Brice Goglin, le Wed 05 Jan 2011 17:17:00 +0100, a écrit : > Le 05/01/2011 17:13, Samuel Thibault a écrit : > > Jeff Squyres, le Wed 05 Jan 2011 15:06:05 +0100, a écrit : > > > >> On Jan 5, 2011, at 8:58 AM, Jeff Squyres wrote: > >> > >> > >>>> Actually, it's usually only supported for read-only data. > >>>> > >>> I'm not sure what you mean -- if this policy is used to allocate memory, > >>> how do I write meaningful data in there that can then be read by multiple > >>> memory localities later? > >>> > >> Ahh... the light clicks on: can you only use REPLICATE with the > >> hwlo_set_*membind* functions (i.e., not the hwloc_alloc* functions)? > >> > > Yes. > > Maybe return EINVAL when passing REPLICATE to alloc_membind?
Well, if we go that way we should add tests in mostly all the src/bind.c functions, then. Samuel
