llvm/test/Transforms/LoopVectorize/AArch64/call-costs.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
; RUN: opt -p loop-vectorize -S %s | FileCheck %s

target triple = "arm64-apple-macosx11.0.0"

define void @fshl_operand_first_order_recurrence(ptr %dst, ptr noalias %src) {
; CHECK-LABEL: define void @fshl_operand_first_order_recurrence(
; CHECK-SAME: ptr [[DST:%.*]], ptr noalias [[SRC:%.*]]) {
; 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:    [[VECTOR_RECUR:%.*]] = phi <2 x i64> [ <i64 poison, i64 0>, %[[VECTOR_PH]] ], [ [[WIDE_LOAD1:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i64, ptr [[SRC]], i64 [[INDEX]]
; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[TMP2]], i64 2
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <2 x i64>, ptr [[TMP2]], align 8
; CHECK-NEXT:    [[WIDE_LOAD1]] = load <2 x i64>, ptr [[TMP5]], align 8
; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <2 x i64> [[VECTOR_RECUR]], <2 x i64> [[WIDE_LOAD]], <2 x i32> <i32 1, i32 2>
; CHECK-NEXT:    [[TMP7:%.*]] = shufflevector <2 x i64> [[WIDE_LOAD]], <2 x i64> [[WIDE_LOAD1]], <2 x i32> <i32 1, i32 2>
; CHECK-NEXT:    [[TMP8:%.*]] = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> splat (i64 1), <2 x i64> [[TMP6]], <2 x i64> splat (i64 1))
; CHECK-NEXT:    [[TMP9:%.*]] = call <2 x i64> @llvm.fshl.v2i64(<2 x i64> splat (i64 1), <2 x i64> [[TMP7]], <2 x i64> splat (i64 1))
; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr inbounds i64, ptr [[DST]], i64 [[INDEX]]
; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[TMP10]], i64 2
; CHECK-NEXT:    store <2 x i64> [[TMP8]], ptr [[TMP10]], align 8
; CHECK-NEXT:    store <2 x i64> [[TMP9]], ptr [[TMP13]], align 8
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], 100
; CHECK-NEXT:    br i1 [[TMP14]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <2 x i64> [[WIDE_LOAD1]], i32 1
; CHECK-NEXT:    br label %[[SCALAR_PH:.*]]
; CHECK:       [[SCALAR_PH]]:
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 100, %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[RECUR:%.*]] = phi i64 [ [[VECTOR_RECUR_EXTRACT]], %[[SCALAR_PH]] ], [ [[L:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[GEP_SRC:%.*]] = getelementptr inbounds i64, ptr [[SRC]], i64 [[IV]]
; CHECK-NEXT:    [[L]] = load i64, ptr [[GEP_SRC]], align 8
; CHECK-NEXT:    [[OR:%.*]] = tail call i64 @llvm.fshl.i64(i64 1, i64 [[RECUR]], i64 1)
; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr inbounds i64, ptr [[DST]], i64 [[IV]]
; CHECK-NEXT:    store i64 [[OR]], ptr [[GEP_DST]], align 8
; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV]], 100
; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label %[[EXIT:.*]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %recur = phi i64 [ 0, %entry ], [ %l, %loop ]
  %gep.src = getelementptr inbounds i64, ptr %src, i64 %iv
  %l = load i64, ptr %gep.src, align 8
  %or = tail call i64 @llvm.fshl.i64(i64 1, i64 %recur, i64 1)
  %gep.dst = getelementptr inbounds i64, ptr %dst, i64 %iv
  store i64 %or, ptr %gep.dst, align 8
  %iv.next = add i64 %iv, 1
  %exitcond.not = icmp eq i64 %iv, 100
  br i1 %exitcond.not, label %exit, label %loop

exit:
  ret void
}

; Test case for https://github.com/llvm/llvm-project/issues/107016.
define void @powi_call(ptr %P) {
; CHECK-LABEL: define void @powi_call(
; CHECK-SAME: ptr [[P:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*:]]
; CHECK-NEXT:    br label %[[VECTOR_PH:.*]]
; CHECK:       [[VECTOR_PH]]:
; CHECK-NEXT:    br label %[[VECTOR_BODY:.*]]
; CHECK:       [[VECTOR_BODY]]:
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <2 x double>, ptr [[P]], align 8
; CHECK-NEXT:    [[TMP3:%.*]] = call <2 x double> @llvm.powi.v2f64.i32(<2 x double> [[WIDE_LOAD]], i32 3)
; CHECK-NEXT:    store <2 x double> [[TMP3]], ptr [[P]], align 8
; CHECK-NEXT:    br label %[[MIDDLE_BLOCK:.*]]
; CHECK:       [[MIDDLE_BLOCK]]:
; CHECK-NEXT:    br label %[[EXIT:.*]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %gep = getelementptr inbounds double, ptr %P, i64 %iv
  %l = load double, ptr %gep
  %powi = tail call double @llvm.powi.f64.i32(double %l, i32 3)
  store double %powi, ptr %gep, align 8
  %iv.next = add i64 %iv, 1
  %ec = icmp eq i64 %iv, 1
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

define void @call_scalarized(ptr noalias %src, ptr noalias %dst) {
; CHECK-LABEL: define void @call_scalarized(
; CHECK-SAME: ptr noalias [[SRC:%.*]], ptr noalias [[DST:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    br label %[[LOOP_HEADER:.*]]
; CHECK:       [[LOOP_HEADER]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 100, %[[ENTRY]] ], [ [[IV_NEXT:%.*]], %[[LOOP_LATCH:.*]] ]
; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], -1
; CHECK-NEXT:    [[GEP_SRC:%.*]] = getelementptr double, ptr [[SRC]], i64 [[IV_NEXT]]
; CHECK-NEXT:    [[L:%.*]] = load double, ptr [[GEP_SRC]], align 8
; CHECK-NEXT:    [[CMP295:%.*]] = fcmp une double [[L]], 4.000000e+00
; CHECK-NEXT:    [[CMP299:%.*]] = fcmp ugt double [[L]], 0.000000e+00
; CHECK-NEXT:    [[OR_COND:%.*]] = or i1 [[CMP295]], [[CMP299]]
; CHECK-NEXT:    br i1 [[OR_COND]], label %[[LOOP_LATCH]], label %[[THEN:.*]]
; CHECK:       [[THEN]]:
; CHECK-NEXT:    [[SQRT:%.*]] = call double @llvm.sqrt.f64(double [[L]])
; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr double, ptr [[DST]], i64 [[IV_NEXT]]
; CHECK-NEXT:    store double [[SQRT]], ptr [[GEP_DST]], align 8
; CHECK-NEXT:    br label %[[LOOP_LATCH]]
; CHECK:       [[LOOP_LATCH]]:
; CHECK-NEXT:    [[TOBOOL_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], 0
; CHECK-NEXT:    br i1 [[TOBOOL_NOT]], label %[[EXIT:.*]], label %[[LOOP_HEADER]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop.header

loop.header:
  %iv = phi i64 [ 100, %entry ], [ %iv.next, %loop.latch ]
  %iv.next = add i64 %iv, -1
  %gep.src = getelementptr double, ptr %src, i64 %iv.next
  %l = load double, ptr %gep.src, align 8
  %cmp295 = fcmp une double %l, 4.000000e+00
  %cmp299 = fcmp ugt double %l, 0.000000e+00
  %or.cond = or i1 %cmp295, %cmp299
  br i1 %or.cond, label %loop.latch, label %then

then:
  %sqrt = call double @llvm.sqrt.f64(double %l)
  %gep.dst = getelementptr double, ptr %dst, i64 %iv.next
  store double %sqrt, ptr %gep.dst, align 8
  br label %loop.latch

loop.latch:
  %tobool.not = icmp eq i64 %iv.next, 0
  br i1 %tobool.not, label %exit, label %loop.header

exit:
  ret void
}

define void @call_forced_scalar(ptr %src.1, ptr %src.2, ptr noalias %dst.1, ptr noalias %dst.2) {
; CHECK-LABEL: define void @call_forced_scalar(
; CHECK-SAME: ptr [[SRC_1:%.*]], ptr [[SRC_2:%.*]], ptr noalias [[DST_1:%.*]], ptr noalias [[DST_2:%.*]]) {
; CHECK-NEXT:  [[ENTRY:.*]]:
; CHECK-NEXT:    br label %[[LOOP:.*]]
; CHECK:       [[LOOP]]:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, %[[ENTRY]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[SRC_1]], align 4
; CHECK-NEXT:    [[SMAX:%.*]] = tail call i32 @llvm.smax.i32(i32 [[TMP0]], i32 0)
; CHECK-NEXT:    [[UMIN:%.*]] = tail call i32 @llvm.umin.i32(i32 [[SMAX]], i32 1)
; CHECK-NEXT:    [[UMIN_EXT:%.*]] = zext i32 [[UMIN]] to i64
; CHECK-NEXT:    [[GEP_SRC_2:%.*]] = getelementptr i8, ptr [[SRC_2]], i64 [[UMIN_EXT]]
; CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr [[GEP_SRC_2]], align 1
; CHECK-NEXT:    [[L_EXT:%.*]] = zext i8 [[TMP1]] to i32
; CHECK-NEXT:    [[MUL:%.*]] = mul i32 3, [[L_EXT]]
; CHECK-NEXT:    store i32 [[MUL]], ptr [[DST_1]], align 4
; CHECK-NEXT:    [[GEP_DST_2:%.*]] = getelementptr i32, ptr [[DST_2]], i64 [[IV]]
; CHECK-NEXT:    store i32 0, ptr [[GEP_DST_2]], align 4
; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 0
; CHECK-NEXT:    br i1 [[EC]], label %[[EXIT:.*]], label %[[LOOP]]
; CHECK:       [[EXIT]]:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %0 = load i32, ptr %src.1, align 4
  %smax = tail call i32 @llvm.smax.i32(i32 %0, i32 0)
  %umin = tail call i32 @llvm.umin.i32(i32 %smax, i32 1)
  %umin.ext = zext i32 %umin to i64
  %gep.src.2 = getelementptr i8, ptr %src.2, i64 %umin.ext
  %1 = load i8, ptr %gep.src.2, align 1
  %l.ext = zext i8 %1 to i32
  %mul = mul i32 3, %l.ext
  store i32 %mul, ptr %dst.1, align 4
  %gep.dst.2 = getelementptr i32, ptr %dst.2, i64 %iv
  store i32 0, ptr %gep.dst.2, align 4
  %iv.next = add i64 %iv, 1
  %ec = icmp eq i64 %iv.next, 0
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

declare i32 @llvm.smax.i32(i32, i32)
declare i32 @llvm.umin.i32(i32, i32)
declare double @llvm.sqrt.f64(double)
declare double @llvm.powi.f64.i32(double, i32)
declare i64 @llvm.fshl.i64(i64, i64, i64)
;.
; 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]]}
;.