[llvm] [AMDGPU] Disable inline constants for pseudo scalar transcendentals (PR #104395)

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Thu Aug 15 00:28:12 PDT 2024


================
@@ -5776,6 +5776,18 @@ bool SIInstrInfo::isOperandLegal(const MachineInstr &MI, unsigned OpIdx,
           return false;
       }
     }
+  } else if (ST.hasNoF16PseudoScalarTransInlineConstants() && isVOP3(MI) &&
----------------
arsenm wrote:

ST.hasNoF16PseudoScalarTransInlineConstants() && isVOP3(MI) are both redundant with the opcode check. Move the opcode switch into an isF16PseudoScalarTrans helper? 

https://github.com/llvm/llvm-project/pull/104395


More information about the llvm-commits mailing list