From mboxrd@z Thu Jan  1 00:00:00 1970
From: holler@ahsoftware.de (Alexander Holler)
Date: Fri, 02 Aug 2013 17:35:38 +0200
Subject: [PATCH] RFC: interrupt consistency check for OF GPIO IRQs
In-Reply-To: <51FB8292.9090000@ahsoftware.de>
References: <1375101368-17645-1-git-send-email-linus.walleij@linaro.org>
 <CACxGe6uNJNMdZ1HYGgCtLM2t2TWGCR6btjOHVmP=vDXXvvFfRg@mail.gmail.com>
 <CACRpkdYjJL6FZwEbB7XLnb9QtjB=3LeE9Ny6KQ4PkM3zpdNqqg@mail.gmail.com>
 <51F8CC35.1070704@collabora.co.uk> <51FB8292.9090000@ahsoftware.de>
Message-ID: <51FBD1CA.1000202@ahsoftware.de>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Am 02.08.2013 11:57, schrieb Alexander Holler:

> There must have been a bug in the patch too. I've also added that
> iinterrupt-parent stuff (with the same flags as used by the driver) and
> just have let the driver call
> 
> request_threaded_irq(gpio_to_irq(gpio), flags);
> 
> without the gpio_request()/input() before. And request_threaded_irq()

Not to forget that it would/will be a big problem, if you will forbid
drivers to use gpio_request() when they use the gpio as irq (as it
was/is necessary now).

With my kernel 3.10.4 with some patches on top, I see
# git grep gpio_to_irq | wc -l
678

So there are a lot of places where

gpio_request();
gpio_direction_input()
request_irq();

would have been changed to

#ifdef CONFIG_OF_GPIO
gpio_request();
gpio_direction_input()
#endif
request_irq();

to not fail during runtime (hopefully with some descriptive error
message if it fails).

Regards,

Alexander Holler