[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Gcl-devel] Re: Misc. re: 2.6.7

From: Robert Dodier
Subject: [Gcl-devel] Re: Misc. re: 2.6.7
Date: Sat, 26 Aug 2006 09:38:42 -0600

Hello Camm, you wrote:

si::newline is a funtion which currently appears unused in the
debugging code.

OK, thanks for the info. I resolved the Maxima problem
(I believe) by not calling si::newline (and calling a local
version instead, as with other Lisp implementations).

The large number segfault issue has also been resolved in 2.6.8pre,
though the division takes essentially forever.  In 2.7.0, we use gmp's
gcd for bignums, so your example finishes in seconds.  Given the time
2.7.0 is taking, perhaps it is worth migrating this one feature back
into 2.6.8.  Comments most welcome.

Well, operations with numbers like 255^(2^22) are fairly
rare, so I don't think it is worth the trouble to backport
the patch from 2.7.0.

Regarding the windows gazonk files issue, as I do not have access to
such a machine, could one please check the variable
compiler::*keep-gaz*?  These files are supposed to be deleted
automatically unless this is set.

In Maxima / GCL 2.6.7 on Linux, I find
compiler::*keep-gaz* is nil. So I guess that's OK.
I don't have a Windows box at hand, maybe someone
else can check that.

I notice the GCL version in the bug report is 2.6.5.
Was *keep-gaz* introduced after that?

The *read-base* error I've just fixed in 2.6.8pre by
backporting a function from cvs head.


Are these all of our issues outstanding?

Well, the one other Maxima-related issue that I can see
(after scanning the GCL bug list) is bug #14367,
dribble file has newlines in wrong places.

Thanks a lot for all your hard work!

Robert Dodier

reply via email to

[Prev in Thread] Current Thread [Next in Thread]