Re-land [openmp] Fix warnings when building on Windows with latest MSVC or Clang...

[llvm-project.git] / llvm / test / Transforms / LoopVectorize / AArch64 / masked-call.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=loop-vectorize,instsimplify -force-vector-interleave=1 -S | FileCheck %s --check-prefixes=TFNONE
; RUN: opt < %s -passes=loop-vectorize,instsimplify,simplifycfg -force-vector-interleave=1 -prefer-predicate-over-epilogue=predicate-dont-vectorize -S | FileCheck %s --check-prefixes=TFCOMMON,TFALWAYS
; RUN: opt < %s -passes=loop-vectorize,instsimplify,simplifycfg -force-vector-interleave=1 -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -S | FileCheck %s --check-prefixes=TFCOMMON,TFFALLBACK
; RUN: opt < %s -passes=loop-vectorize,instsimplify,simplifycfg -force-vector-interleave=2 -prefer-predicate-over-epilogue=predicate-dont-vectorize -S | FileCheck %s --check-prefixes=TFA_INTERLEAVE

target triple = "aarch64-unknown-linux-gnu"

; A call whose argument must be widened. We check that tail folding uses the
; primary mask, and that without tail folding we synthesize an all-true mask.
define void @test_widen(ptr noalias %a, ptr readnone %b) #4 {
; TFNONE-LABEL: @test_widen(
; TFNONE-NEXT:  entry:
; TFNONE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFNONE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1025, [[TMP1]]
; TFNONE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; TFNONE:       vector.ph:
; TFNONE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFNONE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1025, [[TMP3]]
; TFNONE-NEXT:    [[N_VEC:%.*]] = sub i64 1025, [[N_MOD_VF]]
; TFNONE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; TFNONE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFNONE:       vector.body:
; TFNONE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFNONE-NEXT:    [[TMP6:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 2 x i64>, ptr [[TMP6]], align 8
; TFNONE-NEXT:    [[TMP7:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_LOAD]], <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; TFNONE-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    store <vscale x 2 x i64> [[TMP7]], ptr [[TMP8]], align 8
; TFNONE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; TFNONE-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; TFNONE-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; TFNONE:       middle.block:
; TFNONE-NEXT:    br i1 false, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
; TFNONE:       scalar.ph:
; TFNONE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; TFNONE-NEXT:    br label [[FOR_BODY:%.*]]
; TFNONE:       for.body:
; TFNONE-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; TFNONE-NEXT:    [[GEP:%.*]] = getelementptr i64, ptr [[B]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    [[LOAD:%.*]] = load i64, ptr [[GEP]], align 8
; TFNONE-NEXT:    [[CALL:%.*]] = call i64 @foo(i64 [[LOAD]]) #[[ATTR3:[0-9]+]]
; TFNONE-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    store i64 [[CALL]], ptr [[ARRAYIDX]], align 8
; TFNONE-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; TFNONE-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 1025
; TFNONE-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
; TFNONE:       for.cond.cleanup:
; TFNONE-NEXT:    ret void
;
; TFCOMMON-LABEL: @test_widen(
; TFCOMMON-NEXT:  entry:
; TFCOMMON-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFCOMMON-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFCOMMON-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFCOMMON-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFCOMMON-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFCOMMON-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFCOMMON-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFCOMMON-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFCOMMON-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFCOMMON-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 2
; TFCOMMON-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFCOMMON-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFCOMMON:       vector.body:
; TFCOMMON-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFCOMMON-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFCOMMON-NEXT:    [[TMP7:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFCOMMON-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP7]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i64> poison)
; TFCOMMON-NEXT:    [[TMP8:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFCOMMON-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFCOMMON-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP8]], ptr [[TMP9]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFCOMMON-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP6]]
; TFCOMMON-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFCOMMON-NEXT:    [[TMP10:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFCOMMON-NEXT:    [[TMP11:%.*]] = extractelement <vscale x 2 x i1> [[TMP10]], i32 0
; TFCOMMON-NEXT:    br i1 [[TMP11]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; TFCOMMON:       for.cond.cleanup:
; TFCOMMON-NEXT:    ret void
;
; TFA_INTERLEAVE-LABEL: @test_widen(
; TFA_INTERLEAVE-NEXT:  entry:
; TFA_INTERLEAVE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFA_INTERLEAVE-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFA_INTERLEAVE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFA_INTERLEAVE-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFA_INTERLEAVE-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 2
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY1:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP8]], i64 1025)
; TFA_INTERLEAVE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFA_INTERLEAVE:       vector.body:
; TFA_INTERLEAVE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT4:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK2:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY1]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT5:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP10:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP11:%.*]] = mul i64 [[TMP10]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP12:%.*]] = getelementptr i64, ptr [[TMP9]], i64 [[TMP11]]
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP9]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i64> poison)
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD3:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP12]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x i64> poison)
; TFA_INTERLEAVE-NEXT:    [[TMP13:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFA_INTERLEAVE-NEXT:    [[TMP14:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD3]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]])
; TFA_INTERLEAVE-NEXT:    [[TMP15:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP16:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP17:%.*]] = mul i64 [[TMP16]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i64, ptr [[TMP15]], i64 [[TMP17]]
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP13]], ptr [[TMP15]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP14]], ptr [[TMP18]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]])
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT:%.*]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT4]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[TMP19:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP20:%.*]] = mul i64 [[TMP19]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP21:%.*]] = add i64 [[INDEX_NEXT]], [[TMP20]]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT5]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP21]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[TMP22:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP23:%.*]] = extractelement <vscale x 2 x i1> [[TMP22]], i32 0
; TFA_INTERLEAVE-NEXT:    br i1 [[TMP23]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; TFA_INTERLEAVE:       for.cond.cleanup:
; TFA_INTERLEAVE-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
  %gep = getelementptr i64, ptr %b, i64 %indvars.iv
  %load = load i64, ptr %gep
  %call = call i64 @foo(i64 %load) #1
  %arrayidx = getelementptr inbounds i64, ptr %a, i64 %indvars.iv
  store i64 %call, ptr %arrayidx
  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
  %exitcond = icmp eq i64 %indvars.iv.next, 1025
  br i1 %exitcond, label %for.cond.cleanup, label %for.body

for.cond.cleanup:
  ret void
}

; Check that a simple conditional call can be vectorized.
define void @test_if_then(ptr noalias %a, ptr readnone %b) #4 {
; TFNONE-LABEL: @test_if_then(
; TFNONE-NEXT:  entry:
; TFNONE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFNONE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1025, [[TMP1]]
; TFNONE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; TFNONE:       vector.ph:
; TFNONE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFNONE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1025, [[TMP3]]
; TFNONE-NEXT:    [[N_VEC:%.*]] = sub i64 1025, [[N_MOD_VF]]
; TFNONE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; TFNONE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFNONE:       vector.body:
; TFNONE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFNONE-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 2 x i64>, ptr [[TMP6]], align 8
; TFNONE-NEXT:    [[TMP7:%.*]] = icmp ugt <vscale x 2 x i64> [[WIDE_LOAD]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 50, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; TFNONE-NEXT:    [[TMP8:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_LOAD]], <vscale x 2 x i1> [[TMP7]])
; TFNONE-NEXT:    [[TMP9:%.*]] = xor <vscale x 2 x i1> [[TMP7]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFNONE-NEXT:    [[PREDPHI:%.*]] = select <vscale x 2 x i1> [[TMP9]], <vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> [[TMP8]]
; TFNONE-NEXT:    [[TMP10:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    store <vscale x 2 x i64> [[PREDPHI]], ptr [[TMP10]], align 8
; TFNONE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; TFNONE-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; TFNONE-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; TFNONE:       middle.block:
; TFNONE-NEXT:    br i1 false, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
; TFNONE:       scalar.ph:
; TFNONE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; TFNONE-NEXT:    br label [[FOR_BODY:%.*]]
; TFNONE:       for.body:
; TFNONE-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[IF_END:%.*]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
; TFNONE-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    [[TMP12:%.*]] = load i64, ptr [[ARRAYIDX]], align 8
; TFNONE-NEXT:    [[CMP:%.*]] = icmp ugt i64 [[TMP12]], 50
; TFNONE-NEXT:    br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_END]]
; TFNONE:       if.then:
; TFNONE-NEXT:    [[TMP13:%.*]] = call i64 @foo(i64 [[TMP12]]) #[[ATTR3]]
; TFNONE-NEXT:    br label [[IF_END]]
; TFNONE:       if.end:
; TFNONE-NEXT:    [[TMP14:%.*]] = phi i64 [ [[TMP13]], [[IF_THEN]] ], [ 0, [[FOR_BODY]] ]
; TFNONE-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    store i64 [[TMP14]], ptr [[ARRAYIDX1]], align 8
; TFNONE-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; TFNONE-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 1025
; TFNONE-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; TFNONE:       for.cond.cleanup:
; TFNONE-NEXT:    ret void
;
; TFCOMMON-LABEL: @test_if_then(
; TFCOMMON-NEXT:  entry:
; TFCOMMON-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFCOMMON-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFCOMMON-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFCOMMON-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFCOMMON-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFCOMMON-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFCOMMON-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFCOMMON-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFCOMMON-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFCOMMON-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 2
; TFCOMMON-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFCOMMON-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFCOMMON:       vector.body:
; TFCOMMON-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFCOMMON-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFCOMMON-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFCOMMON-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP7]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i64> poison)
; TFCOMMON-NEXT:    [[TMP8:%.*]] = icmp ugt <vscale x 2 x i64> [[WIDE_MASKED_LOAD]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 50, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; TFCOMMON-NEXT:    [[TMP9:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i1> [[TMP8]], <vscale x 2 x i1> zeroinitializer
; TFCOMMON-NEXT:    [[TMP10:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD]], <vscale x 2 x i1> [[TMP9]])
; TFCOMMON-NEXT:    [[TMP11:%.*]] = xor <vscale x 2 x i1> [[TMP8]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFCOMMON-NEXT:    [[TMP12:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i1> [[TMP11]], <vscale x 2 x i1> zeroinitializer
; TFCOMMON-NEXT:    [[TMP13:%.*]] = or <vscale x 2 x i1> [[TMP9]], [[TMP12]]
; TFCOMMON-NEXT:    [[PREDPHI:%.*]] = select <vscale x 2 x i1> [[TMP12]], <vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> [[TMP10]]
; TFCOMMON-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFCOMMON-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[PREDPHI]], ptr [[TMP14]], i32 8, <vscale x 2 x i1> [[TMP13]])
; TFCOMMON-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP6]]
; TFCOMMON-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFCOMMON-NEXT:    [[TMP15:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFCOMMON-NEXT:    [[TMP16:%.*]] = extractelement <vscale x 2 x i1> [[TMP15]], i32 0
; TFCOMMON-NEXT:    br i1 [[TMP16]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
; TFCOMMON:       for.cond.cleanup:
; TFCOMMON-NEXT:    ret void
;
; TFA_INTERLEAVE-LABEL: @test_if_then(
; TFA_INTERLEAVE-NEXT:  entry:
; TFA_INTERLEAVE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFA_INTERLEAVE-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFA_INTERLEAVE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFA_INTERLEAVE-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFA_INTERLEAVE-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 2
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY1:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP8]], i64 1025)
; TFA_INTERLEAVE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFA_INTERLEAVE:       vector.body:
; TFA_INTERLEAVE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT5:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK2:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY1]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT6:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP10:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP11:%.*]] = mul i64 [[TMP10]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[TMP9]], i64 [[TMP11]]
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP9]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i64> poison)
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD3:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP12]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x i64> poison)
; TFA_INTERLEAVE-NEXT:    [[TMP13:%.*]] = icmp ugt <vscale x 2 x i64> [[WIDE_MASKED_LOAD]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 50, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP14:%.*]] = icmp ugt <vscale x 2 x i64> [[WIDE_MASKED_LOAD3]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 50, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP15:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i1> [[TMP13]], <vscale x 2 x i1> zeroinitializer
; TFA_INTERLEAVE-NEXT:    [[TMP16:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x i1> [[TMP14]], <vscale x 2 x i1> zeroinitializer
; TFA_INTERLEAVE-NEXT:    [[TMP17:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD]], <vscale x 2 x i1> [[TMP15]])
; TFA_INTERLEAVE-NEXT:    [[TMP18:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD3]], <vscale x 2 x i1> [[TMP16]])
; TFA_INTERLEAVE-NEXT:    [[TMP19:%.*]] = xor <vscale x 2 x i1> [[TMP13]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP20:%.*]] = xor <vscale x 2 x i1> [[TMP14]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP21:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i1> [[TMP19]], <vscale x 2 x i1> zeroinitializer
; TFA_INTERLEAVE-NEXT:    [[TMP22:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x i1> [[TMP20]], <vscale x 2 x i1> zeroinitializer
; TFA_INTERLEAVE-NEXT:    [[TMP23:%.*]] = or <vscale x 2 x i1> [[TMP15]], [[TMP21]]
; TFA_INTERLEAVE-NEXT:    [[TMP24:%.*]] = or <vscale x 2 x i1> [[TMP16]], [[TMP22]]
; TFA_INTERLEAVE-NEXT:    [[PREDPHI:%.*]] = select <vscale x 2 x i1> [[TMP21]], <vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> [[TMP17]]
; TFA_INTERLEAVE-NEXT:    [[PREDPHI4:%.*]] = select <vscale x 2 x i1> [[TMP22]], <vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> [[TMP18]]
; TFA_INTERLEAVE-NEXT:    [[TMP25:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP26:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP27:%.*]] = mul i64 [[TMP26]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP28:%.*]] = getelementptr inbounds i64, ptr [[TMP25]], i64 [[TMP27]]
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[PREDPHI]], ptr [[TMP25]], i32 8, <vscale x 2 x i1> [[TMP23]])
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[PREDPHI4]], ptr [[TMP28]], i32 8, <vscale x 2 x i1> [[TMP24]])
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT:%.*]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT5]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[TMP29:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP30:%.*]] = mul i64 [[TMP29]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP31:%.*]] = add i64 [[INDEX_NEXT]], [[TMP30]]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT6]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP31]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[TMP32:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP33:%.*]] = extractelement <vscale x 2 x i1> [[TMP32]], i32 0
; TFA_INTERLEAVE-NEXT:    br i1 [[TMP33]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
; TFA_INTERLEAVE:       for.cond.cleanup:
; TFA_INTERLEAVE-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i64 [ %indvars.iv.next, %if.end ], [ 0, %entry ]
  %arrayidx = getelementptr inbounds i64, ptr %a, i64 %indvars.iv
  %0 = load i64, ptr %arrayidx, align 8
  %cmp = icmp ugt i64 %0, 50
  br i1 %cmp, label %if.then, label %if.end

if.then:
  %1 = call i64 @foo(i64 %0) #1
  br label %if.end

if.end:
  %2 = phi i64 [%1, %if.then], [0, %for.body]
  %arrayidx1 = getelementptr inbounds i64, ptr %b, i64 %indvars.iv
  store i64 %2, ptr %arrayidx1, align 8
  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
  %exitcond = icmp eq i64 %indvars.iv.next, 1025
  br i1 %exitcond, label %for.cond.cleanup, label %for.body

for.cond.cleanup:
  ret void
}

; This checks the ability to handle masking of an if-then-else CFG with
; calls inside the conditional blocks. Although one of the calls has a
; uniform parameter and the metadata lists a uniform variant, right now
; we just see a splat of the parameter instead. More work needed.
define void @test_widen_if_then_else(ptr noalias %a, ptr readnone %b) #4 {
; TFNONE-LABEL: @test_widen_if_then_else(
; TFNONE-NEXT:  entry:
; TFNONE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFNONE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1025, [[TMP1]]
; TFNONE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; TFNONE:       vector.ph:
; TFNONE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFNONE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1025, [[TMP3]]
; TFNONE-NEXT:    [[N_VEC:%.*]] = sub i64 1025, [[N_MOD_VF]]
; TFNONE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; TFNONE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFNONE:       vector.body:
; TFNONE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFNONE-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 2 x i64>, ptr [[TMP6]], align 8
; TFNONE-NEXT:    [[TMP7:%.*]] = icmp ugt <vscale x 2 x i64> [[WIDE_LOAD]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 50, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; TFNONE-NEXT:    [[TMP8:%.*]] = xor <vscale x 2 x i1> [[TMP7]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFNONE-NEXT:    [[TMP9:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i1> [[TMP8]])
; TFNONE-NEXT:    [[TMP10:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_LOAD]], <vscale x 2 x i1> [[TMP7]])
; TFNONE-NEXT:    [[PREDPHI:%.*]] = select <vscale x 2 x i1> [[TMP8]], <vscale x 2 x i64> [[TMP9]], <vscale x 2 x i64> [[TMP10]]
; TFNONE-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    store <vscale x 2 x i64> [[PREDPHI]], ptr [[TMP11]], align 8
; TFNONE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; TFNONE-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; TFNONE-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; TFNONE:       middle.block:
; TFNONE-NEXT:    br i1 false, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
; TFNONE:       scalar.ph:
; TFNONE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; TFNONE-NEXT:    br label [[FOR_BODY:%.*]]
; TFNONE:       for.body:
; TFNONE-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[IF_END:%.*]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
; TFNONE-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    [[TMP13:%.*]] = load i64, ptr [[ARRAYIDX]], align 8
; TFNONE-NEXT:    [[CMP:%.*]] = icmp ugt i64 [[TMP13]], 50
; TFNONE-NEXT:    br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]]
; TFNONE:       if.then:
; TFNONE-NEXT:    [[TMP14:%.*]] = call i64 @foo(i64 [[TMP13]]) #[[ATTR4:[0-9]+]]
; TFNONE-NEXT:    br label [[IF_END]]
; TFNONE:       if.else:
; TFNONE-NEXT:    [[TMP15:%.*]] = call i64 @foo(i64 0) #[[ATTR4]]
; TFNONE-NEXT:    br label [[IF_END]]
; TFNONE:       if.end:
; TFNONE-NEXT:    [[TMP16:%.*]] = phi i64 [ [[TMP14]], [[IF_THEN]] ], [ [[TMP15]], [[IF_ELSE]] ]
; TFNONE-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    store i64 [[TMP16]], ptr [[ARRAYIDX1]], align 8
; TFNONE-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; TFNONE-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 1025
; TFNONE-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
; TFNONE:       for.cond.cleanup:
; TFNONE-NEXT:    ret void
;
; TFCOMMON-LABEL: @test_widen_if_then_else(
; TFCOMMON-NEXT:  entry:
; TFCOMMON-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFCOMMON-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFCOMMON-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFCOMMON-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFCOMMON-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFCOMMON-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFCOMMON-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFCOMMON-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFCOMMON-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFCOMMON-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 2
; TFCOMMON-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFCOMMON-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFCOMMON:       vector.body:
; TFCOMMON-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFCOMMON-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFCOMMON-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFCOMMON-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP7]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i64> poison)
; TFCOMMON-NEXT:    [[TMP8:%.*]] = icmp ugt <vscale x 2 x i64> [[WIDE_MASKED_LOAD]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 50, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; TFCOMMON-NEXT:    [[TMP9:%.*]] = xor <vscale x 2 x i1> [[TMP8]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFCOMMON-NEXT:    [[TMP10:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i1> [[TMP9]], <vscale x 2 x i1> zeroinitializer
; TFCOMMON-NEXT:    [[TMP11:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i1> [[TMP10]])
; TFCOMMON-NEXT:    [[TMP12:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i1> [[TMP8]], <vscale x 2 x i1> zeroinitializer
; TFCOMMON-NEXT:    [[TMP13:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD]], <vscale x 2 x i1> [[TMP12]])
; TFCOMMON-NEXT:    [[TMP14:%.*]] = or <vscale x 2 x i1> [[TMP10]], [[TMP12]]
; TFCOMMON-NEXT:    [[PREDPHI:%.*]] = select <vscale x 2 x i1> [[TMP10]], <vscale x 2 x i64> [[TMP11]], <vscale x 2 x i64> [[TMP13]]
; TFCOMMON-NEXT:    [[TMP15:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFCOMMON-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[PREDPHI]], ptr [[TMP15]], i32 8, <vscale x 2 x i1> [[TMP14]])
; TFCOMMON-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP6]]
; TFCOMMON-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFCOMMON-NEXT:    [[TMP16:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFCOMMON-NEXT:    [[TMP17:%.*]] = extractelement <vscale x 2 x i1> [[TMP16]], i32 0
; TFCOMMON-NEXT:    br i1 [[TMP17]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; TFCOMMON:       for.cond.cleanup:
; TFCOMMON-NEXT:    ret void
;
; TFA_INTERLEAVE-LABEL: @test_widen_if_then_else(
; TFA_INTERLEAVE-NEXT:  entry:
; TFA_INTERLEAVE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFA_INTERLEAVE-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFA_INTERLEAVE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFA_INTERLEAVE-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFA_INTERLEAVE-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 2
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY1:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP8]], i64 1025)
; TFA_INTERLEAVE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFA_INTERLEAVE:       vector.body:
; TFA_INTERLEAVE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT5:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK2:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY1]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT6:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP10:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP11:%.*]] = mul i64 [[TMP10]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[TMP9]], i64 [[TMP11]]
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP9]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i64> poison)
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD3:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP12]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x i64> poison)
; TFA_INTERLEAVE-NEXT:    [[TMP13:%.*]] = icmp ugt <vscale x 2 x i64> [[WIDE_MASKED_LOAD]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 50, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP14:%.*]] = icmp ugt <vscale x 2 x i64> [[WIDE_MASKED_LOAD3]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 50, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP15:%.*]] = xor <vscale x 2 x i1> [[TMP13]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP16:%.*]] = xor <vscale x 2 x i1> [[TMP14]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP17:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i1> [[TMP15]], <vscale x 2 x i1> zeroinitializer
; TFA_INTERLEAVE-NEXT:    [[TMP18:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x i1> [[TMP16]], <vscale x 2 x i1> zeroinitializer
; TFA_INTERLEAVE-NEXT:    [[TMP19:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i1> [[TMP17]])
; TFA_INTERLEAVE-NEXT:    [[TMP20:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i1> [[TMP18]])
; TFA_INTERLEAVE-NEXT:    [[TMP21:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i1> [[TMP13]], <vscale x 2 x i1> zeroinitializer
; TFA_INTERLEAVE-NEXT:    [[TMP22:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x i1> [[TMP14]], <vscale x 2 x i1> zeroinitializer
; TFA_INTERLEAVE-NEXT:    [[TMP23:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD]], <vscale x 2 x i1> [[TMP21]])
; TFA_INTERLEAVE-NEXT:    [[TMP24:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD3]], <vscale x 2 x i1> [[TMP22]])
; TFA_INTERLEAVE-NEXT:    [[TMP25:%.*]] = or <vscale x 2 x i1> [[TMP17]], [[TMP21]]
; TFA_INTERLEAVE-NEXT:    [[TMP26:%.*]] = or <vscale x 2 x i1> [[TMP18]], [[TMP22]]
; TFA_INTERLEAVE-NEXT:    [[PREDPHI:%.*]] = select <vscale x 2 x i1> [[TMP17]], <vscale x 2 x i64> [[TMP19]], <vscale x 2 x i64> [[TMP23]]
; TFA_INTERLEAVE-NEXT:    [[PREDPHI4:%.*]] = select <vscale x 2 x i1> [[TMP18]], <vscale x 2 x i64> [[TMP20]], <vscale x 2 x i64> [[TMP24]]
; TFA_INTERLEAVE-NEXT:    [[TMP27:%.*]] = getelementptr inbounds i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP28:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP29:%.*]] = mul i64 [[TMP28]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP30:%.*]] = getelementptr inbounds i64, ptr [[TMP27]], i64 [[TMP29]]
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[PREDPHI]], ptr [[TMP27]], i32 8, <vscale x 2 x i1> [[TMP25]])
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[PREDPHI4]], ptr [[TMP30]], i32 8, <vscale x 2 x i1> [[TMP26]])
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT:%.*]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT5]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[TMP31:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP32:%.*]] = mul i64 [[TMP31]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP33:%.*]] = add i64 [[INDEX_NEXT]], [[TMP32]]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT6]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP33]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[TMP34:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP35:%.*]] = extractelement <vscale x 2 x i1> [[TMP34]], i32 0
; TFA_INTERLEAVE-NEXT:    br i1 [[TMP35]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; TFA_INTERLEAVE:       for.cond.cleanup:
; TFA_INTERLEAVE-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i64 [ %indvars.iv.next, %if.end ], [ 0, %entry ]
  %arrayidx = getelementptr inbounds i64, ptr %a, i64 %indvars.iv
  %0 = load i64, ptr %arrayidx, align 8
  %cmp = icmp ugt i64 %0, 50
  br i1 %cmp, label %if.then, label %if.else

if.then:
  %1 = call i64 @foo(i64 %0) #0
  br label %if.end

if.else:
  %2 = call i64 @foo(i64 0) #0
  br label %if.end

if.end:
  %3 = phi i64 [%1, %if.then], [%2, %if.else]
  %arrayidx1 = getelementptr inbounds i64, ptr %b, i64 %indvars.iv
  store i64 %3, ptr %arrayidx1, align 8
  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
  %exitcond = icmp eq i64 %indvars.iv.next, 1025
  br i1 %exitcond, label %for.cond.cleanup, label %for.body

for.cond.cleanup:
  ret void
}

; A call whose argument must be widened, where the vector variant does not have
; a mask. Forcing tail folding results in no vectorized call, whereas an
; unpredicated body with scalar tail can use the unmasked variant.
define void @test_widen_nomask(ptr noalias %a, ptr readnone %b) #4 {
; TFNONE-LABEL: @test_widen_nomask(
; TFNONE-NEXT:  entry:
; TFNONE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFNONE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1025, [[TMP1]]
; TFNONE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; TFNONE:       vector.ph:
; TFNONE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFNONE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1025, [[TMP3]]
; TFNONE-NEXT:    [[N_VEC:%.*]] = sub i64 1025, [[N_MOD_VF]]
; TFNONE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; TFNONE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFNONE:       vector.body:
; TFNONE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFNONE-NEXT:    [[TMP6:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 2 x i64>, ptr [[TMP6]], align 8
; TFNONE-NEXT:    [[TMP7:%.*]] = call <vscale x 2 x i64> @foo_vector_nomask(<vscale x 2 x i64> [[WIDE_LOAD]])
; TFNONE-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    store <vscale x 2 x i64> [[TMP7]], ptr [[TMP8]], align 8
; TFNONE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; TFNONE-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; TFNONE-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; TFNONE:       middle.block:
; TFNONE-NEXT:    br i1 false, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
; TFNONE:       scalar.ph:
; TFNONE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; TFNONE-NEXT:    br label [[FOR_BODY:%.*]]
; TFNONE:       for.body:
; TFNONE-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; TFNONE-NEXT:    [[GEP:%.*]] = getelementptr i64, ptr [[B]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    [[LOAD:%.*]] = load i64, ptr [[GEP]], align 8
; TFNONE-NEXT:    [[CALL:%.*]] = call i64 @foo(i64 [[LOAD]]) #[[ATTR5:[0-9]+]]
; TFNONE-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    store i64 [[CALL]], ptr [[ARRAYIDX]], align 8
; TFNONE-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; TFNONE-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 1025
; TFNONE-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
; TFNONE:       for.cond.cleanup:
; TFNONE-NEXT:    ret void
;
; TFALWAYS-LABEL: @test_widen_nomask(
; TFALWAYS-NEXT:  entry:
; TFALWAYS-NEXT:    br label [[FOR_BODY:%.*]]
; TFALWAYS:       for.body:
; TFALWAYS-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; TFALWAYS-NEXT:    [[GEP:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDVARS_IV]]
; TFALWAYS-NEXT:    [[LOAD:%.*]] = load i64, ptr [[GEP]], align 8
; TFALWAYS-NEXT:    [[CALL:%.*]] = call i64 @foo(i64 [[LOAD]]) #[[ATTR5:[0-9]+]]
; TFALWAYS-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDVARS_IV]]
; TFALWAYS-NEXT:    store i64 [[CALL]], ptr [[ARRAYIDX]], align 8
; TFALWAYS-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; TFALWAYS-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 1025
; TFALWAYS-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]]
; TFALWAYS:       for.cond.cleanup:
; TFALWAYS-NEXT:    ret void
;
; TFFALLBACK-LABEL: @test_widen_nomask(
; TFFALLBACK-NEXT:  entry:
; TFFALLBACK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFFALLBACK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFFALLBACK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1025, [[TMP1]]
; TFFALLBACK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; TFFALLBACK:       vector.ph:
; TFFALLBACK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFFALLBACK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFFALLBACK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1025, [[TMP3]]
; TFFALLBACK-NEXT:    [[N_VEC:%.*]] = sub i64 1025, [[N_MOD_VF]]
; TFFALLBACK-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; TFFALLBACK-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; TFFALLBACK-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFFALLBACK:       vector.body:
; TFFALLBACK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFFALLBACK-NEXT:    [[TMP6:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFFALLBACK-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 2 x i64>, ptr [[TMP6]], align 8
; TFFALLBACK-NEXT:    [[TMP7:%.*]] = call <vscale x 2 x i64> @foo_vector_nomask(<vscale x 2 x i64> [[WIDE_LOAD]])
; TFFALLBACK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFFALLBACK-NEXT:    store <vscale x 2 x i64> [[TMP7]], ptr [[TMP8]], align 8
; TFFALLBACK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; TFFALLBACK-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; TFFALLBACK-NEXT:    br i1 [[TMP9]], label [[SCALAR_PH]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; TFFALLBACK:       scalar.ph:
; TFFALLBACK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[N_VEC]], [[VECTOR_BODY]] ]
; TFFALLBACK-NEXT:    br label [[FOR_BODY:%.*]]
; TFFALLBACK:       for.body:
; TFFALLBACK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; TFFALLBACK-NEXT:    [[GEP:%.*]] = getelementptr i64, ptr [[B]], i64 [[INDVARS_IV]]
; TFFALLBACK-NEXT:    [[LOAD:%.*]] = load i64, ptr [[GEP]], align 8
; TFFALLBACK-NEXT:    [[CALL:%.*]] = call i64 @foo(i64 [[LOAD]]) #[[ATTR5:[0-9]+]]
; TFFALLBACK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDVARS_IV]]
; TFFALLBACK-NEXT:    store i64 [[CALL]], ptr [[ARRAYIDX]], align 8
; TFFALLBACK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; TFFALLBACK-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 1025
; TFFALLBACK-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; TFFALLBACK:       for.cond.cleanup:
; TFFALLBACK-NEXT:    ret void
;
; TFA_INTERLEAVE-LABEL: @test_widen_nomask(
; TFA_INTERLEAVE-NEXT:  entry:
; TFA_INTERLEAVE-NEXT:    br label [[FOR_BODY:%.*]]
; TFA_INTERLEAVE:       for.body:
; TFA_INTERLEAVE-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[GEP:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDVARS_IV]]
; TFA_INTERLEAVE-NEXT:    [[LOAD:%.*]] = load i64, ptr [[GEP]], align 8
; TFA_INTERLEAVE-NEXT:    [[CALL:%.*]] = call i64 @foo(i64 [[LOAD]]) #[[ATTR5:[0-9]+]]
; TFA_INTERLEAVE-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDVARS_IV]]
; TFA_INTERLEAVE-NEXT:    store i64 [[CALL]], ptr [[ARRAYIDX]], align 8
; TFA_INTERLEAVE-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; TFA_INTERLEAVE-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 1025
; TFA_INTERLEAVE-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]]
; TFA_INTERLEAVE:       for.cond.cleanup:
; TFA_INTERLEAVE-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
  %gep = getelementptr i64, ptr %b, i64 %indvars.iv
  %load = load i64, ptr %gep
  %call = call i64 @foo(i64 %load) #2
  %arrayidx = getelementptr inbounds i64, ptr %a, i64 %indvars.iv
  store i64 %call, ptr %arrayidx
  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
  %exitcond = icmp eq i64 %indvars.iv.next, 1025
  br i1 %exitcond, label %for.cond.cleanup, label %for.body

for.cond.cleanup:
  ret void
}

; If both masked and unmasked options are present, we expect to see tail folding
; use the masked version and unpredicated body with scalar tail use the unmasked
; version.
define void @test_widen_optmask(ptr noalias %a, ptr readnone %b) #4 {
; TFNONE-LABEL: @test_widen_optmask(
; TFNONE-NEXT:  entry:
; TFNONE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFNONE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1025, [[TMP1]]
; TFNONE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; TFNONE:       vector.ph:
; TFNONE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFNONE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1025, [[TMP3]]
; TFNONE-NEXT:    [[N_VEC:%.*]] = sub i64 1025, [[N_MOD_VF]]
; TFNONE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; TFNONE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFNONE:       vector.body:
; TFNONE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFNONE-NEXT:    [[TMP6:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 2 x i64>, ptr [[TMP6]], align 8
; TFNONE-NEXT:    [[TMP7:%.*]] = call <vscale x 2 x i64> @foo_vector_nomask(<vscale x 2 x i64> [[WIDE_LOAD]])
; TFNONE-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    store <vscale x 2 x i64> [[TMP7]], ptr [[TMP8]], align 8
; TFNONE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; TFNONE-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; TFNONE-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
; TFNONE:       middle.block:
; TFNONE-NEXT:    br i1 false, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
; TFNONE:       scalar.ph:
; TFNONE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; TFNONE-NEXT:    br label [[FOR_BODY:%.*]]
; TFNONE:       for.body:
; TFNONE-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; TFNONE-NEXT:    [[GEP:%.*]] = getelementptr i64, ptr [[B]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    [[LOAD:%.*]] = load i64, ptr [[GEP]], align 8
; TFNONE-NEXT:    [[CALL:%.*]] = call i64 @foo(i64 [[LOAD]]) #[[ATTR6:[0-9]+]]
; TFNONE-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    store i64 [[CALL]], ptr [[ARRAYIDX]], align 8
; TFNONE-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; TFNONE-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 1025
; TFNONE-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
; TFNONE:       for.cond.cleanup:
; TFNONE-NEXT:    ret void
;
; TFALWAYS-LABEL: @test_widen_optmask(
; TFALWAYS-NEXT:  entry:
; TFALWAYS-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFALWAYS-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFALWAYS-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFALWAYS-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFALWAYS-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFALWAYS-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFALWAYS-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFALWAYS-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFALWAYS-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFALWAYS-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 2
; TFALWAYS-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFALWAYS-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFALWAYS:       vector.body:
; TFALWAYS-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFALWAYS-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFALWAYS-NEXT:    [[TMP7:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFALWAYS-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP7]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i64> poison)
; TFALWAYS-NEXT:    [[TMP8:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFALWAYS-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFALWAYS-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP8]], ptr [[TMP9]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFALWAYS-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP6]]
; TFALWAYS-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFALWAYS-NEXT:    [[TMP10:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFALWAYS-NEXT:    [[TMP11:%.*]] = extractelement <vscale x 2 x i1> [[TMP10]], i32 0
; TFALWAYS-NEXT:    br i1 [[TMP11]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; TFALWAYS:       for.cond.cleanup:
; TFALWAYS-NEXT:    ret void
;
; TFFALLBACK-LABEL: @test_widen_optmask(
; TFFALLBACK-NEXT:  entry:
; TFFALLBACK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFFALLBACK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFFALLBACK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFFALLBACK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFFALLBACK-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFFALLBACK-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFFALLBACK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFFALLBACK-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFFALLBACK-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFFALLBACK-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 2
; TFFALLBACK-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFFALLBACK-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFFALLBACK:       vector.body:
; TFFALLBACK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFFALLBACK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFFALLBACK-NEXT:    [[TMP7:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFFALLBACK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP7]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i64> poison)
; TFFALLBACK-NEXT:    [[TMP8:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFFALLBACK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFFALLBACK-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP8]], ptr [[TMP9]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFFALLBACK-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP6]]
; TFFALLBACK-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFFALLBACK-NEXT:    [[TMP10:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFFALLBACK-NEXT:    [[TMP11:%.*]] = extractelement <vscale x 2 x i1> [[TMP10]], i32 0
; TFFALLBACK-NEXT:    br i1 [[TMP11]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
; TFFALLBACK:       for.cond.cleanup:
; TFFALLBACK-NEXT:    ret void
;
; TFA_INTERLEAVE-LABEL: @test_widen_optmask(
; TFA_INTERLEAVE-NEXT:  entry:
; TFA_INTERLEAVE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFA_INTERLEAVE-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFA_INTERLEAVE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFA_INTERLEAVE-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFA_INTERLEAVE-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 2
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY1:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP8]], i64 1025)
; TFA_INTERLEAVE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFA_INTERLEAVE:       vector.body:
; TFA_INTERLEAVE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT4:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK2:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY1]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT5:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[B:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP10:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP11:%.*]] = mul i64 [[TMP10]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP12:%.*]] = getelementptr i64, ptr [[TMP9]], i64 [[TMP11]]
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP9]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x i64> poison)
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD3:%.*]] = call <vscale x 2 x i64> @llvm.masked.load.nxv2i64.p0(ptr [[TMP12]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x i64> poison)
; TFA_INTERLEAVE-NEXT:    [[TMP13:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFA_INTERLEAVE-NEXT:    [[TMP14:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[WIDE_MASKED_LOAD3]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]])
; TFA_INTERLEAVE-NEXT:    [[TMP15:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP16:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP17:%.*]] = mul i64 [[TMP16]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i64, ptr [[TMP15]], i64 [[TMP17]]
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP13]], ptr [[TMP15]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP14]], ptr [[TMP18]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]])
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT:%.*]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT4]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[TMP19:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP20:%.*]] = mul i64 [[TMP19]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP21:%.*]] = add i64 [[INDEX_NEXT]], [[TMP20]]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT5]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP21]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[TMP22:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP23:%.*]] = extractelement <vscale x 2 x i1> [[TMP22]], i32 0
; TFA_INTERLEAVE-NEXT:    br i1 [[TMP23]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; TFA_INTERLEAVE:       for.cond.cleanup:
; TFA_INTERLEAVE-NEXT:    ret void
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
  %gep = getelementptr i64, ptr %b, i64 %indvars.iv
  %load = load i64, ptr %gep
  %call = call i64 @foo(i64 %load) #3
  %arrayidx = getelementptr inbounds i64, ptr %a, i64 %indvars.iv
  store i64 %call, ptr %arrayidx
  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
  %exitcond = icmp eq i64 %indvars.iv.next, 1025
  br i1 %exitcond, label %for.cond.cleanup, label %for.body

for.cond.cleanup:
  ret void
}


; An fmuladd intrinsic followed by a call; we want to make sure we correctly
; pick up the second call and assign a vector variant to it.
define double @test_widen_fmuladd_and_call(ptr noalias %a, ptr readnone %b, double %m) #4 {
; TFNONE-LABEL: @test_widen_fmuladd_and_call(
; TFNONE-NEXT:  entry:
; TFNONE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFNONE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1025, [[TMP1]]
; TFNONE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; TFNONE:       vector.ph:
; TFNONE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFNONE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1025, [[TMP3]]
; TFNONE-NEXT:    [[N_VEC:%.*]] = sub i64 1025, [[N_MOD_VF]]
; TFNONE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; TFNONE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; TFNONE-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x double> poison, double [[M:%.*]], i64 0
; TFNONE-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x double> [[BROADCAST_SPLATINSERT]], <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
; TFNONE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFNONE:       vector.body:
; TFNONE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFNONE-NEXT:    [[VEC_PHI:%.*]] = phi double [ 0.000000e+00, [[VECTOR_PH]] ], [ [[TMP11:%.*]], [[VECTOR_BODY]] ]
; TFNONE-NEXT:    [[TMP6:%.*]] = getelementptr double, ptr [[B:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 2 x double>, ptr [[TMP6]], align 8
; TFNONE-NEXT:    [[TMP7:%.*]] = fmul <vscale x 2 x double> [[WIDE_LOAD]], [[BROADCAST_SPLAT]]
; TFNONE-NEXT:    [[TMP8:%.*]] = fptoui <vscale x 2 x double> [[WIDE_LOAD]] to <vscale x 2 x i64>
; TFNONE-NEXT:    [[TMP9:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[TMP8]], <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; TFNONE-NEXT:    [[TMP10:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFNONE-NEXT:    store <vscale x 2 x i64> [[TMP9]], ptr [[TMP10]], align 8
; TFNONE-NEXT:    [[TMP11]] = call double @llvm.vector.reduce.fadd.nxv2f64(double [[VEC_PHI]], <vscale x 2 x double> [[TMP7]])
; TFNONE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; TFNONE-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; TFNONE-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
; TFNONE:       middle.block:
; TFNONE-NEXT:    br i1 false, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
; TFNONE:       scalar.ph:
; TFNONE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; TFNONE-NEXT:    [[BC_MERGE_RDX:%.*]] = phi double [ 0.000000e+00, [[ENTRY]] ], [ [[TMP11]], [[MIDDLE_BLOCK]] ]
; TFNONE-NEXT:    br label [[FOR_BODY:%.*]]
; TFNONE:       for.body:
; TFNONE-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; TFNONE-NEXT:    [[FMA_SUM:%.*]] = phi double [ [[BC_MERGE_RDX]], [[SCALAR_PH]] ], [ [[MULADD:%.*]], [[FOR_BODY]] ]
; TFNONE-NEXT:    [[GEP:%.*]] = getelementptr double, ptr [[B]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    [[LOAD:%.*]] = load double, ptr [[GEP]], align 8
; TFNONE-NEXT:    [[MULADD]] = tail call double @llvm.fmuladd.f64(double [[LOAD]], double [[M]], double [[FMA_SUM]])
; TFNONE-NEXT:    [[TOINT:%.*]] = fptoui double [[LOAD]] to i64
; TFNONE-NEXT:    [[CALL:%.*]] = call i64 @foo(i64 [[TOINT]]) #[[ATTR3]]
; TFNONE-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDVARS_IV]]
; TFNONE-NEXT:    store i64 [[CALL]], ptr [[ARRAYIDX]], align 8
; TFNONE-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; TFNONE-NEXT:    [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 1025
; TFNONE-NEXT:    br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP13:![0-9]+]]
; TFNONE:       for.cond.cleanup:
; TFNONE-NEXT:    [[MULADD_LCSSA:%.*]] = phi double [ [[MULADD]], [[FOR_BODY]] ], [ [[TMP11]], [[MIDDLE_BLOCK]] ]
; TFNONE-NEXT:    ret double [[MULADD_LCSSA]]
;
; TFALWAYS-LABEL: @test_widen_fmuladd_and_call(
; TFALWAYS-NEXT:  entry:
; TFALWAYS-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFALWAYS-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFALWAYS-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFALWAYS-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFALWAYS-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFALWAYS-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFALWAYS-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFALWAYS-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFALWAYS-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFALWAYS-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 2
; TFALWAYS-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFALWAYS-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x double> poison, double [[M:%.*]], i64 0
; TFALWAYS-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x double> [[BROADCAST_SPLATINSERT]], <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
; TFALWAYS-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFALWAYS:       vector.body:
; TFALWAYS-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFALWAYS-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFALWAYS-NEXT:    [[VEC_PHI:%.*]] = phi double [ 0.000000e+00, [[ENTRY]] ], [ [[TMP13:%.*]], [[VECTOR_BODY]] ]
; TFALWAYS-NEXT:    [[TMP7:%.*]] = getelementptr double, ptr [[B:%.*]], i64 [[INDEX]]
; TFALWAYS-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x double> @llvm.masked.load.nxv2f64.p0(ptr [[TMP7]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x double> poison)
; TFALWAYS-NEXT:    [[TMP8:%.*]] = fmul <vscale x 2 x double> [[WIDE_MASKED_LOAD]], [[BROADCAST_SPLAT]]
; TFALWAYS-NEXT:    [[TMP9:%.*]] = fptoui <vscale x 2 x double> [[WIDE_MASKED_LOAD]] to <vscale x 2 x i64>
; TFALWAYS-NEXT:    [[TMP10:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[TMP9]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFALWAYS-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFALWAYS-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP10]], ptr [[TMP11]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFALWAYS-NEXT:    [[TMP12:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x double> [[TMP8]], <vscale x 2 x double> shufflevector (<vscale x 2 x double> insertelement (<vscale x 2 x double> poison, double -0.000000e+00, i64 0), <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer)
; TFALWAYS-NEXT:    [[TMP13]] = call double @llvm.vector.reduce.fadd.nxv2f64(double [[VEC_PHI]], <vscale x 2 x double> [[TMP12]])
; TFALWAYS-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP6]]
; TFALWAYS-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFALWAYS-NEXT:    [[TMP14:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFALWAYS-NEXT:    [[TMP15:%.*]] = extractelement <vscale x 2 x i1> [[TMP14]], i32 0
; TFALWAYS-NEXT:    br i1 [[TMP15]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; TFALWAYS:       for.cond.cleanup:
; TFALWAYS-NEXT:    ret double [[TMP13]]
;
; TFFALLBACK-LABEL: @test_widen_fmuladd_and_call(
; TFFALLBACK-NEXT:  entry:
; TFFALLBACK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFFALLBACK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; TFFALLBACK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFFALLBACK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; TFFALLBACK-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFFALLBACK-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFFALLBACK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFFALLBACK-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFFALLBACK-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFFALLBACK-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 2
; TFFALLBACK-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFFALLBACK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x double> poison, double [[M:%.*]], i64 0
; TFFALLBACK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x double> [[BROADCAST_SPLATINSERT]], <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
; TFFALLBACK-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFFALLBACK:       vector.body:
; TFFALLBACK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFFALLBACK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFFALLBACK-NEXT:    [[VEC_PHI:%.*]] = phi double [ 0.000000e+00, [[ENTRY]] ], [ [[TMP13:%.*]], [[VECTOR_BODY]] ]
; TFFALLBACK-NEXT:    [[TMP7:%.*]] = getelementptr double, ptr [[B:%.*]], i64 [[INDEX]]
; TFFALLBACK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x double> @llvm.masked.load.nxv2f64.p0(ptr [[TMP7]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x double> poison)
; TFFALLBACK-NEXT:    [[TMP8:%.*]] = fmul <vscale x 2 x double> [[WIDE_MASKED_LOAD]], [[BROADCAST_SPLAT]]
; TFFALLBACK-NEXT:    [[TMP9:%.*]] = fptoui <vscale x 2 x double> [[WIDE_MASKED_LOAD]] to <vscale x 2 x i64>
; TFFALLBACK-NEXT:    [[TMP10:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[TMP9]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFFALLBACK-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFFALLBACK-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP10]], ptr [[TMP11]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFFALLBACK-NEXT:    [[TMP12:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x double> [[TMP8]], <vscale x 2 x double> shufflevector (<vscale x 2 x double> insertelement (<vscale x 2 x double> poison, double -0.000000e+00, i64 0), <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer)
; TFFALLBACK-NEXT:    [[TMP13]] = call double @llvm.vector.reduce.fadd.nxv2f64(double [[VEC_PHI]], <vscale x 2 x double> [[TMP12]])
; TFFALLBACK-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP6]]
; TFFALLBACK-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFFALLBACK-NEXT:    [[TMP14:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFFALLBACK-NEXT:    [[TMP15:%.*]] = extractelement <vscale x 2 x i1> [[TMP14]], i32 0
; TFFALLBACK-NEXT:    br i1 [[TMP15]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; TFFALLBACK:       for.cond.cleanup:
; TFFALLBACK-NEXT:    ret double [[TMP13]]
;
; TFA_INTERLEAVE-LABEL: @test_widen_fmuladd_and_call(
; TFA_INTERLEAVE-NEXT:  entry:
; TFA_INTERLEAVE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
; TFA_INTERLEAVE-NEXT:    [[N_RND_UP:%.*]] = add i64 1025, [[TMP4]]
; TFA_INTERLEAVE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
; TFA_INTERLEAVE-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; TFA_INTERLEAVE-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 4
; TFA_INTERLEAVE-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 2
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 0, i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_ENTRY1:%.*]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP8]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x double> poison, double [[M:%.*]], i64 0
; TFA_INTERLEAVE-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x double> [[BROADCAST_SPLATINSERT]], <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
; TFA_INTERLEAVE-NEXT:    br label [[VECTOR_BODY:%.*]]
; TFA_INTERLEAVE:       vector.body:
; TFA_INTERLEAVE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDEX_NEXT4:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK2:%.*]] = phi <vscale x 2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY1]], [[ENTRY]] ], [ [[ACTIVE_LANE_MASK_NEXT5:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[VEC_PHI:%.*]] = phi double [ 0.000000e+00, [[ENTRY]] ], [ [[TMP26:%.*]], [[VECTOR_BODY]] ]
; TFA_INTERLEAVE-NEXT:    [[TMP9:%.*]] = getelementptr double, ptr [[B:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP10:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP11:%.*]] = mul i64 [[TMP10]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP12:%.*]] = getelementptr double, ptr [[TMP9]], i64 [[TMP11]]
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 2 x double> @llvm.masked.load.nxv2f64.p0(ptr [[TMP9]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x double> poison)
; TFA_INTERLEAVE-NEXT:    [[WIDE_MASKED_LOAD3:%.*]] = call <vscale x 2 x double> @llvm.masked.load.nxv2f64.p0(ptr [[TMP12]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x double> poison)
; TFA_INTERLEAVE-NEXT:    [[TMP13:%.*]] = fmul <vscale x 2 x double> [[WIDE_MASKED_LOAD]], [[BROADCAST_SPLAT]]
; TFA_INTERLEAVE-NEXT:    [[TMP14:%.*]] = fmul <vscale x 2 x double> [[WIDE_MASKED_LOAD3]], [[BROADCAST_SPLAT]]
; TFA_INTERLEAVE-NEXT:    [[TMP15:%.*]] = fptoui <vscale x 2 x double> [[WIDE_MASKED_LOAD]] to <vscale x 2 x i64>
; TFA_INTERLEAVE-NEXT:    [[TMP16:%.*]] = fptoui <vscale x 2 x double> [[WIDE_MASKED_LOAD3]] to <vscale x 2 x i64>
; TFA_INTERLEAVE-NEXT:    [[TMP17:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[TMP15]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFA_INTERLEAVE-NEXT:    [[TMP18:%.*]] = call <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64> [[TMP16]], <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]])
; TFA_INTERLEAVE-NEXT:    [[TMP19:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]]
; TFA_INTERLEAVE-NEXT:    [[TMP20:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP21:%.*]] = mul i64 [[TMP20]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP22:%.*]] = getelementptr inbounds i64, ptr [[TMP19]], i64 [[TMP21]]
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP17]], ptr [[TMP19]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK]])
; TFA_INTERLEAVE-NEXT:    call void @llvm.masked.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP18]], ptr [[TMP22]], i32 8, <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]])
; TFA_INTERLEAVE-NEXT:    [[TMP23:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK]], <vscale x 2 x double> [[TMP13]], <vscale x 2 x double> shufflevector (<vscale x 2 x double> insertelement (<vscale x 2 x double> poison, double -0.000000e+00, i64 0), <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP24:%.*]] = call double @llvm.vector.reduce.fadd.nxv2f64(double [[VEC_PHI]], <vscale x 2 x double> [[TMP23]])
; TFA_INTERLEAVE-NEXT:    [[TMP25:%.*]] = select <vscale x 2 x i1> [[ACTIVE_LANE_MASK2]], <vscale x 2 x double> [[TMP14]], <vscale x 2 x double> shufflevector (<vscale x 2 x double> insertelement (<vscale x 2 x double> poison, double -0.000000e+00, i64 0), <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP26]] = call double @llvm.vector.reduce.fadd.nxv2f64(double [[TMP24]], <vscale x 2 x double> [[TMP25]])
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT:%.*]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[INDEX_NEXT4]] = add i64 [[INDEX]], [[TMP6]]
; TFA_INTERLEAVE-NEXT:    [[TMP27:%.*]] = call i64 @llvm.vscale.i64()
; TFA_INTERLEAVE-NEXT:    [[TMP28:%.*]] = mul i64 [[TMP27]], 2
; TFA_INTERLEAVE-NEXT:    [[TMP29:%.*]] = add i64 [[INDEX_NEXT]], [[TMP28]]
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[INDEX_NEXT]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[ACTIVE_LANE_MASK_NEXT5]] = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64 [[TMP29]], i64 1025)
; TFA_INTERLEAVE-NEXT:    [[TMP30:%.*]] = xor <vscale x 2 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer)
; TFA_INTERLEAVE-NEXT:    [[TMP31:%.*]] = extractelement <vscale x 2 x i1> [[TMP30]], i32 0
; TFA_INTERLEAVE-NEXT:    br i1 [[TMP31]], label [[FOR_COND_CLEANUP:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; TFA_INTERLEAVE:       for.cond.cleanup:
; TFA_INTERLEAVE-NEXT:    ret double [[TMP26]]
;
entry:
  br label %for.body

for.body:
  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
  %fma_sum = phi double [ 0.000000e+00, %entry ], [ %muladd, %for.body ]
  %gep = getelementptr double, ptr %b, i64 %indvars.iv
  %load = load double, ptr %gep
  %muladd = tail call double @llvm.fmuladd.f64(double %load, double %m, double %fma_sum)
  %toint = fptoui double %load to i64
  %call = call i64 @foo(i64 %toint) #1
  %arrayidx = getelementptr inbounds i64, ptr %a, i64 %indvars.iv
  store i64 %call, ptr %arrayidx
  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
  %exitcond = icmp eq i64 %indvars.iv.next, 1025
  br i1 %exitcond, label %for.cond.cleanup, label %for.body

for.cond.cleanup:
  ret double %muladd
}

declare i64 @foo(i64)
declare double @llvm.fmuladd.f64(double, double, double)

;; scalable vector variants of foo
declare <vscale x 2 x i64> @foo_uniform(i64, <vscale x 2 x i1>)
declare <vscale x 2 x i64> @foo_vector(<vscale x 2 x i64>, <vscale x 2 x i1>)
declare <vscale x 2 x i64> @foo_vector_nomask(<vscale x 2 x i64>)

attributes #0 = { nounwind "vector-function-abi-variant"="_ZGVsMxv_foo(foo_vector),_ZGVsMxu_foo(foo_uniform)" }
attributes #1 = { nounwind "vector-function-abi-variant"="_ZGVsMxv_foo(foo_vector)" }
attributes #2 = { nounwind "vector-function-abi-variant"="_ZGVsNxv_foo(foo_vector_nomask)" }
attributes #3 = { nounwind "vector-function-abi-variant"="_ZGVsNxv_foo(foo_vector_nomask),_ZGVsMxv_foo(foo_vector)" }
attributes #4 = { "target-features"="+sve" vscale_range(2,16) "no-trapping-math"="false" }