TheRock/compiler/amd-llvm/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv64 -mattr=+m,+f,+d,+a,+c,+v -verify-machineinstrs -O2 < %s | FileCheck %s --check-prefixes=CHECK,NODEPVL
; RUN: llc -mtriple=riscv64 -mattr=+m,+f,+d,+a,+c,+v,+vl-dependent-latency -verify-machineinstrs -O2 < %s | FileCheck %s --check-prefixes=CHECK,DEPVL

define <vscale x 1 x double> @test1(i64 %avl, <vscale x 1 x double> %a, <vscale x 1 x double> %b) nounwind {
; CHECK-LABEL: test1:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v8, v8, v9
; CHECK-NEXT:    ret
entry:
  %0 = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 2, i64 7)
  %1 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %a,
    <vscale x 1 x double> %b,
    i64 7, i64 %0)
  ret <vscale x 1 x double> %1
}

define <vscale x 1 x double> @test2(i64 %avl, <vscale x 1 x double> %a, <vscale x 1 x double> %b) nounwind {
; CHECK-LABEL: test2:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v8, v8, v9
; CHECK-NEXT:    ret
entry:
  %0 = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 2, i64 7)
  %1 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %a,
    <vscale x 1 x double> %b,
    i64 7, i64 %avl)
  ret <vscale x 1 x double> %1
}

define <vscale x 1 x i64> @test3(i64 %avl, <vscale x 1 x i64> %a, ptr %b, <vscale x 1 x i1> %c) nounwind {
; CHECK-LABEL: test3:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
; CHECK-NEXT:    vle64.v v8, (a1), v0.t
; CHECK-NEXT:    ret
entry:
  %0 = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 3, i64 0)
  %1 = call <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
    <vscale x 1 x i64> %a,
    ptr %b,
    <vscale x 1 x i1> %c,
    i64 %0, i64 1)

  ret <vscale x 1 x i64> %1
}

define <vscale x 1 x i64> @test4(i64 %avl, <vscale x 1 x i64> %a, ptr %b, <vscale x 1 x i1> %c) nounwind {
; CHECK-LABEL: test4:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
; CHECK-NEXT:    vle64.v v8, (a1), v0.t
; CHECK-NEXT:    ret
entry:
  %0 = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 3, i64 0)
  %1 = call <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
    <vscale x 1 x i64> %a,
    ptr %b,
    <vscale x 1 x i1> %c,
    i64 %avl, i64 1)

  ret <vscale x 1 x i64> %1
}

; Make sure we don't insert a vsetvli for the vmand instruction.
define <vscale x 1 x i1> @test5(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %2, i64 %avl) nounwind {
; CHECK-LABEL: test5:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
; CHECK-NEXT:    vmseq.vv v8, v8, v9
; CHECK-NEXT:    vmand.mm v0, v8, v0
; CHECK-NEXT:    ret
entry:
  %vl = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 3, i64 0)
  %a = call <vscale x 1 x i1> @llvm.riscv.vmseq.nxv1i64.i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, i64 %vl)
  %b = call <vscale x 1 x i1> @llvm.riscv.vmand.nxv1i1.i64(<vscale x 1 x i1> %a, <vscale x 1 x i1> %2, i64 %vl)
  ret <vscale x 1 x i1> %b
}

; Make sure we don't insert a vsetvli for the vmor instruction.
define void @test6(ptr nocapture readonly %A, ptr nocapture %B, i64 %n) {
; CHECK-LABEL: test6:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a2, a2, e32, m1, ta, ma
; CHECK-NEXT:    beqz a2, .LBB5_3
; CHECK-NEXT:  # %bb.1: # %for.body.preheader
; CHECK-NEXT:    li a3, 0
; CHECK-NEXT:  .LBB5_2: # %for.body
; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
; CHECK-NEXT:    slli a4, a3, 2
; CHECK-NEXT:    add a5, a0, a4
; CHECK-NEXT:    vle32.v v8, (a5)
; CHECK-NEXT:    add a3, a3, a2
; CHECK-NEXT:    vmsle.vi v9, v8, -3
; CHECK-NEXT:    vmsgt.vi v10, v8, 2
; CHECK-NEXT:    vmor.mm v0, v9, v10
; CHECK-NEXT:    add a4, a4, a1
; CHECK-NEXT:    vse32.v v8, (a4), v0.t
; CHECK-NEXT:    vsetvli a2, a2, e32, m1, ta, ma
; CHECK-NEXT:    bnez a2, .LBB5_2
; CHECK-NEXT:  .LBB5_3: # %for.cond.cleanup
; CHECK-NEXT:    ret
entry:
  %0 = tail call i64 @llvm.riscv.vsetvli.i64(i64 %n, i64 2, i64 0)
  %cmp.not11 = icmp eq i64 %0, 0
  br i1 %cmp.not11, label %for.cond.cleanup, label %for.body

for.cond.cleanup:                                 ; preds = %for.body, %entry
  ret void

for.body:                                         ; preds = %entry, %for.body
  %1 = phi i64 [ %8, %for.body ], [ %0, %entry ]
  %i.012 = phi i64 [ %add, %for.body ], [ 0, %entry ]
  %add.ptr = getelementptr inbounds i32, ptr %A, i64 %i.012
  %2 = bitcast ptr %add.ptr to ptr
  %3 = tail call <vscale x 2 x i32> @llvm.riscv.vle.nxv2i32.i64(<vscale x 2 x i32> poison, ptr %2, i64 %1)
  %4 = tail call <vscale x 2 x i1> @llvm.riscv.vmslt.nxv2i32.i32.i64(<vscale x 2 x i32> %3, i32 -2, i64 %1)
  %5 = tail call <vscale x 2 x i1> @llvm.riscv.vmsgt.nxv2i32.i32.i64(<vscale x 2 x i32> %3, i32 2, i64 %1)
  %6 = tail call <vscale x 2 x i1> @llvm.riscv.vmor.nxv2i1.i64(<vscale x 2 x i1> %4, <vscale x 2 x i1> %5, i64 %1)
  %add.ptr1 = getelementptr inbounds i32, ptr %B, i64 %i.012
  %7 = bitcast ptr %add.ptr1 to ptr
  tail call void @llvm.riscv.vse.mask.nxv2i32.i64(<vscale x 2 x i32> %3, ptr %7, <vscale x 2 x i1> %6, i64 %1)
  %add = add i64 %1, %i.012
  %8 = tail call i64 @llvm.riscv.vsetvli.i64(i64 %1, i64 2, i64 0)
  %cmp.not = icmp eq i64 %8, 0
  br i1 %cmp.not, label %for.cond.cleanup, label %for.body
}

define <vscale x 1 x i64> @test7(<vscale x 1 x i64> %a, i64 %b, <vscale x 1 x i1> %mask) nounwind {
; CHECK-LABEL: test7:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 1, e64, m1, tu, ma
; CHECK-NEXT:    vmv.s.x v8, a0
; CHECK-NEXT:    ret
entry:
  %x = tail call i64 @llvm.riscv.vsetvlimax(i64 3, i64 0)
  %y = call <vscale x 1 x i64> @llvm.riscv.vmv.s.x.nxv1i64(
    <vscale x 1 x i64> %a,
    i64 %b, i64 1)

  ret <vscale x 1 x i64> %y
}

define <vscale x 1 x i64> @test8(<vscale x 1 x i64> %a, i64 %b, <vscale x 1 x i1> %mask) nounwind {
; CHECK-LABEL: test8:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 2, e64, m1, tu, ma
; CHECK-NEXT:    vmv.s.x v8, a0
; CHECK-NEXT:    ret
entry:
  %x = tail call i64 @llvm.riscv.vsetvli(i64 6, i64 3, i64 0)
  %y = call <vscale x 1 x i64> @llvm.riscv.vmv.s.x.nxv1i64(<vscale x 1 x i64> %a, i64 %b, i64 2)
  ret <vscale x 1 x i64> %y
}

define <vscale x 1 x i64> @test9(<vscale x 1 x i64> %a, i64 %b, <vscale x 1 x i1> %mask) nounwind {
; CHECK-LABEL: test9:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 9, e64, m1, tu, mu
; CHECK-NEXT:    vadd.vv v8, v8, v8, v0.t
; CHECK-NEXT:    vmv.s.x v8, a0
; CHECK-NEXT:    ret
entry:
  %x = call <vscale x 1 x i64> @llvm.riscv.vadd.mask.nxv1i64.nxv1i64(
    <vscale x 1 x i64> %a,
    <vscale x 1 x i64> %a,
    <vscale x 1 x i64> %a,
    <vscale x 1 x i1> %mask,
    i64 9,
    i64 0)
  %y = call <vscale x 1 x i64> @llvm.riscv.vmv.s.x.nxv1i64(<vscale x 1 x i64> %x, i64 %b, i64 2)
  ret <vscale x 1 x i64> %y
}

define <vscale x 1 x double> @test10(<vscale x 1 x double> %a, double %b) nounwind {
; CHECK-LABEL: test10:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 1, e64, m1, tu, ma
; CHECK-NEXT:    vfmv.s.f v8, fa0
; CHECK-NEXT:    ret
entry:
  %x = tail call i64 @llvm.riscv.vsetvlimax(i64 3, i64 0)
  %y = call <vscale x 1 x double> @llvm.riscv.vfmv.s.f.nxv1f64(
    <vscale x 1 x double> %a, double %b, i64 1)
  ret <vscale x 1 x double> %y
}

define <vscale x 1 x double> @test11(<vscale x 1 x double> %a, double %b) nounwind {
; CHECK-LABEL: test11:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 2, e64, m1, tu, ma
; CHECK-NEXT:    vfmv.s.f v8, fa0
; CHECK-NEXT:    ret
entry:
  %x = tail call i64 @llvm.riscv.vsetvli(i64 6, i64 3, i64 0)
  %y = call <vscale x 1 x double> @llvm.riscv.vfmv.s.f.nxv1f64(
    <vscale x 1 x double> %a, double %b, i64 2)
  ret <vscale x 1 x double> %y
}

define <vscale x 1 x double> @test12(<vscale x 1 x double> %a, double %b, <vscale x 1 x i1> %mask) nounwind {
; CHECK-LABEL: test12:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 9, e64, m1, tu, mu
; CHECK-NEXT:    vfadd.vv v8, v8, v8, v0.t
; CHECK-NEXT:    vfmv.s.f v8, fa0
; CHECK-NEXT:    ret
entry:
  %x = call <vscale x 1 x double> @llvm.riscv.vfadd.mask.nxv1f64.f64(
    <vscale x 1 x double> %a,
    <vscale x 1 x double> %a,
    <vscale x 1 x double> %a,
    <vscale x 1 x i1> %mask,
    i64 7,
    i64 9,
    i64 0)
  %y = call <vscale x 1 x double> @llvm.riscv.vfmv.s.f.nxv1f64(
    <vscale x 1 x double> %x, double %b, i64 2)
  ret <vscale x 1 x double> %y
}

define <vscale x 1 x double> @test13(<vscale x 1 x double> %a, <vscale x 1 x double> %b) nounwind {
; CHECK-LABEL: test13:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v8, v8, v9
; CHECK-NEXT:    ret
entry:
  %0 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %a,
    <vscale x 1 x double> %b,
    i64 7, i64 -1)
  ret <vscale x 1 x double> %0
}

define <vscale x 1 x double> @test14(i64 %avl, <vscale x 1 x double> %a, <vscale x 1 x double> %b) nounwind {
; CHECK-LABEL: test14:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v8, v8, v9
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v8, v8, v9
; CHECK-NEXT:    ret
entry:
  %vsetvli = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 2, i64 7)
  %f1 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %a,
    <vscale x 1 x double> %b,
    i64 7, i64 1)
  %f2 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %f1,
    <vscale x 1 x double> %b,
    i64 7, i64 %vsetvli)
  ret <vscale x 1 x double> %f2
}

define <vscale x 1 x double> @test15(i64 %avl, <vscale x 1 x double> %a, <vscale x 1 x double> %b) nounwind {
; CHECK-LABEL: test15:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v8, v8, v9
; CHECK-NEXT:    vfadd.vv v8, v8, v9
; CHECK-NEXT:    ret
entry:
  %vsetvli = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 2, i64 7)
  %f1 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %a,
    <vscale x 1 x double> %b,
    i64 7, i64 %avl)
  %f2 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %f1,
    <vscale x 1 x double> %b,
    i64 7, i64 %vsetvli)
  ret <vscale x 1 x double> %f2
}

@gdouble = external global double

define <vscale x 1 x double> @test16(i64 %avl, double %a, <vscale x 1 x double> %b) nounwind {
; CHECK-LABEL: test16:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a0, a0, e64, mf2, ta, ma
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
; CHECK-NEXT:    vfmv.v.f v9, fa0
; CHECK-NEXT:    vfadd.vv v8, v9, v8
; CHECK-NEXT:    ret
entry:
  %vsetvli = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 3, i64 7)

  %head = insertelement <vscale x 1 x double> poison, double %a, i32 0
  %splat = shufflevector <vscale x 1 x double> %head, <vscale x 1 x double> poison, <vscale x 1 x i32> zeroinitializer
  %f2 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %splat,
    <vscale x 1 x double> %b,
    i64 7, i64 %vsetvli)
  ret <vscale x 1 x double> %f2
}

define double @test17(i64 %avl, <vscale x 1 x double> %a, <vscale x 1 x double> %b) nounwind {
; CHECK-LABEL: test17:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
; CHECK-NEXT:    vfmv.f.s fa5, v8
; CHECK-NEXT:    vfadd.vv v8, v8, v9
; CHECK-NEXT:    vfmv.f.s fa4, v8
; CHECK-NEXT:    fadd.d fa0, fa5, fa4
; CHECK-NEXT:    ret
entry:
  %vsetvli = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 2, i64 7)
  %c1 = extractelement <vscale x 1 x double> %a, i32 0
  %f2 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %a,
    <vscale x 1 x double> %b,
    i64 7, i64 %vsetvli)
  %c2 = extractelement <vscale x 1 x double> %f2, i32 0
  %c3 = fadd double %c1, %c2
  ret double %c3
}

define <vscale x 1 x double> @test18(<vscale x 1 x double> %a, double %b) nounwind {
; CHECK-LABEL: test18:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 6, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v9, v8, v8
; CHECK-NEXT:    vsetvli zero, zero, e64, m1, tu, ma
; CHECK-NEXT:    vfmv.s.f v8, fa0
; CHECK-NEXT:    vfmv.s.f v9, fa0
; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v8, v8, v9
; CHECK-NEXT:    ret
entry:
  %x = tail call i64 @llvm.riscv.vsetvli(i64 6, i64 3, i64 0)
  %y = call <vscale x 1 x double> @llvm.riscv.vfmv.s.f.nxv1f64(
    <vscale x 1 x double> %a, double %b, i64 2)
  %f2 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %a,
    <vscale x 1 x double> %a,
    i64 7, i64 %x)
  %y2 = call <vscale x 1 x double> @llvm.riscv.vfmv.s.f.nxv1f64(
    <vscale x 1 x double> %f2, double %b, i64 1)
  %res = fadd <vscale x 1 x double> %y, %y2
  ret <vscale x 1 x double> %res
}

define <vscale x 1 x double> @test19(<vscale x 1 x double> %a, double %b) nounwind {
; CHECK-LABEL: test19:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 1, e64, m8, tu, ma
; CHECK-NEXT:    vmv1r.v v9, v8
; CHECK-NEXT:    vfmv.s.f v9, fa0
; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v8, v9, v8
; CHECK-NEXT:    ret
entry:
  %x = tail call i64 @llvm.riscv.vsetvli(i64 6, i64 3, i64 0)
  %y = call <vscale x 1 x double> @llvm.riscv.vfmv.s.f.nxv1f64(
    <vscale x 1 x double> %a, double %b, i64 2)
  %y2 = fadd <vscale x 1 x double> %y, %a
  ret <vscale x 1 x double> %y2
}

define i64 @avl_forward1(<vscale x 2 x i32> %v, ptr %p) nounwind {
; CHECK-LABEL: avl_forward1:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli a1, 6, e32, m1, ta, ma
; CHECK-NEXT:    vse32.v v8, (a0)
; CHECK-NEXT:    mv a0, a1
; CHECK-NEXT:    ret
entry:
  %vl = tail call i64 @llvm.riscv.vsetvli(i64 6, i64 2, i64 0)
  call void @llvm.riscv.vse.nxv2i32.i64(<vscale x 2 x i32> %v, ptr %p, i64 %vl)
  ret i64 %vl
}

; Incompatible vtype
define i64 @avl_forward1b_neg(<vscale x 2 x i32> %v, ptr %p) nounwind {
; CHECK-LABEL: avl_forward1b_neg:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli a1, 6, e16, m1, ta, ma
; CHECK-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
; CHECK-NEXT:    vse32.v v8, (a0)
; CHECK-NEXT:    mv a0, a1
; CHECK-NEXT:    ret
entry:
  %vl = tail call i64 @llvm.riscv.vsetvli(i64 6, i64 1, i64 0)
  call void @llvm.riscv.vse.nxv2i32.i64(<vscale x 2 x i32> %v, ptr %p, i64 %vl)
  ret i64 %vl
}

define i64 @avl_forward2(<vscale x 2 x i32> %v, ptr %p) nounwind {
; CHECK-LABEL: avl_forward2:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
; CHECK-NEXT:    vse32.v v8, (a0)
; CHECK-NEXT:    mv a0, a1
; CHECK-NEXT:    ret
entry:
  %vl = tail call i64 @llvm.riscv.vsetvlimax(i64 2, i64 0)
  call void @llvm.riscv.vse.nxv2i32.i64(<vscale x 2 x i32> %v, ptr %p, i64 %vl)
  ret i64 %vl
}

; %vl is intentionally used only once
define void @avl_forward3(<vscale x 2 x i32> %v, ptr %p, i64 %reg) nounwind {
; CHECK-LABEL: avl_forward3:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
; CHECK-NEXT:    vse32.v v8, (a0)
; CHECK-NEXT:    ret
entry:
  %vl = tail call i64 @llvm.riscv.vsetvli(i64 %reg, i64 2, i64 0)
  call void @llvm.riscv.vse.nxv2i32.i64(<vscale x 2 x i32> %v, ptr %p, i64 %vl)
  ret void
}

; %vl has multiple uses
define i64 @avl_forward3b(<vscale x 2 x i32> %v, ptr %p, i64 %reg) nounwind {
; CHECK-LABEL: avl_forward3b:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a1, a1, e32, m1, ta, ma
; CHECK-NEXT:    vse32.v v8, (a0)
; CHECK-NEXT:    mv a0, a1
; CHECK-NEXT:    ret
entry:
  %vl = tail call i64 @llvm.riscv.vsetvli(i64 %reg, i64 2, i64 0)
  call void @llvm.riscv.vse.nxv2i32.i64(<vscale x 2 x i32> %v, ptr %p, i64 %vl)
  ret i64 %vl
}

; Like4, but with incompatible VTYPE
define void @avl_forward4(<vscale x 2 x i32> %v, ptr %p, i64 %reg) nounwind {
; CHECK-LABEL: avl_forward4:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a1, a1, e16, m1, ta, ma
; CHECK-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
; CHECK-NEXT:    vse32.v v8, (a0)
; CHECK-NEXT:    ret
entry:
  %vl = tail call i64 @llvm.riscv.vsetvli(i64 %reg, i64 1, i64 0)
  call void @llvm.riscv.vse.nxv2i32.i64(<vscale x 2 x i32> %v, ptr %p, i64 %vl)
  ret void
}

; Like4b, but with incompatible VTYPE
define i64 @avl_forward4b(<vscale x 2 x i32> %v, ptr %p, i64 %reg) nounwind {
; CHECK-LABEL: avl_forward4b:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a1, a1, e16, m1, ta, ma
; CHECK-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
; CHECK-NEXT:    vse32.v v8, (a0)
; CHECK-NEXT:    mv a0, a1
; CHECK-NEXT:    ret
entry:
  %vl = tail call i64 @llvm.riscv.vsetvli(i64 %reg, i64 1, i64 0)
  call void @llvm.riscv.vse.nxv2i32.i64(<vscale x 2 x i32> %v, ptr %p, i64 %vl)
  ret i64 %vl
}

; Fault first loads can modify VL.
; TODO: The VSETVLI of vadd could be removed here.
define <vscale x 1 x i64> @vleNff(ptr %str, i64 %n, i64 %x) {
; CHECK-LABEL: vleNff:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a1, a1, e8, m4, ta, ma
; CHECK-NEXT:    vsetvli zero, a1, e64, m1, ta, ma
; CHECK-NEXT:    vle64ff.v v8, (a0)
; CHECK-NEXT:    vsetvli zero, zero, e64, m1, tu, ma
; CHECK-NEXT:    vadd.vx v8, v8, a2
; CHECK-NEXT:    ret
entry:
  %0 = tail call i64 @llvm.riscv.vsetvli.i64(i64 %n, i64 0, i64 2)
  %1 = bitcast ptr %str to ptr
  %2 = tail call { <vscale x 1 x i64>, i64 } @llvm.riscv.vleff.nxv1i64.i64(<vscale x 1 x i64> poison, ptr %1, i64 %0)
  %3 = extractvalue { <vscale x 1 x i64>, i64 } %2, 0
  %4 = extractvalue { <vscale x 1 x i64>, i64 } %2, 1
  %5 = tail call <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.i64.i64(<vscale x 1 x i64> %3, <vscale x 1 x i64> %3, i64 %x, i64 %4)
  ret <vscale x 1 x i64> %5
}

; Similiar test case, but use same policy for vleff and vadd.
; Note: The test may be redundant if we could fix the TODO of @vleNff.
define <vscale x 1 x i64> @vleNff2(ptr %str, i64 %n, i64 %x) {
; CHECK-LABEL: vleNff2:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a1, a1, e8, m4, ta, ma
; CHECK-NEXT:    vsetvli zero, a1, e64, m1, ta, ma
; CHECK-NEXT:    vle64ff.v v8, (a0)
; CHECK-NEXT:    vadd.vx v8, v8, a2
; CHECK-NEXT:    ret
entry:
  %0 = tail call i64 @llvm.riscv.vsetvli.i64(i64 %n, i64 0, i64 2)
  %1 = bitcast ptr %str to ptr
  %2 = tail call { <vscale x 1 x i64>, i64 } @llvm.riscv.vleff.nxv1i64.i64(<vscale x 1 x i64> poison, ptr %1, i64 %0)
  %3 = extractvalue { <vscale x 1 x i64>, i64 } %2, 0
  %4 = extractvalue { <vscale x 1 x i64>, i64 } %2, 1
  %5 = tail call <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.i64.i64(<vscale x 1 x i64> poison, <vscale x 1 x i64> %3, i64 %x, i64 %4)
  ret <vscale x 1 x i64> %5
}

; Ensure AVL register is alive when forwarding an AVL immediate that does not fit in 5 bits
define <vscale x 2 x i32> @avl_forward5(ptr %addr) {
; CHECK-LABEL: avl_forward5:
; CHECK:       # %bb.0:
; CHECK-NEXT:    li a1, 32
; CHECK-NEXT:    vsetvli a1, a1, e8, m4, ta, ma
; CHECK-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
; CHECK-NEXT:    vle32.v v8, (a0)
; CHECK-NEXT:    ret
  %gvl = tail call i64 @llvm.riscv.vsetvli.i64(i64 32, i64 0, i64 2)
  %ret = tail call <vscale x 2 x i32> @llvm.riscv.vle.nxv2i32.i64(<vscale x 2 x i32> poison, ptr %addr, i64 %gvl)
  ret <vscale x 2 x i32> %ret
}

define <vscale x 1 x double> @test20(i64 %avl, <vscale x 1 x float> %a, <vscale x 1 x float> %b, <vscale x 1 x double> %c) nounwind {
; CHECK-LABEL: test20:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
; CHECK-NEXT:    vfwadd.vv v11, v8, v9
; CHECK-NEXT:    vsetvli zero, zero, e64, m1, ta, ma
; CHECK-NEXT:    vfadd.vv v8, v11, v10
; CHECK-NEXT:    ret
entry:
  %0 = tail call i64 @llvm.riscv.vsetvli(i64 %avl, i64 2, i64 7)
  %1 = tail call <vscale x 1 x double> @llvm.riscv.vfwadd.nxv1f64.nxv1f32.nxv1f32(
    <vscale x 1 x double> poison,
    <vscale x 1 x float> %a,
    <vscale x 1 x float> %b,
    i64 7, i64 %0)
  %2 = tail call <vscale x 1 x double> @llvm.riscv.vfadd.nxv1f64.nxv1f64(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %1,
    <vscale x 1 x double> %c,
    i64 7, i64 %0)
  ret <vscale x 1 x double> %2
}

; This used to fail the machine verifier due to the vsetvli being removed
; while the add was still using it.
define i64 @bad_removal(<2 x i64> %arg) {
; CHECK-LABEL: bad_removal:
; CHECK:       # %bb.0: # %bb
; CHECK-NEXT:    vsetivli a0, 16, e64, m1, ta, ma
; CHECK-NEXT:    vmv.x.s a1, v8
; CHECK-NEXT:    add a0, a0, a1
; CHECK-NEXT:    ret
bb:
  %tmp = extractelement <2 x i64> %arg, i64 0
  %tmp1 = call i64 @llvm.riscv.vsetvli.i64(i64 16, i64 3, i64 0)
  %tmp2 = add i64 %tmp, %tmp1
  ret i64 %tmp2
}

define void @add_v128i8(ptr %x, ptr %y) vscale_range(2,2) {
; CHECK-LABEL: add_v128i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vl8r.v v8, (a1)
; CHECK-NEXT:    vl8r.v v16, (a0)
; CHECK-NEXT:    vsetvli a1, zero, e8, m8, ta, ma
; CHECK-NEXT:    vadd.vv v8, v16, v8
; CHECK-NEXT:    vs8r.v v8, (a0)
; CHECK-NEXT:    ret
  %a = load <128 x i8>, ptr %x
  %b = load <128 x i8>, ptr %y
  %c = add <128 x i8> %a, %b
  store <128 x i8> %c, ptr %x
  ret void
}

define void @add_v16i64(ptr %x, ptr %y) vscale_range(2,2) {
; CHECK-LABEL: add_v16i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vl8re64.v v8, (a0)
; CHECK-NEXT:    vl8re64.v v16, (a1)
; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, ma
; CHECK-NEXT:    vadd.vv v8, v8, v16
; CHECK-NEXT:    vs8r.v v8, (a0)
; CHECK-NEXT:    ret
  %a = load <16 x i64>, ptr %x
  %b = load <16 x i64>, ptr %y
  %c = add <16 x i64> %a, %b
  store <16 x i64> %c, ptr %x
  ret void
}

define <vscale x 2 x float> @fp_reduction_vfmv_s_f(float %0, <vscale x 8 x float> %1, i64  %2) {
; CHECK-LABEL: fp_reduction_vfmv_s_f:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, ma
; CHECK-NEXT:    vfmv.s.f v12, fa0
; CHECK-NEXT:    vfredusum.vs v8, v8, v12
; CHECK-NEXT:    ret
  %4 = tail call <vscale x 8 x float> @llvm.riscv.vfmv.s.f.nxv8f32.i64(<vscale x 8 x float> poison, float %0, i64 %2)
  %5 = tail call <vscale x 2 x float> @llvm.vector.extract.nxv2f32.nxv8f32(<vscale x 8 x float> %4, i64 0)
  %6 = tail call <vscale x 2 x float> @llvm.riscv.vfredusum.nxv2f32.nxv8f32.i64(<vscale x 2 x float> poison, <vscale x 8 x float> %1, <vscale x 2 x float> %5, i64 7, i64 %2)
  ret <vscale x 2 x float> %6
}

define dso_local <vscale x 2 x i32> @int_reduction_vmv_s_x(i32 signext %0, <vscale x 8 x i32> %1, i64 %2) {
; CHECK-LABEL: int_reduction_vmv_s_x:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
; CHECK-NEXT:    vmv.s.x v12, a0
; CHECK-NEXT:    vredsum.vs v8, v8, v12
; CHECK-NEXT:    ret
  %4 = tail call <vscale x 8 x i32> @llvm.riscv.vmv.s.x.nxv8i32.i64(<vscale x 8 x i32> poison, i32 %0, i64 %2)
  %5 = tail call <vscale x 2 x i32> @llvm.vector.extract.nxv2i32.nxv8i32(<vscale x 8 x i32> %4, i64 0)
  %6 = tail call <vscale x 2 x i32> @llvm.riscv.vredsum.nxv2i32.nxv8i32.i64(<vscale x 2 x i32> poison, <vscale x 8 x i32> %1, <vscale x 2 x i32> %5, i64 %2)
  ret <vscale x 2 x i32> %6
}

define <vscale x 2 x i32> @avl_undef1(<vscale x 2 x i32>, <vscale x 2 x i32>, <vscale x 2 x i32>) {
; CHECK-LABEL: avl_undef1:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetivli zero, 1, e32, m1, tu, ma
; CHECK-NEXT:    vadd.vv v8, v9, v10
; CHECK-NEXT:    ret
  %a = call <vscale x 2 x i32> @llvm.riscv.vadd.nxv2i32.nxv2i32(
    <vscale x 2 x i32> %0,
    <vscale x 2 x i32> %1,
    <vscale x 2 x i32> %2,
    i64 poison
  )
  ret <vscale x 2 x i32> %a
}

define i64 @avl_undef2() {
; CHECK-LABEL: avl_undef2:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a0, a0, e32, mf2, ta, ma
; CHECK-NEXT:    ret
  %1 = tail call i64 @llvm.riscv.vsetvli(i64 poison, i64 2, i64 7)
  ret i64 %1
}

define i64 @vsetvli_vleff(ptr %s, i64 %evl) {
; CHECK-LABEL: vsetvli_vleff:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetvli a3, zero, e16, m1, ta, ma
; CHECK-NEXT:    vmv.v.i v8, 0
; CHECK-NEXT:  .LBB37_1: # %while.body
; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
; CHECK-NEXT:    vsetvli zero, a1, e16, m1, tu, ma
; CHECK-NEXT:    vmv1r.v v9, v8
; CHECK-NEXT:    vle16ff.v v9, (a0)
; CHECK-NEXT:    csrr a2, vl
; CHECK-NEXT:    beqz a2, .LBB37_1
; CHECK-NEXT:  # %bb.2: # %while.end
; CHECK-NEXT:    li a0, 0
; CHECK-NEXT:    ret
entry:
  br label %while.cond

while.cond:
  %new_vl.0 = phi i64 [ 0, %entry ], [ %1, %while.body ]
  %cmp = icmp eq i64 %new_vl.0, 0
  br i1 %cmp, label %while.body, label %while.end

while.body:
  %0 = tail call { <vscale x 4 x i16>, i64 } @llvm.riscv.vleff.nxv4i16.i64(<vscale x 4 x i16> zeroinitializer, ptr %s, i64 %evl)
  %1 = extractvalue { <vscale x 4 x i16>, i64 } %0, 1
  br label %while.cond

while.end:
  ret i64 0
}

define <vscale x 1 x i64> @vslideup_vl1(<vscale x 1 x i64> %a) nounwind {
; NODEPVL-LABEL: vslideup_vl1:
; NODEPVL:       # %bb.0: # %entry
; NODEPVL-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
; NODEPVL-NEXT:    vslideup.vi v9, v8, 1
; NODEPVL-NEXT:    vadd.vv v8, v9, v9
; NODEPVL-NEXT:    ret
;
; DEPVL-LABEL: vslideup_vl1:
; DEPVL:       # %bb.0: # %entry
; DEPVL-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
; DEPVL-NEXT:    vslideup.vi v9, v8, 1
; DEPVL-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
; DEPVL-NEXT:    vadd.vv v8, v9, v9
; DEPVL-NEXT:    ret
entry:
  %1 = tail call <vscale x 1 x i64> @llvm.riscv.vslideup(
    <vscale x 1 x i64> poison,
    <vscale x 1 x i64> %a,
    i64 1,
    i64 1,
    i64 3)
  %2 = tail call <vscale x 1 x i64> @llvm.riscv.vadd(
    <vscale x 1 x i64> poison,
    <vscale x 1 x i64> %1,
    <vscale x 1 x i64> %1,
    i64 2)
  ret <vscale x 1 x i64> %2
}

define <vscale x 1 x i64> @vslidedown_vl1(<vscale x 1 x i64> %a) nounwind {
; NODEPVL-LABEL: vslidedown_vl1:
; NODEPVL:       # %bb.0: # %entry
; NODEPVL-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
; NODEPVL-NEXT:    vslidedown.vi v8, v8, 1
; NODEPVL-NEXT:    vadd.vv v8, v8, v8
; NODEPVL-NEXT:    ret
;
; DEPVL-LABEL: vslidedown_vl1:
; DEPVL:       # %bb.0: # %entry
; DEPVL-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
; DEPVL-NEXT:    vslidedown.vi v8, v8, 1
; DEPVL-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
; DEPVL-NEXT:    vadd.vv v8, v8, v8
; DEPVL-NEXT:    ret
entry:
  %1 = tail call <vscale x 1 x i64> @llvm.riscv.vslidedown(
    <vscale x 1 x i64> poison,
    <vscale x 1 x i64> %a,
    i64 1,
    i64 1,
    i64 3)
  %2 = tail call <vscale x 1 x i64> @llvm.riscv.vadd(
    <vscale x 1 x i64> poison,
    <vscale x 1 x i64> %1,
    <vscale x 1 x i64> %1,
    i64 2)
  ret <vscale x 1 x i64> %2
}

define <vscale x 1 x i64> @vmv.v.x_vl1() nounwind {
; NODEPVL-LABEL: vmv.v.x_vl1:
; NODEPVL:       # %bb.0: # %entry
; NODEPVL-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
; NODEPVL-NEXT:    vmv.v.i v8, 1
; NODEPVL-NEXT:    vadd.vv v8, v8, v8
; NODEPVL-NEXT:    ret
;
; DEPVL-LABEL: vmv.v.x_vl1:
; DEPVL:       # %bb.0: # %entry
; DEPVL-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
; DEPVL-NEXT:    vmv.v.i v8, 1
; DEPVL-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
; DEPVL-NEXT:    vadd.vv v8, v8, v8
; DEPVL-NEXT:    ret
entry:
  %1 = tail call <vscale x 1 x i64> @llvm.riscv.vmv.v.x(
    <vscale x 1 x i64> poison,
    i64 1,
    i64 1)
  %2 = tail call <vscale x 1 x i64> @llvm.riscv.vadd(
    <vscale x 1 x i64> poison,
    <vscale x 1 x i64> %1,
    <vscale x 1 x i64> %1,
    i64 2)
  ret <vscale x 1 x i64> %2
}

define <vscale x 1 x double> @vfmv.v.f_vl1(double %f) nounwind {
; CHECK-LABEL: vfmv.v.f_vl1:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
; CHECK-NEXT:    vfmv.s.f v8, fa0
; CHECK-NEXT:    vfadd.vv v8, v8, v8
; CHECK-NEXT:    ret
entry:
  %1 = tail call <vscale x 1 x double> @llvm.riscv.vfmv.v.f(
    <vscale x 1 x double> poison,
    double %f,
    i64 1)
  %2 = tail call <vscale x 1 x double> @llvm.riscv.vfadd(
    <vscale x 1 x double> poison,
    <vscale x 1 x double> %1,
    <vscale x 1 x double> %1,
    i64 7,
    i64 2)
  ret <vscale x 1 x double> %2
}

; The two vsetvlis will be coalesced so the add will be made dead and
; removed. Make sure we shrink the live interval of %x.
define void @non_li_addi(i64 %x, ptr %p) {
; CHECK-LABEL: non_li_addi:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
; CHECK-NEXT:    ret
entry:
  %add = add i64 %x, 1
  %0 = tail call i64 @llvm.riscv.vsetvli(i64 %add, i64 3, i64 0)
  %1 = call <vscale x 8 x i8> @llvm.riscv.vle(<vscale x 8 x i8> poison, ptr %p, i64 %0)
  %2 = tail call i64 @llvm.riscv.vsetvli(i64 1, i64 3, i64 0)
  %3 = tail call { <vscale x 8 x i8>, i64 } @llvm.riscv.vleff(<vscale x 8 x i8> poison, ptr %p, i64 %2)
  ret void
}

; This will create a live interval in such a way we can't coalesce two vsetvlis,
; see the corresponding .mir test for more details. Make sure we check for this
; and don't crash.
define void @coalesce_vl_clobber(ptr %p) {
; CHECK-LABEL: coalesce_vl_clobber:
; CHECK:       # %bb.0: # %entry
; CHECK-NEXT:    li a2, 0
; CHECK-NEXT:    li a1, 0
; CHECK-NEXT:    vsetivli zero, 0, e8, mf2, ta, ma
; CHECK-NEXT:    vmclr.m v8
; CHECK-NEXT:    vmv.v.i v9, 0
; CHECK-NEXT:    vmv1r.v v0, v8
; CHECK-NEXT:    vmerge.vim v9, v9, 1, v0
; CHECK-NEXT:  .LBB43_1: # %vector.body
; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
; CHECK-NEXT:    vsetvli a3, a2, e8, mf8, ta, ma
; CHECK-NEXT:    vmv1r.v v0, v8
; CHECK-NEXT:    slli a1, a1, 32
; CHECK-NEXT:    vmv1r.v v10, v9
; CHECK-NEXT:    srli a1, a1, 32
; CHECK-NEXT:    vsetivli zero, 0, e8, mf2, ta, mu
; CHECK-NEXT:    vslideup.vx v10, v9, a1, v0.t
; CHECK-NEXT:    vsetvli zero, zero, e8, mf2, ta, ma
; CHECK-NEXT:    vmsne.vi v0, v10, 0, v0.t
; CHECK-NEXT:    vsetvli zero, a3, e32, m2, ta, ma
; CHECK-NEXT:    vmv.v.i v10, 0
; CHECK-NEXT:    vse32.v v10, (a0), v0.t
; CHECK-NEXT:    li a2, 1
; CHECK-NEXT:    mv a1, a3
; CHECK-NEXT:    j .LBB43_1
entry:
  br label %vector.body

vector.body:
  %avl = phi i64 [ 0, %entry ], [ 1, %vector.body ]
  %prev.evl = phi i32 [ 0, %entry ], [ %0, %vector.body ]
  %0 = tail call i32 @llvm.experimental.get.vector.length(i64 %avl, i32 1, i1 true)
  %1 = tail call <vscale x 4 x i1> @llvm.experimental.vp.splice(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 0, <vscale x 4 x i1> zeroinitializer, i32 %prev.evl, i32 0)
  tail call void @llvm.vp.store(<vscale x 4 x float> zeroinitializer, ptr %p, <vscale x 4 x i1> %1, i32 %0)
  br label %vector.body
}