Index: llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
===================================================================
--- llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
+++ llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
@@ -474,11 +474,23 @@
   let PrimaryKeyName = "getRISCVVIntrinsicInfo";
 }
 
-class RISCVMaskedPseudo<bits<4> MaskIdx, bit HasTU = true, bit IsTernary = false> {
+// Describes the relation of a masked psuedu to the unmasked variants.
+// (HasTU = true, IsCombined = false)
+//    Has both a TA (unsuffixed) and _TU variant defined.  TU variant
+//    may (or may not) have a vector policy operand.
+// (HasTU = false, IsCombined = false)
+//    No TA (unsuffixed) or _TU variants.  Masked version is only pseudo
+// (HasTU = false, IsCombined = true)
+//    The unsuffixed version has a merge operand; no explicit _TU variant
+//    exists.  If the unsuffixed version has a policy operand, and can thus
+//    represent all policy states.
+// (HasTU = true, IsCombined = true)
+//    Invalid and unused state.
+class RISCVMaskedPseudo<bits<4> MaskIdx, bit HasTU = true, bit IsCombined = false> {
   Pseudo MaskedPseudo = !cast<Pseudo>(NAME);
   Pseudo UnmaskedPseudo = !cast<Pseudo>(!subst("_MASK", "", NAME));
   Pseudo UnmaskedTUPseudo = !cond(HasTU : !cast<Pseudo>(!subst("_MASK", "", NAME # "_TU")),
-                                  IsTernary : UnmaskedPseudo,
+                                  IsCombined : UnmaskedPseudo,
                                   true  :  MaskedPseudo);
   bits<4> MaskOpIdx = MaskIdx;
 }
@@ -914,25 +926,15 @@
 
 class VPseudoNullaryNoMask<VReg RegClass>:
       Pseudo<(outs RegClass:$rd),
-             (ins AVL:$vl, ixlenimm:$sew),
-             []>, RISCVVPseudo {
-  let mayLoad = 0;
-  let mayStore = 0;
-  let hasSideEffects = 0;
-  let HasVLOp = 1;
-  let HasSEWOp = 1;
-}
-
-class VPseudoNullaryNoMaskTU<VReg RegClass>:
-      Pseudo<(outs RegClass:$rd),
-             (ins RegClass:$merge, AVL:$vl, ixlenimm:$sew),
-             []>, RISCVVPseudo {
+             (ins RegClass:$merge, AVL:$vl, ixlenimm:$sew,
+              ixlenimm:$policy), []>, RISCVVPseudo {
   let mayLoad = 0;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = "$rd = $merge";
   let HasVLOp = 1;
   let HasSEWOp = 1;
+  let HasVecPolicyOp = 1;
 }
 
 class VPseudoNullaryMask<VReg RegClass>:
@@ -1815,10 +1817,8 @@
     let VLMul = m.value in {
       def "_V_" # m.MX : VPseudoNullaryNoMask<m.vrclass>,
                          Sched<[WriteVMIdxV_MX, ReadVMask]>;
-      def "_V_" # m.MX # "_TU": VPseudoNullaryNoMaskTU<m.vrclass>,
-                                Sched<[WriteVMIdxV_MX, ReadVMask]>;
       def "_V_" # m.MX # "_MASK" : VPseudoNullaryMask<m.vrclass>,
-                                   RISCVMaskedPseudo</*MaskOpIdx*/ 1>,
+                                   RISCVMaskedPseudo</*MaskOpIdx*/ 1, /*HasTU=*/false, /*IsCombined=*/true>,
                                    Sched<[WriteVMIdxV_MX, ReadVMask]>;
     }
   }
@@ -3062,7 +3062,7 @@
     let isCommutable = Commutable in
     def "_" # MInfo.MX : VPseudoTernaryNoMaskWithPolicy<RetClass, Op1Class, Op2Class, Constraint>;
     def "_" # MInfo.MX # "_MASK" : VPseudoBinaryMaskPolicy<RetClass, Op1Class, Op2Class, Constraint>,
-                                   RISCVMaskedPseudo</*MaskOpIdx*/ 3, /*HasTU*/ false, /*IsTernary*/true>;
+                                   RISCVMaskedPseudo</*MaskOpIdx*/ 3, /*HasTU*/ false, /*IsCombined*/true>;
   }
 }
 
@@ -4311,13 +4311,14 @@
       def : Pat<(vti.Vector (!cast<Intrinsic>(intrinsic)
                             (vti.Vector undef),
                             VLOpFrag)),
-                            (!cast<Instruction>(instruction#"_V_" # vti.LMul.MX # "_TU")
-                            (vti.Vector (IMPLICIT_DEF)), GPR:$vl, vti.Log2SEW)>;
+                            (!cast<Instruction>(instruction#"_V_" # vti.LMul.MX)
+                            (vti.Vector (IMPLICIT_DEF)), GPR:$vl, vti.Log2SEW, TA_MA)>;
       def : Pat<(vti.Vector (!cast<Intrinsic>(intrinsic)
                             (vti.Vector vti.RegClass:$merge),
                             VLOpFrag)),
-                            (!cast<Instruction>(instruction#"_V_" # vti.LMul.MX # "_TU")
-                            vti.RegClass:$merge, GPR:$vl, vti.Log2SEW)>;
+                            (!cast<Instruction>(instruction#"_V_" # vti.LMul.MX)
+                            vti.RegClass:$merge, GPR:$vl, vti.Log2SEW,
+                            TAIL_UNDISTURBED_MASK_UNDISTURBED)>;
       def : Pat<(vti.Vector (!cast<Intrinsic>(intrinsic # "_mask")
                             (vti.Vector vti.RegClass:$merge),
                             (vti.Mask V0), VLOpFrag, (XLenVT timm:$policy))),
Index: llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-buildvec.ll
===================================================================
--- llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-buildvec.ll
+++ llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-buildvec.ll
@@ -154,10 +154,10 @@
 ; CHECK-LABEL: buildvec_vid_stepn3_add3_v4i8:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 4, e8, mf4, ta, ma
-; CHECK-NEXT:    vmv.v.i v9, 3
-; CHECK-NEXT:    vid.v v8
+; CHECK-NEXT:    vid.v v9
+; CHECK-NEXT:    vmv.v.i v8, 3
 ; CHECK-NEXT:    li a0, -3
-; CHECK-NEXT:    vmadd.vx v8, a0, v9
+; CHECK-NEXT:    vmacc.vx v8, a0, v9
 ; CHECK-NEXT:    ret
   ret <4 x i8> <i8 3, i8 0, i8 -3, i8 -6>
 }
@@ -166,10 +166,10 @@
 ; CHECK-LABEL: buildvec_vid_stepn3_addn3_v4i32:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
-; CHECK-NEXT:    vmv.v.i v8, -3
-; CHECK-NEXT:    vid.v v9
+; CHECK-NEXT:    vid.v v8
+; CHECK-NEXT:    vmv.v.i v9, -3
 ; CHECK-NEXT:    li a4, -3
-; CHECK-NEXT:    vmadd.vx v9, a4, v8
+; CHECK-NEXT:    vmacc.vx v9, a4, v8
 ; CHECK-NEXT:    vse32.v v9, (a0)
 ; CHECK-NEXT:    vse32.v v9, (a1)
 ; CHECK-NEXT:    vse32.v v9, (a2)
Index: llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int.ll
===================================================================
--- llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int.ll
+++ llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int.ll
@@ -1440,10 +1440,10 @@
 ; CHECK-NEXT:    vsetivli zero, 6, e16, m1, ta, ma
 ; CHECK-NEXT:    vle16.v v8, (a0)
 ; CHECK-NEXT:    vsetivli zero, 2, e16, mf4, ta, ma
-; CHECK-NEXT:    vmv.v.i v9, 7
-; CHECK-NEXT:    vid.v v10
+; CHECK-NEXT:    vid.v v9
+; CHECK-NEXT:    vmv.v.i v10, 7
 ; CHECK-NEXT:    li a1, -14
-; CHECK-NEXT:    vmadd.vx v10, a1, v9
+; CHECK-NEXT:    vmacc.vx v10, a1, v9
 ; CHECK-NEXT:    vsetivli zero, 2, e16, m1, ta, ma
 ; CHECK-NEXT:    vslidedown.vi v9, v8, 4
 ; CHECK-NEXT:    vsetivli zero, 2, e16, mf4, ta, ma
Index: llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll
===================================================================
--- llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll
+++ llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll
@@ -1157,8 +1157,15 @@
 ; RV64-NEXT:    vrgather.vv v8, v24, v16, v0.t
 ; RV64-NEXT:    vsetivli zero, 5, e64, m4, tu, ma
 ; RV64-NEXT:    vmv.v.v v4, v8
-; RV64-NEXT:    addi a1, a0, 320
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, ma
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 29
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vse64.v v8, (a0)
+; RV64-NEXT:    addi a1, a0, 320
 ; RV64-NEXT:    vse64.v v4, (a1)
 ; RV64-NEXT:    addi a1, a0, 256
 ; RV64-NEXT:    csrr a2, vlenb
@@ -1184,16 +1191,9 @@
 ; RV64-NEXT:    addi a2, a2, 16
 ; RV64-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
 ; RV64-NEXT:    vse64.v v8, (a1)
-; RV64-NEXT:    addi a1, a0, 64
-; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    li a3, 25
-; RV64-NEXT:    mul a2, a2, a3
-; RV64-NEXT:    add a2, sp, a2
-; RV64-NEXT:    addi a2, a2, 16
-; RV64-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
-; RV64-NEXT:    vse64.v v8, (a1)
+; RV64-NEXT:    addi a0, a0, 64
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 29
+; RV64-NEXT:    li a2, 25
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
Index: llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-int-vp.ll
===================================================================
--- llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-int-vp.ll
+++ llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-int-vp.ll
@@ -1796,20 +1796,20 @@
 ; RV32-NEXT:    .cfi_def_cfa_offset 16
 ; RV32-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32-NEXT:    .cfi_offset ra, -4
-; RV32-NEXT:    li a3, 32
-; RV32-NEXT:    vsetvli zero, a3, e32, m8, ta, ma
 ; RV32-NEXT:    lui a2, %hi(.LCPI72_0)
 ; RV32-NEXT:    addi a2, a2, %lo(.LCPI72_0)
+; RV32-NEXT:    li a3, 32
+; RV32-NEXT:    vsetvli zero, a3, e32, m8, ta, ma
 ; RV32-NEXT:    vle32.v v16, (a2)
 ; RV32-NEXT:    mv a2, a0
-; RV32-NEXT:    vid.v v24
-; RV32-NEXT:    vmsltu.vx v12, v24, a1
+; RV32-NEXT:    vmsltu.vx v12, v16, a1
+; RV32-NEXT:    vid.v v16
 ; RV32-NEXT:    vmsltu.vx v13, v16, a1
 ; RV32-NEXT:    vsetivli zero, 8, e8, mf2, ta, ma
-; RV32-NEXT:    vslideup.vi v12, v13, 4
+; RV32-NEXT:    vslideup.vi v13, v12, 4
 ; RV32-NEXT:    li a0, 64
 ; RV32-NEXT:    vsetvli zero, a0, e8, m4, ta, ma
-; RV32-NEXT:    vmand.mm v0, v12, v0
+; RV32-NEXT:    vmand.mm v0, v13, v0
 ; RV32-NEXT:    vmv.v.i v12, 1
 ; RV32-NEXT:    vmerge.vvm v8, v12, v8, v0
 ; RV32-NEXT:    vslidedown.vx v12, v8, a3
@@ -1839,20 +1839,20 @@
 ; RV64-NEXT:    .cfi_def_cfa_offset 16
 ; RV64-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
 ; RV64-NEXT:    .cfi_offset ra, -8
-; RV64-NEXT:    li a3, 32
-; RV64-NEXT:    vsetvli zero, a3, e32, m8, ta, ma
 ; RV64-NEXT:    lui a2, %hi(.LCPI72_0)
 ; RV64-NEXT:    addi a2, a2, %lo(.LCPI72_0)
+; RV64-NEXT:    li a3, 32
+; RV64-NEXT:    vsetvli zero, a3, e32, m8, ta, ma
 ; RV64-NEXT:    vle32.v v16, (a2)
 ; RV64-NEXT:    mv a2, a0
-; RV64-NEXT:    vid.v v24
-; RV64-NEXT:    vmsltu.vx v12, v24, a1
+; RV64-NEXT:    vmsltu.vx v12, v16, a1
+; RV64-NEXT:    vid.v v16
 ; RV64-NEXT:    vmsltu.vx v13, v16, a1
 ; RV64-NEXT:    vsetivli zero, 8, e8, mf2, ta, ma
-; RV64-NEXT:    vslideup.vi v12, v13, 4
+; RV64-NEXT:    vslideup.vi v13, v12, 4
 ; RV64-NEXT:    li a0, 64
 ; RV64-NEXT:    vsetvli zero, a0, e8, m4, ta, ma
-; RV64-NEXT:    vmand.mm v0, v12, v0
+; RV64-NEXT:    vmand.mm v0, v13, v0
 ; RV64-NEXT:    vmv.v.i v12, 1
 ; RV64-NEXT:    vmerge.vvm v8, v12, v8, v0
 ; RV64-NEXT:    vslidedown.vx v12, v8, a3
Index: llvm/test/CodeGen/RISCV/rvv/fixed-vectors-store.ll
===================================================================
--- llvm/test/CodeGen/RISCV/rvv/fixed-vectors-store.ll
+++ llvm/test/CodeGen/RISCV/rvv/fixed-vectors-store.ll
@@ -139,10 +139,10 @@
 ; CHECK-LABEL: store_constant_v2i32:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 2, e32, mf2, ta, ma
-; CHECK-NEXT:    vmv.v.i v8, 3
-; CHECK-NEXT:    vid.v v9
+; CHECK-NEXT:    vid.v v8
+; CHECK-NEXT:    vmv.v.i v9, 3
 ; CHECK-NEXT:    li a1, 3
-; CHECK-NEXT:    vmadd.vx v9, a1, v8
+; CHECK-NEXT:    vmacc.vx v9, a1, v8
 ; CHECK-NEXT:    vse32.v v9, (a0)
 ; CHECK-NEXT:    ret
   store <2 x i32> <i32 3, i32 6>, ptr %p
@@ -215,10 +215,10 @@
 ; CHECK-LABEL: store_constant_v2i8_align1:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 2, e8, mf8, ta, ma
-; CHECK-NEXT:    vmv.v.i v8, 3
-; CHECK-NEXT:    vid.v v9
+; CHECK-NEXT:    vid.v v8
+; CHECK-NEXT:    vmv.v.i v9, 3
 ; CHECK-NEXT:    li a1, 3
-; CHECK-NEXT:    vmadd.vx v9, a1, v8
+; CHECK-NEXT:    vmacc.vx v9, a1, v8
 ; CHECK-NEXT:    vse8.v v9, (a0)
 ; CHECK-NEXT:    ret
   store <2 x i8> <i8 3, i8 6>, ptr %p, align 1
Index: llvm/test/CodeGen/RISCV/rvv/named-vector-shuffle-reverse.ll
===================================================================
--- llvm/test/CodeGen/RISCV/rvv/named-vector-shuffle-reverse.ll
+++ llvm/test/CodeGen/RISCV/rvv/named-vector-shuffle-reverse.ll
@@ -16,11 +16,11 @@
 ; RV32-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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 zero, zero, e16, mf2, ta, ma
-; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
 ; RV32-BITS-UNKNOWN-NEXT:    vrsub.vx v9, v9, a0
 ; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf4, ta, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v9
@@ -33,10 +33,10 @@
 ; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
 ; RV32-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-256-NEXT:    vid.v v9
 ; 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:    vid.v v9
 ; RV32-BITS-256-NEXT:    vrsub.vx v9, v9, a0
 ; RV32-BITS-256-NEXT:    vrgather.vv v10, v8, v9
 ; RV32-BITS-256-NEXT:    vand.vi v8, v10, 1
@@ -48,10 +48,10 @@
 ; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
 ; RV32-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-512-NEXT:    vid.v v9
 ; 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:    vid.v v9
 ; RV32-BITS-512-NEXT:    vrsub.vx v9, v9, a0
 ; RV32-BITS-512-NEXT:    vrgather.vv v10, v8, v9
 ; RV32-BITS-512-NEXT:    vand.vi v8, v10, 1
@@ -63,11 +63,11 @@
 ; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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 zero, zero, e16, mf2, ta, ma
-; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
 ; RV64-BITS-UNKNOWN-NEXT:    vrsub.vx v9, v9, a0
 ; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf4, ta, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v9
@@ -80,10 +80,10 @@
 ; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
 ; RV64-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-256-NEXT:    vid.v v9
 ; 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:    vid.v v9
 ; RV64-BITS-256-NEXT:    vrsub.vx v9, v9, a0
 ; RV64-BITS-256-NEXT:    vrgather.vv v10, v8, v9
 ; RV64-BITS-256-NEXT:    vand.vi v8, v10, 1
@@ -95,10 +95,10 @@
 ; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
 ; RV64-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-512-NEXT:    vid.v v9
 ; 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:    vid.v v9
 ; RV64-BITS-512-NEXT:    vrsub.vx v9, v9, a0
 ; RV64-BITS-512-NEXT:    vrgather.vv v10, v8, v9
 ; RV64-BITS-512-NEXT:    vand.vi v8, v10, 1
@@ -114,11 +114,11 @@
 ; RV32-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e16, m1, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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 zero, zero, e16, m1, ta, ma
-; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
 ; RV32-BITS-UNKNOWN-NEXT:    vrsub.vx v9, v9, a0
 ; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf2, ta, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v9
@@ -131,10 +131,10 @@
 ; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
 ; RV32-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-256-NEXT:    vid.v v9
 ; 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:    vid.v v9
 ; RV32-BITS-256-NEXT:    vrsub.vx v9, v9, a0
 ; RV32-BITS-256-NEXT:    vrgather.vv v10, v8, v9
 ; RV32-BITS-256-NEXT:    vand.vi v8, v10, 1
@@ -146,10 +146,10 @@
 ; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
 ; RV32-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-512-NEXT:    vid.v v9
 ; 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:    vid.v v9
 ; RV32-BITS-512-NEXT:    vrsub.vx v9, v9, a0
 ; RV32-BITS-512-NEXT:    vrgather.vv v10, v8, v9
 ; RV32-BITS-512-NEXT:    vand.vi v8, v10, 1
@@ -161,11 +161,11 @@
 ; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e16, m1, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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 zero, zero, e16, m1, ta, ma
-; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
 ; RV64-BITS-UNKNOWN-NEXT:    vrsub.vx v9, v9, a0
 ; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e8, mf2, ta, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v9
@@ -178,10 +178,10 @@
 ; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
 ; RV64-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-256-NEXT:    vid.v v9
 ; 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:    vid.v v9
 ; RV64-BITS-256-NEXT:    vrsub.vx v9, v9, a0
 ; RV64-BITS-256-NEXT:    vrgather.vv v10, v8, v9
 ; RV64-BITS-256-NEXT:    vand.vi v8, v10, 1
@@ -193,10 +193,10 @@
 ; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
 ; RV64-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-512-NEXT:    vid.v v9
 ; 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:    vid.v v9
 ; RV64-BITS-512-NEXT:    vrsub.vx v9, v9, a0
 ; RV64-BITS-512-NEXT:    vrgather.vv v10, v8, v9
 ; RV64-BITS-512-NEXT:    vand.vi v8, v10, 1
@@ -304,11 +304,11 @@
 ; RV32-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e16, m4, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v12
 ; 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 zero, zero, e16, m4, ta, ma
-; 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, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v12
@@ -321,10 +321,10 @@
 ; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
 ; RV32-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-256-NEXT:    vid.v v10
 ; 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:    vid.v v10
 ; RV32-BITS-256-NEXT:    vrsub.vx v10, v10, a0
 ; RV32-BITS-256-NEXT:    vrgather.vv v12, v8, v10
 ; RV32-BITS-256-NEXT:    vand.vi v8, v12, 1
@@ -336,10 +336,10 @@
 ; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
 ; RV32-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-512-NEXT:    vid.v v10
 ; 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:    vid.v v10
 ; RV32-BITS-512-NEXT:    vrsub.vx v10, v10, a0
 ; RV32-BITS-512-NEXT:    vrgather.vv v12, v8, v10
 ; RV32-BITS-512-NEXT:    vand.vi v8, v12, 1
@@ -351,11 +351,11 @@
 ; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e16, m4, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v12
 ; 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 zero, zero, e16, m4, ta, ma
-; 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, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v12
@@ -368,10 +368,10 @@
 ; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
 ; RV64-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-256-NEXT:    vid.v v10
 ; 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:    vid.v v10
 ; RV64-BITS-256-NEXT:    vrsub.vx v10, v10, a0
 ; RV64-BITS-256-NEXT:    vrgather.vv v12, v8, v10
 ; RV64-BITS-256-NEXT:    vand.vi v8, v12, 1
@@ -383,10 +383,10 @@
 ; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
 ; RV64-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-512-NEXT:    vid.v v10
 ; 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:    vid.v v10
 ; RV64-BITS-512-NEXT:    vrsub.vx v10, v10, a0
 ; RV64-BITS-512-NEXT:    vrgather.vv v12, v8, v10
 ; RV64-BITS-512-NEXT:    vand.vi v8, v12, 1
@@ -402,11 +402,11 @@
 ; RV32-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e16, m8, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v16
 ; 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 zero, zero, e16, m8, ta, ma
-; 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, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v12, v8, v16
@@ -419,10 +419,10 @@
 ; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
 ; RV32-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-256-NEXT:    vid.v v12
 ; 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:    vid.v v12
 ; RV32-BITS-256-NEXT:    vrsub.vx v12, v12, a0
 ; RV32-BITS-256-NEXT:    vrgather.vv v16, v8, v12
 ; RV32-BITS-256-NEXT:    vand.vi v8, v16, 1
@@ -434,10 +434,10 @@
 ; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
 ; RV32-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-512-NEXT:    vid.v v12
 ; 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:    vid.v v12
 ; RV32-BITS-512-NEXT:    vrsub.vx v12, v12, a0
 ; RV32-BITS-512-NEXT:    vrgather.vv v16, v8, v12
 ; RV32-BITS-512-NEXT:    vand.vi v8, v16, 1
@@ -449,11 +449,11 @@
 ; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli zero, zero, e16, m8, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v16
 ; 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 zero, zero, e16, m8, ta, ma
-; 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, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v12, v8, v16
@@ -466,10 +466,10 @@
 ; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
 ; RV64-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-256-NEXT:    vid.v v12
 ; 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:    vid.v v12
 ; RV64-BITS-256-NEXT:    vrsub.vx v12, v12, a0
 ; RV64-BITS-256-NEXT:    vrgather.vv v16, v8, v12
 ; RV64-BITS-256-NEXT:    vand.vi v8, v16, 1
@@ -481,10 +481,10 @@
 ; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
 ; RV64-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-512-NEXT:    vid.v v12
 ; 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:    vid.v v12
 ; RV64-BITS-512-NEXT:    vrsub.vx v12, v12, a0
 ; RV64-BITS-512-NEXT:    vrgather.vv v16, v8, v12
 ; RV64-BITS-512-NEXT:    vand.vi v8, v16, 1
@@ -500,11 +500,11 @@
 ; RV32-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v16
 ; 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, ma
-; 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, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v28, v8, v16
@@ -519,10 +519,10 @@
 ; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
 ; RV32-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-256-NEXT:    vid.v v16
 ; 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:    vid.v v16
 ; RV32-BITS-256-NEXT:    vrsub.vx v16, v16, a0
 ; RV32-BITS-256-NEXT:    vrgather.vv v24, v8, v16
 ; RV32-BITS-256-NEXT:    vand.vi v8, v24, 1
@@ -534,11 +534,11 @@
 ; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
 ; RV32-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV32-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
+; RV32-BITS-512-NEXT:    vid.v v16
 ; 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, ma
-; RV32-BITS-512-NEXT:    vid.v v16
 ; RV32-BITS-512-NEXT:    vrsub.vx v16, v16, a0
 ; RV32-BITS-512-NEXT:    vrgather.vv v28, v8, v16
 ; RV32-BITS-512-NEXT:    vrgather.vv v24, v12, v16
@@ -552,11 +552,11 @@
 ; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-UNKNOWN-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v16
 ; 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, ma
-; 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, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v28, v8, v16
@@ -571,10 +571,10 @@
 ; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
 ; RV64-BITS-256-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-256-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-256-NEXT:    vid.v v16
 ; 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:    vid.v v16
 ; RV64-BITS-256-NEXT:    vrsub.vx v16, v16, a0
 ; RV64-BITS-256-NEXT:    vrgather.vv v24, v8, v16
 ; RV64-BITS-256-NEXT:    vand.vi v8, v24, 1
@@ -586,11 +586,11 @@
 ; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
 ; RV64-BITS-512-NEXT:    vmv.v.i v8, 0
 ; RV64-BITS-512-NEXT:    vmerge.vim v8, v8, 1, v0
+; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
+; RV64-BITS-512-NEXT:    vid.v v16
 ; 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, ma
-; RV64-BITS-512-NEXT:    vid.v v16
 ; RV64-BITS-512-NEXT:    vrsub.vx v16, v16, a0
 ; RV64-BITS-512-NEXT:    vrgather.vv v28, v8, v16
 ; RV64-BITS-512-NEXT:    vrgather.vv v24, v12, v16
@@ -609,11 +609,11 @@
 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:    vsetvli a0, zero, e16, mf4, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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, ma
-; 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, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
@@ -622,11 +622,11 @@
 ;
 ; RV32-BITS-256-LABEL: reverse_nxv1i8:
 ; RV32-BITS-256:       # %bb.0:
+; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
+; RV32-BITS-256-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -634,11 +634,11 @@
 ;
 ; RV32-BITS-512-LABEL: reverse_nxv1i8:
 ; RV32-BITS-512:       # %bb.0:
+; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
+; RV32-BITS-512-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -646,11 +646,11 @@
 ;
 ; RV64-BITS-UNKNOWN-LABEL: reverse_nxv1i8:
 ; RV64-BITS-UNKNOWN:       # %bb.0:
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e16, mf4, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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, ma
-; 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, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
@@ -659,11 +659,11 @@
 ;
 ; RV64-BITS-256-LABEL: reverse_nxv1i8:
 ; RV64-BITS-256:       # %bb.0:
+; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
+; RV64-BITS-256-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -671,11 +671,11 @@
 ;
 ; RV64-BITS-512-LABEL: reverse_nxv1i8:
 ; RV64-BITS-512:       # %bb.0:
+; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
+; RV64-BITS-512-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -687,11 +687,11 @@
 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:    vsetvli a0, zero, e16, mf2, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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, ma
-; 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, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
@@ -700,11 +700,11 @@
 ;
 ; RV32-BITS-256-LABEL: reverse_nxv2i8:
 ; RV32-BITS-256:       # %bb.0:
+; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
+; RV32-BITS-256-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -712,11 +712,11 @@
 ;
 ; RV32-BITS-512-LABEL: reverse_nxv2i8:
 ; RV32-BITS-512:       # %bb.0:
+; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
+; RV32-BITS-512-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -724,11 +724,11 @@
 ;
 ; RV64-BITS-UNKNOWN-LABEL: reverse_nxv2i8:
 ; RV64-BITS-UNKNOWN:       # %bb.0:
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e16, mf2, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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, ma
-; 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, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
@@ -737,11 +737,11 @@
 ;
 ; RV64-BITS-256-LABEL: reverse_nxv2i8:
 ; RV64-BITS-256:       # %bb.0:
+; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
+; RV64-BITS-256-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -749,11 +749,11 @@
 ;
 ; RV64-BITS-512-LABEL: reverse_nxv2i8:
 ; RV64-BITS-512:       # %bb.0:
+; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
+; RV64-BITS-512-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -765,11 +765,11 @@
 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:    vsetvli a0, zero, e16, m1, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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, ma
-; 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, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
@@ -778,11 +778,11 @@
 ;
 ; RV32-BITS-256-LABEL: reverse_nxv4i8:
 ; RV32-BITS-256:       # %bb.0:
+; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
+; RV32-BITS-256-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -790,11 +790,11 @@
 ;
 ; RV32-BITS-512-LABEL: reverse_nxv4i8:
 ; RV32-BITS-512:       # %bb.0:
+; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
+; RV32-BITS-512-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -802,11 +802,11 @@
 ;
 ; RV64-BITS-UNKNOWN-LABEL: reverse_nxv4i8:
 ; RV64-BITS-UNKNOWN:       # %bb.0:
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v9
 ; 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, ma
-; 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, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v9, v8, v10
@@ -815,11 +815,11 @@
 ;
 ; RV64-BITS-256-LABEL: reverse_nxv4i8:
 ; RV64-BITS-256:       # %bb.0:
+; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
+; RV64-BITS-256-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -827,11 +827,11 @@
 ;
 ; RV64-BITS-512-LABEL: reverse_nxv4i8:
 ; RV64-BITS-512:       # %bb.0:
+; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
+; RV64-BITS-512-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -915,11 +915,11 @@
 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:    vsetvli a0, zero, e16, m4, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v12
 ; 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, ma
-; 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, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v12
@@ -928,11 +928,11 @@
 ;
 ; RV32-BITS-256-LABEL: reverse_nxv16i8:
 ; RV32-BITS-256:       # %bb.0:
+; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
+; RV32-BITS-256-NEXT:    vid.v v10
 ; 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, ma
-; 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
@@ -940,11 +940,11 @@
 ;
 ; RV32-BITS-512-LABEL: reverse_nxv16i8:
 ; RV32-BITS-512:       # %bb.0:
+; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
+; RV32-BITS-512-NEXT:    vid.v v10
 ; 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, ma
-; 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
@@ -952,11 +952,11 @@
 ;
 ; RV64-BITS-UNKNOWN-LABEL: reverse_nxv16i8:
 ; RV64-BITS-UNKNOWN:       # %bb.0:
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e16, m4, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v12
 ; 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, ma
-; 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, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v10, v8, v12
@@ -965,11 +965,11 @@
 ;
 ; RV64-BITS-256-LABEL: reverse_nxv16i8:
 ; RV64-BITS-256:       # %bb.0:
+; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
+; RV64-BITS-256-NEXT:    vid.v v10
 ; 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, ma
-; 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
@@ -977,11 +977,11 @@
 ;
 ; RV64-BITS-512-LABEL: reverse_nxv16i8:
 ; RV64-BITS-512:       # %bb.0:
+; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
+; RV64-BITS-512-NEXT:    vid.v v10
 ; 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, ma
-; 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
@@ -993,11 +993,11 @@
 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:    vsetvli a0, zero, e16, m8, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v16
 ; 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, ma
-; 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, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v12, v8, v16
@@ -1006,11 +1006,11 @@
 ;
 ; RV32-BITS-256-LABEL: reverse_nxv32i8:
 ; RV32-BITS-256:       # %bb.0:
+; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
+; RV32-BITS-256-NEXT:    vid.v v12
 ; 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, ma
-; 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
@@ -1018,11 +1018,11 @@
 ;
 ; RV32-BITS-512-LABEL: reverse_nxv32i8:
 ; RV32-BITS-512:       # %bb.0:
+; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
+; RV32-BITS-512-NEXT:    vid.v v12
 ; 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, ma
-; 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
@@ -1030,11 +1030,11 @@
 ;
 ; RV64-BITS-UNKNOWN-LABEL: reverse_nxv32i8:
 ; RV64-BITS-UNKNOWN:       # %bb.0:
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v16
 ; 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, ma
-; 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, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v12, v8, v16
@@ -1043,11 +1043,11 @@
 ;
 ; RV64-BITS-256-LABEL: reverse_nxv32i8:
 ; RV64-BITS-256:       # %bb.0:
+; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
+; RV64-BITS-256-NEXT:    vid.v v12
 ; 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, ma
-; 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
@@ -1055,11 +1055,11 @@
 ;
 ; RV64-BITS-512-LABEL: reverse_nxv32i8:
 ; RV64-BITS-512:       # %bb.0:
+; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
+; RV64-BITS-512-NEXT:    vid.v v12
 ; 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, ma
-; 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
@@ -1071,11 +1071,11 @@
 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:    vsetvli a0, zero, e16, m8, ta, ma
+; RV32-BITS-UNKNOWN-NEXT:    vid.v v16
 ; 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, ma
-; 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, ma
 ; RV32-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v20, v8, v24
@@ -1085,11 +1085,11 @@
 ;
 ; RV32-BITS-256-LABEL: reverse_nxv64i8:
 ; RV32-BITS-256:       # %bb.0:
+; RV32-BITS-256-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
+; RV32-BITS-256-NEXT:    vid.v v16
 ; 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, ma
-; 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
@@ -1097,11 +1097,11 @@
 ;
 ; RV32-BITS-512-LABEL: reverse_nxv64i8:
 ; RV32-BITS-512:       # %bb.0:
+; RV32-BITS-512-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
+; RV32-BITS-512-NEXT:    vid.v v16
 ; 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, ma
-; 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
@@ -1110,11 +1110,11 @@
 ;
 ; RV64-BITS-UNKNOWN-LABEL: reverse_nxv64i8:
 ; RV64-BITS-UNKNOWN:       # %bb.0:
+; RV64-BITS-UNKNOWN-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
+; RV64-BITS-UNKNOWN-NEXT:    vid.v v16
 ; 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, ma
-; 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, ma
 ; RV64-BITS-UNKNOWN-NEXT:    vrgatherei16.vv v20, v8, v24
@@ -1124,11 +1124,11 @@
 ;
 ; RV64-BITS-256-LABEL: reverse_nxv64i8:
 ; RV64-BITS-256:       # %bb.0:
+; RV64-BITS-256-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
+; RV64-BITS-256-NEXT:    vid.v v16
 ; 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, ma
-; 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
@@ -1136,11 +1136,11 @@
 ;
 ; RV64-BITS-512-LABEL: reverse_nxv64i8:
 ; RV64-BITS-512:       # %bb.0:
+; RV64-BITS-512-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
+; RV64-BITS-512-NEXT:    vid.v v16
 ; 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, ma
-; 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
@@ -1153,11 +1153,11 @@
 define <vscale x 1 x i16> @reverse_nxv1i16(<vscale x 1 x i16> %a) {
 ; CHECK-LABEL: reverse_nxv1i16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1169,11 +1169,11 @@
 define <vscale x 2 x i16> @reverse_nxv2i16(<vscale x 2 x i16> %a) {
 ; CHECK-LABEL: reverse_nxv2i16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1185,11 +1185,11 @@
 define <vscale x 4 x i16> @reverse_nxv4i16(<vscale x 4 x i16> %a) {
 ; CHECK-LABEL: reverse_nxv4i16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1216,11 +1216,11 @@
 define <vscale x 16 x i16> @reverse_nxv16i16(<vscale x 16 x i16> %a) {
 ; CHECK-LABEL: reverse_nxv16i16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-NEXT:    vid.v v12
 ; 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, ma
-; 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
@@ -1232,11 +1232,11 @@
 define <vscale x 32 x i16> @reverse_nxv32i16(<vscale x 32 x i16> %a) {
 ; CHECK-LABEL: reverse_nxv32i16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-NEXT:    vid.v v16
 ; 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, ma
-; 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
@@ -1248,11 +1248,11 @@
 define <vscale x 1 x i32> @reverse_nxv1i32(<vscale x 1 x i32> %a) {
 ; CHECK-LABEL: reverse_nxv1i32:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1264,11 +1264,11 @@
 define <vscale x 2 x i32> @reverse_nxv2i32(<vscale x 2 x i32> %a) {
 ; CHECK-LABEL: reverse_nxv2i32:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1280,11 +1280,11 @@
 define <vscale x 4 x i32> @reverse_nxv4i32(<vscale x 4 x i32> %a) {
 ; CHECK-LABEL: reverse_nxv4i32:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vid.v v10
 ; 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, ma
-; 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
@@ -1311,11 +1311,11 @@
 define <vscale x 16 x i32> @reverse_nxv16i32(<vscale x 16 x i32> %a) {
 ; CHECK-LABEL: reverse_nxv16i32:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT:    vid.v v16
 ; 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, ma
-; 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
@@ -1327,11 +1327,11 @@
 define <vscale x 1 x i64> @reverse_nxv1i64(<vscale x 1 x i64> %a) {
 ; CHECK-LABEL: reverse_nxv1i64:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1343,11 +1343,11 @@
 define <vscale x 2 x i64> @reverse_nxv2i64(<vscale x 2 x i64> %a) {
 ; CHECK-LABEL: reverse_nxv2i64:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vid.v v10
 ; 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, ma
-; 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
@@ -1359,11 +1359,11 @@
 define <vscale x 4 x i64> @reverse_nxv4i64(<vscale x 4 x i64> %a) {
 ; CHECK-LABEL: reverse_nxv4i64:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vid.v v12
 ; 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, ma
-; 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
@@ -1394,11 +1394,11 @@
 define <vscale x 1 x half> @reverse_nxv1f16(<vscale x 1 x half> %a) {
 ; CHECK-LABEL: reverse_nxv1f16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1410,11 +1410,11 @@
 define <vscale x 2 x half> @reverse_nxv2f16(<vscale x 2 x half> %a) {
 ; CHECK-LABEL: reverse_nxv2f16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1426,11 +1426,11 @@
 define <vscale x 4 x half> @reverse_nxv4f16(<vscale x 4 x half> %a) {
 ; CHECK-LABEL: reverse_nxv4f16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1457,11 +1457,11 @@
 define <vscale x 16 x half> @reverse_nxv16f16(<vscale x 16 x half> %a) {
 ; CHECK-LABEL: reverse_nxv16f16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-NEXT:    vid.v v12
 ; 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, ma
-; 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
@@ -1473,11 +1473,11 @@
 define <vscale x 32 x half> @reverse_nxv32f16(<vscale x 32 x half> %a) {
 ; CHECK-LABEL: reverse_nxv32f16:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-NEXT:    vid.v v16
 ; 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, ma
-; 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
@@ -1489,11 +1489,11 @@
 define <vscale x 1 x float> @reverse_nxv1f32(<vscale x 1 x float> %a) {
 ; CHECK-LABEL: reverse_nxv1f32:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1505,11 +1505,11 @@
 define <vscale x 2 x float> @reverse_nxv2f32(<vscale x 2 x float> %a) {
 ; CHECK-LABEL: reverse_nxv2f32:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1521,11 +1521,11 @@
 define <vscale x 4 x float> @reverse_nxv4f32(<vscale x 4 x float> %a) {
 ; CHECK-LABEL: reverse_nxv4f32:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vid.v v10
 ; 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, ma
-; 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
@@ -1552,11 +1552,11 @@
 define <vscale x 16 x float> @reverse_nxv16f32(<vscale x 16 x float> %a) {
 ; CHECK-LABEL: reverse_nxv16f32:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT:    vid.v v16
 ; 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, ma
-; 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
@@ -1568,11 +1568,11 @@
 define <vscale x 1 x double> @reverse_nxv1f64(<vscale x 1 x double> %a) {
 ; CHECK-LABEL: reverse_nxv1f64:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vid.v v9
 ; 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, ma
-; 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
@@ -1584,11 +1584,11 @@
 define <vscale x 2 x double> @reverse_nxv2f64(<vscale x 2 x double> %a) {
 ; CHECK-LABEL: reverse_nxv2f64:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vid.v v10
 ; 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, ma
-; 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
@@ -1600,11 +1600,11 @@
 define <vscale x 4 x double> @reverse_nxv4f64(<vscale x 4 x double> %a) {
 ; CHECK-LABEL: reverse_nxv4f64:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vid.v v12
 ; 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, ma
-; 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
@@ -1633,11 +1633,11 @@
 define <vscale x 3 x i64> @reverse_nxv3i64(<vscale x 3 x i64> %a) {
 ; CHECK-LABEL: reverse_nxv3i64:
 ; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vid.v v12
 ; 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, ma
-; CHECK-NEXT:    vid.v v12
 ; CHECK-NEXT:    vrsub.vx v12, v12, a0
 ; CHECK-NEXT:    vrgather.vv v16, v8, v12
 ; CHECK-NEXT:    vmv1r.v v8, v17
Index: llvm/test/CodeGen/RISCV/rvv/stepvector.ll
===================================================================
--- llvm/test/CodeGen/RISCV/rvv/stepvector.ll
+++ llvm/test/CodeGen/RISCV/rvv/stepvector.ll
@@ -501,8 +501,8 @@
 ; RV32-NEXT:    lui a0, 797989
 ; RV32-NEXT:    addi a0, a0, -683
 ; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; RV32-NEXT:    addi a0, sp, 8
-; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v8, (a0), zero
 ; RV32-NEXT:    vid.v v16
 ; RV32-NEXT:    vmul.vv v8, v16, v8
@@ -552,8 +552,8 @@
 ; RV32-NEXT:    sw zero, 12(sp)
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; RV32-NEXT:    addi a0, sp, 8
-; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vid.v v8
 ; RV32-NEXT:    vadd.vv v16, v8, v16
@@ -562,9 +562,9 @@
 ;
 ; RV64-LABEL: stepvector_nxv16i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    csrr a0, vlenb
-; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vid.v v8
+; RV64-NEXT:    csrr a0, vlenb
 ; RV64-NEXT:    vadd.vx v16, v8, a0
 ; RV64-NEXT:    ret
   %v = call <vscale x 16 x i64> @llvm.experimental.stepvector.nxv16i64()
@@ -627,13 +627,13 @@
 ;
 ; RV64-LABEL: mul_stepvector_nxv16i64:
 ; RV64:       # %bb.0: # %entry
-; RV64-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
-; RV64-NEXT:    vid.v v8
-; RV64-NEXT:    li a0, 3
-; RV64-NEXT:    vmul.vx v8, v8, a0
 ; RV64-NEXT:    csrr a0, vlenb
 ; RV64-NEXT:    slli a1, a0, 1
 ; RV64-NEXT:    add a0, a1, a0
+; RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
+; RV64-NEXT:    vid.v v8
+; RV64-NEXT:    li a1, 3
+; RV64-NEXT:    vmul.vx v8, v8, a1
 ; RV64-NEXT:    vadd.vx v16, v8, a0
 ; RV64-NEXT:    ret
 entry:
@@ -667,8 +667,8 @@
 ; RV32-NEXT:    mulhu a0, a0, a2
 ; RV32-NEXT:    add a0, a0, a1
 ; RV32-NEXT:    sw a0, 12(sp)
+; RV32-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; RV32-NEXT:    addi a0, sp, 8
-; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v8, (a0), zero
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vid.v v24