[Bug 112230] [CI][RESUME]igt at kms_plane_lowres@pipe-* - skip - Current mode not tall enough; plane would still be onscreen after switch to 10x7, SKIP

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=112230

Matt Roper <matthew.d.roper at intel.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |NOTABUG
                 CC|                            |matthew.d.roper at intel.com

--- Comment #2 from Matt Roper <matthew.d.roper at intel.com> ---
This skip is expected/intentional.  The display you have connected to DP-2 is
running at a very small 848x480 resolution when the driver inherits the display
configuration from the BIOS:

<7>[   13.983855] [drm:drm_client_firmware_config.isra.8] connector DP-2 on
[CRTC:152:pipe B]: 848x480

I think your monitor is larger than this, but this is the mode that the BIOS
decided to setup and that we inherited when our driver initialized (it's
possible that a flaky connection led to link training failures in the BIOS and
caused the BIOS to choose this as the "best" possible mode).

Given that tiny mode, it makes sense for this test to skip since this test
relies on the current mode being a "big" mode and that the test can switch to a
smaller mode during its testing.  Specifically, this test places a small
cursor/sprite plane in the lower left corner of the screen, then switches to a
"lowres" mode that is enough smaller that the plane's position will now be
completely offscreen and clipped.  When looking for a "lowres" mode, the test
walks down the driver's modelist looking for one that's enough smaller than the
current mode that it will cause the plane to be offscreen.  If no suitable
lowres mode is found on the driver's modelist, the driver will check whether a
standard 10x7 mode is good enough to meet this criteria; if it isn't then the
test gives up and skips since there's nothing meaningful that can be tested.

Given that your current mode is 848x480, you'd need a mode with a vertical
resolution of 480-256=224 or lower.  Ville has some patches that make some
changes to this test, including changing the plane size from 256 to 64, but
even with those patches, you'd still need a vertical resolution of 480-64=416
or lower.  So I don't think his patches will make a difference for the failures
here, even though they're doing other general cleanup to this test.

The test and driver appear to be working as expected here, so marking this as
NOTABUG.  Realistically the monitor probably shouldn't be initialized to such a
small 848x480 mode by the BIOS, but as I mentioned before, there's a good
chance that's just due to a bad connection that causes link training to fail.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20191113/a260f4b6/attachment.html>