[RISCV] Support mask policy for RVV IR intrinsics.

[khchen]

Summary

Add the IsValidMaskPolicy flag to indicate the operations
result would be effected by the mask policy. (ex. mask operations).

It means RISCVInsertVSETVLI should decide the mask policy according
by mask policy operand or passthru operand.
If IsValidMaskPolicy is false (ex. nomask, store, reduction operations.),
the mask policy could be either mask undisturbed or agnostic.
Currently, RISCVInsertVSETVLI sets ValidMaskPolicy operation default to
use MA, otherwise to MU to keep the current mask policy would not changed
for nomask operstions.

Add masked-tama, masked-tamu, masked-tuma and masked-tumu test cases.
I didn't add all operations because most of implementations are using
the same pseudo multiclass. Some test maybe be duplicated in different
tests. (ex. masked vmacc with tumu shows in vmacc-rv32.ll and masked-tumu.)
I think having different tests only for policy would make the test
clear.

[craig.topper]

llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp

510

We don't need a ternary operator here. IsValidMaskPolicy is a bool. You can do bool MaskAgnostic = IsValidMaskPolicy

llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td

1155

VPseudoBinaryMask appears to only be used by VPseudoTernary which is used by reductions. But reduction shouldn't have IsValidMaskPolicy?

[craig.topper]

llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h

96

I wonder if "UsesMaskPolicy" might be better?

[khchen]

Address Craig's comments, thanks!

[khchen]

llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td

1155

Yes, I missed it again, thanks!

[khchen]

update comment and refine code.

[khchen]

ping.

[khchen]

ping.

[craig.topper]

llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h

93

nomask-> unmasked to keep with the terminology in the spec

Comma after "store"

120

Is 0 now TAIL_UNDISTURBED_MASK_UNDISTURBED? And 1 TAIL_AGNOSTIC_MASK_UNDISTURBED, etc. Or maybe we just remove TAIL_UNDISTURBED?

llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp

512

Is this a FIXME suggestion?

llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td

45

Is this used? These names are also now ambiguous and tablegen can't easily or them

[khchen]

Address Craig's comments. Thanks!

[craig.topper]

LGTM other than those last comments.

llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp

512

clang-format

516–517

use uint64_t so the assert won't allow negative numbers?

[arcbbb]

llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h

120

Sorry I didn't notice this change.
I saw there are two places using 0 as TAIL_UNDISTURBED.
https://github.com/llvm/llvm-project/blob/main/llvm/lib/Target/RISCV/RISCVInstrInfoVVLPatterns.td
It looks like policy is a 2-bit value representing TU_MU, TA_MU, TU_MA, and TA_MA as below

TU_MU	0
TA_MU	1
TU_MA	2
TA_MA	3

Maybe we still need the TU definition in enum.

[khchen]

llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h

120

Do you mean in order to make code more consistent (enum in RISCVBaseInfo.h and TAIL_UNDISTURBED defined in td) we need to add TAIL_UNDISTURBED back to enum?
I think it's ok to me, but we should also update them as explicitly name like TU_MU or TAIL_UNDISTURBED_MASK_UNDISTURBED since we start model mask policy now?

[arcbbb]

llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h

120

I got it. I thought 0 was undefined at my first glance on the enum.
I think it's fine to leave a comment here.

llvm/include/llvm/IR/IntrinsicsRISCV.td

class RISCVUSLoadFF
         : Intrinsic<[llvm_anyvector_ty, llvm_anyint_ty],
                     [LLVMMatchType<0>,
                      LLVMPointerType<LLVMMatchType<0>>, LLVMMatchType<1>],
                     [NoCapture<ArgIndex<1>>]>,
                     RISCVVIntrinsic {
     let VLOperand = 2;
   }
   // For unit stride load with mask
-  // Input: (maskedoff, pointer, mask, vl, ta)
+  // Input: (maskedoff, pointer, mask, vl, policy)
   class RISCVUSLoadMask
         : Intrinsic<[llvm_anyvector_ty ],
                     [LLVMMatchType<0>,
                      LLVMPointerType<LLVMMatchType<0>>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                      llvm_anyint_ty, LLVMMatchType<1>],
                     [NoCapture<ArgIndex<1>>, ImmArg<ArgIndex<4>>, IntrReadMem]>,
                     RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For unit stride fault-only-first load with mask
-  // Input: (maskedoff, pointer, mask, vl, ta)
+  // Input: (maskedoff, pointer, mask, vl, policy)
   // Output: (data, vl)
   // NOTE: We model this with default memory properties since we model writing
   // VL as a side effect. IntrReadMem, IntrHasSideEffects does not work.
   class RISCVUSLoadFFMask
         : Intrinsic<[llvm_anyvector_ty, llvm_anyint_ty],
                     [LLVMMatchType<0>,
                      LLVMPointerType<LLVMMatchType<0>>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                      LLVMMatchType<1>, LLVMMatchType<1>],
                     [NoCapture<ArgIndex<1>>, ImmArg<ArgIndex<4>>]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For strided load with passthru operand
   // Input: (passthru, pointer, stride, vl)
   class RISCVSLoad
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>,
                      LLVMPointerType<LLVMMatchType<0>>,
                      llvm_anyint_ty, LLVMMatchType<1>],
                     [NoCapture<ArgIndex<1>>, IntrReadMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For strided load with mask
-  // Input: (maskedoff, pointer, stride, mask, vl, ta)
+  // Input: (maskedoff, pointer, stride, mask, vl, policy)
   class RISCVSLoadMask
         : Intrinsic<[llvm_anyvector_ty ],
                     [LLVMMatchType<0>,
                      LLVMPointerType<LLVMMatchType<0>>, llvm_anyint_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, LLVMMatchType<1>,
                      LLVMMatchType<1>],
                     [NoCapture<ArgIndex<1>>, ImmArg<ArgIndex<5>>, IntrReadMem]>,
                     RISCVVIntrinsic {
     let VLOperand = 4;
   }
   // For indexed load with passthru operand
   // Input: (passthru, pointer, index, vl)
   class RISCVILoad
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>,
                      LLVMPointerType<LLVMMatchType<0>>,
                      llvm_anyvector_ty, llvm_anyint_ty],
                     [NoCapture<ArgIndex<1>>, IntrReadMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For indexed load with mask
-  // Input: (maskedoff, pointer, index, mask, vl, ta)
+  // Input: (maskedoff, pointer, index, mask, vl, policy)
   class RISCVILoadMask
         : Intrinsic<[llvm_anyvector_ty ],
                     [LLVMMatchType<0>,
                      LLVMPointerType<LLVMMatchType<0>>, llvm_anyvector_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<2>],
                     [NoCapture<ArgIndex<1>>, ImmArg<ArgIndex<5>>, IntrReadMem]>,
                     RISCVVIntrinsic {
let TargetPrefix = "riscv" in {
   // Input: (passthru, vector_in, vl)
   class RISCVUnaryAANoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 2;
   }
   // For destination vector type is the same as first source vector (with mask).
-  // Input: (vector_in, mask, vl, ta)
+  // Input: (vector_in, vector_in, mask, vl, policy)
   class RISCVUnaryAAMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<1>],
                     [ImmArg<ArgIndex<4>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
-  class RISCVUnaryAAMaskTU
+  // Input: (passthru, vector_in, vector_in, mask, vl)
+  class RISCVCompress
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For destination vector type is the same as first and second source vector.
   // Input: (vector_in, vector_in, vl)
   class RISCVBinaryAAANoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 2;
   }
   // For destination vector type is the same as first and second source vector.
   // Input: (passthru, vector_in, int_vector_in, vl)
   class RISCVRGatherVVNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>,
                      LLVMVectorOfBitcastsToInt<0>, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For destination vector type is the same as first and second source vector.
-  // Input: (vector_in, vector_in, int_vector_in, vl, ta)
+  // Input: (vector_in, vector_in, int_vector_in, vl, policy)
   class RISCVRGatherVVMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, LLVMVectorOfBitcastsToInt<0>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<1>],
                     [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 4;
   }
   // Input: (passthru, vector_in, int16_vector_in, vl)
   class RISCVRGatherEI16VVNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i16_ty>,
                      llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For destination vector type is the same as first and second source vector.
-  // Input: (vector_in, vector_in, int16_vector_in, vl, ta)
+  // Input: (vector_in, vector_in, int16_vector_in, vl, policy)
   class RISCVRGatherEI16VVMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i16_ty>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<1>],
                     [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 4;
   }
   // For destination vector type is the same as first source vector, and the
   // second operand is XLen.
   // Input: (passthru, vector_in, xlen_in, vl)
   class RISCVGatherVXNoMask
         : Intrinsic<[llvm_anyvector_ty],
-                    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty, 
+                    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty,
                      LLVMMatchType<1>],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For destination vector type is the same as first source vector (with mask).
   // Second operand is XLen.
-  // Input: (maskedoff, vector_in, xlen_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, xlen_in, mask, vl, policy)
   class RISCVGatherVXMask
        : Intrinsic<[llvm_anyvector_ty],
                    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty,
                     LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, LLVMMatchType<1>,
                     LLVMMatchType<1>],
                    [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 4;
   }
   // For destination vector type is the same as first source vector.
   // Input: (passthru, vector_in, vector_in/scalar_in, vl)
   class RISCVBinaryAAXNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                      llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let SplatOperand = 2;
     let VLOperand = 3;
   }
   // For destination vector type is the same as first source vector (with mask).
-  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, policy)
   class RISCVBinaryAAXMask
        : Intrinsic<[llvm_anyvector_ty],
                    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                     LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                     LLVMMatchType<2>],
                    [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let SplatOperand = 2;
     let VLOperand = 4;
   }
   // For destination vector type is the same as first source vector. The
   // second source operand must match the destination type or be an XLen scalar.
   // Input: (passthru, vector_in, vector_in/scalar_in, vl)
   class RISCVBinaryAAShiftNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                      llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For destination vector type is the same as first source vector (with mask).
   // The second source operand must match the destination type or be an XLen scalar.
-  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, policy)
   class RISCVBinaryAAShiftMask
        : Intrinsic<[llvm_anyvector_ty],
                    [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                     LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                     LLVMMatchType<2>],
                    [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 4;
   }
   // For destination vector type is NOT the same as first source vector.
   // Input: (passthru, vector_in, vector_in/scalar_in, vl)
   class RISCVBinaryABXNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_any_ty,
                      llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let SplatOperand = 2;
     let VLOperand = 3;
   }
   // For destination vector type is NOT the same as first source vector (with mask).
-  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, policy)
   class RISCVBinaryABXMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<3>],
                     [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let SplatOperand = 2;
     let VLOperand = 4;
   }
   // For destination vector type is NOT the same as first source vector. The
   // second source operand must match the destination type or be an XLen scalar.
   // Input: (passthru, vector_in, vector_in/scalar_in, vl)
   class RISCVBinaryABShiftNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_any_ty,
                      llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For destination vector type is NOT the same as first source vector (with mask).
   // The second source operand must match the destination type or be an XLen scalar.
-  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, policy)
   class RISCVBinaryABShiftMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<3>],
                     [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 4;
   }
class RISCVSaturatingBinaryAAXNoMask
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                      llvm_anyint_ty],
                     [IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
     let SplatOperand = 2;
     let VLOperand = 3;
   }
   // For Saturating binary operations with mask.
   // The destination vector type is the same as first source vector.
-  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, policy)
   class RISCVSaturatingBinaryAAXMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<2>],
                     [ImmArg<ArgIndex<5>>, IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
     let SplatOperand = 2;
     let VLOperand = 4;
   }
   // For Saturating binary operations.
   // The destination vector type is the same as first source vector.
   // The second source operand matches the destination type or is an XLen scalar.
   // Input: (passthru, vector_in, vector_in/scalar_in, vl)
   class RISCVSaturatingBinaryAAShiftNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                      llvm_anyint_ty],
                     [IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For Saturating binary operations with mask.
   // The destination vector type is the same as first source vector.
   // The second source operand matches the destination type or is an XLen scalar.
-  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, policy)
   class RISCVSaturatingBinaryAAShiftMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<2>],
                     [ImmArg<ArgIndex<5>>, IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
     let VLOperand = 4;
   }
   // For Saturating binary operations.
   // The destination vector type is NOT the same as first source vector.
   // The second source operand matches the destination type or is an XLen scalar.
   // Input: (passthru, vector_in, vector_in/scalar_in, vl)
   class RISCVSaturatingBinaryABShiftNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_any_ty,
                      llvm_anyint_ty],
                     [IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
   // For Saturating binary operations with mask.
   // The destination vector type is NOT the same as first source vector (with mask).
   // The second source operand matches the destination type or is an XLen scalar.
-  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, vector_in/scalar_in, mask, vl, policy)
   class RISCVSaturatingBinaryABShiftMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_any_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<3>],
                     [ImmArg<ArgIndex<5>>, IntrNoMem, IntrHasSideEffects]>, RISCVVIntrinsic {
     let VLOperand = 4;
   }
+  // Input: (vector_in, vector_in, vector_in/scalar_in, vl)
   class RISCVTernaryAAAXNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty,
                      LLVMMatchType<1>],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
+  // Input: (vector_in, vector_in, vector_in/scalar_in, mask, vl, policy)
   class RISCVTernaryAAAXMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, LLVMMatchType<0>, llvm_anyint_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                      LLVMMatchType<1>, LLVMMatchType<1>],
                     [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 4;
   }
+  // NoMask Vector Multiply-Add operations, its first operand can not be undef.
   // Input: (vector_in, vector_in/scalar, vector_in, vl, policy)
   class RISCVTernaryAAXANoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_any_ty, LLVMMatchType<0>,
                      llvm_anyint_ty, LLVMMatchType<2>],
                     [ImmArg<ArgIndex<4>>, IntrNoMem]>, RISCVVIntrinsic {
     let SplatOperand = 1;
     let VLOperand = 3;
   }
+  // Mask Vector Multiply-Add operations, its first operand can not be undef.
   // Input: (vector_in, vector_in/scalar, vector_in, mask, vl, policy
   class RISCVTernaryAAXAMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_any_ty, LLVMMatchType<0>,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                      llvm_anyint_ty, LLVMMatchType<2>],
                     [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let SplatOperand = 1;
     let VLOperand = 4;
   }
+  // NoMask Widening Vector Multiply-Add operations, its first operand can not be undef.
   // Input: (vector_in, vector_in/scalar, vector_in, vl, policy)
   class RISCVTernaryWideNoMask
         : Intrinsic< [llvm_anyvector_ty],
                      [LLVMMatchType<0>, llvm_any_ty, llvm_anyvector_ty,
                       llvm_anyint_ty, LLVMMatchType<3>],
                      [ImmArg<ArgIndex<4>>, IntrNoMem] >, RISCVVIntrinsic {
     let SplatOperand = 1;
     let VLOperand = 3;
   }
+  // Mask Widening Vector Multiply-Add operations, its first operand can not be undef.
   // Input: (vector_in, vector_in/scalar, vector_in, mask, vl, policy
   class RISCVTernaryWideMask
         : Intrinsic< [llvm_anyvector_ty],
                      [LLVMMatchType<0>, llvm_any_ty, llvm_anyvector_ty,
                       LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                       llvm_anyint_ty, LLVMMatchType<3>],
                      [ImmArg<ArgIndex<5>>, IntrNoMem]>, RISCVVIntrinsic {
     let SplatOperand = 1;
let TargetPrefix = "riscv" in {
   // Input: (passthru, vector_in, vl)
   class RISCVUnaryABNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 2;
   }
   // For destination vector type is NOT the same as source vector (with mask).
-  // Input: (maskedoff, vector_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, mask, vl, policy)
   class RISCVUnaryABMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty,
                      LLVMScalarOrSameVectorWidth<1, llvm_i1_ty>,
                      llvm_anyint_ty, LLVMMatchType<2>],
                     [ImmArg<ArgIndex<4>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
let TargetPrefix = "riscv" in {
   // Input: (passthru, vector_in, vl)
   class RISCVConversionNoMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty, llvm_anyint_ty],
                     [IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 2;
   }
   // For Conversion unary operations with mask.
-  // Input: (maskedoff, vector_in, mask, vl, ta)
+  // Input: (maskedoff, vector_in, mask, vl, policy)
   class RISCVConversionMask
         : Intrinsic<[llvm_anyvector_ty],
                     [LLVMMatchType<0>, llvm_anyvector_ty,
                      LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>, llvm_anyint_ty,
                      LLVMMatchType<2>],
                     [ImmArg<ArgIndex<4>>, IntrNoMem]>, RISCVVIntrinsic {
     let VLOperand = 3;
   }
 
   // For unit stride segment load
   // Input: (pointer, vl)
   class RISCVUSSegLoad<int nf>
         : Intrinsic<!listconcat([llvm_anyvector_ty], !listsplat(LLVMMatchType<0>,
                                 !add(nf, -1))),
                     [LLVMPointerToElt<0>, llvm_anyint_ty],
                     [NoCapture<ArgIndex<0>>, IntrReadMem]>, RISCVVIntrinsic {
     let VLOperand = 1;
   }
   // For unit stride segment load with mask
-  // Input: (maskedoff, pointer, mask, vl, ta)
+  // Input: (maskedoff, pointer, mask, vl, policy)
   class RISCVUSSegLoadMask<int nf>
         : Intrinsic<!listconcat([llvm_anyvector_ty], !listsplat(LLVMMatchType<0>,
                                 !add(nf, -1))),
                     !listconcat(!listsplat(LLVMMatchType<0>, nf),
                                 [LLVMPointerToElt<0>,
                                  LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                  llvm_anyint_ty, LLVMMatchType<1>]),
                     [ImmArg<ArgIndex<!add(nf, 3)>>, NoCapture<ArgIndex<nf>>, IntrReadMem]>,
let TargetPrefix = "riscv" in {
   class RISCVUSSegLoadFF<int nf>
         : Intrinsic<!listconcat([llvm_anyvector_ty], !listsplat(LLVMMatchType<0>,
                                 !add(nf, -1)), [llvm_anyint_ty]),
                     [LLVMPointerToElt<0>, LLVMMatchType<1>],
                     [NoCapture<ArgIndex<0>>]>, RISCVVIntrinsic {
     let VLOperand = 1;
   }
   // For unit stride fault-only-first segment load with mask
-  // Input: (maskedoff, pointer, mask, vl, ta)
+  // Input: (maskedoff, pointer, mask, vl, policy)
   // Output: (data, vl)
   // NOTE: We model this with default memory properties since we model writing
   // VL as a side effect. IntrReadMem, IntrHasSideEffects does not work.
   class RISCVUSSegLoadFFMask<int nf>
         : Intrinsic<!listconcat([llvm_anyvector_ty], !listsplat(LLVMMatchType<0>,
                                 !add(nf, -1)), [llvm_anyint_ty]),
                     !listconcat(!listsplat(LLVMMatchType<0>, nf),
                      [LLVMPointerToElt<0>,
let TargetPrefix = "riscv" in {
   class RISCVSSegLoad<int nf>
         : Intrinsic<!listconcat([llvm_anyvector_ty], !listsplat(LLVMMatchType<0>,
                                 !add(nf, -1))),
                     [LLVMPointerToElt<0>, llvm_anyint_ty, LLVMMatchType<1>],
                     [NoCapture<ArgIndex<0>>, IntrReadMem]>, RISCVVIntrinsic {
     let VLOperand = 2;
   }
   // For stride segment load with mask
-  // Input: (maskedoff, pointer, offset, mask, vl, ta)
+  // Input: (maskedoff, pointer, offset, mask, vl, policy)
   class RISCVSSegLoadMask<int nf>
         : Intrinsic<!listconcat([llvm_anyvector_ty], !listsplat(LLVMMatchType<0>,
                                 !add(nf, -1))),
                     !listconcat(!listsplat(LLVMMatchType<0>, nf),
                                 [LLVMPointerToElt<0>,
                                  llvm_anyint_ty,
                                  LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                  LLVMMatchType<1>, LLVMMatchType<1>]),
                     [ImmArg<ArgIndex<!add(nf, 4)>>, NoCapture<ArgIndex<nf>>, IntrReadMem]>,
                     RISCVVIntrinsic {
     let VLOperand = !add(nf, 3);
   }
 
   // For indexed segment load
   // Input: (pointer, index, vl)
   class RISCVISegLoad<int nf>
         : Intrinsic<!listconcat([llvm_anyvector_ty], !listsplat(LLVMMatchType<0>,
                                 !add(nf, -1))),
                     [LLVMPointerToElt<0>, llvm_anyvector_ty, llvm_anyint_ty],
                     [NoCapture<ArgIndex<0>>, IntrReadMem]>, RISCVVIntrinsic {
     let VLOperand = 2;
   }
   // For indexed segment load with mask
-  // Input: (maskedoff, pointer, index, mask, vl, ta)
+  // Input: (maskedoff, pointer, index, mask, vl, policy)
   class RISCVISegLoadMask<int nf>
         : Intrinsic<!listconcat([llvm_anyvector_ty], !listsplat(LLVMMatchType<0>,
                                 !add(nf, -1))),
                     !listconcat(!listsplat(LLVMMatchType<0>, nf),
                                 [LLVMPointerToElt<0>,
                                  llvm_anyvector_ty,
                                  LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                  llvm_anyint_ty, LLVMMatchType<2>]),

   defm vslide1down : RISCVBinaryAAX;
   defm vfslide1up : RISCVBinaryAAX;
   defm vfslide1down : RISCVBinaryAAX;
 
   defm vrgather_vv : RISCVRGatherVV;
   defm vrgather_vx : RISCVRGatherVX;
   defm vrgatherei16_vv : RISCVRGatherEI16VV;
 
-  def "int_riscv_vcompress" : RISCVUnaryAAMaskTU;
+  def "int_riscv_vcompress" : RISCVCompress;
 
   defm vaaddu : RISCVSaturatingBinaryAAX;
   defm vaadd : RISCVSaturatingBinaryAAX;
   defm vasubu : RISCVSaturatingBinaryAAX;
   defm vasub : RISCVSaturatingBinaryAAX;
 
   defm vsmul : RISCVSaturatingBinaryAAX;
 

llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h

enum {
   // explicit operand. Used by RVV Pseudos.
   HasVecPolicyOpShift = HasVLOpShift + 1,
   HasVecPolicyOpMask = 1 << HasVecPolicyOpShift,
 
   // Is this instruction a vector widening reduction instruction. Used by RVV
   // Pseudos.
   IsRVVWideningReductionShift = HasVecPolicyOpShift + 1,
   IsRVVWideningReductionMask = 1 << IsRVVWideningReductionShift,
+
+  // Does this instruction care about mask policy. If it is not, the mask policy
+  // could be either agnostic or undisturbed. For example, nomask, store and
+  // reduction operations result would not be affected by mask policy, so
+  // compiler has free to select either one.
+  IsValidMaskPolicyShift = IsRVVWideningReductionShift + 1,
+  IsValidMaskPolicyMask = 1 << IsValidMaskPolicyShift,
 };
 
 // Match with the definitions in RISCVInstrFormatsV.td
 enum VConstraintType {
   NoConstraint = 0,
   VS2Constraint = 0b001,
   VS1Constraint = 0b010,
   VMConstraint = 0b100,
 };
 
 enum VLMUL : uint8_t {
   LMUL_1 = 0,
   LMUL_2,
   LMUL_4,
   LMUL_8,
   LMUL_RESERVED,
   LMUL_F8,
   LMUL_F4,
   LMUL_F2
 };
 
 enum {
   TAIL_UNDISTURBED = 0,
   TAIL_AGNOSTIC = 1,
+  MASK_AGNOSTIC = 2,
 };
 
 // Helper functions to read TSFlags.
 /// \returns the format of the instruction.
 static inline unsigned getFormat(uint64_t TSFlags) {
   return (TSFlags & InstFormatMask) >> InstFormatShift;
 }
 /// \returns the constraint for the instruction.

 /// \returns true if there is a vector policy operand for this instruction.
 static inline bool hasVecPolicyOp(uint64_t TSFlags) {
   return TSFlags & HasVecPolicyOpMask;
 }
 /// \returns true if it is a vector widening reduction instruction.
 static inline bool isRVVWideningReduction(uint64_t TSFlags) {
   return TSFlags & IsRVVWideningReductionMask;
 }
+/// \returns true if mask policy is valid for the instruction.
+static inline bool IsValidMaskPolicy(uint64_t TSFlags) {
+  return TSFlags & IsValidMaskPolicyMask;
+}
 
 // RISC-V Specific Machine Operand Flags
 enum {
   MO_None = 0,
   MO_CALL = 1,
   MO_PLT = 2,
   MO_LO = 3,
   MO_HI = 4,

llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp

 }
 
 static VSETVLIInfo computeInfoForInstr(const MachineInstr &MI, uint64_t TSFlags,
                                        const MachineRegisterInfo *MRI) {
   VSETVLIInfo InstrInfo;
   unsigned NumOperands = MI.getNumExplicitOperands();
   bool HasPolicy = RISCVII::hasVecPolicyOp(TSFlags);
 
-  // Default to tail agnostic unless the destination is tied to a source.
-  // Unless the source is undef. In that case the user would have some control
-  // over the tail values. Some pseudo instructions force a tail agnostic policy
-  // despite having a tied def.
+  // Default to tail and mask agnostic unless the destination is tied to a
+  // source. Unless the source is undef. In that case the user would have some
+  // control over the policy values. Some pseudo instructions force a tail
+  // agnostic policy despite having a tied def.
   bool ForceTailAgnostic = RISCVII::doesForceTailAgnostic(TSFlags);
+  bool IsValidMaskPolicy = RISCVII::IsValidMaskPolicy(TSFlags);
   bool TailAgnostic = true;
+  // Could we look at above or below instructions to choose the matched
+  // mask policy to reduce vsetvli instructions?
+  // Currently we choose mask undisturbed because usually the mask operations
+  // are mask undisturbed.
+  bool MaskAgnostic = IsValidMaskPolicy ? true : false;
   // If the instruction has policy argument, use the argument.
+  unsigned UseOpIdx;
   if (HasPolicy) {
     const MachineOperand &Op = MI.getOperand(MI.getNumExplicitOperands() - 1);
     TailAgnostic = Op.getImm() & 0x1;
-  }
-
-  unsigned UseOpIdx;
-  if (!(ForceTailAgnostic || (HasPolicy && TailAgnostic)) &&
-      MI.isRegTiedToUseOperand(0, &UseOpIdx)) {
+    MaskAgnostic = Op.getImm() & 0x2;
+  } else if (!ForceTailAgnostic && MI.isRegTiedToUseOperand(0, &UseOpIdx)) {
     TailAgnostic = false;
+    if (IsValidMaskPolicy)
+      MaskAgnostic = false;
     // If the tied operand is an IMPLICIT_DEF we can keep TailAgnostic.
     const MachineOperand &UseMO = MI.getOperand(UseOpIdx);
     MachineInstr *UseMI = MRI->getVRegDef(UseMO.getReg());
     if (UseMI) {
       UseMI = elideCopies(UseMI, MRI);
-      if (UseMI && UseMI->isImplicitDef())
+      if (UseMI && UseMI->isImplicitDef()) {
         TailAgnostic = true;
+        if (IsValidMaskPolicy)
+          MaskAgnostic = true;
+      }
     }
   }
 
   // Remove the tail policy so we can find the SEW and VL.
   if (HasPolicy)
     --NumOperands;
 
   RISCVII::VLMUL VLMul = RISCVII::getLMul(TSFlags);
if (VLOp.isImm()) {
         InstrInfo.setAVLReg(RISCV::X0);
       else
         InstrInfo.setAVLImm(Imm);
     } else {
       InstrInfo.setAVLReg(VLOp.getReg());
     }
   } else
     InstrInfo.setAVLReg(RISCV::NoRegister);
-  InstrInfo.setVTYPE(VLMul, SEW, /*TailAgnostic*/ TailAgnostic,
-                     /*MaskAgnostic*/ false, MaskRegOp, StoreOp, ScalarMovOp);
+  InstrInfo.setVTYPE(VLMul, SEW, TailAgnostic, MaskAgnostic, MaskRegOp, StoreOp,
+                     ScalarMovOp);
 
   return InstrInfo;
 }
 
 void RISCVInsertVSETVLI::insertVSETVLI(MachineBasicBlock &MBB, MachineInstr &MI,
                                        const VSETVLIInfo &Info,
                                        const VSETVLIInfo &PrevInfo) {
   DebugLoc DL = MI.getDebugLoc();

llvm/lib/Target/RISCV/RISCVInstrFormats.td

class RVInst<dag outs, dag ins, string opcodestr, string argstr,
 
   bit HasVLOp = 0;
   let TSFlags{15} = HasVLOp;
 
   bit HasVecPolicyOp = 0;
   let TSFlags{16} = HasVecPolicyOp;
 
   bit IsRVVWideningReduction = 0;
-  let TSFlags{17} =  IsRVVWideningReduction;
+  let TSFlags{17} = IsRVVWideningReduction;
+
+  bit IsValidMaskPolicy = 0;
+  let TSFlags{18} = IsValidMaskPolicy;
 }
 
 // Pseudo instructions
 class Pseudo<dag outs, dag ins, list<dag> pattern, string opcodestr = "", string argstr = "">
     : RVInst<outs, ins, opcodestr, argstr, pattern, InstFormatPseudo> {
   let isPseudo = 1;
   let isCodeGenOnly = 1;
 }

llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td

 
 def DecImm : SDNodeXForm<imm, [{
   return CurDAG->getTargetConstant(N->getSExtValue() - 1, SDLoc(N),
                                    N->getValueType(0));
 }]>;
 
 defvar TAIL_UNDISTURBED = 0;
 defvar TAIL_AGNOSTIC = 1;
+defvar MASK_AGNOSTIC = 2;
 
 //===----------------------------------------------------------------------===//
 // Utilities.
 //===----------------------------------------------------------------------===//
 
 // This class describes information associated to the LMUL.
 class LMULInfo<int lmul, int oct, VReg regclass, VReg wregclass,
                VReg f2regclass, VReg f4regclass, VReg f8regclass, string mx> {
class VPseudoUSLoadMask<VReg RetClass, int EEW, bit isFF> :
   let mayLoad = 1;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = "$rd = $merge";
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
   let HasVecPolicyOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 class VPseudoSLoadNoMask<VReg RetClass, int EEW>:
       Pseudo<(outs RetClass:$rd),
              (ins GPR:$rs1, GPR:$rs2, AVL:$vl, ixlenimm:$sew),[]>,
       RISCVVPseudo,
       RISCVVLE</*Masked*/0, /*TU*/0, /*Strided*/1, /*FF*/0, log2<EEW>.val, VLMul> {
class VPseudoSLoadMask<VReg RetClass, int EEW>:
   let mayLoad = 1;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = "$rd = $merge";
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
   let HasVecPolicyOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 class VPseudoILoadNoMask<VReg RetClass, VReg IdxClass, int EEW, bits<3> LMUL,
                          bit Ordered, bit EarlyClobber>:
       Pseudo<(outs RetClass:$rd),
              (ins GPR:$rs1, IdxClass:$rs2, AVL:$vl,
               ixlenimm:$sew),[]>,
class VPseudoILoadMask<VReg RetClass, VReg IdxClass, int EEW, bits<3> LMUL,
   let mayLoad = 1;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = !if(!eq(EarlyClobber, 1), "@earlyclobber $rd, $rd = $merge", "$rd = $merge");
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
   let HasVecPolicyOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 class VPseudoUSStoreNoMask<VReg StClass, int EEW>:
       Pseudo<(outs),
               (ins StClass:$rd, GPR:$rs1, AVL:$vl, ixlenimm:$sew),[]>,
       RISCVVPseudo,
       RISCVVSE</*Masked*/0, /*Strided*/0, log2<EEW>.val, VLMul> {
class VPseudoNullaryMask<VReg RegClass>:
               ixlenimm:$sew), []>, RISCVVPseudo {
   let mayLoad = 0;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints ="$rd = $merge";
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 // Nullary for pseudo instructions. They are expanded in
 // RISCVExpandPseudoInsts pass.
 class VPseudoNullaryPseudoM<string BaseInst>
        : Pseudo<(outs VR:$rd), (ins AVL:$vl, ixlenimm:$sew), []>,
        RISCVVPseudo {
class VPseudoUnaryMask<VReg RetClass, VReg OpClass, string Constraint = ""> :
         RISCVVPseudo {
   let mayLoad = 0;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = Join<[Constraint, "$rd = $merge"], ",">.ret;
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 class VPseudoUnaryMaskTA<VReg RetClass, VReg OpClass, string Constraint = ""> :
         Pseudo<(outs GetVRegNoV0<RetClass>.R:$rd),
                (ins GetVRegNoV0<RetClass>.R:$merge, OpClass:$rs2,
                     VMaskOp:$vm, AVL:$vl, ixlenimm:$sew, ixlenimm:$policy), []>,
         RISCVVPseudo {
   let mayLoad = 0;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = Join<[Constraint, "$rd = $merge"], ",">.ret;
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
   let HasVecPolicyOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 // mask unary operation without maskedoff
 class VPseudoMaskUnarySOutMask:
         Pseudo<(outs GPR:$rd),
                (ins VR:$rs1, VMaskOp:$vm, AVL:$vl, ixlenimm:$sew), []>,
         RISCVVPseudo {
class VPseudoBinaryMask<VReg RetClass,
         RISCVVPseudo {
   let mayLoad = 0;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = Join<[Constraint, "$rd = $merge"], ",">.ret;
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 class VPseudoBinaryMaskPolicy<VReg RetClass,
                               RegisterClass Op1Class,
                               DAGOperand Op2Class,
                               string Constraint> :
         Pseudo<(outs GetVRegNoV0<RetClass>.R:$rd),
                 (ins GetVRegNoV0<RetClass>.R:$merge,
                      Op1Class:$rs2, Op2Class:$rs1,
                      VMaskOp:$vm, AVL:$vl, ixlenimm:$sew, ixlenimm:$policy), []>,
         RISCVVPseudo {
   let mayLoad = 0;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = Join<[Constraint, "$rd = $merge"], ",">.ret;
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
   let HasVecPolicyOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 // Like VPseudoBinaryMask, but output can be V0.
 class VPseudoBinaryMOutMask<VReg RetClass,
                             RegisterClass Op1Class,
                             DAGOperand Op2Class,
                             string Constraint> :
         Pseudo<(outs RetClass:$rd),
                 (ins RetClass:$merge,
                      Op1Class:$rs2, Op2Class:$rs1,
                      VMaskOp:$vm, AVL:$vl, ixlenimm:$sew), []>,
         RISCVVPseudo {
   let mayLoad = 0;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = Join<[Constraint, "$rd = $merge"], ",">.ret;
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
+  // FIXME: In current design, we would not change the mask policy, so
+  // IsValidMaskPolicy is false. We could fix after add the policy operand.
+  let IsValidMaskPolicy = 0;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 // Special version of VPseudoBinaryMask where we pretend the first source is
 // tied to the destination so we can workaround the earlyclobber constraint.
 // This allows maskedoff and rs2 to be the same register.
 class VPseudoTiedBinaryMask<VReg RetClass,
                             DAGOperand Op2Class,
                             string Constraint> :
         Pseudo<(outs GetVRegNoV0<RetClass>.R:$rd),
                 (ins GetVRegNoV0<RetClass>.R:$merge,
                      Op2Class:$rs1,
                      VMaskOp:$vm, AVL:$vl, ixlenimm:$sew, ixlenimm:$policy), []>,
         RISCVVPseudo {
   let mayLoad = 0;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = Join<[Constraint, "$rd = $merge"], ",">.ret;
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 0; // Merge is also rs2.
   let HasVecPolicyOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 class VPseudoBinaryCarryIn<VReg RetClass,
                            VReg Op1Class,
                            DAGOperand Op2Class,
                            LMULInfo MInfo,
                            bit CarryIn,
class VPseudoUSSegLoadMask<VReg RetClass, int EEW, bits<4> NF, bit isFF>:
   let mayLoad = 1;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = "$rd = $merge";
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
   let HasVecPolicyOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 class VPseudoSSegLoadNoMask<VReg RetClass, int EEW, bits<4> NF>:
       Pseudo<(outs RetClass:$rd),
              (ins GPR:$rs1, GPR:$offset, AVL:$vl, ixlenimm:$sew),[]>,
       RISCVVPseudo,
       RISCVVLSEG<NF, /*Masked*/0, /*Strided*/1, /*FF*/0, log2<EEW>.val, VLMul> {
class VPseudoSSegLoadMask<VReg RetClass, int EEW, bits<4> NF>:
   let mayLoad = 1;
   let mayStore = 0;
   let hasSideEffects = 0;
   let Constraints = "$rd = $merge";
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
   let HasVecPolicyOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 class VPseudoISegLoadNoMask<VReg RetClass, VReg IdxClass, int EEW, bits<3> LMUL,
                             bits<4> NF, bit Ordered>:
       Pseudo<(outs RetClass:$rd),
              (ins GPR:$rs1, IdxClass:$offset, AVL:$vl, ixlenimm:$sew),[]>,
       RISCVVPseudo,
class VPseudoISegLoadMask<VReg RetClass, VReg IdxClass, int EEW, bits<3> LMUL,
   let hasSideEffects = 0;
   // For vector indexed segment loads, the destination vector register groups
   // cannot overlap the source vector register group
   let Constraints = "@earlyclobber $rd, $rd = $merge";
   let HasVLOp = 1;
   let HasSEWOp = 1;
   let HasMergeOp = 1;
   let HasVecPolicyOp = 1;
+  let IsValidMaskPolicy = 1;
   let BaseInstr = !cast<Instruction>(PseudoToVInst<NAME>.VInst);
 }
 
 class VPseudoUSSegStoreNoMask<VReg ValClass, int EEW, bits<4> NF>:
       Pseudo<(outs),
              (ins ValClass:$rd, GPR:$rs1, AVL:$vl, ixlenimm:$sew),[]>,
       RISCVVPseudo,
       RISCVVSSEG<NF, /*Masked*/0, /*Strided*/0, log2<EEW>.val, VLMul> {

llvm/test/CodeGen/RISCV/rvv/masked-tama.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+v,+zfh \
+; RUN:   -verify-machineinstrs -target-abi=ilp32d | FileCheck %s --check-prefix=RV32
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+v,+zfh \
+; RUN:   -verify-machineinstrs -target-abi=lp64d | FileCheck %s --check-prefix=RV64
+
+declare <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vle_mask_nxv1i64_nxv1i64(<vscale x 1 x i64>* %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vle_mask_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e64, m1, ta, ma
+; RV32-NEXT:    vle64.v v8, (a0), v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vle_mask_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e64, m1, ta, ma
+; RV64-NEXT:    vle64.v v8, (a0), v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
+    <vscale x 1 x i64> undef,
+    <vscale x 1 x i64>* %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare { <vscale x 1 x i64>, iXLen } @llvm.riscv.vleff.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vleff_mask_v_nxv1i64_nxv1i64(<vscale x 1 x i64>* %0, <vscale x 1 x i1> %1, iXLen %2, iXLen* %3) nounwind {
+; RV32-LABEL: intrinsic_vleff_mask_v_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e64, m1, ta, ma
+; RV32-NEXT:    vle64ff.v v8, (a0), v0.t
+; RV32-NEXT:    csrr a0, vl
+; RV32-NEXT:    sw a0, 0(a2)
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vleff_mask_v_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e64, m1, ta, ma
+; RV64-NEXT:    vle64ff.v v8, (a0), v0.t
+; RV64-NEXT:    csrr a0, vl
+; RV64-NEXT:    sd a0, 0(a2)
+; RV64-NEXT:    ret
+entry:
+  %a = call { <vscale x 1 x i64>, iXLen } @llvm.riscv.vleff.mask.nxv1i64(
+    <vscale x 1 x i64> undef,
+    <vscale x 1 x i64>* %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+  %b = extractvalue { <vscale x 1 x i64>, iXLen } %a, 0
+  %c = extractvalue { <vscale x 1 x i64>, iXLen } %a, 1
+  store iXLen %c, iXLen* %3
+
+  ret <vscale x 1 x i64> %b
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vlse.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  iXLen,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vlse_mask_v_nxv1i64_nxv1i64(<vscale x 1 x i64>* %0, iXLen %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vlse_mask_v_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a2, e64, m1, ta, ma
+; RV32-NEXT:    vlse64.v v8, (a0), a1, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vlse_mask_v_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a2, e64, m1, ta, ma
+; RV64-NEXT:    vlse64.v v8, (a0), a1, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vlse.mask.nxv1i64(
+    <vscale x 1 x i64> undef,
+    <vscale x 1 x i64>* %0,
+    iXLen %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vluxei.mask.nxv1i8.nxv1iXLen(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>*,
+  <vscale x 1 x iXLen>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vluxei_mask_v_nxv1i8_nxv1i8_nxv1iXLen(<vscale x 1 x i8>* %0, <vscale x 1 x iXLen> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vluxei.mask.nxv1i8.nxv1iXLen(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8>* %0,
+    <vscale x 1 x iXLen> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vadd.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vadd.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwadd.mask.nxv1i16.nxv1i8.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vwadd.vv v10, v8, v9, v0.t
+; RV32-NEXT:    vmv1r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vwadd.vv v10, v8, v9, v0.t
+; RV64-NEXT:    vmv1r.v v8, v10
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwadd.mask.nxv1i16.nxv1i8.nxv1i8(
+    <vscale x 1 x i16> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrsub.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, ta, ma
+; RV32-NEXT:    vrsub.vx v8, v8, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, ta, ma
+; RV64-NEXT:    vrsub.vx v8, v8, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrsub.mask.nxv1i8.i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    i8 %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsub.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vadd.vi v8, v8, 9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vadd.vi v8, v8, 9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsub.mask.nxv1i8.i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    i8 -9,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf2_nxv1i64(<vscale x 1 x i32> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf2_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; RV32-NEXT:    vzext.vf2 v9, v8, v0.t
+; RV32-NEXT:    vmv.v.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf2_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; RV64-NEXT:    vzext.vf2 v9, v8, v0.t
+; RV64-NEXT:    vmv.v.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> undef,
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i16(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf4_nxv1i64(<vscale x 1 x i16> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf4_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; RV32-NEXT:    vzext.vf4 v9, v8, v0.t
+; RV32-NEXT:    vmv.v.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf4_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; RV64-NEXT:    vzext.vf4 v9, v8, v0.t
+; RV64-NEXT:    vmv.v.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i16(
+    <vscale x 1 x i64> undef,
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x i64> %a
+}
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i8(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf8_nxv1i64(<vscale x 1 x i8> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf8_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; RV32-NEXT:    vzext.vf8 v9, v8, v0.t
+; RV32-NEXT:    vmv.v.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf8_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; RV64-NEXT:    vzext.vf8 v9, v8, v0.t
+; RV64-NEXT:    vmv.v.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i8(
+    <vscale x 1 x i64> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vand.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vand.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vand.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vand.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsll.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vsll.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vsll.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsll.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vnsra.wv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vnsra.wv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vmin.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vmin.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vmin.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vmin.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vdiv.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vdiv.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vdiv.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vdiv.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vwmul.vv v10, v8, v9, v0.t
+; RV32-NEXT:    vmv1r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vwmul.vv v10, v8, v9, v0.t
+; RV64-NEXT:    vmv1r.v v8, v10
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8(
+    <vscale x 1 x i16> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8>  @intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vmacc.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vmacc.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vwmacc.vv v10, v8, v9, v0.t
+; RV32-NEXT:    vmv1r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vwmacc.vv v10, v8, v9, v0.t
+; RV64-NEXT:    vmv1r.v v8, v10
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vsadd.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vsadd.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vaadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vaadd.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vaadd.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vaadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsmul.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vsmul.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vsmul.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsmul.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vssrl.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vssrl.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vssrl.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vssrl.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vnclip.mask.nxv1i8.nxv1i16.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vnclip.wv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vnclip.wv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vnclip.mask.nxv1i8.nxv1i16.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfadd.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfadd.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwadd.mask.nxv1f32.nxv1f16.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfwadd.vv v10, v8, v9, v0.t
+; RV32-NEXT:    vmv1r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfwadd.vv v10, v8, v9, v0.t
+; RV64-NEXT:    vmv1r.v v8, v10
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwadd.mask.nxv1f32.nxv1f16.nxv1f16(
+    <vscale x 1 x float> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmul.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfmul.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfmul.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmul.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfdiv.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfdiv.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfdiv.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfdiv.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrdiv.mask.nxv1f16.f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  half,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfrdiv.vf v8, v8, fa0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfrdiv.vf v8, v8, fa0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrdiv.mask.nxv1f16.f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    half %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwmul.mask.nxv1f32.nxv1f16.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfwmul.vv v10, v8, v9, v0.t
+; RV32-NEXT:    vmv1r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfwmul.vv v10, v8, v9, v0.t
+; RV64-NEXT:    vmv1r.v v8, v10
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwmul.mask.nxv1f32.nxv1f16.nxv1f16(
+    <vscale x 1 x float> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x half>  @intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfmacc.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfmacc.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x float>  @intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfwmacc.vv v10, v8, v9, v0.t
+; RV32-NEXT:    vmv1r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfwmacc.vv v10, v8, v9, v0.t
+; RV64-NEXT:    vmv1r.v v8, v10
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16(
+    <vscale x 1 x float> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3);
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfsqrt.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfsqrt.v v8, v8, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfsqrt.v v8, v8, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfsqrt.mask.nxv1f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrsqrt7.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfrsqrt7.v v8, v8, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfrsqrt7.v v8, v8, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrsqrt7.mask.nxv1f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrec7.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfrec7.v v8, v8, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfrec7.v v8, v8, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrec7.mask.nxv1f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmin.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfmin.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfmin.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmin.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfsgnj.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfsgnj.vv v8, v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfsgnj.vv v8, v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfsgnj.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vfclass.mask.nxv1i16(
+  <vscale x 1 x i16>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vfclass_mask_v_nxv1i16_nxv1f16(
+; RV32-LABEL: intrinsic_vfclass_mask_v_nxv1i16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfclass.v v8, v8, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfclass_mask_v_nxv1i16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfclass.v v8, v8, v0.t
+; RV64-NEXT:    ret
+  <vscale x 1 x half> %0,
+  <vscale x 1 x i1> %1,
+  iXLen %2) nounwind {
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vfclass.mask.nxv1i16(
+    <vscale x 1 x i16> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vfcvt.xu.f.v.mask.nxv1i16.nxv1f16(
+  <vscale x 1 x i16>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfcvt.xu.f.v v8, v8, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfcvt.xu.f.v v8, v8, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vfcvt.xu.f.v.mask.nxv1i16.nxv1f16(
+    <vscale x 1 x i16> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfcvt.f.x.v.mask.nxv1f16.nxv1i16(
+  <vscale x 1 x half>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16(<vscale x 1 x i16> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfcvt.f.x.v v8, v8, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfcvt.f.x.v v8, v8, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfcvt.f.x.v.mask.nxv1f16.nxv1i16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vfwcvt.xu.f.v.mask.nxv1i32.nxv1f16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i32> @intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfwcvt.xu.f.v v9, v8, v0.t
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfwcvt.xu.f.v v9, v8, v0.t
+; RV64-NEXT:    vmv1r.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vfwcvt.xu.f.v.mask.nxv1i32.nxv1f16(
+    <vscale x 1 x i32> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfwcvt.f.x.v.mask.nxv1f16.nxv1i8(
+  <vscale x 1 x half>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vfwcvt.f.x.v v9, v8, v0.t
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vfwcvt.f.x.v v9, v8, v0.t
+; RV64-NEXT:    vmv1r.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfwcvt.f.x.v.mask.nxv1f16.nxv1i8(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwcvt.f.f.v.mask.nxv1f32.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfwcvt.f.f.v v9, v8, v0.t
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfwcvt.f.f.v v9, v8, v0.t
+; RV64-NEXT:    vmv1r.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwcvt.f.f.v.mask.nxv1f32.nxv1f16(
+    <vscale x 1 x float> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vfncvt.xu.f.w.mask.nxv1i8.nxv1f16(
+  <vscale x 1 x i8>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vfncvt.xu.f.w v9, v8, v0.t
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vfncvt.xu.f.w v9, v8, v0.t
+; RV64-NEXT:    vmv1r.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vfncvt.xu.f.w.mask.nxv1i8.nxv1f16(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfncvt.f.x.w.mask.nxv1f16.nxv1i32(
+  <vscale x 1 x half>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32(<vscale x 1 x i32> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfncvt.f.x.w v9, v8, v0.t
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfncvt.f.x.w v9, v8, v0.t
+; RV64-NEXT:    vmv1r.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfncvt.f.x.w.mask.nxv1f16.nxv1i32(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfncvt.f.f.w.mask.nxv1f16.nxv1f32(
+  <vscale x 1 x half>,
+  <vscale x 1 x float>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32(<vscale x 1 x float> %0, <vscale x 1 x i1> %1, iXLen %2) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfncvt.f.f.w v9, v8, v0.t
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfncvt.f.f.w v9, v8, v0.t
+; RV64-NEXT:    vmv1r.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfncvt.f.f.w.mask.nxv1f16.nxv1f32(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x float> %0,
+    <vscale x 1 x i1> %1,
+    iXLen %2, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  iXLen,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, iXLen %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, ta, ma
+; RV32-NEXT:    vslideup.vx v9, v8, a0, v0.t
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, ta, ma
+; RV64-NEXT:    vslideup.vx v9, v8, a0, v0.t
+; RV64-NEXT:    vmv1r.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    iXLen %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vslide1up.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, i8 %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, ta, ma
+; RV32-NEXT:    vslide1up.vx v9, v8, a0, v0.t
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, ta, ma
+; RV64-NEXT:    vslide1up.vx v9, v8, a0, v0.t
+; RV64-NEXT:    vmv1r.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vslide1up.mask.nxv1i8.i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    i8 %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfslide1up.mask.nxv1f16.f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  half,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16(<vscale x 1 x half> %0, half %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV32-NEXT:    vfslide1up.vf v9, v8, fa0, v0.t
+; RV32-NEXT:    vmv1r.v v8, v9
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma
+; RV64-NEXT:    vfslide1up.vf v9, v8, fa0, v0.t
+; RV64-NEXT:    vmv1r.v v8, v9
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfslide1up.mask.nxv1f16.f16(
+    <vscale x 1 x half> undef,
+    <vscale x 1 x half> %0,
+    half %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrgather.vv.mask.nxv1i8.iXLen(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vrgather.vv v10, v8, v9, v0.t
+; RV32-NEXT:    vmv1r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vrgather.vv v10, v8, v9, v0.t
+; RV64-NEXT:    vmv1r.v v8, v10
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrgather.vv.mask.nxv1i8.iXLen(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrgatherei16.vv.mask.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i16> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV32-NEXT:    vrgatherei16.vv v10, v8, v9, v0.t
+; RV32-NEXT:    vmv1r.v v8, v10
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; RV64-NEXT:    vrgatherei16.vv v10, v8, v9, v0.t
+; RV64-NEXT:    vmv1r.v v8, v10
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrgatherei16.vv.mask.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 3)
+
+  ret <vscale x 1 x i8> %a
+}

llvm/test/CodeGen/RISCV/rvv/masked-tamu.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+v,+zfh \
+; RUN:   -verify-machineinstrs -target-abi=ilp32d | FileCheck %s --check-prefix=RV32
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+v,+zfh \
+; RUN:   -verify-machineinstrs -target-abi=lp64d | FileCheck %s --check-prefix=RV64
+
+declare <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vle_mask_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64>* %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vle_mask_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e64, m1, ta, mu
+; RV32-NEXT:    vle64.v v8, (a0), v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vle_mask_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e64, m1, ta, mu
+; RV64-NEXT:    vle64.v v8, (a0), v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64>* %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare { <vscale x 1 x i64>, iXLen } @llvm.riscv.vleff.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vleff_mask_v_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64>* %1, <vscale x 1 x i1> %2, iXLen %3, iXLen* %4) nounwind {
+; RV32-LABEL: intrinsic_vleff_mask_v_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e64, m1, ta, mu
+; RV32-NEXT:    vle64ff.v v8, (a0), v0.t
+; RV32-NEXT:    csrr a0, vl
+; RV32-NEXT:    sw a0, 0(a2)
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vleff_mask_v_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e64, m1, ta, mu
+; RV64-NEXT:    vle64ff.v v8, (a0), v0.t
+; RV64-NEXT:    csrr a0, vl
+; RV64-NEXT:    sd a0, 0(a2)
+; RV64-NEXT:    ret
+entry:
+  %a = call { <vscale x 1 x i64>, iXLen } @llvm.riscv.vleff.mask.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64>* %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+  %b = extractvalue { <vscale x 1 x i64>, iXLen } %a, 0
+  %c = extractvalue { <vscale x 1 x i64>, iXLen } %a, 1
+  store iXLen %c, iXLen* %4
+
+  ret <vscale x 1 x i64> %b
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vlse.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  iXLen,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vlse_mask_v_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64>* %1, iXLen %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vlse_mask_v_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a2, e64, m1, ta, mu
+; RV32-NEXT:    vlse64.v v8, (a0), a1, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vlse_mask_v_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a2, e64, m1, ta, mu
+; RV64-NEXT:    vlse64.v v8, (a0), a1, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vlse.mask.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64>* %1,
+    iXLen %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vluxei.mask.nxv1i8.nxv1iXLen(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>*,
+  <vscale x 1 x iXLen>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vluxei_mask_v_nxv1i8_nxv1i8_nxv1iXLen(<vscale x 1 x i8> %0, <vscale x 1 x i8>* %1, <vscale x 1 x iXLen> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vluxei.mask.nxv1i8.nxv1iXLen(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8>* %1,
+    <vscale x 1 x iXLen> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwadd.mask.nxv1i16.nxv1i8.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vwadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vwadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwadd.mask.nxv1i16.nxv1i8.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrsub.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i8 %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, ta, mu
+; RV32-NEXT:    vrsub.vx v8, v9, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, ta, mu
+; RV64-NEXT:    vrsub.vx v8, v9, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrsub.mask.nxv1i8.i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i8 %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsub.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vadd.vi v8, v9, 9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vadd.vi v8, v9, 9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsub.mask.nxv1i8.i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i8 -9,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf2_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i32> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf2_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; RV32-NEXT:    vzext.vf2 v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf2_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; RV64-NEXT:    vzext.vf2 v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i16(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf4_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i16> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf4_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; RV32-NEXT:    vzext.vf4 v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf4_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; RV64-NEXT:    vzext.vf4 v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i16(
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x i64> %a
+}
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i8(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf8_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i8> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf8_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; RV32-NEXT:    vzext.vf8 v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf8_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; RV64-NEXT:    vzext.vf8 v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i8(
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vand.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vand.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vand.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vand.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsll.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vsll.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vsll.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsll.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i16> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vnsra.wv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vnsra.wv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vmin.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vmin.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vmin.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vmin.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vdiv.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vdiv.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vdiv.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vdiv.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vwmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vwmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8>  @intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vwmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vwmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vsadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vsadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vaadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vaadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vaadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vaadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsmul.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vsmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vsmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsmul.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vssrl.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vssrl.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vssrl.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vssrl.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vnclip.mask.nxv1i8.nxv1i16.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i16> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vnclip.wv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vnclip.wv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vnclip.mask.nxv1i8.nxv1i16.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwadd.mask.nxv1f32.nxv1f16.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfwadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfwadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwadd.mask.nxv1f32.nxv1f16.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmul.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmul.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfdiv.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfdiv.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfdiv.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfdiv.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrdiv.mask.nxv1f16.f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  half,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, half %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfrdiv.vf v8, v9, fa0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfrdiv.vf v8, v9, fa0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrdiv.mask.nxv1f16.f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    half %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwmul.mask.nxv1f32.nxv1f16.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfwmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfwmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwmul.mask.nxv1f32.nxv1f16.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x half>  @intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x float>  @intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfwmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfwmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1);
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfsqrt.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfsqrt.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfsqrt.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfsqrt.mask.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrsqrt7.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfrsqrt7.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfrsqrt7.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrsqrt7.mask.nxv1f16(
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrec7.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfrec7.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfrec7.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrec7.mask.nxv1f16(
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmin.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfmin.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfmin.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmin.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfsgnj.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfsgnj.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfsgnj.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfsgnj.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vfclass.mask.nxv1i16(
+  <vscale x 1 x i16>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vfclass_mask_v_nxv1i16_nxv1f16(
+; RV32-LABEL: intrinsic_vfclass_mask_v_nxv1i16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfclass.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfclass_mask_v_nxv1i16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfclass.v v8, v9, v0.t
+; RV64-NEXT:    ret
+  <vscale x 1 x i16> %0,
+  <vscale x 1 x half> %1,
+  <vscale x 1 x i1> %2,
+  iXLen %3) nounwind {
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vfclass.mask.nxv1i16(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vfcvt.xu.f.v.mask.nxv1i16.nxv1f16(
+  <vscale x 1 x i16>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16(<vscale x 1 x i16> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfcvt.xu.f.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfcvt.xu.f.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vfcvt.xu.f.v.mask.nxv1i16.nxv1f16(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfcvt.f.x.v.mask.nxv1f16.nxv1i16(
+  <vscale x 1 x half>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16(<vscale x 1 x half> %0, <vscale x 1 x i16> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfcvt.f.x.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfcvt.f.x.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfcvt.f.x.v.mask.nxv1f16.nxv1i16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vfwcvt.xu.f.v.mask.nxv1i32.nxv1f16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i32> @intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16(<vscale x 1 x i32> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfwcvt.xu.f.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfwcvt.xu.f.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vfwcvt.xu.f.v.mask.nxv1i32.nxv1f16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfwcvt.f.x.v.mask.nxv1f16.nxv1i8(
+  <vscale x 1 x half>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8(<vscale x 1 x half> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vfwcvt.f.x.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vfwcvt.f.x.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfwcvt.f.x.v.mask.nxv1f16.nxv1i8(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwcvt.f.f.v.mask.nxv1f32.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfwcvt.f.f.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfwcvt.f.f.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwcvt.f.f.v.mask.nxv1f32.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vfncvt.xu.f.w.mask.nxv1i8.nxv1f16(
+  <vscale x 1 x i8>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16(<vscale x 1 x i8> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vfncvt.xu.f.w v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vfncvt.xu.f.w v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vfncvt.xu.f.w.mask.nxv1i8.nxv1f16(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfncvt.f.x.w.mask.nxv1f16.nxv1i32(
+  <vscale x 1 x half>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32(<vscale x 1 x half> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfncvt.f.x.w v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfncvt.f.x.w v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfncvt.f.x.w.mask.nxv1f16.nxv1i32(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfncvt.f.f.w.mask.nxv1f16.nxv1f32(
+  <vscale x 1 x half>,
+  <vscale x 1 x float>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32(<vscale x 1 x half> %0, <vscale x 1 x float> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfncvt.f.f.w v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfncvt.f.f.w v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfncvt.f.f.w.mask.nxv1f16.nxv1f32(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x float> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  iXLen,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, iXLen %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, ta, mu
+; RV32-NEXT:    vslideup.vx v8, v9, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, ta, mu
+; RV64-NEXT:    vslideup.vx v8, v9, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    iXLen %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vslide1up.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i8 %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, ta, mu
+; RV32-NEXT:    vslide1up.vx v8, v9, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, ta, mu
+; RV64-NEXT:    vslide1up.vx v8, v9, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vslide1up.mask.nxv1i8.i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i8 %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfslide1up.mask.nxv1f16.f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  half,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, half %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV32-NEXT:    vfslide1up.vf v8, v9, fa0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; RV64-NEXT:    vfslide1up.vf v8, v9, fa0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfslide1up.mask.nxv1f16.f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    half %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrgather.vv.mask.nxv1i8.iXLen(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vrgather.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vrgather.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrgather.vv.mask.nxv1i8.iXLen(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrgatherei16.vv.mask.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV32-NEXT:    vrgatherei16.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; RV64-NEXT:    vrgatherei16.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrgatherei16.vv.mask.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 1)
+
+  ret <vscale x 1 x i8> %a
+}

llvm/test/CodeGen/RISCV/rvv/masked-tuma.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+v,+zfh \
+; RUN:   -verify-machineinstrs -target-abi=ilp32d | FileCheck %s --check-prefix=RV32
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+v,+zfh \
+; RUN:   -verify-machineinstrs -target-abi=lp64d | FileCheck %s --check-prefix=RV64
+
+declare <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vle_mask_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64>* %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vle_mask_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e64, m1, tu, ma
+; RV32-NEXT:    vle64.v v8, (a0), v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vle_mask_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e64, m1, tu, ma
+; RV64-NEXT:    vle64.v v8, (a0), v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64>* %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare { <vscale x 1 x i64>, iXLen } @llvm.riscv.vleff.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vleff_mask_v_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64>* %1, <vscale x 1 x i1> %2, iXLen %3, iXLen* %4) nounwind {
+; RV32-LABEL: intrinsic_vleff_mask_v_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e64, m1, tu, ma
+; RV32-NEXT:    vle64ff.v v8, (a0), v0.t
+; RV32-NEXT:    csrr a0, vl
+; RV32-NEXT:    sw a0, 0(a2)
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vleff_mask_v_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e64, m1, tu, ma
+; RV64-NEXT:    vle64ff.v v8, (a0), v0.t
+; RV64-NEXT:    csrr a0, vl
+; RV64-NEXT:    sd a0, 0(a2)
+; RV64-NEXT:    ret
+entry:
+  %a = call { <vscale x 1 x i64>, iXLen } @llvm.riscv.vleff.mask.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64>* %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+  %b = extractvalue { <vscale x 1 x i64>, iXLen } %a, 0
+  %c = extractvalue { <vscale x 1 x i64>, iXLen } %a, 1
+  store iXLen %c, iXLen* %4
+
+  ret <vscale x 1 x i64> %b
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vlse.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  iXLen,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vlse_mask_v_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64>* %1, iXLen %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vlse_mask_v_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a2, e64, m1, tu, ma
+; RV32-NEXT:    vlse64.v v8, (a0), a1, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vlse_mask_v_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a2, e64, m1, tu, ma
+; RV64-NEXT:    vlse64.v v8, (a0), a1, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vlse.mask.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64>* %1,
+    iXLen %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vluxei.mask.nxv1i8.nxv1iXLen(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>*,
+  <vscale x 1 x iXLen>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vluxei_mask_v_nxv1i8_nxv1i8_nxv1iXLen(<vscale x 1 x i8> %0, <vscale x 1 x i8>* %1, <vscale x 1 x iXLen> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vluxei.mask.nxv1i8.nxv1iXLen(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8>* %1,
+    <vscale x 1 x iXLen> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwadd.mask.nxv1i16.nxv1i8.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vwadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vwadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwadd.mask.nxv1i16.nxv1i8.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrsub.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i8 %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, tu, ma
+; RV32-NEXT:    vrsub.vx v8, v9, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, tu, ma
+; RV64-NEXT:    vrsub.vx v8, v9, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrsub.mask.nxv1i8.i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i8 %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsub.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vadd.vi v8, v9, 9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vadd.vi v8, v9, 9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsub.mask.nxv1i8.i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i8 -9,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf2_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i32> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf2_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, tu, ma
+; RV32-NEXT:    vzext.vf2 v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf2_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, tu, ma
+; RV64-NEXT:    vzext.vf2 v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i16(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf4_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i16> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf4_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, tu, ma
+; RV32-NEXT:    vzext.vf4 v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf4_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, tu, ma
+; RV64-NEXT:    vzext.vf4 v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i16(
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x i64> %a
+}
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i8(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf8_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i8> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf8_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, tu, ma
+; RV32-NEXT:    vzext.vf8 v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf8_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, tu, ma
+; RV64-NEXT:    vzext.vf8 v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i8(
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vand.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vand.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vand.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vand.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsll.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vsll.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vsll.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsll.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i16> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vnsra.wv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vnsra.wv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vmin.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vmin.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vmin.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vmin.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vdiv.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vdiv.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vdiv.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vdiv.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vwmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vwmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8>  @intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vwmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vwmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vsadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vsadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vaadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vaadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vaadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vaadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsmul.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vsmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vsmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsmul.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vssrl.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vssrl.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vssrl.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vssrl.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vnclip.mask.nxv1i8.nxv1i16.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i16> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vnclip.wv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vnclip.wv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vnclip.mask.nxv1i8.nxv1i16.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwadd.mask.nxv1f32.nxv1f16.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfwadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfwadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwadd.mask.nxv1f32.nxv1f16.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmul.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmul.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfdiv.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfdiv.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfdiv.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfdiv.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrdiv.mask.nxv1f16.f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  half,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, half %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfrdiv.vf v8, v9, fa0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfrdiv.vf v8, v9, fa0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrdiv.mask.nxv1f16.f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    half %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwmul.mask.nxv1f32.nxv1f16.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfwmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfwmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwmul.mask.nxv1f32.nxv1f16.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x half>  @intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x float>  @intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfwmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfwmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2);
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfsqrt.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfsqrt.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfsqrt.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfsqrt.mask.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrsqrt7.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfrsqrt7.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfrsqrt7.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrsqrt7.mask.nxv1f16(
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrec7.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfrec7.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfrec7.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrec7.mask.nxv1f16(
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmin.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfmin.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfmin.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmin.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfsgnj.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfsgnj.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfsgnj.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfsgnj.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vfclass.mask.nxv1i16(
+  <vscale x 1 x i16>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vfclass_mask_v_nxv1i16_nxv1f16(
+; RV32-LABEL: intrinsic_vfclass_mask_v_nxv1i16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfclass.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfclass_mask_v_nxv1i16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfclass.v v8, v9, v0.t
+; RV64-NEXT:    ret
+  <vscale x 1 x i16> %0,
+  <vscale x 1 x half> %1,
+  <vscale x 1 x i1> %2,
+  iXLen %3) nounwind {
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vfclass.mask.nxv1i16(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vfcvt.xu.f.v.mask.nxv1i16.nxv1f16(
+  <vscale x 1 x i16>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16(<vscale x 1 x i16> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfcvt.xu.f.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfcvt.xu.f.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vfcvt.xu.f.v.mask.nxv1i16.nxv1f16(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfcvt.f.x.v.mask.nxv1f16.nxv1i16(
+  <vscale x 1 x half>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16(<vscale x 1 x half> %0, <vscale x 1 x i16> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfcvt.f.x.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfcvt.f.x.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfcvt.f.x.v.mask.nxv1f16.nxv1i16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vfwcvt.xu.f.v.mask.nxv1i32.nxv1f16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i32> @intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16(<vscale x 1 x i32> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfwcvt.xu.f.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfwcvt.xu.f.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vfwcvt.xu.f.v.mask.nxv1i32.nxv1f16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfwcvt.f.x.v.mask.nxv1f16.nxv1i8(
+  <vscale x 1 x half>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8(<vscale x 1 x half> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vfwcvt.f.x.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vfwcvt.f.x.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfwcvt.f.x.v.mask.nxv1f16.nxv1i8(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwcvt.f.f.v.mask.nxv1f32.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfwcvt.f.f.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfwcvt.f.f.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwcvt.f.f.v.mask.nxv1f32.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vfncvt.xu.f.w.mask.nxv1i8.nxv1f16(
+  <vscale x 1 x i8>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16(<vscale x 1 x i8> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vfncvt.xu.f.w v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vfncvt.xu.f.w v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vfncvt.xu.f.w.mask.nxv1i8.nxv1f16(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfncvt.f.x.w.mask.nxv1f16.nxv1i32(
+  <vscale x 1 x half>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32(<vscale x 1 x half> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfncvt.f.x.w v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfncvt.f.x.w v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfncvt.f.x.w.mask.nxv1f16.nxv1i32(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfncvt.f.f.w.mask.nxv1f16.nxv1f32(
+  <vscale x 1 x half>,
+  <vscale x 1 x float>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32(<vscale x 1 x half> %0, <vscale x 1 x float> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfncvt.f.f.w v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfncvt.f.f.w v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfncvt.f.f.w.mask.nxv1f16.nxv1f32(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x float> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  iXLen,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, iXLen %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, tu, ma
+; RV32-NEXT:    vslideup.vx v8, v9, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, tu, ma
+; RV64-NEXT:    vslideup.vx v8, v9, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    iXLen %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vslide1up.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i8 %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, tu, ma
+; RV32-NEXT:    vslide1up.vx v8, v9, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, tu, ma
+; RV64-NEXT:    vslide1up.vx v8, v9, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vslide1up.mask.nxv1i8.i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i8 %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfslide1up.mask.nxv1f16.f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  half,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, half %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV32-NEXT:    vfslide1up.vf v8, v9, fa0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, ma
+; RV64-NEXT:    vfslide1up.vf v8, v9, fa0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfslide1up.mask.nxv1f16.f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    half %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrgather.vv.mask.nxv1i8.iXLen(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vrgather.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vrgather.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrgather.vv.mask.nxv1i8.iXLen(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrgatherei16.vv.mask.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV32-NEXT:    vrgatherei16.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; RV64-NEXT:    vrgatherei16.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrgatherei16.vv.mask.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 2)
+
+  ret <vscale x 1 x i8> %a
+}

llvm/test/CodeGen/RISCV/rvv/masked-tumu.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+v,+zfh \
+; RUN:   -verify-machineinstrs -target-abi=ilp32d | FileCheck %s --check-prefix=RV32
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+v,+zfh \
+; RUN:   -verify-machineinstrs -target-abi=lp64d | FileCheck %s --check-prefix=RV64
+
+declare <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vle_mask_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64>* %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vle_mask_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e64, m1, tu, mu
+; RV32-NEXT:    vle64.v v8, (a0), v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vle_mask_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e64, m1, tu, mu
+; RV64-NEXT:    vle64.v v8, (a0), v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vle.mask.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64>* %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare { <vscale x 1 x i64>, iXLen } @llvm.riscv.vleff.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vleff_mask_v_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64>* %1, <vscale x 1 x i1> %2, iXLen %3, iXLen* %4) nounwind {
+; RV32-LABEL: intrinsic_vleff_mask_v_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e64, m1, tu, mu
+; RV32-NEXT:    vle64ff.v v8, (a0), v0.t
+; RV32-NEXT:    csrr a0, vl
+; RV32-NEXT:    sw a0, 0(a2)
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vleff_mask_v_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e64, m1, tu, mu
+; RV64-NEXT:    vle64ff.v v8, (a0), v0.t
+; RV64-NEXT:    csrr a0, vl
+; RV64-NEXT:    sd a0, 0(a2)
+; RV64-NEXT:    ret
+entry:
+  %a = call { <vscale x 1 x i64>, iXLen } @llvm.riscv.vleff.mask.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64>* %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+  %b = extractvalue { <vscale x 1 x i64>, iXLen } %a, 0
+  %c = extractvalue { <vscale x 1 x i64>, iXLen } %a, 1
+  store iXLen %c, iXLen* %4
+
+  ret <vscale x 1 x i64> %b
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vlse.mask.nxv1i64(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i64>*,
+  iXLen,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vlse_mask_v_nxv1i64_nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64>* %1, iXLen %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vlse_mask_v_nxv1i64_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a2, e64, m1, tu, mu
+; RV32-NEXT:    vlse64.v v8, (a0), a1, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vlse_mask_v_nxv1i64_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a2, e64, m1, tu, mu
+; RV64-NEXT:    vlse64.v v8, (a0), a1, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vlse.mask.nxv1i64(
+    <vscale x 1 x i64> %0,
+    <vscale x 1 x i64>* %1,
+    iXLen %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vluxei.mask.nxv1i8.nxv1iXLen(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>*,
+  <vscale x 1 x iXLen>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vluxei_mask_v_nxv1i8_nxv1i8_nxv1iXLen(<vscale x 1 x i8> %0, <vscale x 1 x i8>* %1, <vscale x 1 x iXLen> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vluxei.mask.nxv1i8.nxv1iXLen(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8>* %1,
+    <vscale x 1 x iXLen> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwadd.mask.nxv1i16.nxv1i8.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vwadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwadd_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vwadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwadd.mask.nxv1i16.nxv1i8.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrsub.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i8 %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, tu, mu
+; RV32-NEXT:    vrsub.vx v8, v9, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrsub_mask_vx_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, tu, mu
+; RV64-NEXT:    vrsub.vx v8, v9, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrsub.mask.nxv1i8.i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i8 %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsub.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vadd.vi v8, v9, 9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsub_mask_vi_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vadd.vi v8, v9, 9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsub.mask.nxv1i8.i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i8 -9,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i32(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf2_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i32> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf2_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
+; RV32-NEXT:    vzext.vf2 v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf2_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
+; RV64-NEXT:    vzext.vf2 v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i32(
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i32> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i16(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf4_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i16> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf4_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
+; RV32-NEXT:    vzext.vf4 v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf4_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
+; RV64-NEXT:    vzext.vf4 v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i16(
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x i64> %a
+}
+declare <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i8(
+  <vscale x 1 x i64>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i64> @intrinsic_vzext_mask_vf8_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i8> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vzext_mask_vf8_nxv1i64:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
+; RV32-NEXT:    vzext.vf8 v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vzext_mask_vf8_nxv1i64:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e64, m1, tu, mu
+; RV64-NEXT:    vzext.vf8 v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vzext.mask.nxv1i64.nxv1i8(
+    <vscale x 1 x i64> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x i64> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vand.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vand.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vand_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vand.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vand.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsll.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vsll.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsll_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vsll.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsll.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i16> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vnsra.wv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vnsra_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vnsra.wv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vnsra.mask.nxv1i8.nxv1i16.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vmin.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vmin.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vmin_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vmin.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vmin.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vdiv.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vdiv.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vdiv_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vdiv.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vdiv.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vwmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwmul_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vwmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmul.mask.nxv1i16.nxv1i8.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8>  @intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vmacc_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vmacc.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i16>  @intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8(<vscale x 1 x i16> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vwmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vwmacc_mask_vv_nxv1i16_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vwmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vwmacc.mask.nxv1i16.nxv1i8(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vsadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vsadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vaadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vaadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vaadd_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vaadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vaadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vsmul.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vsmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vsmul_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vsmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsmul.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vssrl.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vssrl.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vssrl_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vssrl.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vssrl.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vnclip.mask.nxv1i8.nxv1i16.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i16> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vnclip.wv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vnclip_mask_wv_nxv1i8_nxv1i16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vnclip.wv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vnclip.mask.nxv1i8.nxv1i16.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfadd_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfadd.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwadd.mask.nxv1f32.nxv1f16.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfwadd.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwadd_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfwadd.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwadd.mask.nxv1f32.nxv1f16.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmul.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmul_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmul.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfdiv.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfdiv.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfdiv_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfdiv.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfdiv.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrdiv.mask.nxv1f16.f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  half,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, half %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfrdiv.vf v8, v9, fa0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrdiv_mask_vf_nxv1f16_nxv1f16_f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfrdiv.vf v8, v9, fa0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrdiv.mask.nxv1f16.f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    half %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwmul.mask.nxv1f32.nxv1f16.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfwmul.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwmul_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfwmul.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwmul.mask.nxv1f32.nxv1f16.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x half>  @intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmacc_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmacc.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x float>  @intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfwmacc.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwmacc_mask_vv_nxv1f32_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfwmacc.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwmacc.mask.nxv1f32.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0);
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfsqrt.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfsqrt.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfsqrt_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfsqrt.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfsqrt.mask.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrsqrt7.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfrsqrt7.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrsqrt7_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfrsqrt7.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrsqrt7.mask.nxv1f16(
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfrec7.mask.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfrec7.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfrec7_mask_v_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfrec7.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfrec7.mask.nxv1f16(
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %0,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfmin.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfmin.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfmin_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfmin.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfmin.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfsgnj.mask.nxv1f16.nxv1f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfsgnj.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfsgnj_mask_vv_nxv1f16_nxv1f16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfsgnj.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfsgnj.mask.nxv1f16.nxv1f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x half> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vfclass.mask.nxv1i16(
+  <vscale x 1 x i16>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vfclass_mask_v_nxv1i16_nxv1f16(
+; RV32-LABEL: intrinsic_vfclass_mask_v_nxv1i16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfclass.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfclass_mask_v_nxv1i16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfclass.v v8, v9, v0.t
+; RV64-NEXT:    ret
+  <vscale x 1 x i16> %0,
+  <vscale x 1 x half> %1,
+  <vscale x 1 x i1> %2,
+  iXLen %3) nounwind {
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vfclass.mask.nxv1i16(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x i16> @llvm.riscv.vfcvt.xu.f.v.mask.nxv1i16.nxv1f16(
+  <vscale x 1 x i16>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i16> @intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16(<vscale x 1 x i16> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfcvt.xu.f.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfcvt_mask_xu.f.v_nxv1i16_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfcvt.xu.f.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i16> @llvm.riscv.vfcvt.xu.f.v.mask.nxv1i16.nxv1f16(
+    <vscale x 1 x i16> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x i16> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfcvt.f.x.v.mask.nxv1f16.nxv1i16(
+  <vscale x 1 x half>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16(<vscale x 1 x half> %0, <vscale x 1 x i16> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfcvt.f.x.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfcvt_mask_f.x.v_nxv1f16_nxv1i16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfcvt.f.x.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfcvt.f.x.v.mask.nxv1f16.nxv1i16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i16> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i32> @llvm.riscv.vfwcvt.xu.f.v.mask.nxv1i32.nxv1f16(
+  <vscale x 1 x i32>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i32> @intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16(<vscale x 1 x i32> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfwcvt.xu.f.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_xu.f.v_nxv1i32_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfwcvt.xu.f.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i32> @llvm.riscv.vfwcvt.xu.f.v.mask.nxv1i32.nxv1f16(
+    <vscale x 1 x i32> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x i32> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfwcvt.f.x.v.mask.nxv1f16.nxv1i8(
+  <vscale x 1 x half>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8(<vscale x 1 x half> %0, <vscale x 1 x i8> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vfwcvt.f.x.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_f.x.v_nxv1f16_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vfwcvt.f.x.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfwcvt.f.x.v.mask.nxv1f16.nxv1i8(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x float> @llvm.riscv.vfwcvt.f.f.v.mask.nxv1f32.nxv1f16(
+  <vscale x 1 x float>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x float> @intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16(<vscale x 1 x float> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfwcvt.f.f.v v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfwcvt_mask_f.f.v_nxv1f32_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfwcvt.f.f.v v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x float> @llvm.riscv.vfwcvt.f.f.v.mask.nxv1f32.nxv1f16(
+    <vscale x 1 x float> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x float> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vfncvt.xu.f.w.mask.nxv1i8.nxv1f16(
+  <vscale x 1 x i8>,
+  <vscale x 1 x half>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16(<vscale x 1 x i8> %0, <vscale x 1 x half> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vfncvt.xu.f.w v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_xu.f.w_nxv1i8_nxv1f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vfncvt.xu.f.w v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vfncvt.xu.f.w.mask.nxv1i8.nxv1f16(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x half> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfncvt.f.x.w.mask.nxv1f16.nxv1i32(
+  <vscale x 1 x half>,
+  <vscale x 1 x i32>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32(<vscale x 1 x half> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfncvt.f.x.w v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_f.x.w_nxv1f16_nxv1i32:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfncvt.f.x.w v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfncvt.f.x.w.mask.nxv1f16.nxv1i32(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x i32> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfncvt.f.f.w.mask.nxv1f16.nxv1f32(
+  <vscale x 1 x half>,
+  <vscale x 1 x float>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32(<vscale x 1 x half> %0, <vscale x 1 x float> %1, <vscale x 1 x i1> %2, iXLen %3) nounwind {
+; RV32-LABEL: intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfncvt.f.f.w v8, v9, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfncvt_mask_f.f.w_nxv1f16_nxv1f32:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfncvt.f.f.w v8, v9, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfncvt.f.f.w.mask.nxv1f16.nxv1f32(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x float> %1,
+    <vscale x 1 x i1> %2,
+    iXLen %3, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  iXLen,
+  <vscale x 1 x i1>,
+  iXLen, iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, iXLen %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, tu, mu
+; RV32-NEXT:    vslideup.vx v8, v9, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vslideup_mask_vx_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, tu, mu
+; RV64-NEXT:    vslideup.vx v8, v9, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vslideup.mask.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    iXLen %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vslide1up.mask.nxv1i8.i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i8,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i8 %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a1, e8, mf8, tu, mu
+; RV32-NEXT:    vslide1up.vx v8, v9, a0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vslide1up_mask_vx_nxv1i8_nxv1i8_i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a1, e8, mf8, tu, mu
+; RV64-NEXT:    vslide1up.vx v8, v9, a0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vslide1up.mask.nxv1i8.i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i8 %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x half> @llvm.riscv.vfslide1up.mask.nxv1f16.f16(
+  <vscale x 1 x half>,
+  <vscale x 1 x half>,
+  half,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x half> @intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16(<vscale x 1 x half> %0, <vscale x 1 x half> %1, half %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV32-NEXT:    vfslide1up.vf v8, v9, fa0, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vfslide1up_mask_vf_nxv1f16_nxv1f16_f16:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e16, mf4, tu, mu
+; RV64-NEXT:    vfslide1up.vf v8, v9, fa0, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x half> @llvm.riscv.vfslide1up.mask.nxv1f16.f16(
+    <vscale x 1 x half> %0,
+    <vscale x 1 x half> %1,
+    half %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x half> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrgather.vv.mask.nxv1i8.iXLen(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vrgather.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrgather_mask_vv_nxv1i8_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vrgather.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrgather.vv.mask.nxv1i8.iXLen(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}
+
+declare <vscale x 1 x i8> @llvm.riscv.vrgatherei16.vv.mask.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i16>,
+  <vscale x 1 x i1>,
+  iXLen,
+  iXLen);
+
+define <vscale x 1 x i8> @intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, <vscale x 1 x i16> %2, <vscale x 1 x i1> %3, iXLen %4) nounwind {
+; RV32-LABEL: intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8:
+; RV32:       # %bb.0: # %entry
+; RV32-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV32-NEXT:    vrgatherei16.vv v8, v9, v10, v0.t
+; RV32-NEXT:    ret
+;
+; RV64-LABEL: intrinsic_vrgatherei16_mask_vv_nxv1i8_nxv1i8:
+; RV64:       # %bb.0: # %entry
+; RV64-NEXT:    vsetvli zero, a0, e8, mf8, tu, mu
+; RV64-NEXT:    vrgatherei16.vv v8, v9, v10, v0.t
+; RV64-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vrgatherei16.vv.mask.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i16> %2,
+    <vscale x 1 x i1> %3,
+    iXLen %4, iXLen 0)
+
+  ret <vscale x 1 x i8> %a
+}

llvm/test/CodeGen/RISCV/rvv/vadd-policy.ll

-; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs \
-; RUN:   --riscv-no-aliases < %s | FileCheck %s
-
-declare <vscale x 8 x i8> @llvm.riscv.vadd.nxv8i8.nxv8i8(
-  <vscale x 8 x i8>,
-  <vscale x 8 x i8>,
-  <vscale x 8 x i8>,
-  i64);
-
-define <vscale x 8 x i8> @intrinsic_vadd_vv_nxv8i8_nxv8i8_nxv8i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, i64 %2) nounwind {
-; CHECK-LABEL: intrinsic_vadd_vv_nxv8i8_nxv8i8_nxv8i8:
-; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vadd.vv v8, v8, v9
-; CHECK-NEXT:    jalr zero, 0(ra)
-entry:
-  %a = call <vscale x 8 x i8> @llvm.riscv.vadd.nxv8i8.nxv8i8(
-    <vscale x 8 x i8> undef,
-    <vscale x 8 x i8> %0,
-    <vscale x 8 x i8> %1,
-    i64 %2)
-
-  ret <vscale x 8 x i8> %a
-}
-
-declare <vscale x 8 x i8> @llvm.riscv.vadd.mask.nxv8i8.nxv8i8(
-  <vscale x 8 x i8>,
-  <vscale x 8 x i8>,
-  <vscale x 8 x i8>,
-  <vscale x 8 x i1>,
-  i64, i64);
-
-define <vscale x 8 x i8> @intrinsic_vadd_mask_tu(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
-; CHECK-LABEL: intrinsic_vadd_mask_tu:
-; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, tu, mu
-; CHECK-NEXT:    vadd.vv v8, v9, v10, v0.t
-; CHECK-NEXT:    jalr zero, 0(ra)
-entry:
-  %a = call <vscale x 8 x i8> @llvm.riscv.vadd.mask.nxv8i8.nxv8i8(
-    <vscale x 8 x i8> %0,
-    <vscale x 8 x i8> %1,
-    <vscale x 8 x i8> %2,
-    <vscale x 8 x i1> %3,
-    i64 %4, i64 0)
-
-  ret <vscale x 8 x i8> %a
-}
-
-define <vscale x 8 x i8> @intrinsic_vadd_mask_ta(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i1> %3, i64 %4) nounwind {
-; CHECK-LABEL: intrinsic_vadd_mask_ta:
-; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vadd.vv v8, v9, v10, v0.t
-; CHECK-NEXT:    jalr zero, 0(ra)
-entry:
-  %a = call <vscale x 8 x i8> @llvm.riscv.vadd.mask.nxv8i8.nxv8i8(
-    <vscale x 8 x i8> %0,
-    <vscale x 8 x i8> %1,
-    <vscale x 8 x i8> %2,
-    <vscale x 8 x i1> %3,
-    i64 %4, i64 1)
-
-  ret <vscale x 8 x i8> %a
-}
-