On Tue, May 02, 2023 at 10:00:16AM +0200, Jan Kara wrote:
> On Sat 29-04-23 02:43:32, Kirill A . Shutemov wrote:
> > I think I found relevant snippet of code that solves similar issue.
> > get_futex_key() uses RCU to stabilize page->mapping after GUP_fast:
> >
> >
> > /*
> > * The associated futex object in this case is the inode and
> > * the page->mapping must be traversed. Ordinarily this should
> > * be stabilised under page lock but it's not strictly
> > * necessary in this case as we just want to pin the inode, not
> > * update the radix tree or anything like that.
> > *
> > * The RCU read lock is taken as the inode is finally freed
> > * under RCU. If the mapping still matches expectations then the
> > * mapping->host can be safely accessed as being a valid inode.
> > */
> > rcu_read_lock();
> >
> > if (READ_ONCE(page->mapping) != mapping) {
> > rcu_read_unlock();
> > put_page(page);
> >
> > goto again;
> > }
> >
> > inode = READ_ONCE(mapping->host);
> > if (!inode) {
> > rcu_read_unlock();
> > put_page(page);
> >
> > goto again;
> > }
> >
> > I think something similar can be used inside GUP_fast too.
>
> Yeah, inodes (and thus struct address_space) is RCU protected these days so
> grabbing RCU lock in gup_fast() will get you enough protection for checking
> aops if you are careful (like the futex code is).
GUP_fast has IRQs disabled per definition, there is no need to then also
use rcu_read_lock().
