Question on irq autoprobe

From: Ming Lei
Date: Thu Sep 13 2012 - 22:02:10 EST


I see the below comments on probe_irq_off:

* BUGS: When used in a module (which arguably shouldn't happen)
* nothing prevents two IRQ probe callers from overlapping. The
* results of this are non-optimal.

But from the code of probe_irq_on and probe_irq_off, the mutex of
probing_active is held during the whole irq probe procedure, so
I don't understand why the above said that 'nothing prevents two
IRQ probe callers from overlapping", and why isn't the mutex of
probing_active enough to avoid the overlapping?

Ming Lei
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at