summaryrefslogtreecommitdiff
path: root/src/mesa/pipe
AgeCommit message (Collapse)Author
2008-02-13handle temporaries in llvm code generated pathsZack Rusin
2008-02-13implement addZack Rusin
2008-02-13implement mulZack Rusin
2008-02-13get mov workingZack Rusin
2008-02-13fix fetching input elements and generate a ret on endZack Rusin
2008-02-13add functiontype for the llvm native vs entry pointZack Rusin
2008-02-13disable llvm code in fsZack Rusin
2008-02-12stop generate llvm entry pointsZack Rusin
entrypoints are useless because we use the same paths as all other code. also simplify llvm swizzling code
2008-02-12llvm ir code to fetch the elementsZack Rusin
2008-02-12gallium: initial code for wide/AA line drawingBrian
2008-02-12gallium: clean-up, simplification of mipmapped texturesBrian
Remove pipe_texture->first_level (always implicitly zero). This means there's never any unused mipmap levels at the top. In the state tracker, we no longer re-layout mipmapped textures if the MinLod/MaxLod texture parameters change. It's up to the driver to obey the pipe_sampler->min/max_lod clamps.
2008-02-11gallium: update comment about buffer map flagsBrian
2008-02-11gallium: fix fragment/vertex typoBrian
2008-02-11remove seperate llvm vs entry pointsZack Rusin
they're not necessary anymore. we use the same paths as sse and tgsi code
2008-02-11start genering soa type code in llvm pathsZack Rusin
2008-02-10disable llvm for fragment shaders for nowZack Rusin
2008-02-10rewrite the way cpu engine is handledZack Rusin
2008-02-10rewrite the way we handle ir in llvm codeZack Rusin
introduce intermediate step gallivm_ir before compiling it down to the final llvm ir.
2008-02-10add a stub of a lowering passZack Rusin
2008-02-08fix comment typosBrian
2008-02-08gallium: added draw_flush() call in softpipe_bind_sampler_state()Brian
2008-02-08failover: several fixes to failover pipe moduleJerome Glisse
2008-02-07tgsi: Fall back to interpreter instead of assert(0) on unimplemented SSE code.Michel Dänzer
2008-02-07pipebuffer: Fix reversed assertion.Michel Dänzer
2008-02-06Cell: silence unused var warningsBrian
2008-02-06gallium: #include p_debug.h since we use assertBrian
2008-02-06Cell: use mem_dup()Brian
2008-02-06gallium: use mem_dup()Brian
2008-02-06gallium: added mem_dup()Brian
2008-02-06gallium: change pipe->texture_create() to operate like the CSO functionsBrian
Now, pass in a template object and return a new object.
2008-02-06gallium: add bitmap/drawpixels texcoord bias supportBrian
The state tracker will call pipe->get_paramf(PIPE_CAP_BITMAP_TEXCOORD_BIAS) to get a bias factor for adjusting the texcoords used in bitmap/drawpixels. This allows us to compensate for small differences in rasterization from one device to another.
2008-02-07gallium: Bring latest fixes.José Fonseca
2008-02-06gallium: Use p_debug.h instead of non-portable stdio.h/assert.h functions.José Fonseca
2008-02-06gallium: Add forgotten return type.José Fonseca
2008-02-06gallium: Cross-platform debugging helpers.José Fonseca
2008-02-06gallium: Portability guidelines.José Fonseca
2008-02-05Cell: remove accidentally added OPT_FLAGS linesBrian
2008-02-05Cell: SIMD-ize tri_linear_coeff(), use vector float for vertex attributes in ↵Brian
struct vertex_header
2008-02-05Cell: re-enable inlined vertex buffersBrian
Vertex data must be on a 16-byte address/offset so SIMD operations will work properly in the SPU code.
2008-02-05Cell: remove dummy fields, update/add some commentsBrian
2008-02-05Cell: added cell_batch_alloc_aligned()Brian
2008-02-05Use _transpose_matrix4x4 from Cell SDK instead of my own versionIan Romanick
2008-02-05Add some debug messagesIan Romanick
2008-02-05More semi-trivial vectorization in the shader VMIan Romanick
2008-02-05Vectorize vertex pullerIan Romanick
2008-02-05Vectorize all micro opsIan Romanick
Fold single instruction micro ops inline. Remove unused micro ops.
2008-02-05gallium: Use align_free to free aligned memory.Michal Krol
2008-02-04Cell: fix some alignment issues by aligning commands to 8-byte boundariesBrian
Contributed by Ian Romanick. Also, temporarily disable inlined vertex buffers. They need to be 16-byte aligned...
2008-02-04Cell: don't use VEC_LITERAL macro, doesn't work w/ SDK 3.0Brian
2008-02-04Cell: don't use VEC_LITERAL macro, doesn't work w/ SDK 3.0Brian