-4: Selecting pipe B
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     DP-4: SetCrtc pipe B, fb 68, src (0, 0), mode 3840x2160
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: }
(kms_pipe_crc_basic:8766) igt-debugfs-DEBUG: Using generic frame CRC ABI
(kms_pipe_crc_basic:8766) DEBUG: CRC for this fb: 008b243c 00000000 00000000 00000000 00000000
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: B.0: plane_set_fb(0)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-4: set_pipe(any)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-4: set_pipe(B)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: DP-4: Selecting pipe B
(kms_pipe_crc_basic:8766) DEBUG: Clearing the fb with color (0.00,1.00,1.00)
(kms_pipe_crc_basic:8766) igt-fb-DEBUG: igt_create_fb_with_bo_size(width=3840, height=2160, format=0x34325258, tiling=0x0, size=0)
(kms_pipe_crc_basic:8766) igt-fb-DEBUG: igt_create_fb_with_bo_size(handle=1, pitch=15360)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: B.0: plane_set_fb(68)
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: commit {
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     DP-4: Selecting pipe B
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display:     DP-4: SetCrtc pipe B, fb 68, src (0, 0), mode 3840x2160
(kms_pipe_crc_basic:8766) igt-kms-DEBUG: display: }
(kms_pipe_crc_basic:8766) igt-debugfs-DEBUG: Using generic frame CRC ABI
(kms_pipe_crc_basic:8766) DEBUG: CRC for this fb: 8b17108a 00000000 00000000 00000000 00000000
(kms_pipe_crc_basic:8766) CRITICAL: Test assertion failure function test_read_crc_for_output, file kms_pipe_crc_basic.c:168:
(kms_pipe_crc_basic:8766) CRITICAL: Failed assertion: crcs[j].frame + 1 == crcs[j + 1].frame
(kms_pipe_crc_basic:8766) CRITICAL: error: 20240 != 20241
****  END  ****
Environment	
PIGLIT_PLATFORM="mixed_glx_egl" PIGLIT_SOURCE_DIR="/opt/igt/piglit"
Command	
/opt/igt/tests/kms_pipe_crc_basic --run-subtest read-crc-pipe-B-frame-sequence
Comment 2 Dorota Czaplejewicz 2017-03-10 16:27:41 UTC
160 runs passed without error.

Kernel: drm-tip: 2017y-03m-10d-13h-03m-29s with CI_DRM_2320
CPU: Celeron(R) CPU G3900 @ 2.80GHz
GPU: 00:02.0 0300: 8086:1902 (rev 06) (prog-if 00 [VGA controller])
System: LENOVO ideacentre 510S-08ISH
BIOS Revision: 0.6
OS: Fedora 25
Comment 4 Martin Peres 2017-03-23 10:03:52 UTC
igt@kms_pipe_crc_basic@read-crc-pipe-b-frame-sequence: Failure rate 2/107 run(s) (1%)

igt@kms_pipe_crc_basic@nonblocking-crc-pipe-c-frame-sequence: Failure rate 1/12 run(s) (8%)

Raising the priority because it is seen in CI
Comment 6 Jani Saarinen 2017-04-21 15:38:01 UTC
For:
kms_pipe_crc_basic@read-crc-pipe-b-frame-sequence
Last seen: 2017-03-21
Statistics: Failure rate 2/265 run(s) (0%)

kms_pipe_crc_basic@nonblocking-crc-pipe-c-frame-sequence
Last seen: 2017-03-21
Statistics: Failure rate 1/170 run(s) (0%)
Comment 7 Jani Saarinen 2017-04-26 09:44:37 UTC
CLosing,m as not seen lately, whitelisting for CI.
Comment 8 Martin Peres 2017-04-26 14:04:38 UTC
(In reply to Jani Saarinen from comment #7)
> CLosing,m as not seen lately, whitelisting for CI.

What about https://intel-gfx-ci.01.org/CI/Patchwork_4496/fi-skl-6770hq/igt@kms_pipe_crc_basic@read-crc-pipe-a-frame-sequence.html ?
Comment 9 Jani Saarinen 2017-04-28 08:25:35 UTC
Seen again on pw: https://patchwork.freedesktop.org/series/22937/ 
Marked for CI
Comment 10 Mika Kahola 2017-04-28 13:07:40 UTC
Created attachment 131119 [details] [review]
Reset GPU before running the tests

I ran a test on this problematic fi-skl-6770hq machine for 200 times without any issues. So, maybe we have some garbage left from previous tests that we should clean up. The patch proposes a GPU reset before running the tests. Of course, this slows down the test execution time but I think this would be a worthwhile to give it a go and run full BAT testing for few hundred times with this patch enabled.
Comment 11 Jani Saarinen 2017-05-08 13:53:34 UTC
Mika, any updates on this. was run taken?

For igt@kms_pipe_crc_basic@read-crc-pipe-b-frame-sequence	
Last seen still: 2017-03-21
Statistics: Failure rate 2/308 run(s) (0%)

igt@kms_pipe_crc_basic@nonblocking-crc-pipe-c-frame-sequence
Last seen still: 2017-03-21
Statistics: Failure rate 1/213 run(s) (0%)
Comment 12 Ricardo 2017-05-09 17:17:24 UTC
Adding tag into "Whiteboard" field - ReadyForDev
The bug still active
*Status is correct
*Platform is included
*Feature is included
*Priority and Severity correctly set
*Logs included
Comment 13 Jani Saarinen 2017-06-06 09:36:25 UTC
Mika, any updates on this?
Comment 14 Mika Kahola 2017-06-06 10:11:39 UTC
I'm running tests on CI farm2 with the patch applied.
Comment 15 Jani Saarinen 2017-06-07 10:33:51 UTC
These are really hard to reproduce. Might be that we just need to wait few times still and close if not reproduced.
Comment 16 Mika Kahola 2017-06-14 11:46:19 UTC
Unable to replicate the issue and the issue hasn't surfaced on CI runs either.


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.