[Bug 53235] New: The 32-bit mfplat:mfplat crashes on the TestBot debian11 VM
https://bugs.winehq.org/show_bug.cgi?id=53235 Bug ID: 53235 Summary: The 32-bit mfplat:mfplat crashes on the TestBot debian11 VM Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: mfplat Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- The 32-bit mfplat:mfplat crashes on the TestBot debian11 VM. The crash is either a EXCEPTION_WINE_ASSERTION: mfplat.c:922: Test marked todo: Unexpected sub type {30323449-0000-0010-8000-00aa00389b71}. Unhandled exception: assertion failed in 32-bit code (0xf7f45559). Backtrace: =>0 0xf7f45559 __kernel_vsyscall+0x9() in [vdso].so (0x0065e42c) 1 0xf7d15e02 in libc.so.6 (+0x34e02) (0x0065e42c) 2 0xf7cfe306 in libc.so.6 (+0x1d306) (0xf7ec6000) 3 0x5d3f43d8 in libllvm-11.so.1 (+0x75f3d8) (0x0065e728) 4 0x5d3f41b0 in libllvm-11.so.1 (+0x75f1b0) (0x5fe78762) 5 0x7420656c in libxvidcore.so.4 (+0xa656c) (0x62616e55) 0xf7f45559 __kernel_vsyscall+0x9 in [vdso].so: popl %ebp Or a plain NULL pointer dereference: Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x70baff74).Backtrace: =>0 0x70baff74 memset+0x24(dst=0x000000000, c=0, n=0x4000) [Z:\home\winetest\tools\testbot\var\wine\dlls\msvcrt\string.c:3150] in ucrtbase (0x0357f908) 1 0x715283e3 wined3d_texture_load_location+0x423(texture=001AC9E0, sub_resource_idx=0, context=001A9A30, location=0x8) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\texture.c:853] in wined3d (0x0357f9a8) 2 0x715173ae surface_cpu_blt+0x9ee(dst_texture=001AC9E0, dst_sub_resource_idx=0, dst_box=0357FB38, src_texture=001AC588, src_sub_resource_idx=0, src_box=0357FB50, flags=0x20000000, fx=0357FB68, filter=WINED3D_TEXF_POINT) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\surface.c:752] in wined3d (0x0357faf8) 3 0x71518ea2 cpu_blitter_blit+0xc2(blitter=001AC3C0, op=WINED3D_BLIT_OP_RAW_BLIT, context=001A9A30, src_texture=001AC588, src_sub_resource_idx=0, src_location=0x10, src_rect=0357FE10, dst_texture=001AC9E0, dst_sub_resource_idx=<is not available>, dst_location=0x8, dst_rect=0357FE20, color_key=00000000, filter=WINED3D_TEXF_POINT, resolve_format=00000000) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\surface.c:1328] in wined3d (0x0357fb98) 4 0x715368a2 ffp_blitter_blit+0xa2(blitter=001AC3D8, op=WINED3D_BLIT_OP_RAW_BLIT, context=001A9A30, src_texture=001AC588, src_sub_resource_idx=0, src_location=0x10, src_rect=0357FE10, dst_texture=001AC9E0, dst_sub_resource_idx=0, dst_location=0x8, dst_rect=0357FE20, colour_key=00000000, filter=WINED3D_TEXF_POINT, resolve_format=00000000) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\texture.c:6213] in wined3d (0x0357fc58) 5 0x714dc91d glsl_blitter_blit+0x1ad(blitter=001AC3F0, op=WINED3D_BLIT_OP_RAW_BLIT, context=001A9A30, src_texture=001AC588, src_sub_resource_idx=0, src_location=0x10, src_rect=0357FE10, dst_texture=001AC9E0, dst_sub_resource_idx=0, dst_location=0x8, dst_rect=0357FE20, colour_key=00000000, filter=WINED3D_TEXF_POINT, resolve_format=00000000) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\glsl_shader.c:13270] in wined3d (0x0357fd78) 6 0x7151a181 texture2d_blt+0x3c1(dst_texture=001AC9E0, dst_sub_resource_idx=0, dst_box=0479C3C8, src_texture=001AC588, src_sub_resource_idx=0, src_box=0479C3E8, flags=0x20000000, fx=0479C404, filter=WINED3D_TEXF_POINT) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\surface.c:1639] in wined3d (0x0357fe48) 7 0x7149b07d wined3d_cs_exec_blt_sub_resource+0xcd(cs=04790020, data=0479C3BC) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\cs.c:2579] in wined3d (0x0357fec8) 8 0x7149cb2b wined3d_cs_command_unlock(queue=<internal error>, cs=<internal error>) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\cs.c:3310] in wined3d (0x0357ff28) 9 0x7149cb2b wined3d_cs_execute_next+0x53(queue=<internal error>, cs=<internal error>) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\cs.c:3309] in wined3d (0x0357ff28) 10 0x7149cb2b wined3d_cs_run+0x11b(ctx=<internal error>) [Z:\home\winetest\tools\testbot\var\wine\dlls\wined3d\cs.c:3375] in wined3d (0x0357ff28) 11 0x7b62a170 in kernel32 (+0x2a170) (0x0357ff48) 12 0x7bc5a937 in ntdll (+0x5a937) (0x0357ff5c) 13 0x7bc5aff0 RtlCreateUserThread(entry=7149CA10, arg=04790020) [Z:\home\winetest\tools\testbot\var\wine\dlls\ntdll\thread.c:261] in ntdll (0x0357ffec) 0x70baff74 memset+0x24 [Z:\home\winetest\tools\testbot\var\wine\dlls\msvcrt\string.c:3150] in ucrtbase: movl %eax,0x0(%ebx) 3150 *(unaligned_ui64 *)(d + 0) = v; https://test.winehq.org/data/patterns.html#mfplat:mfplat This is most likely the same issue as the d2d1:d2d1 (bug 53231) and evr:evr (bug 53201) crashes. Still, this time the crash: * Does not happen on my Debian 11 + Intel & KDE box * Does not happen on my Debian 11 + QXL + single monitor VM * Happens on the TestBot's Debian 11 + QXL + dual monitor VM (probably slightly different library versions) * Does not happen on the TestBot's Debian Testing + QXL + dual monitor VM Also the crashes started when the debian11 VM was updated and moved to the gpu1 VM host. -- 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=53235 François Gouget <fgouget(a)codeweavers.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Keywords| |source, testcase -- 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=53235 Zhiyi Zhang <zzhang(a)codeweavers.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |zzhang(a)codeweavers.com Assignee|wine-bugs(a)winehq.org |zzhang(a)codeweavers.com -- 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=53235 Giovanni Mascellani <gmascellani(a)codeweavers.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |gmascellani(a)codeweavers.com --- Comment #1 from Giovanni Mascellani <gmascellani(a)codeweavers.com> --- This seems to be an unhandled memory allocation failure: 01f8:err:d3d:wined3d_debug_callback 001AFFA8: "GL_OUT_OF_MEMORY in glBufferStorage". 01f8:err:d3d:wined3d_debug_callback 001AFFA8: "GL_INVALID_VALUE in glMapBufferRange(offset 0 + length 67108864 > buffer_size 0)". 01f8:err:d3d:wined3d_allocator_chunk_gl_map Failed to map chunk memory. 01f8:err:d3d:wined3d_bo_gl_map Failed to map chunk. 01f8:err:d3d:wined3d_context_gl_map_bo_address Failed to map bo. wine: Unhandled page fault on write access to 00000000 at address 70BAFD94 (thread 01f8), starting debugger... BTW, on the testbot running with +d3d gives: 00f8:trace:d3d:wined3d_adapter_init_gl_caps GL_RENDERER: "llvmpipe (LLVM 11.0.1, 256 bits)". 00f8:trace:d3d:wined3d_adapter_init_gl_caps GL_VENDOR: "Mesa/X.org". 00f8:trace:d3d:wined3d_adapter_init_gl_caps GL_VERSION: "4.5 (Core Profile) Mesa 20.3.5". While on my computer (using "LIBGL_ALWAYS_SOFTWARE=1 GALLIUM_DRIVER=llvmpipe" to force llvmpipe) I get: 0024:trace:d3d:4P:714CB0E3:ine/dlls/wined3d/adapter_gl.c:3413:wined3d_adapter_init_gl_caps GL_RENDERER: "llvmpipe (LLVM 14.0.4, 256 bits)". 0024:trace:d3d:4P:714CB0E3:ine/dlls/wined3d/adapter_gl.c:3421:wined3d_adapter_init_gl_caps GL_VENDOR: "Mesa/X.org". 0024:trace:d3d:4P:714CB0E3:ine/dlls/wined3d/adapter_gl.c:3430:wined3d_adapter_init_gl_caps GL_VERSION: "4.5 (Core Profile) Mesa 22.0.5". So it might be a bug that was fixed between Mesa 20.3.5 and Mesa 22.0.5. In theory Wine should gracefully handle the failed allocation anyway. -- 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=53235 --- Comment #2 from Zhiyi Zhang <zzhang(a)codeweavers.com> --- (In reply to Giovanni Mascellani from comment #1)
This seems to be an unhandled memory allocation failure:
It's a regression from D3DKMTOpenAdapterFromLuid() using too much memory and resource leaks in the tests. LLVMPIPE may get cleverer in the newer versions so they don't 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.
https://bugs.winehq.org/show_bug.cgi?id=53235 Zhiyi Zhang <zzhang(a)codeweavers.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |FIXED Status|NEW |RESOLVED Assignee|zzhang(a)codeweavers.com |wine-bugs(a)winehq.org Fixed by SHA1| |2e0a56257a15af8dee6ebad55f1 | |80591b1c1e18c --- Comment #3 from Zhiyi Zhang <zzhang(a)codeweavers.com> --- Fixed by 2e0a56257a15af8dee6ebad55f180591b1c1e18c -- 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=53235 Alexandre Julliard <julliard(a)winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED --- Comment #4 from Alexandre Julliard <julliard(a)winehq.org> --- Closing bugs fixed in 7.12. -- 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