On Jun 12, 2013, at 5:47 PM, Jason Gunthorpe <jguntho...@obsidianresearch.com> wrote:
> Someone has to finish the ummunotify rewrite Roland > started. Realistically MPI is going to be the only user, can someone > from the MPI world do this? 1. I tried to ask what needed to be done at the beginning of this thread and didn't get much of an answer. 2. We've (all) been asking for this functionality *for years*; I even helped with the first implementation. Can't the verbs community finish it? :-) MPI is probably your biggest customer, after all... >> ...but this is not how people write applications. Real apps use >> malloc (and some direct mmap, and perhaps even some shared memory). > > *shrug* I used MAP_FIXED for some RDMA regions in my IB verbs apps, > specifically to create specalized high-performance memory > structures. But you're not a chemist writing Fortran code to effect n-body simulations. The target audience for MPI is scientists and engineers who are not (and should not be) network / systems developers. They're focusing on their formulae and applications -- as they should be. > It isn't a general purpose technique for non-RDMA apps - but > especially when combined with ODP it is useful in some places. I have no doubt that ODP solves problems for someone. It just doesn't seem to solve the very-long-standing MPI issues with verbs and registration caches. -- Jeff Squyres jsquy...@cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/ -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html