summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* gallium: clarify shift behavior with shift count >= 32Roland Scheidegger2013-08-021-12/+18
| | | | | | | | | | | | | | | | | | Previously, nothing was said what happens with shift counts exceeding bit width of the values to shift. In theory 3 behaviors are possible: 1) undefined (classic c definition) 2) just shift out all bits (so result is zero, or -1 potentially for ashr) 3) mask the shift count to bit width - 1 API's either require 3) or are ok with 1). In particular, GLSL (as well as a couple uninteresting legacy GL extensions) is happy with undefined, whereas both OpenCL and d3d10 require 3). Consequently, most hw also implements 3). So, for simplicity we just specify that 3) is required rather than saying undefined and then needing state trackers to work around it. Also while here specify shift count as a vector, not scalar. As far as I can tell this was a doc bug, neither state trackers nor drivers used scalar shift count. Reviewed-by: Jose Fonseca <[email protected]>
* nvc0: properly align NVE4_COMPUTE_MP_TEMP_SIZESamuel Pitoiset2013-07-312-2/+3
| | | | | | | | | MP_TEMP_SIZE must be aligned to 0x8000, while TEMP_SIZE on NVE4_3D must be aligned to 0x20000, so perform both alignments to be sure we allocate enough space (actually the bo will most likely use 128 KiB pages and not aligning to that would be a waste anyway). Cc: "9.2" [email protected]
* gallivm: use nearest rounding for float->unorm24 conversionRoland Scheidegger2013-07-311-4/+6
| | | | | | | | | | | | | Previously we were using truncation, which gives the correct result only for numbers in [0.5-1.0] range (because there's no mantissa bits to do any rounding there). This is frequently hit (and probably only used there) when converting fragment depth to depth format (d24s8 etc.) or otherwise dealing with depth format. v2: as spotted by Jose, get rid of extra type (src_type is already unsigned). Reviewed-by: Jose Fonseca <[email protected]>
* gallium/util: reformat, comment util_get_offset()Brian Paul2013-07-311-3/+7
| | | | Reviewed-by: Roland Scheidegger <[email protected]>
* gallium/util: comments, var renaming in u_inlines.hBrian Paul2013-07-311-13/+48
| | | | | | | | | | | | The variable 'usage' was being used for two different things. Sometimes for PIPE_USAGE_x and other times for PIPE_TRANSFER_x. This renames usage to access when we're talking about PIPE_TRANSFER_x flags. Plus, add a bunch of comments to remind us what's going on. Also, use unsigned for PIPE_TRANSFER_x bitmask to be consistent with other places. And add a missing const qualifier. Reviewed-by: Roland Scheidegger <[email protected]>
* softpipe: use new softpipe_resource_data() accessorBrian Paul2013-07-313-4/+20
| | | | | | | | | We should probably be using map()/unmap() when accessing resource data, but this is a little better. v2: assert that the resource is not a display target, per Jose. Reviewed-by: José Fonseca <[email protected]>
* softpipe: don't ignore pipe_constant_buffer::buffer_offsetBrian Paul2013-07-311-3/+5
| | | | | | | | | | | | | This was never a problem since the Mesa state tracker always gives us a user-space constant buffer with buffer_offset=0. But if another state tracker ever gave us a "HW" constant buffer with non-zero buffer_offset we'd mis-render. Also, use the correct buffer size. And move an assertion to the top of the function. Reviewed-by: Marek Olšák <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* gallium/docs: clarify definition of PIPE_CAP_USER_CONSTANT_BUFFERS, etcBrian Paul2013-07-311-5/+8
| | | | | | | | | | | The cap means _can_ accept user-space constant buffers; it doesn't mean _only_ accepts user-space constant buffers. v2: also update the PIPE_CAP_USER_VERTEX_BUFFERS and PIPE_CAP_USER_INDEX_BUFFERS descriptions as well. Per Jose. Reviewed-by: Marek Olšák <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* st/dri: add a new driconf option disable_shader_bit_encoding for UnigineMarek Olšák2013-07-304-1/+6
| | | | | | | Now Unigine Heaven 3.0 finally works with r600g. Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* mesa,glsl,st/dri: add a new driconf option force_glsl_version for UnigineMarek Olšák2013-07-304-7/+12
| | | | | | | | See documentation in mtypes.h. Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Brian Paul <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* st/dri: remove driOptionCache from dri_context in favor of dri_screenMarek Olšák2013-07-302-13/+1
| | | | | | | There is no reason to have this duplicated. Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* st/dri: move enabling postprocessing to dri_screenMarek Olšák2013-07-304-16/+18
| | | | | | | The driconf options are global. Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* st/dri: remove more unused driconf optionsMarek Olšák2013-07-301-6/+1
| | | | | | | vblank_mode is read by dri_util.c and falls under the "dri2" driver name, which is not connected to the actual Mesa/Gallium driver in any way. Reviewed-by: Brian Paul <[email protected]>
* st/dri: implement the driconf option force_s3tc_enable properlyMarek Olšák2013-07-305-12/+23
| | | | Reviewed-by: Brian Paul <[email protected]>
* driconf: remove the unused option allow_large_texturesMarek Olšák2013-07-301-2/+1
| | | | | | Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Brian Paul <[email protected]> Reviewed-by: Ian Romanick <[email protected]>
* st/dri: support the driconf option disable_blend_func_extendedMarek Olšák2013-07-304-3/+8
| | | | | | | This is needed for Unigine. Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* st/osmesa: initialize disable_glsl_line_continuationsMarek Olšák2013-07-301-0/+1
| | | | | Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* gallium/postprocessing: convert blits to pipe->blitMarek Olšák2013-07-305-25/+54
| | | | | | | | | PP saves current states to cso_context and then util_blit_pixels does the same. cso_context doesn't like that and the original state is not correctly restored. Cc: [email protected] Reviewed-by: Brian Paul <[email protected]>
* gallium/postprocessing: fix shader parsingMarek Olšák2013-07-301-2/+2
| | | | | | | | | | tokens was converted to a pointer, which made the Elements macro return 1. Broken by e87fc11cac696881469a57955af2ac7b4929a2c7. Cc: [email protected] Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* Revert "r300g: Give CLIP_DISABLE another try"Marek Olšák2013-07-302-3/+2
| | | | | | | | This reverts commit e866bd1adea2c3b4971ad68e69c644752f2ab7b6. https://bugs.freedesktop.org/show_bug.cgi?id=57875 Cc: [email protected]
* r600g/compute: Added missing address space checking of kernel parametersJonathan Charest2013-07-301-3/+2
| | | | | | | | | | | To have non-static buffers in local memory, it is necessary to pass them as arguments to the kernel. For r600, the correct lds size must be set to the SQ_LDS_ALLOC register. The correct size is the clover size plus the size reported by the compiler. Reviewed-by: Tom Stellard <[email protected]>
* clover: Added missing address space checking of kernel parameters v2Jonathan Charest2013-07-301-15/+31
| | | | | | | | | | Here is an updated patch with no line wrapping and respecting 80-column limit (for my changes). v2: Tom Stellard - Create global arguments for constant buffers so we don't break r600g. Reviewed-by: Tom Stellard <[email protected]>
* gallivm: (trivial) git rid of assertion in float->uint conversion codeRoland Scheidegger2013-07-291-2/+3
| | | | | | | Commit 8c3d3622d9ce2fd2a8f46084ab8153d708fa5b09 introduced a new assertion, but since it causes lp_test_conv failures remove it again and let's hope we don't really hit bugs caused by the potentially bogus code (it is possible the assert() caught some cases which work correctly too).
* nvc0: force use of correct firmware fileMaarten Lankhorst2013-07-281-1/+1
| | | | Signed-off-by: Maarten Lankhorst <[email protected]>
* gallivm: fix float->SNORM conversionRoland Scheidegger2013-07-272-11/+47
| | | | | | | | | | Just like the UNORM case we need to use round to nearest, not trunc. (There's also another problem, we're using the formula for SNORM->float which will produce a value below -1.0 for the most negative value which according to both OpenGL and d3d10 would need clamping. However, no actual failures have been observed due to that hence keep cheating on that.) Reviewed-by: Jose Fonseca <[email protected]>
* util: don't flush overflowing values to infinity in half-float conversionRoland Scheidegger2013-07-272-9/+17
| | | | | | | | | | | | | | | | | | I am not able to find _any_ rounding behavior specified for OpenGL for float to half-float conversions. However, it is specified for fp11/fp10 which suggests round to next finite value but round-to-zero would also be allowed, but finite values must not be flushed to infinity in either case. Hence I believe it makes sense to do the same for half-floats too. We could probably also use round-to-zero consistently, which is in fact required by d3d10 (but it doesn't seem to matter much). Does not match the mesa core function doing the same though (which is saying it was built to match intel gpus which I don't believe for a second as it would cause failures in d3d10, moreover the PRM (for ivy bridge, not listed in older manuals) while not specifying rounding behavior clearly states finite numbers are never flushed to infinity). Reviewed-by: Jose Fonseca <[email protected]>
* tgsi: handle texel swizzles correctly for d3d10-style sample opcodesRoland Scheidegger2013-07-271-5/+35
| | | | | | | Same as for gallivm (though these don't quite work correctly in softpipe, so untested). Reviewed-by: Jose Fonseca <[email protected]>
* gallivm: handle texel swizzles correctly for d3d10-style sample opcodesRoland Scheidegger2013-07-272-0/+29
| | | | | | | | | | | unlike OpenGL, the texel swizzle is embedded in the instruction, so honor that. (Technically we now execute both the sampler_view swizzle and the per-instruction swizzle but this should be quite ok.) v2: add documentation note as it's not obvious. Reviewed-by: Jose Fonseca <[email protected]>
* st/clover: Allow double precision operationsNiels Ole Salscheider2013-07-251-0/+1
| | | | | | | Pass "cl_khr_fp64" preprocessor definition to clang Signed-off-by: Niels Ole Salscheider <[email protected]> Reviewed-by: Tom Stellard <[email protected]>
* gallium/vl: add prime supportDave Airlie2013-07-261-1/+19
| | | | | | | | | | | | | | | | This fixes the dri2 opening to check if DRI_PRIME is set, and picks the correct drm device path to open, this along with a change to libvdpau allows vdpauinfo to work at least, Martin Peres tested with nouveau, and there seems to be a further issue with final displaying, it only works sometimes, but this patch is at least necessary to help debug further. Signed-off-by: Dave Airlie <[email protected]> Cc: [email protected] Reviewed-by: Christian König <[email protected]> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=67283 Tested-by: Armin K. <[email protected]>
* xa: move surface to ref/unref apiJerome Glisse2013-07-253-2/+18
| | | | | | | This make ddx life easier. Signed-off-by: Jerome Glisse <[email protected]> Signed-off-by: Rob Clark <[email protected]>
* xa: let ddx handle flushJerome Glisse2013-07-252-9/+2
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* xa: export a common context flush functionJerome Glisse2013-07-254-7/+17
| | | | | | First step before moving flushing inside the ddx. Signed-off-by: Jerome Glisse <[email protected]>
* xa: add handle type parameter to get handleJerome Glisse2013-07-252-2/+18
| | | | | | Allow to retrieve non shared handle. Signed-off-by: Jerome Glisse <[email protected]>
* xa: add xa_surface_from_handle()Rob Clark2013-07-252-4/+46
| | | | | | | | | | | For freedreno DDX, we have to create the scanout GEM bo in a special way (until we have our own KMS/DRM kernel driver.. and even then for phones/tablets you probably need to use the android drivers if you don't want to port the lcd panel driver support). The easiest way to handle this is let the DDX create the scanout bo, and then create the xa surface from that. Signed-off-by: Rob Clark <[email protected]>
* gallivm: Remove NoFramePointerElimNonLeaf for LLVM >= 3.4.Vinson Lee2013-07-251-0/+2
| | | | | | | | TargetOptions::NoFramePointerElimNonLeaf was removed in LLVM 3.4 r187093. Signed-off-by: Vinson Lee <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* draw: fix vertex id computationZack Rusin2013-07-255-13/+37
| | | | | | | | | | | | | vertex id has to be unaffected by the start index (i.e. when calling draw arrays with start_index = 5, the first vertex_id has to still be 0, not 5) and it has to be equal to the index when performing indexed rendering (in which case it has to be unaffected by the index bias). This fixes our behavior. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]> Reviewed-by: Brian Paul <[email protected]> Reviewed-by: Jose Fonseca <[email protected]>
* draw: cleanup and fix instance id computationZack Rusin2013-07-254-13/+8
| | | | | | | | | | | | | | The instance id system value always starts at 0, even if the specified start instance is larger than 0. Instead of implicitly setting instance id to instance id plus start instance and then having to subtract instance id when computing the buffer offsets lets just set instance id to the proper instance id. This fixes instance id computation and cleansup buffer offset computation. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]> Reviewed-by: Brian Paul <[email protected]> Reviewed-by: Jose Fonseca <[email protected]>
* gallivm: Remove dead code in lp_build_compare_ext.Vinson Lee2013-07-241-6/+0
| | | | | | | | | | There are earlier returns for PIPE_FUNC_NEVER and PIPE_FUNC_ALWAYS. The switch value of 'func' cannot be either of those values. Fixes "Logically dead code" defects reported by Coverity. Signed-off-by: Vinson Lee <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* nv50,nvc0: s/uint16/uint32 for constant buffer offsetChristoph Bumiller2013-07-242-2/+2
| | | | | | | | Looks like a thinko, "Hey, constant buffers can be at most 64 KiB in size, offset can't be larger." But it can, of course. I think piglit lacks a test for UBO and BindBufferRange that tests if it actually works.
* draw: always call util_cpu_detect() in draw context creation.Roland Scheidegger2013-07-241-1/+4
| | | | | | | | | | | | | | | | | | | | Since disabling denorms in draw_vbo() we require the util_cpu_caps to be initialized there. Hence add another util_cpu_detect() call in draw_create_context() which should ensure this. (There is another call in draw_get_option_use_llvm() which only gets called with x86 (not x86_64) but calling it always there wouldn't help since it most likely wouldn't get called when compiling without llvm, so leave it alone there.) This fixes https://bugs.freedesktop.org/show_bug.cgi?id=66806. (Because util_cpu_caps wasn't initialized when first calling util_fpstate_get() hence it returning zero, but it would later get initialized by rtasm translate code hence when draw call returned it unmasked all exceptions by calling util_fpstate_set(). This was happening only with DRAW_USE_LLVM=0 or not compiling with llvm, otherwise the llvm init code was calling it on time too.) Reviewed-by: Jose Fonseca <[email protected]> Reviewed-by: Zack Rusin <[email protected]> Tested-by: Vinson Lee <[email protected]>
* gallium/util: Fix detection of AVX cpu capsAndre Heider2013-07-231-2/+25
| | | | | | | | | | | | | | | | | For AVX it's not sufficient to only rely on the cpuid flags. If the CPU supports these extensions, but the OS doesn't, issuing these insns will trigger an undefined opcode exception. In addition to the AVX cpuid bit we also need to: * test cpuid for OSXSAVE support * XGETBV to check if the OS saves/restores AVX regs on context switches See "Detecting Availability and Support" at http://software.intel.com/en-us/articles/introduction-to-intel-advanced-vector-extensions Signed-off-by: Andre Heider <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]> Reviewed-by: José Fonseca <[email protected]>
* clover: Respect kernel argument alignment restrictions.Francisco Jerez2013-07-222-2/+19
| | | | | Cc: [email protected] Reviewed-by: Tom Stellard <[email protected]>
* clover: Extend kernel arguments for differing host and device data types.Francisco Jerez2013-07-222-4/+56
| | | | | | | Loosely based on a similar patch by Tom Stellard. Cc: [email protected] Reviewed-by: Tom Stellard <[email protected]>
* clover: Byte-swap kernel arguments when host and device endianness differ.Francisco Jerez2013-07-221-37/+65
| | | | | Cc: [email protected] Reviewed-by: Tom Stellard <[email protected]>
* clover: Add kernel argument fields to allow differing host/target data types.Francisco Jerez2013-07-221-2/+23
| | | | | | | Loosely based on a similar patch by Tom Stellard. Cc: [email protected] Reviewed-by: Tom Stellard <[email protected]>
* clover: Pass corresponding module::argument to kernel::argument::bind().Francisco Jerez2013-07-222-84/+61
| | | | | | | | And remove size information from most kernel::argument derived classes, it's no longer going to be necessary. Cc: [email protected] Reviewed-by: Tom Stellard <[email protected]>
* clover: Return correct value for CL_DEVICE_ENDIAN_LITTLETom Stellard2013-07-223-1/+8
| | | | | | | | Query the driver using PIPE_CAP_ENDIANNESS rather than always returning true. Cc: [email protected] Reviewed-by: Francisco Jerez <[email protected]>
* gallium: Add PIPE_CAP_ENDIANNESSTom Stellard2013-07-2214-1/+38
| | | | | | Cc: [email protected] [ Francisco Jerez: Fix "PIPE_ENDIAN_SMALL" in the documentation, define PIPE_ENDIAN_NATIVE. ]
* build: Remove unused EGL_PLATFORMS.Matt Turner2013-07-221-1/+0
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-27 22:53:44 +0000