> On Wed, Dec 23, 2020 at 07:09:10PM -0800, Nadav Amit wrote: > > I think there are other cases in which Andy’s concern is relevant > > (MADV_PAGEOUT). I didn't try to figure how it would help MADV_PAGEOUT. MADV_PAGEOUT sounds cool feature, but maybe it'd need a way to flush the invalidates out and a take a static key to enable the buildup of those ranges? I wouldn't like to slow down the fast paths even for MADV_PAGEOUT, and the same applies to uffd-wp and softdirty in fact. On Wed, Dec 23, 2020 at 08:34:00PM -0700, Yu Zhao wrote: > That patch only demonstrate a rough idea and I should have been > elaborate: if we ever decide to go that direction, we only need to > worry about "jumping through hoops", because the final patch (set) > I have in mind would not only have the build time optimization Andrea > suggested but also include runtime optimizations like skipping > do_swap_page() path and (!PageAnon() || page_mapcount > 1). Rest > assured, the performance impact on do_wp_page() from occasionally an > additional TLB flush on top of a page copy is negligible. Agreed, I just posted something in that direction. Feel free to refactor it, it's just a tentative implementation to show something that may deliver all the performance we need in all cases involved without slowing down the fast paths of non-uffd and non-softdirty (well 1 branch). > On Wed, Dec 23, 2020 at 07:09:10PM -0800, Nadav Amit wrote: > > Perhaps holding some small bitmap based on part of the deferred flushed > > pages (e.g., bits 12-17 of the address or some other kind of a single > > hash-function bloom-filter) would be more performant to avoid (most) The concern here aren't only the page faults having to run the bloom filter, but how to manage the RAM storage pointed by the bloomfilter or whatever index into the storage, which would slowdown mprotect. Granted that mprotect is slow to begin with, but the idea we can't make it any slower to make MADV_PAGEOUT or uffd-wp or clear_refs run faster since it's too important and too frequent in comparison. Just to restrict the potential false positive IPI caused by page_count inevitable inaccuracies to uffd-wp and softdirty runtimes, a simple check on vm_flags should be enough. Thanks, Andrea
