[PATCH] D84194: [AMDGPU] Correct the number of SGPR blocks used for GFX9

Ronak Chauhan via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Jul 21 08:29:01 PDT 2020


rochauha marked an inline comment as done.
rochauha added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp:437-438
   // SGPRBlocks is actual number of SGPR blocks minus 1.
-  return NumSGPRs / getSGPREncodingGranule(STI) - 1;
+  unsigned NumSGPRBlocks = NumSGPRs / getSGPREncodingGranule(STI) - 1;
+  return isGFX9(*STI) ? NumSGPRBlocks * 2 : NumSGPRBlocks;
 }
----------------
foad wrote:
> Why have you changed this?
To follow the computation of `GRANULATED_WAVEFRONT_SGPR_COUNT` for GFX9, as mentioned in https://llvm.org/docs/AMDGPUUsage.html#amdgpu-amdhsa-compute-pgm-rsrc1-gfx6-gfx10-table


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D84194/new/

https://reviews.llvm.org/D84194





More information about the llvm-commits mailing list