[AArch64][SME2][SVE2p1] Choose strided or contiguous loads

[MattDevereau]

Summary

Lower to the strided/contiguous addressing mode of
ld1/ldnt1 instructions depending on register allocation.

[dtemirbulatov]

LGTM.

llvm/lib/Target/AArch64/AArch64ExpandPseudoInsts.cpp

public:
   bool runOnMachineFunction(MachineFunction &Fn) override;
 
   StringRef getPassName() const override { return AARCH64_EXPAND_PSEUDO_NAME; }
 
 private:
   bool expandMBB(MachineBasicBlock &MBB);
   bool expandMI(MachineBasicBlock &MBB, MachineBasicBlock::iterator MBBI,
                 MachineBasicBlock::iterator &NextMBBI);
+  bool expandMultiVecPseudo(MachineBasicBlock &MBB,
+                            MachineBasicBlock::iterator MBBI,
+                            TargetRegisterClass ContiguousClass,
+                            TargetRegisterClass StridedClass,
+                            unsigned ContiguousOpc, unsigned StridedOpc);
   bool expandMOVImm(MachineBasicBlock &MBB, MachineBasicBlock::iterator MBBI,
                     unsigned BitSize);
 
   bool expand_DestructiveOp(MachineInstr &MI, MachineBasicBlock &MBB,
                             MachineBasicBlock::iterator MBBI);
   bool expandCMP_SWAP(MachineBasicBlock &MBB, MachineBasicBlock::iterator MBBI,
                       unsigned LdarOp, unsigned StlrOp, unsigned CmpOp,
                       unsigned ExtendImm, unsigned ZeroReg,
for (unsigned i = 4; i < MI.getNumOperands(); ++i)
     MIB.add(MI.getOperand(i));
 
   BuildMI(SMBB, DL, TII->get(AArch64::B)).addMBB(EndBB);
 
   MI.eraseFromParent();
   return EndBB;
 }
 
+bool AArch64ExpandPseudo::expandMultiVecPseudo(
+    MachineBasicBlock &MBB, MachineBasicBlock::iterator MBBI,
+    TargetRegisterClass ContiguousClass, TargetRegisterClass StridedClass,
+    unsigned ContiguousOp, unsigned StridedOpc) {
+  MachineInstr &MI = *MBBI;
+  Register Tuple = MI.getOperand(0).getReg();
+
+  auto ContiguousRange = ContiguousClass.getRegisters();
+  auto StridedRange = StridedClass.getRegisters();
+  unsigned Opc;
+  if ((std::find(ContiguousRange.begin(), ContiguousRange.end(),
+                 Tuple.asMCReg()) != std::end(ContiguousRange))) {
+    Opc = ContiguousOp;
+  } else if ((std::find(StridedRange.begin(), StridedRange.end(),
+                        Tuple.asMCReg()) != std::end(StridedRange))) {
+    Opc = StridedOpc;
+  } else
+    llvm_unreachable("Cannot expand Multi-Vector pseudo");
+
+  MachineInstrBuilder MIB = BuildMI(MBB, MBBI, MI.getDebugLoc(), TII->get(Opc))
+                                .add(MI.getOperand(0))
+                                .add(MI.getOperand(1))
+                                .add(MI.getOperand(2))
+                                .add(MI.getOperand(3));
+  transferImpOps(MI, MIB, MIB);
+  MI.eraseFromParent();
+  return true;
+}
+
 /// If MBBI references a pseudo instruction that should be expanded here,
 /// do the expansion and return true.  Otherwise return false.
 bool AArch64ExpandPseudo::expandMI(MachineBasicBlock &MBB,
                                    MachineBasicBlock::iterator MBBI,
                                    MachineBasicBlock::iterator &NextMBBI) {
   MachineInstr &MI = *MBBI;
   unsigned Opcode = MI.getOpcode();
 
if (MI.getOperand(0).getReg() != MI.getOperand(1).getReg()) {
            .add(MI.getOperand(0))
            .addReg(AArch64::XZR)
            .add(MI.getOperand(1))
            .addImm(0);
      }
      MI.eraseFromParent();
      return true;
    }
+   case AArch64::LD1B_2Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LD1B_2Z_IMM, AArch64::LD1B_2Z_STRIDED_IMM);
+   case AArch64::LD1H_2Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LD1H_2Z_IMM, AArch64::LD1H_2Z_STRIDED_IMM);
+   case AArch64::LD1W_2Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LD1W_2Z_IMM, AArch64::LD1W_2Z_STRIDED_IMM);
+   case AArch64::LD1D_2Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LD1D_2Z_IMM, AArch64::LD1D_2Z_STRIDED_IMM);
+   case AArch64::LDNT1B_2Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LDNT1B_2Z_IMM, AArch64::LDNT1B_2Z_STRIDED_IMM);
+   case AArch64::LDNT1H_2Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LDNT1H_2Z_IMM, AArch64::LDNT1H_2Z_STRIDED_IMM);
+   case AArch64::LDNT1W_2Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LDNT1W_2Z_IMM, AArch64::LDNT1W_2Z_STRIDED_IMM);
+   case AArch64::LDNT1D_2Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LDNT1D_2Z_IMM, AArch64::LDNT1D_2Z_STRIDED_IMM);
+   case AArch64::LD1B_2Z_PSEUDO:
+     return expandMultiVecPseudo(MBB, MBBI, AArch64::ZPR2RegClass,
+                                 AArch64::ZPR2StridedRegClass, AArch64::LD1B_2Z,
+                                 AArch64::LD1B_2Z_STRIDED);
+   case AArch64::LD1H_2Z_PSEUDO:
+     return expandMultiVecPseudo(MBB, MBBI, AArch64::ZPR2RegClass,
+                                 AArch64::ZPR2StridedRegClass, AArch64::LD1H_2Z,
+                                 AArch64::LD1H_2Z_STRIDED);
+   case AArch64::LD1W_2Z_PSEUDO:
+     return expandMultiVecPseudo(MBB, MBBI, AArch64::ZPR2RegClass,
+                                 AArch64::ZPR2StridedRegClass, AArch64::LD1W_2Z,
+                                 AArch64::LD1W_2Z_STRIDED);
+   case AArch64::LD1D_2Z_PSEUDO:
+     return expandMultiVecPseudo(MBB, MBBI, AArch64::ZPR2RegClass,
+                                 AArch64::ZPR2StridedRegClass, AArch64::LD1D_2Z,
+                                 AArch64::LD1D_2Z_STRIDED);
+   case AArch64::LDNT1B_2Z_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LDNT1B_2Z, AArch64::LDNT1B_2Z_STRIDED);
+   case AArch64::LDNT1H_2Z_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LDNT1H_2Z, AArch64::LDNT1H_2Z_STRIDED);
+   case AArch64::LDNT1W_2Z_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LDNT1W_2Z, AArch64::LDNT1W_2Z_STRIDED);
+   case AArch64::LDNT1D_2Z_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR2RegClass, AArch64::ZPR2StridedRegClass,
+         AArch64::LDNT1D_2Z, AArch64::LDNT1D_2Z_STRIDED);
+   case AArch64::LD1B_4Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LD1B_4Z_IMM, AArch64::LD1B_4Z_STRIDED_IMM);
+   case AArch64::LD1H_4Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LD1H_4Z_IMM, AArch64::LD1H_4Z_STRIDED_IMM);
+   case AArch64::LD1W_4Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LD1W_4Z_IMM, AArch64::LD1W_4Z_STRIDED_IMM);
+   case AArch64::LD1D_4Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LD1D_4Z_IMM, AArch64::LD1D_4Z_STRIDED_IMM);
+   case AArch64::LDNT1B_4Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LDNT1B_4Z_IMM, AArch64::LDNT1B_4Z_STRIDED_IMM);
+   case AArch64::LDNT1H_4Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LDNT1H_4Z_IMM, AArch64::LDNT1H_4Z_STRIDED_IMM);
+   case AArch64::LDNT1W_4Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LDNT1W_4Z_IMM, AArch64::LDNT1W_4Z_STRIDED_IMM);
+   case AArch64::LDNT1D_4Z_IMM_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LDNT1D_4Z_IMM, AArch64::LDNT1D_4Z_STRIDED_IMM);
+   case AArch64::LD1B_4Z_PSEUDO:
+     return expandMultiVecPseudo(MBB, MBBI, AArch64::ZPR4RegClass,
+                                 AArch64::ZPR4StridedRegClass, AArch64::LD1B_4Z,
+                                 AArch64::LD1B_4Z_STRIDED);
+   case AArch64::LD1H_4Z_PSEUDO:
+     return expandMultiVecPseudo(MBB, MBBI, AArch64::ZPR4RegClass,
+                                 AArch64::ZPR4StridedRegClass, AArch64::LD1H_4Z,
+                                 AArch64::LD1H_4Z_STRIDED);
+   case AArch64::LD1W_4Z_PSEUDO:
+     return expandMultiVecPseudo(MBB, MBBI, AArch64::ZPR4RegClass,
+                                 AArch64::ZPR4StridedRegClass, AArch64::LD1W_4Z,
+                                 AArch64::LD1W_4Z_STRIDED);
+   case AArch64::LD1D_4Z_PSEUDO:
+     return expandMultiVecPseudo(MBB, MBBI, AArch64::ZPR4RegClass,
+                                 AArch64::ZPR4StridedRegClass, AArch64::LD1D_4Z,
+                                 AArch64::LD1D_4Z_STRIDED);
+   case AArch64::LDNT1B_4Z_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LDNT1B_4Z, AArch64::LDNT1B_4Z_STRIDED);
+   case AArch64::LDNT1H_4Z_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LDNT1H_4Z, AArch64::LDNT1H_4Z_STRIDED);
+   case AArch64::LDNT1W_4Z_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LDNT1W_4Z, AArch64::LDNT1W_4Z_STRIDED);
+   case AArch64::LDNT1D_4Z_PSEUDO:
+     return expandMultiVecPseudo(
+         MBB, MBBI, AArch64::ZPR4RegClass, AArch64::ZPR4StridedRegClass,
+         AArch64::LDNT1D_4Z, AArch64::LDNT1D_4Z_STRIDED);
   }
   return false;
 }
 
 /// Iterate over the instructions in basic block MBB and expand any
 /// pseudo instructions.  Return true if anything was modified.
 bool AArch64ExpandPseudo::expandMBB(MachineBasicBlock &MBB) {
   bool Modified = false;

llvm/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp

case Intrinsic::aarch64_sve_ld2_sret: {
         SelectPredicatedLoad(Node, 2, 3, AArch64::LD2D_IMM, AArch64::LD2D,
                              true);
         return;
       }
       break;
     }
     case Intrinsic::aarch64_sve_ld1_pn_x2: {
       if (VT == MVT::nxv16i8) {
-        SelectContiguousMultiVectorLoad(Node, 2, 0, AArch64::LD1B_2Z_IMM, AArch64::LD1B_2Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(
+              Node, 2, 0, AArch64::LD1B_2Z_IMM_PSEUDO, AArch64::LD1B_2Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 2, 0, AArch64::LD1B_2Z_IMM,
+                                          AArch64::LD1B_2Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv8i16 || VT == MVT::nxv8f16 ||
                  VT == MVT::nxv8bf16) {
-        SelectContiguousMultiVectorLoad(Node, 2, 1, AArch64::LD1H_2Z_IMM, AArch64::LD1H_2Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(
+              Node, 2, 1, AArch64::LD1H_2Z_IMM_PSEUDO, AArch64::LD1H_2Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 2, 1, AArch64::LD1H_2Z_IMM,
+                                          AArch64::LD1H_2Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv4i32 || VT == MVT::nxv4f32) {
-        SelectContiguousMultiVectorLoad(Node, 2, 2, AArch64::LD1W_2Z_IMM, AArch64::LD1W_2Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(
+              Node, 2, 2, AArch64::LD1W_2Z_IMM_PSEUDO, AArch64::LD1W_2Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 2, 2, AArch64::LD1W_2Z_IMM,
+                                          AArch64::LD1W_2Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv2i64 || VT == MVT::nxv2f64) {
-        SelectContiguousMultiVectorLoad(Node, 2, 3, AArch64::LD1D_2Z_IMM, AArch64::LD1D_2Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(
+              Node, 2, 3, AArch64::LD1D_2Z_IMM_PSEUDO, AArch64::LD1D_2Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 2, 3, AArch64::LD1D_2Z_IMM,
+                                          AArch64::LD1D_2Z);
+        else
+          break;
         return;
       }
       break;
     }
     case Intrinsic::aarch64_sve_ld1_pn_x4: {
       if (VT == MVT::nxv16i8) {
-        SelectContiguousMultiVectorLoad(Node, 4, 0, AArch64::LD1B_4Z_IMM, AArch64::LD1B_4Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(
+              Node, 4, 0, AArch64::LD1B_4Z_IMM_PSEUDO, AArch64::LD1B_4Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 4, 0, AArch64::LD1B_4Z_IMM,
+                                          AArch64::LD1B_4Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv8i16 || VT == MVT::nxv8f16 ||
                  VT == MVT::nxv8bf16) {
-        SelectContiguousMultiVectorLoad(Node, 4, 1, AArch64::LD1H_4Z_IMM, AArch64::LD1H_4Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(
+              Node, 4, 1, AArch64::LD1H_4Z_IMM_PSEUDO, AArch64::LD1H_4Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 4, 1, AArch64::LD1H_4Z_IMM,
+                                          AArch64::LD1H_4Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv4i32 || VT == MVT::nxv4f32) {
-        SelectContiguousMultiVectorLoad(Node, 4, 2, AArch64::LD1W_4Z_IMM, AArch64::LD1W_4Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(
+              Node, 4, 2, AArch64::LD1W_4Z_IMM_PSEUDO, AArch64::LD1W_4Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 4, 2, AArch64::LD1W_4Z_IMM,
+                                          AArch64::LD1W_4Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv2i64 || VT == MVT::nxv2f64) {
-        SelectContiguousMultiVectorLoad(Node, 4, 3, AArch64::LD1D_4Z_IMM, AArch64::LD1D_4Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(
+              Node, 4, 3, AArch64::LD1D_4Z_IMM_PSEUDO, AArch64::LD1D_4Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 4, 3, AArch64::LD1D_4Z_IMM,
+                                          AArch64::LD1D_4Z);
+        else
+          break;
         return;
       }
       break;
     }
     case Intrinsic::aarch64_sve_ldnt1_pn_x2: {
       if (VT == MVT::nxv16i8) {
-        SelectContiguousMultiVectorLoad(Node, 2, 0, AArch64::LDNT1B_2Z_IMM, AArch64::LDNT1B_2Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(Node, 2, 0,
+                                          AArch64::LDNT1B_2Z_IMM_PSEUDO,
+                                          AArch64::LDNT1B_2Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 2, 0, AArch64::LDNT1B_2Z_IMM,
+                                          AArch64::LDNT1B_2Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv8i16 || VT == MVT::nxv8f16 ||
                  VT == MVT::nxv8bf16) {
-        SelectContiguousMultiVectorLoad(Node, 2, 1, AArch64::LDNT1H_2Z_IMM, AArch64::LDNT1H_2Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(Node, 2, 1,
+                                          AArch64::LDNT1H_2Z_IMM_PSEUDO,
+                                          AArch64::LDNT1H_2Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 2, 1, AArch64::LDNT1H_2Z_IMM,
+                                          AArch64::LDNT1H_2Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv4i32 || VT == MVT::nxv4f32) {
-        SelectContiguousMultiVectorLoad(Node, 2, 2, AArch64::LDNT1W_2Z_IMM, AArch64::LDNT1W_2Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(Node, 2, 2,
+                                          AArch64::LDNT1W_2Z_IMM_PSEUDO,
+                                          AArch64::LDNT1W_2Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 2, 2, AArch64::LDNT1W_2Z_IMM,
+                                          AArch64::LDNT1W_2Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv2i64 || VT == MVT::nxv2f64) {
-        SelectContiguousMultiVectorLoad(Node, 2, 3, AArch64::LDNT1D_2Z_IMM, AArch64::LDNT1D_2Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(Node, 2, 3,
+                                          AArch64::LDNT1D_2Z_IMM_PSEUDO,
+                                          AArch64::LDNT1D_2Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 2, 3, AArch64::LDNT1D_2Z_IMM,
+                                          AArch64::LDNT1D_2Z);
+        else
+          break;
         return;
       }
       break;
     }
     case Intrinsic::aarch64_sve_ldnt1_pn_x4: {
       if (VT == MVT::nxv16i8) {
-        SelectContiguousMultiVectorLoad(Node, 4, 0, AArch64::LDNT1B_4Z_IMM, AArch64::LDNT1B_4Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(Node, 4, 0,
+                                          AArch64::LDNT1B_4Z_IMM_PSEUDO,
+                                          AArch64::LDNT1B_4Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 4, 0, AArch64::LDNT1B_4Z_IMM,
+                                          AArch64::LDNT1B_4Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv8i16 || VT == MVT::nxv8f16 ||
                  VT == MVT::nxv8bf16) {
-        SelectContiguousMultiVectorLoad(Node, 4, 1, AArch64::LDNT1H_4Z_IMM, AArch64::LDNT1H_4Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(Node, 4, 1,
+                                          AArch64::LDNT1H_4Z_IMM_PSEUDO,
+                                          AArch64::LDNT1H_4Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 4, 1, AArch64::LDNT1H_4Z_IMM,
+                                          AArch64::LDNT1H_4Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv4i32 || VT == MVT::nxv4f32) {
-        SelectContiguousMultiVectorLoad(Node, 4, 2, AArch64::LDNT1W_4Z_IMM, AArch64::LDNT1W_4Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(Node, 4, 2,
+                                          AArch64::LDNT1W_4Z_IMM_PSEUDO,
+                                          AArch64::LDNT1W_4Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 4, 2, AArch64::LDNT1W_4Z_IMM,
+                                          AArch64::LDNT1W_4Z);
+        else
+          break;
         return;
       } else if (VT == MVT::nxv2i64 || VT == MVT::nxv2f64) {
-        SelectContiguousMultiVectorLoad(Node, 4, 3, AArch64::LDNT1D_4Z_IMM, AArch64::LDNT1D_4Z);
+        if (Subtarget->hasSME2())
+          SelectContiguousMultiVectorLoad(Node, 4, 3,
+                                          AArch64::LDNT1D_4Z_IMM_PSEUDO,
+                                          AArch64::LDNT1D_4Z_PSEUDO);
+        else if (Subtarget->hasSVE2p1())
+          SelectContiguousMultiVectorLoad(Node, 4, 3, AArch64::LDNT1D_4Z_IMM,
+                                          AArch64::LDNT1D_4Z);
+        else
+          break;
         return;
       }
       break;
     }
     case Intrinsic::aarch64_sve_ld3_sret: {
       if (VT == MVT::nxv16i8) {
         SelectPredicatedLoad(Node, 3, 0, AArch64::LD3B_IMM, AArch64::LD3B,
                              true);

llvm/lib/Target/AArch64/AArch64RegisterInfo.td

def ZPR2Strided : RegisterClass<"AArch64", [untyped], 256,
                                 (add ZStridedPairsLo, ZStridedPairsHi)>  {
   let Size = 256;
 }
 def ZPR4Strided : RegisterClass<"AArch64", [untyped], 512,
                                 (add ZStridedQuadsLo, ZStridedQuadsHi)>  {
   let Size = 512;
 }
 
+def ZPR2StridedOrContiguous : RegisterClass<"AArch64", [untyped], 256,
+                                (add ZStridedPairsLo, ZStridedPairsHi,
+                                (decimate ZSeqPairs, 2))> {
+  let Size = 256;
+}
 
 class ZPRVectorListStrided<int ElementWidth, int NumRegs, int Stride>
     : ZPRVectorList<ElementWidth, NumRegs> {
   let Name = "SVEVectorListStrided" # NumRegs # "x" # ElementWidth;
   let DiagnosticType = "Invalid" # Name;
   let PredicateMethod = "isTypedVectorListStrided<RegKind::SVEDataVector, "
                         # NumRegs # "," # Stride # "," # ElementWidth # ">";
   let RenderMethod = "addStridedVectorListOperands<" # NumRegs # ">";
def ZZ_s_strided
       : RegisterOperand<ZPR2Strided, "printTypedVectorList<0,'s'>"> {
     let ParserMatchClass = ZPRVectorListStrided<32, 2, 8>;
   }
 
   def ZZ_d_strided
       : RegisterOperand<ZPR2Strided, "printTypedVectorList<0,'d'>"> {
     let ParserMatchClass = ZPRVectorListStrided<64, 2, 8>;
   }
+
+  def ZZ_b_strided_and_contiguous
+      : RegisterOperand<ZPR2StridedOrContiguous, "printTypedVectorList<0,'b'>">;
+  def ZZ_h_strided_and_contiguous
+      : RegisterOperand<ZPR2StridedOrContiguous, "printTypedVectorList<0,'h'>">;
+  def ZZ_s_strided_and_contiguous
+      : RegisterOperand<ZPR2StridedOrContiguous, "printTypedVectorList<0,'s'>">;
+  def ZZ_d_strided_and_contiguous
+      : RegisterOperand<ZPR2StridedOrContiguous, "printTypedVectorList<0,'d'>">;
+}
+
+def ZPR4StridedOrContiguous : RegisterClass<"AArch64", [untyped], 512,
+                                (add ZStridedQuadsLo, ZStridedQuadsHi,
+                                (decimate ZSeqQuads, 4))>  {
+  let Size = 512;
 }
 
 let EncoderMethod = "EncodeZPR4StridedRegisterClass",
     DecoderMethod = "DecodeZPR4StridedRegisterClass" in {
   def ZZZZ_b_strided
       : RegisterOperand<ZPR4Strided, "printTypedVectorList<0,'b'>"> {
     let ParserMatchClass = ZPRVectorListStrided<8, 4, 4>;
   }
 
   def ZZZZ_h_strided
       : RegisterOperand<ZPR4Strided, "printTypedVectorList<0,'h'>"> {
     let ParserMatchClass = ZPRVectorListStrided<16, 4, 4>;
   }
 
   def ZZZZ_s_strided
       : RegisterOperand<ZPR4Strided, "printTypedVectorList<0,'s'>"> {
     let ParserMatchClass = ZPRVectorListStrided<32, 4, 4>;
   }
 
   def ZZZZ_d_strided
       : RegisterOperand<ZPR4Strided, "printTypedVectorList<0,'d'>"> {
     let ParserMatchClass = ZPRVectorListStrided<64, 4, 4>;
   }
+
+  def ZZZZ_b_strided_and_contiguous
+      : RegisterOperand<ZPR4StridedOrContiguous, "printTypedVectorList<0,'b'>">;
+  def ZZZZ_h_strided_and_contiguous
+      : RegisterOperand<ZPR4StridedOrContiguous, "printTypedVectorList<0,'h'>">;
+  def ZZZZ_s_strided_and_contiguous
+      : RegisterOperand<ZPR4StridedOrContiguous, "printTypedVectorList<0,'s'>">;
+  def ZZZZ_d_strided_and_contiguous
+      : RegisterOperand<ZPR4StridedOrContiguous, "printTypedVectorList<0,'d'>">;
 }
 
 class ZPRExtendAsmOperand<string ShiftExtend, int RegWidth, int Scale,
                           bit ScaleAlwaysSame = 0b0> : AsmOperandClass {
   let Name = "ZPRExtend" # ShiftExtend # RegWidth # Scale
                          # !if(ScaleAlwaysSame, "Only", "");
 
   let PredicateMethod = "isSVEDataVectorRegWithShiftExtend<"

llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td

 defm SQCVTN_Z2Z_StoH  : sve2p1_multi_vec_extract_narrow<"sqcvtn", 0b00, int_aarch64_sve_sqcvtn_x2>;
 defm UQCVTN_Z2Z_StoH  : sve2p1_multi_vec_extract_narrow<"uqcvtn", 0b01, int_aarch64_sve_uqcvtn_x2>;
 defm SQCVTUN_Z2Z_StoH : sve2p1_multi_vec_extract_narrow<"sqcvtun", 0b10, int_aarch64_sve_sqcvtun_x2>;
 defm SQRSHRN_Z2ZI_StoH  : sve2p1_multi_vec_shift_narrow<"sqrshrn", 0b101, int_aarch64_sve_sqrshrn_x2>;
 defm UQRSHRN_Z2ZI_StoH  : sve2p1_multi_vec_shift_narrow<"uqrshrn", 0b111, int_aarch64_sve_uqrshrn_x2>;
 defm SQRSHRUN_Z2ZI_StoH : sve2p1_multi_vec_shift_narrow<"sqrshrun", 0b001, int_aarch64_sve_sqrshrun_x2>;
 
 // Load to two registers
-def LD1B_2Z        : sve2p1_mem_cld_ss_2z<"ld1b", 0b00, 0b0, ZZ_b_mul_r, GPR64shifted8>;
-def LD1H_2Z        : sve2p1_mem_cld_ss_2z<"ld1h", 0b01, 0b0, ZZ_h_mul_r, GPR64shifted16>;
-def LD1W_2Z        : sve2p1_mem_cld_ss_2z<"ld1w", 0b10, 0b0, ZZ_s_mul_r, GPR64shifted32>;
-def LD1D_2Z        : sve2p1_mem_cld_ss_2z<"ld1d", 0b11, 0b0, ZZ_d_mul_r, GPR64shifted64>;
-defm LD1B_2Z_IMM   : sve2p1_mem_cld_si_2z<"ld1b", 0b00, 0b0, ZZ_b_mul_r>;
-defm LD1H_2Z_IMM   : sve2p1_mem_cld_si_2z<"ld1h", 0b01, 0b0, ZZ_h_mul_r>;
-defm LD1W_2Z_IMM   : sve2p1_mem_cld_si_2z<"ld1w", 0b10, 0b0, ZZ_s_mul_r>;
-defm LD1D_2Z_IMM   : sve2p1_mem_cld_si_2z<"ld1d", 0b11, 0b0, ZZ_d_mul_r>;
-def LDNT1B_2Z      : sve2p1_mem_cld_ss_2z<"ldnt1b", 0b00, 0b1, ZZ_b_mul_r, GPR64shifted8>;
-def LDNT1H_2Z      : sve2p1_mem_cld_ss_2z<"ldnt1h", 0b01, 0b1, ZZ_h_mul_r, GPR64shifted16>;
-def LDNT1W_2Z      : sve2p1_mem_cld_ss_2z<"ldnt1w", 0b10, 0b1, ZZ_s_mul_r, GPR64shifted32>;
-def LDNT1D_2Z      : sve2p1_mem_cld_ss_2z<"ldnt1d", 0b11, 0b1, ZZ_d_mul_r, GPR64shifted64>;
-defm LDNT1B_2Z_IMM : sve2p1_mem_cld_si_2z<"ldnt1b", 0b00, 0b1, ZZ_b_mul_r>;
-defm LDNT1H_2Z_IMM : sve2p1_mem_cld_si_2z<"ldnt1h", 0b01, 0b1, ZZ_h_mul_r>;
-defm LDNT1W_2Z_IMM : sve2p1_mem_cld_si_2z<"ldnt1w", 0b10, 0b1, ZZ_s_mul_r>;
-defm LDNT1D_2Z_IMM : sve2p1_mem_cld_si_2z<"ldnt1d", 0b11, 0b1, ZZ_d_mul_r>;
+defm LD1B_2Z       : sve2p1_mem_cld_ss_2z<"ld1b", 0b00, 0b0, ZZ_b_mul_r, GPR64shifted8, ZZ_b_strided_and_contiguous>;
+defm LD1H_2Z       : sve2p1_mem_cld_ss_2z<"ld1h", 0b01, 0b0, ZZ_h_mul_r, GPR64shifted16, ZZ_h_strided_and_contiguous>;
+defm LD1W_2Z       : sve2p1_mem_cld_ss_2z<"ld1w", 0b10, 0b0, ZZ_s_mul_r, GPR64shifted32, ZZ_s_strided_and_contiguous>;
+defm LD1D_2Z       : sve2p1_mem_cld_ss_2z<"ld1d", 0b11, 0b0, ZZ_d_mul_r, GPR64shifted64, ZZ_d_strided_and_contiguous>;
+defm LD1B_2Z_IMM   : sve2p1_mem_cld_si_2z<"ld1b", 0b00, 0b0, ZZ_b_mul_r, ZZ_b_strided_and_contiguous>;
+defm LD1H_2Z_IMM   : sve2p1_mem_cld_si_2z<"ld1h", 0b01, 0b0, ZZ_h_mul_r, ZZ_h_strided_and_contiguous>;
+defm LD1W_2Z_IMM   : sve2p1_mem_cld_si_2z<"ld1w", 0b10, 0b0, ZZ_s_mul_r, ZZ_s_strided_and_contiguous>;
+defm LD1D_2Z_IMM   : sve2p1_mem_cld_si_2z<"ld1d", 0b11, 0b0, ZZ_d_mul_r, ZZ_d_strided_and_contiguous>;
+defm LDNT1B_2Z     : sve2p1_mem_cld_ss_2z<"ldnt1b", 0b00, 0b1, ZZ_b_mul_r, GPR64shifted8, ZZ_b_strided_and_contiguous>;
+defm LDNT1H_2Z     : sve2p1_mem_cld_ss_2z<"ldnt1h", 0b01, 0b1, ZZ_h_mul_r, GPR64shifted16, ZZ_h_strided_and_contiguous>;
+defm LDNT1W_2Z     : sve2p1_mem_cld_ss_2z<"ldnt1w", 0b10, 0b1, ZZ_s_mul_r, GPR64shifted32, ZZ_s_strided_and_contiguous>;
+defm LDNT1D_2Z     : sve2p1_mem_cld_ss_2z<"ldnt1d", 0b11, 0b1, ZZ_d_mul_r, GPR64shifted64, ZZ_d_strided_and_contiguous>;
+defm LDNT1B_2Z_IMM : sve2p1_mem_cld_si_2z<"ldnt1b", 0b00, 0b1, ZZ_b_mul_r, ZZ_b_strided_and_contiguous>;
+defm LDNT1H_2Z_IMM : sve2p1_mem_cld_si_2z<"ldnt1h", 0b01, 0b1, ZZ_h_mul_r, ZZ_h_strided_and_contiguous>;
+defm LDNT1W_2Z_IMM : sve2p1_mem_cld_si_2z<"ldnt1w", 0b10, 0b1, ZZ_s_mul_r, ZZ_s_strided_and_contiguous>;
+defm LDNT1D_2Z_IMM : sve2p1_mem_cld_si_2z<"ldnt1d", 0b11, 0b1, ZZ_d_mul_r, ZZ_d_strided_and_contiguous>;
 
 // Load to four registers
-def LD1B_4Z        : sve2p1_mem_cld_ss_4z<"ld1b", 0b00, 0b0, ZZZZ_b_mul_r, GPR64shifted8>;
-def LD1H_4Z        : sve2p1_mem_cld_ss_4z<"ld1h", 0b01, 0b0, ZZZZ_h_mul_r, GPR64shifted16>;
-def LD1W_4Z        : sve2p1_mem_cld_ss_4z<"ld1w", 0b10, 0b0, ZZZZ_s_mul_r, GPR64shifted32>;
-def LD1D_4Z        : sve2p1_mem_cld_ss_4z<"ld1d", 0b11, 0b0, ZZZZ_d_mul_r, GPR64shifted64>;
-defm LD1B_4Z_IMM   : sve2p1_mem_cld_si_4z<"ld1b", 0b00, 0b0, ZZZZ_b_mul_r>;
-defm LD1H_4Z_IMM   : sve2p1_mem_cld_si_4z<"ld1h", 0b01, 0b0, ZZZZ_h_mul_r>;
-defm LD1W_4Z_IMM   : sve2p1_mem_cld_si_4z<"ld1w", 0b10, 0b0, ZZZZ_s_mul_r>;
-defm LD1D_4Z_IMM   : sve2p1_mem_cld_si_4z<"ld1d", 0b11, 0b0, ZZZZ_d_mul_r>;
-def LDNT1B_4Z      : sve2p1_mem_cld_ss_4z<"ldnt1b", 0b00, 0b1, ZZZZ_b_mul_r, GPR64shifted8>;
-def LDNT1H_4Z      : sve2p1_mem_cld_ss_4z<"ldnt1h", 0b01, 0b1, ZZZZ_h_mul_r, GPR64shifted16>;
-def LDNT1W_4Z      : sve2p1_mem_cld_ss_4z<"ldnt1w", 0b10, 0b1, ZZZZ_s_mul_r, GPR64shifted32>;
-def LDNT1D_4Z      : sve2p1_mem_cld_ss_4z<"ldnt1d", 0b11, 0b1, ZZZZ_d_mul_r, GPR64shifted64>;
-defm LDNT1B_4Z_IMM : sve2p1_mem_cld_si_4z<"ldnt1b", 0b00, 0b1, ZZZZ_b_mul_r>;
-defm LDNT1H_4Z_IMM : sve2p1_mem_cld_si_4z<"ldnt1h", 0b01, 0b1, ZZZZ_h_mul_r>;
-defm LDNT1W_4Z_IMM : sve2p1_mem_cld_si_4z<"ldnt1w", 0b10, 0b1, ZZZZ_s_mul_r>;
-defm LDNT1D_4Z_IMM : sve2p1_mem_cld_si_4z<"ldnt1d", 0b11, 0b1, ZZZZ_d_mul_r>;
+defm LD1B_4Z       : sve2p1_mem_cld_ss_4z<"ld1b", 0b00, 0b0, ZZZZ_b_mul_r, GPR64shifted8, ZZZZ_b_strided_and_contiguous>;
+defm LD1H_4Z       : sve2p1_mem_cld_ss_4z<"ld1h", 0b01, 0b0, ZZZZ_h_mul_r, GPR64shifted16, ZZZZ_h_strided_and_contiguous>;
+defm LD1W_4Z       : sve2p1_mem_cld_ss_4z<"ld1w", 0b10, 0b0, ZZZZ_s_mul_r, GPR64shifted32, ZZZZ_s_strided_and_contiguous>;
+defm LD1D_4Z       : sve2p1_mem_cld_ss_4z<"ld1d", 0b11, 0b0, ZZZZ_d_mul_r, GPR64shifted64, ZZZZ_d_strided_and_contiguous>;
+defm LD1B_4Z_IMM   : sve2p1_mem_cld_si_4z<"ld1b", 0b00, 0b0, ZZZZ_b_mul_r, ZZZZ_b_strided_and_contiguous>;
+defm LD1H_4Z_IMM   : sve2p1_mem_cld_si_4z<"ld1h", 0b01, 0b0, ZZZZ_h_mul_r, ZZZZ_h_strided_and_contiguous>;
+defm LD1W_4Z_IMM   : sve2p1_mem_cld_si_4z<"ld1w", 0b10, 0b0, ZZZZ_s_mul_r, ZZZZ_s_strided_and_contiguous>;
+defm LD1D_4Z_IMM   : sve2p1_mem_cld_si_4z<"ld1d", 0b11, 0b0, ZZZZ_d_mul_r, ZZZZ_d_strided_and_contiguous>;
+defm LDNT1B_4Z     : sve2p1_mem_cld_ss_4z<"ldnt1b", 0b00, 0b1, ZZZZ_b_mul_r, GPR64shifted8, ZZZZ_b_strided_and_contiguous>;
+defm LDNT1H_4Z     : sve2p1_mem_cld_ss_4z<"ldnt1h", 0b01, 0b1, ZZZZ_h_mul_r, GPR64shifted16, ZZZZ_h_strided_and_contiguous>;
+defm LDNT1W_4Z     : sve2p1_mem_cld_ss_4z<"ldnt1w", 0b10, 0b1, ZZZZ_s_mul_r, GPR64shifted32, ZZZZ_s_strided_and_contiguous>;
+defm LDNT1D_4Z     : sve2p1_mem_cld_ss_4z<"ldnt1d", 0b11, 0b1, ZZZZ_d_mul_r, GPR64shifted64, ZZZZ_d_strided_and_contiguous>;
+defm LDNT1B_4Z_IMM : sve2p1_mem_cld_si_4z<"ldnt1b", 0b00, 0b1, ZZZZ_b_mul_r, ZZZZ_b_strided_and_contiguous>;
+defm LDNT1H_4Z_IMM : sve2p1_mem_cld_si_4z<"ldnt1h", 0b01, 0b1, ZZZZ_h_mul_r, ZZZZ_h_strided_and_contiguous>;
+defm LDNT1W_4Z_IMM : sve2p1_mem_cld_si_4z<"ldnt1w", 0b10, 0b1, ZZZZ_s_mul_r, ZZZZ_s_strided_and_contiguous>;
+defm LDNT1D_4Z_IMM : sve2p1_mem_cld_si_4z<"ldnt1d", 0b11, 0b1, ZZZZ_d_mul_r, ZZZZ_d_strided_and_contiguous>;
 
 // Stores of two registers
 def ST1B_2Z        : sve2p1_mem_cst_ss_2z<"st1b", 0b00, 0b0, ZZ_b_mul_r, GPR64shifted8>;
 def ST1H_2Z        : sve2p1_mem_cst_ss_2z<"st1h", 0b01, 0b0, ZZ_h_mul_r, GPR64shifted16>;
 def ST1W_2Z        : sve2p1_mem_cst_ss_2z<"st1w", 0b10, 0b0, ZZ_s_mul_r, GPR64shifted32>;
 def ST1D_2Z        : sve2p1_mem_cst_ss_2z<"st1d", 0b11, 0b0, ZZ_d_mul_r, GPR64shifted64>;
 defm ST1B_2Z_IMM   : sve2p1_mem_cst_si_2z<"st1b", 0b00, 0b0, ZZ_b_mul_r>;
 defm ST1H_2Z_IMM   : sve2p1_mem_cst_si_2z<"st1h", 0b01, 0b0, ZZ_h_mul_r>;

llvm/lib/Target/AArch64/SVEInstrFormats.td

class sve2p1_mem_cld_ss_2z<string mnemonic, bits<2> msz, bit n,
   let Inst{9-5} = Rn;
   let Inst{4-1} = Zt;
   let Inst{0}   = n;
 
   let hasSideEffects = 0;
   let mayLoad = 1;
 }
 
+multiclass sve2p1_mem_cld_ss_2z<string mnemonic, bits<2> msz, bit n,
+                         RegisterOperand vector_ty, RegisterOperand gpr_ty, RegisterOperand vector_pseudo_ty> {
+  def NAME # _PSEUDO : Pseudo<(outs vector_pseudo_ty:$Zt), (ins PNRAny_p8to15:$PNg, GPR64sp:$Rn, gpr_ty:$Rm), []>;
+  def NAME : sve2p1_mem_cld_ss_2z<mnemonic, msz, n, vector_ty, gpr_ty>;
+}
+
 // SME2 multi-vec contiguous load (scalar plus immediate, two registers)
 class sve2p1_mem_cld_si_2z<string mnemonic, bits<2> msz, bit n,
                          RegisterOperand vector_ty>
     : I<(outs vector_ty:$Zt),
         (ins PNRAny_p8to15:$PNg, GPR64sp:$Rn, simm4s2:$imm4),
         mnemonic, "\t$Zt, $PNg/z, [$Rn, $imm4, mul vl]",
         "", []>, Sched<[]> {
   bits<4> Zt;
class sve2p1_mem_cld_si_2z<string mnemonic, bits<2> msz, bit n,
   let Inst{4-1}   = Zt;
   let Inst{0}     = n;
 
   let hasSideEffects = 0;
   let mayLoad = 1;
 }
 
 multiclass sve2p1_mem_cld_si_2z<string mnemonic, bits<2> msz, bit n,
-                              RegisterOperand vector_ty> {
+                              RegisterOperand vector_ty, RegisterOperand vector_pseudo_ty> {
   def NAME : sve2p1_mem_cld_si_2z<mnemonic, msz, n, vector_ty>;
-
   def : InstAlias<mnemonic # " $Zt, $PNg/z, [$Rn]",
                   (!cast<Instruction>(NAME) vector_ty:$Zt, PNRAny_p8to15:$PNg, GPR64sp:$Rn, 0), 1>;
+  def NAME # _PSEUDO : Pseudo<(outs vector_pseudo_ty:$Zt), (ins PNRAny_p8to15:$PNg, GPR64sp:$Rn, simm4s2:$imm4), []>;
 }
 
 // SME2 multi-vec contiguous load (scalar plus scalar, four registers)
 class sve2p1_mem_cld_ss_4z<string mnemonic, bits<2> msz, bit n,
                          RegisterOperand vector_ty, RegisterOperand gpr_ty>
     : I<(outs vector_ty:$Zt),
         (ins PNRAny_p8to15:$PNg, GPR64sp:$Rn, gpr_ty:$Rm),
         mnemonic, "\t$Zt, $PNg/z, [$Rn, $Rm]",
class sve2p1_mem_cld_ss_4z<string mnemonic, bits<2> msz, bit n,
   let Inst{4-2} = Zt;
   let Inst{1}   = 0b0;
   let Inst{0}   = n;
 
   let hasSideEffects = 0;
   let mayLoad = 1;
 }
 
+multiclass sve2p1_mem_cld_ss_4z<string mnemonic, bits<2> msz, bit n,
+                         RegisterOperand vector_ty, RegisterOperand gpr_ty, RegisterOperand vector_pseudo_ty> {
+  def NAME # _PSEUDO : Pseudo<(outs vector_pseudo_ty:$Zt), (ins PNRAny_p8to15:$PNg, GPR64sp:$Rn, gpr_ty:$Rm), []>;
+  def NAME : sve2p1_mem_cld_ss_4z<mnemonic, msz, n, vector_ty, gpr_ty>;
+}
+
 // SME2 multi-vec contiguous load (scalar plus immediate, four registers)
 class sve2p1_mem_cld_si_4z<string mnemonic, bits<2> msz, bit n,
                          RegisterOperand vector_ty>
     : I<(outs vector_ty:$Zt),
         (ins PNRAny_p8to15:$PNg, GPR64sp:$Rn, simm4s4:$imm4),
         mnemonic, "\t$Zt, $PNg/z, [$Rn, $imm4, mul vl]",
         "", []>, Sched<[]> {
   bits<3> Zt;
class sve2p1_mem_cld_si_4z<string mnemonic, bits<2> msz, bit n,
   let Inst{1}     = 0b0;
   let Inst{0}     = n;
 
   let hasSideEffects = 0;
   let mayLoad = 1;
 }
 
 multiclass sve2p1_mem_cld_si_4z<string mnemonic, bits<2> msz, bit n,
-                              RegisterOperand vector_ty> {
+                              RegisterOperand vector_ty, RegisterOperand vector_pseudo_ty> {
   def NAME : sve2p1_mem_cld_si_4z<mnemonic, msz, n, vector_ty>;
-
   def : InstAlias<mnemonic # " $Zt, $PNg/z, [$Rn]",
                   (!cast<Instruction>(NAME) vector_ty:$Zt, PNRAny_p8to15:$PNg, GPR64sp:$Rn, 0), 1>;
+  def NAME # _PSEUDO : Pseudo<(outs vector_pseudo_ty:$Zt), (ins PNRAny_p8to15:$PNg, GPR64sp:$Rn, simm4s4:$imm4), []>;
 }
 
-
 // SME2 multi-vec contiguous store (scalar plus scalar, two registers)
 class sve2p1_mem_cst_ss_2z<string mnemonic, bits<2> msz, bit n,
                            RegisterOperand vector_ty, RegisterOperand gpr_ty>
     : I<(outs ),
         (ins vector_ty:$Zt, PNRAny_p8to15:$PNg, GPR64sp:$Rn, gpr_ty:$Rm),
         mnemonic, "\t$Zt, $PNg, [$Rn, $Rm]",
         "", []>, Sched<[]> {
   bits<4> Zt;

llvm/test/CodeGen/AArch64/sme2-intrinsics-ld1.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme2 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=STRIDED
+; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve2p1 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CONTIGUOUS
+
+define <vscale x 32 x i8> @ld1_x2_i8_z0_z8(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; CHECK-LABEL: ld1_x2_i8_z0_z8:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1b { z0.b, z8.b }, pn8/z, [x0]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z8.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x2_i8_z0_z8:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1b { z0.b, z8.b }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x2_i8_z0_z8:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1b { z0.b, z1.b }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld1.pn.x2.nxv16i8(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 0
+  %v0 = call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 1
+  %v1 = call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> %v0, <vscale x 16 x i8> %res.v1, i64 16)
+  ret <vscale x 32 x i8> %v1
+}
+
+define <vscale x 32 x i8> @ld1_x2_i8_z0_z8_scalar(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; CHECK-LABEL: ld1_x2_i8_z0_z8_scalar:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1b { z0.b, z8.b }, pn8/z, [x0, x1]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z8.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x2_i8_z0_z8_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1b { z0.b, z8.b }, pn8/z, [x0, x1]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x2_i8_z0_z8_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1b { z0.b, z1.b }, pn8/z, [x0, x1]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i8, ptr %ptr, i64 %index
+  %res = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld1.pn.x2.nxv16i8(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 0
+  %v0 = call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 1
+  %v1 = call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> %v0, <vscale x 16 x i8> %res.v1, i64 16)
+  ret <vscale x 32 x i8> %v1
+}
+
+define <vscale x 16 x i16> @ld1_x2_i16_z0_z8(<vscale x 8 x i16> %unused, <vscale x 8 x i16> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; CHECK-LABEL: ld1_x2_i16_z0_z8:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1h { z0.h, z8.h }, pn8/z, [x0]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z8.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x2_i16_z0_z8:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1h { z0.h, z8.h }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x2_i16_z0_z8:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1h { z0.h, z1.h }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld1.pn.x2.nxv8i16(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 0
+  %v0 = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 1
+  %v1 = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> %v0, <vscale x 8 x i16> %res.v1, i64 8)
+  ret <vscale x 16 x i16> %v1
+}
+
+define <vscale x 16 x i16> @ld1_x2_i16_z0_z8_scalar(<vscale x 8 x i16> %unused, <vscale x 8 x i16> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; CHECK-LABEL: ld1_x2_i16_z0_z8_scalar:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1h { z0.h, z8.h }, pn8/z, [x0, x1, lsl #1]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z8.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x2_i16_z0_z8_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1h { z0.h, z8.h }, pn8/z, [x0, x1, lsl #1]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x2_i16_z0_z8_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1h { z0.h, z1.h }, pn8/z, [x0, x1, lsl #1]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i16, ptr %ptr, i64 %index
+  %res = call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld1.pn.x2.nxv8i16(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 0
+  %v0 = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 1
+  %v1 = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> %v0, <vscale x 8 x i16> %res.v1, i64 8)
+  ret <vscale x 16 x i16> %v1
+}
+
+define <vscale x 8 x i32> @ld1_x2_i32_z0_z8(<vscale x 4 x i32> %unused, <vscale x 4 x i32> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; CHECK-LABEL: ld1_x2_i32_z0_z8:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1w { z0.s, z8.s }, pn8/z, [x0]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z8.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x2_i32_z0_z8:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1w { z0.s, z8.s }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x2_i32_z0_z8:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1w { z0.s, z1.s }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld1.pn.x2.nxv4i32(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 0
+  %v0 = call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 1
+  %v1 = call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> %v0, <vscale x 4 x i32> %res.v1, i64 4)
+  ret <vscale x 8 x i32> %v1
+}
+
+define <vscale x 8 x i32> @ld1_x2_i32_z0_z8_scalar(<vscale x 4 x i32> %unused, <vscale x 4 x i32> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; CHECK-LABEL: ld1_x2_i32_z0_z8_scalar:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1w { z0.s, z8.s }, pn8/z, [x0, x1, lsl #2]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z8.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x2_i32_z0_z8_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1w { z0.s, z8.s }, pn8/z, [x0, x1, lsl #2]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x2_i32_z0_z8_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1w { z0.s, z1.s }, pn8/z, [x0, x1, lsl #2]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i32, ptr %ptr, i64 %index
+  %res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld1.pn.x2.nxv4i32(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 0
+  %v0 = call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 1
+  %v1 = call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> %v0, <vscale x 4 x i32> %res.v1, i64 4)
+  ret <vscale x 8 x i32> %v1
+}
+
+define <vscale x 4 x i64> @ld1_x2_i64_z0_z8(<vscale x 2 x i64> %unused, <vscale x 2 x i64> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; CHECK-LABEL: ld1_x2_i64_z0_z8:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1d { z0.d, z8.d }, pn8/z, [x0]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z8.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x2_i64_z0_z8:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1d { z0.d, z8.d }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x2_i64_z0_z8:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1d { z0.d, z1.d }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld1.pn.x2.nxv2i64(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 0
+  %v0 = call <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64> poison, <vscale x 2 x i64> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 1
+  %v1 = call <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64> %v0, <vscale x 2 x i64> %res.v1, i64 2)
+  ret <vscale x 4 x i64> %v1
+}
+
+define <vscale x 4 x i64> @ld1_x2_i64_z0_z8_scalar(<vscale x 2 x i64> %unused, <vscale x 2 x i64> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; CHECK-LABEL: ld1_x2_i64_z0_z8_scalar:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1d { z0.d, z8.d }, pn8/z, [x0, x1, lsl #3]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z8.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x2_i64_z0_z8_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1d { z0.d, z8.d }, pn8/z, [x0, x1, lsl #3]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x2_i64_z0_z8_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1d { z0.d, z1.d }, pn8/z, [x0, x1, lsl #3]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i64, ptr %ptr, i64 %index
+  %res = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld1.pn.x2.nxv2i64(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 0
+  %v0 = call <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64> poison, <vscale x 2 x i64> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 1
+  %v1 = call <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64> %v0, <vscale x 2 x i64> %res.v1, i64 2)
+  ret <vscale x 4 x i64> %v1
+}
+
+define <vscale x 64 x i8> @ld1_x4_i8_z0_z4_z8_z12(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; CHECK-LABEL: ld1_x4_i8_z0_z4_z8_z12:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1b { z0.b, z4.b, z8.b, z12.b }, pn8/z, [x0]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z2.d, z8.d
+; CHECK-NEXT:    mov z3.d, z12.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z4.d
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x4_i8_z0_z4_z8_z12:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1b { z0.b, z4.b, z8.b, z12.b }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x4_i8_z0_z4_z8_z12:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1b { z0.b - z3.b }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld1.pn.x4.nxv16i8(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 0
+  %v0 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> poison, <vscale x 16 x i8> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 1
+  %v1 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v0, <vscale x 16 x i8> %res.v1, i64 16)
+  %res.v2 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 2
+  %v2 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v1, <vscale x 16 x i8> %res.v2, i64 32)
+  %res.v3 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 3
+  %v3 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v2, <vscale x 16 x i8> %res.v3, i64 48)
+  ret <vscale x 64 x i8> %v3
+}
+
+define <vscale x 64 x i8> @ld1_x4_i8_z0_z4_z8_z12_scalar(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; CHECK-LABEL: ld1_x4_i8_z0_z4_z8_z12_scalar:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1b { z0.b, z4.b, z8.b, z12.b }, pn8/z, [x0, x1]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z2.d, z8.d
+; CHECK-NEXT:    mov z3.d, z12.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z4.d
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x4_i8_z0_z4_z8_z12_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1b { z0.b, z4.b, z8.b, z12.b }, pn8/z, [x0, x1]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x4_i8_z0_z4_z8_z12_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1b { z0.b - z3.b }, pn8/z, [x0, x1]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i8, ptr %ptr, i64 %index
+  %res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld1.pn.x4.nxv16i8(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 0
+  %v0 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> poison, <vscale x 16 x i8> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 1
+  %v1 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v0, <vscale x 16 x i8> %res.v1, i64 16)
+  %res.v2 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 2
+  %v2 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v1, <vscale x 16 x i8> %res.v2, i64 32)
+  %res.v3 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 3
+  %v3 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v2, <vscale x 16 x i8> %res.v3, i64 48)
+  ret <vscale x 64 x i8> %v3
+}
+
+define <vscale x 32 x i16> @ld1_x4_i16_z0_z4_z8_z12(<vscale x 8 x i16> %unused, <vscale x 8 x i16> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; CHECK-LABEL: ld1_x4_i16_z0_z4_z8_z12:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1h { z0.h, z4.h, z8.h, z12.h }, pn8/z, [x0]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z2.d, z8.d
+; CHECK-NEXT:    mov z3.d, z12.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z4.d
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x4_i16_z0_z4_z8_z12:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1h { z0.h, z4.h, z8.h, z12.h }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x4_i16_z0_z4_z8_z12:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1h { z0.h - z3.h }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld1.pn.x4.nxv8i16(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 0
+  %v0 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> poison, <vscale x 8 x i16> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 1
+  %v1 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v0, <vscale x 8 x i16> %res.v1, i64 8)
+  %res.v2 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 2
+  %v2 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v1, <vscale x 8 x i16> %res.v2, i64 16)
+  %res.v3 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 3
+  %v3 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v2, <vscale x 8 x i16> %res.v3, i64 24)
+  ret <vscale x 32 x i16> %v3
+}
+
+define <vscale x 32 x i16> @ld1_x4_i16_z0_z4_z8_z12_scalar(<vscale x 8 x i16> %unused, <vscale x 8 x i16> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; CHECK-LABEL: ld1_x4_i16_z0_z4_z8_z12_scalar:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    lsl x8, x1, #1
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1h { z0.h, z4.h, z8.h, z12.h }, pn8/z, [x0, x8, lsl #1]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z2.d, z8.d
+; CHECK-NEXT:    mov z3.d, z12.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z4.d
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x4_i16_z0_z4_z8_z12_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1h { z0.h, z4.h, z8.h, z12.h }, pn8/z, [x0, x1, lsl #1]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x4_i16_z0_z4_z8_z12_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1h { z0.h - z3.h }, pn8/z, [x0, x1, lsl #1]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i16, ptr %ptr, i64 %index
+  %res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld1.pn.x4.nxv8i16(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 0
+  %v0 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> poison, <vscale x 8 x i16> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 1
+  %v1 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v0, <vscale x 8 x i16> %res.v1, i64 8)
+  %res.v2 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 2
+  %v2 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v1, <vscale x 8 x i16> %res.v2, i64 16)
+  %res.v3 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 3
+  %v3 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v2, <vscale x 8 x i16> %res.v3, i64 24)
+  ret <vscale x 32 x i16> %v3
+}
+
+define <vscale x 16 x i32> @ld1_x4_i32_z0_z4_z8_z12(<vscale x 4 x i32> %unused, <vscale x 4 x i32> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; CHECK-LABEL: ld1_x4_i32_z0_z4_z8_z12:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1w { z0.s, z4.s, z8.s, z12.s }, pn8/z, [x0]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z2.d, z8.d
+; CHECK-NEXT:    mov z3.d, z12.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z4.d
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x4_i32_z0_z4_z8_z12:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1w { z0.s, z4.s, z8.s, z12.s }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x4_i32_z0_z4_z8_z12:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1w { z0.s - z3.s }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld1.pn.x4.nxv4i32(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 0
+  %v0 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> poison, <vscale x 4 x i32> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 1
+  %v1 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v0, <vscale x 4 x i32> %res.v1, i64 4)
+  %res.v2 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 2
+  %v2 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v1, <vscale x 4 x i32> %res.v2, i64 8)
+  %res.v3 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 3
+  %v3 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v2, <vscale x 4 x i32> %res.v3, i64 12)
+  ret <vscale x 16 x i32> %v3
+}
+
+define <vscale x 16 x i32> @ld1_x4_i32_z0_z4_z8_z12_scalar(<vscale x 4 x i32> %unused, <vscale x 4 x i32> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; CHECK-LABEL: ld1_x4_i32_z0_z4_z8_z12_scalar:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    lsl x8, x1, #2
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1w { z0.s, z4.s, z8.s, z12.s }, pn8/z, [x0, x8, lsl #2]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z2.d, z8.d
+; CHECK-NEXT:    mov z3.d, z12.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z4.d
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x4_i32_z0_z4_z8_z12_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1w { z0.s, z4.s, z8.s, z12.s }, pn8/z, [x0, x1, lsl #2]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x4_i32_z0_z4_z8_z12_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1w { z0.s - z3.s }, pn8/z, [x0, x1, lsl #2]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i32, ptr %ptr, i64 %index
+  %res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld1.pn.x4.nxv4i32(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 0
+  %v0 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> poison, <vscale x 4 x i32> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 1
+  %v1 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v0, <vscale x 4 x i32> %res.v1, i64 4)
+  %res.v2 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 2
+  %v2 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v1, <vscale x 4 x i32> %res.v2, i64 8)
+  %res.v3 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 3
+  %v3 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v2, <vscale x 4 x i32> %res.v3, i64 12)
+  ret <vscale x 16 x i32> %v3
+}
+
+define <vscale x 8 x i64> @ld1_x4_i64_z0_z4_z8_z12(<vscale x 2 x i64> %unused, <vscale x 2 x i64> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; CHECK-LABEL: ld1_x4_i64_z0_z4_z8_z12:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1d { z0.d, z4.d, z8.d, z12.d }, pn8/z, [x0]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z2.d, z8.d
+; CHECK-NEXT:    mov z3.d, z12.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z4.d
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x4_i64_z0_z4_z8_z12:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1d { z0.d, z4.d, z8.d, z12.d }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x4_i64_z0_z4_z8_z12:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1d { z0.d - z3.d }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld1.pn.x4.nxv2i64(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 0
+  %v0 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> poison, <vscale x 2 x i64> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 1
+  %v1 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v0, <vscale x 2 x i64> %res.v1, i64 2)
+  %res.v2 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 2
+  %v2 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v1, <vscale x 2 x i64> %res.v2, i64 4)
+  %res.v3 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 3
+  %v3 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v2, <vscale x 2 x i64> %res.v3, i64 6)
+  ret <vscale x 8 x i64> %v3
+}
+
+define <vscale x 8 x i64> @ld1_x4_i64_z0_z4_z8_z12_scalar(<vscale x 2 x i64> %unused, <vscale x 2 x i64> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; CHECK-LABEL: ld1_x4_i64_z0_z4_z8_z12_scalar:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT:    addvl sp, sp, #-17
+; CHECK-NEXT:    lsl x8, x1, #3
+; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CHECK-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; CHECK-NEXT:    mov p8.b, p0.b
+; CHECK-NEXT:    ld1d { z0.d, z4.d, z8.d, z12.d }, pn8/z, [x0, x8, lsl #3]
+; CHECK-NEXT:    //APP
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    //NO_APP
+; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CHECK-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z2.d, z8.d
+; CHECK-NEXT:    mov z3.d, z12.d
+; CHECK-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; CHECK-NEXT:    mov z1.d, z4.d
+; CHECK-NEXT:    addvl sp, sp, #17
+; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT:    ret
+; STRIDED-LABEL: ld1_x4_i64_z0_z4_z8_z12_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ld1d { z0.d, z4.d, z8.d, z12.d }, pn8/z, [x0, x1, lsl #3]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ld1_x4_i64_z0_z4_z8_z12_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ld1d { z0.d - z3.d }, pn8/z, [x0, x1, lsl #3]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i64, ptr %ptr, i64 %index
+  %res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld1.pn.x4.nxv2i64(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 0
+  %v0 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> poison, <vscale x 2 x i64> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 1
+  %v1 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v0, <vscale x 2 x i64> %res.v1, i64 2)
+  %res.v2 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 2
+  %v2 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v1, <vscale x 2 x i64> %res.v2, i64 4)
+  %res.v3 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 3
+  %v3 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v2, <vscale x 2 x i64> %res.v3, i64 6)
+  ret <vscale x 8 x i64> %v3
+}
+
+declare <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8>, <vscale x 16 x i8>, i64)
+declare <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16>, <vscale x 8 x i16>, i64)
+declare <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32>, <vscale x 4 x i32>, i64)
+declare <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64>, <vscale x 2 x i64>, i64)
+declare <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8>, <vscale x 16 x i8>, i64)
+declare <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16>, <vscale x 8 x i16>, i64)
+declare <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32>, <vscale x 4 x i32>, i64)
+declare <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64>, <vscale x 2 x i64>, i64)
+declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld1.pn.x2.nxv16i8(target("aarch64.svcount"), ptr)
+declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld1.pn.x2.nxv8i16(target("aarch64.svcount"), ptr)
+declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld1.pn.x2.nxv4i32(target("aarch64.svcount"), ptr)
+declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld1.pn.x2.nxv2i64(target("aarch64.svcount"), ptr)
+declare { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld1.pn.x4.nxv16i8(target("aarch64.svcount"), ptr)
+declare { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld1.pn.x4.nxv8i16(target("aarch64.svcount"), ptr)
+declare { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld1.pn.x4.nxv4i32(target("aarch64.svcount"), ptr)
+declare { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld1.pn.x4.nxv2i64(target("aarch64.svcount"), ptr)

llvm/test/CodeGen/AArch64/sme2-intrinsics-ldnt1.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme2 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=STRIDED
+; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve2p1 -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CONTIGUOUS
+
+define <vscale x 32 x i8> @ldnt1_x2_i8_z0_z8(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; STRIDED-LABEL: ldnt1_x2_i8_z0_z8:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1b { z0.b, z8.b }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x2_i8_z0_z8:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1b { z0.b, z1.b }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv16i8(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 0
+  %v0 = call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 1
+  %v1 = call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> %v0, <vscale x 16 x i8> %res.v1, i64 16)
+  ret <vscale x 32 x i8> %v1
+}
+
+define <vscale x 32 x i8> @ldnt1_x2_i8_z0_z8_scalar(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; STRIDED-LABEL: ldnt1_x2_i8_z0_z8_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1b { z0.b, z8.b }, pn8/z, [x0, x1]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x2_i8_z0_z8_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1b { z0.b, z1.b }, pn8/z, [x0, x1]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i8, ptr %ptr, i64 %index
+  %res = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv16i8(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 0
+  %v0 = call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> poison, <vscale x 16 x i8> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 1
+  %v1 = call <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8> %v0, <vscale x 16 x i8> %res.v1, i64 16)
+  ret <vscale x 32 x i8> %v1
+}
+
+define <vscale x 16 x i16> @ldnt1_x2_i16_z0_z8(<vscale x 8 x i16> %unused, <vscale x 8 x i16> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; STRIDED-LABEL: ldnt1_x2_i16_z0_z8:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1h { z0.h, z8.h }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x2_i16_z0_z8:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1h { z0.h, z1.h }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv8i16(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 0
+  %v0 = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 1
+  %v1 = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> %v0, <vscale x 8 x i16> %res.v1, i64 8)
+  ret <vscale x 16 x i16> %v1
+}
+
+define <vscale x 16 x i16> @ldnt1_x2_i16_z0_z8_scalar(<vscale x 8 x i16> %unused, <vscale x 8 x i16> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; STRIDED-LABEL: ldnt1_x2_i16_z0_z8_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1h { z0.h, z8.h }, pn8/z, [x0, x1, lsl #1]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x2_i16_z0_z8_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1h { z0.h, z1.h }, pn8/z, [x0, x1, lsl #1]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i16, ptr %ptr, i64 %index
+  %res = call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv8i16(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 0
+  %v0 = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> poison, <vscale x 8 x i16> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 1
+  %v1 = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16> %v0, <vscale x 8 x i16> %res.v1, i64 8)
+  ret <vscale x 16 x i16> %v1
+}
+
+define <vscale x 8 x i32> @ldnt1_x2_i32_z0_z8(<vscale x 4 x i32> %unused, <vscale x 4 x i32> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; STRIDED-LABEL: ldnt1_x2_i32_z0_z8:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1w { z0.s, z8.s }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x2_i32_z0_z8:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1w { z0.s, z1.s }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv4i32(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 0
+  %v0 = call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 1
+  %v1 = call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> %v0, <vscale x 4 x i32> %res.v1, i64 4)
+  ret <vscale x 8 x i32> %v1
+}
+
+define <vscale x 8 x i32> @ldnt1_x2_i32_z0_z8_scalar(<vscale x 4 x i32> %unused, <vscale x 4 x i32> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; STRIDED-LABEL: ldnt1_x2_i32_z0_z8_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1w { z0.s, z8.s }, pn8/z, [x0, x1, lsl #2]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x2_i32_z0_z8_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1w { z0.s, z1.s }, pn8/z, [x0, x1, lsl #2]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i32, ptr %ptr, i64 %index
+  %res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv4i32(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 0
+  %v0 = call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> poison, <vscale x 4 x i32> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 1
+  %v1 = call <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32> %v0, <vscale x 4 x i32> %res.v1, i64 4)
+  ret <vscale x 8 x i32> %v1
+}
+
+define <vscale x 4 x i64> @ldnt1_x2_i64_z0_z8(<vscale x 2 x i64> %unused, <vscale x 2 x i64> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; STRIDED-LABEL: ldnt1_x2_i64_z0_z8:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1d { z0.d, z8.d }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x2_i64_z0_z8:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1d { z0.d, z1.d }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv2i64(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 0
+  %v0 = call <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64> poison, <vscale x 2 x i64> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 1
+  %v1 = call <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64> %v0, <vscale x 2 x i64> %res.v1, i64 2)
+  ret <vscale x 4 x i64> %v1
+}
+
+define <vscale x 4 x i64> @ldnt1_x2_i64_z0_z8_scalar(<vscale x 2 x i64> %unused, <vscale x 2 x i64> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; STRIDED-LABEL: ldnt1_x2_i64_z0_z8_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1d { z0.d, z8.d }, pn8/z, [x0, x1, lsl #3]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z8.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x2_i64_z0_z8_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-16
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-2
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1d { z0.d, z1.d }, pn8/z, [x0, x1, lsl #3]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #2
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #16
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i64, ptr %ptr, i64 %index
+  %res = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv2i64(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z4},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z12},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 0
+  %v0 = call <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64> poison, <vscale x 2 x i64> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 1
+  %v1 = call <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64> %v0, <vscale x 2 x i64> %res.v1, i64 2)
+  ret <vscale x 4 x i64> %v1
+}
+
+define <vscale x 64 x i8> @ldnt1_x4_i8_z0_z4_z8_z12(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; STRIDED-LABEL: ldnt1_x4_i8_z0_z4_z8_z12:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1b { z0.b, z4.b, z8.b, z12.b }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x4_i8_z0_z4_z8_z12:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1b { z0.b - z3.b }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv16i8(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 0
+  %v0 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> poison, <vscale x 16 x i8> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 1
+  %v1 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v0, <vscale x 16 x i8> %res.v1, i64 16)
+  %res.v2 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 2
+  %v2 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v1, <vscale x 16 x i8> %res.v2, i64 32)
+  %res.v3 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 3
+  %v3 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v2, <vscale x 16 x i8> %res.v3, i64 48)
+  ret <vscale x 64 x i8> %v3
+}
+
+define <vscale x 64 x i8> @ldnt1_x4_i8_z0_z4_z8_z12_scalar(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; STRIDED-LABEL: ldnt1_x4_i8_z0_z4_z8_z12_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1b { z0.b, z4.b, z8.b, z12.b }, pn8/z, [x0, x1]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x4_i8_z0_z4_z8_z12_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1b { z0.b - z3.b }, pn8/z, [x0, x1]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i8, ptr %ptr, i64 %index
+  %res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv16i8(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 0
+  %v0 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> poison, <vscale x 16 x i8> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 1
+  %v1 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v0, <vscale x 16 x i8> %res.v1, i64 16)
+  %res.v2 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 2
+  %v2 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v1, <vscale x 16 x i8> %res.v2, i64 32)
+  %res.v3 = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res, 3
+  %v3 = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8> %v2, <vscale x 16 x i8> %res.v3, i64 48)
+  ret <vscale x 64 x i8> %v3
+}
+
+define <vscale x 32 x i16> @ldnt1_x4_i16_z0_z4_z8_z12(<vscale x 8 x i16> %unused, <vscale x 8 x i16> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; STRIDED-LABEL: ldnt1_x4_i16_z0_z4_z8_z12:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1h { z0.h, z4.h, z8.h, z12.h }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x4_i16_z0_z4_z8_z12:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1h { z0.h - z3.h }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv8i16(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 0
+  %v0 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> poison, <vscale x 8 x i16> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 1
+  %v1 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v0, <vscale x 8 x i16> %res.v1, i64 8)
+  %res.v2 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 2
+  %v2 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v1, <vscale x 8 x i16> %res.v2, i64 16)
+  %res.v3 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 3
+  %v3 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v2, <vscale x 8 x i16> %res.v3, i64 24)
+  ret <vscale x 32 x i16> %v3
+}
+
+define <vscale x 32 x i16> @ldnt1_x4_i16_z0_z4_z8_z12_scalar(<vscale x 8 x i16> %unused, <vscale x 8 x i16> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; STRIDED-LABEL: ldnt1_x4_i16_z0_z4_z8_z12_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1h { z0.h, z4.h, z8.h, z12.h }, pn8/z, [x0, x1, lsl #1]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x4_i16_z0_z4_z8_z12_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1h { z0.h - z3.h }, pn8/z, [x0, x1, lsl #1]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i16, ptr %ptr, i64 %index
+  %res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv8i16(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 0
+  %v0 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> poison, <vscale x 8 x i16> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 1
+  %v1 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v0, <vscale x 8 x i16> %res.v1, i64 8)
+  %res.v2 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 2
+  %v2 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v1, <vscale x 8 x i16> %res.v2, i64 16)
+  %res.v3 = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res, 3
+  %v3 = call <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16> %v2, <vscale x 8 x i16> %res.v3, i64 24)
+  ret <vscale x 32 x i16> %v3
+}
+
+define <vscale x 16 x i32> @ldnt1_x4_i32_z0_z4_z8_z12(<vscale x 4 x i32> %unused, <vscale x 4 x i32> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; STRIDED-LABEL: ldnt1_x4_i32_z0_z4_z8_z12:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1w { z0.s, z4.s, z8.s, z12.s }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x4_i32_z0_z4_z8_z12:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1w { z0.s - z3.s }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv4i32(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 0
+  %v0 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> poison, <vscale x 4 x i32> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 1
+  %v1 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v0, <vscale x 4 x i32> %res.v1, i64 4)
+  %res.v2 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 2
+  %v2 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v1, <vscale x 4 x i32> %res.v2, i64 8)
+  %res.v3 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 3
+  %v3 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v2, <vscale x 4 x i32> %res.v3, i64 12)
+  ret <vscale x 16 x i32> %v3
+}
+
+define <vscale x 16 x i32> @ldnt1_x4_i32_z0_z4_z8_z12_scalar(<vscale x 4 x i32> %unused, <vscale x 4 x i32> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; STRIDED-LABEL: ldnt1_x4_i32_z0_z4_z8_z12_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1w { z0.s, z4.s, z8.s, z12.s }, pn8/z, [x0, x1, lsl #2]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x4_i32_z0_z4_z8_z12_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1w { z0.s - z3.s }, pn8/z, [x0, x1, lsl #2]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i32, ptr %ptr, i64 %index
+  %res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv4i32(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 0
+  %v0 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> poison, <vscale x 4 x i32> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 1
+  %v1 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v0, <vscale x 4 x i32> %res.v1, i64 4)
+  %res.v2 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 2
+  %v2 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v1, <vscale x 4 x i32> %res.v2, i64 8)
+  %res.v3 = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res, 3
+  %v3 = call <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32> %v2, <vscale x 4 x i32> %res.v3, i64 12)
+  ret <vscale x 16 x i32> %v3
+}
+
+define <vscale x 8 x i64> @ldnt1_x4_i64_z0_z4_z8_z12(<vscale x 2 x i64> %unused, <vscale x 2 x i64> %z1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
+; STRIDED-LABEL: ldnt1_x4_i64_z0_z4_z8_z12:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1d { z0.d, z4.d, z8.d, z12.d }, pn8/z, [x0]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x4_i64_z0_z4_z8_z12:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1d { z0.d - z3.d }, pn8/z, [x0]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv2i64(target("aarch64.svcount") %pn, ptr %ptr)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 0
+  %v0 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> poison, <vscale x 2 x i64> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 1
+  %v1 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v0, <vscale x 2 x i64> %res.v1, i64 2)
+  %res.v2 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 2
+  %v2 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v1, <vscale x 2 x i64> %res.v2, i64 4)
+  %res.v3 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 3
+  %v3 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v2, <vscale x 2 x i64> %res.v3, i64 6)
+  ret <vscale x 8 x i64> %v3
+}
+
+define <vscale x 8 x i64> @ldnt1_x4_i64_z0_z4_z8_z12_scalar(<vscale x 2 x i64> %unused, <vscale x 2 x i64> %z1, target("aarch64.svcount") %pn, ptr %ptr, i64 %index) nounwind {
+; STRIDED-LABEL: ldnt1_x4_i64_z0_z4_z8_z12_scalar:
+; STRIDED:       // %bb.0:
+; STRIDED-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; STRIDED-NEXT:    addvl sp, sp, #-17
+; STRIDED-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; STRIDED-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z12, [sp, #12, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z11, [sp, #13, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z10, [sp, #14, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z9, [sp, #15, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    str z8, [sp, #16, mul vl] // 16-byte Folded Spill
+; STRIDED-NEXT:    mov p8.b, p0.b
+; STRIDED-NEXT:    ldnt1d { z0.d, z4.d, z8.d, z12.d }, pn8/z, [x0, x1, lsl #3]
+; STRIDED-NEXT:    //APP
+; STRIDED-NEXT:    nop
+; STRIDED-NEXT:    //NO_APP
+; STRIDED-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; STRIDED-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z2.d, z8.d
+; STRIDED-NEXT:    mov z3.d, z12.d
+; STRIDED-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z12, [sp, #12, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z11, [sp, #13, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z10, [sp, #14, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z9, [sp, #15, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    ldr z8, [sp, #16, mul vl] // 16-byte Folded Reload
+; STRIDED-NEXT:    mov z1.d, z4.d
+; STRIDED-NEXT:    addvl sp, sp, #17
+; STRIDED-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; STRIDED-NEXT:    ret
+;
+; CONTIGUOUS-LABEL: ldnt1_x4_i64_z0_z4_z8_z12_scalar:
+; CONTIGUOUS:       // %bb.0:
+; CONTIGUOUS-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-15
+; CONTIGUOUS-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z23, [sp, #1, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z22, [sp, #2, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z21, [sp, #3, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z20, [sp, #4, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z19, [sp, #5, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z18, [sp, #6, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z17, [sp, #7, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z16, [sp, #8, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z15, [sp, #9, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z14, [sp, #10, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z13, [sp, #11, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z11, [sp, #12, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z10, [sp, #13, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    str z9, [sp, #14, mul vl] // 16-byte Folded Spill
+; CONTIGUOUS-NEXT:    addvl sp, sp, #-4
+; CONTIGUOUS-NEXT:    mov p8.b, p0.b
+; CONTIGUOUS-NEXT:    ldnt1d { z0.d - z3.d }, pn8/z, [x0, x1, lsl #3]
+; CONTIGUOUS-NEXT:    str z0, [sp]
+; CONTIGUOUS-NEXT:    str z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    str z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    str z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    //APP
+; CONTIGUOUS-NEXT:    nop
+; CONTIGUOUS-NEXT:    //NO_APP
+; CONTIGUOUS-NEXT:    ldr z0, [sp]
+; CONTIGUOUS-NEXT:    ldr z1, [sp, #1, mul vl]
+; CONTIGUOUS-NEXT:    ldr z2, [sp, #2, mul vl]
+; CONTIGUOUS-NEXT:    ldr z3, [sp, #3, mul vl]
+; CONTIGUOUS-NEXT:    addvl sp, sp, #4
+; CONTIGUOUS-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z23, [sp, #1, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z22, [sp, #2, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z21, [sp, #3, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z20, [sp, #4, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z19, [sp, #5, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z18, [sp, #6, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z17, [sp, #7, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z16, [sp, #8, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z15, [sp, #9, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z14, [sp, #10, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z13, [sp, #11, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z11, [sp, #12, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z10, [sp, #13, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    ldr z9, [sp, #14, mul vl] // 16-byte Folded Reload
+; CONTIGUOUS-NEXT:    addvl sp, sp, #15
+; CONTIGUOUS-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
+; CONTIGUOUS-NEXT:    ret
+  %base = getelementptr i64, ptr %ptr, i64 %index
+  %res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv2i64(target("aarch64.svcount") %pn, ptr %base)
+  call void asm sideeffect "nop", "~{z1},~{z2},~{z3},~{z5},~{z6},~{z7},~{z9},~{z10},~{z11},~{z13},~{z14},~{z15},~{z16},~{z17},~{z18},~{z19},~{z20},~{z21},~{z22},~{z23},~{z24},~{z25},~{z26},~{z27},~{z28},~{z29},~{z30},~{z31}"() nounwind
+  %res.v0 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 0
+  %v0 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> poison, <vscale x 2 x i64> %res.v0, i64 0)
+  %res.v1 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 1
+  %v1 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v0, <vscale x 2 x i64> %res.v1, i64 2)
+  %res.v2 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 2
+  %v2 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v1, <vscale x 2 x i64> %res.v2, i64 4)
+  %res.v3 = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res, 3
+  %v3 = call <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64> %v2, <vscale x 2 x i64> %res.v3, i64 6)
+  ret <vscale x 8 x i64> %v3
+}
+
+declare <vscale x 32 x i8> @llvm.vector.insert.nxv32i8.nxv16i8(<vscale x 32 x i8>, <vscale x 16 x i8>, i64)
+declare <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.nxv8i16(<vscale x 16 x i16>, <vscale x 8 x i16>, i64)
+declare <vscale x 8 x i32> @llvm.vector.insert.nxv8i32.nxv4i32(<vscale x 8 x i32>, <vscale x 4 x i32>, i64)
+declare <vscale x 4 x i64> @llvm.vector.insert.nxv4i64.nxv2i64(<vscale x 4 x i64>, <vscale x 2 x i64>, i64)
+declare <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.nxv16i8(<vscale x 64 x i8>, <vscale x 16 x i8>, i64)
+declare <vscale x 32 x i16> @llvm.vector.insert.nxv32i16.nxv8i16(<vscale x 32 x i16>, <vscale x 8 x i16>, i64)
+declare <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x i32>, <vscale x 4 x i32>, i64)
+declare <vscale x 8 x i64> @llvm.vector.insert.nxv8i64.nxv2i64(<vscale x 8 x i64>, <vscale x 2 x i64>, i64)
+declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv16i8(target("aarch64.svcount"), ptr)
+declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv8i16(target("aarch64.svcount"), ptr)
+declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv4i32(target("aarch64.svcount"), ptr)
+declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ldnt1.pn.x2.nxv2i64(target("aarch64.svcount"), ptr)
+declare { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv16i8(target("aarch64.svcount"), ptr)
+declare { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv8i16(target("aarch64.svcount"), ptr)
+declare { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv4i32(target("aarch64.svcount"), ptr)
+declare { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ldnt1.pn.x4.nxv2i64(target("aarch64.svcount"), ptr)