What happens to this list after tomorrow? (i.e., general@lists.openfabrics.org
) Will mails bounce?
The intent is that all mails to the general list should be sent to
the linux-rdma list instead, right?
--
Jeff Squyres
jsquy...@cisco.com
and infrastructure.
On Sep 29, 2009, at 12:24 PM, Jon Mason wrote:
On Tue, Sep 29, 2009 at 12:16:39PM -0400, Jeff Squyres wrote:
What happens to this list after tomorrow? (i.e.,
general@lists.openfabrics.org) Will mails bounce?
The intent is that all mails to the general list should be sent
comments/suggestions. Thanks.
-jeff
Jeff Squyres wrote:
What happens to this list after tomorrow? (i.e.,
general@lists.openfabrics.org
) Will mails bounce?
The intent is that all mails to the general list should be sent to
the linux-rdma list instead, right?
--
Jeff Squyres
jsquy
code that everyone was happy with:
- Roland likes it (and anticipated that the kernel community would be
receptive to)
- we like it
- performs correctly
Hope that helps.
--
Jeff Squyres
jsquy...@cisco.com
___
general mailing list
general
--
Jeff Squyres
jsquy...@cisco.com
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
will choose at run time whether to use ptmalloc2 or the
ummunotify stuff (i.e., the --enable-mca-no-build... step won't be
necessary when all is said and done; a default OMPI Linux build will
do the Right Things).
Thanks.
--
Jeff Squyres
jsquy...@cisco.com
, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
jsquy...@cisco.com
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http
on the interface design. Also thanks to Jeff Squyres
jsquy...@cisco.com for prototyping support for this in Open MPI,
which
helped find several bugs during development.
Signed-off-by: Roland Dreier rola...@cisco.com
---
drivers/char/Kconfig| 12 ++
drivers/char/Makefile |1
://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org
- there is a request from Sun to have bug fixes only
release since there is a critical bug that prevent Lustre to run over
OFED 1.4 and 1.4.1
3. RDMAoE
4. MPI in OFED: decision in XWG was not to change for 1.5 and bring it
as a BOF in SC09 with the customers
Tziporet
--
Jeff Squyres
Cisco
an OpenFabrics software stack mean -- is that specific to
a release series such as 1.4.x? Or does it effectively mean the
entire OF stack, regardless of version? I ask because the MPI change
is proposed for v1.5. Since 1.5 doesn't exist yet, is it covered by
14.1 or 14.2?
--
Jeff
.
My $0.02: self-created rules are only useful if they're followed. If
they're not followed, they should be changed or thrown out. :-)
-Original Message-
From: Jeff Squyres [mailto:jsquy...@cisco.com]
Sent: Tuesday, June 16, 2009 7:39 AM
To: Ryan, Jim
Cc: Tziporet Koren; Woodruff
, but some
clarification of the rules might be useful.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman
started, but I don't have a timeline
for delivery.
An obvious concern to me is having code submitted prior to a spec
being agreed
Jim
-Original Message-
From: ewg-boun...@lists.openfabrics.org [mailto:ewg-boun...@lists.openfabrics.org
] On Behalf Of Jeff Squyres
Sent: Tuesday, June
:
1. EWG needs to come into compliance with the OF bylaws
2. The bylaws need to be changed to match what EWG does (has done)
3. Throw the bylaws out (rules are useless if they don't guide what
you do)
4. Some combination of 1 and 2
My $0.02. I leave it to others to figure out what to do.
--
Jeff
implies that he feels the
same way. :-)
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo
OFED
would be insane to remove MPI from its testing/QA/release process.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit
dledf...@redhat.com
To: Jeff Squyres (jsquyres)
Cc: John Russo john.ru...@qlogic.com; EWG openfabrics-...@openib.org;
OpenFabrics General general@lists.openfabrics.org
Sent: Sun Jun 07 12:58:55 2009
Subject: Re: [ofa-general] RE: [ewg] RFC: Do we wish to take MPI out ofOFED?
On Fri, 2009-06-05
-general
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
that this is a minor point, but it is still
worth noting -- I've always thought it strange that people use MPI to
monitor and test their OF-based networks...)
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http
?
Is a process responsible for guaranteeing that it umn_unregister()s
everything before exiting, or will all pending registrations be
cleaned up/unregistered/whatever when a process exits?
--
Jeff Squyres
Cisco Systems
___
general mailing list
general
.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
, if libibverbs'
reg caching is optional, then the verbs-based app can choose to use it
or their own scheme).
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
... 0x3fff
can change even though it is still registered, then we're screwed --
we have no way of knowing that this is now invalid (Open MPI, at least
-- can't speak for others).
Is there a way to detect condition this in userspace?
--
Jeff Squyres
Cisco Systems
to 0.
See my prior mail for a more complex example of our cache's behavior.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit
Why are the RPM version numbers the same between rc5 and the current
1.4.1 nightlies?
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe
port_lmc: 0x00
I don't see any obvious errors occurring in syslog or dmesg.
What could cause this failure?
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi
[mailto:ewg-boun...@lists.openfabrics.org
] On Behalf Of Jeff Squyres
Sent: Wednesday, May 13, 2009 11:34 AM
To: OpenFabrics General; OpenFabrics EWG
Subject: [ewg] /dev/infiniband/rdma_cm not created
I'm running on rhel4u6 with the 1.4.1 nightly from last night and
sometimes /dev/infiniband/rdma_cm
On May 13, 2009, at 2:54 PM, Jeff Squyres wrote:
[11:51] svbu-mpi005:/etc/udev/rules.d % /sbin/lsmod | grep rdma
[11:51] svbu-mpi005:/etc/udev/rules.d %
What would cause it to not be loaded? I *assumed* (but didn't
check) that it is loaded as part of OFED's /etc/init.d/openibd
.ELsmp SMP gcc-3.4
[r...@svbu-mpi055 ~]#
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo
, 62 May 13 12:17 /dev/infiniband/rdma_cm
[r...@svbu-mpi055 ~]#
Something must be going wrong during the bootup. I'm unfortunately
several thousand miles from the server and don't have a serial
console. I guess I'll insert some initlog's in /etc/init.d/openibd...
--
Jeff Squyres
Cisco Systems
:18 PM, Jeff Squyres (jsquyres) wrote:
On May 13, 2009, at 3:12 PM, Woodruff, Robert J wrote:
Check to see if some other driver failed to load.
I think I have seen before that if another driver
fails to load, the start script bails out and
does not load the other drivers.
Perhaps try doing
On May 4, 2009, at 8:25 PM, Jeff Squyres (jsquyres) wrote:
It was suggested today that a teleconference to discuss these issues
might be much more useful (an hour-long teleconference can save a
week's worth of emails!). This will be a technical call to discuss
memory registration issues
the
new memory again.
MVAPICH: do you guys do similar things?
(I don't know if HP/Scali/Intel will comment on their registration
cache schemes)
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http
in libibverbs can be delayed since it can be added
after the API will the kernel is avilable
Fair enough.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo
and accounting.
=
I refrained from a specific new API proposal; let's argue over these
ideas first and see if we can come to consensus. If so, specific API
proposals can follow.
--
Jeff Squyres
Cisco Systems
Hello ,
Jeff Squyres invites you to attend this online meeting.
Topic: Verbs memory registration
Date: Monday, May 11, 2009
Time: 12:00 pm, Eastern Daylight Time (GMT -04:00, New York)
Meeting Number: 203 642 533
Meeting Password: verbs
Please click the link below to see more information
for it)
- R.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
of
sockets programmers to verbs if you effectively restrict large
messages to only be allocated/freed by the network layer (kinda
defeats the point of RDMA if you have to copy large messages, right?).
--
Jeff Squyres
Cisco Systems
___
general mailing
On Apr 30, 2009, at 5:25 PM, arkady kanevsky wrote:
are the MPI applications that are broken are the ones which is
malloc/free
instead of MPI_ALLOC calls?
Yes.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
space
notifier has been proposed or implemented in the kernel it has been a
disaster.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe
those instead of
MPI_ALLOC_MEM+copy).
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo
complex than common sockets usage.Verbs -- and the additional
baggage that it requires for performance, like registration caches and
memory allocation hooking -- does not currently meet this requirement.
--
Jeff Squyres
Cisco Systems
___
general
...@openfabrics.org
Cc: i...@lists.openfabrics.org; Paul Grun; Paul Gray; OFA Marketing
Working Group; Wayne Augsburger; Andy Grover; Richard Frank; asa...@voltaire.com
; Jeff Squyres; Mikkel Hagen;general@lists.openfabrics.org; Scott
Friedman; b...@voltaire.com; Sumanta Chatterjee; Roland Dreier
supporting Open MPI. :-\
People won't migrate, nor will main-line MPI benchmarks. Customers
want top performance out-of-the-box with their MPI (which is not
unreasonable). Users have used malloc() for 10+ years, and other
networks don't require the use of MPI_ALLOC_MEM.
--
Jeff Squyres
layered scenarios,
it's not easy to use these calls (e.g., if an MPI-enabled
computational library may re-use user-provided buffers because they're
so large, etc.).
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
. There are definitely code
paths in glibc where, even if you use the mallopt() hints, memory
*can* (will) be returned to the OS. This led Open MPI to change its
memory allocation / intercept scheme in 1.3.2. See:
http://www.open-mpi.org/community/lists/announce/2009/03/0029.php
--
Jeff Squyres
Cisco
this might just end up as new verbs
anyway -- not a new middleware library.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit
consistent by preventing the kernel from returning
new mmaps that overlap existing memory registrations.
Jason
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo
hooks.
Yuck.
It's memory management. And that belongs in the kernel.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http
need *something* that is safe and sane. Right
now, we don't have that.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit
is unacceptably stupid and I'm actually
amazed there's any resistance to fixing the problem.
I'm a little amazed that it's gone this long without being fixed (I
know I spoke about this exact issue at Sonoma 3 years ago!).
--
Jeff Squyres
Cisco Systems
...@lists.openfabrics.org; Paul Grun; Paul Gray; OFA Marketing
Working Group; Wayne Augsburger; Andy Grover; Richard Frank; asa...@voltaire.com
; Jeff Squyres; Mikkel Hagen;general@lists.openfabrics.org; Scott
Friedman; b...@voltaire.com; Sumanta Chatterjee; Roland Dreier
Subject: [mwg] Re: RDMA
(...) {
int *a = malloc(...);
MPI_Init(...);
MPI_Send(a, ...);
...
}
Re-reading your brief text; I'm wondering if I missed the zen of what
you're trying to suggest...? If I'm off the mark, can you explain
more? Thanks.
--
Jeff Squyres
Cisco Systems
your issue.
Sorry for not explaining better; it's complicated to explain and I
assumed that most people were somewhat familiar with the MPI issues
already. Bad assumption on my part...
--
Jeff Squyres
Cisco Systems
___
general mailing list
general
Is anyone going to comment on this? I'm surprised / disappointed that
it's been over 2 weeks with *no* comments.
Roland can't lead *every* discussion...
On Apr 13, 2009, at 12:07 PM, Jeff Squyres wrote:
The following is a proposal from several MPI implementations to the
OpenFabrics
] problem?
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
___
general mailing list
general
On Apr 23, 2009, at 5:29 PM, Chris Worley wrote:
There is an ofed_info command that returns the version. The
output is not
too well-formed, but it should have the info.
... but the distro makers don't include it :(
Doh! Yes, that's a big bummer. :-(
--
Jeff Squyres
Cisco Systems
/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
/messages' for reading:
No such file or directory when I run openibd (it failed for another
reason).
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
CM-like packages)
Indeed, it seems like the only HPC-level packages should really be the
MPI packages -- *everything* else is basic functionality for
OpenFabrics... Right?
--
Jeff Squyres
Cisco Systems
___
general mailing list
general
). I assume that a proposal along these lines this would
be a [much] larger debate in the OpenFabrics community, and further
assume that the proposal above would be a smaller debate and actually
have a chance of being implemented in the not-distant future.
(/me puts on fire suit)
Thoughts?
--
Jeff
want (in terms of a concrete
proposal, as opposed to a list of requirements that were presented at
Sonoma), I'll send something here.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org
/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe
On Mar 5, 2009, at 5:08 AM, Tziporet Koren wrote:
Can this be fixed in Open MPI next RC?
Hrm; I thought we had fixed this. I'll check into it.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http
asking about? OMPI v1.3.1 is readying for release;
*possibly* this week (50/50 chance of that).
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
___
ewg mailing list
e...@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
--
Jeff Squyres
Cisco Systems
___
general mailing list
general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo
@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
/mailman/listinfo/ewg
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
complex/detailed mechanisms (perhaps remarkably like today's
mechanisms).
I've heard similar comments before but this too will take significant
where-with-all IMO.
Ditto my above remarks. :-)
--
Jeff Squyres
Cisco Systems
___
general mailing list
])
return -EINVAL;
+ if (!(file-device-ib_dev-uverbs_cmd_mask (1ull
hdr.command)))
+ return -ENOSYS;
+
if (!file-ucontext
hdr.command != IB_USER_VERBS_CMD_GET_CONTEXT)
return -EINVAL;
--
Jeff Squyres
Cisco Systems
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
On Nov 14, 2008, at 12:08 PM, Tziporet Koren wrote:
I don't think we need a meeting on Monday (I personally will not be
able to attend)
Ok. Unless, I hear differently by COB today (US Eastern time), I'll
cancel the phone bridge for Monday.
--
Jeff Squyres
Cisco Systems
On Sep 17, 2008, at 1:25 PM, Jeff Squyres wrote:
We do not use rdma_get_devices() because our wireup scheme is both
modular (it may use RDMA CM or it may use something else) and
separate from the RDMA device discovery process. Specifically, the
PD that we have allocated is from
this new RDMA CM wireup problem that
has forced me to set it at 7.
However, as I mentioned before, this is complex code, so it's quite
possible (likely?) that I have a bug in the code somewhere. I was
posting here looking for any possible insights into why this could
happen.
--
Jeff
appreciated; many
thanks.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
. But it is
complex code, so there certainly can be a bug in this area.
The thing that is weird for me is that setting rnr_retry to 7 makes it
work.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http
have to be subscribed to
post, sorry.
Enjoy.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman
.tar.[gz|bz2].
Can these be added to the OFED 1.4 release? (pulling from git or
using the tarball is fine -- whatever is most convenient)
Thanks.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http
typically
need to do. Is anyone working this?
Jim
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
] On Behalf Of Ryan, Jim
Sent: Monday, August 18, 2008 4:03 PM
To: Bill Boas; 'Jeff Squyres'; 'OpenFabrics EWG'; 'OpenFabrics
General'
Subject: RE: [ewg] STOP
by spam with a forged from address of [EMAIL PROTECTED]
It seems disabling the [EMAIL PROTECTED] address would fix things for
now,
and if there were some way to stop mailman from forwarding bounce
messages, that would deal with the backscatter issue more permanently.
- R.
--
Jeff Squyres
Cisco
important
security fixes since v2.1.5) is v2.1.11. Someone needs to fix this
ASAP.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
___
general mailing list
general
.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
) {
- goto err;
- }
+ if (dev-fd 0)
+ goto err2;
free(dev_path);
return dev;
-err:
- free(dev_path);
err2:
+ free(dev_path);
+err1:
free(dev);
return NULL;
}
--
Jeff Squyres
Cisco Systems
just the uverbs name.
That was just a convenient way to get the correct file name, but was
never updated once libibverbs supported non-IB devices.
Gotcha.
Do you have a guesstimate on when you could have a patch available?
--
Jeff Squyres
Cisco Systems
On Jul 21, 2008, at 5:01 PM, Sean Hefty wrote:
Do you have a guesstimate on when you could have a patch available?
I will try to get something out by the end of the week.
Many thanks, Sean. I can give it a whirl on my systems if you need
some testing.
--
Jeff Squyres
Cisco Systems
, though).
2. Or are my major/minor numbers incorrect for the devices that I
created manually? If the major/minor device numbers were created by
the OS upon bootup (as they should be -- there's an open OpenFabrics
bugzilla ticket about this), would they be correct?
--
Jeff Squyres
Cisco
On Jul 15, 2008, at 6:52 AM, Tziporet Koren wrote:
- What is the proposed release schedule for OFED 1.4?
Alpha - next week and Beta by end of July.
Do you really mean alpha one week and beta the next?
Is there a target final release date for OFED v1.4?
--
Jeff Squyres
Cisco Systems
the same issue.
There's a high probability that we're doing something wrong in OMPI's
IBCM support.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit
.
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
--
Jeff Squyres
Cisco Systems
]
vendor_id = 0x1fc1,0x1077
vendor_part_id = 16,29216
use_eager_rdma = 1
mtu = 4096
-
Translation: vendor_id values 0x1fc1 and 0x1077.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org
the
changes in OMPI v1.2.7 (coming out RSN) and v1.3 (still have a little
time).
Heck, I could just hedge our bets and put in QL's OUI in the file and
let that *also* be checked if the debate / implementation-release
cycle on this change is going to take a while...
What's QL's OUI?
--
Jeff Squyres
)?
That would be nice, but it doesn't solve my immediate problem. :-)
Paul Hargrove told me (off list) that he ended up looping over
max_inline_data values testing to see if ibv_create_qp() fails with
EINVAL. Yuck!
--
Jeff Squyres
Cisco Systems
large, ibv_create_qp will simply fail. So
how can we figure out what the max_inline_data value is for Mellanox
adapters at run-time? (I'd prefer not to repeatedly trying to create
a QP with increasing values of max_inline_data until it fails).
Thanks!
--
Jeff Squyres
Cisco Systems
currently-available RNICs (May 2008) do not support RNR
retry. Specifically: current RNICs will terminate a QP connection if
a SEND arrives with no corresponding pre-posted receive.
--
Jeff Squyres
Cisco Systems
___
general mailing list
general
1 - 100 of 172 matches
Mail list logo