Re: [OMPI devel] how to add a component in the ompi?

2010-03-10 Thread Jeff Squyres
Once you add a directory under ompi/mca/btl/ and add the relevant files, then the next time you run "autogen.sh", it should just "find" the component and add it to the configure and build process. You should not need to edit ompi/mca/btl/Makefile.am yourself. Have a look at these wiki pages; t

Re: [OMPI devel] RFC: Merge tmp fault recovery branch into trunk

2010-03-10 Thread Josh Hursey
Wesley, Thanks for catching that oversight. Below are the MCA parameters that you should need at the moment: # # Use the C/R Process Migration Recovery Supervisor recos_base_enable=1 # Only use the 'rsh' launcher, other launchers will be supported later plm=rsh

Re: [OMPI devel] RFC: Merge tmp fault recovery branch into trunk

2010-03-10 Thread Wesley Bland
Josh, You mentioned some MCA parameters that you would include in the email, but I don't see those parameters anywhere. Could you please put those in here to make testing easier for people. Wesley On Wed, Mar 10, 2010 at 1:26 PM, Josh Hursey wrote: > Yesterday evening George, Thomas and I dis

Re: [OMPI devel] RFC: Merge tmp fault recovery branch into trunk

2010-03-10 Thread Josh Hursey
Yesterday evening George, Thomas and I discussed some of their concerns about this RFC at the MPI Forum meeting. After the discussion, we seemed to be in agreement that the RecoS framework is a good idea and the concepts and fixes in this RFC should move forward with a couple of notes: - They

Re: [OMPI devel] RFC: increase default AC/AM/LT requirements

2010-03-10 Thread Rainer Keller
Hey Jeff, as said I'd be in favor of updating. This would get rid of some of the tests in autogen, speeding up things. Regards, Rainer On Wednesday 10 March 2010 12:07:56 pm Jeff Squyres wrote: > Brian and I chatted about this in person -- he doesn't have too strong of > an opinion here. > > We

Re: [OMPI devel] RFC: Rename --enable-*-threads andENABLE*THREAD*(take 2)

2010-03-10 Thread Ralph Castain
No problem with me. Why don't we then modify this RFC to simply eliminate the --enable-opal-progress-thread option, and leave the rest as-is so we can enable the opal thread machinery without enabling MPI thread multiple? I agree that having an ORTE-level thread makes the most sense, if it is ne

Re: [OMPI devel] RFC: Rename --enable-*-threads andENABLE*THREAD*(take 2)

2010-03-10 Thread George Bosilca
There was way too much information on this thread that I was looking for ;) And a lot of misunderstandings ... If we want to allow ORTE to be on his own thread, then we should clearly banish the progress_thread from this equation. I would prefer ORTE to be as separated from the rest of the MPI

Re: [OMPI devel] RFC: increase default AC/AM/LT requirements

2010-03-10 Thread Jeff Squyres
Brian and I chatted about this in person -- he doesn't have too strong of an opinion here. We checked versions shipped in RHEL: RHEL4: AC 2.59, AM 1.9.2, LT 1.5.6 RHEL5: AC 2.59, AM 1.9.2, LT 1.5.22 Meaning: they're both really ancient. I personally don't mind forcing developers to have more m

[OMPI devel] how to add a component in the ompi?

2010-03-10 Thread hu yaohui
Hi Jeff & All i want to add a new component in the ompi, 1: i make a dir ~/mca/btl/ht 2:Then,i have made sure some basic functions i need to implement. such as: mca_btl_ht_add_procs, mca_btl_ht_del_procs, mca_btl_ht_alloc mca_btl_ht_free mca_btl_ht_finalize. 3:after these functions,i must compile t