> No, there is no mmap. Like this: > > u64 my_counter = 0; > > ibv_set_mmu_counter(verbs, &my_counter); > [..] > while (my_counter != last_my_counter) { > last_my_counter = my_counter; > ibv_get_mmu_notifications(verbs, ...); // <- I am a memory barrier as > well > } > > The kernel 'syscall' ibv_set_mmu_counter would bind the given verbs to > the 8 byte counter you specified without having to the mmap thing. As > I understand it this is what perfevents does.
I was trying to look at how perf events handles this, and AFAICT it looks like kernel/perf_event.c just supports mmap(). Can you expand on what you meant here? (I was trying to figure out how one would handle the case where userspace gives us a counter in highmem -- doing kmap_atomic() seems to be to only option but then I'm not sure if I want to deal with that...) -- Roland Dreier <rola...@cisco.com> || For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/index.html -- 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