[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: execution speed in *oct files
From: |
John W. Eaton |
Subject: |
Re: execution speed in *oct files |
Date: |
Tue, 8 Jun 1999 12:17:01 -0500 (CDT) |
On 8-Jun-1999, Thomas Walter <address@hidden> wrote:
| I think some of those reading this mailing list know about 'GSL', the
| 'GNU scientific library'. They encoded the blas in C if I understand
| the code right. Have a large number of random number generators and
| distribution generators.
|
| Perhaps anybody with more knewledge about the needs for octave may
| have a look at it. The current version of 'gsl' as tar archive can be
| found at
| ftp://alpha.gnu.org/gnu/gsl-0.4.1.tar.gz
The TODO file for this library says
These are the public domain packages we should convert to C and put
under the GPL:
fftpack -- mostly done (sine and cosine transforms still to do, want
to add 2-d ffts, N-d ffts, convolutions)
quadpack -- in progress (Brian)
specfun -- in progress (Jerry)
ranlib -- add the stuff we don't have
minpack --
odepack --
blas --
starpac --
dassl --
eispack --
lapack --
linpack --
any others ???
I do understand that desire to have a better interface to these
subroutines (that's why I decided to wrap them in C++ classes) but I
don't fully understand the motivation for converting the original
Fortran code to C. My reasoning for leaving the original code mostly
alone was along the lines of `if it ain't broke, don't fix it'.
I'm assuming that these translations are done at least partly, if not
completely, by hand, so if someone does make a bug fix or release a
new version of any of these packages (although relatively infrequent,
it does happen -- daspk is a newer version of dassl, for example) then
you have to work the fixes back into your translated version instead
of just updating the Fortran sources. To me, that seems like a lot of
unnecessary work and also maybe asking for some trouble.
I do understand that mixed-language programming can be somewhat
tricky, but if all you ever want to do is call Fortran from C or C++,
it is not really all that bad in practice. So why go to all the
trouble? A quick look at the GSL docs and sources didn't turn up any
rationale or motivation for doing the translation. Can anyone shed
any light on this?
Thanks,
jwe
---------------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL. To ensure
that development continues, see www.che.wisc.edu/octave/giftform.html
Instructions for unsubscribing: www.che.wisc.edu/octave/archive.html
---------------------------------------------------------------------
- execution speed in *oct files, (continued)
Re: execution speed in *oct files, A. Scottedward Hodel, 1999/06/07
Re: execution speed in *oct files, A+A Adler, 1999/06/07
RE: execution speed in *oct files, Van den Eynde Gert, 1999/06/09