Messages in this thread | | | Date | Tue, 22 Sep 1998 19:16:00 -0400 | From | "Theodore Y. Ts'o" <> | Subject | Re: Today Linus redesigns the networking driver interface (was Re: tulip driver in ...) |
| |
Date: Tue, 22 Sep 1998 09:23:29 -0700 (PDT) From: Linus Torvalds <torvalds@transmeta.com>
(Aggregation tends to be bad for latency, which is why it's hard: there needs to be some way for the interrupt handler to tell the system that "I want bh's run _now_, because I got an important packet or I'm close to filling up the queues").
I don't know if it's really worth pursuing, though.
This would be really nice for SLIP/PPP processing. We could have the drivers only run the BH when a SLIP or PPP framing character was processed. This gives us the benefits of aggregation without the downside of latency, since we kick the BH at exactly the right time --- when we've gotten a full packet worth of data, and can send it to the network layer.
Furthermore, it shouldn't be that hard to have interrupt drivers return a flag indicating whether BH's should run now or not. It's something that would be really nice for 2.3.
- Ted
- 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/
| |