Charm++ parallel programming system

[Ryan Mokos <ryan.mokos AT gmail.com>]

Hi Xuehan,

As the message states, this is a potential error, but it may not be critical to the completion of your simulation.  If you're running BigNetSim on multiple physical PEs, rollbacks may occur that will cancel this potential error condition.  Does your simulation complete anyway, or does it end prematurely with some kind of error?  And are you running BigNetSim sequentially, on 1 PE, or in parallel, on multiple PEs?

Also, you can run with an extra command line parameter, -check, that will tell you how many unexecuted events are left for each time line.  If no events are listed (it's ok if event 0 is listed for every time line except time line 0), or if some events at the beginning of the simulation are listed because you started the simulation at a skip point, then you're fine.  Otherwise, there may be a problem with your traces or a bug in BigNetSim.

Ryan

On Thu, Jun 23, 2011 at 9:44 AM, Xuehan Xu <xxhdx1985126 AT gmail.com> wrote:

Dear sirs:

               I modified the BigNeiSim to do some new simulation. But I got this error: "POTENTIAL ERROR: [16] Event 1 "msgep" already done!"

               What should I do to deal with it? Please help me. Thank you:-)

_______________________________________________
charm mailing list
charm AT cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/charm

_______________________________________________
ppl mailing list
ppl AT cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/ppl