On 18 May 2009 at 19:57, Christophe Prud'homme wrote:
| Adam
|
| here is what I get at link time with the current (3.0) petsc version
|
| /usr/lib/libpetsc.a(sprng.o): In function
| `PetscRandomGetValueImaginary_Sprng':
| (.text+0xe1): undefined reference to `get_rn_dbl_simple_mpi'
| /usr/lib/libpetsc.a(sprng.o): In function
| `PetscRandomGetValueImaginary_Sprng':
| (.text+0xfa): undefined reference to `get_rn_dbl_simple_mpi'
| /usr/lib/libpetsc.a(sprng.o): In function `PetscRandomGetValueReal_Sprng':
| (.text+0x142): undefined reference to `get_rn_dbl_simple_mpi'
| /usr/lib/libpetsc.a(sprng.o): In function `PetscRandomGetValueReal_Sprng':
| (.text+0x15b): undefined reference to `get_rn_dbl_simple_mpi'
| /usr/lib/libpetsc.a(sprng.o): In function `PetscRandomGetValue_Sprng':
| (.text+0x1a3): undefined reference to `get_rn_dbl_simple_mpi'
| /usr/lib/libpetsc.a(sprng.o):(.text+0x1bc): more undefined references to
| `get_rn_dbl_simple_mpi' follow
| /usr/lib/libpetsc.a(sprng.o): In function `PetscRandomSeed_Sprng':
| (.text+0x1ef): undefined reference to `init_rng_simple_mpi'
|
| Given Dirk's comment regarding sprng, perhaps we should disable it for now ?
So is the following a concise summary
sprng2 available in debian, and used by r-cran-rsprng
sprng4 _not yet_ available in debian, but possibly useful
for PetSc ?
I.e. no overlap? As I understand the sprong website, versions 2 and 4 are
not plug-and-play. So if you guys have use of for sprng _version 4_ you may
just have to package it, no? I could then try to port the R package
forward and we could all get rid of the old sprng2
Nope. Just ran rdepends on it and there is another user of the v2 api:
e...@ron:~$ apt-cache rdepends libsprng2
libsprng2
Reverse Depends:
tree-puzzle
tree-ppuzzle
r-cran-rsprng
libsprng2-doc
libsprng2-dev
e...@ron:~$
Dirk.
|
| Best regards
| C.
| On Monday 18 May 2009 19:18:49 Adam C Powell IV wrote:
| > Hi Christian,
| >
| > AFAICT, PETSc is finding make_new_seed_mpi, and deciding to build with
| > it, right? Oh, but that's one of the very few mpi functions in there, I
| > see.
| >
| > I'd be happy to disable sprng support in PETSc if it's causing problems.
| > Right now I'm chasing down another problem causing grave bug 529303 (and
| > possibly 529162 in a reverse-depend), so I will probably upload a new
| > version soon.
| >
| > -Adam
| >
| > On Mon, 2009-05-18 at 18:07 +0200, Christophe Prud'homme wrote:
| > > Dirk
| > >
| > > I am not really interested (at the moment) but Petsc 3.0 in Debian is
| > > compiled with sprng support (I CC Adam Powell the maintainer of Petsc)
| > > but Petsc cannot find the proper MPI sprng routine.
| > >
| > > Adam
| > > should we disable sprng support in Petsc ? with this Petsc 3.0 is
| > > still broken in Debian
| > >
| > > Best regards
| > > C.
| > >
| > > On Monday 18 May 2009 16:21:29 Dirk Eddelbuettel wrote:
| > > > On 18 May 2009 at 15:50, Christophe Prud'homme wrote:
| > > > | Package: sprng
| > > > | Severity: wishlist
| > > > |
| > > > | --- Please enter the report below this line. ---
| > > > | Hi Dirk
| > > > |
| > > > | it seems there exists a sprng4 (which uses the
| > >
| > > autoconf/automake/libtool)
| > >
| > > > | and in debian only sprng2 is available.
| > > > | Moreover it seems that sprng has support for MPI, would it be
| > >
| > > possible to
| > >
| > > > | add mpi support ?
| > > > |
| > > > | Unfortunately sprng has not been updated wrt recent version of g++
| > >
| > > (4.3
| > >
| > > > | and 4.4)
| > > >
| > > > sprng is, politely speaking, a mess.
| > > >
| > > > I forced my way around sprng2 with some help from others (to make it
| > >
| > > create
| > >
| > > > a shared library) as I needed a distributed random number generator
| > >
| > > for use
| > >
| > > > by R with the snow (r-cran-snow) framework for distributed
| > >
| > > computing. I
| > >
| > > > never used sprng as a library.
| > > >
| > > > I am aware that there are newer versions, and I have at times tried,
| > >
| > > and
| > >
| > > > failed, to package them. Would you like to tackle this jointly?
| > > >
| > > > Otherwise, from R's side there is another parallel rng provided by
| > > > http://cran.r-project.org/web/packages/rlecuyer/index.html
| > > >
| > > > Pierre L'Ecuyer is quite well known for RNG work, maybe his code
| > >
| > > would
| > >
| > > > suitable for a general purpose library too ? I have been thinking
| > >
| > > about
| > >
| > > > packaging this and the retiring r-cran-rsprng and the sprng package.
| > > >
| > > > Thought or comments?
| > > >
| > > > Dirk
| > > >
| > > > | Best regards
| > > > | C.
| > > > |
| > > > | --- System information. ---
| > > > | Architecture: amd64
| > > > | Kernel: Linux 2.6.29-2-amd64
| > > > |
| > > > | Debian Release: squeeze/sid
| > > > | 500 unstable ftp.fr.debian.org
| > > > | 500 stable volatile.debian.org
| > > > | 500 stable security.debian.org
| > > > | 500 stable ftp.fr.debian.org
| > > > |
| > > > | --- Package information. ---
| > > > | Package's Depends field is empty.
| > > > |
| > > > | Package's Recommends field is empty.
| > > > |
| > > > | Package's Suggests field is empty.
|
| --
| Christophe Prud'homme
| Universit de Grenoble [email protected]
| LJK - Room 55 Tel: +33476635497
| 51, rue des Math matiques Fax: +33476631263
| BP53 38041 Grenoble Cedex 9
| <http://ljk.imag.fr/membres/Christophe.Prudhomme/>
|
--
Three out of two people have difficulties with fractions.
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
