Re: [Tinycc-devel] Generated code optimization?

[Anton Ertl]

=?ISO-8859-1?Q?Zolt=E1n_L=F6rincz?= wrote:
> 
> Hello Anton, 
> thank you very much for your help. 
> 
> I looked into the home page of VMGen. 
> Is it possible to generate only the interpreter without rewriting the
> code generator ?

Should mostly be possible.

> Can you tell me how big would be (around) the
> generated interpreter if i have a register based machine with 300-400
> opcodes?

>From my experience, the interpreter engine would be around 16KB.

> My problem is that the interpreter has to be as small as possible. I
> think if VMGen generates the interpreter based on advanced techniques
> it means that it will generate bigger code.
> The interpreter will run on an embedded processor (MK68). This
> processor is slow and the system has small memory which has to be
> managed efficiently.

Ok, Vmgen was not designed for space efficiency.  The interpreter will
not be larger than others, but it does not employ techniques for
making the interpreted code smaller: by default, opcodes and their
arguments take a machine word each, but we should be able to fix that,
too (but at some cost in run-time).

Static superinstructions would cost more interpreter space, but save
on interpreted-program space.

- anton