* Lisandro Dalcin [2007-07-30 18:19:21]:
> On 7/30/07, George Bosilca wrote:
> > In the data-type section there is an advice to implementors that
> > state that a copy can simply increase the reference count if
> > applicable. So, we might want to apply the same logic here ...
> BTW, you just
Brian Barrett wrote:
> I think I have this fixed (using the $file_system_[foo] variables).
> Thanks for all the bug reports. I've committed to the trunk and am
> building a new nightly tarball right now. Can you try it to make
> sure I have all your issues fixed?
>
I reverted all my modif
Hi,
In ompi/mca/io/romio/romio/adio/Makefile.am there is an entry missing
for PVFS2.
Index: Makefile.am
===
--- Makefile.am (revision 9851)
+++ Makefile.am (working copy)
@@ -139,7 +139,7 @@
libadio_la_LIBADD = \
common/
nt it from building pvfs along with pvfs2.
It's not a good solution anyway, because $FILE_SYSTEM contains things
as "nfs ufs pvfs pvfs2"
Really, the correct fix is just to use $file_system_pvfs,
$file_system_pvfs2, ...
Greetings,
Dries
Dries Kimpe wrote:
>
> Sorry to
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Sorry to keep you busy,
but I found another problem with the Open MPI <-> ROMIO integration:
mca/io/romio/romio/Makefile.in:
# Open MPI: setup the AM_CONDITIONALs to build the different adio devices
AM_CONDITIONAL(BUILD_GRIDFTP, [test -n "`echo $
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
There is another small problem in that file:
- --with-io-romio-flags is not being passed either. This is needed to pass
--with-filesystems for
enabling ROMIO support for more than NFS and UFS.
Problem is the
io_romio_flags=
line clearing the
Hi,
When trying to build Open MPI (r9843) with romio configured with pfvs2
support,
I noticed that the configure from romio was failing because of missing
libraries.
Because the pvfs2 library was not in the standard search path, I needed to
specify custom
LDFLAGS and LIBS for the toplevel
Dries Kimpe wrote:
>
>
> The problem is that the OpenMPI group takes ROMIO as-is from an older
> release. Then they pass
> MPI_COMBINER_SUBARRAY as-is to ROMIO, which in that release it does not
> understand. Then ROMIO blows up.
>
>
> I'll try recompiling
