The arrows in the picture look right to me.
I would read the figure (as it is right now) as "The tools starts mpiexec". whereas the text reads "The tools is started on the mpiexec executable".
Step 4 is "the tool starts or attached to the mpiexec process". Step 1 is "the tool takes the mpiexec executable name as an argument". The picture and the text are correct as written. I'm in a bit of a hurry at the moment (gotta get to a sunset) so if you need more of an explanation it will need to wait until late when I get back from vacation. For now, I think it is correct as-is. Cheers, John D. Martin Schulz wrote:
Hi Marc-Andre,
Thanks for the comments - all fixed and in SVN, except for the comment in the picture, which I will leave to John.
Martin
On Sep 1, 2010, at 2:02 AM, Marc-Andr� Hermanns wrote:
Hi all,
apologies for my late comments.
I refer to the SVN version on Monday morning.
p.1, l.33: wasnt -> wasn't
p.1, l.44: current state state -> current state
p.2, l.22: the tool read symbol � and trace ... -> the tool reads symbol � and traces or the tools read symbol � and trace
p.2, l.43: attaches to MPI processes -> attaches to the MPI processes
p.8, l.30: attach Colorgreenmode -> attach mode
p.12, l.23 (sorry that I obviously missed this in previous drafts) In the figure 8.1 directly with the first step, shouldn't the arrow be pointing into the other direction? I would read the figure (as it is right now) as "The tools starts mpiexec". whereas the text reads "The tools is started on the mpiexec executable". Could someone help to clarify this?
p.18, l.41: be reported it to -> be reported to
p.21, l.1: the tool need only release -> the tool needs only to release
Cheers, Marc-Andre
On 28.08.2010, at 01:02, Martin Schulz wrote:
Hi all,
Attached is the final version of the MPIR document, as we are planning to present it to the MPI forum. Please speak up if you have any final comments or concerns. Otherwise, we are planning to submit this version on Monday to the forum for review for the September meeting.
Martin
<mpi-report.pdf>
________________________________________________________________________ Martin Schulz, [email protected], http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA
_______________________________________________ Mpi3-tools mailing list [email protected] http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools -- Marc-Andre Hermanns German Research School for Simulation Sciences GmbH c/o Laboratory for Parallel Programming 52056 Aachen | Germany
Tel +49 241 80 99753 Fax +49 241 80 6 99753 Web www.grs-sim.de
Members: Forschungszentrum J�lich GmbH | RWTH Aachen University Registered in the commercial register of the local court of D�ren (Amtsgericht D�ren) under registration number HRB 5268 Registered office: J�lich Executive board: Prof. Dr. Heiner M�ller-Krumbhaar | Prof. Marek Behr, Ph.D.
<smime.p7s><ATT00001..txt>
________________________________________________________________________ Martin Schulz, [email protected], http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA
_______________________________________________ Mpi3-tools mailing list [email protected] http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools