On Apr 26, 2013, at 7:40 PM, Orion Poplawski wrote:
> On 04/25/2013 04:48 AM, Jeff Squyres (jsquyres) wrote:
>> Orion --
>>
>> I would just caution against using Open MPI with a thread-enabled libevent.
>> In most cases, the performance impact wouldn't matter, but the whole point
>> of MPI i
On 04/25/2013 04:48 AM, Jeff Squyres (jsquyres) wrote:
Orion --
I would just caution against using Open MPI with a thread-enabled libevent. In
most cases, the performance impact wouldn't matter, but the whole point of MPI
is to be high performance. The current 1.7 series does *not* use a
th
Simply a typo - fixed in r28407
On Apr 26, 2013, at 4:04 PM, Jeff Squyres (jsquyres) wrote:
> It looks like the value of the MCA param mpi_param_check was flipped from
> true to false in r28236, the recent MCA param update.
>
> Nathan -- did you mean to flip this value? It means that we're ru
It looks like the value of the MCA param mpi_param_check was flipped from true
to false in r28236, the recent MCA param update.
Nathan -- did you mean to flip this value? It means that we're running without
MPI API param checking by default, which is a change from all prior versions of
Open MP
Ok, we can probably do this.
Is anyone else interested?
On Apr 24, 2013, at 1:25 PM, "Barrett, Brian W" wrote:
> I could probably do Monday afternoon and Tuesday morning? The problem with
> Friday afternoon is that it means I can't grab the Friday afternoon flights
> home and have to stay u
WHAT: Turn on libevent thread support by default
WHY: Async progress relies on progress threads, some of which involve the
event library
WHEN: As this was discussed at the last developer's meeting and at earlier
meetings at UTK, and
all were in agreement at those times, a s
I talked to jeff a while ago and he set up a temp bitbucket for that. I
submited some suggestions about the code there, so I'm not sure about the
revisio of the CMR. Anyway, this is my university mail, and I can also be
reached at alexma at mellanox.com... I'm on mike's team now.
Alex
On 25 Apr 20
