[PATCH] D119696: [AMDGPU] Improve v_cmpx usage on GFX10.3.

Thomas Symalla via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Feb 14 05:50:09 PST 2022


tsymalla added inline comments.


================
Comment at: llvm/test/CodeGen/AMDGPU/wqm.ll:385-386
 ; GFX10-W32-NEXT:    v_mbcnt_hi_u32_b32 v0, -1, v0
-; GFX10-W32-NEXT:    v_cmp_gt_u32_e32 vcc_lo, 16, v0
+; GFX10-W32-NEXT:    v_cmpx_gt_u32_e64 16, v0
 ; GFX10-W32-NEXT:    v_mov_b32_e32 v0, 0
 ; GFX10-W32-NEXT:    s_cbranch_execz .LBB9_2
----------------
sebastian-ne wrote:
> I think the v_cmpx instruction should be inserted the the place of the s_and_saveexec, not at the place of the v_cmp. Otherwise this v_mov gets executed with the wrong exec mask.
Thanks, going to fix this.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D119696/new/

https://reviews.llvm.org/D119696



More information about the llvm-commits mailing list