Age | Commit message (Collapse) | Author | |
---|---|---|---|
2009-08-29 | llvmpipe: SoA blending. | José Fonseca | |
Throughput seems to be 4x higher. | |||
2009-08-29 | llvmpipe: Centralize lp_build_context initialization. | José Fonseca | |
2009-08-29 | llvmpipe: Optimize blend swizzles by using bitmasks instead of shuffles for ↵ | José Fonseca | |
ubytes. | |||
2009-08-29 | llvmpipe: Separate constant building. | José Fonseca | |
2009-08-29 | llvmpipe: Move type support functions into a separate file. | José Fonseca | |
2009-08-29 | llvmpipe: Minor comments / debug mnemonics. | José Fonseca | |
2009-08-29 | llvmpipe: Improve generated 8bit blending code. | José Fonseca | |
2009-08-29 | llvmpipe: Specialize arithmetic operations. | José Fonseca | |
2009-08-29 | llvmpipe: Introduce a custom typing system. | José Fonseca | |
Straightforward representation of floating-point/fixed-point/integer, normalized/scaled, signed/unsigned SIMD vector types. | |||
2009-08-29 | llvmpipe: Blending. | José Fonseca | |
The code |