> > If rcvbuf < skb->truesize you won't have way to receive something without
> > my patch. My changes allow always a packet to be queued in the receiver
> > and so it avoids deadlocking in the receiver.
Could someone explain, why we limit sk->rcvbuf from below at all?
It is very nice to have rcvbuf=0 and it means exactly, that we
do not want to receive anything.
Being useless (but still valid?) with TCP, this feature is very useful
on datagram sockets, I even used it somewhere.
So, why not to remove this check from sock.c and not to announce
that anyone setting rcvbuf to small number gets exactly the thing
which he ordered?
Alexey
-
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/