[PATCH] D96614: [AMDGPU][GlobalISel] Transform (fsub (fmul x, y), z) -> (fma x, y, -z)

[Mirko Brkusanin via Phabricator via llvm-commits]

mbrkusanin added inline comments.


================
Comment at: llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp:4687-4691
+      Register Src3 = Builder.buildFNeg(DstTy, RHS->getOperand(0).getReg())
+                             .getReg(0);
+      Builder.buildInstr(PreferredFusedOpcode, {MI.getOperand(0).getReg()},
+                         {LHS->getOperand(1).getReg(),
+                          LHS->getOperand(2).getReg(), Src3});
----------------
Rename Src3 to NegZ so it matches the comment above.
Use B not Builder.


================
Comment at: llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp:4700-4704
+      Register Src1 = Builder.buildFNeg(DstTy, RHS->getOperand(1).getReg())
+                             .getReg(0);
+      Builder.buildInstr(PreferredFusedOpcode, {MI.getOperand(0).getReg()},
+                         {Src1, RHS->getOperand(2).getReg(),
+                          LHS->getOperand(0).getReg()});
----------------
Rename Src1 to NegY so it matches the comment above.
Use B not Builder.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D96614/new/

https://reviews.llvm.org/D96614