WineHQ Bugzilla – Bug 35580 – d3d8:device test_display_formats() fails on Windows 8

Bugzilla

Last modified: 2014-02-20 12:37:23 UTC

d3d8:device test_display_formats() fails on Windows 8

Bug 35580 - d3d8:device test_display_formats() fails on Windows 8


Summary:	d3d8:device test_display_formats() fails on Windows 8

Status:	CLOSED FIXED

AppDB:	Show Apps affected by this bug
Product:	Wine
Classification:	Unclassified
Component:	d3d
Version:	1.7.12
Hardware:	x86 Windows

Importance:	P2 normal
Target Milestone:	---
Assigned To:	Mr. Bugs

URL:	http://test.winehq.org/data/tests/d3d...
Keywords:	regression, source

Depends on:
Blocks:
	Show dependency tree

Reported:	2014-02-11 11:19 UTC by François Gouget
Modified:	2014-02-20 12:37 UTC (History)
CC List:	0 users

See Also:
Regression SHA1:	1438c74b744d1e06b2efde04e050971d51ed11f8
Fixed by SHA1:	bdb9a9e4f0f5394dffbdb40f113dacf8cca98c0e
Distribution:	---
Staged patchset:

Attachments
Add tracing to the d3d9 test (2.49 KB, patch) 2014-02-12 11:52 UTC, Ken Thomases	Details \| Diff
Acer laptop's log from job 4831 (16.02 KB, text/plain) 2014-02-12 12:24 UTC, François Gouget	Details
View All Add an attachment (back traces, logs, proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description François Gouget 2014-02-11 11:19:41 UTC

The failing test is:
device.c:1056: Test failed: Got unexpected hr 0 for D3DFMT_R5G6B5 / D3DFMT_R5G6B5, windowed 0, should_pass 0.

This test fails on three Windows 8 machines but apparently not on any other Windows version:
 * dr-w8-E1-1200
 * My Acer laptop running Windows 8.1 on real hardware with an Intel HD4000 graphics card.
 * The WineTestBot w864 VM.

http://test.winehq.org/data/tests/d3d8:device.html

The commit that introduced this test is:

commit 1438c74b744d1e06b2efde04e050971d51ed11f8
Author: Henri Verbeet <hverbeet@codeweavers.com>
Date:   Wed Jan 29 19:28:36 2014 +0100

    d3d8/tests: Port test_display_formats() to d3d8.

Comment 1 François Gouget 2014-02-11 11:30:17 UTC

d3d9:device has the same test and it also fails on my Acer laptop and on the w864 VM.

The commit for that one is:

commit 71d018b25b8ee8a4454b047188775e527a33a79e
Author: Henri Verbeet <hverbeet@codeweavers.com>
Date:   Wed Jan 29 19:28:37 2014 +0100

    d3d9/tests: Cleanup test_display_formats() a bit.

Comment 2 François Gouget 2014-02-11 11:32:57 UTC

On my Acer laptop the d3d9:device test runs through a few more formats than on the VMs and thus has a few more failures. Specifically:

device.c:2350: Test failed: Got unexpected hr 0 for D3DFMT_R5G6B5 / D3DFMT_R5G6B5, windowed 0, should_pass 0.
device.c:2350: Test failed: Got unexpected hr 0 for D3DFMT_X1R5G5B5 / D3DFMT_R5G6B5, windowed 0x1, should_pass 0.
device.c:2350: Test failed: Got unexpected hr 0 for D3DFMT_X1R5G5B5 / D3DFMT_X8R8G8B8, windowed 0x1, should_pass 0.
device.c:2350: Test failed: Got unexpected hr 0 for D3DFMT_X1R5G5B5 / D3DFMT_A8R8G8B8, windowed 0x1, should_pass 0.
device.c:2350: Test failed: Got unexpected hr 0 for D3DFMT_X1R5G5B5 / D3DFMT_A2R10G10B10, windowed 0x1, should_pass 0.

On the same machine the d3d8:device test had only the one failure.

Comment 3 Henri Verbeet 2014-02-11 11:34:14 UTC

Did that really start failing after 71d018b25b8ee8a4454b047188775e527a33a79e? I think that should be 6f498c4946087337956a60eb499559b061b015c1, or possibly even earlier.

Comment 4 François Gouget 2014-02-11 12:26:48 UTC

On my laptop I see the failure on the 2014/01/29 test results:
http://test.winehq.org/data/a1b963490a80f92aa3359f5ee97215c0b3064c22/win8_fg-acer64-w81-t32/d3d8:device.html

If I understand the patch correctly, prior to this date the test failure would result in a 'CheckDeviceType...' message. But I don't see it on the 2014/01/28 test results:
http://test.winehq.org/data/6f498c4946087337956a60eb499559b061b015c1/win8_newtb-w864-32/d3d8:device.html

Before this date the test results are sparse but just to be thorough, here are two older test results from my laptop that both have a single unrelated test failure (one that persists to this day, hint hint!):

2014/01/15
http://test.winehq.org/data/2b8e900ac4f1cfce6b3234967b4931a524114ddf/win8_fg-acer64-w81-t32/d3d8:device.html

2014/01/13
http://test.winehq.org/data/6bd6193cb13177bc012141458703fa5632ba5c29/win8_fg-acer64-w81-t32/d3d8:device.html

Comment 5 Ken Thomases 2014-02-12 11:52:48 UTC

Created attachment 47515 [details]
Add tracing to the d3d9 test

Francois, please collect the tracing from the test with this patch applied, on your Acer laptop.  The patch also disables all of the other tests to just focus on the one of interest.

The compiled test binary is available from <https://newtestbot.winehq.org/JobDetails.pl?Key=4831>.

Comment 6 François Gouget 2014-02-12 12:24:01 UTC

Created attachment 47516 [details]
Acer laptop's log from job 4831

I ran both the 32-bit and 64-bit versions and they returned the exact same output. For reference, the laptop is running Windows 8.1 64-bit French Home Edition.

Comment 7 Ken Thomases 2014-02-12 14:48:54 UTC

I have submitted patches to mark the Win8 64-bit behavior as broken:

http://source.winehq.org/patches/data/102417
http://source.winehq.org/patches/data/102418

Comment 8 Henri Verbeet 2014-02-15 07:29:21 UTC

Should no longer happen since commit bdb9a9e4f0f5394dffbdb40f113dacf8cca98c0e.

Comment 9 François Gouget 2014-02-17 03:37:32 UTC

Yep, this is fixed (at least on my Acer and on the w864 VM).

Comment 10 Alexandre Julliard 2014-02-20 12:37:23 UTC

Closing bugs fixed in 1.7.13.