[VE] Support symbol with offset in assembly

[kaz7]

Summary

Change MCExpr to support Aurora VE's modifiers. Change asmparser to use
existing MCExpr parser (parseExpression) to parse an expression contining
symbols with modifiers and offsets. Also add several regression tests
of MC layer.

[simoll]

Please consider the coding style suggestions (in particular for lower-casing function names). Otw, LGTM.

[kaz7]

Change function names following suggestions.

[simoll]

Thanks!

llvm/include/llvm/MC/MCExpr.h

enum VariantKind : uint16_t {
     VK_AMDGPU_GOTPCREL32_LO, // symbol@gotpcrel32@lo
     VK_AMDGPU_GOTPCREL32_HI, // symbol@gotpcrel32@hi
     VK_AMDGPU_REL32_LO,      // symbol@rel32@lo
     VK_AMDGPU_REL32_HI,      // symbol@rel32@hi
     VK_AMDGPU_REL64,         // symbol@rel64
     VK_AMDGPU_ABS32_LO,      // symbol@abs32@lo
     VK_AMDGPU_ABS32_HI,      // symbol@abs32@hi
 
+    VK_VE_HI32,        // symbol@hi
+    VK_VE_LO32,        // symbol@lo
+    VK_VE_PC_HI32,     // symbol@pc_hi
+    VK_VE_PC_LO32,     // symbol@pc_lo
+    VK_VE_GOT_HI32,    // symbol@got_hi
+    VK_VE_GOT_LO32,    // symbol@got_lo
+    VK_VE_GOTOFF_HI32, // symbol@gotoff_hi
+    VK_VE_GOTOFF_LO32, // symbol@gotoff_lo
+    VK_VE_PLT_HI32,    // symbol@plt_hi
+    VK_VE_PLT_LO32,    // symbol@plt_lo
+    VK_VE_TLS_GD_HI32, // symbol@tls_gd_hi
+    VK_VE_TLS_GD_LO32, // symbol@tls_gd_lo
+    VK_VE_TPOFF_HI32,  // symbol@tpoff_hi
+    VK_VE_TPOFF_LO32,  // symbol@tpoff_lo
+
     VK_TPREL,
     VK_DTPREL
   };
 
 private:
   /// The symbol being referenced.
   const MCSymbol *Symbol;
 

llvm/lib/MC/MCExpr.cpp

StringRef MCSymbolRefExpr::getVariantKindName(VariantKind Kind) {
   case VK_WASM_TBREL: return "TBREL";
   case VK_AMDGPU_GOTPCREL32_LO: return "gotpcrel32@lo";
   case VK_AMDGPU_GOTPCREL32_HI: return "gotpcrel32@hi";
   case VK_AMDGPU_REL32_LO: return "rel32@lo";
   case VK_AMDGPU_REL32_HI: return "rel32@hi";
   case VK_AMDGPU_REL64: return "rel64";
   case VK_AMDGPU_ABS32_LO: return "abs32@lo";
   case VK_AMDGPU_ABS32_HI: return "abs32@hi";
+  case VK_VE_HI32: return "hi";
+  case VK_VE_LO32: return "lo";
+  case VK_VE_PC_HI32: return "pc_hi";
+  case VK_VE_PC_LO32: return "pc_lo";
+  case VK_VE_GOT_HI32: return "got_hi";
+  case VK_VE_GOT_LO32: return "got_lo";
+  case VK_VE_GOTOFF_HI32: return "gotoff_hi";
+  case VK_VE_GOTOFF_LO32: return "gotoff_lo";
+  case VK_VE_PLT_HI32: return "plt_hi";
+  case VK_VE_PLT_LO32: return "plt_lo";
+  case VK_VE_TLS_GD_HI32: return "tls_gd_hi";
+  case VK_VE_TLS_GD_LO32: return "tls_gd_lo";
+  case VK_VE_TPOFF_HI32: return "tpoff_hi";
+  case VK_VE_TPOFF_LO32: return "tpoff_lo";
   }
   llvm_unreachable("Invalid variant kind");
 }
 
 MCSymbolRefExpr::VariantKind
 MCSymbolRefExpr::getVariantKindForName(StringRef Name) {
   return StringSwitch<VariantKind>(Name.lower())
     .Case("dtprel", VK_DTPREL)
     .Case("dtpoff", VK_DTPOFF)
     .Case("got", VK_GOT)
     .Case("gotoff", VK_GOTOFF)
     .Case("gotrel", VK_GOTREL)
     .Case("pcrel", VK_PCREL)
     .Case("gotpcrel", VK_GOTPCREL)
     .Case("gottpoff", VK_GOTTPOFF)
     .Case("indntpoff", VK_INDNTPOFF)
return StringSwitch<VariantKind>(Name.lower())
     .Case("mbrel", VK_WASM_MBREL)
     .Case("gotpcrel32@lo", VK_AMDGPU_GOTPCREL32_LO)
     .Case("gotpcrel32@hi", VK_AMDGPU_GOTPCREL32_HI)
     .Case("rel32@lo", VK_AMDGPU_REL32_LO)
     .Case("rel32@hi", VK_AMDGPU_REL32_HI)
     .Case("rel64", VK_AMDGPU_REL64)
     .Case("abs32@lo", VK_AMDGPU_ABS32_LO)
     .Case("abs32@hi", VK_AMDGPU_ABS32_HI)
+    .Case("hi", VK_VE_HI32)
+    .Case("lo", VK_VE_LO32)
+    .Case("pc_hi", VK_VE_PC_HI32)
+    .Case("pc_lo", VK_VE_PC_LO32)
+    .Case("got_hi", VK_VE_GOT_HI32)
+    .Case("got_lo", VK_VE_GOT_LO32)
+    .Case("gotoff_hi", VK_VE_GOTOFF_HI32)
+    .Case("gotoff_lo", VK_VE_GOTOFF_LO32)
+    .Case("plt_hi", VK_VE_PLT_HI32)
+    .Case("plt_lo", VK_VE_PLT_LO32)
+    .Case("tls_gd_hi", VK_VE_TLS_GD_HI32)
+    .Case("tls_gd_lo", VK_VE_TLS_GD_LO32)
+    .Case("tpoff_hi", VK_VE_TPOFF_HI32)
+    .Case("tpoff_lo", VK_VE_TPOFF_LO32)
     .Default(VK_Invalid);
 }
 
 void MCSymbolRefExpr::printVariantKind(raw_ostream &OS) const {
   if (useParensForSymbolVariant())
     OS << '(' << MCSymbolRefExpr::getVariantKindName(getKind()) << ')';
   else
     OS << '@' << MCSymbolRefExpr::getVariantKindName(getKind());

llvm/lib/Target/VE/AsmParser/VEAsmParser.cpp

 //===-- VEAsmParser.cpp - Parse VE assembly to MCInst instructions --------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #include "MCTargetDesc/VEMCExpr.h"
 #include "MCTargetDesc/VEMCTargetDesc.h"
 #include "TargetInfo/VETargetInfo.h"
 #include "VE.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/Twine.h"
 #include "llvm/MC/MCContext.h"
OperandMatchResultTy tryParseRegister(unsigned &RegNo, SMLoc &StartLoc,
                                         SMLoc &EndLoc) override;
   bool ParseInstruction(ParseInstructionInfo &Info, StringRef Name,
                         SMLoc NameLoc, OperandVector &Operands) override;
   bool ParseDirective(AsmToken DirectiveID) override;
 
   unsigned validateTargetOperandClass(MCParsedAsmOperand &Op,
                                       unsigned Kind) override;
 
-  // Helper function to parse and generate identifier with relocation.
-  const MCExpr *parseIdentifier(StringRef Identifier);
   // Custom parse functions for VE specific operands.
   OperandMatchResultTy parseMEMOperand(OperandVector &Operands);
   OperandMatchResultTy parseMEMAsOperand(OperandVector &Operands);
   OperandMatchResultTy parseCCOpOperand(OperandVector &Operands);
   OperandMatchResultTy parseRDOpOperand(OperandVector &Operands);
   OperandMatchResultTy parseMImmOperand(OperandVector &Operands);
   OperandMatchResultTy parseOperand(OperandVector &Operands, StringRef Name);
   OperandMatchResultTy parseVEAsmOperand(std::unique_ptr<VEOperand> &Operand);
+
+  // Helper function to parse expression with a symbol.
+  const MCExpr *ExtractModifierFromExpr(const MCExpr *E,
+                                        VEMCExpr::VariantKind &Variant);
+  const MCExpr *FixupVariantKind(const MCExpr *E);
+  bool ParseExpression(const MCExpr *&EVal);
+
   // Split the mnemonic stripping conditional code and quantifiers
   StringRef splitMnemonic(StringRef Name, SMLoc NameLoc,
                           OperandVector *Operands);
 
 public:
   VEAsmParser(const MCSubtargetInfo &sti, MCAsmParser &parser,
               const MCInstrInfo &MII, const MCTargetOptions &Options)
       : MCTargetAsmParser(Options, sti, MII), Parser(parser) {
bool VEAsmParser::ParseInstruction(ParseInstructionInfo &Info, StringRef Name,
   return false;
 }
 
 bool VEAsmParser::ParseDirective(AsmToken DirectiveID) {
   // Let the MC layer to handle other directives.
   return true;
 }
 
-const MCExpr *VEAsmParser::parseIdentifier(StringRef Identifier) {
-  StringRef Modifier;
-  // Search @modifiers like "symbol@hi".
-  size_t at = Identifier.rfind('@');
-  if (at != 0 || at != StringRef::npos) {
-    std::pair<StringRef, StringRef> Pair = Identifier.rsplit("@");
-    if (!Pair.first.empty() && !Pair.second.empty()) {
-      Identifier = Pair.first;
-      Modifier = Pair.second;
-    }
-  }
-  MCSymbol *Sym = getContext().getOrCreateSymbol(Identifier);
-  const MCExpr *Res =
-      MCSymbolRefExpr::create(Sym, MCSymbolRefExpr::VK_None, getContext());
-  VEMCExpr::VariantKind VK = VEMCExpr::parseVariantKind(Modifier);
-  if (VK == VEMCExpr::VK_VE_None) {
-    // Create identifier using default variant kind
-    VEMCExpr::VariantKind Kind = VEMCExpr::VK_VE_REFLONG;
-    return VEMCExpr::create(Kind, Res, getContext());
+/// Extract \code @lo32/@hi32/etc \endcode modifier from expression.
+/// Recursively scan the expression and check for VK_VE_HI32/LO32/etc
+/// symbol variants.  If all symbols with modifier use the same
+/// variant, return the corresponding VEMCExpr::VariantKind,
+/// and a modified expression using the default symbol variant.
+/// Otherwise, return NULL.
+const MCExpr *
+VEAsmParser::ExtractModifierFromExpr(const MCExpr *E,
+                                     VEMCExpr::VariantKind &Variant) {
+  MCContext &Context = getParser().getContext();
+  Variant = VEMCExpr::VK_VE_None;
+
+  switch (E->getKind()) {
+  case MCExpr::Target:
+  case MCExpr::Constant:
+    return nullptr;
+
+  case MCExpr::SymbolRef: {
+    const MCSymbolRefExpr *SRE = cast<MCSymbolRefExpr>(E);
+
+    switch (SRE->getKind()) {
+    case MCSymbolRefExpr::VK_None:
+      // Use VK_VE_REFLONG to a symbol without modifiers.
+      Variant = VEMCExpr::VK_VE_REFLONG;
+      break;
+    case MCSymbolRefExpr::VK_VE_HI32:
+      Variant = VEMCExpr::VK_VE_HI32;
+      break;
+    case MCSymbolRefExpr::VK_VE_LO32:
+      Variant = VEMCExpr::VK_VE_LO32;
+      break;
+    case MCSymbolRefExpr::VK_VE_PC_HI32:
+      Variant = VEMCExpr::VK_VE_PC_HI32;
+      break;
+    case MCSymbolRefExpr::VK_VE_PC_LO32:
+      Variant = VEMCExpr::VK_VE_PC_LO32;
+      break;
+    case MCSymbolRefExpr::VK_VE_GOT_HI32:
+      Variant = VEMCExpr::VK_VE_GOT_HI32;
+      break;
+    case MCSymbolRefExpr::VK_VE_GOT_LO32:
+      Variant = VEMCExpr::VK_VE_GOT_LO32;
+      break;
+    case MCSymbolRefExpr::VK_VE_GOTOFF_HI32:
+      Variant = VEMCExpr::VK_VE_GOTOFF_HI32;
+      break;
+    case MCSymbolRefExpr::VK_VE_GOTOFF_LO32:
+      Variant = VEMCExpr::VK_VE_GOTOFF_LO32;
+      break;
+    case MCSymbolRefExpr::VK_VE_PLT_HI32:
+      Variant = VEMCExpr::VK_VE_PLT_HI32;
+      break;
+    case MCSymbolRefExpr::VK_VE_PLT_LO32:
+      Variant = VEMCExpr::VK_VE_PLT_LO32;
+      break;
+    case MCSymbolRefExpr::VK_VE_TLS_GD_HI32:
+      Variant = VEMCExpr::VK_VE_TLS_GD_HI32;
+      break;
+    case MCSymbolRefExpr::VK_VE_TLS_GD_LO32:
+      Variant = VEMCExpr::VK_VE_TLS_GD_LO32;
+      break;
+    case MCSymbolRefExpr::VK_VE_TPOFF_HI32:
+      Variant = VEMCExpr::VK_VE_TPOFF_HI32;
+      break;
+    case MCSymbolRefExpr::VK_VE_TPOFF_LO32:
+      Variant = VEMCExpr::VK_VE_TPOFF_LO32;
+      break;
+    default:
+      return nullptr;
+    }
+
+    return MCSymbolRefExpr::create(&SRE->getSymbol(), Context);
+  }
+
+  case MCExpr::Unary: {
+    const MCUnaryExpr *UE = cast<MCUnaryExpr>(E);
+    const MCExpr *Sub = ExtractModifierFromExpr(UE->getSubExpr(), Variant);
+    if (!Sub)
+      return nullptr;
+    return MCUnaryExpr::create(UE->getOpcode(), Sub, Context);
+  }
+
+  case MCExpr::Binary: {
+    const MCBinaryExpr *BE = cast<MCBinaryExpr>(E);
+    VEMCExpr::VariantKind LHSVariant, RHSVariant;
+    const MCExpr *LHS = ExtractModifierFromExpr(BE->getLHS(), LHSVariant);
+    const MCExpr *RHS = ExtractModifierFromExpr(BE->getRHS(), RHSVariant);
+
+    if (!LHS && !RHS)
+      return nullptr;
+
+    if (!LHS)
+      LHS = BE->getLHS();
+    if (!RHS)
+      RHS = BE->getRHS();
+
+    if (LHSVariant == VEMCExpr::VK_VE_None)
+      Variant = RHSVariant;
+    else if (RHSVariant == VEMCExpr::VK_VE_None)
+      Variant = LHSVariant;
+    else if (LHSVariant == RHSVariant)
+      Variant = LHSVariant;
+    else
+      return nullptr;
+
+    return MCBinaryExpr::create(BE->getOpcode(), LHS, RHS, Context);
   }
-  return VEMCExpr::create(VK, Res, getContext());
+  }
+
+  llvm_unreachable("Invalid expression kind!");
+}
+
+const MCExpr *VEAsmParser::FixupVariantKind(const MCExpr *E) {
+  MCContext &Context = getParser().getContext();
+
+  switch (E->getKind()) {
+  case MCExpr::Target:
+  case MCExpr::Constant:
+  case MCExpr::SymbolRef:
+    return E;
+
+  case MCExpr::Unary: {
+    const MCUnaryExpr *UE = cast<MCUnaryExpr>(E);
+    const MCExpr *Sub = FixupVariantKind(UE->getSubExpr());
+    if (Sub == UE->getSubExpr())
+      return E;
+    return MCUnaryExpr::create(UE->getOpcode(), Sub, Context);
+  }
+
+  case MCExpr::Binary: {
+    const MCBinaryExpr *BE = cast<MCBinaryExpr>(E);
+    const MCExpr *LHS = FixupVariantKind(BE->getLHS());
+    const MCExpr *RHS = FixupVariantKind(BE->getRHS());
+    if (LHS == BE->getLHS() && RHS == BE->getRHS())
+      return E;
+    return MCBinaryExpr::create(BE->getOpcode(), LHS, RHS, Context);
+  }
+  }
+
+  llvm_unreachable("Invalid expression kind!");
+}
+
+/// ParseExpression.  This differs from the default "parseExpression" in that
+/// it handles modifiers.
+bool VEAsmParser::ParseExpression(const MCExpr *&EVal) {
+  // Handle \code symbol @lo32/@hi32/etc \endcode.
+  if (getParser().parseExpression(EVal))
+    return true;
+
+  // Convert MCSymbolRefExpr with VK_* to MCExpr with VK_*.
+  EVal = FixupVariantKind(EVal);
+  VEMCExpr::VariantKind Variant;
+  const MCExpr *E = ExtractModifierFromExpr(EVal, Variant);
+  if (E)
+    EVal = VEMCExpr::create(Variant, E, getParser().getContext());
+
+  return false;
 }
 
 OperandMatchResultTy VEAsmParser::parseMEMOperand(OperandVector &Operands) {
   LLVM_DEBUG(dbgs() << "parseMEMOperand\n");
   const AsmToken &Tok = Parser.getTok();
   SMLoc S = Tok.getLoc();
   SMLoc E = Tok.getEndLoc();
   // Parse ASX format
   //   disp
   //   disp(, base)
   //   disp(index)
   //   disp(index, base)
   //   (, base)
   //   (index)
   //   (index, base)
 
   std::unique_ptr<VEOperand> Offset;
   switch (getLexer().getKind()) {
   default:
     return MatchOperand_NoMatch;
 
   case AsmToken::Minus:
   case AsmToken::Integer:
-  case AsmToken::Dot: {
+  case AsmToken::Dot:
+  case AsmToken::Identifier: {
     const MCExpr *EVal;
-    if (!getParser().parseExpression(EVal, E))
+    if (!ParseExpression(EVal))
       Offset = VEOperand::CreateImm(EVal, S, E);
     else
       return MatchOperand_NoMatch;
     break;
   }
 
-  case AsmToken::Identifier: {
-    StringRef Identifier;
-    if (!getParser().parseIdentifier(Identifier)) {
-      E = SMLoc::getFromPointer(Parser.getTok().getLoc().getPointer() - 1);
-      const MCExpr *EVal = parseIdentifier(Identifier);
-
-      E = SMLoc::getFromPointer(Parser.getTok().getLoc().getPointer() - 1);
-      Offset = VEOperand::CreateImm(EVal, S, E);
-    }
-    break;
-  }
-
   case AsmToken::LParen:
     // empty disp (= 0)
     Offset =
         VEOperand::CreateImm(MCConstantExpr::create(0, getContext()), S, E);
     break;
   }
 
   switch (getLexer().getKind()) {
OperandMatchResultTy VEAsmParser::parseMEMAsOperand(OperandVector &Operands) {
   unsigned BaseReg = VE::NoRegister;
   std::unique_ptr<VEOperand> Offset;
   switch (getLexer().getKind()) {
   default:
     return MatchOperand_NoMatch;
 
   case AsmToken::Minus:
   case AsmToken::Integer:
-  case AsmToken::Dot: {
+  case AsmToken::Dot:
+  case AsmToken::Identifier: {
     const MCExpr *EVal;
-    if (!getParser().parseExpression(EVal, E))
+    if (!ParseExpression(EVal))
       Offset = VEOperand::CreateImm(EVal, S, E);
     else
       return MatchOperand_NoMatch;
     break;
   }
 
   case AsmToken::Percent:
     if (ParseRegister(BaseReg, S, E))
case AsmToken::Percent:
     unsigned RegNo;
     if (tryParseRegister(RegNo, S, E) == MatchOperand_Success)
       Op = VEOperand::CreateReg(RegNo, S, E);
     break;
 
   case AsmToken::Minus:
   case AsmToken::Integer:
   case AsmToken::Dot:
-    if (!getParser().parseExpression(EVal, E))
+  case AsmToken::Identifier:
+    if (!ParseExpression(EVal))
       Op = VEOperand::CreateImm(EVal, S, E);
     break;
-
-  case AsmToken::Identifier: {
-    StringRef Identifier;
-    if (!getParser().parseIdentifier(Identifier)) {
-      E = SMLoc::getFromPointer(Parser.getTok().getLoc().getPointer() - 1);
-      MCSymbol *Sym = getContext().getOrCreateSymbol(Identifier);
-
-      const MCExpr *Res =
-          MCSymbolRefExpr::create(Sym, MCSymbolRefExpr::VK_None, getContext());
-      Op = VEOperand::CreateImm(Res, S, E);
-    }
-    break;
-  }
   }
   return (Op) ? MatchOperand_Success : MatchOperand_ParseFail;
 }
 
 // Force static initialization.
 extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeVEAsmParser() {
   RegisterMCAsmParser<VEAsmParser> A(getTheVETarget());
 }

llvm/test/MC/VE/sym-br.s

+# RUN: llvm-mc -triple=ve %s -o - | FileCheck %s
+# RUN: llvm-mc -triple=ve -filetype=obj %s -o - | llvm-objdump -r - | FileCheck %s --check-prefix=CHECK-OBJ
+
+        b.l.t tgt(, %s1)
+        b.l.t tgt+24(, %s1)
+# CHECK: b.l.t tgt(, %s1)
+# CHECK-NEXT: b.l.t tgt+24(, %s1)
+
+# CHECK-OBJ: 0 R_VE_REFLONG tgt
+# CHECK-OBJ-NEXT: 8 R_VE_REFLONG tgt+0x18

llvm/test/MC/VE/symbols.s

 # RUN: llvm-mc -triple=ve %s -o - | FileCheck %s
 # RUN: llvm-mc -triple=ve -filetype=obj %s -o - | llvm-objdump -r - | FileCheck %s --check-prefix=CHECK-OBJ
 
         lea %s0, var
         lea %s1, var@lo
         and %s1, %s1, (32)0
         lea.sl %s1, var@hi(, %s1)
+        lea %s1, var+8@lo
+        and %s1, %s1, (32)0
+        lea.sl %s1, var+8@hi(, %s1)
 # CHECK: lea %s0, var
 # CHECK-NEXT: lea %s1, var@lo
 # CHECK-NEXT: and %s1, %s1, (32)0
 # CHECK-NEXT: lea.sl %s1, var@hi(, %s1)
+# CHECK-NEXT: lea %s1, var+8@lo
+# CHECK-NEXT: and %s1, %s1, (32)0
+# CHECK-NEXT: lea.sl %s1, var+8@hi(, %s1)
 
 # CHECK-OBJ: 0 R_VE_REFLONG var
 # CHECK-OBJ-NEXT: 8 R_VE_LO32 var
 # CHECK-OBJ-NEXT: 18 R_VE_HI32 var
+# CHECK-OBJ-NEXT: 20 R_VE_LO32 var+0x8
+# CHECK-OBJ-NEXT: 30 R_VE_HI32 var+0x8