; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -mattr=+m,+experimental-v,+f,+d,+experimental-zfh -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32-BITS-UNKNOWN
; RUN: llc -mtriple=riscv32 -mattr=+m,+experimental-v,+f,+d,+experimental-zfh -riscv-v-vector-bits-max=256 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32-BITS-256
; RUN: llc -mtriple=riscv32 -mattr=+m,+experimental-v,+f,+d,+experimental-zfh -riscv-v-vector-bits-max=512 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32-BITS-512
; RUN: llc -mtriple=riscv64 -mattr=+m,+experimental-v,+f,+d,+experimental-zfh -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64-BITS-UNKNOWN
; RUN: llc -mtriple=riscv64 -mattr=+m,+experimental-v,+f,+d,+experimental-zfh -riscv-v-vector-bits-max=256 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64-BITS-256
; RUN: llc -mtriple=riscv64 -mattr=+m,+experimental-v,+f,+d,+experimental-zfh -riscv-v-vector-bits-max=512 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64-BITS-512

;
; VECTOR_REVERSE - masks
;
; FIXME: Support for mask vectors

;define <vscale x 2 x i1> @reverse_nxv2i1(<vscale x 2 x i1> %a) {
;  %res = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> %a)
;  ret <vscale x 2 x i1> %res
;}
;
;define <vscale x 4 x i1> @reverse_nxv4i1(<vscale x 4 x i1> %a) {
;  %res = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
;  ret <vscale x 4 x i1> %res
;}
;
;define <vscale x 8 x i1> @reverse_nxv8i1(<vscale x 8 x i1> %a) {
;  %res = call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> %a)
;  ret <vscale x 8 x i1> %res
;}
;
;define <vscale x 16 x i1> @reverse_nxv16i1(<vscale x 16 x i1> %a) {
;  %res = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> %a)
;  ret <vscale x 16 x i1> %res
;}
;
;define <vscale x 32 x i1> @reverse_nxv32i1(<vscale x 32 x i1> %a) {
;  %res = call <vscale x 32 x i1> @llvm.experimental.vector.reverse.nxv32i1(<vscale x 32 x i1> %a)
;  ret <vscale x 32 x i1> %res
;}
;
;define <vscale x 64 x i1> @reverse_nxv64i1(<vscale x 64 x i1> %a) {
;  %res = call <vscale x 64 x i1> @llvm.experimental.vector.reverse.nxv64i1(<vscale x 64 x i1> %a)
;  ret <vscale x 64 x i1> %res
;}


;
; VECTOR_REVERSE - integer
;

define <vscale x 1 x i8> @reverse_nxv1i8(<vscale x 1 x i8> %a) {
; RV32-BITS-UNKNOWN-LABEL: reverse_nxv1i8:
; RV32-BITS-UNKNOWN:       # %bb.0:
; RV32-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV32-BITS-UNKNOWN-NEXT:    srli a0, a0, 3
; RV32-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV32-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, mf4, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
; RV32-BITS-UNKNOWN-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf8, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
; RV32-BITS-UNKNOWN-NEXT:    vmv1r.v v8, v9
; RV32-BITS-UNKNOWN-NEXT:    ret
;
; RV32-BITS-256-LABEL: reverse_nxv1i8:
; RV32-BITS-256:       # %bb.0:
; RV32-BITS-256-NEXT:    csrr a0, vlenb
; RV32-BITS-256-NEXT:    srli a0, a0, 3
; RV32-BITS-256-NEXT:    addi a0, a0, -1
; RV32-BITS-256-NEXT:    vsetvli a1, zero, e8, mf8, ta, mu
; RV32-BITS-256-NEXT:    vid.v v9
; RV32-BITS-256-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-256-NEXT:    vrgather.vv v9, v8, v10
; RV32-BITS-256-NEXT:    vmv1r.v v8, v9
; RV32-BITS-256-NEXT:    ret
;
; RV32-BITS-512-LABEL: reverse_nxv1i8:
; RV32-BITS-512:       # %bb.0:
; RV32-BITS-512-NEXT:    csrr a0, vlenb
; RV32-BITS-512-NEXT:    srli a0, a0, 3
; RV32-BITS-512-NEXT:    addi a0, a0, -1
; RV32-BITS-512-NEXT:    vsetvli a1, zero, e8, mf8, ta, mu
; RV32-BITS-512-NEXT:    vid.v v9
; RV32-BITS-512-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-512-NEXT:    vrgather.vv v9, v8, v10
; RV32-BITS-512-NEXT:    vmv1r.v v8, v9
; RV32-BITS-512-NEXT:    ret
;
; RV64-BITS-UNKNOWN-LABEL: reverse_nxv1i8:
; RV64-BITS-UNKNOWN:       # %bb.0:
; RV64-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV64-BITS-UNKNOWN-NEXT:    srli a0, a0, 3
; RV64-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV64-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, mf4, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
; RV64-BITS-UNKNOWN-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf8, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
; RV64-BITS-UNKNOWN-NEXT:    vmv1r.v v8, v9
; RV64-BITS-UNKNOWN-NEXT:    ret
;
; RV64-BITS-256-LABEL: reverse_nxv1i8:
; RV64-BITS-256:       # %bb.0:
; RV64-BITS-256-NEXT:    csrr a0, vlenb
; RV64-BITS-256-NEXT:    srli a0, a0, 3
; RV64-BITS-256-NEXT:    addi a0, a0, -1
; RV64-BITS-256-NEXT:    vsetvli a1, zero, e8, mf8, ta, mu
; RV64-BITS-256-NEXT:    vid.v v9
; RV64-BITS-256-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-256-NEXT:    vrgather.vv v9, v8, v10
; RV64-BITS-256-NEXT:    vmv1r.v v8, v9
; RV64-BITS-256-NEXT:    ret
;
; RV64-BITS-512-LABEL: reverse_nxv1i8:
; RV64-BITS-512:       # %bb.0:
; RV64-BITS-512-NEXT:    csrr a0, vlenb
; RV64-BITS-512-NEXT:    srli a0, a0, 3
; RV64-BITS-512-NEXT:    addi a0, a0, -1
; RV64-BITS-512-NEXT:    vsetvli a1, zero, e8, mf8, ta, mu
; RV64-BITS-512-NEXT:    vid.v v9
; RV64-BITS-512-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-512-NEXT:    vrgather.vv v9, v8, v10
; RV64-BITS-512-NEXT:    vmv1r.v v8, v9
; RV64-BITS-512-NEXT:    ret
  %res = call <vscale x 1 x i8> @llvm.experimental.vector.reverse.nxv1i8(<vscale x 1 x i8> %a)
  ret <vscale x 1 x i8> %res
}

define <vscale x 2 x i8> @reverse_nxv2i8(<vscale x 2 x i8> %a) {
; RV32-BITS-UNKNOWN-LABEL: reverse_nxv2i8:
; RV32-BITS-UNKNOWN:       # %bb.0:
; RV32-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV32-BITS-UNKNOWN-NEXT:    srli a0, a0, 2
; RV32-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV32-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, mf2, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
; RV32-BITS-UNKNOWN-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf4, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
; RV32-BITS-UNKNOWN-NEXT:    vmv1r.v v8, v9
; RV32-BITS-UNKNOWN-NEXT:    ret
;
; RV32-BITS-256-LABEL: reverse_nxv2i8:
; RV32-BITS-256:       # %bb.0:
; RV32-BITS-256-NEXT:    csrr a0, vlenb
; RV32-BITS-256-NEXT:    srli a0, a0, 2
; RV32-BITS-256-NEXT:    addi a0, a0, -1
; RV32-BITS-256-NEXT:    vsetvli a1, zero, e8, mf4, ta, mu
; RV32-BITS-256-NEXT:    vid.v v9
; RV32-BITS-256-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-256-NEXT:    vrgather.vv v9, v8, v10
; RV32-BITS-256-NEXT:    vmv1r.v v8, v9
; RV32-BITS-256-NEXT:    ret
;
; RV32-BITS-512-LABEL: reverse_nxv2i8:
; RV32-BITS-512:       # %bb.0:
; RV32-BITS-512-NEXT:    csrr a0, vlenb
; RV32-BITS-512-NEXT:    srli a0, a0, 2
; RV32-BITS-512-NEXT:    addi a0, a0, -1
; RV32-BITS-512-NEXT:    vsetvli a1, zero, e8, mf4, ta, mu
; RV32-BITS-512-NEXT:    vid.v v9
; RV32-BITS-512-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-512-NEXT:    vrgather.vv v9, v8, v10
; RV32-BITS-512-NEXT:    vmv1r.v v8, v9
; RV32-BITS-512-NEXT:    ret
;
; RV64-BITS-UNKNOWN-LABEL: reverse_nxv2i8:
; RV64-BITS-UNKNOWN:       # %bb.0:
; RV64-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV64-BITS-UNKNOWN-NEXT:    srli a0, a0, 2
; RV64-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV64-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, mf2, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
; RV64-BITS-UNKNOWN-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf4, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
; RV64-BITS-UNKNOWN-NEXT:    vmv1r.v v8, v9
; RV64-BITS-UNKNOWN-NEXT:    ret
;
; RV64-BITS-256-LABEL: reverse_nxv2i8:
; RV64-BITS-256:       # %bb.0:
; RV64-BITS-256-NEXT:    csrr a0, vlenb
; RV64-BITS-256-NEXT:    srli a0, a0, 2
; RV64-BITS-256-NEXT:    addi a0, a0, -1
; RV64-BITS-256-NEXT:    vsetvli a1, zero, e8, mf4, ta, mu
; RV64-BITS-256-NEXT:    vid.v v9
; RV64-BITS-256-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-256-NEXT:    vrgather.vv v9, v8, v10
; RV64-BITS-256-NEXT:    vmv1r.v v8, v9
; RV64-BITS-256-NEXT:    ret
;
; RV64-BITS-512-LABEL: reverse_nxv2i8:
; RV64-BITS-512:       # %bb.0:
; RV64-BITS-512-NEXT:    csrr a0, vlenb
; RV64-BITS-512-NEXT:    srli a0, a0, 2
; RV64-BITS-512-NEXT:    addi a0, a0, -1
; RV64-BITS-512-NEXT:    vsetvli a1, zero, e8, mf4, ta, mu
; RV64-BITS-512-NEXT:    vid.v v9
; RV64-BITS-512-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-512-NEXT:    vrgather.vv v9, v8, v10
; RV64-BITS-512-NEXT:    vmv1r.v v8, v9
; RV64-BITS-512-NEXT:    ret
  %res = call <vscale x 2 x i8> @llvm.experimental.vector.reverse.nxv2i8(<vscale x 2 x i8> %a)
  ret <vscale x 2 x i8> %res
}

define <vscale x 4 x i8> @reverse_nxv4i8(<vscale x 4 x i8> %a) {
; RV32-BITS-UNKNOWN-LABEL: reverse_nxv4i8:
; RV32-BITS-UNKNOWN:       # %bb.0:
; RV32-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV32-BITS-UNKNOWN-NEXT:    srli a0, a0, 1
; RV32-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV32-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m1, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
; RV32-BITS-UNKNOWN-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf2, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
; RV32-BITS-UNKNOWN-NEXT:    vmv1r.v v8, v9
; RV32-BITS-UNKNOWN-NEXT:    ret
;
; RV32-BITS-256-LABEL: reverse_nxv4i8:
; RV32-BITS-256:       # %bb.0:
; RV32-BITS-256-NEXT:    csrr a0, vlenb
; RV32-BITS-256-NEXT:    srli a0, a0, 1
; RV32-BITS-256-NEXT:    addi a0, a0, -1
; RV32-BITS-256-NEXT:    vsetvli a1, zero, e8, mf2, ta, mu
; RV32-BITS-256-NEXT:    vid.v v9
; RV32-BITS-256-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-256-NEXT:    vrgather.vv v9, v8, v10
; RV32-BITS-256-NEXT:    vmv1r.v v8, v9
; RV32-BITS-256-NEXT:    ret
;
; RV32-BITS-512-LABEL: reverse_nxv4i8:
; RV32-BITS-512:       # %bb.0:
; RV32-BITS-512-NEXT:    csrr a0, vlenb
; RV32-BITS-512-NEXT:    srli a0, a0, 1
; RV32-BITS-512-NEXT:    addi a0, a0, -1
; RV32-BITS-512-NEXT:    vsetvli a1, zero, e8, mf2, ta, mu
; RV32-BITS-512-NEXT:    vid.v v9
; RV32-BITS-512-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-512-NEXT:    vrgather.vv v9, v8, v10
; RV32-BITS-512-NEXT:    vmv1r.v v8, v9
; RV32-BITS-512-NEXT:    ret
;
; RV64-BITS-UNKNOWN-LABEL: reverse_nxv4i8:
; RV64-BITS-UNKNOWN:       # %bb.0:
; RV64-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV64-BITS-UNKNOWN-NEXT:    srli a0, a0, 1
; RV64-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV64-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m1, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
; RV64-BITS-UNKNOWN-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf2, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
; RV64-BITS-UNKNOWN-NEXT:    vmv1r.v v8, v9
; RV64-BITS-UNKNOWN-NEXT:    ret
;
; RV64-BITS-256-LABEL: reverse_nxv4i8:
; RV64-BITS-256:       # %bb.0:
; RV64-BITS-256-NEXT:    csrr a0, vlenb
; RV64-BITS-256-NEXT:    srli a0, a0, 1
; RV64-BITS-256-NEXT:    addi a0, a0, -1
; RV64-BITS-256-NEXT:    vsetvli a1, zero, e8, mf2, ta, mu
; RV64-BITS-256-NEXT:    vid.v v9
; RV64-BITS-256-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-256-NEXT:    vrgather.vv v9, v8, v10
; RV64-BITS-256-NEXT:    vmv1r.v v8, v9
; RV64-BITS-256-NEXT:    ret
;
; RV64-BITS-512-LABEL: reverse_nxv4i8:
; RV64-BITS-512:       # %bb.0:
; RV64-BITS-512-NEXT:    csrr a0, vlenb
; RV64-BITS-512-NEXT:    srli a0, a0, 1
; RV64-BITS-512-NEXT:    addi a0, a0, -1
; RV64-BITS-512-NEXT:    vsetvli a1, zero, e8, mf2, ta, mu
; RV64-BITS-512-NEXT:    vid.v v9
; RV64-BITS-512-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-512-NEXT:    vrgather.vv v9, v8, v10
; RV64-BITS-512-NEXT:    vmv1r.v v8, v9
; RV64-BITS-512-NEXT:    ret
  %res = call <vscale x 4 x i8> @llvm.experimental.vector.reverse.nxv4i8(<vscale x 4 x i8> %a)
  ret <vscale x 4 x i8> %res
}

define <vscale x 8 x i8> @reverse_nxv8i8(<vscale x 8 x i8> %a) {
; RV32-BITS-UNKNOWN-LABEL: reverse_nxv8i8:
; RV32-BITS-UNKNOWN:       # %bb.0:
; RV32-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV32-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV32-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m2, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vid.v v10
; RV32-BITS-UNKNOWN-NEXT:    vrsub.vx v10, v10, a0
; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, m1, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
; RV32-BITS-UNKNOWN-NEXT:    vmv.v.v v8, v9
; RV32-BITS-UNKNOWN-NEXT:    ret
;
; RV32-BITS-256-LABEL: reverse_nxv8i8:
; RV32-BITS-256:       # %bb.0:
; RV32-BITS-256-NEXT:    csrr a0, vlenb
; RV32-BITS-256-NEXT:    addi a0, a0, -1
; RV32-BITS-256-NEXT:    vsetvli a1, zero, e8, m1, ta, mu
; RV32-BITS-256-NEXT:    vid.v v9
; RV32-BITS-256-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-256-NEXT:    vrgather.vv v9, v8, v10
; RV32-BITS-256-NEXT:    vmv.v.v v8, v9
; RV32-BITS-256-NEXT:    ret
;
; RV32-BITS-512-LABEL: reverse_nxv8i8:
; RV32-BITS-512:       # %bb.0:
; RV32-BITS-512-NEXT:    csrr a0, vlenb
; RV32-BITS-512-NEXT:    addi a0, a0, -1
; RV32-BITS-512-NEXT:    vsetvli a1, zero, e8, m1, ta, mu
; RV32-BITS-512-NEXT:    vid.v v9
; RV32-BITS-512-NEXT:    vrsub.vx v10, v9, a0
; RV32-BITS-512-NEXT:    vrgather.vv v9, v8, v10
; RV32-BITS-512-NEXT:    vmv.v.v v8, v9
; RV32-BITS-512-NEXT:    ret
;
; RV64-BITS-UNKNOWN-LABEL: reverse_nxv8i8:
; RV64-BITS-UNKNOWN:       # %bb.0:
; RV64-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV64-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV64-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m2, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vid.v v10
; RV64-BITS-UNKNOWN-NEXT:    vrsub.vx v10, v10, a0
; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, m1, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
; RV64-BITS-UNKNOWN-NEXT:    vmv.v.v v8, v9
; RV64-BITS-UNKNOWN-NEXT:    ret
;
; RV64-BITS-256-LABEL: reverse_nxv8i8:
; RV64-BITS-256:       # %bb.0:
; RV64-BITS-256-NEXT:    csrr a0, vlenb
; RV64-BITS-256-NEXT:    addi a0, a0, -1
; RV64-BITS-256-NEXT:    vsetvli a1, zero, e8, m1, ta, mu
; RV64-BITS-256-NEXT:    vid.v v9
; RV64-BITS-256-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-256-NEXT:    vrgather.vv v9, v8, v10
; RV64-BITS-256-NEXT:    vmv.v.v v8, v9
; RV64-BITS-256-NEXT:    ret
;
; RV64-BITS-512-LABEL: reverse_nxv8i8:
; RV64-BITS-512:       # %bb.0:
; RV64-BITS-512-NEXT:    csrr a0, vlenb
; RV64-BITS-512-NEXT:    addi a0, a0, -1
; RV64-BITS-512-NEXT:    vsetvli a1, zero, e8, m1, ta, mu
; RV64-BITS-512-NEXT:    vid.v v9
; RV64-BITS-512-NEXT:    vrsub.vx v10, v9, a0
; RV64-BITS-512-NEXT:    vrgather.vv v9, v8, v10
; RV64-BITS-512-NEXT:    vmv.v.v v8, v9
; RV64-BITS-512-NEXT:    ret
  %res = call <vscale x 8 x i8> @llvm.experimental.vector.reverse.nxv8i8(<vscale x 8 x i8> %a)
  ret <vscale x 8 x i8> %res
}

define <vscale x 16 x i8> @reverse_nxv16i8(<vscale x 16 x i8> %a) {
; RV32-BITS-UNKNOWN-LABEL: reverse_nxv16i8:
; RV32-BITS-UNKNOWN:       # %bb.0:
; RV32-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV32-BITS-UNKNOWN-NEXT:    slli a0, a0, 1
; RV32-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV32-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m4, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vid.v v12
; RV32-BITS-UNKNOWN-NEXT:    vrsub.vx v12, v12, a0
; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, m2, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v12
; RV32-BITS-UNKNOWN-NEXT:    vmv.v.v v8, v10
; RV32-BITS-UNKNOWN-NEXT:    ret
;
; RV32-BITS-256-LABEL: reverse_nxv16i8:
; RV32-BITS-256:       # %bb.0:
; RV32-BITS-256-NEXT:    csrr a0, vlenb
; RV32-BITS-256-NEXT:    slli a0, a0, 1
; RV32-BITS-256-NEXT:    addi a0, a0, -1
; RV32-BITS-256-NEXT:    vsetvli a1, zero, e8, m2, ta, mu
; RV32-BITS-256-NEXT:    vid.v v10
; RV32-BITS-256-NEXT:    vrsub.vx v12, v10, a0
; RV32-BITS-256-NEXT:    vrgather.vv v10, v8, v12
; RV32-BITS-256-NEXT:    vmv.v.v v8, v10
; RV32-BITS-256-NEXT:    ret
;
; RV32-BITS-512-LABEL: reverse_nxv16i8:
; RV32-BITS-512:       # %bb.0:
; RV32-BITS-512-NEXT:    csrr a0, vlenb
; RV32-BITS-512-NEXT:    slli a0, a0, 1
; RV32-BITS-512-NEXT:    addi a0, a0, -1
; RV32-BITS-512-NEXT:    vsetvli a1, zero, e8, m2, ta, mu
; RV32-BITS-512-NEXT:    vid.v v10
; RV32-BITS-512-NEXT:    vrsub.vx v12, v10, a0
; RV32-BITS-512-NEXT:    vrgather.vv v10, v8, v12
; RV32-BITS-512-NEXT:    vmv.v.v v8, v10
; RV32-BITS-512-NEXT:    ret
;
; RV64-BITS-UNKNOWN-LABEL: reverse_nxv16i8:
; RV64-BITS-UNKNOWN:       # %bb.0:
; RV64-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV64-BITS-UNKNOWN-NEXT:    slli a0, a0, 1
; RV64-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV64-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m4, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vid.v v12
; RV64-BITS-UNKNOWN-NEXT:    vrsub.vx v12, v12, a0
; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, m2, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v12
; RV64-BITS-UNKNOWN-NEXT:    vmv.v.v v8, v10
; RV64-BITS-UNKNOWN-NEXT:    ret
;
; RV64-BITS-256-LABEL: reverse_nxv16i8:
; RV64-BITS-256:       # %bb.0:
; RV64-BITS-256-NEXT:    csrr a0, vlenb
; RV64-BITS-256-NEXT:    slli a0, a0, 1
; RV64-BITS-256-NEXT:    addi a0, a0, -1
; RV64-BITS-256-NEXT:    vsetvli a1, zero, e8, m2, ta, mu
; RV64-BITS-256-NEXT:    vid.v v10
; RV64-BITS-256-NEXT:    vrsub.vx v12, v10, a0
; RV64-BITS-256-NEXT:    vrgather.vv v10, v8, v12
; RV64-BITS-256-NEXT:    vmv.v.v v8, v10
; RV64-BITS-256-NEXT:    ret
;
; RV64-BITS-512-LABEL: reverse_nxv16i8:
; RV64-BITS-512:       # %bb.0:
; RV64-BITS-512-NEXT:    csrr a0, vlenb
; RV64-BITS-512-NEXT:    slli a0, a0, 1
; RV64-BITS-512-NEXT:    addi a0, a0, -1
; RV64-BITS-512-NEXT:    vsetvli a1, zero, e8, m2, ta, mu
; RV64-BITS-512-NEXT:    vid.v v10
; RV64-BITS-512-NEXT:    vrsub.vx v12, v10, a0
; RV64-BITS-512-NEXT:    vrgather.vv v10, v8, v12
; RV64-BITS-512-NEXT:    vmv.v.v v8, v10
; RV64-BITS-512-NEXT:    ret
  %res = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> %a)
  ret <vscale x 16 x i8> %res
}

define <vscale x 32 x i8> @reverse_nxv32i8(<vscale x 32 x i8> %a) {
; RV32-BITS-UNKNOWN-LABEL: reverse_nxv32i8:
; RV32-BITS-UNKNOWN:       # %bb.0:
; RV32-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV32-BITS-UNKNOWN-NEXT:    slli a0, a0, 2
; RV32-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV32-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m8, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vid.v v16
; RV32-BITS-UNKNOWN-NEXT:    vrsub.vx v16, v16, a0
; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, m4, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v12, v8, v16
; RV32-BITS-UNKNOWN-NEXT:    vmv.v.v v8, v12
; RV32-BITS-UNKNOWN-NEXT:    ret
;
; RV32-BITS-256-LABEL: reverse_nxv32i8:
; RV32-BITS-256:       # %bb.0:
; RV32-BITS-256-NEXT:    csrr a0, vlenb
; RV32-BITS-256-NEXT:    slli a0, a0, 2
; RV32-BITS-256-NEXT:    addi a0, a0, -1
; RV32-BITS-256-NEXT:    vsetvli a1, zero, e8, m4, ta, mu
; RV32-BITS-256-NEXT:    vid.v v12
; RV32-BITS-256-NEXT:    vrsub.vx v16, v12, a0
; RV32-BITS-256-NEXT:    vrgather.vv v12, v8, v16
; RV32-BITS-256-NEXT:    vmv.v.v v8, v12
; RV32-BITS-256-NEXT:    ret
;
; RV32-BITS-512-LABEL: reverse_nxv32i8:
; RV32-BITS-512:       # %bb.0:
; RV32-BITS-512-NEXT:    csrr a0, vlenb
; RV32-BITS-512-NEXT:    slli a0, a0, 2
; RV32-BITS-512-NEXT:    addi a0, a0, -1
; RV32-BITS-512-NEXT:    vsetvli a1, zero, e8, m4, ta, mu
; RV32-BITS-512-NEXT:    vid.v v12
; RV32-BITS-512-NEXT:    vrsub.vx v16, v12, a0
; RV32-BITS-512-NEXT:    vrgather.vv v12, v8, v16
; RV32-BITS-512-NEXT:    vmv.v.v v8, v12
; RV32-BITS-512-NEXT:    ret
;
; RV64-BITS-UNKNOWN-LABEL: reverse_nxv32i8:
; RV64-BITS-UNKNOWN:       # %bb.0:
; RV64-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV64-BITS-UNKNOWN-NEXT:    slli a0, a0, 2
; RV64-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV64-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m8, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vid.v v16
; RV64-BITS-UNKNOWN-NEXT:    vrsub.vx v16, v16, a0
; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, m4, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v12, v8, v16
; RV64-BITS-UNKNOWN-NEXT:    vmv.v.v v8, v12
; RV64-BITS-UNKNOWN-NEXT:    ret
;
; RV64-BITS-256-LABEL: reverse_nxv32i8:
; RV64-BITS-256:       # %bb.0:
; RV64-BITS-256-NEXT:    csrr a0, vlenb
; RV64-BITS-256-NEXT:    slli a0, a0, 2
; RV64-BITS-256-NEXT:    addi a0, a0, -1
; RV64-BITS-256-NEXT:    vsetvli a1, zero, e8, m4, ta, mu
; RV64-BITS-256-NEXT:    vid.v v12
; RV64-BITS-256-NEXT:    vrsub.vx v16, v12, a0
; RV64-BITS-256-NEXT:    vrgather.vv v12, v8, v16
; RV64-BITS-256-NEXT:    vmv.v.v v8, v12
; RV64-BITS-256-NEXT:    ret
;
; RV64-BITS-512-LABEL: reverse_nxv32i8:
; RV64-BITS-512:       # %bb.0:
; RV64-BITS-512-NEXT:    csrr a0, vlenb
; RV64-BITS-512-NEXT:    slli a0, a0, 2
; RV64-BITS-512-NEXT:    addi a0, a0, -1
; RV64-BITS-512-NEXT:    vsetvli a1, zero, e8, m4, ta, mu
; RV64-BITS-512-NEXT:    vid.v v12
; RV64-BITS-512-NEXT:    vrsub.vx v16, v12, a0
; RV64-BITS-512-NEXT:    vrgather.vv v12, v8, v16
; RV64-BITS-512-NEXT:    vmv.v.v v8, v12
; RV64-BITS-512-NEXT:    ret
  %res = call <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8> %a)
  ret <vscale x 32 x i8> %res
}

define <vscale x 64 x i8> @reverse_nxv64i8(<vscale x 64 x i8> %a) {
; RV32-BITS-UNKNOWN-LABEL: reverse_nxv64i8:
; RV32-BITS-UNKNOWN:       # %bb.0:
; RV32-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV32-BITS-UNKNOWN-NEXT:    slli a0, a0, 2
; RV32-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV32-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m8, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vid.v v16
; RV32-BITS-UNKNOWN-NEXT:    vrsub.vx v24, v16, a0
; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, m4, ta, mu
; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v20, v8, v24
; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v16, v12, v24
; RV32-BITS-UNKNOWN-NEXT:    vmv8r.v v8, v16
; RV32-BITS-UNKNOWN-NEXT:    ret
;
; RV32-BITS-256-LABEL: reverse_nxv64i8:
; RV32-BITS-256:       # %bb.0:
; RV32-BITS-256-NEXT:    csrr a0, vlenb
; RV32-BITS-256-NEXT:    slli a0, a0, 3
; RV32-BITS-256-NEXT:    addi a0, a0, -1
; RV32-BITS-256-NEXT:    vsetvli a1, zero, e8, m8, ta, mu
; RV32-BITS-256-NEXT:    vid.v v16
; RV32-BITS-256-NEXT:    vrsub.vx v24, v16, a0
; RV32-BITS-256-NEXT:    vrgather.vv v16, v8, v24
; RV32-BITS-256-NEXT:    vmv.v.v v8, v16
; RV32-BITS-256-NEXT:    ret
;
; RV32-BITS-512-LABEL: reverse_nxv64i8:
; RV32-BITS-512:       # %bb.0:
; RV32-BITS-512-NEXT:    csrr a0, vlenb
; RV32-BITS-512-NEXT:    slli a0, a0, 2
; RV32-BITS-512-NEXT:    addi a0, a0, -1
; RV32-BITS-512-NEXT:    vsetvli a1, zero, e8, m4, ta, mu
; RV32-BITS-512-NEXT:    vid.v v16
; RV32-BITS-512-NEXT:    vrsub.vx v24, v16, a0
; RV32-BITS-512-NEXT:    vrgather.vv v20, v8, v24
; RV32-BITS-512-NEXT:    vrgather.vv v16, v12, v24
; RV32-BITS-512-NEXT:    vmv8r.v v8, v16
; RV32-BITS-512-NEXT:    ret
;
; RV64-BITS-UNKNOWN-LABEL: reverse_nxv64i8:
; RV64-BITS-UNKNOWN:       # %bb.0:
; RV64-BITS-UNKNOWN-NEXT:    csrr a0, vlenb
; RV64-BITS-UNKNOWN-NEXT:    slli a0, a0, 2
; RV64-BITS-UNKNOWN-NEXT:    addi a0, a0, -1
; RV64-BITS-UNKNOWN-NEXT:    vsetvli a1, zero, e16, m8, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vid.v v16
; RV64-BITS-UNKNOWN-NEXT:    vrsub.vx v24, v16, a0
; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, m4, ta, mu
; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v20, v8, v24
; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v16, v12, v24
; RV64-BITS-UNKNOWN-NEXT:    vmv8r.v v8, v16
; RV64-BITS-UNKNOWN-NEXT:    ret
;
; RV64-BITS-256-LABEL: reverse_nxv64i8:
; RV64-BITS-256:       # %bb.0:
; RV64-BITS-256-NEXT:    csrr a0, vlenb
; RV64-BITS-256-NEXT:    slli a0, a0, 3
; RV64-BITS-256-NEXT:    addi a0, a0, -1
; RV64-BITS-256-NEXT:    vsetvli a1, zero, e8, m8, ta, mu
; RV64-BITS-256-NEXT:    vid.v v16
; RV64-BITS-256-NEXT:    vrsub.vx v24, v16, a0
; RV64-BITS-256-NEXT:    vrgather.vv v16, v8, v24
; RV64-BITS-256-NEXT:    vmv.v.v v8, v16
; RV64-BITS-256-NEXT:    ret
;
; RV64-BITS-512-LABEL: reverse_nxv64i8:
; RV64-BITS-512:       # %bb.0:
; RV64-BITS-512-NEXT:    csrr a0, vlenb
; RV64-BITS-512-NEXT:    slli a0, a0, 2
; RV64-BITS-512-NEXT:    addi a0, a0, -1
; RV64-BITS-512-NEXT:    vsetvli a1, zero, e8, m4, ta, mu
; RV64-BITS-512-NEXT:    vid.v v16
; RV64-BITS-512-NEXT:    vrsub.vx v24, v16, a0
; RV64-BITS-512-NEXT:    vrgather.vv v20, v8, v24
; RV64-BITS-512-NEXT:    vrgather.vv v16, v12, v24
; RV64-BITS-512-NEXT:    vmv8r.v v8, v16
; RV64-BITS-512-NEXT:    ret
  %res = call <vscale x 64 x i8> @llvm.experimental.vector.reverse.nxv64i8(<vscale x 64 x i8> %a)
  ret <vscale x 64 x i8> %res
}

define <vscale x 1 x i16> @reverse_nxv1i16(<vscale x 1 x i16> %a) {
; CHECK-LABEL: reverse_nxv1i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 3
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, mf4, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 1 x i16> @llvm.experimental.vector.reverse.nxv1i16(<vscale x 1 x i16> %a)
  ret <vscale x 1 x i16> %res
}

define <vscale x 2 x i16> @reverse_nxv2i16(<vscale x 2 x i16> %a) {
; CHECK-LABEL: reverse_nxv2i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 2
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, mf2, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16> %a)
  ret <vscale x 2 x i16> %res
}

define <vscale x 4 x i16> @reverse_nxv4i16(<vscale x 4 x i16> %a) {
; CHECK-LABEL: reverse_nxv4i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16> %a)
  ret <vscale x 4 x i16> %res
}

define <vscale x 8 x i16> @reverse_nxv8i16(<vscale x 8 x i16> %a) {
; CHECK-LABEL: reverse_nxv8i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, mu
; CHECK-NEXT:    vid.v v10
; CHECK-NEXT:    vrsub.vx v12, v10, a0
; CHECK-NEXT:    vrgather.vv v10, v8, v12
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16> %a)
  ret <vscale x 8 x i16> %res
}

define <vscale x 16 x i16> @reverse_nxv16i16(<vscale x 16 x i16> %a) {
; CHECK-LABEL: reverse_nxv16i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    slli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, m4, ta, mu
; CHECK-NEXT:    vid.v v12
; CHECK-NEXT:    vrsub.vx v16, v12, a0
; CHECK-NEXT:    vrgather.vv v12, v8, v16
; CHECK-NEXT:    vmv.v.v v8, v12
; CHECK-NEXT:    ret
  %res = call <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16> %a)
  ret <vscale x 16 x i16> %res
}

define <vscale x 32 x i16> @reverse_nxv32i16(<vscale x 32 x i16> %a) {
; CHECK-LABEL: reverse_nxv32i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    slli a0, a0, 2
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, m8, ta, mu
; CHECK-NEXT:    vid.v v16
; CHECK-NEXT:    vrsub.vx v24, v16, a0
; CHECK-NEXT:    vrgather.vv v16, v8, v24
; CHECK-NEXT:    vmv.v.v v8, v16
; CHECK-NEXT:    ret
  %res = call <vscale x 32 x i16> @llvm.experimental.vector.reverse.nxv32i16(<vscale x 32 x i16> %a)
  ret <vscale x 32 x i16> %res
}

define <vscale x 1 x i32> @reverse_nxv1i32(<vscale x 1 x i32> %a) {
; CHECK-LABEL: reverse_nxv1i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 3
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 1 x i32> @llvm.experimental.vector.reverse.nxv1i32(<vscale x 1 x i32> %a)
  ret <vscale x 1 x i32> %res
}

define <vscale x 2 x i32> @reverse_nxv2i32(<vscale x 2 x i32> %a) {
; CHECK-LABEL: reverse_nxv2i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 2
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 2 x i32> @llvm.experimental.vector.reverse.nxv2i32(<vscale x 2 x i32> %a)
  ret <vscale x 2 x i32> %res
}

define <vscale x 4 x i32> @reverse_nxv4i32(<vscale x 4 x i32> %a) {
; CHECK-LABEL: reverse_nxv4i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
; CHECK-NEXT:    vid.v v10
; CHECK-NEXT:    vrsub.vx v12, v10, a0
; CHECK-NEXT:    vrgather.vv v10, v8, v12
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
  ret <vscale x 4 x i32> %res
}

define <vscale x 8 x i32> @reverse_nxv8i32(<vscale x 8 x i32> %a) {
; CHECK-LABEL: reverse_nxv8i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
; CHECK-NEXT:    vid.v v12
; CHECK-NEXT:    vrsub.vx v16, v12, a0
; CHECK-NEXT:    vrgather.vv v12, v8, v16
; CHECK-NEXT:    vmv.v.v v8, v12
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> %a)
  ret <vscale x 8 x i32> %res
}

define <vscale x 16 x i32> @reverse_nxv16i32(<vscale x 16 x i32> %a) {
; CHECK-LABEL: reverse_nxv16i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    slli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, m8, ta, mu
; CHECK-NEXT:    vid.v v16
; CHECK-NEXT:    vrsub.vx v24, v16, a0
; CHECK-NEXT:    vrgather.vv v16, v8, v24
; CHECK-NEXT:    vmv.v.v v8, v16
; CHECK-NEXT:    ret
  %res = call <vscale x 16 x i32> @llvm.experimental.vector.reverse.nxv16i32(<vscale x 16 x i32> %a)
  ret <vscale x 16 x i32> %res
}

define <vscale x 1 x i64> @reverse_nxv1i64(<vscale x 1 x i64> %a) {
; CHECK-LABEL: reverse_nxv1i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 3
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 1 x i64> @llvm.experimental.vector.reverse.nxv1i64(<vscale x 1 x i64> %a)
  ret <vscale x 1 x i64> %res
}

define <vscale x 2 x i64> @reverse_nxv2i64(<vscale x 2 x i64> %a) {
; CHECK-LABEL: reverse_nxv2i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 2
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
; CHECK-NEXT:    vid.v v10
; CHECK-NEXT:    vrsub.vx v12, v10, a0
; CHECK-NEXT:    vrgather.vv v10, v8, v12
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %res = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> %a)
  ret <vscale x 2 x i64> %res
}

define <vscale x 4 x i64> @reverse_nxv4i64(<vscale x 4 x i64> %a) {
; CHECK-LABEL: reverse_nxv4i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
; CHECK-NEXT:    vid.v v12
; CHECK-NEXT:    vrsub.vx v16, v12, a0
; CHECK-NEXT:    vrgather.vv v12, v8, v16
; CHECK-NEXT:    vmv.v.v v8, v12
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64> %a)
  ret <vscale x 4 x i64> %res
}

define <vscale x 8 x i64> @reverse_nxv8i64(<vscale x 8 x i64> %a) {
; CHECK-LABEL: reverse_nxv8i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
; CHECK-NEXT:    vid.v v16
; CHECK-NEXT:    vrsub.vx v24, v16, a0
; CHECK-NEXT:    vrgather.vv v16, v8, v24
; CHECK-NEXT:    vmv.v.v v8, v16
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x i64> @llvm.experimental.vector.reverse.nxv8i64(<vscale x 8 x i64> %a)
  ret <vscale x 8 x i64> %res
}

;
; VECTOR_REVERSE - floating point
;

define <vscale x 1 x half> @reverse_nxv1f16(<vscale x 1 x half> %a) {
; CHECK-LABEL: reverse_nxv1f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 3
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, mf4, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 1 x half> @llvm.experimental.vector.reverse.nxv1f16(<vscale x 1 x half> %a)
  ret <vscale x 1 x half> %res
}

define <vscale x 2 x half> @reverse_nxv2f16(<vscale x 2 x half> %a) {
; CHECK-LABEL: reverse_nxv2f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 2
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, mf2, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half> %a)
  ret <vscale x 2 x half> %res
}

define <vscale x 4 x half> @reverse_nxv4f16(<vscale x 4 x half> %a) {
; CHECK-LABEL: reverse_nxv4f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half> %a)
  ret <vscale x 4 x half> %res
}

define <vscale x 8 x half> @reverse_nxv8f16(<vscale x 8 x half> %a) {
; CHECK-LABEL: reverse_nxv8f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, mu
; CHECK-NEXT:    vid.v v10
; CHECK-NEXT:    vrsub.vx v12, v10, a0
; CHECK-NEXT:    vrgather.vv v10, v8, v12
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half> %a)
  ret <vscale x 8 x half> %res
}

define <vscale x 16 x half> @reverse_nxv16f16(<vscale x 16 x half> %a) {
; CHECK-LABEL: reverse_nxv16f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    slli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, m4, ta, mu
; CHECK-NEXT:    vid.v v12
; CHECK-NEXT:    vrsub.vx v16, v12, a0
; CHECK-NEXT:    vrgather.vv v12, v8, v16
; CHECK-NEXT:    vmv.v.v v8, v12
; CHECK-NEXT:    ret
  %res = call <vscale x 16 x half> @llvm.experimental.vector.reverse.nxv16f16(<vscale x 16 x half> %a)
  ret <vscale x 16 x half> %res
}

define <vscale x 32 x half> @reverse_nxv32f16(<vscale x 32 x half> %a) {
; CHECK-LABEL: reverse_nxv32f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    slli a0, a0, 2
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e16, m8, ta, mu
; CHECK-NEXT:    vid.v v16
; CHECK-NEXT:    vrsub.vx v24, v16, a0
; CHECK-NEXT:    vrgather.vv v16, v8, v24
; CHECK-NEXT:    vmv.v.v v8, v16
; CHECK-NEXT:    ret
  %res = call <vscale x 32 x half> @llvm.experimental.vector.reverse.nxv32f16(<vscale x 32 x half> %a)
  ret <vscale x 32 x half> %res
}

define <vscale x 1 x float> @reverse_nxv1f32(<vscale x 1 x float> %a) {
; CHECK-LABEL: reverse_nxv1f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 3
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 1 x float> @llvm.experimental.vector.reverse.nxv1f32(<vscale x 1 x float> %a)
  ret <vscale x 1 x float> %res
}

define <vscale x 2 x float> @reverse_nxv2f32(<vscale x 2 x float> %a) {
; CHECK-LABEL: reverse_nxv2f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 2
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float> %a)
  ret <vscale x 2 x float> %res
}

define <vscale x 4 x float> @reverse_nxv4f32(<vscale x 4 x float> %a) {
; CHECK-LABEL: reverse_nxv4f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, mu
; CHECK-NEXT:    vid.v v10
; CHECK-NEXT:    vrsub.vx v12, v10, a0
; CHECK-NEXT:    vrgather.vv v10, v8, v12
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
  ret <vscale x 4 x float> %res
}

define <vscale x 8 x float> @reverse_nxv8f32(<vscale x 8 x float> %a) {
; CHECK-LABEL: reverse_nxv8f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, mu
; CHECK-NEXT:    vid.v v12
; CHECK-NEXT:    vrsub.vx v16, v12, a0
; CHECK-NEXT:    vrgather.vv v12, v8, v16
; CHECK-NEXT:    vmv.v.v v8, v12
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x float> @llvm.experimental.vector.reverse.nxv8f32(<vscale x 8 x float> %a)
  ret <vscale x 8 x float> %res
}

define <vscale x 16 x float> @reverse_nxv16f32(<vscale x 16 x float> %a) {
; CHECK-LABEL: reverse_nxv16f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    slli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e32, m8, ta, mu
; CHECK-NEXT:    vid.v v16
; CHECK-NEXT:    vrsub.vx v24, v16, a0
; CHECK-NEXT:    vrgather.vv v16, v8, v24
; CHECK-NEXT:    vmv.v.v v8, v16
; CHECK-NEXT:    ret
  %res = call <vscale x 16 x float> @llvm.experimental.vector.reverse.nxv16f32(<vscale x 16 x float> %a)
  ret <vscale x 16 x float> %res
}

define <vscale x 1 x double> @reverse_nxv1f64(<vscale x 1 x double> %a) {
; CHECK-LABEL: reverse_nxv1f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 3
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e64, m1, ta, mu
; CHECK-NEXT:    vid.v v9
; CHECK-NEXT:    vrsub.vx v10, v9, a0
; CHECK-NEXT:    vrgather.vv v9, v8, v10
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %res = call <vscale x 1 x double> @llvm.experimental.vector.reverse.nxv1f64(<vscale x 1 x double> %a)
  ret <vscale x 1 x double> %res
}

define <vscale x 2 x double> @reverse_nxv2f64(<vscale x 2 x double> %a) {
; CHECK-LABEL: reverse_nxv2f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 2
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, mu
; CHECK-NEXT:    vid.v v10
; CHECK-NEXT:    vrsub.vx v12, v10, a0
; CHECK-NEXT:    vrgather.vv v10, v8, v12
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %res = call <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double> %a)
  ret <vscale x 2 x double> %res
}

define <vscale x 4 x double> @reverse_nxv4f64(<vscale x 4 x double> %a) {
; CHECK-LABEL: reverse_nxv4f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    srli a0, a0, 1
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e64, m4, ta, mu
; CHECK-NEXT:    vid.v v12
; CHECK-NEXT:    vrsub.vx v16, v12, a0
; CHECK-NEXT:    vrgather.vv v12, v8, v16
; CHECK-NEXT:    vmv.v.v v8, v12
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double> %a)
  ret <vscale x 4 x double> %res
}

define <vscale x 8 x double> @reverse_nxv8f64(<vscale x 8 x double> %a) {
; CHECK-LABEL: reverse_nxv8f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a0, vlenb
; CHECK-NEXT:    addi a0, a0, -1
; CHECK-NEXT:    vsetvli a1, zero, e64, m8, ta, mu
; CHECK-NEXT:    vid.v v16
; CHECK-NEXT:    vrsub.vx v24, v16, a0
; CHECK-NEXT:    vrgather.vv v16, v8, v24
; CHECK-NEXT:    vmv.v.v v8, v16
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x double> @llvm.experimental.vector.reverse.nxv8f64(<vscale x 8 x double> %a)
  ret <vscale x 8 x double> %res
}

declare <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1>)
declare <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1>)
declare <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1>)
declare <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1>)
declare <vscale x 32 x i1> @llvm.experimental.vector.reverse.nxv32i1(<vscale x 32 x i1>)
declare <vscale x 64 x i1> @llvm.experimental.vector.reverse.nxv64i1(<vscale x 64 x i1>)
declare <vscale x 1 x i8> @llvm.experimental.vector.reverse.nxv1i8(<vscale x 1 x i8>)
declare <vscale x 2 x i8> @llvm.experimental.vector.reverse.nxv2i8(<vscale x 2 x i8>)
declare <vscale x 4 x i8> @llvm.experimental.vector.reverse.nxv4i8(<vscale x 4 x i8>)
declare <vscale x 8 x i8> @llvm.experimental.vector.reverse.nxv8i8(<vscale x 8 x i8>)
declare <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8>)
declare <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8>)
declare <vscale x 64 x i8> @llvm.experimental.vector.reverse.nxv64i8(<vscale x 64 x i8>)
declare <vscale x 1 x i16> @llvm.experimental.vector.reverse.nxv1i16(<vscale x 1 x i16>)
declare <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16>)
declare <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16>)
declare <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16>)
declare <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16>)
declare <vscale x 32 x i16> @llvm.experimental.vector.reverse.nxv32i16(<vscale x 32 x i16>)
declare <vscale x 1 x i32> @llvm.experimental.vector.reverse.nxv1i32(<vscale x 1 x i32>)
declare <vscale x 2 x i32> @llvm.experimental.vector.reverse.nxv2i32(<vscale x 2 x i32>)
declare <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32>)
declare <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32>)
declare <vscale x 16 x i32> @llvm.experimental.vector.reverse.nxv16i32(<vscale x 16 x i32>)
declare <vscale x 1 x i64> @llvm.experimental.vector.reverse.nxv1i64(<vscale x 1 x i64>)
declare <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64>)
declare <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64>)
declare <vscale x 8 x i64> @llvm.experimental.vector.reverse.nxv8i64(<vscale x 8 x i64>)
declare <vscale x 1 x half> @llvm.experimental.vector.reverse.nxv1f16(<vscale x 1 x half>)
declare <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half>)
declare <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half>)
declare <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half>)
declare <vscale x 16 x half> @llvm.experimental.vector.reverse.nxv16f16(<vscale x 16 x half>)
declare <vscale x 32 x half> @llvm.experimental.vector.reverse.nxv32f16(<vscale x 32 x half>)
declare <vscale x 1 x float> @llvm.experimental.vector.reverse.nxv1f32(<vscale x 1 x float>)
declare <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float>)
declare <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float>)
declare <vscale x 8 x float> @llvm.experimental.vector.reverse.nxv8f32(<vscale x 8 x float>)
declare <vscale x 16 x float> @llvm.experimental.vector.reverse.nxv16f32(<vscale x 16 x float>)
declare <vscale x 1 x double> @llvm.experimental.vector.reverse.nxv1f64(<vscale x 1 x double>)
declare <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double>)
declare <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double>)
declare <vscale x 8 x double> @llvm.experimental.vector.reverse.nxv8f64(<vscale x 8 x double>)