Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Revert "Temporarily Revert "Add basic loop fusion pass."" | Eric Christopher | 2019-04-17 | 1 | -0/+203 |
| | | | | | | | | The reversion apparently deleted the test/Transforms directory. Will be re-reverting again. llvm-svn: 358552 | ||||
* | Temporarily Revert "Add basic loop fusion pass." | Eric Christopher | 2019-04-17 | 1 | -203/+0 |
| | | | | | | | | As it's causing some bot failures (and per request from kbarton). This reverts commit r358543/ab70da07286e618016e78247e4a24fcb84077fda. llvm-svn: 358546 | ||||
* | Fix vectorization of canonicalize | Matt Arsenault | 2018-09-17 | 1 | -0/+17 |
| | | | | llvm-svn: 342390 | ||||
* | AMDGPU: Make v2i16/v2f16 legal on VI | Matt Arsenault | 2018-05-22 | 1 | -26/+17 |
| | | | | | | | | | | | | This usually results in better code. Fixes using inline asm with short2, and also fixes having a different ABI for function parameters between VI and gfx9. Partially cleans up the mess used for lowering of the d16 operations. Making v4f16 legal will help clean this up more, but this requires additional work. llvm-svn: 332953 | ||||
* | AMDGPU: Allow vectorization of packed types | Matt Arsenault | 2017-06-20 | 1 | -0/+195 |
llvm-svn: 305844 |