From: Oliver Stannard <oliver.stannard@arm.com>
Date: Mon, 18 May 2015 16:23:33 +0000 (+0000)
Subject: [LLVM - ARM/AArch64] Add ACLE special register intrinsics
X-Git-Url: http://demsky.eecs.uci.edu/git/?a=commitdiff_plain;h=d811b4bacb44065e4a5b7b477e35ec7e5a02acbd;p=oota-llvm.git

[LLVM - ARM/AArch64] Add ACLE special register intrinsics

This patch implements LLVM support for the ACLE special register intrinsics in
section 10.1, __arm_{w,r}sr{,p,64}.

This patch is intended to lower the read/write_register instrinsics, used to
implement the special register intrinsics in the clang patch for special
register intrinsics (see http://reviews.llvm.org/D9697), to ARM specific
instructions MRC,MCR,MSR etc. to allow reading an writing of coprocessor
registers in AArch32 and AArch64. This is done by inspecting the register
string passed to the intrinsic and then lowering to the appropriate
instruction.

Patch by Luke Cheeseman.

Differential Revision: http://reviews.llvm.org/D9699



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@237579 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp b/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
index 3c5d32f1749..d4f69e88c2e 100644
--- a/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
+++ b/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
@@ -4054,10 +4054,9 @@ SelectionDAGBuilder::visitIntrinsicCall(const CallInst &I, unsigned Intrinsic) {
   case Intrinsic::write_register: {
     Value *Reg = I.getArgOperand(0);
     Value *RegValue = I.getArgOperand(1);
-    SDValue Chain = getValue(RegValue).getOperand(0);
     SDValue RegName =
         DAG.getMDNode(cast<MDNode>(cast<MetadataAsValue>(Reg)->getMetadata()));
-    DAG.setRoot(DAG.getNode(ISD::WRITE_REGISTER, sdl, MVT::Other, Chain,
+    DAG.setRoot(DAG.getNode(ISD::WRITE_REGISTER, sdl, MVT::Other, getRoot(),
                             RegName, getValue(RegValue)));
     return nullptr;
   }
diff --git a/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp b/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
index 78a2021f79a..bdf64c6f980 100644
--- a/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
+++ b/lib/Target/AArch64/AArch64ISelDAGToDAG.cpp
@@ -156,6 +156,9 @@ public:
 
   SDNode *SelectLIBM(SDNode *N);
 
+  SDNode *SelectReadRegister(SDNode *N);
+  SDNode *SelectWriteRegister(SDNode *N);
+
 // Include the pieces autogenerated from the target description.
 #include "AArch64GenDAGISel.inc"
 
@@ -2114,6 +2117,115 @@ AArch64DAGToDAGISel::SelectCVTFixedPosOperand(SDValue N, SDValue &FixedPos,
   return true;
 }
 
+// Inspects a register string of the form o0:op1:CRn:CRm:op2 gets the fields
+// of the string and obtains the integer values from them and combines these
+// into a single value to be used in the MRS/MSR instruction.
+static int getIntOperandFromRegisterString(StringRef RegString) {
+  SmallVector<StringRef, 5> Fields;
+  RegString.split(Fields, ":");
+
+  if (Fields.size() == 1)
+    return -1;
+
+  assert(Fields.size() == 5
+            && "Invalid number of fields in read register string");
+
+  SmallVector<int, 5> Ops;
+  bool AllIntFields = true;
+
+  for (StringRef Field : Fields) {
+    unsigned IntField;
+    AllIntFields &= !Field.getAsInteger(10, IntField);
+    Ops.push_back(IntField);
+  }
+
+  assert(AllIntFields &&
+          "Unexpected non-integer value in special register string.");
+
+  // Need to combine the integer fields of the string into a single value
+  // based on the bit encoding of MRS/MSR instruction.
+  return (Ops[0] << 14) | (Ops[1] << 11) | (Ops[2] << 7) |
+         (Ops[3] << 3) | (Ops[4]);
+}
+
+// Lower the read_register intrinsic to an MRS instruction node if the special
+// register string argument is either of the form detailed in the ALCE (the
+// form described in getIntOperandsFromRegsterString) or is a named register
+// known by the MRS SysReg mapper.
+SDNode *AArch64DAGToDAGISel::SelectReadRegister(SDNode *N) {
+  const MDNodeSDNode *MD = dyn_cast<MDNodeSDNode>(N->getOperand(0));
+  const MDString *RegString = dyn_cast<MDString>(MD->getMD()->getOperand(0));
+  SDLoc DL(N);
+
+  int Reg = getIntOperandFromRegisterString(RegString->getString());
+  if (Reg != -1)
+    return CurDAG->getMachineNode(AArch64::MRS, DL, N->getSimpleValueType(0),
+                                  CurDAG->getTargetConstant(Reg, DL, MVT::i32));
+
+  // Use the sysreg mapper to map the remaining possible strings to the
+  // value for the register to be used for the instruction operand.
+  AArch64SysReg::MRSMapper mapper;
+  bool IsValidSpecialReg;
+  Reg = mapper.fromString(RegString->getString(),
+                          Subtarget->getFeatureBits(),
+                          IsValidSpecialReg);
+  if (IsValidSpecialReg)
+    return CurDAG->getMachineNode(AArch64::MRS, DL, N->getSimpleValueType(0),
+                                  CurDAG->getTargetConstant(Reg, DL, MVT::i32));
+  return nullptr;
+}
+
+// Lower the write_register intrinsic to an MSR instruction node if the special
+// register string argument is either of the form detailed in the ALCE (the
+// form described in getIntOperandsFromRegsterString) or is a named register
+// known by the MSR SysReg mapper.
+SDNode *AArch64DAGToDAGISel::SelectWriteRegister(SDNode *N) {
+  const MDNodeSDNode *MD = dyn_cast<MDNodeSDNode>(N->getOperand(1));
+  const MDString *RegString = dyn_cast<MDString>(MD->getMD()->getOperand(0));
+  SDLoc DL(N);
+
+  int Reg = getIntOperandFromRegisterString(RegString->getString());
+  if (Reg != -1)
+    return CurDAG->getMachineNode(AArch64::MSR, DL, MVT::Other,
+                                  CurDAG->getTargetConstant(Reg, DL, MVT::i32),
+                                  N->getOperand(2), N->getOperand(0));
+
+  // Check if the register was one of those allowed as the pstatefield value in
+  // the MSR (immediate) instruction. To accept the values allowed in the
+  // pstatefield for the MSR (immediate) instruction, we also require that an
+  // immediate value has been provided as an argument, we know that this is
+  // the case as it has been ensured by semantic checking.
+  AArch64PState::PStateMapper PMapper;
+  bool IsValidSpecialReg;
+  Reg = PMapper.fromString(RegString->getString(),
+                           Subtarget->getFeatureBits(),
+                           IsValidSpecialReg);
+  if (IsValidSpecialReg) {
+    assert (isa<ConstantSDNode>(N->getOperand(2))
+              && "Expected a constant integer expression.");
+    uint64_t Immed = cast<ConstantSDNode>(N->getOperand(2))->getZExtValue();
+    return CurDAG->getMachineNode(AArch64::MSRpstate, DL, MVT::Other,
+                                  CurDAG->getTargetConstant(Reg, DL, MVT::i32),
+                                  CurDAG->getTargetConstant(Immed, DL, MVT::i16),
+                                  N->getOperand(0));
+  }
+
+  // Use the sysreg mapper to attempt to map the remaining possible strings
+  // to the value for the register to be used for the MSR (register)
+  // instruction operand.
+  AArch64SysReg::MSRMapper Mapper;
+  Reg = Mapper.fromString(RegString->getString(),
+                          Subtarget->getFeatureBits(),
+                          IsValidSpecialReg);
+
+  if (IsValidSpecialReg)
+    return CurDAG->getMachineNode(AArch64::MSR, DL, MVT::Other,
+                                  CurDAG->getTargetConstant(Reg, DL, MVT::i32),
+                                  N->getOperand(2), N->getOperand(0));
+
+  return nullptr;
+}
+
 SDNode *AArch64DAGToDAGISel::Select(SDNode *Node) {
   // Dump information about the Node being selected
   DEBUG(errs() << "Selecting: ");
@@ -2135,6 +2247,16 @@ SDNode *AArch64DAGToDAGISel::Select(SDNode *Node) {
   default:
     break;
 
+  case ISD::READ_REGISTER:
+    if (SDNode *Res = SelectReadRegister(Node))
+      return Res;
+    break;
+
+  case ISD::WRITE_REGISTER:
+    if (SDNode *Res = SelectWriteRegister(Node))
+      return Res;
+    break;
+
   case ISD::ADD:
     if (SDNode *I = SelectMLAV64LaneV128(Node))
       return I;
diff --git a/lib/Target/AArch64/AArch64ISelLowering.cpp b/lib/Target/AArch64/AArch64ISelLowering.cpp
index 6251d4a5d26..eca3297c7b0 100644
--- a/lib/Target/AArch64/AArch64ISelLowering.cpp
+++ b/lib/Target/AArch64/AArch64ISelLowering.cpp
@@ -4064,7 +4064,8 @@ unsigned AArch64TargetLowering::getRegisterByName(const char* RegName,
                        .Default(0);
   if (Reg)
     return Reg;
-  report_fatal_error("Invalid register name global variable");
+  report_fatal_error(Twine("Invalid register name \""
+                              + StringRef(RegName)  + "\"."));
 }
 
 SDValue AArch64TargetLowering::LowerRETURNADDR(SDValue Op,
diff --git a/lib/Target/ARM/ARMISelDAGToDAG.cpp b/lib/Target/ARM/ARMISelDAGToDAG.cpp
index 4405625e47c..e9acca5c050 100644
--- a/lib/Target/ARM/ARMISelDAGToDAG.cpp
+++ b/lib/Target/ARM/ARMISelDAGToDAG.cpp
@@ -15,6 +15,7 @@
 #include "ARMBaseInstrInfo.h"
 #include "ARMTargetMachine.h"
 #include "MCTargetDesc/ARMAddressingModes.h"
+#include "llvm/ADT/StringSwitch.h"
 #include "llvm/CodeGen/MachineFrameInfo.h"
 #include "llvm/CodeGen/MachineFunction.h"
 #include "llvm/CodeGen/MachineInstrBuilder.h"
@@ -251,6 +252,9 @@ private:
   // Select special operations if node forms integer ABS pattern
   SDNode *SelectABSOp(SDNode *N);
 
+  SDNode *SelectReadRegister(SDNode *N);
+  SDNode *SelectWriteRegister(SDNode *N);
+
   SDNode *SelectInlineAsm(SDNode *N);
 
   SDNode *SelectConcatVector(SDNode *N);
@@ -2457,6 +2461,18 @@ SDNode *ARMDAGToDAGISel::Select(SDNode *N) {
 
   switch (N->getOpcode()) {
   default: break;
+  case ISD::WRITE_REGISTER: {
+    SDNode *ResNode = SelectWriteRegister(N);
+    if (ResNode)
+      return ResNode;
+    break;
+  }
+  case ISD::READ_REGISTER: {
+    SDNode *ResNode = SelectReadRegister(N);
+    if (ResNode)
+      return ResNode;
+    break;
+  }
   case ISD::INLINEASM: {
     SDNode *ResNode = SelectInlineAsm(N);
     if (ResNode)
@@ -3336,6 +3352,413 @@ SDNode *ARMDAGToDAGISel::Select(SDNode *N) {
   return SelectCode(N);
 }
 
+// Inspect a register string of the form
+// cp<coprocessor>:<opc1>:c<CRn>:c<CRm>:<opc2> (32bit) or
+// cp<coprocessor>:<opc1>:c<CRm> (64bit) inspect the fields of the string
+// and obtain the integer operands from them, adding these operands to the
+// provided vector.
+static void getIntOperandsFromRegisterString(StringRef RegString,
+                                             SelectionDAG *CurDAG, SDLoc DL,
+                                             std::vector<SDValue>& Ops) {
+  SmallVector<StringRef, 5> Fields;
+  RegString.split(Fields, ":");
+
+  if (Fields.size() > 1) {
+    bool AllIntFields = true;
+
+    for (StringRef Field : Fields) {
+      // Need to trim out leading 'cp' characters and get the integer field.
+      unsigned IntField;
+      AllIntFields &= !Field.trim("CPcp").getAsInteger(10, IntField);
+      Ops.push_back(CurDAG->getTargetConstant(IntField, DL, MVT::i32));
+    }
+
+    assert(AllIntFields &&
+            "Unexpected non-integer value in special register string.");
+  }
+}
+
+// Maps a Banked Register string to its mask value. The mask value returned is
+// for use in the MRSbanked / MSRbanked instruction nodes as the Banked Register
+// mask operand, which expresses which register is to be used, e.g. r8, and in
+// which mode it is to be used, e.g. usr. Returns -1 to signify that the string
+// was invalid.
+static inline int getBankedRegisterMask(StringRef RegString) {
+  return StringSwitch<int>(RegString.lower())
+          .Case("r8_usr", 0x00)
+          .Case("r9_usr", 0x01)
+          .Case("r10_usr", 0x02)
+          .Case("r11_usr", 0x03)
+          .Case("r12_usr", 0x04)
+          .Case("sp_usr", 0x05)
+          .Case("lr_usr", 0x06)
+          .Case("r8_fiq", 0x08)
+          .Case("r9_fiq", 0x09)
+          .Case("r10_fiq", 0x0a)
+          .Case("r11_fiq", 0x0b)
+          .Case("r12_fiq", 0x0c)
+          .Case("sp_fiq", 0x0d)
+          .Case("lr_fiq", 0x0e)
+          .Case("lr_irq", 0x10)
+          .Case("sp_irq", 0x11)
+          .Case("lr_svc", 0x12)
+          .Case("sp_svc", 0x13)
+          .Case("lr_abt", 0x14)
+          .Case("sp_abt", 0x15)
+          .Case("lr_und", 0x16)
+          .Case("sp_und", 0x17)
+          .Case("lr_mon", 0x1c)
+          .Case("sp_mon", 0x1d)
+          .Case("elr_hyp", 0x1e)
+          .Case("sp_hyp", 0x1f)
+          .Case("spsr_fiq", 0x2e)
+          .Case("spsr_irq", 0x30)
+          .Case("spsr_svc", 0x32)
+          .Case("spsr_abt", 0x34)
+          .Case("spsr_und", 0x36)
+          .Case("spsr_mon", 0x3c)
+          .Case("spsr_hyp", 0x3e)
+          .Default(-1);
+}
+
+// Maps a MClass special register string to its value for use in the
+// t2MRS_M / t2MSR_M instruction nodes as the SYSm value operand.
+// Returns -1 to signify that the string was invalid.
+static inline int getMClassRegisterSYSmValueMask(StringRef RegString) {
+  return StringSwitch<int>(RegString.lower())
+          .Case("apsr", 0x0)
+          .Case("iapsr", 0x1)
+          .Case("eapsr", 0x2)
+          .Case("xpsr", 0x3)
+          .Case("ipsr", 0x5)
+          .Case("epsr", 0x6)
+          .Case("iepsr", 0x7)
+          .Case("msp", 0x8)
+          .Case("psp", 0x9)
+          .Case("primask", 0x10)
+          .Case("basepri", 0x11)
+          .Case("basepri_max", 0x12)
+          .Case("faultmask", 0x13)
+          .Case("control", 0x14)
+          .Default(-1);
+}
+
+// The flags here are common to those allowed for apsr in the A class cores and
+// those allowed for the special registers in the M class cores. Returns a
+// value representing which flags were present, -1 if invalid.
+static inline int getMClassFlagsMask(StringRef Flags) {
+  if (Flags.empty())
+    return 0x3;
+
+  return StringSwitch<int>(Flags)
+          .Case("g", 0x1)
+          .Case("nzcvq", 0x2)
+          .Case("nzcvqg", 0x3)
+          .Default(-1);
+}
+
+static int getMClassRegisterMask(StringRef Reg, StringRef Flags, bool IsRead,
+                                 const ARMSubtarget *Subtarget) {
+  // Ensure that the register (without flags) was a valid M Class special
+  // register.
+  int SYSmvalue = getMClassRegisterSYSmValueMask(Reg);
+  if (SYSmvalue == -1)
+    return -1;
+
+  // basepri, basepri_max and faultmask are only valid for V7m.
+  if (!Subtarget->hasV7Ops() && SYSmvalue >= 0x11 && SYSmvalue <= 0x13)
+    return -1;
+
+  // If it was a read then we won't be expecting flags and so at this point
+  // we can return the mask.
+  if (IsRead) {
+    assert (Flags.empty() && "Unexpected flags for reading M class register.");
+    return SYSmvalue;
+  }
+
+  // We know we are now handling a write so need to get the mask for the flags.
+  int Mask = getMClassFlagsMask(Flags);
+
+  // Only apsr, iapsr, eapsr, xpsr can have flags. The other register values
+  // shouldn't have flags present.
+  if ((SYSmvalue < 0x4 && Mask == -1) || (SYSmvalue > 0x4 && !Flags.empty()))
+    return -1;
+
+  // The _g and _nzcvqg versions are only valid if the DSP extension is
+  // available.
+  if (!Subtarget->hasThumb2DSP() && (Mask & 0x2))
+    return -1;
+
+  // The register was valid so need to put the mask in the correct place
+  // (the flags need to be in bits 11-10) and combine with the SYSmvalue to
+  // construct the operand for the instruction node.
+  if (SYSmvalue < 0x4)
+    return SYSmvalue | Mask << 10;
+
+  return SYSmvalue;
+}
+
+static int getARClassRegisterMask(StringRef Reg, StringRef Flags) {
+  // The mask operand contains the special register (R Bit) in bit 4, whether
+  // the register is spsr (R bit is 1) or one of cpsr/apsr (R bit is 0), and
+  // bits 3-0 contains the fields to be accessed in the special register, set by
+  // the flags provided with the register.
+  int Mask = 0;
+  if (Reg == "apsr") {
+    // The flags permitted for apsr are the same flags that are allowed in
+    // M class registers. We get the flag value and then shift the flags into
+    // the correct place to combine with the mask.
+    Mask = getMClassFlagsMask(Flags);
+    if (Mask == -1)
+      return -1;
+    return Mask << 2;
+  }
+
+  if (Reg != "cpsr" && Reg != "spsr") {
+    return -1;
+  }
+
+  // This is the same as if the flags were "fc"
+  if (Flags.empty() || Flags == "all")
+    return Mask | 0x9;
+
+  // Inspect the supplied flags string and set the bits in the mask for
+  // the relevant and valid flags allowed for cpsr and spsr.
+  for (char Flag : Flags) {
+    int FlagVal;
+    switch (Flag) {
+      case 'c':
+        FlagVal = 0x1;
+        break;
+      case 'x':
+        FlagVal = 0x2;
+        break;
+      case 's':
+        FlagVal = 0x4;
+        break;
+      case 'f':
+        FlagVal = 0x8;
+        break;
+      default:
+        FlagVal = 0;
+    }
+
+    // This avoids allowing strings where the same flag bit appears twice.
+    if (!FlagVal || (Mask & FlagVal))
+      return -1;
+    Mask |= FlagVal;
+  }
+
+  // If the register is spsr then we need to set the R bit.
+  if (Reg == "spsr")
+    Mask |= 0x10;
+
+  return Mask;
+}
+
+// Lower the read_register intrinsic to ARM specific DAG nodes
+// using the supplied metadata string to select the instruction node to use
+// and the registers/masks to construct as operands for the node.
+SDNode *ARMDAGToDAGISel::SelectReadRegister(SDNode *N){
+  const MDNodeSDNode *MD = dyn_cast<MDNodeSDNode>(N->getOperand(0));
+  const MDString *RegString = dyn_cast<MDString>(MD->getMD()->getOperand(0));
+  bool IsThumb2 = Subtarget->isThumb2();
+  SDLoc DL(N);
+
+  std::vector<SDValue> Ops;
+  getIntOperandsFromRegisterString(RegString->getString(), CurDAG, DL, Ops);
+
+  if (!Ops.empty()) {
+    // If the special register string was constructed of fields (as defined
+    // in the ACLE) then need to lower to MRC node (32 bit) or
+    // MRRC node(64 bit), we can make the distinction based on the number of
+    // operands we have.
+    unsigned Opcode;
+    SmallVector<EVT, 2> ResTypes;
+    if (Ops.size() == 5){
+      Opcode = IsThumb2 ? ARM::t2MRC : ARM::MRC;
+      ResTypes = { MVT::i32 };
+    } else {
+      assert(Ops.size() == 3 &&
+              "Invalid number of fields in special register string.");
+      Opcode = IsThumb2 ? ARM::t2MRRC : ARM::MRRC;
+      ResTypes = { MVT::i32, MVT::i32 };
+    }
+
+    Ops.push_back(getAL(CurDAG, DL));
+    Ops.push_back(CurDAG->getRegister(0, MVT::i32));
+    return CurDAG->getMachineNode(Opcode, DL, ResTypes, Ops);
+  }
+
+  std::string SpecialReg = RegString->getString().lower();
+
+  int BankedReg = getBankedRegisterMask(SpecialReg);
+  if (BankedReg != -1) {
+    Ops = { CurDAG->getTargetConstant(BankedReg, DL, MVT::i32),
+            getAL(CurDAG, DL), CurDAG->getRegister(0, MVT::i32) };
+    return CurDAG->getMachineNode(IsThumb2 ? ARM::t2MRSbanked : ARM::MRSbanked,
+                                  DL, MVT::i32, Ops);
+  }
+
+  // The VFP registers are read by creating SelectionDAG nodes with opcodes
+  // corresponding to the register that is being read from. So we switch on the
+  // string to find which opcode we need to use.
+  unsigned Opcode = StringSwitch<unsigned>(SpecialReg)
+                    .Case("fpscr", ARM::VMRS)
+                    .Case("fpexc", ARM::VMRS_FPEXC)
+                    .Case("fpsid", ARM::VMRS_FPSID)
+                    .Case("mvfr0", ARM::VMRS_MVFR0)
+                    .Case("mvfr1", ARM::VMRS_MVFR1)
+                    .Case("mvfr2", ARM::VMRS_MVFR2)
+                    .Case("fpinst", ARM::VMRS_FPINST)
+                    .Case("fpinst2", ARM::VMRS_FPINST2)
+                    .Default(0);
+
+  // If an opcode was found then we can lower the read to a VFP instruction.
+  if (Opcode) {
+    if (!Subtarget->hasVFP2())
+      return nullptr;
+    if (Opcode == ARM::VMRS_MVFR2 && !Subtarget->hasFPARMv8())
+      return nullptr;
+
+    Ops = { getAL(CurDAG, DL), CurDAG->getRegister(0, MVT::i32) };
+    return CurDAG->getMachineNode(Opcode, DL, MVT::i32, Ops);
+  }
+
+  // If the target is M Class then need to validate that the register string
+  // is an acceptable value, so check that a mask can be constructed from the
+  // string.
+  if (Subtarget->isMClass()) {
+    int SYSmValue = getMClassRegisterMask(SpecialReg, "", true, Subtarget);
+    if (SYSmValue == -1)
+      return nullptr;
+
+    SDValue Ops[] = { CurDAG->getTargetConstant(SYSmValue, DL, MVT::i32),
+                      getAL(CurDAG, DL),
+                      CurDAG->getRegister(0, MVT::i32) };
+    return CurDAG->getMachineNode(ARM::t2MRS_M, DL, MVT::i32, Ops);
+  }
+
+  // Here we know the target is not M Class so we need to check if it is one
+  // of the remaining possible values which are apsr, cpsr or spsr.
+  if (SpecialReg == "apsr" || SpecialReg == "cpsr") {
+    Ops = { getAL(CurDAG, DL), CurDAG->getRegister(0, MVT::i32) };
+    return CurDAG->getMachineNode(IsThumb2 ? ARM::t2MRS_AR : ARM::MRS, DL,
+                                  MVT::i32, Ops);
+  }
+
+  if (SpecialReg == "spsr") {
+    Ops = { getAL(CurDAG, DL), CurDAG->getRegister(0, MVT::i32) };
+    return CurDAG->getMachineNode(IsThumb2 ? ARM::t2MRSsys_AR : ARM::MRSsys,
+                                  DL, MVT::i32, Ops);
+  }
+
+  return nullptr;
+}
+
+// Lower the write_register intrinsic to ARM specific DAG nodes
+// using the supplied metadata string to select the instruction node to use
+// and the registers/masks to use in the nodes
+SDNode *ARMDAGToDAGISel::SelectWriteRegister(SDNode *N){
+  const MDNodeSDNode *MD = dyn_cast<MDNodeSDNode>(N->getOperand(1));
+  const MDString *RegString = dyn_cast<MDString>(MD->getMD()->getOperand(0));
+  bool IsThumb2 = Subtarget->isThumb2();
+  SDLoc DL(N);
+
+  std::vector<SDValue> Ops;
+  getIntOperandsFromRegisterString(RegString->getString(), CurDAG, DL, Ops);
+
+  if (!Ops.empty()) {
+    // If the special register string was constructed of fields (as defined
+    // in the ACLE) then need to lower to MCR node (32 bit) or
+    // MCRR node(64 bit), we can make the distinction based on the number of
+    // operands we have.
+    unsigned Opcode;
+    if (Ops.size() == 5) {
+      Opcode = IsThumb2 ? ARM::t2MCR : ARM::MCR;
+      Ops.insert(Ops.begin()+2, N->getOperand(2));
+    } else {
+      assert(Ops.size() == 3 &&
+              "Invalid number of fields in special register string.");
+      Opcode = IsThumb2 ? ARM::t2MCRR : ARM::MCRR;
+      SDValue WriteValue[] = { N->getOperand(2), N->getOperand(3) };
+      Ops.insert(Ops.begin()+2, WriteValue, WriteValue+2);
+    }
+
+    Ops.push_back(getAL(CurDAG, DL));
+    Ops.push_back(CurDAG->getRegister(0, MVT::i32));
+    Ops.push_back(N->getOperand(0));
+
+    return CurDAG->getMachineNode(Opcode, DL, MVT::Other, Ops);
+  }
+
+  std::string SpecialReg = RegString->getString().lower();
+  int BankedReg = getBankedRegisterMask(SpecialReg);
+  if (BankedReg != -1) {
+    Ops = { CurDAG->getTargetConstant(BankedReg, DL, MVT::i32), N->getOperand(2),
+            getAL(CurDAG, DL), CurDAG->getRegister(0, MVT::i32),
+            N->getOperand(0) };
+    return CurDAG->getMachineNode(IsThumb2 ? ARM::t2MSRbanked : ARM::MSRbanked,
+                                  DL, MVT::Other, Ops);
+  }
+
+  // The VFP registers are written to by creating SelectionDAG nodes with
+  // opcodes corresponding to the register that is being written. So we switch
+  // on the string to find which opcode we need to use.
+  unsigned Opcode = StringSwitch<unsigned>(SpecialReg)
+                    .Case("fpscr", ARM::VMSR)
+                    .Case("fpexc", ARM::VMSR_FPEXC)
+                    .Case("fpsid", ARM::VMSR_FPSID)
+                    .Case("fpinst", ARM::VMSR_FPINST)
+                    .Case("fpinst2", ARM::VMSR_FPINST2)
+                    .Default(0);
+
+  if (Opcode) {
+    if (!Subtarget->hasVFP2())
+      return nullptr;
+    Ops = { N->getOperand(2), getAL(CurDAG, DL),
+            CurDAG->getRegister(0, MVT::i32), N->getOperand(0) };
+    return CurDAG->getMachineNode(Opcode, DL, MVT::Other, Ops);
+  }
+
+  SmallVector<StringRef, 5> Fields;
+  StringRef(SpecialReg).split(Fields, "_", 1, false);
+  std::string Reg = Fields[0].str();
+  StringRef Flags = Fields.size() == 2 ? Fields[1] : "";
+
+  // If the target was M Class then need to validate the special register value
+  // and retrieve the mask for use in the instruction node.
+  if (Subtarget->isMClass()) {
+    // basepri_max gets split so need to correct Reg and Flags.
+    if (SpecialReg == "basepri_max") {
+      Reg = SpecialReg;
+      Flags = "";
+    }
+    int SYSmValue = getMClassRegisterMask(Reg, Flags, false, Subtarget);
+    if (SYSmValue == -1)
+      return nullptr;
+
+    SDValue Ops[] = { CurDAG->getTargetConstant(SYSmValue, DL, MVT::i32),
+                      N->getOperand(2), getAL(CurDAG, DL),
+                      CurDAG->getRegister(0, MVT::i32), N->getOperand(0) };
+    return CurDAG->getMachineNode(ARM::t2MSR_M, DL, MVT::Other, Ops);
+  }
+
+  // We then check to see if a valid mask can be constructed for one of the
+  // register string values permitted for the A and R class cores. These values
+  // are apsr, spsr and cpsr; these are also valid on older cores.
+  int Mask = getARClassRegisterMask(Reg, Flags);
+  if (Mask != -1) {
+    Ops = { CurDAG->getTargetConstant(Mask, DL, MVT::i32), N->getOperand(2),
+            getAL(CurDAG, DL), CurDAG->getRegister(0, MVT::i32),
+            N->getOperand(0) };
+    return CurDAG->getMachineNode(IsThumb2 ? ARM::t2MSR_AR : ARM::MSR,
+                                  DL, MVT::Other, Ops);
+  }
+
+  return nullptr;
+}
+
 SDNode *ARMDAGToDAGISel::SelectInlineAsm(SDNode *N){
   std::vector<SDValue> AsmNodeOperands;
   unsigned Flag, Kind;
diff --git a/lib/Target/ARM/ARMISelLowering.cpp b/lib/Target/ARM/ARMISelLowering.cpp
index 5218234abeb..4b1fd22b4ef 100644
--- a/lib/Target/ARM/ARMISelLowering.cpp
+++ b/lib/Target/ARM/ARMISelLowering.cpp
@@ -426,6 +426,9 @@ ARMTargetLowering::ARMTargetLowering(const TargetMachine &TM,
   setOperationAction(ISD::ConstantFP, MVT::f32, Custom);
   setOperationAction(ISD::ConstantFP, MVT::f64, Custom);
 
+  setOperationAction(ISD::READ_REGISTER, MVT::i64, Custom);
+  setOperationAction(ISD::WRITE_REGISTER, MVT::i64, Custom);
+
   if (Subtarget->hasNEON()) {
     addDRTypeForNEON(MVT::v2f32);
     addDRTypeForNEON(MVT::v8i8);
@@ -2379,6 +2382,24 @@ bool ARMTargetLowering::mayBeEmittedAsTailCall(CallInst *CI) const {
   return !Subtarget->isThumb1Only();
 }
 
+// Trying to write a 64 bit value so need to split into two 32 bit values first,
+// and pass the lower and high parts through.
+static SDValue LowerWRITE_REGISTER(SDValue Op, SelectionDAG &DAG) {
+  SDLoc DL(Op);
+  SDValue WriteValue = Op->getOperand(2);
+
+  // This function is only supposed to be called for i64 type argument.
+  assert(WriteValue.getValueType() == MVT::i64
+          && "LowerWRITE_REGISTER called for non-i64 type argument.");
+
+  SDValue Lo = DAG.getNode(ISD::EXTRACT_ELEMENT, DL, MVT::i32, WriteValue,
+                           DAG.getConstant(0, DL, MVT::i32));
+  SDValue Hi = DAG.getNode(ISD::EXTRACT_ELEMENT, DL, MVT::i32, WriteValue,
+                           DAG.getConstant(1, DL, MVT::i32));
+  SDValue Ops[] = { Op->getOperand(0), Op->getOperand(1), Lo, Hi };
+  return DAG.getNode(ISD::WRITE_REGISTER, DL, MVT::Other, Ops);
+}
+
 // ConstantPool, JumpTable, GlobalAddress, and ExternalSymbol are lowered as
 // their target counterpart wrapped in the ARMISD::Wrapper node. Suppose N is
 // one of the above mentioned nodes. It has to be wrapped because otherwise
@@ -4086,9 +4107,27 @@ unsigned ARMTargetLowering::getRegisterByName(const char* RegName,
                        .Default(0);
   if (Reg)
     return Reg;
-  report_fatal_error("Invalid register name global variable");
+  report_fatal_error(Twine("Invalid register name \""
+                              + StringRef(RegName)  + "\"."));
 }
 
+// Result is 64 bit value so split into two 32 bit values and return as a
+// pair of values.
+static SDValue ExpandREAD_REGISTER(SDNode *N, SelectionDAG &DAG) {
+  SDLoc DL(N);
+  SDValue Op = N->getOperand(0);
+
+  // This function is only supposed to be called for i64 type destination.
+  assert(N->getValueType(0) == MVT::i64
+          && "ExpandREAD_REGISTER called for non-i64 type result.");
+
+  SDValue Read = DAG.getNode(ISD::READ_REGISTER, DL,
+                             DAG.getVTList(MVT::i32, MVT::i32), Op);
+  return DAG.getNode(ISD::BUILD_PAIR, DL, MVT::i64, Read.getValue(0),
+                     Read.getValue(1));
+}
+
+
 /// ExpandBITCAST - If the target supports VFP, this function is called to
 /// expand a bit convert where either the source or destination type is i64 to
 /// use a VMOVDRR or VMOVRRD node.  This should not be done when the non-i64
@@ -6356,6 +6395,7 @@ static void ReplaceREADCYCLECOUNTER(SDNode *N,
 SDValue ARMTargetLowering::LowerOperation(SDValue Op, SelectionDAG &DAG) const {
   switch (Op.getOpcode()) {
   default: llvm_unreachable("Don't know how to custom lower this!");
+  case ISD::WRITE_REGISTER: return LowerWRITE_REGISTER(Op, DAG);
   case ISD::ConstantPool:  return LowerConstantPool(Op, DAG);
   case ISD::BlockAddress:  return LowerBlockAddress(Op, DAG);
   case ISD::GlobalAddress:
@@ -6440,6 +6480,9 @@ void ARMTargetLowering::ReplaceNodeResults(SDNode *N,
   switch (N->getOpcode()) {
   default:
     llvm_unreachable("Don't know how to custom expand this!");
+  case ISD::READ_REGISTER:
+    Res = ExpandREAD_REGISTER(N, DAG);
+    break;
   case ISD::BITCAST:
     Res = ExpandBITCAST(N, DAG);
     break;
@@ -7914,7 +7957,7 @@ static SDValue AddCombineToVPADDL(SDNode *N, SDValue N0, SDValue N1,
   // Get widened type and narrowed type.
   MVT widenType;
   unsigned numElem = VT.getVectorNumElements();
-  
+
   EVT inputLaneType = Vec.getValueType().getVectorElementType();
   switch (inputLaneType.getSimpleVT().SimpleTy) {
     case MVT::i8: widenType = MVT::getVectorVT(MVT::i16, numElem); break;
diff --git a/lib/Target/ARM/ARMInstrInfo.td b/lib/Target/ARM/ARMInstrInfo.td
index 778fd17137f..53263cb2524 100644
--- a/lib/Target/ARM/ARMInstrInfo.td
+++ b/lib/Target/ARM/ARMInstrInfo.td
@@ -5039,10 +5039,8 @@ def : ARMV5TPat<(int_arm_mrc2 imm:$cop, imm:$opc1, imm:$CRn,
                               imm:$CRm, imm:$opc2),
                 (MRC2 imm:$cop, imm:$opc1, imm:$CRn, imm:$CRm, imm:$opc2)>;
 
-class MovRRCopro<string opc, bit direction, list<dag> pattern = []>
-  : ABI<0b1100, (outs), (ins p_imm:$cop, imm0_15:$opc1,
-        GPRnopc:$Rt, GPRnopc:$Rt2, c_imm:$CRm),
-        NoItinerary, opc, "\t$cop, $opc1, $Rt, $Rt2, $CRm", pattern> {
+class MovRRCopro<string opc, bit direction, dag oops, dag iops, list<dag> pattern = []>
+  : ABI<0b1100, oops, iops, NoItinerary, opc, "\t$cop, $opc1, $Rt, $Rt2, $CRm", pattern> {
   let Inst{23-21} = 0b010;
   let Inst{20} = direction;
 
@@ -5060,9 +5058,13 @@ class MovRRCopro<string opc, bit direction, list<dag> pattern = []>
 }
 
 def MCRR : MovRRCopro<"mcrr", 0 /* from ARM core register to coprocessor */,
+                      (outs), (ins p_imm:$cop, imm0_15:$opc1, GPRnopc:$Rt, GPRnopc:$Rt2,
+                      c_imm:$CRm),
                       [(int_arm_mcrr imm:$cop, imm:$opc1, GPRnopc:$Rt,
                                      GPRnopc:$Rt2, imm:$CRm)]>;
-def MRRC : MovRRCopro<"mrrc", 1 /* from coprocessor to ARM core register */>;
+def MRRC : MovRRCopro<"mrrc", 1 /* from coprocessor to ARM core register */,
+                      (outs GPRnopc:$Rt, GPRnopc:$Rt2),
+                      (ins p_imm:$cop, imm0_15:$opc1, c_imm:$CRm), []>;
 
 class MovRRCopro2<string opc, bit direction, list<dag> pattern = []>
   : ABXI<0b1100, (outs), (ins p_imm:$cop, imm0_15:$opc1,
diff --git a/lib/Target/ARM/ARMInstrThumb2.td b/lib/Target/ARM/ARMInstrThumb2.td
index 814b524b2bc..cfe671095e7 100644
--- a/lib/Target/ARM/ARMInstrThumb2.td
+++ b/lib/Target/ARM/ARMInstrThumb2.td
@@ -4141,11 +4141,9 @@ class t2MovRCopro<bits<4> Op, string opc, bit direction, dag oops, dag iops,
   let Inst{19-16} = CRn;
 }
 
-class t2MovRRCopro<bits<4> Op, string opc, bit direction,
+class t2MovRRCopro<bits<4> Op, string opc, bit direction, dag oops, dag iops,
                    list<dag> pattern = []>
-  : T2Cop<Op, (outs),
-          (ins p_imm:$cop, imm0_15:$opc1, GPR:$Rt, GPR:$Rt2, c_imm:$CRm),
-          opc, "\t$cop, $opc1, $Rt, $Rt2, $CRm", pattern> {
+  : T2Cop<Op, oops, iops, opc, "\t$cop, $opc1, $Rt, $Rt2, $CRm", pattern> {
   let Inst{27-24} = 0b1100;
   let Inst{23-21} = 0b010;
   let Inst{20} = direction;
@@ -4210,19 +4208,23 @@ def : T2v6Pat<(int_arm_mrc2 imm:$cop, imm:$opc1, imm:$CRn, imm:$CRm, imm:$opc2),
 
 
 /* from ARM core register to coprocessor */
-def t2MCRR : t2MovRRCopro<0b1110, "mcrr", 0,
-                        [(int_arm_mcrr imm:$cop, imm:$opc1, GPR:$Rt, GPR:$Rt2,
+def t2MCRR : t2MovRRCopro<0b1110, "mcrr", 0, (outs),
+                         (ins p_imm:$cop, imm0_15:$opc1, GPR:$Rt, GPR:$Rt2, c_imm:$CRm),
+                         [(int_arm_mcrr imm:$cop, imm:$opc1, GPR:$Rt, GPR:$Rt2,
                                        imm:$CRm)]>;
-def t2MCRR2 : t2MovRRCopro<0b1111, "mcrr2", 0,
-                           [(int_arm_mcrr2 imm:$cop, imm:$opc1, GPR:$Rt,
+def t2MCRR2 : t2MovRRCopro<0b1111, "mcrr2", 0, (outs),
+                          (ins p_imm:$cop, imm0_15:$opc1, GPR:$Rt, GPR:$Rt2, c_imm:$CRm),
+                          [(int_arm_mcrr2 imm:$cop, imm:$opc1, GPR:$Rt,
                                            GPR:$Rt2, imm:$CRm)]> {
   let Predicates = [IsThumb2, PreV8];
 }
 
 /* from coprocessor to ARM core register */
-def t2MRRC : t2MovRRCopro<0b1110, "mrrc", 1>;
+def t2MRRC : t2MovRRCopro<0b1110, "mrrc", 1, (outs GPR:$Rt, GPR:$Rt2),
+                                             (ins p_imm:$cop, imm0_15:$opc1, c_imm:$CRm)>;
 
-def t2MRRC2 : t2MovRRCopro<0b1111, "mrrc2", 1> {
+def t2MRRC2 : t2MovRRCopro<0b1111, "mrrc2", 1, (outs GPR:$Rt, GPR:$Rt2),
+                                               (ins p_imm:$cop, imm0_15:$opc1, c_imm:$CRm)> {
   let Predicates = [IsThumb2, PreV8];
 }
 
diff --git a/test/CodeGen/AArch64/arm64-named-reg-alloc.ll b/test/CodeGen/AArch64/arm64-named-reg-alloc.ll
index 0c564544a53..5d48c17e128 100644
--- a/test/CodeGen/AArch64/arm64-named-reg-alloc.ll
+++ b/test/CodeGen/AArch64/arm64-named-reg-alloc.ll
@@ -4,7 +4,7 @@
 define i32 @get_stack() nounwind {
 entry:
 ; FIXME: Include an allocatable-specific error message
-; CHECK: Invalid register name global variable
+; CHECK: Invalid register name "x5".
 	%sp = call i32 @llvm.read_register.i32(metadata !0)
   ret i32 %sp
 }
diff --git a/test/CodeGen/AArch64/arm64-named-reg-notareg.ll b/test/CodeGen/AArch64/arm64-named-reg-notareg.ll
index 759bc15807b..8a5fd6f1ac8 100644
--- a/test/CodeGen/AArch64/arm64-named-reg-notareg.ll
+++ b/test/CodeGen/AArch64/arm64-named-reg-notareg.ll
@@ -3,7 +3,7 @@
 
 define i32 @get_stack() nounwind {
 entry:
-; CHECK: Invalid register name global variable
+; CHECK: Invalid register name "notareg".
 	%sp = call i32 @llvm.read_register.i32(metadata !0)
   ret i32 %sp
 }
diff --git a/test/CodeGen/AArch64/special-reg.ll b/test/CodeGen/AArch64/special-reg.ll
new file mode 100644
index 00000000000..91c32158d42
--- /dev/null
+++ b/test/CodeGen/AArch64/special-reg.ll
@@ -0,0 +1,48 @@
+; RUN: llc < %s -mtriple=aarch64-none-eabi -mcpu=cortex-a57 2>&1 | FileCheck %s
+
+define i64 @read_encoded_register() nounwind {
+entry:
+; CHECK-LABEL: read_encoded_register:
+; CHECK: mrs x0, S1_2_C3_C4_5
+  %reg = call i64 @llvm.read_register.i64(metadata !0)
+  ret i64 %reg
+}
+
+define i64 @read_daif() nounwind {
+entry:
+; CHECK-LABEL: read_daif:
+; CHECK: mrs x0, DAIF
+  %reg = call i64 @llvm.read_register.i64(metadata !1)
+  ret i64 %reg
+}
+
+define void @write_encoded_register(i64 %x) nounwind {
+entry:
+; CHECK-LABEL: write_encoded_register:
+; CHECK: msr S1_2_C3_C4_5, x0
+  call void @llvm.write_register.i64(metadata !0, i64 %x)
+  ret void
+}
+
+define void @write_daif(i64 %x) nounwind {
+entry:
+; CHECK-LABEL: write_daif:
+; CHECK: msr DAIF, x0
+  call void @llvm.write_register.i64(metadata !1, i64 %x)
+  ret void
+}
+
+define void @write_daifset() nounwind {
+entry:
+; CHECK-LABEL: write_daifset:
+; CHECK: msr DAIFSET, #2
+  call void @llvm.write_register.i64(metadata !2, i64 2)
+  ret void
+}
+
+declare i64 @llvm.read_register.i64(metadata) nounwind
+declare void @llvm.write_register.i64(metadata, i64) nounwind
+
+!0 = !{!"1:2:3:4:5"}
+!1 = !{!"daif"}
+!2 = !{!"daifset"}
diff --git a/test/CodeGen/ARM/named-reg-alloc.ll b/test/CodeGen/ARM/named-reg-alloc.ll
index 380cf39734f..d41fa64882c 100644
--- a/test/CodeGen/ARM/named-reg-alloc.ll
+++ b/test/CodeGen/ARM/named-reg-alloc.ll
@@ -4,7 +4,7 @@
 define i32 @get_stack() nounwind {
 entry:
 ; FIXME: Include an allocatable-specific error message
-; CHECK: Invalid register name global variable
+; CHECK: Invalid register name "r5".
 	%sp = call i32 @llvm.read_register.i32(metadata !0)
   ret i32 %sp
 }
diff --git a/test/CodeGen/ARM/named-reg-notareg.ll b/test/CodeGen/ARM/named-reg-notareg.ll
index 3ac03f4fdaa..45cb38f30f3 100644
--- a/test/CodeGen/ARM/named-reg-notareg.ll
+++ b/test/CodeGen/ARM/named-reg-notareg.ll
@@ -3,7 +3,7 @@
 
 define i32 @get_stack() nounwind {
 entry:
-; CHECK: Invalid register name global variable
+; CHECK: Invalid register name "notareg".
 	%sp = call i32 @llvm.read_register.i32(metadata !0)
   ret i32 %sp
 }
diff --git a/test/CodeGen/ARM/special-reg-acore.ll b/test/CodeGen/ARM/special-reg-acore.ll
new file mode 100644
index 00000000000..f63d5ddecfc
--- /dev/null
+++ b/test/CodeGen/ARM/special-reg-acore.ll
@@ -0,0 +1,78 @@
+; RUN: llc < %s -mtriple=arm-none-eabi -mcpu=cortex-a8 2>&1 | FileCheck %s --check-prefix=ACORE
+; RUN: not llc < %s -mtriple=thumb-none-eabi -mcpu=cortex-m4 2>&1 | FileCheck %s --check-prefix=MCORE
+
+; MCORE: LLVM ERROR: Invalid register name "cpsr".
+
+define i32 @read_cpsr() nounwind {
+  ; ACORE-LABEL: read_cpsr:
+  ; ACORE: mrs r0, apsr
+	%reg = call i32 @llvm.read_register.i32(metadata !1)
+  ret i32 %reg
+}
+
+define i32 @read_aclass_registers() nounwind {
+entry:
+  ; ACORE-LABEL: read_aclass_registers:
+  ; ACORE: mrs r0, apsr
+  ; ACORE: mrs r1, spsr
+
+  %0 = call i32 @llvm.read_register.i32(metadata !0)
+  %1 = call i32 @llvm.read_register.i32(metadata !1)
+  %add1 = add i32 %1, %0
+  %2 = call i32 @llvm.read_register.i32(metadata !2)
+  %add2 = add i32 %add1, %2
+  ret i32 %add2
+}
+
+define void @write_aclass_registers(i32 %x) nounwind {
+entry:
+  ; ACORE-LABEL: write_aclass_registers:
+  ; ACORE:   msr APSR_nzcvq, r0
+  ; ACORE:   msr APSR_g, r0
+  ; ACORE:   msr APSR_nzcvqg, r0
+  ; ACORE:   msr CPSR_c, r0
+  ; ACORE:   msr CPSR_x, r0
+  ; ACORE:   msr APSR_g, r0
+  ; ACORE:   msr APSR_nzcvq, r0
+  ; ACORE:   msr CPSR_fsxc, r0
+  ; ACORE:   msr SPSR_c, r0
+  ; ACORE:   msr SPSR_x, r0
+  ; ACORE:   msr SPSR_s, r0
+  ; ACORE:   msr SPSR_f, r0
+  ; ACORE:   msr SPSR_fsxc, r0
+
+  call void @llvm.write_register.i32(metadata !3, i32 %x)
+  call void @llvm.write_register.i32(metadata !4, i32 %x)
+  call void @llvm.write_register.i32(metadata !5, i32 %x)
+  call void @llvm.write_register.i32(metadata !6, i32 %x)
+  call void @llvm.write_register.i32(metadata !7, i32 %x)
+  call void @llvm.write_register.i32(metadata !8, i32 %x)
+  call void @llvm.write_register.i32(metadata !9, i32 %x)
+  call void @llvm.write_register.i32(metadata !10, i32 %x)
+  call void @llvm.write_register.i32(metadata !11, i32 %x)
+  call void @llvm.write_register.i32(metadata !12, i32 %x)
+  call void @llvm.write_register.i32(metadata !13, i32 %x)
+  call void @llvm.write_register.i32(metadata !14, i32 %x)
+  call void @llvm.write_register.i32(metadata !15, i32 %x)
+  ret void
+}
+
+declare i32 @llvm.read_register.i32(metadata) nounwind
+declare void @llvm.write_register.i32(metadata, i32) nounwind
+
+!0 = !{!"apsr"}
+!1 = !{!"cpsr"}
+!2 = !{!"spsr"}
+!3 = !{!"apsr_nzcvq"}
+!4 = !{!"apsr_g"}
+!5 = !{!"apsr_nzcvqg"}
+!6 = !{!"cpsr_c"}
+!7 = !{!"cpsr_x"}
+!8 = !{!"cpsr_s"}
+!9 = !{!"cpsr_f"}
+!10 = !{!"cpsr_cxsf"}
+!11 = !{!"spsr_c"}
+!12 = !{!"spsr_x"}
+!13 = !{!"spsr_s"}
+!14 = !{!"spsr_f"}
+!15 = !{!"spsr_cxsf"}
diff --git a/test/CodeGen/ARM/special-reg-mcore.ll b/test/CodeGen/ARM/special-reg-mcore.ll
new file mode 100644
index 00000000000..686da0f6b83
--- /dev/null
+++ b/test/CodeGen/ARM/special-reg-mcore.ll
@@ -0,0 +1,143 @@
+; RUN: llc < %s -mtriple=thumb-none-eabi -mcpu=cortex-m4 2>&1 | FileCheck %s --check-prefix=MCORE
+; RUN: not llc < %s -mtriple=thumb-none-eabi -mcpu=cortex-m3 2>&1 | FileCheck %s --check-prefix=M3CORE
+; RUN: not llc < %s -mtriple=arm-none-eabi -mcpu=cortex-a8 2>&1 | FileCheck %s --check-prefix=ACORE
+
+; ACORE: LLVM ERROR: Invalid register name "control".
+; M3CORE: LLVM ERROR: Invalid register name "control".
+
+define i32 @read_mclass_registers() nounwind {
+entry:
+  ; MCORE-LABEL: read_mclass_registers:
+  ; MCORE:   mrs r0, apsr
+  ; MCORE:   mrs r1, iapsr
+  ; MCORE:   mrs r1, eapsr
+  ; MCORE:   mrs r1, xpsr
+  ; MCORE:   mrs r1, ipsr
+  ; MCORE:   mrs r1, epsr
+  ; MCORE:   mrs r1, iepsr
+  ; MCORE:   mrs r1, msp
+  ; MCORE:   mrs r1, psp
+  ; MCORE:   mrs r1, primask
+  ; MCORE:   mrs r1, basepri
+  ; MCORE:   mrs r1, basepri_max
+  ; MCORE:   mrs r1, faultmask
+  ; MCORE:   mrs r1, control
+
+  %0 = call i32 @llvm.read_register.i32(metadata !0)
+  %1 = call i32 @llvm.read_register.i32(metadata !4)
+  %add1 = add i32 %1, %0
+  %2 = call i32 @llvm.read_register.i32(metadata !8)
+  %add2 = add i32 %add1, %2
+  %3 = call i32 @llvm.read_register.i32(metadata !12)
+  %add3 = add i32 %add2, %3
+  %4 = call i32 @llvm.read_register.i32(metadata !16)
+  %add4 = add i32 %add3, %4
+  %5 = call i32 @llvm.read_register.i32(metadata !17)
+  %add5 = add i32 %add4, %5
+  %6 = call i32 @llvm.read_register.i32(metadata !18)
+  %add6 = add i32 %add5, %6
+  %7 = call i32 @llvm.read_register.i32(metadata !19)
+  %add7 = add i32 %add6, %7
+  %8 = call i32 @llvm.read_register.i32(metadata !20)
+  %add8 = add i32 %add7, %8
+  %9 = call i32 @llvm.read_register.i32(metadata !21)
+  %add9 = add i32 %add8, %9
+  %10 = call i32 @llvm.read_register.i32(metadata !22)
+  %add10 = add i32 %add9, %10
+  %11 = call i32 @llvm.read_register.i32(metadata !23)
+  %add11 = add i32 %add10, %11
+  %12 = call i32 @llvm.read_register.i32(metadata !24)
+  %add12 = add i32 %add11, %12
+  %13 = call i32 @llvm.read_register.i32(metadata !25)
+  %add13 = add i32 %add12, %13
+  ret i32 %add13
+}
+
+define void @write_mclass_registers(i32 %x) nounwind {
+entry:
+  ; MCORE-LABEL: write_mclass_registers:
+  ; MCORE:   msr apsr_nzcvqg, r0
+  ; MCORE:   msr apsr_nzcvq, r0
+  ; MCORE:   msr apsr_g, r0
+  ; MCORE:   msr apsr_nzcvqg, r0
+  ; MCORE:   msr iapsr_nzcvqg, r0
+  ; MCORE:   msr iapsr_nzcvq, r0
+  ; MCORE:   msr iapsr_g, r0
+  ; MCORE:   msr iapsr_nzcvqg, r0
+  ; MCORE:   msr eapsr_nzcvqg, r0
+  ; MCORE:   msr eapsr_nzcvq, r0
+  ; MCORE:   msr eapsr_g, r0
+  ; MCORE:   msr eapsr_nzcvqg, r0
+  ; MCORE:   msr xpsr_nzcvqg, r0
+  ; MCORE:   msr xpsr_nzcvq, r0
+  ; MCORE:   msr xpsr_g, r0
+  ; MCORE:   msr xpsr_nzcvqg, r0
+  ; MCORE:   msr ipsr, r0
+  ; MCORE:   msr epsr, r0
+  ; MCORE:   msr iepsr, r0
+  ; MCORE:   msr msp, r0
+  ; MCORE:   msr psp, r0
+  ; MCORE:   msr primask, r0
+  ; MCORE:   msr basepri, r0
+  ; MCORE:   msr basepri_max, r0
+  ; MCORE:   msr faultmask, r0
+  ; MCORE:   msr control, r0
+
+  call void @llvm.write_register.i32(metadata !0, i32 %x)
+  call void @llvm.write_register.i32(metadata !1, i32 %x)
+  call void @llvm.write_register.i32(metadata !2, i32 %x)
+  call void @llvm.write_register.i32(metadata !3, i32 %x)
+  call void @llvm.write_register.i32(metadata !4, i32 %x)
+  call void @llvm.write_register.i32(metadata !5, i32 %x)
+  call void @llvm.write_register.i32(metadata !6, i32 %x)
+  call void @llvm.write_register.i32(metadata !7, i32 %x)
+  call void @llvm.write_register.i32(metadata !8, i32 %x)
+  call void @llvm.write_register.i32(metadata !9, i32 %x)
+  call void @llvm.write_register.i32(metadata !10, i32 %x)
+  call void @llvm.write_register.i32(metadata !11, i32 %x)
+  call void @llvm.write_register.i32(metadata !12, i32 %x)
+  call void @llvm.write_register.i32(metadata !13, i32 %x)
+  call void @llvm.write_register.i32(metadata !14, i32 %x)
+  call void @llvm.write_register.i32(metadata !15, i32 %x)
+  call void @llvm.write_register.i32(metadata !16, i32 %x)
+  call void @llvm.write_register.i32(metadata !17, i32 %x)
+  call void @llvm.write_register.i32(metadata !18, i32 %x)
+  call void @llvm.write_register.i32(metadata !19, i32 %x)
+  call void @llvm.write_register.i32(metadata !20, i32 %x)
+  call void @llvm.write_register.i32(metadata !21, i32 %x)
+  call void @llvm.write_register.i32(metadata !22, i32 %x)
+  call void @llvm.write_register.i32(metadata !23, i32 %x)
+  call void @llvm.write_register.i32(metadata !24, i32 %x)
+  call void @llvm.write_register.i32(metadata !25, i32 %x)
+  ret void
+}
+
+declare i32 @llvm.read_register.i32(metadata) nounwind
+declare void @llvm.write_register.i32(metadata, i32) nounwind
+
+!0 = !{!"apsr"}
+!1 = !{!"apsr_nzcvq"}
+!2 = !{!"apsr_g"}
+!3 = !{!"apsr_nzcvqg"}
+!4 = !{!"iapsr"}
+!5 = !{!"iapsr_nzcvq"}
+!6 = !{!"iapsr_g"}
+!7 = !{!"iapsr_nzcvqg"}
+!8 = !{!"eapsr"}
+!9 = !{!"eapsr_nzcvq"}
+!10 = !{!"eapsr_g"}
+!11 = !{!"eapsr_nzcvqg"}
+!12 = !{!"xpsr"}
+!13 = !{!"xpsr_nzcvq"}
+!14 = !{!"xpsr_g"}
+!15 = !{!"xpsr_nzcvqg"}
+!16 = !{!"ipsr"}
+!17 = !{!"epsr"}
+!18 = !{!"iepsr"}
+!19 = !{!"msp"}
+!20 = !{!"psp"}
+!21 = !{!"primask"}
+!22 = !{!"basepri"}
+!23 = !{!"basepri_max"}
+!24 = !{!"faultmask"}
+!25 = !{!"control"}
diff --git a/test/CodeGen/ARM/special-reg.ll b/test/CodeGen/ARM/special-reg.ll
new file mode 100644
index 00000000000..8b8243ec712
--- /dev/null
+++ b/test/CodeGen/ARM/special-reg.ll
@@ -0,0 +1,78 @@
+; RUN: llc < %s -mtriple=arm-none-eabi -mcpu=cortex-a8 2>&1 | FileCheck %s --check-prefix=ARM --check-prefix=ACORE
+; RUN: llc < %s -mtriple=thumb-none-eabi -mcpu=cortex-m4 2>&1 | FileCheck %s --check-prefix=ARM --check-prefix=MCORE
+
+define i32 @read_i32_encoded_register() nounwind {
+entry:
+; ARM-LABEL: read_i32_encoded_register:
+; ARM: mrc p1, #2, r0, c3, c4, #5
+	%reg = call i32 @llvm.read_register.i32(metadata !0)
+  ret i32 %reg
+}
+
+define i64 @read_i64_encoded_register() nounwind {
+entry:
+; ARM-LABEL: read_i64_encoded_register:
+; ARM: mrrc p1, #2, r0, r1, c3
+	%reg = call i64 @llvm.read_register.i64(metadata !1)
+  ret i64 %reg
+}
+
+define i32 @read_apsr() nounwind {
+entry:
+; ARM-LABEL: read_apsr:
+; ARM: mrs r0, apsr
+	%reg = call i32 @llvm.read_register.i32(metadata !2)
+  ret i32 %reg
+}
+
+define i32 @read_fpscr() nounwind {
+entry:
+; ARM-LABEL: read_fpscr:
+; ARM: vmrs r0, fpscr
+	%reg = call i32 @llvm.read_register.i32(metadata !3)
+  ret i32 %reg
+}
+
+define void @write_i32_encoded_register(i32 %x) nounwind {
+entry:
+; ARM-LABEL: write_i32_encoded_register:
+; ARM: mcr p1, #2, r0, c3, c4, #5
+	call void @llvm.write_register.i32(metadata !0, i32 %x)
+  ret void
+}
+
+define void @write_i64_encoded_register(i64 %x) nounwind {
+entry:
+; ARM-LABEL: write_i64_encoded_register:
+; ARM: mcrr p1, #2, r0, r1, c3
+	call void @llvm.write_register.i64(metadata !1, i64 %x)
+  ret void
+}
+
+define void @write_apsr(i32 %x) nounwind {
+entry:
+; ARM-LABEL: write_apsr:
+; ACORE: msr APSR_nzcvq, r0
+; MCORE: msr apsr_nzcvq, r0
+	call void @llvm.write_register.i32(metadata !4, i32 %x)
+  ret void
+}
+
+define void @write_fpscr(i32 %x) nounwind {
+entry:
+; ARM-LABEL: write_fpscr:
+; ARM: vmsr fpscr, r0
+	call void @llvm.write_register.i32(metadata !3, i32 %x)
+  ret void
+}
+
+declare i32 @llvm.read_register.i32(metadata) nounwind
+declare i64 @llvm.read_register.i64(metadata) nounwind
+declare void @llvm.write_register.i32(metadata, i32) nounwind
+declare void @llvm.write_register.i64(metadata, i64) nounwind
+
+!0 = !{!"cp1:2:c3:c4:5"}
+!1 = !{!"cp1:2:c3"}
+!2 = !{!"apsr"}
+!3 = !{!"fpscr"}
+!4 = !{!"apsr_nzcvq"}