[PATCH] arm: at91: fix build-error for board-dt

[PATCH] arm: at91: fix build-error for board-dt

[Wolfram Sang]

From: Wolfram Sang <wolfram@the-dreams.de>

When build-testing i2c-patches, I encountered:

arch/arm/mach-at91/built-in.o:(.arch.info.init+0x34): undefined reference to `at91sam926x_timer'

(using allnoconfig and roughly only MACH_AT91SAM_DT enabled)

Since this boardfile doesn't belong to a specific architecture, directly
add the only supported timer to compilation.

Signed-off-by: Wolfram Sang <wolfram@the-dreams.de>
---
 arch/arm/mach-at91/Makefile |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/mach-at91/Makefile b/arch/arm/mach-at91/Makefile
index 8512e53..d7c17fc 100644
--- a/arch/arm/mach-at91/Makefile
+++ b/arch/arm/mach-at91/Makefile
@@ -79,7 +79,7 @@ obj-$(CONFIG_MACH_SNAPPER_9260)	+= board-snapper9260.o
 obj-$(CONFIG_MACH_AT91SAM9M10G45EK) += board-sam9m10g45ek.o
 
 # AT91SAM board with device-tree
-obj-$(CONFIG_MACH_AT91SAM_DT) += board-dt.o
+obj-$(CONFIG_MACH_AT91SAM_DT) += board-dt.o at91sam926x_time.o
 
 # AT91X40 board-specific support
 obj-$(CONFIG_MACH_AT91EB01)	+= board-eb01.o
-- 
1.7.10

[PATCH] arm: at91: fix build-error for board-dt

[Jean-Christophe PLAGNIOL-VILLARD]

On 12:28 Sat 21 Apr     , Wolfram Sang wrote:
> From: Wolfram Sang <wolfram@the-dreams.de>
> 
> When build-testing i2c-patches, I encountered:
> 
> arch/arm/mach-at91/built-in.o:(.arch.info.init+0x34): undefined reference to `at91sam926x_timer'
> 
> (using allnoconfig and roughly only MACH_AT91SAM_DT enabled)
> 
> Since this boardfile doesn't belong to a specific architecture, directly
> add the only supported timer to compilation.
> 
> Signed-off-by: Wolfram Sang <wolfram@the-dreams.de>
no this is selected by the soc

I do not want the timer be compiled un condtionally

Best Regards,
J.

[PATCH] arm: at91: fix build-error for board-dt

[Wolfram Sang]

On Sat, Apr 21, 2012 at 01:40:01PM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> On 12:28 Sat 21 Apr     , Wolfram Sang wrote:
> > From: Wolfram Sang <wolfram@the-dreams.de>
> > 
> > When build-testing i2c-patches, I encountered:
> > 
> > arch/arm/mach-at91/built-in.o:(.arch.info.init+0x34): undefined reference to `at91sam926x_timer'
> > 
> > (using allnoconfig and roughly only MACH_AT91SAM_DT enabled)
> > 
> > Since this boardfile doesn't belong to a specific architecture, directly
> > add the only supported timer to compilation.
> > 
> > Signed-off-by: Wolfram Sang <wolfram@the-dreams.de>
> no this is selected by the soc
> 
> I do not want the timer be compiled un condtionally

Ok, we can discard this patch and see it as a bug report.
Since you know best what you'd like, would you send a patch?

Thanks,

   Wolfram

-- 
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120421/6f706c9f/attachment.sig>

[PATCH] arm: at91: fix build-error for board-dt

[Jean-Christophe PLAGNIOL-VILLARD]

On 16:09 Sat 21 Apr     , Wolfram Sang wrote:
> On Sat, Apr 21, 2012 at 01:40:01PM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > On 12:28 Sat 21 Apr     , Wolfram Sang wrote:
> > > From: Wolfram Sang <wolfram@the-dreams.de>
> > > 
> > > When build-testing i2c-patches, I encountered:
> > > 
> > > arch/arm/mach-at91/built-in.o:(.arch.info.init+0x34): undefined reference to `at91sam926x_timer'
> > > 
> > > (using allnoconfig and roughly only MACH_AT91SAM_DT enabled)
> > > 
> > > Since this boardfile doesn't belong to a specific architecture, directly
> > > add the only supported timer to compilation.
> > > 
> > > Signed-off-by: Wolfram Sang <wolfram@the-dreams.de>
> > no this is selected by the soc
> > 
> > I do not want the timer be compiled un condtionally
> 
> Ok, we can discard this patch and see it as a bug report.
> Since you know best what you'd like, would you send a patch?
I'll fix on 3.5 by make it depends on AT91SAM9

Best Regards,
J.

[PATCH] arm: at91: fix build-error for board-dt

[Arnd Bergmann]

On Saturday 21 April 2012, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > Ok, we can discard this patch and see it as a bug report.
> > Since you know best what you'd like, would you send a patch?
> I'll fix on 3.5 by make it depends on AT91SAM9

I think it should be fixed for v3.4, especially since it's so easy
to trigger and the fix will be trivial.

	Arnd

[PATCH] arm: at91: fix build-error for board-dt

[Jean-Christophe PLAGNIOL-VILLARD]

On 18:39 Sat 21 Apr     , Arnd Bergmann wrote:
> On Saturday 21 April 2012, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > Ok, we can discard this patch and see it as a bug report.
> > > Since you know best what you'd like, would you send a patch?
> > I'll fix on 3.5 by make it depends on AT91SAM9
> 
> I think it should be fixed for v3.4, especially since it's so easy
> to trigger and the fix will be trivial.
The symbol does not exist before

and the board-dt is sam9 only

Best Regards,
J.

[PATCH] arm: at91: fix build-error for board-dt

[Arnd Bergmann]

On Sunday 22 April 2012, Jean-Christophe PLAGNIOL-VILLARD wrote:
> On 18:39 Sat 21 Apr     , Arnd Bergmann wrote:
> > On Saturday 21 April 2012, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > > Ok, we can discard this patch and see it as a bug report.
> > > > Since you know best what you'd like, would you send a patch?
> > > I'll fix on 3.5 by make it depends on AT91SAM9
> > 
> > I think it should be fixed for v3.4, especially since it's so easy
> > to trigger and the fix will be trivial.
> The symbol does not exist before
> 
> and the board-dt is sam9 only

How about this patch for v3.4 then? It ensures that we don't try to build
ARCH_AT91RM9200 or ARCH_AT91X40 with MACH_AT91SAM_DT, which would be broken
in v3.4-rc4 as Wolfram pointed out.

8<----
ARM: at91: fix DT-enabled builds for non-DT chips

The ARCH_AT91RM9200 and ARCH_AT91X40 variants do not support booting with
DT yet, so we must not let the user select MACH_AT91SAM_DT in order
to avoid build errors.

It's still possible to enable USE_OF, and we should provide a few stub
functions for that case for restart and clock_init.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>

diff --git a/arch/arm/mach-at91/Kconfig b/arch/arm/mach-at91/Kconfig
index 45db05d..b9d9410 100644
--- a/arch/arm/mach-at91/Kconfig
+++ b/arch/arm/mach-at91/Kconfig
@@ -457,6 +457,7 @@ endif
 
 # ----------------------------------------------------------
 
+if !ARCH_AT91RM9200 && !ARCH_AT91X40
 comment "Generic Board Type"
 
 config MACH_AT91SAM_DT
@@ -465,6 +466,7 @@ config MACH_AT91SAM_DT
 	help
 	  Select this if you want to experiment device-tree with
 	  an Atmel Evaluation Kit.
+endif
 
 # ----------------------------------------------------------
 
diff --git a/arch/arm/mach-at91/generic.h b/arch/arm/mach-at91/generic.h
index dd9b346..1355b44 100644
--- a/arch/arm/mach-at91/generic.h
+++ b/arch/arm/mach-at91/generic.h
@@ -55,7 +55,8 @@ extern void __init at91sam9g45_set_console_clock(int id);
 extern int __init at91_clock_init(unsigned long main_clock);
 extern int __init at91_dt_clock_init(void);
 #else
-static int inline at91_clock_init(unsigned long main_clock) { return 0; }
+static inline int at91_clock_init(unsigned long main_clock) { return 0; }
+static inline int at91_dt_clock_init(void) { return 0; }
 #endif
 struct device;
 
@@ -68,8 +69,16 @@ extern void at91sam9_idle(void);
 
 /* reset */
 extern void at91_ioremap_rstc(u32 base_addr);
+#ifdef CONFIG_AT91_SAM9_ALT_RESET
 extern void at91sam9_alt_restart(char, const char *);
+#else
+static inline void at91sam9_alt_restart(char c, const char *s) { }
+#endif
+#ifdef CONFIG_AT91_SAM9G45_RESET
 extern void at91sam9g45_restart(char, const char *);
+#else
+static inline void at91sam9g45_restart(char c, const char *s) { }
+#endif
 
 /* shutdown */
 extern void at91_ioremap_shdwc(u32 base_addr);

[PATCH] arm: at91: fix build-error for board-dt

[Wolfram Sang]

> How about this patch for v3.4 then? It ensures that we don't try to build
> ARCH_AT91RM9200 or ARCH_AT91X40 with MACH_AT91SAM_DT, which would be broken
> in v3.4-rc4 as Wolfram pointed out.
> 
> 8<----
> ARM: at91: fix DT-enabled builds for non-DT chips
> 
> The ARCH_AT91RM9200 and ARCH_AT91X40 variants do not support booting with
> DT yet, so we must not let the user select MACH_AT91SAM_DT in order
> to avoid build errors.
> 
> It's still possible to enable USE_OF, and we should provide a few stub
> functions for that case for restart and clock_init.
> 
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Thanks! In case it goes in:

Reported-by: Wolfram Sang <w.sang@pengutronix.de>

-- 
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120422/638bd040/attachment.sig>

[PATCH] arm: at91: fix build-error for board-dt

[Jean-Christophe PLAGNIOL-VILLARD]

On 13:39 Sun 22 Apr     , Arnd Bergmann wrote:
> On Sunday 22 April 2012, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > On 18:39 Sat 21 Apr     , Arnd Bergmann wrote:
> > > On Saturday 21 April 2012, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > > > Ok, we can discard this patch and see it as a bug report.
> > > > > Since you know best what you'd like, would you send a patch?
> > > > I'll fix on 3.5 by make it depends on AT91SAM9
> > > 
> > > I think it should be fixed for v3.4, especially since it's so easy
> > > to trigger and the fix will be trivial.
> > The symbol does not exist before
> > 
> > and the board-dt is sam9 only
> 
> How about this patch for v3.4 then? It ensures that we don't try to build
> ARCH_AT91RM9200 or ARCH_AT91X40 with MACH_AT91SAM_DT, which would be broken
> in v3.4-rc4 as Wolfram pointed out.
no I don't like it either

I prefer to apply this patch from 3.5
1441bd325bbbcd38d190b2444481b23cdf70069a ARM: at91: add SOC_AT91SAM9 kconfig
option to factorise select

and then fix it

so the board will be build only for sam9

Best Regards,
J.