May 2021 - Wine-bugs - list.winehq.org

[Bug 48175] New: AION (64 bit) - crashes in CrySystem32.dll.CryFree() due to high memory pointers allocated
by WineHQ Bugzilla 01 Dec '24

01 Dec '24

https://bugs.winehq.org/show_bug.cgi?id=48175 Bug ID: 48175 Summary: AION (64 bit) - crashes in CrySystem32.dll.CryFree() due to high memory pointers allocated Product: Wine Version: 4.20 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: gofmanp(a)gmail.com Distribution: --- Created attachment 65809 --> https://bugs.winehq.org/attachment.cgi?id=65809 Force down-top memory allocation Split off from bug #40334. The crash happens after the original issue in Bug #40334 is solved, after the loading screen or very soon after login. The game expects memory pointers from HeapAlloc and VirtualAlloc to fit in the certain range. It is fine with ~16GB range as per my observations, that is, up to 0x3ffffffff address. I am attaching the patch which is confirmed to fix the issue with AION. The patch consists of two, the first one is not exactly related but the second one depends on it. At the present moment the first one is passing the review upstream. The solution in the second patch is to force virtual memory allocation return lower addresses when they are available (unless top-down allocation is explicitly requested), that's how it seem to work on Windows. The patch does it essentially the same way as the allocation with zero_bits mask is performed. -- 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 16302] New: [oblivion] can not skip the intro movie by press ESC
by wine-bugs＠winehq.org 27 Nov '24

27 Nov '24

http://bugs.winehq.org/show_bug.cgi?id=16302 Summary: [oblivion] can not skip the intro movie by press ESC Product: Wine Version: 1.1.8 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: wangbinrobin(a)yahoo.com.cn when runing Oblivion on Windows, we can skip the intro movie by pressing ESC, but when runing Oblivion by wine, we can not skip it by ESC. it is quite a long time to wait the intro movie end. The graphic card I use is ATI Radeon 2600, driver 8.10 -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email 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 45

[Bug 45871] New: Several apps in .NET472 crash with unhandled exception.\nException at System.Windows.Media.Composition.DUCE+Channel.SyncFlush ()\n at System.Windows.Media.MediaContext.Complet"...( native d3dcompiler works around it)
by wine-bugs＠winehq.org 25 Nov '24

25 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=45871 Bug ID: 45871 Summary: Several apps in .NET472 crash with unhandled exception.\nException at System.Windows.Media.Composition.DUCE+Channel.SyncFlus h()\n at System.Windows.Media.MediaContext.Complet"...(native d3dcompiler works around it) Product: Wine Version: 3.16 Hardware: x86 OS: Linux Status: NEW Keywords: dotnet, download Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: xerox.xerox2000x(a)gmail.com Distribution: --- Created attachment 62374 --> https://bugs.winehq.org/attachment.cgi?id=62374 part of crashlog Hi Seen this for a bunch of programs while trying to run them with installed dotnet472. The bug seems to be hidden for several apps as they ship their owm d3dcompiler_47, but some don`t I found out, and they crash Something like: 01c2:fixme:d3dcompiler:make_assignment Check for casts in the lhs. 01c2:err:d3dcompiler:compile_shader HLSL shader parsing failed. 01c2:err:ntdll:vDbgPrintExWithPrefix 65: MIL FAILURE: Unexpected HRESULT 0x88760b59 in caller: intermediate rendering error 01c2:err:ntdll:vDbgPrintExWithPrefix 65: MIL FAILURE: Unexpected HRESULT 0x88760b59 in caller: The render thread failed unexpectedly. Then crash Example: Slackinstaller.exe: winetricks dotnet452 win7 corefonts, then start installer with builtin d3dcomiler_47 works fine. However winetricks dotnet472 win7 corefonts, then start installer with builtin d3dcomiler_47 --> crash (Note: for Slackinstller one also needs Staging-patches for bug 40613) -- 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 24

[Bug 10472] New: File open dialog buffer too short for multiple file selection
by wine-bugs＠winehq.org 23 Nov '24

23 Nov '24

http://bugs.winehq.org/show_bug.cgi?id=10472 Summary: File open dialog buffer too short for multiple file selection Product: Wine Version: 0.9.49. Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: minor Priority: P2 Component: wine-comctl32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: freireclaudio(a)yahoo.com.ar The file open dialog seems to be using a buffer which is too short for multiple file selection. While Windows itself has this bug/feature, wine's version is much more annoying since I'm unable to select over 6 files usually (using long but not excessive names). A dynamic buffer (with a reasonable limit for security reasons perhaps) would be preferrable, but even a bigger buffer would help. I think windows' buffer is 4k. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching all bug changes.

2 18

[Bug 42606] New: wine doesn' t *fully* respect locale settings in some corner cases
by wine-bugs＠winehq.org 22 Nov '24

22 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=42606 Bug ID: 42606 Summary: wine doesn't *fully* respect locale settings in some corner cases Product: Wine Version: 2.2 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: galtgendo(a)o2.pl Distribution: --- This time it's a problem even if this is the only app currently running in the prefix. Download address for the engine: http://kikyou.info/dl_redirect.php?%2Ftvp%2Ffiles%2Fkr2_232r2.zip Instruction for setting up a project: http://kirikirikag.sourceforge.net/contents/Prepare.html (instructions seem to be for a different archive, but get the point across) So, when running 'LANG=ja_JP.utf8 wine krkr.eXe'(as it's different from my standard locale) and starting that project, most of the things seem to work correctly (as I've got the standard overrides for Gothic/Mincho set already in place), yet... When right clicking the engine console (not the game window), there's mojibake in that menu. WINEDEBUG=font log shows that for some reason CreateFontIndirectExW for Tahoma is called with charset determined by my standard locale, instead of the Japanese one. That somehow results in the text being treated as encoded in that locale's Windows codepage, instead of cp932 (or is that vice versa ?). -- 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 8

[Bug 48426] New: Wine-Gecko can't be installed.
by WineHQ Bugzilla 21 Nov '24

21 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=48426 Bug ID: 48426 Summary: Wine-Gecko can't be installed. Product: Wine-staging Version: 5.0-rc4 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: winebugs(a)evolution-hosting.eu CC: leslie_alistair(a)hotmail.com, z.figura12(a)gmail.com Distribution: --- wine-staging 64/32 is asking to install Gecko, but the installation attempt hangs at "loading down" with no progress. -- 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 49933] New: Alt+Tab doesn't work with gog installers
by WineHQ Bugzilla 20 Nov '24

20 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=49933 Bug ID: 49933 Summary: Alt+Tab doesn't work with gog installers Product: Wine Version: 5.18 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: rtentser(a)yandex.ru Distribution: --- I need to minimize other windows to see progress. When i try to alttab to an installer i'm getting "00f0:fixme:win:WINNLSEnableIME hwnd 0001011C enable 0: stub!" in terminal. When i try to alttab out of it i'm getting "00f0:fixme:win:WINNLSEnableIME hwnd 0001011C enable -1: stub!" So, i'm guessing, implementing WINNLSEnableIME will help. -- 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 71

[Bug 41392] New: directory.c: match_filename always false if mask contains "*?"
by wine-bugs＠winehq.org 19 Nov '24

19 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=41392 Bug ID: 41392 Summary: directory.c: match_filename always false if mask contains "*?" Product: Wine Version: 1.9.19 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: lcmecmac(a)gmail.com Distribution: --- A mask containing '*' followed by '?' will not match any files. For example, "dir *?" does not show any files in a non-empty directory. -- 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 1

[Bug 42407] New: LdrUnloadDll may recursion unload same dll
by wine-bugs＠winehq.org 19 Nov '24

19 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=42407 Bug ID: 42407 Summary: LdrUnloadDll may recursion unload same dll Product: Wine Version: 1.9.0 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: xiewei(a)deepin.com Distribution: --- Created attachment 57219 --> https://bugs.winehq.org/attachment.cgi?id=57219 LdrUnloadDll log file It may cause QQ crash.LdrUnloadDll funcation recursion unload Flash.ocx,and dsound.dll ldr.LoadCount to 0,then dsound.dll be unload.But QQ process is still using dsound.dll. Type: EXCEPTION_ACCESS_VIOLATION Error: Read address 0x7985C309 Address: 7985C309 -- 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 8

[Bug 40972] New: RtlVerifyVersionInfo returns wrong results
by wine-bugs＠winehq.org 19 Nov '24

19 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=40972 Bug ID: 40972 Summary: RtlVerifyVersionInfo returns wrong results Product: Wine Version: 1.9.14 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: marcinwiaz(a)poczta.onet.pl Distribution: --- Created attachment 55126 --> https://bugs.winehq.org/attachment.cgi?id=55126 Tiny test program - EXE and source (Note: this a different problem than in Bug 36143) Steps to reproduce: 1) Download and unpack Reproduce.exe, which is attached to this bug report 2) Configure Wine to emulate Windows 2003 (which has version number 5.2) for Reproduce.exe 3) Launch Reproduce.exe - you will see in the console window: CurrentOS is: 5.2 RequestedOS is >= than: 5.19 CurrentOS is < than RequestedOS (which is WRONG!) 4) Run true Microsoft Windows 2003 Server 5) Launch Reproduce.exe - you will see in the console window: CurrentOS is: 5.2 RequestedOS is >= than: 5.19 CurrentOS is >= than RequestedOS Explanation: Minor version numbers should be treated similarly to digits after a decimal point - so when comparing software versions: version 5.1 is < than version 5.2 version 5.11 is < than version 5.2 (note that 11 is > than 2 !) version 5.19 is < than version 5.2 (note that 19 is > than 2 !) So when comparing for VER_MINORVERSION or VER_SERVICEPACKMINOR inside the "VerifyVersionInfo" API, Microsoft uses a lexical comparison - which works as: lstrcmp(tostring(dwMinorVersion1),tostring(dwMinorVersion2)) or lstrcmp(tostring(wServicePackMinor1),tostring(wServicePackMinor2)) Please note, that it's still different than comparing real numbers, because: 5.20 = 5.2 but lstrcmp("5.20","5.2") != 0 So lstrcmp-like function must be used for VER_MINORVERSION and VER_SERVICEPACKMINOR checking inside the "RtlVerifyVersionInfo" function (which is called by the "VerifyVersionInfo" API in Wine) to be compatible with Microsoft Windows. Possible fix: in "version.c" file, add a new "version_compare_values_lexical" function. It should be based on the original "version_compare_values" function, but should use "lstrcmp" calls instead of integer comparisons. This "version_compare_values_lexical" function should be called instead of "version_compare_values" when checking for VER_MINORVERSION and VER_SERVICEPACKMINOR inside the "RtlVerifyVersionInfo" function. -- 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 5