Thanks for the quick and helpful response.
I'm willing to make the changes if only I were more confident I
understood the issues.
What you call the first usage of num_procs_alive seems to count "local
procs alive or about to be launched". It tests on "child->alive ||
opal_dss.compare(&job,
On 7/12/11 4:21 PM, "Ralph Castain" wrote:
>On Jul 12, 2011, at 12:29 PM, Barrett, Brian W wrote:
>
>> On 7/11/11 4:31 PM, "Ralph Castain" wrote:
>>
>>> On Jul 11, 2011, at 2:51 PM, Barrett, Brian W wrote:
>>>
Hi all -
When libevent was made its own component last fall, it appe
On Jul 12, 2011, at 12:29 PM, Barrett, Brian W wrote:
> On 7/11/11 4:31 PM, "Ralph Castain" wrote:
>
>> On Jul 11, 2011, at 2:51 PM, Barrett, Brian W wrote:
>>
>>> Hi all -
>>>
>>> When libevent was made its own component last fall, it appears that the
>>> function renames and visibility set
Well, yes and no.
Yes - the value definitely needs to be computed in all cases since it is used
later on.
No - the way it is computed is only correct for that one usage. The later usage
(in the block that starts with 0 < opal_sys_limits.num_procs)) needs a
completely different value.
The curr
This makes sense to me.
Brian
On 7/1/11 8:45 AM, "Mike Dubman" wrote:
>WHAT: Adding communicator add/delete callbacks to MTL.
>WHY: MTL will be able to separate messages on different contexts.
>WHEN: On trunk (later on v1.5 as well), Tuesday telconf, 5 July 2011
>TIMEOUT: Tuesday telconf, 12 Ju
The function orte_odls_base_default_launch_local() has a variable
num_procs_alive that is basically initialized like this:
if ( oversubscribed ) {
...
} else {
num_procs_alive = ...;
}
Specifically, if the "oversubscribed" test passes, the variable is not
initialize
On 7/11/11 4:31 PM, "Ralph Castain" wrote:
>On Jul 11, 2011, at 2:51 PM, Barrett, Brian W wrote:
>
>> Hi all -
>>
>> When libevent was made its own component last fall, it appears that the
>> function renames and visibility settings were lost. This is proving
>> rather problematic for a projec
That is quite possible. ARM support was one of the advertised features in
1.5.3; it really should have been there. Plus, a bunch of other random bugs
have been fixed -- we've been waiting for an IB-related bug to get fixed, but
it isn't moving fast, so perhaps we should just do a 1.5.4 release
