From: Herbert Xu > On Wed, Apr 09, 2014 at 10:10:16AM +0000, David Laight wrote: > > From: Herbert Xu > > ... > > > This patch picks the second option and moves all broadcast handling > > > bar the trivial case of packets going to a single interface into > > > a work queue. Obviously there also needs to be a limit on how > > > many broadcast packets we postpone in this way. I've arbitrarily > > > chosen tx_queue_len of the master device as the limit (act_mirred > > > also happens to use this parameter in a similar way). > > > > > > In order to ensure we don't exceed the backlog queue we will use > > > netif_rx_ni instead of netif_rx for broadcast packets. > > > > Should you limit the number of broadcasts queued for transmit > > on each interface as well as the number of postponed broadcasts. > > > > It probably isn't a good idea to completely fill an interface's > > transmit queue with broadcasts. > > These are *received* packets so I don't see how they're going > to fill up the transmit queues. I was thinking of a bridge - where the packets get transmitted. In this case they get put on the interfaces receive queue, but the same thing applies. Maybe there isn't actually a queue at that point? David -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html