On Tuesday, February 2, 2016, Brice Goglin wrote:
> Le 02/02/2016 15:21, Jeff Squyres (jsquyres) a écrit :
> > On Feb 2, 2016, at 9:00 AM, Dave Love > wrote:
> >> Now that MX support has been dropped, is there an alternative for fast
> Ethernet?
> > There are several options for low latency ethe
On Feb 2, 2016, at 12:15 PM, Number Cruncher
wrote:
>
> Thanks for the info. I'll probably go with insisting on libibverbs.
>
> Does seem a bit contrary to the very high modularity in OpenMPI that
> essentially a Cisco-specific module introduces a libmpi dependency where
> openib never did...
"Jeff Squyres (jsquyres)" writes:
> This functionality is there to overcome a bug in libibverbs (that
> prints a dire warning about Cisco usNIC devices not being supported).
It wasn't clear, but I meant there are also other libraries sucked in,
like hwloc and opensm, and that's not new in 1.10.
Thanks for the info. I'll probably go with insisting on libibverbs.
Does seem a bit contrary to the very high modularity in OpenMPI that
essentially a Cisco-specific module introduces a libmpi dependency where
openib never did
Simon
On 02/02/16 01:26, Gilles Gouaillardet wrote:
Simon,
Le 02/02/2016 15:21, Jeff Squyres (jsquyres) a écrit :
> On Feb 2, 2016, at 9:00 AM, Dave Love wrote:
>> Now that MX support has been dropped, is there an alternative for fast
>> Ethernet?
> There are several options for low latency ethernet, but they're all
> vendor-based solutions (e.g., my co
This functionality is there to overcome a bug in libibverbs (that prints a dire
warning about Cisco usNIC devices not being supported).
However, I can see how this additional linkage is undesirable. We can probably
flip the default on this component to not build by default -- but leave it
ther
Thanks Peter,
this is just a workaround for a bug we just identified, the fix will come
soon
Cheers,
Gilles
On Tuesday, February 2, 2016, Peter Wind wrote:
> That worked!
>
> i.e with the changed you proposed the code gives the right result.
>
> That was efficient work, thank you Gilles :)
>
On Feb 2, 2016, at 9:00 AM, Dave Love wrote:
>
> Now that MX support has been dropped, is there an alternative for fast
> Ethernet?
There are several options for low latency ethernet, but they're all
vendor-based solutions (e.g., my company's usNIC solution).
Note that MX support was dropped
Number Cruncher writes:
> Having compiled various recent Open MPI sources with the same
> "configure" options, I've noticed that the "libmpi.so" shared library
> from 1.10.1 now depends itself directly on
> libibverbs.so.1. Previously, 1.10.0 for example, only plugins such as
> mca_btl_openib.so
That worked!
i.e with the changed you proposed the code gives the right result.
That was efficient work, thank you Gilles :)
Best wishes,
Peter
- Original Message -
> Thanks Peter,
> that is quite unexpected ...
> let s try an other workaround, can you replace
> integer
Now that MX support has been dropped, is there an alternative for fast
Ethernet?
Thanks Peter,
that is quite unexpected ...
let s try an other workaround, can you replace
integer:: comm_group
with
integer:: comm_group, comm_tmp
and
call MPI_COMM_SPLIT(comm, irank*2/num_procs, irank, comm_group, ierr)
with
call MPI_COMM_SPLIT(comm, irank*2/num
Thanks Gilles,
I get the following output (I guess it is not what you wanted?).
Peter
$ mpirun --mca osc pt2pt -np 4 a.out
--
A requested component was not found, or was unable to be opened. This
means that this compon
Peter,
at first glance, your test program looks correct.
can you please try to run
mpirun --mca osc pt2pt -np 4 ...
I might have identified a bug with the sm osc component.
Cheers,
Gilles
On Tuesday, February 2, 2016, Peter Wind wrote:
> Enclosed is a short (< 100 lines) fortran code examp
Enclosed is a short (< 100 lines) fortran code example that uses shared memory.
It seems to me it behaves wrongly if openmpi is used.
Compiled with SGI/mpt , it gives the right result.
To fail, the code must be run on a single node.
It creates two groups of 2 processes each. Within each group mem
15 matches
Mail list logo
