[Bug 59561] New: VICE: GTK version of x86sc.exe crashes at startup.
http://bugs.winehq.org/show_bug.cgi?id=59561 Bug ID: 59561 Summary: VICE: GTK version of x86sc.exe crashes at startup. Product: Wine Version: 11.4 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@list.winehq.org Reporter: bernhardu@mailbox.org Distribution: --- Created attachment 80616 --> http://bugs.winehq.org/attachment.cgi?id=80616 vice_gtk_3.10_winehq-11.4_backtrace.txt https://sourceforge.net/projects/vice-emu/files/releases/binaries/windows/GT... Maybe related logging immediately before the crash: fixme:dxgi:DXGID3D10CreateDevice Ignoring flags 0x20. fixme:d3d:wined3d_swapchain_init The application requested more than one back buffer, this is not properly supported. Please configure the application to use double buffering (1 back buffer) if possible. ... fixme:d3d11:d3d11_device_context_SwapDeviceContextState D3D10 interface emulation not fully implemented yet! It crashes with following Backtrace: =>0 ID2D1Image_AddRef() [include/d2d1.h:3254] in d2d1 1 d2d_effect_SetInput+0x21() [dlls/d2d1/effect.c:2805] in d2d1 2 0x0000014022e58a in x64sc Unfortunately upstream builds are stripped so lacking debug information. But I guess following location is calling this SetInput with the context->render_bitmap being a NULL pointer: https://github.com/VICE-Team/svn-mirror/blob/main/vice/src/arch/gtk3/directx... (The SDL version from SDL2VICE-3.10-win64.zip seems to work find.) -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 Bernhard Übelacker <bernhardu@mailbox.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download, source Distribution|--- |Debian URL| |https://sourceforge.net/pro | |jects/vice-emu/files/releas | |es/binaries/windows/GTK3VIC | |E-3.10-win64.zip/download -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #1 from Bernhard Übelacker <bernhardu@mailbox.org> --- If building x64sc.exe with below modification makes the user interface show up without it crashing. The actual emulation is still not shown. And unfortunately the menus and user interface shows some graphical glitches like truncated texts, so it looks like GTK3 seems to have another issue with wine. (A version provided by msys2 is used for building x64sc.exe.) - context->d2d_effect_strip_alpha->SetInput(0, context->render_bitmap); + if (context->render_bitmap) context->d2d_effect_strip_alpha->SetInput(0, context->render_bitmap); -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #2 from Nikolay Sivov <bunglehead@gmail.com> --- For some reason it fails much earlier for me. Could you attach +d3d,+d3d11,+d2d log, please? -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #3 from Bernhard Übelacker <bernhardu@mailbox.org> --- Created attachment 80623 --> http://bugs.winehq.org/attachment.cgi?id=80623 vice_gtk_3.10_winehq-11.4_+d3d,+d3d11,+d2d_backtrace.txt (In reply to Nikolay Sivov from comment #2)
For some reason it fails much earlier for me. Could you attach +d3d,+d3d11,+d2d log, please?
Thanks for taking a look, attached file is with +d3d,+d3d11,+d2d. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #4 from Nikolay Sivov <bunglehead@gmail.com> --- I don't see an attempt to do CreateBitmap(), so it probably fails earlier, and d2d crash is a consequence. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #5 from Bernhard Übelacker <bernhardu@mailbox.org> --- (In reply to Nikolay Sivov from comment #4)
I don't see an attempt to do CreateBitmap(), so it probably fails earlier, and d2d crash is a consequence.
In the source there is one CreateBitmap in `build_render_bitmap`. But that get just called when context->d2d_device_context is set. And context->d2d_device_context gets set in build_directx_resources. But build_directx_resources gets called after `build_render_bitmap`. Therefore I don't see how this would behave different in Windows. Could one explanation be SetInput is early returning when given a NULL bitmap? Doing so makes the crash go away, but still no emulation shows up. I did some experiments with github actions and received this vice package with debug information :-) https://github.com/bernhardu/vice-svn-mirror/releases/tag/test3 -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #6 from Nikolay Sivov <bunglehead@gmail.com> --- Yes, quick test shows that SetInput(NULL) does not 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #7 from Bernhard Übelacker <bernhardu@mailbox.org> --- Created attachment 80627 --> http://bugs.winehq.org/attachment.cgi?id=80627 59561_d2d1_Allow_value_NULL_in_d2d_effect_SetInput.patch (In reply to Nikolay Sivov from comment #6)
Yes, quick test shows that SetInput(NULL) does not crash.
So something like this patch may be enough for submission? -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #8 from Nikolay Sivov <bunglehead@gmail.com> --- (In reply to Bernhard Übelacker from comment #7)
Created attachment 80627 [details] 59561_d2d1_Allow_value_NULL_in_d2d_effect_SetInput.patch
(In reply to Nikolay Sivov from comment #6)
Yes, quick test shows that SetInput(NULL) does not crash.
So something like this patch may be enough for submission?
Sure, this looks fine. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #9 from Bernhard Übelacker <bernhardu@mailbox.org> --- Submitted the attached patch as merge request: https://gitlab.winehq.org/wine/wine/-/merge_requests/10475 -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 Nikolay Sivov <bunglehead@gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |d2d -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 --- Comment #10 from Nikolay Sivov <bunglehead@gmail.com> --- Should we close this? -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 Bernhard Übelacker <bernhardu@mailbox.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |c2f0d81c7b91c8327fb8bf7ad53 | |ed637abedd3e5 Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #11 from Bernhard Übelacker <bernhardu@mailbox.org> --- (In reply to Nikolay Sivov from comment #10)
Should we close this?
Yes, the crash in x64sc.exe no longer shows up, resolving fixed. I am going to put the next issue of the distorted user interface into a new report. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 Bernhard Übelacker <bernhardu@mailbox.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|VICE: GTK version of |VICE: GTK version of |x86sc.exe crashes at |x64sc.exe crashes at |startup. |startup. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 Bernhard Übelacker <bernhardu@mailbox.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Blocks| |59579 -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 Alexandre Julliard <julliard@winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED --- Comment #12 from Alexandre Julliard <julliard@winehq.org> --- Closing bugs fixed in 11.6. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=59561 Bernhard Übelacker <bernhardu@mailbox.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Blocks| |59629 -- 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.
participants (1)
-
WineHQ Bugzilla