Bug 78426

Summary: [IVB/HSW/BYT]Some subcases of igt/gem_exec_parse fails
Product: DRI Reporter: Guo Jinxian <jinxianx.guo>
Component: DRM/IntelAssignee: Intel GFX Bugs mailing list <intel-gfx-bugs>
Status: CLOSED FIXED QA Contact: Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity: normal    
Priority: medium CC: intel-gfx-bugs
Version: unspecified   
Hardware: Other   
OS: All   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
dmesg
none
dmesg on fixes branch
none
dmesg none

Description Guo Jinxian 2014-05-08 08:39:10 UTC 
Created attachment 98666 [details]
dmesg

*System Environment:
--------------------------
Regression: No.  
The cases always fails

Non-working platforms: BYT

 *kernel: 
--------------------------
-nightly: dd28119c31cf06fc4c3bb548699018a91e45a676 (fails)
-queued: 1cf0ba14740d96fbf6f58a201f000a34b74f4725 (fails)
    Author: Chris Wilson <chris@chris-wilson.co.uk>
    Date:   Mon May 5 09:07:33 2014 +0100

    drm/i915: Flush request queue when waiting for ring space

    During the review of

    commit 1f70999f9052f5a1b0ce1a55aff3808f2ec9fe42
    Author: Chris Wilson <chris@chris-wilson.co.uk>
    Date:   Mon Jan 27 22:43:07 2014 +0000

        drm/i915: Prevent recursion by retiring requests when the ring is full

    Ville raised the point that our interaction with request->tail was
    likely to foul up other uses elsewhere (such as hang check comparing
    ACTHD against requests).

    However, we also need to restore the implicit retire requests that certain
    test cases depend upon (e.g. igt/gem_exec_lut_handle), this raises the
    spectre that the ppgtt will randomly call i915_gpu_idle() and recurse
    back into intel_ring_begin().

    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78023
    Reviewed-by: Brad Volkin <bradley.d.volkin@intel.com>
    [danvet: Remove now unused 'tail' variable as spotted by Brad.]
    Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>


-fixes: e4c610fe051579ba0a1fadf339905b0231c6ef94 (skips)
    Author: Egbert Eich <eich@suse.de>
    Date:   Fri Apr 11 19:07:44 2014 +0200

    drm/i915/SDVO: For sysfs link put directory and target in correct order

    When linking the i2c sysfs file into the connector's directory
    pass directory and link target in the right order.
    This code was introduced with:

      commit 931c1c26983b4f84e33b78579fc8d57e4a14c6b4
      Author: Imre Deak <imre.deak@intel.com>
      Date:   Tue Feb 11 17:12:51 2014 +0200

        drm/i915: sdvo: add i2c sysfs symlink to the connector's directory

        This is the same what we do for DP connectors, so make things more
        consistent.

    Signed-off-by: Egbert Eich <eich@suse.de>
    Reviewed-by: Imre Deak <imre.deak@intel.com>
    Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
    Signed-off-by: Jani Nikula <jani.nikula@intel.com>


 *Bug detailed description:
-----------------------------
Some subcases of igt/gem_exec_parse below fails
igt/gem_exec_parse/basic-rejected
igt/gem_exec_parse/batch-without-end
igt/gem_exec_parse/bitmasks
igt/gem_exec_parse/oacontrol-tracking
igt/gem_exec_parse/registers


Output:
 ./gem_exec_parse --run-subtest registers
IGT-Version: 1.6-g7935bbd (x86_64) (Linux: 3.15.0-rc3_drm-intel-next-queued_1cf0ba_20140508+ x86_64)
Test assertion failure function exec_batch, file gem_exec_parse.c:134:
Last errno: 0, Success
Failed assertion: expected_ret == 0
Subtest registers: FAIL

The result on -fixes was skipped:
./gem_exec_parse --run-subtest registers
IGT-Version: 1.6-g7935bbd (x86_64) (Linux: 3.15.0-rc3_drm-intel-fixes_e4c610_20140508+ x86_64)
Test requirement not met in function __real_main210, file gem_exec_parse.c:222:
Last errno: 22, Invalid argument
Test requirement: (!(!rc && parser_version > 0))
Subtest registers: SKIP


 *Reproduce steps:
---------------------------- 
1. ./gem_exec_parse --run-subtest registers
Comment 1 Guo Jinxian 2014-05-08 08:40:10 UTC 
Created attachment 98667 [details]
dmesg on fixes branch
Comment 2 Daniel Vetter 2014-05-15 16:16:54 UTC 
This is known to be broken on -fixes. Please only reopen if there are failures with gem_exec_param on byt on -nightly.
Comment 3 Guo Jinxian 2014-08-14 05:38:27 UTC 
Created attachment 104602 [details]
dmesg

(In reply to comment #2)
> This is known to be broken on -fixes. Please only reopen if there are
> failures with gem_exec_param on byt on -nightly.

Test failed on latest -nightly(da31e7c60be217316278a055dd3f91c33913270f) on byt-m

root@x-bytm02:/GFX/Test/Intel_gpu_tools/intel-gpu-tools/tests# ./gem_exec_parse
IGT-Version: 1.7-geda904c (x86_64) (Linux: 3.16.0_drm-intel-nightly_da31e7_20140814+ x86_64)
Subtest basic-allowed: SUCCESS
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest basic-rejected: FAIL
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest registers: FAIL
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest bitmasks: FAIL
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest batch-without-end: FAIL
Subtest cmd-crossing-page: SUCCESS
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest oacontrol-tracking: FAIL
Comment 4 Guo Jinxian 2014-08-14 07:17:15 UTC 
This bug is able to reproduce on IVB and HSW platforms.

[root@x-ivb9 tests]# ./gem_exec_parse
IGT-Version: 1.7-g5c7bcb1 (x86_64) (Linux: 3.16.0_drm-intel-nightly_da31e7_20140814_debug+ x86_64)
Subtest basic-allowed: SUCCESS
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest basic-rejected: FAIL
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest registers: FAIL
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest bitmasks: FAIL
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest batch-without-end: FAIL
Subtest cmd-crossing-page: SUCCESS
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 == -22
Subtest oacontrol-tracking: FAIL
Comment 5 lu hua 2014-10-22 01:56:30 UTC 
It works well on -nightly kernel.
It also happens on -fixes kernel(commit f114040e3ea6e0737)

run ./gem_exec_parse --run-subtest registers
output:
IGT-Version: 1.8-gbba1cd0 (x86_64) (Linux: 3.18.0-rc1_drm-intel-fixes_f11404_20141021_debug+ x86_64)
Test assertion failure function exec_batch, file gem_exec_parse.c:135:
Failed assertion: __gem_execbuf(fd, &execbuf) == expected_ret
error: 0 != -22
Subtest registers: FAIL (0.000s)
Comment 6 Rodrigo Vivi 2015-01-15 19:03:20 UTC 
If it works well on -nightly it was probably fixed already. Feel free to reopen if it still happens.
Comment 7 Ding Heng 2015-01-20 08:19:42 UTC 
pass on BYT/HSW/IVB with nightly latest baranch:
d2a1764437d8b4c30948704ff6c64e4bbfd1df7c(2015-01-20)

change state to verified.
Comment 8 Elizabeth 2017-10-06 14:38:13 UTC 
Closing old verified.

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.