On May 9, 2007, at 1:37 AM, Or Gerlitz wrote:
Doing a bit of zoom out from the how to make ofed's udapl work for
ompi thread, my thinking is that the ompi udapl btl enablement is
actually only the first step, where for production/longterm/etc you
want to have an rdmacm btl.
I think this
On Wed, 2007-05-09 at 08:37 +0300, Or Gerlitz wrote:
Andrew Friedley wrote:
Jeff Squyres wrote:
FWIW, yes, adding RDMA CM support has actually been on my to-do list
for a while, but it keeps getting bumped by higher priority items.
It would be *much* better if some iWARP companies got
On May 9, 2007, at 10:30 AM, Steve Wise wrote:
Agreed. enabling udapl will get OMPI over iwarp immediately (and
hopefully in ofed-1.2). Post ofed-1.2, I think OMPI _should_ create a
rdma-cm btl. That's the plan...
Yes and no. Please see my other reply about an rdma cm BTL...
--
Jeff
In the FYI category...
There was discussion about the udpal BTL over OFED today on the
weekly developer teleconference (per my earlier post, a user is
reporting that it doesn't work). Andrew Friedley is going to work
with the Sun developers -- he thinks he might know where the problem
Well I've tried OMPI on ofed-1.2 udapl today and it doesn't work. I'm
debugging now.
Here's part of the problem (from ompi/btl/udapl/btl_udapl.c):
/* TODO - big bad evil hack! */
/* uDAPL doesn't ever seem to keep track of ports with addresses. This
becomes a problem when
Chelsio's gonna pony up the resources to get this work done asap. Do
you have any thoughts on how we can collaborate on this project? I'm
familiar with mvapich, not ompi, so I need to go do some homework.
But
any pointers on the connection setup design for ompi would be great.
Steve Wise wrote:
1) OMPI shouldn't be stepping on the ia_address.
stongly agree
2) OFA udapl should probably be explicitly binding local cm_ids to port
zero.
current implementation uses port zero on ep_create and ia_open.
3) dat_ep_query() should be returning the correct port
On Tue, 2007-05-08 at 12:55 -0700, Arlin Davis wrote:
Steve Wise wrote:
1) OMPI shouldn't be stepping on the ia_address.
stongly agree
2) OFA udapl should probably be explicitly binding local cm_ids to port
zero.
current implementation uses port zero on ep_create and ia_open.
] On Behalf Of Steve Wise
Sent: Tuesday, May 08, 2007 12:52 PM
To: Andrew Friedley
Cc: [EMAIL PROTECTED]; Open MPI Developers; general; Asgeir Eiriksson
Subject: Re: [OMPI devel] [ofa-general] OpenMPI and RDMA-CM
Chelsio's gonna pony up the resources to get this work done asap.
Do you have any
Andrew Friedley wrote:
Jeff Squyres wrote:
FWIW, yes, adding RDMA CM support has actually been on my to-do list
for a while, but it keeps getting bumped by higher priority items.
It would be *much* better if some iWARP companies got involved in
Open MPI...
Hmm I'm interested. I've already
Also, there appears to be a DAPL BTL in OMPI. Is this BTL complete and
enabled for the ofed-1.2 udapl library?
Steve.
On Mon, 2007-05-07 at 17:09 -0500, Steve Wise wrote:
On Sat, 2007-04-28 at 16:20 -0400, Jeff Squyres wrote:
You'd probably be better asking this question on the Open MPI
On May 7, 2007, at 6:52 PM, Steve Wise wrote:
Also, there appears to be a DAPL BTL in OMPI. Is this BTL complete
and
enabled for the ofed-1.2 udapl library?
Yes, it is complete and is well-tested in Solaris.
It is not well tested in Linux/OFED (we've been concentrating on the
verbs
12 matches
Mail list logo
