Yes, I think I may have been responsible for that code in 2.0.x.. I was
surprised that it looked like it did in 2.1.x when I checked.
> Outside of the ACK cases, the other bogus case as I mentioned is for
> clones of real outgoing data packets, used for transmission, the SOCK
> gets charged again for the entire data area when in reality only
> sizeof(struct sk_buff) more memory is being tied up due to the clone.
I think this is bogus. I have a strong memory of _not_ updating the socket
counters at all for cloned skb's, exactly because that would have gotten
the counts wrong. The thing used to set "skb->truesize" to zero for this
reason.
I think. It's been a long time since I worked on that code, I have to
admit.
Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu