On Tue, 2008-09-16 at 16:05 -0400, Jeff Squyres wrote:
> On Sep 16, 2008, at 3:23 PM, Davis, Arlin R wrote:
>
> >> 2. Due to some uninteresting race conditions, we only allow
> >> connections to be made "in one direction" (the lower (IP address,
> >> port) tuple is the initiator). If the "wrong"
Arlin, you usually put out such nice stuff. But the install-exec-hook
you wrote into Makefile.am (as well as the uninstall-hook) are both
busted (not to mention the absolute wrong thing to do anyway).
When using automake, all of your install directives that automake
creates automatically have $(D
On Sep 16, 2008, at 3:23 PM, Davis, Arlin R wrote:
2. Due to some uninteresting race conditions, we only allow
connections to be made "in one direction" (the lower (IP address,
port) tuple is the initiator). If the "wrong" MPI process desires to
make a connection, it makes a bogus QP and initia
>
>Random notes:
>
>1. Open MPI uses a progress thread in its OpenFabrics support for the
>RDMA CM. rdma_connect() is initiated from the main thread, but all
>other events are handled from the progress thread.
>
>2. Due to some uninteresting race conditions, we only allow
>connections to
thanks, applied
___
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
Greetings. I'm trying to finish up RDMA CM support in Open MPI, but
am running into a problem on IB that I have been beating my head over
for about a week and I can't figure it out (seem to work fine on
iWARP). I know Sean is out on sabbatical; I'm hoping that someone
will have an insight
thanks, queued for 2.6.28.
___
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
Yes it was.
On Tue, Sep 16, 2008 at 1:53 AM, Roland Dreier <[EMAIL PROTECTED]> wrote:
> Looks good... I assume this has been tested and fixes the issue?
>
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman
> > nesqp in rem_ref_cm_node() is assigned before use so this
> change really > doesn't matter. Do you want me to take it out?
>
> I can drop that chunk myself... It doesn't really seems like
> a cleanup to me to add a redundant assignment, so if it
> doesn't fix anything, I'll drop it.
Oka
> nesqp in rem_ref_cm_node() is assigned before use so this change really
> doesn't matter. Do you want me to take it out?
I can drop that chunk myself... It doesn't really seems like a cleanup
to me to add a redundant assignment, so if it doesn't fix anything, I'll
drop it.
- R.
Hi Hal,
On 09:52 Tue 16 Sep , Hal Rosenstock wrote:
> > +void
> > +osm_mgrp_send_delete_notice(IN osm_subn_t * const p_subn,
> > + IN osm_log_t * const p_log, IN osm_mgrp_t *
> > p_mgrp)
> > +{
> > + mgrp_send_notice(p_subn, p_log, p_mgrp, 67);
> > }
>
> Any
> > - struct nes_qp *nesqp;
> > + struct nes_qp *nesqp = NULL;
>
> Is this really just a cleanup, or is this properly part of
> the previous patch?
>
nesqp in rem_ref_cm_node() is assigned before use so this change really
doesn't matter. Do you want me to take it out?
Chien
__
OFED meeting summary for Sep 15, 2008 on 1.4 status
Summary:
- 1.4-rc2 will be done by end of the week for the interop event
- Reviewed the testing matrix. Feedback is good but wish to add more
info
Details:
==
1. Bugs review: Reviewed critica
> I think some applications need an IP interface which could be ethernet
> or IPoIB or any other network. Others don't need anything else as they
> use the IB CM or RDMA CM.
>
> There are specific mailing lists and documentation for mvapich2 and
> pvfs2.
>
> -- Hal
Thanks Hal !!!
I will follow up
Hello,
I'm building an application (Apache Qpid) that includes
ofed/include/rdma/rdma_cma.h and receive the following error message
during "make all". (The detailed error message is at the end of this
message.)
/usr/local/ofed/include/infiniband/verbs.h:209: error: comma at end of
enumerator li
Sasha,
On Mon, Sep 15, 2008 at 6:40 AM, Sasha Khapyorsky <[EMAIL PROTECTED]> wrote:
>
> Simplify flows, remove unused and mean less osm_mgrp_init() functions,
> consolidate notice sending functions.
>
> Signed-off-by: Sasha Khapyorsky <[EMAIL PROTECTED]>
> ---
> opensm/include/opensm/osm_multicas
On Fri, Sep 12, 2008 at 3:24 PM, Kumar, Amit H. <[EMAIL PROTECTED]> wrote:
> Thank you Hal & Joe for your prompt reply.
>
> Two more questions:
> If ifconfig for IB is just for IPoIB, is it okay to bring down this
> interface(ib0) and still be able to run applications like Mvapich2 and pvfs2
> ??
Or Gerlitz wrote:
Tziporet, what firmware is going to support this?
FW - 2.6.0
Tziporet
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://open
This email was generated automatically, please do not reply
git_url: git://git.openfabrics.org/ofed_1_4/linux-2.6.git
git_branch: ofed_kernel
Common build parameters:
Passed:
Passed on i686 with linux-2.6.16
Passed on i686 with linux-2.6.19
Passed on i686 with linux-2.6.18
Passed on i686 with
Yossi Etigin wrote:
Fix bug #1172: If a pkey for an interface is not found during
initialization, then poll_timer is left uninitialized. When the
device is brought down, ipoib tries to del_timer_sync() it. This
call hangs in an infinite loop in lock_timer_base(), because
timer_base is NULL. We sh
20 matches
Mail list logo
