Interpreter POC

This is a poc of an stack based interpreter a bit like java with a custom. The assembler can take source files as input and will output a byc file as it's output.

Description

This is a 64bit stack machine, still figuring out if it'll be Risc or Cisc.

Assembler

The syntax to write for this assembler is fairly simple(lazy), whitespaces are ignore, comments too:

;Don't try executing this code, I'm just showing off the syntax
start: 		;Labels on single lines like this have to finish with a colon

	push 0x59	;supports hexadecimal
    push 932	;supports decimal
    add
    bz end		
    dup
    bnz start
end ret		;Labels preceding an instruction this don't need a colon

Virtual Machine

Usage

• "-ra" switch is to define the size of the return address stack(as to be decimal)
• "-stk" switch is to define the size of the general stack(as to be decimal) Example: "./vm -ra 38 -stk 40 "

Code

Self explanatory, executable bytecode resides there

Stacks

• Data stack: this stack is used for data manipulation
• Return address stack: this stack saves the return addresses

Instructions

9 bytes instructions. Only 23 supported for the moment ('i' means there's an immediate):

Other

• Syscall: currently unimplemented
• Vmentry: exits the vm

Stack operations

• Push: push an immediate on the stack
• Pop: pops a value off the stack
• Ldi: loads i variable on the top of the stack
• Sti: stores value on the top of the stack to i variable
• Dup: duplicates the value on top of the stack

Arithmetic

• Add: adds two values on top of the stack
• Sub: subs two values on top of the stack
• Addi: adds the value on top of the stack with the operand
• Subi: subs the value on top of the stack with the operand
• Mul: mul two values on top of the stack
• Muli: multiply the value on top of the stack with the operand

Logical

• Xor: xors two values on top of the stack
• Or: ors two values on top of the stack
• And: ands two values on top of the stack
• Not: nots the value on top of the stack
• Shl: shifts the value on top of the stack by the second value on top of the stack to the left
• Shr: shifts the value on top of the stack by the second value on top of the stack to the right

Branch

• Ret: return to the return address
• BL: branch and links
• Bz: subs the two values on top of the stack and branch if zero branch, the offset as to be specified
• Bnz: subs the two values on top of the stack and branch if not zero, the offset as to be specified

Registers

There are a few registers, they're not general purpose they can not be used for calculation or whatever

• Sp: the stack pointer that keeps track of the top of the stack
• Bp: the base register that keeps track of the base of a stack frame for each functions
• Pc: program counter, it points to the executed instruction
• Ra: just like the Sp but for the stack for the return addresses

Issues

No proper error handling, if you write bad written code the vm might segfault or whatever, except all kind of exceptions

Advantages

• :3

Disadvantages

• Size: please do not talk to me about the size of the bytecode