[PATCH] D144313: [AMDGPU] Improve the lowering of raw_buffer_load_{i8,i16} and struct_buffer_load_{i8,i16} intrinsics

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUCombinerHelper.cpp:391
+  Register Op0Reg = MI.getOperand(1).getReg();
+  MachineInstr *SubwordBufferLoad = (*(MRI.def_begin(Op0Reg))).getParent();
+  Register SignExtendInsnDst = MI.getOperand(0).getReg();
----------------
getVRegDef?


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUCombinerHelper.cpp:393
+  Register SignExtendInsnDst = MI.getOperand(0).getReg();
+  B.setInsertPt(*(MI.getParent()), MI);
+
----------------
The calling combiner code should really set the insert point for you, not sure why we still don't do that


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D144313/new/

https://reviews.llvm.org/D144313