* Xander Huff | 2015-05-07 10:38:10 [-0500]: >The driver currently registers a pair of irq handlers using >request_threaded_irq(), however the synchronization mechanism >between the hardirq and the threadedirq handler is a regular spinlock. > >Unfortunately, this breaks PREEMPT_RT builds, where a spinlock can >sleep, and is thus not able to be acquired from a hardirq handler. >Because the set of operations performed under the spinlock is already >minimal and bounded, it is a suitable candidate for being a raw_spinlock. > >This patch should not impact behavior on !PREEMPT_RT builds. > >Signed-off-by: Xander Huff <xander.huff@xxxxxx> >Reviewed-by: Joe Hershberger <joe.hershberger@xxxxxx> >Reviewed-by: Ben Shelton <ben.shelton@xxxxxx> >Reviewed-by: Josh Cartwright <joshc@xxxxxx> Do you have any cyclictest numbers how much the latency increases on heavy iio usage? Traditionally the primary handler should be very quick / short. Sebastian -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
