March 2022 - Wine-bugs - list.winehq.org

[Bug 52743] New: regression cannot write to _fdopen'ed stdout
by WineHQ Bugzilla 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=52743 Bug ID: 52743 Summary: regression cannot write to _fdopen'ed stdout Product: Wine Version: 7.5 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: xantares09(a)hotmail.com Distribution: --- since wine 7.2, opening stdout with _fdopen and printing into it crashes ``` #include <stdio.h> #include <io.h> #include <fcntl.h> int main() { FILE * m_stdout = _fdopen(_fileno(stdout), "wb"); fprintf(m_stdout, "foo\n"); fclose(m_stdout); return 0; } $ x86_64-w64-mingw32-gcc main.c $ wine a.exe ... wine: Unhandled page fault on read access to 0000000000000050 at address 0000000170059685 (thread 0128), starting debugger... ... ``` For context, this is what qt-xmlpatterns does: https://github.com/qt/qtxmlpatterns/blob/dev/tools/xmlpatterns/main.cpp#L184 -- 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 17

[Bug 48045] New: comdlg32:filedlg crashes or times out randomly
by WineHQ Bugzilla 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=48045 Bug ID: 48045 Summary: comdlg32:filedlg crashes or times out randomly Product: Wine Version: unspecified Hardware: x86 OS: Windows Status: NEW Severity: normal Priority: P2 Component: comdlg32 Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Created attachment 65586 --> https://bugs.winehq.org/attachment.cgi?id=65586 comdlg32:filedlg: Traces to debug random crash / timeouts. comdlg32:filedlg crashes randomly on all Windows versions from Vista to Windows 8 while Windows 10 (at least 1507, 1709 and 1809) gets random timeouts. https://test.winehq.org/data/tests/comdlg32:filedlg.html When successful comdlg32:filedlg does not print much so the standard reports are pretty much useless to spot the place where the test gets stuck on Windows 10, assuming it gets stuck in a specific place rather than just being slow. As for the crashes, they don't trigger the tests exc_filter() so there is no indication of the last successful ok() call. Turning on 'Report successful tests' may be preventing the crash (or these runs just got lucky). It's possible to reproduce the crash while adding the traces in the attached patch. This is not entirely fruitful as the crashes don't seem to always happen in the same place. test_extension() is the place where most issues happen though: https://testbot.winehq.org/JobDetails.pl?Key=59060 -> wvistau64: Crash in test_extension(). Based on the test_extension_helper() traces, it happened in the loop with i=1 so for "TestFilter (*.abc;)\0*.abc;\0". https://testbot.winehq.org/JobDetails.pl?Key=59086 -> w2008s64+exe: Crash on or after test_extension() return https://testbot.winehq.org/JobDetails.pl?Key=59088 -> wvistau64+exe: Crash in test_DialogCancel() https://testbot.winehq.org/JobDetails.pl?Key=59091 -> wvistau64_fr: Spent 39s in test_null_filename() but no timeout https://testbot.winehq.org/JobDetails.pl?Key=59139 -> wvistau64: Crash in test_null_filename() -> wvistau64_fr: Timeout in test_extension() filedlg.c:1048: filter=[TestFilter (*.ab?) lpstrDefExt=NULL] https://testbot.winehq.org/JobDetails.pl?Key=59151 -> wvistau64: Crash in test_extension() filedlg.c:1048: filter=[TestFilter (*.abc.def)] -> wvistau64_fr: Timeout in or after test_mru() filedlg.c:637: 19 test_ok filedlg.c:904: 29 test_getfolderpath filedlg.c:986: 30 test_mru -> w1064v1809_zh_CN: Two failures in test_extension() filedlg.c:1053: Test failed: TestFilter (*.pt*;*.abc): expected TRUE filedlg.c:1059: Test failed: TestFilter (*.pt*;*.abc): Filename is deadbeef, expected deadbeef.xyz Memory corruption, maybe happening quite early in the test? -- 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 52554] New: Serbian locale mapping cause crash on multiple installshield wizards
by WineHQ Bugzilla 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=52554 Bug ID: 52554 Summary: Serbian locale mapping cause crash on multiple installshield wizards Product: Wine Version: 7.2 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: lei.pero(a)gmail.com Distribution: --- Since commits: kernelbase: Map LANG_SERBIAN_NEUTRAL in ConvertDefaultLocale(). f51e44c1802338cdd41b38efe2757b642b619d6f ntdll: Map LANG_SERBIAN_NEUTRAL in RtlLocaleNameToLcid(). db3e08770fe9bfb6f06a71761e48a40fe9764058 starting with wine-6.2 - when using locale sr_RS.UTF-8, some (specifically older) installers do crash, for example 3DMark03, preventing installation, passing LANG=en_US.UTF-8 EV works as expected with no crash. Reverting those two specific commits solves the issue. Or changing line in: dlls/ntdll/locale.c from: case MAKELANGID( LANG_SERBIAN, SUBLANG_NEUTRAL ): *lcid = LANG_SERBIAN_NEUTRAL; to: case MAKELANGID( LANG_SERBIAN, SUBLANG_NEUTRAL ): *lcid = MAKELCID( 0x02, SORT_DEFAULT ); solves the issue as well, but that basically renders all those commits useless I assume? On the topic, but unrelated to this bug, I can see changes in dlls/kernel32/nls/srl.nls, with commit: kernel32: Update sr-Latn locale definition. 3b3dfda59951b0f42e297f2b9a31ded04a98d4b4 that doesn't seem right to me, srl.nls should stand for Serbian Latin I assume, srm.nls probably stands for m as Montenegro. So, if I'm correct, addition of srm.nls with commit: kernel32: Add sr-Latn-RS locale definition. db2666e9d20f80968ff6b4b0ea1deae20c3c368b should have changes made to srl.lns. I don't know how this is solved for English for example, since neutral should have a specific code not related to any country where language is used (e.g. US, UK, CA etc.), and the code it seems to be sr_CS as it is changed for sr_RS latin locale, where it shouldn't be changed I assume. -- 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 8

[Bug 43995] New: Uplay - Assassin's Creed 4 Black Flag won't start
by wine-bugs＠winehq.org 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=43995 Bug ID: 43995 Summary: Uplay - Assassin's Creed 4 Black Flag won't start Product: Wine-staging Version: 2.20 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: mi.bar.2001(a)gmail.com CC: erich.e.hoover(a)wine-staging.com, michael(a)fds-team.de, sebastian(a)fds-team.de Distribution: --- Created attachment 59652 --> https://bugs.winehq.org/attachment.cgi?id=59652 Terminal log from `wine AC4BFSP.exe` After clicking the "Play" button in Ubisoft Game Launcher (or launching `wine AC4BFSP.exe`), a window pops up informing, that the game is starting, but it disappears and the game doesn't launch. I don't have the Steam version, just the Uplay one. I've set Windows version to XP, wineprefix is 32-bit. Wine version will probably be visible when I post this, but I'm going to mention it just to be safe. Using Wine 2.20 staging -- 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.

2 38

[Bug 51181] New: d3d10core:d3d10core fails systematically on AMD GPUs
by WineHQ Bugzilla 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=51181 Bug ID: 51181 Summary: d3d10core:d3d10core fails systematically on AMD GPUs Product: Wine Version: 6.8 Hardware: x86-64 OS: Windows Status: NEW Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com test_instanced_draw() fails systematically in d3d10core:d3d10core on AMD GPUs (and sometimes there is a crash that follows): d3d10core.c:9275: Test failed: Got 0xfff0f010, expected 0xff80f010 at (160, 0), sub-resource 0. d3d10core.c:9275: Test failed: Got 0xfff0f040, expected 0xff80f040 at (240, 0), sub-resource 0. d3d10core.c:9275: Test failed: Got 0xffaaaacc, expected 0xffbbaacc at (480, 0), sub-resource 0. d3d10core.c:9275: Test failed: Got 0xffaaaa90, expected 0xffbbaa90 at (560, 0), sub-resource 0. Notice how this failure does not happen on the machines that have other GPUs such as cw-gtx560 or QEmu VMs, and how this does not depend on the Windows version or Radeon driver version for that matter: https://test.winehq.org/data/patterns.html#d3d10core:d3d10core So it looks like the test expects fails to account for the Radeon driver results for some reason. The commit that introduced this test is: commit fcd549345d96109179b125d31baf9e9073ecf643 Author: Józef Kucia <jkucia(a)codeweavers.com> AuthorDate: Mon Nov 6 10:55:20 2017 +0100 Commit: Alexandre Julliard <julliard(a)winehq.org> CommitDate: Mon Nov 6 19:37:09 2017 +0100 d3d10core/tests: Add test for SV_InstanceID. Signed-off-by: Józef Kucia <jkucia(a)codeweavers.com> Signed-off-by: Henri Verbeet <hverbeet(a)codeweavers.com> Signed-off-by: Alexandre Julliard <julliard(a)winehq.org> -- 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 8

[Bug 52559] New: kernel32:resource times out since 2021-12-29 at two win10 testbot systems.
by WineHQ Bugzilla 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=52559 Bug ID: 52559 Summary: kernel32:resource times out since 2021-12-29 at two win10 testbot systems. Product: Wine Version: 7.2 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: kernel32 Assignee: wine-bugs(a)winehq.org Reporter: bernhardu(a)mailbox.org Distribution: --- I wondered because the windows test patterns showed timeouts for the kernel32:resource [1] since some days. I tried a run with verbose timestamps and got [2]. There it shows gaps of ~6 seconds, and it looks like the delay originates from an execution of EndUpdateResourceA [3]. Unfortunately there are a lot calls for this function and therefore the test times out. Searching the net led me to the autohotkey forums where this function was associated with windows defender. Is windows defender active at these VMs? [1] https://test.winehq.org/data/tests/kernel32:resource.html [2] https://testbot.winehq.org/JobDetails.pl?Key=108231&f201=exe32.report#k201 [3] https://source.winehq.org/git/wine.git/blob/781277de62822c2bac23dc892a99665… -- 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 52079] New: oleacc:main crashes randomly on Windows 10 1709+
by WineHQ Bugzilla 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=52079 Bug ID: 52079 Summary: oleacc:main crashes randomly on Windows 10 1709+ Product: Wine Version: unspecified Hardware: x86-64 OS: Windows Status: NEW Severity: normal Priority: P2 Component: oleacc Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com oleacc:main crashes randomly on Windows 10 1709+ in both 32- and 64-bit: main.c:1041: Test failed: V_VT(&v) = 0 main.c:1048: Test failed: V_VT(&v) = 0 main.c:1049: Test failed: V_DISPATCH(&v) = 0000000000000000 main.c:1049: this is the last test seen before the exception 1050:main: unhandled exception c0000005 at 0000000000401597 oleacc:main:1050 done (-1073741819) in 0s Test failed: crash (c0000005) https://test.winehq.org/data/patterns.html#oleacc:main Unsurprisingly a bisect shows that the tests started failing with the commit that introduced the test: commit 1a3db363c6b4065af532a8db1e6309be753b1781 Author: Connor McAdams <cmcadams(a)codeweavers.com> Date: Mon Sep 20 18:03:31 2021 +0200 oleacc: Add Client_get_accFocus tests. Signed-off-by: Connor McAdams <cmcadams(a)codeweavers.com> Signed-off-by: Piotr Caban <piotr(a)codeweavers.com> Signed-off-by: Alexandre Julliard <julliard(a)winehq.org> At the time the crash happened after line 1037 which points to line 1039 as a likely culprit: 1032 /* Set focus to each child window. */ 1033 SetFocus(btn); 1034 hr = IAccessible_get_accFocus(acc, &v); 1035 ok(hr == S_OK, "hr %#x\n", hr); 1036 ok(V_VT(&v) == VT_DISPATCH, "V_VT(&v) = %d\n", V_VT(&v)); 1037 ok(V_DISPATCH(&v) != NULL, "V_DISPATCH(&v) = %p\n", V_DISPATCH(&v)); 1038 1039 hr = IDispatch_QueryInterface(V_DISPATCH(&v), &IID_IOleWindow, (void**)&ow); 1040 ok(hr == S_OK, "got %x\n", hr); That call has been moved to _check_acc_hwnd() and so the crash now happens in the first QueryInterface() call in that function: hr = IUnknown_QueryInterface(unk, &IID_IOleWindow, (void**)&ow); called from check_acc_hwnd((IUnknown*)V_DISPATCH(&v), btn); -- 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 51407] New: kernel32:time's test_GetCalendarInfo() fails in Hindi
by WineHQ Bugzilla 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=51407 Bug ID: 51407 Summary: kernel32:time's test_GetCalendarInfo() fails in Hindi Product: Wine Version: 6.10 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: kernel32 Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- Created attachment 70263 --> https://bugs.winehq.org/attachment.cgi?id=70263 Broken fix and traces Three GetCalendarInfo() checks fail in Hindi: https://test.winehq.org/data/patterns.html#kernel32:time time.c:728: Test failed: GetCalendarInfoA failed err 87 time.c:731: Test failed: GetCalendarInfoA failed err 87 time.c:738: Test failed: got 0, expected 7 The immediate cause for the failure is that Hindi is one of the few languages for which NLS_IsUnicodeOnlyLcid() returns true. In turn this causes the ANSI GetCalendarInfoA() to return ERROR_INVALID_PARAMETER. The last failure just reflects the fact that the two calls to GetCalendarInfoA() should not have failed. Only Hindi fails and only in Wine. But before declaring this a Wine bug we need to look at Windows. * For some reason probably related to SetWinLocale [1] the system locale of w10pro64_hi is still English. * This causes both the ANSI and Unicode GetCalendarInfo() to return English names (Monday). * Manually setting the system locale to Hindi is possible on Windows 10 2004. What's interesting is the disabled and checked-by-default checkbox in that dialog: [X] Beta: Use Unicode UTF-8 for worldwide language support * And indeed doing that and running the test again results in GetCalendarInfoA() returning UTF-8 strings on Windows! time.c:730: ret=19 sdayname1="\xe0\xa4\xb8\xe0\xa5\x8b\xe0\xa4\xae\xe0\xa4\xb5\xe0\xa4\xbe\xe0\xa4\xb0" सोमवार time.c:735: ret2=19 sdayname1="\xe0\xa4\xb8\xe0\xa5\x8b\xe0\xa4\xae\xe0\xa4\xb5\xe0\xa4\xbe\xe0\xa4\xb0" सोमवार time.c:742: ret2=19 sdayname1=L"\0938\094b\092e\0935\093e\0930" * The beta moniker strongly suggests that this is a new option. Indeed selecting Hindi as the system locale was not possible in Windows 8.1 for instance. * This is probably also why SetWinLocale has trouble setting the system locale to Hindi. So the Windows behavior can be summarized as such: * Return localized values for non-Unicode-only locales. * In older Windows versions setting the system locale to a Unicode-only locale is impossible. * In newer Windows versions setting the system locale to a Unicode-only locale is possible and then it uses UTF-8. Commenting out the Hindi line in NLS_IsUnicodeOnlyLcid() fixes the failures in Wine (see attached patch). But this fix is wrong because it causes GetCalendarInfoA() to return question marks instead of intelligible strings. (it also impacts many APIs but that's probably ok based on the Windows behavior) So Wine should do one of the following: 1. Stick to a 'safe' system locale when $LANG is set to a Unicode-only locale. Typically this would be English. 2. Or use UTF-8 when the system locale is a Unicode-only locale. Option 2 would actually make a lot of sense on Unix systems but it is also more likely to break Windows applications. So option 1 may be the safer (and simpler?) option (for now). [1] https://source.winehq.org/git/tools.git/blob/HEAD:/testbot/bin/SetWinLocale… -- 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 7

[Bug 43208] New: Assassin's Creed IV - Black Flag Hangs tightly
by wine-bugs＠winehq.org 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=43208 Bug ID: 43208 Summary: Assassin's Creed IV - Black Flag Hangs tightly Product: Wine-staging Version: 2.10 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: tall89(a)mail.ru CC: erich.e.hoover(a)wine-staging.com, michael(a)fds-team.de, sebastian(a)fds-team.de Distribution: --- When you switch to the Big Inagua location, the game hangs tight, you have to do Cntrl + Atl + Delete to remove the task, then the game closes, then you receive an error message, the program ACBF.exe has a serious error, the application will be closed, we apologize for the inconvenience WINEDEBUG=fixme-all,err+loaddll,err+dll,err+file,err+reg ------ Program AC4BFSP.exe ------ err:winsock:WSAIoctl -> SIO_ADDRESS_LIST_CHANGE request failed with status 0x2733 err:winedevice:async_create_driver failed to create driver L"ps6ajtsb": c0000135 err:eventlog:ReportEventW L"mDNSCoreReceiveResponse: Received from 192.168.0.102:5353 16 aspire89-P55A-UD4.local. AAAA FE80:0000:0000:0000:0A12:F0CA:1A84:437A" err:eventlog:ReportEventW L"mDNSCoreReceiveResponse: ProbeCount 2; will deregister 4 aspire89-P55A-UD4.local. Addr 192.168.0.102" err:eventlog:ReportEventW L"Local Hostname aspire89-P55A-UD4.local already in use; will try aspire89-P55A-UD4-2.local instead" err:eventlog:ReportEventW L"mDNSCoreReceiveResponse: Received from 192.168.0.102:5353 25 102.0.168.192.in-addr.arpa. PTR aspire89-P55A-UD4.local." err:eventlog:ReportEventW L"mDNSCoreReceiveResponse: Unexpected conflict discarding 27 102.0.168.192.in-addr.arpa. PTR aspire89-P55A-UD4-2.local." err:ole:CoInitializeEx Attempt to change threading model of this apartment from multi-threaded to apartment threaded err:d3d:wined3d_debug_callback 0x721acb8: "GL_INVALID_OPERATION error generated. <program> has not been linked, or is not a program object.". err:d3d:wined3d_debug_callback 0x721acb8: "GL_INVALID_OPERATION error generated. <program> object is not successfully linked, or is not a program object.". err:d3d:wined3d_debug_callback 0x721acb8: "GL_INVALID_OPERATION error generated. <program> has not been linked, or is not a program object.". err:d3d:wined3d_debug_callback 0x721acb8: "GL_INVALID_OPERATION error generated. <program> object is not successfully linked, or is not a program object.". err:d3d:wined3d_debug_callback 0x721acb8: "GL_INVALID_OPERATION error generated. <program> has not been linked, or is not a program object.". err:d3d:wined3d_debug_callback 0x721acb8: "GL_INVALID_OPERATION error generated. <program> object is not successfully linked, or is not a program object.". wine: Unhandled page fault on read access to 0x00000063 at address 0x8cfb7f (thread 0047), starting debugger... wine: Unhandled page fault at address 0x7bc5d590 (thread 0012), starting debugger... err:seh:start_debugger Couldn't start debugger ("winedbg --auto 14 168") (1115) Read the Wine Developers Guide on how to set up winedbg or another debugger -- 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.

2 14

[Bug 51460] New: oleaut32:vartest has a todo because Wine uses an outdated currency symbol for Swiss Francs
by WineHQ Bugzilla 06 May '22

06 May '22

https://bugs.winehq.org/show_bug.cgi?id=51460 Bug ID: 51460 Summary: oleaut32:vartest has a todo because Wine uses an outdated currency symbol for Swiss Francs Product: Wine Version: 6.10 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: oleaut32 Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- Wine uses uses "SFr." as the currency symbol for Swiss francs instead of "CHF". According to Wikipedia: "The use of SFr. for Swiss Franc and fr.sv. are outdated" and are not recognized as correct by the "Secrétariat d'État à l'économie". https://en.wikipedia.org/wiki/Swiss_franc This forces the use of a todo in oleaut32:vartest: lcid = MAKELCID(MAKELANGID(LANG_FRENCH,SUBLANG_FRENCH_SWISS),SORT_DEFAULT); WCONVERT(L"3CHF", NUMPRS_CURRENCY|NUMPRS_USE_ALL); todo_wine EXPECT(1,NUMPRS_CURRENCY|NUMPRS_USE_ALL,NUMPRS_CURRENCY,4,0,0); And it may cause oleaut32:varformat to fail when run in a Swiss locale. The "SFr." currency symbol comes from nls/(de|fr|it)s.nls which are generated from Unicode tables by tools/make_unicode so it's not clear how to fix them. -- 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