[PATCH] D69735: [amdgpu] Fix known bits compuation on `MUL_I24`/`MUL_U24`.

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

What was the symptom of this?



================
Comment at: llvm/test/CodeGen/AMDGPU/computeNumSignBits-mul.ll:163
+; GFX9: v_mov_b32_e32 v0, 0
+; GFX9-NEXT:    s_setpc_b64 s[30:31]
+define i32 @known_bits_mul24() {
----------------
Drop the return address register


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D69735/new/

https://reviews.llvm.org/D69735