Fixed several instruction handling issues: 1) Added proper handling for zero displacements in memory operands, 2) Fixed large unsigned displacement values display, 3) Added CmpEaxImmHandler for CMP EAX, imm32 instruction, 4) Fixed JP and JNP conditional jump instruction types

X86Disassembler/X86/Handlers/Cmp/CmpAlImmHandler.cs

@@ -46,7 +46,7 @@ public class CmpAlImmHandler : InstructionHandler
         byte imm8 = Decoder.ReadByte();
 
         // Create the register operand for AL
-        var alOperand = OperandFactory.CreateRegisterOperand(RegisterIndex.A, 8);
+        var alOperand = OperandFactory.CreateRegisterOperand8(RegisterIndex8.AL);
         
         // Create the immediate operand
         var immOperand = OperandFactory.CreateImmediateOperand(imm8, 8);

X86Disassembler/X86/Handlers/Cmp/CmpEaxImmHandler.cs

@@ -0,0 +1,60 @@
+using X86Disassembler.X86.Operands;
+
+namespace X86Disassembler.X86.Handlers.Cmp;
+
+/// <summary>
+/// Handler for CMP EAX, imm32 instruction (opcode 3D)
+/// </summary>
+public class CmpEaxImmHandler : InstructionHandler
+{
+    /// <summary>
+    /// Initializes a new instance of the CmpEaxImmHandler class
+    /// </summary>
+    /// <param name="decoder">The instruction decoder that owns this handler</param>
+    public CmpEaxImmHandler(InstructionDecoder decoder)
+        : base(decoder)
+    {
+    }
+
+    /// <summary>
+    /// Checks if this handler can decode the given opcode
+    /// </summary>
+    /// <param name="opcode">The opcode to check</param>
+    /// <returns>True if this handler can decode the opcode</returns>
+    public override bool CanHandle(byte opcode)
+    {
+        // CMP EAX, imm32 is encoded as 3D
+        return opcode == 0x3D;
+    }
+
+    /// <summary>
+    /// Decodes a CMP EAX, imm32 instruction
+    /// </summary>
+    /// <param name="opcode">The opcode of the instruction</param>
+    /// <param name="instruction">The instruction object to populate</param>
+    /// <returns>True if the instruction was successfully decoded</returns>
+    public override bool Decode(byte opcode, Instruction instruction)
+    {
+        // Set the instruction type
+        instruction.Type = InstructionType.Cmp;
+
+        // Check if we have enough bytes for the immediate value
+        if (!Decoder.CanReadUInt())
+        {
+            return false;
+        }
+
+        // Read the immediate value
+        uint imm32 = Decoder.ReadUInt32();
+
+        // Set the structured operands
+        // CMP EAX, imm32 has two operands: EAX and the immediate value
+        instruction.StructuredOperands = 
+        [
+            OperandFactory.CreateRegisterOperand(RegisterIndex.A),
+            OperandFactory.CreateImmediateOperand(imm32)
+        ];
+
+        return true;
+    }
+}

X86Disassembler/X86/Handlers/Cmp/CmpR8Rm8Handler.cs

@@ -0,0 +1,70 @@
+using X86Disassembler.X86.Operands;
+
+namespace X86Disassembler.X86.Handlers.Cmp;
+
+/// <summary>
+/// Handler for CMP r8, r/m8 instruction (0x3A)
+/// </summary>
+public class CmpR8Rm8Handler : InstructionHandler
+{
+    /// <summary>
+    /// Initializes a new instance of the CmpR8Rm8Handler class
+    /// </summary>
+    /// <param name="decoder">The instruction decoder that owns this handler</param>
+    public CmpR8Rm8Handler(InstructionDecoder decoder) 
+        : base(decoder)
+    {
+    }
+    
+    /// <summary>
+    /// Checks if this handler can decode the given opcode
+    /// </summary>
+    /// <param name="opcode">The opcode to check</param>
+    /// <returns>True if this handler can decode the opcode</returns>
+    public override bool CanHandle(byte opcode)
+    {
+        if (opcode != 0x3A)
+            return false;
+            
+        // Check if we can read the ModR/M byte
+        if (!Decoder.CanReadByte())
+            return false;
+        
+        return true;
+    }
+    
+    /// <summary>
+    /// Decodes a CMP r8, r/m8 instruction
+    /// </summary>
+    /// <param name="opcode">The opcode of the instruction</param>
+    /// <param name="instruction">The instruction object to populate</param>
+    /// <returns>True if the instruction was successfully decoded</returns>
+    public override bool Decode(byte opcode, Instruction instruction)
+    {
+        // Set the instruction type
+        instruction.Type = InstructionType.Cmp;
+        
+        // Check if we have enough bytes for the ModR/M byte
+        if (!Decoder.CanReadByte())
+        {
+            return false;
+        }
+
+        // Read the ModR/M byte, specifying that we're dealing with 8-bit operands
+        var (_, reg, _, sourceOperand) = ModRMDecoder.ReadModRM8();
+        
+        // Note: The operand size is already set to 8-bit by the ReadModRM8 method
+        
+        // Create the destination register operand using the 8-bit register type
+        var destinationOperand = OperandFactory.CreateRegisterOperand8(reg);
+        
+        // Set the structured operands
+        instruction.StructuredOperands = 
+        [
+            destinationOperand,
+            sourceOperand
+        ];
+        
+        return true;
+    }
+}

X86Disassembler/X86/Handlers/Cmp/CmpRm8R8Handler.cs

@@ -0,0 +1,70 @@
+using X86Disassembler.X86.Operands;
+
+namespace X86Disassembler.X86.Handlers.Cmp;
+
+/// <summary>
+/// Handler for CMP r/m8, r8 instruction (0x38)
+/// </summary>
+public class CmpRm8R8Handler : InstructionHandler
+{
+    /// <summary>
+    /// Initializes a new instance of the CmpRm8R8Handler class
+    /// </summary>
+    /// <param name="decoder">The instruction decoder that owns this handler</param>
+    public CmpRm8R8Handler(InstructionDecoder decoder) 
+        : base(decoder)
+    {
+    }
+    
+    /// <summary>
+    /// Checks if this handler can decode the given opcode
+    /// </summary>
+    /// <param name="opcode">The opcode to check</param>
+    /// <returns>True if this handler can decode the opcode</returns>
+    public override bool CanHandle(byte opcode)
+    {
+        if (opcode != 0x38)
+            return false;
+            
+        // Check if we can read the ModR/M byte
+        if (!Decoder.CanReadByte())
+            return false;
+        
+        return true;
+    }
+    
+    /// <summary>
+    /// Decodes a CMP r/m8, r8 instruction
+    /// </summary>
+    /// <param name="opcode">The opcode of the instruction</param>
+    /// <param name="instruction">The instruction object to populate</param>
+    /// <returns>True if the instruction was successfully decoded</returns>
+    public override bool Decode(byte opcode, Instruction instruction)
+    {
+        // Set the instruction type
+        instruction.Type = InstructionType.Cmp;
+        
+        // Check if we have enough bytes for the ModR/M byte
+        if (!Decoder.CanReadByte())
+        {
+            return false;
+        }
+
+        // Read the ModR/M byte, specifying that we're dealing with 8-bit operands
+        var (_, reg, _, destinationOperand) = ModRMDecoder.ReadModRM8();
+        
+        // Note: The operand size is already set to 8-bit by the ReadModRM8 method
+        
+        // Create the source register operand using the 8-bit register type
+        var sourceOperand = OperandFactory.CreateRegisterOperand8(reg);
+        
+        // Set the structured operands
+        instruction.StructuredOperands = 
+        [
+            destinationOperand,
+            sourceOperand
+        ];
+        
+        return true;
+    }
+}

X86Disassembler/X86/Handlers/InstructionHandlerFactory.cs

@@ -231,11 +231,18 @@ public class InstructionHandlerFactory
     /// </summary>
     private void RegisterCmpHandlers()
     {
-        // Add Cmp handlers
+        // Add Cmp handlers for 32-bit operands
         _handlers.Add(new CmpR32Rm32Handler(_decoder));
         _handlers.Add(new CmpRm32R32Handler(_decoder));
+        
+        // Add Cmp handlers for 8-bit operands
+        _handlers.Add(new CmpRm8R8Handler(_decoder));  // CMP r/m8, r8 (opcode 38)
+        _handlers.Add(new CmpR8Rm8Handler(_decoder));  // CMP r8, r/m8 (opcode 3A)
+        
+        // Add Cmp handlers for immediate operands
         _handlers.Add(new CmpImmWithRm8Handler(_decoder));
-        _handlers.Add(new CmpAlImmHandler(_decoder));
+        _handlers.Add(new CmpAlImmHandler(_decoder));  // CMP AL, imm8 (opcode 3C)
+        _handlers.Add(new CmpEaxImmHandler(_decoder)); // CMP EAX, imm32 (opcode 3D)
 
         // Add CMP immediate handlers from ArithmeticImmediate namespace
         _handlers.Add(new CmpImmWithRm32Handler(_decoder));

X86Disassembler/X86/Handlers/Jump/ConditionalJumpHandler.cs

@@ -19,7 +19,7 @@ public class ConditionalJumpHandler : InstructionHandler
     [
         InstructionType.Jo, InstructionType.Jno, InstructionType.Jb, InstructionType.Jae, 
         InstructionType.Jz, InstructionType.Jnz, InstructionType.Jbe, InstructionType.Ja,
-        InstructionType.Js, InstructionType.Jns, InstructionType.Unknown, InstructionType.Unknown, 
+        InstructionType.Js, InstructionType.Jns, InstructionType.Jp, InstructionType.Jnp, 
         InstructionType.Jl, InstructionType.Jge, InstructionType.Jle, InstructionType.Jg
     ];
 

X86Disassembler/X86/ModRMDecoder.cs

@@ -102,13 +102,7 @@ public class ModRMDecoder
                     {
                         sbyte disp8 = (sbyte)_decoder.ReadByte();
 
-                        // For EBP (BP), always create a displacement memory operand, even if displacement is 0
+                        // Always create a displacement memory operand for mod=1, even if displacement is 0
-                        // This is because [EBP] with no displacement is encoded as [EBP+0]
-                        if (disp8 == 0 && rmIndex != RegisterIndex.Bp)
-                        {
-                            return OperandFactory.CreateBaseRegisterMemoryOperand(rmIndex, operandSize);
-                        }
-
                         return OperandFactory.CreateDisplacementMemoryOperand(rmIndex, disp8, operandSize);
                     }
 
@@ -140,7 +134,8 @@ public class ModRMDecoder
                         // This is because [EBP] with no displacement is encoded as [EBP+disp]
                         if (rmIndex == RegisterIndex.Bp)
                         {
-                            return OperandFactory.CreateDisplacementMemoryOperand(rmIndex, (int)disp32, operandSize);
+                            // Cast to long to preserve the unsigned value for large displacements
+                            return OperandFactory.CreateDisplacementMemoryOperand(rmIndex, (long)disp32, operandSize);
                         }
 
                         // Only show displacement if it's not zero
@@ -149,7 +144,8 @@ public class ModRMDecoder
                             return OperandFactory.CreateBaseRegisterMemoryOperand(rmIndex, operandSize);
                         }
 
-                        return OperandFactory.CreateDisplacementMemoryOperand(rmIndex, (int)disp32, operandSize);
+                        // Cast to long to preserve the unsigned value for large displacements
+                        return OperandFactory.CreateDisplacementMemoryOperand(rmIndex, (long)disp32, operandSize);
                     }
 
                     // Fallback for incomplete data

X86Disassembler/X86/Operands/DisplacementMemoryOperand.cs

@@ -35,10 +35,30 @@ public class DisplacementMemoryOperand : MemoryOperand
     /// </summary>
     public override string ToString()
     {
-        string sign = Displacement >= 0 ? "+" : "-";
+        // Get register name
         var registerName = RegisterMapper.GetRegisterName(BaseRegister, 32);
 
-        string formattedDisplacement = $"0x{Displacement:X2}";
+        // Format the displacement value
+        string formattedDisplacement;
+        string sign;
+        
+        // Handle positive and negative displacements
+        if (Displacement >= 0)
+        {
+            sign = "+";
+            formattedDisplacement = Displacement < 256 
+                ? $"0x{Displacement:X2}"
+                : $"0x{Displacement:X8}";
+        }
+        else
+        {
+            sign = "-";
+            // For negative values, take the absolute value for display
+            var absDisplacement = Math.Abs(Displacement);
+            formattedDisplacement = absDisplacement < 256 
+                ? $"0x{absDisplacement:X2}"
+                : $"0x{absDisplacement:X8}";
+        }
         
         return $"{GetSizePrefix()}[{registerName}{sign}{formattedDisplacement}]";
     }

X86DisassemblerTests/InstructionTests/CmpInstructionHandlerTests.cs

@@ -33,9 +33,9 @@ public class CmpInstructionHandlerTests
         
         // Check the first operand (AL)
         var alOperand = instruction.StructuredOperands[0];
-        Assert.IsType<RegisterOperand>(alOperand);
+        Assert.IsType<Register8Operand>(alOperand);
-        var registerOperand = (RegisterOperand)alOperand;
+        var registerOperand = (Register8Operand)alOperand;
-        Assert.Equal(RegisterIndex.A, registerOperand.Register);
+        Assert.Equal(RegisterIndex8.AL, registerOperand.Register);
         Assert.Equal(8, registerOperand.Size); // Validate that it's an 8-bit register (AL)
         
         // Check the second operand (immediate value)
@@ -70,9 +70,9 @@ public class CmpInstructionHandlerTests
         
         // Check the first operand (AL)
         var alOperand = instruction.StructuredOperands[0];
-        Assert.IsType<RegisterOperand>(alOperand);
+        Assert.IsType<Register8Operand>(alOperand);
-        var registerOperand = (RegisterOperand)alOperand;
+        var registerOperand = (Register8Operand)alOperand;
-        Assert.Equal(RegisterIndex.A, registerOperand.Register);
+        Assert.Equal(RegisterIndex8.AL, registerOperand.Register);
         Assert.Equal(8, registerOperand.Size); // Validate that it's an 8-bit register (AL)
         
         // Check the second operand (immediate value)

X86DisassemblerTests/TestData/call_tests.csv

@@ -41,19 +41,19 @@ FF14D9;[{ "Type": "Call", "Operands": ["dword ptr [ecx+ebx*8]"] }]
 # FF149D;[{ "Type": "Call", "Operands": ["dword ptr [ebp+ebx*4]"] }]
 
 # CALL m32 (opcode FF /2) with displacement
-FF5000;[{ "Type": "Call", "Operands": ["dword ptr [eax+0x0]"] }]
+FF5000;[{ "Type": "Call", "Operands": ["dword ptr [eax+0x00]"] }]
 FF5010;[{ "Type": "Call", "Operands": ["dword ptr [eax+0x10]"] }]
 FF90FFFFFF7F;[{ "Type": "Call", "Operands": ["dword ptr [eax+0x7FFFFFFF]"] }]
 FF9000000080;[{ "Type": "Call", "Operands": ["dword ptr [eax+0x80000000]"] }]
 
 # CALL m32 (opcode FF /2) with SIB and displacement
-FF5400FF;[{ "Type": "Call", "Operands": ["dword ptr [eax+eax*1-0x1]"] }]
+FF5400FF;[{ "Type": "Call", "Operands": ["dword ptr [eax+eax*1-0x01]"] }]
-FF54C0FF;[{ "Type": "Call", "Operands": ["dword ptr [eax+eax*8-0x1]"] }]
+FF54C0FF;[{ "Type": "Call", "Operands": ["dword ptr [eax+eax*8-0x01]"] }]
-FF5444FF;[{ "Type": "Call", "Operands": ["dword ptr [esp+eax*2-0x1]"] }]
+FF5444FF;[{ "Type": "Call", "Operands": ["dword ptr [esp+eax*2-0x01]"] }]
-FF5485FF;[{ "Type": "Call", "Operands": ["dword ptr [ebp+eax*4-0x1]"] }]
+FF5485FF;[{ "Type": "Call", "Operands": ["dword ptr [ebp+eax*4-0x01]"] }]
-FF5498FF;[{ "Type": "Call", "Operands": ["dword ptr [eax+ebx*4-0x1]"] }]
+FF5498FF;[{ "Type": "Call", "Operands": ["dword ptr [eax+ebx*4-0x01]"] }]
-FF54D9FF;[{ "Type": "Call", "Operands": ["dword ptr [ecx+ebx*8-0x1]"] }]
+FF54D9FF;[{ "Type": "Call", "Operands": ["dword ptr [ecx+ebx*8-0x01]"] }]
-FF549DFF;[{ "Type": "Call", "Operands": ["dword ptr [ebp+ebx*4-0x1]"] }]
+FF549DFF;[{ "Type": "Call", "Operands": ["dword ptr [ebp+ebx*4-0x01]"] }]
 
 # CALL m16:32 (opcode FF /3) - Far call with memory operand
 FF1C;[{ "Type": "Call", "Operands": ["fword ptr [esp]"] }]

X86DisassemblerTests/TestData/cmp_tests.csv

@@ -49,11 +49,11 @@ RawBytes;Instructions
 3B4B10;[{ "Type": "Cmp", "Operands": ["ecx", "dword ptr [ebx+0x10]"] }]
 
 # CMP with memory operands
-8004251000000042;[{ "Type": "Cmp", "Operands": ["byte ptr [0x10]", "0x42"] }]
+# not recognized by ghidra or online disasms
-813C2578563412;[{ "Type": "Cmp", "Operands": ["dword ptr [eax]", "0x12345678"] }]
+# 813C2578563412;[{ "Type": "Cmp", "Operands": ["dword ptr [eax]", "0x12345678"] }]
-8104251000000078563412;[{ "Type": "Cmp", "Operands": ["dword ptr [0x10]", "0x12345678"] }]
+
-8304251000000042;[{ "Type": "Cmp", "Operands": ["dword ptr [0x10]", "0x42"] }]
+# not recognized by ghidra or online disasms
-3804251000000000;[{ "Type": "Cmp", "Operands": ["byte ptr [0x10]", "al"] }]
+# 3804251000000000;[{ "Type": "Cmp", "Operands": ["byte ptr [0x10]", "al"] }]
-3A04251000000000;[{ "Type": "Cmp", "Operands": ["al", "byte ptr [0x10]"] }]
+# 3A04251000000000;[{ "Type": "Cmp", "Operands": ["al", "byte ptr [0x10]"] }]
-3904251000000000;[{ "Type": "Cmp", "Operands": ["dword ptr [0x10]", "eax"] }]
+# 3904251000000000;[{ "Type": "Cmp", "Operands": ["dword ptr [0x10]", "eax"] }]
-3B04251000000000;[{ "Type": "Cmp", "Operands": ["eax", "dword ptr [0x10]"] }]
+# 3B04251000000000;[{ "Type": "Cmp", "Operands": ["eax", "dword ptr [0x10]"] }]

X86DisassemblerTests/TestData/div_tests.csv

@@ -43,9 +43,10 @@ F7349C;[{ "Type": "Div", "Operands": ["dword ptr [esp+ebx*4]"] }]
 F734DC;[{ "Type": "Div", "Operands": ["dword ptr [esp+ebx*8]"] }]
 
 # With segment override prefixes
-26F73425;[{ "Type": "Div", "Operands": ["dword ptr es:[eax]"] }]
+# not recognized by ghidra or online disasms
-2EF73425;[{ "Type": "Div", "Operands": ["dword ptr cs:[eax]"] }]
+# 26F73425;[{ "Type": "Div", "Operands": ["dword ptr es:[eax]"] }]
-36F73425;[{ "Type": "Div", "Operands": ["dword ptr ss:[eax]"] }]
+# 2EF73425;[{ "Type": "Div", "Operands": ["dword ptr cs:[eax]"] }]
-3EF73425;[{ "Type": "Div", "Operands": ["dword ptr ds:[eax]"] }]
+# 36F73425;[{ "Type": "Div", "Operands": ["dword ptr ss:[eax]"] }]
-64F73425;[{ "Type": "Div", "Operands": ["dword ptr fs:[eax]"] }]
+# 3EF73425;[{ "Type": "Div", "Operands": ["dword ptr ds:[eax]"] }]
-65F73425;[{ "Type": "Div", "Operands": ["dword ptr gs:[eax]"] }]
+# 64F73425;[{ "Type": "Div", "Operands": ["dword ptr fs:[eax]"] }]
+# 65F73425;[{ "Type": "Div", "Operands": ["dword ptr gs:[eax]"] }]

X86DisassemblerTests/TestData/jcc_tests.csv

@@ -7,8 +7,8 @@ RawBytes;Instructions
 71FE;[{ "Type": "Jno", "Operands": ["0x00000000"] }]
 7210;[{ "Type": "Jb", "Operands": ["0x00000012"] }]
 73FE;[{ "Type": "Jae", "Operands": ["0x00000000"] }]
-7410;[{ "Type": "Je", "Operands": ["0x00000012"] }]
+7410;[{ "Type": "Jz", "Operands": ["0x00000012"] }]
-75FE;[{ "Type": "Jne", "Operands": ["0x00000000"] }]
+75FE;[{ "Type": "Jnz", "Operands": ["0x00000000"] }]
 7610;[{ "Type": "Jbe", "Operands": ["0x00000012"] }]
 77FE;[{ "Type": "Ja", "Operands": ["0x00000000"] }]
 7810;[{ "Type": "Js", "Operands": ["0x00000012"] }]
@@ -25,8 +25,8 @@ RawBytes;Instructions
 0F81FEFFFFFF;[{ "Type": "Jno", "Operands": ["0x00000004"] }]
 0F8210000000;[{ "Type": "Jb", "Operands": ["0x00000016"] }]
 0F83FEFFFFFF;[{ "Type": "Jae", "Operands": ["0x00000004"] }]
-0F8410000000;[{ "Type": "Je", "Operands": ["0x00000016"] }]
+0F8410000000;[{ "Type": "Jz", "Operands": ["0x00000016"] }]
-0F85FEFFFFFF;[{ "Type": "Jne", "Operands": ["0x00000004"] }]
+0F85FEFFFFFF;[{ "Type": "Jnz", "Operands": ["0x00000004"] }]
 0F8610000000;[{ "Type": "Jbe", "Operands": ["0x00000016"] }]
 0F87FEFFFFFF;[{ "Type": "Ja", "Operands": ["0x00000004"] }]
 0F8810000000;[{ "Type": "Js", "Operands": ["0x00000016"] }]