[llvm] 98dd390 - AMDGPU: Use removeAllRegUnitsForPhysReg()

[Ruiling Song via llvm-commits]

Author: Ruiling Song
Date: 2022-03-15T10:28:27+08:00
New Revision: 98dd390573dc0a350ce2aa79e31d57055700747d

URL: https://github.com/llvm/llvm-project/commit/98dd390573dc0a350ce2aa79e31d57055700747d
DIFF: https://github.com/llvm/llvm-project/commit/98dd390573dc0a350ce2aa79e31d57055700747d.diff

LOG: AMDGPU: Use removeAllRegUnitsForPhysReg()

I met the issue here when working on something else.
Actually we have already reserved EXEC, but it looks
like the register coalescer is causing the sub-register
of EXEC appears in LiveIntervals. I have not looked
deeper why register coalscer have such behavior, but
removeAllRegUnitsForPhysReg() is the right way.

Reviewed By: critson, foad, arsenm

Differential Revision: https://reviews.llvm.org/D117014

Added: 
    

Modified: 
    llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp b/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
index 720fc213f77cd..efc480e79aa6c 100644
--- a/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
+++ b/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
@@ -1588,11 +1588,11 @@ bool SIWholeQuadMode::runOnMachineFunction(MachineFunction &MF) {
   // Physical registers like SCC aren't tracked by default anyway, so just
   // removing the ranges we computed is the simplest option for maintaining
   // the analysis results.
-  LIS->removeRegUnit(*MCRegUnitIterator(MCRegister::from(AMDGPU::SCC), TRI));
+  LIS->removeAllRegUnitsForPhysReg(AMDGPU::SCC);
 
   // If we performed any kills then recompute EXEC
   if (!KillInstrs.empty())
-    LIS->removeRegUnit(*MCRegUnitIterator(AMDGPU::EXEC, TRI));
+    LIS->removeAllRegUnitsForPhysReg(AMDGPU::EXEC);
 
   return true;
 }