No need for SM2 right now. We need a change to the communicator, before we
can bring this over, and have just gotten around to addressing this yet.
Rich
On 7/21/08 3:40 PM, "Jeff Squyres" wrote:
> Should these 2 components be in v1.3?
>
> --
> Jeff Squyres
> Cisco
Thanks! I'll delete from the trunk an v1.3.
On Jul 21, 2008, at 6:36 PM, Torsten Hoefler wrote:
Hi Jeff,
Should these 2 components be in v1.3?
for LibNBC: no. This was a test to implement blocking colls on top of
non-blocking when the tuned implementation was lacking some ops (e.g.,
On Jul 21, 2008, at 6:57 PM, Brian W. Barrett wrote:
I guess I don't understand. I thought there were three versions in
every
component -- the MCA version, the framework version, and the component
version. The first two should determine if the component can safely
be
loaded and the third
On Sun, 20 Jul 2008 21:13:48 +0200, Ralf Wildenhues
wrote:
> * Funda Wang wrote on Sun, Jul 20, 2008 at 05:29:57AM CEST:
>> I'm currently building openmpi 1.2.6 under Mandriva cooker, and its
>> default LDFLAGS is "-Wl,--as-needed -Wl,--no-undefined".
>>
>> But openmpi
> After starting, we decided that changing the MCA base revision number
> to 2.0.0 also meant changing *ALL* the framework version numbers.
> This is because the same components from framework compiled with
> MCA base version 1.x.x would not be binary compatible when compiled
> with MCA
Hi Jeff,
> Should these 2 components be in v1.3?
for LibNBC: no. This was a test to implement blocking colls on top of
non-blocking when the tuned implementation was lacking some ops (e.g.,
Allgather). Tuned now efficiently implements most collectives which
makes the blocking version of LibNBC
Short version:
==
MCA base component changes were a bit more extensive than we
anticipated. Please test the following hg tree on your systems before
we bring it into the SVN trunk this Friday evening:
http://www.open-mpi.org/hg/hgwebdir.cgi/jsquyres/mca-base-updates/
Should these 2 components be in v1.3?
--
Jeff Squyres
Cisco Systems