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

[Bug 53298] New: Check boxes don't scale properly in Qt application with light theme
by WineHQ Bugzilla 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=53298 Bug ID: 53298 Summary: Check boxes don't scale properly in Qt application with light theme Product: Wine Version: 7.12 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: uxtheme Assignee: wine-bugs(a)winehq.org Reporter: rpisl(a)seznam.cz Distribution: --- Created attachment 72666 --> https://bugs.winehq.org/attachment.cgi?id=72666 Check box glitch Attached screenshots shows two UI glitches in Qt application (tested on Wireshark and verified on another one) with theming enabled. 1. Check boxes (and also radio buttons) change size on mouse over when DPI is higher than 96 (tested with DPI set to 192). 2. Disabled check box is not visually distinguishable from enabled one. Normally, disabled checkbox has gray background if theming is disabled. -- 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 53341] New: uxtheme:system depends on GetDpiForSystem() which is missing on Windows 10 1507 and earlier
by WineHQ Bugzilla 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=53341 Bug ID: 53341 Summary: uxtheme:system depends on GetDpiForSystem() which is missing on Windows 10 1507 and earlier Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: uxtheme Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- uxtheme:system depends on GetDpiForSystem() which is missing on Windows 10 1507 and earlier: uxtheme:system start - system.c:0: Tests skipped: missing entry point (details below) | uxtheme_test.exe - Entry Point Not Found | The procedure entry point GetDpiForSystem could not be located in the dynamic link library C:\Users\Public\Documents\uxtheme_test.exe. https://test.winehq.org/data/patterns.html#uxtheme:system The dependency was introduced in the commit below: commit e2daffda25bf79ec3313349b85a33e29d202b1de Author: Zhiyi Zhang <zzhang(a)codeweavers.com> AuthorDate: Mon Jul 4 22:08:25 2022 +0800 uxtheme/tests: Add more GetThemePartSize() tests in different DPI. To test that GetThemePartSize() should always report theme part size in system DPI if the theme handle is open without a valid window handle. Previous tests succeeded because the cur rent DPI happens to be 96. Tested manually on Windows 10 for 96, 120, 144, 168, 192 and 216 DPI. Signed-off-by: Zhiyi Zhang <zzhang(a)codeweavers.com> -- 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 51462] New: SnowRunner crashes when changing settings
by WineHQ Bugzilla 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=51462 Bug ID: 51462 Summary: SnowRunner crashes when changing settings Product: Wine Version: 6.12 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: andrey.goosev(a)gmail.com Distribution: --- 0274:err:ole:StdMarshalImpl_MarshalInterface Failed to create ifstub, hr 0x80004002 0274:err:ole:CoMarshalInterface Failed to marshal the interface {9c6b4cb0-23f8-49cc-a3ed-45a55000a6d2}, hr 0x80004002 04a4:fixme:ole:NdrClearOutParameters (000000000011D808,00007FC8EFEAA0A0,000000000011DAB8): stub 04a4:fixme:ole:CoCreateInstanceEx no instance created for interface {9c6b4cb0-23f8-49cc-a3ed-45a55000a6d2} of class {a65b8071-3bfe-4213-9a5b-491da4461ca7}, hr 0x80004002. Changing as example Antialiasing, Lightshafts, Mirrors settings leads to crash with a pop-up window DirectX Diagnostic Tool Usage: dxdiag [/whql:off | /whql:on] [/t filename | /x filename] wine-6.12-194-g2ffcf440218 -- 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 2

[Bug 52391] New: No sound effects in Hyperdimension Neptunia Re;Birth 1 since import of FAudio
by WineHQ Bugzilla 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=52391 Bug ID: 52391 Summary: No sound effects in Hyperdimension Neptunia Re;Birth 1 since import of FAudio Product: Wine Version: 7.0-rc5 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: xactengine Assignee: wine-bugs(a)winehq.org Reporter: j-r(a)online.de Distribution: --- Since the import of FAudio Hyperdimension Neptunia Re;Birth 1 (GOG version) triggers the following FAudio assertion since the import of FAUDIO in commit 6b45777121fcd50d9b7d5b649c6710063b549892 already on the title screen while highlighting or selecting a menu entry Assertion failed: (voice->src.wmadec != NULL && (pBufferWMA != NULL || (voice->src.format->wFormatTag == FAUDIO_FORMAT_XMAUDIO2)) || (voice->src.wmadec == NULL && (pBufferWMA == NULL && voice->src.format->wFormatTag != FAUDIO_FORMAT_XMAUDIO2) ), file libs/faudio/src/FAudio.c, line 2456 Note that this assertion was changed in the import of FAudio 21.12 in da89825f98e4f0cb3875735345a455512b36b665 (erronously declared as import of 21.11), without effect on this game (contrary to what https://bugs.winehq.org/show_bug.cgi?id=50757#c2 claims). There is still no sound in 7.0-rc5 (and reverting babaf916ab12c7a9128409f717878919ccc5826b which hides the assertion shows it still triggering the extended assertion condition). Sound works before 6b45777121fcd50d9b7d5b649c6710063b549892 (also see https://bugs.winehq.org/show_bug.cgi?id=50416 for caveats). Unfortunately 6b45777121fcd50d9b7d5b649c6710063b549892 suffers from frequent crashing already on the title screen (haven't tested ingame; might be the same as https://bugs.winehq.org/show_bug.cgi?id=50757; despite it claiming to be fixed with exactly that commit; afaict nobody ever confirmed it being fixed) . Fortunately this crashing seems to be fixed or at least much reduced with 7.0-rc5 on my (very old and slow) system. Sound works fine in 7.0-rc5 after winetricks xact and manually setting xactengine3_7 to native. Not sure about the relation to this bug, but 7.0-rc5 is showing the following error whenever the game attempts to play a sound 00c8:err:ole:com_get_class_object class {2eeb4adf-4578-4d10-bca7-bb955f56320a} not registered 00c8:err:ole:com_get_class_object no class object {2eeb4adf-4578-4d10-bca7-bb955f56320a} could be created for context 0x1 which I think wasn't there before the regression. -- 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 20

[Bug 53344] New: Legacy TLS applications: GnuTLS priority string overrides are ignored in schan_create_session() (patch included)
by WineHQ Bugzilla 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=53344 Bug ID: 53344 Summary: Legacy TLS applications: GnuTLS priority string overrides are ignored in schan_create_session() (patch included) Product: Wine Version: 7.12 Hardware: x86-64 URL: https://support.hpe.com/connect/s/softwaredetails?lang uage=en_US&softwareId=MTX_bc8e3ffa59904ec3b505d9964d OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: secur32 Assignee: wine-bugs(a)winehq.org Reporter: cflwxa(a)knowledgemd.com Distribution: Debian Created attachment 72727 --> https://bugs.winehq.org/attachment.cgi?id=72727 trace+secur32,warn+secur32 and GNUTLS_DEBUG_LEVEL=11 I use Wine devel 7.12 from the winehq repos and winetricks 20220411 from github. Context: I am trying to run the "HPE Lights-Out Standalone Remote Console" (HPLOCONS.exe, https://support.hpe.com/connect/s/softwaredetails?language=en_US&softwareId…) to connect to an iLO 3 interface of an HP ProLiant server. iLO 3 only supports legacy ciphers (3DES, RC4). Because the Wine schannel implementation uses GnuTLS, I added an override config like so: sudo mkdir -p /etc/gnutls sudo tee /etc/gnutls/config_hplocons <<EOF [overrides] default-priority-string = NORMAL:+3DES-CBC:+ARCFOUR-128 EOF Then I set up a wine prefix like so: env WINEARCH=win32 WINEPREFIX=~/.local/share/wineprefixes/hplocons wineboot --init env WINEARCH=win32 WINEPREFIX=~/.local/share/wineprefixes/hplocons winetricks -q dotnet471 d3dcompiler_47 corefonts fontfix winver= wget https://downloads.hpe.com/pub/softlib2/software1/pubsw-windows/p390407056/v… env WINEARCH=win32 WINEPREFIX=~/.local/share/wineprefixes/hplocons wine Setup.exe And successfully run HPLOCONS: env WINEARCH=win32 WINEPREFIX=~/.local/share/wineprefixes/hplocons WINEDEBUG=trace+secur32,warn+secur32 GNUTLS_SYSTEM_PRIORITY_FILE=/etc/gnutls/config_hplocons GNUTLS_DEBUG_LEVEL=11 LC_ALL=C wine start 'C:\Program Files\Hewlett Packard Enterprise\HPE iLO Integrated Remote Console\HPLOCONS.exe' When trying to connect to iLO, I get "The request was aborted: Could not create SSL/TLS secure channel." The trace logs and gnutls debug logs (attached) show that schan_create_session() in dlls\secur32\schannel_gnutls.c ignores my priority string override and instead, uses "NORMAL" without TLS 1.3 ciphersuites. A look at the source code confirms that schan_create_session() does indeed not incorporate the system defaults. Also note that process_attach() sets GNUTLS_SYSTEM_PRIORITY_FILE to "/dev/null" if it was unset, which has the consequence that everything in the default /etc/gnutls/config file is always ignored. I further confirmed that this is a Wine issue by running gnutls-cli: # Succeeds env GNUTLS_SYSTEM_PRIORITY_FILE=/etc/gnutls/config_hplocons gnutls-cli --insecure <iLO-IP> # Fails with "Received alert [40]: Handshake failed" gnutls-cli --insecure <iLO-IP> To be able to run legacy applications, I think Wine should respect GnuTLS overrides. I attached an untested patch proposal. Note that this patch does not change behaviour if the default priority string is "NORMAL". Also note that Wine would require gnutls 3.6.3 due to the use of gnutls_set_default_priority_append(). Perhaps check_supported_protocols() should also be patched in a similar way. -- 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 14

[Bug 53231] New: d2d1:d2d1 - test_shared_bitmap() crashes on the TestBot VMs
by WineHQ Bugzilla 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=53231 Bug ID: 53231 Summary: d2d1:d2d1 - test_shared_bitmap() crashes on the TestBot VMs Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- d2d1:d2d1 - test_shared_bitmap() crashes on the TestBot VMs: d2d1.c:5179: Test marked todo: 6: Got unexpected hr 0. Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x6ac0f8ef). [...] Backtrace: =>0 0x6ac0f8ef d2d_device_context_Clear+0x7f(iface=<internal error>, colour=<internal error>) [Z:\home\winetest\tools\testbot\var\wine\dlls\d2d1\device.c:1647] in d2d1 (0x0066fc78) 1 0x00414d75 in d2d1_test (+0x14d75) (0x0066fd68) 2 0x00434613 in d2d1_test (+0x34613) (0x0066fdd8) 3 0x00434983 in d2d1_test (+0x34983) (0x0066fe38) 4 0x0043618d in d2d1_test (+0x3618d) (0x0066fee8) 5 0x00435fbf in d2d1_test (+0x35fbf) (0x0066ff30) 6 0x7b62a170 in kernel32 (+0x2a170) (0x0066ff48) 7 0x7bc5a937 in ntdll (+0x5a937) (0x0066ff5c) 8 0x7bc5aff0 RtlCreateUserThread(entry=00435F40, arg=7FFD1000) [Z:\home\winetest\tools\testbot\var\wine\dlls\ntdll\thread.c:261] in ntdll (0x0066ffec) 0x6ac0f8ef d2d_device_context_Clear+0x7f [Z:\home\winetest\tools\testbot\var\wine\dlls\d2d1\device.c:1647] in d2d1: fsts 0x0(%eax) 1647 vs_cb_data->transform_geometry._11 = 1.0f; https://test.winehq.org/data/patterns.html#d2d1:d2d1 The crash still happens even if all the tests are commented out except test_shared_bitmap(). Furthermore adding traces shows that it happens in either of the two locations below: https://gitlab.winehq.org/wine/wine/-/blob/f6f66661b3fabdcd6cf2f4999b3029e7… /* DXGI surface render target and WIC bitmap render target, same factory. */ hr = ID2D1Factory_CreateWicBitmapRenderTarget(factory1, wic_bitmap2, &desc, &rt2); // crashed here or https://gitlab.winehq.org/wine/wine/-/blob/f6f66661b3fabdcd6cf2f4999b3029e7… ID2D1GdiInteropRenderTarget_Release(interop); hr = ID2D1Factory_CreateWicBitmapRenderTarget(factory2, wic_bitmap2, &desc, &rt2); // crashed here The crashes only happen with 32-bit code. d2d1:d2d1 also crashes on my box with an Intel GPU and the crash happens in the same two locations but the backtrace is different. -- 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 53157] New: --without-mingw builds on Mac can't unwind exceptions
by WineHQ Bugzilla 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=53157 Bug ID: 53157 Summary: --without-mingw builds on Mac can't unwind exceptions Product: Wine Version: 7.10 Hardware: x86-64 OS: Mac OS X Status: NEW Severity: normal Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: stefan(a)codeweavers.com I am trying to get a non-mingw build on MacOS to run my games to get debug symbols for Wine DLLs in Instruments for performance profiling. One of the issues I ran into is that SEH exceptions thrown in the PE side of our code can't be handled. Wine runs into further nested exceptions trying to unwind the stack and crashes. This is a regression introduced by the following commit: 264bd4fc84020f217807f1e59d3bfa9302e5110c is the first bad commit commit 264bd4fc84020f217807f1e59d3bfa9302e5110c Author: Martin Storsjo <martin(a)martin.st> Date: Fri May 29 13:51:04 2020 +0300 wine/asm.h: Don't define __ASM_CFI when using clang as cross compiler. Reverting this commit on top of Wine 7.10 fixes this particular problem. Getting a Wine build like this is a bit tricky. clang does not support -mabi=ms, and gcc can't build winemac.drv because of Apple-isms in the system headers that Apple never upstreamed to FSF gcc. (But they did have it in their own gcc before they switched to clang). To get a non-mingw gcc'ish Wine I do roughly the following: 1) Use wine-7.10, not master, there is yet another regression since then 2) Install gcc-11 with homebrew 3) ~/wine/configure --enable-win64 --without-x --without-mingw --disable-tests CC=gcc-11 4) Build the thing 5) Edit Makefile, replace CC = gcc-11 with CC = clang 6) cd into dlls/winemac.drv, run make 7) optionally do the same in dlls/winecoreaudio.drv 8) optionally undo the Makefile changes if you plan to recompile e.g. ntdll The game in question is Rocket League; That adds extra troubles with Steam (32 bit, so needs Mojave). I think the problem is independent of Rocket League though. The exception causing the crash is OutputDebugStringA. I don't think the problem is specific to this game though. -- 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 43167] New: Witcher 3 will not start
by wine-bugs＠winehq.org 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=43167 Bug ID: 43167 Summary: Witcher 3 will not start Product: Wine Version: 2.10 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: kallekadakas(a)gmail.com Distribution: --- Created attachment 58423 --> https://bugs.winehq.org/attachment.cgi?id=58423 Output.txt I apologize for the vague title, but there is very little info on why it fails. Terminal output is attached to the case. When clicking play on TW3 in Steam, it goes to "running" and then fails with the output attached. Arch64 i7-4790k gtx1070 Wine2.10 with dx11>opengl4.5 registry fix applied. -- 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 12

[Bug 51129] New: Function name conflict (ldap_connect) with OpenLDAP 2.5
by WineHQ Bugzilla 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=51129 Bug ID: 51129 Summary: Function name conflict (ldap_connect) with OpenLDAP 2.5 Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: wldap32 Assignee: wine-bugs(a)winehq.org Reporter: sergio.durigan(a)canonical.com Distribution: --- Hello, Wine installs the "/usr/include/wine/wine/windows/winldap.h" file, which exports the ldap_connect function: ULONG CDECL ldap_connect(LDAP*,struct l_timeval*); With the recent OpenLDAP 2.5 release, we now have an "ldap_connection" function being exported by it as well (with a different prototype): https://git.openldap.org/openldap/openldap/-/blame/OPENLDAP_REL_ENG_2_5_4/i… This will need to be fixed in order to prevent linking problems with OpenLDAP. -- 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 22

[Bug 53104] New: Wireshark can't start capturing packets
by WineHQ Bugzilla 15 Jul '22

15 Jul '22

https://bugs.winehq.org/show_bug.cgi?id=53104 Bug ID: 53104 Summary: Wireshark can't start capturing packets Product: Wine Version: 7.10 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: wpcap Assignee: wine-bugs(a)winehq.org Reporter: rpisl(a)seznam.cz Distribution: --- Created attachment 72539 --> https://bugs.winehq.org/attachment.cgi?id=72539 Patch for wpcap While testing bug 53066 I just wanted to find out what would be needed to make Wireshark working in Wine. There are two obvious problems: 1. pcap_create() doesn't translate interface identifier to the real name 2. pcap_dispatch() is not implemented I fixed these problems, however, packet window still remains empty after capture is started while the temporary capture file is filled with captured packets. If this temporary file is later opened, Wireshark shows all packets captured. Attached is the patch if someone wants to push this further. "setcap cap_net_raw,cap_net_admin=eip wine64-preloader" is needed to give Wine permissions to capture packets. -- 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