February 2019 - Wine-Bugs - list.winehq.org

[Bug 46435] New: F1 2011 crashes with a messed up rendering in DX9 mode
by wine-bugs＠winehq.org March 3, 2019

March 3, 2019

https://bugs.winehq.org/show_bug.cgi?id=46435 Bug ID: 46435 Summary: F1 2011 crashes with a messed up rendering in DX9 mode Product: Wine Version: 4.0-rc5 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: andrey.goosev(a)gmail.com Distribution: --- Created attachment 63213 --> https://bugs.winehq.org/attachment.cgi?id=63213 backtraces d3dx9_43 - builtin fixme:d3dx:D3DXGetShaderConstantTableEx Flags (0x20000) are not handled, yet! Crashes immediately after race starts. Backtrace is different each time. Rendering is normal and no crash with d3dx9_43 native,builtin -- 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 9

[Bug 44485] New: makes Delphi 7 debugger generate new exceptions by itself (regression)
by wine-bugs＠winehq.org March 3, 2019

March 3, 2019

https://bugs.winehq.org/show_bug.cgi?id=44485 Bug ID: 44485 Summary: makes Delphi 7 debugger generate new exceptions by itself (regression) Product: Wine Version: unspecified Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: development(a)winomega.com Distribution: --- Created attachment 60433 --> https://bugs.winehq.org/attachment.cgi?id=60433 WINEDEBUG=+seh Starting with commit a75e3f734d32bd98df9017f1a94819d2b6dca85d, the Delphi 7 debugger behaves erratically after catching an exception generated by the debugged program. It generates new floating point arithmetic exceptions, making debugging impossible and forces the debugger application to abort. The test program is a trivial application which raises an exception when clicking on a button. I'm attaching a WINEDEBUG=+seh log. Also, here's an annotated excerpt describing what I'm doing when each of the traces show up: I start the program and click on the button. Then get this: trace:seh:raise_exception code=eedfade flags=1 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=00450767 trace:seh:raise_exception info[1]=00473dbc trace:seh:raise_exception info[2]=004736c0 trace:seh:raise_exception info[3]=00426c58 trace:seh:raise_exception info[4]=0033ef44 trace:seh:raise_exception info[5]=0033ef08 trace:seh:raise_exception info[6]=0033edc4 trace:seh:raise_exception eax=7b428e39 ebx=00000018 ecx=0033ed30 edx=0033eda8 esi=0033ef08 edi=0033ef44 trace:seh:raise_exception ebp=0033ed88 esp=0033ed04 cs=0023 ds=13002b es=4d43002b fs=0063 gs=006b flags=00000212 I press F9 (run) to resume execution. Then get these: trace:seh:call_stack_handlers calling handler at 0x433637 code=eedfade flags=1 trace:seh:call_stack_handlers handler at 0x433637 returned 1 trace:seh:call_stack_handlers calling handler at 0x433648 code=eedfade flags=1 trace:seh:__regs_RtlUnwind code=eedfade flags=3 trace:seh:__regs_RtlUnwind calling handler at 0x7bc81710 code=eedfade flags=3 trace:seh:__regs_RtlUnwind handler at 0x7bc81710 returned 1 trace:seh:__regs_RtlUnwind calling handler at 0x433637 code=eedfade flags=3 trace:seh:raise_exception code=eedfae1 flags=0 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=0043362c trace:seh:raise_exception eax=7b428e39 ebx=0043362c ecx=0033e3d0 edx=0033e448 esi=00000000 edi=0033ed1c trace:seh:raise_exception ebp=0033e428 esp=0033e3a4 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00000212 trace:seh:__regs_RtlUnwind handler at 0x403844 returned 1 trace:seh:raise_exception code=eedfae0 flags=0 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=0043364d trace:seh:raise_exception eax=7b428e39 ebx=0043364d ecx=0033e890 edx=0033e90c esi=00000000 edi=0033f104 trace:seh:raise_exception ebp=0033e8e8 esp=0033e864 cs=330023 ds=002b es=002b fs=0063 gs=006b flags=00000212 trace:seh:raise_exception code=c0000092 flags=0 addr=0x7ecdb6be ip=7ecdb6c1 tid=0031 trace:seh:raise_exception eax=00117f68 ebx=7ed32000 ecx=0013e398 edx=0013e398 esi=0033dc40 edi=00117f68 trace:seh:raise_exception ebp=0033db98 esp=0033db70 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00010202 trace:seh:call_stack_handlers calling handler at 0x44f399 code=c0000092 flags=0 trace:seh:call_stack_handlers handler at 0x44f399 returned 1 trace:seh:call_stack_handlers calling handler at 0x44f467 code=c0000092 flags=0 trace:seh:call_stack_handlers handler at 0x44f467 returned 1 trace:seh:call_stack_handlers calling handler at 0x403664 code=c0000092 flags=0 trace:seh:call_stack_handlers handler at 0x403664 returned 1 trace:seh:call_stack_handlers calling handler at 0x433637 code=c0000092 flags=0 trace:seh:call_stack_handlers handler at 0x433637 returned 1 trace:seh:call_stack_handlers calling handler at 0x433648 code=c0000092 flags=0 trace:seh:raise_exception code=eedfae4 flags=0 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=0033d84c trace:seh:raise_exception info[1]=00473dfc trace:seh:raise_exception eax=7b428e39 ebx=00000004 ecx=0033d690 edx=0033d70c esi=0033d84c edi=0033db18 trace:seh:raise_exception ebp=0033d6e8 esp=0033d664 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00000212 A MessageBox is shown: "Debugger Exception Notification: Project xxx.exe raised exception class EInvalidOp with message 'Invalid floating point operation'. Process stopped. Use Step or Run to continue." I click OK. Debugger stops at vcl/Controls.pas class, at CallWindowProc; it appears as if the exception is triggered in CallWindowProc itself, but I can't tell for sure as the debugger environment is unreliable (if I were to single-step, it keeps triggering new exceptions). I press F9 again.... trace:seh:__regs_RtlUnwind code=c0000092 flags=2 trace:seh:__regs_RtlUnwind calling handler at 0x7bc81710 code=c0000092 flags=2 trace:seh:__regs_RtlUnwind handler at 0x7bc81710 returned 1 trace:seh:__regs_RtlUnwind calling handler at 0x44f399 code=c0000092 flags=2 trace:seh:raise_exception code=eedfae1 flags=0 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=0044f33f trace:seh:raise_exception eax=7b428e39 ebx=0044f33f ecx=0033d1b0 edx=0033d228 esi=00000000 edi=0033db18 trace:seh:raise_exception ebp=0033d208 esp=0033d184 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00000216 trace:seh:__regs_RtlUnwind handler at 0x403844 returned 1 trace:seh:__regs_RtlUnwind calling handler at 0x44f467 code=c0000092 flags=2 trace:seh:raise_exception code=eedfae1 flags=0 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=0044f453 trace:seh:raise_exception eax=7b428e39 ebx=0044f453 ecx=0033d1b0 edx=0033d228 esi=00000000 edi=0033db18 trace:seh:raise_exception ebp=0033d208 esp=0033d184 cs=110023 ds=002b es=33002b fs=0063 gs=7bc8006b flags=00000216 trace:seh:__regs_RtlUnwind handler at 0x403844 returned 1 trace:seh:__regs_RtlUnwind calling handler at 0x403664 code=c0000092 flags=2 trace:seh:raise_exception code=eedfae1 flags=0 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=00403669 trace:seh:raise_exception eax=7b428e39 ebx=00403669 ecx=0033d1b0 edx=0033d228 esi=00000000 edi=0033db18 trace:seh:raise_exception ebp=0033d208 esp=0033d184 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00000216 trace:seh:__regs_RtlUnwind handler at 0x403844 returned 1 trace:seh:__regs_RtlUnwind calling handler at 0x433637 code=c0000092 flags=2 trace:seh:raise_exception code=eedfae1 flags=0 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=0043362c trace:seh:raise_exception eax=7b428e39 ebx=0043362c ecx=0033d1b0 edx=0033d228 esi=00000000 edi=0033db18 trace:seh:raise_exception ebp=0033d208 esp=0033d184 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00000216 trace:seh:__regs_RtlUnwind handler at 0x403844 returned 1 trace:seh:raise_exception code=eedfae0 flags=0 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=0043364d trace:seh:raise_exception eax=7b428e39 ebx=0043364d ecx=0033d670 edx=0033d6ec esi=00000000 edi=0033fb94 trace:seh:raise_exception ebp=0033d6c8 esp=0033d644 cs=330023 ds=002b es=002b fs=0063 gs=006b flags=00000216 trace:seh:raise_exception code=c0000092 flags=0 addr=0x7ece39b7 ip=7ece39d6 tid=0031 trace:seh:raise_exception eax=00000003 ebx=7ed32000 ecx=7b43c5ac edx=00000000 esi=00141214 edi=00000001 trace:seh:raise_exception ebp=0033ca18 esp=0033c8a0 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00010297 trace:seh:call_stack_handlers calling handler at 0x44f399 code=c0000092 flags=0 trace:seh:call_stack_handlers handler at 0x44f399 returned 1 trace:seh:call_stack_handlers calling handler at 0x44f467 code=c0000092 flags=0 trace:seh:call_stack_handlers handler at 0x44f467 returned 1 trace:seh:call_stack_handlers calling handler at 0x403664 code=c0000092 flags=0 trace:seh:call_stack_handlers handler at 0x403664 returned 1 trace:seh:call_stack_handlers calling handler at 0x44f159 code=c0000092 flags=0 trace:seh:raise_exception code=eedfae4 flags=0 addr=0x7b43c5ac ip=7b43c5ac tid=0031 trace:seh:raise_exception info[0]=0033c57c trace:seh:raise_exception info[1]=00473e40 trace:seh:raise_exception eax=7b428e39 ebx=00000004 ecx=0033c3c0 edx=0033c43c esi=0033c57c edi=0033c848 trace:seh:raise_exception ebp=0033c418 esp=0033c394 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00000212 Again: "Debugger Exception Notification: Project xxx.exe raised exception class EInvalidOp with message 'Invalid floating point operation'. Process stopped. Use Step or Run to continue." I press OK. Debugged test app is terminated this time. Before exitting a seh error shows up: err:seh:setup_exception_record stack overflow 952 bytes in thread 0031 eip f73fb162 esp 00240f78 stack 0x240000-0x241000-0x340000 -- 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 9

[Bug 20104] New: Fujiprint/Fotokasten albums cannot be created due to missing richtext query interface
by wine-bugs＠winehq.org March 3, 2019

March 3, 2019

http://bugs.winehq.org/show_bug.cgi?id=20104 Summary: Fujiprint/Fotokasten albums cannot be created due to missing richtext query interface Product: Wine Version: 1.1.29 Platform: PC URL: http://www.fotokasten.de/index.php?ftkPage=static&ftkV iew=ftksoftware OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: richedit AssignedTo: wine-bugs(a)winehq.org ReportedBy: cweiske(a)cweiske.de Running the "fotokasten" photo album creation software fails to create certain photo books with the following error: > fixme:richedit:fnTextSrv_QueryInterface Unknown interface: > {01c25500-4268-11d1-883a-3c8b00c10000} > fixme:richedit:fnTextSrv_OnTxInplaceDeactivate 0x3a1e8e0: STUB How to reproduce: 1. Run software 2. Click on tab "Fotobücher" 3. Click on "Fotobücher" icon 4. Click on "Fun" 5. Click on "Fun A4" 6. A loading window appears but is hidden soon after. When using Fotobücher->Fotobücher->Soft->A4, the richtext window has not to be shown, and I get to the next window (meaning it works without richtext). -- 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.

1 9

[Bug 46677] New: Test Drive Unlimited 2 Crash at begins
by wine-bugs＠winehq.org March 3, 2019

March 3, 2019

https://bugs.winehq.org/show_bug.cgi?id=46677 Bug ID: 46677 Summary: Test Drive Unlimited 2 Crash at begins Product: Wine Version: 4.2 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: mrdeathjr28(a)yahoo.es Distribution: --- Created attachment 63629 --> https://bugs.winehq.org/attachment.cgi?id=63629 backtrace Game shows credits first screen and after this crash -- 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 3

[Bug 46629] New: Unravel Two crashes crashes during loading
by wine-bugs＠winehq.org March 3, 2019

March 3, 2019

https://bugs.winehq.org/show_bug.cgi?id=46629 Bug ID: 46629 Summary: Unravel Two crashes crashes during loading Product: Wine-staging Version: 4.1 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: ahmed.com(a)protonmail.com CC: leslie_alistair(a)hotmail.com, z.figura12(a)gmail.com Distribution: --- Created attachment 63528 --> https://bugs.winehq.org/attachment.cgi?id=63528 Unravel Two.log Unravel two crashes at the beginning of loading. Any suggestions will be appreciated. -- 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 46705] New: Revenant GOG don't running
by wine-bugs＠winehq.org March 3, 2019

March 3, 2019

https://bugs.winehq.org/show_bug.cgi?id=46705 Bug ID: 46705 Summary: Revenant GOG don't running Product: Wine Version: 3.21 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: evjeniy(a)mail.ru Distribution: --- Created attachment 63665 --> https://bugs.winehq.org/attachment.cgi?id=63665 setup.txt Wine version (staging) I use environment variables: WINEARCH = win32 WINEPREFIX = clean winecfg defaults Additional libraries: winetricks d3dx9_36 directplay quartz At the end of the installation process (similar to installing directx9.0c) "Runtime error (at 73: 1358):" "Runtime error (at 210: 800): Out Of Stack Range." "Runtime error (at 210: 806): Out Of Stack Range." "Out of Global Vars range." Console output (setup.txt) In winecfg I put the checkbox "Graphics -) Emulate a virtual desktop 640x480" During launch: FATAL ERROR. "DirectX Error D3DERR_SURFACENOTINVIDMEM in d: \ revenant \ DirectDraw.cpp at line 333" Console output (start.txt) -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 4

[Bug 10913] New: start.exe doesn't support optional process title argument
by wine-bugs＠winehq.org March 3, 2019

March 3, 2019

http://bugs.winehq.org/show_bug.cgi?id=10913 Summary: start.exe doesn't support optional process title argument Product: Wine Version: CVS/GIT Platform: Other URL: http://jedit.org OS/Version: other Status: NEW Keywords: download Severity: normal Priority: P2 Component: wine-programs AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com Running real Java apps under the latest Sun win32 JRE on Wine seems like a fine stress test, even though in many cases users should run with the Linux JRE instead. The Sun JRE and jEdit install ok, but jedit has many startup problems. Here's the first one. Invoking it with the batch file provided by jEdit fails. Here's what's in the batch file: start "jEdit startup" "c:\windows\system32\javaw.exe" -Xms64M -Xmx192M -jar "C:\Program Files\jEdit\jedit.jar" -reuseview %* And here's what happens when you run it: $ cd ~/.wine/drive_c/Program Files/jEdit $ ~/wine-git/wine cmd /c jedit.bat fixme:exec:SHELL_execute flags ignored: 0x00000500 trace:process:CreateProcessW app (null) cmdline L"jEdit startup c:\\windows\\system32\\javaw.exe -Xms64M -Xmx192M -jar \"C:\\Program Files\\jEdit\\jedit.jar\" -reuseview" trace:process:find_exe_file looking for L"jEdit" trace:process:find_exe_file Trying native exe L"c:\\windows\\system32\\jEdit.exe" ... trace:process:find_exe_file Trying built-in exe L"c:\\windows\\system32\\jEdit startup.exe" ... trace:process:find_exe_file looking for L"jEdit startup c:\\windows\\system32\\javaw.exe -Xms64M -Xmx192M -jar \"C:\\Program Files\\jEdit\\jedit.jar\" -reuseview" Application could not be started, or no application associated with the specified file. ShellExecuteEx failed: File not found So it seems that start takes an optional process title argument. Sure enough, http://www.microsoft.com/resources/documentation/windows/xp/all/proddocs/en… documents it. Who knew? -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching all bug changes.

1 20

[Bug 42508] New: start.exe does not detect its title argument when it should ( breaking .e.g URL opening in League of Legends)
by wine-bugs＠winehq.org March 3, 2019

March 3, 2019

https://bugs.winehq.org/show_bug.cgi?id=42508 Bug ID: 42508 Summary: start.exe does not detect its title argument when it should (breaking .e.g URL opening in League of Legends) Product: Wine Version: unspecified Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: programs Assignee: wine-bugs(a)winehq.org Reporter: hydratech(a)gmail.com Distribution: --- Created attachment 57397 --> https://bugs.winehq.org/attachment.cgi?id=57397 Trace showing League of Legends Client attempting to open a URL. When start.exe is invoked, its first argument in quotes that is not an argument of the application to be executed should be interpreted as the console title. Wine's start.exe fails to do so. -- How to reproduce -- Running the following line in Wine's cmd should open a URL in your browser: Z:\>start "" https://winehq.org/ the actual result is an application not found error. On Windows the same line opens the website as expected. -- Why this is a bug -- I know this was marked as fixed in bug 10913 but the reasoning was incorrect. The argument for stating the bug was fixed was, that when \"escaping your quotes\" in cmd to the title was parsed. This is however not consistent with with Windows behaviour. As explained at the end of that bug report, if you escape your quotes in Wine's cmd start.exe parses the argument as the console title, i.e.: Z:\>start \"hello\" cmd I will now clarify why this is incorrect by showing Windows' behaviour. The following command lines are not equivalent on Windows: C:\>start hello.exe vs. C:\>start "hello.exe" The latter will open a new cmd window titled hello.exe whereas the former attempts to execute an application named hello.exe. In other words cmd's start built-in parses its own command-line checking wether quotes where used. In fact escaping the quotes using back-slashes is invalid as it tries start a process named \hello.exe\. I have tested this on both Windows 7 and Windows XP. In other words, this is invalid: C:>start \"hello.exe\" To illustrate how unintuitive this Windows behaviour is: C:>start "C:\Program Files\Application\hello.exe" Few would guess this actually opens a console titled as such, instead of running an application. That's why I can understand the maintainers would think the bug was fixed. Wine will however correctly emulate unescaped quote-senstive behaviour with the echo built-in: Z:\>echo hoi hoi Z:\>echo "hoi" "hoi" We can see echo being "aware" of the quotes and correctly displaying them; behaviour identical to Windows' echo built-in. Looking at the wine source, start.exe relies on its "argv" parameter instead of using GetCommandLine() and thereby accessing the unmodified command-line. As a consequence the difference between a quoted argument and a non-quoted argument is lost as CreateProcess strips the quotes off the parameters when passing them to execve(). To quote the source of Wine's cmd.exe (wcmdmain.c): /* Can't use argc/argv as it will have stripped quotes from parameters * meaning cmd.exe /C echo "quoted string" is impossible */ It should now be evident that start.exe should either become a wrapper around cmd.exe's built-in start function, invoking cmd and let cmd itself implement start.exe's actual behaviour, or implement GetCommandLine() based argument parsing similar to cmd.exe's within start.exe, leading to duplicated command-line parsing code. Both are not ideal solutions. -- Why does this bug matter? -- There are actually Windows applications relying on this behaviour to open URLs in the user's browser. As an example, League of Legends executes the following shell command to open a URL: cmd.exe /c start "" "<url here>" In other words, League instructs start to launch an untitled shell window with a URL as application, effectively causing the shell to pass it to the browser. However, wine's start.exe interprets this as League trying to execute an empty string as executable name, with a URL as argument. I have attached the relevant output running League with WINEDEBUG=cmd,start,process,exec set. I will speculate that more applications will run into this exact same issue. League's client is based on CEF and it might very well be CEF which implements this style of external URL opening, furthermore there is documentation recommending the use of empty quotes as console title argument when invoking start to prevent it from accidentally parsing follow up arguments in quotes as the console title, think "C:\Program Files\etc...", establishing this as a pattern that might well have been adopted by other applications. See: https://ss64.com/nt/start.html I will also say that, given some instruction, I will be glad to help fixing this bug and write any necessary code. There is actually another issue with current start.exe's console title parsing which I will submit as a separate issue. That issue is however, significantly easier to fix. Finally this also means wine's current handling of command-line arguments might need fixing as: $ wine start '""' "http://winehq.og" will break in a fixed version of 'start.exe', while it should work. For this I will file yet another issue, although the necessity of this fix is debatable. -- 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 13

[Bug 46741] New: Cemu says "This program has encountered a serious problem " after opening Mario Maker
by wine-bugs＠winehq.org March 2, 2019

March 2, 2019

https://bugs.winehq.org/show_bug.cgi?id=46741 Bug ID: 46741 Summary: Cemu says "This program has encountered a serious problem" after opening Mario Maker Product: Wine Version: 4.0-rc5 Hardware: x86 OS: Mac OS X Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: 2021duffykod(a)isd47.org I'm using Cemu version 1.15.2 and I have open GL 4.1 on my Mac. Also I'm opening the file "Block.rpx" from the mario maker rom. -- 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 3

[Bug 46707] New: Microsoft Word 2019 freeze the whole screen on nouveau driver with CSMT enabled
by wine-bugs＠winehq.org March 1, 2019

March 1, 2019

https://bugs.winehq.org/show_bug.cgi?id=46707 Bug ID: 46707 Summary: Microsoft Word 2019 freeze the whole screen on nouveau driver with CSMT enabled Product: Wine Version: 4.2 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: minor Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: peathot(a)hotmail.com Distribution: --- Created attachment 63673 --> https://bugs.winehq.org/attachment.cgi?id=63673 Console output with CSMT enabled When using Microsoft Word 2019 (version 1902, build 11328.20100, installed via Office 365) for a while, the program, followed by the whole screen, will freeze. The mouse pointer is still movable, and I can SSH into the machine and kill WINWORD.EXE. After a moment, the screen will become usable again. By disabling CSMT in registry (HKCU/Software/Wine/Direct3D, csmt=dword:00000000), the problem doesn't occurred. The console output for both when CSMT is enabled and disabled is attached. Also, the kernel messages when CSMT is enabled is attached (there's no kernel messages when CSMT is disabled). This is Wine 4.2 on Xubuntu 18.04. The machine is Mac Mini early 2009 with Nvidia Geforce 9400m [1]. The graphic stack is mesa + nouveau (i.e. Nvidia proprietary driver is not installed). The MS Office is installed using the instructions from [2]. The first registry workaround is also used. Actually, the kernel messages suggests that it's nouveau fault. From my research, it seems like nouveau doesn't like OpenGL calls from multiple thread [3][4]. If it's actually nouveau's fault, I guess we could disable CSMT by default if we detect nouveau driver? [1] Yes, it's a bit old. But this same machine can run this software on Windows 10. So, it's not the problem. [2] https://appdb.winehq.org/objectManager.php?sClass=version&iId=35527&iTestin… [3] https://bugs.freedesktop.org/show_bug.cgi?id=92438#c39 [4] https://bugreports.qt.io/browse/QTBUG-73715 -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 4