linux-mips
[Top] [All Lists]

Re: [PATCH 1/8] arch/mips/txx9: introduce missing kfree, iounmap

To: julia@diku.dk
Subject: Re: [PATCH 1/8] arch/mips/txx9: introduce missing kfree, iounmap
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Wed, 16 Sep 2009 00:45:25 +0900 (JST)
Cc: ralf@linux-mips.org, linux-mips@linux-mips.org, linux-kernel@vger.kernel.org, kernel-janitors@vger.kernel.org
In-reply-to: <Pine.LNX.4.64.0909150701170.11392@ask.diku.dk>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <20090914.003321.160496287.anemo@mba.ocn.ne.jp> <Pine.LNX.4.64.0909132113520.31000@ask.diku.dk> <Pine.LNX.4.64.0909150701170.11392@ask.diku.dk>
Sender: linux-mips-bounce@linux-mips.org
On Tue, 15 Sep 2009 07:03:42 +0200 (CEST), Julia Lawall <julia@diku.dk> wrote:
> > +out_pdev:
> > +   platform_device_put(pdev);
> > +out_gpio:
> > +   gpio_remove(&iocled->chip);
> 
> I just noticed that the prototype of gpio_remove has __must_check
> I don't think there is anything to check here; since the thing is not 
> fully initialized here, it is unlikely to be busy.  Should there be (void) 
> in front of it?

The void casting does not silence the warning.  How about this?

        if (gpiochip_remove(&iocled->chip))
                return;

In general, memory leak would be less serious than crash or data
corruption ;)

---
Atsushi Nemoto

<Prev in Thread] Current Thread [Next in Thread>