I found the bug in otfprofile.
When ompi/contrib/vt/vt/extlib/otf/tools/otfprofile/otfprofile.cpp is
compiled with the PGI C++ compiler, two "expected an identifier"
errors occur:
"/opt/pgi/linux86-64/10.3/include/omp.h", line 41: error: expected an
identifier
extern int omp_get_
My head hurts from working on this! I just realized is for
OpenMP, not OpenMPI. So, of course the PGI is used. I still
don't know why otfprofile is failing, but at least that explains why
OpenMPI-1.5rc5 has no .
Sorry for the noise.
Larry Baker
US Geological Survey
650-329-5608
ba...@
Hi
When I run the attached program with the following arguments, the size of MPI
processes keep increasing alarmingly (I saw that the size grew from 18M to 12G
in under 10 minutes) making me suspect that there is a major memory leak:
mpirun -am ft-enable-cr --mca coll basic -np 2
If I run thi
Ah, I see, I misunderstood how you organized things. Never mind :).
Brian
On Aug 31, 2010, at 2:48 PM, Ralph Castain wrote:
> I believe it is there, but correct me if not. All of the component
> configure.m4s depend on a test for sockaddr_in, which I believe RS will fail
> - yes?
>
> On Tue
I should also have noted that:
1. all of the opal_ifxxx functions still have their "no-op" versions when
struct sockaddr_in is not present.
2. we still "lazy discover" interfaces, so the new opal_if framework is
-not- opened unless you call one of the opal_ifxxx functions and get a
non-no-op vers
I believe it is there, but correct me if not. All of the component
configure.m4s depend on a test for sockaddr_in, which I believe RS will fail
- yes?
On Tue, Aug 31, 2010 at 2:42 PM, Barrett, Brian W wrote:
> Ralph -
>
> My Red Storm dev cluster isn't working right now, but I didn't see an
> opt
Ralph -
My Red Storm dev cluster isn't working right now, but I didn't see an option
for no if support at all. Did I miss it, or are we going to have to add it
later?
Thanks,
Brian
On Aug 31, 2010, at 2:28 PM, Ralph Castain wrote:
> Per a discussion on the devel telecon a week or two ago, I
Per a discussion on the devel telecon a week or two ago, I have refactored
the opal/util/if.c code into a static framework (ala the installdirs
implementation) so that the various cases are dealt with in configure.m4
instead of a mass of interwoven #if-#else clauses. Jeff will undoubtedly do
some c
I created a new ticket for this issue. It is
https://svn.open-mpi.org/trac/ompi/ticket/2560
Please go there for future details on this issue. But to quickly
summarize, I would like to
go with George's first recommendation that uses the "addl" command as
the "xaddl"
creates a second set of p
The source of this problem is configure does not properly set up the
LIBS variable in 1.5rc5 (compared to 1.4.2):
[root@hydra src]# diff openmpi-{1.4.2,1.5rc5}/config.log | grep -w
LIBS=
< LIBS='-lnsl -lutil -lpthread'
> LIBS='-lnsl -lutil'
Larry Baker
US Geological Survey
650-329-5608
b
The make of OpenMPI 1.5rc5 fails for PGI 10.3 in otfprofile:
Making all in otfprofile
make[9]: Entering directory `/usr/local/src/openmpi-1.5rc5/ompi/
contrib/vt/vt/extlib/otf/tools/otfprofile'
CXXotfprofile-otfprofile.o
"/opt/pgi/linux86-64/10.3/include/omp.h", line 41: error: expected
Yes, they are in a bitbucket repo:
https://bitbucket.org/rhc/ompi-agen
On Aug 31, 2010, at 9:08 AM, Shamis, Pavel wrote:
> Jeff,
>
> Is the autogen changes are public available? I would like to see the code.
>
> Thanks.
>
>
>
> On Aug 16, 2010, at 10:55 AM, Jeff Squyres wrote:
>
>> I just
Jeff,
Is the autogen changes are public available? I would like to see the code.
Thanks.
On Aug 16, 2010, at 10:55 AM, Jeff Squyres wrote:
> I just wanted to give the community a heads up that Ralph, Brian, and I are
> revamping autogen in a Mercurial branch. I don't know the exact timeline
13 matches
Mail list logo
