[Bug 52884] New: Adobe Lightroom Classic 10.4 + 11.1 crash in ??0CTJPEGWriter@@QEAA@XZ+0x2459d()
https://bugs.winehq.org/show_bug.cgi?id=52884 Bug ID: 52884 Summary: Adobe Lightroom Classic 10.4 + 11.1 crash in ??0CTJPEGWriter@@QEAA(a)XZ+0x2459d() Product: Wine Version: 7.6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: bugs(a)bzatek.net Distribution: --- Created attachment 72268 --> https://bugs.winehq.org/attachment.cgi?id=72268 backtrace
Unhandled exception: page fault on read access to 0x0000000000000018 in 64-bit code (0x0000000020301a52). Backtrace: =>0 0x0000000020301a52 EntryPoint+0x5fd5fe() in cameraraw (0x0000000000117bd0) 1 0x00000000203069c4 EntryPoint+0x602570() in cameraraw (0x0000000000117bd0) 2 0x000000001fb1aa7d ??0CTJPEGWriter@@QEAA(a)XZ+0x2459d() in cameraraw (0x0000000000117bd0) 0x0000000020301a52 cameraraw+0xaf1a52: cmpb $0,0x0000000000000018(%rsi)
good: wine-7.3 (vkd3d not compiled in), wine-6.22 bad: wine-7.4 (vkd3d bundled and enabled, vulkan enabled), wine-7.6 Clean prefix, only vcrun2015 installed. I've seen this crash in Lightroom Classic 10.4 and 11.1 under different prefixes with more libraries in. Those prefixes running wine-6.22 were stable for months with daily usage. GPU usage is disabled in Lightroom Preferences. Crashes often right after startup or on a first image manipulation. Notable messages from the console:
0124:fixme:dxgi:dxgi_adapter_QueryVideoMemoryInfo Returning fake video memory info. 0124:fixme:dxgi:dxgi_adapter_SetVideoMemoryReservation iface 000000005CA3F5F0, node_index 0, segment_group 0, reservation 0x40000000 stub! 0124:fixme:dxgi:dxgi_adapter_RegisterVideoMemoryBudgetChangeNotificationEvent iface 000000005CA7A050, event 0000000000000728, cookie 000000005CA79F18 stub! wine: Unhandled page fault on read access to 0000000000000018 at address 0000000020301A52 (thread 0124), starting debugger... 0060:err:virtual:virtual_setup_exception stack overflow 1232 bytes in thread 0060 addr 0x7f8e8c1eef87 stack 0x940b30 (0x940000-0x941000-0xa40000)
Hardware used: Skylake Xeon, AMD Radeon Pro WX 2100 (polaris12, LLVM 13.0.1, DRM 3.44, 5.16.8-zen-ge527425d9e93) (0x6995), mesa-22.0.0-rc2, Vulkan enabled, etc. -- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 Zeb Figura <z.figura12(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |z.figura12(a)gmail.com --- Comment #1 from Zeb Figura <z.figura12(a)gmail.com> --- Yeah, the bundled vkd3d is a good guess, especially given that it crashes after RegisterVideoMemoryBudgetChangeNotificationEvent(), which looks to be an API introduced with d3d12. Trying to return S_OK from that function would probably help confirm it, if you have the time and ability to build from source. Disabling d3d12 in winecfg may also help confirm it, or serve as a workaround. -- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 --- Comment #2 from Tomáš Bžatek <bugs(a)bzatek.net> --- Thanks, disabling d3d12.dll made it seemingly work (incredible to see apps working just fine with missing libraries). Also changed the return code to S_OK as suggested and managed to get further! Seems to be reasonably stable now, GPU acceleration is now indicated as enabled though the performance is roughly on the same level as before (GDI rendering only), i.e. rather slow (still around ~3 fps even for basic image zooming and panning).
0124:fixme:dxgi:dxgi_adapter_QueryVideoMemoryInfo Returning fake video memory info. 0124:fixme:dxgi:dxgi_adapter_SetVideoMemoryReservation iface 00000000546510E0, node_index 0, segment_group 0, reservation 0x40000000 stub! 0124:fixme:dxgi:dxgi_adapter_RegisterVideoMemoryBudgetChangeNotificationEvent iface 000000005333E3E0, event 0000000000000620, cookie 000000005333CE18 stub! [tbzatek was here]
This pattern repeats quite often but looks non-fatal, rendering looks okay:
0124:fixme:dxgi:d3d12_swapchain_init Ignoring swap effect 0x4. 0124:fixme:dxgi:wined3d_swapchain_desc_from_dxgi Ignoring scaling 0x1. 0124:fixme:dxgi:wined3d_swapchain_desc_from_dxgi Ignoring alpha mode 0. 0124:fixme:vulkan:X11DRV_vkCreateWin32SurfaceKHR Application requires child window rendering, which is not implemented yet! 0124:fixme:dxgi:hresult_from_vk_result Unhandled VkResult -9.
This is a good first step. I'm open for further debugging, testing patches, etc. Lightroom would greatly benefit from GPU acceleration. -- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 --- Comment #3 from Tomáš Bžatek <bugs(a)bzatek.net> --- Created attachment 72280 --> https://bugs.winehq.org/attachment.cgi?id=72280 Lightroom Preferences dialog -- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 --- Comment #4 from Tomáš Bžatek <bugs(a)bzatek.net> --- Btw. that was with wine-7.7 Installing gdiplus32.dll through winetricks seems to alter the rendering path a little, no functional differences observed, messages from my previous post are still valid.
0154:fixme:dxgi:dxgi_surface_GetDC iface 00007F64781A8490, discard 0, hdc 000000000011E378 semi-stub! 0154:fixme:d2d:d2d_device_context_DrawGeometry Ignoring stroke style 00007F6478329860. 0154:fixme:d2d:d2d_path_geometry_Stream iface 00007F647834ED50, sink 00007F647837F7C8 stub! 0154:fixme:d2d:d2d_device_context_PushLayer iface 000000003DF0CB80, layer_parameters 000000000011D898, layer 00007F647837F210 stub! 0154:fixme:d2d:d2d_device_context_PopLayer iface 000000003DF0CB80 stub! 0154:fixme:d2d:d2d_device_context_DrawGeometry Ignoring stroke style 00007F6478329860. 0154:fixme:d2d:d2d_device_context_DrawGeometry Ignoring stroke style 00007F6478329860.
-- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 --- Comment #5 from Zeb Figura <z.figura12(a)gmail.com> --- (In reply to Tomáš Bžatek from comment #2)
Thanks, disabling d3d12.dll made it seemingly work (incredible to see apps working just fine with missing libraries).
Often (as here) the libraries just aren't available on older versions of Windows, so it broadly makes sense.
Also changed the return code to S_OK as suggested and managed to get further! Seems to be reasonably stable now, GPU acceleration is now indicated as enabled though the performance is roughly on the same level as before (GDI rendering only), i.e. rather slow (still around ~3 fps even for basic image zooming and panning).
...
0124:fixme:vulkan:X11DRV_vkCreateWin32SurfaceKHR Application requires child window rendering, which is not implemented yet! 0124:fixme:dxgi:hresult_from_vk_result Unhandled VkResult -9.
Right, so it needs child window rendering, aka bug 45277. I guess we end up going through GDI if that's not available, but that would be rather slow. There are some patches linked to that bug that might help, although that area is being thrashed a lot lately so I'm not sure if there's anything that currently applies... -- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 Ken Sharp <imwellcushtymelike(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Severity|major |minor -- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 Zeb Figura <z.figura12(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Severity|minor |normal -- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 Neko-san <nekoNexus(a)protonmail.ch> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |nekoNexus(a)protonmail.ch -- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 Tomáš Bžatek <bugs(a)bzatek.net> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution|--- |WORKSFORME --- Comment #6 from Tomáš Bžatek <bugs(a)bzatek.net> --- Testing wine-8.8, appears to be stable, no patches needed anymore. -- 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.
https://bugs.winehq.org/show_bug.cgi?id=52884 Gijs Vermeulen <gijsvrm(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED --- Comment #7 from Gijs Vermeulen <gijsvrm(a)gmail.com> --- Closing. -- 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