Re: Just-in-Time Predereferencing Patch

On Tue, Dec 11, 2001 at 10:09:07AM -0500, Gregor N. Purdy wrote:
> -        runops_generic(core, interpreter, pc);
> +        if ((interpreter->flags & PARROT_PREDEREF_FLAG) != 0) {
> +          size_t offset = pc - (opcode_t *)interpreter->code->byte_code;
> +
> +          if (!interpreter->prederef_code) {
> +            interpreter->prederef_code = (void **)calloc(interpreter->code->byte_code_size, sizeof(void *));
> +          }
> +
> +          runops_prederef(interpreter, pc, interpreter->prederef_code + offset);
> +        }
> +        else {
> +          runops_generic(core, interpreter, pc);
> +        }
> +    }
> +
> +    if (interpreter->prederef_code) {
> +      free(interpreter->prederef_code);
> +      interpreter->prederef_code = NULL;
>  

I haven't tried this, so feel free to shoot me down, but is your 22%
speed-up based on an unpatched Parrot, or a patched Parrot without the
-P flag? I'm trying to work out whether or not this patch actually slows
down the general case, which wouldn't be pleasant.

-- 
Timesharing just doesn't work. -K. Thompson, 1982.

• Just-in-Time Predereferencing Patch by Gregor N. Purdy
• Re: Just-in-Time Predereferencing Patch by Simon Cozens
• Re: Just-in-Time Predereferencing Patch by Gregor N. Purdy