summaryrefslogtreecommitdiff
path: root/src/gallium/drivers/llvmpipe
AgeCommit message (Collapse)Author
2009-08-29llvmpipe: keep texture stride 16byte alignedKeith Whitwell
Fixes crash when resizing gears window.
2009-08-29llvmpipe: Squash the quad pipeline.José Fonseca
It had been reduced to one fixed stage.
2009-08-29llvmpipe: s/uni/scalar/.José Fonseca
More obvious name.
2009-08-29llvmpipe: Add a bunch of comments.José Fonseca
Description/rationale/to-do items, while I still remember them...
2009-08-29llvmpipe: Split off vs stuff from lp_state_fs.c.José Fonseca
lp_state_fs.c is already too big without it.
2009-08-29llvmpipe: Avoid duplicate clamping when converting between types.José Fonseca
2009-08-29llvmpipe: Generate the fragment pipeline into a single function.José Fonseca
Still hackish. Will document and optimize later.
2009-08-29llvmpipe: Get conversions working correctly for all integer types.José Fonseca
2009-08-29llvmpipe: Delete the quad polygon stipple stage.José Fonseca
Not used now -- stipple done by the draw module. May code generate later.
2009-08-29llvmpipe: Early depth testing.José Fonseca
2009-08-29llvmpipe: Disassemble beyond referred code beyond the RET instruction.José Fonseca
2009-08-29llvmpipe: Split control flow function declarations and notes.José Fonseca
2009-08-29llvmpipe: Back port recent softpipe-opt improvements from Keith.José Fonseca
2009-08-29llvmpipe: Don't limit the number of tiles in the cache.José Fonseca
2009-08-29llvmpipe: Clean dead tile cache code.José Fonseca
2009-08-29llvmpipe: Simplify tile clears. Use pipe_fill_rect instead of scratch tile.José Fonseca
2009-08-29llvmpipe: Start hiding llvmpipe_cached_tile.José Fonseca
2009-08-29llvmpipe: Drop depth/stencil support from tile cache.José Fonseca
2009-08-29llvmpipe: Code generate the depth test, and include in the shader.José Fonseca
Only 32bit depth/stencil surfaces supported for now. Stencil ops not implemented yet.
2009-08-29llvmpipe: Update texture block from format.José Fonseca
2009-08-29llvmpipe: Depth test to LLVM IR translation.José Fonseca
2009-08-29llvmpipe: Centralize mask update logic.José Fonseca
2009-08-29llvmpipe: Clamped float to unsigned norm via mantissa manipulation.José Fonseca
2009-08-29llvmpipe: Several fixes of the conversion test.José Fonseca
2009-08-29llvmpipe: Generate instrinsics for integer comparisons.José Fonseca
It is a little messy, given the available instrucions form SIMD integer comparisons is rather limited.
2009-08-29llvmpipe: Mention the ability to get microbenchmarks from the unit tests.José Fonseca
Even better would be to have a -h/--help option to the tests...
2009-08-29llvmpipe: Code generate alpha testing and append to generated fragment shader.José Fonseca
2009-08-29llvmpipe: frag shader commentsBrian Paul
2009-08-29llvmpipe: blend commentsBrian Paul
2009-08-29llvmpipe: comments about blend generatorBrian Paul
2009-08-29llvmpipe: basic commentsBrian Paul
2009-08-29llvmpipe: Hack to force the line/point rasterization also fit the blend ↵José Fonseca
function expectations.
2009-08-29llvmpipe: Refer my udis86 branch which decodes more opcodes.José Fonseca
2009-08-29llvmpipe: Code generate the position interpolation.José Fonseca
2009-08-29llvmpipe: Simpler variant of lp_build_broadcast_scalar.José Fonseca
2009-08-29llvmpipe: Put the position coefficients together with the inputs.José Fonseca
The automatic search'n'replace leaves lp_setup.c a bit ugly, but this code will be eventually code generated as well.
2009-08-29llvmpipe: Implement saturation.José Fonseca
2009-08-29llvmpipe: Put color/depth directly into the quad structure.José Fonseca
2009-08-29llvmpipe: prefix point/line/tri_setup() functions with llvmpipe_Brian Paul
Avoid symbol collisions with softpipe.
2009-08-29llvmpipe: updated Makefile C_SOURCESBrian Paul
2009-08-29llvmpipe: Allow to build without udis86.José Fonseca
2009-08-29llvmpipe: Drop the FETCH/STORE macros.José Fonseca
2009-08-29llvmpipe: Debug helper function to name llvm intermediate values.José Fonseca
2009-08-29llvmpipe: use align_malloc for all structs containing ALIGN16 membersKeith Whitwell
Unless the struct is allocated aligned, aligning the members isn't very helpful.
2009-08-29llvmpipe: Document that llvm 2.5 is expected.José Fonseca
2009-08-29llvmpipe: Briefly describe current status.José Fonseca
2009-08-29llvmpipe: Blend in place.José Fonseca
2009-08-29llvmpipe: Code generate logic ops.José Fonseca
2009-08-29llvmpipe: Rasterize the quads according to what the blend function expects.José Fonseca
2009-08-29llvmpipe: Tile in SoA format identical to the blender function output.José Fonseca