wine-devel

wine-devel@winehq.org

2 participants
84469 discussions

Wine staging 8.5 release
by Alistair Leslie-Hughes 01 Apr '23

01 Apr '23

Binary packages for various distributions will be available from: https://www.winehq.org/download Summary since last release * Rebased to current wine 8.5 (537 patches are applied to wine vanilla) The patchinstall.sh script is no longer available and users should use the staging/patchinstall.py instead. Upstreamed (Either directly from staging or fixed with a similar patch). * dinput: Allow mapping of controls based of Genre type. Added: * MR 2449: Allow Diable IV to run without reporting errors. * [53826] Attempting to install / update battle.net fails with Qt error. Updated: * mfplat-streaming-support Where can you help * Run Steam/Battle.net/GOG/UPlay/Epic * Test your favorite game. * Test your favorite applications. * Improve staging patches and get them accepted upstream. * Suggest patches to be included in staging. As always, if you find a bug, please report it via https://bugs.winehq.org Best Regards Alistair.

1 0

Should we stop linking to FAudio for our XAudio implementation?
by Zebediah Figura 29 Mar '23

29 Mar '23

Hi all, This may seem something of a controversial question, but I am concerned that FAudio is not a good fit for Wine, mainly because of its licensing considerations. Recently I found a bug in FAudio wherein it does not properly handle unaligned buffers. While in the process of drafting a fix, I was stymied multiple times by its rather broken resampler, and when I sought to try to fix that, I found that it was a rather difficult endeavour, which led me to propose replacing it with an external library such as GStreamer. Unfortunately FAudio cannot, counterintuitively, link to LGPL code, because it needs to be able to run on embedded platforms, which (for reasons I do not understand) cannot ship any LGPL code. While I can to some degree respect the decision of FAudio to target platforms that forbid copyleft licensing, I do not think this aligns well with the interests of the Wine project—a project which specifically chose copyleft licensing. More personally, I do not want to have to fix the FAudio resampler or implement another resampler with a non-copyleft license (and I am not even particularly sure that I can, given my familiarity with resamplers in LGPL projects such as GStreamer and Wine). Besides this, it is probably worth noting that we do not seem to gain much from continuing to rely on FAudio. There would be the usual benefit of relying on upstream development, but development on FAudio seems very scarce, especially when restricted to parts relevant to Wine, and much of that is from Wine contributors anyway. (And, of course, due to our bundling practices, there is hardly the usual benefit of size or modularity in linking to external libraries.) Of course we would be well served by simply importing the current version of FAudio; I do not mean to propose that we reimplement XAudio entirely, but rather begin to modify it in ways that would not be acceptable upstream. This would mean replacing the resampler, as well as perhaps also the mixer, and probably with the ones from GStreamer, which is already a dependency of ours, and which has a very featureful resampler and mixer, and a much higher rate of development overall. Do others in the Wine community have thoughts or opinions on this matter? ἔρρωσθε, Zeb

4 8

TestBot News: Missing dlls and VM updates
by Francois Gouget 19 Mar '23

19 Mar '23

I reviewed the tests that are skipped because the dll is missing on Windows and I found 4 tests that are never run on Windows which makes them a bit pointless: * d3dcompiler_46 I can install version 43 and 47 from dxwebsetup.exe without trouble. But d3dcompiler_46.dll? No idea. * d3drm Is this a deprecated Direct 3D 8 dll? dxwebsetup.exe does not install it. * msvcr70 As far as I know this dll is part of the Visual Studio 7.0 runtime. But that runtime is nowhere to be found. * msvcrtd The test look like they succeed on test.winehq.org but in fact all they do is skip the tests: debug.c:45: Tests skipped: LoadLibraryA failed to load msvcrtd.dll with GLE=126 (where 126 == ERROR_MOD_NOT_FOUND) That's the debug version of the Visual C++ C library. As such it's not shipped by the runtime redistributables. Also, which C runtime does it correspond to? (iirc it was already there in the 90s) Note: I'd rather not download these dlls from the many 'not.trojan.horse.dlls.trust.me.com' websites. VM updates: * debiant The gst-plugins-base1.0-dev maintainer intentionally removed multi-arch support in version 1.20.3-2. As a result GStreamer support went missing in 32-bit builds when Wine's detection code was tweaked recently, which in turn caused a lot of new failures (bug 53977). So I hacked multi-arch support back in and the 32-bit tests are back to normal. Another point is that updating the X11 / Mesa packages causes the X server to crash during the Direct 3D tests. So for now that VM is frozen in time. That's all worrying for the future Debian 12. * w11pro64 I added a bunch of Visual Studio C runtime dlls, including msvcr71 which I got through the .Net 1.1 framework. I also added a number of Direct 3D dlls that don't ship with Windows by default. * w1064v1709 It seems this snapshot was not entirely idle so I retook it after Windows Update finished installing the already downloaded updates (see bug 54158). This does not seem to have made much of a difference :-( (particularly for bug 53227) * w1064v1909 Windows Defender was not disabled and was intefering with the tests from time to time. That's because 1909 is the first Windows version where Defender cannot be disabled and where I have to exclude the WineTest folder instead. This is fixed now. * w1064v2004 This snapshot also had some trouble with Windows Update not being idle so I retook it. This does not seem to have made much of a difference either :-( -- Francois Gouget <fgouget(a)codeweavers.com>

4 12

Wine staging 8.4 release
by Alistair Leslie-Hughes 18 Mar '23

18 Mar '23

Binary packages for various distributions will be available from: https://www.winehq.org/download Summary since last release * Rebased to current wine 8.4 (532 patches are applied to wine vanilla) The patchinstall.sh script is no longer available and users should use the staging/patchinstall.py instead. Upstreamed (Either directly from staging or fixed with a similar patch). * None Added: * None Updated: * None Where can you help * Run Steam/Battle.net/GOG/UPlay/Epic * Test your favorite game. * Test your favorite applications. * Improve staging patches and get them accepted upstream. * Suggest patches to be included in staging. As always, if you find a bug, please report it via https://bugs.winehq.org Best Regards Alistair.

1 0

wine-7.0.2-rc1 Available
by Michael Stefaniuc 16 Mar '23

16 Mar '23

Hello, I have pushed wine-7.0.2-rc1 (f25a27c2e57c9f8c061746a813ace405fbd39073) to https://github.com/mstefani/wine-stable/tree/oldstable https://people.redhat.com/mstefani/wine/stable/ And created the draft merge request https://gitlab.winehq.org/wine/wine/-/merge_requests/2402 Nothing fancy so here we go: What's new in this release: - Various bug fixes ---------------------------------------------------------------- Bugs fixed in 7.0.2 (total 43): - #36711 Sid Meier's Civilization IV crashes while loading saved game (GOG.com complete, Colonization, Warlords, Beyond the Sword)('winetricks -q msxml3' is a workaround) - #39346 Lego Rock Raiders crashes on startup (d3drm_device_init doesn't handle device version 3 case) - #40444 Find doesn't jump to a found item in WinSCP when editing a file - #42751 Multiple applications crash on unimplemented function msvcr{100,110}.dll.?_Schedule@_StructuredTaskCollection@details@Concurrency@@QAEXPAV_UnrealizedChore@23@@Z (Gas Guzzlers Combat Carnage, Clip Studio Paint, Cube World, Intel ACAT) - #42857 Can't select a fixture in FreeStyler's fixture editor - #44814 cygwin's bsdtar.exe needs a better ntdll.NtQueryEaFile stub - #46822 Edit control in ADL search dialog gets initially not drawn in DC++ 0.868, regression - #47375 Multiple applications crash on unimplemented function msvcr120.dll.?_Schedule@_StructuredTaskCollection@details@Concurrency@@QEAAXPEAV_UnrealizedChore@23@@Z (Photoshop C 2018, CLIP Studio Paint) - #48487 MSYS based Git 2.25 installer reports 'could not decode hex <hexstring>' (needs support for CryptStringToBinaryW CRYPT_STRING_HEX) - #48600 Bloomberg Terminal needs unimplemented function dbghelp.dll.SymSrvGetFileIndexInfo - #49285 PNotesPortable crashes inside WritePrivateProfileStructW - #50948 taskmgr.exe: wrong memory usage unit (GB => MB) - #51788 windowscodecs:wmpformat test_decode() fails randomly - #51842 include/winnt.h:403:50: error: size of unnamed array is negative - #52457 CNG Encryption Failure (BCryptEncrypt) - #52534 ListView: multi select never sends LVN_ODSTATECHANGED - #52545 ACE rights parser should allow octal and decimal formats - #52686 DateAdd not implemented - #52831 Kernel32::GetSystemPowerStatus returns invalid data if /sys/class/power_supply/BAT0 is missing - #52841 Leverless arcade controller SOCD cleaning does not work - #52972 wineusb.sys: add_usb_device erroneously releases critical section on failure path - #52975 VTFEdit fails to start: err:module:LdrInitializeThunk "MSVCP80.dll" failed to initialize, aborting - #53003 XM6 TypeG: some MIDI note-on messages aren't sent to MIDI devices (MIDI running status isn't supported in the MIDI mapper) - #53005 quartz:systemclock Fails after timeGetTime() wraps - #53027 Port Royale 4 crashes with wined3d_texture* - #53032 winedevice.exe segfaults on exit when built with GCC - #53051 Kvaser software crashes on unimplemented function ntoskrnl.exe.ExNotifyCallback - #53123 ListView: Ctrl+Shift+LMouse multiselect on LVS_OWNERDATA listviews erroneously sends LVN_ODSTATECHANGED - #53169 Rich edit control becomes unstable or trips assertion after ITextRange::SetFont is called - #53170 Rich edit control becomes unstable or trips assertion after changing TextFont properties - #53234 nsi:nsi - test_ndis_ifinfo() fails randomly on Windows 10 1507 - #53310 Roblox Fails With HttpSendRequest - #53335 Crash when converting an rich text editor already containing text into a password input - #53455 shell32:recyclebin - test_query_recyclebin() gets unexpected recyclebin size in Wine - #53569 postgresql installer 9.3 needs correct string length returned from fso.GetTempName - #53587 comctl32 GetWindowSubclass crash in Notepad++ (v8.4.3 and 8.4.4 only) - #53614 KeePassXC crashes on unimplemented function msvcp140.dll.?ReportUnhandledError@_ExceptionHolder@details@Concurrency@@AEAAXXZ - #53645 AIMP 3 causes memory leaks - #53706 NtDeviceIoControlFile() is quite slow when called with and invalid handle - #53714 SQLGetInfo(W) doesn't fill StringLength when InfoValue is NULL - #53800 CRYPT_GetBuiltinDecoder Unsupported decoder for lpszStructType 1.3.6.1.4.1.311.2.1.4 - #53821 winbase.h: ReOpenFile declaration missing (and not available in any other header) - #53832 wbemprox: where-clause seems to be case-sensitive ---------------------------------------------------------------- Changes since 7.0.1: Akihiro Sagawa (2): midimap: Handle MIDI running status. mciqtz32: Fix thread handle leak if a second play. Alex Henrie (3): ntoskrnl: Add ExNotifyCallback stub. ntdll: Don't hard-code the battery and AC adapter names on Linux. user32: Plug icon handle leak in MDI_AugmentFrameMenu (Coverity). Alexandre Julliard (4): shell32: Don't return an uninitialized pointer on error in enum_trash_items(). urlmon/tests: Fix race condition in abort test. mountmgr: Don't crash if connecting to dbus fails. gdiplus: Get the compositing mode directly from the graphics object. Alistair Leslie-Hughes (8): d3drm: Support IDirect3D3 when creating device. ntdll: Check return parameter before use. ntdll: RtlFirstFreeAce only return FALSE on error. ntdll/tests: Add RtlFirstFreeAce tests. dsdmo: Add Echo FX Support. dsdmo: Add Compressor FX Support. odbc: SQLGetInfoW allow NULL parameter InfoValue. crypt32: Only report Unimplemented decoder when not found in external dll. Angelo Haller (9): comctl32/tests: Add hold_key and release_key functions. comctl32/tests: Add multi select tests for ownerdata listviews. comctl32/tests: Add change sequence for ownerdata listviews. comctl32/listview: Send one deselect all items notification for LVS_OWNERDATA listviews. comctl32/listview: Move sending LVN_ODSTATECHANGED notifications to a function. comctl32/listview: Send LVN_ODSTATECHANGED only for LVS_OWNERDATA listviews. comctl32/listview: Send LVN_ODSTATECHANGED notification for LVS_OWNERDATA listview on selection changes. comctl32/listview: Don't send LVN_ODSTATECHANGED for empty ranges. comctl32/tests: Add more ownerdata listview tests. Claire Girka (1): winepulse: Fix memory leak in pulse_create_stream in some error cases. Daniel Lehman (1): kernelbase: Pass va_list copy to internal RtlFormatMessage. David Curtiss (1): ws2_32: Allow getsockname after AcceptEx. David Kahurani (2): win32u: Use allocator mechanism to allocate DC attributes. ieframe: Fix memory leak in open_dde_url. Dmitry Timoshkov (2): comctl32/tests: Add iItem and uChanged to LVN_ITEMCHANGING/LVN_ITEMCHANGED context. sechost: Support decimal or octal access rights in ConvertStringSecurityDescriptorToSecurityDescriptor. Eric Pouech (2): wldap32: Fix compilation in 32-bit. winedbg: Fix retrieving integral values. Esme Povirk (1): wmphoto: Zero user buffer before calling jxrlib Copy. Fabian Maurer (1): kernel32: Prevent null pointer dereference in WritePrivateProfileStructW. Gijs Vermeulen (2): dbghelp: Add SymSrvGetFileIndexInfo stub. msvcp140: Add _ExceptionHolder::ReportUnhandledError stub. Hans Leidekker (4): wininet: Remove assertion that is no longer valid in Windows 10. wininet: Always reset the second input buffer in netcon_secure_connect_setup(). bcrypt: Force symmetric key reset if necessary. wbemprox: WQL string comparisons are case insensitive. Hugh McMaster (1): conhost: Set correct face name when the specified font is not available. Huw D. M. Davies (1): nsi: Use the correct members for the error and discard counts. Jeff Smith (2): d3drm/tests: Clean up some magic values. d3drm: Fix components of palette built when loading texture from file. Jinoh Kang (3): wineusb.sys: Remove erroneous call to LeaveCriticalSection() in add_usb_device. riched20: Commit and wrap paragraphs after changing text properties. riched20: Ensure MEPF_COMPLEX is unset when in password input mode. Kevin Puetz (2): winecrt0: Remove free_delay_imports. vbscript: Don't abort when trying to TRACE() functions with many arguments. Limstash Wong (1): taskmgr: Fix the unit in memory usage graph on performance tab. Louis Lenders (1): msvcr80: Update manifest resource. Martin Storsjö (2): include: Add declarations of _alloca and alloca for MSVC mode. ntdll: Add SEH unwind info in ARM assembly functions/trampolines. Mihail Ivanchev (1): include: Add prototype for ReOpenFile(). Nikolay Sivov (2): vbscript: Implement DateSerial(). vbscript: Implement DateAdd(). Paul Gofman (3): msxml3: Use node_get_text() in domelem_get_nodeTypedValue() if no type is present. crypt32: Support CRYPT_STRING_HEX in CryptStringToBinary(). winhttp/tests: Avoid race condition on closing connection in server_thread(). Piotr Caban (4): msvcr120: Add missing _StructuredTaskCollection destructor spec file entry. msvcrt: Fix _wputenv_s return value on error. msvcrt: Fix _putenv_s return value on error. msvcrt: Fix _wputenv_s invalid argument handling. Robert Wilhelm (4): scrrun: Added DateCreated property for IFile. scrrun: Add test for correct string length returned from GetTempName. scrrun: Correct string length returned from GetTempName. scrrun: Avoid memory leak in error case. Rémi Bernon (2): winebus.sys: Implement SOCD neutral cleaning for hatswitches. server: Avoid using pointer value after realloc. Santino Mazza (3): riched20/tests: When selection is out of view, the scrollbar should move. riched20: Ensure that cursor is visible when selecting text. jscript: Return ParseProcedureText when compile fails. Sebastian Lackner (1): ntdll: Improve stub of NtQueryEaFile. Tim Clem (1): kernelbase: Don't assume the length of the process image name. Torge Matthies (8): msvcr110: Add _StructuredTaskCollection::_StructuredTaskCollection_ctor stub. msvcr120: Add _StructuredTaskCollection::_StructuredTaskCollection_dtor stub. msvcr100: Add _StructuredTaskCollection::_Schedule stub. msvcr110: Add _StructuredTaskCollection::_Schedule_loc stub. msvcr100: Implement invalid_multiple_scheduling exception. msvcr100: Factor out the mapping of a context to a scheduler. msvcr100: Implement _StructuredTaskCollection::_Schedule and _Schedule_loc. msvcp90: Fix segfault in bad_alloc exception default constructor. Zebediah Figura (7): quartz/systemclock: Use timeGetTime() to retrieve the current time. wined3d: Correctly handle 3D textures in wined3d_view_{load, invalidate}_location(). ntdll: Add a fast path for NtDeviceIoControlFile(INVALID_HANDLE_VALUE). xslt: Use win32config.h from the correct directory. ntdll: Calculate the necessary length after calling snprintf() in SystemWineVersionInformation. msvcrt: Add SEH annotations for sse2_memcpy(). libwine: Avoid comparing the result of pointer arithmetic to zero. Zhiyi Zhang (14): shell32: Correct SHOpenFolderAndSelectItems() prototype. include: Add SHOpenFolderAndSelectItems() declaration and flags. shell32/tests: Add SHOpenFolderAndSelectItems() tests. shell32: Implement SHOpenFolderAndSelectItems(). comctl32/tests: Add GetWindowSubclass() tests. comctl32: Check reference data pointer before using it. comctl32/tests: Test that changing focus for edit controls doesn't repaint immediately. comctl32/edit: Don't repaint immediately when getting focus. winex11.drv: Handle X errors from glXCopyContext(). winex11.drv: Don't compare error event serial if it's zero. winegstreamer: Always check the return value from mf_media_type_from_wg_format(). gdi32/tests: Add more tests for GetTextMetrics(). win32u: Choose a smaller ppem to avoid exceeding the requested font height. win32u: Fix a memory leak. bye michael

1 0

fileapi.h / Kernel32: GetTempPath2A/W() implementation
by Kevin Connor Arpe 16 Mar '23

16 Mar '23

Hello, This is my first time posting to this mailing list. (This email is a bit long... sorry!) I am a developer writing Win32 code using Wine on Debian Linux. I tried to use the Win32 API function GetTempPath2W(), but my Wine version does not support it. I checked out the Wine source code and found GetTempPathW() here: dlls/kernelbase/file.c ... but did not find GetTempPath2W(). First, if this function is already in the pipeline (another feature branch, etc.), please let me know... and ignore the rest of this email! :-) From GetTempPathW() docs: https://learn.microsoft.com/en-us/windows/win32/api/fileapi/nf-fileapi-gett… "Note: Apps should call GetTempPath2 instead of GetTempPath." From GetTempPath2W() docs: "When calling this function from a process running as SYSTEM it will return the path C:\Windows\SystemTemp" The implementation looks rather straightforward except this part: "process running as SYSTEM". How to detect this? I found a good blog post from (the legend) Raymond Chen: https://devblogs.microsoft.com/oldnewthing/20210106-00/?p=104669 Sample code looks like this: << #include <wil/token_helpers.h> bool DoesTokenRepresentSid(HANDLE token, WELL_KNOWN_SID_TYPE type) { // maps to GetTokenInformation(token, TokenUser, ...); auto user = wil::get_token_information<TOKEN_USER>(token); return !!IsWellKnownSid(user->User.Sid, type); } bool IsCurrentProcessRunningAsSystem() { return DoesTokenRepresentSid(GetCurrentProcessToken(), WinLocalSystemSid); } bool IsCurrentThreadRunningAsSystem() { return DoesTokenRepresentSid(GetCurrentThreadEffectiveToken(), WinLocalSystemSid); } >> Reading the WIL source code on GitHub, I understand the required Win32 calls. In short, I plan to re-write the above same code using pure Win32 code (remove the WIL requirement). My questions: 1. Do you agree with Raymond Chen's technique for Wine source code? (Or: Is there a better way to do it for Wine source code?) 2. Should my SYSTEM account test use IsCurrentProcessRunningAsSystem() or IsCurrentThreadRunningAsSystem()? If I read the official docs literally, I think "process" not "thread". Please advise. 3. Reading the WIL code: TOKEN_INFORMATION_CLASS TokenUser appears to require this pattern: 1. Call GetTokenInformation() to get required buffer size 2. malloc buffer 3. Call GetTokenInformation() again with buffer 4. Check: WinLocalSystemSid == ((TOKEN_USER *) buffer)->User.Sid 5. free buffer 1. Is there a way to avoid the above steps? malloc+free seems like overkill to decide if the current user is SYSTEM! 2. Do I misunderstand the WIL code? Reading the Wine code for: GetTokenInformation() -> NtQueryInformationToken(): TOKEN_USER appears to be fixed size. Why does WIL think TOKEN_USER is not fixed size? I am confused! :-) Kind regards, Kevin Connor ARPE Tokyo, Japan

4 3

dxcore.dll or dxgi?
by James McDonnell 16 Mar '23

16 Mar '23

Hi-d ho winerinos, I have made some changes to add a dxcore.dll on a personal branch. I want to upstream the changes, however I'm unsure of whether I should create a new dll or just append it to dxgi. Dxcore overlaps with dxgi, but presumably has separate functionality as well. Thoughts? Here is my branch if anyone wanted to review the commits. https://gitlab.winehq.org/ElementalWarrior/wine/-/commits/affinity-photo2 Cheers, James

3 2

Wine staging 8.3 release
by Alistair Leslie-Hughes 04 Mar '23

04 Mar '23

Binary packages for various distributions will be available from: https://www.winehq.org/download Summary since last release * Rebased to current wine 8.3 (532 patches are applied to wine vanilla) The patchinstall.sh script is no longer available and users should use the staging/patchinstall.py instead. Upstreamed (Either directly from staging or fixed with a similar patch). * ntdll: Add helper function to delete free blocks. * ntdll: Improve heap allocation performance. (v2) Added: * [52012] Implement _vcomp_for_dynamic_init_i8. Updated: * None Where can you help * Run Steam/Battle.net/GOG/UPlay/Epic * Test your favorite game. * Test your favorite applications. * Improve staging patches and get them accepted upstream. * Suggest patches to be included in staging. As always, if you find a bug, please report it via https://bugs.winehq.org Best Regards Alistair.

1 0

Upcoming vkd3d 1.7 release
by Henri Verbeet 24 Feb '23

24 Feb '23

Hi all, The tentative release date for vkd3d 1.7 is March 22, in slightly under 4 weeks. Please test your applications for any regressions, review any API changes since vkd3d 1.6, and so on. While starting today it will become increasingly harder to make major code changes until the release, improvements to the API documentation and tests are welcome as always. For vkd3d 1.8, the release following 1.7, we're currently aiming for a date in late June. Regards, Henri

1 0

[PATCH] user32/tests: fixed size passed to ToUnicode
by Marcus Meissner 24 Feb '23

24 Feb '23

Signed-off-by: Marcus Meissner <marcus(a)jet.franken.de> --- dlls/user32/tests/input.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/dlls/user32/tests/input.c b/dlls/user32/tests/input.c index ee39b11fc12..be41078bf29 100644 --- a/dlls/user32/tests/input.c +++ b/dlls/user32/tests/input.c @@ -3186,7 +3186,7 @@ static void test_ToAscii(void) ok(character == '\r', "ToAscii for Return was %i (expected 13)\n", character); wstr[0] = 0; - ret = ToUnicode('A', SC_A, state, wstr, sizeof(wstr), 0); + ret = ToUnicode('A', SC_A, state, wstr, ARRAY_SIZE(wstr), 0); ok(ret == 1, "ToUnicode(A) returned %i, expected 1\n", ret); str[0] = '\0'; -- 2.39.2

2 2

Wine staging 8.2 release
by Alistair Leslie-Hughes 18 Feb '23

18 Feb '23

Binary packages for various distributions will be available from: https://www.winehq.org/download Summary since last release * Rebased to current wine 8.2 (531 patches are applied to wine vanilla) The patchinstall.sh script is no longer available and users should use the staging/patchinstall.py instead. For those that wanted to use ODBC in windows applications, this is the release for you. For those using ADO to connect to a dababase via ODBC, you will still need to use the native version of msado15 and msdasql. Through some of the bugs have been identified and fixed in this wine release there is still more work to be done. Some cavets on the ODBC support * Only the Unicode drivers for MySql/PostgreSQL were tested. * Not all functions have been mapped, so lodge a bug if something is missing. Developers. If anybody has time to have a look over or reivew the patchset, that would be great. Don't be scared of the 43 patches, most of them are calls into the loaded driver (boiler plate stuff). Users. If you are a user of unixODBC with wine, it would be great if you could post on bug 54499, the database used or if you prefer to use the standard windows driver. Known issues. * Using SQLDriverConnectW to connect to an ANSI ODBC driver will fail (works on windows). * The Driver DLL is possible loaded multiple times (Should have map of driver/function table). Upstreamed (Either directly from staging or fixed with a similar patch). * None Added: * [52233] Black NPC's skin or minor rendering glitches in multiple games (Batman: Arkham Knight, Steel Division: Normandy 44) * [48220] dxruby needs WAVE form loading implemented * [45135] Black Rockman Shooter has no sound, implementation missing in dmime * [34751] Aura: Fate of the Ages: sounds aren't played, but music works fine * [9027] No sound for rise of nations - all versions * [54499] Native ODBC drivers should be able be used. Updated: * Install scripts updated to Python 3 * winex11-Fixed-scancodes (was winex11-key_translation) * ntdll-NtDevicePath * ntdll-Junction_Points Where can you help * Run Steam/Battle.net/GOG/UPlay/Epic * Test your favorite game. * Test your favorite applications. * Improve staging patches and get them accepted upstream. * Suggest patches to be included in staging. As always, if you find a bug, please report it via https://bugs.winehq.org Best Regards Alistair.

1 0

How do I package the output of wine as a deb file?
by zhengxianwei 18 Feb '23

18 Feb '23

I've compiled wine on a 64-core machine, and I want to package it up and send it to another machine for installation. I checked 2 pages, but still not sure how to do this https://wiki.winehq.org/Building_Wine https://wiki.winehq.org/Packaging

1 0

Debugging channel weirdness
by Zebediah Figura 14 Feb '23

14 Feb '23

Hi all, I was running into some odd behaviour while debugging—specifically, I found that specifying the d3d channel in WINEDEBUG was cutting my performance in half, even when disabling logging, and even when it should have been functionally identical to a run without the d3d channel specified. I was able to "fix" this inconsistency with this diff: diff --git a/dlls/ntdll/thread.c b/dlls/ntdll/thread.c index 1bd8f900d22..f0e4b824918 100644 --- a/dlls/ntdll/thread.c +++ b/dlls/ntdll/thread.c @@ -103,7 +103,11 @@ unsigned char __cdecl __wine_dbg_get_channel_flags( struct __wine_debug_channel { pos = (min + max) / 2; res = strcmp( channel->name, debug_options[pos].name ); - if (!res) return debug_options[pos].flags; + if (!res) + { + if (channel->flags & (1 << __WINE_DBCL_INIT)) channel->flags = debug_options[pos].flags; + return debug_options[pos].flags; + } if (res < 0) max = pos - 1; else min = pos + 1; } Which makes sense to some degree, but leaves three questions unanswered, which I was hoping someone might be able to cluebat me on: (1) Why does setting channel->flags affect performance at all? I'm thoroughly baffled by this one. (2) Why was this diff not applied in the first place? (Simple oversight?) (3) Why does __WINE_DBCL_INIT even exist, if its only purpose is to guard out a single assignment? --Zeb

2 6

Hangover Update
by André Zwing 11 Feb '23

11 Feb '23

Hi, As you might know, Hangover development stalled years ago waiting for the WoW64 support in Wine to complete. With Wine 8.0 (and even better with Jaceks wow branch) this is now done. In October I started to connect Qemu to that WoW64 support in my spare time and it progressed until I finally was able to publish some of that work today. It's still missing instructions and even some patches, but it should give a good introduction on what this will look like. In fact I started to get wowarmhw running, so I could do the first steps on a x86-64 machine to run ARM32 binaries. But exactly this part is missing for now as things changed since then. So in case I made you curious, here's the code :) https://github.com/AndreRH/hangover

1 0

Compiling issues on macOS after 7.20 on ARM
by Jason Millard 10 Feb '23

10 Feb '23

Hello. I’m using parts of Wine (specifically the vbscript engine) while porting Visual Pinball to be cross platform. I have been using the WIDL utility extensively and it has worked great. I wanted to keep my patches up to date so I don’t fall behind, however I am having issues compiling wine after version 7.20. This was my typical compile: brew install llvm llvm-mingw git clone -b wine-7.20 git://source.winehq.org/git/wine.git git clone git@github.com:vpinball/pinmame.git cd wine export MACOSX_DEPLOYMENT_TARGET="10.14" export LDFLAGS="-Wl,-rpath,/opt/X11/lib" export PATH="$(brew --prefix llvm)/bin:$PATH" export PATH="$(brew --prefix bison)/bin:$PATH" ./configure --without-freetype make -j10 tools/widl/widl -o ../vpinmame_i.h --nostdinc -Ldlls/\* -Iinclude -D__WINESRC__ -D_UCRT ../pinmame/src/win32com/VPinMAME.idl Now with anything past 7.20 I’m getting: configure: error: PE cross-compilation is required for ARM64, please install clang/llvm-dlltool/lld, or llvm-mingw. Is there anyway to disable this? Thanks Jason

2 4

Wine staging 8.1 release
by Alistair Leslie-Hughes 03 Feb '23

03 Feb '23

Binary packages for various distributions will be available from: https://www.winehq.org/download Summary since last release * Rebased to current wine 8.1 (475 patches are applied to wine vanilla) Upstreamed (Either directly from staging or fixed with a similar patch). * rpcrt4: Avoid implicit cast of interface pointer. * include: Add _XHR enum values * ntdll-Builtin_Prot (removed as PE conversion is complete) Added: * None Updated: * ntdll-Placeholders Where can you help * Run Steam/Battle.net/GOG/UPlay/Epic * Test your favorite game. * Test your favorite applications. * Improve staging patches and get them accepted upstream. * Suggest patches to be included in staging. As always, if you find a bug, please report it via https://bugs.winehq.org Best Regards Alistair.

1 0

Re: How should one report GitLab CI bugs?
by Francois Gouget 26 Jan '23

26 Jan '23

On Thu, 19 Jan 2023, Alexandre Julliard wrote: [...] > Would it mean failures that only happen on Gitlab CI? Or failures that > happen in CI but also elsewhere? Either way, the distinction doesn't > seem all that useful to me. The idea was that some Wine developers may be more interested in fixing bugs causing false positives in the new CI system (GitLab CI) than in the old one (TestBot). So identifying those in Bugzilla may help them find targets to work on. But of course that assumes that set of developers is non-empty. -- Francois Gouget <fgouget(a)codeweavers.com>

1 0

Re: Alexandre Julliard : shell32: Allocate a full SHITEMID to avoid a compiler warning.
by Eric Pouech 26 Jan '23

26 Jan '23

Le 25/01/2023 à 22:22, Alexandre Julliard a écrit : > Module: wine > Branch: master > Commit: c844ebdb0ae8994ee4f3d1d3c1697ece06f52a4e > URL: https://gitlab.winehq.org/wine/wine/-/commit/c844ebdb0ae8994ee4f3d1d3c1697e… > > Author: Alexandre Julliard <julliard(a)winehq.org> > Date: Tue Jan 24 22:22:52 2023 +0100 > > shell32: Allocate a full SHITEMID to avoid a compiler warning. > > --- > > dlls/shell32/shfldr_fs.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/dlls/shell32/shfldr_fs.c b/dlls/shell32/shfldr_fs.c > index 662d8b52d5b..d47911dbef7 100644 > --- a/dlls/shell32/shfldr_fs.c > +++ b/dlls/shell32/shfldr_fs.c > @@ -347,7 +347,7 @@ IShellFolder_fnParseDisplayName (IShellFolder2 * iface, > /* Special case for the root folder. */ > if (!wcsicmp( szPath, L"\\\\?\\unix\\" )) > { > - *ppidl = SHAlloc(sizeof(USHORT)); > + *ppidl = SHAlloc(sizeof(*ppidl)); shouldn't it be sizeof(**ppidl) instead to better match the patch description?

1 0

Problem with the wine-gitlab mailing list archive?
by Francois Gouget 26 Jan '23

26 Jan '23

The wine-gitlab mailing list archive shows no new message since last Tuesday (2022-10-18). Yet I did receive emails on that list since then. https://www.winehq.org/mailman3/hyperkitty/list/wine-gitlab@winehq.org/ What's going on? -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ Cahn's Axiom: When all else fails, read the instructions.

4 7

dmloader:dmloader crash on gitlab-debian-64
by Francois Gouget 26 Jan '23

26 Jan '23

On test.winehq.org I see that dmloader:dmloader crashed on gitlab-debian-64 due to an assertion (2023-01-12): 060c:loader: 104 tests executed (0 marked as todo, 0 as flaky, 0 failures), 0 skipped. dmloader:loader:060c done (-2147483391) in 0s 88B Test failed: crash (80000101) https://test.winehq.org/data/patterns.html#dmloader:loader I'd like to know if this is caused by the one assert() in stream_get_chunk(). A backtrace could help in that regard but: * The test report on test.winehq.org only has the test output, not Wine's stderr traces which is where a backtrace would be. http://winetest.dolphin/data/047118247ce489a7138428c8b2cd5692ed04337e/linux… * And the GitLab log has neither the test output nor the Wine traces. https://gitlab.winehq.org/wine/wine/-/jobs/8996 So are the Wine traces available somewhere for GitLab's nightly WineTest runs? -- Francois Gouget <fgouget(a)codeweavers.com>

2 2

Wine staging 8.0 release
by Alistair Leslie-Hughes 25 Jan '23

25 Jan '23

Binary packages for various distributions will be available from: https://www.winehq.org/download Summary since last release * Rebased to current wine 8.0 (478 patches are applied to wine vanilla) Upstreamed (Either directly from staging or fixed with a similar patch). * None Added: * None Updated: * None Where can you help * Run Steam/Battle.net/GOG/UPlay/Epic * Test your favorite game. * Test your favorite applications. * Improve staging patches and get them accepted upstream. * Suggest patches to be included in staging. As always, if you find a bug, please report it via https://bugs.winehq.org Best Regards Alistair.

1 0

8.0 release notes (draft)
by Alexandre Julliard 24 Jan '23

24 Jan '23

Folks, The final 8.0 will be released shortly. Thanks to everybody who helped fix regressions! We didn't quite make it back to the 7.0 count, but we've still made very good progress, with over 160 bugs fixed during code freeze. Below is the draft of the release notes, please have a look and let me know if anything is incorrect or missing. Thanks to Jacek, Nikolay, Rémi and Zeb for their help with this! -- Alexandre Julliard julliard(a)winehq.org ---------------------------------------------------------------- What's new in Wine 8.0 ====================== *** PE modules - After 4 years of work, the PE conversion is finally complete: all modules can be built in PE format. This is an important milestone on the road to supporting various features such as copy protection, 32-bit applications on 64-bit hosts, Windows debuggers, x86 applications on ARM, etc. However, some modules still perform direct calls between the PE and the Unix part, instead of going through the NT system call interface. The remaining direct calls will be removed during the Wine 8.x development phase. - A special syscall dispatcher is used for PE -> Unix transitions, to avoid the overhead of a full NT system call. This minimizes the performance impact of the new architecture, in particular for the OpenGL and Vulkan libraries. - Building mixed Windows/Unix libraries in ELF format (.dll.so libraries) is still supported for use in Winelib applications. However, such applications won't support features enabled by the NT syscall interface, such as WoW64 without 32-bit libraries. *** WoW64 - WoW64 thunks are implemented for essentially all Unix libraries, enabling a 32-bit PE module to call a 64-bit Unix library. Once the remaining direct PE/Unix calls have been removed, this will make it possible to run 32-bit Windows applications without any 32-bit Unix library. - When the 32-bit Wine loader isn't found, 32-bit applications are started in the new "Windows-like" WoW64 mode (where 32-bit code runs inside a 64-bit host process). This mode can be enabled by building with the '--enable-archs' configure option. Since in case of configuration errors it is possible for it to be triggered inadvertently, applications started in this mode print the warning "starting in experimental wow64 mode". *** Graphics - The "Light" theme is enabled in the default configuration, for a more modern look. Other themes can still be configured through WineCfg. - The graphics drivers (winex11.drv, winemac.drv, wineandroid.drv) are converted to run on the Unix side of the syscall boundary, and interface with the Unix side of the Win32u library. - The Print Processor architecture is implemented. This will be used to avoid direct PE<->Unix calls in the printer driver. - Effects are supported in Direct2D, including description parsing and a number of core objects. - Command lists recording and playback are implemented in Direct2D. - The Vulkan driver supports up to version 1.3.237 of the Vulkan spec. *** Direct3D - Many optimizations related to streaming map acceleration are implemented, in common code as well as in the GL renderer. Depending on the application, this may result in major performance improvements. - Adapter video memory budget change notifications are implemented. - The Vulkan renderer supports setting multiple viewports and scissor rectangles. - Several Direct3D 10 and 11 features, which were already implemented, are accurately reported through capability flags: - Per-format support for vertex and index buffers. - Per-format support for all capabilities via D3D11_FEATURE_FORMAT_SUPPORT. - No-overwrite maps on shader resources and constant buffers. - Partial range binding of constant buffers. - For convenience, a new WINE_D3D_CONFIG environment variable may be used instead of modifying the HKEY_CURRENT_USER\Software\Wine\Direct3D registry key. The value is a comma- or semicolon-separated list of key-value pairs, e.g.: WINE_D3D_CONFIG="renderer=vulkan;VideoPciVendorID=0xc0de" If an individual setting is specified in both the environment variable and the registry, the former takes precedence. - The Direct3D graphics card database recognizes more graphics cards. - The Vulkan renderer limits the maximum Direct3D feature level based on available Vulkan features. *** Audio / Video - The MPEG-1 audio decoder filter for layers 1 and 2, and the separate filter for layer 3, are both implemented on top of GStreamer. - The ASF reader filter is implemented. - The OpenAL32.dll wrapper library is removed, in favor of the native OpenAL32.dll library shipped by Windows applications. *** Media Foundation - Content type resolution is improved in the Media Foundation Player. - Rate control is implemented. - The Enhanced Video Renderer default mixer and presenter are better supported. - The DirectShow filter is implemented in the Enhanced Video Renderer. - An initial framework is implemented for the Writer encoding API. - The topology loader is better supported. *** Input devices - Controller hotplug support is greatly improved, and controller removal and insertion are correctly dispatched to applications. - Driving wheel device detection and reporting is better implemented, using the SDL library and HID usages as hints for the device types. - Force feedback effect support is improved, specifically with driving wheels controllers. - The HID Haptics specification is used for left / right motor rumble and trigger rumble. It makes it possible for device drivers to expose haptics support over HidRaw, and overcome EvDev limitations. - The Joystick Control Panel is redesigned, with new graphics and a dedicated view for XInput gamepads. - Sony DualShock and DualSense controllers are supported when the hidraw backend is used, and exposed to applications in the same way as they are on Windows. - The Windows.Gaming.Input module is introduced, one of the first WinRT modules in Wine, implementing a new programming interface to access gamepads, joysticks and driving wheel devices. For this new input API, hotplug notifications are also implemented, as well as force feedback effects and haptics. This API is the first one to support trigger rumble. *** Internationalization - A proper locale database, in Windows locale.nls format, is generated from the Unicode CLDR database. The list of supported locales is essentially identical to that of the latest Windows version. - Unicode string comparison is based on the Windows Sortkey database and algorithms instead of the Unicode Collation Algorithm, yielding more compatible results. - Characters from high Unicode planes (in UTF-16 encoding) are supported in most functions, in particular for case mappings, BiDi formatting, and character shaping. - The UTF-8 encoding is supported as Ansi codepage. It is enabled when requested by the application's manifest. - Unicode character tables are based on version 15.0.0 of the Unicode Standard. - The timezone data is generated from the IANA timezone database version 2022g. *** Text and fonts - Font linking is enabled for most system fonts, which should fix missing glyphs in CJK locales. - Font fallback in DirectWrite is reworked, with additional support for an large number of scripts. *** Kernel - The ApiSetSchema database is implemented, replacing all the api-ms-* forwarding modules. This reduces both disk space and address space usage. - DOS file attributes are persistent, and stored on disk using the filesystem's extended attributes, in a format compatible with Samba. *** Internet and networking - The Online Certificate Status Protocol (OCSP) is implemented. - More EcmaScript features are supported in JScript standard-compliant mode. - The JScript garbage collector is implemented. - Accessibility support is enabled in Wine Gecko package. - Web Storage, Performance object and more event objects are implemented in MSHTML. *** Mono / .NET - The Mono engine is updated to version 7.4.0. The list of changes can be viewed at https://github.com/madewokherd/wine-mono/releases/tag/wine-mono-7.4.0 *** Builtin applications - All the builtin applications use Common Controls version 6, which enables theming and high-DPI rendering by default. - The Wine Debugger (winedbg) prints thread names in the 'info thread' listing, for applications that set the names of their threads. All internal Wine threads also set their own name. - The QWORD type is supported in the REGEDIT and REG registry tools. - Notepad has a status bar displaying the current cursor position. The Goto Line functionality is also implemented. - The builtin console programs print data in the OEM codepage, for better compatibility with applications reading their output. - The Service Control tool (sc.exe) supports the 'query' command. *** Development tools - The PE conversion has made it necessary to implement a new mechanism to run static C++ constructors in Winelib ELF applications. Unfortunately, this requires the corresponding modules to be rebuilt with Wine 8.0. The rebuilt modules will still run correctly on older Wine versions, but older modules will no longer execute static constructors when run on Wine 8.0. - In regression tests, individual checks can be marked as "flaky". This is useful for checks that depend on external factors such as network timings. The errors are still reported but not counted as failures. - Test failure messages can be printed in color by setting the WINETEST_COLOR environment variable, to make them easier to spot in the test log. - The Resource Compiler (WRC) and Message Compiler (WMC) use the locale.nls database to detect languages and codepages, and thus support all the Windows locales. - The WineDump tool supports printing the contents of Enhanced Metafile spool files, as well as more contents from PDB files. *** Build infrastructure - PE binaries can be built for multiple architectures from the same build tree, using the '--enable-archs' option to configure, for example '--enable-archs=i386,x86_64'. This can be used to enable the experimental WoW64 mode, where 32-bit applications can run without any 32-bit host libraries. - Data types that are defined as 'long' on Windows are defined as 'long' instead of 'int' on all platforms with a 32-bit long type. This is more compatible, but it can cause printf format warnings with Unix compilers. In Winelib code, it can be disabled by defining WINE_NO_LONG_TYPES. - Import libraries can be generated without using dlltool, by passing the '--without-dlltool' option to winebuild. This is used to avoid dlltool bugs with delay import libraries. - Libraries that contain only resources and no code can be built with the '--data-only' option to winegcc, which makes them smaller and more efficient to load. *** Bundled libraries - Faudio is updated to the upstream release 22.11. - LCMS2 is updated to the upstream release 2.14. - LibJPEG is updated to the upstream release 9e. - LibMPG123 is updated to the upstream release 1.31.1. - LibPng is updated to the upstream release 1.6.39. - LibTiff is updated to the upstream release 4.4.0. - LibXml2 is updated to the upstream release 2.10.3. - LibXslt is updated to the upstream release 1.1.37. - Zlib is updated to the upstream release 1.2.13. *** Platform-specific features - The experimental WoW64 architecture is supported on macOS versions from 10.15. - 64-bit time_t is supported on 32-bit Linux with recent Glibc versions. - Exception unwinding on ARM platforms is extended with EHABI support on ARM32 and Dwarf support on ARM64. *** Miscellaneous - RSA encryption and the RSA-PSS signing algorithm are implemented. - An initial version of the UI Automation API is implemented. - C++11 syntax is supported in function name demangling. *** External dependencies - The vkd3d and LDAP libraries are bundled in the source tree and built as PE. The corresponding Unix libraries are no longer needed. - The OpenAL library is no longer used.

3 2

Wine staging 8.0-rc5 release
by Alistair Leslie-Hughes 22 Jan '23

22 Jan '23

Binary packages for various distributions will be available from: https://www.winehq.org/download Summary since last release * Rebased to current wine 8.0-rc5 (478 patches are applied to wine vanilla) Upstreamed (Either directly from staging or fixed with a similar patch). * None Added: * None Updated: * None Where can you help * Run Steam/Battle.net/GOG/UPlay/Epic * Test your favorite game. * Test your favorite applications. * Improve staging patches and get them accepted upstream. * Suggest patches to be included in staging. As always, if you find a bug, please report it via https://bugs.winehq.org Best Regards Alistair.

1 0

[WINE-CE] Wine-CE Can Run Cross Instruction Architecture Windows App
by fanwj＠mail.ustc.edu.cn 19 Jan '23

19 Jan '23

Wine-CE (Chimera Edition) is a compatibility layer capable of running Windows applications on Cross-architecture paltform of Linux. It is base on Wine Project and Qemu Project. Project Address: https://gitee.com/fanwenjie/wine-ce TEST App: Sword and Fairy 3 Demonstration: https://www.bilibili.com/video/BV1Kd4y157Lm

2 1

How should one report GitLab CI bugs?
by Francois Gouget 18 Jan '23

18 Jan '23

How should one report GitLab CI bugs? I'm thinking of two types of bugs. 1. Wine test bugs that specifically happen on the GitLab CI infrastructure. See for instance bug 54334 which describes a set of qasf:asfreader failures that has, so far, only been seen in the Linux GitLab CI. I'm thinking some Wine developers may be particularly interested in getting such bugs fixed and thus would want to have a way to find them. Would a 'gitlabci' or just 'ci' keyword be appropriate? 2. Bugs and missing features in the GitLab CI infrastructure itself. For instance the failures that gitlab-win10-21h2 gets in the ole32:clipboard, riched20:richole and user32:msg tests strongly suggest that the AMD driver was installed with its Adrenalin GUI. Since it is known to interfere with the clipboard this could be considered a bug in that Windows 10 VM configuration. (Alternatively one could argue the tests should be modified to be resistant to such interference which would move this issue to the first category.) Which product should one pick in Bugzilla for reporting such bugs? They would not fit in 'Wine' product or any of the other existing products except 'Wine-Testbot' if one considers the GitLab CI to be the new TestBot (though I think that would be quite confusing). -- Francois Gouget <fgouget(a)codeweavers.com>

1 0

← Newer
1
...
8
9
10
11
12
13
14
...
3379
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-devel