> I am not against reporting back (NETDEV_TX_CONTINUE or > NETDEV_TX_GIVE_ME_THE_SAME_Q_AGAIN for example); > what worries me is that the stopping of the queue being complex. As i > said, I may be worrying too much and i will be proved wrong.
I understand, this schema will probably work perfectly for drr or any deficit based qdisc. But what about a prio qdisc? Assuming we have 8 rings and 8 bands, we fill the first ring from the first band and the ring gets full. The driver will signal a 'continue' so we jump to the second band to fill the second ring. Now, if the first ring has slots again how could the driver signal to send packets from the first ring (high prio) again? It could send 7 continues but this sounds just horrible. ;-> > It has to be 1:1 as set by the admin. If they fsck, its their problem. > Whats the worst that could happen if you have less qdiscs and only one > ring gets used? In countries where they allow guns, people are not > stopped from shooting some important appendages off their bodies ;-> I don't disagree but we should at least provide a way to have a prio qdisc we same amount of bands as tx rings without manually configuring that amount. > I am chewing in the background, but i cant see how > reordering will happen even if we got the most stoopid admin on earth > setting things up. It is mostly theoretical but sometimes happens if many fragments pass by and the classification accuracy drops. > > Do we have any hardware which expects filling the rings in a > > RR fashion? > > I am not sure i followed what you are saying above. My question was, is there any hardware which expects us to fill the tx rings using round robin based mechanism instead of by some sort of classification? - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
