[PATCH] D105507: [AMDGPU] Add gfx10 assembler directive to specify shared VGPR count

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp:4880
+    // SharedVGPRCount < 16 checked by PARSE_ENTRY_BITS
+    if (SharedVGPRCount && EnableWavefrontSize32)
+      return TokError("shared_vgpr_count directive not valid on "
----------------
Braces


================
Comment at: llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp:4883
+                      "wavefront size 32");
+    printf("Shared*2+VGPR: %d\n", SharedVGPRCount*2 + VGPRBlocks);
+    if (SharedVGPRCount*2 + VGPRBlocks > 63)
----------------
Leftover debug printing


================
Comment at: llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp:4884
+    printf("Shared*2+VGPR: %d\n", SharedVGPRCount*2 + VGPRBlocks);
+    if (SharedVGPRCount*2 + VGPRBlocks > 63)
+      return TokError("shared_vgpr_count*2 + "
----------------
Braces, spaces around *


================
Comment at: llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp:4887
+                      "compute_pgm_rsrc1.GRANULATED_WORKITEM_VGPR_COUNT "
+                      "cannot exceed 63\n");
+  }
----------------
I don't think this needs the newline


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D105507/new/

https://reviews.llvm.org/D105507