| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel | Matt Arsenault | 2017-03-21 | 1 | -2/+2 |
| | | | | | | | | | | | | | 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/R600: Replace barrier intrinsics | Matt Arsenault | 2016-07-18 | 1 | -49/+9 |
| | | | | | llvm-svn: 275870 | ||||
| * | AMDGPU: Cleanup load tests | Matt Arsenault | 2016-06-02 | 1 | -0/+35 |
| | | | | | | | | | | There are a lot of different kinds of loads to test for, and these were scattered around inconsistently with some redundancy. Try to comprehensively test all loads in a consistent way. llvm-svn: 271571 | ||||
| * | AMDGPU: Run r600 tests last | Matt Arsenault | 2016-05-05 | 1 | -1/+1 |
| | | | | | llvm-svn: 268672 | ||||
| * | AMDGPU: Rework how private buffer passed for HSA | Matt Arsenault | 2015-11-30 | 1 | -2/+2 |
| | | | | | | | | | | | | | | | | | If we know we have stack objects, we reserve the registers that the private buffer resource and wave offset are passed and use them directly. If not, reserve the last 5 SGPRs just in case we need to spill. After register allocation, try to pick the next available registers instead of the last SGPRs, and then insert copies from the inputs to the reserved registers in the progloue. This also only selectively enables all of the input registers which are really required instead of always enabling them. llvm-svn: 254331 | ||||
| * | R600 -> AMDGPU rename | Tom Stellard | 2015-06-13 | 1 | -0/+49 |
| llvm-svn: 239657 | |||||
 |
index : bcm5719-llvm | |
| Project Ortega BCM5719 LLVM | Raptor Computing Systems |
| summaryrefslogtreecommitdiffstats |
|