Skip to Content.
Sympa Menu

charm - Re: [charm] reduction client

charm AT lists.cs.illinois.edu

Subject: Charm++ parallel programming system

List archive

Re: [charm] reduction client


Chronological Thread 
  • From: Pritish Jetley <pjetley2 AT illinois.edu>
  • To: Nicolas Bock <nicolasbock AT gmail.com>, "charm AT cs.uiuc.edu" <charm AT cs.uiuc.edu>
  • Subject: Re: [charm] reduction client
  • Date: Fri, 26 Jul 2013 14:34:55 -0500
  • List-archive: <http://lists.cs.uiuc.edu/pipermail/charm/>
  • List-id: CHARM parallel programming system <charm.cs.uiuc.edu>

Sorry, I meant that the method invoking Work::reduce() should be threaded. 

Pritish


On Fri, Jul 26, 2013 at 2:30 PM, Nicolas Bock <nicolasbock AT gmail.com> wrote:
Just to make sure, you mean that the caller needs to be threaded, but Work::reduce() does not, since it is being called, right? And the reason is that the RTS would block otherwise, right?

nick



On Fri, Jul 26, 2013 at 12:28 PM, Pritish Jetley <pjetley2 AT illinois.edu> wrote:

As long as work::reduce is a threaded entry method, this is ok. The destructor of the resume thread callback suspends the thread in which it is invoked, which is why you must label the calling entry method as threaded.

Pritish

On Jul 26, 2013 11:49 AM, "Nicolas Bock" <nicolasbock AT gmail.com> wrote:
Hi,

is it ok to use a CkCallbackResumeThread() as a reduction client?

CProxy_Work A = CProxy_Work::ckNew();
A.reduce(CkCallbackResumeThread());

void Work::reduce (CkCallback &cb)
{
  /* .... */
  contribute(cb);
}

Thanks,

nick


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





--
Pritish Jetley
Doctoral Candidate, Computer Science
University of Illinois at Urbana-Champaign



Archive powered by MHonArc 2.6.16.

Top of Page