[llvm] r356956 - AMDGPU: Remove unnecessary check for isFullCopy

[Matt Arsenault via llvm-commits]

Author: arsenm
Date: Mon Mar 25 14:28:53 2019
New Revision: 356956

URL: http://llvm.org/viewvc/llvm-project?rev=356956&view=rev
Log:
AMDGPU: Remove unnecessary check for isFullCopy

Subregister indexes are not used for physical register operands, so
isFullCopy is implied by the physical register check.

Modified:
    llvm/trunk/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp

Modified: llvm/trunk/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp?rev=356956&r1=356955&r2=356956&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp (original)
+++ llvm/trunk/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp Mon Mar 25 14:28:53 2019
@@ -76,7 +76,7 @@ static bool isEndCF(const MachineInstr&
 }
 
 static bool isFullExecCopy(const MachineInstr& MI) {
-  return MI.isFullCopy() && MI.getOperand(1).getReg() == AMDGPU::EXEC;
+  return MI.getOperand(1).getReg() == AMDGPU::EXEC;
 }
 
 static unsigned getOrNonExecReg(const MachineInstr &MI,