llvm/test/Transforms/LoopVectorize/forked-pointers.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -passes=loop-vectorize -force-vector-width=4 -S < %s 2>&1 | FileCheck %s

target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"

;;;; Derived from the following C code
;; void forked_ptrs_different_base_same_offset(ptr A, ptr B, ptr C, int *D) {
;;   for (int i=0; i<100; i++) {
;;     if (D[i] != 0) {
;;       C[i] = A[i];
;;     } else {
;;       C[i] = B[i];
;;     }
;;   }
;; }

define void @forked_ptrs_different_base_same_offset(ptr nocapture readonly %Base1, ptr nocapture readonly %Base2, ptr nocapture %Dest, ptr nocapture readonly %Preds) {
; CHECK-LABEL: @forked_ptrs_different_base_same_offset(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[BASE23:%.*]] = ptrtoaddr ptr [[BASE2:%.*]] to i64
; CHECK-NEXT:    [[BASE15:%.*]] = ptrtoaddr ptr [[BASE1:%.*]] to i64
; CHECK-NEXT:    [[PREDS9:%.*]] = ptrtoaddr ptr [[PREDS:%.*]] to i64
; CHECK-NEXT:    [[DEST8:%.*]] = ptrtoaddr ptr [[DEST:%.*]] to i64
; CHECK-NEXT:    br label [[VECTOR_MEMCHECK:%.*]]
; CHECK:       vector.memcheck:
; CHECK-NEXT:    [[TMP0:%.*]] = sub i64 [[DEST8]], [[PREDS9]]
; CHECK-NEXT:    [[TMP1:%.*]] = sub i64 [[TMP0]], 1
; CHECK-NEXT:    [[DIFF_CHECK:%.*]] = icmp ult i64 [[TMP1]], 15
; CHECK-NEXT:    [[TMP4:%.*]] = sub i64 [[DEST8]], [[BASE15]]
; CHECK-NEXT:    [[TMP3:%.*]] = sub i64 [[TMP4]], 1
; CHECK-NEXT:    [[DIFF_CHECK5:%.*]] = icmp ult i64 [[TMP3]], 15
; CHECK-NEXT:    [[DIFF_CHECK4:%.*]] = freeze i1 [[DIFF_CHECK5]]
; CHECK-NEXT:    [[CONFLICT_RDX:%.*]] = or i1 [[DIFF_CHECK]], [[DIFF_CHECK4]]
; CHECK-NEXT:    [[TMP2:%.*]] = sub i64 [[DEST8]], [[BASE23]]
; CHECK-NEXT:    [[TMP5:%.*]] = sub i64 [[TMP2]], 1
; CHECK-NEXT:    [[DIFF_CHECK7:%.*]] = icmp ult i64 [[TMP5]], 15
; CHECK-NEXT:    [[DIFF_CHECK6:%.*]] = freeze i1 [[DIFF_CHECK7]]
; CHECK-NEXT:    [[CONFLICT_RDX7:%.*]] = or i1 [[CONFLICT_RDX]], [[DIFF_CHECK6]]
; CHECK-NEXT:    br i1 [[CONFLICT_RDX7]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x ptr> poison, ptr [[BASE1]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x ptr> [[BROADCAST_SPLATINSERT]], <4 x ptr> poison, <4 x i32> zeroinitializer
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT8:%.*]] = insertelement <4 x ptr> poison, ptr [[BASE2]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT9:%.*]] = shufflevector <4 x ptr> [[BROADCAST_SPLATINSERT8]], <4 x ptr> 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:    [[TMP21:%.*]] = add i64 [[INDEX]], 1
; CHECK-NEXT:    [[TMP22:%.*]] = add i64 [[INDEX]], 2
; CHECK-NEXT:    [[TMP23:%.*]] = add i64 [[INDEX]], 3
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i32, ptr [[PREDS]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP6]], align 4
; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq <4 x i32> [[WIDE_LOAD]], zeroinitializer
; CHECK-NEXT:    [[TMP8:%.*]] = select <4 x i1> [[TMP7]], <4 x ptr> [[BROADCAST_SPLAT]], <4 x ptr> [[BROADCAST_SPLAT9]]
; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <4 x ptr> [[TMP8]], i64 0
; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr inbounds float, ptr [[TMP9]], i64 [[INDEX]]
; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <4 x ptr> [[TMP8]], i64 1
; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds float, ptr [[TMP11]], i64 [[TMP21]]
; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <4 x ptr> [[TMP8]], i64 2
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds float, ptr [[TMP13]], i64 [[TMP22]]
; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <4 x ptr> [[TMP8]], i64 3
; CHECK-NEXT:    [[TMP16:%.*]] = getelementptr inbounds float, ptr [[TMP15]], i64 [[TMP23]]
; CHECK-NEXT:    [[TMP17:%.*]] = load float, ptr [[TMP10]], align 4
; CHECK-NEXT:    [[TMP18:%.*]] = load float, ptr [[TMP12]], align 4
; CHECK-NEXT:    [[TMP19:%.*]] = load float, ptr [[TMP14]], align 4
; CHECK-NEXT:    [[TMP20:%.*]] = load float, ptr [[TMP16]], align 4
; CHECK-NEXT:    [[TMP29:%.*]] = insertelement <4 x float> poison, float [[TMP17]], i64 0
; CHECK-NEXT:    [[TMP30:%.*]] = insertelement <4 x float> [[TMP29]], float [[TMP18]], i64 1
; CHECK-NEXT:    [[TMP31:%.*]] = insertelement <4 x float> [[TMP30]], float [[TMP19]], i64 2
; CHECK-NEXT:    [[TMP24:%.*]] = insertelement <4 x float> [[TMP31]], float [[TMP20]], i64 3
; CHECK-NEXT:    [[TMP25:%.*]] = getelementptr inbounds float, ptr [[DEST]], i64 [[INDEX]]
; CHECK-NEXT:    store <4 x float> [[TMP24]], ptr [[TMP25]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP26:%.*]] = icmp eq i64 [[INDEX_NEXT]], 100
; CHECK-NEXT:    br i1 [[TMP26]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[FOR_COND_CLEANUP:%.*]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.cond.cleanup:
; CHECK-NEXT:    ret void
; CHECK:       for.body:
; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ 0, [[SCALAR_PH]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[PREDS]], i64 [[INDVARS_IV]]
; CHECK-NEXT:    [[TMP27:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
; CHECK-NEXT:    [[CMP1_NOT:%.*]] = icmp eq i32 [[TMP27]], 0
; CHECK-NEXT:    [[SPEC_SELECT:%.*]] = select i1 [[CMP1_NOT]], ptr [[BASE1]], ptr [[BASE2]]
; CHECK-NEXT:    [[DOTSINK_IN:%.*]] = getelementptr inbounds float, ptr [[SPEC_SELECT]], i64 [[INDVARS_IV]]
; CHECK-NEXT:    [[DOTSINK:%.*]] = load float, ptr [[DOTSINK_IN]], align 4
; CHECK-NEXT:    [[TMP28:%.*]] = getelementptr inbounds float, ptr [[DEST]], i64 [[INDVARS_IV]]
; CHECK-NEXT:    store float [[DOTSINK]], ptr [[TMP28]], align 4
; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 100
; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
;
entry:
  br label %for.body

for.cond.cleanup:
  ret void

for.body:
  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
  %arrayidx = getelementptr inbounds i32, ptr %Preds, i64 %indvars.iv
  %0 = load i32, ptr %arrayidx, align 4
  %cmp1.not = icmp eq i32 %0, 0
  %spec.select = select i1 %cmp1.not, ptr %Base2, ptr %Base1
  %.sink.in = getelementptr inbounds float, ptr %spec.select, i64 %indvars.iv
  %.sink = load float, ptr %.sink.in, align 4
  %1 = getelementptr inbounds float, ptr %Dest, i64 %indvars.iv
  store float %.sink, ptr %1, align 4
  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
  %exitcond.not = icmp eq i64 %indvars.iv.next, 100
  br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}

;;;; Derived from the following C code
;; void forked_ptrs_same_base_different_offset(ptr A, ptr B, int *C) {
;;   int offset;
;;   for (int i = 0; i < 100; i++) {
;;     if (C[i] != 0)
;;       offset = i;
;;     else
;;       offset = i+1;
;;     B[i] = A[offset];
;;   }
;; }

define void @forked_ptrs_same_base_different_offset(ptr nocapture readonly %Base, ptr nocapture %Dest, ptr nocapture readonly %Preds) {
; CHECK-LABEL: @forked_ptrs_same_base_different_offset(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.cond.cleanup:
; CHECK-NEXT:    ret void
; CHECK:       for.body:
; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; CHECK-NEXT:    [[I_014:%.*]] = phi i32 [ 0, [[ENTRY]] ], [ [[ADD:%.*]], [[FOR_BODY]] ]
; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[PREDS:%.*]], i64 [[INDVARS_IV]]
; CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
; CHECK-NEXT:    [[CMP1_NOT:%.*]] = icmp eq i32 [[TMP0]], 0
; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; CHECK-NEXT:    [[ADD]] = add nuw nsw i32 [[I_014]], 1
; CHECK-NEXT:    [[TMP1:%.*]] = trunc i64 [[INDVARS_IV]] to i32
; CHECK-NEXT:    [[OFFSET_0:%.*]] = select i1 [[CMP1_NOT]], i32 [[ADD]], i32 [[TMP1]]
; CHECK-NEXT:    [[IDXPROM213:%.*]] = zext i32 [[OFFSET_0]] to i64
; CHECK-NEXT:    [[ARRAYIDX3:%.*]] = getelementptr inbounds float, ptr [[BASE:%.*]], i64 [[IDXPROM213]]
; CHECK-NEXT:    [[TMP2:%.*]] = load float, ptr [[ARRAYIDX3]], align 4
; CHECK-NEXT:    [[ARRAYIDX5:%.*]] = getelementptr inbounds float, ptr [[DEST:%.*]], i64 [[INDVARS_IV]]
; CHECK-NEXT:    store float [[TMP2]], ptr [[ARRAYIDX5]], align 4
; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 100
; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]]
;
entry:
  br label %for.body

for.cond.cleanup:
  ret void

for.body:
  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
  %i.014 = phi i32 [ 0, %entry ], [ %add, %for.body ]
  %arrayidx = getelementptr inbounds i32, ptr %Preds, i64 %indvars.iv
  %0 = load i32, ptr %arrayidx, align 4
  %cmp1.not = icmp eq i32 %0, 0
  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
  %add = add nuw nsw i32 %i.014, 1
  %1 = trunc i64 %indvars.iv to i32
  %offset.0 = select i1 %cmp1.not, i32 %add, i32 %1
  %idxprom213 = zext i32 %offset.0 to i64
  %arrayidx3 = getelementptr inbounds float, ptr %Base, i64 %idxprom213
  %2 = load float, ptr %arrayidx3, align 4
  %arrayidx5 = getelementptr inbounds float, ptr %Dest, i64 %indvars.iv
  store float %2, ptr %arrayidx5, align 4
  %exitcond.not = icmp eq i64 %indvars.iv.next, 100
  br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}