[PATCH] D69661: [AMDGPU] Fix vccz after v_readlane/v_readfirstlane to vcc_lo/hi

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/test/CodeGen/AMDGPU/vccz-corrupt-bug-workaround.mir:124
+# GFX9:  $vcc = S_MOV_B64 $vcc
+# CHECK-NOT: $vcc = S_MOV_B64 $vcc
+# CHECK: S_CBRANCH_VCCNZ %bb.1, implicit killed $vcc
----------------
CHECK-NEXT would be less fragile


================
Comment at: llvm/test/CodeGen/AMDGPU/vccz-corrupt-bug-workaround.mir:141
+# CHECK: INLINEASM &"; def vcc", 1, 10, implicit-def $vcc
+# CHECK-NOT: $vcc = S_MOV_B64 $vcc
+# CHECK: S_CBRANCH_VCCNZ %bb.1, implicit killed $vcc
----------------
ditto


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D69661/new/

https://reviews.llvm.org/D69661