On Mon, Aug 02, 2021 at 02:02:36PM +0300, Nikolay Aleksandrov wrote: > >> Actually I believe there is still a bug in 52e4bec15546 even with this fix. > >> The flag can change after the dst has been read in br_switchdev_fdb_notify() > >> so in theory you could still do a null pointer dereference. fdb_notify() > >> can be called from a few places without locking. The code shouldn't dereference > >> the dst based on the flag. > > > > Are you thinking of a specific code path that triggers a race between > > (a) a writer side doing WRITE_ONCE(fdb->dst, NULL) and then > > set_bit(BR_FDB_LOCAL, &fdb->flags), exactly in this order, and > > Visible order is not guaranteed, there are no barriers neither at writer nor reader > sides, especially when used without locking. So we cannot make any assumptions > about the order visibility of these writes. > > > (b) a reader side catching that fdb exactly in between the above 2 > > statements, through fdb_notify or otherwise (br_fdb_replay)? > > > > Because I don't see any. > > > > Plus, I am a bit nervous about protecting against theoretical/unproven > > races in a way that masks real bugs, as we would be doing if I add an > > extra check in br_fdb_replay_one and br_switchdev_fdb_notify against the > > case where an entry has fdb->dst == NULL but not BR_FDB_LOCAL. > > > > The bits are _not_ visible atomically with the setting of ->dst. It is obvious > you must not dereference anything based on them, they are only indications when used > outside of locked regions and code must be able to deal with inconsistencies as that > is implied by the way they're used. It is a clear and obvious bug dereferencing based > on a bit that can change in parallel without any memory ordering guarantees. Ok, I will send a separate patch for that. > You are not "masking" anything, but fixing what is currently buggy use of fdb bits. I am "masking" in the sense that the bug I am fixing here was not obvious to me until it triggered a NPD. That would stop happening with the patch I'm about to send, but maybe there are still bridge UAPI functions that do not validate the 'permanent' flag from FDB entries. > As I already said - this doesn't fix the null deref bug completely, in fact it fixes a different > inconsistency, before at worst you'd get blackholed traffic for such entries now > you get a null pointer dereference.
