[AMDGPU] Resubmit SDWA peephole: enable by default
Reviewers: vpykhtin, rampitec, arsenm
Subscribers: qcolombet, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye
Differential Revision: https://reviews.llvm.org/D31671
llvm-svn: 299654
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.rint.f16.ll b/llvm/test/CodeGen/AMDGPU/llvm.rint.f16.ll
index f04d3dc..f566556 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.rint.f16.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.rint.f16.ll
@@ -26,19 +26,20 @@
; GCN-LABEL: {{^}}rint_v2f16
; GCN: buffer_load_dword v[[A_V2_F16:[0-9]+]]
; SI: v_cvt_f32_f16_e32 v[[A_F32_0:[0-9]+]], v[[A_V2_F16]]
-; SI: v_lshrrev_b32_e32 v[[A_F16_1:[0-9]+]], 16, v[[A_V2_F16]]
+; SI: v_lshrrev_b32_e32 v[[A_F16_1:[0-9]+]], 16, v[[A_V2_F16]]
; SI: v_cvt_f32_f16_e32 v[[A_F32_1:[0-9]+]], v[[A_F16_1]]
; SI: v_rndne_f32_e32 v[[R_F32_0:[0-9]+]], v[[A_F32_0]]
; SI: v_cvt_f16_f32_e32 v[[R_F16_0:[0-9]+]], v[[R_F32_0]]
; SI: v_rndne_f32_e32 v[[R_F32_1:[0-9]+]], v[[A_F32_1]]
; SI: v_cvt_f16_f32_e32 v[[R_F16_1:[0-9]+]], v[[R_F32_1]]
+; SI-DAG: v_lshlrev_b32_e32 v[[R_F16_HI:[0-9]+]], 16, v[[R_F16_1]]
+; SI-NOT: v_and_b32
+; SI: v_or_b32_e32 v[[R_V2_F16:[0-9]+]], v[[R_F16_HI]], v[[R_F16_0]]
-; VI-DAG: v_lshrrev_b32_e32 v[[A_F16_1:[0-9]+]], 16, v[[A_V2_F16]]
-; VI-DAG: v_rndne_f16_e32 v[[R_F16_0:[0-9]+]], v[[A_V2_F16]]
-; VI-DAG: v_rndne_f16_e32 v[[R_F16_1:[0-9]+]], v[[A_F16_1]]
-; SIVI-DAG: v_lshlrev_b32_e32 v[[R_F16_HI:[0-9]+]], 16, v[[R_F16_1]]
+; VI-DAG: v_rndne_f16_e32 v[[R_F16_0:[0-9]+]], v[[A_V2_F16]]
+; VI-DAG: v_rndne_f16_sdwa v[[R_F16_1:[0-9]+]], v[[A_V2_F16]] dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1
; VI-NOT: v_and_b32
-; SIVI: v_or_b32_e32 v[[R_V2_F16:[0-9]+]], v[[R_F16_HI]], v[[R_F16_0]]
+; VI: v_or_b32_e32 v[[R_V2_F16:[0-9]+]], v[[R_F16_1]], v[[R_F16_0]]
; GFX9: v_rndne_f16_e32 v[[R_F16_0:[0-9]+]], v[[A_V2_F16]]
; GFX9: v_lshrrev_b32_e32 v[[A_F16_1:[0-9]+]], 16, v[[A_V2_F16]]