Bug 81209 - [BDW/BYT Bisected semaphores] igt/gem_gtt_hog fails, gem_reset_stats and more
Summary: [BDW/BYT Bisected semaphores] igt/gem_gtt_hog fails, gem_reset_stats and more
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: Other All
: highest major
Assignee: Elio
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
: 81317 81831 (view as bug list)
Depends on:
Blocks:
 
Reported: 2014-07-11 06:43 UTC by Guo Jinxian
Modified: 2017-07-24 22:53 UTC (History)
2 users (show)

See Also:
i915 platform: BYT
i915 features:


Attachments
dmesg (92.02 KB, text/plain)
2014-07-11 06:55 UTC, Guo Jinxian
no flags Details
i915 error state (273.88 KB, text/plain)
2014-07-22 01:14 UTC, Guo Jinxian
no flags Details
error state with "i915.semaphore=0" (276.15 KB, text/plain)
2014-07-23 00:55 UTC, Guo Jinxian
no flags Details
error state with "i915.semaphore=1" (277.18 KB, text/plain)
2014-07-23 00:56 UTC, Guo Jinxian
no flags Details
dmesg (103.77 KB, text/plain)
2014-07-23 02:26 UTC, Guo Jinxian
no flags Details
error state with "i915.semaphores=0" (25 bytes, text/plain)
2014-07-23 02:27 UTC, Guo Jinxian
no flags Details
dmesg (83.80 KB, text/plain)
2014-07-28 01:25 UTC, Guo Jinxian
no flags Details

Description Guo Jinxian 2014-07-11 06:43:11 UTC
==System Environment==
--------------------------
Regression: Yes. 
Good commit on -next-queued: 57127efaf1adc5e0bb2d3fa5a01bd97f71af7836

Non-working platforms: BDW

==kernel==
--------------------------
origin/drm-intel-nightly: 140ef1831108dabeeb68bcaa046c2aee44990eb3(fails)
    drm-intel-nightly: 2014y-07m-10d-22h-32m-44s integration manifest
origin/drm-intel-next-queued: 7b3c29f6fd374640266d82aafe6929761469e27b(fails)
    drm/i915: Make the RPS interrupt generation mask handle the vlv wa
origin/drm-intel-fixes: 92ae62076957c5904509f755eea0075ad60f74c6(works)
    Revert "drm/i915: Don't set the 8to6 dither flag when not scaling"

==Bug detailed description==
igt/gem_gtt_hog fails

Output:
"[root@x-bdw01 tests]# time ./gem_gtt_hog
IGT-Version: 1.7-g6b986a2 (x86_64) (Linux: 3.16.0-rc4_drm-intel-nightly_140ef1_20140711+ x86_64)
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:153:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function run, file gem_gtt_hog.c:152:
Last errno: 0, Success
Failed assertion: x == canary
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:158:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function run, file gem_gtt_hog.c:152:
Last errno: 0, Success
Failed assertion: x == canary
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function __real_main156, file gem_gtt_hog.c:187:
Last errno: 0, Success
Failed assertion: status == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function run, file gem_gtt_hog.c:152:
Last errno: 0, Success
Failed assertion: x == canary
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0

real    7m24.025s
user    0m0.003s
sys     0m0.022s
[root@x-bdw01 tests]# Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0

==Reproduce steps==
---------------------------- 
1. ./gem_gtt_hog
Comment 1 Guo Jinxian 2014-07-11 06:55:25 UTC
Created attachment 102597 [details]
dmesg
Comment 2 Chris Wilson 2014-07-11 07:19:00 UTC
Please attach the error state as that may provide the vital clue, and please bisect.
Comment 3 Guo Jinxian 2014-07-16 05:16:48 UTC
This bug still able to reproduce on latest -nightly() while running test igt/kms_flip/flip-vs-panning-vs-hang

./kms_flip --run-subtest flip-vs-panning-vs-hang
IGT-Version: 1.7-g8d60b82 (x86_64) (Linux: 3.16.0-rc5_drm-intel-nightly_778206_20140716+ x86_64)
Using monotonic timestamps
Beginning flip-vs-panning-vs-hang on crtc 7, connector 18
  1920x1080 60 1920 1966 1996 2080 1080 1082 1086 1112 0xa 0x48 138780
..
flip-vs-panning-vs-hang on crtc 7, connector 18: PASSED

Beginning flip-vs-panning-vs-hang on crtc 11, connector 18
  1920x1080 60 1920 1966 1996 2080 1080 1082 1086 1112 0xa 0x48 138780
...
flip-vs-panning-vs-hang on crtc 11, connector 18: PASSED

Beginning flip-vs-panning-vs-hang on crtc 15, connector 18
  1920x1080 60 1920 1966 1996 2080 1080 1082 1086 1112 0xa 0x48 138780
...
flip-vs-panning-vs-hang on crtc 15, connector 18: PASSED

Subtest flip-vs-panning-vs-hang: SUCCESS
Test assertion failure function gem_quiescent_gpu, file drmtest.c:149:
Last errno: 5, Input/output error
Failed assertion: drmIoctl((fd), ((((1U) << (((0+8)+8)+14)) | ((('d')) << (0+8)) | (((0x40 + 0x29)) << 0) | ((((sizeof(struct drm_i915_gem_execbuffer2)))) << ((0+8)+8)))), (&execbuf)) == 0
kms_flip: igt_core.c:651: igt_fail: Assertion `!test_with_subtests || in_fixture' failed.
Aborted (core dumped)
Comment 4 Guo Jinxian 2014-07-16 08:37:45 UTC
(In reply to comment #2)
> Please attach the error state as that may provide the vital clue, and please
> bisect.

521e62e49a42661a4ee0102644517dbe2f100a23 is the first bad commit

commit 521e62e49a42661a4ee0102644517dbe2f100a23
Author:     Rodrigo Vivi <rodrigo.vivi@intel.com>
AuthorDate: Mon Jun 30 09:53:44 2014 -0700
Commit:     Daniel Vetter <daniel.vetter@ffwll.ch>
CommitDate: Mon Jul 7 23:16:56 2014 +0200

    drm/i915: Enable semaphores on BDW

    Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
    Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

The error state file is too big to bugzilla, I will send it via email.
Comment 5 Chris Wilson 2014-07-17 07:02:25 UTC
Error-state says: IPEHR 0x54300005 on render ring. Those semaphores are not serialising properly. Fix or revert.
Comment 6 Rodrigo Vivi 2014-07-18 18:30:52 UTC
Hi Guo, can you please send me the error state?
Comment 7 Guo Jinxian 2014-07-21 01:15:59 UTC
(In reply to comment #6)
> Hi Guo, can you please send me the error state?

I forwarded the error state to you by email. Please check you email. Thanks.
Comment 8 Chris Wilson 2014-07-21 07:00:07 UTC
*** Bug 81317 has been marked as a duplicate of this bug. ***
Comment 9 Ben Widawsky 2014-07-21 22:34:44 UTC
There are some things in the error state that look slightly suspicious (it kind of looks like mesa is also running).

Since we cannot reproduce it locally, can you please generate another error state? The compressed error state should fit as an attachment.

Thanks.
Comment 10 Guo Jinxian 2014-07-22 01:14:54 UTC
Created attachment 103244 [details]
i915 error state

(In reply to comment #9)
> There are some things in the error state that look slightly suspicious (it
> kind of looks like mesa is also running).
> 
> Since we cannot reproduce it locally, can you please generate another error
> state? The compressed error state should fit as an attachment.
> 
> Thanks.

Update i915 error state.
Comment 11 Guo Jinxian 2014-07-23 00:55:55 UTC
Created attachment 103308 [details]
error state with "i915.semaphore=0"
Comment 12 Guo Jinxian 2014-07-23 00:56:26 UTC
Created attachment 103309 [details]
error state with "i915.semaphore=1"
Comment 13 Ben Widawsky 2014-07-23 01:25:16 UTC
Something is not right with the semaphore = 0 error state. Semaphores signals and waits are still being emitted. Can you please confirm you correct set the module parameter for that test by reproducing, and providing the dmesg and new error state?
Comment 14 Guo Jinxian 2014-07-23 02:26:43 UTC
Created attachment 103311 [details]
dmesg

(In reply to comment #13)
> Something is not right with the semaphore = 0 error state. Semaphores
> signals and waits are still being emitted. Can you please confirm you
> correct set the module parameter for that test by reproducing, and providing
> the dmesg and new error state?

I misspelled "i915.semaphores=0" to "i915.semaphore=0" last time. With correct parameter "i915.semaphores=0", the test was passed on latest -nightly(411fa8b275ee903df6c07976af4eebe5815646a4)

[root@x-bdw01 tests]# time ./gem_gtt_hog
IGT-Version: 1.7-g70e6ed9 (x86_64) (Linux: 3.16.0-rc6_drm-intel-nightly_411fa8_20140722+ x86_64)
Time to execute 64 children:            93781.680ms

real    1m33.837s
user    0m0.018s
sys     0m0.804s
Comment 15 Guo Jinxian 2014-07-23 02:27:43 UTC
Created attachment 103312 [details]
error state with "i915.semaphores=0"

(In reply to comment #14)
> Created attachment 103311 [details]
> dmesg
> 
> (In reply to comment #13)
> > Something is not right with the semaphore = 0 error state. Semaphores
> > signals and waits are still being emitted. Can you please confirm you
> > correct set the module parameter for that test by reproducing, and providing
> > the dmesg and new error state?
> 
> I misspelled "i915.semaphores=0" to "i915.semaphore=0" last time. With
> correct parameter "i915.semaphores=0", the test was passed on latest
> -nightly(411fa8b275ee903df6c07976af4eebe5815646a4)
> 
> [root@x-bdw01 tests]# time ./gem_gtt_hog
> IGT-Version: 1.7-g70e6ed9 (x86_64) (Linux:
> 3.16.0-rc6_drm-intel-nightly_411fa8_20140722+ x86_64)
> Time to execute 64 children:            93781.680ms
> 
> real    1m33.837s
> user    0m0.018s
> sys     0m0.804s

Update error state with correct parameter
Comment 16 Guo Jinxian 2014-07-28 01:25:38 UTC
Created attachment 103569 [details]
dmesg

Test failed on latest -nightly(e967a525207bd40ab446e2f809907039f88e66f3). but the failure is a litter difference. I am not sure if they are the same issue.

[root@x-bdw01 tests]# time ./gem_gtt_hog
IGT-Version: 1.7-gfcbc502 (x86_64) (Linux: 3.16.0-rc6_drm-intel-nightly_e967a5_20140727+ x86_64)

Test assertion failure function run, file gem_gtt_hog.c:152:
Failed assertion: x == canary
Test assertion failure function run, file gem_gtt_hog.c:152:
Failed assertion: x == canary
Test assertion failure function run, file gem_gtt_hog.c:152:
Failed assertion: x == canary
Test assertion failure function run, file gem_gtt_hog.c:152:
Failed assertion: x == canary

Test assertion failure function __real_main156, file gem_gtt_hog.c:187:
Failed assertion: status == 0

real    9m37.636s
user    0m0.007s
sys     0m0.179s
[root@x-bdw01 tests]#
[root@x-bdw01 tests]# echo $?
99
Comment 17 Yi Sun 2014-07-28 08:28:38 UTC
It seems not exactly the same as the one when we filing the bug.
can we close this bug and file a new one to track new issue?
Comment 18 Chris Wilson 2014-09-02 06:43:13 UTC
*** Bug 81831 has been marked as a duplicate of this bug. ***
Comment 19 Rodrigo Vivi 2014-09-24 19:55:20 UTC
Is this still happening on latest -nightly?

It contains the semaphores disabled by default, but per latest duplicated mark it seems it isn't anymore related to semaphores, right?
Comment 20 Guo Jinxian 2014-09-25 08:22:12 UTC
(In reply to comment #19)
> Is this still happening on latest -nightly?
> 
> It contains the semaphores disabled by default, but per latest duplicated
> mark it seems it isn't anymore related to semaphores, right?

The failure unable to reproduce on latest -nightly(0f7cc12c94e3a3033a46ce41bed55e8b6b35561b)

root@x-bdw05:/GFX/Test/Intel_gpu_tools/intel-gpu-tools/tests# ./gem_gtt_hog
IGT-Version: 1.7-g5f16ef6 (x86_64) (Linux: 3.17.0-rc6_drm-intel-nightly_0f7cc1_20140925+ x86_64)
Time to execute 64 children:            72740.048ms
root@x-bdw05:/GFX/Test/Intel_gpu_tools/intel-gpu-tools/tests# echo $?
0
Comment 21 Chris Wilson 2014-09-25 08:25:25 UTC
(In reply to comment #19) 
> It contains the semaphores disabled by default, but per latest duplicated
> mark it seems it isn't anymore related to semaphores, right?

No. The last duplicate was because gem_seqno_wrap failed with semaphores enabled but passed with semaphores disabled...
Comment 22 Rodrigo Vivi 2014-09-25 21:04:43 UTC
Ok, so with semaphores disabled we can close this for now.

Created a Jira task to investigate semaphores further later:
https://jira01.devtools.intel.com/browse/VIZ-4400
Comment 23 Guo Jinxian 2014-09-26 01:28:16 UTC
Verified.
Comment 24 Jari Tahvanainen 2016-10-19 09:35:16 UTC
Closing verified+fixed.
Comment 25 Elio 2016-11-23 22:45:17 UTC
The problem with gem_seqno_wrap  still happening on BYT with following configuration:

Platform BYT: Acer Aspire XC-603
CPU: Intel(R) Pentium(R) CPU  J2900  @ 2.41GHz (family 6, model 55, stepping 8)
Motherboard: Aspire XC-603
GPU: Intel® HD Graphics - Intel Corporation Atom Processor Z36xxx/Z37xxx Series Graphics & Display (rev 0e)

Software
Bios: P11-B2
Linux distribution: Ubuntu 16.04 64 bits

Kernel: 4.9.0-rc4 91e164f branch drm-intel-nightly from http://cgit.freedesktop.org/drm-intel/ 
  commit 91e164fea17d3e5366048b6eae3c6eea4e14e9fe
  Author: Ville Syrjälä <ville.syrjala@linux.intel.com>
  Date:   Mon Nov 14 16:31:06 2016 +0200
  drm-intel-nightly: 2016y-11m-14d-14h-30m-30s UTC integration manifest

libdrm-2.4.71-13 670f1e4 from git://anongit.freedesktop.org/mesa/drm
mesa: mesa-13.0.0 df1b0a5 from git://anongit.freedesktop.org/mesa/mesa
cairo 1.15.2 db8a7f1 from git://anongit.freedesktop.org/cairo
xorg-server-1.18.99.902-2 7513da4 from git://git.freedesktop.org/git/xorg/xserver
xf86-video-intel 2.99.917-731 d1d14f2 from git://git.freedesktop.org/git/xorg/driver/xf86-video-intel
libva-1.7.2-40 3a7547b from git://git.freedesktop.org/git/vaapi/libva 
vaapi-intel-driver: 1.7.2-157 55a538c from git://git.freedesktop.org/git/vaapi/intel-driver
intel-gpu-tools-1.16-132 773ac7c from http://anongit.freedesktop.org/git/xorg/app/intel-gpu-tools.git

External screen: DELL U2312HM (VGA)

errors:

sudo ./gem_seqno_wrap
IGT-Version: 1.16-g40a3ada (x86_64) (Linux: 4.9.0-rc6-nightly+ x86_64)
seqno readback differs rb:0x0 vs w:0x1
(gem_seqno_wrap:6004) CRITICAL: Test assertion failure function preset_run_once, file gem_seqno_wrap.c:383:
(gem_seqno_wrap:6004) CRITICAL: Failed assertion: write_seqno(1) == 0
(gem_seqno_wrap:6004) CRITICAL: error: -1 != 0
Stack trace:
  #0 [__igt_fail_assert+0xf1]
  #1 [main+0x382]
  #2 [__libc_start_main+0xf0]
  #3 [_start+0x29]
  #4 [<unknown>+0x29]
Test gem_seqno_wrap failed.
**** DEBUG ****
(gem_seqno_wrap:6004) DEBUG: next_seqno set to: 0x1
(gem_seqno_wrap:6004) DEBUG: next_seqno: 0x0
(gem_seqno_wrap:6004) INFO: seqno readback differs rb:0x0 vs w:0x1
(gem_seqno_wrap:6004) CRITICAL: Test assertion failure function preset_run_once, file gem_seqno_wrap.c:383:
(gem_seqno_wrap:6004) CRITICAL: Failed assertion: write_seqno(1) == 0
(gem_seqno_wrap:6004) CRITICAL: error: -1 != 0
****  END  ****
FAIL (0.024s)

No dmesg available.
[49295.265477] [IGT] gem_seqno_wrap: executing
[49295.289616] [IGT] gem_seqno_wrap: exiting, ret=99
Comment 26 Chris Wilson 2016-11-24 09:46:44 UTC
(In reply to Elio from comment #25)
> The problem with gem_seqno_wrap  still happening on BYT with following
> configuration:

This was a new regression. It should have been a new bug.
Comment 27 Chris Wilson 2016-11-24 12:49:18 UTC
commit 9607ae79710afb453173b90d5bf564788a6e09b1
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Nov 24 09:47:52 2016 +0000

    drm/i915/debugfs: Increment return value of gt.next_seqno
    
    The i915_next_seqno read value is to be the next seqno used by the
    kernel. However, in the conversion to atomics ops for gt.next_seqno, in
    commit 28176ef4cfa5 ("drm/i915: Reserve space in the global seqno during
    request allocation"), this was changed from a post-increment to a
    pre-increment. This increment was missed from the value reported by
    debugfs, so in effect it was reporting the current seqno (last
    assigned), not the next seqno.
Comment 28 yann 2016-11-24 13:18:56 UTC
(In reply to Chris Wilson from comment #27)
> commit 9607ae79710afb453173b90d5bf564788a6e09b1
> Author: Chris Wilson <chris@chris-wilson.co.uk>
> Date:   Thu Nov 24 09:47:52 2016 +0000
> 
>     drm/i915/debugfs: Increment return value of gt.next_seqno
>     
>     The i915_next_seqno read value is to be the next seqno used by the
>     kernel. However, in the conversion to atomics ops for gt.next_seqno, in
>     commit 28176ef4cfa5 ("drm/i915: Reserve space in the global seqno during
>     request allocation"), this was changed from a post-increment to a
>     pre-increment. This increment was missed from the value reported by
>     debugfs, so in effect it was reporting the current seqno (last
>     assigned), not the next seqno.

Elio, please re-test and confirm that the issue you posted here is fixed then.
Moreover, don't forget to create new bug when defect cause is different.


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.