I'm writing a USB serial device driver and I'm running into some
puzzling behavior. At reasonably high data rates (about 3 Mbs) I
occasionally get a driver lockup. At these rates I ordinarily receive
a sequence of throttle/unthrottle calls, but once in a while I get
them in the other order. I.e.:
throttle
unthrottle
throttle
unthrottle
unthrottle
throttle
The n_tty line discipline uses an atomic test and set flag to send the
throttle first and then the unthrottle, but my driver actually
receives them in the other order. It appears to me that there is no
mechanism in place to prevent this race if the throttling code and the
unthrottling code are running on different processors.
I confess that some of the interactions between the driver, the line
discipline, and the tty core remain a bit murky to me, and I could
easily have misunderstood how it fits together. Is it a problem in
the line discipline, my driver, or me?
I'm running 2.6.30-rc3-0058-g6b35ca0 (pulled from Greg KH's tree a
week or so ago) on a dual processor machine.
Thanks for any advice,
Rob.
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
