On Nov 29 2010, George Bosilca wrote:
>
> If your code doesn't exactly what is described in the code snippet
> attached to your previous email, then you can safely ignore the warning.
> In fact, any fork done prior to the communication is a non-issue, but it
> is difficult to identify. Therefore, we output the warning as soon as we
> detect a fork after MPI_Init.
USUALLY a non-issue.
> You can find more information about the usage of fork in Open MPI at
> http://www.open-mpi.de/faq/?category=tuning#fork-warning
Yes. It says about as much as can be said, except between consenting
geeks in private! Well, I suppose that this mailing list counts :-)
I have had trouble with systems that sent signals to process groups
and/or all processes with a particular controlling terminal, and even
operating system 'super groups', where the location of the fork won't
save you. I can't see how to describe that issue in terms that make
sense to most users.
Regards,
Nick Maclaren.
|
![[Flag]](../../../../../community/mirrors/flag-us.gif){kind=small}