llvm/test/Transforms/LoopVectorize/bzip_reverse_loops.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --filter-out-after "^scalar.ph:" --version 6
; RUN: opt < %s -passes=loop-vectorize -force-vector-interleave=1 -force-vector-width=4 -S -enable-if-conversion | FileCheck %s

target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"

define void @fc(ptr nocapture %p, i32 %n, i32 %size) {
; CHECK-LABEL: define void @fc(
; CHECK-SAME: ptr captures(none) [[P:%.*]], i32 [[N:%.*]], i32 [[SIZE:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[TMP0:%.*]] = add i32 [[N]], -1
; CHECK-NEXT:    [[TMP1:%.*]] = zext i32 [[TMP0]] to i64
; CHECK-NEXT:    [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 4
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 4
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP3:%.*]] = trunc i64 [[N_VEC]] to i32
; CHECK-NEXT:    [[TMP4:%.*]] = sub i32 [[N]], [[TMP3]]
; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[N_VEC]], -2
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP5]]
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[SIZE]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> 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:    [[TMP7:%.*]] = mul i64 [[INDEX]], -2
; CHECK-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP7]]
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i16, ptr [[NEXT_GEP]], i64 -1
; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i16, ptr [[TMP8]], i64 -3
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i16>, ptr [[TMP9]], align 2
; CHECK-NEXT:    [[REVERSE:%.*]] = shufflevector <4 x i16> [[WIDE_LOAD]], <4 x i16> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; CHECK-NEXT:    [[TMP10:%.*]] = zext <4 x i16> [[REVERSE]] to <4 x i32>
; CHECK-NEXT:    [[TMP11:%.*]] = icmp ult <4 x i32> [[TMP10]], [[BROADCAST_SPLAT]]
; CHECK-NEXT:    [[TMP12:%.*]] = sub <4 x i32> [[TMP10]], [[BROADCAST_SPLAT]]
; CHECK-NEXT:    [[TMP13:%.*]] = trunc <4 x i32> [[TMP12]] to <4 x i16>
; CHECK-NEXT:    [[PREDPHI:%.*]] = select <4 x i1> [[TMP11]], <4 x i16> zeroinitializer, <4 x i16> [[TMP13]]
; CHECK-NEXT:    [[REVERSE1:%.*]] = shufflevector <4 x i16> [[PREDPHI]], <4 x i16> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; CHECK-NEXT:    store <4 x i16> [[REVERSE1]], ptr [[TMP9]], align 2
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP14]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], [[DO_END:label %.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
;
entry:
  br label %do.body

do.body:
  %n.addr.0 = phi i32 [ %n, %entry ], [ %dec, %cond.end ]
  %p.addr.0 = phi ptr [ %p, %entry ], [ %incdec.ptr, %cond.end ]
  %incdec.ptr = getelementptr inbounds i16, ptr %p.addr.0, i64 -1
  %0 = load i16, ptr %incdec.ptr, align 2
  %conv = zext i16 %0 to i32
  %cmp = icmp ult i32 %conv, %size
  br i1 %cmp, label %cond.end, label %cond.true

cond.true:
  %sub = sub i32 %conv, %size
  %phitmp = trunc i32 %sub to i16
  br label %cond.end

cond.end:
  %cond = phi i16 [ %phitmp, %cond.true ], [ 0, %do.body ]
  store i16 %cond, ptr %incdec.ptr, align 2
  %dec = add i32 %n.addr.0, -1
  %tobool = icmp eq i32 %dec, 0
  br i1 %tobool, label %do.end, label %do.body

do.end:
  ret void
}

define void @example1(ptr nocapture %a, i32 %n, i32 %wsize) {
; CHECK-LABEL: define void @example1(
; CHECK-SAME: ptr captures(none) [[A:%.*]], i32 [[N:%.*]], i32 [[WSIZE:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    [[TMP0:%.*]] = add nsw i32 [[N]], -1
; CHECK-NEXT:    [[TMP1:%.*]] = zext i32 [[TMP0]] to i64
; CHECK-NEXT:    [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 4
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 4
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP3:%.*]] = trunc i64 [[N_VEC]] to i32
; CHECK-NEXT:    [[TMP4:%.*]] = sub i32 [[N]], [[TMP3]]
; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[N_VEC]], -4
; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr i8, ptr [[A]], i64 [[TMP5]]
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[WSIZE]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> 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:    [[TMP7:%.*]] = mul i64 [[INDEX]], -4
; CHECK-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[A]], i64 [[TMP7]]
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[NEXT_GEP]], i64 -1
; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, ptr [[TMP8]], i64 -3
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP9]], align 4
; CHECK-NEXT:    [[REVERSE:%.*]] = shufflevector <4 x i32> [[WIDE_LOAD]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; CHECK-NEXT:    [[TMP10:%.*]] = icmp slt <4 x i32> [[REVERSE]], [[BROADCAST_SPLAT]]
; CHECK-NEXT:    [[TMP11:%.*]] = sub nsw <4 x i32> [[REVERSE]], [[BROADCAST_SPLAT]]
; CHECK-NEXT:    [[TMP12:%.*]] = select <4 x i1> [[TMP10]], <4 x i32> zeroinitializer, <4 x i32> [[TMP11]]
; CHECK-NEXT:    [[REVERSE1:%.*]] = shufflevector <4 x i32> [[TMP12]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; CHECK-NEXT:    store <4 x i32> [[REVERSE1]], ptr [[TMP9]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP13]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], [[DO_END:label %.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
;
entry:
  br label %do.body

do.body:
  %n.addr.0 = phi i32 [ %n, %entry ], [ %dec, %do.body ]
  %p.0 = phi ptr [ %a, %entry ], [ %incdec.ptr, %do.body ]
  %incdec.ptr = getelementptr inbounds i32, ptr %p.0, i64 -1
  %0 = load i32, ptr %incdec.ptr, align 4
  %cmp = icmp slt i32 %0, %wsize
  %sub = sub nsw i32 %0, %wsize
  %cond = select i1 %cmp, i32 0, i32 %sub
  store i32 %cond, ptr %incdec.ptr, align 4
  %dec = add nsw i32 %n.addr.0, -1
  %tobool = icmp eq i32 %dec, 0
  br i1 %tobool, label %do.end, label %do.body

do.end:
  ret void
}