Hi, On Tue, Apr 14, 2020 at 12:05:15PM +0300, Andy Shevchenko wrote: > On Mon, Apr 13, 2020 at 11:48 PM Sebastian Reichel > <sebastian.reichel@xxxxxxxxxxxxx> wrote: > > On Mon, Apr 13, 2020 at 10:28:19PM +0200, saravanan sekar wrote: > > > On 13/04/20 10:10 pm, Andy Shevchenko wrote: > > > > On Mon, Apr 13, 2020 at 8:37 PM Saravanan Sekar <sravanhome@xxxxxxxxx> wrote: > > ... > > > > > > + irq = platform_get_irq(to_platform_device(pdev->dev.parent), 0); > > > > Why not to use temporary variable dev? > > > > > > > > This should be platform_get_irq_optional(). > > > > > > Platform_get_irq in turn calls platform_get_irq_optional. It was suggested > > > by Lee and is it mandatory to change it? > > > > platform_get_irq is fine. > > I don't think so. It will spill an error in case there is no IRQ or > error happened. I suppose even for an optional IRQ we want an error message when an error happens (i.e. IRQ is specified, but not used because $reason). > So, either is should be _optional, or below conditional simply wrong, should be > if (irq < 0) > return irq; In other words: Making the irq mandatory. Sounds reasonable to me considering the driver code. Without IRQ userspace needs to poll to know about error states like thermal shutdown. -- Sebastian > > > > > > + if (irq) { > > > > But this must be > > > > if (irq > 0) > > > > or you will also try to continue with error codes. > > > > > > > + ret = devm_request_irq(dev, irq, mp2629_irq_handler, > > > > > + IRQF_TRIGGER_RISING, "mp2629-charger", > > > > > + charger); > > > > > + if (ret) { > > > > > + dev_err(dev, "failed to request gpio IRQ\n"); > > > > > + goto iio_fail; > > > > > + } > > > > > + } > > > > > +} > > -- > With Best Regards, > Andy Shevchenko
Attachment:
signature.asc
Description: PGP signature