[Python-Dev] Why is Bytecode the way it is?

[Guido van Rossum]

> While recently goofing around with the bytecode, I thought of doing 
> something like this:
> 
>   case LOAD_CONST:
>       x = GETITEM(consts, oparg);
>       Py_INCREF(x);
> +     if (*next_instr == RETURN_VAL) {
> +         retval = x;
> +         why = WHY_RETURN;
> +         goto fast_block_end;
> +     }
>       PUSH(x);
>       goto fast_next_opcode;
> 
> This would skip the stack and a trip through the loop without changing 
> the parser or the bytecode,  and with a minimal amount of added code or 
> overhead. This could (of course) be applied to other opcodes, too.

At the cost of an extra pointer dereference and jump, and usually for
naught (the uses of LOAD_CONST not in a return statement must
certainly vastly outnumber those in return statements).

Have you timed this?  I doubt you can measure a consistent performance
increase across platforms.

--Guido van Rossum (home page: http://www.python.org/~guido/)