| commit | 51e02409f0220c796d34f72b3a4b8ba3d3f34cb9 | |
| author | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | |
| Wed, 27 Apr 2022 23:28:44 +0000 (27 16:28 -0700) | ||
| committer | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | |
| Fri, 29 Apr 2022 18:14:11 +0000 (29 11:14 -0700) | ||
| tree | da948ebc16f1d61bb07f48909be0bb1567dd7c82 | treesnapshot (tar.gz zip) |
| parent | 5a7936401c0a5a8959273108051b094ffba84b8c | commitdiff |
[AMDGPU] Produce waitcounts for LDS DMA
MUBUF and FLAT LDS DMA operations need a wait on vmcnt before LDS written
can be accessed. A load from LDS to VMEM does not need a wait.
Differential Revision: https://reviews.llvm.org/D124626
MUBUF and FLAT LDS DMA operations need a wait on vmcnt before LDS written
can be accessed. A load from LDS to VMEM does not need a wait.
Differential Revision: https://reviews.llvm.org/D124626
| llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp | diffblobblamehistory | |
| llvm/test/CodeGen/AMDGPU/lds-dma-waitcnt.mir | [new file with mode: 0644] | blob |