October 2024 - Wine-bugs - list.winehq.org

[Bug 56372] New: musl based exp2() gives very inaccurate results on i686
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=56372 Bug ID: 56372 Summary: musl based exp2() gives very inaccurate results on i686 Product: Wine Version: 8.6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: msvcrt Assignee: wine-bugs(a)winehq.org Reporter: martin(a)martin.st CC: piotr(a)codeweavers.com Regression SHA1: ca0643875961ef62c76e0b9cbd0433002a948d9b Distribution: --- Since switching the msvcrt implementation of exp2() from internal to musl based, in ca0643875961ef62c76e0b9cbd0433002a948d9b, the accuracy on i686 has become quite bad, to the point of breaking a handful of tests in ffmpeg. The issue can be reproduced with the following test snippet: --- #include <stdio.h> #include <math.h> double input = 5.979471; int main(int argc, char* argv[]) { printf("exp2(%f) = %f\n", input, exp2(input)); return 0; } --- Prior to this commit, and with later versions on e.g. x86_64, the output is: exp2(5.979471) = 63.095753 However, on i686, the output now is: exp2(5.979471) = 62.968753 This is an error of around 0.2%, which is a couple orders of magnitudes larger than expected/tolerated. It's curious how this issue only appears on i686 though, i.e. it doesn't look like an issue with the musl exp2() implementation algorithm, but some component of it misbehaving on i686. -- 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 11

[Bug 56781] New: srcrrun: Dictionary setting item to object fails
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=56781 Bug ID: 56781 Summary: srcrrun: Dictionary setting item to object fails Product: Wine Version: 9.4 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: scrrun Assignee: wine-bugs(a)winehq.org Reporter: francisdb(a)gmail.com Distribution: --- Reproducer in vbscript Class C End Class Dim cls1: Set cls1 = new C Dim cls2: Set cls2 = new C Dim dict: Set dict = createobject("scripting.dictionary") Dim key: key = "test" dict.Add key, cls1 Set dict.Item(key) = cls2 Looks like this is not implemented https://gitlab.winehq.org/wine/wine/-/blob/master/dlls/scrrun/dictionary.c?… Without understanding the code this should maybe increase the reference count and delegate to dictionary_put_Item? -- 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 49703] New: Ghost Recon fails to start
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=49703 Bug ID: 49703 Summary: Ghost Recon fails to start Product: Wine Version: 5.14 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: opengl Assignee: wine-bugs(a)winehq.org Reporter: sramey40(a)gmail.com Distribution: --- The original Tom Clancy's Ghost Recon (on CD) seems to install fine, except for this error at the end of installation: wine start Setup.exe [sean(a)sean-pc-arch GRECON]$ 00d4:fixme:file:NtLockFile I/O completion on lock not implemented yet 00d4:fixme:sfc:SRSetRestorePointA 02EB66F0 02EB51E0 00d4:fixme:sfc:SRSetRestorePointA 02EB66F0 02EB51E0 00d4:err:ole:CoReleaseMarshalData StdMarshal ReleaseMarshalData failed with error 0x8001011d ^C But, when I try to run the game, nothing happens and this is the output on console: wine start GhostRecon.exe [sean(a)sean-pc-arch Ghost Recon]$ 00c0:err:winediag:wined3d_dll_init Setting multithreaded command stream to 0. ^C ^C marks where I hit ctrl-c to stop the process, as it seemed hung. I'm using a 32bit wineprefix with Windows 2000 or XP. I can't really do regression testing, as I can't get an older package of wine to install properly, and I also can't get wine to compile for even as new a version as wine-4.0. It complains about multiple definitions of hInstance and fails on compiling crypt32. Only the 5.0 releases will compile, and I have this bug in them. I've also noticed that a lot of older games seem broken. Seems like nobody has tested these games in a while. The latest test of the Ghost Recon CD version recorded in the App database is wine 1.5. Recently I had somebody help me out with Age of Empires 2: The Age of Kings CD version, and we discovered that the safedisc 1.x copy protection support was broken in wine since version 3.0 I think. Probably a good idea to try running some old games, especially ones on CD, to check for compatibility regressions. -- 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 55421] New: Fallout Tactics graphics glitches only with winehq-devel
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=55421 Bug ID: 55421 Summary: Fallout Tactics graphics glitches only with winehq-devel Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: darkranger_red(a)hotmail.com Distribution: --- The game 'Fallout Tactics: Brotherhood of Steel' is fully playable with Wine. But recently I am not sure from which Wine version, the game shows some noticeable graphics glitches, which the previous Wine version doesn't have. So I uninstalled Wine from Fedora's repo, and use the WineHQ's repo instead. It turns out that the game's graphics runs normally with winehq-stable(8.0.2), but has some glitches with winehq-devel(8.13). The attached screenshots show the launcher menu differences between Wine 8.0.2 and 8.13. Besides that, the in-game graphics also has more glitches with winehq-devel, but usually happen in very short time and difficult to capture the screenshots. OS: Fedora Linux 38 (Workstation Edition) x86_64 Kernel: 6.4.7-200.fc38.x86_64 Uptime: 2 hours, 9 mins Packages: 2703 (rpm) Shell: bash 5.2.15 Resolution: 1920x1080 DE: GNOME 44.3 WM: Mutter WM Theme: Adwaita Theme: Adwaita [GTK2/3] Icons: Adwaita [GTK2/3] Terminal: gnome-terminal CPU: AMD FX-8350 (8) @ 4.000GHz GPU: NVIDIA GeForce GTX 660 Memory: 5130MiB / 15932MiB -- 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 21

[Bug 56357] New: Zero sized writes using WriteProcessMemory succeed on Windows, but fail on Wine. testcase source
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=56357 Bug ID: 56357 Summary: Zero sized writes using WriteProcessMemory succeed on Windows, but fail on Wine. testcase source Product: Wine Version: 9.2 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: kernel32 Assignee: wine-bugs(a)winehq.org Reporter: admin(a)sewer56.dev Distribution: --- Created attachment 76082 --> https://bugs.winehq.org/attachment.cgi?id=76082 Precompiled sample C program. [ testcase source ] Zero sized writes using WriteProcessMemory succeed on Windows, but fail on Wine. Build the following program: ```c #include <windows.h> #include <stdio.h> int main() { // Allocate a buffer in memory char* buffer = (char*) VirtualAlloc(NULL, 1024, MEM_COMMIT | MEM_RESERVE, PAGE_READWRITE); if (buffer == NULL) { printf("Memory allocation failed\n"); return 1; } // Sample data to write (not actually used due to 0-byte length) const char* data = "Sample data"; SIZE_T writtenBytes = 0; // Get a handle to the current process HANDLE processHandle = GetCurrentProcess(); // Attempt to write memory with 0 byte length BOOL result = WriteProcessMemory(processHandle, buffer, data, 0, &writtenBytes); if (result == FALSE) { printf("WriteProcessMemory failed: %lu\n", GetLastError()); } else { printf("WriteProcessMemory succeeded, but 0 bytes were written.\n"); } VirtualFree(buffer, 0, MEM_RELEASE); return 0; } ``` Commandline: ``` x86_64-w64-mingw32-gcc write_process_memory.c -o write_process_memory.exe -lkernel32 ``` When running on Linux, this prints `WriteProcessMemory failed: 87`. When running on Windows, this succeeds `WriteProcessMemory succeeded, but 0 bytes were written`. Tested on x86_64 Archlinux using latest git version of Wine (wine-9.2-286-g232b18d820e). ----------- Extra Notes: I encountered this bug while unit testing a Rust DLL Injection library against Wine: https://github.com/OpenByteDev/dll-syringe/pull/19 -- 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 56054] New: Microsoft Safety Scanner crashes on exit on unimplemented function tbs.dll.GetDeviceIDString
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=56054 Bug ID: 56054 Summary: Microsoft Safety Scanner crashes on exit on unimplemented function tbs.dll.GetDeviceIDString Product: Wine Version: 9.0-rc2 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: alexhenrie24(a)gmail.com Distribution: --- Created attachment 75710 --> https://bugs.winehq.org/attachment.cgi?id=75710 Terminal output All you have to do to reproduce the crash is open the program and immediately close it. $ sha256sum MSERT.exe 8a885021b8c41fb8578cfa50a0c3a706c6b1b7fb7fd3faddcc48bd8eb7f8f091 -- 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 49089] New: nProtect Anti-Virus/Spyware 4.0 'tkpl2k64.sys' crashes on unimplemented function 'fltmgr.sys.FltBuildDefaultSecurityDescriptor'
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=49089 Bug ID: 49089 Summary: nProtect Anti-Virus/Spyware 4.0 'tkpl2k64.sys' crashes on unimplemented function 'fltmgr.sys.FltBuildDefaultSecurityDescriptor' Product: Wine Version: 5.7 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: ntoskrnl Assignee: wine-bugs(a)winehq.org Reporter: focht(a)gmx.net Distribution: --- Hello folks, encountered while revisiting bug 47170 Download: https://web.archive.org/web/20160510225518/http://avsd.nprotect.net/avs40/s… --- snip --- $ WINEDEBUG=+seh,+loaddll,+process wine ./nProtectSetup_AVS40.exe ... 0244:trace:loaddll:load_native_dll Loaded L"C:\\windows\\system32\\FLTMGR.SYS" at 0xe50000: PE builtin 0244:trace:loaddll:load_native_dll Loaded L"C:\\windows\\system32\\tkpl2k64.sys" at 0xe10000: native 0244:trace:seh:raise_exception code=c0000005 flags=0 addr=0xe38108 ip=e38108 tid=0244 0244:trace:seh:raise_exception info[0]=0000000000000000 0244:trace:seh:raise_exception info[1]=fffff78000000320 0244:trace:seh:raise_exception rax=fffff78000000320 rbx=0000000000e380dc rcx=00000000007d5050 rdx=00000000007d51b8 0244:trace:seh:raise_exception rsi=0000000000cef94c rdi=00000000007d3ea8 rbp=00000000000fbff8 rsp=0000000000cef8f8 0244:trace:seh:raise_exception r8=0000000000e26100 r9=00002b992ddfa232 r10=000000000023584c r11=00000000000fc0c0 0244:trace:seh:raise_exception r12=00000000007d5050 r13=00007fffffea4000 r14=00000000007d51b8 r15=0000000000000000 0244:trace:seh:call_vectored_handlers calling handler at 0x22cde0 code=c0000005 flags=0 0244:trace:seh:call_vectored_handlers handler at 0x22cde0 returned ffffffff 0244:trace:seh:raise_exception code=c0000096 flags=0 addr=0xe12eb0 ip=e12eb0 tid=0244 0244:trace:seh:raise_exception rax=fffffffffe537b79 rbx=0000000000e380dc rcx=00000000007d5050 rdx=00000000007d51b8 0244:trace:seh:raise_exception rsi=0000000000cef94c rdi=00000000007d3ea8 rbp=00000000000fbff8 rsp=0000000000cef838 0244:trace:seh:raise_exception r8=0000ffffffffffff r9=00002b992ddfa232 r10=000000000023584c r11=00000000000fc0c0 0244:trace:seh:raise_exception r12=00000000007d5050 r13=00007fffffea4000 r14=00000000007d51b8 r15=0000000000000000 0244:trace:seh:call_vectored_handlers calling handler at 0x22cde0 code=c0000096 flags=0 0244:trace:seh:call_vectored_handlers handler at 0x22cde0 returned ffffffff 0244:trace:seh:raise_exception code=c0000096 flags=0 addr=0xe12eb0 ip=e12eb0 tid=0244 0244:trace:seh:raise_exception rax=0000000000950330 rbx=0000000000e380dc rcx=0000000000000000 rdx=0000000000e222a0 0244:trace:seh:raise_exception rsi=0000000000cef94c rdi=00000000007d3ea8 rbp=00000000000fbff8 rsp=0000000000cef7d8 0244:trace:seh:raise_exception r8=00000000009503a2 r9=0000000000000016 r10=0000000000000000 r11=00000000009503d0 0244:trace:seh:raise_exception r12=00000000007d5050 r13=00007fffffea4000 r14=00000000007d51b8 r15=0000000000000000 0244:trace:seh:call_vectored_handlers calling handler at 0x22cde0 code=c0000096 flags=0 0244:trace:seh:call_vectored_handlers handler at 0x22cde0 returned ffffffff 0244:fixme:ntdll:NtQuerySystemInformation info_class SystemModuleInformation stub! 0244:fixme:ntoskrnl:PsSetCreateProcessNotifyRoutine stub: 0000000000E19C30 0 0244:fixme:fltmgr:FltRegisterFilter (00000000007D5050,0000000000E24D30,0000000000E26228): stub 0244:fixme:fltmgr:FltStartFiltering (00000000DEADBEAF): stub 0244:trace:seh:raise_exception code=80000100 flags=1 addr=0x7b00f665 ip=7b00f665 tid=0244 0244:trace:seh:raise_exception info[0]=0000000000e59000 0244:trace:seh:raise_exception info[1]=0000000000e59119 wine: Call from 0x7b00f665 to unimplemented function fltmgr.sys.FltBuildDefaultSecurityDescriptor, aborting --- snip --- Microsoft docs: https://docs.microsoft.com/en-us/windows-hardware/drivers/ddi/fltkernel/nf-… --- snip --- $ winedump -j import ~/.wine/drive_c/windows/system32/tkpl2k64.sys Contents of /home/focht/.wine/drive_c/windows/system32/tkpl2k64.sys: 98056 bytes Import Table size: 0000003c offset 00014738 ntoskrnl.exe Hint/Name Table: 000281E8 TimeDateStamp: 00000000 (Thu Jan 1 01:00:00 1970) ForwarderChain: 00000000 First thunk RVA: 00014070 Thunk Ordn Name 00014070 332 IoCreateDevice 00014078 1490 towlower 00014080 870 ProbeForRead 00014088 1452 _wcsnicmp 00014090 633 KeSetEvent 00014098 1322 ZwCreateFile 000140a0 1504 wcsrchr 000140a8 722 MmMapLockedPagesSpecifyCache 000140b0 143 ExSystemTimeToLocalTime 000140b8 964 PsTerminateSystemThread 000140c0 661 KeWaitForSingleObject 000140c8 1006 RtlCopyUnicodeString 000140d0 1360 ZwOpenProcess 000140d8 710 MmIsAddressValid 000140e0 850 ObfDereferenceObject 000140e8 1502 wcsncmp 000140f0 1428 ZwWriteFile 000140f8 49 DbgPrint 00014100 1450 _wcsicmp 00014108 558 KeInitializeEvent 00014110 613 KeReleaseSpinLock 00014118 524 KeAcquireSpinLockRaiseToDpc 00014120 531 KeBugCheckEx 00014128 885 PsGetCurrentProcessId 00014130 890 PsGetCurrentThreadId 00014138 341 IoCreateSymbolicLink 00014140 965 PsThreadType 00014148 842 ObReferenceObjectByHandle 00014150 1317 ZwClose 00014158 84 ExEventObjectType 00014160 878 PsCreateSystemThread 00014168 952 PsSetCreateProcessNotifyRoutine 00014170 402 IoIs32bitProcess 00014178 351 IoDeleteDevice 00014180 1086 RtlInitUnicodeString 00014188 1192 RtlTimeToTimeFields 00014190 353 IoDeleteSymbolicLink 00014198 1208 RtlUnicodeToMultiByteN 000141a0 974 RtlAnsiCharToUnicodeChar 000141a8 751 MmUnmapLockedPages 000141b0 1441 _stricmp 000141b8 1443 _strnicmp 000141c0 70 ExAllocatePoolWithTag 000141c8 88 ExFreePoolWithTag 000141d0 976 RtlAnsiStringToUnicodeString 000141d8 95 ExInitializeNPagedLookasideList 000141e0 937 PsLookupProcessByProcessId 000141e8 1390 ZwQuerySymbolicLinkObject 000141f0 160 ExpInterlockedPushEntrySList 000141f8 706 MmGetSystemRoutineAddress 00014200 1082 RtlInitAnsiString 00014208 1202 RtlUnicodeStringToAnsiString 00014210 1391 ZwQuerySystemInformation 00014218 159 ExpInterlockedPopEntrySList 00014220 690 MmBuildMdlForNonPagedPool 00014228 1364 ZwOpenSymbolicLinkObject 00014230 370 IoFreeMdl 00014238 655 KeUnstackDetachProcess 00014240 1061 RtlFreeUnicodeString 00014248 840 ObQueryNameString 00014250 1483 strncpy 00014258 1392 ZwQueryValueKey 00014260 377 IoGetCurrentProcess 00014268 502 IofCompleteRequest 00014270 114 ExQueryDepthSList 00014278 1057 RtlFreeAnsiString 00014280 928 PsGetVersion 00014288 647 KeStackAttachProcess 00014290 74 ExDeleteNPagedLookasideList 00014298 307 IoAllocateMdl 000142a0 1359 ZwOpenKey 000142a8 1430 __C_specific_handler offset 0001474c FLTMGR.SYS Hint/Name Table: 00028178 TimeDateStamp: 00000000 (Thu Jan 1 01:00:00 1970) ForwarderChain: 00000000 First thunk RVA: 00014000 Thunk Ordn Name 00014000 55 FltEnumerateVolumes 00014008 109 FltObjectDereference 00014010 152 FltStartFiltering 00014018 130 FltRegisterFilter 00014020 12 FltBuildDefaultSecurityDescriptor 00014028 29 FltCloseCommunicationPort 00014030 160 FltUnregisterFilter 00014038 121 FltQueryInformationFile 00014040 62 FltFreeSecurityDescriptor 00014048 33 FltCreateCommunicationPort 00014050 28 FltCloseClientPort 00014058 13 FltCancelFileOpen 00014060 95 FltGetVolumeName Done dumping /home/focht/.wine/drive_c/windows/system32/tkpl2k64.sys --- snip --- $ sha1sum nProtectSetup_AVS40.exe 913b33ab5c9477539d4d65b9f89e67be1a6b6c13 nProtectSetup_AVS40.exe $ du -sh nProtectSetup_AVS40.exe 36M nProtectSetup_AVS40.exe $ wine --version wine-5.7-177-gad1fad8a94 Regards -- 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 51599] New: cmd.exe incorrectly parses an all-whitespace line followed by a closing parenthesis
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=51599 Bug ID: 51599 Summary: cmd.exe incorrectly parses an all-whitespace line followed by a closing parenthesis Product: Wine Version: 6.14 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: programs Assignee: wine-bugs(a)winehq.org Reporter: alexhenrie24(a)gmail.com Distribution: --- Created attachment 70449 --> https://bugs.winehq.org/attachment.cgi?id=70449 Example batch file The attached batch file works on Windows, but Wine says: Can't recognize ')' as an internal or external command, or batch script. Wine's cmd.exe doesn't have a problem with entirely blank lines, but something about having spaces on the line is tripping it up. -- 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 56367] New: Tomb Raider 3 GOG crashes at start
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=56367 Bug ID: 56367 Summary: Tomb Raider 3 GOG crashes at start Product: Wine Version: 9.3 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: d3d Assignee: wine-bugs(a)winehq.org Reporter: dkk089(a)gmail.com Distribution: --- Created attachment 76108 --> https://bugs.winehq.org/attachment.cgi?id=76108 Tomb3.exe backtrace with Wine 9.3 Tomb Raider 3 GOG crashes in Wine 9.3 with the attached backtrace. Last known working version in AppDB is Wine 4.2. Working upwards from that identified Wine 5.7 as the last working release, and bisecting between 5.7 and 5.8 identified commit 100ca518055c0e9edb68aa370f15a2768cf2c74f as the cause. Reverting this commit from current master (1b32ac45f821ee1fe06a3dc4f903a81a190216c7) makes the game work normally. Tested with Windows version set to 98, WINEARCH is win32. This is also reproducible with the demo version available at https://archive.org/details/TombRaiderIiiAdventuresOfLaraCroftDemo . -- 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 10

[Bug 55897] New: cpython 3.12.0 crashes due to unimplemented CopyFile2
by WineHQ Bugzilla 16 Apr '25

16 Apr '25

https://bugs.winehq.org/show_bug.cgi?id=55897 Bug ID: 55897 Summary: cpython 3.12.0 crashes due to unimplemented CopyFile2 Product: Wine Version: 8.20 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: kernel32 Assignee: wine-bugs(a)winehq.org Reporter: tobias-winehq(a)23.gs Distribution: --- Newer Python versions (starting with 3.12.0) contain an optimized implementation of shutil.copy2 (https://devdocs.io/python~3.11/library/shutil#shutil.copy2) which uses kernel32.CopyFile2 in the background (Implementation: https://github.com/python/cpython/pull/105086/files#diff-db8ac5932616071392…) This leads to crashes if this method is used. Unfortunatly, this method is used by pip in some cases, so a random `wine pip install pysftp` leads to a crash... To reproduce: 1. Install Python 3.12.0 from https://www.python.org/ 2. Run `wine python -c 'import shutil; shutil.copy2("a", "b")'` -- 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 19