implement memory operand parsing.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@73875 91177308-0d34-0410-b5e6-96231b3b80d8

tools/llvm-mc/AsmParser.cpp

@@ -48,10 +48,12 @@ void AsmParser::EatToEndOfStatement() {
     Lexer.Lex();
 }
 
+
 struct AsmParser::X86Operand {
   enum {
     Register,
-    Immediate
+    Immediate,
+    Memory
   } Kind;
   
   union {
@@ -63,6 +65,14 @@ struct AsmParser::X86Operand {
       // FIXME: Should be a general expression.
       int64_t Val;
     } Imm;
+    
+    struct {
+      unsigned SegReg;
+      int64_t Disp;     // FIXME: Should be a general expression.
+      unsigned BaseReg;
+      unsigned Scale;
+      unsigned ScaleReg;
+    } Mem;
   };
   
   static X86Operand CreateReg(unsigned RegNo) {
@@ -77,33 +87,39 @@ struct AsmParser::X86Operand {
     Res.Imm.Val = Val;
     return Res;
   }
+  static X86Operand CreateMem(unsigned SegReg, int64_t Disp, unsigned BaseReg,
+                              unsigned Scale, unsigned ScaleReg) {
+    X86Operand Res;
+    Res.Kind = Memory;
+    Res.Mem.SegReg   = SegReg;
+    Res.Mem.Disp     = Disp;
+    Res.Mem.BaseReg  = BaseReg;
+    Res.Mem.Scale    = Scale;
+    Res.Mem.ScaleReg = ScaleReg;
+    return Res;
+  }
 };
 
 bool AsmParser::ParseX86Operand(X86Operand &Op) {
   switch (Lexer.getKind()) {
   default:
-    return TokError("unknown token at start of instruction operand");
+    return ParseX86MemOperand(Op);
   case asmtok::Register:
     // FIXME: Decode reg #.
+    // FIXME: if a segment register, this could either be just the seg reg, or
+    // the start of a memory operand.
     Op = X86Operand::CreateReg(0);
     Lexer.Lex(); // Eat register.
     return false;
-  case asmtok::Dollar:
+  case asmtok::Dollar: {
     // $42 -> immediate.
     Lexer.Lex();
-    // FIXME: Parse an arbitrary expression here, like $(4+5)
-    if (Lexer.isNot(asmtok::IntVal))
+    int64_t Val;
+    if (ParseExpression(Val))
       return TokError("expected integer constant");
-    
-    Op = X86Operand::CreateReg(Lexer.getCurIntVal());
-    Lexer.Lex(); // Eat register.
-    return false;
-  case asmtok::Identifier:
-    // This is a label, this should be parsed as part of an expression, to
-    // handle things like LFOO+4
-    Op = X86Operand::CreateImm(0); // FIXME.
-    Lexer.Lex(); // Eat identifier.
+    Op = X86Operand::CreateReg(Val);
     return false;
+  }
       
   //case asmtok::Star:
   // * %eax
@@ -112,7 +128,107 @@ bool AsmParser::ParseX86Operand(X86Operand &Op) {
   }
 }
 
+/// ParseX86MemOperand: segment: disp(basereg, indexreg, scale)
+bool AsmParser::ParseX86MemOperand(X86Operand &Op) {
+  // FIXME: If SegReg ':'  (e.g. %gs:), eat and remember.
+  unsigned SegReg = 0;
+  
+  
+  // We have to disambiguate a parenthesized expression "(4+5)" from the start
+  // of a memory operand with a missing displacement "(%ebx)" or "(,%eax)".  The
+  // only way to do this without lookahead is to eat the ( and see what is after
+  // it.
+  int64_t Disp = 0;
+  if (Lexer.isNot(asmtok::LParen)) {
+    if (ParseExpression(Disp)) return true;
+    
+    // After parsing the base expression we could either have a parenthesized
+    // memory address or not.  If not, return now.  If so, eat the (.
+    if (Lexer.isNot(asmtok::LParen)) {
+      Op = X86Operand::CreateMem(SegReg, Disp, 0, 0, 0);
+      return false;
+    }
+    
+    // Eat the '('.
+    Lexer.Lex();
+  } else {
+    // Okay, we have a '('.  We don't know if this is an expression or not, but
+    // so we have to eat the ( to see beyond it.
+    Lexer.Lex(); // Eat the '('.
+    
+    if (Lexer.is(asmtok::Register) || Lexer.is(asmtok::Comma)) {
+      // Nothing to do here, fall into the code below with the '(' part of the
+      // memory operand consumed.
+    } else {
+      // FIXME: Call ParseParenExpression with the leading ( consumed.
+      return TokError("FIXME: Paren expr not implemented yet!");
+    }
+  }
+  
+  // If we reached here, then we just ate the ( of the memory operand.  Process
+  // the rest of the memory operand.
+  unsigned BaseReg = 0, ScaleReg = 0, Scale = 0;
+  
+  if (Lexer.is(asmtok::Register)) {
+    BaseReg = 123; // FIXME: decode reg #
+    Lexer.Lex();  // eat the register.
+  }
+  
+  if (Lexer.is(asmtok::Comma)) {
+    Lexer.Lex(); // eat the comma.
+    
+    if (Lexer.is(asmtok::Register)) {
+      ScaleReg = 123; // FIXME: decode reg #
+      Lexer.Lex();  // eat the register.
+      Scale = 1;      // If not specified, the scale defaults to 1.
+    }
+    
+    if (Lexer.is(asmtok::Comma)) {
+      Lexer.Lex(); // eat the comma.
 
+      // If present, get and validate scale amount.
+      if (Lexer.is(asmtok::IntVal)) {
+        int64_t ScaleVal = Lexer.getCurIntVal();
+        if (ScaleVal != 1 && ScaleVal != 2 && ScaleVal != 4 && ScaleVal != 8)
+          return TokError("scale factor in address must be 1, 2, 4 or 8");
+        Lexer.Lex();  // eat the scale.
+        Scale = (unsigned)ScaleVal;
+      }
+    }
+  }
+  
+  // Ok, we've eaten the memory operand, verify we have a ')' and eat it too.
+  if (Lexer.isNot(asmtok::RParen))
+    return TokError("unexpected token in memory operand");
+  Lexer.Lex(); // Eat the ')'.
+  
+  Op = X86Operand::CreateMem(SegReg, Disp, BaseReg, Scale, ScaleReg);
+  return false;
+}
+
+
+/// ParseExpression - Parse an expression and return it.
+/// FIXME: This should handle real expressions, we do something trivial for now.
+bool AsmParser::ParseExpression(int64_t &Res) {
+  switch (Lexer.getKind()) {
+  default:
+    return TokError("unknown token in expression");
+  case asmtok::Identifier:
+    // This is a label, this should be parsed as part of an expression, to
+    // handle things like LFOO+4
+    Res = 0; // FIXME.
+    Lexer.Lex(); // Eat identifier.
+    return false;
+  case asmtok::IntVal:
+    Res = Lexer.getCurIntVal();
+    Lexer.Lex(); // Eat identifier.
+    return false;
+  }
+}
+  
+  
+  
+  
 /// ParseStatement:
 ///   ::= EndOfStatement
 ///   ::= Label* Directive ...Operands... EndOfStatement

tools/llvm-mc/AsmParser.h

@@ -37,6 +37,8 @@ private:
   void EatToEndOfStatement();
   
   bool ParseX86Operand(X86Operand &Op);
+  bool ParseX86MemOperand(X86Operand &Op);
+  bool ParseExpression(int64_t &Res);
 };
 
 } // end namespace llvm