| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | [AMDGPU] Allow SDWA in instructions with immediates and SGPRs | Stanislav Mekhanoshin | 2017-05-30 | 1 | -1/+1 |
| | | | | | | | | | | | | | | | | | An encoding does not allow to use SDWA in an instruction with scalar operands, either literals or SGPRs. That is however possible to copy these operands into a VGPR first. Several copies of the value are produced if multiple SDWA conversions were done. To cleanup MachineLICM (to hoist copies out of loops), MachineCSE (to remove duplicate copies) and SIFoldOperands (to replace SGPR to VGPR copy with immediate copy right to the VGPR) runs are added after the SDWA pass. Differential Revision: https://reviews.llvm.org/D33583 llvm-svn: 304219 | ||||
| * | AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel | Matt Arsenault | 2017-03-21 | 1 | -5/+5 |
| | | | | | | | | | | | | | Currently the default C calling convention functions are treated the same as compute kernels. Make this explicit so the default calling convention can be changed to a non-kernel. Converted with perl -pi -e 's/define void/define amdgpu_kernel void/' on the relevant test directories (and undoing in one place that actually wanted a non-kernel). llvm-svn: 298444 | ||||
| * | AMDGPU: Support commuting with immediate in src0 | Matt Arsenault | 2016-09-08 | 1 | -1/+1 |
| | | | | | llvm-svn: 280970 | ||||
| * | AMDGPU: Add fp legacy instruction intrinsics | Matt Arsenault | 2016-07-26 | 1 | -0/+54 |
| This could use some additional optimization work to use mad/mac legacy. llvm-svn: 276764 | |||||
 |
index : bcm5719-llvm | |
| Project Ortega BCM5719 LLVM | Raptor Computing Systems |
| summaryrefslogtreecommitdiffstats |
|