[llvm] AMDGPU/GlobalISel: Fix inst-selection of ballot (PR #109986)
Petar Avramovic via llvm-commits
llvm-commits at lists.llvm.org
Mon Sep 30 05:17:41 PDT 2024
================
@@ -89,7 +90,8 @@ define amdgpu_cs i32 @branch_divergent_ballot_ne_zero_non_compare(i32 %v) {
; CHECK: ; %bb.0:
; CHECK-NEXT: v_and_b32_e32 v0, 1, v0
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc_lo, 0, v0
-; CHECK-NEXT: s_cmp_eq_u32 vcc_lo, 0
+; CHECK-NEXT: s_and_b32 s0, vcc_lo, exec_lo
----------------
petar-avramovic wrote:
This is really not obvious, Input is "vgpr to vcc trunc" that is selected as compare. Will become better after new reg bank select.
https://github.com/llvm/llvm-project/pull/109986
More information about the llvm-commits
mailing list