llvm/test/Transforms/LoopVectorize/induction-unroll-novec.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --version 5
; RUN: opt -passes=loop-vectorize -force-vector-interleave=2 -force-vector-width=1 -S %s | 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"

; Test for PR54427.
define void @test_nonconst_start_and_step(ptr %dst, i32 %start, i32 %step, i64 %N) {
; CHECK-LABEL: define void @test_nonconst_start_and_step(
; CHECK-SAME: ptr [[DST:%.*]], i32 [[START:%.*]], i32 [[STEP:%.*]], i64 [[N:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    [[NEG_STEP:%.*]] = sub i32 0, [[STEP]]
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N]], 2
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N]], 2
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]]
; CHECK-NEXT:    [[DOTCAST:%.*]] = trunc i64 [[N_VEC]] to i32
; CHECK-NEXT:    [[TMP1:%.*]] = mul i32 [[DOTCAST]], [[NEG_STEP]]
; CHECK-NEXT:    [[TMP4:%.*]] = add i32 [[START]], [[TMP1]]
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[INDUCTION4:%.*]] = add i64 [[INDEX]], 1
; CHECK-NEXT:    [[TMP2:%.*]] = trunc i64 [[INDEX]] to i32
; CHECK-NEXT:    [[TMP3:%.*]] = mul i32 [[TMP2]], [[NEG_STEP]]
; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i32 [[START]], [[TMP3]]
; CHECK-NEXT:    [[TMP5:%.*]] = mul i32 1, [[NEG_STEP]]
; CHECK-NEXT:    [[INDUCTION2:%.*]] = add i32 [[OFFSET_IDX]], [[TMP5]]
; CHECK-NEXT:    [[TMP6:%.*]] = sub nsw i32 [[OFFSET_IDX]], [[STEP]]
; CHECK-NEXT:    [[TMP7:%.*]] = sub nsw i32 [[INDUCTION2]], [[STEP]]
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[INDEX]]
; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[INDUCTION4]]
; CHECK-NEXT:    store i32 [[TMP6]], ptr [[TMP8]], align 2
; CHECK-NEXT:    store i32 [[TMP7]], ptr [[TMP9]], align 2
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP11]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT:    [[BC_RESUME_VAL2:%.*]] = phi i32 [ [[TMP4]], %[[MIDDLE_BLOCK]] ], [ [[START]], %[[ENTRY]] ]
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[PRIMARY_IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[PRIMARY_IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[IV_DOWN:%.*]] = phi i32 [ [[BC_RESUME_VAL2]], %[[SCALAR_PH]] ], [ [[IV_DOWN_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[IV_DOWN_NEXT]] = sub nsw i32 [[IV_DOWN]], [[STEP]]
; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[PRIMARY_IV]]
; CHECK-NEXT:    store i32 [[IV_DOWN_NEXT]], ptr [[GEP_DST]], align 2
; CHECK-NEXT:    [[PRIMARY_IV_NEXT]] = add nuw nsw i64 [[PRIMARY_IV]], 1
; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[PRIMARY_IV_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[EXITCOND]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %primary.iv = phi i64 [ 0, %entry ], [ %primary.iv.next, %loop ]
  %iv.down = phi i32 [ %start, %entry ], [ %iv.down.next, %loop ]
  %iv.down.next = sub nsw i32 %iv.down, %step
  %gep.dst = getelementptr inbounds i32, ptr %dst, i64 %primary.iv
  store i32 %iv.down.next, ptr %gep.dst, align 2
  %primary.iv.next = add nuw nsw i64 %primary.iv, 1
  %exitcond = icmp eq i64 %primary.iv.next, %N
  br i1 %exitcond, label %exit, label %loop

exit:
  ret void
}