Re: Lisp vs. Java vs. C++ speed comparison time? [LONG]

Greetings!

Valery Pipin <pip@iszf.irk.ru> writes:

> On Friday 09 July 2004 01:04, Camm Maguire wrote:
> > Greetings!
> >
> > "Vadim V. Zhytnikov" <vvzhy@mail.ru> writes:
> > > Valery Pipin writes:
> > > > On Saturday 03 July 2004 01:42, Camm Maguire wrote:
> > > > <snipped>
> 
> > > > Is that the reason why  I do not observe the memory leakage with maxima
> > > > computations  that done with cmucl implementation. In opposite, the gcl
> > > > implementation can consume so much memory that I force to restart the
> > > > x-session or even to reboot PC.
> > >
> > > I don't think this is memory leak.  Gcl memory layout is less
> > > compact compared to one of cmucl. The same computation on GCL
> > > require more RAM and it starts swapping earlier.
> >
> > This is an interesting observation.  If you can quantify this with
> > some reproducible examples, we might be able to take a look at it.
> >
> > Take care,
> 
> I don't find problems like that "memory leakage" with the new gcl-2.6.3.

Yes, we know of no leaks -- of course please let us know if you find one.

> Instead gcl-maxima simply stops (with "bind-stack-overflow")  when encounter 
> the piece of code needed the extensive memory usage. The example is given by 

Try (setq si::*multiply-stacks* 2) or 4 or whatever at the toplevel.

> http://www.iszf.irk.ru/~pip/maxima/test/alm_dr.mc. To test it you need other 
> files in http://www.iszf.irk.ru/~pip/maxima/test like backrh.mac,ruleh.mac.
> There are  the correct versions of coeflist(format).lisp  from contrib/format
> and lc2kdt.mac from tensor.
> The correct example obtained with cmucl is given by alp_dr.ts( for texmacs).
> 

It would be helpful for us to collate any memory related (or other for that
matter) problems in gcl vis a vis other systems to aid in development
going forward.  We already know that we have a three word cons, which
could likely be compacted to two words with considerable effort.
Likewise, stack and heap are statically defined in various ways -- the
max heap at compile time, the max stack at toplevel.  Alternate and
preferable ways of handling these items, if suggested, can be placed
on our todo list for future releases.

Take care,

>    
> _____________
> best regards,
> Valery
> 
> 
> 

-- 
Camm Maguire			     			camm@enhanced.com
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah