January 2021 - Wine-Bugs - list.winehq.org

[Bug 46940] New: GTA 5 clients cannot connect to each other behind home router
by wine-bugs＠winehq.org Jan. 25, 2021

Jan. 25, 2021

https://bugs.winehq.org/show_bug.cgi?id=46940 Bug ID: 46940 Summary: GTA 5 clients cannot connect to each other behind home router Product: Wine Version: 4.5 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: electrodexsnet(a)gmail.com Distribution: --- When the game is running, it seems to be unable to do anything that requires the ability to host a direct connection. In most circumstances this is not noticeable as the game will fall back to using rockstar servers when direct is not available. However, if you have multiple players behind the same home router, they will experience symptoms of this issue, as the game expects to be able to use p2p/direct connectivity between clients on a home network. This includes: * Inability to join an invite-only or friend-only session hosted by either * Inability to join a public (or hosted by another player) session if either player is already in it; as soon as one connects, the other cannot join, and vice-versa. This can be worked around consistently by disabling UPnP on the home router. Workaround only works for joining public sessions. * Inability to join a heist hosted by either (even when started from a public or 3rd-party hosted session) These were tested and confirmed with 4 accounts at 4 computers at 2 locations. Tested with respective firewalls enabled/disabled. Players at different locations could connect to each other without issue; players at the same location could not. This issue does not manifest if one of the computers is Windows [tested W10 build 1803]. It can join or be joined to local friend-only/private sessions as you'd expect. ====Steps used to produce logs (on Majaro Linux)==== (Did steps on 2 computers running separate instances of GTA 5) sudo pacman -S --needed --noconfirm wine-staging wine-mono wine_gecko winetricks lib32-gtk3 lib32-mpg123 mpg123 yay WINEPREFIX=~/.wine_vulkan winetricks -q allfonts steam yay -S --noconfirm dxvk-bin WINEPREFIX=~/.wine_vulkan setup_dxvk install WINEDLLOVERRIDES=winedbg.exe=d WINEPREFIX=~/.wine_vulkan wine Steam.exe Launch game in Story Mode, then launch a Friend-only Online session. Once both are at this point, try connecting either of them to the other. GTA On-screen Messages: Alert: Getting GTA Online session details Alert: Unable to connect to game session. ======================== This affects both wine and wine-staging, but I could not get logs from vanilla wine as GTA 5 in recent versions does not get past the socialclub launcher. This also affects Proton. I've limited the logs scope to just wine for this report, but you can find the related Proton reports here: https://github.com/ValveSoftware/Proton/issues/37#issuecomment-445510575 https://github.com/ValveSoftware/Proton/issues/37#issuecomment-478411346 -- 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 14

[Bug 31881] New: Silverlight 5.x can't individualize
by wine-bugs＠winehq.org Jan. 25, 2021

Jan. 25, 2021

http://bugs.winehq.org/show_bug.cgi?id=31881 Bug #: 31881 Summary: Silverlight 5.x can't individualize Product: Wine Version: 1.5.14 Platform: x86 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: ehoover(a)mines.edu Depends on: 31589 Classification: Unclassified After working around Bug #31589 (attachment 41923) Silverlight 4.x is able to individualize; however, Silverlight 5.x still has problems. It appears to create the individualization files and then delete them after it fails to connect to a named pipe: 0059:KERNEL32.CreateFileA(020ab990 "\\\\.\\pipe\\lrpc\\keysvc",c0000000,00000000,00000000,00000003,00000000,00000000) ret=7e87340e 0059:Ret KERNEL32.CreateFileA() retval=ffffffff ret=7e87340e This named pipe business is not attempted with Silverlight 4.x, so creating this pipe would probably do the trick. A little googling finds this documentation: http://www.hsc.fr/ressources/articles/win_net_srv/msrpc_cryptsvc.html -- 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 18

[Bug 7106] Need for Speed 3 autorun crashes when starting setup
by WineHQ Bugzilla Jan. 25, 2021

Jan. 25, 2021

https://bugs.winehq.org/show_bug.cgi?id=7106 --- Comment #31 from Chebanenko Igor <chebanenkoigor93(a)gmail.com> --- Creating a Child Process with Redirected Input and Output https://docs.microsoft.com/en-us/windows/win32/procthread/creating-a-child-… Get console handle https://stackoverflow.com/questions/3859276/get-console-handle How to run child process in the same console in C# (part 1) https://csharp.today/how-to-run-child-process-in-the-same-console-in-c/ -- 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 0

[Bug 7106] Need for Speed 3 autorun crashes when starting setup
by WineHQ Bugzilla Jan. 25, 2021

Jan. 25, 2021

https://bugs.winehq.org/show_bug.cgi?id=7106 --- Comment #30 from Chebanenko Igor <chebanenkoigor93(a)gmail.com> --- (In reply to Anastasius Focht from comment #29) > Hello Igor, > > --- quote --- > Wine 6.0 was released. Any updates? > --- quote --- > > the bug is still marked as 'staged' which means the patch or another > solution has not been merged into mainline yet. Use Wine-Staging if you want > to play this game. > > Regards I found this: Console Handles https://docs.microsoft.com/en-us/windows/console/console-handles Console-Docs (GitHub) https://github.com/MicrosoftDocs/Console-Docs How to detect if Console.In (stdin) has been redirected? https://stackoverflow.com/questions/3453220/how-to-detect-if-console-in-std… Console Handles and Standard Handles https://github.com/rprichard/win32-console-docs But I don't know if that 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 0

[Bug 47832] New: FindFirstFileExW believes every directory entry has been read if NtQueryDirectoryFile underfills buffer
by WineHQ Bugzilla Jan. 25, 2021

Jan. 25, 2021

https://bugs.winehq.org/show_bug.cgi?id=47832 Bug ID: 47832 Summary: FindFirstFileExW believes every directory entry has been read if NtQueryDirectoryFile underfills buffer Product: Wine Version: unspecified Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: kernel32 Assignee: wine-bugs(a)winehq.org Reporter: qsniyg(a)mail.com Distribution: --- When using USVFS (through ModOrganizer 2), it will hook NtQueryDirectoryFile, and its implementation will occasionally underfill the buffer, even if there are more directory entries to be read. Due to this check in FindFirstFileExW: if (!has_wildcard || info->data_len < info->data_size - max_entry_size) { ... info->data_size = 0; /* we read everything */ } It will stop FindNextFileW from reading more. Windows doesn't seem to exhibit this behavior. Removing the `info->data_len < info->data_size - max_entry_size` check allows it to work properly. Unfortunately I wasn't able to find any reference to why that check exists. My best guess after looking through the source code is that it's an optimization, as the code should still run fine if wine's NtQueryDirectoryFile implementation underfills the buffer (therefore indicating that there are no more files needed to be read). So far I've tested many various applications with the check patched out, with no obvious 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 15

[Bug 48445] New: Linux Mint Cinnamon
by WineHQ Bugzilla Jan. 25, 2021

Jan. 25, 2021

https://bugs.winehq.org/show_bug.cgi?id=48445 Bug ID: 48445 Summary: Linux Mint Cinnamon Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: henadzitru(a)yandex.ru Distribution: --- Created attachment 66225 --> https://bugs.winehq.org/attachment.cgi?id=66225 game Crashes when you install the game World of Tanks -- 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 50040] New: crash of MSMoney
by WineHQ Bugzilla Jan. 24, 2021

Jan. 24, 2021

https://bugs.winehq.org/show_bug.cgi?id=50040 Bug ID: 50040 Summary: crash of MSMoney Product: Wine Version: 3.0.4 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: fbmines(a)gmail.com Distribution: --- Created attachment 68488 --> https://bugs.winehq.org/attachment.cgi?id=68488 crash log Hello, MsMoney run with Wine 3.0.4 and earlier versions crashed randomly but quite frequently when calling a function of the Investment windows or a report, specially the Historic values of the Investments statements. Attached the log of the crash. Thanks for your 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 6

[Bug 33679] New: WordPerfect X6 trial does not install
by wine-bugs＠winehq.org Jan. 24, 2021

Jan. 24, 2021

http://bugs.winehq.org/show_bug.cgi?id=33679 Bug #: 33679 Summary: WordPerfect X6 trial does not install Product: Wine Version: 1.5.30 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: blocker Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: erik(a)familiendam.dk Classification: Unclassified Created attachment 44584 --> http://bugs.winehq.org/attachment.cgi?id=44584 Trace result from the installation attempt. Tried installing a trial version of WordPerfect Office X6 I downloaded from Corel's homepage on 29-may-13. It unpacks the .msi files, but after this it fails, getting no further. I first tried with wine 1.4.0 since that is what is provided as default when you install wine on a Ubuntu 12.10 dist, subsequently I updated it to 1.5.30, but, alas, to no avail - same result. -- 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 10

[Bug 50556] New: Wine-Gecko debug channel 'gecko' doesn't work (win32 API 'SetEnvironmentVariable' has no effect on CRT 'getenv')
by WineHQ Bugzilla Jan. 24, 2021

Jan. 24, 2021

https://bugs.winehq.org/show_bug.cgi?id=50556 Bug ID: 50556 Summary: Wine-Gecko debug channel 'gecko' doesn't work (win32 API 'SetEnvironmentVariable' has no effect on CRT 'getenv') Product: Wine Version: 6.0 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: mshtml Assignee: wine-bugs(a)winehq.org Reporter: focht(a)gmx.net Distribution: --- Hello folks, found by chance while checking some issues with Wine-Gecko. Many years ago one had to manually set/export 'NSPR_LOG_MODULES=all:5' environment variable to get more verbosity from Gecko engine. Since commit https://source.winehq.org/git/wine.git/commitdiff/f9cbbab53b27ab5b9801d5635… ("mshtml: Added gecko debug channel to control Gecko logging."), part of Wine 1.1.12 release, this wasn't necessary anymore. After that, 'WINEDEBUG=+gecko' achieves the same. Wine source: https://source.winehq.org/git/wine.git/blob/2d4dd4252b0cf6526b3cc8194cce642… --- snip --- 443 static void set_environment(LPCWSTR gre_path) 444 { 445 size_t len, gre_path_len; 446 int debug_level = 0; 447 WCHAR *path, buf[20]; 448 const WCHAR *ptr; 449 450 SetEnvironmentVariableW(L"XPCOM_DEBUG_BREAK", L"warn"); 451 452 if(TRACE_ON(gecko)) 453 debug_level = 5; 454 else if(WARN_ON(gecko)) 455 debug_level = 2; 456 else if(ERR_ON(gecko)) 457 debug_level = 1; 458 459 swprintf(buf, ARRAY_SIZE(buf), L"all:%d", debug_level); 460 SetEnvironmentVariableW(L"NSPR_LOG_MODULES", buf); 461 462 len = GetEnvironmentVariableW(L"PATH", NULL, 0); 463 gre_path_len = lstrlenW(gre_path); 464 path = heap_alloc((len+gre_path_len+1)*sizeof(WCHAR)); 465 if(!path) 466 return; 467 GetEnvironmentVariableW(L"PATH", path, len); 468 469 /* We have to modify PATH as xul.dll loads other DLLs from this directory. */ 470 if(!(ptr = wcsstr(path, gre_path)) 471 || (ptr > path && *(ptr-1) != ';') 472 || (ptr[gre_path_len] && ptr[gre_path_len] != ';')) { 473 if(len) 474 path[len-1] = ';'; 475 lstrcpyW(path+len, gre_path); 476 SetEnvironmentVariableW(L"PATH", path); 477 } 478 heap_free(path); 479 } --- snip --- It worked until Wine-Gecko was built with msvcrt/UCRT runtime. Wine-Gecko 2.47.1, part of Wine 5.0-rc1 release. https://source.winehq.org/git/wine.git/commitdiff/5f0b5d350566a46f0f999e4cf… ("mshtml: Update to Wine Gecko 2.47.1.") --- snip --- $ WINEDEBUG=+relay,+gecko wine iexplore.exe ... 031c:Call KERNEL32.SetEnvironmentVariableW(01836084 L"NSPR_LOG_MODULES",0031cfb0 L"all:5") ret=017c2ea1 ... 031c:Ret KERNEL32.SetEnvironmentVariableW() retval=00000001 ret=017c2ea1 ... 031c:Call ucrtbase.getenv(0b5d46b0 "NSPR_LOG_MODULES") ret=0b58b3f9 031c:Ret ucrtbase.getenv() retval=00000000 ret=0b58b3f9 ... --- snip --- The workaround using external (shell) environment variable: --- snip --- $ NSPR_LOG_MODULES=all:5 WINEDEBUG=+relay wine iexplore.exe ... 01d0:Call KERNEL32.SetEnvironmentVariableW(01836084 L"NSPR_LOG_MODULES",0031cfb0 L"all:1") ret=017c2ea1 ... 01d0:Ret KERNEL32.SetEnvironmentVariableW() retval=00000001 ret=017c2ea1 ... 01d0:Call ucrtbase.getenv(093d46b0 "NSPR_LOG_MODULES") ret=0938b3f9 01d0:Ret ucrtbase.getenv() retval=001389d6 ret=0938b3f9 01d0:Call ucrtbase.strlen(001389d6 "all:5") ret=09385b52 01d0:Ret ucrtbase.strlen() retval=00000005 ret=09385b52 ... --- snip --- 'getenv' makes a copy of the environment variable block of the process on startup. Any later changes using win32 API 'SetEnvironmentVariable' are not reflected in the block of variables used by 'getenv' by design. https://source.winehq.org/git/wine.git/blob/2d4dd4252b0cf6526b3cc8194cce642… --- snip --- 28 /********************************************************************* 29 * getenv (MSVCRT.@) 30 */ 31 char * CDECL getenv(const char *name) 32 { 33 char **env; 34 unsigned int length=strlen(name); 35 36 for (env = MSVCRT__environ; *env; env++) 37 { 38 char *str = *env; 39 char *pos = strchr(str,'='); 40 if (pos && ((pos - str) == length) && !_strnicmp(str,name,length)) 41 { 42 TRACE("(%s): got %s\n", debugstr_a(name), debugstr_a(pos + 1)); 43 return pos + 1; 44 } 45 } 46 return NULL; 47 } --- snip --- https://source.winehq.org/git/wine.git/blob/2d4dd4252b0cf6526b3cc8194cce642… --- snip --- 402 /* INTERNAL: Since we can't rely on Winelib startup code calling w/getmainargs, 403 * we initialise data values during DLL loading. When called by a native 404 * program we simply return the data we've already initialised. This also means 405 * you can call multiple times without leaking 406 */ 407 void msvcrt_init_args(void) 408 { 409 OSVERSIONINFOW osvi; 410 411 MSVCRT__acmdln = _strdup( GetCommandLineA() ); 412 MSVCRT__wcmdln = _wcsdup( GetCommandLineW() ); 413 initial_wargv = cmdline_to_argv( GetCommandLineW(), &initial_argc ); 414 MSVCRT___argc = initial_argc; 415 MSVCRT___wargv = initial_wargv; 416 MSVCRT___argv = build_argv( initial_wargv ); 417 418 TRACE("got %s, wide = %s argc=%d\n", debugstr_a(MSVCRT__acmdln), 419 debugstr_w(MSVCRT__wcmdln),MSVCRT___argc); 420 421 osvi.dwOSVersionInfoSize = sizeof(OSVERSIONINFOW); 422 GetVersionExW( &osvi ); 423 MSVCRT__winver = (osvi.dwMajorVersion << 8) | osvi.dwMinorVersion; 424 MSVCRT__winmajor = osvi.dwMajorVersion; 425 MSVCRT__winminor = osvi.dwMinorVersion; 426 MSVCRT__osver = osvi.dwBuildNumber; 427 MSVCRT__osplatform = osvi.dwPlatformId; 428 TRACE( "winver %08x winmajor %08x winminor %08x osver %08x\n", 429 MSVCRT__winver, MSVCRT__winmajor, MSVCRT__winminor, MSVCRT__osver); 430 #ifdef _CRTDLL 431 CRTDLL__baseversion_dll = (GetVersion() >> 16); 432 CRTDLL__basemajor_dll = CRTDLL__baseversion_dll >> 8; 433 CRTDLL__baseminor_dll = CRTDLL__baseversion_dll & 0xff; 434 #endif 435 436 MSVCRT__HUGE = HUGE_VAL; 437 MSVCRT___setlc_active = 0; 438 MSVCRT___unguarded_readlc_active = 0; 439 MSVCRT__fmode = _O_TEXT; 440 441 MSVCRT__environ = msvcrt_SnapshotOfEnvironmentA(NULL); 442 MSVCRT___initenv = msvcrt_SnapshotOfEnvironmentA(NULL); 443 MSVCRT___winitenv = msvcrt_SnapshotOfEnvironmentW(NULL); ... --- snip --- $ wine --version wine-6.0-202-gc7de9f2daa8 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 3

[Bug 7106] Need for Speed 3 autorun crashes when starting setup
by WineHQ Bugzilla Jan. 24, 2021

Jan. 24, 2021

https://bugs.winehq.org/show_bug.cgi?id=7106 Anastasius Focht <focht(a)gmx.net> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |focht(a)gmx.net --- Comment #29 from Anastasius Focht <focht(a)gmx.net> --- Hello Igor, --- quote --- Wine 6.0 was released. Any updates? --- quote --- the bug is still marked as 'staged' which means the patch or another solution has not been merged into mainline yet. Use Wine-Staging if you want to play this game. 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 0