[RISCV] Attach VK_RISCV_CALL to symbols upon creation

[lewis-revill]

Summary

This patch replaces the addition of VK_RISCV_CALL in RISCVMCCodeEmitter by creating the RISCVMCExpr when tail/call are parsed, or in the codegen case when the callee symbols are created.

This required adding a new CallSymbol operand to allow only adding VK_RISCV_CALL to tail/call instructions.

This patch will allow further expansion of parsing and codegen to easily include PLT symbols which must generate the R_RISCV_CALL_PLT relocation.

[jrtc27]

lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp

288

I don't think we want to allow VK_RISCV_None, as expandFunctionCall won't add a VK_RISCV_CALL wrapper so we won't get an R_RISCV_CALL?

[lewis-revill]

lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp

288

Good catch, I'll fix that.

[lewis-revill]

Do not allow VK_RISCV_None on call symbols

[lewis-revill]

Rebased.

[asb]

I know this isn't a behaviour change vs the previous bare_symbol, but I note that GNU as currently accepts call with constants, e.g. call 1234. Should we be doing the same?

[jrtc27]

(Added dependency due to both adding to RISCVBaseInfo.h)

[jrtc27]

Hi Lewis, please rebase this after rL351823 (the Call node is now a riscv_call node in PseudoCll's pattern.

[lewis-revill]

Rebased

[asb]

Thanks Lewis, this looks good to me. We should later revisit whether we should accept call 1234 like GNU as does.

lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp

#include "RISCVGenAsmMatcher.inc"
 
   OperandMatchResultTy parseCSRSystemRegister(OperandVector &Operands);
   OperandMatchResultTy parseImmediate(OperandVector &Operands);
   OperandMatchResultTy parseRegister(OperandVector &Operands,
                                      bool AllowParens = false);
   OperandMatchResultTy parseMemOpBaseReg(OperandVector &Operands);
   OperandMatchResultTy parseOperandWithModifier(OperandVector &Operands);
   OperandMatchResultTy parseBareSymbol(OperandVector &Operands);
+  OperandMatchResultTy parseCallSymbol(OperandVector &Operands);
   OperandMatchResultTy parseJALOffset(OperandVector &Operands);
 
   bool parseOperand(OperandVector &Operands, StringRef Mnemonic);
 
   bool parseDirectiveOption();
 
   void setFeatureBits(uint64_t Feature, StringRef FeatureString) {
     if (!(getSTI().getFeatureBits()[Feature])) {
bool isBareSymbol() const {
     RISCVMCExpr::VariantKind VK;
     // Must be of 'immediate' type but not a constant.
     if (!isImm() || evaluateConstantImm(getImm(), Imm, VK))
       return false;
     return RISCVAsmParser::classifySymbolRef(getImm(), VK, Imm) &&
            VK == RISCVMCExpr::VK_RISCV_None;
   }
 
+  bool isCallSymbol() const {
+    int64_t Imm;
+    RISCVMCExpr::VariantKind VK;
+    // Must be of 'immediate' type but not a constant.
+    if (!isImm() || evaluateConstantImm(getImm(), Imm, VK))
+      return false;
+    return RISCVAsmParser::classifySymbolRef(getImm(), VK, Imm) &&
+           VK == RISCVMCExpr::VK_RISCV_CALL;
+  }
+
   bool isCSRSystemRegister() const { return isSystemRegister(); }
 
   /// Return true if the operand is a valid for the fence instruction e.g.
   /// ('iorw').
   bool isFenceArg() const {
     if (!isImm())
       return false;
     const MCExpr *Val = getImm();
case Match_InvalidFRMArg: {
     return Error(
         ErrorLoc,
         "operand must be a valid floating point rounding mode mnemonic");
   }
   case Match_InvalidBareSymbol: {
     SMLoc ErrorLoc = ((RISCVOperand &)*Operands[ErrorInfo]).getStartLoc();
     return Error(ErrorLoc, "operand must be a bare symbol name");
   }
+  case Match_InvalidCallSymbol: {
+    SMLoc ErrorLoc = ((RISCVOperand &)*Operands[ErrorInfo]).getStartLoc();
+    return Error(ErrorLoc, "operand must be a bare symbol name");
+  }
   }
 
   llvm_unreachable("Unknown match type detected!");
 }
 
 bool RISCVAsmParser::ParseRegister(unsigned &RegNo, SMLoc &StartLoc,
                                    SMLoc &EndLoc) {
   const AsmToken &Tok = getParser().getTok();
if (getParser().parseIdentifier(Identifier))
     return MatchOperand_ParseFail;
 
   MCSymbol *Sym = getContext().getOrCreateSymbol(Identifier);
   Res = MCSymbolRefExpr::create(Sym, MCSymbolRefExpr::VK_None, getContext());
   Operands.push_back(RISCVOperand::createImm(Res, S, E, isRV64()));
   return MatchOperand_Success;
 }
 
+OperandMatchResultTy RISCVAsmParser::parseCallSymbol(OperandVector &Operands) {
+  SMLoc S = getLoc();
+  SMLoc E = SMLoc::getFromPointer(S.getPointer() - 1);
+  const MCExpr *Res;
+
+  if (getLexer().getKind() != AsmToken::Identifier)
+    return MatchOperand_NoMatch;
+
+  StringRef Identifier;
+  if (getParser().parseIdentifier(Identifier))
+    return MatchOperand_ParseFail;
+
+  MCSymbol *Sym = getContext().getOrCreateSymbol(Identifier);
+  Res = MCSymbolRefExpr::create(Sym, MCSymbolRefExpr::VK_None, getContext());
+  Res = RISCVMCExpr::create(Res, RISCVMCExpr::VK_RISCV_CALL, getContext());
+  Operands.push_back(RISCVOperand::createImm(Res, S, E, isRV64()));
+  return MatchOperand_Success;
+}
+
 OperandMatchResultTy RISCVAsmParser::parseJALOffset(OperandVector &Operands) {
   // Parsing jal operands is fiddly due to the `jal foo` and `jal ra, foo`
   // both being acceptable forms. When parsing `jal ra, foo` this function
   // will be called for the `ra` register operand in an attempt to match the
   // single-operand alias. parseJALOffset must fail for this case. It would
   // seem logical to try parse the operand using parseImmediate and return
   // NoMatch if the next token is a comma (meaning we must be parsing a jal in
   // the second form rather than the first). We can't do this as there's no

lib/Target/RISCV/MCTargetDesc/RISCVMCCodeEmitter.cpp

void RISCVMCCodeEmitter::expandFunctionCall(const MCInst &MI, raw_ostream &OS,
                                             const MCSubtargetInfo &STI) const {
   MCInst TmpInst;
   MCOperand Func = MI.getOperand(0);
   unsigned Ra = (MI.getOpcode() == RISCV::PseudoTAIL) ? RISCV::X6 : RISCV::X1;
   uint32_t Binary;
 
   assert(Func.isExpr() && "Expected expression");
 
-  const MCExpr *Expr = Func.getExpr();
-
-  // Create function call expression CallExpr for AUIPC.
-  const MCExpr *CallExpr =
-      RISCVMCExpr::create(Expr, RISCVMCExpr::VK_RISCV_CALL, Ctx);
+  const MCExpr *CallExpr = Func.getExpr();
 
   // Emit AUIPC Ra, Func with R_RISCV_CALL relocation type.
   TmpInst = MCInstBuilder(RISCV::AUIPC)
                 .addReg(Ra)
                 .addOperand(MCOperand::createExpr(CallExpr));
   Binary = getBinaryCodeForInstr(TmpInst, Fixups, STI);
   support::endian::write(OS, Binary, support::little);
 

lib/Target/RISCV/RISCVISelLowering.cpp

for (auto &Reg : RegsToPass) {
     Chain = DAG.getCopyToReg(Chain, DL, Reg.first, Reg.second, Glue);
     Glue = Chain.getValue(1);
   }
 
   // If the callee is a GlobalAddress/ExternalSymbol node, turn it into a
   // TargetGlobalAddress/TargetExternalSymbol node so that legalize won't
   // split it and then direct call can be matched by PseudoCALL.
   if (GlobalAddressSDNode *S = dyn_cast<GlobalAddressSDNode>(Callee)) {
-    Callee = DAG.getTargetGlobalAddress(S->getGlobal(), DL, PtrVT, 0, 0);
+    Callee = DAG.getTargetGlobalAddress(S->getGlobal(), DL, PtrVT, 0,
+                                        RISCVII::MO_CALL);
   } else if (ExternalSymbolSDNode *S = dyn_cast<ExternalSymbolSDNode>(Callee)) {
-    Callee = DAG.getTargetExternalSymbol(S->getSymbol(), PtrVT, 0);
+    Callee =
+        DAG.getTargetExternalSymbol(S->getSymbol(), PtrVT, RISCVII::MO_CALL);
   }
 
   // The first call operand is the chain and the second is the target address.
   SmallVector<SDValue, 8> Ops;
   Ops.push_back(Chain);
   Ops.push_back(Callee);
 
   // Add argument registers to the end of the list so that they are

lib/Target/RISCV/RISCVInstrInfo.td

def BareSymbol : AsmOperandClass {
   let ParserMethod = "parseBareSymbol";
 }
 
 // A bare symbol.
 def bare_symbol : Operand<XLenVT> {
   let ParserMatchClass = BareSymbol;
 }
 
+def CallSymbol : AsmOperandClass {
+  let Name = "CallSymbol";
+  let RenderMethod = "addImmOperands";
+  let DiagnosticType = "InvalidCallSymbol";
+  let ParserMethod = "parseCallSymbol";
+}
+
+// A bare symbol used in call/tail only.
+def call_symbol : Operand<XLenVT> {
+  let ParserMatchClass = CallSymbol;
+}
+
 def CSRSystemRegister : AsmOperandClass {
   let Name = "CSRSystemRegister";
   let ParserMethod = "parseCSRSystemRegister";
   let DiagnosticType = "InvalidCSRSystemRegister";
 }
 
 def csr_sysreg : Operand<XLenVT> {
   let ParserMatchClass = CSRSystemRegister;

 
 // PseudoCALL is a pseudo instruction which will eventually expand to auipc
 // and jalr while encoding. This is desirable, as an auipc+jalr pair with
 // R_RISCV_CALL and R_RISCV_RELAX relocations can be be relaxed by the linker
 // if the offset fits in a signed 21-bit immediate.
 // Define AsmString to print "call" when compile with -S flag.
 // Define isCodeGenOnly = 0 to support parsing assembly "call" instruction.
 let isCall = 1, Defs = [X1], isCodeGenOnly = 0 in
-def PseudoCALL : Pseudo<(outs), (ins bare_symbol:$func),
+def PseudoCALL : Pseudo<(outs), (ins call_symbol:$func),
                         [(Call tglobaladdr:$func)]> {
   let AsmString = "call\t$func";
 }
 
 def : Pat<(Call texternalsym:$func), (PseudoCALL texternalsym:$func)>;
 
 def : Pat<(URetFlag), (URET X0, X0)>;
 def : Pat<(SRetFlag), (SRET X0, X0)>;
 def : Pat<(MRetFlag), (MRET X0, X0)>;
 
 let isCall = 1, Defs = [X1] in
 def PseudoCALLIndirect : Pseudo<(outs), (ins GPR:$rs1), [(Call GPR:$rs1)]>,
                          PseudoInstExpansion<(JALR X1, GPR:$rs1, 0)>;
 
 let isBarrier = 1, isReturn = 1, isTerminator = 1 in
 def PseudoRET : Pseudo<(outs), (ins), [(RetFlag)]>,
                 PseudoInstExpansion<(JALR X0, X1, 0)>;
 
 // PseudoTAIL is a pseudo instruction similar to PseudoCALL and will eventually
 // expand to auipc and jalr while encoding.
 // Define AsmString to print "tail" when compile with -S flag.
 let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1, Uses = [X2],
     isCodeGenOnly = 0 in
-def PseudoTAIL : Pseudo<(outs), (ins bare_symbol:$dst), []> {
+def PseudoTAIL : Pseudo<(outs), (ins call_symbol:$dst), []> {
   let AsmString = "tail\t$dst";
 }
 
 let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1, Uses = [X2] in
 def PseudoTAILIndirect : Pseudo<(outs), (ins GPRTC:$rs1), [(Tail GPRTC:$rs1)]>,
                          PseudoInstExpansion<(JALR X0, GPR:$rs1, 0)>;
 
 def : Pat<(Tail (iPTR tglobaladdr:$dst)),

lib/Target/RISCV/RISCVMCInstLower.cpp

static MCOperand lowerSymbolOperand(const MachineOperand &MO, MCSymbol *Sym,
   RISCVMCExpr::VariantKind Kind;
 
   switch (MO.getTargetFlags()) {
   default:
     llvm_unreachable("Unknown target flag on GV operand");
   case RISCVII::MO_None:
     Kind = RISCVMCExpr::VK_RISCV_None;
     break;
+  case RISCVII::MO_CALL:
+    Kind = RISCVMCExpr::VK_RISCV_CALL;
+    break;
   case RISCVII::MO_LO:
     Kind = RISCVMCExpr::VK_RISCV_LO;
     break;
   case RISCVII::MO_HI:
     Kind = RISCVMCExpr::VK_RISCV_HI;
     break;
   case RISCVII::MO_PCREL_LO:
     Kind = RISCVMCExpr::VK_RISCV_PCREL_LO;

lib/Target/RISCV/Utils/RISCVBaseInfo.h

enum {
   InstFormatCJ = 16,
   InstFormatOther = 17,
 
   InstFormatMask = 31
 };
 
 enum {
   MO_None,
+  MO_CALL,
   MO_LO,
   MO_HI,
   MO_PCREL_LO,
   MO_PCREL_HI,
   MO_GOT_HI,
 };
 } // namespace RISCVII