[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20160615024355.GB4049@shlinux2>
Date: Wed, 15 Jun 2016 10:43:55 +0800
From: Peter Chen <hzpeterchen@...il.com>
To: Arnd Bergmann <arnd@...db.de>
Cc: kbuild test robot <lkp@...el.com>,
Ben Dooks <ben.dooks@...ethink.co.uk>, kbuild-all@...org,
linux-kernel@...ts.codethink.co.uk,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Philipp Zabel <p.zabel@...gutronix.de>,
Alan Stern <stern@...land.harvard.edu>,
Peter Chen <peter.chen@...escale.com>,
linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] USB: core: fix missing include <linux/usb/of.h>
On Tue, Jun 14, 2016 at 01:08:08PM +0200, Arnd Bergmann wrote:
> On Wednesday, June 8, 2016 3:04:27 AM CEST kbuild test robot wrote:
> > >> drivers/usb/core/of.c:32:21: error: redefinition of 'usb_of_get_child_node'
> > struct device_node *usb_of_get_child_node(struct device_node *parent,
> > ^~~~~~~~~~~~~~~~~~~~~
> > In file included from drivers/usb/core/of.c:21:0:
> > include/linux/usb/of.h:36:35: note: previous definition of 'usb_of_get_child_node' was here
> > static inline struct device_node *usb_of_get_child_node
> > ^~~~~~~~~~~~~~~~~~~~~
> >
> > vim +/usb_of_get_child_node +32 drivers/usb/core/of.c
> >
> > 69bec725 Peter Chen 2016-02-19 26 * @portnum: the port number which device is connecting
> > 69bec725 Peter Chen 2016-02-19 27 *
> > 69bec725 Peter Chen 2016-02-19 28 * Find the node from device tree according to its port number.
> > 69bec725 Peter Chen 2016-02-19 29 *
> > 69bec725 Peter Chen 2016-02-19 30 * Return: On success, a pointer to the device node, %NULL on failure.
> > 69bec725 Peter Chen 2016-02-19 31 */
> > 69bec725 Peter Chen 2016-02-19 @32 struct device_node *usb_of_get_child_node(struct device_node *parent,
> > 69bec725 Peter Chen 2016-02-19 33 int portnum)
> > 69bec725 Peter Chen 2016-02-19 34 {
> > 69bec725 Peter Chen 2016-02-19 35 struct device_node *node;
> >
> >
>
> I think what we want here is to make the compilation of of.o conditional on
> CONFIG_OF, so we get only one of the two definitions.
>
> Arnd
Thanks, Arnd. It is the correct solution, I will send patch soon.
--
Best Regards,
Peter Chen
Powered by blists - more mailing lists