wine-bugs February 2021

wine-bugs@winehq.org

1 participants
6956 discussions

[Bug 20056] New: The Westerner: crash follows bogus DSOUND:SetFormat call
by wine-bugs＠winehq.org 26 Feb '21

26 Feb '21

http://bugs.winehq.org/show_bug.cgi?id=20056 Summary: The Westerner: crash follows bogus DSOUND:SetFormat call Product: Wine Version: 1.1.29 Platform: PC OS/Version: Mac OS X 10.5 Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-dsound AssignedTo: wine-bugs(a)winehq.org ReportedBy: hoehle(a)users.sourceforge.net When DSOUND_PrimarySetFormat is given a broken format, it will nevertheless HeapReAlloc() and thus forget the old format. The Westerner happens to call DSOUND_PrimarySetFormat with a completely bogus format. Why? I don't know. DSOUND_PrimarySetFormat (formattag=0x0001,chans=0, samplerate=0,bytespersec=0, blockalign=0,bitspersamp=0,cbSize=0) >From there, two execution traces are possible, each leading to a crash. A) ReopenDevice fails as WAVE_Open refuses the bogus new format. The device is left in floating state, neither open nor closed, and with a bogus format description in device->pwfx. A subsequent DSOUND_PerformMix() will choke on the broken format. DSOUND_ReopenDevice waveOutOpen failed DSOUND_PrimarySetFormat DSOUND_ReopenDevice failed: 80070057 DSOUND_PerformMix (0x213620) DSOUND_PrimaryGetPosition pwplay=0, pwqueue=0 DSOUND_PrimaryGetPosition playpos = 0, writepos = 0 ( time=8688) DSOUND_PerformMix primary playpos=0, writepos=0, clrpos=0, mixpos=0, buflen=65536 wine: Unhandled division by zero -- likely BytesPerSec=0 B) One winmm:WAVE_Open initiated by DSOUND_PrimaryOpen succeeds. trace:winmm:WAVE_Open wFormatTag=1, nChannels=3, nSamplesPerSec=96000, nAvgBytesPerSec=576000, nBlockAlign=6, wBitsPerSample=16 The original format structure is still bogus (wBitsPerSample=0) and PrimaryOpen crashes in a division by zero in this case as well. This is the case I have to investigate further, as I thought my recent patches to wodOpen in wineoss.drv and winecoreaudio.drv for bug #12349 and bug #19901 would have taken care of it. Apparently there are several WAVE_FORMAT structures floating around that one does not know which one to correct. I also need to find out why the app keeps crashing now while I had it working in the past (IIRC I did not use the demo). -- 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 5

[Bug 19124] New: The Westerner: dsound heap overflow prevented by warn+heap
by wine-bugs＠winehq.org 26 Feb '21

26 Feb '21

http://bugs.winehq.org/show_bug.cgi?id=19124 Summary: The Westerner: dsound heap overflow prevented by warn+heap Product: Wine Version: 1.1.24 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-dsound AssignedTo: wine-bugs(a)winehq.org ReportedBy: hoehle(a)users.sourceforge.net Created an attachment (id=22087) --> (http://bugs.winehq.org/attachment.cgi?id=22087) short backtrace The Westerner crashes without WINEDEBUG=warn+heap in DSOUND_bufpos_to_mixpos() 100 DWORD ret = pos * 32 / device->pwfx->wBitsPerSample; So either the application or dsound is guilty of heap corruption / producing broken data. With the memory fence installed by using WINEDEBUG=warn+heap, the application seems to work "normally", except 2 lines are repeated very often and always in pair: err:dsound:DSOUND_MixInBuffer length not a multiple of block size, len = 5120, block size = 64626 warn:heap:allocate_large_block Could not allocate block for fc720000 bytes I tried out ALSA, emulation, 22050Hz 8bit, or default full HW (and also OSS IIRC). No difference. Note that in my test data from 2009-06-14, I required a native quartz.dll, but one also gets this crash with pure Wine components, before quartz gets used. Using Ubuntu 8.10 on Intel/SigmalTel/AC'97 audio HW. -- 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 16

[Bug 21785] New: The Westerner: no background in menu + errors when loading
by wine-bugs＠winehq.org 26 Feb '21

26 Feb '21

http://bugs.winehq.org/show_bug.cgi?id=21785 Summary: The Westerner: no background in menu + errors when loading Product: Wine Version: 1.1.38 Platform: x86 OS/Version: Mac OS X 10.5 Status: UNCONFIRMED Severity: minor Priority: P2 Component: gdiplus AssignedTo: wine-bugs(a)winehq.org ReportedBy: hoehle(a)users.sourceforge.net commit db3d756c1e4b03b0fb33f2863ab3d2275dcf00c2 Author: Vincent Povirk <vincent(a)codeweavers.com> Date: Sun Jan 24 15:18:46 2010 -0600 gdiplus: Prefer the builtin version. causes 2 bugs in The Westerner. 1. The graphics background and mouse are missing from the main menu, after displaying this log a hundred times: fixme:gdiplus:GdipImageRotateFlip (0x0 6) stub err:d3d_surface:surface_prepare_system_memory Surface without memory or pbo has SFLAG_INSYSMEM set! 2. Afterwards, starting or loading a game (via the cursor keys), the app displays many, many requesters, each time about a different file, like this: error: no es pontencia de 2 "RData\Textures\maderas\madera_fuelle.tif w:0 h:0 I didn't click through all of them to see what would be left to see. What's surprising is that there is a GdiPlus.dll in the app's directory. The app's installer must have added it as I found no note of adding one myself in my installation notes. This setup worked fine until Vincent's commit changed the priorities. I was not using any override except for quartz, as my AppDB entry explains. Unlike bug #21604 about gdiplus, the above fixme+err suggest that the missing functionality is clearly identified and could be added to the builtin gdiplus, so that it becomes useable with The Westerner. -- 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 23

[Bug 40763] New: League of legends crashes when changing screen resolution
by wine-bugs＠winehq.org 26 Feb '21

26 Feb '21

https://bugs.winehq.org/show_bug.cgi?id=40763 Bug ID: 40763 Summary: League of legends crashes when changing screen resolution Product: Wine Version: 1.9.11 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: scrimekiler(a)yahoo.fr Distribution: --- Created attachment 54659 --> https://bugs.winehq.org/attachment.cgi?id=54659 Standard debug output Since a few League of legends versions (or wine-staging versions), when I try to change the game resolution, the game crashes. The game client detects it and asks to reconnect, with success. (and asking if we want to send a bug report to them). I didn't see anything special in the default log output, but I added it as attachment anyway. Should I add more specific WINEDEBUG channels ? I will also add the game logs as specified in the screenshot. Using wine staging 1.9.11 -- 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 6

[Bug 50724] New: Tomb Raider (2013) enabling tessellation leads to crash with Vulkan renderer
by WineHQ Bugzilla 25 Feb '21

25 Feb '21

https://bugs.winehq.org/show_bug.cgi?id=50724 Bug ID: 50724 Summary: Tomb Raider (2013) enabling tessellation leads to crash with Vulkan renderer Product: Wine Version: 6.2 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: --- err:vkd3d_dxbc_compiler_put_symbol: Failed to insert symbol entry (register 0x6, 0). err:vkd3d_dxbc_compiler_put_symbol: Failed to insert symbol entry (register 0x6, 1). err:vkd3d_dxbc_compiler_put_symbol: Failed to insert symbol entry (register 0x6, 2). err:vkd3d_dxbc_compiler_put_symbol: Failed to insert symbol entry (register 0x6, 3). err:vkd3d_dxbc_compiler_put_symbol: Failed to insert symbol entry (register 0x6, 7). wine-6.2-320-gee4e33ed9ac vkd3d-1.2-180-g87a3712 -- 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 50676] New: Unicode characters do not display in Dialog boxes
by WineHQ Bugzilla 25 Feb '21

25 Feb '21

https://bugs.winehq.org/show_bug.cgi?id=50676 Bug ID: 50676 Summary: Unicode characters do not display in Dialog boxes Product: Wine Version: 6.0 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: lunarlambda(a)gmail.com Distribution: --- Created attachment 69392 --> https://bugs.winehq.org/attachment.cgi?id=69392 Screenshot of how the dialog looks in Wine I have a program which uses the CreateDialog macro to create a dialog from a template specified in a resource script. The resource file uses japanese text, and while both menus and window titles display correctly, dialog elements do not, simply replacing all japanese characters with '?'. A screenshot of how the dialog looks on Windows 10 with the system language set to English: https://cdn.discordapp.com/attachments/698182658890399794/81126958083171944… The program uses the -A APIs, and has a manifest setting the default code page to UTF-8, as described here: https://docs.microsoft.com/en-us/windows/uwp/design/globalizing/use-utf8-co… I also tried recompiling the program to use CreateDialogW instead, but it made no difference. -- 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 50648] New: No backtrace available for crashing apps
by WineHQ Bugzilla 24 Feb '21

24 Feb '21

https://bugs.winehq.org/show_bug.cgi?id=50648 Bug ID: 50648 Summary: No backtrace available for crashing apps Product: Wine-Testbot Version: unspecified Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: unknown Assignee: wine-bugs(a)winehq.org Reporter: ralf.habacker(a)freenet.de Distribution: --- At https://testbot.winehq.org/JobDetails.pl?Key=85326&f301=exe64.report#k301 it was reported that a Windows application crashed for an unknown reason. Since the referenced Windows variant is not available locally, further debugging is not possible or only possible with great difficulty. It would be helpful if the testbot would provide a backtrace in these cases. -- 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 24 Feb '21

24 Feb '21

https://bugs.winehq.org/show_bug.cgi?id=7106 Olivier F. R. Dierick <o.dierick(a)piezo-forte.be> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |o.dierick(a)piezo-forte.be --- Comment #36 from Olivier F. R. Dierick <o.dierick(a)piezo-forte.be> --- > Something about Staging patchset. Looks like that page is with patch that > redirects invalid console handles from NT to DOS,but in some "dirty" way. Is > that possible to reimplement it,to make it more NT-oriented? > > For example: reimplement it,using functions like > KTHREAD,CSR_THREAD,CSR_PROCESS? Hello, Linux/Unix doesn't provide those functions. They would have to be implemented in Wine. That patch uses its own implementation because there is no other. If Wine would implement functions with those names, it would end up with the same code anyway, so moving that code in functions with those names doesn't change anything and is not important. 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

[Bug 7106] Need for Speed 3 autorun crashes when starting setup
by WineHQ Bugzilla 24 Feb '21

24 Feb '21

https://bugs.winehq.org/show_bug.cgi?id=7106 --- Comment #35 from Chebanenko Igor <chebanenkoigor93(a)gmail.com> --- Something about Staging patchset. Looks like that page is with patch that redirects invalid console handles from NT to DOS,but in some "dirty" way. Is that possible to reimplement it,to make it more NT-oriented? For example: reimplement it,using functions like KTHREAD,CSR_THREAD,CSR_PROCESS? Is that possible? CSR_THREAD https://www.geoffchappell.com/studies/windows/win32/csrsrv/api/process/thre… CSR_PROCESS https://www.geoffchappell.com/studies/windows/win32/csrsrv/api/process/proc… KTHREAD https://www.geoffchappell.com/studies/windows/km/ntoskrnl/inc/ntos/ke/kthre… -- 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 50720] New: Cannot apply patch series that modifies and then renames a file
by WineHQ Bugzilla 24 Feb '21

24 Feb '21

https://bugs.winehq.org/show_bug.cgi?id=50720 Bug ID: 50720 Summary: Cannot apply patch series that modifies and then renames a file Product: Wine-Testbot Version: unspecified Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: unknown Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- If part 1 of a patch series modifies a file and then part 2 renames that file, git apply misapplies or fails to apply the combined (concatenated) patch used by the TestBot. See for instance: https://www.winehq.org/pipermail/wine-devel/2021-February/181490.html When a Git commit renames and modifies a file, the resulting Git diff does both in one operation: diff --git a/dlls/winegstreamer/gstdemux.c b/dlls/winegstreamer/quartz_parser.c similarity index 97% rename from dlls/winegstreamer/gstdemux.c rename to dlls/winegstreamer/quartz_parser.c index 84e6bcd61e4..e9e7c3a0f9f 100644 --- a/dlls/winegstreamer/gstdemux.c +++ b/dlls/winegstreamer/quartz_parser.c ... But the combined patches that the TestBot produces for patch series results in multiple "diff --git" entries, one to modify the old file, and then one to rename it. That results in "git apply" either mishandling the patch (apparently it performs a global analysis and renames the file before patching the old file, maybe to handle patches with A->B + B->A renames), or failing to apply it altogether. Potential fixes: * Use "patch -p1". It handles these combined patches as expected. It may however be somewhat more permissive than "git apply" in other scenarios which may not be desirable. * Split the combined patch into its individual components before "git applying" them one by one. * Store the patches in mbox format so that the combined patch is in mbox format too, and use "git am". This can be tested with the following wtbsuite tests: [2/2] WTBS S10.2 - ...before renaming it in the next part (amstream:amstream). [7/7] WTBS S5.7 - Check for rename+addition+removal conflicts (msi:suminfo). -- 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

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-bugs February 2021