[mpich-discuss] Some inconsistencies with universe size (r9823)

Yauheni Zelenko zelenko at cadence.com
Thu May 3 16:50:40 CDT 2012


Thank you for help!

I'll try new version.

Eugene.
________________________________________
From: Pavan Balaji [balaji at mcs.anl.gov]
Sent: Thursday, May 03, 2012 2:29 PM
To: mpich-discuss at mcs.anl.gov
Cc: Yauheni Zelenko
Subject: Re: [mpich-discuss] Some inconsistencies with universe size (r9823)

Thanks for reporting the error.  You are right, this was a regression.
I've fixed it in r9832.

  -- Pavan

On 05/03/2012 04:09 PM, Yauheni Zelenko wrote:
> Hi!
>
> I.m trying to experiment with dynamic spawning in r9823.
>
> My parent process relies on universe size to launch children processes (universe size - 1 of them). on remaining resources.
>
> I launch parent process with
>
> mpiexec -host "host1,host2,host3,host4" -n 1<parent process name>
>
> and
>
> mpiexec -host "host1:4" -n 1<parent process name>
>
> In first case universe size is 1, on second - 4.
>
> MPICH2 1.4 provides universe size 4 for both cases.
>
> Eugene.
>
> _______________________________________________
> mpich-discuss mailing list     mpich-discuss at mcs.anl.gov
> To manage subscription options or unsubscribe:
> https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss

--
Pavan Balaji
http://www.mcs.anl.gov/~balaji


More information about the mpich-discuss mailing list