Hi
> > I used your test code to confirm it also fails on our trunk -
> > it looks like someone got the reference count wrong when
> > creating/destructing groups.
>
> No, the code is not MPI compliant.
>
> The culprit is line 254 in the test code where Siegmar manually
> copied the group_comm_wo
Ah - cool! Thanks!
On Jan 19, 2013, at 7:19 AM, George Bosilca wrote:
> On Jan 19, 2013, at 15:44 , Ralph Castain wrote:
>
>> I used your test code to confirm it also fails on our trunk - it looks like
>> someone got the reference count wrong when creating/destructing groups.
>
> No, the cod
On Jan 19, 2013, at 15:44 , Ralph Castain wrote:
> I used your test code to confirm it also fails on our trunk - it looks like
> someone got the reference count wrong when creating/destructing groups.
No, the code is not MPI compliant.
The culprit is line 254 in the test code where Siegmar man
I'll look into that next week.
Edgar
On 1/19/2013 8:44 AM, Ralph Castain wrote:
> I used your test code to confirm it also fails on our trunk - it looks like
> someone got the reference count wrong when creating/destructing groups.
>
> Afraid I'll have to defer to the authors of that code area..
I used your test code to confirm it also fails on our trunk - it looks like
someone got the reference count wrong when creating/destructing groups.
Afraid I'll have to defer to the authors of that code area...
On Jan 19, 2013, at 1:27 AM, Siegmar Gross
wrote:
> Hi
>
> I have installed openm
