Hi, On Wed, Nov 23, 2011 at 04:00:48PM +0200, Felipe Balbi wrote: > Then, when moving to devicetree, we can list > IRQs as 1, 2, 3. It's the only way to have a > sane devicetree actually. > > Signed-off-by: Felipe Balbi <[email protected]> > --- > drivers/cbus/retu.c | 42 +++++++++++++++++++++++++++--------------- > 1 files changed, 27 insertions(+), 15 deletions(-) > > diff --git a/drivers/cbus/retu.c b/drivers/cbus/retu.c > index 25fa405..f25e0a3 100644 > --- a/drivers/cbus/retu.c > +++ b/drivers/cbus/retu.c > @@ -25,7 +25,7 @@ > > #include <linux/module.h> > #include <linux/init.h> > - > +#include <linux/irqdomain.h> > #include <linux/slab.h> > #include <linux/kernel.h> > #include <linux/errno.h> > @@ -46,6 +46,7 @@ struct retu { > struct mutex mutex; > struct device *dev; > > + struct irq_domain irq_domain; > struct irq_chip irq_chip; > > int irq_base; > @@ -199,11 +200,9 @@ static irqreturn_t retu_irq_handler(int irq, void *_retu) > > while (idr) { > unsigned long pending = __ffs(idr); > - unsigned int irq; > > idr &= ~BIT(pending); > - irq = pending + retu->irq_base; > - handle_nested_irq(irq); > + handle_nested_irq(pending);
This is particular, I'm not sure if it's correct. Not sure if we should use hw IRQ number of linux IRQ number. A boot test should be enough to verify though. -- balbi
signature.asc
Description: Digital signature
