[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20140514235232.57bbc9df@wiggum>
Date: Wed, 14 May 2014 23:52:32 +0200
From: Michael Büsch <m@...s.ch>
To: abdoulaye berthe <berthe.ab@...il.com>
Cc: linus.walleij@...aro.org, gnurou@...il.com,
linux-gpio@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] gpio: gpio-bt8xx: fix compilation warning
On Wed, 14 May 2014 23:43:36 +0200
abdoulaye berthe <berthe.ab@...il.com> wrote:
> this fixes a compilation warning by checking
> the retun value of gpiochip_remove()
Really, I would rather change the gpiochip_remove return type to void.
What is the point of printing a useless error message in case
of a failure and then continuing as always.
This is not fixing a compiler warning, it is a workaround. And a bad one, indeed.
How are we supposed to handle a gpiochip_remove failure correctly
in a pcidev remove handler?
> Signed-off-by: abdoulaye berthe <berthe.ab@...il.com>
> ---
> drivers/gpio/gpio-bt8xx.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpio/gpio-bt8xx.c b/drivers/gpio/gpio-bt8xx.c
> index ecb3ca2d..5480229 100644
> --- a/drivers/gpio/gpio-bt8xx.c
> +++ b/drivers/gpio/gpio-bt8xx.c
> @@ -240,7 +240,8 @@ static void bt8xxgpio_remove(struct pci_dev *pdev)
> {
> struct bt8xxgpio *bg = pci_get_drvdata(pdev);
>
> - gpiochip_remove(&bg->gpio);
> + if (gpiochip_remove(&bg->gpio))
> + dev_info(&pdev->dev, "gpiochip_remove() failed.\n");
>
> bgwrite(0, BT848_INT_MASK);
> bgwrite(~0x0, BT848_INT_STAT);
--
Michael.
----
Please use PGP/GPG encryption.
Key-ID: F532BE1D908D8B0E
--------
Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)
Powered by blists - more mailing lists