New subject: [Bug 53212] d3d11:d3d11 - test_clear_image_unordered_access_view() fails in the TestBot VMs

20 Jun 2022


      https://bugs.winehq.org/show_bug.cgi?id=53212
Bug ID: 53212
           Summary: d3d11:d3d11 - test_clear_image_unordered_access_view()
                    fails in the TestBot VMs
           Product: Wine
           Version: unspecified
          Hardware: x86-64
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: directx-d3d
          Assignee: wine-bugs@winehq.org
          Reporter: fgouget@codeweavers.com
      Distribution: ---
d3d11:d3d11 - test_clear_image_unordered_access_view() fails in the TestBot
VMs:
d3d11.c:16719: Test failed: Dim 0, Test 15: At layer 0, (0,0,0), expected
0x00c01001, got 0x00000000.
d3d11.c:16719: Test failed: Dim 1, Test 15: At layer 0, (0,0,0), expected
0x00c01001, got 0x00000000.
d3d11.c:16719: Test failed: Dim 2, Test 15: At layer 0, (0,0,0), expected
0x00c01001, got 0x00000000.
https://test.winehq.org/data/patterns.html#d3d11:d3d11
These failures don't happen on my box which has an Intel graphics card instead
of QEmu's QXL graphics card.
Note also that these failures still happen on my debian11 VM which has a single
monitor instead of the TestBot VMs' multi-monitor setup.
So this seems to be related to the VMs QXL GPU.
Despite being systematic these failures someimes cause false positives because
of the test's multi-threading: sometimes these failures come before the
test_texture() failures (Test 60:), and most of the time they come after. The
order change causes the TestBot's diffing algorithm to consider them to be new
failures if they come in an order it has not seen before.
-- 
Do not reply to this email, post in Bugzilla using the
above URL to reply.
You are receiving this mail because:
You are watching all bug changes.

[Bug 53212] New: d3d11:d3d11 - test_clear_image_unordered_access_view() fails in the TestBot VMs