[Ecls-list] Status ECL + Maxima

[Juan Jose Garcia-Ripoll]

On Sun, Aug 22, 2010 at 12:13 AM, Robert Dodier <robert.dodier at gmail.com>wrote:

> I see those errors in rtest16 (and no others).
> I am working w/ Maxima and ECL from their respective
> CVS repos as of today, on Linux (Centos 4).
> If I'm not mistaken, those are recent in origin.
> I looked into it a few months ago; I concluded that the problem
> was something to do with special variables.
>

Ok, I made a tiny fix to the function that decides whether variables may be
changed or not, and now ECL/CVS passes all tests except for one

Error summary:
Error found in /Users/jjgarcia/build/maxima.new/tests/rtest14.mac, problem:
(144)
1 test failed out of 8,554 total tests.
real time : 300.540 secs
run time  : 294.800 secs
gc count  : 94 times
consed    : 1223353065 bytes

Incidentally, there is about a 10% speedup with respect to 10.4.1 on the
same platform (OS X)

Error summary:
Error found in /Users/jjgarcia/build/maxima.bak/tests/rtest14.mac, problem:
(144)
Errors found in /Users/jjgarcia/build/maxima.bak/tests/rtest16.mac,
problems:
(383 384)
3 tests failed out of 8,553 total tests.
real time : 347.305 secs
run time  : 335.393 secs
gc count  : 94 times
consed    : 1336525441 bytes

I would not have expected much due to the high level nature of the code in
Maxima and its pervasive use of special variables, but it is something.

Juanjo

-- 
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/ecl-devel/attachments/20100822/4a364303/attachment.html>