[Ecls-list] ECL shootout

[Gabriel Dos Reis]

On Tue, Jun 30, 2009 at 12:36 PM, Juan Jose
Garcia-Ripoll<juanjose.garciaripoll at googlemail.com> wrote:
> 2009/6/30 Gabriel Dos Reis <gdr at integrable-solutions.net>:
>> Definitely -- however, the documentation of the base code is still shallow.
>
> Sure, I do not expect you to go and change the whole compiler.

Many of the problems I've reported in the past involved changing the compiler.
And I suspect, more along those lines will come.

> As I
> said, there are already a lot of things to improve in the base
> library.

OK.  Le me say more about uses of Lisp in OpenAxiom.
There are heavy uses of generic sequence functions (map, elt, etc.)
As we've discussed in the past, CL library spec is not very useful
for OpenAxiom.  For instance, we want IEEE-756 floats, we want
portable semantics for pathnames (much closer to POSIX than
what CL spec offers).  We do use symbols (interning, hashing, etc).
and string catenation.  In summary, we use only a tiny subset of
CL, and that small core is primarily used as a virtual machine.
You would be surprised to see however primitive the generated Lisp
looks like :-)  Ah, we CONS a lot.

I'm not claiming that use is typical of performance problems in ECL.  I'm
just offering data points (e.g. not artificially made up.)

-- Gaby