| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | AMDGPU: Make i64 loads/stores promote to v2i32 | Matt Arsenault | 2016-05-02 | 1 | -2/+2 |
| | | | | | | | | | | | | | Now that unaligned access expansion should not attempt to produce i64 accesses, we can remove the hack in PreprocessISelDAG where this is done. This allows splitting i64 private accesses while allowing the new add nodes indexing the vector components can be folded with the base pointer arithmetic. llvm-svn: 268293 | ||||
| * | AMDGPU: Materialize sign bits with bfrev | Matt Arsenault | 2016-03-11 | 1 | -0/+158 |
| If a constant is the same as the reverse of an inline immediate, this is 4 bytes smaller than having to embed a 32-bit literal. llvm-svn: 263201 | |||||
 |
index : bcm5719-llvm | |
| Project Ortega BCM5719 LLVM | Raptor Computing Systems |
| summaryrefslogtreecommitdiffstats |
|