; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=loop-vectorize -mtriple riscv64 -mattr=+v,+f -S 2>%t | FileCheck %s

; A collection of fairly basic functional tests when both fixed and scalable vectorization is
; allowed.  The primary goal of this is check for crashes during cost modeling, but it also
; exercises the default heuristics in a useful way.

define void @vector_add(ptr noalias nocapture %a, i64 %v, i64 %n) {
; CHECK-LABEL: @vector_add(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[V:%.*]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.vp.load.nxv2i64.p0(ptr align 8 [[TMP6]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP10]])
; CHECK-NEXT:    [[TMP8:%.*]] = add <vscale x 2 x i64> [[WIDE_LOAD]], [[BROADCAST_SPLAT]]
; CHECK-NEXT:    call void @llvm.vp.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP8]], ptr align 8 [[TMP6]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP10]])
; CHECK-NEXT:    [[TMP11:%.*]] = zext i32 [[TMP10]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP11]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP11]]
; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
  %arrayidx = getelementptr inbounds i64, ptr %a, i64 %iv
  %elem = load i64, ptr %arrayidx
  %add = add i64 %elem, %v
  store i64 %add, ptr %arrayidx
  %iv.next = add nuw nsw i64 %iv, 1
  %exitcond.not = icmp eq i64 %iv.next, 1024
  br i1 %exitcond.not, label %for.end, label %for.body

for.end:
  ret void
}

; Same as above, but with op type of i32.  We currently have a bug around
; etype=ELEN profitability in the vectorizer, and having a smaller element
; width test allows us to highlight different aspects of codegen.
define void @vector_add_i32(ptr noalias nocapture %a, i32 %v, i64 %n) {
; CHECK-LABEL: @vector_add_i32(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[V:%.*]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr align 4 [[TMP6]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP10]])
; CHECK-NEXT:    [[TMP8:%.*]] = add <vscale x 4 x i32> [[WIDE_LOAD]], [[BROADCAST_SPLAT]]
; CHECK-NEXT:    call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[TMP8]], ptr align 4 [[TMP6]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP10]])
; CHECK-NEXT:    [[TMP11:%.*]] = zext i32 [[TMP10]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP11]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP11]]
; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
  %arrayidx = getelementptr inbounds i32, ptr %a, i64 %iv
  %elem = load i32, ptr %arrayidx
  %add = add i32 %elem, %v
  store i32 %add, ptr %arrayidx
  %iv.next = add nuw nsw i64 %iv, 1
  %exitcond.not = icmp eq i64 %iv.next, 1024
  br i1 %exitcond.not, label %for.end, label %for.body

for.end:
  ret void
}


; a[b[i]] += v, mostly to exercise scatter/gather costing
; TODO: Currently fails to vectorize due to a memory conflict
define void @indexed_add(ptr noalias nocapture %a, ptr noalias nocapture %b, i64 %v, i64 %n) {
; CHECK-LABEL: @indexed_add(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.body:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
; CHECK-NEXT:    [[BADDR:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 [[IV]]
; CHECK-NEXT:    [[AIDX:%.*]] = load i64, ptr [[BADDR]], align 8
; CHECK-NEXT:    [[AADDR:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[AIDX]]
; CHECK-NEXT:    [[ELEM:%.*]] = load i64, ptr [[AADDR]], align 8
; CHECK-NEXT:    [[ADD:%.*]] = add i64 [[ELEM]], [[V:%.*]]
; CHECK-NEXT:    store i64 [[ADD]], ptr [[AADDR]], align 8
; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], 1024
; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_END:%.*]], label [[FOR_BODY]]
; CHECK:       for.end:
; CHECK-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
  %baddr = getelementptr inbounds i64, ptr %b, i64 %iv
  %aidx = load i64, ptr %baddr
  %aaddr = getelementptr inbounds i64, ptr %a, i64 %aidx
  %elem = load i64, ptr %aaddr
  %add = add i64 %elem, %v
  store i64 %add, ptr %aaddr
  %iv.next = add nuw nsw i64 %iv, 1
  %exitcond.not = icmp eq i64 %iv.next, 1024
  br i1 %exitcond.not, label %for.end, label %for.body

for.end:
  ret void
}

; a[b[i]] = v, exercise scatter support
define void @indexed_store(ptr noalias nocapture %a, ptr noalias nocapture %b, i64 %v, i64 %n) {
; CHECK-LABEL: @indexed_store(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[V:%.*]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.vp.load.nxv2i64.p0(ptr align 8 [[TMP6]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP10]])
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], <vscale x 2 x i64> [[WIDE_LOAD]]
; CHECK-NEXT:    call void @llvm.vp.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[BROADCAST_SPLAT]], <vscale x 2 x ptr> align 8 [[TMP8]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP10]])
; CHECK-NEXT:    [[TMP11:%.*]] = zext i32 [[TMP10]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP11]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP11]]
; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
  %baddr = getelementptr inbounds i64, ptr %b, i64 %iv
  %aidx = load i64, ptr %baddr
  %aaddr = getelementptr inbounds i64, ptr %a, i64 %aidx
  store i64 %v, ptr %aaddr
  %iv.next = add nuw nsw i64 %iv, 1
  %exitcond.not = icmp eq i64 %iv.next, 1024
  br i1 %exitcond.not, label %for.end, label %for.body

for.end:
  ret void
}

define i64 @indexed_load(ptr noalias nocapture %a, ptr noalias nocapture %b, i64 %v, i64 %n) {
; CHECK-LABEL: @indexed_load(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br 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:    [[VEC_PHI:%.*]] = phi <vscale x 2 x i64> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP9:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP12:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.vp.load.nxv2i64.p0(ptr align 8 [[TMP6]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP12]])
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], <vscale x 2 x i64> [[WIDE_LOAD]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 2 x i64> @llvm.vp.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> align 8 [[TMP8]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP12]])
; CHECK-NEXT:    [[TMP13:%.*]] = add <vscale x 2 x i64> [[VEC_PHI]], [[WIDE_MASKED_GATHER]]
; CHECK-NEXT:    [[TMP9]] = call <vscale x 2 x i64> @llvm.vp.merge.nxv2i64(<vscale x 2 x i1> splat (i1 true), <vscale x 2 x i64> [[TMP13]], <vscale x 2 x i64> [[VEC_PHI]], i32 [[TMP12]])
; CHECK-NEXT:    [[TMP10:%.*]] = zext i32 [[TMP12]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP10]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP10]]
; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[TMP11:%.*]] = call i64 @llvm.vector.reduce.add.nxv2i64(<vscale x 2 x i64> [[TMP9]])
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    ret i64 [[TMP11]]
;
entry:
  br label %for.body

for.body:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
  %sum = phi i64 [0, %entry], [%sum.next, %for.body]
  %baddr = getelementptr inbounds i64, ptr %b, i64 %iv
  %aidx = load i64, ptr %baddr
  %aaddr = getelementptr inbounds i64, ptr %a, i64 %aidx
  %elem = load i64, ptr %aaddr
  %iv.next = add nuw nsw i64 %iv, 1
  %sum.next = add i64 %sum, %elem
  %exitcond.not = icmp eq i64 %iv.next, 1024
  br i1 %exitcond.not, label %for.end, label %for.body

for.end:
  ret i64 %sum.next
}

define void @splat_int(ptr noalias nocapture %a, i64 %v, i64 %n) {
; CHECK-LABEL: @splat_int(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[V:%.*]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; CHECK-NEXT:    call void @llvm.vp.store.nxv2i64.p0(<vscale x 2 x i64> [[BROADCAST_SPLAT]], ptr align 8 [[TMP6]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP10]])
; CHECK-NEXT:    [[TMP7:%.*]] = zext i32 [[TMP10]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP7]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP7]]
; CHECK-NEXT:    [[TMP8:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
  %arrayidx = getelementptr inbounds i64, ptr %a, i64 %iv
  store i64 %v, ptr %arrayidx
  %iv.next = add nuw nsw i64 %iv, 1
  %exitcond.not = icmp eq i64 %iv.next, 1024
  br i1 %exitcond.not, label %for.end, label %for.body

for.end:
  ret void
}

define void @splat_ptr(ptr noalias nocapture %a, ptr %v, i64 %n) {
; CHECK-LABEL: @splat_ptr(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x ptr> poison, ptr [[V:%.*]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x ptr> [[BROADCAST_SPLATINSERT]], <vscale x 2 x ptr> poison, <vscale x 2 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; CHECK-NEXT:    call void @llvm.vp.store.nxv2p0.p0(<vscale x 2 x ptr> [[BROADCAST_SPLAT]], ptr align 8 [[TMP6]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP10]])
; CHECK-NEXT:    [[TMP7:%.*]] = zext i32 [[TMP10]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP7]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP7]]
; CHECK-NEXT:    [[TMP8:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.end:
; CHECK-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
  %arrayidx = getelementptr inbounds i64, ptr %a, i64 %iv
  store ptr %v, ptr %arrayidx
  %iv.next = add nuw nsw i64 %iv, 1
  %exitcond.not = icmp eq i64 %iv.next, 1024
  br i1 %exitcond.not, label %for.end, label %for.body

for.end:
  ret void
}