From: Andres Salomon Subject: Re: [PATCH 4/4] Move of_node[attach, detach] declarations to linux/of.h Date: Thu, 19 Nov 2009 00:35:17 -0500 Message-ID: <20091119003517.677a5b02@mycelium.queued.net> References: <4B030FC2.9070401@austin.ibm.com> <4B03629A.9050503@austin.ibm.com> <4B046B92.3060108@austin.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4B046B92.3060108-V7BBcbaFuwjMbYB6QlFGEg@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org Errors-To: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org To: devicetree-discuss-mnsaURCQ41sdnm+yROfE0A@public.gmane.org List-Id: devicetree@vger.kernel.org On Wed, 18 Nov 2009 15:48:02 -0600 Nathan Fontenot wrote: > Merge the declarations of of_attach_node and of_detach_node from the > asm/prom.h headers of powerpc and microblaze into linux/of.h. > > This update also requires adding linux/of.h to the include list for > powerpc/platforms/pseries/reconfig.h. > > Updated patch that should apply cleanly to test-devicetree, now with > diffstat output. > > Signed-off-by: Nathan Fontenot > --- > arch/microblaze/include/asm/prom.h | 4 ---- > arch/powerpc/include/asm/prom.h | 4 ---- > arch/powerpc/platforms/pseries/reconfig.c | 1 + > include/linux/of.h | 6 ++++++ > 4 files changed, 7 insertions(+), 8 deletions(-) > > Index: test-devicetree/arch/microblaze/include/asm/prom.h > =================================================================== > --- test-devicetree.orig/arch/microblaze/include/asm/prom.h > 2009-11-18 17:34:36.000000000 -0600 +++ > test-devicetree/arch/microblaze/include/asm/prom.h 2009-11-18 > 19:26:53.000000000 -0600 @@ -39,10 +39,6 @@ > extern rwlock_t devtree_lock; /* temporary while merging */ > > -/* For updating the device tree at runtime */ > -extern void of_attach_node(struct device_node *); > -extern void of_detach_node(struct device_node *); > - Minor points, but I'd recommend moving the declarations at the same time as the definitions. That is, move the of_attach_node declaration in patch1 and of_detach_node declaration in patch2. > /* Other Prototypes */ > extern int early_uartlite_console(void); > > Index: test-devicetree/arch/powerpc/include/asm/prom.h > =================================================================== > --- test-devicetree.orig/arch/powerpc/include/asm/prom.h > 2009-11-18 17:34:36.000000000 -0600 +++ > test-devicetree/arch/powerpc/include/asm/prom.h 2009-11-18 > 19:27:32.000000000 -0600 @@ -34,10 +34,6 @@ > #define HAVE_ARCH_DEVTREE_FIXUPS > > -/* For updating the device tree at runtime */ > -extern void of_attach_node(struct device_node *); > -extern void of_detach_node(struct device_node *); > - > #ifdef CONFIG_PPC32 > /* > * PCI <-> OF matching functions > Index: test-devicetree/arch/powerpc/platforms/pseries/reconfig.c > =================================================================== > --- > test-devicetree.orig/arch/powerpc/platforms/pseries/reconfig.c > 2009-11-18 17:23:40.000000000 -0600 +++ > test-devicetree/arch/powerpc/platforms/pseries/reconfig.c > 2009-11-18 19:26:53.000000000 -0600 @@ -15,6 +15,7 @@ #include > #include #include > +#include > > #include > #include > Index: test-devicetree/include/linux/of.h > =================================================================== > --- test-devicetree.orig/include/linux/of.h 2009-11-18 > 17:34:36.000000000 -0600 +++ > test-devicetree/include/linux/of.h 2009-11-18 > 19:26:53.000000000 -0600 @@ -187,4 +187,10 @@ const char *list_name, > const char *cells_name, int index, struct device_node **out_node, > const void **out_args); > +#ifdef CONFIG_OF_DYNAMIC > +/* For updating the device tree at runtime */ > +extern void of_attach_node(struct device_node *); > +extern void of_detach_node(struct device_node *); > +#endif > + > #endif /* _LINUX_OF_H */ The trend seems to be a different header file per config option. "of_dynamic.h" would be the logic extension of that..