Bug#807666: nmu: mpich_3.1-6

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Bug#807666: nmu: mpich_3.1-6

Samuel Thibault-8
Package: release.debian.org
Severity: normal
User: [hidden email]
Usertags: binnmu

Hello,

Compiling c++ programs against mpich currently fails in sid:

In file included from /usr/include/mpich/mpi.h:2175:0,
                 from test.c:1:
/usr/include/mpich/mpicxx.h:22:4: error: #error 'Please use the same version of GCC and g++ for compiling MPICH and user MPI programs'
 #  error 'Please use the same version of GCC and g++ for compiling MPICH and user MPI programs'

Apparently mpich got built with 5.2, and sid currently has 5.3, thus the
failure.  Rebuilding mpich with current gcc 5.3 should be fixing it.

Samuel

nmu mpich_3.1-6 . ANY . unstable . -m "Rebuild against g++ 5.3"

-- System Information:
Debian Release: stretch/sid
  APT prefers testing
  APT policy: (990, 'testing'), (500, 'oldoldstable'), (500, 'buildd-unstable'), (500, 'unstable'), (500, 'stable'), (500, 'oldstable'), (1, 'buildd-experimental'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.3.0 (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

--
Samuel
 Créer une hiérarchie supplementaire pour remedier à un problème (?) de
 dispersion est d'une logique digne des Shadocks.
 * BT in: Guide du Cabaliste Usenet - La Cabale vote oui (les Shadocks aussi) *

Reply | Threaded
Open this post in threaded view
|

Bug#807666: nmu: mpich_3.1-6

Matthias Klose
On 11.12.2015 14:50, Samuel Thibault wrote:

> Package: release.debian.org
> Severity: normal
> User: [hidden email]
> Usertags: binnmu
>
> Hello,
>
> Compiling c++ programs against mpich currently fails in sid:
>
> In file included from /usr/include/mpich/mpi.h:2175:0,
>                   from test.c:1:
> /usr/include/mpich/mpicxx.h:22:4: error: #error 'Please use the same version of GCC and g++ for compiling MPICH and user MPI programs'
>   #  error 'Please use the same version of GCC and g++ for compiling MPICH and user MPI programs'
>
> Apparently mpich got built with 5.2, and sid currently has 5.3, thus the
> failure.  Rebuilding mpich with current gcc 5.3 should be fixing it.
>
> Samuel
>
> nmu mpich_3.1-6 . ANY . unstable . -m "Rebuild against g++ 5.3"

better fix mpich to not give this error on minor releases. This was correct
before GCC 5, not for GCC 5 and newer versions.

Reply | Threaded
Open this post in threaded view
|

Bug#807666: nmu: mpich_3.1-6

Emilio Pozuelo Monfort-4
In reply to this post by Samuel Thibault-8
Control: reassign -1 src:mpich
Control: severity -1 serious
Control: retitle -1 mpich: overly restrictive GCC check

On 11/12/15 14:50, Samuel Thibault wrote:

> Package: release.debian.org
> Severity: normal
> User: [hidden email]
> Usertags: binnmu
>
> Hello,
>
> Compiling c++ programs against mpich currently fails in sid:
>
> In file included from /usr/include/mpich/mpi.h:2175:0,
>                  from test.c:1:
> /usr/include/mpich/mpicxx.h:22:4: error: #error 'Please use the same version of GCC and g++ for compiling MPICH and user MPI programs'
>  #  error 'Please use the same version of GCC and g++ for compiling MPICH and user MPI programs'
>
> Apparently mpich got built with 5.2, and sid currently has 5.3, thus the
> failure.  Rebuilding mpich with current gcc 5.3 should be fixing it.
>
> Samuel
>
> nmu mpich_3.1-6 . ANY . unstable . -m "Rebuild against g++ 5.3"

This sounds like a too restrictive check, especially now that a 5.2 -> 5.3 GCC
bump is similar to a 4.9.2 -> 4.9.3 one (i.e. a "micro" or bugfix update). I'm
reassigning so this gets fixed in mpich.

Cheers,
Emilio

Reply | Threaded
Open this post in threaded view
|

Bug#807666: reopen 807666, it should be fixed properly

Ana Guerrero Lopez
In reply to this post by Samuel Thibault-8
unarchive 807666
reopen 807666
notfixed 807666 mpich/3.2-1~exp1
found 807666 3.3-2
forwarded 807666 https://lists.mpich.org/pipermail/discuss/2019-March/011160.html
kthxbye

Hi,

This bug wasn't closed properly. While a rebuild of the package with
the new upload fixed the problem temporarily, this is a problem
that must be solved in mpich as Matthias and Emilio pointed out.

I have sent a patch to upstream that should fix the issue:
https://lists.mpich.org/pipermail/discuss/2019-March/011160.html

Cheers,
Ana