I have to ask each driver if it can complete a read/write/error if there
is a hup or other event outstanding. What you are suggesting means every
time there is an event I need to ask each device twice, once including a
queue once not including a queue. I have to have conditionals in the drivers
and I slow down the normal path - which is waiting.
Signal based I/O is definitely the right approach for scaling
-
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/