[PATCH] D105507: [AMDGPU] Add gfx10 assembler directive to specify shared VGPR count
Tony Tye via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Mon Feb 7 14:05:14 PST 2022
t-tye added inline comments.
================
Comment at: llvm/docs/AMDGPUUsage.rst:4285-4287
+ 3:0 4 bits SHARED_VGPR_COUNT Number of shared VGPR blocks for wavefront size 64. Granularity of 8.
+ Value 0-15, representing 0-120 VGPRs.
+ compute_pgm_rsrc1.vgprs + shared_vgpr_count*2 cannot exceed 63.
----------------
Suggest reword to:
```
Number of shared VGPR blocks for wavefront size 64 when executing in subvector mode. For wavefront size 64 the value is 0-15, representing 0-120 VGPRs (granularity of 8), such that (compute_pgm_rsrc1.vgprs +1)*4 + shared_vgpr_count*8 does not exceed 256. For wavefront size 32 must be 0.
```
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D105507/new/
https://reviews.llvm.org/D105507
More information about the llvm-commits
mailing list