> o clear-backlog-2 -> fix a SMP race in the common
> device backlog (where the irq
> queue packets that will be
> processed from net_bh()).
Network fixes go via DaveM.
And I'm not accepting this one until Andrea shows me why ANK's clever
trick there, which is _specifically_ to avoid grabbing the lock in the
top-level list traversal code, does not work correctly.
Andrea, interrupts only add to one end of the list, and BH's are
disabled so nobody can remove anything on us. Read that again,
_nobody_ can remove entries from the list while we are in the BH
atomic. It is the same reason why the scheduler locking works the
way it does.
this is why ANK walks the list the way he does and only grabs the lock
when we have to actually modify something.
Later,
David S. Miller
davem@redhat.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/