On Thu, Apr 18, 2019 at 3:42 AM Jan Kara <jack@xxxxxxx> wrote: > > Except that this solution (biasing everyone in bio) would _more complex_ > > it is only conceptualy appealing. The changes are on the other hand much > > deeper and much riskier but you decided to ignore that and focus on some- > > thing i was just giving as an example. > > Yeah, after going and reading several places like fs/iomap.c, fs/mpage.c, > drivers/md/dm-io.c I agree with you. The places that are not doing direct > IO usually just don't hold any page reference that could be directly > attributed to the bio (and they don't drop it when bio finishes). They > rather use other means (like PageLocked, PageWriteback) to make sure the > page stays alive so mandating gup-pin reference for all pages attached to a > bio would require a lot of reworking of places that are not related to our > problem and currently work just fine. So I withdraw my suggestion. Nice in > theory, too much work in practice ;). Is it though? We already have BIO_NO_PAGE_REF, so it seems it would be a useful cleanup to have all locations that don't participate in page references use that existing flag and then teach all other locations to use gup-pinned pages.
