; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+experimental-zvfh,+v,+m -target-abi=ilp32d -riscv-v-vector-bits-min=128 \
; RUN:   -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32
; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+experimental-zvfh,+v,+m -target-abi=lp64d -riscv-v-vector-bits-min=128 \
; RUN:   -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64

declare <4 x i1> @llvm.vp.merge.v4i1(<4 x i1>, <4 x i1>, <4 x i1>, i32)

define <4 x i1> @vpmerge_vv_v4i1(<4 x i1> %va, <4 x i1> %vb, <4 x i1> %m, i32 zeroext %evl) {
; RV32-LABEL: vpmerge_vv_v4i1:
; RV32:       # %bb.0:
; RV32-NEXT:    vsetivli zero, 4, e32, m1, ta, mu
; RV32-NEXT:    vid.v v10
; RV32-NEXT:    vmsltu.vx v10, v10, a0
; RV32-NEXT:    vmand.mm v9, v9, v10
; RV32-NEXT:    vmandn.mm v8, v8, v9
; RV32-NEXT:    vmand.mm v9, v0, v9
; RV32-NEXT:    vmor.mm v0, v9, v8
; RV32-NEXT:    ret
;
; RV64-LABEL: vpmerge_vv_v4i1:
; RV64:       # %bb.0:
; RV64-NEXT:    vsetivli zero, 4, e64, m2, ta, mu
; RV64-NEXT:    vid.v v10
; RV64-NEXT:    vmsltu.vx v12, v10, a0
; RV64-NEXT:    vmand.mm v9, v9, v12
; RV64-NEXT:    vmandn.mm v8, v8, v9
; RV64-NEXT:    vmand.mm v9, v0, v9
; RV64-NEXT:    vmor.mm v0, v9, v8
; RV64-NEXT:    ret
  %v = call <4 x i1> @llvm.vp.merge.v4i1(<4 x i1> %m, <4 x i1> %va, <4 x i1> %vb, i32 %evl)
  ret <4 x i1> %v
}

declare <2 x i8> @llvm.vp.merge.v2i8(<2 x i1>, <2 x i8>, <2 x i8>, i32)

define <2 x i8> @vpmerge_vv_v2i8(<2 x i8> %va, <2 x i8> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v2i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <2 x i8> @llvm.vp.merge.v2i8(<2 x i1> %m, <2 x i8> %va, <2 x i8> %vb, i32 %evl)
  ret <2 x i8> %v
}

define <2 x i8> @vpmerge_vx_v2i8(i8 %a, <2 x i8> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v2i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e8, mf8, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x i8> poison, i8 %a, i32 0
  %va = shufflevector <2 x i8> %elt.head, <2 x i8> poison, <2 x i32> zeroinitializer
  %v = call <2 x i8> @llvm.vp.merge.v2i8(<2 x i1> %m, <2 x i8> %va, <2 x i8> %vb, i32 %evl)
  ret <2 x i8> %v
}

define <2 x i8> @vpmerge_vi_v2i8(<2 x i8> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v2i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x i8> poison, i8 2, i32 0
  %va = shufflevector <2 x i8> %elt.head, <2 x i8> poison, <2 x i32> zeroinitializer
  %v = call <2 x i8> @llvm.vp.merge.v2i8(<2 x i1> %m, <2 x i8> %va, <2 x i8> %vb, i32 %evl)
  ret <2 x i8> %v
}

declare <4 x i8> @llvm.vp.merge.v4i8(<4 x i1>, <4 x i8>, <4 x i8>, i32)

define <4 x i8> @vpmerge_vv_v4i8(<4 x i8> %va, <4 x i8> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v4i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x i8> @llvm.vp.merge.v4i8(<4 x i1> %m, <4 x i8> %va, <4 x i8> %vb, i32 %evl)
  ret <4 x i8> %v
}

define <4 x i8> @vpmerge_vx_v4i8(i8 %a, <4 x i8> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v4i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e8, mf4, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x i8> poison, i8 %a, i32 0
  %va = shufflevector <4 x i8> %elt.head, <4 x i8> poison, <4 x i32> zeroinitializer
  %v = call <4 x i8> @llvm.vp.merge.v4i8(<4 x i1> %m, <4 x i8> %va, <4 x i8> %vb, i32 %evl)
  ret <4 x i8> %v
}

define <4 x i8> @vpmerge_vi_v4i8(<4 x i8> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v4i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x i8> poison, i8 2, i32 0
  %va = shufflevector <4 x i8> %elt.head, <4 x i8> poison, <4 x i32> zeroinitializer
  %v = call <4 x i8> @llvm.vp.merge.v4i8(<4 x i1> %m, <4 x i8> %va, <4 x i8> %vb, i32 %evl)
  ret <4 x i8> %v
}

declare <6 x i8> @llvm.vp.merge.v6i8(<6 x i1>, <6 x i8>, <6 x i8>, i32)

define <6 x i8> @vpmerge_vv_v6i8(<6 x i8> %va, <6 x i8> %vb, <6 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v6i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <6 x i8> @llvm.vp.merge.v6i8(<6 x i1> %m, <6 x i8> %va, <6 x i8> %vb, i32 %evl)
  ret <6 x i8> %v
}

define <6 x i8> @vpmerge_vx_v6i8(i8 %a, <6 x i8> %vb, <6 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v6i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e8, mf2, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <6 x i8> poison, i8 %a, i32 0
  %va = shufflevector <6 x i8> %elt.head, <6 x i8> poison, <6 x i32> zeroinitializer
  %v = call <6 x i8> @llvm.vp.merge.v6i8(<6 x i1> %m, <6 x i8> %va, <6 x i8> %vb, i32 %evl)
  ret <6 x i8> %v
}

define <6 x i8> @vpmerge_vi_v6i8(<6 x i8> %vb, <6 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v6i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <6 x i8> poison, i8 2, i32 0
  %va = shufflevector <6 x i8> %elt.head, <6 x i8> poison, <6 x i32> zeroinitializer
  %v = call <6 x i8> @llvm.vp.merge.v6i8(<6 x i1> %m, <6 x i8> %va, <6 x i8> %vb, i32 %evl)
  ret <6 x i8> %v
}

declare <8 x i7> @llvm.vp.merge.v8i7(<8 x i1>, <8 x i7>, <8 x i7>, i32)

define <8 x i7> @vpmerge_vv_v8i7(<8 x i7> %va, <8 x i7> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v8i7:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <8 x i7> @llvm.vp.merge.v8i7(<8 x i1> %m, <8 x i7> %va, <8 x i7> %vb, i32 %evl)
  ret <8 x i7> %v
}

define <8 x i7> @vpmerge_vx_v8i7(i7 %a, <8 x i7> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v8i7:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e8, mf2, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x i7> poison, i7 %a, i32 0
  %va = shufflevector <8 x i7> %elt.head, <8 x i7> poison, <8 x i32> zeroinitializer
  %v = call <8 x i7> @llvm.vp.merge.v8i7(<8 x i1> %m, <8 x i7> %va, <8 x i7> %vb, i32 %evl)
  ret <8 x i7> %v
}

define <8 x i7> @vpmerge_vi_v8i7(<8 x i7> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v8i7:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x i7> poison, i7 2, i32 0
  %va = shufflevector <8 x i7> %elt.head, <8 x i7> poison, <8 x i32> zeroinitializer
  %v = call <8 x i7> @llvm.vp.merge.v8i7(<8 x i1> %m, <8 x i7> %va, <8 x i7> %vb, i32 %evl)
  ret <8 x i7> %v
}

declare <8 x i8> @llvm.vp.merge.v8i8(<8 x i1>, <8 x i8>, <8 x i8>, i32)

define <8 x i8> @vpmerge_vv_v8i8(<8 x i8> %va, <8 x i8> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v8i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <8 x i8> @llvm.vp.merge.v8i8(<8 x i1> %m, <8 x i8> %va, <8 x i8> %vb, i32 %evl)
  ret <8 x i8> %v
}

define <8 x i8> @vpmerge_vx_v8i8(i8 %a, <8 x i8> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v8i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e8, mf2, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x i8> poison, i8 %a, i32 0
  %va = shufflevector <8 x i8> %elt.head, <8 x i8> poison, <8 x i32> zeroinitializer
  %v = call <8 x i8> @llvm.vp.merge.v8i8(<8 x i1> %m, <8 x i8> %va, <8 x i8> %vb, i32 %evl)
  ret <8 x i8> %v
}

define <8 x i8> @vpmerge_vi_v8i8(<8 x i8> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v8i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x i8> poison, i8 2, i32 0
  %va = shufflevector <8 x i8> %elt.head, <8 x i8> poison, <8 x i32> zeroinitializer
  %v = call <8 x i8> @llvm.vp.merge.v8i8(<8 x i1> %m, <8 x i8> %va, <8 x i8> %vb, i32 %evl)
  ret <8 x i8> %v
}

declare <16 x i8> @llvm.vp.merge.v16i8(<16 x i1>, <16 x i8>, <16 x i8>, i32)

define <16 x i8> @vpmerge_vv_v16i8(<16 x i8> %va, <16 x i8> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v16i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, m1, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <16 x i8> @llvm.vp.merge.v16i8(<16 x i1> %m, <16 x i8> %va, <16 x i8> %vb, i32 %evl)
  ret <16 x i8> %v
}

define <16 x i8> @vpmerge_vx_v16i8(i8 %a, <16 x i8> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v16i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e8, m1, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x i8> poison, i8 %a, i32 0
  %va = shufflevector <16 x i8> %elt.head, <16 x i8> poison, <16 x i32> zeroinitializer
  %v = call <16 x i8> @llvm.vp.merge.v16i8(<16 x i1> %m, <16 x i8> %va, <16 x i8> %vb, i32 %evl)
  ret <16 x i8> %v
}

define <16 x i8> @vpmerge_vi_v16i8(<16 x i8> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v16i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e8, m1, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x i8> poison, i8 2, i32 0
  %va = shufflevector <16 x i8> %elt.head, <16 x i8> poison, <16 x i32> zeroinitializer
  %v = call <16 x i8> @llvm.vp.merge.v16i8(<16 x i1> %m, <16 x i8> %va, <16 x i8> %vb, i32 %evl)
  ret <16 x i8> %v
}

declare <2 x i16> @llvm.vp.merge.v2i16(<2 x i1>, <2 x i16>, <2 x i16>, i32)

define <2 x i16> @vpmerge_vv_v2i16(<2 x i16> %va, <2 x i16> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v2i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <2 x i16> @llvm.vp.merge.v2i16(<2 x i1> %m, <2 x i16> %va, <2 x i16> %vb, i32 %evl)
  ret <2 x i16> %v
}

define <2 x i16> @vpmerge_vx_v2i16(i16 %a, <2 x i16> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v2i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e16, mf4, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x i16> poison, i16 %a, i32 0
  %va = shufflevector <2 x i16> %elt.head, <2 x i16> poison, <2 x i32> zeroinitializer
  %v = call <2 x i16> @llvm.vp.merge.v2i16(<2 x i1> %m, <2 x i16> %va, <2 x i16> %vb, i32 %evl)
  ret <2 x i16> %v
}

define <2 x i16> @vpmerge_vi_v2i16(<2 x i16> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v2i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x i16> poison, i16 2, i32 0
  %va = shufflevector <2 x i16> %elt.head, <2 x i16> poison, <2 x i32> zeroinitializer
  %v = call <2 x i16> @llvm.vp.merge.v2i16(<2 x i1> %m, <2 x i16> %va, <2 x i16> %vb, i32 %evl)
  ret <2 x i16> %v
}

declare <4 x i16> @llvm.vp.merge.v4i16(<4 x i1>, <4 x i16>, <4 x i16>, i32)

define <4 x i16> @vpmerge_vv_v4i16(<4 x i16> %va, <4 x i16> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v4i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x i16> @llvm.vp.merge.v4i16(<4 x i1> %m, <4 x i16> %va, <4 x i16> %vb, i32 %evl)
  ret <4 x i16> %v
}

define <4 x i16> @vpmerge_vx_v4i16(i16 %a, <4 x i16> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v4i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e16, mf2, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x i16> poison, i16 %a, i32 0
  %va = shufflevector <4 x i16> %elt.head, <4 x i16> poison, <4 x i32> zeroinitializer
  %v = call <4 x i16> @llvm.vp.merge.v4i16(<4 x i1> %m, <4 x i16> %va, <4 x i16> %vb, i32 %evl)
  ret <4 x i16> %v
}

define <4 x i16> @vpmerge_vi_v4i16(<4 x i16> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v4i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x i16> poison, i16 2, i32 0
  %va = shufflevector <4 x i16> %elt.head, <4 x i16> poison, <4 x i32> zeroinitializer
  %v = call <4 x i16> @llvm.vp.merge.v4i16(<4 x i1> %m, <4 x i16> %va, <4 x i16> %vb, i32 %evl)
  ret <4 x i16> %v
}

declare <8 x i16> @llvm.vp.merge.v8i16(<8 x i1>, <8 x i16>, <8 x i16>, i32)

define <8 x i16> @vpmerge_vv_v8i16(<8 x i16> %va, <8 x i16> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v8i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, m1, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <8 x i16> @llvm.vp.merge.v8i16(<8 x i1> %m, <8 x i16> %va, <8 x i16> %vb, i32 %evl)
  ret <8 x i16> %v
}

define <8 x i16> @vpmerge_vx_v8i16(i16 %a, <8 x i16> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v8i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e16, m1, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x i16> poison, i16 %a, i32 0
  %va = shufflevector <8 x i16> %elt.head, <8 x i16> poison, <8 x i32> zeroinitializer
  %v = call <8 x i16> @llvm.vp.merge.v8i16(<8 x i1> %m, <8 x i16> %va, <8 x i16> %vb, i32 %evl)
  ret <8 x i16> %v
}

define <8 x i16> @vpmerge_vi_v8i16(<8 x i16> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v8i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, m1, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x i16> poison, i16 2, i32 0
  %va = shufflevector <8 x i16> %elt.head, <8 x i16> poison, <8 x i32> zeroinitializer
  %v = call <8 x i16> @llvm.vp.merge.v8i16(<8 x i1> %m, <8 x i16> %va, <8 x i16> %vb, i32 %evl)
  ret <8 x i16> %v
}

declare <16 x i16> @llvm.vp.merge.v16i16(<16 x i1>, <16 x i16>, <16 x i16>, i32)

define <16 x i16> @vpmerge_vv_v16i16(<16 x i16> %va, <16 x i16> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v16i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, m2, tu, mu
; CHECK-NEXT:    vmerge.vvm v10, v10, v8, v0
; CHECK-NEXT:    vmv2r.v v8, v10
; CHECK-NEXT:    ret
  %v = call <16 x i16> @llvm.vp.merge.v16i16(<16 x i1> %m, <16 x i16> %va, <16 x i16> %vb, i32 %evl)
  ret <16 x i16> %v
}

define <16 x i16> @vpmerge_vx_v16i16(i16 %a, <16 x i16> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v16i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e16, m2, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x i16> poison, i16 %a, i32 0
  %va = shufflevector <16 x i16> %elt.head, <16 x i16> poison, <16 x i32> zeroinitializer
  %v = call <16 x i16> @llvm.vp.merge.v16i16(<16 x i1> %m, <16 x i16> %va, <16 x i16> %vb, i32 %evl)
  ret <16 x i16> %v
}

define <16 x i16> @vpmerge_vi_v16i16(<16 x i16> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v16i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, m2, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x i16> poison, i16 2, i32 0
  %va = shufflevector <16 x i16> %elt.head, <16 x i16> poison, <16 x i32> zeroinitializer
  %v = call <16 x i16> @llvm.vp.merge.v16i16(<16 x i1> %m, <16 x i16> %va, <16 x i16> %vb, i32 %evl)
  ret <16 x i16> %v
}

declare <2 x i32> @llvm.vp.merge.v2i32(<2 x i1>, <2 x i32>, <2 x i32>, i32)

define <2 x i32> @vpmerge_vv_v2i32(<2 x i32> %va, <2 x i32> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v2i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <2 x i32> @llvm.vp.merge.v2i32(<2 x i1> %m, <2 x i32> %va, <2 x i32> %vb, i32 %evl)
  ret <2 x i32> %v
}

define <2 x i32> @vpmerge_vx_v2i32(i32 %a, <2 x i32> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v2i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e32, mf2, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x i32> poison, i32 %a, i32 0
  %va = shufflevector <2 x i32> %elt.head, <2 x i32> poison, <2 x i32> zeroinitializer
  %v = call <2 x i32> @llvm.vp.merge.v2i32(<2 x i1> %m, <2 x i32> %va, <2 x i32> %vb, i32 %evl)
  ret <2 x i32> %v
}

define <2 x i32> @vpmerge_vi_v2i32(<2 x i32> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v2i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x i32> poison, i32 2, i32 0
  %va = shufflevector <2 x i32> %elt.head, <2 x i32> poison, <2 x i32> zeroinitializer
  %v = call <2 x i32> @llvm.vp.merge.v2i32(<2 x i1> %m, <2 x i32> %va, <2 x i32> %vb, i32 %evl)
  ret <2 x i32> %v
}

declare <4 x i32> @llvm.vp.merge.v4i32(<4 x i1>, <4 x i32>, <4 x i32>, i32)

define <4 x i32> @vpmerge_vv_v4i32(<4 x i32> %va, <4 x i32> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v4i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m1, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x i32> @llvm.vp.merge.v4i32(<4 x i1> %m, <4 x i32> %va, <4 x i32> %vb, i32 %evl)
  ret <4 x i32> %v
}

define <4 x i32> @vpmerge_vx_v4i32(i32 %a, <4 x i32> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v4i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e32, m1, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x i32> poison, i32 %a, i32 0
  %va = shufflevector <4 x i32> %elt.head, <4 x i32> poison, <4 x i32> zeroinitializer
  %v = call <4 x i32> @llvm.vp.merge.v4i32(<4 x i1> %m, <4 x i32> %va, <4 x i32> %vb, i32 %evl)
  ret <4 x i32> %v
}

define <4 x i32> @vpmerge_vi_v4i32(<4 x i32> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v4i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m1, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x i32> poison, i32 2, i32 0
  %va = shufflevector <4 x i32> %elt.head, <4 x i32> poison, <4 x i32> zeroinitializer
  %v = call <4 x i32> @llvm.vp.merge.v4i32(<4 x i1> %m, <4 x i32> %va, <4 x i32> %vb, i32 %evl)
  ret <4 x i32> %v
}

declare <8 x i32> @llvm.vp.merge.v8i32(<8 x i1>, <8 x i32>, <8 x i32>, i32)

define <8 x i32> @vpmerge_vv_v8i32(<8 x i32> %va, <8 x i32> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v8i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m2, tu, mu
; CHECK-NEXT:    vmerge.vvm v10, v10, v8, v0
; CHECK-NEXT:    vmv2r.v v8, v10
; CHECK-NEXT:    ret
  %v = call <8 x i32> @llvm.vp.merge.v8i32(<8 x i1> %m, <8 x i32> %va, <8 x i32> %vb, i32 %evl)
  ret <8 x i32> %v
}

define <8 x i32> @vpmerge_vx_v8i32(i32 %a, <8 x i32> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v8i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e32, m2, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x i32> poison, i32 %a, i32 0
  %va = shufflevector <8 x i32> %elt.head, <8 x i32> poison, <8 x i32> zeroinitializer
  %v = call <8 x i32> @llvm.vp.merge.v8i32(<8 x i1> %m, <8 x i32> %va, <8 x i32> %vb, i32 %evl)
  ret <8 x i32> %v
}

define <8 x i32> @vpmerge_vi_v8i32(<8 x i32> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v8i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m2, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x i32> poison, i32 2, i32 0
  %va = shufflevector <8 x i32> %elt.head, <8 x i32> poison, <8 x i32> zeroinitializer
  %v = call <8 x i32> @llvm.vp.merge.v8i32(<8 x i1> %m, <8 x i32> %va, <8 x i32> %vb, i32 %evl)
  ret <8 x i32> %v
}

declare <16 x i32> @llvm.vp.merge.v16i32(<16 x i1>, <16 x i32>, <16 x i32>, i32)

define <16 x i32> @vpmerge_vv_v16i32(<16 x i32> %va, <16 x i32> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v16i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m4, tu, mu
; CHECK-NEXT:    vmerge.vvm v12, v12, v8, v0
; CHECK-NEXT:    vmv4r.v v8, v12
; CHECK-NEXT:    ret
  %v = call <16 x i32> @llvm.vp.merge.v16i32(<16 x i1> %m, <16 x i32> %va, <16 x i32> %vb, i32 %evl)
  ret <16 x i32> %v
}

define <16 x i32> @vpmerge_vx_v16i32(i32 %a, <16 x i32> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vx_v16i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a1, e32, m4, tu, mu
; CHECK-NEXT:    vmerge.vxm v8, v8, a0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x i32> poison, i32 %a, i32 0
  %va = shufflevector <16 x i32> %elt.head, <16 x i32> poison, <16 x i32> zeroinitializer
  %v = call <16 x i32> @llvm.vp.merge.v16i32(<16 x i1> %m, <16 x i32> %va, <16 x i32> %vb, i32 %evl)
  ret <16 x i32> %v
}

define <16 x i32> @vpmerge_vi_v16i32(<16 x i32> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v16i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m4, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x i32> poison, i32 2, i32 0
  %va = shufflevector <16 x i32> %elt.head, <16 x i32> poison, <16 x i32> zeroinitializer
  %v = call <16 x i32> @llvm.vp.merge.v16i32(<16 x i1> %m, <16 x i32> %va, <16 x i32> %vb, i32 %evl)
  ret <16 x i32> %v
}

declare <2 x i64> @llvm.vp.merge.v2i64(<2 x i1>, <2 x i64>, <2 x i64>, i32)

define <2 x i64> @vpmerge_vv_v2i64(<2 x i64> %va, <2 x i64> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v2i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <2 x i64> @llvm.vp.merge.v2i64(<2 x i1> %m, <2 x i64> %va, <2 x i64> %vb, i32 %evl)
  ret <2 x i64> %v
}

define <2 x i64> @vpmerge_vx_v2i64(i64 %a, <2 x i64> %vb, <2 x i1> %m, i32 zeroext %evl) {
; RV32-LABEL: vpmerge_vx_v2i64:
; RV32:       # %bb.0:
; RV32-NEXT:    addi sp, sp, -16
; RV32-NEXT:    .cfi_def_cfa_offset 16
; RV32-NEXT:    sw a1, 12(sp)
; RV32-NEXT:    sw a0, 8(sp)
; RV32-NEXT:    addi a0, sp, 8
; RV32-NEXT:    vsetivli zero, 2, e64, m1, ta, mu
; RV32-NEXT:    vlse64.v v9, (a0), zero
; RV32-NEXT:    vsetvli zero, a2, e64, m1, tu, mu
; RV32-NEXT:    vmerge.vvm v8, v8, v9, v0
; RV32-NEXT:    addi sp, sp, 16
; RV32-NEXT:    ret
;
; RV64-LABEL: vpmerge_vx_v2i64:
; RV64:       # %bb.0:
; RV64-NEXT:    vsetvli zero, a1, e64, m1, tu, mu
; RV64-NEXT:    vmerge.vxm v8, v8, a0, v0
; RV64-NEXT:    ret
  %elt.head = insertelement <2 x i64> poison, i64 %a, i32 0
  %va = shufflevector <2 x i64> %elt.head, <2 x i64> poison, <2 x i32> zeroinitializer
  %v = call <2 x i64> @llvm.vp.merge.v2i64(<2 x i1> %m, <2 x i64> %va, <2 x i64> %vb, i32 %evl)
  ret <2 x i64> %v
}

define <2 x i64> @vpmerge_vi_v2i64(<2 x i64> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v2i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x i64> poison, i64 2, i32 0
  %va = shufflevector <2 x i64> %elt.head, <2 x i64> poison, <2 x i32> zeroinitializer
  %v = call <2 x i64> @llvm.vp.merge.v2i64(<2 x i1> %m, <2 x i64> %va, <2 x i64> %vb, i32 %evl)
  ret <2 x i64> %v
}

declare <4 x i64> @llvm.vp.merge.v4i64(<4 x i1>, <4 x i64>, <4 x i64>, i32)

define <4 x i64> @vpmerge_vv_v4i64(<4 x i64> %va, <4 x i64> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v4i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, tu, mu
; CHECK-NEXT:    vmerge.vvm v10, v10, v8, v0
; CHECK-NEXT:    vmv2r.v v8, v10
; CHECK-NEXT:    ret
  %v = call <4 x i64> @llvm.vp.merge.v4i64(<4 x i1> %m, <4 x i64> %va, <4 x i64> %vb, i32 %evl)
  ret <4 x i64> %v
}

define <4 x i64> @vpmerge_vx_v4i64(i64 %a, <4 x i64> %vb, <4 x i1> %m, i32 zeroext %evl) {
; RV32-LABEL: vpmerge_vx_v4i64:
; RV32:       # %bb.0:
; RV32-NEXT:    addi sp, sp, -16
; RV32-NEXT:    .cfi_def_cfa_offset 16
; RV32-NEXT:    sw a1, 12(sp)
; RV32-NEXT:    sw a0, 8(sp)
; RV32-NEXT:    addi a0, sp, 8
; RV32-NEXT:    vsetivli zero, 4, e64, m2, ta, mu
; RV32-NEXT:    vlse64.v v10, (a0), zero
; RV32-NEXT:    vsetvli zero, a2, e64, m2, tu, mu
; RV32-NEXT:    vmerge.vvm v8, v8, v10, v0
; RV32-NEXT:    addi sp, sp, 16
; RV32-NEXT:    ret
;
; RV64-LABEL: vpmerge_vx_v4i64:
; RV64:       # %bb.0:
; RV64-NEXT:    vsetvli zero, a1, e64, m2, tu, mu
; RV64-NEXT:    vmerge.vxm v8, v8, a0, v0
; RV64-NEXT:    ret
  %elt.head = insertelement <4 x i64> poison, i64 %a, i32 0
  %va = shufflevector <4 x i64> %elt.head, <4 x i64> poison, <4 x i32> zeroinitializer
  %v = call <4 x i64> @llvm.vp.merge.v4i64(<4 x i1> %m, <4 x i64> %va, <4 x i64> %vb, i32 %evl)
  ret <4 x i64> %v
}

define <4 x i64> @vpmerge_vi_v4i64(<4 x i64> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v4i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x i64> poison, i64 2, i32 0
  %va = shufflevector <4 x i64> %elt.head, <4 x i64> poison, <4 x i32> zeroinitializer
  %v = call <4 x i64> @llvm.vp.merge.v4i64(<4 x i1> %m, <4 x i64> %va, <4 x i64> %vb, i32 %evl)
  ret <4 x i64> %v
}

declare <8 x i64> @llvm.vp.merge.v8i64(<8 x i1>, <8 x i64>, <8 x i64>, i32)

define <8 x i64> @vpmerge_vv_v8i64(<8 x i64> %va, <8 x i64> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v8i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m4, tu, mu
; CHECK-NEXT:    vmerge.vvm v12, v12, v8, v0
; CHECK-NEXT:    vmv4r.v v8, v12
; CHECK-NEXT:    ret
  %v = call <8 x i64> @llvm.vp.merge.v8i64(<8 x i1> %m, <8 x i64> %va, <8 x i64> %vb, i32 %evl)
  ret <8 x i64> %v
}

define <8 x i64> @vpmerge_vx_v8i64(i64 %a, <8 x i64> %vb, <8 x i1> %m, i32 zeroext %evl) {
; RV32-LABEL: vpmerge_vx_v8i64:
; RV32:       # %bb.0:
; RV32-NEXT:    addi sp, sp, -16
; RV32-NEXT:    .cfi_def_cfa_offset 16
; RV32-NEXT:    sw a1, 12(sp)
; RV32-NEXT:    sw a0, 8(sp)
; RV32-NEXT:    addi a0, sp, 8
; RV32-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
; RV32-NEXT:    vlse64.v v12, (a0), zero
; RV32-NEXT:    vsetvli zero, a2, e64, m4, tu, mu
; RV32-NEXT:    vmerge.vvm v8, v8, v12, v0
; RV32-NEXT:    addi sp, sp, 16
; RV32-NEXT:    ret
;
; RV64-LABEL: vpmerge_vx_v8i64:
; RV64:       # %bb.0:
; RV64-NEXT:    vsetvli zero, a1, e64, m4, tu, mu
; RV64-NEXT:    vmerge.vxm v8, v8, a0, v0
; RV64-NEXT:    ret
  %elt.head = insertelement <8 x i64> poison, i64 %a, i32 0
  %va = shufflevector <8 x i64> %elt.head, <8 x i64> poison, <8 x i32> zeroinitializer
  %v = call <8 x i64> @llvm.vp.merge.v8i64(<8 x i1> %m, <8 x i64> %va, <8 x i64> %vb, i32 %evl)
  ret <8 x i64> %v
}

define <8 x i64> @vpmerge_vi_v8i64(<8 x i64> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v8i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m4, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x i64> poison, i64 2, i32 0
  %va = shufflevector <8 x i64> %elt.head, <8 x i64> poison, <8 x i32> zeroinitializer
  %v = call <8 x i64> @llvm.vp.merge.v8i64(<8 x i1> %m, <8 x i64> %va, <8 x i64> %vb, i32 %evl)
  ret <8 x i64> %v
}

declare <16 x i64> @llvm.vp.merge.v16i64(<16 x i1>, <16 x i64>, <16 x i64>, i32)

define <16 x i64> @vpmerge_vv_v16i64(<16 x i64> %va, <16 x i64> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v16i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m8, tu, mu
; CHECK-NEXT:    vmerge.vvm v16, v16, v8, v0
; CHECK-NEXT:    vmv8r.v v8, v16
; CHECK-NEXT:    ret
  %v = call <16 x i64> @llvm.vp.merge.v16i64(<16 x i1> %m, <16 x i64> %va, <16 x i64> %vb, i32 %evl)
  ret <16 x i64> %v
}

define <16 x i64> @vpmerge_vx_v16i64(i64 %a, <16 x i64> %vb, <16 x i1> %m, i32 zeroext %evl) {
; RV32-LABEL: vpmerge_vx_v16i64:
; RV32:       # %bb.0:
; RV32-NEXT:    addi sp, sp, -16
; RV32-NEXT:    .cfi_def_cfa_offset 16
; RV32-NEXT:    sw a1, 12(sp)
; RV32-NEXT:    sw a0, 8(sp)
; RV32-NEXT:    addi a0, sp, 8
; RV32-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
; RV32-NEXT:    vlse64.v v16, (a0), zero
; RV32-NEXT:    vsetvli zero, a2, e64, m8, tu, mu
; RV32-NEXT:    vmerge.vvm v8, v8, v16, v0
; RV32-NEXT:    addi sp, sp, 16
; RV32-NEXT:    ret
;
; RV64-LABEL: vpmerge_vx_v16i64:
; RV64:       # %bb.0:
; RV64-NEXT:    vsetvli zero, a1, e64, m8, tu, mu
; RV64-NEXT:    vmerge.vxm v8, v8, a0, v0
; RV64-NEXT:    ret
  %elt.head = insertelement <16 x i64> poison, i64 %a, i32 0
  %va = shufflevector <16 x i64> %elt.head, <16 x i64> poison, <16 x i32> zeroinitializer
  %v = call <16 x i64> @llvm.vp.merge.v16i64(<16 x i1> %m, <16 x i64> %va, <16 x i64> %vb, i32 %evl)
  ret <16 x i64> %v
}

define <16 x i64> @vpmerge_vi_v16i64(<16 x i64> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vi_v16i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m8, tu, mu
; CHECK-NEXT:    vmerge.vim v8, v8, 2, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x i64> poison, i64 2, i32 0
  %va = shufflevector <16 x i64> %elt.head, <16 x i64> poison, <16 x i32> zeroinitializer
  %v = call <16 x i64> @llvm.vp.merge.v16i64(<16 x i1> %m, <16 x i64> %va, <16 x i64> %vb, i32 %evl)
  ret <16 x i64> %v
}

declare <2 x half> @llvm.vp.merge.v2f16(<2 x i1>, <2 x half>, <2 x half>, i32)

define <2 x half> @vpmerge_vv_v2f16(<2 x half> %va, <2 x half> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v2f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <2 x half> @llvm.vp.merge.v2f16(<2 x i1> %m, <2 x half> %va, <2 x half> %vb, i32 %evl)
  ret <2 x half> %v
}

define <2 x half> @vpmerge_vf_v2f16(half %a, <2 x half> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v2f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x half> poison, half %a, i32 0
  %va = shufflevector <2 x half> %elt.head, <2 x half> poison, <2 x i32> zeroinitializer
  %v = call <2 x half> @llvm.vp.merge.v2f16(<2 x i1> %m, <2 x half> %va, <2 x half> %vb, i32 %evl)
  ret <2 x half> %v
}

declare <4 x half> @llvm.vp.merge.v4f16(<4 x i1>, <4 x half>, <4 x half>, i32)

define <4 x half> @vpmerge_vv_v4f16(<4 x half> %va, <4 x half> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v4f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x half> @llvm.vp.merge.v4f16(<4 x i1> %m, <4 x half> %va, <4 x half> %vb, i32 %evl)
  ret <4 x half> %v
}

define <4 x half> @vpmerge_vf_v4f16(half %a, <4 x half> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v4f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x half> poison, half %a, i32 0
  %va = shufflevector <4 x half> %elt.head, <4 x half> poison, <4 x i32> zeroinitializer
  %v = call <4 x half> @llvm.vp.merge.v4f16(<4 x i1> %m, <4 x half> %va, <4 x half> %vb, i32 %evl)
  ret <4 x half> %v
}

declare <8 x half> @llvm.vp.merge.v8f16(<8 x i1>, <8 x half>, <8 x half>, i32)

define <8 x half> @vpmerge_vv_v8f16(<8 x half> %va, <8 x half> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v8f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, m1, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <8 x half> @llvm.vp.merge.v8f16(<8 x i1> %m, <8 x half> %va, <8 x half> %vb, i32 %evl)
  ret <8 x half> %v
}

define <8 x half> @vpmerge_vf_v8f16(half %a, <8 x half> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v8f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, m1, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x half> poison, half %a, i32 0
  %va = shufflevector <8 x half> %elt.head, <8 x half> poison, <8 x i32> zeroinitializer
  %v = call <8 x half> @llvm.vp.merge.v8f16(<8 x i1> %m, <8 x half> %va, <8 x half> %vb, i32 %evl)
  ret <8 x half> %v
}

declare <16 x half> @llvm.vp.merge.v16f16(<16 x i1>, <16 x half>, <16 x half>, i32)

define <16 x half> @vpmerge_vv_v16f16(<16 x half> %va, <16 x half> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v16f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, m2, tu, mu
; CHECK-NEXT:    vmerge.vvm v10, v10, v8, v0
; CHECK-NEXT:    vmv2r.v v8, v10
; CHECK-NEXT:    ret
  %v = call <16 x half> @llvm.vp.merge.v16f16(<16 x i1> %m, <16 x half> %va, <16 x half> %vb, i32 %evl)
  ret <16 x half> %v
}

define <16 x half> @vpmerge_vf_v16f16(half %a, <16 x half> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v16f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, m2, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x half> poison, half %a, i32 0
  %va = shufflevector <16 x half> %elt.head, <16 x half> poison, <16 x i32> zeroinitializer
  %v = call <16 x half> @llvm.vp.merge.v16f16(<16 x i1> %m, <16 x half> %va, <16 x half> %vb, i32 %evl)
  ret <16 x half> %v
}

declare <2 x float> @llvm.vp.merge.v2f32(<2 x i1>, <2 x float>, <2 x float>, i32)

define <2 x float> @vpmerge_vv_v2f32(<2 x float> %va, <2 x float> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v2f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <2 x float> @llvm.vp.merge.v2f32(<2 x i1> %m, <2 x float> %va, <2 x float> %vb, i32 %evl)
  ret <2 x float> %v
}

define <2 x float> @vpmerge_vf_v2f32(float %a, <2 x float> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v2f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x float> poison, float %a, i32 0
  %va = shufflevector <2 x float> %elt.head, <2 x float> poison, <2 x i32> zeroinitializer
  %v = call <2 x float> @llvm.vp.merge.v2f32(<2 x i1> %m, <2 x float> %va, <2 x float> %vb, i32 %evl)
  ret <2 x float> %v
}

declare <4 x float> @llvm.vp.merge.v4f32(<4 x i1>, <4 x float>, <4 x float>, i32)

define <4 x float> @vpmerge_vv_v4f32(<4 x float> %va, <4 x float> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v4f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m1, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x float> @llvm.vp.merge.v4f32(<4 x i1> %m, <4 x float> %va, <4 x float> %vb, i32 %evl)
  ret <4 x float> %v
}

define <4 x float> @vpmerge_vf_v4f32(float %a, <4 x float> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v4f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m1, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x float> poison, float %a, i32 0
  %va = shufflevector <4 x float> %elt.head, <4 x float> poison, <4 x i32> zeroinitializer
  %v = call <4 x float> @llvm.vp.merge.v4f32(<4 x i1> %m, <4 x float> %va, <4 x float> %vb, i32 %evl)
  ret <4 x float> %v
}

declare <8 x float> @llvm.vp.merge.v8f32(<8 x i1>, <8 x float>, <8 x float>, i32)

define <8 x float> @vpmerge_vv_v8f32(<8 x float> %va, <8 x float> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v8f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m2, tu, mu
; CHECK-NEXT:    vmerge.vvm v10, v10, v8, v0
; CHECK-NEXT:    vmv2r.v v8, v10
; CHECK-NEXT:    ret
  %v = call <8 x float> @llvm.vp.merge.v8f32(<8 x i1> %m, <8 x float> %va, <8 x float> %vb, i32 %evl)
  ret <8 x float> %v
}

define <8 x float> @vpmerge_vf_v8f32(float %a, <8 x float> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v8f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m2, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x float> poison, float %a, i32 0
  %va = shufflevector <8 x float> %elt.head, <8 x float> poison, <8 x i32> zeroinitializer
  %v = call <8 x float> @llvm.vp.merge.v8f32(<8 x i1> %m, <8 x float> %va, <8 x float> %vb, i32 %evl)
  ret <8 x float> %v
}

declare <16 x float> @llvm.vp.merge.v16f32(<16 x i1>, <16 x float>, <16 x float>, i32)

define <16 x float> @vpmerge_vv_v16f32(<16 x float> %va, <16 x float> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v16f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m4, tu, mu
; CHECK-NEXT:    vmerge.vvm v12, v12, v8, v0
; CHECK-NEXT:    vmv4r.v v8, v12
; CHECK-NEXT:    ret
  %v = call <16 x float> @llvm.vp.merge.v16f32(<16 x i1> %m, <16 x float> %va, <16 x float> %vb, i32 %evl)
  ret <16 x float> %v
}

define <16 x float> @vpmerge_vf_v16f32(float %a, <16 x float> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v16f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m4, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x float> poison, float %a, i32 0
  %va = shufflevector <16 x float> %elt.head, <16 x float> poison, <16 x i32> zeroinitializer
  %v = call <16 x float> @llvm.vp.merge.v16f32(<16 x i1> %m, <16 x float> %va, <16 x float> %vb, i32 %evl)
  ret <16 x float> %v
}

declare <2 x double> @llvm.vp.merge.v2f64(<2 x i1>, <2 x double>, <2 x double>, i32)

define <2 x double> @vpmerge_vv_v2f64(<2 x double> %va, <2 x double> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v2f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
; CHECK-NEXT:    vmerge.vvm v9, v9, v8, v0
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <2 x double> @llvm.vp.merge.v2f64(<2 x i1> %m, <2 x double> %va, <2 x double> %vb, i32 %evl)
  ret <2 x double> %v
}

define <2 x double> @vpmerge_vf_v2f64(double %a, <2 x double> %vb, <2 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v2f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <2 x double> poison, double %a, i32 0
  %va = shufflevector <2 x double> %elt.head, <2 x double> poison, <2 x i32> zeroinitializer
  %v = call <2 x double> @llvm.vp.merge.v2f64(<2 x i1> %m, <2 x double> %va, <2 x double> %vb, i32 %evl)
  ret <2 x double> %v
}

declare <4 x double> @llvm.vp.merge.v4f64(<4 x i1>, <4 x double>, <4 x double>, i32)

define <4 x double> @vpmerge_vv_v4f64(<4 x double> %va, <4 x double> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v4f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, tu, mu
; CHECK-NEXT:    vmerge.vvm v10, v10, v8, v0
; CHECK-NEXT:    vmv2r.v v8, v10
; CHECK-NEXT:    ret
  %v = call <4 x double> @llvm.vp.merge.v4f64(<4 x i1> %m, <4 x double> %va, <4 x double> %vb, i32 %evl)
  ret <4 x double> %v
}

define <4 x double> @vpmerge_vf_v4f64(double %a, <4 x double> %vb, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v4f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <4 x double> poison, double %a, i32 0
  %va = shufflevector <4 x double> %elt.head, <4 x double> poison, <4 x i32> zeroinitializer
  %v = call <4 x double> @llvm.vp.merge.v4f64(<4 x i1> %m, <4 x double> %va, <4 x double> %vb, i32 %evl)
  ret <4 x double> %v
}

declare <8 x double> @llvm.vp.merge.v8f64(<8 x i1>, <8 x double>, <8 x double>, i32)

define <8 x double> @vpmerge_vv_v8f64(<8 x double> %va, <8 x double> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v8f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m4, tu, mu
; CHECK-NEXT:    vmerge.vvm v12, v12, v8, v0
; CHECK-NEXT:    vmv4r.v v8, v12
; CHECK-NEXT:    ret
  %v = call <8 x double> @llvm.vp.merge.v8f64(<8 x i1> %m, <8 x double> %va, <8 x double> %vb, i32 %evl)
  ret <8 x double> %v
}

define <8 x double> @vpmerge_vf_v8f64(double %a, <8 x double> %vb, <8 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v8f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m4, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <8 x double> poison, double %a, i32 0
  %va = shufflevector <8 x double> %elt.head, <8 x double> poison, <8 x i32> zeroinitializer
  %v = call <8 x double> @llvm.vp.merge.v8f64(<8 x i1> %m, <8 x double> %va, <8 x double> %vb, i32 %evl)
  ret <8 x double> %v
}

declare <16 x double> @llvm.vp.merge.v16f64(<16 x i1>, <16 x double>, <16 x double>, i32)

define <16 x double> @vpmerge_vv_v16f64(<16 x double> %va, <16 x double> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vv_v16f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m8, tu, mu
; CHECK-NEXT:    vmerge.vvm v16, v16, v8, v0
; CHECK-NEXT:    vmv8r.v v8, v16
; CHECK-NEXT:    ret
  %v = call <16 x double> @llvm.vp.merge.v16f64(<16 x i1> %m, <16 x double> %va, <16 x double> %vb, i32 %evl)
  ret <16 x double> %v
}

define <16 x double> @vpmerge_vf_v16f64(double %a, <16 x double> %vb, <16 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v16f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m8, tu, mu
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <16 x double> poison, double %a, i32 0
  %va = shufflevector <16 x double> %elt.head, <16 x double> poison, <16 x i32> zeroinitializer
  %v = call <16 x double> @llvm.vp.merge.v16f64(<16 x i1> %m, <16 x double> %va, <16 x double> %vb, i32 %evl)
  ret <16 x double> %v
}

declare <32 x double> @llvm.vp.merge.v32f64(<32 x i1>, <32 x double>, <32 x double>, i32)

define <32 x double> @vpmerge_vv_v32f64(<32 x double> %va, <32 x double> %vb, <32 x i1> %m, i32 zeroext %evl) {
; RV32-LABEL: vpmerge_vv_v32f64:
; RV32:       # %bb.0:
; RV32-NEXT:    addi sp, sp, -16
; RV32-NEXT:    .cfi_def_cfa_offset 16
; RV32-NEXT:    csrr a1, vlenb
; RV32-NEXT:    li a3, 24
; RV32-NEXT:    mul a1, a1, a3
; RV32-NEXT:    sub sp, sp, a1
; RV32-NEXT:    addi a1, a0, 128
; RV32-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
; RV32-NEXT:    vle64.v v24, (a1)
; RV32-NEXT:    csrr a1, vlenb
; RV32-NEXT:    slli a1, a1, 3
; RV32-NEXT:    add a1, sp, a1
; RV32-NEXT:    addi a1, a1, 16
; RV32-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
; RV32-NEXT:    addi a3, a2, -16
; RV32-NEXT:    vmv1r.v v1, v0
; RV32-NEXT:    csrr a1, vlenb
; RV32-NEXT:    slli a1, a1, 4
; RV32-NEXT:    add a1, sp, a1
; RV32-NEXT:    addi a1, a1, 16
; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
; RV32-NEXT:    addi a1, sp, 16
; RV32-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
; RV32-NEXT:    li a1, 0
; RV32-NEXT:    bltu a2, a3, .LBB79_2
; RV32-NEXT:  # %bb.1:
; RV32-NEXT:    mv a1, a3
; RV32-NEXT:  .LBB79_2:
; RV32-NEXT:    vle64.v v8, (a0)
; RV32-NEXT:    vsetivli zero, 2, e8, mf4, ta, mu
; RV32-NEXT:    vslidedown.vi v0, v1, 2
; RV32-NEXT:    vsetvli zero, a1, e64, m8, tu, mu
; RV32-NEXT:    li a0, 16
; RV32-NEXT:    csrr a1, vlenb
; RV32-NEXT:    slli a1, a1, 4
; RV32-NEXT:    add a1, sp, a1
; RV32-NEXT:    addi a1, a1, 16
; RV32-NEXT:    vl8re8.v v24, (a1) # Unknown-size Folded Reload
; RV32-NEXT:    csrr a1, vlenb
; RV32-NEXT:    slli a1, a1, 3
; RV32-NEXT:    add a1, sp, a1
; RV32-NEXT:    addi a1, a1, 16
; RV32-NEXT:    vl8re8.v v16, (a1) # Unknown-size Folded Reload
; RV32-NEXT:    vmerge.vvm v16, v16, v24, v0
; RV32-NEXT:    bltu a2, a0, .LBB79_4
; RV32-NEXT:  # %bb.3:
; RV32-NEXT:    li a2, 16
; RV32-NEXT:  .LBB79_4:
; RV32-NEXT:    vsetvli zero, a2, e64, m8, tu, mu
; RV32-NEXT:    vmv1r.v v0, v1
; RV32-NEXT:    addi a0, sp, 16
; RV32-NEXT:    vl8re8.v v24, (a0) # Unknown-size Folded Reload
; RV32-NEXT:    vmerge.vvm v8, v8, v24, v0
; RV32-NEXT:    csrr a0, vlenb
; RV32-NEXT:    li a1, 24
; RV32-NEXT:    mul a0, a0, a1
; RV32-NEXT:    add sp, sp, a0
; RV32-NEXT:    addi sp, sp, 16
; RV32-NEXT:    ret
;
; RV64-LABEL: vpmerge_vv_v32f64:
; RV64:       # %bb.0:
; RV64-NEXT:    addi sp, sp, -16
; RV64-NEXT:    .cfi_def_cfa_offset 16
; RV64-NEXT:    csrr a1, vlenb
; RV64-NEXT:    slli a1, a1, 4
; RV64-NEXT:    sub sp, sp, a1
; RV64-NEXT:    addi a1, a0, 128
; RV64-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
; RV64-NEXT:    vle64.v v24, (a1)
; RV64-NEXT:    addi a3, a2, -16
; RV64-NEXT:    addi a1, sp, 16
; RV64-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
; RV64-NEXT:    vmv1r.v v1, v0
; RV64-NEXT:    csrr a1, vlenb
; RV64-NEXT:    slli a1, a1, 3
; RV64-NEXT:    add a1, sp, a1
; RV64-NEXT:    addi a1, a1, 16
; RV64-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
; RV64-NEXT:    li a1, 0
; RV64-NEXT:    bltu a2, a3, .LBB79_2
; RV64-NEXT:  # %bb.1:
; RV64-NEXT:    mv a1, a3
; RV64-NEXT:  .LBB79_2:
; RV64-NEXT:    vle64.v v8, (a0)
; RV64-NEXT:    vsetivli zero, 2, e8, mf4, ta, mu
; RV64-NEXT:    vslidedown.vi v0, v1, 2
; RV64-NEXT:    vsetvli zero, a1, e64, m8, tu, mu
; RV64-NEXT:    li a0, 16
; RV64-NEXT:    addi a1, sp, 16
; RV64-NEXT:    vl8re8.v v16, (a1) # Unknown-size Folded Reload
; RV64-NEXT:    vmerge.vvm v24, v24, v16, v0
; RV64-NEXT:    bltu a2, a0, .LBB79_4
; RV64-NEXT:  # %bb.3:
; RV64-NEXT:    li a2, 16
; RV64-NEXT:  .LBB79_4:
; RV64-NEXT:    vsetvli zero, a2, e64, m8, tu, mu
; RV64-NEXT:    vmv1r.v v0, v1
; RV64-NEXT:    csrr a0, vlenb
; RV64-NEXT:    slli a0, a0, 3
; RV64-NEXT:    add a0, sp, a0
; RV64-NEXT:    addi a0, a0, 16
; RV64-NEXT:    vl8re8.v v16, (a0) # Unknown-size Folded Reload
; RV64-NEXT:    vmerge.vvm v8, v8, v16, v0
; RV64-NEXT:    vmv8r.v v16, v24
; RV64-NEXT:    csrr a0, vlenb
; RV64-NEXT:    slli a0, a0, 4
; RV64-NEXT:    add sp, sp, a0
; RV64-NEXT:    addi sp, sp, 16
; RV64-NEXT:    ret
  %v = call <32 x double> @llvm.vp.merge.v32f64(<32 x i1> %m, <32 x double> %va, <32 x double> %vb, i32 %evl)
  ret <32 x double> %v
}

define <32 x double> @vpmerge_vf_v32f64(double %a, <32 x double> %vb, <32 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vpmerge_vf_v32f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    addi a2, a0, -16
; CHECK-NEXT:    vmv1r.v v24, v0
; CHECK-NEXT:    li a1, 0
; CHECK-NEXT:    bltu a0, a2, .LBB80_2
; CHECK-NEXT:  # %bb.1:
; CHECK-NEXT:    mv a1, a2
; CHECK-NEXT:  .LBB80_2:
; CHECK-NEXT:    vsetivli zero, 2, e8, mf4, ta, mu
; CHECK-NEXT:    vslidedown.vi v0, v24, 2
; CHECK-NEXT:    vsetvli zero, a1, e64, m8, tu, mu
; CHECK-NEXT:    li a1, 16
; CHECK-NEXT:    vfmerge.vfm v16, v16, fa0, v0
; CHECK-NEXT:    bltu a0, a1, .LBB80_4
; CHECK-NEXT:  # %bb.3:
; CHECK-NEXT:    li a0, 16
; CHECK-NEXT:  .LBB80_4:
; CHECK-NEXT:    vsetvli zero, a0, e64, m8, tu, mu
; CHECK-NEXT:    vmv1r.v v0, v24
; CHECK-NEXT:    vfmerge.vfm v8, v8, fa0, v0
; CHECK-NEXT:    ret
  %elt.head = insertelement <32 x double> poison, double %a, i32 0
  %va = shufflevector <32 x double> %elt.head, <32 x double> poison, <32 x i32> zeroinitializer
  %v = call <32 x double> @llvm.vp.merge.v32f64(<32 x i1> %m, <32 x double> %va, <32 x double> %vb, i32 %evl)
  ret <32 x double> %v
}