llvm/llvm/test/Transforms/LoopVectorize/uniform-blend.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
; RUN: opt -passes=loop-vectorize -force-vector-width=4 -S %s | FileCheck %s

@dst = external global [32 x i16], align 1

define void @blend_uniform_iv_trunc(i1 %c) {
; CHECK-LABEL: define void @blend_uniform_iv_trunc(
; CHECK-SAME: i1 [[C:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    br i1 false, label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = trunc i64 [[INDEX]] to i16
; CHECK-NEXT:    [[TMP1:%.*]] = add i16 [[TMP0]], 0
; CHECK-NEXT:    [[PREDPHI:%.*]] = select i1 [[C]], i16 [[TMP1]], i16 undef
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i16 [[PREDPHI]]
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i16, ptr [[TMP2]], i32 0
; CHECK-NEXT:    store <4 x i16> zeroinitializer, ptr [[TMP3]], align 2
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq i64 [[INDEX_NEXT]], 32
; CHECK-NEXT:    br i1 [[TMP4]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    br i1 true, label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 32, %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT:    br label %[[LOOP_HEADER:.*]]
; CHECK:       [[LOOP_HEADER]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP_LATCH:.*]] ]
; CHECK-NEXT:    [[IV_TRUNC_2:%.*]] = trunc i64 [[IV]] to i16
; CHECK-NEXT:    br i1 [[C]], label %[[LOOP_NEXT:.*]], label %[[LOOP_LATCH]]
; CHECK:       [[LOOP_NEXT]]:
; CHECK-NEXT:    br label %[[LOOP_LATCH]]
; CHECK:       [[LOOP_LATCH]]:
; CHECK-NEXT:    [[BLEND:%.*]] = phi i16 [ undef, %[[LOOP_HEADER]] ], [ [[IV_TRUNC_2]], %[[LOOP_NEXT]] ]
; CHECK-NEXT:    [[DST_PTR:%.*]] = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i16 [[BLEND]]
; CHECK-NEXT:    store i16 0, ptr [[DST_PTR]], align 2
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT:    [[CMP439:%.*]] = icmp ult i64 [[IV]], 31
; CHECK-NEXT:    br i1 [[CMP439]], label %[[LOOP_HEADER]], label %[[EXIT]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop.header

loop.header:                                      ; preds = %loop.latch, %entry
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.latch ]
  %iv.trunc.2 = trunc i64 %iv to i16
  br i1 %c, label %loop.next, label %loop.latch

loop.next:                                        ; preds = %loop.header
  br label %loop.latch

loop.latch:                                       ; preds = %loop.next, %loop.header
  %blend = phi i16 [ undef, %loop.header ], [ %iv.trunc.2, %loop.next ]
  %dst.ptr = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i16 %blend
  store i16 0, ptr %dst.ptr
  %iv.next = add nuw nsw i64 %iv, 1
  %cmp439 = icmp ult i64 %iv, 31
  br i1 %cmp439, label %loop.header, label %exit

exit:                                             ; preds = %loop.latch
  ret void
}

define void @blend_uniform_iv(i1 %c) {
; CHECK-LABEL: define void @blend_uniform_iv(
; CHECK-SAME: i1 [[C:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    br i1 false, label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT:    [[PREDPHI:%.*]] = select i1 [[C]], i64 [[TMP0]], i64 undef
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i64 [[PREDPHI]]
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i16, ptr [[TMP1]], i32 0
; CHECK-NEXT:    store <4 x i16> zeroinitializer, ptr [[TMP2]], align 2
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], 32
; CHECK-NEXT:    br i1 [[TMP3]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    br i1 true, label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 32, %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT:    br label %[[LOOP_HEADER:.*]]
; CHECK:       [[LOOP_HEADER]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP_LATCH:.*]] ]
; CHECK-NEXT:    br i1 [[C]], label %[[LOOP_NEXT:.*]], label %[[LOOP_LATCH]]
; CHECK:       [[LOOP_NEXT]]:
; CHECK-NEXT:    br label %[[LOOP_LATCH]]
; CHECK:       [[LOOP_LATCH]]:
; CHECK-NEXT:    [[BLEND:%.*]] = phi i64 [ undef, %[[LOOP_HEADER]] ], [ [[IV]], %[[LOOP_NEXT]] ]
; CHECK-NEXT:    [[DST_PTR:%.*]] = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i64 [[BLEND]]
; CHECK-NEXT:    store i16 0, ptr [[DST_PTR]], align 2
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT:    [[CMP439:%.*]] = icmp ult i64 [[IV]], 31
; CHECK-NEXT:    br i1 [[CMP439]], label %[[LOOP_HEADER]], label %[[EXIT]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;

entry:
  br label %loop.header

loop.header:                                      ; preds = %loop.latch, %entry
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.latch ]
  br i1 %c, label %loop.next, label %loop.latch

loop.next:                                        ; preds = %loop.header
  br label %loop.latch

loop.latch:                                       ; preds = %loop.next, %loop.header
  %blend = phi i64 [ undef, %loop.header ], [ %iv, %loop.next ]
  %dst.ptr = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i64 %blend
  store i16 0, ptr %dst.ptr
  %iv.next = add nuw nsw i64 %iv, 1
  %cmp439 = icmp ult i64 %iv, 31
  br i1 %cmp439, label %loop.header, label %exit

exit:                                             ; preds = %loop.latch
  ret void
}

define void @blend_chain_iv(i1 %c) {
; CHECK-LABEL: define void @blend_chain_iv(
; CHECK-SAME: i1 [[C:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    br i1 false, label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i1> poison, i1 [[C]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i1> [[BROADCAST_SPLATINSERT]], <4 x i1> poison, <4 x i32> zeroinitializer
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 0, i64 1, i64 2, i64 3>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = select <4 x i1> [[BROADCAST_SPLAT]], <4 x i1> [[BROADCAST_SPLAT]], <4 x i1> zeroinitializer
; CHECK-NEXT:    [[PREDPHI:%.*]] = select <4 x i1> [[TMP0]], <4 x i64> [[VEC_IND]], <4 x i64> undef
; CHECK-NEXT:    [[PREDPHI1:%.*]] = select <4 x i1> [[BROADCAST_SPLAT]], <4 x i64> [[PREDPHI]], <4 x i64> undef
; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <4 x i64> [[PREDPHI1]], i32 0
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i64 [[TMP1]]
; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <4 x i64> [[PREDPHI1]], i32 1
; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i64 [[TMP3]]
; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <4 x i64> [[PREDPHI1]], i32 2
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i64 [[TMP5]]
; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <4 x i64> [[PREDPHI1]], i32 3
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i64 [[TMP7]]
; CHECK-NEXT:    store i16 0, ptr [[TMP2]], align 2
; CHECK-NEXT:    store i16 0, ptr [[TMP4]], align 2
; CHECK-NEXT:    store i16 0, ptr [[TMP6]], align 2
; CHECK-NEXT:    store i16 0, ptr [[TMP8]], align 2
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i64> [[VEC_IND]], <i64 4, i64 4, i64 4, i64 4>
; CHECK-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], 32
; CHECK-NEXT:    br i1 [[TMP9]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    br i1 true, label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 32, %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT:    br label %[[LOOP_HEADER:.*]]
; CHECK:       [[LOOP_HEADER]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP_LATCH:.*]] ]
; CHECK-NEXT:    br i1 [[C]], label %[[LOOP_NEXT:.*]], label %[[LOOP_LATCH]]
; CHECK:       [[LOOP_NEXT]]:
; CHECK-NEXT:    br i1 [[C]], label %[[LOOP_NEXT_2:.*]], label %[[LOOP_NEXT_3:.*]]
; CHECK:       [[LOOP_NEXT_2]]:
; CHECK-NEXT:    br label %[[LOOP_NEXT_3]]
; CHECK:       [[LOOP_NEXT_3]]:
; CHECK-NEXT:    [[BLEND_1:%.*]] = phi i64 [ undef, %[[LOOP_NEXT]] ], [ [[IV]], %[[LOOP_NEXT_2]] ]
; CHECK-NEXT:    br label %[[LOOP_LATCH]]
; CHECK:       [[LOOP_LATCH]]:
; CHECK-NEXT:    [[BLEND:%.*]] = phi i64 [ undef, %[[LOOP_HEADER]] ], [ [[BLEND_1]], %[[LOOP_NEXT_3]] ]
; CHECK-NEXT:    [[DST_PTR:%.*]] = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i64 [[BLEND]]
; CHECK-NEXT:    store i16 0, ptr [[DST_PTR]], align 2
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT:    [[CMP439:%.*]] = icmp ult i64 [[IV]], 31
; CHECK-NEXT:    br i1 [[CMP439]], label %[[LOOP_HEADER]], label %[[EXIT]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;

entry:
  br label %loop.header

loop.header:                                      ; preds = %loop.latch, %entry
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.latch ]
  br i1 %c, label %loop.next, label %loop.latch

loop.next:                                        ; preds = %loop.header
  br i1 %c, label %loop.next.2, label %loop.next.3

loop.next.2:
  br label %loop.next.3

loop.next.3:
  %blend.1 = phi i64 [ undef, %loop.next ], [ %iv, %loop.next.2 ]
  br label %loop.latch

loop.latch:                                       ; preds = %loop.next, %loop.header
  %blend = phi i64 [ undef, %loop.header ], [ %blend.1, %loop.next.3 ]
  %dst.ptr = getelementptr inbounds [32 x i16], ptr @dst, i16 0, i64 %blend
  store i16 0, ptr %dst.ptr
  %iv.next = add nuw nsw i64 %iv, 1
  %cmp439 = icmp ult i64 %iv, 31
  br i1 %cmp439, label %loop.header, label %exit

exit:                                             ; preds = %loop.latch
  ret void
}
;.
; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
; CHECK: [[LOOP5]] = distinct !{[[LOOP5]], [[META2]], [[META1]]}
; CHECK: [[LOOP6]] = distinct !{[[LOOP6]], [[META1]], [[META2]]}
; CHECK: [[LOOP7]] = distinct !{[[LOOP7]], [[META2]], [[META1]]}
;.