Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

From: Tim Prins (tprins_at_[hidden])
Date: 2007-10-08 14:11:42


I don't think so (in fact the bookkeeping overhead of the non-blocking
receive is a slight detriment).

Right now modex information is only exchanged during init and during
spawn/dynamics operations (and I do not see that changing at any point
soon). So I think the only use of the non-blocking receive would be to
ask for the information before the exchange has been done. I don't know
if this is helpful to anyone. My guess is not since no one does it.

Tim

Andrew Friedley wrote:
> Tim Prins wrote:
>> Hi,
>>
>> I am working on implementing the RSL. Part of this is changing the modex
>> to use the process attribute system in the RSL. I had designed this
>> system to to include a non-blocking interface.
>>
>> However, I have looked again and noticed that nobody is using the
>> non-blocking modex receive. Because of this I am inclined to not have
>> such an interface in the RSL.
>
> hmm, would using a non-blocking modex recv improve performance in any
> way, or have any other useful impacts? If so, I would use it.
>
> Andrew
> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel