March 2023 - Wine-bugs - list.winehq.org

[Bug 54637] New: riched20:txtsrv - test_TxGetNaturalSize fails if system GUI font's glyph widths are wider than expected by the test
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=54637 Bug ID: 54637 Summary: riched20:txtsrv - test_TxGetNaturalSize fails if system GUI font's glyph widths are wider than expected by the test Product: Wine Version: 8.3 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: richedit Assignee: wine-bugs(a)winehq.org Reporter: jinoh.kang.kr(a)gmail.com CC: huw(a)codeweavers.com, jzeng(a)codeweavers.com Regression SHA1: 6e264bbd1ddf0d271d4506462ecd4f20f4a20597 Distribution: --- Today, test_TxGetNaturalSize() creates a pop-up window with a fixed size (extent) of 100 x 100. The test function then proceeds to compute the natural size of rich edit control that fits the sample text "TestSomeText" and compare it to the RECT calculated by DrawText. It appears that this test fails if the text width exceeds the width of the test window's client area. In this case, DrawText() with DT_WORDBREAK breaks the text into the two lines due to text wrapping; however, Rich Edit's ITextServices::TxGetNaturalSize implementation does not seem to perform text wrapping on overflow. This results in extent mismatch. In conclusion, the test may fail if the rendered width of the sample text "TestSomeText" is larger than what the test expects. This depends on the current font used for DEFAULT_GUI_FONT. Omitting DT_WORDWRAP for the DrawText() call seems to fix this test failure. -- 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.

1 3

[Bug 53526] New: kernel32:sync - test_timer_queue() occasionally fails to delete the timer on Windows 10
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=53526 Bug ID: 53526 Summary: kernel32:sync - test_timer_queue() occasionally fails to delete the timer on Windows 10 Product: Wine Version: unspecified Hardware: x86-64 OS: Windows Status: NEW Severity: normal Priority: P2 Component: kernel32 Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com kernel32:sync - test_timer_queue() occasionally fails to delete the timer on Windows 10: sync.c:1135: Test failed: DeleteTimerQueueTimer https://test.winehq.org/data/patterns.html#kernel32:sync This failure happened twice: * 2022-05-27 w10pro64-de-64 * 2022-08-04 w10pro64-he-64 -- 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.

1 3

[Bug 53818] New: foobar2000 crashes shortly after startup
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=53818 Bug ID: 53818 Summary: foobar2000 crashes shortly after startup Product: Wine Version: 7.19 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: matthias.grobarek(a)posteo.de Distribution: --- Created attachment 73344 --> https://bugs.winehq.org/attachment.cgi?id=73344 winedbg output After starting foobar2000, it quickly crashes after a few seconds, no matter what you do. Wine 7.18 and 7.17 are not affected. I’m using wine-staging (no matter the version) on Gentoo Linux (kernel 6.0.2), compiled for both 32bit and 64bit, with the following flags activated: x, alsa, fontconfig, mshtml, gstreamer, mingw, mscoree, gettext, openal, opengl, sdl, gnutls, freetype, udev, dbus, unwind, usb, vulkan, xcomposite. I attached the output of winedbg and also the crash log and dump that foobar2000 itself produces. If you need any more information, please tell me what exactly I should do. -- 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.

1 12

[Bug 53488] New: The dxgi:dxgi output is too big on debiant
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=53488 Bug ID: 53488 Summary: The dxgi:dxgi output is too big on debiant Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: d3d Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- The dxgi:dxgi output is too big in Wine. It used to be about 24 KB but the commit below pushed it to 34 KB, over the 32 KB limit. Then on 2022-07-22 another commit improved the situation on debian11 but not on debiant where the size has not changed since the commit below. commit 184ff3bfbb4be3e3fbcfaaba3ec7095d2d885882 Author: Nikolay Sivov <nsivov(a)codeweavers.com> Date: Sat Jun 11 19:36:20 2022 +0300 dxgi: Create DXGI resource object, optionally supporting surface interfaces. Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> dlls/d3d11/d3d11_private.h | 4 +- dlls/d3d11/tests/d3d11.c | 7 - dlls/d3d11/texture.c | 130 +++++++------- dlls/dxgi/Makefile.in | 2 +- dlls/dxgi/device.c | 24 +-- dlls/dxgi/dxgi_private.h | 11 +- dlls/dxgi/resource.c | 423 +++++++++++++++++++++++++++++++++++++++++++++ dlls/dxgi/surface.c | 296 ------------------------------- dlls/dxgi/tests/dxgi.c | 34 ++-- include/wine/winedxgi.idl | 7 +- 10 files changed, 526 insertions(+), 412 deletions(-) create mode 100644 dlls/dxgi/resource.c delete mode 100644 dlls/dxgi/surface.c -- 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.

1 3

[Bug 54058] New: user32:input - test_ToAscii() fails in the Hindi UTF-8 locale
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=54058 Bug ID: 54058 Summary: user32:input - test_ToAscii() fails in the Hindi UTF-8 locale Product: Wine Version: unspecified Hardware: x86-64 OS: Windows Status: NEW Severity: normal Priority: P2 Component: user32 Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Created attachment 73613 --> https://bugs.winehq.org/attachment.cgi?id=73613 user32/tests: Skip a ToAscii() test in UTF-8 code pages. user32:input - test_ToAscii() fails in the Hindi UTF-8 locale: input.c:3133: Test failed: ToAscii for character 'A' didn't return 1 (was 0) See https://test.winehq.org/data/patterns.html#user32:input Note that this did not happen before 2022-11-21 because some unknown Windows quirk caused the UserDefaultLCID and ThreadLocale to be reset to English before the test was started. In any case it seems ToAscii() does not depend just on the keyboard layout but also on the current code page which makes sense. So then maybe we should limit the test to non-UTF-8 code pages as in the attached patch? -- 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.

1 5

[Bug 51848] New: Performance Regression in Secondhand Lands
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=51848 Bug ID: 51848 Summary: Performance Regression in Secondhand Lands Product: Wine Version: 6.18 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: escomk3(a)hotmail.com Distribution: --- After 4261369e5d8 [1], FPS will be near zero, and the game area will be black, with the actual should-be content sometimes appearing and disappearing. Standard output with default'ish debug channels doesn't show any difference with a build prior to this commit. 1. https://source.winehq.org/git/wine.git/commit/4261369e5d8f1d71e8d2074747515… -- 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.

1 18

[Bug 52948] New: gdi32:font - test_EnumFonts() fails on Arial Bold on Windows in Russian
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=52948 Bug ID: 52948 Summary: gdi32:font - test_EnumFonts() fails on Arial Bold on Windows in Russian Product: Wine Version: unspecified Hardware: x86-64 OS: Windows Status: NEW Severity: normal Priority: P2 Component: gdi32 Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com gdi32:font - test_EnumFonts() fails on Arial Bold on Windows in Russian: font.c:5409: Test failed: font Arial Bold is not enumerated font.c:5412: Test failed: expected FW_BOLD got 400 font.c:5419: Test failed: font Arial Bold Italic is not enumerated font.c:5421: Test failed: expected Arial got Arial Bold font.c:5422: Test failed: expected FW_BOLD got 400 https://test.winehq.org/data/patterns.html#gdi32:font Maybe this can happen in other locales, but among those that are tested this only happens with the Russian one (w10pro64-ru). -- 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.

1 3

[Bug 53182] New: shell32:shelllink - A save(NULL, TRUE) fails randomly in test_load_save() on Wine
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=53182 Bug ID: 53182 Summary: shell32:shelllink - A save(NULL, TRUE) fails randomly in test_load_save() on Wine Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: shell32 Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- A save(NULL, TRUE) fails randomly in test_load_save() on Wine: shelllink.c:775: Test failed: save(NULL, TRUE) failed (0x80070020) https://test.winehq.org/data/patterns.html#shell32:shelllink This is probably the same issue as bug 28290, the only difference being that it does not happen at a point that leads to a NULL pointer access. -- 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.

1 3

[Bug 54078] New: ntdll:pipe - test_blocking() sometimes fails in Wine when the pipe is not signaled
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=54078 Bug ID: 54078 Summary: ntdll:pipe - test_blocking() sometimes fails in Wine when the pipe is not signaled Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- ntdll:pipe - test_blocking() sometimes fails in Wine when the pipe is not signaled: pipe.c:1769: Test marked todo: client is not signaled pipe.c:1751: Test failed: pipe is not signaled pipe.c:1769: Test marked todo: client is not signaled pipe.c:1770: Test failed: pipe is not signaled See https://test.winehq.org/data/patterns.html#ntdll:pipe Here is this failure's nightly WineTest run history: * There was no failure in June and July. * The first known failure happened on 2022-08-13, the next on 2022-08-18. * The third failure happened on 2022-09-23. * The fourth on 2022-11-04 which was followed by 9 more in November. So it's likely something changed in early August-ish that introduced this failure. But there has been a further change in November that makes it more frequent (maybe the latter is because the TestBot now runs two VMs at a time so the host is more busy, though that started a bit earlier, on 2022-10-25). This failure also happens in the GitLab CI. -- 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.

1 4

[Bug 54553] New: mmdevapi:propstore - The 32-bit test_setvalue_on_wow64() fails on Windows 10 2004+
by WineHQ Bugzilla 17 Mar '23

17 Mar '23

https://bugs.winehq.org/show_bug.cgi?id=54553 Bug ID: 54553 Summary: mmdevapi:propstore - The 32-bit test_setvalue_on_wow64() fails on Windows 10 2004+ Product: Wine Version: unspecified Hardware: x86-64 OS: Windows Status: NEW Severity: normal Priority: P2 Component: mmdevapi Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com mmdevapi:propstore - The 32-bit test_setvalue_on_wow64() fails on Windows 10 2004+: propstore.c:186: Test failed: Wrong error when opening mmdevices Render key: 0 See propstore.c:186: Test failed: Wrong error when opening mmdevices Render key: 0 As far as I can tell what this means is that the assumption that "should NOT find the key in 32-bit view" is incorrect on Windows 10 2004+. So maybe the old behavior should be considered to be broken() nowadays? -- 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.

1 4