[VP] Strided loads/stores

[loralb]

Summary

This patch introduces two new experimental IR intrinsics and SDAG nodes to represent vector strided loads and stores.

[craig.topper]

llvm/include/llvm/IR/VPIntrinsics.def

217

Is this an existing mistake in the documentation? Can you just push this as an NFC commit? No need to bury it in this patch.

llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp

7969

Since this accesses multiple disjoint memory locations, the Size is not MemVT.getStoreSize(). It probably needs to always be unknown. I guess it would be if the stride were statically known to be 0, but not sure it is worth special casing that.

8102

Same as load, the Size can't be based on the store type.

8158

Same

llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

7449

This Size doesn't make sense. I think it needs to be unknown.

7470

Same

llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.h

574

Can these be SmallVectorImpl<SDValue>?

[rogfer01]

llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

7437

Does anyone know if we should be passing AAInfo here?

(Non-stride) VPLoad doesn't do that, so we can fix both in a later change if necessary.

[loralb]

ChangeLog:

• add documentation for the new intrinsics
• use MemoryLocation::UnknownSize
• use SmallVectorImpl<SDValue> in functions signatures

[loralb]

llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.h

574

I'll fix visitVPLoadGather and visitVPStoreScatter above in a later change.

[loralb]

Changelog:

• ensure default alignment behaviour reflects changes in D114423
• align MemoryLocation retrieval for VPStridedLoad to VPLoad (see D115036)

[simoll]

llvm/lib/IR/IntrinsicInst.cpp

495

This will require an explicit return type because of opaque pointers, as in 6213f1dd03e27fd2483f8ac775a346de8e873573

[loralb]

Changelog:

• use explicit return type in Intrinsic::getDeclaration() for experimental_vp_strided_load
• get first argument type from first operand in Intrinsic::getDeclaration() for experimental_vp_strided_store

[frasercrmck]

llvm/docs/LangRef.rst

19747

I think mentioning that the align parameter can be used (and correctly handling that further down in SelectionDAG) would be good.

[loralb]

Changelog:

• Mention that the align parameter attribute can be provided for the base pointer operands

[loralb]

llvm/docs/LangRef.rst

19735

Reflecting on how the strided versions work, I was thinking if, semantically speaking, it would be better to define the pointer operand as a pointer to the scalar element type of the data/return operand instead of a pointer to a vector (this is only relevant while opaque pointers are not the default)

[simoll]

llvm/docs/LangRef.rst

19735

Agreed. Typed pointers will go away but we shouldn't be sloppy here.

[loralb]

Changelog:

• Changed the pointer operand type to match the scalar elements type of the data/return operand

[simoll]

llvm/docs/LangRef.rst

19763

The stride can have a different number of bits than the ptr.
To make the definition unambiguous, we should mention that the stride is always interpreted as a signed integer and all arithmetic occurs in the pointer type.
This implies that an over-sized stride is truncated and a less-bits-than-the-address stride is always sign extended to the address type.

[loralb]

Changelog:

• Update documentation
• Align with recent changes to vp_load and vp_store

[simoll]

LGTM

[loralb]

I do not have commit access, could anyone commit this for me please?

[loralb]

Thanks!

llvm/docs/LangRef.rst

 .. code-block:: text
 
      call void @llvm.vp.store.v8i8.p0v8i8(<8 x i8> %val, <8 x i8>* align 4 %ptr, <8 x i1> %mask, i32 %evl)
      ;; For all lanes below %evl, the call above is lane-wise equivalent to the call below.
 
      call void @llvm.masked.store.v8i8.p0v8i8(<8 x i8> %val, <8 x i8>* %ptr, i32 4, <8 x i1> %mask)
 
 
+.. _int_experimental_vp_strided_load:
+
+'``llvm.experimental.vp.strided.load``' Intrinsic
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+This is an overloaded intrinsic.
+
+::
+
+    declare <4 x float> @llvm.experimental.vp.strided.load.v4f32.p0v4f32.i64(<4 x float>* %ptr, i64 %stride, <4 x i1> %mask, i32 %evl)
+    declare <vscale x 2 x i16> @llvm.experimental.vp.strided.load.nxv2i16.p0nxv2i16.i64(<vscale x 2 x i16>* %ptr, i64 %stride, <vscale x 2 x i1> %mask, i32 %evl)
+
+Overview:
+"""""""""
+
+The '``llvm.experimental.vp.strided.load``' intrinsic loads, into a vector, scalar values from
+memory locations evenly spaced apart by '``stride``' number of bytes, starting from '``ptr``'.
+
+Arguments:
+""""""""""
+
+The first operand is the base pointer for the load. The second operand is the stride
+value expressed in bytes. The third operand is a vector of boolean values
+with the same number of elements as the return type. The fourth is the explicit
+vector length of the operation. The return type and underlying type of the base pointer
+are the same vector types.
+
+Semantics:
+""""""""""
+
+The '``llvm.experimental.vp.strided.load``' intrinsic loads, into a vector, multiple scalar
+values from memory in the same way as the :ref:`llvm.vp.gather <int_vp_gather>` intrinsic,
+where the vector of pointers is in the form:
+
+   ``%ptrs = <%ptr, %ptr + %stride, %ptr + 2 * %stride, ... >``,
+
+with %ptr previously casted to a pointer `i8`.
+
+Examples:
+"""""""""
+
+.. code-block:: text
+
+   %r = call <8 x i64> @llvm.experimental.vp.strided.load.v8i64.v8p0i64.i64(<8 x i64>* %ptr, i64 %stride, <8 x i64> %mask, i32 %evl)
+   ;; The operation can also be expressed like this:
+
+   %addr = bitcast <8 x i64>* %ptr to i8*
+   ;; Create a vector of pointers %addrs in the form:
+   ;; %addrs = <%addr, %addr + %stride, %addr + 2 * %stride, ...>
+   %ptrs = bitcast <8 x i8* > %addrs to <8 x i64* >
+   %also.r = call <8 x i64> @llvm.vp.gather.v8i64.v8p0i64(<8 x i64* > %ptrs, <8 x i64> %mask, i32 %evl)
+
+
+.. _int_experimental_vp_strided_store:
+
+'``llvm.experimental.vp.strided.store``' Intrinsic
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+This is an overloaded intrinsic.
+
+::
+
+    declare void @llvm.experimental.vp.strided.store.v4f32.p0v4f32.i64(<4 x float> %val, <4 x float>* %ptr, i64 %stride, <4 x i1> %mask, i32 %evl)
+    declare void @llvm.experimental.vp.strided.store.nxv2i16.p0nxv2i16.i64(<vscale x 2 x i16> %val, <vscale x 2 x i16>* %ptr, i64 %stride, <vscale x 2 x i1> %mask, i32 %evl)
+
+Overview:
+"""""""""
+
+The '``@llvm.experimental.vp.strided.store``' intrinsic stores the elements of
+'``val``' into memory locations evenly spaced apart by '``stride``' number of
+bytes, starting from '``ptr``'.
+
+Arguments:
+""""""""""
+
+The first operand is the vector value to be written to memory. The second
+operand is the base pointer for the store. It has the same underlying type as
+the value operand. The third operand is the stride value expressed in bytes.
+The fourth operand is a vector of boolean values with the same number of elements
+as the return type. The fifth is the explicit vector length of the operation.
+
+Semantics:
+""""""""""
+
+The '``llvm.experimental.vp.strided.store``' intrinsic stores the elements of
+'``val``' in the same way as the :ref:`llvm.vp.scatter <int_vp_scatter>` intrinsic,
+where the vector of pointers is in the form:
+
+  ``%ptrs = <%ptr, %ptr + %stride, %ptr + 2 * %stride, ... >``,
+
+with %ptr previously casted to a pointer `i8`.
+
+Examples:
+"""""""""
+
+.. code-block:: text
+
+   call void @llvm.experimental.vp.strided.store.v8i64.v8p0i64.i64(<8 x i64> %val, <8 x i64>* %ptr, i64 %stride, <8 x i1> %mask, i32 %evl)
+   ;; The operation can also be expressed like this:
+
+   %addr = bitcast <8 x i64>* %ptr to i8*
+   ;; Create a vector of pointers %addrs in the form:
+   ;; %addrs = <%addr, %addr + %stride, %addr + 2 * %stride, ...>
+   %ptrs = bitcast <8 x i8* > %addrs to <8 x i64* >
+   call void @llvm.vp.scatter.v8i64.v8p0i64(<8 x i64> %val, <8 x i64*> %ptrs, <8 x i1> %mask, i32 %evl)
+
+
 .. _int_vp_gather:
 
 '``llvm.vp.gather``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Syntax:
 """""""
 This is an overloaded intrinsic.

llvm/include/llvm/CodeGen/SelectionDAG.h

SDValue getTruncStoreVP(SDValue Chain, const SDLoc &dl, SDValue Val,
                           MachineMemOperand::Flags MMOFlags,
                           const AAMDNodes &AAInfo, bool IsCompressing = false);
   SDValue getTruncStoreVP(SDValue Chain, const SDLoc &dl, SDValue Val,
                           SDValue Ptr, SDValue Mask, SDValue EVL, EVT SVT,
                           MachineMemOperand *MMO, bool IsCompressing = false);
   SDValue getIndexedStoreVP(SDValue OrigStore, const SDLoc &dl, SDValue Base,
                             SDValue Offset, ISD::MemIndexedMode AM);
 
+  SDValue getStridedLoadVP(ISD::MemIndexedMode AM, ISD::LoadExtType ExtType,
+                           EVT VT, const SDLoc &DL, SDValue Chain, SDValue Ptr,
+                           SDValue Offset, SDValue Stride, SDValue Mask,
+                           SDValue EVL, MachinePointerInfo PtrInfo, EVT MemVT,
+                           Align Alignment, MachineMemOperand::Flags MMOFlags,
+                           const AAMDNodes &AAInfo,
+                           const MDNode *Ranges = nullptr,
+                           bool IsExpanding = false);
+  inline SDValue getStridedLoadVP(
+      ISD::MemIndexedMode AM, ISD::LoadExtType ExtType, EVT VT, const SDLoc &DL,
+      SDValue Chain, SDValue Ptr, SDValue Offset, SDValue Stride, SDValue Mask,
+      SDValue EVL, MachinePointerInfo PtrInfo, EVT MemVT,
+      MaybeAlign Alignment = MaybeAlign(),
+      MachineMemOperand::Flags MMOFlags = MachineMemOperand::MONone,
+      const AAMDNodes &AAInfo = AAMDNodes(), const MDNode *Ranges = nullptr,
+      bool IsExpanding = false) {
+    // Ensures that codegen never sees a None Alignment.
+    return getStridedLoadVP(AM, ExtType, VT, DL, Chain, Ptr, Offset, Stride,
+                            Mask, EVL, PtrInfo, MemVT,
+                            Alignment.getValueOr(getEVTAlign(MemVT)), MMOFlags,
+                            AAInfo, Ranges, IsExpanding);
+  }
+  SDValue getStridedLoadVP(ISD::MemIndexedMode AM, ISD::LoadExtType ExtType,
+                           EVT VT, const SDLoc &DL, SDValue Chain, SDValue Ptr,
+                           SDValue Offset, SDValue Stride, SDValue Mask,
+                           SDValue EVL, EVT MemVT, MachineMemOperand *MMO,
+                           bool IsExpanding = false);
+  SDValue getStridedLoadVP(EVT VT, const SDLoc &DL, SDValue Chain, SDValue Ptr,
+                           SDValue Stride, SDValue Mask, SDValue EVL,
+                           MachinePointerInfo PtrInfo, MaybeAlign Alignment,
+                           MachineMemOperand::Flags MMOFlags,
+                           const AAMDNodes &AAInfo,
+                           const MDNode *Ranges = nullptr,
+                           bool IsExpanding = false);
+  SDValue getStridedLoadVP(EVT VT, const SDLoc &DL, SDValue Chain, SDValue Ptr,
+                           SDValue Stride, SDValue Mask, SDValue EVL,
+                           MachineMemOperand *MMO, bool IsExpanding = false);
+  SDValue
+  getExtStridedLoadVP(ISD::LoadExtType ExtType, const SDLoc &DL, EVT VT,
+                      SDValue Chain, SDValue Ptr, SDValue Stride, SDValue Mask,
+                      SDValue EVL, MachinePointerInfo PtrInfo, EVT MemVT,
+                      MaybeAlign Alignment, MachineMemOperand::Flags MMOFlags,
+                      const AAMDNodes &AAInfo, bool IsExpanding = false);
+  SDValue getExtStridedLoadVP(ISD::LoadExtType ExtType, const SDLoc &DL, EVT VT,
+                              SDValue Chain, SDValue Ptr, SDValue Stride,
+                              SDValue Mask, SDValue EVL, EVT MemVT,
+                              MachineMemOperand *MMO, bool IsExpanding = false);
+  SDValue getIndexedStridedLoadVP(SDValue OrigLoad, const SDLoc &DL,
+                                  SDValue Base, SDValue Offset,
+                                  ISD::MemIndexedMode AM);
+  SDValue getStridedStoreVP(SDValue Chain, const SDLoc &DL, SDValue Val,
+                            SDValue Ptr, SDValue Stride, SDValue Mask,
+                            SDValue EVL, MachinePointerInfo PtrInfo,
+                            Align Alignment, MachineMemOperand::Flags MMOFlags,
+                            const AAMDNodes &AAInfo = AAMDNodes(),
+                            bool IsCompressing = false);
+  SDValue getStridedStoreVP(SDValue Chain, const SDLoc &DL, SDValue Val,
+                            SDValue Ptr, SDValue Stride, SDValue Mask,
+                            SDValue EVL, MachineMemOperand *MMO,
+                            bool IsCompressing = false);
+  SDValue getTruncStridedStoreVP(SDValue Chain, const SDLoc &DL, SDValue Val,
+                                 SDValue Ptr, SDValue Stride, SDValue Mask,
+                                 SDValue EVL, MachinePointerInfo PtrInfo,
+                                 EVT SVT, Align Alignment,
+                                 MachineMemOperand::Flags MMOFlags,
+                                 const AAMDNodes &AAInfo,
+                                 bool IsCompressing = false);
+  SDValue getTruncStridedStoreVP(SDValue Chain, const SDLoc &DL, SDValue Val,
+                                 SDValue Ptr, SDValue Stride, SDValue Mask,
+                                 SDValue EVL, EVT SVT, MachineMemOperand *MMO,
+                                 bool IsCompressing = false);
+  SDValue getIndexedStridedStoreVP(SDValue OrigStore, const SDLoc &DL,
+                                   SDValue Base, SDValue Offset,
+                                   ISD::MemIndexedMode AM);
+
   SDValue getGatherVP(SDVTList VTs, EVT VT, const SDLoc &dl,
                       ArrayRef<SDValue> Ops, MachineMemOperand *MMO,
                       ISD::MemIndexType IndexType);
   SDValue getScatterVP(SDVTList VTs, EVT VT, const SDLoc &dl,
                        ArrayRef<SDValue> Ops, MachineMemOperand *MMO,
                        ISD::MemIndexType IndexType);
 
   SDValue getMaskedLoad(EVT VT, const SDLoc &dl, SDValue Chain, SDValue Base,

llvm/include/llvm/CodeGen/SelectionDAGNodes.h

class MemSDNodeBitfields {
     uint16_t IsNonTemporal : 1;
     uint16_t IsDereferenceable : 1;
     uint16_t IsInvariant : 1;
   };
   enum { NumMemSDNodeBits = NumSDNodeBits + 4 };
 
   class LSBaseSDNodeBitfields {
     friend class LSBaseSDNode;
-    friend class VPLoadStoreSDNode;
+    friend class VPBaseLoadStoreSDNode;
     friend class MaskedLoadStoreSDNode;
     friend class MaskedGatherScatterSDNode;
     friend class VPGatherScatterSDNode;
 
     uint16_t : NumMemSDNodeBits;
 
     // This storage is shared between disparate class hierarchies to hold an
     // enumeration specific to the class hierarchy in use.
     //   LSBaseSDNode => enum ISD::MemIndexedMode
     //   VPLoadStoreBaseSDNode => enum ISD::MemIndexedMode
     //   MaskedLoadStoreBaseSDNode => enum ISD::MemIndexedMode
     //   VPGatherScatterSDNode => enum ISD::MemIndexType
     //   MaskedGatherScatterSDNode => enum ISD::MemIndexType
     uint16_t AddressingMode : 3;
   };
   enum { NumLSBaseSDNodeBits = NumMemSDNodeBits + 3 };
 
   class LoadSDNodeBitfields {
     friend class LoadSDNode;
     friend class VPLoadSDNode;
+    friend class VPStridedLoadSDNode;
     friend class MaskedLoadSDNode;
     friend class MaskedGatherSDNode;
     friend class VPGatherSDNode;
 
     uint16_t : NumLSBaseSDNodeBits;
 
     uint16_t ExtTy : 2; // enum ISD::LoadExtType
     uint16_t IsExpanding : 1;
   };
 
   class StoreSDNodeBitfields {
     friend class StoreSDNode;
     friend class VPStoreSDNode;
+    friend class VPStridedStoreSDNode;
     friend class MaskedStoreSDNode;
     friend class MaskedScatterSDNode;
     friend class VPScatterSDNode;
 
     uint16_t : NumLSBaseSDNodeBits;
 
     uint16_t IsTruncating : 1;
     uint16_t IsCompressing : 1;
public:
   const SDValue &getChain() const { return getOperand(0); }
 
   const SDValue &getBasePtr() const {
     switch (getOpcode()) {
     case ISD::STORE:
     case ISD::VP_STORE:
     case ISD::MSTORE:
     case ISD::VP_SCATTER:
+    case ISD::EXPERIMENTAL_VP_STRIDED_STORE:
       return getOperand(2);
     case ISD::MGATHER:
     case ISD::MSCATTER:
       return getOperand(3);
     default:
       return getOperand(1);
     }
   }
static bool classof(const SDNode *N) {
     case ISD::MLOAD:
     case ISD::MSTORE:
     case ISD::MGATHER:
     case ISD::MSCATTER:
     case ISD::VP_LOAD:
     case ISD::VP_STORE:
     case ISD::VP_GATHER:
     case ISD::VP_SCATTER:
+    case ISD::EXPERIMENTAL_VP_STRIDED_LOAD:
+    case ISD::EXPERIMENTAL_VP_STRIDED_STORE:
       return true;
     default:
       return N->isMemIntrinsic() || N->isTargetMemoryOpcode();
     }
   }
 };
 
 /// This is an SDNode representing atomic operations.
public:
   const SDValue &getBasePtr() const { return getOperand(2); }
   const SDValue &getOffset() const { return getOperand(3); }
 
   static bool classof(const SDNode *N) {
     return N->getOpcode() == ISD::STORE;
   }
 };
 
-/// This base class is used to represent VP_LOAD and VP_STORE nodes
-class VPLoadStoreSDNode : public MemSDNode {
+/// This base class is used to represent VP_LOAD, VP_STORE,
+/// EXPERIMENTAL_VP_STRIDED_LOAD and EXPERIMENTAL_VP_STRIDED_STORE nodes
+class VPBaseLoadStoreSDNode : public MemSDNode {
 public:
   friend class SelectionDAG;
 
-  VPLoadStoreSDNode(ISD::NodeType NodeTy, unsigned Order, const DebugLoc &dl,
-                    SDVTList VTs, ISD::MemIndexedMode AM, EVT MemVT,
+  VPBaseLoadStoreSDNode(ISD::NodeType NodeTy, unsigned Order,
+                        const DebugLoc &DL, SDVTList VTs,
+                        ISD::MemIndexedMode AM, EVT MemVT,
                         MachineMemOperand *MMO)
-      : MemSDNode(NodeTy, Order, dl, VTs, MemVT, MMO) {
+      : MemSDNode(NodeTy, Order, DL, VTs, MemVT, MMO) {
     LSBaseSDNodeBits.AddressingMode = AM;
     assert(getAddressingMode() == AM && "Value truncated");
   }
 
-  // VPLoadSDNode (Chain, Ptr, Offset, Mask, EVL)
-  // VPStoreSDNode (Chain, Data, Ptr, Offset, Mask, EVL)
+  // VPStridedStoreSDNode (Chain, Data, Ptr,    Offset, Stride, Mask, EVL)
+  // VPStoreSDNode        (Chain, Data, Ptr,    Offset, Mask,   EVL)
+  // VPStridedLoadSDNode  (Chain, Ptr,  Offset, Stride, Mask,   EVL)
+  // VPLoadSDNode         (Chain, Ptr,  Offset, Mask,   EVL)
   // Mask is a vector of i1 elements;
   // the type of EVL is TLI.getVPExplicitVectorLengthTy().
   const SDValue &getOffset() const {
-    return getOperand(getOpcode() == ISD::VP_LOAD ? 2 : 3);
+    return getOperand((getOpcode() == ISD::EXPERIMENTAL_VP_STRIDED_LOAD ||
+                       getOpcode() == ISD::VP_LOAD)
+                          ? 2
+                          : 3);
   }
   const SDValue &getBasePtr() const {
-    return getOperand(getOpcode() == ISD::VP_LOAD ? 1 : 2);
+    return getOperand((getOpcode() == ISD::EXPERIMENTAL_VP_STRIDED_LOAD ||
+                       getOpcode() == ISD::VP_LOAD)
+                          ? 1
+                          : 2);
   }
   const SDValue &getMask() const {
-    return getOperand(getOpcode() == ISD::VP_LOAD ? 3 : 4);
+    switch (getOpcode()) {
+    default:
+      llvm_unreachable("Invalid opcode");
+    case ISD::VP_LOAD:
+      return getOperand(3);
+    case ISD::VP_STORE:
+    case ISD::EXPERIMENTAL_VP_STRIDED_LOAD:
+      return getOperand(4);
+    case ISD::EXPERIMENTAL_VP_STRIDED_STORE:
+      return getOperand(5);
+    }
   }
   const SDValue &getVectorLength() const {
-    return getOperand(getOpcode() == ISD::VP_LOAD ? 4 : 5);
+    switch (getOpcode()) {
+    default:
+      llvm_unreachable("Invalid opcode");
+    case ISD::VP_LOAD:
+      return getOperand(4);
+    case ISD::VP_STORE:
+    case ISD::EXPERIMENTAL_VP_STRIDED_LOAD:
+      return getOperand(5);
+    case ISD::EXPERIMENTAL_VP_STRIDED_STORE:
+      return getOperand(6);
+    }
   }
 
   /// Return the addressing mode for this load or store:
   /// unindexed, pre-inc, pre-dec, post-inc, or post-dec.
   ISD::MemIndexedMode getAddressingMode() const {
     return static_cast<ISD::MemIndexedMode>(LSBaseSDNodeBits.AddressingMode);
   }
 
   /// Return true if this is a pre/post inc/dec load/store.
   bool isIndexed() const { return getAddressingMode() != ISD::UNINDEXED; }
 
   /// Return true if this is NOT a pre/post inc/dec load/store.
   bool isUnindexed() const { return getAddressingMode() == ISD::UNINDEXED; }
 
   static bool classof(const SDNode *N) {
-    return N->getOpcode() == ISD::VP_LOAD || N->getOpcode() == ISD::VP_STORE;
+    return N->getOpcode() == ISD::EXPERIMENTAL_VP_STRIDED_LOAD ||
+           N->getOpcode() == ISD::EXPERIMENTAL_VP_STRIDED_STORE ||
+           N->getOpcode() == ISD::VP_LOAD || N->getOpcode() == ISD::VP_STORE;
   }
 };
 
 /// This class is used to represent a VP_LOAD node
-class VPLoadSDNode : public VPLoadStoreSDNode {
+class VPLoadSDNode : public VPBaseLoadStoreSDNode {
 public:
   friend class SelectionDAG;
 
   VPLoadSDNode(unsigned Order, const DebugLoc &dl, SDVTList VTs,
                ISD::MemIndexedMode AM, ISD::LoadExtType ETy, bool isExpanding,
                EVT MemVT, MachineMemOperand *MMO)
-      : VPLoadStoreSDNode(ISD::VP_LOAD, Order, dl, VTs, AM, MemVT, MMO) {
+      : VPBaseLoadStoreSDNode(ISD::VP_LOAD, Order, dl, VTs, AM, MemVT, MMO) {
     LoadSDNodeBits.ExtTy = ETy;
     LoadSDNodeBits.IsExpanding = isExpanding;
   }
 
   ISD::LoadExtType getExtensionType() const {
     return static_cast<ISD::LoadExtType>(LoadSDNodeBits.ExtTy);
   }
 
   const SDValue &getBasePtr() const { return getOperand(1); }
   const SDValue &getOffset() const { return getOperand(2); }
   const SDValue &getMask() const { return getOperand(3); }
   const SDValue &getVectorLength() const { return getOperand(4); }
 
   static bool classof(const SDNode *N) {
     return N->getOpcode() == ISD::VP_LOAD;
   }
   bool isExpandingLoad() const { return LoadSDNodeBits.IsExpanding; }
 };
 
+/// This class is used to represent an EXPERIMENTAL_VP_STRIDED_LOAD node.
+class VPStridedLoadSDNode : public VPBaseLoadStoreSDNode {
+public:
+  friend class SelectionDAG;
+
+  VPStridedLoadSDNode(unsigned Order, const DebugLoc &DL, SDVTList VTs,
+                      ISD::MemIndexedMode AM, ISD::LoadExtType ETy,
+                      bool IsExpanding, EVT MemVT, MachineMemOperand *MMO)
+      : VPBaseLoadStoreSDNode(ISD::EXPERIMENTAL_VP_STRIDED_LOAD, Order, DL, VTs,
+                              AM, MemVT, MMO) {
+    LoadSDNodeBits.ExtTy = ETy;
+    LoadSDNodeBits.IsExpanding = IsExpanding;
+  }
+
+  ISD::LoadExtType getExtensionType() const {
+    return static_cast<ISD::LoadExtType>(LoadSDNodeBits.ExtTy);
+  }
+
+  const SDValue &getBasePtr() const { return getOperand(1); }
+  const SDValue &getOffset() const { return getOperand(2); }
+  const SDValue &getStride() const { return getOperand(3); }
+  const SDValue &getMask() const { return getOperand(4); }
+  const SDValue &getVectorLength() const { return getOperand(5); }
+
+  static bool classof(const SDNode *N) {
+    return N->getOpcode() == ISD::EXPERIMENTAL_VP_STRIDED_LOAD;
+  }
+  bool isExpandingLoad() const { return LoadSDNodeBits.IsExpanding; }
+};
+
 /// This class is used to represent a VP_STORE node
-class VPStoreSDNode : public VPLoadStoreSDNode {
+class VPStoreSDNode : public VPBaseLoadStoreSDNode {
 public:
   friend class SelectionDAG;
 
   VPStoreSDNode(unsigned Order, const DebugLoc &dl, SDVTList VTs,
                 ISD::MemIndexedMode AM, bool isTrunc, bool isCompressing,
                 EVT MemVT, MachineMemOperand *MMO)
-      : VPLoadStoreSDNode(ISD::VP_STORE, Order, dl, VTs, AM, MemVT, MMO) {
+      : VPBaseLoadStoreSDNode(ISD::VP_STORE, Order, dl, VTs, AM, MemVT, MMO) {
     StoreSDNodeBits.IsTruncating = isTrunc;
     StoreSDNodeBits.IsCompressing = isCompressing;
   }
 
   /// Return true if this is a truncating store.
   /// For integers this is the same as doing a TRUNCATE and storing the result.
   /// For floats, it is the same as doing an FP_ROUND and storing the result.
   bool isTruncatingStore() const { return StoreSDNodeBits.IsTruncating; }
public:
   const SDValue &getMask() const { return getOperand(4); }
   const SDValue &getVectorLength() const { return getOperand(5); }
 
   static bool classof(const SDNode *N) {
     return N->getOpcode() == ISD::VP_STORE;
   }
 };
 
+/// This class is used to represent an EXPERIMENTAL_VP_STRIDED_STORE node.
+class VPStridedStoreSDNode : public VPBaseLoadStoreSDNode {
+public:
+  friend class SelectionDAG;
+
+  VPStridedStoreSDNode(unsigned Order, const DebugLoc &DL, SDVTList VTs,
+                       ISD::MemIndexedMode AM, bool IsTrunc, bool IsCompressing,
+                       EVT MemVT, MachineMemOperand *MMO)
+      : VPBaseLoadStoreSDNode(ISD::EXPERIMENTAL_VP_STRIDED_STORE, Order, DL,
+                              VTs, AM, MemVT, MMO) {
+    StoreSDNodeBits.IsTruncating = IsTrunc;
+    StoreSDNodeBits.IsCompressing = IsCompressing;
+  }
+
+  /// Return true if this is a truncating store.
+  /// For integers this is the same as doing a TRUNCATE and storing the result.
+  /// For floats, it is the same as doing an FP_ROUND and storing the result.
+  bool isTruncatingStore() const { return StoreSDNodeBits.IsTruncating; }
+
+  /// Returns true if the op does a compression to the vector before storing.
+  /// The node contiguously stores the active elements (integers or floats)
+  /// in src (those with their respective bit set in writemask k) to unaligned
+  /// memory at base_addr.
+  bool isCompressingStore() const { return StoreSDNodeBits.IsCompressing; }
+
+  const SDValue &getValue() const { return getOperand(1); }
+  const SDValue &getBasePtr() const { return getOperand(2); }
+  const SDValue &getOffset() const { return getOperand(3); }
+  const SDValue &getStride() const { return getOperand(4); }
+  const SDValue &getMask() const { return getOperand(5); }
+  const SDValue &getVectorLength() const { return getOperand(6); }
+
+  static bool classof(const SDNode *N) {
+    return N->getOpcode() == ISD::EXPERIMENTAL_VP_STRIDED_STORE;
+  }
+};
+
 /// This base class is used to represent MLOAD and MSTORE nodes
 class MaskedLoadStoreSDNode : public MemSDNode {
 public:
   friend class SelectionDAG;
 
   MaskedLoadStoreSDNode(ISD::NodeType NodeTy, unsigned Order,
                         const DebugLoc &dl, SDVTList VTs,
                         ISD::MemIndexedMode AM, EVT MemVT,

llvm/include/llvm/IR/Intrinsics.td

 
 def int_vp_scatter: DefaultAttrsIntrinsic<[],
                               [ llvm_anyvector_ty,
                                 LLVMVectorOfAnyPointersToElt<0>,
                                 LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                 llvm_i32_ty],
                               [ IntrArgMemOnly, IntrNoSync, IntrWillReturn ]>; // TODO allow IntrNoCapture for vectors of pointers
 
+// Experimental strided memory accesses
+def int_experimental_vp_strided_store : DefaultAttrsIntrinsic<[],
+                             [ llvm_anyvector_ty,
+                               LLVMAnyPointerType<LLVMMatchType<0>>,
+                               llvm_anyint_ty, // Stride in bytes
+                               LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                               llvm_i32_ty],
+                             [ NoCapture<ArgIndex<1>>, IntrNoSync, IntrWriteMem, IntrArgMemOnly, IntrWillReturn ]>;
+
+def int_experimental_vp_strided_load  : DefaultAttrsIntrinsic<[llvm_anyvector_ty],
+                             [ LLVMAnyPointerType<LLVMMatchType<0>>,
+                               llvm_anyint_ty, // Stride in bytes
+                               LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                               llvm_i32_ty],
+                             [ NoCapture<ArgIndex<0>>, IntrNoSync, IntrReadMem, IntrWillReturn, IntrArgMemOnly ]>;
+
 // Speculatable Binary operators
 let IntrProperties = [IntrSpeculatable, IntrNoMem, IntrNoSync, IntrWillReturn] in {
   def int_vp_add : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
                              [ LLVMMatchType<0>,
                                LLVMMatchType<0>,
                                LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                llvm_i32_ty]>;
   def int_vp_sub : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],

llvm/include/llvm/IR/VPIntrinsics.def

 // llvm.vp.frem(x,y,mask,vlen)
 HELPER_REGISTER_BINARY_FP_VP(frem, VP_FREM, FRem)
 
 #undef HELPER_REGISTER_BINARY_FP_VP
 
 ///// } Floating-Point Arithmetic
 
 ///// Memory Operations {
 // llvm.vp.store(val,ptr,mask,vlen)
 BEGIN_REGISTER_VP_INTRINSIC(vp_store, 2, 3)
 // chain = VP_STORE chain,val,base,offset,mask,evl
 BEGIN_REGISTER_VP_SDNODE(VP_STORE, 0, vp_store, 4, 5)
 VP_PROPERTY_FUNCTIONAL_OPC(Store)
 VP_PROPERTY_FUNCTIONAL_INTRINSIC(masked_store)
 VP_PROPERTY_MEMOP(1, 0)
 END_REGISTER_VP(vp_store, VP_STORE)
 
+// llvm.experimental.vp.strided.store(val,ptr,stride,mask,vlen)
+BEGIN_REGISTER_VP_INTRINSIC(experimental_vp_strided_store, 3, 4)
+// chain = EXPERIMENTAL_VP_STRIDED_STORE chain,val,base,offset,stride,mask,evl
+BEGIN_REGISTER_VP_SDNODE(EXPERIMENTAL_VP_STRIDED_STORE, 0, experimental_vp_strided_store, 5, 6)
+VP_PROPERTY_MEMOP(1, 0)
+END_REGISTER_VP(experimental_vp_strided_store, EXPERIMENTAL_VP_STRIDED_STORE)
+
 // llvm.vp.scatter(ptr,val,mask,vlen)
 BEGIN_REGISTER_VP_INTRINSIC(vp_scatter, 2, 3)
 // chain = VP_SCATTER chain,val,base,indices,scale,mask,evl
 BEGIN_REGISTER_VP_SDNODE(VP_SCATTER, -1, vp_scatter, 5, 6)
 VP_PROPERTY_FUNCTIONAL_INTRINSIC(masked_scatter)
 VP_PROPERTY_MEMOP(1, 0)
 END_REGISTER_VP(vp_scatter, VP_SCATTER)
 
 // llvm.vp.load(ptr,mask,vlen)
 BEGIN_REGISTER_VP_INTRINSIC(vp_load, 1, 2)
 // val,chain = VP_LOAD chain,base,offset,mask,evl
 BEGIN_REGISTER_VP_SDNODE(VP_LOAD, -1, vp_load, 3, 4)
 VP_PROPERTY_FUNCTIONAL_OPC(Load)
 VP_PROPERTY_FUNCTIONAL_INTRINSIC(masked_load)
 VP_PROPERTY_MEMOP(0, None)
 END_REGISTER_VP(vp_load, VP_LOAD)
 
+// llvm.experimental.vp.strided.load(ptr,stride,mask,vlen)
+BEGIN_REGISTER_VP_INTRINSIC(experimental_vp_strided_load, 2, 3)
+// chain = EXPERIMENTAL_VP_STRIDED_LOAD chain,base,offset,stride,mask,evl
+BEGIN_REGISTER_VP_SDNODE(EXPERIMENTAL_VP_STRIDED_LOAD, -1, experimental_vp_strided_load, 4, 5)
+VP_PROPERTY_MEMOP(0, None)
+END_REGISTER_VP(experimental_vp_strided_load, EXPERIMENTAL_VP_STRIDED_LOAD)
+
 // llvm.vp.gather(ptr,mask,vlen)
 BEGIN_REGISTER_VP_INTRINSIC(vp_gather, 1, 2)
 // val,chain = VP_GATHER chain,base,indices,scale,mask,evl
 BEGIN_REGISTER_VP_SDNODE(VP_GATHER, -1, vp_gather, 4, 5)
 VP_PROPERTY_FUNCTIONAL_INTRINSIC(masked_gather)
 VP_PROPERTY_MEMOP(0, None)
 END_REGISTER_VP(vp_gather, VP_GATHER)
 

llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp

Action = TLI.getOperationAction(
         Node->getOpcode(),
         cast<VPScatterSDNode>(Node)->getValue().getValueType());
     break;
   case ISD::VP_STORE:
     Action = TLI.getOperationAction(
         Node->getOpcode(),
         cast<VPStoreSDNode>(Node)->getValue().getValueType());
     break;
+  case ISD::EXPERIMENTAL_VP_STRIDED_STORE:
+    Action = TLI.getOperationAction(
+        Node->getOpcode(),
+        cast<VPStridedStoreSDNode>(Node)->getValue().getValueType());
+    break;
   case ISD::VECREDUCE_FADD:
   case ISD::VECREDUCE_FMUL:
   case ISD::VECREDUCE_ADD:
   case ISD::VECREDUCE_MUL:
   case ISD::VECREDUCE_AND:
   case ISD::VECREDUCE_OR:
   case ISD::VECREDUCE_XOR:
   case ISD::VECREDUCE_SMAX:

llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp

static void AddNodeIDCustom(FoldingSetNodeID &ID, const SDNode *N) {
   }
   case ISD::VP_STORE: {
     const VPStoreSDNode *EST = cast<VPStoreSDNode>(N);
     ID.AddInteger(EST->getMemoryVT().getRawBits());
     ID.AddInteger(EST->getRawSubclassData());
     ID.AddInteger(EST->getPointerInfo().getAddrSpace());
     break;
   }
+  case ISD::EXPERIMENTAL_VP_STRIDED_LOAD: {
+    const VPStridedLoadSDNode *SLD = cast<VPStridedLoadSDNode>(N);
+    ID.AddInteger(SLD->getMemoryVT().getRawBits());
+    ID.AddInteger(SLD->getRawSubclassData());
+    ID.AddInteger(SLD->getPointerInfo().getAddrSpace());
+    break;
+  }
+  case ISD::EXPERIMENTAL_VP_STRIDED_STORE: {
+    const VPStridedStoreSDNode *SST = cast<VPStridedStoreSDNode>(N);
+    ID.AddInteger(SST->getMemoryVT().getRawBits());
+    ID.AddInteger(SST->getRawSubclassData());
+    ID.AddInteger(SST->getPointerInfo().getAddrSpace());
+    break;
+  }
   case ISD::VP_GATHER: {
     const VPGatherSDNode *EG = cast<VPGatherSDNode>(N);
     ID.AddInteger(EG->getMemoryVT().getRawBits());
     ID.AddInteger(EG->getRawSubclassData());
     ID.AddInteger(EG->getPointerInfo().getAddrSpace());
     break;
   }
   case ISD::VP_SCATTER: {
SDValue SelectionDAG::getIndexedStoreVP(SDValue OrigStore, const SDLoc &dl,
 
   CSEMap.InsertNode(N, IP);
   InsertNode(N);
   SDValue V(N, 0);
   NewSDValueDbgMsg(V, "Creating new node: ", this);
   return V;
 }
 
+SDValue SelectionDAG::getStridedLoadVP(
+    ISD::MemIndexedMode AM, ISD::LoadExtType ExtType, EVT VT, const SDLoc &DL,
+    SDValue Chain, SDValue Ptr, SDValue Offset, SDValue Stride, SDValue Mask,
+    SDValue EVL, MachinePointerInfo PtrInfo, EVT MemVT, Align Alignment,
+    MachineMemOperand::Flags MMOFlags, const AAMDNodes &AAInfo,
+    const MDNode *Ranges, bool IsExpanding) {
+  assert(Chain.getValueType() == MVT::Other && "Invalid chain type");
+
+  MMOFlags |= MachineMemOperand::MOLoad;
+  assert((MMOFlags & MachineMemOperand::MOStore) == 0);
+  // If we don't have a PtrInfo, infer the trivial frame index case to simplify
+  // clients.
+  if (PtrInfo.V.isNull())
+    PtrInfo = InferPointerInfo(PtrInfo, *this, Ptr, Offset);
+
+  uint64_t Size = MemoryLocation::UnknownSize;
+  MachineFunction &MF = getMachineFunction();
+  MachineMemOperand *MMO = MF.getMachineMemOperand(PtrInfo, MMOFlags, Size,
+                                                   Alignment, AAInfo, Ranges);
+  return getStridedLoadVP(AM, ExtType, VT, DL, Chain, Ptr, Offset, Stride, Mask,
+                          EVL, MemVT, MMO, IsExpanding);
+}
+
+SDValue SelectionDAG::getStridedLoadVP(
+    ISD::MemIndexedMode AM, ISD::LoadExtType ExtType, EVT VT, const SDLoc &DL,
+    SDValue Chain, SDValue Ptr, SDValue Offset, SDValue Stride, SDValue Mask,
+    SDValue EVL, EVT MemVT, MachineMemOperand *MMO, bool IsExpanding) {
+  if (VT == MemVT) {
+    ExtType = ISD::NON_EXTLOAD;
+  } else if (ExtType == ISD::NON_EXTLOAD) {
+    assert(VT == MemVT && "Non-extending load from different memory type!");
+  } else {
+    // Extending load.
+    assert(MemVT.getScalarType().bitsLT(VT.getScalarType()) &&
+           "Should only be an extending load, not truncating!");
+    assert(VT.isInteger() == MemVT.isInteger() &&
+           "Cannot convert from FP to Int or Int -> FP!");
+    assert(VT.isVector() == MemVT.isVector() &&
+           "Cannot use an ext load to convert to or from a vector!");
+    assert((!VT.isVector() ||
+            VT.getVectorElementCount() == MemVT.getVectorElementCount()) &&
+           "Cannot use an ext load to change the number of vector elements!");
+  }
+
+  bool Indexed = AM != ISD::UNINDEXED;
+  assert((Indexed || Offset.isUndef()) && "Unindexed load with an offset!");
+
+  SDValue Ops[] = {Chain, Ptr, Offset, Stride, Mask, EVL};
+  SDVTList VTs = Indexed ? getVTList(VT, Ptr.getValueType(), MVT::Other)
+                         : getVTList(VT, MVT::Other);
+  FoldingSetNodeID ID;
+  AddNodeIDNode(ID, ISD::EXPERIMENTAL_VP_STRIDED_LOAD, VTs, Ops);
+  ID.AddInteger(VT.getRawBits());
+  ID.AddInteger(getSyntheticNodeSubclassData<VPStridedLoadSDNode>(
+      DL.getIROrder(), VTs, AM, ExtType, IsExpanding, MemVT, MMO));
+  ID.AddInteger(MMO->getPointerInfo().getAddrSpace());
+
+  void *IP = nullptr;
+  if (SDNode *E = FindNodeOrInsertPos(ID, DL, IP)) {
+    cast<VPStridedLoadSDNode>(E)->refineAlignment(MMO);
+    return SDValue(E, 0);
+  }
+
+  auto *N =
+      newSDNode<VPStridedLoadSDNode>(DL.getIROrder(), DL.getDebugLoc(), VTs, AM,
+                                     ExtType, IsExpanding, MemVT, MMO);
+  createOperands(N, Ops);
+  CSEMap.InsertNode(N, IP);
+  InsertNode(N);
+  SDValue V(N, 0);
+  NewSDValueDbgMsg(V, "Creating new node: ", this);
+  return V;
+}
+
+SDValue SelectionDAG::getStridedLoadVP(
+    EVT VT, const SDLoc &DL, SDValue Chain, SDValue Ptr, SDValue Stride,
+    SDValue Mask, SDValue EVL, MachinePointerInfo PtrInfo, MaybeAlign Alignment,
+    MachineMemOperand::Flags MMOFlags, const AAMDNodes &AAInfo,
+    const MDNode *Ranges, bool IsExpanding) {
+  SDValue Undef = getUNDEF(Ptr.getValueType());
+  return getStridedLoadVP(ISD::UNINDEXED, ISD::NON_EXTLOAD, VT, DL, Chain, Ptr,
+                          Undef, Stride, Mask, EVL, PtrInfo, VT, Alignment,
+                          MMOFlags, AAInfo, Ranges, IsExpanding);
+}
+
+SDValue SelectionDAG::getStridedLoadVP(EVT VT, const SDLoc &DL, SDValue Chain,
+                                       SDValue Ptr, SDValue Stride,
+                                       SDValue Mask, SDValue EVL,
+                                       MachineMemOperand *MMO,
+                                       bool IsExpanding) {
+  SDValue Undef = getUNDEF(Ptr.getValueType());
+  return getStridedLoadVP(ISD::UNINDEXED, ISD::NON_EXTLOAD, VT, DL, Chain, Ptr,
+                          Undef, Stride, Mask, EVL, VT, MMO, IsExpanding);
+}
+
+SDValue SelectionDAG::getExtStridedLoadVP(
+    ISD::LoadExtType ExtType, const SDLoc &DL, EVT VT, SDValue Chain,
+    SDValue Ptr, SDValue Stride, SDValue Mask, SDValue EVL,
+    MachinePointerInfo PtrInfo, EVT MemVT, MaybeAlign Alignment,
+    MachineMemOperand::Flags MMOFlags, const AAMDNodes &AAInfo,
+    bool IsExpanding) {
+  SDValue Undef = getUNDEF(Ptr.getValueType());
+  return getStridedLoadVP(ISD::UNINDEXED, ExtType, VT, DL, Chain, Ptr, Undef,
+                          Stride, Mask, EVL, PtrInfo, MemVT, Alignment,
+                          MMOFlags, AAInfo, nullptr, IsExpanding);
+}
+
+SDValue SelectionDAG::getExtStridedLoadVP(
+    ISD::LoadExtType ExtType, const SDLoc &DL, EVT VT, SDValue Chain,
+    SDValue Ptr, SDValue Stride, SDValue Mask, SDValue EVL, EVT MemVT,
+    MachineMemOperand *MMO, bool IsExpanding) {
+  SDValue Undef = getUNDEF(Ptr.getValueType());
+  return getStridedLoadVP(ISD::UNINDEXED, ExtType, VT, DL, Chain, Ptr, Undef,
+                          Stride, Mask, EVL, MemVT, MMO, IsExpanding);
+}
+
+SDValue SelectionDAG::getIndexedStridedLoadVP(SDValue OrigLoad, const SDLoc &DL,
+                                              SDValue Base, SDValue Offset,
+                                              ISD::MemIndexedMode AM) {
+  auto *SLD = cast<VPStridedLoadSDNode>(OrigLoad);
+  assert(SLD->getOffset().isUndef() &&
+         "Strided load is already a indexed load!");
+  // Don't propagate the invariant or dereferenceable flags.
+  auto MMOFlags =
+      SLD->getMemOperand()->getFlags() &
+      ~(MachineMemOperand::MOInvariant | MachineMemOperand::MODereferenceable);
+  return getStridedLoadVP(
+      AM, SLD->getExtensionType(), OrigLoad.getValueType(), DL, SLD->getChain(),
+      Base, Offset, SLD->getStride(), SLD->getMask(), SLD->getVectorLength(),
+      SLD->getPointerInfo(), SLD->getMemoryVT(), SLD->getAlign(), MMOFlags,
+      SLD->getAAInfo(), nullptr, SLD->isExpandingLoad());
+}
+
+SDValue SelectionDAG::getStridedStoreVP(
+    SDValue Chain, const SDLoc &DL, SDValue Val, SDValue Ptr, SDValue Stride,
+    SDValue Mask, SDValue EVL, MachinePointerInfo PtrInfo, Align Alignment,
+    MachineMemOperand::Flags MMOFlags, const AAMDNodes &AAInfo,
+    bool IsCompressing) {
+  assert(Chain.getValueType() == MVT::Other && "Invalid chain type");
+
+  MMOFlags |= MachineMemOperand::MOStore;
+  assert((MMOFlags & MachineMemOperand::MOLoad) == 0);
+
+  if (PtrInfo.V.isNull())
+    PtrInfo = InferPointerInfo(PtrInfo, *this, Ptr);
+
+  MachineFunction &MF = getMachineFunction();
+  uint64_t Size = MemoryLocation::UnknownSize;
+  MachineMemOperand *MMO =
+      MF.getMachineMemOperand(PtrInfo, MMOFlags, Size, Alignment, AAInfo);
+  return getStridedStoreVP(Chain, DL, Val, Ptr, Stride, Mask, EVL, MMO,
+                           IsCompressing);
+}
+
+SDValue SelectionDAG::getStridedStoreVP(SDValue Chain, const SDLoc &DL,
+                                        SDValue Val, SDValue Ptr,
+                                        SDValue Stride, SDValue Mask,
+                                        SDValue EVL, MachineMemOperand *MMO,
+                                        bool IsCompressing) {
+  assert(Chain.getValueType() == MVT::Other && "Invalid chain type");
+  EVT VT = Val.getValueType();
+  SDVTList VTs = getVTList(MVT::Other);
+  SDValue Undef = getUNDEF(Ptr.getValueType());
+  SDValue Ops[] = {Chain, Val, Ptr, Undef, Stride, Mask, EVL};
+  FoldingSetNodeID ID;
+  AddNodeIDNode(ID, ISD::EXPERIMENTAL_VP_STRIDED_STORE, VTs, Ops);
+  ID.AddInteger(VT.getRawBits());
+  ID.AddInteger(getSyntheticNodeSubclassData<VPStridedStoreSDNode>(
+      DL.getIROrder(), VTs, ISD::UNINDEXED, false, IsCompressing, VT, MMO));
+  ID.AddInteger(MMO->getPointerInfo().getAddrSpace());
+
+  void *IP = nullptr;
+  if (SDNode *E = FindNodeOrInsertPos(ID, DL, IP)) {
+    cast<VPStridedStoreSDNode>(E)->refineAlignment(MMO);
+    return SDValue(E, 0);
+  }
+
+  auto *N = newSDNode<VPStridedStoreSDNode>(DL.getIROrder(), DL.getDebugLoc(),
+                                            VTs, ISD::UNINDEXED, false,
+                                            IsCompressing, VT, MMO);
+  createOperands(N, Ops);
+  CSEMap.InsertNode(N, IP);
+  InsertNode(N);
+  SDValue V(N, 0);
+  NewSDValueDbgMsg(V, "Creating new node: ", this);
+  return V;
+}
+
+SDValue SelectionDAG::getTruncStridedStoreVP(
+    SDValue Chain, const SDLoc &DL, SDValue Val, SDValue Ptr, SDValue Stride,
+    SDValue Mask, SDValue EVL, MachinePointerInfo PtrInfo, EVT SVT,
+    Align Alignment, MachineMemOperand::Flags MMOFlags, const AAMDNodes &AAInfo,
+    bool IsCompressing) {
+  assert(Chain.getValueType() == MVT::Other && "Invalid chain type");
+
+  MMOFlags |= MachineMemOperand::MOStore;
+  assert((MMOFlags & MachineMemOperand::MOLoad) == 0);
+
+  if (PtrInfo.V.isNull())
+    PtrInfo = InferPointerInfo(PtrInfo, *this, Ptr);
+
+  MachineFunction &MF = getMachineFunction();
+  MachineMemOperand *MMO = MF.getMachineMemOperand(
+      PtrInfo, MMOFlags, MemoryLocation::UnknownSize, Alignment, AAInfo);
+  return getTruncStridedStoreVP(Chain, DL, Val, Ptr, Stride, Mask, EVL, SVT,
+                                MMO, IsCompressing);
+}
+
+SDValue SelectionDAG::getTruncStridedStoreVP(SDValue Chain, const SDLoc &DL,
+                                             SDValue Val, SDValue Ptr,
+                                             SDValue Stride, SDValue Mask,
+                                             SDValue EVL, EVT SVT,
+                                             MachineMemOperand *MMO,
+                                             bool IsCompressing) {
+  EVT VT = Val.getValueType();
+
+  assert(Chain.getValueType() == MVT::Other && "Invalid chain type");
+  if (VT == SVT)
+    return getStridedStoreVP(Chain, DL, Val, Ptr, Stride, Mask, EVL, MMO,
+                             IsCompressing);
+
+  assert(SVT.getScalarType().bitsLT(VT.getScalarType()) &&
+         "Should only be a truncating store, not extending!");
+  assert(VT.isInteger() == SVT.isInteger() && "Can't do FP-INT conversion!");
+  assert(VT.isVector() == SVT.isVector() &&
+         "Cannot use trunc store to convert to or from a vector!");
+  assert((!VT.isVector() ||
+          VT.getVectorElementCount() == SVT.getVectorElementCount()) &&
+         "Cannot use trunc store to change the number of vector elements!");
+
+  SDVTList VTs = getVTList(MVT::Other);
+  SDValue Undef = getUNDEF(Ptr.getValueType());
+  SDValue Ops[] = {Chain, Val, Ptr, Undef, Stride, Mask, EVL};
+  FoldingSetNodeID ID;
+  AddNodeIDNode(ID, ISD::EXPERIMENTAL_VP_STRIDED_STORE, VTs, Ops);
+  ID.AddInteger(SVT.getRawBits());
+  ID.AddInteger(getSyntheticNodeSubclassData<VPStridedStoreSDNode>(
+      DL.getIROrder(), VTs, ISD::UNINDEXED, true, IsCompressing, SVT, MMO));
+  ID.AddInteger(MMO->getPointerInfo().getAddrSpace());
+  void *IP = nullptr;
+  if (SDNode *E = FindNodeOrInsertPos(ID, DL, IP)) {
+    cast<VPStridedStoreSDNode>(E)->refineAlignment(MMO);
+    return SDValue(E, 0);
+  }
+  auto *N = newSDNode<VPStridedStoreSDNode>(DL.getIROrder(), DL.getDebugLoc(),
+                                            VTs, ISD::UNINDEXED, true,
+                                            IsCompressing, SVT, MMO);
+  createOperands(N, Ops);
+
+  CSEMap.InsertNode(N, IP);
+  InsertNode(N);
+  SDValue V(N, 0);
+  NewSDValueDbgMsg(V, "Creating new node: ", this);
+  return V;
+}
+
+SDValue SelectionDAG::getIndexedStridedStoreVP(SDValue OrigStore,
+                                               const SDLoc &DL, SDValue Base,
+                                               SDValue Offset,
+                                               ISD::MemIndexedMode AM) {
+  auto *SST = cast<VPStridedStoreSDNode>(OrigStore);
+  assert(SST->getOffset().isUndef() &&
+         "Strided store is already an indexed store!");
+  SDVTList VTs = getVTList(Base.getValueType(), MVT::Other);
+  SDValue Ops[] = {
+      SST->getChain(), SST->getValue(),       Base, Offset, SST->getStride(),
+      SST->getMask(),  SST->getVectorLength()};
+  FoldingSetNodeID ID;
+  AddNodeIDNode(ID, ISD::EXPERIMENTAL_VP_STRIDED_STORE, VTs, Ops);
+  ID.AddInteger(SST->getMemoryVT().getRawBits());
+  ID.AddInteger(SST->getRawSubclassData());
+  ID.AddInteger(SST->getPointerInfo().getAddrSpace());
+  void *IP = nullptr;
+  if (SDNode *E = FindNodeOrInsertPos(ID, DL, IP))
+    return SDValue(E, 0);
+
+  auto *N = newSDNode<VPStridedStoreSDNode>(
+      DL.getIROrder(), DL.getDebugLoc(), VTs, AM, SST->isTruncatingStore(),
+      SST->isCompressingStore(), SST->getMemoryVT(), SST->getMemOperand());
+  createOperands(N, Ops);
+
+  CSEMap.InsertNode(N, IP);
+  InsertNode(N);
+  SDValue V(N, 0);
+  NewSDValueDbgMsg(V, "Creating new node: ", this);
+  return V;
+}
+
 SDValue SelectionDAG::getGatherVP(SDVTList VTs, EVT VT, const SDLoc &dl,
                                   ArrayRef<SDValue> Ops, MachineMemOperand *MMO,
                                   ISD::MemIndexType IndexType) {
   assert(Ops.size() == 6 && "Incompatible number of operands");
 
   FoldingSetNodeID ID;
   AddNodeIDNode(ID, ISD::VP_GATHER, VTs, Ops);
   ID.AddInteger(VT.getRawBits());

llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.h

void visitInlineAsm(const CallBase &Call,
                       const BasicBlock *EHPadBB = nullptr);
   void visitIntrinsicCall(const CallInst &I, unsigned Intrinsic);
   void visitTargetIntrinsic(const CallInst &I, unsigned Intrinsic);
   void visitConstrainedFPIntrinsic(const ConstrainedFPIntrinsic &FPI);
   void visitVPLoadGather(const VPIntrinsic &VPIntrin, EVT VT,
                          SmallVector<SDValue, 7> &OpValues, bool IsGather);
   void visitVPStoreScatter(const VPIntrinsic &VPIntrin,
                            SmallVector<SDValue, 7> &OpValues, bool IsScatter);
+  void visitVPStridedLoad(const VPIntrinsic &VPIntrin, EVT VT,
+                          SmallVectorImpl<SDValue> &OpValues);
+  void visitVPStridedStore(const VPIntrinsic &VPIntrin,
+                           SmallVectorImpl<SDValue> &OpValues);
   void visitVectorPredicationIntrinsic(const VPIntrinsic &VPIntrin);
 
   void visitVAStart(const CallInst &I);
   void visitVAArg(const VAArgInst &I);
   void visitVAEnd(const CallInst &I);
   void visitVACopy(const CallInst &I);
   void visitStackmap(const CallInst &I);
   void visitPatchpoint(const CallBase &CB, const BasicBlock *EHPadBB = nullptr);

llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

ST = DAG.getScatterVP(DAG.getVTList(MVT::Other), VT, DL,
                           {getMemoryRoot(), OpValues[0], Base, Index, Scale,
                            OpValues[2], OpValues[3]},
                           MMO, IndexType);
   }
   DAG.setRoot(ST);
   setValue(&VPIntrin, ST);
 }
 
+void SelectionDAGBuilder::visitVPStridedLoad(
+    const VPIntrinsic &VPIntrin, EVT VT, SmallVectorImpl<SDValue> &OpValues) {
+  SDLoc DL = getCurSDLoc();
+  Value *PtrOperand = VPIntrin.getArgOperand(0);
+  MaybeAlign Alignment = VPIntrin.getPointerAlignment();
+  if (!Alignment)
+    Alignment = DAG.getEVTAlign(VT.getScalarType());
+  AAMDNodes AAInfo = VPIntrin.getAAMetadata();
+  const MDNode *Ranges = VPIntrin.getMetadata(LLVMContext::MD_range);
+  MemoryLocation ML = MemoryLocation::getAfter(PtrOperand, AAInfo);
+  bool AddToChain = !AA || !AA->pointsToConstantMemory(ML);
+  SDValue InChain = AddToChain ? DAG.getRoot() : DAG.getEntryNode();
+  MachineMemOperand *MMO = DAG.getMachineFunction().getMachineMemOperand(
+      MachinePointerInfo(PtrOperand), MachineMemOperand::MOLoad,
+      MemoryLocation::UnknownSize, *Alignment, AAInfo, Ranges);
+
+  SDValue LD = DAG.getStridedLoadVP(VT, DL, InChain, OpValues[0], OpValues[1],
+                                    OpValues[2], OpValues[3], MMO,
+                                    false /*IsExpanding*/);
+
+  if (AddToChain)
+    PendingLoads.push_back(LD.getValue(1));
+  setValue(&VPIntrin, LD);
+}
+
+void SelectionDAGBuilder::visitVPStridedStore(
+    const VPIntrinsic &VPIntrin, SmallVectorImpl<SDValue> &OpValues) {
+  SDLoc DL = getCurSDLoc();
+  Value *PtrOperand = VPIntrin.getArgOperand(1);
+  EVT VT = OpValues[0].getValueType();
+  MaybeAlign Alignment = VPIntrin.getPointerAlignment();
+  if (!Alignment)
+    Alignment = DAG.getEVTAlign(VT.getScalarType());
+  AAMDNodes AAInfo = VPIntrin.getAAMetadata();
+  MachineMemOperand *MMO = DAG.getMachineFunction().getMachineMemOperand(
+      MachinePointerInfo(PtrOperand), MachineMemOperand::MOStore,
+      MemoryLocation::UnknownSize, *Alignment, AAInfo);
+
+  SDValue ST = DAG.getStridedStoreVP(
+      getMemoryRoot(), DL, OpValues[0], OpValues[1], OpValues[2], OpValues[3],
+      OpValues[4], MMO, false /* IsCompressing */);
+
+  DAG.setRoot(ST);
+  setValue(&VPIntrin, ST);
+}
+
 void SelectionDAGBuilder::visitVectorPredicationIntrinsic(
     const VPIntrinsic &VPIntrin) {
   SDLoc DL = getCurSDLoc();
   unsigned Opcode = getISDForVPIntrinsic(VPIntrin);
 
   SmallVector<EVT, 4> ValueVTs;
   const TargetLowering &TLI = DAG.getTargetLoweringInfo();
   ComputeValueVTs(TLI, DAG.getDataLayout(), VPIntrin.getType(), ValueVTs);
default: {
     setValue(&VPIntrin, Result);
     break;
   }
   case ISD::VP_LOAD:
   case ISD::VP_GATHER:
     visitVPLoadGather(VPIntrin, ValueVTs[0], OpValues,
                       Opcode == ISD::VP_GATHER);
     break;
+  case ISD::EXPERIMENTAL_VP_STRIDED_LOAD:
+    visitVPStridedLoad(VPIntrin, ValueVTs[0], OpValues);
+    break;
   case ISD::VP_STORE:
   case ISD::VP_SCATTER:
     visitVPStoreScatter(VPIntrin, OpValues, Opcode == ISD::VP_SCATTER);
     break;
+  case ISD::EXPERIMENTAL_VP_STRIDED_STORE:
+    visitVPStridedStore(VPIntrin, OpValues);
+    break;
   }
 }
 
 SDValue SelectionDAGBuilder::lowerStartEH(SDValue Chain,
                                           const BasicBlock *EHPadBB,
                                           MCSymbol *&BeginLabel) {
   MachineFunction &MF = DAG.getMachineFunction();
   MachineModuleInfo &MMI = MF.getMMI();

llvm/lib/IR/IntrinsicInst.cpp

Function *VPIntrinsic::getDeclarationForParams(Module *M, Intrinsic::ID VPID,
   }
   case Intrinsic::vp_select:
     VPFunc = Intrinsic::getDeclaration(M, VPID, {Params[1]->getType()});
     break;
   case Intrinsic::vp_load:
     VPFunc = Intrinsic::getDeclaration(
         M, VPID, {ReturnType, Params[0]->getType()});
     break;
+  case Intrinsic::experimental_vp_strided_load:
+    VPFunc = Intrinsic::getDeclaration(
+        M, VPID, {ReturnType, Params[0]->getType(), Params[1]->getType()});
+    break;
   case Intrinsic::vp_gather:
     VPFunc = Intrinsic::getDeclaration(
         M, VPID, {ReturnType, Params[0]->getType()});
     break;
   case Intrinsic::vp_store:
     VPFunc = Intrinsic::getDeclaration(
         M, VPID, {Params[0]->getType(), Params[1]->getType()});
     break;
+  case Intrinsic::experimental_vp_strided_store:
+    VPFunc = Intrinsic::getDeclaration(
+        M, VPID,
+        {Params[0]->getType(), Params[1]->getType(), Params[2]->getType()});
+    break;
   case Intrinsic::vp_scatter:
     VPFunc = Intrinsic::getDeclaration(
         M, VPID, {Params[0]->getType(), Params[1]->getType()});
     break;
   }
   assert(VPFunc && "Could not declare VP intrinsic");
   return VPFunc;
 }

llvm/unittests/IR/VPIntrinsicTest.cpp

std::unique_ptr<Module> createVPDeclarationModule() {
 
     const char *BinaryFPOpcodes[] = {"fadd", "fsub", "fmul", "fdiv", "frem"};
     for (const char *BinaryFPOpcode : BinaryFPOpcodes)
       Str << " declare <8 x float> @llvm.vp." << BinaryFPOpcode
           << ".v8f32(<8 x float>, <8 x float>, <8 x i1>, i32) ";
 
     Str << " declare void @llvm.vp.store.v8i32.p0v8i32(<8 x i32>, <8 x i32>*, "
            "<8 x i1>, i32) ";
+    Str << "declare void "
+           "@llvm.experimental.vp.strided.store.v8i32.p0v8i32.i32(<8 x i32>, "
+           "<8 x i32>*, i32, <8 x i1>, i32) ";
     Str << " declare void @llvm.vp.scatter.v8i32.v8p0i32(<8 x i32>, <8 x "
            "i32*>, <8 x i1>, i32) ";
     Str << " declare <8 x i32> @llvm.vp.load.v8i32.p0v8i32(<8 x i32>*, <8 x "
            "i1>, i32) ";
+    Str << "declare <8 x i32> "
+           "@llvm.experimental.vp.strided.load.v8i32.p0v8i32.i32(<8 x i32>*, "
+           "i32, <8 x i1>, i32) ";
     Str << " declare <8 x i32> @llvm.vp.gather.v8i32.v8p0i32(<8 x i32*>, <8 x "
            "i1>, i32) ";
 
     for (const char *ReductionOpcode : ReductionIntOpcodes)
       Str << " declare i32 @llvm.vp.reduce." << ReductionOpcode
           << ".v8i32(i32, <8 x i32>, <8 x i1>, i32) ";
 
     for (const char *ReductionOpcode : ReductionFPOpcodes)