On Wed, 9 Oct 2019 15:21:13 -0700, Eric Dumazet wrote: > sk_add_backlog() callers usually read sk->sk_rcvbuf without > owning the socket lock. This means sk_rcvbuf value can > be changed by other cpus, and KCSAN complains. > > Add READ_ONCE() annotations to document the lockless nature > of these reads. > > Note that writes over sk_rcvbuf should also use WRITE_ONCE(), > but this will be done in separate patches to ease stable > backports (if we decide this is relevant for stable trees). > > [...] > > Signed-off-by: Eric Dumazet <eduma...@google.com> > Reported-by: syzbot <syzkal...@googlegroups.com>
Applied, thanks. There was a minor conflict on net/sctp/input.c here, hopefully resolved correctly.
