[llvm] [AMDGPU] Fix bad removal of s_delay_alu (PR #145728)
Jay Foad via llvm-commits
llvm-commits at lists.llvm.org
Thu Jun 26 00:13:21 PDT 2025
================
@@ -3070,6 +3070,7 @@ define amdgpu_kernel void @flat_atomic_inc_ret_i32_offset_addr64(ptr %out, ptr %
; GFX12-NEXT: v_dual_mov_b32 v0, s0 :: v_dual_mov_b32 v1, s1
; GFX12-NEXT: v_add_co_u32 v0, vcc_lo, v0, v2
; GFX12-NEXT: s_wait_alu 0xfffd
+; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_2)
----------------
jayfoad wrote:
Not your fault, but I think this delay is unnecessary because of the fast-forward path from add-with-carry-out to add-with-carry-in.
https://github.com/llvm/llvm-project/pull/145728
More information about the llvm-commits
mailing list