Re-subjecting this thread... Hmm. The WHOLE_SYSTEM flag is confusing. It's only defined in the docs in terms of negative logic.
Is it true that if you don't supply WHOLE_SYSTEM, then - cpuset may not be equal to complete_cpuset - cpuset is comprised only of PUs that you can bind to - cpuset is already the logical AND of allowed_cpuset and online_cpuset - (ditto for nodeset / complete_nodeset) And if you *do* supply WHOLE_SYSTEM, then - cpuset will be equal to complete_cpuset - cpuset is not necessarily the logical AND of allowed_cpuset and online_cpuset Is that correct? On Dec 13, 2011, at 1:01 PM, Brice Goglin wrote: > Le 13/12/2011 18:47, Samuel Thibault a écrit : >> I'd say that some people might want WHOLE_SYSTEM while still needing >> the bindable cpuset. > > Let's wait for those people to complain before adding a 8th > cpuset/nodeset to the object structure. If they do complain and they > really don't want to AND the cpusets manually, we'll think about it. > > Brice > > _______________________________________________ > hwloc-devel mailing list > [email protected] > http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel -- Jeff Squyres [email protected] For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/
