Fri, May 19, 2023 at 02:00:54PM +0300, Matti Vaittinen kirjoitti: > The fwnode_irq_get() and the fwnode_irq_get_byname() return 0 upon > device-tree IRQ mapping failure. This is contradicting the > fwnode_irq_get_byname() function documentation and can potentially be a > source of errors like: > > int probe(...) { > ... > > irq = fwnode_irq_get_byname(); > if (irq <= 0) > return irq; > > ... > } > > Here we do correctly check the return value from fwnode_irq_get_byname() > but the driver probe will now return success. (There was already one > such user in-tree). > > Change the fwnode_irq_get_byname() to work as documented and make also the > fwnode_irq_get() follow same common convention returning a negative errno > upon failure. Reviewed-by: Andy Shevchenko <andy.shevchenko@xxxxxxxxx> > Fixes: ca0acb511c21 ("device property: Add fwnode_irq_get_byname") > Suggested-by: Sakari Ailus <sakari.ailus@xxxxxxxxxxxxxxx> > Suggested-by: Jonathan Cameron <jic23@xxxxxxxxxx> > Signed-off-by: Matti Vaittinen <mazziesaccount@xxxxxxxxx> > > --- > I dropped the existing reviewed-by tags because change to > fwnode_irq_get() was added. > > Revision history: > v3 => v4: > - Change also the fwnode_irq_get() > --- > drivers/base/property.c | 12 +++++++++--- > 1 file changed, 9 insertions(+), 3 deletions(-) > > diff --git a/drivers/base/property.c b/drivers/base/property.c > index f6117ec9805c..8c40abed7852 100644 > --- a/drivers/base/property.c > +++ b/drivers/base/property.c > @@ -987,12 +987,18 @@ EXPORT_SYMBOL(fwnode_iomap); > * @fwnode: Pointer to the firmware node > * @index: Zero-based index of the IRQ > * > - * Return: Linux IRQ number on success. Other values are determined > - * according to acpi_irq_get() or of_irq_get() operation. > + * Return: Linux IRQ number on success. Negative errno on failure. > */ > int fwnode_irq_get(const struct fwnode_handle *fwnode, unsigned int index) > { > - return fwnode_call_int_op(fwnode, irq_get, index); > + int ret; > + > + ret = fwnode_call_int_op(fwnode, irq_get, index); > + /* We treat mapping errors as invalid case */ > + if (ret == 0) > + return -EINVAL; Not sure if this is the best choice, perhaps -EEXIST or -ENOENT might be better, but it's just a spoken up thought. > + return ret; > } > EXPORT_SYMBOL(fwnode_irq_get); -- With Best Regards, Andy Shevchenko