[PATCH] D157624: [AMDGPU] Treat KIMM32 and KIMM16 operand types as noninlinable

Mirko Brkusanin via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Fri Aug 11 09:54:53 PDT 2023


mbrkusanin added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/SIInstrInfo.cpp:5268-5269
         }
-      } else if (InstDesc.operands()[i].OperandType == AMDGPU::OPERAND_KIMM32 ||
-                 (AMDGPU::isSISrcOperand(InstDesc, i) &&
-                  !isInlineConstant(Op, InstDesc.operands()[i]))) {
+      } else if (AMDGPU::isSISrcOperand(InstDesc, i) &&
+                 !isInlineConstant(Op, InstDesc.operands()[i])) {
         if (!LiteralLimit--)
----------------
This check here was why test mentioned in https://reviews.llvm.org/D157103#4579608 failed, but only for fmaak_f16 not f32 version. Previously we only checked for KIMM32, but with this patch it's redundant to check either.



CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D157624/new/

https://reviews.llvm.org/D157624



More information about the llvm-commits mailing list