summaryrefslogtreecommitdiff
path: root/src/exa_wm_noca.g4b
AgeCommit message (Collapse)AuthorFilesLines
2008-04-10Revert "Compute pixel values directly into data port"Keith Packard1-8/+4
This reverts commit 346cf57deabb4c336612df4c13650a87b5ef6775. Mixing randr transforms and video caused screen corruption for Render operations. No, I don't understand why.
2008-04-10Compute pixel values directly into data portKeith Packard1-4/+8
Instead of leaving pixel values in src_sample registers, compute the pixel values directl to the data port to save 8 moves. This cannot work when no computation is done as there is both no way to wait for the sampler to finish and because the sampler returns data in a different order from that required by the data port (sigh).
2008-04-01Shrink WM thread to 32 registers and 1024 scratch space.Keith Packard1-4/+4
Saving registers means we can run more in parallel.
2008-03-31Fix composite with mask using new compositing thread codeKeith Packard1-4/+4
Clean up register allocation to never overlap Always write 4 values for each texture vertex.
2008-03-31Use m4 to clean up gen4 asm progs. Start adding projective transform support.Keith Packard1-0/+4
Use macros for register names, modularize functions into separate files.