April 2019 - Wine-bugs - list.winehq.org

[Bug 46905] New: Multiple eventlog monitor applications go haywire ( endless console spam, live-loop CPU) due to advapi32.ReadEventLogA/ W returning lasterror ERROR_CALL_NOT_IMPLEMENTED ( use ERROR_HANDLE_EOF instead)
by wine-bugs＠winehq.org 25 Nov '19

25 Nov '19

https://bugs.winehq.org/show_bug.cgi?id=46905 Bug ID: 46905 Summary: Multiple eventlog monitor applications go haywire (endless console spam, live-loop CPU) due to advapi32.ReadEventLogA/W returning lasterror ERROR_CALL_NOT_IMPLEMENTED (use ERROR_HANDLE_EOF instead) Product: Wine Version: 4.4 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: advapi32 Assignee: wine-bugs(a)winehq.org Reporter: focht(a)gmx.net Distribution: --- Hello folks, noticed this while looking at multiple SIMATIC WinCC V15.1 Runtime (installer) bugs -> bug 4690x After the point when multiple services got installed and configured as 'autostart', each time a prefix is bootstrapped (running an app), one of the services begins to endlessly spam the console and churn one cpu core. --- snip --- .. 0057:fixme:advapi:ReadEventLogW (0xcafe4242,0x00000006,0xffffffff,0xa2de0,0x00000200,0x307f5dc,0x307f5d8) stub 0057:fixme:advapi:GetOldestEventLogRecord (0xcafe4242,0x307f6d4) stub 0057:fixme:advapi:GetNumberOfEventLogRecords (0xcafe4242,0x307f6d0) stub 0057:fixme:advapi:ReadEventLogW (0xcafe4242,0x00000006,0xffffffff,0xa2de0,0x00000200,0x307f5dc,0x307f5d8) stub 0057:fixme:advapi:GetOldestEventLogRecord (0xcafe4242,0x307f6d4) stub ... <endless console spam in quick succession> --- snip --- Debugger session: --- snip --- Wine-dbg> info process pid threads executable (all id:s are in hex) 0000019a 1 'winedbg.exe' 00000013 4 'explorer.exe' 0000000e 10 'services.exe' 0000006b 5 \_ 'rpcss.exe' 00000065 6 \_ 'svchost.exe' 0000004a 4 \_ 'winedevice.exe' 00000041 4 \_ 'TraceConceptX.exe' 00000035 11 \_ 's7oiehsx64.exe' 0000003c 2 \_ 'S7EpaSRV64x.exe' 00000030 3 \_ 'plugplay.exe' 0000002a 4 \_ 'winedevice.exe' 0000001b 3 \_ 'SmartServer.exe' 00000008 1 'Start.exe' 00000080 10 \_ 'SIA2.exe' Wine-dbg> info thread ... 00000035 s7oiehsx64.exe 0000005b 0 00000058 0 00000057 0 00000056 0 00000055 0 00000053 0 00000052 2 00000051 15 00000039 0 00000038 0 00000036 0 Wine-dbg> attach 0x35 0x00007ff7ccea2b32 __make_stacks_executable+0xa1b2 in libpthread.so.0: syscall Wine-dbg>info share Module Address Debug info Name (91 modules) PE 7e0000- 807000 Export s7epascp64x PE 820000- 845000 Export s7oniepg64x PE 870000- 901000 Export s7ontcp PE 910000- 992000 Export s7epaapi64x PE 9a0000- 9b8000 Export s7epromapi64x PE 9c0000- f25000 Export mfc110u PE 2790000- 27a4000 Export s7otsiex64 PE 27c0000- 28af000 Export s7pcadpsrv ELF 7b400000- 7b895000 Dwarf kernel32<elf> \-PE 7b460000- 7b895000 \ kernel32 ELF 7bc00000- 7bdd4000 Dwarf ntdll<elf> \-PE 7bc80000- 7bdd4000 \ ntdll ELF 7c000000- 7c004000 Dwarf <wine-loader> PE 140000000- 140033000 Export s7oiehsx64 PE 180000000- 18002c000 Export s7onlinx64 ... Wine-dbg>b GetOldestEventLogRecord Breakpoint 1 at 0x00007ff7be6c96f0 GetOldestEventLogRecord [/home/focht/projects/wine/mainline-src/dlls/advapi32/eventlog.c:374] in advapi32 Wine-dbg>c Stopped on breakpoint 1 at 0x00007ff7be6c96f0 GetOldestEventLogRecord [/home/focht/projects/wine/mainline-src/dlls/advapi32/eventlog.c:374] in advapi32 GetOldestEventLogRecord () at /home/focht/projects/wine/mainline-src/dlls/advapi32/eventlog.c:374 374 { Wine-dbg>bt Backtrace: =>0 0x00007ff7be6c96f0 GetOldestEventLogRecord() [/home/focht/projects/wine/mainline-src/dlls/advapi32/eventlog.c:374] in advapi32 (0x000000000307f7a0) 1 0x00000000027e8467 EntryPoint+0xffffffffffffffff() in s7pcadpsrv (0x000000000307f7a0) 2 0x00000000027c4187 EntryPoint+0xffffffffffffffff() in s7pcadpsrv (0x000000000307f7a0) 3 0x00000000027c4b64 EntryPoint+0xffffffffffffffff() in s7pcadpsrv (0x000000000307fbc0) 4 0x00000000027c52df EntryPoint+0xffffffffffffffff() in s7pcadpsrv (0x000000000307fbc0) 5 0x00000000027d8e8b EntryPoint+0xffffffffffffffff() in s7pcadpsrv (0x000000000307fbc0) 6 0x00000000027d8cc0 EntryPoint+0xffffffffffffffff() in s7pcadpsrv (0x000000000307ffd0) 7 0x00000000027dcbe0 EntryPoint+0xffffffffffffffff() in s7pcadpsrv (0x000000000307ffd0) 8 0x00000000028108d0 EntryPoint+0xffffffffffffffff() in s7pcadpsrv (0x000000000307ffd0) 9 0x00000000028163ff EntryPoint+0x418a() in s7pcadpsrv (0x000000000307ffd0) 10 0x00000000028165a6 EntryPoint+0x4331() in s7pcadpsrv (0x000000000307ffd0) 11 0x000000007bd310ee call_thread_func+0x116() [/home/focht/projects/wine/mainline-src/dlls/ntdll/signal_x86_64.c:4454] in ntdll (0x000000000307ffd0) --- snip --- Trace log: --- snip --- $ WINEDEBUG=+seh,+relay,+service wine notepad >>log.txt 2>&1 ... 0095:Ret PE DLL (proc=0x2812274,module=0x27c0000 L"s7pcadpsrv.dll",reason=THREAD_ATTACH,res=(nil)) retval=1 0095:Starting thread proc 0x28164a4 (arg=0xdd6c0) ... 0095:Call KERNEL32.LoadLibraryExW(0306f790 L"C:\\Program Files\\Common Files\\Siemens\\Automation\\Simatic OAM\\bin\\s7pcadpsrv.dll",00000000,00000000) ret=02821d4d 0095:Ret KERNEL32.LoadLibraryExW() retval=027c0000 ret=02821d4d 0095:Call KERNEL32.SetThreadpoolWait(000dd6c0,00000188,00000000) ret=0281800e 0095:Ret KERNEL32.SetThreadpoolWait() retval=00000000 ret=0281800e 0095:Call KERNEL32.SetEvent(00000154) ret=0283d44c 0095:Ret KERNEL32.SetEvent() retval=00000001 ret=0283d44c ... 0095:Call advapi32.OpenEventLogW(00000000,000afef0 L"System") ret=027e8247 0095:fixme:advapi:OpenEventLogW ((null),L"System") stub 0095:Ret advapi32.OpenEventLogW() retval=cafe4242 ret=027e8247 ... 0095:Call KERNEL32.SetEvent(00000130) ret=027d98c2 0095:Ret KERNEL32.SetEvent() retval=00000001 ret=027d98c2 ... 0095:Call advapi32.GetOldestEventLogRecord(cafe4242,0306f6d4) ret=027e8467 0095:fixme:advapi:GetOldestEventLogRecord (0xcafe4242,0x306f6d4) stub 0095:Ret advapi32.GetOldestEventLogRecord() retval=00000001 ret=027e8467 ... 0095:Call advapi32.GetNumberOfEventLogRecords(cafe4242,0306f6d0) ret=027e84f7 0095:fixme:advapi:GetNumberOfEventLogRecords (0xcafe4242,0x306f6d0) stub 0095:Ret advapi32.GetNumberOfEventLogRecords() retval=00000001 ret=027e84f7 ... 0095:Call advapi32.ReadEventLogW(cafe4242,00000006,ffffffff,000de970,00000200,0306f5dc,0306f5d8) ret=027e85c3 0095:fixme:advapi:ReadEventLogW (0xcafe4242,0x00000006,0xffffffff,0xde970,0x00000200,0x306f5dc,0x306f5d8) stub 0095:Ret advapi32.ReadEventLogW() retval=00000000 ret=027e85c3 ... 0095:Call KERNEL32.GetLastError() ret=0281b6e4 0095:Ret KERNEL32.GetLastError() retval=00000078 ret=0281b6e4 ... 0095:Call KERNEL32.RaiseException(e06d7363,00000001,00000004,0306f560) ret=02812f70 0095:trace:seh:NtRaiseException code=e06d7363 flags=1 addr=0x7b497b7e ip=7b497b7e tid=0095 0095:trace:seh:NtRaiseException info[0]=0000000019930520 0095:trace:seh:NtRaiseException info[1]=000000000306f5e0 0095:trace:seh:NtRaiseException info[2]=00000000028919d0 0095:trace:seh:NtRaiseException info[3]=00000000027c0000 0095:trace:seh:NtRaiseException rax=000000007b478128 rbx=00000000028919d0 rcx=000000000306f350 rdx=0000000000000020 0095:trace:seh:NtRaiseException rsi=000000000306f560 rdi=000000000306f370 rbp=000000000306f4a0 rsp=000000000306f330 0095:trace:seh:NtRaiseException r8=0000000000000004 r9=000000000306f560 r10=0000000000000000 r11=0000000000000000 0095:trace:seh:NtRaiseException r12=00000000ffffffff r13=0000000002897898 r14=00000000cafe4242 r15=0000000000000006 ... 0095:Call ntdll.RtlUnwindEx(0306f8a0,027c52df,0306dd40,00000000,0306dde0,0306eb10) ret=02813589 0095:trace:seh:RtlUnwindEx code=80000029 flags=3 end_frame=0x306f8a0 target_ip=0x27c52df rip=000000007bd2f75d ... <sequence repeats again endlessly> ... 0095:Call advapi32.GetOldestEventLogRecord(cafe4242,0306f6d4) ret=027e8467 0095:fixme:advapi:GetOldestEventLogRecord (0xcafe4242,0x306f6d4) stub 0095:Ret advapi32.GetOldestEventLogRecord() retval=00000001 ret=027e8467 ... --- snip --- Not really an issue but worth mentioning: Wine's current stub logic with regards to event log entry numbers is a bit inconsistent when looking at old commit comments. * GetOldestEventLogRecord -> *OldestRecord = 0; return TRUE; (= success, 1 event) * GetNumberOfEventLogRecords -> *NumberOfRecords = 0; return TRUE; (= success, 0 events) GetOldestEventLogRecord -> "0" is a valid record number = 1 event. https://docs.microsoft.com/en-us/windows/desktop/EventLog/event-log-records --- quote --- Information about each event is stored in the event log in an event log record. The event log record includes time, type, and category information. For more information, see the EVENTLOGRECORD structure. The RecordNumber member of EVENTLOGRECORD contains the record number for the event log record. The very first record written to an event log is record number 1, and other records are numbered sequentially. If the record number reaches ULONG_MAX, the next record number will be 0, not 1; however, you use zero to seek to the record. If the Retention registry value is set to zero, the event records are overwritten when the maximum log size is reached. Therefore, the oldest record in an event log may not be record number 1. To identify the oldest record in the log, call the GetOldestEventLogRecord function. You can then call the GetNumberOfEventLogRecords function and add the returned value to the oldest record number to determine the newest record. You can read individual records from the event log using the ReadEventLog function. For more information, see Querying for Event Information. --- quote --- https://source.winehq.org/git/wine.git/commit/860c90a32217fb37a3b6c19ecd57f… --- quote --- author Hans Leidekker <hans(a)it.vu.nl> Wed, 23 Mar 2005 13:09:27 +0000 (13:09 +0000) committer Alexandre Julliard <julliard(a)winehq.org> Wed, 23 Mar 2005 13:09:27 +0000 (13:09 +0000) commit 860c90a32217fb37a3b6c19ecd57f4651085f3e1 Pretend there are 0 log records. Improve traces. --- quote --- But again, that's not the real problem. The service code does something similar like this one (pretty standard way of watching the eventlog for events). https://github.com/matrix207/VC/blob/master/eventlog/EventLogMonitor/EventL… https://www.google.com/search?q=ReadEventLog+ERROR_HANDLE_EOF Wine lacks event log implementation entirely, hence it returns lasterror 'ERROR_CALL_NOT_IMPLEMENTED' in 'ReadEventLogA/W'. https://source.winehq.org/git/wine.git/blob/HEAD:/dlls/advapi32/eventlog.c#… --- snip --- 610 /****************************************************************************** 611 * ReadEventLogW [ADVAPI32.@] 612 * 613 * See ReadEventLogA. 614 */ 615 BOOL WINAPI ReadEventLogW( HANDLE hEventLog, DWORD dwReadFlags, DWORD dwRecordOffset, 616 LPVOID lpBuffer, DWORD nNumberOfBytesToRead, DWORD *pnBytesRead, DWORD *pnMinNumberOfBytesNeeded ) 617 { 618 FIXME("(%p,0x%08x,0x%08x,%p,0x%08x,%p,%p) stub\n", hEventLog, dwReadFlags, 619 dwRecordOffset, lpBuffer, nNumberOfBytesToRead, pnBytesRead, pnMinNumberOfBytesNeeded); 620 621 SetLastError(ERROR_CALL_NOT_IMPLEMENTED); 622 return FALSE; 623 } --- snip --- Unfortunately various applications can't handle this lasterror code. They only check for 'ERROR_INSUFFICIENT_BUFFER' and 'ERROR_HANDLE_EOF' lasterrors and go haywire otherwise. To help these applications I propose to change the lasterror to 'ERROR_HANDLE_EOF'. With the change applied, the service no longer misbehaves: --- snip --- ... 0095:Call advapi32.GetOldestEventLogRecord(cafe4242,0306f6d4) ret=027e8467 0095:fixme:advapi:GetOldestEventLogRecord (0xcafe4242,0x306f6d4) stub 0095:Ret advapi32.GetOldestEventLogRecord() retval=00000001 ret=027e8467 ... 0095:Call advapi32.GetNumberOfEventLogRecords(cafe4242,0306f6d0) ret=027e84f7 0095:fixme:advapi:GetNumberOfEventLogRecords (0xcafe4242,0x306f6d0) stub 0095:Ret advapi32.GetNumberOfEventLogRecords() retval=00000001 ret=027e84f7 ... 0095:Call advapi32.ReadEventLogW(cafe4242,00000006,ffffffff,000dd530,00000200,0306f5dc,0306f5d8) ret=027e85c3 0095:fixme:advapi:ReadEventLogW (0xcafe4242,0x00000006,0xffffffff,0xdd530,0x00000200,0x306f5dc,0x306f5d8) stub 0095:Ret advapi32.ReadEventLogW() retval=00000000 ret=027e85c3 ... 0095:Call KERNEL32.GetLastError() ret=027c4690 0095:Ret KERNEL32.GetLastError() retval=00000026 ret=027c4690 ... 0095:Call advapi32.NotifyChangeEventLog(cafe4242,00000190) ret=027e8657 0095:fixme:advapi:NotifyChangeEventLog (0xcafe4242,0x190) stub 0095:Ret advapi32.NotifyChangeEventLog() retval=00000001 ret=027e8657 ... 0095:Call KERNEL32.WaitForMultipleObjects(00000002,000aea30,00000000,ffffffff) ret=027e694d <blocked> --- snip --- The blocking behaviour is intended and by design. It also avoids any further console spam/cpu bound polling due to APIs failing. $ sha1sum SIMATIC_WinCC_Runtime_Advanced_V15_1.exe db1f97bb648b62fa1c5d974d7f2bcb6b4a9fd786 SIMATIC_WinCC_Runtime_Advanced_V15_1.exe $ du -sh SIMATIC_WinCC_Runtime_Advanced_V15_1.exe 1.3G SIMATIC_WinCC_Runtime_Advanced_V15_1.exe $ wine --version wine-4.4-188-gc988910cae 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.

2 7

[Bug 43286] New: Qt 5.7.0 based: Cannot read UDP packets: "errno 90, ( Message too long)"
by wine-bugs＠winehq.org 25 Nov '19

25 Nov '19

https://bugs.winehq.org/show_bug.cgi?id=43286 Bug ID: 43286 Summary: Qt 5.7.0 based: Cannot read UDP packets: "errno 90, (Message too long)" Product: Wine Version: 2.11 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: winsock Assignee: wine-bugs(a)winehq.org Reporter: winehq(a)supradigital.org Distribution: --- Created attachment 58625 --> https://bugs.winehq.org/attachment.cgi?id=58625 Binary, source code, debug log In any program using Qt 5.7.0, if you try to read a udp datagram packet, the call would always return size 0 and leave you with an empty peer-address/port as well. I discovered the problem in the launcher for my game which uses UDP to ping various servers. I wrote a minimal UDP server to which you can connect with netcat (or anything) and it easily reproduces the problem. Attached you'll find a small archive containing: - minimal udp test server - source code for that server - debug log with +winsock showing the error - the client netcat command to cause that error Note: I have not checked a simple raw UDP server using the winsock API directly. The problem exists at least since wine v1.6.2 -- 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 28196] New: Test Drive Unlimited: Water is always rendered on top of everything.
by wine-bugs＠winehq.org 25 Nov '19

25 Nov '19

http://bugs.winehq.org/show_bug.cgi?id=28196 Summary: Test Drive Unlimited: Water is always rendered on top of everything. Product: Wine Version: 1.3.26 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d AssignedTo: wine-bugs(a)winehq.org ReportedBy: samurai_no_densetsu(a)yahoo.es On Test Drive Unlimited water is always rendered on top of any other object, no matter how far or high the object is, in fact one can see the rivers and sea over the road/mountains being miles away from the sea/river. This issue happens since the very first runnable wine version (1.1.x), and I belive that it wasn't reported earlier because it was confused by most users with bug 7991, but definitely not the same issue. I have tested it with the following hardware (DX10 capable) and the bug shows clearly: ------------------------[ snip ]------------------------------------ direct rendering: Yes server glx version string: 1.4 client glx version string: 1.4 GLX version: 1.4 OpenGL vendor string: NVIDIA Corporation OpenGL renderer string: GeForce 9300M GS/PCI/SSE2 OpenGL version string: 3.3.0 NVIDIA 270.41.06 OpenGL shading language version string: 3.30 NVIDIA via Cg compiler ------------------------[ snip ]------------------------------------ There are also reports on the appdb of users with a Geforce 9600GT or ATI HD4870 having the same issue (both cards are DX10 capable), and a lot of screenshots with the "flood". -- 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 36

[Bug 46453] New: The Testament of Sherlock Holmes crashes while loading level with builtin d3dx9_43
by wine-bugs＠winehq.org 25 Nov '19

25 Nov '19

https://bugs.winehq.org/show_bug.cgi?id=46453 Bug ID: 46453 Summary: The Testament of Sherlock Holmes crashes while loading level with builtin d3dx9_43 Product: Wine Version: 3.20 Hardware: x86-64 OS: Linux Status: NEW Keywords: regression Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: andrey.goosev(a)gmail.com CC: hverbeet(a)gmail.com Regression SHA1: 949dbbd31f450178c90ea8267097a975b77c3219 Distribution: --- fixme:dbghelp_dwarf:dwarf2_parse_subprogram Unhandled Tag type 0x1 at ctx(0x33b4ec,L"d3dx9_43<elf>"), for debug_info(abbrev:0x6d3303ac,symt:0x6d27b524) fixme:dbghelp_dwarf:dwarf2_parse_const_type Unsupported children Set d3dx9_43 to native,builtin helps. -- 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 9

[Bug 47082] New: Unhandled exception: unimplemented function authz.dll.AuthzFreeResourceManager called in 32-bit code (0x7b445f5b).
by wine-bugs＠winehq.org 25 Nov '19

25 Nov '19

https://bugs.winehq.org/show_bug.cgi?id=47082 Bug ID: 47082 Summary: Unhandled exception: unimplemented function authz.dll.AuthzFreeResourceManager called in 32-bit code (0x7b445f5b). Product: Wine Version: 4.6 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: terry(a)beam.ltd.uk Distribution: --- Created attachment 64274 --> https://bugs.winehq.org/attachment.cgi?id=64274 Backtrace An un-implemented function when using a program called passpad.exe an MRI imaging viewer: Clarity English, OsiriX Viewer. This is using Fedora29 wine. -- 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 7

[Bug 39614] New: Wine initialises a Critical Section for Mutexes by NULL instead of (void*)-1
by wine-bugs＠winehq.org 25 Nov '19

25 Nov '19

https://bugs.winehq.org/show_bug.cgi?id=39614 Bug ID: 39614 Summary: Wine initialises a Critical Section for Mutexes by NULL instead of (void*)-1 Product: Wine Version: 1.7.55 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: EoD(a)xmw.de Distribution: --- After initialising a critical section for mutexes by using static CRITICAL_SECTION Mutex = {(PCRITICAL_SECTION_DEBUG)-1, -1, 0, 0, 0, 0}; wine crashed in "RtlpWaitForCriticalSection()". According to http://locklessinc.com/articles/pthreads_on_windows/ this is a valid way of initialising the critical section. However, Wine seems to initialize it by NULL instead of (void*)-1: http://source.winehq.org/git/wine.git/blob/e8b9bda53ac39c913ddc93de0308f0c8… This is a re-reporting of an issue on github: https://github.com/apitrace/apitrace/issues/397 with a backtrace in https://gist.github.com/EoD/cf759116a4af8744bdb7 -- 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 7

[Bug 38421] New: Windows Media Player 9 & 10: mp3 sound plays too fast
by wine-bugs＠winehq.org 25 Nov '19

25 Nov '19

https://bugs.winehq.org/show_bug.cgi?id=38421 Bug ID: 38421 Summary: Windows Media Player 9 & 10: mp3 sound plays too fast Product: Wine Version: 1.7.34 Hardware: x86 OS: Linux Status: NEW Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: dimesio(a)earthlink.net Distribution: --- To reproduce, install WMP10 with winetricks and play an mp3 file. Setting version based on regression test results, but the problem is still present in wine-1.7.40-156-gba00e8e. Regression test results: 5c91d011bbdff0cac6535c866ef5498b1ddc42c7 is the first bad commit commit 5c91d011bbdff0cac6535c866ef5498b1ddc42c7 Author: Nikolay Sivov <nsivov(a)codeweavers.com> Date: Mon Dec 29 00:13:56 2014 +0300 msdmo: Registry API functions don't return HRESULT codes. :040000 040000 dd337587dde868375ad75d4fd6adcc590453375c 35cf0352b6c68488c4b17f2064575eec35eb25aa M dlls The patch can't be reverted cleanly. The results seemed odd, so I ran the regression test a second time and got the same result. Winetricks quartz devenum works around it. -- 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 15

[Bug 47023] New: Warframe in-game screenshot fails if screenshot directory already exists
by wine-bugs＠winehq.org 25 Nov '19

25 Nov '19

https://bugs.winehq.org/show_bug.cgi?id=47023 Bug ID: 47023 Summary: Warframe in-game screenshot fails if screenshot directory already exists Product: Wine Version: 4.6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: shell32 Assignee: wine-bugs(a)winehq.org Reporter: wine.bugzilla(a)johnthomson.fastmail.com.au Distribution: --- Created attachment 64181 --> https://bugs.winehq.org/attachment.cgi?id=64181 SHCreateDirectoryExW already exists demo application source The Warframe in-game screenshot (F6) utility fails to save a screenshot if the directory it tries to use already exists. These are: $user\My Pictures\\Warframe $warframe_install_location\\Downloaded\\Public\\Captures $warframe_install_location\\Downloaded\\Public $user\\Local Settings\\Application Data\\Warframe\\Captures If one of these folders does not exist, it is created, and one screenshot can be saved. I believe the issue stems from SHCreateDirectoryExW returning ERROR_CANCELLED, but have not had time to compile Wine with a fix yet. 0037:Call shell32.SHCreateDirectoryExW(00020060,09df9270 L"C:\\users\\john\\My Pictures\\Warframe",00000000) ret=1401353f4 ... 0037:fixme:shell:SHCreateDirectoryExW Show system error message, creating path L"C:\\users\\john\\My Pictures\\Warframe", failed with error 183 0037:Ret shell32.SHCreateDirectoryExW() retval=000004c7 ret=1401353f4 When the folder exists, SHCreateDirectoryExW returns with ERROR_CANCELLED 1223 (0x4C7) https://github.com/wine-mirror/wine/blob/fabde842ede40fc53cf6f5743d5b477841… My simple test program with a non-null Hwnd returned error ERROR_ALREADY_EXISTS 183 on Windows 7, and ERROR_CANCELLED 1223 in Wine, but I am not confident in my quick testing. The MSDN docs do not mention this https://docs.microsoft.com/en-us/windows/desktop/api/shlobj_core/nf-shlobj_… I would like for someone to test the return value of SHCreateDirectoryExW with a Hwnd for an exiting directory on Windows. -- 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 11

[Bug 46655] New: Adobe Photoshop CC 2015.5: Tabs show completely black
by wine-bugs＠winehq.org 25 Nov '19

25 Nov '19

https://bugs.winehq.org/show_bug.cgi?id=46655 Bug ID: 46655 Summary: Adobe Photoshop CC 2015.5: Tabs show completely black Product: Wine Version: 4.0 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: gdi32 Assignee: wine-bugs(a)winehq.org Reporter: winepala(a)tradermail.info Distribution: --- Created attachment 63587 --> https://bugs.winehq.org/attachment.cgi?id=63587 screenshot This is a regression since wine-3.0 where I last saw it working. Photoshop panels render to pitch black but can still be interacted with. Confirmed broken in wine/stable (4.0) and wine/master (currently tag:wine-4.2) I did a git bisect from wine-3.0 (good) to wine-3.10 (bad) and landed on these two commits. The last one I could not tag as it causes Photoshop to crash. commit 99b64f74320ba11b282a33aad2edc28551501875 (refs/bisect/bad) Date: Wed Apr 4 15:12:55 2018 +0200 gdi32: Build the region object directly instead of making a copy in CreatePolyPolygonRgn. commit 97f996744eafc4ad640ca3b9d97016625443c8cd (HEAD) (crashes) Date: Wed Apr 4 12:38:29 2018 +0200 gdi32: Pre-allocate the points array in CreatePolyPolygonRgn. -- 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 18

[Bug 22909] New: Drawn text overlapps after the second character.
by wine-bugs＠winehq.org 24 Nov '19

24 Nov '19

http://bugs.winehq.org/show_bug.cgi?id=22909 Summary: Drawn text overlapps after the second character. Product: Wine Version: 1.1.43 Platform: x86 URL: http://www.subsystems.com/tecfe.zip OS/Version: Linux Status: NEW Keywords: download, regression Severity: normal Priority: P2 Component: gdi32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: leslie_alistair(a)hotmail.com CC: aric(a)codeweavers.com Created an attachment (id=28340) --> (http://bugs.winehq.org/attachment.cgi?id=28340) Screenshot Open the TECF Demo app and open the demo.rtf file. See screenshot of results. Good: 1.1.42 Bad: 1.1.43 The issue appears to be caused by the change in the gdi32/bidi.c or changes in usp10 between the above versions. -- 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 13