[X86][AsmParser][MS-compatability] Binary/Unary operators enhancements

[coby]

Summary

Introducing MOD binary operator
https://msdn.microsoft.com/en-us/library/hha180wt.aspx

Enhancing unary operators NEG and NOT, to support more complex patterns

[coby]

ping

[echristo]

Reid is still the best reviewer for this.

[rnk]

lgtm Sorry for the delay, I've been trying to prioritize writing PDBs in LLD to stay in the zone.

lib/Target/X86/AsmParser/X86AsmParser.cpp

542

Thanks, this is a nice improvement!

[coby]

In D33876#791376, @rnk wrote:

lgtm Sorry for the delay, I've been trying to prioritize writing PDBs in LLD to stay in the zone.

np
thanks

lib/Target/X86/AsmParser/X86AsmParser.cpp

static const char OpPrecedence[] = {
   1, // IC_XOR
   2, // IC_AND
   3, // IC_LSHIFT
   3, // IC_RSHIFT
   4, // IC_PLUS
   4, // IC_MINUS
   5, // IC_MULTIPLY
   5, // IC_DIVIDE
-  6, // IC_RPAREN
-  7, // IC_LPAREN
+  5, // IC_MOD
+  6, // IC_NOT
+  7, // IC_NEG
+  8, // IC_RPAREN
+  9, // IC_LPAREN
   0, // IC_IMM
   0  // IC_REGISTER
 };
 
 class X86AsmParser : public MCTargetAsmParser {
   const MCInstrInfo &MII;
   ParseInstructionInfo *InstInfo;
   std::unique_ptr<X86AsmInstrumentation> Instrumentation;
enum InfixCalculatorTok {
     IC_XOR,
     IC_AND,
     IC_LSHIFT,
     IC_RSHIFT,
     IC_PLUS,
     IC_MINUS,
     IC_MULTIPLY,
     IC_DIVIDE,
+    IC_MOD,
+    IC_NOT,
+    IC_NEG,
     IC_RPAREN,
     IC_LPAREN,
     IC_IMM,
     IC_REGISTER
   };
 
   enum IntelOperatorKind {
     IOK_INVALID = 0,
     IOK_LENGTH,
     IOK_SIZE,
     IOK_TYPE,
     IOK_OFFSET
   };
 
   class InfixCalculator {
     typedef std::pair< InfixCalculatorTok, int64_t > ICToken;
     SmallVector<InfixCalculatorTok, 4> InfixOperatorStack;
     SmallVector<ICToken, 4> PostfixStack;
 
+    bool isUnaryOperator(const InfixCalculatorTok Op) {
+      return Op == IC_NEG || Op == IC_NOT;
+    }
+
   public:
     int64_t popOperand() {
       assert (!PostfixStack.empty() && "Poped an empty stack!");
       ICToken Op = PostfixStack.pop_back_val();
       assert ((Op.first == IC_IMM || Op.first == IC_REGISTER)
               && "Expected and immediate or register!");
       return Op.second;
     }
int64_t execute() {
       if (PostfixStack.empty())
         return 0;
 
       SmallVector<ICToken, 16> OperandStack;
       for (unsigned i = 0, e = PostfixStack.size(); i != e; ++i) {
         ICToken Op = PostfixStack[i];
         if (Op.first == IC_IMM || Op.first == IC_REGISTER) {
           OperandStack.push_back(Op);
+        } else if (isUnaryOperator(Op.first)) {
+          assert (OperandStack.size() > 0 && "Too few operands.");
+          ICToken Operand = OperandStack.pop_back_val();
+          assert (Operand.first == IC_IMM &&
+                  "Unary operation with a register!");
+          switch (Op.first) {
+          default:
+            report_fatal_error("Unexpected operator!");
+            break;
+          case IC_NEG:
+            OperandStack.push_back(std::make_pair(IC_IMM, -Operand.second));
+            break;
+          case IC_NOT:
+            OperandStack.push_back(std::make_pair(IC_IMM, ~Operand.second));
+            break;
+          }
         } else {
           assert (OperandStack.size() > 1 && "Too few operands.");
           int64_t Val;
           ICToken Op2 = OperandStack.pop_back_val();
           ICToken Op1 = OperandStack.pop_back_val();
           switch (Op.first) {
           default:
             report_fatal_error("Unexpected operator!");
int64_t execute() {
             break;
           case IC_DIVIDE:
             assert (Op1.first == IC_IMM && Op2.first == IC_IMM &&
                     "Divide operation with an immediate and a register!");
             assert (Op2.second != 0 && "Division by zero!");
             Val = Op1.second / Op2.second;
             OperandStack.push_back(std::make_pair(IC_IMM, Val));
             break;
+          case IC_MOD:
+            assert (Op1.first == IC_IMM && Op2.first == IC_IMM &&
+                    "Modulo operation with an immediate and a register!");
+            Val = Op1.second % Op2.second;
+            OperandStack.push_back(std::make_pair(IC_IMM, Val));
+            break;
           case IC_OR:
             assert (Op1.first == IC_IMM && Op2.first == IC_IMM &&
                     "Or operation with an immediate and a register!");
             Val = Op1.second | Op2.second;
             OperandStack.push_back(std::make_pair(IC_IMM, Val));
             break;
           case IC_XOR:
             assert(Op1.first == IC_IMM && Op2.first == IC_IMM &&
enum IntelExprState {
     IES_AND,
     IES_LSHIFT,
     IES_RSHIFT,
     IES_PLUS,
     IES_MINUS,
     IES_NOT,
     IES_MULTIPLY,
     IES_DIVIDE,
+    IES_MOD,
     IES_LBRAC,
     IES_RBRAC,
     IES_LPAREN,
     IES_RPAREN,
     IES_REGISTER,
     IES_INTEGER,
     IES_IDENTIFIER,
     IES_ERROR
void onPlus() {
       PrevState = CurrState;
     }
     void onMinus() {
       IntelExprState CurrState = State;
       switch (State) {
       default:
         State = IES_ERROR;
         break;
+      case IES_OR:
+      case IES_XOR:
+      case IES_AND:
+      case IES_LSHIFT:
+      case IES_RSHIFT:
       case IES_PLUS:
       case IES_NOT:
       case IES_MULTIPLY:
       case IES_DIVIDE:
+      case IES_MOD:
       case IES_LPAREN:
       case IES_RPAREN:
       case IES_LBRAC:
       case IES_RBRAC:
       case IES_INTEGER:
       case IES_REGISTER:
         State = IES_MINUS;
-        // Only push the minus operator if it is not a unary operator.
-        if (!(CurrState == IES_PLUS || CurrState == IES_MINUS ||
-              CurrState == IES_MULTIPLY || CurrState == IES_DIVIDE ||
-              CurrState == IES_LPAREN || CurrState == IES_LBRAC))
+        // push minus operator if it is not a negate operator
+        if (CurrState == IES_REGISTER || CurrState == IES_RPAREN ||
+            CurrState == IES_INTEGER  || CurrState == IES_RBRAC)
           IC.pushOperator(IC_MINUS);
+        else
+          IC.pushOperator(IC_NEG);
         if (CurrState == IES_REGISTER && PrevState != IES_MULTIPLY) {
           // If we already have a BaseReg, then assume this is the IndexReg with
           // a scale of 1.
           if (!BaseReg) {
             BaseReg = TmpReg;
           } else {
             assert (!IndexReg && "BaseReg/IndexReg already set!");
             IndexReg = TmpReg;
             Scale = 1;
           }
         }
         break;
       }
       PrevState = CurrState;
     }
     void onNot() {
       IntelExprState CurrState = State;
       switch (State) {
       default:
         State = IES_ERROR;
         break;
+      case IES_OR:
+      case IES_XOR:
+      case IES_AND:
+      case IES_LSHIFT:
+      case IES_RSHIFT:
       case IES_PLUS:
+      case IES_MINUS:
       case IES_NOT:
+      case IES_MULTIPLY:
+      case IES_DIVIDE:
+      case IES_MOD:
+      case IES_LPAREN:
+      case IES_LBRAC:
         State = IES_NOT;
+        IC.pushOperator(IC_NOT);
         break;
       }
       PrevState = CurrState;
     }
     void onRegister(unsigned Reg) {
       IntelExprState CurrState = State;
       switch (State) {
       default:
bool onInteger(int64_t TmpInt, StringRef &ErrMsg) {
       case IES_MINUS:
       case IES_NOT:
       case IES_OR:
       case IES_XOR:
       case IES_AND:
       case IES_LSHIFT:
       case IES_RSHIFT:
       case IES_DIVIDE:
+      case IES_MOD:
       case IES_MULTIPLY:
       case IES_LPAREN:
         State = IES_INTEGER;
         if (PrevState == IES_REGISTER && CurrState == IES_MULTIPLY) {
           // Index Register - Register * Scale
           assert (!IndexReg && "IndexReg already set!");
           IndexReg = TmpReg;
           Scale = TmpInt;
           if(Scale != 1 && Scale != 2 && Scale != 4 && Scale != 8) {
             ErrMsg = "scale factor in address must be 1, 2, 4 or 8";
             return true;
           }
           // Get the scale and replace the 'Register * Scale' with '0'.
           IC.popOperator();
-        } else if ((PrevState == IES_PLUS || PrevState == IES_MINUS ||
-                    PrevState == IES_OR || PrevState == IES_AND ||
-                    PrevState == IES_LSHIFT || PrevState == IES_RSHIFT ||
-                    PrevState == IES_MULTIPLY || PrevState == IES_DIVIDE ||
-                    PrevState == IES_LPAREN || PrevState == IES_LBRAC ||
-                    PrevState == IES_NOT || PrevState == IES_XOR) &&
-                   CurrState == IES_MINUS) {
-          // Unary minus.  No need to pop the minus operand because it was never
-          // pushed.
-          IC.pushOperand(IC_IMM, -TmpInt); // Push -Imm.
-        } else if ((PrevState == IES_PLUS || PrevState == IES_MINUS ||
-                    PrevState == IES_OR || PrevState == IES_AND ||
-                    PrevState == IES_LSHIFT || PrevState == IES_RSHIFT ||
-                    PrevState == IES_MULTIPLY || PrevState == IES_DIVIDE ||
-                    PrevState == IES_LPAREN || PrevState == IES_LBRAC ||
-                    PrevState == IES_NOT || PrevState == IES_XOR) &&
-                   CurrState == IES_NOT) {
-          // Unary not.  No need to pop the not operand because it was never
-          // pushed.
-          IC.pushOperand(IC_IMM, ~TmpInt); // Push ~Imm.
         } else {
           IC.pushOperand(IC_IMM, TmpInt);
         }
         break;
       }
       PrevState = CurrState;
       return false;
     }
void onDivide() {
         break;
       case IES_INTEGER:
       case IES_RPAREN:
         State = IES_DIVIDE;
         IC.pushOperator(IC_DIVIDE);
         break;
       }
     }
+    void onMod() {
+      PrevState = State;
+      switch (State) {
+      default:
+        State = IES_ERROR;
+        break;
+      case IES_INTEGER:
+      case IES_RPAREN:
+        State = IES_MOD;
+        IC.pushOperator(IC_MOD);
+        break;
+      }
+    }
     void onLBrac() {
       PrevState = State;
       switch (State) {
       default:
         State = IES_ERROR;
         break;
       case IES_RBRAC:
         State = IES_PLUS;
void onLParen() {
       case IES_NOT:
       case IES_OR:
       case IES_XOR:
       case IES_AND:
       case IES_LSHIFT:
       case IES_RSHIFT:
       case IES_MULTIPLY:
       case IES_DIVIDE:
+      case IES_MOD:
       case IES_LPAREN:
-        // FIXME: We don't handle this type of unary minus or not, yet.
-        if ((PrevState == IES_PLUS || PrevState == IES_MINUS ||
-            PrevState == IES_OR || PrevState == IES_AND ||
-            PrevState == IES_LSHIFT || PrevState == IES_RSHIFT ||
-            PrevState == IES_MULTIPLY || PrevState == IES_DIVIDE ||
-            PrevState == IES_LPAREN || PrevState == IES_LBRAC ||
-            PrevState == IES_NOT || PrevState == IES_XOR) &&
-            (CurrState == IES_MINUS || CurrState == IES_NOT)) {
-          State = IES_ERROR;
-          break;
-        }
         State = IES_LPAREN;
         IC.pushOperator(IC_LPAREN);
         break;
       }
       PrevState = CurrState;
     }
     void onRParen() {
       PrevState = State;
bool X86AsmParser::ParseIntelNamedOperator(StringRef Name, IntelExprStateMachine &SM) {
   else if (Name.equals_lower("shl"))
     SM.onLShift();
   else if (Name.equals_lower("shr"))
     SM.onRShift();
   else if (Name.equals_lower("xor"))
     SM.onXor();
   else if (Name.equals_lower("and"))
     SM.onAnd();
+  else if (Name.equals_lower("mod"))
+    SM.onMod();
   else
     return false;
   return true;
 }
 
 bool X86AsmParser::ParseIntelExpression(IntelExprStateMachine &SM, SMLoc &End) {
   MCAsmParser &Parser = getParser();
   const AsmToken &Tok = Parser.getTok();

test/MC/X86/intel-syntax-bitwise-ops.s

 // CHECK: addl $5, %eax
     add eax, 6 ^ 3
 // CHECK: addl $5, %eax
     add eax, 6 xor 3
 // CHECK: addl $5, %eax
     add eax, 6 XOR 3
 // CHECK: addl $5, %eax
     add eax, 6 XOR 3 shl 1 SHR 1
+// CHECK: movl  $-9, %eax
+    mov eax, not(1 shl 3)
+// CHECK: movl    $-2, %eax
+    mov eax, ~(0x8 shr 3)
+// CHECK: movl  $-4, %eax
+    mov eax,  not(1 or 3)
+// CHECK: movl  $-2, %eax
+    mov eax, -(1 xor 3)
+// CHECK: movl  $-2, %eax
+    mov eax, not(1 and 3)
+// CHECK: movl  $3, %eax
+    mov eax, not(not 3)
+// CHECK: movl  $-3, %eax
+    mov eax, ~(5 mod 3)
+// CHECK: movl  $-2, %eax
+    mov eax, (-5 mod 3)
+