| Summary: | crash in memcpy_texture because of NULL dstAddr (gl_texture_image.Data) | ||
|---|---|---|---|
| Product: | Mesa | Reporter: | Andriy Gapon <avg> |
| Component: | Drivers/DRI/R600 | Assignee: | Default DRI bug account <dri-devel> |
| Status: | RESOLVED WONTFIX | QA Contact: | |
| Severity: | major | ||
| Priority: | medium | ||
| Version: | 7.11 | ||
| Hardware: | x86-64 (AMD64) | ||
| OS: | FreeBSD | ||
| Whiteboard: | |||
| i915 platform: | i915 features: | ||
Just notice that the following message appeared in Xorg.log before the crash: __glXDRIbindTexImage: Failed to register texture offset override Not sure if this is related. Note: classic r600 driver has been abandoned. Please use r600g (gallium driver) instead. Is this still an issue with a newer driver/kernel? Is KMS needed for r600g? (In reply to comment #3) > Is KMS needed for r600g? Yes, KMS is required for r600g. Support for UMS in general is deprecated. That was my impression, thank you for the confirmation. (please see the Platform header of this report) Unfortunately, FreeBSD doesn't have KMS support for ati yet... The classic r600 driver has been abandoned long ago. It was replaced by the Gallium driver r600g. If you have issues with r600g please file a new bug report with component Drivers/Gallium/r600 Thanks. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.
Useful part of stack trace: (gdb) bt #0 0x0000000802e452f6 in memcpy () from /lib/libc.so.7 #1 0x0000000817fff011 in memcpy_texture (ctx=0x82c4a8000, dimensions=0, dstFormat=MESA_FORMAT_ARGB8888_REV, dstAddr=0x0, dstXoffset=0, dstYoffset=0, dstZoffset=0, dstRowStride=4352, dstImageOffsets=0x82b1d78c8, srcWidth=1056, srcHeight=33, srcDepth=1, srcFormat=32993, srcType=5121, srcAddr=0x83beff000, srcPacking=0x82c4b7690) at main/texstore.c:994 #2 0x00000008180082ca in _mesa_texstore_argb8888 (ctx=0x82c4a8000, dims=2, baseInternalFormat=6408, dstFormat=MESA_FORMAT_ARGB8888, dstAddr=0x0, dstXoffset=0, dstYoffset=0, dstZoffset=0, dstRowStride=4352, dstImageOffsets=0x82b1d78c8, srcWidth=1056, srcHeight=33, srcDepth=1, srcFormat=32993, srcType=5121, srcAddr=0x83beff000, srcPacking=0x82c4b7690) at main/texstore.c:1430 #3 0x00000008180091fb in _mesa_texstore (ctx=<optimized out>, dims=<optimized out>, baseInternalFormat=<optimized out>, dstFormat=<optimized out>, dstAddr=<optimized out>, dstXoffset=<optimized out>, dstYoffset=0, dstZoffset=0, dstRowStride=4352, dstImageOffsets=0x82b1d78c8, srcWidth=1056, srcHeight=33, srcDepth=1, srcFormat=32993, srcType=5121, srcAddr=0x83beff000, srcPacking=0x82c4b7690) at main/texstore.c:4478 #4 0x0000000817f8a4c7 in radeon_store_teximage (ctx=0x82c4a8000, dims=2, xoffset=0, yoffset=0, zoffset=0, width=1056, height=33, depth=1, imageSize=0, format=32993, type=5121, pixels=0x83beff000, packing=0x82c4b7690, texObj=0x844c11500, texImage=0x82ba98ec0, compressed=0) at radeon_texture.c:749 #5 0x0000000817f8ab1b in radeon_teximage (ctx=0x82c4a8000, dims=2, target=<optimized out>, level=0, internalFormat=<optimized out>, width=1056, height=33, depth=1, imageSize=0, format=32993, type=5121, pixels=<optimized out>, packing=0x82c4b7690, texObj=0x844c11500, texImage=0x82ba98ec0, compressed=0) at radeon_texture.c:834 #6 0x0000000817f8b6e7 in radeonTexImage2D (ctx=<optimized out>, target=<optimized out>, level=<optimized out>, internalFormat=<optimized out>, width=<optimized out>, height=<optimized out>, border=0, format=32993, type=5121, pixels=0x83beff000, packing=0x82c4b7690, texObj=0x844c11500, texImage=0x82ba98ec0) at radeon_texture.c:867 #7 0x0000000817ffad91 in teximage (ctx=0x82c4a8000, dims=2, target=3553, level=0, internalFormat=4, width=1056, height=33, depth=1, border=0, format=32993, type=5121, pixels=0x83beff000) at main/teximage.c:2505 #8 0x0000000817ffafc9 in _mesa_TexImage2D (target=3553, level=0, internalFormat=4, width=1056, height=33, border=0, format=32993, type=5121, pixels=0x83beff000) at main/teximage.c:2559 #9 0x0000000803e46622 in __glXDRIbindTexImage () from /usr/local/lib/xorg/modules/extensions/libglx.so #10 0x0000000803e3a7d8 in __glXDisp_BindTexImageEXT () from /usr/local/lib/xorg/modules/extensions/libglx.so #11 0x0000000803e3b020 in __glXDisp_VendorPrivate () from /usr/local/lib/xorg/modules/extensions/libglx.so #12 0x0000000803e3d22d in __glXDispatch () from /usr/local/lib/xorg/modules/extensions/libglx.so In frame 5 the following inconsistency can be observed: (gdb) p *(radeonTexObj*)texObj $12 = {base = {Mutex = 0x8363736a0, RefCount = 3, Name = 10721, Target = 3553, Sampler = {Name = 0, RefCount = 0, WrapS = 33071, WrapT = 33071, WrapR = 10497, MinFilter = 9728, MagFilter = 9728, BorderColor = {f = {0, 0, 0, 0}, ui = {0, 0, 0, 0}, i = {0, 0, 0, 0}}, MinLod = -1000, MaxLod = 1000, LodBias = 0, MaxAnisotropy = 1, CompareMode = 0, CompareFunc = 515, CompareFailValue = 0, sRGBDecode = 35401, CubeMapSeamless = 0 '\000', DepthMode = 6409, _CompleteTexture = 0 '\000'}, Priority = 1, BaseLevel = 0, MaxLevel = 1000, _MaxLevel = 10, _MaxLambda = 10, CropRect = {0, 0, 0, 0}, Swizzle = {6403, 6404, 6405, 6406}, _Swizzle = 1672, GenerateMipmap = 0 '\000', _Complete = 1 '\001', _RenderToTexture = 0 '\000', Purgeable = 0 '\000', Image = {{0x82ba98ec0, 0x0 <repeats 14 times>}, {0x0 <repeats 15 times>}, { 0x0 <repeats 15 times>}, {0x0 <repeats 15 times>}, {0x0 <repeats 15 times>}, {0x0 <repeats 15 times>}}, BufferObject = 0x0, BufferObjectFormat = 0, Palette = {InternalFormat = 0, _BaseFormat = 0, Size = 0, TableF = 0x0, TableUB = 0x0, RedSize = 0 '\000', GreenSize = 0 '\000', BlueSize = 0 '\000', AlphaSize = 0 '\000', LuminanceSize = 0 '\000', IntensitySize = 0 '\000'}, DriverData = 0x0}, mt = 0x846d87b00, validated = 0 '\000', minLod = 0, maxLod = 0, override_offset = 3398523648, image_override = 1 '\001', tile_bits = 0, bo = 0x839e88480, pp_txfilter = 0, pp_txformat = 0, pp_txformat_x = 0, pp_txsize = 0, pp_txpitch = 0, pp_border_color = 0, pp_cubic_faces = 0, pp_txfilter_1 = 0, SQ_TEX_RESOURCE0 = 553158401, SQ_TEX_RESOURCE1 = 1744830496, SQ_TEX_RESOURCE2 = 0, SQ_TEX_RESOURCE3 = 1088, SQ_TEX_RESOURCE4 = 101335040, SQ_TEX_RESOURCE5 = 0, SQ_TEX_RESOURCE6 = 2147483648, SQ_TEX_RESOURCE7 = 0, SQ_TEX_SAMPLER0 = 12582930, SQ_TEX_SAMPLER1 = 4228841472, SQ_TEX_SAMPLER2 = 2147483648, TD_PS_SAMPLER0_BORDER_RED = 0, TD_PS_SAMPLER0_BORDER_GREEN = 0, TD_PS_SAMPLER0_BORDER_BLUE = 0, TD_PS_SAMPLER0_BORDER_ALPHA = 0, border_fallback = 0 '\000'} (gdb) p *(radeon_texture_image*)texImage $14 = {base = {InternalFormat = 4, _BaseFormat = 6408, TexFormat = MESA_FORMAT_ARGB8888, Border = 0, Width = 1056, Height = 33, Depth = 1, Width2 = 1056, Height2 = 33, Depth2 = 1, WidthLog2 = 10, HeightLog2 = 5, DepthLog2 = 0, MaxLog2 = 10, WidthScale = 1056, HeightScale = 33, DepthScale = 1, IsClientData = 0 '\000', _IsPowerOfTwo = 0 '\000', TexObject = 0x844c11500, FetchTexelc = 0x81810268f <fetch_texel_float_to_chan>, FetchTexelf = 0x8180fe1d7 <fetch_texel_2d_f_argb8888>, RowStride = 1056, ImageOffsets = 0x82b1d78c8, Data = 0x0, DriverData = 0x0}, mt = 0x0, bo = 0x0, mtlevel = 0, mtface = 0} As you can see (a) texObj has bo set, but (b) texImage has neither bo nor mt set. Because of (a) the Data field is _not_ set in radeon_teximage(), because of (b) the field is not set in radeon_teximage_map() call made from radeon_store_teximage. Thus Data remains NULL. I am not sure if there is a simple logic error somewhere in radeon_teximage or radeon_store_teximage, or if the discrepancy between texImage and texObj means that there is a bug some place else. I see that the code in git master is quite different, but unfortunately I do not have an opportunity to test it.