linux-next: arm tree build failure

linux-next: arm tree build failure

[Stephen Rothwell]

Hi Russell,

Today's linux-next build (x86_64 allmodconfig) failed like this:

drivers/input/touchscreen/ucb1400_ts.c: In function 'ucb1400_ts_detect_irq':
drivers/input/touchscreen/ucb1400_ts.c:333: error: 'NO_IRQ' undeclared (first use in this function)

I applied the following patch.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

>From 3a6383e5556524c941c1421a0e65a0725189ac9e Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 19 Aug 2008 15:49:30 +1000
Subject: [PATCH] ucb1400: restore NO_IRQ definition

This piece of code was removed by commit
2881353061f896c3894c64e39cf2be35b36aa9fc ("[ARM] 5184/1: Split ucb1400_ts
into core and touchscreen").

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/input/touchscreen/ucb1400_ts.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/drivers/input/touchscreen/ucb1400_ts.c b/drivers/input/touchscreen/ucb1400_ts.c
index 006499b..5498662 100644
--- a/drivers/input/touchscreen/ucb1400_ts.c
+++ b/drivers/input/touchscreen/ucb1400_ts.c
@@ -289,6 +289,10 @@ static void ucb1400_ts_close(struct input_dev *idev)
 	ucb1400_reg_write(ucb->ac97, UCB_TS_CR, 0);
 }
 
+#ifndef NO_IRQ
+#define NO_IRQ	0
+#endif
+
 /*
  * Try to probe our interrupt, rather than relying on lots of
  * hard-coded machine dependencies.
-- 
1.5.6.3

Re: linux-next: arm tree build failure

[Marek Vasut]

Dne Tuesday 19 of August 2008 07:53:19 Stephen Rothwell napsal(a):
> Hi Russell,
>
> Today's linux-next build (x86_64 allmodconfig) failed like this:
>
> drivers/input/touchscreen/ucb1400_ts.c: In function
> 'ucb1400_ts_detect_irq': drivers/input/touchscreen/ucb1400_ts.c:333: error:
> 'NO_IRQ' undeclared (first use in this function)
>
> I applied the following patch.
> --
> Cheers,
> Stephen Rothwell                    sfr@canb.auug.org.au
> http://www.canb.auug.org.au/~sfr/
>
> From 3a6383e5556524c941c1421a0e65a0725189ac9e Mon Sep 17 00:00:00 2001
> From: Stephen Rothwell <sfr@canb.auug.org.au>
> Date: Tue, 19 Aug 2008 15:49:30 +1000
> Subject: [PATCH] ucb1400: restore NO_IRQ definition
>
> This piece of code was removed by commit
> 2881353061f896c3894c64e39cf2be35b36aa9fc ("[ARM] 5184/1: Split ucb1400_ts
> into core and touchscreen").
>
> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
> ---
>  drivers/input/touchscreen/ucb1400_ts.c |    4 ++++
>  1 files changed, 4 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/input/touchscreen/ucb1400_ts.c
> b/drivers/input/touchscreen/ucb1400_ts.c index 006499b..5498662 100644
> --- a/drivers/input/touchscreen/ucb1400_ts.c
> +++ b/drivers/input/touchscreen/ucb1400_ts.c
> @@ -289,6 +289,10 @@ static void ucb1400_ts_close(struct input_dev *idev)
>  	ucb1400_reg_write(ucb->ac97, UCB_TS_CR, 0);
>  }
>
> +#ifndef NO_IRQ
> +#define NO_IRQ	0
> +#endif
> +
>  /*
>   * Try to probe our interrupt, rather than relying on lots of
>   * hard-coded machine dependencies.

This is my fault indeed, sorry. I tested only on ARM.

Re: linux-next: arm tree build failure

[Takashi Iwai]

At Tue, 19 Aug 2008 13:16:21 +0200,
Marek Vasut wrote:
> 
> Dne Tuesday 19 of August 2008 07:53:19 Stephen Rothwell napsal(a):
> > Hi Russell,
> >
> > Today's linux-next build (x86_64 allmodconfig) failed like this:
> >
> > drivers/input/touchscreen/ucb1400_ts.c: In function
> > 'ucb1400_ts_detect_irq': drivers/input/touchscreen/ucb1400_ts.c:333: error:
> > 'NO_IRQ' undeclared (first use in this function)
> >
> > I applied the following patch.
> > --
> > Cheers,
> > Stephen Rothwell                    sfr@canb.auug.org.au
> > http://www.canb.auug.org.au/~sfr/
> >
> > From 3a6383e5556524c941c1421a0e65a0725189ac9e Mon Sep 17 00:00:00 2001
> > From: Stephen Rothwell <sfr@canb.auug.org.au>
> > Date: Tue, 19 Aug 2008 15:49:30 +1000
> > Subject: [PATCH] ucb1400: restore NO_IRQ definition
> >
> > This piece of code was removed by commit
> > 2881353061f896c3894c64e39cf2be35b36aa9fc ("[ARM] 5184/1: Split ucb1400_ts
> > into core and touchscreen").
> >
> > Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
> > ---
> >  drivers/input/touchscreen/ucb1400_ts.c |    4 ++++
> >  1 files changed, 4 insertions(+), 0 deletions(-)
> >
> > diff --git a/drivers/input/touchscreen/ucb1400_ts.c
> > b/drivers/input/touchscreen/ucb1400_ts.c index 006499b..5498662 100644
> > --- a/drivers/input/touchscreen/ucb1400_ts.c
> > +++ b/drivers/input/touchscreen/ucb1400_ts.c
> > @@ -289,6 +289,10 @@ static void ucb1400_ts_close(struct input_dev *idev)
> >  	ucb1400_reg_write(ucb->ac97, UCB_TS_CR, 0);
> >  }
> >
> > +#ifndef NO_IRQ
> > +#define NO_IRQ	0
> > +#endif
> > +
> >  /*
> >   * Try to probe our interrupt, rather than relying on lots of
> >   * hard-coded machine dependencies.
> 
> This is my fault indeed, sorry. I tested only on ARM.

I also got a build error with ucb1400 regarding undefined
gpio_get_value() (sorry I lost the exact build log).

The patch below fixes it, but not 100% sure whether it breaks the
build on ARM.  Please check it.


thanks,

Takashi

===

[PATCH] ucb1400 - Remove unneeded inclusion

Including asm-generic/gpio.h causes the compile errors no systems
without GPIO_LIB support.


Signed-off-by: Takashi Iwai <tiwai@suse.de>

diff --git a/include/linux/ucb1400.h b/include/linux/ucb1400.h
index 8c10f31..970473b 100644
--- a/include/linux/ucb1400.h
+++ b/include/linux/ucb1400.h
@@ -24,7 +24,6 @@
 #define _LINUX__UCB1400_H
 
 #include <sound/ac97_codec.h>
-#include <asm-generic/gpio.h>
 #include <linux/mutex.h>
 #include <linux/platform_device.h>
 

Re: linux-next: arm tree build failure

[Russell King]

On Tue, Aug 19, 2008 at 01:21:44PM +0200, Takashi Iwai wrote:
> I also got a build error with ucb1400 regarding undefined
> gpio_get_value() (sorry I lost the exact build log).
> 
> The patch below fixes it, but not 100% sure whether it breaks the
> build on ARM.  Please check it.

There's nothing in ucb1400.h that would require gpio.h, nor in
ucb1400_*.c, so I think it's safe to remove.  Mind if I merge your
patch into Marek's original to avoid bisect problems?

Ditto for Stephen's patch?

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:

Re: linux-next: arm tree build failure

[Takashi Iwai]

At Tue, 19 Aug 2008 17:06:56 +0100,
Russell King wrote:
> 
> On Tue, Aug 19, 2008 at 01:21:44PM +0200, Takashi Iwai wrote:
> > I also got a build error with ucb1400 regarding undefined
> > gpio_get_value() (sorry I lost the exact build log).
> > 
> > The patch below fixes it, but not 100% sure whether it breaks the
> > build on ARM.  Please check it.
> 
> There's nothing in ucb1400.h that would require gpio.h, nor in
> ucb1400_*.c, so I think it's safe to remove.  Mind if I merge your
> patch into Marek's original to avoid bisect problems?

No problem for me.

Thanks!

Takashi

Re: linux-next: arm tree build failure

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 506 bytes --]

Hi Russell,

On Tue, 19 Aug 2008 17:06:56 +0100 Russell King <rmk@arm.linux.org.uk> wrote:
>
> There's nothing in ucb1400.h that would require gpio.h, nor in
> ucb1400_*.c, so I think it's safe to remove.  Mind if I merge your
> patch into Marek's original to avoid bisect problems?
> 
> Ditto for Stephen's patch?

Fine by me - that goes for all merge/build fixes I do for linux-next.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

linux-next: arm tree build failure

[Stephen Rothwell]

Hi Russell,

Today's linux-next build (x86_64 allmodconfig) failed like this:

drivers/video/backlight/corgi_lcd.c:26:36: error: asm/mach/sharpsl_param.h: No such file or directory
drivers/video/backlight/corgi_lcd.c: In function 'lcdtg_set_phadadj':
drivers/video/backlight/corgi_lcd.c:193: error: 'sharpsl_param' undeclared (first use in this function)
drivers/video/backlight/corgi_lcd.c: In function 'corgi_lcd_power_on':
drivers/video/backlight/corgi_lcd.c:240: error: 'sharpsl_param' undeclared (first use in this function)

I previously reported this against the pxa tree.  I have applied the
patch below (which may not be correct).

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Mon, 15 Sep 2008 05:24:27 +1000
Subject: [PATCH] pxa: LCD_CORGI is currently BROKEN

I assume it should really depend on ARM or some such.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/video/backlight/Kconfig |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
index 8854dbb..361240a 100644
--- a/drivers/video/backlight/Kconfig
+++ b/drivers/video/backlight/Kconfig
@@ -27,6 +27,7 @@ config LCD_CLASS_DEVICE
 config LCD_CORGI
 	tristate "LCD Panel support for SHARP corgi/spitz model"
 	depends on LCD_CLASS_DEVICE && SPI_MASTER
+	depends on ARM
 	default n
 	help
 	  Say y here to support the LCD panels usually found on SHARP
-- 
1.5.6.5

Re: linux-next: arm tree build failure

[Eric Miao]

> diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
> index 8854dbb..361240a 100644
> --- a/drivers/video/backlight/Kconfig
> +++ b/drivers/video/backlight/Kconfig
> @@ -27,6 +27,7 @@ config LCD_CLASS_DEVICE
>  config LCD_CORGI
>        tristate "LCD Panel support for SHARP corgi/spitz model"
>        depends on LCD_CLASS_DEVICE && SPI_MASTER
> +       depends on ARM

I think a "depends on PXA_SHARPSL" would be more precise here.

>        default n
>        help
>          Say y here to support the LCD panels usually found on SHARP

Re: linux-next: arm tree build failure

[Eric Miao]

On Fri, Sep 19, 2008 at 2:26 AM, Eric Miao <eric.y.miao@gmail.com> wrote:
>> diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
>> index 8854dbb..361240a 100644
>> --- a/drivers/video/backlight/Kconfig
>> +++ b/drivers/video/backlight/Kconfig
>> @@ -27,6 +27,7 @@ config LCD_CLASS_DEVICE
>>  config LCD_CORGI
>>        tristate "LCD Panel support for SHARP corgi/spitz model"
>>        depends on LCD_CLASS_DEVICE && SPI_MASTER
>> +       depends on ARM
>
> I think a "depends on PXA_SHARPSL" would be more precise here.
>
>>        default n
>>        help
>>          Say y here to support the LCD panels usually found on SHARP
>

Russell,

Since the patch titled:

"lcd: add SPI-based LCD and backlight driver for SHARP corgi/spitz"

is merged in your tree, and I don't know if it hits the mainline or not
(most likely not), I'd suggest to merge this into that one with Stephen's
Signed-off-by. Would you please do that at your convenience, or let
me know if you would like me to.

Thanks.

-- 
Cheers
- eric

Re: linux-next: arm tree build failure

[Russell King]

On Fri, Sep 19, 2008 at 02:33:02AM +0800, Eric Miao wrote:
> On Fri, Sep 19, 2008 at 2:26 AM, Eric Miao <eric.y.miao@gmail.com> wrote:
> >> diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
> >> index 8854dbb..361240a 100644
> >> --- a/drivers/video/backlight/Kconfig
> >> +++ b/drivers/video/backlight/Kconfig
> >> @@ -27,6 +27,7 @@ config LCD_CLASS_DEVICE
> >>  config LCD_CORGI
> >>        tristate "LCD Panel support for SHARP corgi/spitz model"
> >>        depends on LCD_CLASS_DEVICE && SPI_MASTER
> >> +       depends on ARM
> >
> > I think a "depends on PXA_SHARPSL" would be more precise here.
> >
> >>        default n
> >>        help
> >>          Say y here to support the LCD panels usually found on SHARP
> >
> 
> Russell,
> 
> Since the patch titled:
> 
> "lcd: add SPI-based LCD and backlight driver for SHARP corgi/spitz"
> 
> is merged in your tree, and I don't know if it hits the mainline or not
> (most likely not), I'd suggest to merge this into that one with Stephen's
> Signed-off-by. Would you please do that at your convenience, or let
> me know if you would like me to.

I've added your suggested change instead (PXA_SHARPSL makes much more
sense) so it doesn't require any additional sign-offs.  Will push it
out in about half an hour.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:

Re: linux-next: arm tree build failure

[Eric Miao]

On Fri, Sep 19, 2008 at 6:04 AM, Russell King <rmk@arm.linux.org.uk> wrote:
> On Fri, Sep 19, 2008 at 02:33:02AM +0800, Eric Miao wrote:
>> On Fri, Sep 19, 2008 at 2:26 AM, Eric Miao <eric.y.miao@gmail.com> wrote:
>> >> diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
>> >> index 8854dbb..361240a 100644
>> >> --- a/drivers/video/backlight/Kconfig
>> >> +++ b/drivers/video/backlight/Kconfig
>> >> @@ -27,6 +27,7 @@ config LCD_CLASS_DEVICE
>> >>  config LCD_CORGI
>> >>        tristate "LCD Panel support for SHARP corgi/spitz model"
>> >>        depends on LCD_CLASS_DEVICE && SPI_MASTER
>> >> +       depends on ARM
>> >
>> > I think a "depends on PXA_SHARPSL" would be more precise here.
>> >
>> >>        default n
>> >>        help
>> >>          Say y here to support the LCD panels usually found on SHARP
>> >
>>
>> Russell,
>>
>> Since the patch titled:
>>
>> "lcd: add SPI-based LCD and backlight driver for SHARP corgi/spitz"
>>
>> is merged in your tree, and I don't know if it hits the mainline or not
>> (most likely not), I'd suggest to merge this into that one with Stephen's
>> Signed-off-by. Would you please do that at your convenience, or let
>> me know if you would like me to.
>
> I've added your suggested change instead (PXA_SHARPSL makes much more
> sense) so it doesn't require any additional sign-offs.  Will push it
> out in about half an hour.
>

Good! And thanks to Stephen to point this out.

linux-next: arm tree build failure

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 3205 bytes --]

Hi Russell,

Today's linux-next build (arm versatile_defconfig and many other configs)
failed like this:

  AS      usr/initramfs_data.o
arch/arm/include/asm/unwind.h: Assembler messages:
arch/arm/include/asm/unwind.h:68: Error: Bad macro parameter list
arch/arm/include/asm/unwind.h: Assembler messages:
arch/arm/include/asm/unwind.h:68: Error: arch/arm/kernel/entry-common.S:26: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-common.S:27: Error: Bad macro parameter listbad instruction `unwind .cantunwind'
arch/arm/kernel/entry-common.S:44: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:158: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:158: Error: bad instruction `unwind .save{r0-pc}'
arch/arm/kernel/entry-armv.S:202: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:207: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:207: Error: bad instruction `unwind .save{r0-pc}'
arch/arm/kernel/entry-armv.S:235: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:258: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:258: Error: bad instruction `unwind .save{r0-pc}'
arch/arm/kernel/entry-armv.S:286: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:291: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:291: Error: bad instruction `unwind .save{r0-pc}'
arch/arm/kernel/entry-armv.S:329: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:405: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:405: Error: bad instruction `unwind .cantunwind'
arch/arm/kernel/entry-armv.S:432: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:437: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:437: Error: bad instruction `unwind .cantunwind'
arch/arm/kernel/entry-armv.S:463: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:470: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:470: Error: bad instruction `unwind .cantunwind'
arch/arm/kernel/entry-armv.S:498: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:673: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:673: Error: bad instruction `unwind .cantunwind'
arch/arm/kernel/entry-armv.S:686: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:692: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:693: Error: bad instruction `unwind .cantunwind'
arch/arm/kernel/entry-armv.S:697: Error: bad instruction `unwind .fnend'
arch/arm/kernel/entry-armv.S:707: Error: bad instruction `unwind .fnstart'
arch/arm/kernel/entry-armv.S:708: Error: bad instruction `unwind .cantunwind'
arch/arm/kernel/entry-armv.S:738: Error: bad instruction `unwind .fnend'

Presumably caused by commit 7e4e372df35c93e8f2693fd37010cec05c57ff47
("[ARM] 5383/1: unwind: Add core support for ARM stack unwinding").

Does this require a particular binutils version?
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: linux-next: arm tree build failure

[Catalin Marinas]

On Sat, 2009-02-14 at 15:27 +1100, Stephen Rothwell wrote:
> Today's linux-next build (arm versatile_defconfig and many other configs)
> failed like this:
> 
>   AS      usr/initramfs_data.o
> arch/arm/include/asm/unwind.h: Assembler messages:
> arch/arm/include/asm/unwind.h:68: Error: Bad macro parameter list
[...]
> Does this require a particular binutils version?

It looks like constructs like below are only supported by newer
binutils:

	.macro	unwind	instr:vararg
	\instr
	.endm

I'll change it to classic C macros and re-post on the ARM kernel list.

Thanks.

-- 
Catalin

linux-next: arm tree build failure

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 916 bytes --]

Hi Russell,

Today's linux-next build (arm em_x270_defconfig) failed like this:

arch/arm/kernel/sys_oabi-compat.c: In function 'sys_oabi_epoll_wait':
arch/arm/kernel/sys_oabi-compat.c:288: error: implicit declaration of function 'kmalloc'
arch/arm/kernel/sys_oabi-compat.c:288: error: 'GFP_KERNEL' undeclared (first use in this function)
arch/arm/kernel/sys_oabi-compat.c:288: warning: assignment makes pointer from integer without a cast
arch/arm/kernel/sys_oabi-compat.c:301: error: implicit declaration of function 'kfree'
arch/arm/kernel/sys_oabi-compat.c: In function 'sys_oabi_semtimedop':
arch/arm/kernel/sys_oabi-compat.c:324: error: 'GFP_KERNEL' undeclared (first use in this function)
arch/arm/kernel/sys_oabi-compat.c:324: warning: assignment makes pointer from integer without a cast
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: linux-next: arm tree build failure

[Russell King]

On Tue, Apr 14, 2009 at 05:48:14PM +1000, Stephen Rothwell wrote:
> Today's linux-next build (arm em_x270_defconfig) failed like this:
> 
> arch/arm/kernel/sys_oabi-compat.c: In function 'sys_oabi_epoll_wait':
> arch/arm/kernel/sys_oabi-compat.c:288: error: implicit declaration of function 'kmalloc'
> arch/arm/kernel/sys_oabi-compat.c:288: error: 'GFP_KERNEL' undeclared (first use in this function)
> arch/arm/kernel/sys_oabi-compat.c:288: warning: assignment makes pointer from integer without a cast
> arch/arm/kernel/sys_oabi-compat.c:301: error: implicit declaration of function 'kfree'
> arch/arm/kernel/sys_oabi-compat.c: In function 'sys_oabi_semtimedop':
> arch/arm/kernel/sys_oabi-compat.c:324: error: 'GFP_KERNEL' undeclared (first use in this function)
> arch/arm/kernel/sys_oabi-compat.c:324: warning: assignment makes pointer from integer without a cast

It'll take me a while to resolve anything; I've effectively been away since
the 7th, so I've piles of email to read through.  I do notice that other
people have fixes, but the question is what caused this to occur (and
therefore is the fix the right one.)

Expect something within a week.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:

linux-next: arm tree build failure

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1260 bytes --]

Hi Russell,

The next-20090907 and later builds (arm at91cap9adk_defconfig) failed
like this:

arch/arm/mach-at91/at91cap9_devices.c:821: warning: 'struct atmel_ac97_data' declared inside parameter list
arch/arm/mach-at91/at91cap9_devices.c:821: warning: its scope is only this definition or declaration, which is probably not what you want
arch/arm/mach-at91/at91cap9_devices.c:821: error: conflicting types for 'at91_add_device_ac97'
arch/arm/mach-at91/include/mach/board.h:178: note: previous declaration of 'at91_add_device_ac97' was here
arch/arm/mach-at91/board-cap9adk.c:367: error: variable 'cap9adk_ac97_data' has initializer but incomplete type
arch/arm/mach-at91/board-cap9adk.c: In function 'cap9adk_board_init':
arch/arm/mach-at91/board-cap9adk.c:397: warning: passing argument 1 of 'at91_add_device_ac97' from incompatible pointer type
arch/arm/mach-at91/include/mach/board.h:178: note: expected 'struct ac97c_platform_data *' but argument is of type 'struct atmel_ac97_data *'

Caused by commit d656f07a7405e788e1275d0238b67f593615f2bb ("ARM: 5640/1:
This patch modifies the support of AC97 on the at91sam9263 ek board").
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

Re: linux-next: arm tree build failure

[Russell King]

On Thu, Sep 10, 2009 at 02:59:44PM +1000, Stephen Rothwell wrote:
> The next-20090907 and later builds (arm at91cap9adk_defconfig) failed
> like this:
> 
> arch/arm/mach-at91/at91cap9_devices.c:821: warning: 'struct atmel_ac97_data' declared inside parameter list
> arch/arm/mach-at91/at91cap9_devices.c:821: warning: its scope is only this definition or declaration, which is probably not what you want
> arch/arm/mach-at91/at91cap9_devices.c:821: error: conflicting types for 'at91_add_device_ac97'
> arch/arm/mach-at91/include/mach/board.h:178: note: previous declaration of 'at91_add_device_ac97' was here
> arch/arm/mach-at91/board-cap9adk.c:367: error: variable 'cap9adk_ac97_data' has initializer but incomplete type
> arch/arm/mach-at91/board-cap9adk.c: In function 'cap9adk_board_init':
> arch/arm/mach-at91/board-cap9adk.c:397: warning: passing argument 1 of 'at91_add_device_ac97' from incompatible pointer type
> arch/arm/mach-at91/include/mach/board.h:178: note: expected 'struct ac97c_platform_data *' but argument is of type 'struct atmel_ac97_data *'
> 
> Caused by commit d656f07a7405e788e1275d0238b67f593615f2bb ("ARM: 5640/1:
> This patch modifies the support of AC97 on the at91sam9263 ek board").

No idea.  I guess the Atmel folk have split one logical change up between
two trees, and its only been half merged.  I don't see anything in the
patch description which mentions this though.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:

linux-next: arm tree build failure

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 484 bytes --]

Hi Russell,

Yesterday's linux-next (next-20100128) build (arm versatile_defconfig)
failed like this:

arch/arm/mach-versatile/core.c: In function 'versatile_init':
arch/arm/mach-versatile/core.c:838: error: 'i' undeclared (first use in this function)

Looks like a bad merge (commit b7cdd38da36d61d0f42367057c9d04678644317a
"Merge branch 'versatile' into devel").

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

Re: linux-next: arm tree build failure

[Russell King]

On Fri, Jan 29, 2010 at 10:38:37AM +1100, Stephen Rothwell wrote:
> Hi Russell,
> 
> Yesterday's linux-next (next-20100128) build (arm versatile_defconfig)
> failed like this:
> 
> arch/arm/mach-versatile/core.c: In function 'versatile_init':
> arch/arm/mach-versatile/core.c:838: error: 'i' undeclared (first use in this function)
> 
> Looks like a bad merge (commit b7cdd38da36d61d0f42367057c9d04678644317a
> "Merge branch 'versatile' into devel").

Thanks for catching that, now fixed.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:

Re: linux-next: arm tree build failure

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 269 bytes --]

Hi Russell,

On Fri, 29 Jan 2010 23:45:01 +0000 Russell King <rmk+lkml@arm.linux.org.uk> wrote:
>
> Thanks for catching that, now fixed.

Great, thanks.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]