[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20100725130723.a5192793.sfr@canb.auug.org.au>
Date: Sun, 25 Jul 2010 13:07:23 +1000
From: Stephen Rothwell <sfr@...b.auug.org.au>
To: Grant Likely <grant.likely@...retlab.ca>
Cc: monstr@...str.eu, microblaze-uclinux@...e.uq.edu.au,
devicetree-discuss@...ts.ozlabs.org, linux-kernel@...r.kernel.org,
linuxppc-dev@...abs.org, benh@...nel.crashing.org,
sparclinux@...r.kernel.org, davem@...emloft.net
Subject: Re: [PATCH] of: Provide default of_node_to_nid() when CONFIG_NUMA
is not set
Hi Grant,
On Sat, 24 Jul 2010 09:43:31 -0600 Grant Likely <grant.likely@...retlab.ca> wrote:
>
> of_node_to_nid() is only relevant for NUMA. Don't force architectures to
> implement it if CONFIG_NUMA is not set.
Why not just do:
#ifndef of_node_to_nid
static inline int of_node_to_nid(struct device_node *device) { return 0; }
#define of_node_to_nid of_node_to_nid
#endif
in include/linux/of.h
and then add:
#define of_node_to_nid of_node_to_nid
Where it is declared/defined for each arch ...
(it would be nice if all the archs were consistent in where it was
declared).
--
Cheers,
Stephen Rothwell sfr@...b.auug.org.au
http://www.canb.auug.org.au/~sfr/
Content of type "application/pgp-signature" skipped
Powered by blists - more mailing lists