On Dec 6, 2006, at 3:09 PM, Scott Atchley wrote:
Brock and Galen,
We are willing to assist. Our best guess is that OMPI is using the
code in a way different than MPICH-GM does. One of our other
developers who is more comfortable with the GM API is looking into it.
We tried running with HPCC w
On Dec 5, 2006, at 12:32 PM, Rainer Keller wrote:
To check, You may also use:
ssh echo $SHELL
Make sure you quote this properly so that $SHELL is not evaluated on
the local node:
ssh 'echo $SHELL'
--
Jeff Squyres
Server Virtualization Business Unit
Cisco Systems
Not a lot of description I can provide. I'm trying to configure
under OSX 10.4, using the g77 fortran compiler.
Any hints will be most appreciated.
Tom Spraggins
t...@virginia.edu
ompi-output.tar.gz
Description: GNU Zip compressed data
That is wonderful, that fixes the observed problem for running with
OB1. Has a bug for this been filed to get RDMA working on macs?
The only working MPI lib is MPICH-GM as this problem happens with
LAM-7.1.3 also.
So on track for one bug.
Would the person working on the DR PML like m
Anthony,
MPE is built and working. The errors I saw were only related to rlog
and sample. Once
I disabled them there were no more errors.
Thanks for your help :-)
Ryan
On Dec 6, 2006, at 3:44 PM, Anthony Chan wrote:
On Wed, 6 Dec 2006, Ryan Thompson wrote:
Hi Anthony,
I made some p
Something is not clear for me in this discussion. Sometimes the
subject was the DR PML and sometimes the OB1 PML. In fact I'm
completely in the dark ... Which PML fails the HPCC test on MAC ?
When I look at the command line it look like it should be OB1 not DR ...
george.
On Dec 7, 2006
There were two issues here, one found the other. the OB1 works
just fine on OSX on PPC64. the DR PML does not work, there is no
output to STDOUT and the application while you can see the threads in
'top' no progress is ever made in running the application.
The original problem stems
On Dec 7, 2006, at 2:45 PM, Brock Palen wrote:
$ mpirun -np 4 -machinefile hosts -mca btl ^tcp -mca
btl_gm_min_rdma_size $((10*1024*1024)) ./hpcc.ompi.gm
and HPL passes. The problem seems to be in the RDMA fragmenting
code
on OSX. The boundary values at the edges of the fragments are not
On Dec 7, 2006, at 3:14 PM, George Bosilca wrote:
On Dec 7, 2006, at 2:45 PM, Brock Palen wrote:
$ mpirun -np 4 -machinefile hosts -mca btl ^tcp -mca
btl_gm_min_rdma_size $((10*1024*1024)) ./hpcc.ompi.gm
and HPL passes. The problem seems to be in the RDMA fragmenting
code
on OSX. The bound
George,
Using DR was suggested to see if it could find an error. The original
problem was using OB1, and HPL gave failed residuals. The hope was
that DR would pinpoint any problems. It did not and HPL did not
progress at all (the GM counters incremented, but no tests were
completed succes
This is due to a problem in (void *)->(uint64_t_ conversion in OMPI. The
following patch fixes the problem, as would an appropriate cast of pval, I
suspect. The problem is an inappropriate use of ompi_ptr_t. I would guess
that other uses of lval might be suspect also (such as in the Portals c
Thanks Jeff.
It turns out that all our IB blades are EM64T - it's just that some have
i686 OS's and some x86_64 OS's. So I think we'll move to all x86_64
installs on IB hosts. I guess if we make the OpenMPI a 32-bit build, and
link against 32-bit IB drivers (my interpretation of the release no
On Dec 7, 2006, at 5:04 PM, Aaron McDonough wrote:
It turns out that all our IB blades are EM64T - it's just that some
have
i686 OS's and some x86_64 OS's. So I think we'll move to all x86_64
installs on IB hosts. I guess if we make the OpenMPI a 32-bit
build, and
link against 32-bit IB dri
13 matches
Mail list logo
