[RISCV] Add support for fixed vector reductions.

[craig.topper]

Summary

I've included tests that require type legalization to split the
vector. The i64 version of these scalarizes on RV32 due to type
legalization visiting the result before the vector type. So we
have to abort our custom expansion to avoid creating target
specific nodes with an illegal type. Then type legalization ends
up scalarizing. We might be able to fix this by doing custom
splitting for large vectors in our handler to get down to a legal
type.

[craig.topper]

clang-format

[frasercrmck]

Wow yeah that's some heavy scalarization. I think I came across some situations where custom vector legalization may be required, as you say. It just felt odd, though. Do any other targets have to do this, other than simple hi/lo splitting?

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

2447

This (and further down) can now be simplified to = getContainerForFixedLengthVector(VecVT); (after rebasing)

[craig.topper]

rebase

[craig.topper]

In D98102#2610721, @frasercrmck wrote:

Wow yeah that's some heavy scalarization. I think I came across some situations where custom vector legalization may be required, as you say. It just felt odd, though. Do any other targets have to do this, other than simple hi/lo splitting?

Only 3 targets use the VECREDUCE nodes today, AArch64, ARM, and RISCV. X86 does not use them yet. AArch64 doesn't have this problem because their largest integer type is their largest vector element type. ARM doesn't appear to support i64 reductions.

[frasercrmck]

In D98102#2611638, @craig.topper wrote:

Only 3 targets use the VECREDUCE nodes today, AArch64, ARM, and RISCV. X86 does not use them yet. AArch64 doesn't have this problem because their largest integer type is their largest vector element type. ARM doesn't appear to support i64 reductions.

Ah right okay, thanks. Anyway, the changes LGTM. I don't know if the size of the test files is a problem. It's not unprecedented but it still makes me uncomfortable.

llvm/lib/Target/RISCV/RISCVISelLowering.h

 //===-- RISCVISelLowering.h - RISCV DAG Lowering Interface ------*- C++ -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 //
enum NodeType : unsigned {
   // performed over the first vector operand plus the first element of the
   // second vector operand. The first operand is an unconstrained vector type,
   // and the result and second operand's types are expected to be the
   // corresponding full-width LMUL=1 type for the first operand:
   //   nxv8i8 = vecreduce_add nxv32i8, nxv8i8
   //   nxv2i32 = vecreduce_add nxv8i32, nxv2i32
   // The different in types does introduce extra vsetvli instructions but
   // similarly it reduces the number of registers consumed per reduction.
-  VECREDUCE_ADD,
-  VECREDUCE_UMAX,
-  VECREDUCE_SMAX,
-  VECREDUCE_UMIN,
-  VECREDUCE_SMIN,
-  VECREDUCE_AND,
-  VECREDUCE_OR,
-  VECREDUCE_XOR,
-  VECREDUCE_FADD,
-  VECREDUCE_SEQ_FADD,
+  // Also has a mask and VL operand.
+  VECREDUCE_ADD_VL,
+  VECREDUCE_UMAX_VL,
+  VECREDUCE_SMAX_VL,
+  VECREDUCE_UMIN_VL,
+  VECREDUCE_SMIN_VL,
+  VECREDUCE_AND_VL,
+  VECREDUCE_OR_VL,
+  VECREDUCE_XOR_VL,
+  VECREDUCE_FADD_VL,
+  VECREDUCE_SEQ_FADD_VL,
 
   // Vector binary and unary ops with a mask as a third operand, and VL as a
   // fourth operand.
   // FIXME: Can we replace these with ISD::VP_*?
   ADD_VL,
   AND_VL,
   MUL_VL,
   OR_VL,

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

 //===-- RISCVISelLowering.cpp - RISCV DAG Lowering Implementation  --------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 //
if (Subtarget.useRVVForFixedLengthVectors()) {
 
         setOperationAction(ISD::VSELECT, VT, Custom);
 
         setOperationAction(ISD::ANY_EXTEND, VT, Custom);
         setOperationAction(ISD::SIGN_EXTEND, VT, Custom);
         setOperationAction(ISD::ZERO_EXTEND, VT, Custom);
 
         setOperationAction(ISD::BITCAST, VT, Custom);
+
+        // Custom-lower reduction operations to set up the corresponding custom
+        // nodes' operands.
+        setOperationAction(ISD::VECREDUCE_ADD, VT, Custom);
+        setOperationAction(ISD::VECREDUCE_AND, VT, Custom);
+        setOperationAction(ISD::VECREDUCE_OR, VT, Custom);
+        setOperationAction(ISD::VECREDUCE_XOR, VT, Custom);
+        setOperationAction(ISD::VECREDUCE_SMAX, VT, Custom);
+        setOperationAction(ISD::VECREDUCE_SMIN, VT, Custom);
+        setOperationAction(ISD::VECREDUCE_UMAX, VT, Custom);
+        setOperationAction(ISD::VECREDUCE_UMIN, VT, Custom);
       }
 
       for (MVT VT : MVT::fp_fixedlen_vector_valuetypes()) {
         if (!useRVVForFixedLengthVectorVT(VT))
           continue;
 
         // By default everything must be expanded.
         for (unsigned Op = 0; Op < ISD::BUILTIN_OP_END; ++Op)
if (Subtarget.useRVVForFixedLengthVectors()) {
         setOperationAction(ISD::FP_EXTEND, VT, Custom);
 
         for (auto CC : VFPCCToExpand)
           setCondCodeAction(CC, VT, Expand);
 
         setOperationAction(ISD::VSELECT, VT, Custom);
 
         setOperationAction(ISD::BITCAST, VT, Custom);
+
+        setOperationAction(ISD::VECREDUCE_FADD, VT, Custom);
+        setOperationAction(ISD::VECREDUCE_SEQ_FADD, VT, Custom);
       }
     }
   }
 
   // Function alignments.
   const Align FunctionAlignment(Subtarget.hasStdExtC() ? 2 : 4);
   setMinFunctionAlignment(FunctionAlignment);
   setPrefFunctionAlignment(FunctionAlignment);
return MVT::getScalableVectorVT(
       RISCV::RVVBitsPerBlock / VT.getVectorElementType().getSizeInBits());
 }
 
 static unsigned getRVVReductionOp(unsigned ISDOpcode) {
   switch (ISDOpcode) {
   default:
     llvm_unreachable("Unhandled reduction");
   case ISD::VECREDUCE_ADD:
-    return RISCVISD::VECREDUCE_ADD;
+    return RISCVISD::VECREDUCE_ADD_VL;
   case ISD::VECREDUCE_UMAX:
-    return RISCVISD::VECREDUCE_UMAX;
+    return RISCVISD::VECREDUCE_UMAX_VL;
   case ISD::VECREDUCE_SMAX:
-    return RISCVISD::VECREDUCE_SMAX;
+    return RISCVISD::VECREDUCE_SMAX_VL;
   case ISD::VECREDUCE_UMIN:
-    return RISCVISD::VECREDUCE_UMIN;
+    return RISCVISD::VECREDUCE_UMIN_VL;
   case ISD::VECREDUCE_SMIN:
-    return RISCVISD::VECREDUCE_SMIN;
+    return RISCVISD::VECREDUCE_SMIN_VL;
   case ISD::VECREDUCE_AND:
-    return RISCVISD::VECREDUCE_AND;
+    return RISCVISD::VECREDUCE_AND_VL;
   case ISD::VECREDUCE_OR:
-    return RISCVISD::VECREDUCE_OR;
+    return RISCVISD::VECREDUCE_OR_VL;
   case ISD::VECREDUCE_XOR:
-    return RISCVISD::VECREDUCE_XOR;
+    return RISCVISD::VECREDUCE_XOR_VL;
   }
 }
 
 // Take a (supported) standard ISD reduction opcode and transform it to a RISCV
 // reduction opcode. Note that this returns a vector type, which must be
 // further processed to access the scalar result in element 0.
 SDValue RISCVTargetLowering::lowerVECREDUCE(SDValue Op,
                                             SelectionDAG &DAG) const {
   SDLoc DL(Op);
-  assert(Op.getValueType().isSimple() &&
-         Op.getOperand(0).getValueType().isSimple() &&
-         "Unexpected vector-reduce lowering");
   MVT VecVT = Op.getOperand(0).getSimpleValueType();
   MVT VecEltVT = VecVT.getVectorElementType();
+
+  // Avoid creating vectors with illegal type.
+  if (!isTypeLegal(VecVT))
+    return SDValue();
+
   unsigned RVVOpcode = getRVVReductionOp(Op.getOpcode());
-  MVT M1VT = getLMUL1VT(VecVT);
+
+  SDValue Vec = Op.getOperand(0);
+
+  MVT ContainerVT = VecVT;
+  if (VecVT.isFixedLengthVector()) {
+    ContainerVT = getContainerForFixedLengthVector(VecVT);
+    Vec = convertToScalableVector(ContainerVT, Vec, DAG, Subtarget);
+  }
+
+  MVT M1VT = getLMUL1VT(ContainerVT);
+
+  SDValue Mask, VL;
+  std::tie(Mask, VL) = getDefaultVLOps(VecVT, ContainerVT, DL, DAG, Subtarget);
+
+  // FIXME: This is a VLMAX splat which might be too large and can prevent
+  // vsetvli removal.
   SDValue NeutralElem = DAG.getNeutralElement(
       ISD::getVecReduceBaseOpcode(Op.getOpcode()), DL, VecEltVT, SDNodeFlags());
   SDValue IdentitySplat = DAG.getSplatVector(M1VT, DL, NeutralElem);
   SDValue Reduction =
-      DAG.getNode(RVVOpcode, DL, M1VT, Op.getOperand(0), IdentitySplat);
+      DAG.getNode(RVVOpcode, DL, M1VT, Vec, IdentitySplat, Mask, VL);
   SDValue Elt0 = DAG.getNode(ISD::EXTRACT_VECTOR_ELT, DL, VecEltVT, Reduction,
                              DAG.getConstant(0, DL, Subtarget.getXLenVT()));
   return DAG.getSExtOrTrunc(Elt0, DL, Op.getValueType());
 }
 
 // Given a reduction op, this function returns the matching reduction opcode,
 // the vector SDValue and the scalar SDValue required to lower this to a
 // RISCVISD node.
 static std::tuple<unsigned, SDValue, SDValue>
 getRVVFPReductionOpAndOperands(SDValue Op, SelectionDAG &DAG, EVT EltVT) {
   SDLoc DL(Op);
   switch (Op.getOpcode()) {
   default:
     llvm_unreachable("Unhandled reduction");
   case ISD::VECREDUCE_FADD:
-    return std::make_tuple(RISCVISD::VECREDUCE_FADD, Op.getOperand(0),
+    return std::make_tuple(RISCVISD::VECREDUCE_FADD_VL, Op.getOperand(0),
                            DAG.getConstantFP(0.0, DL, EltVT));
   case ISD::VECREDUCE_SEQ_FADD:
-    return std::make_tuple(RISCVISD::VECREDUCE_SEQ_FADD, Op.getOperand(1),
+    return std::make_tuple(RISCVISD::VECREDUCE_SEQ_FADD_VL, Op.getOperand(1),
                            Op.getOperand(0));
   }
 }
 
 SDValue RISCVTargetLowering::lowerFPVECREDUCE(SDValue Op,
                                               SelectionDAG &DAG) const {
   SDLoc DL(Op);
   MVT VecEltVT = Op.getSimpleValueType();
 
   unsigned RVVOpcode;
   SDValue VectorVal, ScalarVal;
   std::tie(RVVOpcode, VectorVal, ScalarVal) =
       getRVVFPReductionOpAndOperands(Op, DAG, VecEltVT);
+  MVT VecVT = VectorVal.getSimpleValueType();
+
+  MVT ContainerVT = VecVT;
+  if (VecVT.isFixedLengthVector()) {
+    ContainerVT = getContainerForFixedLengthVector(VecVT);
+    VectorVal = convertToScalableVector(ContainerVT, VectorVal, DAG, Subtarget);
+  }
 
   MVT M1VT = getLMUL1VT(VectorVal.getSimpleValueType());
+
+  SDValue Mask, VL;
+  std::tie(Mask, VL) = getDefaultVLOps(VecVT, ContainerVT, DL, DAG, Subtarget);
+
+  // FIXME: This is a VLMAX splat which might be too large and can prevent
+  // vsetvli removal.
   SDValue ScalarSplat = DAG.getSplatVector(M1VT, DL, ScalarVal);
-  SDValue Reduction = DAG.getNode(RVVOpcode, DL, M1VT, VectorVal, ScalarSplat);
+  SDValue Reduction =
+      DAG.getNode(RVVOpcode, DL, M1VT, VectorVal, ScalarSplat, Mask, VL);
   return DAG.getNode(ISD::EXTRACT_VECTOR_ELT, DL, VecEltVT, Reduction,
                      DAG.getConstant(0, DL, Subtarget.getXLenVT()));
 }
 
 SDValue RISCVTargetLowering::lowerINSERT_SUBVECTOR(SDValue Op,
                                                    SelectionDAG &DAG) const {
   SDValue Vec = Op.getOperand(0);
   SDValue SubVec = Op.getOperand(1);
void RISCVTargetLowering::ReplaceNodeResults(SDNode *N,
   case ISD::VECREDUCE_XOR:
   case ISD::VECREDUCE_SMAX:
   case ISD::VECREDUCE_UMAX:
   case ISD::VECREDUCE_SMIN:
   case ISD::VECREDUCE_UMIN:
     // The custom-lowering for these nodes returns a vector whose first element
     // is the result of the reduction. Extract its first element and let the
     // legalization for EXTRACT_VECTOR_ELT do the rest of the job.
-    Results.push_back(lowerVECREDUCE(SDValue(N, 0), DAG));
+    if (SDValue V = lowerVECREDUCE(SDValue(N, 0), DAG))
+      Results.push_back(V);
     break;
   }
 }
 
 // A structure to hold one of the bit-manipulation patterns below. Together, a
 // SHL and non-SHL pattern may form a bit-manipulation pair on a single source:
 //   (or (and (shl x, 1), 0xAAAAAAAA),
 //       (and (srl x, 1), 0x55555555))
#define NODE_NAME_CASE(NODE)                                                   \
   NODE_NAME_CASE(READ_VLENB)
   NODE_NAME_CASE(TRUNCATE_VECTOR_VL)
   NODE_NAME_CASE(VLEFF)
   NODE_NAME_CASE(VLEFF_MASK)
   NODE_NAME_CASE(VSLIDEUP_VL)
   NODE_NAME_CASE(VSLIDEDOWN_VL)
   NODE_NAME_CASE(VID_VL)
   NODE_NAME_CASE(VFNCVT_ROD_VL)
-  NODE_NAME_CASE(VECREDUCE_ADD)
-  NODE_NAME_CASE(VECREDUCE_UMAX)
-  NODE_NAME_CASE(VECREDUCE_SMAX)
-  NODE_NAME_CASE(VECREDUCE_UMIN)
-  NODE_NAME_CASE(VECREDUCE_SMIN)
-  NODE_NAME_CASE(VECREDUCE_AND)
-  NODE_NAME_CASE(VECREDUCE_OR)
-  NODE_NAME_CASE(VECREDUCE_XOR)
-  NODE_NAME_CASE(VECREDUCE_FADD)
-  NODE_NAME_CASE(VECREDUCE_SEQ_FADD)
+  NODE_NAME_CASE(VECREDUCE_ADD_VL)
+  NODE_NAME_CASE(VECREDUCE_UMAX_VL)
+  NODE_NAME_CASE(VECREDUCE_SMAX_VL)
+  NODE_NAME_CASE(VECREDUCE_UMIN_VL)
+  NODE_NAME_CASE(VECREDUCE_SMIN_VL)
+  NODE_NAME_CASE(VECREDUCE_AND_VL)
+  NODE_NAME_CASE(VECREDUCE_OR_VL)
+  NODE_NAME_CASE(VECREDUCE_XOR_VL)
+  NODE_NAME_CASE(VECREDUCE_FADD_VL)
+  NODE_NAME_CASE(VECREDUCE_SEQ_FADD_VL)
   NODE_NAME_CASE(ADD_VL)
   NODE_NAME_CASE(AND_VL)
   NODE_NAME_CASE(MUL_VL)
   NODE_NAME_CASE(OR_VL)
   NODE_NAME_CASE(SDIV_VL)
   NODE_NAME_CASE(SHL_VL)
   NODE_NAME_CASE(SREM_VL)
   NODE_NAME_CASE(SRA_VL)

llvm/lib/Target/RISCV/RISCVInstrInfoVSDPatterns.td

 def SplatPat       : ComplexPattern<vAny, 1, "selectVSplat",      [splat_vector, rv32_splat_i64], [], 1>;
 def SplatPat_simm5 : ComplexPattern<vAny, 1, "selectVSplatSimm5", [splat_vector, rv32_splat_i64], [], 2>;
 def SplatPat_uimm5 : ComplexPattern<vAny, 1, "selectVSplatUimm5", [splat_vector, rv32_splat_i64], [], 2>;
 
 class SwapHelper<dag Prefix, dag A, dag B, dag Suffix, bit swap> {
    dag Value = !con(Prefix, !if(swap, B, A), !if(swap, A, B), Suffix);
 }
 
-def SDTRVVVecReduce : SDTypeProfile<1, 2, [
-  SDTCisVec<0>, SDTCisVec<1>, SDTCisSameAs<0, 2>
-]>;
-
-foreach kind = ["ADD", "UMAX", "SMAX", "UMIN", "SMIN", "AND", "OR", "XOR",
-                "FADD", "SEQ_FADD"] in
-  def rvv_vecreduce_#kind : SDNode<"RISCVISD::VECREDUCE_"#kind, SDTRVVVecReduce>;
-
 multiclass VPatUSLoadStoreSDNode<ValueType type,
                                  int sew,
                                  LMULInfo vlmul,
                                  OutPatFrag avl,
                                  VReg reg_class>
 {
   defvar load_instr = !cast<Instruction>("PseudoVLE"#sew#"_V_"#vlmul.MX);
   defvar store_instr = !cast<Instruction>("PseudoVSE"#sew#"_V_"#vlmul.MX);
foreach vtiToWti = AllWidenableIntToFloatVectors in {
     defvar vti = vtiToWti.Vti;
     defvar fwti = vtiToWti.Wti;
     def : Pat<(vti.Vector (vop (fwti.Vector fwti.RegClass:$rs1))),
               (!cast<Instruction>(instruction_name#"_"#vti.LMul.MX)
                   fwti.RegClass:$rs1, vti.AVL, vti.SEW)>;
   }
 }
 
-multiclass VPatReductionSDNode<SDNode vop, string instruction_name, bit is_float> {
-  foreach vti = !if(is_float, AllFloatVectors, AllIntegerVectors) in {
-    defvar vti_m1 = !cast<VTypeInfo>(!if(is_float, "VF", "VI") # vti.SEW # "M1");
-    def: Pat<(vti_m1.Vector (vop (vti.Vector vti.RegClass:$rs1), VR:$rs2)),
-        (!cast<Instruction>(instruction_name#"_VS_"#vti.LMul.MX)
-            (vti_m1.Vector (IMPLICIT_DEF)),
-            (vti.Vector vti.RegClass:$rs1),
-            (vti_m1.Vector VR:$rs2),
-            vti.AVL, vti.SEW)>;
-  }
-}
-
 //===----------------------------------------------------------------------===//
 // Patterns.
 //===----------------------------------------------------------------------===//
 
 let Predicates = [HasStdExtV] in {
 
 // 7.4. Vector Unit-Stride Instructions
 foreach vti = !listconcat(FractionalGroupIntegerVectors,
def : Pat<(vti.Vector (vselect (vti.Mask VMV0:$vm), (SplatPat XLenVT:$rs1),
                  vti.RegClass:$rs2, GPR:$rs1, VMV0:$vm, vti.AVL, vti.SEW)>;
 
   def : Pat<(vti.Vector (vselect (vti.Mask VMV0:$vm), (SplatPat_simm5 simm5:$rs1),
                                                       vti.RegClass:$rs2)),
             (!cast<Instruction>("PseudoVMERGE_VIM_"#vti.LMul.MX)
                  vti.RegClass:$rs2, simm5:$rs1, VMV0:$vm, vti.AVL, vti.SEW)>;
 }
 
-// 15.1. Vector Single-Width Integer Reduction Instructions
-defm "" : VPatReductionSDNode<rvv_vecreduce_ADD,  "PseudoVREDSUM", /*is_float*/0>;
-defm "" : VPatReductionSDNode<rvv_vecreduce_UMAX, "PseudoVREDMAXU", /*is_float*/0>;
-defm "" : VPatReductionSDNode<rvv_vecreduce_SMAX, "PseudoVREDMAX", /*is_float*/0>;
-defm "" : VPatReductionSDNode<rvv_vecreduce_UMIN, "PseudoVREDMINU", /*is_float*/0>;
-defm "" : VPatReductionSDNode<rvv_vecreduce_SMIN, "PseudoVREDMIN", /*is_float*/0>;
-defm "" : VPatReductionSDNode<rvv_vecreduce_AND,  "PseudoVREDAND", /*is_float*/0>;
-defm "" : VPatReductionSDNode<rvv_vecreduce_OR,   "PseudoVREDOR", /*is_float*/0>;
-defm "" : VPatReductionSDNode<rvv_vecreduce_XOR,  "PseudoVREDXOR", /*is_float*/0>;
-
-// 15.3. Vector Single-Width Floating-Point Reduction Instructions
-defm "" : VPatReductionSDNode<rvv_vecreduce_SEQ_FADD, "PseudoVFREDOSUM", /*is_float*/1>;
-defm "" : VPatReductionSDNode<rvv_vecreduce_FADD,     "PseudoVFREDSUM", /*is_float*/1>;
-
 // 16.1. Vector Mask-Register Logical Instructions
 foreach mti = AllMasks in {
   def : Pat<(mti.Mask (and VR:$rs1, VR:$rs2)),
             (!cast<Instruction>("PseudoVMAND_MM_"#mti.LMul.MX)
                  VR:$rs1, VR:$rs2, mti.AVL, mti.SEW)>;
   def : Pat<(mti.Mask (or VR:$rs1, VR:$rs2)),
             (!cast<Instruction>("PseudoVMOR_MM_"#mti.LMul.MX)
                  VR:$rs1, VR:$rs2, mti.AVL, mti.SEW)>;

llvm/lib/Target/RISCV/RISCVInstrInfoVVLPatterns.td

 
 def riscv_trunc_vector_vl : SDNode<"RISCVISD::TRUNCATE_VECTOR_VL",
                                    SDTypeProfile<1, 3, [SDTCisVec<0>,
                                                         SDTCisVec<1>,
                                                         SDTCisSameNumEltsAs<0, 2>,
                                                         SDTCVecEltisVT<2, i1>,
                                                         SDTCisVT<3, XLenVT>]>>;
 
+def SDTRVVVecReduce : SDTypeProfile<1, 4, [
+  SDTCisVec<0>, SDTCisVec<1>, SDTCisSameAs<0, 2>, SDTCVecEltisVT<3, i1>,
+  SDTCisSameNumEltsAs<1, 3>, SDTCisVT<4, XLenVT>
+]>;
+
+foreach kind = ["ADD", "UMAX", "SMAX", "UMIN", "SMIN", "AND", "OR", "XOR",
+                "FADD", "SEQ_FADD"] in
+  def rvv_vecreduce_#kind#_vl : SDNode<"RISCVISD::VECREDUCE_"#kind#"_VL", SDTRVVVecReduce>;
+
 // Ignore the vl operand.
 def SplatFPOp : PatFrag<(ops node:$op),
                         (riscv_vfmv_v_f_vl node:$op, srcvalue)>;
 
 def sew8simm5  : ComplexPattern<XLenVT, 1, "selectRVVSimm5<8>",  []>;
 def sew16simm5 : ComplexPattern<XLenVT, 1, "selectRVVSimm5<16>", []>;
 def sew32simm5 : ComplexPattern<XLenVT, 1, "selectRVVSimm5<32>", []>;
 def sew64simm5 : ComplexPattern<XLenVT, 1, "selectRVVSimm5<64>", []>;
foreach fvtiToFWti = AllWidenableFloatVectors in {
     def : Pat<(fvti.Vector (vop (iwti.Vector iwti.RegClass:$rs1),
                                 (iwti.Mask true_mask),
                                 (XLenVT (VLOp GPR:$vl)))),
               (!cast<Instruction>(instruction_name#"_"#fvti.LMul.MX)
                   iwti.RegClass:$rs1, GPR:$vl, fvti.SEW)>;
   }
 }
 
+multiclass VPatReductionVL<SDNode vop, string instruction_name, bit is_float> {
+  foreach vti = !if(is_float, AllFloatVectors, AllIntegerVectors) in {
+    defvar vti_m1 = !cast<VTypeInfo>(!if(is_float, "VF", "VI") # vti.SEW # "M1");
+    def: Pat<(vti_m1.Vector (vop (vti.Vector vti.RegClass:$rs1), VR:$rs2,
+                                 (vti.Mask true_mask),
+                                 (XLenVT (VLOp GPR:$vl)))),
+        (!cast<Instruction>(instruction_name#"_VS_"#vti.LMul.MX)
+            (vti_m1.Vector (IMPLICIT_DEF)),
+            (vti.Vector vti.RegClass:$rs1),
+            (vti_m1.Vector VR:$rs2),
+            GPR:$vl, vti.SEW)>;
+  }
+}
+
 //===----------------------------------------------------------------------===//
 // Patterns.
 //===----------------------------------------------------------------------===//
 
 let Predicates = [HasStdExtV] in {
 
 // 7.4. Vector Unit-Stride Instructions
 foreach vti = AllVectors in {
foreach vti = AllIntegerVectors in {
   def : Pat<(vti.Vector (riscv_vmv_v_x_vl (ImmPat XLenVT:$imm5),
                                           (XLenVT (VLOp GPR:$vl)))),
             (!cast<Instruction>("PseudoVMV_V_I_"#vti.LMul.MX)
              XLenVT:$imm5, GPR:$vl, vti.SEW)>;
 }
 
 } // Predicates = [HasStdExtV]
 
+// 15.1. Vector Single-Width Integer Reduction Instructions
+let Predicates = [HasStdExtV] in {
+defm "" : VPatReductionVL<rvv_vecreduce_ADD_vl,  "PseudoVREDSUM", /*is_float*/0>;
+defm "" : VPatReductionVL<rvv_vecreduce_UMAX_vl, "PseudoVREDMAXU", /*is_float*/0>;
+defm "" : VPatReductionVL<rvv_vecreduce_SMAX_vl, "PseudoVREDMAX", /*is_float*/0>;
+defm "" : VPatReductionVL<rvv_vecreduce_UMIN_vl, "PseudoVREDMINU", /*is_float*/0>;
+defm "" : VPatReductionVL<rvv_vecreduce_SMIN_vl, "PseudoVREDMIN", /*is_float*/0>;
+defm "" : VPatReductionVL<rvv_vecreduce_AND_vl,  "PseudoVREDAND", /*is_float*/0>;
+defm "" : VPatReductionVL<rvv_vecreduce_OR_vl,   "PseudoVREDOR", /*is_float*/0>;
+defm "" : VPatReductionVL<rvv_vecreduce_XOR_vl,  "PseudoVREDXOR", /*is_float*/0>;
+} // Predicates = [HasStdExtV]
+
+// 15.3. Vector Single-Width Floating-Point Reduction Instructions
+let Predicates = [HasStdExtV, HasStdExtF] in {
+defm "" : VPatReductionVL<rvv_vecreduce_SEQ_FADD_vl, "PseudoVFREDOSUM", /*is_float*/1>;
+defm "" : VPatReductionVL<rvv_vecreduce_FADD_vl,     "PseudoVFREDSUM", /*is_float*/1>;
+} // Predicates = [HasStdExtV, HasStdExtF]
+
 let Predicates = [HasStdExtV, HasStdExtF] in {
 
 // 14.2. Vector Single-Width Floating-Point Add/Subtract Instructions
 defm "" : VPatBinaryFPVL_VV_VF<riscv_fadd_vl, "PseudoVFADD">;
 defm "" : VPatBinaryFPVL_VV_VF<riscv_fsub_vl, "PseudoVFSUB">;
 defm "" : VPatBinaryFPVL_R_VF<riscv_fsub_vl, "PseudoVFRSUB">;
 
 // 14.4. Vector Single-Width Floating-Point Multiply/Divide Instructions

llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-fp.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -target-abi=ilp32d -mattr=+experimental-v,+experimental-zfh,+f,+d -verify-machineinstrs -riscv-v-vector-bits-min=128 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK
+; RUN: llc -mtriple=riscv64 -target-abi=lp64d -mattr=+experimental-v,+experimental-zfh,+f,+d -verify-machineinstrs -riscv-v-vector-bits-min=128  -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK
+
+declare half @llvm.vector.reduce.fadd.v1f16(half, <1 x half>)
+
+define half @vreduce_fadd_v1f16(<1 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_fadd_v1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    flh ft0, 0(a0)
+; CHECK-NEXT:    fadd.h fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <1 x half>, <1 x half>* %x
+  %red = call reassoc half @llvm.vector.reduce.fadd.v1f16(half %s, <1 x half> %v)
+  ret half %red
+}
+
+define half @vreduce_ord_fadd_v1f16(<1 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    flh ft0, 0(a0)
+; CHECK-NEXT:    fadd.h fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <1 x half>, <1 x half>* %x
+  %red = call half @llvm.vector.reduce.fadd.v1f16(half %s, <1 x half> %v)
+  ret half %red
+}
+
+declare half @llvm.vector.reduce.fadd.v2f16(half, <2 x half>)
+
+define half @vreduce_fadd_v2f16(<2 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_fadd_v2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.h fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <2 x half>, <2 x half>* %x
+  %red = call reassoc half @llvm.vector.reduce.fadd.v2f16(half %s, <2 x half> %v)
+  ret half %red
+}
+
+define half @vreduce_ord_fadd_v2f16(<2 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v26, fa0
+; CHECK-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x half>, <2 x half>* %x
+  %red = call half @llvm.vector.reduce.fadd.v2f16(half %s, <2 x half> %v)
+  ret half %red
+}
+
+declare half @llvm.vector.reduce.fadd.v4f16(half, <4 x half>)
+
+define half @vreduce_fadd_v4f16(<4 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_fadd_v4f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.h fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <4 x half>, <4 x half>* %x
+  %red = call reassoc half @llvm.vector.reduce.fadd.v4f16(half %s, <4 x half> %v)
+  ret half %red
+}
+
+define half @vreduce_ord_fadd_v4f16(<4 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v4f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v26, fa0
+; CHECK-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x half>, <4 x half>* %x
+  %red = call half @llvm.vector.reduce.fadd.v4f16(half %s, <4 x half> %v)
+  ret half %red
+}
+
+declare half @llvm.vector.reduce.fadd.v8f16(half, <8 x half>)
+
+define half @vreduce_fadd_v8f16(<8 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_fadd_v8f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.h fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <8 x half>, <8 x half>* %x
+  %red = call reassoc half @llvm.vector.reduce.fadd.v8f16(half %s, <8 x half> %v)
+  ret half %red
+}
+
+define half @vreduce_ord_fadd_v8f16(<8 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v8f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v26, fa0
+; CHECK-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x half>, <8 x half>* %x
+  %red = call half @llvm.vector.reduce.fadd.v8f16(half %s, <8 x half> %v)
+  ret half %red
+}
+
+declare half @llvm.vector.reduce.fadd.v16f16(half, <16 x half>)
+
+define half @vreduce_fadd_v16f16(<16 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_fadd_v16f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vle16.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.h fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <16 x half>, <16 x half>* %x
+  %red = call reassoc half @llvm.vector.reduce.fadd.v16f16(half %s, <16 x half> %v)
+  ret half %red
+}
+
+define half @vreduce_ord_fadd_v16f16(<16 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v16f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vle16.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x half>, <16 x half>* %x
+  %red = call half @llvm.vector.reduce.fadd.v16f16(half %s, <16 x half> %v)
+  ret half %red
+}
+
+declare half @llvm.vector.reduce.fadd.v32f16(half, <32 x half>)
+
+define half @vreduce_fadd_v32f16(<32 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_fadd_v32f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.h fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <32 x half>, <32 x half>* %x
+  %red = call reassoc half @llvm.vector.reduce.fadd.v32f16(half %s, <32 x half> %v)
+  ret half %red
+}
+
+define half @vreduce_ord_fadd_v32f16(<32 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v32f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x half>, <32 x half>* %x
+  %red = call half @llvm.vector.reduce.fadd.v32f16(half %s, <32 x half> %v)
+  ret half %red
+}
+
+declare half @llvm.vector.reduce.fadd.v64f16(half, <64 x half>)
+
+define half @vreduce_fadd_v64f16(<64 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_fadd_v64f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.h fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <64 x half>, <64 x half>* %x
+  %red = call reassoc half @llvm.vector.reduce.fadd.v64f16(half %s, <64 x half> %v)
+  ret half %red
+}
+
+define half @vreduce_ord_fadd_v64f16(<64 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v64f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x half>, <64 x half>* %x
+  %red = call half @llvm.vector.reduce.fadd.v64f16(half %s, <64 x half> %v)
+  ret half %red
+}
+
+declare half @llvm.vector.reduce.fadd.v128f16(half, <128 x half>)
+
+define half @vreduce_fadd_v128f16(<128 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_fadd_v128f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle16.v v16, (a0)
+; CHECK-NEXT:    vfadd.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.h fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <128 x half>, <128 x half>* %x
+  %red = call reassoc half @llvm.vector.reduce.fadd.v128f16(half %s, <128 x half> %v)
+  ret half %red
+}
+
+define half @vreduce_ord_fadd_v128f16(<128 x half>* %x, half %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v128f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, a0, 128
+; CHECK-NEXT:    addi a2, zero, 64
+; CHECK-NEXT:    vsetvli a3, a2, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a1)
+; CHECK-NEXT:    vle16.v v16, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetvli a0, a2, e16,m8,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v16, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, ft0
+; CHECK-NEXT:    vsetvli a0, a2, e16,m8,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x half>, <128 x half>* %x
+  %red = call half @llvm.vector.reduce.fadd.v128f16(half %s, <128 x half> %v)
+  ret half %red
+}
+
+declare float @llvm.vector.reduce.fadd.v1f32(float, <1 x float>)
+
+define float @vreduce_fadd_v1f32(<1 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_fadd_v1f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.s fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <1 x float>, <1 x float>* %x
+  %red = call reassoc float @llvm.vector.reduce.fadd.v1f32(float %s, <1 x float> %v)
+  ret float %red
+}
+
+define float @vreduce_ord_fadd_v1f32(<1 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v1f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v26, fa0
+; CHECK-NEXT:    vsetivli a0, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x float>, <1 x float>* %x
+  %red = call float @llvm.vector.reduce.fadd.v1f32(float %s, <1 x float> %v)
+  ret float %red
+}
+
+declare float @llvm.vector.reduce.fadd.v2f32(float, <2 x float>)
+
+define float @vreduce_fadd_v2f32(<2 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_fadd_v2f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.s fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <2 x float>, <2 x float>* %x
+  %red = call reassoc float @llvm.vector.reduce.fadd.v2f32(float %s, <2 x float> %v)
+  ret float %red
+}
+
+define float @vreduce_ord_fadd_v2f32(<2 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v2f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v26, fa0
+; CHECK-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x float>, <2 x float>* %x
+  %red = call float @llvm.vector.reduce.fadd.v2f32(float %s, <2 x float> %v)
+  ret float %red
+}
+
+declare float @llvm.vector.reduce.fadd.v4f32(float, <4 x float>)
+
+define float @vreduce_fadd_v4f32(<4 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_fadd_v4f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.s fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <4 x float>, <4 x float>* %x
+  %red = call reassoc float @llvm.vector.reduce.fadd.v4f32(float %s, <4 x float> %v)
+  ret float %red
+}
+
+define float @vreduce_ord_fadd_v4f32(<4 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v4f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v26, fa0
+; CHECK-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x float>, <4 x float>* %x
+  %red = call float @llvm.vector.reduce.fadd.v4f32(float %s, <4 x float> %v)
+  ret float %red
+}
+
+declare float @llvm.vector.reduce.fadd.v8f32(float, <8 x float>)
+
+define float @vreduce_fadd_v8f32(<8 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_fadd_v8f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vle32.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.s fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <8 x float>, <8 x float>* %x
+  %red = call reassoc float @llvm.vector.reduce.fadd.v8f32(float %s, <8 x float> %v)
+  ret float %red
+}
+
+define float @vreduce_ord_fadd_v8f32(<8 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v8f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vle32.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x float>, <8 x float>* %x
+  %red = call float @llvm.vector.reduce.fadd.v8f32(float %s, <8 x float> %v)
+  ret float %red
+}
+
+declare float @llvm.vector.reduce.fadd.v16f32(float, <16 x float>)
+
+define float @vreduce_fadd_v16f32(<16 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_fadd_v16f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.s fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <16 x float>, <16 x float>* %x
+  %red = call reassoc float @llvm.vector.reduce.fadd.v16f32(float %s, <16 x float> %v)
+  ret float %red
+}
+
+define float @vreduce_ord_fadd_v16f32(<16 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v16f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x float>, <16 x float>* %x
+  %red = call float @llvm.vector.reduce.fadd.v16f32(float %s, <16 x float> %v)
+  ret float %red
+}
+
+declare float @llvm.vector.reduce.fadd.v32f32(float, <32 x float>)
+
+define float @vreduce_fadd_v32f32(<32 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_fadd_v32f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.s fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <32 x float>, <32 x float>* %x
+  %red = call reassoc float @llvm.vector.reduce.fadd.v32f32(float %s, <32 x float> %v)
+  ret float %red
+}
+
+define float @vreduce_ord_fadd_v32f32(<32 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v32f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x float>, <32 x float>* %x
+  %red = call float @llvm.vector.reduce.fadd.v32f32(float %s, <32 x float> %v)
+  ret float %red
+}
+
+declare float @llvm.vector.reduce.fadd.v64f32(float, <64 x float>)
+
+define float @vreduce_fadd_v64f32(<64 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_fadd_v64f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle32.v v16, (a0)
+; CHECK-NEXT:    vfadd.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.s fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <64 x float>, <64 x float>* %x
+  %red = call reassoc float @llvm.vector.reduce.fadd.v64f32(float %s, <64 x float> %v)
+  ret float %red
+}
+
+define float @vreduce_ord_fadd_v64f32(<64 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v64f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, a0, 128
+; CHECK-NEXT:    addi a2, zero, 32
+; CHECK-NEXT:    vsetvli a3, a2, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a1)
+; CHECK-NEXT:    vle32.v v16, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetvli a0, a2, e32,m8,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v16, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, ft0
+; CHECK-NEXT:    vsetvli a0, a2, e32,m8,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x float>, <64 x float>* %x
+  %red = call float @llvm.vector.reduce.fadd.v64f32(float %s, <64 x float> %v)
+  ret float %red
+}
+
+declare double @llvm.vector.reduce.fadd.v1f64(double, <1 x double>)
+
+define double @vreduce_fadd_v1f64(<1 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_fadd_v1f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; CHECK-NEXT:    vle64.v v25, (a0)
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.d fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <1 x double>, <1 x double>* %x
+  %red = call reassoc double @llvm.vector.reduce.fadd.v1f64(double %s, <1 x double> %v)
+  ret double %red
+}
+
+define double @vreduce_ord_fadd_v1f64(<1 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v1f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; CHECK-NEXT:    vle64.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v26, fa0
+; CHECK-NEXT:    vsetivli a0, 1, e64,m1,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x double>, <1 x double>* %x
+  %red = call double @llvm.vector.reduce.fadd.v1f64(double %s, <1 x double> %v)
+  ret double %red
+}
+
+declare double @llvm.vector.reduce.fadd.v2f64(double, <2 x double>)
+
+define double @vreduce_fadd_v2f64(<2 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_fadd_v2f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; CHECK-NEXT:    vle64.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.d fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <2 x double>, <2 x double>* %x
+  %red = call reassoc double @llvm.vector.reduce.fadd.v2f64(double %s, <2 x double> %v)
+  ret double %red
+}
+
+define double @vreduce_ord_fadd_v2f64(<2 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v2f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; CHECK-NEXT:    vle64.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v26, fa0
+; CHECK-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v25, v26
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x double>, <2 x double>* %x
+  %red = call double @llvm.vector.reduce.fadd.v2f64(double %s, <2 x double> %v)
+  ret double %red
+}
+
+declare double @llvm.vector.reduce.fadd.v4f64(double, <4 x double>)
+
+define double @vreduce_fadd_v4f64(<4 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_fadd_v4f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; CHECK-NEXT:    vle64.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.d fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <4 x double>, <4 x double>* %x
+  %red = call reassoc double @llvm.vector.reduce.fadd.v4f64(double %s, <4 x double> %v)
+  ret double %red
+}
+
+define double @vreduce_ord_fadd_v4f64(<4 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v4f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; CHECK-NEXT:    vle64.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x double>, <4 x double>* %x
+  %red = call double @llvm.vector.reduce.fadd.v4f64(double %s, <4 x double> %v)
+  ret double %red
+}
+
+declare double @llvm.vector.reduce.fadd.v8f64(double, <8 x double>)
+
+define double @vreduce_fadd_v8f64(<8 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_fadd_v8f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; CHECK-NEXT:    vle64.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.d fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <8 x double>, <8 x double>* %x
+  %red = call reassoc double @llvm.vector.reduce.fadd.v8f64(double %s, <8 x double> %v)
+  ret double %red
+}
+
+define double @vreduce_ord_fadd_v8f64(<8 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v8f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; CHECK-NEXT:    vle64.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x double>, <8 x double>* %x
+  %red = call double @llvm.vector.reduce.fadd.v8f64(double %s, <8 x double> %v)
+  ret double %red
+}
+
+declare double @llvm.vector.reduce.fadd.v16f64(double, <16 x double>)
+
+define double @vreduce_fadd_v16f64(<16 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_fadd_v16f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; CHECK-NEXT:    vle64.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.d fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <16 x double>, <16 x double>* %x
+  %red = call reassoc double @llvm.vector.reduce.fadd.v16f64(double %s, <16 x double> %v)
+  ret double %red
+}
+
+define double @vreduce_ord_fadd_v16f64(<16 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v16f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; CHECK-NEXT:    vle64.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x double>, <16 x double>* %x
+  %red = call double @llvm.vector.reduce.fadd.v16f64(double %s, <16 x double> %v)
+  ret double %red
+}
+
+declare double @llvm.vector.reduce.fadd.v32f64(double, <32 x double>)
+
+define double @vreduce_fadd_v32f64(<32 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_fadd_v32f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; CHECK-NEXT:    vle64.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle64.v v16, (a0)
+; CHECK-NEXT:    vfadd.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; CHECK-NEXT:    vfredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    fadd.d fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <32 x double>, <32 x double>* %x
+  %red = call reassoc double @llvm.vector.reduce.fadd.v32f64(double %s, <32 x double> %v)
+  ret double %red
+}
+
+define double @vreduce_ord_fadd_v32f64(<32 x double>* %x, double %s) {
+; CHECK-LABEL: vreduce_ord_fadd_v32f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, a0, 128
+; CHECK-NEXT:    vsetivli a2, 16, e64,m8,ta,mu
+; CHECK-NEXT:    vle64.v v8, (a1)
+; CHECK-NEXT:    vle64.v v16, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, fa0
+; CHECK-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v16, v25
+; CHECK-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s ft0, v25
+; CHECK-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.v.f v25, ft0
+; CHECK-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; CHECK-NEXT:    vfredosum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; CHECK-NEXT:    vfmv.f.s fa0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x double>, <32 x double>* %x
+  %red = call double @llvm.vector.reduce.fadd.v32f64(double %s, <32 x double> %v)
+  ret double %red
+}

llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-int.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+m,+experimental-v -verify-machineinstrs -riscv-v-vector-bits-min=128 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV32
+; RUN: llc -mtriple=riscv64 -mattr=+m,+experimental-v -verify-machineinstrs -riscv-v-vector-bits-min=128  -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,RV64
+
+declare i8 @llvm.vector.reduce.add.v1i8(<1 x i8>)
+
+define i8 @vreduce_add_v1i8(<1 x i8>* %x) {
+; CHECK-LABEL: vreduce_add_v1i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i8>, <1 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.add.v1i8(<1 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.add.v2i8(<2 x i8>)
+
+define i8 @vreduce_add_v2i8(<2 x i8>* %x) {
+; CHECK-LABEL: vreduce_add_v2i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i8>, <2 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.add.v4i8(<4 x i8>)
+
+define i8 @vreduce_add_v4i8(<4 x i8>* %x) {
+; CHECK-LABEL: vreduce_add_v4i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i8>, <4 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.add.v8i8(<8 x i8>)
+
+define i8 @vreduce_add_v8i8(<8 x i8>* %x) {
+; CHECK-LABEL: vreduce_add_v8i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i8>, <8 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.add.v16i8(<16 x i8>)
+
+define i8 @vreduce_add_v16i8(<16 x i8>* %x) {
+; CHECK-LABEL: vreduce_add_v16i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i8>, <16 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.add.v32i8(<32 x i8>)
+
+define i8 @vreduce_add_v32i8(<32 x i8>* %x) {
+; CHECK-LABEL: vreduce_add_v32i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vle8.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i8>, <32 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.add.v32i8(<32 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.add.v64i8(<64 x i8>)
+
+define i8 @vreduce_add_v64i8(<64 x i8>* %x) {
+; CHECK-LABEL: vreduce_add_v64i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i8>, <64 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.add.v64i8(<64 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.add.v128i8(<128 x i8>)
+
+define i8 @vreduce_add_v128i8(<128 x i8>* %x) {
+; CHECK-LABEL: vreduce_add_v128i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i8>, <128 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.add.v128i8(<128 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.add.v256i8(<256 x i8>)
+
+define i8 @vreduce_add_v256i8(<256 x i8>* %x) {
+; CHECK-LABEL: vreduce_add_v256i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle8.v v16, (a0)
+; CHECK-NEXT:    vadd.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <256 x i8>, <256 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.add.v256i8(<256 x i8> %v)
+  ret i8 %red
+}
+
+declare i16 @llvm.vector.reduce.add.v1i16(<1 x i16>)
+
+define i16 @vreduce_add_v1i16(<1 x i16>* %x) {
+; CHECK-LABEL: vreduce_add_v1i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i16>, <1 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.add.v1i16(<1 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.add.v2i16(<2 x i16>)
+
+define i16 @vreduce_add_v2i16(<2 x i16>* %x) {
+; CHECK-LABEL: vreduce_add_v2i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i16>, <2 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.add.v2i16(<2 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.add.v4i16(<4 x i16>)
+
+define i16 @vreduce_add_v4i16(<4 x i16>* %x) {
+; CHECK-LABEL: vreduce_add_v4i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i16>, <4 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.add.v4i16(<4 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.add.v8i16(<8 x i16>)
+
+define i16 @vreduce_add_v8i16(<8 x i16>* %x) {
+; CHECK-LABEL: vreduce_add_v8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i16>, <8 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.add.v8i16(<8 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.add.v16i16(<16 x i16>)
+
+define i16 @vreduce_add_v16i16(<16 x i16>* %x) {
+; CHECK-LABEL: vreduce_add_v16i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vle16.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i16>, <16 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.add.v16i16(<16 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.add.v32i16(<32 x i16>)
+
+define i16 @vreduce_add_v32i16(<32 x i16>* %x) {
+; CHECK-LABEL: vreduce_add_v32i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i16>, <32 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.add.v32i16(<32 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.add.v64i16(<64 x i16>)
+
+define i16 @vreduce_add_v64i16(<64 x i16>* %x) {
+; CHECK-LABEL: vreduce_add_v64i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i16>, <64 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.add.v64i16(<64 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.add.v128i16(<128 x i16>)
+
+define i16 @vreduce_add_v128i16(<128 x i16>* %x) {
+; CHECK-LABEL: vreduce_add_v128i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle16.v v16, (a0)
+; CHECK-NEXT:    vadd.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i16>, <128 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.add.v128i16(<128 x i16> %v)
+  ret i16 %red
+}
+
+declare i32 @llvm.vector.reduce.add.v1i32(<1 x i32>)
+
+define i32 @vreduce_add_v1i32(<1 x i32>* %x) {
+; CHECK-LABEL: vreduce_add_v1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i32>, <1 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.add.v1i32(<1 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.add.v2i32(<2 x i32>)
+
+define i32 @vreduce_add_v2i32(<2 x i32>* %x) {
+; CHECK-LABEL: vreduce_add_v2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i32>, <2 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.add.v2i32(<2 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.add.v4i32(<4 x i32>)
+
+define i32 @vreduce_add_v4i32(<4 x i32>* %x) {
+; CHECK-LABEL: vreduce_add_v4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i32>, <4 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.add.v8i32(<8 x i32>)
+
+define i32 @vreduce_add_v8i32(<8 x i32>* %x) {
+; CHECK-LABEL: vreduce_add_v8i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vle32.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i32>, <8 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.add.v16i32(<16 x i32>)
+
+define i32 @vreduce_add_v16i32(<16 x i32>* %x) {
+; CHECK-LABEL: vreduce_add_v16i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i32>, <16 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.add.v32i32(<32 x i32>)
+
+define i32 @vreduce_add_v32i32(<32 x i32>* %x) {
+; CHECK-LABEL: vreduce_add_v32i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i32>, <32 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.add.v32i32(<32 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.add.v64i32(<64 x i32>)
+
+define i32 @vreduce_add_v64i32(<64 x i32>* %x) {
+; CHECK-LABEL: vreduce_add_v64i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle32.v v16, (a0)
+; CHECK-NEXT:    vadd.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredsum.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i32>, <64 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.add.v64i32(<64 x i32> %v)
+  ret i32 %red
+}
+
+declare i64 @llvm.vector.reduce.add.v1i64(<1 x i64>)
+
+define i64 @vreduce_add_v1i64(<1 x i64>* %x) {
+; RV32-LABEL: vreduce_add_v1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_add_v1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <1 x i64>, <1 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.add.v1i64(<1 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.add.v2i64(<2 x i64>)
+
+define i64 @vreduce_add_v2i64(<2 x i64>* %x) {
+; RV32-LABEL: vreduce_add_v2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v26, 0
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vredsum.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_add_v2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v26, 0
+; RV64-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV64-NEXT:    vredsum.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i64>, <2 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.add.v4i64(<4 x i64>)
+
+define i64 @vreduce_add_v4i64(<4 x i64>* %x) {
+; RV32-LABEL: vreduce_add_v4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV32-NEXT:    vle64.v v26, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vredsum.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_add_v4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV64-NEXT:    vle64.v v26, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV64-NEXT:    vredsum.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i64>, <4 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.add.v8i64(<8 x i64>)
+
+define i64 @vreduce_add_v8i64(<8 x i64>* %x) {
+; RV32-LABEL: vreduce_add_v8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV32-NEXT:    vle64.v v28, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vredsum.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_add_v8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV64-NEXT:    vle64.v v28, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV64-NEXT:    vredsum.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i64>, <8 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.add.v16i64(<16 x i64>)
+
+define i64 @vreduce_add_v16i64(<16 x i64>* %x) {
+; RV32-LABEL: vreduce_add_v16i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV32-NEXT:    vredsum.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_add_v16i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredsum.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i64>, <16 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.add.v32i64(<32 x i64>)
+
+define i64 @vreduce_add_v32i64(<32 x i64>* %x) {
+; RV32-LABEL: vreduce_add_v32i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle64.v v16, (a0)
+; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    vsetivli a0, 8, e64,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vadd.vv v28, v8, v16
+; RV32-NEXT:    vsetivli a0, 4, e64,m4,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v28, 4
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vadd.vv v26, v28, v8
+; RV32-NEXT:    vsetivli a0, 2, e64,m2,ta,mu
+; RV32-NEXT:    vslidedown.vi v28, v26, 2
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vadd.vv v25, v26, v28
+; RV32-NEXT:    vsetivli a0, 1, e64,m1,ta,mu
+; RV32-NEXT:    vslidedown.vi v26, v25, 1
+; RV32-NEXT:    vadd.vv v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_add_v32i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v16, (a0)
+; RV64-NEXT:    vadd.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredsum.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i64>, <32 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.add.v32i64(<32 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.add.v64i64(<64 x i64>)
+
+define i64 @vreduce_add_v64i64(<64 x i64>* %x) nounwind {
+; RV32-LABEL: vreduce_add_v64i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi sp, sp, -352
+; RV32-NEXT:    sw ra, 348(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s0, 344(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s1, 340(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s2, 336(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s3, 332(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s4, 328(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s5, 324(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s6, 320(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s7, 316(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s8, 312(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s9, 308(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s10, 304(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s11, 300(sp) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    sub sp, sp, a1
+; RV32-NEXT:    addi a1, a0, 384
+; RV32-NEXT:    addi s1, zero, 32
+; RV32-NEXT:    vsetvli a2, s1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a1)
+; RV32-NEXT:    vsetivli a1, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 296(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 292(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 288(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 284(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 280(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 276(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 272(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 268(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 264(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 260(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 256(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 252(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 248(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 244(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 240(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 236(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 232(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 228(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 224(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 220(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 216(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 212(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 208(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 204(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 200(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 196(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 224
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 216
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 208
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 224
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 192(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 216
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 188(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 208
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 184(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a1, v24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 180(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 176(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 172(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a1, s1, e32,m8,ta,mu
+; RV32-NEXT:    addi a1, a0, 256
+; RV32-NEXT:    vle32.v v8, (a1)
+; RV32-NEXT:    vsetivli a1, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 224
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 216
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 208
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 200
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 192
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 184
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 176
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 168
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 160
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 152
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 144
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 136
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 7
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 120
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 112
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 104
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 96
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 88
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 80
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 72
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 6
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 56
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 48
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 40
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 24
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 4
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 224
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 168(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 216
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 164(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 208
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 160(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 200
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 156(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 192
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 152(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 184
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 148(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 176
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 144(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 168
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 140(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 160
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 136(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 152
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 132(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 144
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 128(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 136
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 124(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 7
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 120(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 120
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 116(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 112
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 112(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 104
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 108(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 96
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 104(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 88
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 100(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 80
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 96(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 72
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 92(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 6
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 88(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 56
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 84(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 48
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 80(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 40
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 76(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 72(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 24
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 68(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 4
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 64(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 60(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 56(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a1, v24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 52(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 48(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 44(a2) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a1, s1, e32,m8,ta,mu
+; RV32-NEXT:    addi a1, a0, 128
+; RV32-NEXT:    vle32.v v8, (a1)
+; RV32-NEXT:    vsetivli a1, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 224
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 216
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 208
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 200
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 192
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 184
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 176
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 168
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 160
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 152
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 144
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 136
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 7
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 120
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 112
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 104
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 96
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 88
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 80
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 72
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 6
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 56
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 48
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 40
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 24
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 4
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 224
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 40(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 216
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 36(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 208
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 32(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 200
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 28(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 192
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 24(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 184
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 20(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 176
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 16(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 168
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw a1, 12(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 160
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s11, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 152
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s10, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 144
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s8, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 136
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s7, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 7
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s6, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 120
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s5, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 112
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s4, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 104
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s3, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 96
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s9, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 88
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s2, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 80
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t6, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 72
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t5, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 6
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t4, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 56
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t3, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 48
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t2, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 40
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t1, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 24
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a7, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 4
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a6, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a5, v0
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a4, v0
+; RV32-NEXT:    vmv.x.s a3, v24
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    vsetvli s1, s1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a0)
+; RV32-NEXT:    vsetivli a0, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 3
+; RV32-NEXT:    vslidedown.vi v24, v8, 1
+; RV32-NEXT:    vmv.x.s a0, v16
+; RV32-NEXT:    vmv.x.s s1, v24
+; RV32-NEXT:    add a0, s1, a0
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    vmv.x.s ra, v16
+; RV32-NEXT:    add ra, s1, ra
+; RV32-NEXT:    sltu s1, ra, s1
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add s1, ra, s1
+; RV32-NEXT:    sltu s0, s1, ra
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add s0, s1, s0
+; RV32-NEXT:    sltu s1, s0, s1
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add s1, s0, s1
+; RV32-NEXT:    sltu s0, s1, s0
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add s0, s1, s0
+; RV32-NEXT:    sltu s1, s0, s1
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add s1, s0, s1
+; RV32-NEXT:    sltu s0, s1, s0
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add s0, s1, s0
+; RV32-NEXT:    sltu s1, s0, s1
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add s1, s0, s1
+; RV32-NEXT:    sltu s0, s1, s0
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add s0, s1, s0
+; RV32-NEXT:    sltu s1, s0, s1
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add s1, s0, s1
+; RV32-NEXT:    sltu s0, s1, s0
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add s0, s1, s0
+; RV32-NEXT:    sltu s1, s0, s1
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add s1, s0, s1
+; RV32-NEXT:    sltu s0, s1, s0
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add s0, s1, s0
+; RV32-NEXT:    sltu s1, s0, s1
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    vmv.x.s s1, v16
+; RV32-NEXT:    add s1, s0, s1
+; RV32-NEXT:    sltu s0, s1, s0
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    vmv.x.s s0, v16
+; RV32-NEXT:    add a0, a0, s0
+; RV32-NEXT:    vslidedown.vi v8, v8, 30
+; RV32-NEXT:    vmv.x.s s0, v8
+; RV32-NEXT:    add s0, s1, s0
+; RV32-NEXT:    sltu s1, s0, s1
+; RV32-NEXT:    add a0, a0, s1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    add a1, s0, a2
+; RV32-NEXT:    sltu a2, a1, s0
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    add a0, a0, a3
+; RV32-NEXT:    add a2, a1, a4
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    add a0, a0, a5
+; RV32-NEXT:    add a1, a2, a6
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    add a0, a0, a7
+; RV32-NEXT:    add a2, a1, t0
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    add a0, a0, t1
+; RV32-NEXT:    add a1, a2, t2
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    add a0, a0, t3
+; RV32-NEXT:    add a2, a1, t4
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    add a0, a0, t5
+; RV32-NEXT:    add a1, a2, t6
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    add a0, a0, s2
+; RV32-NEXT:    add a2, a1, s9
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    add a0, a0, s3
+; RV32-NEXT:    add a1, a2, s4
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    add a0, a0, s5
+; RV32-NEXT:    add a2, a1, s6
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    add a0, a0, s7
+; RV32-NEXT:    add a1, a2, s8
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    add a0, a0, s10
+; RV32-NEXT:    add a2, a1, s11
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 12(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 16(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 20(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 24(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 28(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 32(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 36(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 40(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 44(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 48(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 52(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 56(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 60(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 64(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 68(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 72(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 76(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 80(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 84(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 88(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 92(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 96(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 100(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 104(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 108(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 112(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 116(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 120(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 124(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 128(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 132(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 136(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 140(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 144(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 148(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 152(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 156(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 160(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 164(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 168(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 172(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 176(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 180(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 184(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 188(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 192(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 196(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 200(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 204(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 208(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 212(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 216(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 220(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 224(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 228(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 232(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 236(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 240(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 244(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 248(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 252(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 256(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 260(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 264(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 268(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 272(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 276(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 280(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a1, a2
+; RV32-NEXT:    sltu a1, a2, a1
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 284(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    lw a1, 288(a1) # 4-byte Folded Reload
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    sltu a2, a1, a2
+; RV32-NEXT:    add a0, a0, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 292(a2) # 4-byte Folded Reload
+; RV32-NEXT:    add a2, a0, a2
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a0, a0, a3
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    lw a0, 296(a0) # 4-byte Folded Reload
+; RV32-NEXT:    add a0, a1, a0
+; RV32-NEXT:    sltu a1, a0, a1
+; RV32-NEXT:    add a1, a2, a1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add sp, sp, a2
+; RV32-NEXT:    lw s11, 300(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s10, 304(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s9, 308(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s8, 312(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s7, 316(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s6, 320(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s5, 324(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s4, 328(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s3, 332(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s2, 336(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s1, 340(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s0, 344(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw ra, 348(sp) # 4-byte Folded Reload
+; RV32-NEXT:    addi sp, sp, 352
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_add_v64i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a1, a0, 384
+; RV64-NEXT:    vle64.v v16, (a1)
+; RV64-NEXT:    addi a1, a0, 256
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v24, (a0)
+; RV64-NEXT:    vle64.v v0, (a1)
+; RV64-NEXT:    vadd.vv v16, v24, v16
+; RV64-NEXT:    vadd.vv v8, v8, v0
+; RV64-NEXT:    vadd.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredsum.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i64>, <64 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.add.v64i64(<64 x i64> %v)
+  ret i64 %red
+}
+
+declare i8 @llvm.vector.reduce.and.v1i8(<1 x i8>)
+
+define i8 @vreduce_and_v1i8(<1 x i8>* %x) {
+; CHECK-LABEL: vreduce_and_v1i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i8>, <1 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.and.v1i8(<1 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.and.v2i8(<2 x i8>)
+
+define i8 @vreduce_and_v2i8(<2 x i8>* %x) {
+; CHECK-LABEL: vreduce_and_v2i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i8>, <2 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.and.v2i8(<2 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.and.v4i8(<4 x i8>)
+
+define i8 @vreduce_and_v4i8(<4 x i8>* %x) {
+; CHECK-LABEL: vreduce_and_v4i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i8>, <4 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.and.v4i8(<4 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.and.v8i8(<8 x i8>)
+
+define i8 @vreduce_and_v8i8(<8 x i8>* %x) {
+; CHECK-LABEL: vreduce_and_v8i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i8>, <8 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.and.v8i8(<8 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.and.v16i8(<16 x i8>)
+
+define i8 @vreduce_and_v16i8(<16 x i8>* %x) {
+; CHECK-LABEL: vreduce_and_v16i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i8>, <16 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.and.v16i8(<16 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.and.v32i8(<32 x i8>)
+
+define i8 @vreduce_and_v32i8(<32 x i8>* %x) {
+; CHECK-LABEL: vreduce_and_v32i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vle8.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i8>, <32 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.and.v32i8(<32 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.and.v64i8(<64 x i8>)
+
+define i8 @vreduce_and_v64i8(<64 x i8>* %x) {
+; CHECK-LABEL: vreduce_and_v64i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i8>, <64 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.and.v64i8(<64 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.and.v128i8(<128 x i8>)
+
+define i8 @vreduce_and_v128i8(<128 x i8>* %x) {
+; CHECK-LABEL: vreduce_and_v128i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i8>, <128 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.and.v128i8(<128 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.and.v256i8(<256 x i8>)
+
+define i8 @vreduce_and_v256i8(<256 x i8>* %x) {
+; CHECK-LABEL: vreduce_and_v256i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle8.v v16, (a0)
+; CHECK-NEXT:    vand.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <256 x i8>, <256 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.and.v256i8(<256 x i8> %v)
+  ret i8 %red
+}
+
+declare i16 @llvm.vector.reduce.and.v1i16(<1 x i16>)
+
+define i16 @vreduce_and_v1i16(<1 x i16>* %x) {
+; CHECK-LABEL: vreduce_and_v1i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i16>, <1 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.and.v1i16(<1 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.and.v2i16(<2 x i16>)
+
+define i16 @vreduce_and_v2i16(<2 x i16>* %x) {
+; CHECK-LABEL: vreduce_and_v2i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i16>, <2 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.and.v2i16(<2 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.and.v4i16(<4 x i16>)
+
+define i16 @vreduce_and_v4i16(<4 x i16>* %x) {
+; CHECK-LABEL: vreduce_and_v4i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i16>, <4 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.and.v4i16(<4 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.and.v8i16(<8 x i16>)
+
+define i16 @vreduce_and_v8i16(<8 x i16>* %x) {
+; CHECK-LABEL: vreduce_and_v8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i16>, <8 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.and.v8i16(<8 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.and.v16i16(<16 x i16>)
+
+define i16 @vreduce_and_v16i16(<16 x i16>* %x) {
+; CHECK-LABEL: vreduce_and_v16i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vle16.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i16>, <16 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.and.v16i16(<16 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.and.v32i16(<32 x i16>)
+
+define i16 @vreduce_and_v32i16(<32 x i16>* %x) {
+; CHECK-LABEL: vreduce_and_v32i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i16>, <32 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.and.v32i16(<32 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.and.v64i16(<64 x i16>)
+
+define i16 @vreduce_and_v64i16(<64 x i16>* %x) {
+; CHECK-LABEL: vreduce_and_v64i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i16>, <64 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.and.v64i16(<64 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.and.v128i16(<128 x i16>)
+
+define i16 @vreduce_and_v128i16(<128 x i16>* %x) {
+; CHECK-LABEL: vreduce_and_v128i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle16.v v16, (a0)
+; CHECK-NEXT:    vand.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i16>, <128 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.and.v128i16(<128 x i16> %v)
+  ret i16 %red
+}
+
+declare i32 @llvm.vector.reduce.and.v1i32(<1 x i32>)
+
+define i32 @vreduce_and_v1i32(<1 x i32>* %x) {
+; CHECK-LABEL: vreduce_and_v1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i32>, <1 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.and.v1i32(<1 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.and.v2i32(<2 x i32>)
+
+define i32 @vreduce_and_v2i32(<2 x i32>* %x) {
+; CHECK-LABEL: vreduce_and_v2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i32>, <2 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.and.v2i32(<2 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.and.v4i32(<4 x i32>)
+
+define i32 @vreduce_and_v4i32(<4 x i32>* %x) {
+; CHECK-LABEL: vreduce_and_v4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i32>, <4 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.and.v4i32(<4 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.and.v8i32(<8 x i32>)
+
+define i32 @vreduce_and_v8i32(<8 x i32>* %x) {
+; CHECK-LABEL: vreduce_and_v8i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vle32.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i32>, <8 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.and.v8i32(<8 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.and.v16i32(<16 x i32>)
+
+define i32 @vreduce_and_v16i32(<16 x i32>* %x) {
+; CHECK-LABEL: vreduce_and_v16i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i32>, <16 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.and.v16i32(<16 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.and.v32i32(<32 x i32>)
+
+define i32 @vreduce_and_v32i32(<32 x i32>* %x) {
+; CHECK-LABEL: vreduce_and_v32i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i32>, <32 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.and.v32i32(<32 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.and.v64i32(<64 x i32>)
+
+define i32 @vreduce_and_v64i32(<64 x i32>* %x) {
+; CHECK-LABEL: vreduce_and_v64i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle32.v v16, (a0)
+; CHECK-NEXT:    vand.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredand.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i32>, <64 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.and.v64i32(<64 x i32> %v)
+  ret i32 %red
+}
+
+declare i64 @llvm.vector.reduce.and.v1i64(<1 x i64>)
+
+define i64 @vreduce_and_v1i64(<1 x i64>* %x) {
+; RV32-LABEL: vreduce_and_v1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_and_v1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <1 x i64>, <1 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.and.v1i64(<1 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.and.v2i64(<2 x i64>)
+
+define i64 @vreduce_and_v2i64(<2 x i64>* %x) {
+; RV32-LABEL: vreduce_and_v2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v26, -1
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vredand.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_and_v2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v26, -1
+; RV64-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV64-NEXT:    vredand.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i64>, <2 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.and.v2i64(<2 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.and.v4i64(<4 x i64>)
+
+define i64 @vreduce_and_v4i64(<4 x i64>* %x) {
+; RV32-LABEL: vreduce_and_v4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV32-NEXT:    vle64.v v26, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, -1
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vredand.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_and_v4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV64-NEXT:    vle64.v v26, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV64-NEXT:    vredand.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i64>, <4 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.and.v4i64(<4 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.and.v8i64(<8 x i64>)
+
+define i64 @vreduce_and_v8i64(<8 x i64>* %x) {
+; RV32-LABEL: vreduce_and_v8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV32-NEXT:    vle64.v v28, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, -1
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vredand.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_and_v8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV64-NEXT:    vle64.v v28, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV64-NEXT:    vredand.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i64>, <8 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.and.v8i64(<8 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.and.v16i64(<16 x i64>)
+
+define i64 @vreduce_and_v16i64(<16 x i64>* %x) {
+; RV32-LABEL: vreduce_and_v16i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, -1
+; RV32-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV32-NEXT:    vredand.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_and_v16i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredand.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i64>, <16 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.and.v16i64(<16 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.and.v32i64(<32 x i64>)
+
+define i64 @vreduce_and_v32i64(<32 x i64>* %x) {
+; RV32-LABEL: vreduce_and_v32i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle64.v v16, (a0)
+; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    vsetivli a0, 8, e64,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vand.vv v28, v8, v16
+; RV32-NEXT:    vsetivli a0, 4, e64,m4,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v28, 4
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vand.vv v26, v28, v8
+; RV32-NEXT:    vsetivli a0, 2, e64,m2,ta,mu
+; RV32-NEXT:    vslidedown.vi v28, v26, 2
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vand.vv v25, v26, v28
+; RV32-NEXT:    vsetivli a0, 1, e64,m1,ta,mu
+; RV32-NEXT:    vslidedown.vi v26, v25, 1
+; RV32-NEXT:    vand.vv v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_and_v32i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v16, (a0)
+; RV64-NEXT:    vand.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredand.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i64>, <32 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.and.v32i64(<32 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.and.v64i64(<64 x i64>)
+
+define i64 @vreduce_and_v64i64(<64 x i64>* %x) nounwind {
+; RV32-LABEL: vreduce_and_v64i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi sp, sp, -160
+; RV32-NEXT:    sw ra, 156(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s0, 152(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s1, 148(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s2, 144(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s3, 140(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s4, 136(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s5, 132(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s6, 128(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s7, 124(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s8, 120(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s9, 116(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s10, 112(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s11, 108(sp) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 120
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    sub sp, sp, a1
+; RV32-NEXT:    addi a2, a0, 384
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v8, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 104(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 100(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 96(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 92(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 88(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 84(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 80(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 76(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 72(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 68(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 64(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 60(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 56(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 52(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 3
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 48(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 44(a3) # 4-byte Folded Spill
+; RV32-NEXT:    addi a2, a0, 256
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v16, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v24, v16, 31
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 40(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 29
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 36(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 27
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 32(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 25
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 28(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 23
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 24(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 21
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 20(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 19
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 16(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 17
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 12(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 15
+; RV32-NEXT:    vmv.x.s s11, v24
+; RV32-NEXT:    vslidedown.vi v24, v16, 13
+; RV32-NEXT:    vmv.x.s s10, v24
+; RV32-NEXT:    vslidedown.vi v8, v16, 11
+; RV32-NEXT:    vslidedown.vi v24, v16, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v0, v16, 3
+; RV32-NEXT:    vslidedown.vi v24, v16, 1
+; RV32-NEXT:    vmv.x.s s9, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s8, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s7, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s6, v8
+; RV32-NEXT:    vmv.x.s s5, v0
+; RV32-NEXT:    vmv.x.s s4, v24
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 128
+; RV32-NEXT:    vle32.v v0, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v0, 31
+; RV32-NEXT:    vslidedown.vi v24, v0, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 3
+; RV32-NEXT:    vs8r.v v24, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 1
+; RV32-NEXT:    vmv.x.s s3, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s2, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s0, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t6, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t5, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t4, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t3, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t2, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t1, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t0, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a7, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a6, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a5, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a4, v8
+; RV32-NEXT:    vl8re8.v v8, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a3, v8
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    vsetvli a1, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v24, (a0)
+; RV32-NEXT:    vsetivli a0, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v24, 2
+; RV32-NEXT:    vmv.x.s a0, v24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 2
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 2
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi s1, zero, 112
+; RV32-NEXT:    mul a1, a1, s1
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 2
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    and a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 3
+; RV32-NEXT:    vslidedown.vi v16, v24, 1
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    vmv.x.s ra, v16
+; RV32-NEXT:    and a1, ra, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 5
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 7
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 9
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 11
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 13
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 15
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 17
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 19
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 21
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 23
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 25
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 27
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 29
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 31
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    and a1, a1, s1
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    and a1, a1, a3
+; RV32-NEXT:    and a1, a1, a4
+; RV32-NEXT:    and a1, a1, a5
+; RV32-NEXT:    and a1, a1, a6
+; RV32-NEXT:    and a1, a1, a7
+; RV32-NEXT:    and a1, a1, t0
+; RV32-NEXT:    and a1, a1, t1
+; RV32-NEXT:    and a1, a1, t2
+; RV32-NEXT:    and a1, a1, t3
+; RV32-NEXT:    and a1, a1, t4
+; RV32-NEXT:    and a1, a1, t5
+; RV32-NEXT:    and a1, a1, t6
+; RV32-NEXT:    and a1, a1, s0
+; RV32-NEXT:    and a1, a1, s2
+; RV32-NEXT:    and a1, a1, s3
+; RV32-NEXT:    and a1, a1, s4
+; RV32-NEXT:    and a1, a1, s5
+; RV32-NEXT:    and a1, a1, s6
+; RV32-NEXT:    and a1, a1, s7
+; RV32-NEXT:    and a1, a1, s8
+; RV32-NEXT:    and a1, a1, s9
+; RV32-NEXT:    and a1, a1, s10
+; RV32-NEXT:    and a1, a1, s11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 12(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 16(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 20(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 24(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 28(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 32(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 36(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 40(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 44(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 48(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 52(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 56(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 60(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 64(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 68(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 72(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 76(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 80(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 84(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 88(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 92(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 96(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 100(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 104(a2) # 4-byte Folded Reload
+; RV32-NEXT:    and a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add sp, sp, a2
+; RV32-NEXT:    lw s11, 108(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s10, 112(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s9, 116(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s8, 120(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s7, 124(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s6, 128(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s5, 132(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s4, 136(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s3, 140(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s2, 144(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s1, 148(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s0, 152(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw ra, 156(sp) # 4-byte Folded Reload
+; RV32-NEXT:    addi sp, sp, 160
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_and_v64i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a1, a0, 384
+; RV64-NEXT:    vle64.v v16, (a1)
+; RV64-NEXT:    addi a1, a0, 256
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v24, (a0)
+; RV64-NEXT:    vle64.v v0, (a1)
+; RV64-NEXT:    vand.vv v16, v24, v16
+; RV64-NEXT:    vand.vv v8, v8, v0
+; RV64-NEXT:    vand.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredand.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i64>, <64 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.and.v64i64(<64 x i64> %v)
+  ret i64 %red
+}
+
+declare i8 @llvm.vector.reduce.or.v1i8(<1 x i8>)
+
+define i8 @vreduce_or_v1i8(<1 x i8>* %x) {
+; CHECK-LABEL: vreduce_or_v1i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i8>, <1 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.or.v1i8(<1 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.or.v2i8(<2 x i8>)
+
+define i8 @vreduce_or_v2i8(<2 x i8>* %x) {
+; CHECK-LABEL: vreduce_or_v2i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i8>, <2 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.or.v2i8(<2 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.or.v4i8(<4 x i8>)
+
+define i8 @vreduce_or_v4i8(<4 x i8>* %x) {
+; CHECK-LABEL: vreduce_or_v4i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i8>, <4 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.or.v4i8(<4 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.or.v8i8(<8 x i8>)
+
+define i8 @vreduce_or_v8i8(<8 x i8>* %x) {
+; CHECK-LABEL: vreduce_or_v8i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i8>, <8 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.or.v8i8(<8 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.or.v16i8(<16 x i8>)
+
+define i8 @vreduce_or_v16i8(<16 x i8>* %x) {
+; CHECK-LABEL: vreduce_or_v16i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i8>, <16 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.or.v16i8(<16 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.or.v32i8(<32 x i8>)
+
+define i8 @vreduce_or_v32i8(<32 x i8>* %x) {
+; CHECK-LABEL: vreduce_or_v32i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vle8.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i8>, <32 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.or.v32i8(<32 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.or.v64i8(<64 x i8>)
+
+define i8 @vreduce_or_v64i8(<64 x i8>* %x) {
+; CHECK-LABEL: vreduce_or_v64i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i8>, <64 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.or.v64i8(<64 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.or.v128i8(<128 x i8>)
+
+define i8 @vreduce_or_v128i8(<128 x i8>* %x) {
+; CHECK-LABEL: vreduce_or_v128i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i8>, <128 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.or.v128i8(<128 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.or.v256i8(<256 x i8>)
+
+define i8 @vreduce_or_v256i8(<256 x i8>* %x) {
+; CHECK-LABEL: vreduce_or_v256i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle8.v v16, (a0)
+; CHECK-NEXT:    vor.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <256 x i8>, <256 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.or.v256i8(<256 x i8> %v)
+  ret i8 %red
+}
+
+declare i16 @llvm.vector.reduce.or.v1i16(<1 x i16>)
+
+define i16 @vreduce_or_v1i16(<1 x i16>* %x) {
+; CHECK-LABEL: vreduce_or_v1i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i16>, <1 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.or.v1i16(<1 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.or.v2i16(<2 x i16>)
+
+define i16 @vreduce_or_v2i16(<2 x i16>* %x) {
+; CHECK-LABEL: vreduce_or_v2i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i16>, <2 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.or.v2i16(<2 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.or.v4i16(<4 x i16>)
+
+define i16 @vreduce_or_v4i16(<4 x i16>* %x) {
+; CHECK-LABEL: vreduce_or_v4i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i16>, <4 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.or.v4i16(<4 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.or.v8i16(<8 x i16>)
+
+define i16 @vreduce_or_v8i16(<8 x i16>* %x) {
+; CHECK-LABEL: vreduce_or_v8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i16>, <8 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.or.v8i16(<8 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.or.v16i16(<16 x i16>)
+
+define i16 @vreduce_or_v16i16(<16 x i16>* %x) {
+; CHECK-LABEL: vreduce_or_v16i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vle16.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i16>, <16 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.or.v16i16(<16 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.or.v32i16(<32 x i16>)
+
+define i16 @vreduce_or_v32i16(<32 x i16>* %x) {
+; CHECK-LABEL: vreduce_or_v32i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i16>, <32 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.or.v32i16(<32 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.or.v64i16(<64 x i16>)
+
+define i16 @vreduce_or_v64i16(<64 x i16>* %x) {
+; CHECK-LABEL: vreduce_or_v64i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i16>, <64 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.or.v64i16(<64 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.or.v128i16(<128 x i16>)
+
+define i16 @vreduce_or_v128i16(<128 x i16>* %x) {
+; CHECK-LABEL: vreduce_or_v128i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle16.v v16, (a0)
+; CHECK-NEXT:    vor.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i16>, <128 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.or.v128i16(<128 x i16> %v)
+  ret i16 %red
+}
+
+declare i32 @llvm.vector.reduce.or.v1i32(<1 x i32>)
+
+define i32 @vreduce_or_v1i32(<1 x i32>* %x) {
+; CHECK-LABEL: vreduce_or_v1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i32>, <1 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.or.v1i32(<1 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.or.v2i32(<2 x i32>)
+
+define i32 @vreduce_or_v2i32(<2 x i32>* %x) {
+; CHECK-LABEL: vreduce_or_v2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i32>, <2 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.or.v2i32(<2 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.or.v4i32(<4 x i32>)
+
+define i32 @vreduce_or_v4i32(<4 x i32>* %x) {
+; CHECK-LABEL: vreduce_or_v4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i32>, <4 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.or.v4i32(<4 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.or.v8i32(<8 x i32>)
+
+define i32 @vreduce_or_v8i32(<8 x i32>* %x) {
+; CHECK-LABEL: vreduce_or_v8i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vle32.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i32>, <8 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.or.v8i32(<8 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.or.v16i32(<16 x i32>)
+
+define i32 @vreduce_or_v16i32(<16 x i32>* %x) {
+; CHECK-LABEL: vreduce_or_v16i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i32>, <16 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.or.v16i32(<16 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.or.v32i32(<32 x i32>)
+
+define i32 @vreduce_or_v32i32(<32 x i32>* %x) {
+; CHECK-LABEL: vreduce_or_v32i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i32>, <32 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.or.v32i32(<32 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.or.v64i32(<64 x i32>)
+
+define i32 @vreduce_or_v64i32(<64 x i32>* %x) {
+; CHECK-LABEL: vreduce_or_v64i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle32.v v16, (a0)
+; CHECK-NEXT:    vor.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i32>, <64 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.or.v64i32(<64 x i32> %v)
+  ret i32 %red
+}
+
+declare i64 @llvm.vector.reduce.or.v1i64(<1 x i64>)
+
+define i64 @vreduce_or_v1i64(<1 x i64>* %x) {
+; RV32-LABEL: vreduce_or_v1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_or_v1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <1 x i64>, <1 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.or.v1i64(<1 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.or.v2i64(<2 x i64>)
+
+define i64 @vreduce_or_v2i64(<2 x i64>* %x) {
+; RV32-LABEL: vreduce_or_v2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v26, 0
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vredor.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_or_v2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v26, 0
+; RV64-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV64-NEXT:    vredor.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i64>, <2 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.or.v2i64(<2 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.or.v4i64(<4 x i64>)
+
+define i64 @vreduce_or_v4i64(<4 x i64>* %x) {
+; RV32-LABEL: vreduce_or_v4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV32-NEXT:    vle64.v v26, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vredor.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_or_v4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV64-NEXT:    vle64.v v26, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV64-NEXT:    vredor.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i64>, <4 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.or.v4i64(<4 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.or.v8i64(<8 x i64>)
+
+define i64 @vreduce_or_v8i64(<8 x i64>* %x) {
+; RV32-LABEL: vreduce_or_v8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV32-NEXT:    vle64.v v28, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vredor.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_or_v8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV64-NEXT:    vle64.v v28, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV64-NEXT:    vredor.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i64>, <8 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.or.v8i64(<8 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.or.v16i64(<16 x i64>)
+
+define i64 @vreduce_or_v16i64(<16 x i64>* %x) {
+; RV32-LABEL: vreduce_or_v16i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV32-NEXT:    vredor.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_or_v16i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredor.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i64>, <16 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.or.v16i64(<16 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.or.v32i64(<32 x i64>)
+
+define i64 @vreduce_or_v32i64(<32 x i64>* %x) {
+; RV32-LABEL: vreduce_or_v32i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle64.v v16, (a0)
+; RV32-NEXT:    vor.vv v8, v8, v16
+; RV32-NEXT:    vsetivli a0, 8, e64,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vor.vv v28, v8, v16
+; RV32-NEXT:    vsetivli a0, 4, e64,m4,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v28, 4
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vor.vv v26, v28, v8
+; RV32-NEXT:    vsetivli a0, 2, e64,m2,ta,mu
+; RV32-NEXT:    vslidedown.vi v28, v26, 2
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vor.vv v25, v26, v28
+; RV32-NEXT:    vsetivli a0, 1, e64,m1,ta,mu
+; RV32-NEXT:    vslidedown.vi v26, v25, 1
+; RV32-NEXT:    vor.vv v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_or_v32i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v16, (a0)
+; RV64-NEXT:    vor.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredor.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i64>, <32 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.or.v32i64(<32 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.or.v64i64(<64 x i64>)
+
+define i64 @vreduce_or_v64i64(<64 x i64>* %x) nounwind {
+; RV32-LABEL: vreduce_or_v64i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi sp, sp, -160
+; RV32-NEXT:    sw ra, 156(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s0, 152(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s1, 148(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s2, 144(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s3, 140(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s4, 136(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s5, 132(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s6, 128(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s7, 124(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s8, 120(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s9, 116(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s10, 112(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s11, 108(sp) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 120
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    sub sp, sp, a1
+; RV32-NEXT:    addi a2, a0, 384
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v8, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 104(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 100(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 96(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 92(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 88(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 84(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 80(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 76(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 72(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 68(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 64(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 60(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 56(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 52(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 3
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 48(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 44(a3) # 4-byte Folded Spill
+; RV32-NEXT:    addi a2, a0, 256
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v16, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v24, v16, 31
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 40(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 29
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 36(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 27
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 32(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 25
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 28(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 23
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 24(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 21
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 20(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 19
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 16(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 17
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 12(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 15
+; RV32-NEXT:    vmv.x.s s11, v24
+; RV32-NEXT:    vslidedown.vi v24, v16, 13
+; RV32-NEXT:    vmv.x.s s10, v24
+; RV32-NEXT:    vslidedown.vi v8, v16, 11
+; RV32-NEXT:    vslidedown.vi v24, v16, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v0, v16, 3
+; RV32-NEXT:    vslidedown.vi v24, v16, 1
+; RV32-NEXT:    vmv.x.s s9, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s8, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s7, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s6, v8
+; RV32-NEXT:    vmv.x.s s5, v0
+; RV32-NEXT:    vmv.x.s s4, v24
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 128
+; RV32-NEXT:    vle32.v v0, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v0, 31
+; RV32-NEXT:    vslidedown.vi v24, v0, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 3
+; RV32-NEXT:    vs8r.v v24, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 1
+; RV32-NEXT:    vmv.x.s s3, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s2, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s0, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t6, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t5, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t4, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t3, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t2, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t1, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t0, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a7, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a6, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a5, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a4, v8
+; RV32-NEXT:    vl8re8.v v8, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a3, v8
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    vsetvli a1, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v24, (a0)
+; RV32-NEXT:    vsetivli a0, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v24, 2
+; RV32-NEXT:    vmv.x.s a0, v24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 2
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 2
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi s1, zero, 112
+; RV32-NEXT:    mul a1, a1, s1
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 2
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    or a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 3
+; RV32-NEXT:    vslidedown.vi v16, v24, 1
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    vmv.x.s ra, v16
+; RV32-NEXT:    or a1, ra, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 5
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 7
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 9
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 11
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 13
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 15
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 17
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 19
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 21
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 23
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 25
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 27
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 29
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 31
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    or a1, a1, s1
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    or a1, a1, a3
+; RV32-NEXT:    or a1, a1, a4
+; RV32-NEXT:    or a1, a1, a5
+; RV32-NEXT:    or a1, a1, a6
+; RV32-NEXT:    or a1, a1, a7
+; RV32-NEXT:    or a1, a1, t0
+; RV32-NEXT:    or a1, a1, t1
+; RV32-NEXT:    or a1, a1, t2
+; RV32-NEXT:    or a1, a1, t3
+; RV32-NEXT:    or a1, a1, t4
+; RV32-NEXT:    or a1, a1, t5
+; RV32-NEXT:    or a1, a1, t6
+; RV32-NEXT:    or a1, a1, s0
+; RV32-NEXT:    or a1, a1, s2
+; RV32-NEXT:    or a1, a1, s3
+; RV32-NEXT:    or a1, a1, s4
+; RV32-NEXT:    or a1, a1, s5
+; RV32-NEXT:    or a1, a1, s6
+; RV32-NEXT:    or a1, a1, s7
+; RV32-NEXT:    or a1, a1, s8
+; RV32-NEXT:    or a1, a1, s9
+; RV32-NEXT:    or a1, a1, s10
+; RV32-NEXT:    or a1, a1, s11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 12(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 16(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 20(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 24(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 28(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 32(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 36(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 40(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 44(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 48(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 52(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 56(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 60(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 64(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 68(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 72(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 76(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 80(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 84(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 88(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 92(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 96(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 100(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 104(a2) # 4-byte Folded Reload
+; RV32-NEXT:    or a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add sp, sp, a2
+; RV32-NEXT:    lw s11, 108(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s10, 112(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s9, 116(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s8, 120(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s7, 124(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s6, 128(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s5, 132(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s4, 136(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s3, 140(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s2, 144(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s1, 148(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s0, 152(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw ra, 156(sp) # 4-byte Folded Reload
+; RV32-NEXT:    addi sp, sp, 160
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_or_v64i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a1, a0, 384
+; RV64-NEXT:    vle64.v v16, (a1)
+; RV64-NEXT:    addi a1, a0, 256
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v24, (a0)
+; RV64-NEXT:    vle64.v v0, (a1)
+; RV64-NEXT:    vor.vv v16, v24, v16
+; RV64-NEXT:    vor.vv v8, v8, v0
+; RV64-NEXT:    vor.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredor.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i64>, <64 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.or.v64i64(<64 x i64> %v)
+  ret i64 %red
+}
+
+declare i8 @llvm.vector.reduce.xor.v1i8(<1 x i8>)
+
+define i8 @vreduce_xor_v1i8(<1 x i8>* %x) {
+; CHECK-LABEL: vreduce_xor_v1i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i8>, <1 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.xor.v1i8(<1 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.xor.v2i8(<2 x i8>)
+
+define i8 @vreduce_xor_v2i8(<2 x i8>* %x) {
+; CHECK-LABEL: vreduce_xor_v2i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i8>, <2 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.xor.v2i8(<2 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.xor.v4i8(<4 x i8>)
+
+define i8 @vreduce_xor_v4i8(<4 x i8>* %x) {
+; CHECK-LABEL: vreduce_xor_v4i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i8>, <4 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.xor.v4i8(<4 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.xor.v8i8(<8 x i8>)
+
+define i8 @vreduce_xor_v8i8(<8 x i8>* %x) {
+; CHECK-LABEL: vreduce_xor_v8i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i8>, <8 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.xor.v8i8(<8 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.xor.v16i8(<16 x i8>)
+
+define i8 @vreduce_xor_v16i8(<16 x i8>* %x) {
+; CHECK-LABEL: vreduce_xor_v16i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i8>, <16 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.xor.v16i8(<16 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.xor.v32i8(<32 x i8>)
+
+define i8 @vreduce_xor_v32i8(<32 x i8>* %x) {
+; CHECK-LABEL: vreduce_xor_v32i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vle8.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i8>, <32 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.xor.v32i8(<32 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.xor.v64i8(<64 x i8>)
+
+define i8 @vreduce_xor_v64i8(<64 x i8>* %x) {
+; CHECK-LABEL: vreduce_xor_v64i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i8>, <64 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.xor.v64i8(<64 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.xor.v128i8(<128 x i8>)
+
+define i8 @vreduce_xor_v128i8(<128 x i8>* %x) {
+; CHECK-LABEL: vreduce_xor_v128i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i8>, <128 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.xor.v128i8(<128 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.xor.v256i8(<256 x i8>)
+
+define i8 @vreduce_xor_v256i8(<256 x i8>* %x) {
+; CHECK-LABEL: vreduce_xor_v256i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle8.v v16, (a0)
+; CHECK-NEXT:    vxor.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <256 x i8>, <256 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.xor.v256i8(<256 x i8> %v)
+  ret i8 %red
+}
+
+declare i16 @llvm.vector.reduce.xor.v1i16(<1 x i16>)
+
+define i16 @vreduce_xor_v1i16(<1 x i16>* %x) {
+; CHECK-LABEL: vreduce_xor_v1i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i16>, <1 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.xor.v1i16(<1 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.xor.v2i16(<2 x i16>)
+
+define i16 @vreduce_xor_v2i16(<2 x i16>* %x) {
+; CHECK-LABEL: vreduce_xor_v2i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i16>, <2 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.xor.v2i16(<2 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.xor.v4i16(<4 x i16>)
+
+define i16 @vreduce_xor_v4i16(<4 x i16>* %x) {
+; CHECK-LABEL: vreduce_xor_v4i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i16>, <4 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.xor.v4i16(<4 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.xor.v8i16(<8 x i16>)
+
+define i16 @vreduce_xor_v8i16(<8 x i16>* %x) {
+; CHECK-LABEL: vreduce_xor_v8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i16>, <8 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.xor.v8i16(<8 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.xor.v16i16(<16 x i16>)
+
+define i16 @vreduce_xor_v16i16(<16 x i16>* %x) {
+; CHECK-LABEL: vreduce_xor_v16i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vle16.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i16>, <16 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.xor.v16i16(<16 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.xor.v32i16(<32 x i16>)
+
+define i16 @vreduce_xor_v32i16(<32 x i16>* %x) {
+; CHECK-LABEL: vreduce_xor_v32i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i16>, <32 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.xor.v32i16(<32 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.xor.v64i16(<64 x i16>)
+
+define i16 @vreduce_xor_v64i16(<64 x i16>* %x) {
+; CHECK-LABEL: vreduce_xor_v64i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i16>, <64 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.xor.v64i16(<64 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.xor.v128i16(<128 x i16>)
+
+define i16 @vreduce_xor_v128i16(<128 x i16>* %x) {
+; CHECK-LABEL: vreduce_xor_v128i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle16.v v16, (a0)
+; CHECK-NEXT:    vxor.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i16>, <128 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.xor.v128i16(<128 x i16> %v)
+  ret i16 %red
+}
+
+declare i32 @llvm.vector.reduce.xor.v1i32(<1 x i32>)
+
+define i32 @vreduce_xor_v1i32(<1 x i32>* %x) {
+; CHECK-LABEL: vreduce_xor_v1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i32>, <1 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.xor.v1i32(<1 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.xor.v2i32(<2 x i32>)
+
+define i32 @vreduce_xor_v2i32(<2 x i32>* %x) {
+; CHECK-LABEL: vreduce_xor_v2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i32>, <2 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.xor.v2i32(<2 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.xor.v4i32(<4 x i32>)
+
+define i32 @vreduce_xor_v4i32(<4 x i32>* %x) {
+; CHECK-LABEL: vreduce_xor_v4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i32>, <4 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.xor.v4i32(<4 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.xor.v8i32(<8 x i32>)
+
+define i32 @vreduce_xor_v8i32(<8 x i32>* %x) {
+; CHECK-LABEL: vreduce_xor_v8i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vle32.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i32>, <8 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.xor.v8i32(<8 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.xor.v16i32(<16 x i32>)
+
+define i32 @vreduce_xor_v16i32(<16 x i32>* %x) {
+; CHECK-LABEL: vreduce_xor_v16i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i32>, <16 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.xor.v16i32(<16 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.xor.v32i32(<32 x i32>)
+
+define i32 @vreduce_xor_v32i32(<32 x i32>* %x) {
+; CHECK-LABEL: vreduce_xor_v32i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i32>, <32 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.xor.v32i32(<32 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.xor.v64i32(<64 x i32>)
+
+define i32 @vreduce_xor_v64i32(<64 x i32>* %x) {
+; CHECK-LABEL: vreduce_xor_v64i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle32.v v16, (a0)
+; CHECK-NEXT:    vxor.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredxor.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i32>, <64 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.xor.v64i32(<64 x i32> %v)
+  ret i32 %red
+}
+
+declare i64 @llvm.vector.reduce.xor.v1i64(<1 x i64>)
+
+define i64 @vreduce_xor_v1i64(<1 x i64>* %x) {
+; RV32-LABEL: vreduce_xor_v1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_xor_v1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <1 x i64>, <1 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.xor.v1i64(<1 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.xor.v2i64(<2 x i64>)
+
+define i64 @vreduce_xor_v2i64(<2 x i64>* %x) {
+; RV32-LABEL: vreduce_xor_v2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v26, 0
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vredxor.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_xor_v2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v26, 0
+; RV64-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV64-NEXT:    vredxor.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i64>, <2 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.xor.v2i64(<2 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.xor.v4i64(<4 x i64>)
+
+define i64 @vreduce_xor_v4i64(<4 x i64>* %x) {
+; RV32-LABEL: vreduce_xor_v4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV32-NEXT:    vle64.v v26, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vredxor.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_xor_v4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV64-NEXT:    vle64.v v26, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV64-NEXT:    vredxor.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i64>, <4 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.xor.v4i64(<4 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.xor.v8i64(<8 x i64>)
+
+define i64 @vreduce_xor_v8i64(<8 x i64>* %x) {
+; RV32-LABEL: vreduce_xor_v8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV32-NEXT:    vle64.v v28, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vredxor.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_xor_v8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV64-NEXT:    vle64.v v28, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV64-NEXT:    vredxor.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i64>, <8 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.xor.v8i64(<8 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.xor.v16i64(<16 x i64>)
+
+define i64 @vreduce_xor_v16i64(<16 x i64>* %x) {
+; RV32-LABEL: vreduce_xor_v16i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV32-NEXT:    vredxor.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_xor_v16i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredxor.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i64>, <16 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.xor.v16i64(<16 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.xor.v32i64(<32 x i64>)
+
+define i64 @vreduce_xor_v32i64(<32 x i64>* %x) {
+; RV32-LABEL: vreduce_xor_v32i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle64.v v16, (a0)
+; RV32-NEXT:    vxor.vv v8, v8, v16
+; RV32-NEXT:    vsetivli a0, 8, e64,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vxor.vv v28, v8, v16
+; RV32-NEXT:    vsetivli a0, 4, e64,m4,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v28, 4
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vxor.vv v26, v28, v8
+; RV32-NEXT:    vsetivli a0, 2, e64,m2,ta,mu
+; RV32-NEXT:    vslidedown.vi v28, v26, 2
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vxor.vv v25, v26, v28
+; RV32-NEXT:    vsetivli a0, 1, e64,m1,ta,mu
+; RV32-NEXT:    vslidedown.vi v26, v25, 1
+; RV32-NEXT:    vxor.vv v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_xor_v32i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v16, (a0)
+; RV64-NEXT:    vxor.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredxor.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i64>, <32 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.xor.v32i64(<32 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.xor.v64i64(<64 x i64>)
+
+define i64 @vreduce_xor_v64i64(<64 x i64>* %x) nounwind {
+; RV32-LABEL: vreduce_xor_v64i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi sp, sp, -160
+; RV32-NEXT:    sw ra, 156(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s0, 152(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s1, 148(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s2, 144(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s3, 140(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s4, 136(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s5, 132(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s6, 128(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s7, 124(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s8, 120(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s9, 116(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s10, 112(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s11, 108(sp) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 120
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    sub sp, sp, a1
+; RV32-NEXT:    addi a2, a0, 384
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v8, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 104(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 100(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 96(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 92(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 88(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 84(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 80(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 76(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 72(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 68(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 64(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 60(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 56(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 52(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 3
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 48(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 44(a3) # 4-byte Folded Spill
+; RV32-NEXT:    addi a2, a0, 256
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v16, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v24, v16, 31
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 40(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 29
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 36(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 27
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 32(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 25
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 28(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 23
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 24(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 21
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 20(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 19
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 16(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 17
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 120
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 12(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 15
+; RV32-NEXT:    vmv.x.s s11, v24
+; RV32-NEXT:    vslidedown.vi v24, v16, 13
+; RV32-NEXT:    vmv.x.s s10, v24
+; RV32-NEXT:    vslidedown.vi v8, v16, 11
+; RV32-NEXT:    vslidedown.vi v24, v16, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v16, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v0, v16, 3
+; RV32-NEXT:    vslidedown.vi v24, v16, 1
+; RV32-NEXT:    vmv.x.s s9, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s8, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s7, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s6, v8
+; RV32-NEXT:    vmv.x.s s5, v0
+; RV32-NEXT:    vmv.x.s s4, v24
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 128
+; RV32-NEXT:    vle32.v v0, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v0, 31
+; RV32-NEXT:    vslidedown.vi v24, v0, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 3
+; RV32-NEXT:    vs8r.v v24, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v0, 1
+; RV32-NEXT:    vmv.x.s s3, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s2, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s0, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t6, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t5, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t4, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t3, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t2, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t1, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t0, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a7, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a6, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a5, v8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a4, v8
+; RV32-NEXT:    vl8re8.v v8, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a3, v8
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    vsetvli a1, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v24, (a0)
+; RV32-NEXT:    vsetivli a0, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v24, 2
+; RV32-NEXT:    vmv.x.s a0, v24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vmv.x.s a1, v0
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 2
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v0, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 2
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi s1, zero, 112
+; RV32-NEXT:    mul a1, a1, s1
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    vl8re8.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 2
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 4
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 6
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 8
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 10
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 12
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 14
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 16
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 18
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 20
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 22
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 24
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 26
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 28
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v16, 30
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    xor a0, a0, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 3
+; RV32-NEXT:    vslidedown.vi v16, v24, 1
+; RV32-NEXT:    vmv.x.s a1, v8
+; RV32-NEXT:    vmv.x.s ra, v16
+; RV32-NEXT:    xor a1, ra, a1
+; RV32-NEXT:    vslidedown.vi v8, v24, 5
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 7
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 9
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 11
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 13
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 15
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 17
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 19
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 21
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 23
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 25
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 27
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 29
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    vslidedown.vi v8, v24, 31
+; RV32-NEXT:    vmv.x.s s1, v8
+; RV32-NEXT:    xor a1, a1, s1
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    xor a1, a1, a3
+; RV32-NEXT:    xor a1, a1, a4
+; RV32-NEXT:    xor a1, a1, a5
+; RV32-NEXT:    xor a1, a1, a6
+; RV32-NEXT:    xor a1, a1, a7
+; RV32-NEXT:    xor a1, a1, t0
+; RV32-NEXT:    xor a1, a1, t1
+; RV32-NEXT:    xor a1, a1, t2
+; RV32-NEXT:    xor a1, a1, t3
+; RV32-NEXT:    xor a1, a1, t4
+; RV32-NEXT:    xor a1, a1, t5
+; RV32-NEXT:    xor a1, a1, t6
+; RV32-NEXT:    xor a1, a1, s0
+; RV32-NEXT:    xor a1, a1, s2
+; RV32-NEXT:    xor a1, a1, s3
+; RV32-NEXT:    xor a1, a1, s4
+; RV32-NEXT:    xor a1, a1, s5
+; RV32-NEXT:    xor a1, a1, s6
+; RV32-NEXT:    xor a1, a1, s7
+; RV32-NEXT:    xor a1, a1, s8
+; RV32-NEXT:    xor a1, a1, s9
+; RV32-NEXT:    xor a1, a1, s10
+; RV32-NEXT:    xor a1, a1, s11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 12(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 16(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 20(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 24(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 28(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 32(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 36(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 40(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 44(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 48(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 52(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 56(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 60(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 64(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 68(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 72(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 76(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 80(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 84(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 88(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 92(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 96(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 100(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 104(a2) # 4-byte Folded Reload
+; RV32-NEXT:    xor a1, a1, a2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add sp, sp, a2
+; RV32-NEXT:    lw s11, 108(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s10, 112(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s9, 116(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s8, 120(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s7, 124(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s6, 128(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s5, 132(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s4, 136(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s3, 140(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s2, 144(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s1, 148(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s0, 152(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw ra, 156(sp) # 4-byte Folded Reload
+; RV32-NEXT:    addi sp, sp, 160
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_xor_v64i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a1, a0, 384
+; RV64-NEXT:    vle64.v v16, (a1)
+; RV64-NEXT:    addi a1, a0, 256
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v24, (a0)
+; RV64-NEXT:    vle64.v v0, (a1)
+; RV64-NEXT:    vxor.vv v16, v24, v16
+; RV64-NEXT:    vxor.vv v8, v8, v0
+; RV64-NEXT:    vxor.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredxor.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i64>, <64 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.xor.v64i64(<64 x i64> %v)
+  ret i64 %red
+}
+
+declare i8 @llvm.vector.reduce.smin.v1i8(<1 x i8>)
+
+define i8 @vreduce_smin_v1i8(<1 x i8>* %x) {
+; CHECK-LABEL: vreduce_smin_v1i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i8>, <1 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smin.v1i8(<1 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smin.v2i8(<2 x i8>)
+
+define i8 @vreduce_smin_v2i8(<2 x i8>* %x) {
+; CHECK-LABEL: vreduce_smin_v2i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    addi a0, zero, 127
+; CHECK-NEXT:    vsetvli a1, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vredmin.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i8>, <2 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smin.v2i8(<2 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smin.v4i8(<4 x i8>)
+
+define i8 @vreduce_smin_v4i8(<4 x i8>* %x) {
+; CHECK-LABEL: vreduce_smin_v4i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    addi a0, zero, 127
+; CHECK-NEXT:    vsetvli a1, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vredmin.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i8>, <4 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smin.v4i8(<4 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smin.v8i8(<8 x i8>)
+
+define i8 @vreduce_smin_v8i8(<8 x i8>* %x) {
+; CHECK-LABEL: vreduce_smin_v8i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    addi a0, zero, 127
+; CHECK-NEXT:    vsetvli a1, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vredmin.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i8>, <8 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smin.v8i8(<8 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smin.v16i8(<16 x i8>)
+
+define i8 @vreduce_smin_v16i8(<16 x i8>* %x) {
+; CHECK-LABEL: vreduce_smin_v16i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    addi a0, zero, 127
+; CHECK-NEXT:    vsetvli a1, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vredmin.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i8>, <16 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smin.v16i8(<16 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smin.v32i8(<32 x i8>)
+
+define i8 @vreduce_smin_v32i8(<32 x i8>* %x) {
+; CHECK-LABEL: vreduce_smin_v32i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vle8.v v26, (a0)
+; CHECK-NEXT:    addi a0, zero, 127
+; CHECK-NEXT:    vsetvli a2, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vredmin.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i8>, <32 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smin.v32i8(<32 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smin.v64i8(<64 x i8>)
+
+define i8 @vreduce_smin_v64i8(<64 x i8>* %x) {
+; CHECK-LABEL: vreduce_smin_v64i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a0)
+; CHECK-NEXT:    addi a0, zero, 127
+; CHECK-NEXT:    vsetvli a2, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vredmin.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i8>, <64 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smin.v64i8(<64 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smin.v128i8(<128 x i8>)
+
+define i8 @vreduce_smin_v128i8(<128 x i8>* %x) {
+; CHECK-LABEL: vreduce_smin_v128i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, zero, 127
+; CHECK-NEXT:    vsetvli a2, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredmin.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i8>, <128 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smin.v128i8(<128 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smin.v256i8(<256 x i8>)
+
+define i8 @vreduce_smin_v256i8(<256 x i8>* %x) {
+; CHECK-LABEL: vreduce_smin_v256i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle8.v v16, (a0)
+; CHECK-NEXT:    vmin.vv v8, v8, v16
+; CHECK-NEXT:    addi a0, zero, 127
+; CHECK-NEXT:    vsetvli a2, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredmin.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <256 x i8>, <256 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smin.v256i8(<256 x i8> %v)
+  ret i8 %red
+}
+
+declare i16 @llvm.vector.reduce.smin.v1i16(<1 x i16>)
+
+define i16 @vreduce_smin_v1i16(<1 x i16>* %x) {
+; CHECK-LABEL: vreduce_smin_v1i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i16>, <1 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smin.v1i16(<1 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smin.v2i16(<2 x i16>)
+
+define i16 @vreduce_smin_v2i16(<2 x i16>* %x) {
+; RV32-LABEL: vreduce_smin_v2i16:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; RV32-NEXT:    vle16.v v25, (a0)
+; RV32-NEXT:    lui a0, 8
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v26, a0
+; RV32-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v2i16:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; RV64-NEXT:    vle16.v v25, (a0)
+; RV64-NEXT:    lui a0, 8
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v26, a0
+; RV64-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i16>, <2 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smin.v2i16(<2 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smin.v4i16(<4 x i16>)
+
+define i16 @vreduce_smin_v4i16(<4 x i16>* %x) {
+; RV32-LABEL: vreduce_smin_v4i16:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; RV32-NEXT:    vle16.v v25, (a0)
+; RV32-NEXT:    lui a0, 8
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v26, a0
+; RV32-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v4i16:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; RV64-NEXT:    vle16.v v25, (a0)
+; RV64-NEXT:    lui a0, 8
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v26, a0
+; RV64-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i16>, <4 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smin.v4i16(<4 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smin.v8i16(<8 x i16>)
+
+define i16 @vreduce_smin_v8i16(<8 x i16>* %x) {
+; RV32-LABEL: vreduce_smin_v8i16:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; RV32-NEXT:    vle16.v v25, (a0)
+; RV32-NEXT:    lui a0, 8
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v26, a0
+; RV32-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v8i16:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; RV64-NEXT:    vle16.v v25, (a0)
+; RV64-NEXT:    lui a0, 8
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v26, a0
+; RV64-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i16>, <8 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smin.v8i16(<8 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smin.v16i16(<16 x i16>)
+
+define i16 @vreduce_smin_v16i16(<16 x i16>* %x) {
+; RV32-LABEL: vreduce_smin_v16i16:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; RV32-NEXT:    vle16.v v26, (a0)
+; RV32-NEXT:    lui a0, 8
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v16i16:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; RV64-NEXT:    vle16.v v26, (a0)
+; RV64-NEXT:    lui a0, 8
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i16>, <16 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smin.v16i16(<16 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smin.v32i16(<32 x i16>)
+
+define i16 @vreduce_smin_v32i16(<32 x i16>* %x) {
+; RV32-LABEL: vreduce_smin_v32i16:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; RV32-NEXT:    vle16.v v28, (a0)
+; RV32-NEXT:    lui a0, 8
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a2, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v32i16:
+; RV64:       # %bb.0:
+; RV64-NEXT:    addi a1, zero, 32
+; RV64-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; RV64-NEXT:    vle16.v v28, (a0)
+; RV64-NEXT:    lui a0, 8
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a2, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i16>, <32 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smin.v32i16(<32 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smin.v64i16(<64 x i16>)
+
+define i16 @vreduce_smin_v64i16(<64 x i16>* %x) {
+; RV32-LABEL: vreduce_smin_v64i16:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi a1, zero, 64
+; RV32-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; RV32-NEXT:    vle16.v v8, (a0)
+; RV32-NEXT:    lui a0, 8
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a2, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v64i16:
+; RV64:       # %bb.0:
+; RV64-NEXT:    addi a1, zero, 64
+; RV64-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; RV64-NEXT:    vle16.v v8, (a0)
+; RV64-NEXT:    lui a0, 8
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a2, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i16>, <64 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smin.v64i16(<64 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smin.v128i16(<128 x i16>)
+
+define i16 @vreduce_smin_v128i16(<128 x i16>* %x) {
+; RV32-LABEL: vreduce_smin_v128i16:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi a1, zero, 64
+; RV32-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; RV32-NEXT:    vle16.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle16.v v16, (a0)
+; RV32-NEXT:    vmin.vv v8, v8, v16
+; RV32-NEXT:    lui a0, 8
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a2, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v128i16:
+; RV64:       # %bb.0:
+; RV64-NEXT:    addi a1, zero, 64
+; RV64-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; RV64-NEXT:    vle16.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle16.v v16, (a0)
+; RV64-NEXT:    vmin.vv v8, v8, v16
+; RV64-NEXT:    lui a0, 8
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a2, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <128 x i16>, <128 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smin.v128i16(<128 x i16> %v)
+  ret i16 %red
+}
+
+declare i32 @llvm.vector.reduce.smin.v1i32(<1 x i32>)
+
+define i32 @vreduce_smin_v1i32(<1 x i32>* %x) {
+; CHECK-LABEL: vreduce_smin_v1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i32>, <1 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smin.v1i32(<1 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smin.v2i32(<2 x i32>)
+
+define i32 @vreduce_smin_v2i32(<2 x i32>* %x) {
+; RV32-LABEL: vreduce_smin_v2i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; RV32-NEXT:    vle32.v v25, (a0)
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v26, a0
+; RV32-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v2i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; RV64-NEXT:    vle32.v v25, (a0)
+; RV64-NEXT:    lui a0, 524288
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v26, a0
+; RV64-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i32>, <2 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smin.v2i32(<2 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smin.v4i32(<4 x i32>)
+
+define i32 @vreduce_smin_v4i32(<4 x i32>* %x) {
+; RV32-LABEL: vreduce_smin_v4i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; RV32-NEXT:    vle32.v v25, (a0)
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v26, a0
+; RV32-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v4i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; RV64-NEXT:    vle32.v v25, (a0)
+; RV64-NEXT:    lui a0, 524288
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v26, a0
+; RV64-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i32>, <4 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smin.v4i32(<4 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smin.v8i32(<8 x i32>)
+
+define i32 @vreduce_smin_v8i32(<8 x i32>* %x) {
+; RV32-LABEL: vreduce_smin_v8i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; RV32-NEXT:    vle32.v v26, (a0)
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v8i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; RV64-NEXT:    vle32.v v26, (a0)
+; RV64-NEXT:    lui a0, 524288
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i32>, <8 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smin.v8i32(<8 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smin.v16i32(<16 x i32>)
+
+define i32 @vreduce_smin_v16i32(<16 x i32>* %x) {
+; RV32-LABEL: vreduce_smin_v16i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; RV32-NEXT:    vle32.v v28, (a0)
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v16i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; RV64-NEXT:    vle32.v v28, (a0)
+; RV64-NEXT:    lui a0, 524288
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i32>, <16 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smin.v16i32(<16 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smin.v32i32(<32 x i32>)
+
+define i32 @vreduce_smin_v32i32(<32 x i32>* %x) {
+; RV32-LABEL: vreduce_smin_v32i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a0)
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a2, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v32i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    addi a1, zero, 32
+; RV64-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV64-NEXT:    vle32.v v8, (a0)
+; RV64-NEXT:    lui a0, 524288
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a2, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i32>, <32 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smin.v32i32(<32 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smin.v64i32(<64 x i32>)
+
+define i32 @vreduce_smin_v64i32(<64 x i32>* %x) {
+; RV32-LABEL: vreduce_smin_v64i32:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle32.v v16, (a0)
+; RV32-NEXT:    vmin.vv v8, v8, v16
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vsetvli a2, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v64i32:
+; RV64:       # %bb.0:
+; RV64-NEXT:    addi a1, zero, 32
+; RV64-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV64-NEXT:    vle32.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle32.v v16, (a0)
+; RV64-NEXT:    vmin.vv v8, v8, v16
+; RV64-NEXT:    lui a0, 524288
+; RV64-NEXT:    addiw a0, a0, -1
+; RV64-NEXT:    vsetvli a2, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i32>, <64 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smin.v64i32(<64 x i32> %v)
+  ret i32 %red
+}
+
+declare i64 @llvm.vector.reduce.smin.v1i64(<1 x i64>)
+
+define i64 @vreduce_smin_v1i64(<1 x i64>* %x) {
+; RV32-LABEL: vreduce_smin_v1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <1 x i64>, <1 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smin.v1i64(<1 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smin.v2i64(<2 x i64>)
+
+define i64 @vreduce_smin_v2i64(<2 x i64>* %x) {
+; RV32-LABEL: vreduce_smin_v2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v26, -1
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsll.vx v26, v26, a1
+; RV32-NEXT:    vsrl.vx v26, v26, a1
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vmv.v.x v27, a0
+; RV32-NEXT:    vsll.vx v27, v27, a1
+; RV32-NEXT:    vor.vv v26, v26, v27
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    addi a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v26, a0
+; RV64-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i64>, <2 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smin.v2i64(<2 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smin.v4i64(<4 x i64>)
+
+define i64 @vreduce_smin_v4i64(<4 x i64>* %x) {
+; RV32-LABEL: vreduce_smin_v4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV32-NEXT:    vle64.v v26, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, -1
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsll.vx v25, v25, a1
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vmv.v.x v28, a0
+; RV32-NEXT:    vsll.vx v28, v28, a1
+; RV32-NEXT:    vor.vv v25, v25, v28
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV64-NEXT:    vle64.v v26, (a0)
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    addi a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i64>, <4 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smin.v4i64(<4 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smin.v8i64(<8 x i64>)
+
+define i64 @vreduce_smin_v8i64(<8 x i64>* %x) {
+; RV32-LABEL: vreduce_smin_v8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV32-NEXT:    vle64.v v28, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, -1
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsll.vx v25, v25, a1
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vmv.v.x v26, a0
+; RV32-NEXT:    vsll.vx v26, v26, a1
+; RV32-NEXT:    vor.vv v25, v25, v26
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV64-NEXT:    vle64.v v28, (a0)
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    addi a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i64>, <8 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smin.v8i64(<8 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smin.v16i64(<16 x i64>)
+
+define i64 @vreduce_smin_v16i64(<16 x i64>* %x) {
+; RV32-LABEL: vreduce_smin_v16i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, -1
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsll.vx v25, v25, a1
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    addi a0, a0, -1
+; RV32-NEXT:    vmv.v.x v26, a0
+; RV32-NEXT:    vsll.vx v26, v26, a1
+; RV32-NEXT:    vor.vv v25, v25, v26
+; RV32-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV32-NEXT:    vredmin.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v16i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    addi a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i64>, <16 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smin.v16i64(<16 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smin.v32i64(<32 x i64>)
+
+define i64 @vreduce_smin_v32i64(<32 x i64>* %x) {
+; RV32-LABEL: vreduce_smin_v32i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle64.v v16, (a0)
+; RV32-NEXT:    vmin.vv v8, v8, v16
+; RV32-NEXT:    vsetivli a0, 8, e64,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vmin.vv v28, v8, v16
+; RV32-NEXT:    vsetivli a0, 4, e64,m4,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v28, 4
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vmin.vv v26, v28, v8
+; RV32-NEXT:    vsetivli a0, 2, e64,m2,ta,mu
+; RV32-NEXT:    vslidedown.vi v28, v26, 2
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vmin.vv v25, v26, v28
+; RV32-NEXT:    vsetivli a0, 1, e64,m1,ta,mu
+; RV32-NEXT:    vslidedown.vi v26, v25, 1
+; RV32-NEXT:    vmin.vv v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smin_v32i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v16, (a0)
+; RV64-NEXT:    vmin.vv v8, v8, v16
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    addi a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i64>, <32 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smin.v32i64(<32 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smin.v64i64(<64 x i64>)
+
+define i64 @vreduce_smin_v64i64(<64 x i64>* %x) nounwind {
+; RV32-LABEL: vreduce_smin_v64i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi sp, sp, -480
+; RV32-NEXT:    sw ra, 476(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s0, 472(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s1, 468(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s2, 464(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s3, 460(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s4, 456(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s5, 452(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s6, 448(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s7, 444(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s8, 440(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s9, 436(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s10, 432(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s11, 428(sp) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    sub sp, sp, a1
+; RV32-NEXT:    addi a2, a0, 384
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 320(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 324(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 312(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 316(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 304(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 308(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 296(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 300(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 288(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 292(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 280(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 284(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 272(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 276(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 264(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 268(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 256(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 260(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 248(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 252(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 240(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 244(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 232(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 236(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 388(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 228(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 384(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 224(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 216(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 220(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 212(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 424(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 256
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 380(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 208(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 200(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 204(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 196(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 420(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 376(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 192(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 184(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 188(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 180(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 416(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 372(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 176(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 168(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 172(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 164(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 412(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 368(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 160(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 364(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 156(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 148(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 152(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 144(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 408(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 360(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 140(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 356(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 136(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 128(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 132(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 128
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 124(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 404(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 352(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 120(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 112(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 116(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s0, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 400(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 348(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 108(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 344(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 104(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 340(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 100(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 336(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 84(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 92(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 96(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s9, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 396(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 332(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 80(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 72(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 76(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 64(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 68(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 56(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 60(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s7, v0
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 392(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 328(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 48(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a1, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a0)
+; RV32-NEXT:    vsetivli a0, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 224
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 216
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 208
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 200
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 192
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 184
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 176
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 168
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 160
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 152
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 144
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 136
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 7
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 120
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 112
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 104
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 96
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 88
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 80
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 72
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 6
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 56
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 48
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 40
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 5
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 3
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 24
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 224
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 40(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 216
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 44(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 208
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 32(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 200
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 36(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 192
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 28(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 184
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 12(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 176
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 24(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 168
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t2, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 160
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 20(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 152
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a6, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 144
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 16(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 136
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t0, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 7
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s2, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 120
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t3, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 112
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s8, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 104
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s11, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 96
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s ra, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 88
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s1, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 80
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 72
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t1, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 6
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 56
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 48
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s6, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 40
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s3, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 5
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a7, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s10, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 3
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t4, v0
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a4, v0
+; RV32-NEXT:    vmv.x.s a0, v8
+; RV32-NEXT:    vmv.x.s a3, v24
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t6, zero, 24
+; RV32-NEXT:    mul a2, a2, t6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t6, v8
+; RV32-NEXT:    mv s4, a0
+; RV32-NEXT:    blt a1, a3, .LBB154_1
+; RV32-NEXT:    j .LBB154_358
+; RV32-NEXT:  .LBB154_1:
+; RV32-NEXT:    bltu a0, t6, .LBB154_2
+; RV32-NEXT:    j .LBB154_359
+; RV32-NEXT:  .LBB154_2:
+; RV32-NEXT:    beq a1, a3, .LBB154_3
+; RV32-NEXT:    j .LBB154_360
+; RV32-NEXT:  .LBB154_3:
+; RV32-NEXT:    blt a1, a3, .LBB154_4
+; RV32-NEXT:    j .LBB154_361
+; RV32-NEXT:  .LBB154_4:
+; RV32-NEXT:    mv a3, a0
+; RV32-NEXT:    blt a1, a4, .LBB154_5
+; RV32-NEXT:    j .LBB154_362
+; RV32-NEXT:  .LBB154_5:
+; RV32-NEXT:    bltu a0, t4, .LBB154_6
+; RV32-NEXT:    j .LBB154_363
+; RV32-NEXT:  .LBB154_6:
+; RV32-NEXT:    beq a1, a4, .LBB154_8
+; RV32-NEXT:  .LBB154_7:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB154_8:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw s0, 88(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw s9, 52(a2) # 4-byte Folded Spill
+; RV32-NEXT:    mv t4, s7
+; RV32-NEXT:    blt a1, a4, .LBB154_9
+; RV32-NEXT:    j .LBB154_364
+; RV32-NEXT:  .LBB154_9:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, s10, .LBB154_10
+; RV32-NEXT:    j .LBB154_365
+; RV32-NEXT:  .LBB154_10:
+; RV32-NEXT:    bltu a0, a7, .LBB154_11
+; RV32-NEXT:    j .LBB154_366
+; RV32-NEXT:  .LBB154_11:
+; RV32-NEXT:    beq a1, s10, .LBB154_12
+; RV32-NEXT:    j .LBB154_367
+; RV32-NEXT:  .LBB154_12:
+; RV32-NEXT:    blt a1, s10, .LBB154_13
+; RV32-NEXT:    j .LBB154_368
+; RV32-NEXT:  .LBB154_13:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, s3, .LBB154_15
+; RV32-NEXT:  .LBB154_14:
+; RV32-NEXT:    mv a4, s6
+; RV32-NEXT:  .LBB154_15:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s10, 84(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s6, .LBB154_16
+; RV32-NEXT:    j .LBB154_369
+; RV32-NEXT:  .LBB154_16:
+; RV32-NEXT:    beq a1, s3, .LBB154_17
+; RV32-NEXT:    j .LBB154_370
+; RV32-NEXT:  .LBB154_17:
+; RV32-NEXT:    blt a1, s3, .LBB154_18
+; RV32-NEXT:    j .LBB154_371
+; RV32-NEXT:  .LBB154_18:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, a5, .LBB154_19
+; RV32-NEXT:    j .LBB154_372
+; RV32-NEXT:  .LBB154_19:
+; RV32-NEXT:    bltu a0, t5, .LBB154_20
+; RV32-NEXT:    j .LBB154_373
+; RV32-NEXT:  .LBB154_20:
+; RV32-NEXT:    beq a1, a5, .LBB154_21
+; RV32-NEXT:    j .LBB154_374
+; RV32-NEXT:  .LBB154_21:
+; RV32-NEXT:    blt a1, a5, .LBB154_22
+; RV32-NEXT:    j .LBB154_375
+; RV32-NEXT:  .LBB154_22:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, t1, .LBB154_23
+; RV32-NEXT:    j .LBB154_376
+; RV32-NEXT:  .LBB154_23:
+; RV32-NEXT:    bltu a0, s5, .LBB154_24
+; RV32-NEXT:    j .LBB154_377
+; RV32-NEXT:  .LBB154_24:
+; RV32-NEXT:    beq a1, t1, .LBB154_26
+; RV32-NEXT:  .LBB154_25:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB154_26:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s5, 104(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, t1, .LBB154_28
+; RV32-NEXT:  # %bb.27:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:  .LBB154_28:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, s1, .LBB154_30
+; RV32-NEXT:  # %bb.29:
+; RV32-NEXT:    mv a4, ra
+; RV32-NEXT:  .LBB154_30:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t1, 100(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, ra, .LBB154_31
+; RV32-NEXT:    j .LBB154_378
+; RV32-NEXT:  .LBB154_31:
+; RV32-NEXT:    beq a1, s1, .LBB154_32
+; RV32-NEXT:    j .LBB154_379
+; RV32-NEXT:  .LBB154_32:
+; RV32-NEXT:    blt a1, s1, .LBB154_33
+; RV32-NEXT:    j .LBB154_380
+; RV32-NEXT:  .LBB154_33:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, s11, .LBB154_35
+; RV32-NEXT:  .LBB154_34:
+; RV32-NEXT:    mv a4, s8
+; RV32-NEXT:  .LBB154_35:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t5, 48(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s8, .LBB154_37
+; RV32-NEXT:  # %bb.36:
+; RV32-NEXT:    mv a0, s8
+; RV32-NEXT:  .LBB154_37:
+; RV32-NEXT:    beq a1, s11, .LBB154_39
+; RV32-NEXT:  # %bb.38:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB154_39:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a7, 132(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, s11, .LBB154_41
+; RV32-NEXT:  # %bb.40:
+; RV32-NEXT:    mv a1, s11
+; RV32-NEXT:  .LBB154_41:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, t3, .LBB154_43
+; RV32-NEXT:  # %bb.42:
+; RV32-NEXT:    mv a4, s2
+; RV32-NEXT:  .LBB154_43:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s3, 96(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s11, 92(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s2, .LBB154_44
+; RV32-NEXT:    j .LBB154_381
+; RV32-NEXT:  .LBB154_44:
+; RV32-NEXT:    beq a1, t3, .LBB154_45
+; RV32-NEXT:    j .LBB154_382
+; RV32-NEXT:  .LBB154_45:
+; RV32-NEXT:    blt a1, t3, .LBB154_47
+; RV32-NEXT:  .LBB154_46:
+; RV32-NEXT:    mv a1, t3
+; RV32-NEXT:  .LBB154_47:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s0, 16(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, t0, .LBB154_49
+; RV32-NEXT:  # %bb.48:
+; RV32-NEXT:    mv a2, s0
+; RV32-NEXT:  .LBB154_49:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s1, 24(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 20(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s0, .LBB154_51
+; RV32-NEXT:  # %bb.50:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:  .LBB154_51:
+; RV32-NEXT:    beq a1, t0, .LBB154_53
+; RV32-NEXT:  # %bb.52:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_53:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s8, 116(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, t0, .LBB154_55
+; RV32-NEXT:  # %bb.54:
+; RV32-NEXT:    mv a1, t0
+; RV32-NEXT:  .LBB154_55:
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    blt a1, a6, .LBB154_57
+; RV32-NEXT:  # %bb.56:
+; RV32-NEXT:    mv s0, a4
+; RV32-NEXT:  .LBB154_57:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t0, 152(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s6, 80(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 28(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a4, .LBB154_58
+; RV32-NEXT:    j .LBB154_383
+; RV32-NEXT:  .LBB154_58:
+; RV32-NEXT:    beq a1, a6, .LBB154_59
+; RV32-NEXT:    j .LBB154_384
+; RV32-NEXT:  .LBB154_59:
+; RV32-NEXT:    blt a1, a6, .LBB154_60
+; RV32-NEXT:    j .LBB154_385
+; RV32-NEXT:  .LBB154_60:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, t2, .LBB154_62
+; RV32-NEXT:  .LBB154_61:
+; RV32-NEXT:    mv a4, s1
+; RV32-NEXT:  .LBB154_62:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 36(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a6, 32(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s1, .LBB154_63
+; RV32-NEXT:    j .LBB154_386
+; RV32-NEXT:  .LBB154_63:
+; RV32-NEXT:    beq a1, t2, .LBB154_64
+; RV32-NEXT:    j .LBB154_387
+; RV32-NEXT:  .LBB154_64:
+; RV32-NEXT:    blt a1, t2, .LBB154_66
+; RV32-NEXT:  .LBB154_65:
+; RV32-NEXT:    mv a1, t2
+; RV32-NEXT:  .LBB154_66:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t6, 108(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s1, 12(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, s1, .LBB154_67
+; RV32-NEXT:    j .LBB154_388
+; RV32-NEXT:  .LBB154_67:
+; RV32-NEXT:    bltu a0, a5, .LBB154_68
+; RV32-NEXT:    j .LBB154_389
+; RV32-NEXT:  .LBB154_68:
+; RV32-NEXT:    beq a1, s1, .LBB154_69
+; RV32-NEXT:    j .LBB154_390
+; RV32-NEXT:  .LBB154_69:
+; RV32-NEXT:    blt a1, s1, .LBB154_70
+; RV32-NEXT:    j .LBB154_391
+; RV32-NEXT:  .LBB154_70:
+; RV32-NEXT:    mv a5, a0
+; RV32-NEXT:    blt a1, s0, .LBB154_72
+; RV32-NEXT:  .LBB154_71:
+; RV32-NEXT:    mv a5, a6
+; RV32-NEXT:  .LBB154_72:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t3, 120(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 44(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 40(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a6, .LBB154_73
+; RV32-NEXT:    j .LBB154_392
+; RV32-NEXT:  .LBB154_73:
+; RV32-NEXT:    beq a1, s0, .LBB154_74
+; RV32-NEXT:    j .LBB154_393
+; RV32-NEXT:  .LBB154_74:
+; RV32-NEXT:    blt a1, s0, .LBB154_75
+; RV32-NEXT:    j .LBB154_394
+; RV32-NEXT:  .LBB154_75:
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    blt a1, s1, .LBB154_77
+; RV32-NEXT:  .LBB154_76:
+; RV32-NEXT:    mv s0, a2
+; RV32-NEXT:  .LBB154_77:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw ra, 172(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s2, 76(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s7, 72(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 68(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a6, 64(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 60(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 56(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, a2, .LBB154_395
+; RV32-NEXT:  # %bb.78:
+; RV32-NEXT:    bne a1, s1, .LBB154_396
+; RV32-NEXT:  .LBB154_79:
+; RV32-NEXT:    bge a1, s1, .LBB154_397
+; RV32-NEXT:  .LBB154_80:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, t5, .LBB154_82
+; RV32-NEXT:  .LBB154_81:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a2, a2, t2
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 328(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_82:
+; RV32-NEXT:    csrr t2, vlenb
+; RV32-NEXT:    addi s0, zero, 232
+; RV32-NEXT:    mul t2, t2, s0
+; RV32-NEXT:    add t2, sp, t2
+; RV32-NEXT:    lw t2, 208(t2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr s0, vlenb
+; RV32-NEXT:    addi s1, zero, 232
+; RV32-NEXT:    mul s0, s0, s1
+; RV32-NEXT:    add s0, sp, s0
+; RV32-NEXT:    lw s4, 140(s0) # 4-byte Folded Reload
+; RV32-NEXT:    csrr s0, vlenb
+; RV32-NEXT:    addi s1, zero, 232
+; RV32-NEXT:    mul s0, s0, s1
+; RV32-NEXT:    add s0, sp, s0
+; RV32-NEXT:    lw s9, 328(s0) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s9, .LBB154_398
+; RV32-NEXT:  # %bb.83:
+; RV32-NEXT:    bne a1, t5, .LBB154_399
+; RV32-NEXT:  .LBB154_84:
+; RV32-NEXT:    blt a1, t5, .LBB154_86
+; RV32-NEXT:  .LBB154_85:
+; RV32-NEXT:    mv a1, t5
+; RV32-NEXT:  .LBB154_86:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr t5, vlenb
+; RV32-NEXT:    addi s0, zero, 232
+; RV32-NEXT:    mul t5, t5, s0
+; RV32-NEXT:    add t5, sp, t5
+; RV32-NEXT:    lw s1, 392(t5) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, s1, .LBB154_88
+; RV32-NEXT:  # %bb.87:
+; RV32-NEXT:    mv a2, t4
+; RV32-NEXT:  .LBB154_88:
+; RV32-NEXT:    bltu a0, t4, .LBB154_90
+; RV32-NEXT:  # %bb.89:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:  .LBB154_90:
+; RV32-NEXT:    csrr t4, vlenb
+; RV32-NEXT:    addi t5, zero, 232
+; RV32-NEXT:    mul t4, t4, t5
+; RV32-NEXT:    add t4, sp, t4
+; RV32-NEXT:    lw t4, 392(t4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB154_92
+; RV32-NEXT:  # %bb.91:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_92:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t4, zero, 232
+; RV32-NEXT:    mul a2, a2, t4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 392(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a2, .LBB154_400
+; RV32-NEXT:  # %bb.93:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a5, .LBB154_401
+; RV32-NEXT:  .LBB154_94:
+; RV32-NEXT:    bgeu a0, a3, .LBB154_402
+; RV32-NEXT:  .LBB154_95:
+; RV32-NEXT:    beq a1, a5, .LBB154_97
+; RV32-NEXT:  .LBB154_96:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_97:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 52(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a5, .LBB154_403
+; RV32-NEXT:  # %bb.98:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a4, .LBB154_404
+; RV32-NEXT:  .LBB154_99:
+; RV32-NEXT:    bgeu a0, a6, .LBB154_405
+; RV32-NEXT:  .LBB154_100:
+; RV32-NEXT:    beq a1, a4, .LBB154_102
+; RV32-NEXT:  .LBB154_101:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_102:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 136(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a4, .LBB154_406
+; RV32-NEXT:  # %bb.103:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s2, .LBB154_407
+; RV32-NEXT:  .LBB154_104:
+; RV32-NEXT:    bgeu a0, s7, .LBB154_408
+; RV32-NEXT:  .LBB154_105:
+; RV32-NEXT:    bne a1, s2, .LBB154_409
+; RV32-NEXT:  .LBB154_106:
+; RV32-NEXT:    bge a1, s2, .LBB154_410
+; RV32-NEXT:  .LBB154_107:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s6, .LBB154_109
+; RV32-NEXT:  .LBB154_108:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 332(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_109:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s7, 332(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s7, .LBB154_411
+; RV32-NEXT:  # %bb.110:
+; RV32-NEXT:    bne a1, s6, .LBB154_412
+; RV32-NEXT:  .LBB154_111:
+; RV32-NEXT:    blt a1, s6, .LBB154_113
+; RV32-NEXT:  .LBB154_112:
+; RV32-NEXT:    mv a1, s6
+; RV32-NEXT:  .LBB154_113:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a5, 396(a4) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a5, .LBB154_115
+; RV32-NEXT:  # %bb.114:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB154_115:
+; RV32-NEXT:    bltu a0, a3, .LBB154_117
+; RV32-NEXT:  # %bb.116:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB154_117:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t4, 396(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 88(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB154_119
+; RV32-NEXT:  # %bb.118:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_119:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 396(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 112(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a2, .LBB154_413
+; RV32-NEXT:  # %bb.120:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s3, .LBB154_414
+; RV32-NEXT:  .LBB154_121:
+; RV32-NEXT:    bgeu a0, s11, .LBB154_415
+; RV32-NEXT:  .LBB154_122:
+; RV32-NEXT:    bne a1, s3, .LBB154_416
+; RV32-NEXT:  .LBB154_123:
+; RV32-NEXT:    bge a1, s3, .LBB154_417
+; RV32-NEXT:  .LBB154_124:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s10, .LBB154_126
+; RV32-NEXT:  .LBB154_125:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 336(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_126:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s11, 336(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s11, .LBB154_418
+; RV32-NEXT:  # %bb.127:
+; RV32-NEXT:    bne a1, s10, .LBB154_419
+; RV32-NEXT:  .LBB154_128:
+; RV32-NEXT:    bge a1, s10, .LBB154_420
+; RV32-NEXT:  .LBB154_129:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, t1, .LBB154_131
+; RV32-NEXT:  .LBB154_130:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 340(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_131:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s10, 340(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s10, .LBB154_421
+; RV32-NEXT:  # %bb.132:
+; RV32-NEXT:    bne a1, t1, .LBB154_422
+; RV32-NEXT:  .LBB154_133:
+; RV32-NEXT:    bge a1, t1, .LBB154_423
+; RV32-NEXT:  .LBB154_134:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s5, .LBB154_136
+; RV32-NEXT:  .LBB154_135:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 344(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_136:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t1, 344(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, t1, .LBB154_424
+; RV32-NEXT:  # %bb.137:
+; RV32-NEXT:    bne a1, s5, .LBB154_425
+; RV32-NEXT:  .LBB154_138:
+; RV32-NEXT:    mv t1, t2
+; RV32-NEXT:    bge a1, s5, .LBB154_426
+; RV32-NEXT:  .LBB154_139:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, t6, .LBB154_141
+; RV32-NEXT:  .LBB154_140:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 348(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_141:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s5, 348(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s5, .LBB154_427
+; RV32-NEXT:  # %bb.142:
+; RV32-NEXT:    bne a1, t6, .LBB154_428
+; RV32-NEXT:  .LBB154_143:
+; RV32-NEXT:    blt a1, t6, .LBB154_145
+; RV32-NEXT:  .LBB154_144:
+; RV32-NEXT:    mv a1, t6
+; RV32-NEXT:  .LBB154_145:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 400(a4) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a4, .LBB154_147
+; RV32-NEXT:  # %bb.146:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:  .LBB154_147:
+; RV32-NEXT:    bltu a0, a5, .LBB154_149
+; RV32-NEXT:  # %bb.148:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:  .LBB154_149:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t4, 400(a4) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 124(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB154_151
+; RV32-NEXT:  # %bb.150:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_151:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 400(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 128(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a2, .LBB154_429
+; RV32-NEXT:  # %bb.152:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s8, .LBB154_430
+; RV32-NEXT:  .LBB154_153:
+; RV32-NEXT:    bgeu a0, a3, .LBB154_431
+; RV32-NEXT:  .LBB154_154:
+; RV32-NEXT:    bne a1, s8, .LBB154_432
+; RV32-NEXT:  .LBB154_155:
+; RV32-NEXT:    bge a1, s8, .LBB154_433
+; RV32-NEXT:  .LBB154_156:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, t3, .LBB154_158
+; RV32-NEXT:  .LBB154_157:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 352(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_158:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s9, 352(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s9, .LBB154_434
+; RV32-NEXT:  # %bb.159:
+; RV32-NEXT:    bne a1, t3, .LBB154_435
+; RV32-NEXT:  .LBB154_160:
+; RV32-NEXT:    blt a1, t3, .LBB154_162
+; RV32-NEXT:  .LBB154_161:
+; RV32-NEXT:    mv a1, t3
+; RV32-NEXT:  .LBB154_162:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 404(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a3, .LBB154_164
+; RV32-NEXT:  # %bb.163:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB154_164:
+; RV32-NEXT:    bltu a0, a4, .LBB154_166
+; RV32-NEXT:  # %bb.165:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB154_166:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s10, 404(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 164(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s10, .LBB154_168
+; RV32-NEXT:  # %bb.167:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_168:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 404(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 144(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a2, .LBB154_436
+; RV32-NEXT:  # %bb.169:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a7, .LBB154_437
+; RV32-NEXT:  .LBB154_170:
+; RV32-NEXT:    bgeu a0, a5, .LBB154_438
+; RV32-NEXT:  .LBB154_171:
+; RV32-NEXT:    beq a1, a7, .LBB154_173
+; RV32-NEXT:  .LBB154_172:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_173:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 156(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a7, .LBB154_175
+; RV32-NEXT:  # %bb.174:
+; RV32-NEXT:    mv a1, a7
+; RV32-NEXT:  .LBB154_175:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s1, .LBB154_177
+; RV32-NEXT:  # %bb.176:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 356(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_177:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s8, 356(a6) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s2, 196(a6) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw a7, 220(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s8, .LBB154_439
+; RV32-NEXT:  # %bb.178:
+; RV32-NEXT:    bne a1, s1, .LBB154_440
+; RV32-NEXT:  .LBB154_179:
+; RV32-NEXT:    bge a1, s1, .LBB154_441
+; RV32-NEXT:  .LBB154_180:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s4, .LBB154_182
+; RV32-NEXT:  .LBB154_181:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 360(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_182:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw t5, 360(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, t5, .LBB154_442
+; RV32-NEXT:  # %bb.183:
+; RV32-NEXT:    bne a1, s4, .LBB154_443
+; RV32-NEXT:  .LBB154_184:
+; RV32-NEXT:    blt a1, s4, .LBB154_186
+; RV32-NEXT:  .LBB154_185:
+; RV32-NEXT:    mv a1, s4
+; RV32-NEXT:  .LBB154_186:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s0, 408(a6) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, s0, .LBB154_188
+; RV32-NEXT:  # %bb.187:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB154_188:
+; RV32-NEXT:    bltu a0, a4, .LBB154_190
+; RV32-NEXT:  # %bb.189:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB154_190:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s11, 408(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s11, .LBB154_192
+; RV32-NEXT:  # %bb.191:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_192:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 408(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 148(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a2, .LBB154_444
+; RV32-NEXT:  # %bb.193:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, t0, .LBB154_445
+; RV32-NEXT:  .LBB154_194:
+; RV32-NEXT:    bgeu a0, a4, .LBB154_446
+; RV32-NEXT:  .LBB154_195:
+; RV32-NEXT:    beq a1, t0, .LBB154_197
+; RV32-NEXT:  .LBB154_196:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_197:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 168(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, t0, .LBB154_199
+; RV32-NEXT:  # %bb.198:
+; RV32-NEXT:    mv a1, t0
+; RV32-NEXT:  .LBB154_199:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a5, .LBB154_201
+; RV32-NEXT:  # %bb.200:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 364(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_201:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a6, a6, t0
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw t0, 364(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, t0, .LBB154_203
+; RV32-NEXT:  # %bb.202:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:  .LBB154_203:
+; RV32-NEXT:    beq a1, a5, .LBB154_205
+; RV32-NEXT:  # %bb.204:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_205:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t0, 200(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 184(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 204(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a5, .LBB154_207
+; RV32-NEXT:  # %bb.206:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB154_207:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 160(a5) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a5, .LBB154_209
+; RV32-NEXT:  # %bb.208:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 368(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_209:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s4, 368(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s4, .LBB154_447
+; RV32-NEXT:  # %bb.210:
+; RV32-NEXT:    bne a1, a5, .LBB154_448
+; RV32-NEXT:  .LBB154_211:
+; RV32-NEXT:    blt a1, a5, .LBB154_213
+; RV32-NEXT:  .LBB154_212:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB154_213:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 412(a5) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a5, .LBB154_215
+; RV32-NEXT:  # %bb.214:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB154_215:
+; RV32-NEXT:    bltu a0, a3, .LBB154_217
+; RV32-NEXT:  # %bb.216:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB154_217:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t5, 412(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 216(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t5, .LBB154_219
+; RV32-NEXT:  # %bb.218:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_219:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 412(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 180(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a2, .LBB154_449
+; RV32-NEXT:  # %bb.220:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, ra, .LBB154_450
+; RV32-NEXT:  .LBB154_221:
+; RV32-NEXT:    bgeu a0, a4, .LBB154_451
+; RV32-NEXT:  .LBB154_222:
+; RV32-NEXT:    beq a1, ra, .LBB154_224
+; RV32-NEXT:  .LBB154_223:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_224:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 176(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, ra, .LBB154_226
+; RV32-NEXT:  # %bb.225:
+; RV32-NEXT:    mv a1, ra
+; RV32-NEXT:  .LBB154_226:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a4, .LBB154_228
+; RV32-NEXT:  # %bb.227:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 372(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_228:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s7, 372(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s7, .LBB154_452
+; RV32-NEXT:  # %bb.229:
+; RV32-NEXT:    bne a1, a4, .LBB154_453
+; RV32-NEXT:  .LBB154_230:
+; RV32-NEXT:    blt a1, a4, .LBB154_232
+; RV32-NEXT:  .LBB154_231:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB154_232:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 416(a4) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a4, .LBB154_234
+; RV32-NEXT:  # %bb.233:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB154_234:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a6, 244(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a3, .LBB154_236
+; RV32-NEXT:  # %bb.235:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB154_236:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s9, 416(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t5, 264(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 212(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s9, .LBB154_238
+; RV32-NEXT:  # %bb.237:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_238:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 416(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 188(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a2, .LBB154_454
+; RV32-NEXT:  # %bb.239:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a3, .LBB154_455
+; RV32-NEXT:  .LBB154_240:
+; RV32-NEXT:    bgeu a0, s0, .LBB154_456
+; RV32-NEXT:  .LBB154_241:
+; RV32-NEXT:    bne a1, a3, .LBB154_457
+; RV32-NEXT:  .LBB154_242:
+; RV32-NEXT:    blt a1, a3, .LBB154_244
+; RV32-NEXT:  .LBB154_243:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB154_244:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 192(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a3, .LBB154_246
+; RV32-NEXT:  # %bb.245:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a2, a2, t2
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 376(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_246:
+; RV32-NEXT:    csrr t2, vlenb
+; RV32-NEXT:    addi t3, zero, 232
+; RV32-NEXT:    mul t2, t2, t3
+; RV32-NEXT:    add t2, sp, t2
+; RV32-NEXT:    lw s11, 376(t2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s11, .LBB154_458
+; RV32-NEXT:  # %bb.247:
+; RV32-NEXT:    bne a1, a3, .LBB154_459
+; RV32-NEXT:  .LBB154_248:
+; RV32-NEXT:    blt a1, a3, .LBB154_250
+; RV32-NEXT:  .LBB154_249:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB154_250:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 420(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a3, .LBB154_252
+; RV32-NEXT:  # %bb.251:
+; RV32-NEXT:    mv a2, s2
+; RV32-NEXT:  .LBB154_252:
+; RV32-NEXT:    bltu a0, s2, .LBB154_254
+; RV32-NEXT:  # %bb.253:
+; RV32-NEXT:    mv a0, s2
+; RV32-NEXT:  .LBB154_254:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t2, 420(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t3, zero, 232
+; RV32-NEXT:    mul a3, a3, t3
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t4, 232(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t2, .LBB154_256
+; RV32-NEXT:  # %bb.255:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_256:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 420(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a2, .LBB154_258
+; RV32-NEXT:  # %bb.257:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:  .LBB154_258:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 228(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, s1, .LBB154_460
+; RV32-NEXT:  # %bb.259:
+; RV32-NEXT:    bgeu a0, t0, .LBB154_461
+; RV32-NEXT:  .LBB154_260:
+; RV32-NEXT:    bne a1, s1, .LBB154_462
+; RV32-NEXT:  .LBB154_261:
+; RV32-NEXT:    bge a1, s1, .LBB154_463
+; RV32-NEXT:  .LBB154_262:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, t1, .LBB154_264
+; RV32-NEXT:  .LBB154_263:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a2, a2, t0
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 380(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_264:
+; RV32-NEXT:    csrr t0, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul t0, t0, t2
+; RV32-NEXT:    add t0, sp, t0
+; RV32-NEXT:    lw t0, 380(t0) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, t0, .LBB154_464
+; RV32-NEXT:  # %bb.265:
+; RV32-NEXT:    bne a1, t1, .LBB154_465
+; RV32-NEXT:  .LBB154_266:
+; RV32-NEXT:    blt a1, t1, .LBB154_268
+; RV32-NEXT:  .LBB154_267:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:  .LBB154_268:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr t0, vlenb
+; RV32-NEXT:    addi t1, zero, 232
+; RV32-NEXT:    mul t0, t0, t1
+; RV32-NEXT:    add t0, sp, t0
+; RV32-NEXT:    lw s0, 424(t0) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, s0, .LBB154_270
+; RV32-NEXT:  # %bb.269:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB154_270:
+; RV32-NEXT:    bltu a0, a4, .LBB154_272
+; RV32-NEXT:  # %bb.271:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB154_272:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a4, a4, t0
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t3, 424(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t3, .LBB154_274
+; RV32-NEXT:  # %bb.273:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_274:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 424(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a4, a4, t0
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 224(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a2, .LBB154_466
+; RV32-NEXT:  # %bb.275:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a7, .LBB154_467
+; RV32-NEXT:  .LBB154_276:
+; RV32-NEXT:    bgeu a0, a5, .LBB154_468
+; RV32-NEXT:  .LBB154_277:
+; RV32-NEXT:    bne a1, a7, .LBB154_469
+; RV32-NEXT:  .LBB154_278:
+; RV32-NEXT:    bge a1, a7, .LBB154_470
+; RV32-NEXT:  .LBB154_279:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a4, .LBB154_281
+; RV32-NEXT:  .LBB154_280:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 384(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_281:
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a5, a5, a7
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw t1, 384(a5) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a5, a5, a7
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw t0, 312(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, t1, .LBB154_283
+; RV32-NEXT:  # %bb.282:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:  .LBB154_283:
+; RV32-NEXT:    beq a1, a4, .LBB154_285
+; RV32-NEXT:  # %bb.284:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_285:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 236(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a4, .LBB154_287
+; RV32-NEXT:  # %bb.286:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB154_287:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a3, .LBB154_289
+; RV32-NEXT:  # %bb.288:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 388(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB154_289:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a7, 388(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a7, .LBB154_291
+; RV32-NEXT:  # %bb.290:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:  .LBB154_291:
+; RV32-NEXT:    beq a1, a3, .LBB154_293
+; RV32-NEXT:  # %bb.292:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_293:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t1, 256(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 260(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a7, 272(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 252(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 248(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a3, .LBB154_471
+; RV32-NEXT:  # %bb.294:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s1, .LBB154_472
+; RV32-NEXT:  .LBB154_295:
+; RV32-NEXT:    bgeu a0, t4, .LBB154_473
+; RV32-NEXT:  .LBB154_296:
+; RV32-NEXT:    beq a1, s1, .LBB154_298
+; RV32-NEXT:  .LBB154_297:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_298:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 240(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, s1, .LBB154_474
+; RV32-NEXT:  # %bb.299:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a6, .LBB154_475
+; RV32-NEXT:  .LBB154_300:
+; RV32-NEXT:    bgeu a0, a3, .LBB154_476
+; RV32-NEXT:  .LBB154_301:
+; RV32-NEXT:    bne a1, a6, .LBB154_477
+; RV32-NEXT:  .LBB154_302:
+; RV32-NEXT:    bge a1, a6, .LBB154_478
+; RV32-NEXT:  .LBB154_303:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a4, .LBB154_305
+; RV32-NEXT:  .LBB154_304:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:  .LBB154_305:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a6, 304(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 284(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a5, .LBB154_307
+; RV32-NEXT:  # %bb.306:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:  .LBB154_307:
+; RV32-NEXT:    beq a1, a4, .LBB154_309
+; RV32-NEXT:  # %bb.308:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_309:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 268(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a4, .LBB154_479
+; RV32-NEXT:  # %bb.310:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s0, .LBB154_480
+; RV32-NEXT:  .LBB154_311:
+; RV32-NEXT:    bgeu a0, t1, .LBB154_481
+; RV32-NEXT:  .LBB154_312:
+; RV32-NEXT:    bne a1, s0, .LBB154_482
+; RV32-NEXT:  .LBB154_313:
+; RV32-NEXT:    bge a1, s0, .LBB154_483
+; RV32-NEXT:  .LBB154_314:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a5, .LBB154_484
+; RV32-NEXT:  .LBB154_315:
+; RV32-NEXT:    bgeu a0, t5, .LBB154_485
+; RV32-NEXT:  .LBB154_316:
+; RV32-NEXT:    beq a1, a5, .LBB154_318
+; RV32-NEXT:  .LBB154_317:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_318:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 276(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a5, .LBB154_486
+; RV32-NEXT:  # %bb.319:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a4, .LBB154_487
+; RV32-NEXT:  .LBB154_320:
+; RV32-NEXT:    bgeu a0, a7, .LBB154_488
+; RV32-NEXT:  .LBB154_321:
+; RV32-NEXT:    beq a1, a4, .LBB154_323
+; RV32-NEXT:  .LBB154_322:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_323:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 296(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 292(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a4, .LBB154_325
+; RV32-NEXT:  # %bb.324:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB154_325:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a4, a4, a7
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 280(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a3, .LBB154_489
+; RV32-NEXT:  # %bb.326:
+; RV32-NEXT:    bgeu a0, a4, .LBB154_490
+; RV32-NEXT:  .LBB154_327:
+; RV32-NEXT:    bne a1, a3, .LBB154_491
+; RV32-NEXT:  .LBB154_328:
+; RV32-NEXT:    blt a1, a3, .LBB154_330
+; RV32-NEXT:  .LBB154_329:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB154_330:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 288(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a5, .LBB154_332
+; RV32-NEXT:  # %bb.331:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB154_332:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a3, a3, a7
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s0, 316(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a4, .LBB154_334
+; RV32-NEXT:  # %bb.333:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB154_334:
+; RV32-NEXT:    beq a1, a5, .LBB154_336
+; RV32-NEXT:  # %bb.335:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_336:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 320(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a1, a5, .LBB154_338
+; RV32-NEXT:  # %bb.337:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB154_338:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 300(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a5, .LBB154_492
+; RV32-NEXT:  # %bb.339:
+; RV32-NEXT:    bgeu a0, s1, .LBB154_493
+; RV32-NEXT:  .LBB154_340:
+; RV32-NEXT:    bne a1, a5, .LBB154_494
+; RV32-NEXT:  .LBB154_341:
+; RV32-NEXT:    blt a1, a5, .LBB154_343
+; RV32-NEXT:  .LBB154_342:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB154_343:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 308(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a5, .LBB154_495
+; RV32-NEXT:  # %bb.344:
+; RV32-NEXT:    bgeu a0, a6, .LBB154_496
+; RV32-NEXT:  .LBB154_345:
+; RV32-NEXT:    beq a1, a5, .LBB154_347
+; RV32-NEXT:  .LBB154_346:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB154_347:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 324(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a1, a5, .LBB154_497
+; RV32-NEXT:  # %bb.348:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s0, .LBB154_498
+; RV32-NEXT:  .LBB154_349:
+; RV32-NEXT:    bgeu a0, t0, .LBB154_499
+; RV32-NEXT:  .LBB154_350:
+; RV32-NEXT:    bne a1, s0, .LBB154_500
+; RV32-NEXT:  .LBB154_351:
+; RV32-NEXT:    bge a1, s0, .LBB154_501
+; RV32-NEXT:  .LBB154_352:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a3, .LBB154_502
+; RV32-NEXT:  .LBB154_353:
+; RV32-NEXT:    bgeu a0, a4, .LBB154_503
+; RV32-NEXT:  .LBB154_354:
+; RV32-NEXT:    bne a1, a3, .LBB154_504
+; RV32-NEXT:  .LBB154_355:
+; RV32-NEXT:    blt a1, a3, .LBB154_357
+; RV32-NEXT:  .LBB154_356:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB154_357:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add sp, sp, a2
+; RV32-NEXT:    lw s11, 428(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s10, 432(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s9, 436(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s8, 440(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s7, 444(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s6, 448(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s5, 452(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s4, 456(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s3, 460(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s2, 464(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s1, 468(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s0, 472(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw ra, 476(sp) # 4-byte Folded Reload
+; RV32-NEXT:    addi sp, sp, 480
+; RV32-NEXT:    ret
+; RV32-NEXT:  .LBB154_358:
+; RV32-NEXT:    mv s4, t6
+; RV32-NEXT:    bgeu a0, t6, .LBB154_359
+; RV32-NEXT:    j .LBB154_2
+; RV32-NEXT:  .LBB154_359:
+; RV32-NEXT:    mv a0, t6
+; RV32-NEXT:    bne a1, a3, .LBB154_360
+; RV32-NEXT:    j .LBB154_3
+; RV32-NEXT:  .LBB154_360:
+; RV32-NEXT:    mv a0, s4
+; RV32-NEXT:    bge a1, a3, .LBB154_361
+; RV32-NEXT:    j .LBB154_4
+; RV32-NEXT:  .LBB154_361:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:    mv a3, a0
+; RV32-NEXT:    bge a1, a4, .LBB154_362
+; RV32-NEXT:    j .LBB154_5
+; RV32-NEXT:  .LBB154_362:
+; RV32-NEXT:    mv a3, t4
+; RV32-NEXT:    bgeu a0, t4, .LBB154_363
+; RV32-NEXT:    j .LBB154_6
+; RV32-NEXT:  .LBB154_363:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:    beq a1, a4, .LBB154_505
+; RV32-NEXT:    j .LBB154_7
+; RV32-NEXT:  .LBB154_505:
+; RV32-NEXT:    j .LBB154_8
+; RV32-NEXT:  .LBB154_364:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bge a1, s10, .LBB154_365
+; RV32-NEXT:    j .LBB154_10
+; RV32-NEXT:  .LBB154_365:
+; RV32-NEXT:    mv a4, a7
+; RV32-NEXT:    bgeu a0, a7, .LBB154_366
+; RV32-NEXT:    j .LBB154_11
+; RV32-NEXT:  .LBB154_366:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:    bne a1, s10, .LBB154_367
+; RV32-NEXT:    j .LBB154_12
+; RV32-NEXT:  .LBB154_367:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bge a1, s10, .LBB154_368
+; RV32-NEXT:    j .LBB154_13
+; RV32-NEXT:  .LBB154_368:
+; RV32-NEXT:    mv a1, s10
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, s3, .LBB154_506
+; RV32-NEXT:    j .LBB154_14
+; RV32-NEXT:  .LBB154_506:
+; RV32-NEXT:    j .LBB154_15
+; RV32-NEXT:  .LBB154_369:
+; RV32-NEXT:    mv a0, s6
+; RV32-NEXT:    bne a1, s3, .LBB154_370
+; RV32-NEXT:    j .LBB154_17
+; RV32-NEXT:  .LBB154_370:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bge a1, s3, .LBB154_371
+; RV32-NEXT:    j .LBB154_18
+; RV32-NEXT:  .LBB154_371:
+; RV32-NEXT:    mv a1, s3
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bge a1, a5, .LBB154_372
+; RV32-NEXT:    j .LBB154_19
+; RV32-NEXT:  .LBB154_372:
+; RV32-NEXT:    mv a4, t5
+; RV32-NEXT:    bgeu a0, t5, .LBB154_373
+; RV32-NEXT:    j .LBB154_20
+; RV32-NEXT:  .LBB154_373:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    bne a1, a5, .LBB154_374
+; RV32-NEXT:    j .LBB154_21
+; RV32-NEXT:  .LBB154_374:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bge a1, a5, .LBB154_375
+; RV32-NEXT:    j .LBB154_22
+; RV32-NEXT:  .LBB154_375:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bge a1, t1, .LBB154_376
+; RV32-NEXT:    j .LBB154_23
+; RV32-NEXT:  .LBB154_376:
+; RV32-NEXT:    mv a4, s5
+; RV32-NEXT:    bgeu a0, s5, .LBB154_377
+; RV32-NEXT:    j .LBB154_24
+; RV32-NEXT:  .LBB154_377:
+; RV32-NEXT:    mv a0, s5
+; RV32-NEXT:    beq a1, t1, .LBB154_507
+; RV32-NEXT:    j .LBB154_25
+; RV32-NEXT:  .LBB154_507:
+; RV32-NEXT:    j .LBB154_26
+; RV32-NEXT:  .LBB154_378:
+; RV32-NEXT:    mv a0, ra
+; RV32-NEXT:    bne a1, s1, .LBB154_379
+; RV32-NEXT:    j .LBB154_32
+; RV32-NEXT:  .LBB154_379:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bge a1, s1, .LBB154_380
+; RV32-NEXT:    j .LBB154_33
+; RV32-NEXT:  .LBB154_380:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, s11, .LBB154_508
+; RV32-NEXT:    j .LBB154_34
+; RV32-NEXT:  .LBB154_508:
+; RV32-NEXT:    j .LBB154_35
+; RV32-NEXT:  .LBB154_381:
+; RV32-NEXT:    mv a0, s2
+; RV32-NEXT:    bne a1, t3, .LBB154_382
+; RV32-NEXT:    j .LBB154_45
+; RV32-NEXT:  .LBB154_382:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    blt a1, t3, .LBB154_509
+; RV32-NEXT:    j .LBB154_46
+; RV32-NEXT:  .LBB154_509:
+; RV32-NEXT:    j .LBB154_47
+; RV32-NEXT:  .LBB154_383:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, a6, .LBB154_384
+; RV32-NEXT:    j .LBB154_59
+; RV32-NEXT:  .LBB154_384:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    bge a1, a6, .LBB154_385
+; RV32-NEXT:    j .LBB154_60
+; RV32-NEXT:  .LBB154_385:
+; RV32-NEXT:    mv a1, a6
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a1, t2, .LBB154_510
+; RV32-NEXT:    j .LBB154_61
+; RV32-NEXT:  .LBB154_510:
+; RV32-NEXT:    j .LBB154_62
+; RV32-NEXT:  .LBB154_386:
+; RV32-NEXT:    mv a0, s1
+; RV32-NEXT:    bne a1, t2, .LBB154_387
+; RV32-NEXT:    j .LBB154_64
+; RV32-NEXT:  .LBB154_387:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    blt a1, t2, .LBB154_511
+; RV32-NEXT:    j .LBB154_65
+; RV32-NEXT:  .LBB154_511:
+; RV32-NEXT:    j .LBB154_66
+; RV32-NEXT:  .LBB154_388:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bgeu a0, a5, .LBB154_389
+; RV32-NEXT:    j .LBB154_68
+; RV32-NEXT:  .LBB154_389:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bne a1, s1, .LBB154_390
+; RV32-NEXT:    j .LBB154_69
+; RV32-NEXT:  .LBB154_390:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, s1, .LBB154_391
+; RV32-NEXT:    j .LBB154_70
+; RV32-NEXT:  .LBB154_391:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a5, a0
+; RV32-NEXT:    blt a1, s0, .LBB154_512
+; RV32-NEXT:    j .LBB154_71
+; RV32-NEXT:  .LBB154_512:
+; RV32-NEXT:    j .LBB154_72
+; RV32-NEXT:  .LBB154_392:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, s0, .LBB154_393
+; RV32-NEXT:    j .LBB154_74
+; RV32-NEXT:  .LBB154_393:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bge a1, s0, .LBB154_394
+; RV32-NEXT:    j .LBB154_75
+; RV32-NEXT:  .LBB154_394:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    blt a1, s1, .LBB154_513
+; RV32-NEXT:    j .LBB154_76
+; RV32-NEXT:  .LBB154_513:
+; RV32-NEXT:    j .LBB154_77
+; RV32-NEXT:  .LBB154_395:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    beq a1, s1, .LBB154_79
+; RV32-NEXT:  .LBB154_396:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    blt a1, s1, .LBB154_80
+; RV32-NEXT:  .LBB154_397:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, t5, .LBB154_81
+; RV32-NEXT:    j .LBB154_82
+; RV32-NEXT:  .LBB154_398:
+; RV32-NEXT:    mv a0, s9
+; RV32-NEXT:    beq a1, t5, .LBB154_84
+; RV32-NEXT:  .LBB154_399:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, t5, .LBB154_85
+; RV32-NEXT:    j .LBB154_86
+; RV32-NEXT:  .LBB154_400:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a5, .LBB154_94
+; RV32-NEXT:  .LBB154_401:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a0, a3, .LBB154_95
+; RV32-NEXT:  .LBB154_402:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    bne a1, a5, .LBB154_96
+; RV32-NEXT:    j .LBB154_97
+; RV32-NEXT:  .LBB154_403:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a4, .LBB154_99
+; RV32-NEXT:  .LBB154_404:
+; RV32-NEXT:    mv a2, a6
+; RV32-NEXT:    bltu a0, a6, .LBB154_100
+; RV32-NEXT:  .LBB154_405:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, a4, .LBB154_101
+; RV32-NEXT:    j .LBB154_102
+; RV32-NEXT:  .LBB154_406:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s2, .LBB154_104
+; RV32-NEXT:  .LBB154_407:
+; RV32-NEXT:    mv a2, s7
+; RV32-NEXT:    bltu a0, s7, .LBB154_105
+; RV32-NEXT:  .LBB154_408:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, s2, .LBB154_106
+; RV32-NEXT:  .LBB154_409:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, s2, .LBB154_107
+; RV32-NEXT:  .LBB154_410:
+; RV32-NEXT:    mv a1, s2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s6, .LBB154_108
+; RV32-NEXT:    j .LBB154_109
+; RV32-NEXT:  .LBB154_411:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, s6, .LBB154_111
+; RV32-NEXT:  .LBB154_412:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, s6, .LBB154_112
+; RV32-NEXT:    j .LBB154_113
+; RV32-NEXT:  .LBB154_413:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s3, .LBB154_121
+; RV32-NEXT:  .LBB154_414:
+; RV32-NEXT:    mv a2, s11
+; RV32-NEXT:    bltu a0, s11, .LBB154_122
+; RV32-NEXT:  .LBB154_415:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, s3, .LBB154_123
+; RV32-NEXT:  .LBB154_416:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, s3, .LBB154_124
+; RV32-NEXT:  .LBB154_417:
+; RV32-NEXT:    mv a1, s3
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s10, .LBB154_125
+; RV32-NEXT:    j .LBB154_126
+; RV32-NEXT:  .LBB154_418:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, s10, .LBB154_128
+; RV32-NEXT:  .LBB154_419:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, s10, .LBB154_129
+; RV32-NEXT:  .LBB154_420:
+; RV32-NEXT:    mv a1, s10
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, t1, .LBB154_130
+; RV32-NEXT:    j .LBB154_131
+; RV32-NEXT:  .LBB154_421:
+; RV32-NEXT:    mv a0, s10
+; RV32-NEXT:    beq a1, t1, .LBB154_133
+; RV32-NEXT:  .LBB154_422:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, t1, .LBB154_134
+; RV32-NEXT:  .LBB154_423:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s5, .LBB154_135
+; RV32-NEXT:    j .LBB154_136
+; RV32-NEXT:  .LBB154_424:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:    beq a1, s5, .LBB154_138
+; RV32-NEXT:  .LBB154_425:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    mv t1, t2
+; RV32-NEXT:    blt a1, s5, .LBB154_139
+; RV32-NEXT:  .LBB154_426:
+; RV32-NEXT:    mv a1, s5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, t6, .LBB154_140
+; RV32-NEXT:    j .LBB154_141
+; RV32-NEXT:  .LBB154_427:
+; RV32-NEXT:    mv a0, s5
+; RV32-NEXT:    beq a1, t6, .LBB154_143
+; RV32-NEXT:  .LBB154_428:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, t6, .LBB154_144
+; RV32-NEXT:    j .LBB154_145
+; RV32-NEXT:  .LBB154_429:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s8, .LBB154_153
+; RV32-NEXT:  .LBB154_430:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a0, a3, .LBB154_154
+; RV32-NEXT:  .LBB154_431:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    beq a1, s8, .LBB154_155
+; RV32-NEXT:  .LBB154_432:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, s8, .LBB154_156
+; RV32-NEXT:  .LBB154_433:
+; RV32-NEXT:    mv a1, s8
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, t3, .LBB154_157
+; RV32-NEXT:    j .LBB154_158
+; RV32-NEXT:  .LBB154_434:
+; RV32-NEXT:    mv a0, s9
+; RV32-NEXT:    beq a1, t3, .LBB154_160
+; RV32-NEXT:  .LBB154_435:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, t3, .LBB154_161
+; RV32-NEXT:    j .LBB154_162
+; RV32-NEXT:  .LBB154_436:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a7, .LBB154_170
+; RV32-NEXT:  .LBB154_437:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bltu a0, a5, .LBB154_171
+; RV32-NEXT:  .LBB154_438:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bne a1, a7, .LBB154_172
+; RV32-NEXT:    j .LBB154_173
+; RV32-NEXT:  .LBB154_439:
+; RV32-NEXT:    mv a0, s8
+; RV32-NEXT:    beq a1, s1, .LBB154_179
+; RV32-NEXT:  .LBB154_440:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, s1, .LBB154_180
+; RV32-NEXT:  .LBB154_441:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, s4, .LBB154_181
+; RV32-NEXT:    j .LBB154_182
+; RV32-NEXT:  .LBB154_442:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    beq a1, s4, .LBB154_184
+; RV32-NEXT:  .LBB154_443:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, s4, .LBB154_185
+; RV32-NEXT:    j .LBB154_186
+; RV32-NEXT:  .LBB154_444:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, t0, .LBB154_194
+; RV32-NEXT:  .LBB154_445:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a0, a4, .LBB154_195
+; RV32-NEXT:  .LBB154_446:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, t0, .LBB154_196
+; RV32-NEXT:    j .LBB154_197
+; RV32-NEXT:  .LBB154_447:
+; RV32-NEXT:    mv a0, s4
+; RV32-NEXT:    beq a1, a5, .LBB154_211
+; RV32-NEXT:  .LBB154_448:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, a5, .LBB154_212
+; RV32-NEXT:    j .LBB154_213
+; RV32-NEXT:  .LBB154_449:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, ra, .LBB154_221
+; RV32-NEXT:  .LBB154_450:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a0, a4, .LBB154_222
+; RV32-NEXT:  .LBB154_451:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, ra, .LBB154_223
+; RV32-NEXT:    j .LBB154_224
+; RV32-NEXT:  .LBB154_452:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, a4, .LBB154_230
+; RV32-NEXT:  .LBB154_453:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, a4, .LBB154_231
+; RV32-NEXT:    j .LBB154_232
+; RV32-NEXT:  .LBB154_454:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a3, .LBB154_240
+; RV32-NEXT:  .LBB154_455:
+; RV32-NEXT:    mv a2, s0
+; RV32-NEXT:    bltu a0, s0, .LBB154_241
+; RV32-NEXT:  .LBB154_456:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    beq a1, a3, .LBB154_242
+; RV32-NEXT:  .LBB154_457:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, a3, .LBB154_243
+; RV32-NEXT:    j .LBB154_244
+; RV32-NEXT:  .LBB154_458:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, a3, .LBB154_248
+; RV32-NEXT:  .LBB154_459:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, a3, .LBB154_249
+; RV32-NEXT:    j .LBB154_250
+; RV32-NEXT:  .LBB154_460:
+; RV32-NEXT:    mv a2, t0
+; RV32-NEXT:    bltu a0, t0, .LBB154_260
+; RV32-NEXT:  .LBB154_461:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, s1, .LBB154_261
+; RV32-NEXT:  .LBB154_462:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, s1, .LBB154_262
+; RV32-NEXT:  .LBB154_463:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, t1, .LBB154_263
+; RV32-NEXT:    j .LBB154_264
+; RV32-NEXT:  .LBB154_464:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, t1, .LBB154_266
+; RV32-NEXT:  .LBB154_465:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, t1, .LBB154_267
+; RV32-NEXT:    j .LBB154_268
+; RV32-NEXT:  .LBB154_466:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a7, .LBB154_276
+; RV32-NEXT:  .LBB154_467:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bltu a0, a5, .LBB154_277
+; RV32-NEXT:  .LBB154_468:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    beq a1, a7, .LBB154_278
+; RV32-NEXT:  .LBB154_469:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, a7, .LBB154_279
+; RV32-NEXT:  .LBB154_470:
+; RV32-NEXT:    mv a1, a7
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a4, .LBB154_280
+; RV32-NEXT:    j .LBB154_281
+; RV32-NEXT:  .LBB154_471:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s1, .LBB154_295
+; RV32-NEXT:  .LBB154_472:
+; RV32-NEXT:    mv a2, t4
+; RV32-NEXT:    bltu a0, t4, .LBB154_296
+; RV32-NEXT:  .LBB154_473:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:    bne a1, s1, .LBB154_297
+; RV32-NEXT:    j .LBB154_298
+; RV32-NEXT:  .LBB154_474:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a6, .LBB154_300
+; RV32-NEXT:  .LBB154_475:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a0, a3, .LBB154_301
+; RV32-NEXT:  .LBB154_476:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    beq a1, a6, .LBB154_302
+; RV32-NEXT:  .LBB154_477:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, a6, .LBB154_303
+; RV32-NEXT:  .LBB154_478:
+; RV32-NEXT:    mv a1, a6
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a1, a4, .LBB154_304
+; RV32-NEXT:    j .LBB154_305
+; RV32-NEXT:  .LBB154_479:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s0, .LBB154_311
+; RV32-NEXT:  .LBB154_480:
+; RV32-NEXT:    mv a2, t1
+; RV32-NEXT:    bltu a0, t1, .LBB154_312
+; RV32-NEXT:  .LBB154_481:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:    beq a1, s0, .LBB154_313
+; RV32-NEXT:  .LBB154_482:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, s0, .LBB154_314
+; RV32-NEXT:  .LBB154_483:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a5, .LBB154_315
+; RV32-NEXT:  .LBB154_484:
+; RV32-NEXT:    mv a2, t5
+; RV32-NEXT:    bltu a0, t5, .LBB154_316
+; RV32-NEXT:  .LBB154_485:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    bne a1, a5, .LBB154_317
+; RV32-NEXT:    j .LBB154_318
+; RV32-NEXT:  .LBB154_486:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a4, .LBB154_320
+; RV32-NEXT:  .LBB154_487:
+; RV32-NEXT:    mv a2, a7
+; RV32-NEXT:    bltu a0, a7, .LBB154_321
+; RV32-NEXT:  .LBB154_488:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:    bne a1, a4, .LBB154_322
+; RV32-NEXT:    j .LBB154_323
+; RV32-NEXT:  .LBB154_489:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a0, a4, .LBB154_327
+; RV32-NEXT:  .LBB154_490:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    beq a1, a3, .LBB154_328
+; RV32-NEXT:  .LBB154_491:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, a3, .LBB154_329
+; RV32-NEXT:    j .LBB154_330
+; RV32-NEXT:  .LBB154_492:
+; RV32-NEXT:    mv a2, s1
+; RV32-NEXT:    bltu a0, s1, .LBB154_340
+; RV32-NEXT:  .LBB154_493:
+; RV32-NEXT:    mv a0, s1
+; RV32-NEXT:    beq a1, a5, .LBB154_341
+; RV32-NEXT:  .LBB154_494:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, a5, .LBB154_342
+; RV32-NEXT:    j .LBB154_343
+; RV32-NEXT:  .LBB154_495:
+; RV32-NEXT:    mv a2, a6
+; RV32-NEXT:    bltu a0, a6, .LBB154_345
+; RV32-NEXT:  .LBB154_496:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, a5, .LBB154_346
+; RV32-NEXT:    j .LBB154_347
+; RV32-NEXT:  .LBB154_497:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, s0, .LBB154_349
+; RV32-NEXT:  .LBB154_498:
+; RV32-NEXT:    mv a2, t0
+; RV32-NEXT:    bltu a0, t0, .LBB154_350
+; RV32-NEXT:  .LBB154_499:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, s0, .LBB154_351
+; RV32-NEXT:  .LBB154_500:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a1, s0, .LBB154_352
+; RV32-NEXT:  .LBB154_501:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a1, a3, .LBB154_353
+; RV32-NEXT:  .LBB154_502:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a0, a4, .LBB154_354
+; RV32-NEXT:  .LBB154_503:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    beq a1, a3, .LBB154_355
+; RV32-NEXT:  .LBB154_504:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a1, a3, .LBB154_356
+; RV32-NEXT:    j .LBB154_357
+;
+; RV64-LABEL: vreduce_smin_v64i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a1, a0, 384
+; RV64-NEXT:    vle64.v v16, (a1)
+; RV64-NEXT:    addi a1, a0, 256
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v24, (a0)
+; RV64-NEXT:    vle64.v v0, (a1)
+; RV64-NEXT:    vmin.vv v16, v24, v16
+; RV64-NEXT:    vmin.vv v8, v8, v0
+; RV64-NEXT:    vmin.vv v8, v8, v16
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    addi a0, a0, -1
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredmin.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i64>, <64 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smin.v64i64(<64 x i64> %v)
+  ret i64 %red
+}
+
+declare i8 @llvm.vector.reduce.smax.v1i8(<1 x i8>)
+
+define i8 @vreduce_smax_v1i8(<1 x i8>* %x) {
+; CHECK-LABEL: vreduce_smax_v1i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i8>, <1 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smax.v1i8(<1 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smax.v2i8(<2 x i8>)
+
+define i8 @vreduce_smax_v2i8(<2 x i8>* %x) {
+; CHECK-LABEL: vreduce_smax_v2i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    addi a0, zero, -128
+; CHECK-NEXT:    vsetvli a1, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i8>, <2 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smax.v2i8(<2 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smax.v4i8(<4 x i8>)
+
+define i8 @vreduce_smax_v4i8(<4 x i8>* %x) {
+; CHECK-LABEL: vreduce_smax_v4i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    addi a0, zero, -128
+; CHECK-NEXT:    vsetvli a1, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i8>, <4 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smax.v4i8(<4 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smax.v8i8(<8 x i8>)
+
+define i8 @vreduce_smax_v8i8(<8 x i8>* %x) {
+; CHECK-LABEL: vreduce_smax_v8i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    addi a0, zero, -128
+; CHECK-NEXT:    vsetvli a1, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i8>, <8 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smax.v8i8(<8 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smax.v16i8(<16 x i8>)
+
+define i8 @vreduce_smax_v16i8(<16 x i8>* %x) {
+; CHECK-LABEL: vreduce_smax_v16i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    addi a0, zero, -128
+; CHECK-NEXT:    vsetvli a1, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i8>, <16 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smax.v16i8(<16 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smax.v32i8(<32 x i8>)
+
+define i8 @vreduce_smax_v32i8(<32 x i8>* %x) {
+; CHECK-LABEL: vreduce_smax_v32i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vle8.v v26, (a0)
+; CHECK-NEXT:    addi a0, zero, -128
+; CHECK-NEXT:    vsetvli a2, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i8>, <32 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smax.v32i8(<32 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smax.v64i8(<64 x i8>)
+
+define i8 @vreduce_smax_v64i8(<64 x i8>* %x) {
+; CHECK-LABEL: vreduce_smax_v64i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a0)
+; CHECK-NEXT:    addi a0, zero, -128
+; CHECK-NEXT:    vsetvli a2, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i8>, <64 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smax.v64i8(<64 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smax.v128i8(<128 x i8>)
+
+define i8 @vreduce_smax_v128i8(<128 x i8>* %x) {
+; CHECK-LABEL: vreduce_smax_v128i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, zero, -128
+; CHECK-NEXT:    vsetvli a2, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i8>, <128 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smax.v128i8(<128 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.smax.v256i8(<256 x i8>)
+
+define i8 @vreduce_smax_v256i8(<256 x i8>* %x) {
+; CHECK-LABEL: vreduce_smax_v256i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle8.v v16, (a0)
+; CHECK-NEXT:    vmax.vv v8, v8, v16
+; CHECK-NEXT:    addi a0, zero, -128
+; CHECK-NEXT:    vsetvli a2, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <256 x i8>, <256 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.smax.v256i8(<256 x i8> %v)
+  ret i8 %red
+}
+
+declare i16 @llvm.vector.reduce.smax.v1i16(<1 x i16>)
+
+define i16 @vreduce_smax_v1i16(<1 x i16>* %x) {
+; CHECK-LABEL: vreduce_smax_v1i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i16>, <1 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smax.v1i16(<1 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smax.v2i16(<2 x i16>)
+
+define i16 @vreduce_smax_v2i16(<2 x i16>* %x) {
+; CHECK-LABEL: vreduce_smax_v2i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    lui a0, 1048568
+; CHECK-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i16>, <2 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smax.v2i16(<2 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smax.v4i16(<4 x i16>)
+
+define i16 @vreduce_smax_v4i16(<4 x i16>* %x) {
+; CHECK-LABEL: vreduce_smax_v4i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    lui a0, 1048568
+; CHECK-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i16>, <4 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smax.v4i16(<4 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smax.v8i16(<8 x i16>)
+
+define i16 @vreduce_smax_v8i16(<8 x i16>* %x) {
+; CHECK-LABEL: vreduce_smax_v8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    lui a0, 1048568
+; CHECK-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i16>, <8 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smax.v8i16(<8 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smax.v16i16(<16 x i16>)
+
+define i16 @vreduce_smax_v16i16(<16 x i16>* %x) {
+; CHECK-LABEL: vreduce_smax_v16i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vle16.v v26, (a0)
+; CHECK-NEXT:    lui a0, 1048568
+; CHECK-NEXT:    vsetvli a1, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i16>, <16 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smax.v16i16(<16 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smax.v32i16(<32 x i16>)
+
+define i16 @vreduce_smax_v32i16(<32 x i16>* %x) {
+; CHECK-LABEL: vreduce_smax_v32i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a0)
+; CHECK-NEXT:    lui a0, 1048568
+; CHECK-NEXT:    vsetvli a2, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i16>, <32 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smax.v32i16(<32 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smax.v64i16(<64 x i16>)
+
+define i16 @vreduce_smax_v64i16(<64 x i16>* %x) {
+; CHECK-LABEL: vreduce_smax_v64i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    lui a0, 1048568
+; CHECK-NEXT:    vsetvli a2, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i16>, <64 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smax.v64i16(<64 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.smax.v128i16(<128 x i16>)
+
+define i16 @vreduce_smax_v128i16(<128 x i16>* %x) {
+; CHECK-LABEL: vreduce_smax_v128i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle16.v v16, (a0)
+; CHECK-NEXT:    vmax.vv v8, v8, v16
+; CHECK-NEXT:    lui a0, 1048568
+; CHECK-NEXT:    vsetvli a2, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i16>, <128 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.smax.v128i16(<128 x i16> %v)
+  ret i16 %red
+}
+
+declare i32 @llvm.vector.reduce.smax.v1i32(<1 x i32>)
+
+define i32 @vreduce_smax_v1i32(<1 x i32>* %x) {
+; CHECK-LABEL: vreduce_smax_v1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i32>, <1 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smax.v1i32(<1 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smax.v2i32(<2 x i32>)
+
+define i32 @vreduce_smax_v2i32(<2 x i32>* %x) {
+; CHECK-LABEL: vreduce_smax_v2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    lui a0, 524288
+; CHECK-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i32>, <2 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smax.v2i32(<2 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smax.v4i32(<4 x i32>)
+
+define i32 @vreduce_smax_v4i32(<4 x i32>* %x) {
+; CHECK-LABEL: vreduce_smax_v4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    lui a0, 524288
+; CHECK-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v26, a0
+; CHECK-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i32>, <4 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smax.v4i32(<4 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smax.v8i32(<8 x i32>)
+
+define i32 @vreduce_smax_v8i32(<8 x i32>* %x) {
+; CHECK-LABEL: vreduce_smax_v8i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vle32.v v26, (a0)
+; CHECK-NEXT:    lui a0, 524288
+; CHECK-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i32>, <8 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smax.v8i32(<8 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smax.v16i32(<16 x i32>)
+
+define i32 @vreduce_smax_v16i32(<16 x i32>* %x) {
+; CHECK-LABEL: vreduce_smax_v16i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a0)
+; CHECK-NEXT:    lui a0, 524288
+; CHECK-NEXT:    vsetvli a1, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i32>, <16 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smax.v16i32(<16 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smax.v32i32(<32 x i32>)
+
+define i32 @vreduce_smax_v32i32(<32 x i32>* %x) {
+; CHECK-LABEL: vreduce_smax_v32i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    lui a0, 524288
+; CHECK-NEXT:    vsetvli a2, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i32>, <32 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smax.v32i32(<32 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.smax.v64i32(<64 x i32>)
+
+define i32 @vreduce_smax_v64i32(<64 x i32>* %x) {
+; CHECK-LABEL: vreduce_smax_v64i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle32.v v16, (a0)
+; CHECK-NEXT:    vmax.vv v8, v8, v16
+; CHECK-NEXT:    lui a0, 524288
+; CHECK-NEXT:    vsetvli a2, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.x v25, a0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredmax.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i32>, <64 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.smax.v64i32(<64 x i32> %v)
+  ret i32 %red
+}
+
+declare i64 @llvm.vector.reduce.smax.v1i64(<1 x i64>)
+
+define i64 @vreduce_smax_v1i64(<1 x i64>* %x) {
+; RV32-LABEL: vreduce_smax_v1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smax_v1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <1 x i64>, <1 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smax.v1i64(<1 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smax.v2i64(<2 x i64>)
+
+define i64 @vreduce_smax_v2i64(<2 x i64>* %x) {
+; RV32-LABEL: vreduce_smax_v2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v26, a0
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsll.vx v26, v26, a1
+; RV32-NEXT:    vmv.v.i v27, 0
+; RV32-NEXT:    vsll.vx v27, v27, a1
+; RV32-NEXT:    vsrl.vx v27, v27, a1
+; RV32-NEXT:    vor.vv v26, v27, v26
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vredmax.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smax_v2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v26, a0
+; RV64-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV64-NEXT:    vredmax.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i64>, <2 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smax.v2i64(<2 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smax.v4i64(<4 x i64>)
+
+define i64 @vreduce_smax_v4i64(<4 x i64>* %x) {
+; RV32-LABEL: vreduce_smax_v4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV32-NEXT:    vle64.v v26, (a0)
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsll.vx v25, v25, a1
+; RV32-NEXT:    vmv.v.i v28, 0
+; RV32-NEXT:    vsll.vx v28, v28, a1
+; RV32-NEXT:    vsrl.vx v28, v28, a1
+; RV32-NEXT:    vor.vv v25, v28, v25
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vredmax.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smax_v4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV64-NEXT:    vle64.v v26, (a0)
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV64-NEXT:    vredmax.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i64>, <4 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smax.v4i64(<4 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smax.v8i64(<8 x i64>)
+
+define i64 @vreduce_smax_v8i64(<8 x i64>* %x) {
+; RV32-LABEL: vreduce_smax_v8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV32-NEXT:    vle64.v v28, (a0)
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsll.vx v25, v25, a1
+; RV32-NEXT:    vmv.v.i v26, 0
+; RV32-NEXT:    vsll.vx v26, v26, a1
+; RV32-NEXT:    vsrl.vx v26, v26, a1
+; RV32-NEXT:    vor.vv v25, v26, v25
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vredmax.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smax_v8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV64-NEXT:    vle64.v v28, (a0)
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV64-NEXT:    vredmax.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i64>, <8 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smax.v8i64(<8 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smax.v16i64(<16 x i64>)
+
+define i64 @vreduce_smax_v16i64(<16 x i64>* %x) {
+; RV32-LABEL: vreduce_smax_v16i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    lui a0, 524288
+; RV32-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.x v25, a0
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsll.vx v25, v25, a1
+; RV32-NEXT:    vmv.v.i v26, 0
+; RV32-NEXT:    vsll.vx v26, v26, a1
+; RV32-NEXT:    vsrl.vx v26, v26, a1
+; RV32-NEXT:    vor.vv v25, v26, v25
+; RV32-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV32-NEXT:    vredmax.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smax_v16i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredmax.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i64>, <16 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smax.v16i64(<16 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smax.v32i64(<32 x i64>)
+
+define i64 @vreduce_smax_v32i64(<32 x i64>* %x) {
+; RV32-LABEL: vreduce_smax_v32i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle64.v v16, (a0)
+; RV32-NEXT:    vmax.vv v8, v8, v16
+; RV32-NEXT:    vsetivli a0, 8, e64,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vmax.vv v28, v8, v16
+; RV32-NEXT:    vsetivli a0, 4, e64,m4,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v28, 4
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vmax.vv v26, v28, v8
+; RV32-NEXT:    vsetivli a0, 2, e64,m2,ta,mu
+; RV32-NEXT:    vslidedown.vi v28, v26, 2
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vmax.vv v25, v26, v28
+; RV32-NEXT:    vsetivli a0, 1, e64,m1,ta,mu
+; RV32-NEXT:    vslidedown.vi v26, v25, 1
+; RV32-NEXT:    vmax.vv v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_smax_v32i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v16, (a0)
+; RV64-NEXT:    vmax.vv v8, v8, v16
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredmax.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i64>, <32 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smax.v32i64(<32 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.smax.v64i64(<64 x i64>)
+
+define i64 @vreduce_smax_v64i64(<64 x i64>* %x) nounwind {
+; RV32-LABEL: vreduce_smax_v64i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi sp, sp, -480
+; RV32-NEXT:    sw ra, 476(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s0, 472(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s1, 468(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s2, 464(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s3, 460(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s4, 456(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s5, 452(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s6, 448(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s7, 444(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s8, 440(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s9, 436(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s10, 432(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s11, 428(sp) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    sub sp, sp, a1
+; RV32-NEXT:    addi a2, a0, 384
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 320(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 324(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 312(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 316(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 304(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 308(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 296(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 300(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 288(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 292(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 280(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 284(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 272(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 276(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 264(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 268(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 256(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 260(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 248(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 252(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 240(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 244(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 232(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 236(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 388(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 228(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 384(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 224(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 216(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 220(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 212(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 424(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 256
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 380(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 208(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 200(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 204(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 196(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 420(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 376(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 192(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 184(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 188(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 180(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 416(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 372(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 176(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 168(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 172(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 164(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 412(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 368(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 160(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 364(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 156(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 148(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 152(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 144(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 408(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 360(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 140(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 356(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 136(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 128(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 132(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 128
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 124(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 404(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 352(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 120(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 112(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 116(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s0, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 400(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 348(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 108(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 344(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 104(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 340(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 100(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 336(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 84(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 92(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 96(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s9, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 396(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 332(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 80(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 72(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 76(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 64(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 68(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 56(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 60(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s7, v0
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 392(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 328(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 48(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a1, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a0)
+; RV32-NEXT:    vsetivli a0, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 224
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 216
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 208
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 200
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 192
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 184
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 176
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 168
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 160
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 152
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 144
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 136
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 7
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 120
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 112
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 104
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 96
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 88
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 80
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 72
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 6
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 56
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 48
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 40
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 5
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 3
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 24
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 1
+; RV32-NEXT:    vslidedown.vi v16, v8, 3
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 224
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 40(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 216
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 44(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 208
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 32(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 200
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 36(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 192
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 28(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 184
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 12(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 176
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 24(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 168
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t2, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 160
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 20(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 152
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a6, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 144
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 16(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 136
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t0, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 7
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s2, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 120
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t3, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 112
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s8, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 104
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s11, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 96
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s ra, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 88
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s1, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 80
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 72
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t1, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 6
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 56
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 48
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s6, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 40
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s3, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 5
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a7, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s10, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 3
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t4, v0
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a4, v0
+; RV32-NEXT:    vmv.x.s a0, v8
+; RV32-NEXT:    vmv.x.s a1, v24
+; RV32-NEXT:    vmv.x.s a3, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t6, zero, 24
+; RV32-NEXT:    mul a2, a2, t6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t6, v8
+; RV32-NEXT:    mv s4, a0
+; RV32-NEXT:    blt a3, a1, .LBB185_1
+; RV32-NEXT:    j .LBB185_358
+; RV32-NEXT:  .LBB185_1:
+; RV32-NEXT:    bltu t6, a0, .LBB185_2
+; RV32-NEXT:    j .LBB185_359
+; RV32-NEXT:  .LBB185_2:
+; RV32-NEXT:    beq a1, a3, .LBB185_3
+; RV32-NEXT:    j .LBB185_360
+; RV32-NEXT:  .LBB185_3:
+; RV32-NEXT:    blt a3, a1, .LBB185_4
+; RV32-NEXT:    j .LBB185_361
+; RV32-NEXT:  .LBB185_4:
+; RV32-NEXT:    mv a3, a0
+; RV32-NEXT:    blt a4, a1, .LBB185_5
+; RV32-NEXT:    j .LBB185_362
+; RV32-NEXT:  .LBB185_5:
+; RV32-NEXT:    bltu t4, a0, .LBB185_6
+; RV32-NEXT:    j .LBB185_363
+; RV32-NEXT:  .LBB185_6:
+; RV32-NEXT:    beq a1, a4, .LBB185_8
+; RV32-NEXT:  .LBB185_7:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB185_8:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw s0, 88(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw s9, 52(a2) # 4-byte Folded Spill
+; RV32-NEXT:    mv t4, s7
+; RV32-NEXT:    blt a4, a1, .LBB185_9
+; RV32-NEXT:    j .LBB185_364
+; RV32-NEXT:  .LBB185_9:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt s10, a1, .LBB185_10
+; RV32-NEXT:    j .LBB185_365
+; RV32-NEXT:  .LBB185_10:
+; RV32-NEXT:    bltu a7, a0, .LBB185_11
+; RV32-NEXT:    j .LBB185_366
+; RV32-NEXT:  .LBB185_11:
+; RV32-NEXT:    beq a1, s10, .LBB185_12
+; RV32-NEXT:    j .LBB185_367
+; RV32-NEXT:  .LBB185_12:
+; RV32-NEXT:    blt s10, a1, .LBB185_13
+; RV32-NEXT:    j .LBB185_368
+; RV32-NEXT:  .LBB185_13:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt s3, a1, .LBB185_15
+; RV32-NEXT:  .LBB185_14:
+; RV32-NEXT:    mv a4, s6
+; RV32-NEXT:  .LBB185_15:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s10, 84(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s6, a0, .LBB185_16
+; RV32-NEXT:    j .LBB185_369
+; RV32-NEXT:  .LBB185_16:
+; RV32-NEXT:    beq a1, s3, .LBB185_17
+; RV32-NEXT:    j .LBB185_370
+; RV32-NEXT:  .LBB185_17:
+; RV32-NEXT:    blt s3, a1, .LBB185_18
+; RV32-NEXT:    j .LBB185_371
+; RV32-NEXT:  .LBB185_18:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt a5, a1, .LBB185_19
+; RV32-NEXT:    j .LBB185_372
+; RV32-NEXT:  .LBB185_19:
+; RV32-NEXT:    bltu t5, a0, .LBB185_20
+; RV32-NEXT:    j .LBB185_373
+; RV32-NEXT:  .LBB185_20:
+; RV32-NEXT:    beq a1, a5, .LBB185_21
+; RV32-NEXT:    j .LBB185_374
+; RV32-NEXT:  .LBB185_21:
+; RV32-NEXT:    blt a5, a1, .LBB185_22
+; RV32-NEXT:    j .LBB185_375
+; RV32-NEXT:  .LBB185_22:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt t1, a1, .LBB185_23
+; RV32-NEXT:    j .LBB185_376
+; RV32-NEXT:  .LBB185_23:
+; RV32-NEXT:    bltu s5, a0, .LBB185_24
+; RV32-NEXT:    j .LBB185_377
+; RV32-NEXT:  .LBB185_24:
+; RV32-NEXT:    beq a1, t1, .LBB185_26
+; RV32-NEXT:  .LBB185_25:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB185_26:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s5, 104(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt t1, a1, .LBB185_28
+; RV32-NEXT:  # %bb.27:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:  .LBB185_28:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt s1, a1, .LBB185_30
+; RV32-NEXT:  # %bb.29:
+; RV32-NEXT:    mv a4, ra
+; RV32-NEXT:  .LBB185_30:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t1, 100(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu ra, a0, .LBB185_31
+; RV32-NEXT:    j .LBB185_378
+; RV32-NEXT:  .LBB185_31:
+; RV32-NEXT:    beq a1, s1, .LBB185_32
+; RV32-NEXT:    j .LBB185_379
+; RV32-NEXT:  .LBB185_32:
+; RV32-NEXT:    blt s1, a1, .LBB185_33
+; RV32-NEXT:    j .LBB185_380
+; RV32-NEXT:  .LBB185_33:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt s11, a1, .LBB185_35
+; RV32-NEXT:  .LBB185_34:
+; RV32-NEXT:    mv a4, s8
+; RV32-NEXT:  .LBB185_35:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t5, 48(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s8, a0, .LBB185_37
+; RV32-NEXT:  # %bb.36:
+; RV32-NEXT:    mv a0, s8
+; RV32-NEXT:  .LBB185_37:
+; RV32-NEXT:    beq a1, s11, .LBB185_39
+; RV32-NEXT:  # %bb.38:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB185_39:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a7, 132(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt s11, a1, .LBB185_41
+; RV32-NEXT:  # %bb.40:
+; RV32-NEXT:    mv a1, s11
+; RV32-NEXT:  .LBB185_41:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt t3, a1, .LBB185_43
+; RV32-NEXT:  # %bb.42:
+; RV32-NEXT:    mv a4, s2
+; RV32-NEXT:  .LBB185_43:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s3, 96(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s11, 92(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s2, a0, .LBB185_44
+; RV32-NEXT:    j .LBB185_381
+; RV32-NEXT:  .LBB185_44:
+; RV32-NEXT:    beq a1, t3, .LBB185_45
+; RV32-NEXT:    j .LBB185_382
+; RV32-NEXT:  .LBB185_45:
+; RV32-NEXT:    blt t3, a1, .LBB185_47
+; RV32-NEXT:  .LBB185_46:
+; RV32-NEXT:    mv a1, t3
+; RV32-NEXT:  .LBB185_47:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s0, 16(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt t0, a1, .LBB185_49
+; RV32-NEXT:  # %bb.48:
+; RV32-NEXT:    mv a2, s0
+; RV32-NEXT:  .LBB185_49:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s1, 24(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 20(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s0, a0, .LBB185_51
+; RV32-NEXT:  # %bb.50:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:  .LBB185_51:
+; RV32-NEXT:    beq a1, t0, .LBB185_53
+; RV32-NEXT:  # %bb.52:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_53:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s8, 116(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt t0, a1, .LBB185_55
+; RV32-NEXT:  # %bb.54:
+; RV32-NEXT:    mv a1, t0
+; RV32-NEXT:  .LBB185_55:
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    blt a6, a1, .LBB185_57
+; RV32-NEXT:  # %bb.56:
+; RV32-NEXT:    mv s0, a4
+; RV32-NEXT:  .LBB185_57:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t0, 152(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s6, 80(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 28(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a4, a0, .LBB185_58
+; RV32-NEXT:    j .LBB185_383
+; RV32-NEXT:  .LBB185_58:
+; RV32-NEXT:    beq a1, a6, .LBB185_59
+; RV32-NEXT:    j .LBB185_384
+; RV32-NEXT:  .LBB185_59:
+; RV32-NEXT:    blt a6, a1, .LBB185_60
+; RV32-NEXT:    j .LBB185_385
+; RV32-NEXT:  .LBB185_60:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt t2, a1, .LBB185_62
+; RV32-NEXT:  .LBB185_61:
+; RV32-NEXT:    mv a4, s1
+; RV32-NEXT:  .LBB185_62:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 36(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a6, 32(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s1, a0, .LBB185_63
+; RV32-NEXT:    j .LBB185_386
+; RV32-NEXT:  .LBB185_63:
+; RV32-NEXT:    beq a1, t2, .LBB185_64
+; RV32-NEXT:    j .LBB185_387
+; RV32-NEXT:  .LBB185_64:
+; RV32-NEXT:    blt t2, a1, .LBB185_66
+; RV32-NEXT:  .LBB185_65:
+; RV32-NEXT:    mv a1, t2
+; RV32-NEXT:  .LBB185_66:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t6, 108(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s1, 12(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt s1, a1, .LBB185_67
+; RV32-NEXT:    j .LBB185_388
+; RV32-NEXT:  .LBB185_67:
+; RV32-NEXT:    bltu a5, a0, .LBB185_68
+; RV32-NEXT:    j .LBB185_389
+; RV32-NEXT:  .LBB185_68:
+; RV32-NEXT:    beq a1, s1, .LBB185_69
+; RV32-NEXT:    j .LBB185_390
+; RV32-NEXT:  .LBB185_69:
+; RV32-NEXT:    blt s1, a1, .LBB185_70
+; RV32-NEXT:    j .LBB185_391
+; RV32-NEXT:  .LBB185_70:
+; RV32-NEXT:    mv a5, a0
+; RV32-NEXT:    blt s0, a1, .LBB185_72
+; RV32-NEXT:  .LBB185_71:
+; RV32-NEXT:    mv a5, a6
+; RV32-NEXT:  .LBB185_72:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t3, 120(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 44(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 40(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a6, a0, .LBB185_73
+; RV32-NEXT:    j .LBB185_392
+; RV32-NEXT:  .LBB185_73:
+; RV32-NEXT:    beq a1, s0, .LBB185_74
+; RV32-NEXT:    j .LBB185_393
+; RV32-NEXT:  .LBB185_74:
+; RV32-NEXT:    blt s0, a1, .LBB185_75
+; RV32-NEXT:    j .LBB185_394
+; RV32-NEXT:  .LBB185_75:
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    blt s1, a1, .LBB185_77
+; RV32-NEXT:  .LBB185_76:
+; RV32-NEXT:    mv s0, a2
+; RV32-NEXT:  .LBB185_77:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw ra, 172(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s2, 76(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s7, 72(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 68(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a6, 64(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 60(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 56(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a0, .LBB185_395
+; RV32-NEXT:  # %bb.78:
+; RV32-NEXT:    bne a1, s1, .LBB185_396
+; RV32-NEXT:  .LBB185_79:
+; RV32-NEXT:    bge s1, a1, .LBB185_397
+; RV32-NEXT:  .LBB185_80:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt t5, a1, .LBB185_82
+; RV32-NEXT:  .LBB185_81:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a2, a2, t2
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 328(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_82:
+; RV32-NEXT:    csrr t2, vlenb
+; RV32-NEXT:    addi s0, zero, 232
+; RV32-NEXT:    mul t2, t2, s0
+; RV32-NEXT:    add t2, sp, t2
+; RV32-NEXT:    lw t2, 208(t2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr s0, vlenb
+; RV32-NEXT:    addi s1, zero, 232
+; RV32-NEXT:    mul s0, s0, s1
+; RV32-NEXT:    add s0, sp, s0
+; RV32-NEXT:    lw s4, 140(s0) # 4-byte Folded Reload
+; RV32-NEXT:    csrr s0, vlenb
+; RV32-NEXT:    addi s1, zero, 232
+; RV32-NEXT:    mul s0, s0, s1
+; RV32-NEXT:    add s0, sp, s0
+; RV32-NEXT:    lw s9, 328(s0) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s9, a0, .LBB185_398
+; RV32-NEXT:  # %bb.83:
+; RV32-NEXT:    bne a1, t5, .LBB185_399
+; RV32-NEXT:  .LBB185_84:
+; RV32-NEXT:    blt t5, a1, .LBB185_86
+; RV32-NEXT:  .LBB185_85:
+; RV32-NEXT:    mv a1, t5
+; RV32-NEXT:  .LBB185_86:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr t5, vlenb
+; RV32-NEXT:    addi s0, zero, 232
+; RV32-NEXT:    mul t5, t5, s0
+; RV32-NEXT:    add t5, sp, t5
+; RV32-NEXT:    lw s1, 392(t5) # 4-byte Folded Reload
+; RV32-NEXT:    blt s1, a1, .LBB185_88
+; RV32-NEXT:  # %bb.87:
+; RV32-NEXT:    mv a2, t4
+; RV32-NEXT:  .LBB185_88:
+; RV32-NEXT:    bltu t4, a0, .LBB185_90
+; RV32-NEXT:  # %bb.89:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:  .LBB185_90:
+; RV32-NEXT:    csrr t4, vlenb
+; RV32-NEXT:    addi t5, zero, 232
+; RV32-NEXT:    mul t4, t4, t5
+; RV32-NEXT:    add t4, sp, t4
+; RV32-NEXT:    lw t4, 392(t4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB185_92
+; RV32-NEXT:  # %bb.91:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_92:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t4, zero, 232
+; RV32-NEXT:    mul a2, a2, t4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 392(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a2, a1, .LBB185_400
+; RV32-NEXT:  # %bb.93:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a5, a1, .LBB185_401
+; RV32-NEXT:  .LBB185_94:
+; RV32-NEXT:    bgeu a3, a0, .LBB185_402
+; RV32-NEXT:  .LBB185_95:
+; RV32-NEXT:    beq a1, a5, .LBB185_97
+; RV32-NEXT:  .LBB185_96:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_97:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 52(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a5, a1, .LBB185_403
+; RV32-NEXT:  # %bb.98:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a4, a1, .LBB185_404
+; RV32-NEXT:  .LBB185_99:
+; RV32-NEXT:    bgeu a6, a0, .LBB185_405
+; RV32-NEXT:  .LBB185_100:
+; RV32-NEXT:    beq a1, a4, .LBB185_102
+; RV32-NEXT:  .LBB185_101:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_102:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 136(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a4, a1, .LBB185_406
+; RV32-NEXT:  # %bb.103:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s2, a1, .LBB185_407
+; RV32-NEXT:  .LBB185_104:
+; RV32-NEXT:    bgeu s7, a0, .LBB185_408
+; RV32-NEXT:  .LBB185_105:
+; RV32-NEXT:    bne a1, s2, .LBB185_409
+; RV32-NEXT:  .LBB185_106:
+; RV32-NEXT:    bge s2, a1, .LBB185_410
+; RV32-NEXT:  .LBB185_107:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s6, a1, .LBB185_109
+; RV32-NEXT:  .LBB185_108:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 332(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_109:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s7, 332(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s7, a0, .LBB185_411
+; RV32-NEXT:  # %bb.110:
+; RV32-NEXT:    bne a1, s6, .LBB185_412
+; RV32-NEXT:  .LBB185_111:
+; RV32-NEXT:    blt s6, a1, .LBB185_113
+; RV32-NEXT:  .LBB185_112:
+; RV32-NEXT:    mv a1, s6
+; RV32-NEXT:  .LBB185_113:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a5, 396(a4) # 4-byte Folded Reload
+; RV32-NEXT:    blt a5, a1, .LBB185_115
+; RV32-NEXT:  # %bb.114:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB185_115:
+; RV32-NEXT:    bltu a3, a0, .LBB185_117
+; RV32-NEXT:  # %bb.116:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB185_117:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t4, 396(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 88(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB185_119
+; RV32-NEXT:  # %bb.118:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_119:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 396(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 112(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a2, a1, .LBB185_413
+; RV32-NEXT:  # %bb.120:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s3, a1, .LBB185_414
+; RV32-NEXT:  .LBB185_121:
+; RV32-NEXT:    bgeu s11, a0, .LBB185_415
+; RV32-NEXT:  .LBB185_122:
+; RV32-NEXT:    bne a1, s3, .LBB185_416
+; RV32-NEXT:  .LBB185_123:
+; RV32-NEXT:    bge s3, a1, .LBB185_417
+; RV32-NEXT:  .LBB185_124:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s10, a1, .LBB185_126
+; RV32-NEXT:  .LBB185_125:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 336(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_126:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s11, 336(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s11, a0, .LBB185_418
+; RV32-NEXT:  # %bb.127:
+; RV32-NEXT:    bne a1, s10, .LBB185_419
+; RV32-NEXT:  .LBB185_128:
+; RV32-NEXT:    bge s10, a1, .LBB185_420
+; RV32-NEXT:  .LBB185_129:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt t1, a1, .LBB185_131
+; RV32-NEXT:  .LBB185_130:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 340(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_131:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s10, 340(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s10, a0, .LBB185_421
+; RV32-NEXT:  # %bb.132:
+; RV32-NEXT:    bne a1, t1, .LBB185_422
+; RV32-NEXT:  .LBB185_133:
+; RV32-NEXT:    bge t1, a1, .LBB185_423
+; RV32-NEXT:  .LBB185_134:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s5, a1, .LBB185_136
+; RV32-NEXT:  .LBB185_135:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 344(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_136:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t1, 344(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu t1, a0, .LBB185_424
+; RV32-NEXT:  # %bb.137:
+; RV32-NEXT:    bne a1, s5, .LBB185_425
+; RV32-NEXT:  .LBB185_138:
+; RV32-NEXT:    mv t1, t2
+; RV32-NEXT:    bge s5, a1, .LBB185_426
+; RV32-NEXT:  .LBB185_139:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt t6, a1, .LBB185_141
+; RV32-NEXT:  .LBB185_140:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 348(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_141:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s5, 348(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s5, a0, .LBB185_427
+; RV32-NEXT:  # %bb.142:
+; RV32-NEXT:    bne a1, t6, .LBB185_428
+; RV32-NEXT:  .LBB185_143:
+; RV32-NEXT:    blt t6, a1, .LBB185_145
+; RV32-NEXT:  .LBB185_144:
+; RV32-NEXT:    mv a1, t6
+; RV32-NEXT:  .LBB185_145:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 400(a4) # 4-byte Folded Reload
+; RV32-NEXT:    blt a4, a1, .LBB185_147
+; RV32-NEXT:  # %bb.146:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:  .LBB185_147:
+; RV32-NEXT:    bltu a5, a0, .LBB185_149
+; RV32-NEXT:  # %bb.148:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:  .LBB185_149:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t4, 400(a4) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 124(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB185_151
+; RV32-NEXT:  # %bb.150:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_151:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 400(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 128(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bge a2, a1, .LBB185_429
+; RV32-NEXT:  # %bb.152:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s8, a1, .LBB185_430
+; RV32-NEXT:  .LBB185_153:
+; RV32-NEXT:    bgeu a3, a0, .LBB185_431
+; RV32-NEXT:  .LBB185_154:
+; RV32-NEXT:    bne a1, s8, .LBB185_432
+; RV32-NEXT:  .LBB185_155:
+; RV32-NEXT:    bge s8, a1, .LBB185_433
+; RV32-NEXT:  .LBB185_156:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt t3, a1, .LBB185_158
+; RV32-NEXT:  .LBB185_157:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 352(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_158:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s9, 352(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s9, a0, .LBB185_434
+; RV32-NEXT:  # %bb.159:
+; RV32-NEXT:    bne a1, t3, .LBB185_435
+; RV32-NEXT:  .LBB185_160:
+; RV32-NEXT:    blt t3, a1, .LBB185_162
+; RV32-NEXT:  .LBB185_161:
+; RV32-NEXT:    mv a1, t3
+; RV32-NEXT:  .LBB185_162:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 404(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a3, a1, .LBB185_164
+; RV32-NEXT:  # %bb.163:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB185_164:
+; RV32-NEXT:    bltu a4, a0, .LBB185_166
+; RV32-NEXT:  # %bb.165:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB185_166:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s10, 404(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 164(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s10, .LBB185_168
+; RV32-NEXT:  # %bb.167:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_168:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 404(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 144(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bge a2, a1, .LBB185_436
+; RV32-NEXT:  # %bb.169:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a7, a1, .LBB185_437
+; RV32-NEXT:  .LBB185_170:
+; RV32-NEXT:    bgeu a5, a0, .LBB185_438
+; RV32-NEXT:  .LBB185_171:
+; RV32-NEXT:    beq a1, a7, .LBB185_173
+; RV32-NEXT:  .LBB185_172:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_173:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 156(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a7, a1, .LBB185_175
+; RV32-NEXT:  # %bb.174:
+; RV32-NEXT:    mv a1, a7
+; RV32-NEXT:  .LBB185_175:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s1, a1, .LBB185_177
+; RV32-NEXT:  # %bb.176:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 356(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_177:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s8, 356(a6) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s2, 196(a6) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw a7, 220(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s8, a0, .LBB185_439
+; RV32-NEXT:  # %bb.178:
+; RV32-NEXT:    bne a1, s1, .LBB185_440
+; RV32-NEXT:  .LBB185_179:
+; RV32-NEXT:    bge s1, a1, .LBB185_441
+; RV32-NEXT:  .LBB185_180:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s4, a1, .LBB185_182
+; RV32-NEXT:  .LBB185_181:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 360(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_182:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw t5, 360(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu t5, a0, .LBB185_442
+; RV32-NEXT:  # %bb.183:
+; RV32-NEXT:    bne a1, s4, .LBB185_443
+; RV32-NEXT:  .LBB185_184:
+; RV32-NEXT:    blt s4, a1, .LBB185_186
+; RV32-NEXT:  .LBB185_185:
+; RV32-NEXT:    mv a1, s4
+; RV32-NEXT:  .LBB185_186:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s0, 408(a6) # 4-byte Folded Reload
+; RV32-NEXT:    blt s0, a1, .LBB185_188
+; RV32-NEXT:  # %bb.187:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB185_188:
+; RV32-NEXT:    bltu a4, a0, .LBB185_190
+; RV32-NEXT:  # %bb.189:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB185_190:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s11, 408(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s11, .LBB185_192
+; RV32-NEXT:  # %bb.191:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_192:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 408(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 148(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bge a2, a1, .LBB185_444
+; RV32-NEXT:  # %bb.193:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge t0, a1, .LBB185_445
+; RV32-NEXT:  .LBB185_194:
+; RV32-NEXT:    bgeu a4, a0, .LBB185_446
+; RV32-NEXT:  .LBB185_195:
+; RV32-NEXT:    beq a1, t0, .LBB185_197
+; RV32-NEXT:  .LBB185_196:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_197:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 168(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt t0, a1, .LBB185_199
+; RV32-NEXT:  # %bb.198:
+; RV32-NEXT:    mv a1, t0
+; RV32-NEXT:  .LBB185_199:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a5, a1, .LBB185_201
+; RV32-NEXT:  # %bb.200:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 364(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_201:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a6, a6, t0
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw t0, 364(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bltu t0, a0, .LBB185_203
+; RV32-NEXT:  # %bb.202:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:  .LBB185_203:
+; RV32-NEXT:    beq a1, a5, .LBB185_205
+; RV32-NEXT:  # %bb.204:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_205:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t0, 200(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 184(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 204(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a5, a1, .LBB185_207
+; RV32-NEXT:  # %bb.206:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB185_207:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 160(a5) # 4-byte Folded Reload
+; RV32-NEXT:    blt a5, a1, .LBB185_209
+; RV32-NEXT:  # %bb.208:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 368(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_209:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s4, 368(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s4, a0, .LBB185_447
+; RV32-NEXT:  # %bb.210:
+; RV32-NEXT:    bne a1, a5, .LBB185_448
+; RV32-NEXT:  .LBB185_211:
+; RV32-NEXT:    blt a5, a1, .LBB185_213
+; RV32-NEXT:  .LBB185_212:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB185_213:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 412(a5) # 4-byte Folded Reload
+; RV32-NEXT:    blt a5, a1, .LBB185_215
+; RV32-NEXT:  # %bb.214:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB185_215:
+; RV32-NEXT:    bltu a3, a0, .LBB185_217
+; RV32-NEXT:  # %bb.216:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB185_217:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t5, 412(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 216(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t5, .LBB185_219
+; RV32-NEXT:  # %bb.218:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_219:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 412(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 180(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a2, a1, .LBB185_449
+; RV32-NEXT:  # %bb.220:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge ra, a1, .LBB185_450
+; RV32-NEXT:  .LBB185_221:
+; RV32-NEXT:    bgeu a4, a0, .LBB185_451
+; RV32-NEXT:  .LBB185_222:
+; RV32-NEXT:    beq a1, ra, .LBB185_224
+; RV32-NEXT:  .LBB185_223:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_224:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 176(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt ra, a1, .LBB185_226
+; RV32-NEXT:  # %bb.225:
+; RV32-NEXT:    mv a1, ra
+; RV32-NEXT:  .LBB185_226:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a4, a1, .LBB185_228
+; RV32-NEXT:  # %bb.227:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 372(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_228:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s7, 372(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s7, a0, .LBB185_452
+; RV32-NEXT:  # %bb.229:
+; RV32-NEXT:    bne a1, a4, .LBB185_453
+; RV32-NEXT:  .LBB185_230:
+; RV32-NEXT:    blt a4, a1, .LBB185_232
+; RV32-NEXT:  .LBB185_231:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB185_232:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 416(a4) # 4-byte Folded Reload
+; RV32-NEXT:    blt a4, a1, .LBB185_234
+; RV32-NEXT:  # %bb.233:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB185_234:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a6, 244(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a3, a0, .LBB185_236
+; RV32-NEXT:  # %bb.235:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB185_236:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s9, 416(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t5, 264(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 212(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s9, .LBB185_238
+; RV32-NEXT:  # %bb.237:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_238:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 416(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 188(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a2, a1, .LBB185_454
+; RV32-NEXT:  # %bb.239:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a3, a1, .LBB185_455
+; RV32-NEXT:  .LBB185_240:
+; RV32-NEXT:    bgeu s0, a0, .LBB185_456
+; RV32-NEXT:  .LBB185_241:
+; RV32-NEXT:    bne a1, a3, .LBB185_457
+; RV32-NEXT:  .LBB185_242:
+; RV32-NEXT:    blt a3, a1, .LBB185_244
+; RV32-NEXT:  .LBB185_243:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB185_244:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 192(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a3, a1, .LBB185_246
+; RV32-NEXT:  # %bb.245:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a2, a2, t2
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 376(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_246:
+; RV32-NEXT:    csrr t2, vlenb
+; RV32-NEXT:    addi t3, zero, 232
+; RV32-NEXT:    mul t2, t2, t3
+; RV32-NEXT:    add t2, sp, t2
+; RV32-NEXT:    lw s11, 376(t2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s11, a0, .LBB185_458
+; RV32-NEXT:  # %bb.247:
+; RV32-NEXT:    bne a1, a3, .LBB185_459
+; RV32-NEXT:  .LBB185_248:
+; RV32-NEXT:    blt a3, a1, .LBB185_250
+; RV32-NEXT:  .LBB185_249:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB185_250:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 420(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a3, a1, .LBB185_252
+; RV32-NEXT:  # %bb.251:
+; RV32-NEXT:    mv a2, s2
+; RV32-NEXT:  .LBB185_252:
+; RV32-NEXT:    bltu s2, a0, .LBB185_254
+; RV32-NEXT:  # %bb.253:
+; RV32-NEXT:    mv a0, s2
+; RV32-NEXT:  .LBB185_254:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t2, 420(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t3, zero, 232
+; RV32-NEXT:    mul a3, a3, t3
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t4, 232(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t2, .LBB185_256
+; RV32-NEXT:  # %bb.255:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_256:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 420(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a2, a1, .LBB185_258
+; RV32-NEXT:  # %bb.257:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:  .LBB185_258:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 228(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge s1, a1, .LBB185_460
+; RV32-NEXT:  # %bb.259:
+; RV32-NEXT:    bgeu t0, a0, .LBB185_461
+; RV32-NEXT:  .LBB185_260:
+; RV32-NEXT:    bne a1, s1, .LBB185_462
+; RV32-NEXT:  .LBB185_261:
+; RV32-NEXT:    bge s1, a1, .LBB185_463
+; RV32-NEXT:  .LBB185_262:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt t1, a1, .LBB185_264
+; RV32-NEXT:  .LBB185_263:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a2, a2, t0
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 380(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_264:
+; RV32-NEXT:    csrr t0, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul t0, t0, t2
+; RV32-NEXT:    add t0, sp, t0
+; RV32-NEXT:    lw t0, 380(t0) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu t0, a0, .LBB185_464
+; RV32-NEXT:  # %bb.265:
+; RV32-NEXT:    bne a1, t1, .LBB185_465
+; RV32-NEXT:  .LBB185_266:
+; RV32-NEXT:    blt t1, a1, .LBB185_268
+; RV32-NEXT:  .LBB185_267:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:  .LBB185_268:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr t0, vlenb
+; RV32-NEXT:    addi t1, zero, 232
+; RV32-NEXT:    mul t0, t0, t1
+; RV32-NEXT:    add t0, sp, t0
+; RV32-NEXT:    lw s0, 424(t0) # 4-byte Folded Reload
+; RV32-NEXT:    blt s0, a1, .LBB185_270
+; RV32-NEXT:  # %bb.269:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB185_270:
+; RV32-NEXT:    bltu a4, a0, .LBB185_272
+; RV32-NEXT:  # %bb.271:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB185_272:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a4, a4, t0
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t3, 424(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t3, .LBB185_274
+; RV32-NEXT:  # %bb.273:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_274:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 424(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a4, a4, t0
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 224(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bge a2, a1, .LBB185_466
+; RV32-NEXT:  # %bb.275:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a7, a1, .LBB185_467
+; RV32-NEXT:  .LBB185_276:
+; RV32-NEXT:    bgeu a5, a0, .LBB185_468
+; RV32-NEXT:  .LBB185_277:
+; RV32-NEXT:    bne a1, a7, .LBB185_469
+; RV32-NEXT:  .LBB185_278:
+; RV32-NEXT:    bge a7, a1, .LBB185_470
+; RV32-NEXT:  .LBB185_279:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a4, a1, .LBB185_281
+; RV32-NEXT:  .LBB185_280:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 384(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_281:
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a5, a5, a7
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw t1, 384(a5) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a5, a5, a7
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw t0, 312(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu t1, a0, .LBB185_283
+; RV32-NEXT:  # %bb.282:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:  .LBB185_283:
+; RV32-NEXT:    beq a1, a4, .LBB185_285
+; RV32-NEXT:  # %bb.284:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_285:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 236(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a4, a1, .LBB185_287
+; RV32-NEXT:  # %bb.286:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB185_287:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a3, a1, .LBB185_289
+; RV32-NEXT:  # %bb.288:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 388(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB185_289:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a7, 388(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a7, a0, .LBB185_291
+; RV32-NEXT:  # %bb.290:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:  .LBB185_291:
+; RV32-NEXT:    beq a1, a3, .LBB185_293
+; RV32-NEXT:  # %bb.292:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_293:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t1, 256(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 260(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a7, 272(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 252(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 248(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a3, a1, .LBB185_471
+; RV32-NEXT:  # %bb.294:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s1, a1, .LBB185_472
+; RV32-NEXT:  .LBB185_295:
+; RV32-NEXT:    bgeu t4, a0, .LBB185_473
+; RV32-NEXT:  .LBB185_296:
+; RV32-NEXT:    beq a1, s1, .LBB185_298
+; RV32-NEXT:  .LBB185_297:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_298:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 240(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge s1, a1, .LBB185_474
+; RV32-NEXT:  # %bb.299:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a6, a1, .LBB185_475
+; RV32-NEXT:  .LBB185_300:
+; RV32-NEXT:    bgeu a3, a0, .LBB185_476
+; RV32-NEXT:  .LBB185_301:
+; RV32-NEXT:    bne a1, a6, .LBB185_477
+; RV32-NEXT:  .LBB185_302:
+; RV32-NEXT:    bge a6, a1, .LBB185_478
+; RV32-NEXT:  .LBB185_303:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a4, a1, .LBB185_305
+; RV32-NEXT:  .LBB185_304:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:  .LBB185_305:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a6, 304(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 284(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a5, a0, .LBB185_307
+; RV32-NEXT:  # %bb.306:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:  .LBB185_307:
+; RV32-NEXT:    beq a1, a4, .LBB185_309
+; RV32-NEXT:  # %bb.308:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_309:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 268(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a4, a1, .LBB185_479
+; RV32-NEXT:  # %bb.310:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s0, a1, .LBB185_480
+; RV32-NEXT:  .LBB185_311:
+; RV32-NEXT:    bgeu t1, a0, .LBB185_481
+; RV32-NEXT:  .LBB185_312:
+; RV32-NEXT:    bne a1, s0, .LBB185_482
+; RV32-NEXT:  .LBB185_313:
+; RV32-NEXT:    bge s0, a1, .LBB185_483
+; RV32-NEXT:  .LBB185_314:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a5, a1, .LBB185_484
+; RV32-NEXT:  .LBB185_315:
+; RV32-NEXT:    bgeu t5, a0, .LBB185_485
+; RV32-NEXT:  .LBB185_316:
+; RV32-NEXT:    beq a1, a5, .LBB185_318
+; RV32-NEXT:  .LBB185_317:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_318:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 276(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a5, a1, .LBB185_486
+; RV32-NEXT:  # %bb.319:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a4, a1, .LBB185_487
+; RV32-NEXT:  .LBB185_320:
+; RV32-NEXT:    bgeu a7, a0, .LBB185_488
+; RV32-NEXT:  .LBB185_321:
+; RV32-NEXT:    beq a1, a4, .LBB185_323
+; RV32-NEXT:  .LBB185_322:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_323:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 296(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 292(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a4, a1, .LBB185_325
+; RV32-NEXT:  # %bb.324:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB185_325:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a4, a4, a7
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 280(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bge a3, a1, .LBB185_489
+; RV32-NEXT:  # %bb.326:
+; RV32-NEXT:    bgeu a4, a0, .LBB185_490
+; RV32-NEXT:  .LBB185_327:
+; RV32-NEXT:    bne a1, a3, .LBB185_491
+; RV32-NEXT:  .LBB185_328:
+; RV32-NEXT:    blt a3, a1, .LBB185_330
+; RV32-NEXT:  .LBB185_329:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB185_330:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 288(a3) # 4-byte Folded Reload
+; RV32-NEXT:    blt a5, a1, .LBB185_332
+; RV32-NEXT:  # %bb.331:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB185_332:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a3, a3, a7
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s0, 316(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a4, a0, .LBB185_334
+; RV32-NEXT:  # %bb.333:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB185_334:
+; RV32-NEXT:    beq a1, a5, .LBB185_336
+; RV32-NEXT:  # %bb.335:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_336:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 320(a2) # 4-byte Folded Reload
+; RV32-NEXT:    blt a5, a1, .LBB185_338
+; RV32-NEXT:  # %bb.337:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB185_338:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 300(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a5, a1, .LBB185_492
+; RV32-NEXT:  # %bb.339:
+; RV32-NEXT:    bgeu s1, a0, .LBB185_493
+; RV32-NEXT:  .LBB185_340:
+; RV32-NEXT:    bne a1, a5, .LBB185_494
+; RV32-NEXT:  .LBB185_341:
+; RV32-NEXT:    blt a5, a1, .LBB185_343
+; RV32-NEXT:  .LBB185_342:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB185_343:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 308(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bge a5, a1, .LBB185_495
+; RV32-NEXT:  # %bb.344:
+; RV32-NEXT:    bgeu a6, a0, .LBB185_496
+; RV32-NEXT:  .LBB185_345:
+; RV32-NEXT:    beq a1, a5, .LBB185_347
+; RV32-NEXT:  .LBB185_346:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB185_347:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 324(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bge a5, a1, .LBB185_497
+; RV32-NEXT:  # %bb.348:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s0, a1, .LBB185_498
+; RV32-NEXT:  .LBB185_349:
+; RV32-NEXT:    bgeu t0, a0, .LBB185_499
+; RV32-NEXT:  .LBB185_350:
+; RV32-NEXT:    bne a1, s0, .LBB185_500
+; RV32-NEXT:  .LBB185_351:
+; RV32-NEXT:    bge s0, a1, .LBB185_501
+; RV32-NEXT:  .LBB185_352:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a3, a1, .LBB185_502
+; RV32-NEXT:  .LBB185_353:
+; RV32-NEXT:    bgeu a4, a0, .LBB185_503
+; RV32-NEXT:  .LBB185_354:
+; RV32-NEXT:    bne a1, a3, .LBB185_504
+; RV32-NEXT:  .LBB185_355:
+; RV32-NEXT:    blt a3, a1, .LBB185_357
+; RV32-NEXT:  .LBB185_356:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB185_357:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add sp, sp, a2
+; RV32-NEXT:    lw s11, 428(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s10, 432(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s9, 436(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s8, 440(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s7, 444(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s6, 448(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s5, 452(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s4, 456(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s3, 460(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s2, 464(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s1, 468(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s0, 472(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw ra, 476(sp) # 4-byte Folded Reload
+; RV32-NEXT:    addi sp, sp, 480
+; RV32-NEXT:    ret
+; RV32-NEXT:  .LBB185_358:
+; RV32-NEXT:    mv s4, t6
+; RV32-NEXT:    bgeu t6, a0, .LBB185_359
+; RV32-NEXT:    j .LBB185_2
+; RV32-NEXT:  .LBB185_359:
+; RV32-NEXT:    mv a0, t6
+; RV32-NEXT:    bne a1, a3, .LBB185_360
+; RV32-NEXT:    j .LBB185_3
+; RV32-NEXT:  .LBB185_360:
+; RV32-NEXT:    mv a0, s4
+; RV32-NEXT:    bge a3, a1, .LBB185_361
+; RV32-NEXT:    j .LBB185_4
+; RV32-NEXT:  .LBB185_361:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:    mv a3, a0
+; RV32-NEXT:    bge a4, a1, .LBB185_362
+; RV32-NEXT:    j .LBB185_5
+; RV32-NEXT:  .LBB185_362:
+; RV32-NEXT:    mv a3, t4
+; RV32-NEXT:    bgeu t4, a0, .LBB185_363
+; RV32-NEXT:    j .LBB185_6
+; RV32-NEXT:  .LBB185_363:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:    beq a1, a4, .LBB185_505
+; RV32-NEXT:    j .LBB185_7
+; RV32-NEXT:  .LBB185_505:
+; RV32-NEXT:    j .LBB185_8
+; RV32-NEXT:  .LBB185_364:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bge s10, a1, .LBB185_365
+; RV32-NEXT:    j .LBB185_10
+; RV32-NEXT:  .LBB185_365:
+; RV32-NEXT:    mv a4, a7
+; RV32-NEXT:    bgeu a7, a0, .LBB185_366
+; RV32-NEXT:    j .LBB185_11
+; RV32-NEXT:  .LBB185_366:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:    bne a1, s10, .LBB185_367
+; RV32-NEXT:    j .LBB185_12
+; RV32-NEXT:  .LBB185_367:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bge s10, a1, .LBB185_368
+; RV32-NEXT:    j .LBB185_13
+; RV32-NEXT:  .LBB185_368:
+; RV32-NEXT:    mv a1, s10
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt s3, a1, .LBB185_506
+; RV32-NEXT:    j .LBB185_14
+; RV32-NEXT:  .LBB185_506:
+; RV32-NEXT:    j .LBB185_15
+; RV32-NEXT:  .LBB185_369:
+; RV32-NEXT:    mv a0, s6
+; RV32-NEXT:    bne a1, s3, .LBB185_370
+; RV32-NEXT:    j .LBB185_17
+; RV32-NEXT:  .LBB185_370:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bge s3, a1, .LBB185_371
+; RV32-NEXT:    j .LBB185_18
+; RV32-NEXT:  .LBB185_371:
+; RV32-NEXT:    mv a1, s3
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bge a5, a1, .LBB185_372
+; RV32-NEXT:    j .LBB185_19
+; RV32-NEXT:  .LBB185_372:
+; RV32-NEXT:    mv a4, t5
+; RV32-NEXT:    bgeu t5, a0, .LBB185_373
+; RV32-NEXT:    j .LBB185_20
+; RV32-NEXT:  .LBB185_373:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    bne a1, a5, .LBB185_374
+; RV32-NEXT:    j .LBB185_21
+; RV32-NEXT:  .LBB185_374:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bge a5, a1, .LBB185_375
+; RV32-NEXT:    j .LBB185_22
+; RV32-NEXT:  .LBB185_375:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bge t1, a1, .LBB185_376
+; RV32-NEXT:    j .LBB185_23
+; RV32-NEXT:  .LBB185_376:
+; RV32-NEXT:    mv a4, s5
+; RV32-NEXT:    bgeu s5, a0, .LBB185_377
+; RV32-NEXT:    j .LBB185_24
+; RV32-NEXT:  .LBB185_377:
+; RV32-NEXT:    mv a0, s5
+; RV32-NEXT:    beq a1, t1, .LBB185_507
+; RV32-NEXT:    j .LBB185_25
+; RV32-NEXT:  .LBB185_507:
+; RV32-NEXT:    j .LBB185_26
+; RV32-NEXT:  .LBB185_378:
+; RV32-NEXT:    mv a0, ra
+; RV32-NEXT:    bne a1, s1, .LBB185_379
+; RV32-NEXT:    j .LBB185_32
+; RV32-NEXT:  .LBB185_379:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bge s1, a1, .LBB185_380
+; RV32-NEXT:    j .LBB185_33
+; RV32-NEXT:  .LBB185_380:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt s11, a1, .LBB185_508
+; RV32-NEXT:    j .LBB185_34
+; RV32-NEXT:  .LBB185_508:
+; RV32-NEXT:    j .LBB185_35
+; RV32-NEXT:  .LBB185_381:
+; RV32-NEXT:    mv a0, s2
+; RV32-NEXT:    bne a1, t3, .LBB185_382
+; RV32-NEXT:    j .LBB185_45
+; RV32-NEXT:  .LBB185_382:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    blt t3, a1, .LBB185_509
+; RV32-NEXT:    j .LBB185_46
+; RV32-NEXT:  .LBB185_509:
+; RV32-NEXT:    j .LBB185_47
+; RV32-NEXT:  .LBB185_383:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, a6, .LBB185_384
+; RV32-NEXT:    j .LBB185_59
+; RV32-NEXT:  .LBB185_384:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    bge a6, a1, .LBB185_385
+; RV32-NEXT:    j .LBB185_60
+; RV32-NEXT:  .LBB185_385:
+; RV32-NEXT:    mv a1, a6
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    blt t2, a1, .LBB185_510
+; RV32-NEXT:    j .LBB185_61
+; RV32-NEXT:  .LBB185_510:
+; RV32-NEXT:    j .LBB185_62
+; RV32-NEXT:  .LBB185_386:
+; RV32-NEXT:    mv a0, s1
+; RV32-NEXT:    bne a1, t2, .LBB185_387
+; RV32-NEXT:    j .LBB185_64
+; RV32-NEXT:  .LBB185_387:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    blt t2, a1, .LBB185_511
+; RV32-NEXT:    j .LBB185_65
+; RV32-NEXT:  .LBB185_511:
+; RV32-NEXT:    j .LBB185_66
+; RV32-NEXT:  .LBB185_388:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bgeu a5, a0, .LBB185_389
+; RV32-NEXT:    j .LBB185_68
+; RV32-NEXT:  .LBB185_389:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bne a1, s1, .LBB185_390
+; RV32-NEXT:    j .LBB185_69
+; RV32-NEXT:  .LBB185_390:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge s1, a1, .LBB185_391
+; RV32-NEXT:    j .LBB185_70
+; RV32-NEXT:  .LBB185_391:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a5, a0
+; RV32-NEXT:    blt s0, a1, .LBB185_512
+; RV32-NEXT:    j .LBB185_71
+; RV32-NEXT:  .LBB185_512:
+; RV32-NEXT:    j .LBB185_72
+; RV32-NEXT:  .LBB185_392:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, s0, .LBB185_393
+; RV32-NEXT:    j .LBB185_74
+; RV32-NEXT:  .LBB185_393:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bge s0, a1, .LBB185_394
+; RV32-NEXT:    j .LBB185_75
+; RV32-NEXT:  .LBB185_394:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    blt s1, a1, .LBB185_513
+; RV32-NEXT:    j .LBB185_76
+; RV32-NEXT:  .LBB185_513:
+; RV32-NEXT:    j .LBB185_77
+; RV32-NEXT:  .LBB185_395:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    beq a1, s1, .LBB185_79
+; RV32-NEXT:  .LBB185_396:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    blt s1, a1, .LBB185_80
+; RV32-NEXT:  .LBB185_397:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge t5, a1, .LBB185_81
+; RV32-NEXT:    j .LBB185_82
+; RV32-NEXT:  .LBB185_398:
+; RV32-NEXT:    mv a0, s9
+; RV32-NEXT:    beq a1, t5, .LBB185_84
+; RV32-NEXT:  .LBB185_399:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge t5, a1, .LBB185_85
+; RV32-NEXT:    j .LBB185_86
+; RV32-NEXT:  .LBB185_400:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a5, a1, .LBB185_94
+; RV32-NEXT:  .LBB185_401:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a3, a0, .LBB185_95
+; RV32-NEXT:  .LBB185_402:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    bne a1, a5, .LBB185_96
+; RV32-NEXT:    j .LBB185_97
+; RV32-NEXT:  .LBB185_403:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a4, a1, .LBB185_99
+; RV32-NEXT:  .LBB185_404:
+; RV32-NEXT:    mv a2, a6
+; RV32-NEXT:    bltu a6, a0, .LBB185_100
+; RV32-NEXT:  .LBB185_405:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, a4, .LBB185_101
+; RV32-NEXT:    j .LBB185_102
+; RV32-NEXT:  .LBB185_406:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s2, a1, .LBB185_104
+; RV32-NEXT:  .LBB185_407:
+; RV32-NEXT:    mv a2, s7
+; RV32-NEXT:    bltu s7, a0, .LBB185_105
+; RV32-NEXT:  .LBB185_408:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, s2, .LBB185_106
+; RV32-NEXT:  .LBB185_409:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt s2, a1, .LBB185_107
+; RV32-NEXT:  .LBB185_410:
+; RV32-NEXT:    mv a1, s2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s6, a1, .LBB185_108
+; RV32-NEXT:    j .LBB185_109
+; RV32-NEXT:  .LBB185_411:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, s6, .LBB185_111
+; RV32-NEXT:  .LBB185_412:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge s6, a1, .LBB185_112
+; RV32-NEXT:    j .LBB185_113
+; RV32-NEXT:  .LBB185_413:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s3, a1, .LBB185_121
+; RV32-NEXT:  .LBB185_414:
+; RV32-NEXT:    mv a2, s11
+; RV32-NEXT:    bltu s11, a0, .LBB185_122
+; RV32-NEXT:  .LBB185_415:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, s3, .LBB185_123
+; RV32-NEXT:  .LBB185_416:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt s3, a1, .LBB185_124
+; RV32-NEXT:  .LBB185_417:
+; RV32-NEXT:    mv a1, s3
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s10, a1, .LBB185_125
+; RV32-NEXT:    j .LBB185_126
+; RV32-NEXT:  .LBB185_418:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, s10, .LBB185_128
+; RV32-NEXT:  .LBB185_419:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt s10, a1, .LBB185_129
+; RV32-NEXT:  .LBB185_420:
+; RV32-NEXT:    mv a1, s10
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge t1, a1, .LBB185_130
+; RV32-NEXT:    j .LBB185_131
+; RV32-NEXT:  .LBB185_421:
+; RV32-NEXT:    mv a0, s10
+; RV32-NEXT:    beq a1, t1, .LBB185_133
+; RV32-NEXT:  .LBB185_422:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt t1, a1, .LBB185_134
+; RV32-NEXT:  .LBB185_423:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s5, a1, .LBB185_135
+; RV32-NEXT:    j .LBB185_136
+; RV32-NEXT:  .LBB185_424:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:    beq a1, s5, .LBB185_138
+; RV32-NEXT:  .LBB185_425:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    mv t1, t2
+; RV32-NEXT:    blt s5, a1, .LBB185_139
+; RV32-NEXT:  .LBB185_426:
+; RV32-NEXT:    mv a1, s5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge t6, a1, .LBB185_140
+; RV32-NEXT:    j .LBB185_141
+; RV32-NEXT:  .LBB185_427:
+; RV32-NEXT:    mv a0, s5
+; RV32-NEXT:    beq a1, t6, .LBB185_143
+; RV32-NEXT:  .LBB185_428:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge t6, a1, .LBB185_144
+; RV32-NEXT:    j .LBB185_145
+; RV32-NEXT:  .LBB185_429:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s8, a1, .LBB185_153
+; RV32-NEXT:  .LBB185_430:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a3, a0, .LBB185_154
+; RV32-NEXT:  .LBB185_431:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    beq a1, s8, .LBB185_155
+; RV32-NEXT:  .LBB185_432:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt s8, a1, .LBB185_156
+; RV32-NEXT:  .LBB185_433:
+; RV32-NEXT:    mv a1, s8
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge t3, a1, .LBB185_157
+; RV32-NEXT:    j .LBB185_158
+; RV32-NEXT:  .LBB185_434:
+; RV32-NEXT:    mv a0, s9
+; RV32-NEXT:    beq a1, t3, .LBB185_160
+; RV32-NEXT:  .LBB185_435:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge t3, a1, .LBB185_161
+; RV32-NEXT:    j .LBB185_162
+; RV32-NEXT:  .LBB185_436:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a7, a1, .LBB185_170
+; RV32-NEXT:  .LBB185_437:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bltu a5, a0, .LBB185_171
+; RV32-NEXT:  .LBB185_438:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bne a1, a7, .LBB185_172
+; RV32-NEXT:    j .LBB185_173
+; RV32-NEXT:  .LBB185_439:
+; RV32-NEXT:    mv a0, s8
+; RV32-NEXT:    beq a1, s1, .LBB185_179
+; RV32-NEXT:  .LBB185_440:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt s1, a1, .LBB185_180
+; RV32-NEXT:  .LBB185_441:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge s4, a1, .LBB185_181
+; RV32-NEXT:    j .LBB185_182
+; RV32-NEXT:  .LBB185_442:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    beq a1, s4, .LBB185_184
+; RV32-NEXT:  .LBB185_443:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge s4, a1, .LBB185_185
+; RV32-NEXT:    j .LBB185_186
+; RV32-NEXT:  .LBB185_444:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt t0, a1, .LBB185_194
+; RV32-NEXT:  .LBB185_445:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a4, a0, .LBB185_195
+; RV32-NEXT:  .LBB185_446:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, t0, .LBB185_196
+; RV32-NEXT:    j .LBB185_197
+; RV32-NEXT:  .LBB185_447:
+; RV32-NEXT:    mv a0, s4
+; RV32-NEXT:    beq a1, a5, .LBB185_211
+; RV32-NEXT:  .LBB185_448:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a5, a1, .LBB185_212
+; RV32-NEXT:    j .LBB185_213
+; RV32-NEXT:  .LBB185_449:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt ra, a1, .LBB185_221
+; RV32-NEXT:  .LBB185_450:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a4, a0, .LBB185_222
+; RV32-NEXT:  .LBB185_451:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, ra, .LBB185_223
+; RV32-NEXT:    j .LBB185_224
+; RV32-NEXT:  .LBB185_452:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, a4, .LBB185_230
+; RV32-NEXT:  .LBB185_453:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a4, a1, .LBB185_231
+; RV32-NEXT:    j .LBB185_232
+; RV32-NEXT:  .LBB185_454:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a3, a1, .LBB185_240
+; RV32-NEXT:  .LBB185_455:
+; RV32-NEXT:    mv a2, s0
+; RV32-NEXT:    bltu s0, a0, .LBB185_241
+; RV32-NEXT:  .LBB185_456:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    beq a1, a3, .LBB185_242
+; RV32-NEXT:  .LBB185_457:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a3, a1, .LBB185_243
+; RV32-NEXT:    j .LBB185_244
+; RV32-NEXT:  .LBB185_458:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, a3, .LBB185_248
+; RV32-NEXT:  .LBB185_459:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a3, a1, .LBB185_249
+; RV32-NEXT:    j .LBB185_250
+; RV32-NEXT:  .LBB185_460:
+; RV32-NEXT:    mv a2, t0
+; RV32-NEXT:    bltu t0, a0, .LBB185_260
+; RV32-NEXT:  .LBB185_461:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, s1, .LBB185_261
+; RV32-NEXT:  .LBB185_462:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt s1, a1, .LBB185_262
+; RV32-NEXT:  .LBB185_463:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge t1, a1, .LBB185_263
+; RV32-NEXT:    j .LBB185_264
+; RV32-NEXT:  .LBB185_464:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, t1, .LBB185_266
+; RV32-NEXT:  .LBB185_465:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge t1, a1, .LBB185_267
+; RV32-NEXT:    j .LBB185_268
+; RV32-NEXT:  .LBB185_466:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a7, a1, .LBB185_276
+; RV32-NEXT:  .LBB185_467:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bltu a5, a0, .LBB185_277
+; RV32-NEXT:  .LBB185_468:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    beq a1, a7, .LBB185_278
+; RV32-NEXT:  .LBB185_469:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a7, a1, .LBB185_279
+; RV32-NEXT:  .LBB185_470:
+; RV32-NEXT:    mv a1, a7
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a4, a1, .LBB185_280
+; RV32-NEXT:    j .LBB185_281
+; RV32-NEXT:  .LBB185_471:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s1, a1, .LBB185_295
+; RV32-NEXT:  .LBB185_472:
+; RV32-NEXT:    mv a2, t4
+; RV32-NEXT:    bltu t4, a0, .LBB185_296
+; RV32-NEXT:  .LBB185_473:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:    bne a1, s1, .LBB185_297
+; RV32-NEXT:    j .LBB185_298
+; RV32-NEXT:  .LBB185_474:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a6, a1, .LBB185_300
+; RV32-NEXT:  .LBB185_475:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a3, a0, .LBB185_301
+; RV32-NEXT:  .LBB185_476:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    beq a1, a6, .LBB185_302
+; RV32-NEXT:  .LBB185_477:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt a6, a1, .LBB185_303
+; RV32-NEXT:  .LBB185_478:
+; RV32-NEXT:    mv a1, a6
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bge a4, a1, .LBB185_304
+; RV32-NEXT:    j .LBB185_305
+; RV32-NEXT:  .LBB185_479:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s0, a1, .LBB185_311
+; RV32-NEXT:  .LBB185_480:
+; RV32-NEXT:    mv a2, t1
+; RV32-NEXT:    bltu t1, a0, .LBB185_312
+; RV32-NEXT:  .LBB185_481:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:    beq a1, s0, .LBB185_313
+; RV32-NEXT:  .LBB185_482:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt s0, a1, .LBB185_314
+; RV32-NEXT:  .LBB185_483:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a5, a1, .LBB185_315
+; RV32-NEXT:  .LBB185_484:
+; RV32-NEXT:    mv a2, t5
+; RV32-NEXT:    bltu t5, a0, .LBB185_316
+; RV32-NEXT:  .LBB185_485:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    bne a1, a5, .LBB185_317
+; RV32-NEXT:    j .LBB185_318
+; RV32-NEXT:  .LBB185_486:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a4, a1, .LBB185_320
+; RV32-NEXT:  .LBB185_487:
+; RV32-NEXT:    mv a2, a7
+; RV32-NEXT:    bltu a7, a0, .LBB185_321
+; RV32-NEXT:  .LBB185_488:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:    bne a1, a4, .LBB185_322
+; RV32-NEXT:    j .LBB185_323
+; RV32-NEXT:  .LBB185_489:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a4, a0, .LBB185_327
+; RV32-NEXT:  .LBB185_490:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    beq a1, a3, .LBB185_328
+; RV32-NEXT:  .LBB185_491:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a3, a1, .LBB185_329
+; RV32-NEXT:    j .LBB185_330
+; RV32-NEXT:  .LBB185_492:
+; RV32-NEXT:    mv a2, s1
+; RV32-NEXT:    bltu s1, a0, .LBB185_340
+; RV32-NEXT:  .LBB185_493:
+; RV32-NEXT:    mv a0, s1
+; RV32-NEXT:    beq a1, a5, .LBB185_341
+; RV32-NEXT:  .LBB185_494:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a5, a1, .LBB185_342
+; RV32-NEXT:    j .LBB185_343
+; RV32-NEXT:  .LBB185_495:
+; RV32-NEXT:    mv a2, a6
+; RV32-NEXT:    bltu a6, a0, .LBB185_345
+; RV32-NEXT:  .LBB185_496:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, a5, .LBB185_346
+; RV32-NEXT:    j .LBB185_347
+; RV32-NEXT:  .LBB185_497:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt s0, a1, .LBB185_349
+; RV32-NEXT:  .LBB185_498:
+; RV32-NEXT:    mv a2, t0
+; RV32-NEXT:    bltu t0, a0, .LBB185_350
+; RV32-NEXT:  .LBB185_499:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, s0, .LBB185_351
+; RV32-NEXT:  .LBB185_500:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    blt s0, a1, .LBB185_352
+; RV32-NEXT:  .LBB185_501:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    blt a3, a1, .LBB185_353
+; RV32-NEXT:  .LBB185_502:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a4, a0, .LBB185_354
+; RV32-NEXT:  .LBB185_503:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    beq a1, a3, .LBB185_355
+; RV32-NEXT:  .LBB185_504:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bge a3, a1, .LBB185_356
+; RV32-NEXT:    j .LBB185_357
+;
+; RV64-LABEL: vreduce_smax_v64i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a1, a0, 384
+; RV64-NEXT:    vle64.v v16, (a1)
+; RV64-NEXT:    addi a1, a0, 256
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v24, (a0)
+; RV64-NEXT:    vle64.v v0, (a1)
+; RV64-NEXT:    vmax.vv v16, v24, v16
+; RV64-NEXT:    vmax.vv v8, v8, v0
+; RV64-NEXT:    vmax.vv v8, v8, v16
+; RV64-NEXT:    addi a0, zero, -1
+; RV64-NEXT:    slli a0, a0, 63
+; RV64-NEXT:    vsetvli a1, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.x v25, a0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredmax.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i64>, <64 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.smax.v64i64(<64 x i64> %v)
+  ret i64 %red
+}
+
+declare i8 @llvm.vector.reduce.umin.v1i8(<1 x i8>)
+
+define i8 @vreduce_umin_v1i8(<1 x i8>* %x) {
+; CHECK-LABEL: vreduce_umin_v1i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i8>, <1 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umin.v1i8(<1 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umin.v2i8(<2 x i8>)
+
+define i8 @vreduce_umin_v2i8(<2 x i8>* %x) {
+; CHECK-LABEL: vreduce_umin_v2i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i8>, <2 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umin.v2i8(<2 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umin.v4i8(<4 x i8>)
+
+define i8 @vreduce_umin_v4i8(<4 x i8>* %x) {
+; CHECK-LABEL: vreduce_umin_v4i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i8>, <4 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umin.v4i8(<4 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umin.v8i8(<8 x i8>)
+
+define i8 @vreduce_umin_v8i8(<8 x i8>* %x) {
+; CHECK-LABEL: vreduce_umin_v8i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i8>, <8 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umin.v8i8(<8 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umin.v16i8(<16 x i8>)
+
+define i8 @vreduce_umin_v16i8(<16 x i8>* %x) {
+; CHECK-LABEL: vreduce_umin_v16i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i8>, <16 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umin.v16i8(<16 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umin.v32i8(<32 x i8>)
+
+define i8 @vreduce_umin_v32i8(<32 x i8>* %x) {
+; CHECK-LABEL: vreduce_umin_v32i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vle8.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i8>, <32 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umin.v32i8(<32 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umin.v64i8(<64 x i8>)
+
+define i8 @vreduce_umin_v64i8(<64 x i8>* %x) {
+; CHECK-LABEL: vreduce_umin_v64i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i8>, <64 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umin.v64i8(<64 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umin.v128i8(<128 x i8>)
+
+define i8 @vreduce_umin_v128i8(<128 x i8>* %x) {
+; CHECK-LABEL: vreduce_umin_v128i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i8>, <128 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umin.v128i8(<128 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umin.v256i8(<256 x i8>)
+
+define i8 @vreduce_umin_v256i8(<256 x i8>* %x) {
+; CHECK-LABEL: vreduce_umin_v256i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle8.v v16, (a0)
+; CHECK-NEXT:    vminu.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <256 x i8>, <256 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umin.v256i8(<256 x i8> %v)
+  ret i8 %red
+}
+
+declare i16 @llvm.vector.reduce.umin.v1i16(<1 x i16>)
+
+define i16 @vreduce_umin_v1i16(<1 x i16>* %x) {
+; CHECK-LABEL: vreduce_umin_v1i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i16>, <1 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umin.v1i16(<1 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umin.v2i16(<2 x i16>)
+
+define i16 @vreduce_umin_v2i16(<2 x i16>* %x) {
+; CHECK-LABEL: vreduce_umin_v2i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i16>, <2 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umin.v2i16(<2 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umin.v4i16(<4 x i16>)
+
+define i16 @vreduce_umin_v4i16(<4 x i16>* %x) {
+; CHECK-LABEL: vreduce_umin_v4i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i16>, <4 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umin.v4i16(<4 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umin.v8i16(<8 x i16>)
+
+define i16 @vreduce_umin_v8i16(<8 x i16>* %x) {
+; CHECK-LABEL: vreduce_umin_v8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i16>, <8 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umin.v8i16(<8 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umin.v16i16(<16 x i16>)
+
+define i16 @vreduce_umin_v16i16(<16 x i16>* %x) {
+; CHECK-LABEL: vreduce_umin_v16i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vle16.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i16>, <16 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umin.v16i16(<16 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umin.v32i16(<32 x i16>)
+
+define i16 @vreduce_umin_v32i16(<32 x i16>* %x) {
+; CHECK-LABEL: vreduce_umin_v32i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i16>, <32 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umin.v32i16(<32 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umin.v64i16(<64 x i16>)
+
+define i16 @vreduce_umin_v64i16(<64 x i16>* %x) {
+; CHECK-LABEL: vreduce_umin_v64i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i16>, <64 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umin.v64i16(<64 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umin.v128i16(<128 x i16>)
+
+define i16 @vreduce_umin_v128i16(<128 x i16>* %x) {
+; CHECK-LABEL: vreduce_umin_v128i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle16.v v16, (a0)
+; CHECK-NEXT:    vminu.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i16>, <128 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umin.v128i16(<128 x i16> %v)
+  ret i16 %red
+}
+
+declare i32 @llvm.vector.reduce.umin.v1i32(<1 x i32>)
+
+define i32 @vreduce_umin_v1i32(<1 x i32>* %x) {
+; CHECK-LABEL: vreduce_umin_v1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i32>, <1 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umin.v1i32(<1 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umin.v2i32(<2 x i32>)
+
+define i32 @vreduce_umin_v2i32(<2 x i32>* %x) {
+; CHECK-LABEL: vreduce_umin_v2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i32>, <2 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umin.v2i32(<2 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umin.v4i32(<4 x i32>)
+
+define i32 @vreduce_umin_v4i32(<4 x i32>* %x) {
+; CHECK-LABEL: vreduce_umin_v4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, -1
+; CHECK-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i32>, <4 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umin.v4i32(<4 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umin.v8i32(<8 x i32>)
+
+define i32 @vreduce_umin_v8i32(<8 x i32>* %x) {
+; CHECK-LABEL: vreduce_umin_v8i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vle32.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i32>, <8 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umin.v8i32(<8 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umin.v16i32(<16 x i32>)
+
+define i32 @vreduce_umin_v16i32(<16 x i32>* %x) {
+; CHECK-LABEL: vreduce_umin_v16i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i32>, <16 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umin.v16i32(<16 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umin.v32i32(<32 x i32>)
+
+define i32 @vreduce_umin_v32i32(<32 x i32>* %x) {
+; CHECK-LABEL: vreduce_umin_v32i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i32>, <32 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umin.v32i32(<32 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umin.v64i32(<64 x i32>)
+
+define i32 @vreduce_umin_v64i32(<64 x i32>* %x) {
+; CHECK-LABEL: vreduce_umin_v64i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle32.v v16, (a0)
+; CHECK-NEXT:    vminu.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, -1
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredminu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i32>, <64 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umin.v64i32(<64 x i32> %v)
+  ret i32 %red
+}
+
+declare i64 @llvm.vector.reduce.umin.v1i64(<1 x i64>)
+
+define i64 @vreduce_umin_v1i64(<1 x i64>* %x) {
+; RV32-LABEL: vreduce_umin_v1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umin_v1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <1 x i64>, <1 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umin.v1i64(<1 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umin.v2i64(<2 x i64>)
+
+define i64 @vreduce_umin_v2i64(<2 x i64>* %x) {
+; RV32-LABEL: vreduce_umin_v2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v26, -1
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vredminu.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umin_v2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v26, -1
+; RV64-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV64-NEXT:    vredminu.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i64>, <2 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umin.v2i64(<2 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umin.v4i64(<4 x i64>)
+
+define i64 @vreduce_umin_v4i64(<4 x i64>* %x) {
+; RV32-LABEL: vreduce_umin_v4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV32-NEXT:    vle64.v v26, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, -1
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vredminu.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umin_v4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV64-NEXT:    vle64.v v26, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV64-NEXT:    vredminu.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i64>, <4 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umin.v4i64(<4 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umin.v8i64(<8 x i64>)
+
+define i64 @vreduce_umin_v8i64(<8 x i64>* %x) {
+; RV32-LABEL: vreduce_umin_v8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV32-NEXT:    vle64.v v28, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, -1
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vredminu.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umin_v8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV64-NEXT:    vle64.v v28, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV64-NEXT:    vredminu.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i64>, <8 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umin.v8i64(<8 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umin.v16i64(<16 x i64>)
+
+define i64 @vreduce_umin_v16i64(<16 x i64>* %x) {
+; RV32-LABEL: vreduce_umin_v16i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, -1
+; RV32-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV32-NEXT:    vredminu.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umin_v16i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredminu.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i64>, <16 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umin.v16i64(<16 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umin.v32i64(<32 x i64>)
+
+define i64 @vreduce_umin_v32i64(<32 x i64>* %x) {
+; RV32-LABEL: vreduce_umin_v32i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle64.v v16, (a0)
+; RV32-NEXT:    vminu.vv v8, v8, v16
+; RV32-NEXT:    vsetivli a0, 8, e64,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vminu.vv v28, v8, v16
+; RV32-NEXT:    vsetivli a0, 4, e64,m4,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v28, 4
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vminu.vv v26, v28, v8
+; RV32-NEXT:    vsetivli a0, 2, e64,m2,ta,mu
+; RV32-NEXT:    vslidedown.vi v28, v26, 2
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vminu.vv v25, v26, v28
+; RV32-NEXT:    vsetivli a0, 1, e64,m1,ta,mu
+; RV32-NEXT:    vslidedown.vi v26, v25, 1
+; RV32-NEXT:    vminu.vv v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umin_v32i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v16, (a0)
+; RV64-NEXT:    vminu.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredminu.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i64>, <32 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umin.v32i64(<32 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umin.v64i64(<64 x i64>)
+
+define i64 @vreduce_umin_v64i64(<64 x i64>* %x) nounwind {
+; RV32-LABEL: vreduce_umin_v64i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi sp, sp, -480
+; RV32-NEXT:    sw ra, 476(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s0, 472(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s1, 468(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s2, 464(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s3, 460(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s4, 456(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s5, 452(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s6, 448(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s7, 444(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s8, 440(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s9, 436(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s10, 432(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s11, 428(sp) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    sub sp, sp, a1
+; RV32-NEXT:    addi a2, a0, 384
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 320(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 324(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 312(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 316(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 304(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 308(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 296(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 300(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 288(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 292(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 280(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 284(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 272(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 276(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 264(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 268(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 256(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 260(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 248(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 252(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 240(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 244(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 232(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 236(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 388(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 228(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 384(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 224(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 216(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 220(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 212(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 424(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 256
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 380(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 208(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 200(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 204(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 196(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 420(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 376(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 192(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 184(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 188(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 180(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 416(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 372(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 176(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 168(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 172(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 164(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 412(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 368(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 160(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 364(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 156(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 148(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 152(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 144(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 408(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 360(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 140(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 356(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 136(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 128(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 132(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 128
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 124(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 404(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 352(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 120(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 112(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 116(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s0, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 400(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 348(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 108(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 344(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 104(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 340(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 100(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 336(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 84(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 92(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 96(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s9, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 396(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 332(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 80(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 72(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 76(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 64(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 68(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 56(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 60(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s7, v0
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 392(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 328(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 48(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a1, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a0)
+; RV32-NEXT:    vsetivli a0, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 224
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 216
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 208
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 200
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 192
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 184
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 176
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 168
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 160
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 152
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 144
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 136
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 7
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 120
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 112
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 104
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 96
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 88
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 80
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 72
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 6
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 56
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 48
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 40
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 5
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 3
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 24
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 224
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 40(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 216
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 44(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 208
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 32(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 200
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 36(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 192
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 28(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 184
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 12(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 176
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 24(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 168
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t2, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 160
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 20(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 152
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a6, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 144
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 16(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 136
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t0, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 7
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s2, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 120
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t3, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 112
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s8, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 104
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s11, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 96
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s ra, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 88
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s1, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 80
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 72
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t1, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 6
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 56
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 48
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s6, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 40
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s3, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 5
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a7, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s10, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 3
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t4, v0
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a4, v0
+; RV32-NEXT:    vmv.x.s a0, v8
+; RV32-NEXT:    vmv.x.s a3, v24
+; RV32-NEXT:    vmv.x.s a1, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t6, zero, 24
+; RV32-NEXT:    mul a2, a2, t6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t6, v8
+; RV32-NEXT:    mv s4, a0
+; RV32-NEXT:    bltu a1, a3, .LBB216_1
+; RV32-NEXT:    j .LBB216_358
+; RV32-NEXT:  .LBB216_1:
+; RV32-NEXT:    bltu a0, t6, .LBB216_2
+; RV32-NEXT:    j .LBB216_359
+; RV32-NEXT:  .LBB216_2:
+; RV32-NEXT:    beq a1, a3, .LBB216_3
+; RV32-NEXT:    j .LBB216_360
+; RV32-NEXT:  .LBB216_3:
+; RV32-NEXT:    bltu a1, a3, .LBB216_4
+; RV32-NEXT:    j .LBB216_361
+; RV32-NEXT:  .LBB216_4:
+; RV32-NEXT:    mv a3, a0
+; RV32-NEXT:    bltu a1, a4, .LBB216_5
+; RV32-NEXT:    j .LBB216_362
+; RV32-NEXT:  .LBB216_5:
+; RV32-NEXT:    bltu a0, t4, .LBB216_6
+; RV32-NEXT:    j .LBB216_363
+; RV32-NEXT:  .LBB216_6:
+; RV32-NEXT:    beq a1, a4, .LBB216_8
+; RV32-NEXT:  .LBB216_7:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB216_8:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw s0, 88(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw s9, 52(a2) # 4-byte Folded Spill
+; RV32-NEXT:    mv t4, s7
+; RV32-NEXT:    bltu a1, a4, .LBB216_9
+; RV32-NEXT:    j .LBB216_364
+; RV32-NEXT:  .LBB216_9:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, s10, .LBB216_10
+; RV32-NEXT:    j .LBB216_365
+; RV32-NEXT:  .LBB216_10:
+; RV32-NEXT:    bltu a0, a7, .LBB216_11
+; RV32-NEXT:    j .LBB216_366
+; RV32-NEXT:  .LBB216_11:
+; RV32-NEXT:    beq a1, s10, .LBB216_12
+; RV32-NEXT:    j .LBB216_367
+; RV32-NEXT:  .LBB216_12:
+; RV32-NEXT:    bltu a1, s10, .LBB216_13
+; RV32-NEXT:    j .LBB216_368
+; RV32-NEXT:  .LBB216_13:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, s3, .LBB216_15
+; RV32-NEXT:  .LBB216_14:
+; RV32-NEXT:    mv a4, s6
+; RV32-NEXT:  .LBB216_15:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s10, 84(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s6, .LBB216_16
+; RV32-NEXT:    j .LBB216_369
+; RV32-NEXT:  .LBB216_16:
+; RV32-NEXT:    beq a1, s3, .LBB216_17
+; RV32-NEXT:    j .LBB216_370
+; RV32-NEXT:  .LBB216_17:
+; RV32-NEXT:    bltu a1, s3, .LBB216_18
+; RV32-NEXT:    j .LBB216_371
+; RV32-NEXT:  .LBB216_18:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, a5, .LBB216_19
+; RV32-NEXT:    j .LBB216_372
+; RV32-NEXT:  .LBB216_19:
+; RV32-NEXT:    bltu a0, t5, .LBB216_20
+; RV32-NEXT:    j .LBB216_373
+; RV32-NEXT:  .LBB216_20:
+; RV32-NEXT:    beq a1, a5, .LBB216_21
+; RV32-NEXT:    j .LBB216_374
+; RV32-NEXT:  .LBB216_21:
+; RV32-NEXT:    bltu a1, a5, .LBB216_22
+; RV32-NEXT:    j .LBB216_375
+; RV32-NEXT:  .LBB216_22:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, t1, .LBB216_23
+; RV32-NEXT:    j .LBB216_376
+; RV32-NEXT:  .LBB216_23:
+; RV32-NEXT:    bltu a0, s5, .LBB216_24
+; RV32-NEXT:    j .LBB216_377
+; RV32-NEXT:  .LBB216_24:
+; RV32-NEXT:    beq a1, t1, .LBB216_26
+; RV32-NEXT:  .LBB216_25:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB216_26:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s5, 104(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, t1, .LBB216_28
+; RV32-NEXT:  # %bb.27:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:  .LBB216_28:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, s1, .LBB216_30
+; RV32-NEXT:  # %bb.29:
+; RV32-NEXT:    mv a4, ra
+; RV32-NEXT:  .LBB216_30:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t1, 100(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, ra, .LBB216_31
+; RV32-NEXT:    j .LBB216_378
+; RV32-NEXT:  .LBB216_31:
+; RV32-NEXT:    beq a1, s1, .LBB216_32
+; RV32-NEXT:    j .LBB216_379
+; RV32-NEXT:  .LBB216_32:
+; RV32-NEXT:    bltu a1, s1, .LBB216_33
+; RV32-NEXT:    j .LBB216_380
+; RV32-NEXT:  .LBB216_33:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, s11, .LBB216_35
+; RV32-NEXT:  .LBB216_34:
+; RV32-NEXT:    mv a4, s8
+; RV32-NEXT:  .LBB216_35:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t5, 48(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s8, .LBB216_37
+; RV32-NEXT:  # %bb.36:
+; RV32-NEXT:    mv a0, s8
+; RV32-NEXT:  .LBB216_37:
+; RV32-NEXT:    beq a1, s11, .LBB216_39
+; RV32-NEXT:  # %bb.38:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB216_39:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a7, 132(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, s11, .LBB216_41
+; RV32-NEXT:  # %bb.40:
+; RV32-NEXT:    mv a1, s11
+; RV32-NEXT:  .LBB216_41:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, t3, .LBB216_43
+; RV32-NEXT:  # %bb.42:
+; RV32-NEXT:    mv a4, s2
+; RV32-NEXT:  .LBB216_43:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s3, 96(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s11, 92(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s2, .LBB216_44
+; RV32-NEXT:    j .LBB216_381
+; RV32-NEXT:  .LBB216_44:
+; RV32-NEXT:    beq a1, t3, .LBB216_45
+; RV32-NEXT:    j .LBB216_382
+; RV32-NEXT:  .LBB216_45:
+; RV32-NEXT:    bltu a1, t3, .LBB216_47
+; RV32-NEXT:  .LBB216_46:
+; RV32-NEXT:    mv a1, t3
+; RV32-NEXT:  .LBB216_47:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s0, 16(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, t0, .LBB216_49
+; RV32-NEXT:  # %bb.48:
+; RV32-NEXT:    mv a2, s0
+; RV32-NEXT:  .LBB216_49:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s1, 24(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 20(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s0, .LBB216_51
+; RV32-NEXT:  # %bb.50:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:  .LBB216_51:
+; RV32-NEXT:    beq a1, t0, .LBB216_53
+; RV32-NEXT:  # %bb.52:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_53:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s8, 116(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, t0, .LBB216_55
+; RV32-NEXT:  # %bb.54:
+; RV32-NEXT:    mv a1, t0
+; RV32-NEXT:  .LBB216_55:
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    bltu a1, a6, .LBB216_57
+; RV32-NEXT:  # %bb.56:
+; RV32-NEXT:    mv s0, a4
+; RV32-NEXT:  .LBB216_57:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t0, 152(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s6, 80(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 28(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a4, .LBB216_58
+; RV32-NEXT:    j .LBB216_383
+; RV32-NEXT:  .LBB216_58:
+; RV32-NEXT:    beq a1, a6, .LBB216_59
+; RV32-NEXT:    j .LBB216_384
+; RV32-NEXT:  .LBB216_59:
+; RV32-NEXT:    bltu a1, a6, .LBB216_60
+; RV32-NEXT:    j .LBB216_385
+; RV32-NEXT:  .LBB216_60:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, t2, .LBB216_62
+; RV32-NEXT:  .LBB216_61:
+; RV32-NEXT:    mv a4, s1
+; RV32-NEXT:  .LBB216_62:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 36(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a6, 32(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, s1, .LBB216_63
+; RV32-NEXT:    j .LBB216_386
+; RV32-NEXT:  .LBB216_63:
+; RV32-NEXT:    beq a1, t2, .LBB216_64
+; RV32-NEXT:    j .LBB216_387
+; RV32-NEXT:  .LBB216_64:
+; RV32-NEXT:    bltu a1, t2, .LBB216_66
+; RV32-NEXT:  .LBB216_65:
+; RV32-NEXT:    mv a1, t2
+; RV32-NEXT:  .LBB216_66:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t6, 108(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s1, 12(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, s1, .LBB216_67
+; RV32-NEXT:    j .LBB216_388
+; RV32-NEXT:  .LBB216_67:
+; RV32-NEXT:    bltu a0, a5, .LBB216_68
+; RV32-NEXT:    j .LBB216_389
+; RV32-NEXT:  .LBB216_68:
+; RV32-NEXT:    beq a1, s1, .LBB216_69
+; RV32-NEXT:    j .LBB216_390
+; RV32-NEXT:  .LBB216_69:
+; RV32-NEXT:    bltu a1, s1, .LBB216_70
+; RV32-NEXT:    j .LBB216_391
+; RV32-NEXT:  .LBB216_70:
+; RV32-NEXT:    mv a5, a0
+; RV32-NEXT:    bltu a1, s0, .LBB216_72
+; RV32-NEXT:  .LBB216_71:
+; RV32-NEXT:    mv a5, a6
+; RV32-NEXT:  .LBB216_72:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t3, 120(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 44(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 40(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a6, .LBB216_73
+; RV32-NEXT:    j .LBB216_392
+; RV32-NEXT:  .LBB216_73:
+; RV32-NEXT:    beq a1, s0, .LBB216_74
+; RV32-NEXT:    j .LBB216_393
+; RV32-NEXT:  .LBB216_74:
+; RV32-NEXT:    bltu a1, s0, .LBB216_75
+; RV32-NEXT:    j .LBB216_394
+; RV32-NEXT:  .LBB216_75:
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    bltu a1, s1, .LBB216_77
+; RV32-NEXT:  .LBB216_76:
+; RV32-NEXT:    mv s0, a2
+; RV32-NEXT:  .LBB216_77:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw ra, 172(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s2, 76(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s7, 72(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 68(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a6, 64(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 60(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 56(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, a2, .LBB216_395
+; RV32-NEXT:  # %bb.78:
+; RV32-NEXT:    bne a1, s1, .LBB216_396
+; RV32-NEXT:  .LBB216_79:
+; RV32-NEXT:    bgeu a1, s1, .LBB216_397
+; RV32-NEXT:  .LBB216_80:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, t5, .LBB216_82
+; RV32-NEXT:  .LBB216_81:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a2, a2, t2
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 328(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_82:
+; RV32-NEXT:    csrr t2, vlenb
+; RV32-NEXT:    addi s0, zero, 232
+; RV32-NEXT:    mul t2, t2, s0
+; RV32-NEXT:    add t2, sp, t2
+; RV32-NEXT:    lw t2, 208(t2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr s0, vlenb
+; RV32-NEXT:    addi s1, zero, 232
+; RV32-NEXT:    mul s0, s0, s1
+; RV32-NEXT:    add s0, sp, s0
+; RV32-NEXT:    lw s4, 140(s0) # 4-byte Folded Reload
+; RV32-NEXT:    csrr s0, vlenb
+; RV32-NEXT:    addi s1, zero, 232
+; RV32-NEXT:    mul s0, s0, s1
+; RV32-NEXT:    add s0, sp, s0
+; RV32-NEXT:    lw s9, 328(s0) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s9, .LBB216_398
+; RV32-NEXT:  # %bb.83:
+; RV32-NEXT:    bne a1, t5, .LBB216_399
+; RV32-NEXT:  .LBB216_84:
+; RV32-NEXT:    bltu a1, t5, .LBB216_86
+; RV32-NEXT:  .LBB216_85:
+; RV32-NEXT:    mv a1, t5
+; RV32-NEXT:  .LBB216_86:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr t5, vlenb
+; RV32-NEXT:    addi s0, zero, 232
+; RV32-NEXT:    mul t5, t5, s0
+; RV32-NEXT:    add t5, sp, t5
+; RV32-NEXT:    lw s1, 392(t5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, s1, .LBB216_88
+; RV32-NEXT:  # %bb.87:
+; RV32-NEXT:    mv a2, t4
+; RV32-NEXT:  .LBB216_88:
+; RV32-NEXT:    bltu a0, t4, .LBB216_90
+; RV32-NEXT:  # %bb.89:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:  .LBB216_90:
+; RV32-NEXT:    csrr t4, vlenb
+; RV32-NEXT:    addi t5, zero, 232
+; RV32-NEXT:    mul t4, t4, t5
+; RV32-NEXT:    add t4, sp, t4
+; RV32-NEXT:    lw t4, 392(t4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB216_92
+; RV32-NEXT:  # %bb.91:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_92:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t4, zero, 232
+; RV32-NEXT:    mul a2, a2, t4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 392(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a2, .LBB216_400
+; RV32-NEXT:  # %bb.93:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a5, .LBB216_401
+; RV32-NEXT:  .LBB216_94:
+; RV32-NEXT:    bgeu a0, a3, .LBB216_402
+; RV32-NEXT:  .LBB216_95:
+; RV32-NEXT:    beq a1, a5, .LBB216_97
+; RV32-NEXT:  .LBB216_96:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_97:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 52(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a5, .LBB216_403
+; RV32-NEXT:  # %bb.98:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a4, .LBB216_404
+; RV32-NEXT:  .LBB216_99:
+; RV32-NEXT:    bgeu a0, a6, .LBB216_405
+; RV32-NEXT:  .LBB216_100:
+; RV32-NEXT:    beq a1, a4, .LBB216_102
+; RV32-NEXT:  .LBB216_101:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_102:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 136(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a4, .LBB216_406
+; RV32-NEXT:  # %bb.103:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s2, .LBB216_407
+; RV32-NEXT:  .LBB216_104:
+; RV32-NEXT:    bgeu a0, s7, .LBB216_408
+; RV32-NEXT:  .LBB216_105:
+; RV32-NEXT:    bne a1, s2, .LBB216_409
+; RV32-NEXT:  .LBB216_106:
+; RV32-NEXT:    bgeu a1, s2, .LBB216_410
+; RV32-NEXT:  .LBB216_107:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s6, .LBB216_109
+; RV32-NEXT:  .LBB216_108:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 332(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_109:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s7, 332(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s7, .LBB216_411
+; RV32-NEXT:  # %bb.110:
+; RV32-NEXT:    bne a1, s6, .LBB216_412
+; RV32-NEXT:  .LBB216_111:
+; RV32-NEXT:    bltu a1, s6, .LBB216_113
+; RV32-NEXT:  .LBB216_112:
+; RV32-NEXT:    mv a1, s6
+; RV32-NEXT:  .LBB216_113:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a5, 396(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a5, .LBB216_115
+; RV32-NEXT:  # %bb.114:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB216_115:
+; RV32-NEXT:    bltu a0, a3, .LBB216_117
+; RV32-NEXT:  # %bb.116:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB216_117:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t4, 396(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 88(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB216_119
+; RV32-NEXT:  # %bb.118:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_119:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 396(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 112(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a2, .LBB216_413
+; RV32-NEXT:  # %bb.120:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s3, .LBB216_414
+; RV32-NEXT:  .LBB216_121:
+; RV32-NEXT:    bgeu a0, s11, .LBB216_415
+; RV32-NEXT:  .LBB216_122:
+; RV32-NEXT:    bne a1, s3, .LBB216_416
+; RV32-NEXT:  .LBB216_123:
+; RV32-NEXT:    bgeu a1, s3, .LBB216_417
+; RV32-NEXT:  .LBB216_124:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s10, .LBB216_126
+; RV32-NEXT:  .LBB216_125:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 336(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_126:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s11, 336(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s11, .LBB216_418
+; RV32-NEXT:  # %bb.127:
+; RV32-NEXT:    bne a1, s10, .LBB216_419
+; RV32-NEXT:  .LBB216_128:
+; RV32-NEXT:    bgeu a1, s10, .LBB216_420
+; RV32-NEXT:  .LBB216_129:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, t1, .LBB216_131
+; RV32-NEXT:  .LBB216_130:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 340(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_131:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s10, 340(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s10, .LBB216_421
+; RV32-NEXT:  # %bb.132:
+; RV32-NEXT:    bne a1, t1, .LBB216_422
+; RV32-NEXT:  .LBB216_133:
+; RV32-NEXT:    bgeu a1, t1, .LBB216_423
+; RV32-NEXT:  .LBB216_134:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s5, .LBB216_136
+; RV32-NEXT:  .LBB216_135:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 344(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_136:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t1, 344(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, t1, .LBB216_424
+; RV32-NEXT:  # %bb.137:
+; RV32-NEXT:    bne a1, s5, .LBB216_425
+; RV32-NEXT:  .LBB216_138:
+; RV32-NEXT:    mv t1, t2
+; RV32-NEXT:    bgeu a1, s5, .LBB216_426
+; RV32-NEXT:  .LBB216_139:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, t6, .LBB216_141
+; RV32-NEXT:  .LBB216_140:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 348(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_141:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s5, 348(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s5, .LBB216_427
+; RV32-NEXT:  # %bb.142:
+; RV32-NEXT:    bne a1, t6, .LBB216_428
+; RV32-NEXT:  .LBB216_143:
+; RV32-NEXT:    bltu a1, t6, .LBB216_145
+; RV32-NEXT:  .LBB216_144:
+; RV32-NEXT:    mv a1, t6
+; RV32-NEXT:  .LBB216_145:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 400(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a4, .LBB216_147
+; RV32-NEXT:  # %bb.146:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:  .LBB216_147:
+; RV32-NEXT:    bltu a0, a5, .LBB216_149
+; RV32-NEXT:  # %bb.148:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:  .LBB216_149:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t4, 400(a4) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 124(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB216_151
+; RV32-NEXT:  # %bb.150:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_151:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 400(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 128(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a2, .LBB216_429
+; RV32-NEXT:  # %bb.152:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s8, .LBB216_430
+; RV32-NEXT:  .LBB216_153:
+; RV32-NEXT:    bgeu a0, a3, .LBB216_431
+; RV32-NEXT:  .LBB216_154:
+; RV32-NEXT:    bne a1, s8, .LBB216_432
+; RV32-NEXT:  .LBB216_155:
+; RV32-NEXT:    bgeu a1, s8, .LBB216_433
+; RV32-NEXT:  .LBB216_156:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, t3, .LBB216_158
+; RV32-NEXT:  .LBB216_157:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 352(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_158:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s9, 352(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s9, .LBB216_434
+; RV32-NEXT:  # %bb.159:
+; RV32-NEXT:    bne a1, t3, .LBB216_435
+; RV32-NEXT:  .LBB216_160:
+; RV32-NEXT:    bltu a1, t3, .LBB216_162
+; RV32-NEXT:  .LBB216_161:
+; RV32-NEXT:    mv a1, t3
+; RV32-NEXT:  .LBB216_162:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 404(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a3, .LBB216_164
+; RV32-NEXT:  # %bb.163:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB216_164:
+; RV32-NEXT:    bltu a0, a4, .LBB216_166
+; RV32-NEXT:  # %bb.165:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB216_166:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s10, 404(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 164(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s10, .LBB216_168
+; RV32-NEXT:  # %bb.167:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_168:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 404(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 144(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a2, .LBB216_436
+; RV32-NEXT:  # %bb.169:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a7, .LBB216_437
+; RV32-NEXT:  .LBB216_170:
+; RV32-NEXT:    bgeu a0, a5, .LBB216_438
+; RV32-NEXT:  .LBB216_171:
+; RV32-NEXT:    beq a1, a7, .LBB216_173
+; RV32-NEXT:  .LBB216_172:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_173:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 156(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a7, .LBB216_175
+; RV32-NEXT:  # %bb.174:
+; RV32-NEXT:    mv a1, a7
+; RV32-NEXT:  .LBB216_175:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s1, .LBB216_177
+; RV32-NEXT:  # %bb.176:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 356(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_177:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s8, 356(a6) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s2, 196(a6) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw a7, 220(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s8, .LBB216_439
+; RV32-NEXT:  # %bb.178:
+; RV32-NEXT:    bne a1, s1, .LBB216_440
+; RV32-NEXT:  .LBB216_179:
+; RV32-NEXT:    bgeu a1, s1, .LBB216_441
+; RV32-NEXT:  .LBB216_180:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s4, .LBB216_182
+; RV32-NEXT:  .LBB216_181:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 360(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_182:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw t5, 360(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, t5, .LBB216_442
+; RV32-NEXT:  # %bb.183:
+; RV32-NEXT:    bne a1, s4, .LBB216_443
+; RV32-NEXT:  .LBB216_184:
+; RV32-NEXT:    bltu a1, s4, .LBB216_186
+; RV32-NEXT:  .LBB216_185:
+; RV32-NEXT:    mv a1, s4
+; RV32-NEXT:  .LBB216_186:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s0, 408(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, s0, .LBB216_188
+; RV32-NEXT:  # %bb.187:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB216_188:
+; RV32-NEXT:    bltu a0, a4, .LBB216_190
+; RV32-NEXT:  # %bb.189:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB216_190:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s11, 408(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s11, .LBB216_192
+; RV32-NEXT:  # %bb.191:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_192:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 408(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 148(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a2, .LBB216_444
+; RV32-NEXT:  # %bb.193:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, t0, .LBB216_445
+; RV32-NEXT:  .LBB216_194:
+; RV32-NEXT:    bgeu a0, a4, .LBB216_446
+; RV32-NEXT:  .LBB216_195:
+; RV32-NEXT:    beq a1, t0, .LBB216_197
+; RV32-NEXT:  .LBB216_196:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_197:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 168(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, t0, .LBB216_199
+; RV32-NEXT:  # %bb.198:
+; RV32-NEXT:    mv a1, t0
+; RV32-NEXT:  .LBB216_199:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a5, .LBB216_201
+; RV32-NEXT:  # %bb.200:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 364(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_201:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a6, a6, t0
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw t0, 364(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, t0, .LBB216_203
+; RV32-NEXT:  # %bb.202:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:  .LBB216_203:
+; RV32-NEXT:    beq a1, a5, .LBB216_205
+; RV32-NEXT:  # %bb.204:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_205:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t0, 200(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 184(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 204(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a5, .LBB216_207
+; RV32-NEXT:  # %bb.206:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB216_207:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 160(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a5, .LBB216_209
+; RV32-NEXT:  # %bb.208:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 368(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_209:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s4, 368(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s4, .LBB216_447
+; RV32-NEXT:  # %bb.210:
+; RV32-NEXT:    bne a1, a5, .LBB216_448
+; RV32-NEXT:  .LBB216_211:
+; RV32-NEXT:    bltu a1, a5, .LBB216_213
+; RV32-NEXT:  .LBB216_212:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB216_213:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 412(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a5, .LBB216_215
+; RV32-NEXT:  # %bb.214:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB216_215:
+; RV32-NEXT:    bltu a0, a3, .LBB216_217
+; RV32-NEXT:  # %bb.216:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB216_217:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t5, 412(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 216(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t5, .LBB216_219
+; RV32-NEXT:  # %bb.218:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_219:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 412(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 180(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a2, .LBB216_449
+; RV32-NEXT:  # %bb.220:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, ra, .LBB216_450
+; RV32-NEXT:  .LBB216_221:
+; RV32-NEXT:    bgeu a0, a4, .LBB216_451
+; RV32-NEXT:  .LBB216_222:
+; RV32-NEXT:    beq a1, ra, .LBB216_224
+; RV32-NEXT:  .LBB216_223:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_224:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 176(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, ra, .LBB216_226
+; RV32-NEXT:  # %bb.225:
+; RV32-NEXT:    mv a1, ra
+; RV32-NEXT:  .LBB216_226:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a4, .LBB216_228
+; RV32-NEXT:  # %bb.227:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 372(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_228:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s7, 372(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s7, .LBB216_452
+; RV32-NEXT:  # %bb.229:
+; RV32-NEXT:    bne a1, a4, .LBB216_453
+; RV32-NEXT:  .LBB216_230:
+; RV32-NEXT:    bltu a1, a4, .LBB216_232
+; RV32-NEXT:  .LBB216_231:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB216_232:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 416(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a4, .LBB216_234
+; RV32-NEXT:  # %bb.233:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB216_234:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a6, 244(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a3, .LBB216_236
+; RV32-NEXT:  # %bb.235:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB216_236:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s9, 416(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t5, 264(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 212(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s9, .LBB216_238
+; RV32-NEXT:  # %bb.237:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_238:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 416(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 188(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a2, .LBB216_454
+; RV32-NEXT:  # %bb.239:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a3, .LBB216_455
+; RV32-NEXT:  .LBB216_240:
+; RV32-NEXT:    bgeu a0, s0, .LBB216_456
+; RV32-NEXT:  .LBB216_241:
+; RV32-NEXT:    bne a1, a3, .LBB216_457
+; RV32-NEXT:  .LBB216_242:
+; RV32-NEXT:    bltu a1, a3, .LBB216_244
+; RV32-NEXT:  .LBB216_243:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB216_244:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 192(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a3, .LBB216_246
+; RV32-NEXT:  # %bb.245:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a2, a2, t2
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 376(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_246:
+; RV32-NEXT:    csrr t2, vlenb
+; RV32-NEXT:    addi t3, zero, 232
+; RV32-NEXT:    mul t2, t2, t3
+; RV32-NEXT:    add t2, sp, t2
+; RV32-NEXT:    lw s11, 376(t2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, s11, .LBB216_458
+; RV32-NEXT:  # %bb.247:
+; RV32-NEXT:    bne a1, a3, .LBB216_459
+; RV32-NEXT:  .LBB216_248:
+; RV32-NEXT:    bltu a1, a3, .LBB216_250
+; RV32-NEXT:  .LBB216_249:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB216_250:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 420(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a3, .LBB216_252
+; RV32-NEXT:  # %bb.251:
+; RV32-NEXT:    mv a2, s2
+; RV32-NEXT:  .LBB216_252:
+; RV32-NEXT:    bltu a0, s2, .LBB216_254
+; RV32-NEXT:  # %bb.253:
+; RV32-NEXT:    mv a0, s2
+; RV32-NEXT:  .LBB216_254:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t2, 420(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t3, zero, 232
+; RV32-NEXT:    mul a3, a3, t3
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t4, 232(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t2, .LBB216_256
+; RV32-NEXT:  # %bb.255:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_256:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 420(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a2, .LBB216_258
+; RV32-NEXT:  # %bb.257:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:  .LBB216_258:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 228(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, s1, .LBB216_460
+; RV32-NEXT:  # %bb.259:
+; RV32-NEXT:    bgeu a0, t0, .LBB216_461
+; RV32-NEXT:  .LBB216_260:
+; RV32-NEXT:    bne a1, s1, .LBB216_462
+; RV32-NEXT:  .LBB216_261:
+; RV32-NEXT:    bgeu a1, s1, .LBB216_463
+; RV32-NEXT:  .LBB216_262:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, t1, .LBB216_264
+; RV32-NEXT:  .LBB216_263:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a2, a2, t0
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 380(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_264:
+; RV32-NEXT:    csrr t0, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul t0, t0, t2
+; RV32-NEXT:    add t0, sp, t0
+; RV32-NEXT:    lw t0, 380(t0) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a0, t0, .LBB216_464
+; RV32-NEXT:  # %bb.265:
+; RV32-NEXT:    bne a1, t1, .LBB216_465
+; RV32-NEXT:  .LBB216_266:
+; RV32-NEXT:    bltu a1, t1, .LBB216_268
+; RV32-NEXT:  .LBB216_267:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:  .LBB216_268:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr t0, vlenb
+; RV32-NEXT:    addi t1, zero, 232
+; RV32-NEXT:    mul t0, t0, t1
+; RV32-NEXT:    add t0, sp, t0
+; RV32-NEXT:    lw s0, 424(t0) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, s0, .LBB216_270
+; RV32-NEXT:  # %bb.269:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB216_270:
+; RV32-NEXT:    bltu a0, a4, .LBB216_272
+; RV32-NEXT:  # %bb.271:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB216_272:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a4, a4, t0
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t3, 424(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t3, .LBB216_274
+; RV32-NEXT:  # %bb.273:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_274:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 424(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a4, a4, t0
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 224(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a2, .LBB216_466
+; RV32-NEXT:  # %bb.275:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a7, .LBB216_467
+; RV32-NEXT:  .LBB216_276:
+; RV32-NEXT:    bgeu a0, a5, .LBB216_468
+; RV32-NEXT:  .LBB216_277:
+; RV32-NEXT:    bne a1, a7, .LBB216_469
+; RV32-NEXT:  .LBB216_278:
+; RV32-NEXT:    bgeu a1, a7, .LBB216_470
+; RV32-NEXT:  .LBB216_279:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a4, .LBB216_281
+; RV32-NEXT:  .LBB216_280:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 384(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_281:
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a5, a5, a7
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw t1, 384(a5) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a5, a5, a7
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw t0, 312(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, t1, .LBB216_283
+; RV32-NEXT:  # %bb.282:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:  .LBB216_283:
+; RV32-NEXT:    beq a1, a4, .LBB216_285
+; RV32-NEXT:  # %bb.284:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_285:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 236(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a4, .LBB216_287
+; RV32-NEXT:  # %bb.286:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB216_287:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a3, .LBB216_289
+; RV32-NEXT:  # %bb.288:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 388(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB216_289:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a7, 388(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a7, .LBB216_291
+; RV32-NEXT:  # %bb.290:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:  .LBB216_291:
+; RV32-NEXT:    beq a1, a3, .LBB216_293
+; RV32-NEXT:  # %bb.292:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_293:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t1, 256(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 260(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a7, 272(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 252(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 248(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a3, .LBB216_471
+; RV32-NEXT:  # %bb.294:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s1, .LBB216_472
+; RV32-NEXT:  .LBB216_295:
+; RV32-NEXT:    bgeu a0, t4, .LBB216_473
+; RV32-NEXT:  .LBB216_296:
+; RV32-NEXT:    beq a1, s1, .LBB216_298
+; RV32-NEXT:  .LBB216_297:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_298:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 240(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, s1, .LBB216_474
+; RV32-NEXT:  # %bb.299:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a6, .LBB216_475
+; RV32-NEXT:  .LBB216_300:
+; RV32-NEXT:    bgeu a0, a3, .LBB216_476
+; RV32-NEXT:  .LBB216_301:
+; RV32-NEXT:    bne a1, a6, .LBB216_477
+; RV32-NEXT:  .LBB216_302:
+; RV32-NEXT:    bgeu a1, a6, .LBB216_478
+; RV32-NEXT:  .LBB216_303:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a4, .LBB216_305
+; RV32-NEXT:  .LBB216_304:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:  .LBB216_305:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a6, 304(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 284(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a5, .LBB216_307
+; RV32-NEXT:  # %bb.306:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:  .LBB216_307:
+; RV32-NEXT:    beq a1, a4, .LBB216_309
+; RV32-NEXT:  # %bb.308:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_309:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 268(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a4, .LBB216_479
+; RV32-NEXT:  # %bb.310:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s0, .LBB216_480
+; RV32-NEXT:  .LBB216_311:
+; RV32-NEXT:    bgeu a0, t1, .LBB216_481
+; RV32-NEXT:  .LBB216_312:
+; RV32-NEXT:    bne a1, s0, .LBB216_482
+; RV32-NEXT:  .LBB216_313:
+; RV32-NEXT:    bgeu a1, s0, .LBB216_483
+; RV32-NEXT:  .LBB216_314:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a5, .LBB216_484
+; RV32-NEXT:  .LBB216_315:
+; RV32-NEXT:    bgeu a0, t5, .LBB216_485
+; RV32-NEXT:  .LBB216_316:
+; RV32-NEXT:    beq a1, a5, .LBB216_318
+; RV32-NEXT:  .LBB216_317:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_318:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 276(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a5, .LBB216_486
+; RV32-NEXT:  # %bb.319:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a4, .LBB216_487
+; RV32-NEXT:  .LBB216_320:
+; RV32-NEXT:    bgeu a0, a7, .LBB216_488
+; RV32-NEXT:  .LBB216_321:
+; RV32-NEXT:    beq a1, a4, .LBB216_323
+; RV32-NEXT:  .LBB216_322:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_323:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 296(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 292(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a4, .LBB216_325
+; RV32-NEXT:  # %bb.324:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB216_325:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a4, a4, a7
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 280(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a3, .LBB216_489
+; RV32-NEXT:  # %bb.326:
+; RV32-NEXT:    bgeu a0, a4, .LBB216_490
+; RV32-NEXT:  .LBB216_327:
+; RV32-NEXT:    bne a1, a3, .LBB216_491
+; RV32-NEXT:  .LBB216_328:
+; RV32-NEXT:    bltu a1, a3, .LBB216_330
+; RV32-NEXT:  .LBB216_329:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB216_330:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 288(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a5, .LBB216_332
+; RV32-NEXT:  # %bb.331:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB216_332:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a3, a3, a7
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s0, 316(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a0, a4, .LBB216_334
+; RV32-NEXT:  # %bb.333:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB216_334:
+; RV32-NEXT:    beq a1, a5, .LBB216_336
+; RV32-NEXT:  # %bb.335:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_336:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 320(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a1, a5, .LBB216_338
+; RV32-NEXT:  # %bb.337:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB216_338:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 300(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a5, .LBB216_492
+; RV32-NEXT:  # %bb.339:
+; RV32-NEXT:    bgeu a0, s1, .LBB216_493
+; RV32-NEXT:  .LBB216_340:
+; RV32-NEXT:    bne a1, a5, .LBB216_494
+; RV32-NEXT:  .LBB216_341:
+; RV32-NEXT:    bltu a1, a5, .LBB216_343
+; RV32-NEXT:  .LBB216_342:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB216_343:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 308(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a5, .LBB216_495
+; RV32-NEXT:  # %bb.344:
+; RV32-NEXT:    bgeu a0, a6, .LBB216_496
+; RV32-NEXT:  .LBB216_345:
+; RV32-NEXT:    beq a1, a5, .LBB216_347
+; RV32-NEXT:  .LBB216_346:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB216_347:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 324(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a1, a5, .LBB216_497
+; RV32-NEXT:  # %bb.348:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s0, .LBB216_498
+; RV32-NEXT:  .LBB216_349:
+; RV32-NEXT:    bgeu a0, t0, .LBB216_499
+; RV32-NEXT:  .LBB216_350:
+; RV32-NEXT:    bne a1, s0, .LBB216_500
+; RV32-NEXT:  .LBB216_351:
+; RV32-NEXT:    bgeu a1, s0, .LBB216_501
+; RV32-NEXT:  .LBB216_352:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a3, .LBB216_502
+; RV32-NEXT:  .LBB216_353:
+; RV32-NEXT:    bgeu a0, a4, .LBB216_503
+; RV32-NEXT:  .LBB216_354:
+; RV32-NEXT:    bne a1, a3, .LBB216_504
+; RV32-NEXT:  .LBB216_355:
+; RV32-NEXT:    bltu a1, a3, .LBB216_357
+; RV32-NEXT:  .LBB216_356:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB216_357:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add sp, sp, a2
+; RV32-NEXT:    lw s11, 428(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s10, 432(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s9, 436(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s8, 440(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s7, 444(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s6, 448(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s5, 452(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s4, 456(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s3, 460(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s2, 464(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s1, 468(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s0, 472(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw ra, 476(sp) # 4-byte Folded Reload
+; RV32-NEXT:    addi sp, sp, 480
+; RV32-NEXT:    ret
+; RV32-NEXT:  .LBB216_358:
+; RV32-NEXT:    mv s4, t6
+; RV32-NEXT:    bgeu a0, t6, .LBB216_359
+; RV32-NEXT:    j .LBB216_2
+; RV32-NEXT:  .LBB216_359:
+; RV32-NEXT:    mv a0, t6
+; RV32-NEXT:    bne a1, a3, .LBB216_360
+; RV32-NEXT:    j .LBB216_3
+; RV32-NEXT:  .LBB216_360:
+; RV32-NEXT:    mv a0, s4
+; RV32-NEXT:    bgeu a1, a3, .LBB216_361
+; RV32-NEXT:    j .LBB216_4
+; RV32-NEXT:  .LBB216_361:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:    mv a3, a0
+; RV32-NEXT:    bgeu a1, a4, .LBB216_362
+; RV32-NEXT:    j .LBB216_5
+; RV32-NEXT:  .LBB216_362:
+; RV32-NEXT:    mv a3, t4
+; RV32-NEXT:    bgeu a0, t4, .LBB216_363
+; RV32-NEXT:    j .LBB216_6
+; RV32-NEXT:  .LBB216_363:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:    beq a1, a4, .LBB216_505
+; RV32-NEXT:    j .LBB216_7
+; RV32-NEXT:  .LBB216_505:
+; RV32-NEXT:    j .LBB216_8
+; RV32-NEXT:  .LBB216_364:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bgeu a1, s10, .LBB216_365
+; RV32-NEXT:    j .LBB216_10
+; RV32-NEXT:  .LBB216_365:
+; RV32-NEXT:    mv a4, a7
+; RV32-NEXT:    bgeu a0, a7, .LBB216_366
+; RV32-NEXT:    j .LBB216_11
+; RV32-NEXT:  .LBB216_366:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:    bne a1, s10, .LBB216_367
+; RV32-NEXT:    j .LBB216_12
+; RV32-NEXT:  .LBB216_367:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bgeu a1, s10, .LBB216_368
+; RV32-NEXT:    j .LBB216_13
+; RV32-NEXT:  .LBB216_368:
+; RV32-NEXT:    mv a1, s10
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, s3, .LBB216_506
+; RV32-NEXT:    j .LBB216_14
+; RV32-NEXT:  .LBB216_506:
+; RV32-NEXT:    j .LBB216_15
+; RV32-NEXT:  .LBB216_369:
+; RV32-NEXT:    mv a0, s6
+; RV32-NEXT:    bne a1, s3, .LBB216_370
+; RV32-NEXT:    j .LBB216_17
+; RV32-NEXT:  .LBB216_370:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bgeu a1, s3, .LBB216_371
+; RV32-NEXT:    j .LBB216_18
+; RV32-NEXT:  .LBB216_371:
+; RV32-NEXT:    mv a1, s3
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bgeu a1, a5, .LBB216_372
+; RV32-NEXT:    j .LBB216_19
+; RV32-NEXT:  .LBB216_372:
+; RV32-NEXT:    mv a4, t5
+; RV32-NEXT:    bgeu a0, t5, .LBB216_373
+; RV32-NEXT:    j .LBB216_20
+; RV32-NEXT:  .LBB216_373:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    bne a1, a5, .LBB216_374
+; RV32-NEXT:    j .LBB216_21
+; RV32-NEXT:  .LBB216_374:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bgeu a1, a5, .LBB216_375
+; RV32-NEXT:    j .LBB216_22
+; RV32-NEXT:  .LBB216_375:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bgeu a1, t1, .LBB216_376
+; RV32-NEXT:    j .LBB216_23
+; RV32-NEXT:  .LBB216_376:
+; RV32-NEXT:    mv a4, s5
+; RV32-NEXT:    bgeu a0, s5, .LBB216_377
+; RV32-NEXT:    j .LBB216_24
+; RV32-NEXT:  .LBB216_377:
+; RV32-NEXT:    mv a0, s5
+; RV32-NEXT:    beq a1, t1, .LBB216_507
+; RV32-NEXT:    j .LBB216_25
+; RV32-NEXT:  .LBB216_507:
+; RV32-NEXT:    j .LBB216_26
+; RV32-NEXT:  .LBB216_378:
+; RV32-NEXT:    mv a0, ra
+; RV32-NEXT:    bne a1, s1, .LBB216_379
+; RV32-NEXT:    j .LBB216_32
+; RV32-NEXT:  .LBB216_379:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bgeu a1, s1, .LBB216_380
+; RV32-NEXT:    j .LBB216_33
+; RV32-NEXT:  .LBB216_380:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, s11, .LBB216_508
+; RV32-NEXT:    j .LBB216_34
+; RV32-NEXT:  .LBB216_508:
+; RV32-NEXT:    j .LBB216_35
+; RV32-NEXT:  .LBB216_381:
+; RV32-NEXT:    mv a0, s2
+; RV32-NEXT:    bne a1, t3, .LBB216_382
+; RV32-NEXT:    j .LBB216_45
+; RV32-NEXT:  .LBB216_382:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bltu a1, t3, .LBB216_509
+; RV32-NEXT:    j .LBB216_46
+; RV32-NEXT:  .LBB216_509:
+; RV32-NEXT:    j .LBB216_47
+; RV32-NEXT:  .LBB216_383:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, a6, .LBB216_384
+; RV32-NEXT:    j .LBB216_59
+; RV32-NEXT:  .LBB216_384:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    bgeu a1, a6, .LBB216_385
+; RV32-NEXT:    j .LBB216_60
+; RV32-NEXT:  .LBB216_385:
+; RV32-NEXT:    mv a1, a6
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a1, t2, .LBB216_510
+; RV32-NEXT:    j .LBB216_61
+; RV32-NEXT:  .LBB216_510:
+; RV32-NEXT:    j .LBB216_62
+; RV32-NEXT:  .LBB216_386:
+; RV32-NEXT:    mv a0, s1
+; RV32-NEXT:    bne a1, t2, .LBB216_387
+; RV32-NEXT:    j .LBB216_64
+; RV32-NEXT:  .LBB216_387:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bltu a1, t2, .LBB216_511
+; RV32-NEXT:    j .LBB216_65
+; RV32-NEXT:  .LBB216_511:
+; RV32-NEXT:    j .LBB216_66
+; RV32-NEXT:  .LBB216_388:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bgeu a0, a5, .LBB216_389
+; RV32-NEXT:    j .LBB216_68
+; RV32-NEXT:  .LBB216_389:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bne a1, s1, .LBB216_390
+; RV32-NEXT:    j .LBB216_69
+; RV32-NEXT:  .LBB216_390:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, s1, .LBB216_391
+; RV32-NEXT:    j .LBB216_70
+; RV32-NEXT:  .LBB216_391:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a5, a0
+; RV32-NEXT:    bltu a1, s0, .LBB216_512
+; RV32-NEXT:    j .LBB216_71
+; RV32-NEXT:  .LBB216_512:
+; RV32-NEXT:    j .LBB216_72
+; RV32-NEXT:  .LBB216_392:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, s0, .LBB216_393
+; RV32-NEXT:    j .LBB216_74
+; RV32-NEXT:  .LBB216_393:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bgeu a1, s0, .LBB216_394
+; RV32-NEXT:    j .LBB216_75
+; RV32-NEXT:  .LBB216_394:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    bltu a1, s1, .LBB216_513
+; RV32-NEXT:    j .LBB216_76
+; RV32-NEXT:  .LBB216_513:
+; RV32-NEXT:    j .LBB216_77
+; RV32-NEXT:  .LBB216_395:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    beq a1, s1, .LBB216_79
+; RV32-NEXT:  .LBB216_396:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    bltu a1, s1, .LBB216_80
+; RV32-NEXT:  .LBB216_397:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, t5, .LBB216_81
+; RV32-NEXT:    j .LBB216_82
+; RV32-NEXT:  .LBB216_398:
+; RV32-NEXT:    mv a0, s9
+; RV32-NEXT:    beq a1, t5, .LBB216_84
+; RV32-NEXT:  .LBB216_399:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, t5, .LBB216_85
+; RV32-NEXT:    j .LBB216_86
+; RV32-NEXT:  .LBB216_400:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a5, .LBB216_94
+; RV32-NEXT:  .LBB216_401:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a0, a3, .LBB216_95
+; RV32-NEXT:  .LBB216_402:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    bne a1, a5, .LBB216_96
+; RV32-NEXT:    j .LBB216_97
+; RV32-NEXT:  .LBB216_403:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a4, .LBB216_99
+; RV32-NEXT:  .LBB216_404:
+; RV32-NEXT:    mv a2, a6
+; RV32-NEXT:    bltu a0, a6, .LBB216_100
+; RV32-NEXT:  .LBB216_405:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, a4, .LBB216_101
+; RV32-NEXT:    j .LBB216_102
+; RV32-NEXT:  .LBB216_406:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s2, .LBB216_104
+; RV32-NEXT:  .LBB216_407:
+; RV32-NEXT:    mv a2, s7
+; RV32-NEXT:    bltu a0, s7, .LBB216_105
+; RV32-NEXT:  .LBB216_408:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, s2, .LBB216_106
+; RV32-NEXT:  .LBB216_409:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, s2, .LBB216_107
+; RV32-NEXT:  .LBB216_410:
+; RV32-NEXT:    mv a1, s2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s6, .LBB216_108
+; RV32-NEXT:    j .LBB216_109
+; RV32-NEXT:  .LBB216_411:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, s6, .LBB216_111
+; RV32-NEXT:  .LBB216_412:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, s6, .LBB216_112
+; RV32-NEXT:    j .LBB216_113
+; RV32-NEXT:  .LBB216_413:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s3, .LBB216_121
+; RV32-NEXT:  .LBB216_414:
+; RV32-NEXT:    mv a2, s11
+; RV32-NEXT:    bltu a0, s11, .LBB216_122
+; RV32-NEXT:  .LBB216_415:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, s3, .LBB216_123
+; RV32-NEXT:  .LBB216_416:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, s3, .LBB216_124
+; RV32-NEXT:  .LBB216_417:
+; RV32-NEXT:    mv a1, s3
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s10, .LBB216_125
+; RV32-NEXT:    j .LBB216_126
+; RV32-NEXT:  .LBB216_418:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, s10, .LBB216_128
+; RV32-NEXT:  .LBB216_419:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, s10, .LBB216_129
+; RV32-NEXT:  .LBB216_420:
+; RV32-NEXT:    mv a1, s10
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, t1, .LBB216_130
+; RV32-NEXT:    j .LBB216_131
+; RV32-NEXT:  .LBB216_421:
+; RV32-NEXT:    mv a0, s10
+; RV32-NEXT:    beq a1, t1, .LBB216_133
+; RV32-NEXT:  .LBB216_422:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, t1, .LBB216_134
+; RV32-NEXT:  .LBB216_423:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s5, .LBB216_135
+; RV32-NEXT:    j .LBB216_136
+; RV32-NEXT:  .LBB216_424:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:    beq a1, s5, .LBB216_138
+; RV32-NEXT:  .LBB216_425:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    mv t1, t2
+; RV32-NEXT:    bltu a1, s5, .LBB216_139
+; RV32-NEXT:  .LBB216_426:
+; RV32-NEXT:    mv a1, s5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, t6, .LBB216_140
+; RV32-NEXT:    j .LBB216_141
+; RV32-NEXT:  .LBB216_427:
+; RV32-NEXT:    mv a0, s5
+; RV32-NEXT:    beq a1, t6, .LBB216_143
+; RV32-NEXT:  .LBB216_428:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, t6, .LBB216_144
+; RV32-NEXT:    j .LBB216_145
+; RV32-NEXT:  .LBB216_429:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s8, .LBB216_153
+; RV32-NEXT:  .LBB216_430:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a0, a3, .LBB216_154
+; RV32-NEXT:  .LBB216_431:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    beq a1, s8, .LBB216_155
+; RV32-NEXT:  .LBB216_432:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, s8, .LBB216_156
+; RV32-NEXT:  .LBB216_433:
+; RV32-NEXT:    mv a1, s8
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, t3, .LBB216_157
+; RV32-NEXT:    j .LBB216_158
+; RV32-NEXT:  .LBB216_434:
+; RV32-NEXT:    mv a0, s9
+; RV32-NEXT:    beq a1, t3, .LBB216_160
+; RV32-NEXT:  .LBB216_435:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, t3, .LBB216_161
+; RV32-NEXT:    j .LBB216_162
+; RV32-NEXT:  .LBB216_436:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a7, .LBB216_170
+; RV32-NEXT:  .LBB216_437:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bltu a0, a5, .LBB216_171
+; RV32-NEXT:  .LBB216_438:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bne a1, a7, .LBB216_172
+; RV32-NEXT:    j .LBB216_173
+; RV32-NEXT:  .LBB216_439:
+; RV32-NEXT:    mv a0, s8
+; RV32-NEXT:    beq a1, s1, .LBB216_179
+; RV32-NEXT:  .LBB216_440:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, s1, .LBB216_180
+; RV32-NEXT:  .LBB216_441:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, s4, .LBB216_181
+; RV32-NEXT:    j .LBB216_182
+; RV32-NEXT:  .LBB216_442:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    beq a1, s4, .LBB216_184
+; RV32-NEXT:  .LBB216_443:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, s4, .LBB216_185
+; RV32-NEXT:    j .LBB216_186
+; RV32-NEXT:  .LBB216_444:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, t0, .LBB216_194
+; RV32-NEXT:  .LBB216_445:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a0, a4, .LBB216_195
+; RV32-NEXT:  .LBB216_446:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, t0, .LBB216_196
+; RV32-NEXT:    j .LBB216_197
+; RV32-NEXT:  .LBB216_447:
+; RV32-NEXT:    mv a0, s4
+; RV32-NEXT:    beq a1, a5, .LBB216_211
+; RV32-NEXT:  .LBB216_448:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, a5, .LBB216_212
+; RV32-NEXT:    j .LBB216_213
+; RV32-NEXT:  .LBB216_449:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, ra, .LBB216_221
+; RV32-NEXT:  .LBB216_450:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a0, a4, .LBB216_222
+; RV32-NEXT:  .LBB216_451:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, ra, .LBB216_223
+; RV32-NEXT:    j .LBB216_224
+; RV32-NEXT:  .LBB216_452:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, a4, .LBB216_230
+; RV32-NEXT:  .LBB216_453:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, a4, .LBB216_231
+; RV32-NEXT:    j .LBB216_232
+; RV32-NEXT:  .LBB216_454:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a3, .LBB216_240
+; RV32-NEXT:  .LBB216_455:
+; RV32-NEXT:    mv a2, s0
+; RV32-NEXT:    bltu a0, s0, .LBB216_241
+; RV32-NEXT:  .LBB216_456:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    beq a1, a3, .LBB216_242
+; RV32-NEXT:  .LBB216_457:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, a3, .LBB216_243
+; RV32-NEXT:    j .LBB216_244
+; RV32-NEXT:  .LBB216_458:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, a3, .LBB216_248
+; RV32-NEXT:  .LBB216_459:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, a3, .LBB216_249
+; RV32-NEXT:    j .LBB216_250
+; RV32-NEXT:  .LBB216_460:
+; RV32-NEXT:    mv a2, t0
+; RV32-NEXT:    bltu a0, t0, .LBB216_260
+; RV32-NEXT:  .LBB216_461:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, s1, .LBB216_261
+; RV32-NEXT:  .LBB216_462:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, s1, .LBB216_262
+; RV32-NEXT:  .LBB216_463:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, t1, .LBB216_263
+; RV32-NEXT:    j .LBB216_264
+; RV32-NEXT:  .LBB216_464:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, t1, .LBB216_266
+; RV32-NEXT:  .LBB216_465:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, t1, .LBB216_267
+; RV32-NEXT:    j .LBB216_268
+; RV32-NEXT:  .LBB216_466:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a7, .LBB216_276
+; RV32-NEXT:  .LBB216_467:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bltu a0, a5, .LBB216_277
+; RV32-NEXT:  .LBB216_468:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    beq a1, a7, .LBB216_278
+; RV32-NEXT:  .LBB216_469:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, a7, .LBB216_279
+; RV32-NEXT:  .LBB216_470:
+; RV32-NEXT:    mv a1, a7
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a4, .LBB216_280
+; RV32-NEXT:    j .LBB216_281
+; RV32-NEXT:  .LBB216_471:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s1, .LBB216_295
+; RV32-NEXT:  .LBB216_472:
+; RV32-NEXT:    mv a2, t4
+; RV32-NEXT:    bltu a0, t4, .LBB216_296
+; RV32-NEXT:  .LBB216_473:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:    bne a1, s1, .LBB216_297
+; RV32-NEXT:    j .LBB216_298
+; RV32-NEXT:  .LBB216_474:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a6, .LBB216_300
+; RV32-NEXT:  .LBB216_475:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a0, a3, .LBB216_301
+; RV32-NEXT:  .LBB216_476:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    beq a1, a6, .LBB216_302
+; RV32-NEXT:  .LBB216_477:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, a6, .LBB216_303
+; RV32-NEXT:  .LBB216_478:
+; RV32-NEXT:    mv a1, a6
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a1, a4, .LBB216_304
+; RV32-NEXT:    j .LBB216_305
+; RV32-NEXT:  .LBB216_479:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s0, .LBB216_311
+; RV32-NEXT:  .LBB216_480:
+; RV32-NEXT:    mv a2, t1
+; RV32-NEXT:    bltu a0, t1, .LBB216_312
+; RV32-NEXT:  .LBB216_481:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:    beq a1, s0, .LBB216_313
+; RV32-NEXT:  .LBB216_482:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, s0, .LBB216_314
+; RV32-NEXT:  .LBB216_483:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a5, .LBB216_315
+; RV32-NEXT:  .LBB216_484:
+; RV32-NEXT:    mv a2, t5
+; RV32-NEXT:    bltu a0, t5, .LBB216_316
+; RV32-NEXT:  .LBB216_485:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    bne a1, a5, .LBB216_317
+; RV32-NEXT:    j .LBB216_318
+; RV32-NEXT:  .LBB216_486:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a4, .LBB216_320
+; RV32-NEXT:  .LBB216_487:
+; RV32-NEXT:    mv a2, a7
+; RV32-NEXT:    bltu a0, a7, .LBB216_321
+; RV32-NEXT:  .LBB216_488:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:    bne a1, a4, .LBB216_322
+; RV32-NEXT:    j .LBB216_323
+; RV32-NEXT:  .LBB216_489:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a0, a4, .LBB216_327
+; RV32-NEXT:  .LBB216_490:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    beq a1, a3, .LBB216_328
+; RV32-NEXT:  .LBB216_491:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, a3, .LBB216_329
+; RV32-NEXT:    j .LBB216_330
+; RV32-NEXT:  .LBB216_492:
+; RV32-NEXT:    mv a2, s1
+; RV32-NEXT:    bltu a0, s1, .LBB216_340
+; RV32-NEXT:  .LBB216_493:
+; RV32-NEXT:    mv a0, s1
+; RV32-NEXT:    beq a1, a5, .LBB216_341
+; RV32-NEXT:  .LBB216_494:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, a5, .LBB216_342
+; RV32-NEXT:    j .LBB216_343
+; RV32-NEXT:  .LBB216_495:
+; RV32-NEXT:    mv a2, a6
+; RV32-NEXT:    bltu a0, a6, .LBB216_345
+; RV32-NEXT:  .LBB216_496:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, a5, .LBB216_346
+; RV32-NEXT:    j .LBB216_347
+; RV32-NEXT:  .LBB216_497:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, s0, .LBB216_349
+; RV32-NEXT:  .LBB216_498:
+; RV32-NEXT:    mv a2, t0
+; RV32-NEXT:    bltu a0, t0, .LBB216_350
+; RV32-NEXT:  .LBB216_499:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, s0, .LBB216_351
+; RV32-NEXT:  .LBB216_500:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a1, s0, .LBB216_352
+; RV32-NEXT:  .LBB216_501:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a1, a3, .LBB216_353
+; RV32-NEXT:  .LBB216_502:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a0, a4, .LBB216_354
+; RV32-NEXT:  .LBB216_503:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    beq a1, a3, .LBB216_355
+; RV32-NEXT:  .LBB216_504:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a1, a3, .LBB216_356
+; RV32-NEXT:    j .LBB216_357
+;
+; RV64-LABEL: vreduce_umin_v64i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a1, a0, 384
+; RV64-NEXT:    vle64.v v16, (a1)
+; RV64-NEXT:    addi a1, a0, 256
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v24, (a0)
+; RV64-NEXT:    vle64.v v0, (a1)
+; RV64-NEXT:    vminu.vv v16, v24, v16
+; RV64-NEXT:    vminu.vv v8, v8, v0
+; RV64-NEXT:    vminu.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, -1
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredminu.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i64>, <64 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umin.v64i64(<64 x i64> %v)
+  ret i64 %red
+}
+
+declare i8 @llvm.vector.reduce.umax.v1i8(<1 x i8>)
+
+define i8 @vreduce_umax_v1i8(<1 x i8>* %x) {
+; CHECK-LABEL: vreduce_umax_v1i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i8>, <1 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umax.v1i8(<1 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umax.v2i8(<2 x i8>)
+
+define i8 @vreduce_umax_v2i8(<2 x i8>* %x) {
+; CHECK-LABEL: vreduce_umax_v2i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e8,m1,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i8>, <2 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umax.v2i8(<2 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umax.v4i8(<4 x i8>)
+
+define i8 @vreduce_umax_v4i8(<4 x i8>* %x) {
+; CHECK-LABEL: vreduce_umax_v4i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e8,m1,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i8>, <4 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umax.v4i8(<4 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umax.v8i8(<8 x i8>)
+
+define i8 @vreduce_umax_v8i8(<8 x i8>* %x) {
+; CHECK-LABEL: vreduce_umax_v8i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 8, e8,m1,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i8>, <8 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umax.v8i8(<8 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umax.v16i8(<16 x i8>)
+
+define i8 @vreduce_umax_v16i8(<16 x i8>* %x) {
+; CHECK-LABEL: vreduce_umax_v16i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vle8.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 16, e8,m1,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i8>, <16 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umax.v16i8(<16 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umax.v32i8(<32 x i8>)
+
+define i8 @vreduce_umax_v32i8(<32 x i8>* %x) {
+; CHECK-LABEL: vreduce_umax_v32i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vle8.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m2,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i8>, <32 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umax.v32i8(<32 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umax.v64i8(<64 x i8>)
+
+define i8 @vreduce_umax_v64i8(<64 x i8>* %x) {
+; CHECK-LABEL: vreduce_umax_v64i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vle8.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m4,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i8>, <64 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umax.v64i8(<64 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umax.v128i8(<128 x i8>)
+
+define i8 @vreduce_umax_v128i8(<128 x i8>* %x) {
+; CHECK-LABEL: vreduce_umax_v128i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i8>, <128 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umax.v128i8(<128 x i8> %v)
+  ret i8 %red
+}
+
+declare i8 @llvm.vector.reduce.umax.v256i8(<256 x i8>)
+
+define i8 @vreduce_umax_v256i8(<256 x i8>* %x) {
+; CHECK-LABEL: vreduce_umax_v256i8:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 128
+; CHECK-NEXT:    vsetvli a2, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vle8.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle8.v v16, (a0)
+; CHECK-NEXT:    vmaxu.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e8,m8,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e8,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <256 x i8>, <256 x i8>* %x
+  %red = call i8 @llvm.vector.reduce.umax.v256i8(<256 x i8> %v)
+  ret i8 %red
+}
+
+declare i16 @llvm.vector.reduce.umax.v1i16(<1 x i16>)
+
+define i16 @vreduce_umax_v1i16(<1 x i16>* %x) {
+; CHECK-LABEL: vreduce_umax_v1i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i16>, <1 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umax.v1i16(<1 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umax.v2i16(<2 x i16>)
+
+define i16 @vreduce_umax_v2i16(<2 x i16>* %x) {
+; CHECK-LABEL: vreduce_umax_v2i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e16,m1,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i16>, <2 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umax.v2i16(<2 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umax.v4i16(<4 x i16>)
+
+define i16 @vreduce_umax_v4i16(<4 x i16>* %x) {
+; CHECK-LABEL: vreduce_umax_v4i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e16,m1,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i16>, <4 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umax.v4i16(<4 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umax.v8i16(<8 x i16>)
+
+define i16 @vreduce_umax_v8i16(<8 x i16>* %x) {
+; CHECK-LABEL: vreduce_umax_v8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vle16.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 8, e16,m1,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i16>, <8 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umax.v8i16(<8 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umax.v16i16(<16 x i16>)
+
+define i16 @vreduce_umax_v16i16(<16 x i16>* %x) {
+; CHECK-LABEL: vreduce_umax_v16i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vle16.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e16,m2,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i16>, <16 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umax.v16i16(<16 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umax.v32i16(<32 x i16>)
+
+define i16 @vreduce_umax_v32i16(<32 x i16>* %x) {
+; CHECK-LABEL: vreduce_umax_v32i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vle16.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m4,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i16>, <32 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umax.v32i16(<32 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umax.v64i16(<64 x i16>)
+
+define i16 @vreduce_umax_v64i16(<64 x i16>* %x) {
+; CHECK-LABEL: vreduce_umax_v64i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i16>, <64 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umax.v64i16(<64 x i16> %v)
+  ret i16 %red
+}
+
+declare i16 @llvm.vector.reduce.umax.v128i16(<128 x i16>)
+
+define i16 @vreduce_umax_v128i16(<128 x i16>* %x) {
+; CHECK-LABEL: vreduce_umax_v128i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 64
+; CHECK-NEXT:    vsetvli a2, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle16.v v16, (a0)
+; CHECK-NEXT:    vmaxu.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e16,m8,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e16,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <128 x i16>, <128 x i16>* %x
+  %red = call i16 @llvm.vector.reduce.umax.v128i16(<128 x i16> %v)
+  ret i16 %red
+}
+
+declare i32 @llvm.vector.reduce.umax.v1i32(<1 x i32>)
+
+define i32 @vreduce_umax_v1i32(<1 x i32>* %x) {
+; CHECK-LABEL: vreduce_umax_v1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 1, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <1 x i32>, <1 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umax.v1i32(<1 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umax.v2i32(<2 x i32>)
+
+define i32 @vreduce_umax_v2i32(<2 x i32>* %x) {
+; CHECK-LABEL: vreduce_umax_v2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 2, e32,m1,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <2 x i32>, <2 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umax.v2i32(<2 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umax.v4i32(<4 x i32>)
+
+define i32 @vreduce_umax_v4i32(<4 x i32>* %x) {
+; CHECK-LABEL: vreduce_umax_v4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vle32.v v25, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v26, 0
+; CHECK-NEXT:    vsetivli a0, 4, e32,m1,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v25, v26
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <4 x i32>, <4 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umax.v4i32(<4 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umax.v8i32(<8 x i32>)
+
+define i32 @vreduce_umax_v8i32(<8 x i32>* %x) {
+; CHECK-LABEL: vreduce_umax_v8i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vle32.v v26, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 8, e32,m2,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v26, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <8 x i32>, <8 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umax.v8i32(<8 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umax.v16i32(<16 x i32>)
+
+define i32 @vreduce_umax_v16i32(<16 x i32>* %x) {
+; CHECK-LABEL: vreduce_umax_v16i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli a1, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vle32.v v28, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetivli a0, 16, e32,m4,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v28, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <16 x i32>, <16 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umax.v16i32(<16 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umax.v32i32(<32 x i32>)
+
+define i32 @vreduce_umax_v32i32(<32 x i32>* %x) {
+; CHECK-LABEL: vreduce_umax_v32i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <32 x i32>, <32 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umax.v32i32(<32 x i32> %v)
+  ret i32 %red
+}
+
+declare i32 @llvm.vector.reduce.umax.v64i32(<64 x i32>)
+
+define i32 @vreduce_umax_v64i32(<64 x i32>* %x) {
+; CHECK-LABEL: vreduce_umax_v64i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, zero, 32
+; CHECK-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    addi a0, a0, 128
+; CHECK-NEXT:    vle32.v v16, (a0)
+; CHECK-NEXT:    vmaxu.vv v8, v8, v16
+; CHECK-NEXT:    vsetvli a0, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.v.i v25, 0
+; CHECK-NEXT:    vsetvli a0, a1, e32,m8,ta,mu
+; CHECK-NEXT:    vredmaxu.vs v25, v8, v25
+; CHECK-NEXT:    vsetvli zero, zero, e32,m1,ta,mu
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
+  %v = load <64 x i32>, <64 x i32>* %x
+  %red = call i32 @llvm.vector.reduce.umax.v64i32(<64 x i32> %v)
+  ret i32 %red
+}
+
+declare i64 @llvm.vector.reduce.umax.v1i64(<1 x i64>)
+
+define i64 @vreduce_umax_v1i64(<1 x i64>* %x) {
+; RV32-LABEL: vreduce_umax_v1i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umax_v1i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 1, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <1 x i64>, <1 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umax.v1i64(<1 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umax.v2i64(<2 x i64>)
+
+define i64 @vreduce_umax_v2i64(<2 x i64>* %x) {
+; RV32-LABEL: vreduce_umax_v2i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV32-NEXT:    vle64.v v25, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v26, 0
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vredmaxu.vs v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umax_v2i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 2, e64,m1,ta,mu
+; RV64-NEXT:    vle64.v v25, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v26, 0
+; RV64-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV64-NEXT:    vredmaxu.vs v25, v25, v26
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <2 x i64>, <2 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umax.v2i64(<2 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umax.v4i64(<4 x i64>)
+
+define i64 @vreduce_umax_v4i64(<4 x i64>* %x) {
+; RV32-LABEL: vreduce_umax_v4i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV32-NEXT:    vle64.v v26, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vredmaxu.vs v25, v26, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umax_v4i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 4, e64,m2,ta,mu
+; RV64-NEXT:    vle64.v v26, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV64-NEXT:    vredmaxu.vs v25, v26, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <4 x i64>, <4 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umax.v4i64(<4 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umax.v8i64(<8 x i64>)
+
+define i64 @vreduce_umax_v8i64(<8 x i64>* %x) {
+; RV32-LABEL: vreduce_umax_v8i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV32-NEXT:    vle64.v v28, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vredmaxu.vs v25, v28, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umax_v8i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 8, e64,m4,ta,mu
+; RV64-NEXT:    vle64.v v28, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV64-NEXT:    vredmaxu.vs v25, v28, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <8 x i64>, <8 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umax.v8i64(<8 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umax.v16i64(<16 x i64>)
+
+define i64 @vreduce_umax_v16i64(<16 x i64>* %x) {
+; RV32-LABEL: vreduce_umax_v16i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.v.i v25, 0
+; RV32-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV32-NEXT:    vredmaxu.vs v25, v8, v25
+; RV32-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetivli a2, 1, e64,m1,ta,mu
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umax_v16i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredmaxu.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <16 x i64>, <16 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umax.v16i64(<16 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umax.v32i64(<32 x i64>)
+
+define i64 @vreduce_umax_v32i64(<32 x i64>* %x) {
+; RV32-LABEL: vreduce_umax_v32i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV32-NEXT:    vle64.v v8, (a0)
+; RV32-NEXT:    addi a0, a0, 128
+; RV32-NEXT:    vle64.v v16, (a0)
+; RV32-NEXT:    vmaxu.vv v8, v8, v16
+; RV32-NEXT:    vsetivli a0, 8, e64,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vsetivli a0, 8, e64,m4,ta,mu
+; RV32-NEXT:    vmaxu.vv v28, v8, v16
+; RV32-NEXT:    vsetivli a0, 4, e64,m4,ta,mu
+; RV32-NEXT:    vslidedown.vi v8, v28, 4
+; RV32-NEXT:    vsetivli a0, 4, e64,m2,ta,mu
+; RV32-NEXT:    vmaxu.vv v26, v28, v8
+; RV32-NEXT:    vsetivli a0, 2, e64,m2,ta,mu
+; RV32-NEXT:    vslidedown.vi v28, v26, 2
+; RV32-NEXT:    vsetivli a0, 2, e64,m1,ta,mu
+; RV32-NEXT:    vmaxu.vv v25, v26, v28
+; RV32-NEXT:    vsetivli a0, 1, e64,m1,ta,mu
+; RV32-NEXT:    vslidedown.vi v26, v25, 1
+; RV32-NEXT:    vmaxu.vv v25, v25, v26
+; RV32-NEXT:    vmv.x.s a0, v25
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsrl.vx v25, v25, a1
+; RV32-NEXT:    vmv.x.s a1, v25
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: vreduce_umax_v32i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v16, (a0)
+; RV64-NEXT:    vmaxu.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredmaxu.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <32 x i64>, <32 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umax.v32i64(<32 x i64> %v)
+  ret i64 %red
+}
+
+declare i64 @llvm.vector.reduce.umax.v64i64(<64 x i64>)
+
+define i64 @vreduce_umax_v64i64(<64 x i64>* %x) nounwind {
+; RV32-LABEL: vreduce_umax_v64i64:
+; RV32:       # %bb.0:
+; RV32-NEXT:    addi sp, sp, -480
+; RV32-NEXT:    sw ra, 476(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s0, 472(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s1, 468(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s2, 464(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s3, 460(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s4, 456(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s5, 452(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s6, 448(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s7, 444(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s8, 440(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s9, 436(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s10, 432(sp) # 4-byte Folded Spill
+; RV32-NEXT:    sw s11, 428(sp) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    sub sp, sp, a1
+; RV32-NEXT:    addi a2, a0, 384
+; RV32-NEXT:    addi a1, zero, 32
+; RV32-NEXT:    vsetvli a3, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 320(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 324(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 312(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 316(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 304(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 308(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 296(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 300(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 288(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 292(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 280(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 284(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 272(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 276(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 264(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 268(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 256(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 260(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 248(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 252(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 240(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 244(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 232(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 236(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 388(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 228(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 384(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 224(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 216(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 220(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 212(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 424(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 256
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 380(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 208(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 200(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 204(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 196(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 420(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 376(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 192(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 184(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 188(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 180(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 416(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 372(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 176(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 168(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 172(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 164(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 412(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 368(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 160(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 364(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 156(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 148(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 152(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 144(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 408(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 360(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 140(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 356(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 136(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 128(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 132(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a2, a1, e32,m8,ta,mu
+; RV32-NEXT:    addi a2, a0, 128
+; RV32-NEXT:    vle32.v v8, (a2)
+; RV32-NEXT:    vsetivli a2, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 3
+; RV32-NEXT:    vslidedown.vi v16, v8, 1
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 224
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 124(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 216
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 404(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 208
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 352(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 200
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 120(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 192
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 112(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 184
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 116(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 176
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s0, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 168
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 400(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 160
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 348(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 152
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 108(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 144
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 344(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 136
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 104(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 7
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 340(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 120
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 100(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 112
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 336(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 104
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 84(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 96
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 92(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 88
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 96(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 80
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s9, v0
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 72
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 396(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 332(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 56
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 80(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 48
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 72(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 40
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 76(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 64(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 24
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 68(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 56(a3) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v0, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a2, v0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 60(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s7, v0
+; RV32-NEXT:    vmv.x.s a2, v24
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 392(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 328(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vmv.x.s a2, v16
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    sw a2, 48(a3) # 4-byte Folded Spill
+; RV32-NEXT:    vsetvli a1, a1, e32,m8,ta,mu
+; RV32-NEXT:    vle32.v v8, (a0)
+; RV32-NEXT:    vsetivli a0, 1, e32,m8,ta,mu
+; RV32-NEXT:    vslidedown.vi v16, v8, 30
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 224
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 31
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 216
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 28
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 208
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 29
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 200
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 26
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 192
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 27
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 184
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 24
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 176
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 25
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 168
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 22
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 160
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 23
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 152
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 20
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 144
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 21
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 136
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 18
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 7
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 19
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 120
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 16
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 112
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 17
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 104
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 14
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 96
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 15
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 88
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 12
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 80
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 13
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 72
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 10
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 6
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 11
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 56
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 8
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 48
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 9
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 40
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 6
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 5
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 7
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 4
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 3
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 5
+; RV32-NEXT:    vs8r.v v16, (sp) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v16, v8, 2
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 24
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vs8r.v v16, (a0) # Unknown-size Folded Spill
+; RV32-NEXT:    vslidedown.vi v24, v8, 1
+; RV32-NEXT:    vslidedown.vi v16, v8, 3
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 224
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 40(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 216
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 44(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 208
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 32(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 200
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 36(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 192
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 28(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 184
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 12(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 176
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 24(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 168
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t2, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 160
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 20(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 152
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a6, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 144
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a0, v0
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    addi a2, zero, 232
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    sw a0, 16(a1) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 136
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t0, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 7
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s2, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 120
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t3, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 112
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s8, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 104
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s11, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 96
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s ra, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 88
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s1, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 80
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 72
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t1, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 6
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 56
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a5, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 48
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s6, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    addi a1, zero, 40
+; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s3, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 5
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a7, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s s10, v0
+; RV32-NEXT:    csrr a0, vlenb
+; RV32-NEXT:    slli a0, a0, 3
+; RV32-NEXT:    add a0, sp, a0
+; RV32-NEXT:    vl8re8.v v0, (a0) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t4, v0
+; RV32-NEXT:    vl8re8.v v0, (sp) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s a4, v0
+; RV32-NEXT:    vmv.x.s a0, v8
+; RV32-NEXT:    vmv.x.s a1, v24
+; RV32-NEXT:    vmv.x.s a3, v16
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t6, zero, 24
+; RV32-NEXT:    mul a2, a2, t6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    vl8re8.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.x.s t6, v8
+; RV32-NEXT:    mv s4, a0
+; RV32-NEXT:    bltu a3, a1, .LBB247_1
+; RV32-NEXT:    j .LBB247_358
+; RV32-NEXT:  .LBB247_1:
+; RV32-NEXT:    bltu t6, a0, .LBB247_2
+; RV32-NEXT:    j .LBB247_359
+; RV32-NEXT:  .LBB247_2:
+; RV32-NEXT:    beq a1, a3, .LBB247_3
+; RV32-NEXT:    j .LBB247_360
+; RV32-NEXT:  .LBB247_3:
+; RV32-NEXT:    bltu a3, a1, .LBB247_4
+; RV32-NEXT:    j .LBB247_361
+; RV32-NEXT:  .LBB247_4:
+; RV32-NEXT:    mv a3, a0
+; RV32-NEXT:    bltu a4, a1, .LBB247_5
+; RV32-NEXT:    j .LBB247_362
+; RV32-NEXT:  .LBB247_5:
+; RV32-NEXT:    bltu t4, a0, .LBB247_6
+; RV32-NEXT:    j .LBB247_363
+; RV32-NEXT:  .LBB247_6:
+; RV32-NEXT:    beq a1, a4, .LBB247_8
+; RV32-NEXT:  .LBB247_7:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB247_8:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw s0, 88(a2) # 4-byte Folded Spill
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    sw s9, 52(a2) # 4-byte Folded Spill
+; RV32-NEXT:    mv t4, s7
+; RV32-NEXT:    bltu a4, a1, .LBB247_9
+; RV32-NEXT:    j .LBB247_364
+; RV32-NEXT:  .LBB247_9:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu s10, a1, .LBB247_10
+; RV32-NEXT:    j .LBB247_365
+; RV32-NEXT:  .LBB247_10:
+; RV32-NEXT:    bltu a7, a0, .LBB247_11
+; RV32-NEXT:    j .LBB247_366
+; RV32-NEXT:  .LBB247_11:
+; RV32-NEXT:    beq a1, s10, .LBB247_12
+; RV32-NEXT:    j .LBB247_367
+; RV32-NEXT:  .LBB247_12:
+; RV32-NEXT:    bltu s10, a1, .LBB247_13
+; RV32-NEXT:    j .LBB247_368
+; RV32-NEXT:  .LBB247_13:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu s3, a1, .LBB247_15
+; RV32-NEXT:  .LBB247_14:
+; RV32-NEXT:    mv a4, s6
+; RV32-NEXT:  .LBB247_15:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s10, 84(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s6, a0, .LBB247_16
+; RV32-NEXT:    j .LBB247_369
+; RV32-NEXT:  .LBB247_16:
+; RV32-NEXT:    beq a1, s3, .LBB247_17
+; RV32-NEXT:    j .LBB247_370
+; RV32-NEXT:  .LBB247_17:
+; RV32-NEXT:    bltu s3, a1, .LBB247_18
+; RV32-NEXT:    j .LBB247_371
+; RV32-NEXT:  .LBB247_18:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu a5, a1, .LBB247_19
+; RV32-NEXT:    j .LBB247_372
+; RV32-NEXT:  .LBB247_19:
+; RV32-NEXT:    bltu t5, a0, .LBB247_20
+; RV32-NEXT:    j .LBB247_373
+; RV32-NEXT:  .LBB247_20:
+; RV32-NEXT:    beq a1, a5, .LBB247_21
+; RV32-NEXT:    j .LBB247_374
+; RV32-NEXT:  .LBB247_21:
+; RV32-NEXT:    bltu a5, a1, .LBB247_22
+; RV32-NEXT:    j .LBB247_375
+; RV32-NEXT:  .LBB247_22:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu t1, a1, .LBB247_23
+; RV32-NEXT:    j .LBB247_376
+; RV32-NEXT:  .LBB247_23:
+; RV32-NEXT:    bltu s5, a0, .LBB247_24
+; RV32-NEXT:    j .LBB247_377
+; RV32-NEXT:  .LBB247_24:
+; RV32-NEXT:    beq a1, t1, .LBB247_26
+; RV32-NEXT:  .LBB247_25:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB247_26:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s5, 104(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu t1, a1, .LBB247_28
+; RV32-NEXT:  # %bb.27:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:  .LBB247_28:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu s1, a1, .LBB247_30
+; RV32-NEXT:  # %bb.29:
+; RV32-NEXT:    mv a4, ra
+; RV32-NEXT:  .LBB247_30:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t1, 100(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu ra, a0, .LBB247_31
+; RV32-NEXT:    j .LBB247_378
+; RV32-NEXT:  .LBB247_31:
+; RV32-NEXT:    beq a1, s1, .LBB247_32
+; RV32-NEXT:    j .LBB247_379
+; RV32-NEXT:  .LBB247_32:
+; RV32-NEXT:    bltu s1, a1, .LBB247_33
+; RV32-NEXT:    j .LBB247_380
+; RV32-NEXT:  .LBB247_33:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu s11, a1, .LBB247_35
+; RV32-NEXT:  .LBB247_34:
+; RV32-NEXT:    mv a4, s8
+; RV32-NEXT:  .LBB247_35:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t5, 48(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s8, a0, .LBB247_37
+; RV32-NEXT:  # %bb.36:
+; RV32-NEXT:    mv a0, s8
+; RV32-NEXT:  .LBB247_37:
+; RV32-NEXT:    beq a1, s11, .LBB247_39
+; RV32-NEXT:  # %bb.38:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB247_39:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a7, 132(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s11, a1, .LBB247_41
+; RV32-NEXT:  # %bb.40:
+; RV32-NEXT:    mv a1, s11
+; RV32-NEXT:  .LBB247_41:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu t3, a1, .LBB247_43
+; RV32-NEXT:  # %bb.42:
+; RV32-NEXT:    mv a4, s2
+; RV32-NEXT:  .LBB247_43:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s3, 96(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s11, 92(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s2, a0, .LBB247_44
+; RV32-NEXT:    j .LBB247_381
+; RV32-NEXT:  .LBB247_44:
+; RV32-NEXT:    beq a1, t3, .LBB247_45
+; RV32-NEXT:    j .LBB247_382
+; RV32-NEXT:  .LBB247_45:
+; RV32-NEXT:    bltu t3, a1, .LBB247_47
+; RV32-NEXT:  .LBB247_46:
+; RV32-NEXT:    mv a1, t3
+; RV32-NEXT:  .LBB247_47:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s0, 16(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu t0, a1, .LBB247_49
+; RV32-NEXT:  # %bb.48:
+; RV32-NEXT:    mv a2, s0
+; RV32-NEXT:  .LBB247_49:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s1, 24(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 20(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s0, a0, .LBB247_51
+; RV32-NEXT:  # %bb.50:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:  .LBB247_51:
+; RV32-NEXT:    beq a1, t0, .LBB247_53
+; RV32-NEXT:  # %bb.52:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_53:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s8, 116(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu t0, a1, .LBB247_55
+; RV32-NEXT:  # %bb.54:
+; RV32-NEXT:    mv a1, t0
+; RV32-NEXT:  .LBB247_55:
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    bltu a6, a1, .LBB247_57
+; RV32-NEXT:  # %bb.56:
+; RV32-NEXT:    mv s0, a4
+; RV32-NEXT:  .LBB247_57:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t0, 152(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s6, 80(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 28(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a4, a0, .LBB247_58
+; RV32-NEXT:    j .LBB247_383
+; RV32-NEXT:  .LBB247_58:
+; RV32-NEXT:    beq a1, a6, .LBB247_59
+; RV32-NEXT:    j .LBB247_384
+; RV32-NEXT:  .LBB247_59:
+; RV32-NEXT:    bltu a6, a1, .LBB247_60
+; RV32-NEXT:    j .LBB247_385
+; RV32-NEXT:  .LBB247_60:
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu t2, a1, .LBB247_62
+; RV32-NEXT:  .LBB247_61:
+; RV32-NEXT:    mv a4, s1
+; RV32-NEXT:  .LBB247_62:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 36(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a6, 32(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s1, a0, .LBB247_63
+; RV32-NEXT:    j .LBB247_386
+; RV32-NEXT:  .LBB247_63:
+; RV32-NEXT:    beq a1, t2, .LBB247_64
+; RV32-NEXT:    j .LBB247_387
+; RV32-NEXT:  .LBB247_64:
+; RV32-NEXT:    bltu t2, a1, .LBB247_66
+; RV32-NEXT:  .LBB247_65:
+; RV32-NEXT:    mv a1, t2
+; RV32-NEXT:  .LBB247_66:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t6, 108(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s1, 12(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s1, a1, .LBB247_67
+; RV32-NEXT:    j .LBB247_388
+; RV32-NEXT:  .LBB247_67:
+; RV32-NEXT:    bltu a5, a0, .LBB247_68
+; RV32-NEXT:    j .LBB247_389
+; RV32-NEXT:  .LBB247_68:
+; RV32-NEXT:    beq a1, s1, .LBB247_69
+; RV32-NEXT:    j .LBB247_390
+; RV32-NEXT:  .LBB247_69:
+; RV32-NEXT:    bltu s1, a1, .LBB247_70
+; RV32-NEXT:    j .LBB247_391
+; RV32-NEXT:  .LBB247_70:
+; RV32-NEXT:    mv a5, a0
+; RV32-NEXT:    bltu s0, a1, .LBB247_72
+; RV32-NEXT:  .LBB247_71:
+; RV32-NEXT:    mv a5, a6
+; RV32-NEXT:  .LBB247_72:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t3, 120(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 44(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 40(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a6, a0, .LBB247_73
+; RV32-NEXT:    j .LBB247_392
+; RV32-NEXT:  .LBB247_73:
+; RV32-NEXT:    beq a1, s0, .LBB247_74
+; RV32-NEXT:    j .LBB247_393
+; RV32-NEXT:  .LBB247_74:
+; RV32-NEXT:    bltu s0, a1, .LBB247_75
+; RV32-NEXT:    j .LBB247_394
+; RV32-NEXT:  .LBB247_75:
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    bltu s1, a1, .LBB247_77
+; RV32-NEXT:  .LBB247_76:
+; RV32-NEXT:    mv s0, a2
+; RV32-NEXT:  .LBB247_77:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw ra, 172(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s2, 76(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s7, 72(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 68(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a6, 64(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 60(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 56(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a0, .LBB247_395
+; RV32-NEXT:  # %bb.78:
+; RV32-NEXT:    bne a1, s1, .LBB247_396
+; RV32-NEXT:  .LBB247_79:
+; RV32-NEXT:    bgeu s1, a1, .LBB247_397
+; RV32-NEXT:  .LBB247_80:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu t5, a1, .LBB247_82
+; RV32-NEXT:  .LBB247_81:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a2, a2, t2
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 328(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_82:
+; RV32-NEXT:    csrr t2, vlenb
+; RV32-NEXT:    addi s0, zero, 232
+; RV32-NEXT:    mul t2, t2, s0
+; RV32-NEXT:    add t2, sp, t2
+; RV32-NEXT:    lw t2, 208(t2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr s0, vlenb
+; RV32-NEXT:    addi s1, zero, 232
+; RV32-NEXT:    mul s0, s0, s1
+; RV32-NEXT:    add s0, sp, s0
+; RV32-NEXT:    lw s4, 140(s0) # 4-byte Folded Reload
+; RV32-NEXT:    csrr s0, vlenb
+; RV32-NEXT:    addi s1, zero, 232
+; RV32-NEXT:    mul s0, s0, s1
+; RV32-NEXT:    add s0, sp, s0
+; RV32-NEXT:    lw s9, 328(s0) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s9, a0, .LBB247_398
+; RV32-NEXT:  # %bb.83:
+; RV32-NEXT:    bne a1, t5, .LBB247_399
+; RV32-NEXT:  .LBB247_84:
+; RV32-NEXT:    bltu t5, a1, .LBB247_86
+; RV32-NEXT:  .LBB247_85:
+; RV32-NEXT:    mv a1, t5
+; RV32-NEXT:  .LBB247_86:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr t5, vlenb
+; RV32-NEXT:    addi s0, zero, 232
+; RV32-NEXT:    mul t5, t5, s0
+; RV32-NEXT:    add t5, sp, t5
+; RV32-NEXT:    lw s1, 392(t5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s1, a1, .LBB247_88
+; RV32-NEXT:  # %bb.87:
+; RV32-NEXT:    mv a2, t4
+; RV32-NEXT:  .LBB247_88:
+; RV32-NEXT:    bltu t4, a0, .LBB247_90
+; RV32-NEXT:  # %bb.89:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:  .LBB247_90:
+; RV32-NEXT:    csrr t4, vlenb
+; RV32-NEXT:    addi t5, zero, 232
+; RV32-NEXT:    mul t4, t4, t5
+; RV32-NEXT:    add t4, sp, t4
+; RV32-NEXT:    lw t4, 392(t4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB247_92
+; RV32-NEXT:  # %bb.91:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_92:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t4, zero, 232
+; RV32-NEXT:    mul a2, a2, t4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 392(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a1, .LBB247_400
+; RV32-NEXT:  # %bb.93:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a5, a1, .LBB247_401
+; RV32-NEXT:  .LBB247_94:
+; RV32-NEXT:    bgeu a3, a0, .LBB247_402
+; RV32-NEXT:  .LBB247_95:
+; RV32-NEXT:    beq a1, a5, .LBB247_97
+; RV32-NEXT:  .LBB247_96:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_97:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 52(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a5, a1, .LBB247_403
+; RV32-NEXT:  # %bb.98:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a4, a1, .LBB247_404
+; RV32-NEXT:  .LBB247_99:
+; RV32-NEXT:    bgeu a6, a0, .LBB247_405
+; RV32-NEXT:  .LBB247_100:
+; RV32-NEXT:    beq a1, a4, .LBB247_102
+; RV32-NEXT:  .LBB247_101:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_102:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 136(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a4, a1, .LBB247_406
+; RV32-NEXT:  # %bb.103:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s2, a1, .LBB247_407
+; RV32-NEXT:  .LBB247_104:
+; RV32-NEXT:    bgeu s7, a0, .LBB247_408
+; RV32-NEXT:  .LBB247_105:
+; RV32-NEXT:    bne a1, s2, .LBB247_409
+; RV32-NEXT:  .LBB247_106:
+; RV32-NEXT:    bgeu s2, a1, .LBB247_410
+; RV32-NEXT:  .LBB247_107:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s6, a1, .LBB247_109
+; RV32-NEXT:  .LBB247_108:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 332(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_109:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s7, 332(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s7, a0, .LBB247_411
+; RV32-NEXT:  # %bb.110:
+; RV32-NEXT:    bne a1, s6, .LBB247_412
+; RV32-NEXT:  .LBB247_111:
+; RV32-NEXT:    bltu s6, a1, .LBB247_113
+; RV32-NEXT:  .LBB247_112:
+; RV32-NEXT:    mv a1, s6
+; RV32-NEXT:  .LBB247_113:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a5, 396(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a5, a1, .LBB247_115
+; RV32-NEXT:  # %bb.114:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB247_115:
+; RV32-NEXT:    bltu a3, a0, .LBB247_117
+; RV32-NEXT:  # %bb.116:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB247_117:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t4, 396(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 88(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB247_119
+; RV32-NEXT:  # %bb.118:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_119:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 396(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 112(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a1, .LBB247_413
+; RV32-NEXT:  # %bb.120:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s3, a1, .LBB247_414
+; RV32-NEXT:  .LBB247_121:
+; RV32-NEXT:    bgeu s11, a0, .LBB247_415
+; RV32-NEXT:  .LBB247_122:
+; RV32-NEXT:    bne a1, s3, .LBB247_416
+; RV32-NEXT:  .LBB247_123:
+; RV32-NEXT:    bgeu s3, a1, .LBB247_417
+; RV32-NEXT:  .LBB247_124:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s10, a1, .LBB247_126
+; RV32-NEXT:  .LBB247_125:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 336(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_126:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s11, 336(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s11, a0, .LBB247_418
+; RV32-NEXT:  # %bb.127:
+; RV32-NEXT:    bne a1, s10, .LBB247_419
+; RV32-NEXT:  .LBB247_128:
+; RV32-NEXT:    bgeu s10, a1, .LBB247_420
+; RV32-NEXT:  .LBB247_129:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu t1, a1, .LBB247_131
+; RV32-NEXT:  .LBB247_130:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 340(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_131:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s10, 340(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s10, a0, .LBB247_421
+; RV32-NEXT:  # %bb.132:
+; RV32-NEXT:    bne a1, t1, .LBB247_422
+; RV32-NEXT:  .LBB247_133:
+; RV32-NEXT:    bgeu t1, a1, .LBB247_423
+; RV32-NEXT:  .LBB247_134:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s5, a1, .LBB247_136
+; RV32-NEXT:  .LBB247_135:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 344(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_136:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t1, 344(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu t1, a0, .LBB247_424
+; RV32-NEXT:  # %bb.137:
+; RV32-NEXT:    bne a1, s5, .LBB247_425
+; RV32-NEXT:  .LBB247_138:
+; RV32-NEXT:    mv t1, t2
+; RV32-NEXT:    bgeu s5, a1, .LBB247_426
+; RV32-NEXT:  .LBB247_139:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu t6, a1, .LBB247_141
+; RV32-NEXT:  .LBB247_140:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 348(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_141:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s5, 348(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s5, a0, .LBB247_427
+; RV32-NEXT:  # %bb.142:
+; RV32-NEXT:    bne a1, t6, .LBB247_428
+; RV32-NEXT:  .LBB247_143:
+; RV32-NEXT:    bltu t6, a1, .LBB247_145
+; RV32-NEXT:  .LBB247_144:
+; RV32-NEXT:    mv a1, t6
+; RV32-NEXT:  .LBB247_145:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 400(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a4, a1, .LBB247_147
+; RV32-NEXT:  # %bb.146:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:  .LBB247_147:
+; RV32-NEXT:    bltu a5, a0, .LBB247_149
+; RV32-NEXT:  # %bb.148:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:  .LBB247_149:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t4, 400(a4) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 124(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t4, .LBB247_151
+; RV32-NEXT:  # %bb.150:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_151:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 400(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 128(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a1, .LBB247_429
+; RV32-NEXT:  # %bb.152:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s8, a1, .LBB247_430
+; RV32-NEXT:  .LBB247_153:
+; RV32-NEXT:    bgeu a3, a0, .LBB247_431
+; RV32-NEXT:  .LBB247_154:
+; RV32-NEXT:    bne a1, s8, .LBB247_432
+; RV32-NEXT:  .LBB247_155:
+; RV32-NEXT:    bgeu s8, a1, .LBB247_433
+; RV32-NEXT:  .LBB247_156:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu t3, a1, .LBB247_158
+; RV32-NEXT:  .LBB247_157:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 352(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_158:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s9, 352(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s9, a0, .LBB247_434
+; RV32-NEXT:  # %bb.159:
+; RV32-NEXT:    bne a1, t3, .LBB247_435
+; RV32-NEXT:  .LBB247_160:
+; RV32-NEXT:    bltu t3, a1, .LBB247_162
+; RV32-NEXT:  .LBB247_161:
+; RV32-NEXT:    mv a1, t3
+; RV32-NEXT:  .LBB247_162:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 404(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a3, a1, .LBB247_164
+; RV32-NEXT:  # %bb.163:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB247_164:
+; RV32-NEXT:    bltu a4, a0, .LBB247_166
+; RV32-NEXT:  # %bb.165:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB247_166:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s10, 404(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 164(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s10, .LBB247_168
+; RV32-NEXT:  # %bb.167:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_168:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 404(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 144(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a1, .LBB247_436
+; RV32-NEXT:  # %bb.169:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a7, a1, .LBB247_437
+; RV32-NEXT:  .LBB247_170:
+; RV32-NEXT:    bgeu a5, a0, .LBB247_438
+; RV32-NEXT:  .LBB247_171:
+; RV32-NEXT:    beq a1, a7, .LBB247_173
+; RV32-NEXT:  .LBB247_172:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_173:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 156(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a7, a1, .LBB247_175
+; RV32-NEXT:  # %bb.174:
+; RV32-NEXT:    mv a1, a7
+; RV32-NEXT:  .LBB247_175:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s1, a1, .LBB247_177
+; RV32-NEXT:  # %bb.176:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 356(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_177:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s8, 356(a6) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s2, 196(a6) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a6, a6, a7
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw a7, 220(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s8, a0, .LBB247_439
+; RV32-NEXT:  # %bb.178:
+; RV32-NEXT:    bne a1, s1, .LBB247_440
+; RV32-NEXT:  .LBB247_179:
+; RV32-NEXT:    bgeu s1, a1, .LBB247_441
+; RV32-NEXT:  .LBB247_180:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s4, a1, .LBB247_182
+; RV32-NEXT:  .LBB247_181:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 360(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_182:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw t5, 360(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu t5, a0, .LBB247_442
+; RV32-NEXT:  # %bb.183:
+; RV32-NEXT:    bne a1, s4, .LBB247_443
+; RV32-NEXT:  .LBB247_184:
+; RV32-NEXT:    bltu s4, a1, .LBB247_186
+; RV32-NEXT:  .LBB247_185:
+; RV32-NEXT:    mv a1, s4
+; RV32-NEXT:  .LBB247_186:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s0, 408(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s0, a1, .LBB247_188
+; RV32-NEXT:  # %bb.187:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB247_188:
+; RV32-NEXT:    bltu a4, a0, .LBB247_190
+; RV32-NEXT:  # %bb.189:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB247_190:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw s11, 408(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s11, .LBB247_192
+; RV32-NEXT:  # %bb.191:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_192:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 408(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 148(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a1, .LBB247_444
+; RV32-NEXT:  # %bb.193:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu t0, a1, .LBB247_445
+; RV32-NEXT:  .LBB247_194:
+; RV32-NEXT:    bgeu a4, a0, .LBB247_446
+; RV32-NEXT:  .LBB247_195:
+; RV32-NEXT:    beq a1, t0, .LBB247_197
+; RV32-NEXT:  .LBB247_196:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_197:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 168(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu t0, a1, .LBB247_199
+; RV32-NEXT:  # %bb.198:
+; RV32-NEXT:    mv a1, t0
+; RV32-NEXT:  .LBB247_199:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a5, a1, .LBB247_201
+; RV32-NEXT:  # %bb.200:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 364(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_201:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a6, a6, t0
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw t0, 364(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bltu t0, a0, .LBB247_203
+; RV32-NEXT:  # %bb.202:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:  .LBB247_203:
+; RV32-NEXT:    beq a1, a5, .LBB247_205
+; RV32-NEXT:  # %bb.204:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_205:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t0, 200(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 184(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 204(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a5, a1, .LBB247_207
+; RV32-NEXT:  # %bb.206:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB247_207:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 160(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a5, a1, .LBB247_209
+; RV32-NEXT:  # %bb.208:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 368(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_209:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s4, 368(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s4, a0, .LBB247_447
+; RV32-NEXT:  # %bb.210:
+; RV32-NEXT:    bne a1, a5, .LBB247_448
+; RV32-NEXT:  .LBB247_211:
+; RV32-NEXT:    bltu a5, a1, .LBB247_213
+; RV32-NEXT:  .LBB247_212:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB247_213:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a5, a5, a6
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw a5, 412(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a5, a1, .LBB247_215
+; RV32-NEXT:  # %bb.214:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB247_215:
+; RV32-NEXT:    bltu a3, a0, .LBB247_217
+; RV32-NEXT:  # %bb.216:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB247_217:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t5, 412(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 216(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t5, .LBB247_219
+; RV32-NEXT:  # %bb.218:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_219:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 412(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 180(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a1, .LBB247_449
+; RV32-NEXT:  # %bb.220:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu ra, a1, .LBB247_450
+; RV32-NEXT:  .LBB247_221:
+; RV32-NEXT:    bgeu a4, a0, .LBB247_451
+; RV32-NEXT:  .LBB247_222:
+; RV32-NEXT:    beq a1, ra, .LBB247_224
+; RV32-NEXT:  .LBB247_223:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_224:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 176(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu ra, a1, .LBB247_226
+; RV32-NEXT:  # %bb.225:
+; RV32-NEXT:    mv a1, ra
+; RV32-NEXT:  .LBB247_226:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a4, a1, .LBB247_228
+; RV32-NEXT:  # %bb.227:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a2, a2, a6
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 372(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_228:
+; RV32-NEXT:    csrr a6, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a6, a6, t2
+; RV32-NEXT:    add a6, sp, a6
+; RV32-NEXT:    lw s7, 372(a6) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s7, a0, .LBB247_452
+; RV32-NEXT:  # %bb.229:
+; RV32-NEXT:    bne a1, a4, .LBB247_453
+; RV32-NEXT:  .LBB247_230:
+; RV32-NEXT:    bltu a4, a1, .LBB247_232
+; RV32-NEXT:  .LBB247_231:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB247_232:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 416(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a4, a1, .LBB247_234
+; RV32-NEXT:  # %bb.233:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:  .LBB247_234:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a4, a4, a6
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a6, 244(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a3, a0, .LBB247_236
+; RV32-NEXT:  # %bb.235:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:  .LBB247_236:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s9, 416(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t5, 264(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 212(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, s9, .LBB247_238
+; RV32-NEXT:  # %bb.237:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_238:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 416(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 188(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a1, .LBB247_454
+; RV32-NEXT:  # %bb.239:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a3, a1, .LBB247_455
+; RV32-NEXT:  .LBB247_240:
+; RV32-NEXT:    bgeu s0, a0, .LBB247_456
+; RV32-NEXT:  .LBB247_241:
+; RV32-NEXT:    bne a1, a3, .LBB247_457
+; RV32-NEXT:  .LBB247_242:
+; RV32-NEXT:    bltu a3, a1, .LBB247_244
+; RV32-NEXT:  .LBB247_243:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB247_244:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 192(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a3, a1, .LBB247_246
+; RV32-NEXT:  # %bb.245:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a2, a2, t2
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 376(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_246:
+; RV32-NEXT:    csrr t2, vlenb
+; RV32-NEXT:    addi t3, zero, 232
+; RV32-NEXT:    mul t2, t2, t3
+; RV32-NEXT:    add t2, sp, t2
+; RV32-NEXT:    lw s11, 376(t2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s11, a0, .LBB247_458
+; RV32-NEXT:  # %bb.247:
+; RV32-NEXT:    bne a1, a3, .LBB247_459
+; RV32-NEXT:  .LBB247_248:
+; RV32-NEXT:    bltu a3, a1, .LBB247_250
+; RV32-NEXT:  .LBB247_249:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB247_250:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 420(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a3, a1, .LBB247_252
+; RV32-NEXT:  # %bb.251:
+; RV32-NEXT:    mv a2, s2
+; RV32-NEXT:  .LBB247_252:
+; RV32-NEXT:    bltu s2, a0, .LBB247_254
+; RV32-NEXT:  # %bb.253:
+; RV32-NEXT:    mv a0, s2
+; RV32-NEXT:  .LBB247_254:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t2, 420(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t3, zero, 232
+; RV32-NEXT:    mul a3, a3, t3
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw t4, 232(a3) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t2, .LBB247_256
+; RV32-NEXT:  # %bb.255:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_256:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 420(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a2, a1, .LBB247_258
+; RV32-NEXT:  # %bb.257:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:  .LBB247_258:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 228(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s1, a1, .LBB247_460
+; RV32-NEXT:  # %bb.259:
+; RV32-NEXT:    bgeu t0, a0, .LBB247_461
+; RV32-NEXT:  .LBB247_260:
+; RV32-NEXT:    bne a1, s1, .LBB247_462
+; RV32-NEXT:  .LBB247_261:
+; RV32-NEXT:    bgeu s1, a1, .LBB247_463
+; RV32-NEXT:  .LBB247_262:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu t1, a1, .LBB247_264
+; RV32-NEXT:  .LBB247_263:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a2, a2, t0
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 380(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_264:
+; RV32-NEXT:    csrr t0, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul t0, t0, t2
+; RV32-NEXT:    add t0, sp, t0
+; RV32-NEXT:    lw t0, 380(t0) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu t0, a0, .LBB247_464
+; RV32-NEXT:  # %bb.265:
+; RV32-NEXT:    bne a1, t1, .LBB247_465
+; RV32-NEXT:  .LBB247_266:
+; RV32-NEXT:    bltu t1, a1, .LBB247_268
+; RV32-NEXT:  .LBB247_267:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:  .LBB247_268:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr t0, vlenb
+; RV32-NEXT:    addi t1, zero, 232
+; RV32-NEXT:    mul t0, t0, t1
+; RV32-NEXT:    add t0, sp, t0
+; RV32-NEXT:    lw s0, 424(t0) # 4-byte Folded Reload
+; RV32-NEXT:    bltu s0, a1, .LBB247_270
+; RV32-NEXT:  # %bb.269:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB247_270:
+; RV32-NEXT:    bltu a4, a0, .LBB247_272
+; RV32-NEXT:  # %bb.271:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB247_272:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a4, a4, t0
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw t3, 424(a4) # 4-byte Folded Reload
+; RV32-NEXT:    beq a1, t3, .LBB247_274
+; RV32-NEXT:  # %bb.273:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_274:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 424(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi t0, zero, 232
+; RV32-NEXT:    mul a4, a4, t0
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 224(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a2, a1, .LBB247_466
+; RV32-NEXT:  # %bb.275:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a7, a1, .LBB247_467
+; RV32-NEXT:  .LBB247_276:
+; RV32-NEXT:    bgeu a5, a0, .LBB247_468
+; RV32-NEXT:  .LBB247_277:
+; RV32-NEXT:    bne a1, a7, .LBB247_469
+; RV32-NEXT:  .LBB247_278:
+; RV32-NEXT:    bgeu a7, a1, .LBB247_470
+; RV32-NEXT:  .LBB247_279:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a4, a1, .LBB247_281
+; RV32-NEXT:  .LBB247_280:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 384(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_281:
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a5, a5, a7
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw t1, 384(a5) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a5, a5, a7
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    lw t0, 312(a5) # 4-byte Folded Reload
+; RV32-NEXT:    bltu t1, a0, .LBB247_283
+; RV32-NEXT:  # %bb.282:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:  .LBB247_283:
+; RV32-NEXT:    beq a1, a4, .LBB247_285
+; RV32-NEXT:  # %bb.284:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_285:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 236(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a4, a1, .LBB247_287
+; RV32-NEXT:  # %bb.286:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB247_287:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a3, a1, .LBB247_289
+; RV32-NEXT:  # %bb.288:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a2, 388(a2) # 4-byte Folded Reload
+; RV32-NEXT:  .LBB247_289:
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a7, 388(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a7, a0, .LBB247_291
+; RV32-NEXT:  # %bb.290:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:  .LBB247_291:
+; RV32-NEXT:    beq a1, a3, .LBB247_293
+; RV32-NEXT:  # %bb.292:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_293:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw t1, 256(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s0, 260(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a7, 272(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 252(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 248(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a3, a1, .LBB247_471
+; RV32-NEXT:  # %bb.294:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s1, a1, .LBB247_472
+; RV32-NEXT:  .LBB247_295:
+; RV32-NEXT:    bgeu t4, a0, .LBB247_473
+; RV32-NEXT:  .LBB247_296:
+; RV32-NEXT:    beq a1, s1, .LBB247_298
+; RV32-NEXT:  .LBB247_297:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_298:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 240(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu s1, a1, .LBB247_474
+; RV32-NEXT:  # %bb.299:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a6, a1, .LBB247_475
+; RV32-NEXT:  .LBB247_300:
+; RV32-NEXT:    bgeu a3, a0, .LBB247_476
+; RV32-NEXT:  .LBB247_301:
+; RV32-NEXT:    bne a1, a6, .LBB247_477
+; RV32-NEXT:  .LBB247_302:
+; RV32-NEXT:    bgeu a6, a1, .LBB247_478
+; RV32-NEXT:  .LBB247_303:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a4, a1, .LBB247_305
+; RV32-NEXT:  .LBB247_304:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:  .LBB247_305:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a6, zero, 232
+; RV32-NEXT:    mul a3, a3, a6
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a6, 304(a3) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi t2, zero, 232
+; RV32-NEXT:    mul a3, a3, t2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a3, 284(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a5, a0, .LBB247_307
+; RV32-NEXT:  # %bb.306:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:  .LBB247_307:
+; RV32-NEXT:    beq a1, a4, .LBB247_309
+; RV32-NEXT:  # %bb.308:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_309:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 268(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a4, a1, .LBB247_479
+; RV32-NEXT:  # %bb.310:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s0, a1, .LBB247_480
+; RV32-NEXT:  .LBB247_311:
+; RV32-NEXT:    bgeu t1, a0, .LBB247_481
+; RV32-NEXT:  .LBB247_312:
+; RV32-NEXT:    bne a1, s0, .LBB247_482
+; RV32-NEXT:  .LBB247_313:
+; RV32-NEXT:    bgeu s0, a1, .LBB247_483
+; RV32-NEXT:  .LBB247_314:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a5, a1, .LBB247_484
+; RV32-NEXT:  .LBB247_315:
+; RV32-NEXT:    bgeu t5, a0, .LBB247_485
+; RV32-NEXT:  .LBB247_316:
+; RV32-NEXT:    beq a1, a5, .LBB247_318
+; RV32-NEXT:  .LBB247_317:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_318:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a2, a2, a4
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 276(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a5, a1, .LBB247_486
+; RV32-NEXT:  # %bb.319:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a4, a1, .LBB247_487
+; RV32-NEXT:  .LBB247_320:
+; RV32-NEXT:    bgeu a7, a0, .LBB247_488
+; RV32-NEXT:  .LBB247_321:
+; RV32-NEXT:    beq a1, a4, .LBB247_323
+; RV32-NEXT:  .LBB247_322:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_323:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw s1, 296(a2) # 4-byte Folded Reload
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a2, a2, a5
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a5, 292(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a4, a1, .LBB247_325
+; RV32-NEXT:  # %bb.324:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:  .LBB247_325:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a4, a4, a7
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    lw a4, 280(a4) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a3, a1, .LBB247_489
+; RV32-NEXT:  # %bb.326:
+; RV32-NEXT:    bgeu a4, a0, .LBB247_490
+; RV32-NEXT:  .LBB247_327:
+; RV32-NEXT:    bne a1, a3, .LBB247_491
+; RV32-NEXT:  .LBB247_328:
+; RV32-NEXT:    bltu a3, a1, .LBB247_330
+; RV32-NEXT:  .LBB247_329:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB247_330:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a4, zero, 232
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a4, 288(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a5, a1, .LBB247_332
+; RV32-NEXT:  # %bb.331:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:  .LBB247_332:
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a7, zero, 232
+; RV32-NEXT:    mul a3, a3, a7
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw s0, 316(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a4, a0, .LBB247_334
+; RV32-NEXT:  # %bb.333:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:  .LBB247_334:
+; RV32-NEXT:    beq a1, a5, .LBB247_336
+; RV32-NEXT:  # %bb.335:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_336:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a4, 320(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bltu a5, a1, .LBB247_338
+; RV32-NEXT:  # %bb.337:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB247_338:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 300(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a5, a1, .LBB247_492
+; RV32-NEXT:  # %bb.339:
+; RV32-NEXT:    bgeu s1, a0, .LBB247_493
+; RV32-NEXT:  .LBB247_340:
+; RV32-NEXT:    bne a1, a5, .LBB247_494
+; RV32-NEXT:  .LBB247_341:
+; RV32-NEXT:    bltu a5, a1, .LBB247_343
+; RV32-NEXT:  .LBB247_342:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:  .LBB247_343:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    addi a5, zero, 232
+; RV32-NEXT:    mul a3, a3, a5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    lw a5, 308(a3) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a5, a1, .LBB247_495
+; RV32-NEXT:  # %bb.344:
+; RV32-NEXT:    bgeu a6, a0, .LBB247_496
+; RV32-NEXT:  .LBB247_345:
+; RV32-NEXT:    beq a1, a5, .LBB247_347
+; RV32-NEXT:  .LBB247_346:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:  .LBB247_347:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    lw a3, 324(a2) # 4-byte Folded Reload
+; RV32-NEXT:    bgeu a5, a1, .LBB247_497
+; RV32-NEXT:  # %bb.348:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s0, a1, .LBB247_498
+; RV32-NEXT:  .LBB247_349:
+; RV32-NEXT:    bgeu t0, a0, .LBB247_499
+; RV32-NEXT:  .LBB247_350:
+; RV32-NEXT:    bne a1, s0, .LBB247_500
+; RV32-NEXT:  .LBB247_351:
+; RV32-NEXT:    bgeu s0, a1, .LBB247_501
+; RV32-NEXT:  .LBB247_352:
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a3, a1, .LBB247_502
+; RV32-NEXT:  .LBB247_353:
+; RV32-NEXT:    bgeu a4, a0, .LBB247_503
+; RV32-NEXT:  .LBB247_354:
+; RV32-NEXT:    bne a1, a3, .LBB247_504
+; RV32-NEXT:  .LBB247_355:
+; RV32-NEXT:    bltu a3, a1, .LBB247_357
+; RV32-NEXT:  .LBB247_356:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:  .LBB247_357:
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    addi a3, zero, 232
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add sp, sp, a2
+; RV32-NEXT:    lw s11, 428(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s10, 432(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s9, 436(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s8, 440(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s7, 444(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s6, 448(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s5, 452(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s4, 456(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s3, 460(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s2, 464(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s1, 468(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw s0, 472(sp) # 4-byte Folded Reload
+; RV32-NEXT:    lw ra, 476(sp) # 4-byte Folded Reload
+; RV32-NEXT:    addi sp, sp, 480
+; RV32-NEXT:    ret
+; RV32-NEXT:  .LBB247_358:
+; RV32-NEXT:    mv s4, t6
+; RV32-NEXT:    bgeu t6, a0, .LBB247_359
+; RV32-NEXT:    j .LBB247_2
+; RV32-NEXT:  .LBB247_359:
+; RV32-NEXT:    mv a0, t6
+; RV32-NEXT:    bne a1, a3, .LBB247_360
+; RV32-NEXT:    j .LBB247_3
+; RV32-NEXT:  .LBB247_360:
+; RV32-NEXT:    mv a0, s4
+; RV32-NEXT:    bgeu a3, a1, .LBB247_361
+; RV32-NEXT:    j .LBB247_4
+; RV32-NEXT:  .LBB247_361:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:    mv a3, a0
+; RV32-NEXT:    bgeu a4, a1, .LBB247_362
+; RV32-NEXT:    j .LBB247_5
+; RV32-NEXT:  .LBB247_362:
+; RV32-NEXT:    mv a3, t4
+; RV32-NEXT:    bgeu t4, a0, .LBB247_363
+; RV32-NEXT:    j .LBB247_6
+; RV32-NEXT:  .LBB247_363:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:    beq a1, a4, .LBB247_505
+; RV32-NEXT:    j .LBB247_7
+; RV32-NEXT:  .LBB247_505:
+; RV32-NEXT:    j .LBB247_8
+; RV32-NEXT:  .LBB247_364:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bgeu s10, a1, .LBB247_365
+; RV32-NEXT:    j .LBB247_10
+; RV32-NEXT:  .LBB247_365:
+; RV32-NEXT:    mv a4, a7
+; RV32-NEXT:    bgeu a7, a0, .LBB247_366
+; RV32-NEXT:    j .LBB247_11
+; RV32-NEXT:  .LBB247_366:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:    bne a1, s10, .LBB247_367
+; RV32-NEXT:    j .LBB247_12
+; RV32-NEXT:  .LBB247_367:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bgeu s10, a1, .LBB247_368
+; RV32-NEXT:    j .LBB247_13
+; RV32-NEXT:  .LBB247_368:
+; RV32-NEXT:    mv a1, s10
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu s3, a1, .LBB247_506
+; RV32-NEXT:    j .LBB247_14
+; RV32-NEXT:  .LBB247_506:
+; RV32-NEXT:    j .LBB247_15
+; RV32-NEXT:  .LBB247_369:
+; RV32-NEXT:    mv a0, s6
+; RV32-NEXT:    bne a1, s3, .LBB247_370
+; RV32-NEXT:    j .LBB247_17
+; RV32-NEXT:  .LBB247_370:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bgeu s3, a1, .LBB247_371
+; RV32-NEXT:    j .LBB247_18
+; RV32-NEXT:  .LBB247_371:
+; RV32-NEXT:    mv a1, s3
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bgeu a5, a1, .LBB247_372
+; RV32-NEXT:    j .LBB247_19
+; RV32-NEXT:  .LBB247_372:
+; RV32-NEXT:    mv a4, t5
+; RV32-NEXT:    bgeu t5, a0, .LBB247_373
+; RV32-NEXT:    j .LBB247_20
+; RV32-NEXT:  .LBB247_373:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    bne a1, a5, .LBB247_374
+; RV32-NEXT:    j .LBB247_21
+; RV32-NEXT:  .LBB247_374:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bgeu a5, a1, .LBB247_375
+; RV32-NEXT:    j .LBB247_22
+; RV32-NEXT:  .LBB247_375:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bgeu t1, a1, .LBB247_376
+; RV32-NEXT:    j .LBB247_23
+; RV32-NEXT:  .LBB247_376:
+; RV32-NEXT:    mv a4, s5
+; RV32-NEXT:    bgeu s5, a0, .LBB247_377
+; RV32-NEXT:    j .LBB247_24
+; RV32-NEXT:  .LBB247_377:
+; RV32-NEXT:    mv a0, s5
+; RV32-NEXT:    beq a1, t1, .LBB247_507
+; RV32-NEXT:    j .LBB247_25
+; RV32-NEXT:  .LBB247_507:
+; RV32-NEXT:    j .LBB247_26
+; RV32-NEXT:  .LBB247_378:
+; RV32-NEXT:    mv a0, ra
+; RV32-NEXT:    bne a1, s1, .LBB247_379
+; RV32-NEXT:    j .LBB247_32
+; RV32-NEXT:  .LBB247_379:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bgeu s1, a1, .LBB247_380
+; RV32-NEXT:    j .LBB247_33
+; RV32-NEXT:  .LBB247_380:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu s11, a1, .LBB247_508
+; RV32-NEXT:    j .LBB247_34
+; RV32-NEXT:  .LBB247_508:
+; RV32-NEXT:    j .LBB247_35
+; RV32-NEXT:  .LBB247_381:
+; RV32-NEXT:    mv a0, s2
+; RV32-NEXT:    bne a1, t3, .LBB247_382
+; RV32-NEXT:    j .LBB247_45
+; RV32-NEXT:  .LBB247_382:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bltu t3, a1, .LBB247_509
+; RV32-NEXT:    j .LBB247_46
+; RV32-NEXT:  .LBB247_509:
+; RV32-NEXT:    j .LBB247_47
+; RV32-NEXT:  .LBB247_383:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, a6, .LBB247_384
+; RV32-NEXT:    j .LBB247_59
+; RV32-NEXT:  .LBB247_384:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    bgeu a6, a1, .LBB247_385
+; RV32-NEXT:    j .LBB247_60
+; RV32-NEXT:  .LBB247_385:
+; RV32-NEXT:    mv a1, a6
+; RV32-NEXT:    mv a4, a0
+; RV32-NEXT:    bltu t2, a1, .LBB247_510
+; RV32-NEXT:    j .LBB247_61
+; RV32-NEXT:  .LBB247_510:
+; RV32-NEXT:    j .LBB247_62
+; RV32-NEXT:  .LBB247_386:
+; RV32-NEXT:    mv a0, s1
+; RV32-NEXT:    bne a1, t2, .LBB247_387
+; RV32-NEXT:    j .LBB247_64
+; RV32-NEXT:  .LBB247_387:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bltu t2, a1, .LBB247_511
+; RV32-NEXT:    j .LBB247_65
+; RV32-NEXT:  .LBB247_511:
+; RV32-NEXT:    j .LBB247_66
+; RV32-NEXT:  .LBB247_388:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bgeu a5, a0, .LBB247_389
+; RV32-NEXT:    j .LBB247_68
+; RV32-NEXT:  .LBB247_389:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bne a1, s1, .LBB247_390
+; RV32-NEXT:    j .LBB247_69
+; RV32-NEXT:  .LBB247_390:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu s1, a1, .LBB247_391
+; RV32-NEXT:    j .LBB247_70
+; RV32-NEXT:  .LBB247_391:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a5, a0
+; RV32-NEXT:    bltu s0, a1, .LBB247_512
+; RV32-NEXT:    j .LBB247_71
+; RV32-NEXT:  .LBB247_512:
+; RV32-NEXT:    j .LBB247_72
+; RV32-NEXT:  .LBB247_392:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, s0, .LBB247_393
+; RV32-NEXT:    j .LBB247_74
+; RV32-NEXT:  .LBB247_393:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bgeu s0, a1, .LBB247_394
+; RV32-NEXT:    j .LBB247_75
+; RV32-NEXT:  .LBB247_394:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv s0, a0
+; RV32-NEXT:    bltu s1, a1, .LBB247_513
+; RV32-NEXT:    j .LBB247_76
+; RV32-NEXT:  .LBB247_513:
+; RV32-NEXT:    j .LBB247_77
+; RV32-NEXT:  .LBB247_395:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    beq a1, s1, .LBB247_79
+; RV32-NEXT:  .LBB247_396:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    bltu s1, a1, .LBB247_80
+; RV32-NEXT:  .LBB247_397:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu t5, a1, .LBB247_81
+; RV32-NEXT:    j .LBB247_82
+; RV32-NEXT:  .LBB247_398:
+; RV32-NEXT:    mv a0, s9
+; RV32-NEXT:    beq a1, t5, .LBB247_84
+; RV32-NEXT:  .LBB247_399:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu t5, a1, .LBB247_85
+; RV32-NEXT:    j .LBB247_86
+; RV32-NEXT:  .LBB247_400:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a5, a1, .LBB247_94
+; RV32-NEXT:  .LBB247_401:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a3, a0, .LBB247_95
+; RV32-NEXT:  .LBB247_402:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    bne a1, a5, .LBB247_96
+; RV32-NEXT:    j .LBB247_97
+; RV32-NEXT:  .LBB247_403:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a4, a1, .LBB247_99
+; RV32-NEXT:  .LBB247_404:
+; RV32-NEXT:    mv a2, a6
+; RV32-NEXT:    bltu a6, a0, .LBB247_100
+; RV32-NEXT:  .LBB247_405:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, a4, .LBB247_101
+; RV32-NEXT:    j .LBB247_102
+; RV32-NEXT:  .LBB247_406:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s2, a1, .LBB247_104
+; RV32-NEXT:  .LBB247_407:
+; RV32-NEXT:    mv a2, s7
+; RV32-NEXT:    bltu s7, a0, .LBB247_105
+; RV32-NEXT:  .LBB247_408:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, s2, .LBB247_106
+; RV32-NEXT:  .LBB247_409:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu s2, a1, .LBB247_107
+; RV32-NEXT:  .LBB247_410:
+; RV32-NEXT:    mv a1, s2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s6, a1, .LBB247_108
+; RV32-NEXT:    j .LBB247_109
+; RV32-NEXT:  .LBB247_411:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, s6, .LBB247_111
+; RV32-NEXT:  .LBB247_412:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu s6, a1, .LBB247_112
+; RV32-NEXT:    j .LBB247_113
+; RV32-NEXT:  .LBB247_413:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s3, a1, .LBB247_121
+; RV32-NEXT:  .LBB247_414:
+; RV32-NEXT:    mv a2, s11
+; RV32-NEXT:    bltu s11, a0, .LBB247_122
+; RV32-NEXT:  .LBB247_415:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, s3, .LBB247_123
+; RV32-NEXT:  .LBB247_416:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu s3, a1, .LBB247_124
+; RV32-NEXT:  .LBB247_417:
+; RV32-NEXT:    mv a1, s3
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s10, a1, .LBB247_125
+; RV32-NEXT:    j .LBB247_126
+; RV32-NEXT:  .LBB247_418:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, s10, .LBB247_128
+; RV32-NEXT:  .LBB247_419:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu s10, a1, .LBB247_129
+; RV32-NEXT:  .LBB247_420:
+; RV32-NEXT:    mv a1, s10
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu t1, a1, .LBB247_130
+; RV32-NEXT:    j .LBB247_131
+; RV32-NEXT:  .LBB247_421:
+; RV32-NEXT:    mv a0, s10
+; RV32-NEXT:    beq a1, t1, .LBB247_133
+; RV32-NEXT:  .LBB247_422:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu t1, a1, .LBB247_134
+; RV32-NEXT:  .LBB247_423:
+; RV32-NEXT:    mv a1, t1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s5, a1, .LBB247_135
+; RV32-NEXT:    j .LBB247_136
+; RV32-NEXT:  .LBB247_424:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:    beq a1, s5, .LBB247_138
+; RV32-NEXT:  .LBB247_425:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    mv t1, t2
+; RV32-NEXT:    bltu s5, a1, .LBB247_139
+; RV32-NEXT:  .LBB247_426:
+; RV32-NEXT:    mv a1, s5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu t6, a1, .LBB247_140
+; RV32-NEXT:    j .LBB247_141
+; RV32-NEXT:  .LBB247_427:
+; RV32-NEXT:    mv a0, s5
+; RV32-NEXT:    beq a1, t6, .LBB247_143
+; RV32-NEXT:  .LBB247_428:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu t6, a1, .LBB247_144
+; RV32-NEXT:    j .LBB247_145
+; RV32-NEXT:  .LBB247_429:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s8, a1, .LBB247_153
+; RV32-NEXT:  .LBB247_430:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a3, a0, .LBB247_154
+; RV32-NEXT:  .LBB247_431:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    beq a1, s8, .LBB247_155
+; RV32-NEXT:  .LBB247_432:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu s8, a1, .LBB247_156
+; RV32-NEXT:  .LBB247_433:
+; RV32-NEXT:    mv a1, s8
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu t3, a1, .LBB247_157
+; RV32-NEXT:    j .LBB247_158
+; RV32-NEXT:  .LBB247_434:
+; RV32-NEXT:    mv a0, s9
+; RV32-NEXT:    beq a1, t3, .LBB247_160
+; RV32-NEXT:  .LBB247_435:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu t3, a1, .LBB247_161
+; RV32-NEXT:    j .LBB247_162
+; RV32-NEXT:  .LBB247_436:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a7, a1, .LBB247_170
+; RV32-NEXT:  .LBB247_437:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bltu a5, a0, .LBB247_171
+; RV32-NEXT:  .LBB247_438:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    bne a1, a7, .LBB247_172
+; RV32-NEXT:    j .LBB247_173
+; RV32-NEXT:  .LBB247_439:
+; RV32-NEXT:    mv a0, s8
+; RV32-NEXT:    beq a1, s1, .LBB247_179
+; RV32-NEXT:  .LBB247_440:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu s1, a1, .LBB247_180
+; RV32-NEXT:  .LBB247_441:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu s4, a1, .LBB247_181
+; RV32-NEXT:    j .LBB247_182
+; RV32-NEXT:  .LBB247_442:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    beq a1, s4, .LBB247_184
+; RV32-NEXT:  .LBB247_443:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu s4, a1, .LBB247_185
+; RV32-NEXT:    j .LBB247_186
+; RV32-NEXT:  .LBB247_444:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu t0, a1, .LBB247_194
+; RV32-NEXT:  .LBB247_445:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a4, a0, .LBB247_195
+; RV32-NEXT:  .LBB247_446:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, t0, .LBB247_196
+; RV32-NEXT:    j .LBB247_197
+; RV32-NEXT:  .LBB247_447:
+; RV32-NEXT:    mv a0, s4
+; RV32-NEXT:    beq a1, a5, .LBB247_211
+; RV32-NEXT:  .LBB247_448:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a5, a1, .LBB247_212
+; RV32-NEXT:    j .LBB247_213
+; RV32-NEXT:  .LBB247_449:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu ra, a1, .LBB247_221
+; RV32-NEXT:  .LBB247_450:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a4, a0, .LBB247_222
+; RV32-NEXT:  .LBB247_451:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    bne a1, ra, .LBB247_223
+; RV32-NEXT:    j .LBB247_224
+; RV32-NEXT:  .LBB247_452:
+; RV32-NEXT:    mv a0, s7
+; RV32-NEXT:    beq a1, a4, .LBB247_230
+; RV32-NEXT:  .LBB247_453:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a4, a1, .LBB247_231
+; RV32-NEXT:    j .LBB247_232
+; RV32-NEXT:  .LBB247_454:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a3, a1, .LBB247_240
+; RV32-NEXT:  .LBB247_455:
+; RV32-NEXT:    mv a2, s0
+; RV32-NEXT:    bltu s0, a0, .LBB247_241
+; RV32-NEXT:  .LBB247_456:
+; RV32-NEXT:    mv a0, s0
+; RV32-NEXT:    beq a1, a3, .LBB247_242
+; RV32-NEXT:  .LBB247_457:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a3, a1, .LBB247_243
+; RV32-NEXT:    j .LBB247_244
+; RV32-NEXT:  .LBB247_458:
+; RV32-NEXT:    mv a0, s11
+; RV32-NEXT:    beq a1, a3, .LBB247_248
+; RV32-NEXT:  .LBB247_459:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a3, a1, .LBB247_249
+; RV32-NEXT:    j .LBB247_250
+; RV32-NEXT:  .LBB247_460:
+; RV32-NEXT:    mv a2, t0
+; RV32-NEXT:    bltu t0, a0, .LBB247_260
+; RV32-NEXT:  .LBB247_461:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, s1, .LBB247_261
+; RV32-NEXT:  .LBB247_462:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu s1, a1, .LBB247_262
+; RV32-NEXT:  .LBB247_463:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu t1, a1, .LBB247_263
+; RV32-NEXT:    j .LBB247_264
+; RV32-NEXT:  .LBB247_464:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, t1, .LBB247_266
+; RV32-NEXT:  .LBB247_465:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu t1, a1, .LBB247_267
+; RV32-NEXT:    j .LBB247_268
+; RV32-NEXT:  .LBB247_466:
+; RV32-NEXT:    mv a1, a2
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a7, a1, .LBB247_276
+; RV32-NEXT:  .LBB247_467:
+; RV32-NEXT:    mv a2, a5
+; RV32-NEXT:    bltu a5, a0, .LBB247_277
+; RV32-NEXT:  .LBB247_468:
+; RV32-NEXT:    mv a0, a5
+; RV32-NEXT:    beq a1, a7, .LBB247_278
+; RV32-NEXT:  .LBB247_469:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a7, a1, .LBB247_279
+; RV32-NEXT:  .LBB247_470:
+; RV32-NEXT:    mv a1, a7
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a4, a1, .LBB247_280
+; RV32-NEXT:    j .LBB247_281
+; RV32-NEXT:  .LBB247_471:
+; RV32-NEXT:    mv a1, a3
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s1, a1, .LBB247_295
+; RV32-NEXT:  .LBB247_472:
+; RV32-NEXT:    mv a2, t4
+; RV32-NEXT:    bltu t4, a0, .LBB247_296
+; RV32-NEXT:  .LBB247_473:
+; RV32-NEXT:    mv a0, t4
+; RV32-NEXT:    bne a1, s1, .LBB247_297
+; RV32-NEXT:    j .LBB247_298
+; RV32-NEXT:  .LBB247_474:
+; RV32-NEXT:    mv a1, s1
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a6, a1, .LBB247_300
+; RV32-NEXT:  .LBB247_475:
+; RV32-NEXT:    mv a2, a3
+; RV32-NEXT:    bltu a3, a0, .LBB247_301
+; RV32-NEXT:  .LBB247_476:
+; RV32-NEXT:    mv a0, a3
+; RV32-NEXT:    beq a1, a6, .LBB247_302
+; RV32-NEXT:  .LBB247_477:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu a6, a1, .LBB247_303
+; RV32-NEXT:  .LBB247_478:
+; RV32-NEXT:    mv a1, a6
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bgeu a4, a1, .LBB247_304
+; RV32-NEXT:    j .LBB247_305
+; RV32-NEXT:  .LBB247_479:
+; RV32-NEXT:    mv a1, a4
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s0, a1, .LBB247_311
+; RV32-NEXT:  .LBB247_480:
+; RV32-NEXT:    mv a2, t1
+; RV32-NEXT:    bltu t1, a0, .LBB247_312
+; RV32-NEXT:  .LBB247_481:
+; RV32-NEXT:    mv a0, t1
+; RV32-NEXT:    beq a1, s0, .LBB247_313
+; RV32-NEXT:  .LBB247_482:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu s0, a1, .LBB247_314
+; RV32-NEXT:  .LBB247_483:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a5, a1, .LBB247_315
+; RV32-NEXT:  .LBB247_484:
+; RV32-NEXT:    mv a2, t5
+; RV32-NEXT:    bltu t5, a0, .LBB247_316
+; RV32-NEXT:  .LBB247_485:
+; RV32-NEXT:    mv a0, t5
+; RV32-NEXT:    bne a1, a5, .LBB247_317
+; RV32-NEXT:    j .LBB247_318
+; RV32-NEXT:  .LBB247_486:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a4, a1, .LBB247_320
+; RV32-NEXT:  .LBB247_487:
+; RV32-NEXT:    mv a2, a7
+; RV32-NEXT:    bltu a7, a0, .LBB247_321
+; RV32-NEXT:  .LBB247_488:
+; RV32-NEXT:    mv a0, a7
+; RV32-NEXT:    bne a1, a4, .LBB247_322
+; RV32-NEXT:    j .LBB247_323
+; RV32-NEXT:  .LBB247_489:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a4, a0, .LBB247_327
+; RV32-NEXT:  .LBB247_490:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    beq a1, a3, .LBB247_328
+; RV32-NEXT:  .LBB247_491:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a3, a1, .LBB247_329
+; RV32-NEXT:    j .LBB247_330
+; RV32-NEXT:  .LBB247_492:
+; RV32-NEXT:    mv a2, s1
+; RV32-NEXT:    bltu s1, a0, .LBB247_340
+; RV32-NEXT:  .LBB247_493:
+; RV32-NEXT:    mv a0, s1
+; RV32-NEXT:    beq a1, a5, .LBB247_341
+; RV32-NEXT:  .LBB247_494:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a5, a1, .LBB247_342
+; RV32-NEXT:    j .LBB247_343
+; RV32-NEXT:  .LBB247_495:
+; RV32-NEXT:    mv a2, a6
+; RV32-NEXT:    bltu a6, a0, .LBB247_345
+; RV32-NEXT:  .LBB247_496:
+; RV32-NEXT:    mv a0, a6
+; RV32-NEXT:    bne a1, a5, .LBB247_346
+; RV32-NEXT:    j .LBB247_347
+; RV32-NEXT:  .LBB247_497:
+; RV32-NEXT:    mv a1, a5
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu s0, a1, .LBB247_349
+; RV32-NEXT:  .LBB247_498:
+; RV32-NEXT:    mv a2, t0
+; RV32-NEXT:    bltu t0, a0, .LBB247_350
+; RV32-NEXT:  .LBB247_499:
+; RV32-NEXT:    mv a0, t0
+; RV32-NEXT:    beq a1, s0, .LBB247_351
+; RV32-NEXT:  .LBB247_500:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bltu s0, a1, .LBB247_352
+; RV32-NEXT:  .LBB247_501:
+; RV32-NEXT:    mv a1, s0
+; RV32-NEXT:    mv a2, a0
+; RV32-NEXT:    bltu a3, a1, .LBB247_353
+; RV32-NEXT:  .LBB247_502:
+; RV32-NEXT:    mv a2, a4
+; RV32-NEXT:    bltu a4, a0, .LBB247_354
+; RV32-NEXT:  .LBB247_503:
+; RV32-NEXT:    mv a0, a4
+; RV32-NEXT:    beq a1, a3, .LBB247_355
+; RV32-NEXT:  .LBB247_504:
+; RV32-NEXT:    mv a0, a2
+; RV32-NEXT:    bgeu a3, a1, .LBB247_356
+; RV32-NEXT:    j .LBB247_357
+;
+; RV64-LABEL: vreduce_umax_v64i64:
+; RV64:       # %bb.0:
+; RV64-NEXT:    vsetivli a1, 16, e64,m8,ta,mu
+; RV64-NEXT:    vle64.v v8, (a0)
+; RV64-NEXT:    addi a1, a0, 384
+; RV64-NEXT:    vle64.v v16, (a1)
+; RV64-NEXT:    addi a1, a0, 256
+; RV64-NEXT:    addi a0, a0, 128
+; RV64-NEXT:    vle64.v v24, (a0)
+; RV64-NEXT:    vle64.v v0, (a1)
+; RV64-NEXT:    vmaxu.vv v16, v24, v16
+; RV64-NEXT:    vmaxu.vv v8, v8, v0
+; RV64-NEXT:    vmaxu.vv v8, v8, v16
+; RV64-NEXT:    vsetvli a0, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.v.i v25, 0
+; RV64-NEXT:    vsetivli a0, 16, e64,m8,ta,mu
+; RV64-NEXT:    vredmaxu.vs v25, v8, v25
+; RV64-NEXT:    vsetvli zero, zero, e64,m1,ta,mu
+; RV64-NEXT:    vmv.x.s a0, v25
+; RV64-NEXT:    ret
+  %v = load <64 x i64>, <64 x i64>* %x
+  %red = call i64 @llvm.vector.reduce.umax.v64i64(<64 x i64> %v)
+  ret i64 %red
+}