[Debian-med-packaging] Please review biosquid [Was: ansgar at debian.org: Re: [Help] pvm: ftbfs with GCC-10]

[Andreas Tille]

Hi,

I tried to disable pvm in biosuid.  I think I need a short review.  The
build says:


libtool: compile:  gcc -DHAVE_CONFIG_H -I. -Wdate-time -D_FORTIFY_SOURCE=2 -g -O2 -fdebug-prefix-map=/build/biosquid-1.9g+cvs20050121=. -fstack-protector-strong -Wformat -Werror=format-security -c stopwatch.c  -fPIC -DPIC -o .libs/libsquid_la-stopwatch.o
stopwatch.c:58:10: fatal error: pvm3.h: No such file or directory
   58 | #include <pvm3.h>
      |          ^~~~~~~~

I wonder why this happens since SRE_ENABLE_PVM is not defined and
the code says:

#ifdef SRE_ENABLE_PVM
#include <pvm3.h>
#endif


Seems my C knowledge becomes a bit rusty. :-(

I'd love to disable pvm as simple fix here - extra bonus for replacing
it by MPI.

Kind regards

       Andreas.


[1] https://salsa.debian.org/med-team/biosquid

----- Forwarded message from Ansgar <ansgar at debian.org> -----

Date: Wed, 12 Aug 2020 10:59:20 +0200
From: Ansgar <ansgar at debian.org>
To: Andreas Tille <andreas at an3as.eu>
Cc: 957717 at bugs.debian.org, Debian Mentors List <debian-mentors at lists.debian.org>
Subject: Re: [Help] pvm: ftbfs with GCC-10

Andreas Tille writes:
> while I do not intend to maintain pvm personally some Debian Med package
> depend from it.  Thus I like to see bug #957717 fixed but I need help.

Do people still use PVM for parallel computing?  I thought it was
replaced by MPI decades ago.

If use of PVM is optional, I would consider disabling use of PVM in
reverse dependencies and just remove the unmaintained PVM package from
Debian.

Ansgar


----- End forwarded message -----

-- 
http://fam-tille.de