One more thing...
On Fri, Jul 16, 2010 at 09:20:18AM -0700, Dmitry Torokhov wrote:
> On Fri, Jul 16, 2010 at 07:48:07PM +0800, Xiaolong CHEN wrote:
> > static int __devexit adp5588_remove(struct i2c_client *client)
> > {
> > + int ret;
> > struct adp5588_kpad *kpad = i2c_get_clientdata(client);
> >
> > adp5588_write(client, CFG, 0);
> > free_irq(client->irq, kpad);
> > cancel_delayed_work_sync(&kpad->work);
> > input_unregister_device(kpad->input);
> > + if (kpad->support_gpio)
> > + ret = gpiochip_remove(&kpad->gpio_chip);
You need to actually handle this (add an error message) since we can't
simply ignore errors.
However I think that having gpiochip_remove() __must_check while
gpiochip_add is quite silly.
I'd argue that gpiochip_remove() should wither WARN or BUG if we trying
to remove a chip that still has GPIOs requested and return void since
callers normally can't do anything about the failure. Many simply abort
device teardown leaving device a zombie and report failure up the stack,
to the driver core, where it is ignored. Not very helpful.
--
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
