On 2 Jan 98 at 19:45, Laurent Mirabito wrote:
> Hi again,
> Finally it works fine if I declare a dummy class Tlandau and in it
> only the denlan function. It's nearly as fast as PAW but needs a lot
> of work to implement. Is it possible that in one of the next
> releases of ROOT basic mathematical functions (at least the ones
> that are accessible in PAW) of the cernlib will be implement?
In theory it would be nice to create a separate DLL calling
kernnum.lib and mathlib.lib (for example based on CFortran.h and
kernlib.h + mathlib.h). One needs no translation from Fortran to C++
at all this way.
This approach has one major restriction. The CERNLIB Fortran
subroutines have a habit to communicate each other via common blocks
with one and same name but with the different length. This "habit"
doesn't allow (for some platforms) split CERNLIB by share/dynamic
libraries. All of them must be merged into the single large one (at
least under Windows), let's say "mathlib" and "kernnum". (otherwise
each DLL/share libs will get its own COPY of those COMMON blocks in
the main memory)
From another hand it is nothing to do with ROOT itself so may be
someone from RootTalk is ready to contribute. (or may be one can
demand CERNLIB office itself to produce these DLLs/Share libs ???)
I don't think we should overload the ROOT team with such sort of
"routine" work. It can be done by student. They have been making a
lot of them. If fact the main question here would be "Who will
support / maintain these "CERNLIB" libraries for ROOT ? (for dozen
platforms)". I suppose the maintaining ROOT/C++ version of CENRLIB is
as complicate as those from the original Fortran/C isn't ?
Anyway sometimes ago I was warned by CERNLIB office (by another
non-ROOT reason though) nobody can port and re-destribute the CERNLIB
code without the CERNLIB office official permission.
With my best regards,
Valery
==
=================================================================
Dr. Valeri Faine (Fine)
------------------- Phone: +1 516 344 1043
Brookhaven National Laboratory FAX : +1 516 344 4206
Bldg. 510A /STAR mailto:fine@rsgi01.rhic.bnl.gov
Upton, New York, 11973-5000 http://nicewww.cern.ch/~fine
USA
Dr. Valery Fine Telex : 911621 dubna su
-----------
LCTA/Joint Inst.for Nuclear Res. Phone : +7 09621 6 40 80
141980 Dubna, Moscow region Fax : +7 09621 6 51 45
Russia mailto:fine@main1.jinr.dubna.su
This archive was generated by hypermail 2b29 : Tue Jan 04 2000 - 00:34:29 MET