Bug 93769 - [BAT] kms_force_connector_basic/force-connector-state skipped randomly
Summary: [BAT] kms_force_connector_basic/force-connector-state skipped randomly
Status: CLOSED WORKSFORME
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: DRI git
Hardware: Other All
: highest major
Assignee: Jari Tahvanainen
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-01-19 10:04 UTC by Tvrtko Ursulin
Modified: 2016-09-22 08:13 UTC (History)
3 users (show)

See Also:
i915 platform: IVB, SNB
i915 features: display/eDP


Attachments

Description Tvrtko Ursulin 2016-01-19 10:04:37 UTC
From the CI run of: /archive/results/CI_IGT_test/Patchwork_1179/

Apparently, according to e-mail results this test skipped on ivb-t430s:

Test kms_force_connector_basic:
        Subgroup force-connector-state:
                pass       -> SKIP       (ivb-t430s)

CI page does not provide test output so I can't figure out why it skipped.

Looking at the history at: /archive/results/CI_IGT_test/igt@kms_force_connector_basic@force-connector-state.html

It seems this test did not use to skip on this platform, but it skips on ones with no VGA.
Comment 1 Daniel Vetter 2016-01-19 10:10:42 UTC
Could be that the cable isn't plugged in correctly and falls out occasionally. Tomi, can you pls check?
Comment 2 Tomi Sarvela 2016-01-19 10:12:49 UTC
Lenovo T430s is laptop and doesn't have external monitor connected.
Comment 3 Daniel Vetter 2016-01-27 09:40:34 UTC
(In reply to Tomi Sarvela from comment #2)
> Lenovo T430s is laptop and doesn't have external monitor connected.

In that case there's likely something fishy going on with the connector injection. This stuff shouldn't ever randomly skip ...
Comment 4 Daniel Vetter 2016-02-10 07:48:11 UTC
Note, sometimes this even seems to result in broken EDID checksums, indicating that there's a race somewhere in this code.

See also bug #93226
Comment 5 cprigent 2016-04-21 14:30:23 UTC
tvrtko.ursulin@linux.intel.com, daniel@ffwll.ch,
I checked kms_force_connector_basic@force-connector-state result on several platforms:
Skip: BSW, HSW, BDW, SKL, KBL, BXT
Pass: IVB, SNB

Results are identical over time. Last check is with:
kernel 4.6.0-rc3 d9131d6 drm-intel-nightly from git://anongit.freedesktop.org/drm-intel 
  commit d9131d62d18ba94fb3ca019f1156c22b5f4ce23c
  Author: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
  Date:   Fri Apr 15 14:54:26 2016 +0100
  drm-intel-nightly: 2016y-04m-15d-13h-53m-44s UTC integration manifest

Does it still need to be investigated?
Comment 6 anusha 2016-08-11 21:04:41 UTC
Hi,
Can you run xrandr on your laptop and share the output?
Comment 7 Tomi Sarvela 2016-08-12 07:33:02 UTC
t430s:~$ xrandr
Screen 0: minimum 8 x 8, current 1600 x 900, maximum 32767 x 32767
LVDS1 connected 1600x900+0+0 (normal left inverted right x axis y axis) 310mm x 170mm
   1600x900      60.00*+
   1368x768      60.00  
   1280x720      60.00  
   1024x768      60.00  
   1024x576      60.00  
   960x540       60.00  
   800x600       60.32    56.25  
   864x486       60.00  
   800x450       60.00  
   640x480       59.94  
   720x405       60.00  
   640x360       60.00  
DP1 disconnected (normal left inverted right x axis y axis)
DP2 disconnected (normal left inverted right x axis y axis)
DP3 disconnected (normal left inverted right x axis y axis)
HDMI1 disconnected (normal left inverted right x axis y axis)
HDMI2 disconnected (normal left inverted right x axis y axis)
HDMI3 disconnected (normal left inverted right x axis y axis)
VGA1 disconnected (normal left inverted right x axis y axis)
VIRTUAL1 disconnected (normal left inverted right x axis y axis)
Comment 8 anusha 2016-09-01 17:51:29 UTC
The test looks for a native VGA connector and skips if not found. That is the reason why this test skips on newer platforms. This is not a peculiar behavior, its an expected behavior. We can close this bug.
Comment 9 Chris Wilson 2016-09-01 20:32:08 UTC
(In reply to anusha from comment #8)
> The test looks for a native VGA connector and skips if not found. That is
> the reason why this test skips on newer platforms. This is not a peculiar
> behavior, its an expected behavior. We can close this bug.

That doesn't explain:


Test kms_force_connector_basic:
        Subgroup force-connector-state:
                pass       -> SKIP       (ivb-t430s)

which is what this bug is about, sporadic SKIPs.
Comment 10 Jari Tahvanainen 2016-09-05 07:20:53 UTC
The latest executions on CI system are showing the following:

- on fi-ilk-m540 after 16-Aug-2016 10:47 all execution are having pass
- on fi-ilk-650 after 29-Aug-2016 18:24 (machine attached to CI on 29-Aug-2016) all execution are having SKIP as they should since
Test requirement not met in function main, file kms_force_connector_basic.c:111:
Test requirement: !(vga_connector->connection == DRM_MODE_CONNECTED)
Last errno: 2, No such file or directory
Subtest force-connector-state: SKIP

- on fi-snb-2520m there is one sporadic skip on this case at 26-Aug-2016 11:24 
IGT-Version: 1.15-gc1e0e30 (x86_64) (Linux: 4.8.0-rc3-CI-CI_DRM_1590+ x86_64)
Test requirement not met in function main, file kms_force_connector_basic.c:111:
Test requirement: !(vga_connector->connection == DRM_MODE_CONNECTED)
Last errno: 2, No such file or directory
Subtest force-connector-state: SKIP
while all other execution are having pass e.g.
IGT-Version: 1.16-gf92e47f (x86_64) (Linux: 4.8.0-rc4-CI-CI_DRM_1612+ x86_64)
Subtest force-connector-state: SUCCESS (0.693s)

So one cannot yet say that this sporadic skip does not happen, let's keep this open and follow the situation for couple of the weeks at least.
Comment 11 Jari Tahvanainen 2016-09-22 08:13:20 UTC
Marking this resolved+Worksforme (and will close) this since sporadic skip is not visible anymore. 

Latest skip on ivb-t430s visible on 23-Apr-2016 (CI_IGT_test/Patchwork_2017/ivb-t430s), no skips on Patchwork testing after this hw was renames as fi-ivb-3520m.
Root cause unfortunately not known for these old flip-flops.

There is one time SKIP on CI_IGT_test/CI_DRM_1590/fi-snb-2520m 26-Aug-2016, for which we do not know root cause. All executions after that are Success/Pass.


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.