ARM: imx: Explicitly select CONFIG_SPI

From: Fabio Estevam <fabio.estevam@freescale.com>

Since 3.18-rc1 CONFIG_SPI is no longer automatically selected for 
imx_v4_v5_defconfig and imx_v6_v7_defconfig, so in order to keep the same
behaviour as previous kernel versions and avoid regressions, let's explicitly
select CONFIG_SPI.

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
---
Shawn,

If possible, please apply this as a bug fix for 3.18.

 arch/arm/configs/imx_v4_v5_defconfig | 1 +
 arch/arm/configs/imx_v6_v7_defconfig | 1 +
 2 files changed, 2 insertions(+)

On Wed, Oct 29, 2014 at 07:19:28PM -0200, Fabio Estevam wrote:
> From: Fabio Estevam <fabio.estevam@freescale.com>
> 
> Since 3.18-rc1 CONFIG_SPI is no longer automatically selected for 
> imx_v4_v5_defconfig and imx_v6_v7_defconfig, so in order to keep the same
> behaviour as previous kernel versions and avoid regressions, let's explicitly
> select CONFIG_SPI.
> 
> Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
> ---
> Shawn,
> 
> If possible, please apply this as a bug fix for 3.18.

Can you please mention the commit that introduced the regression in the
commit log?

Shawn

Hi Shawn,

On Mon, Nov 3, 2014 at 4:26 AM, Shawn Guo <shawn.guo@linaro.org> wrote:
> On Wed, Oct 29, 2014 at 07:19:28PM -0200, Fabio Estevam wrote:
>> From: Fabio Estevam <fabio.estevam@freescale.com>
>>
>> Since 3.18-rc1 CONFIG_SPI is no longer automatically selected for
>> imx_v4_v5_defconfig and imx_v6_v7_defconfig, so in order to keep the same
>> behaviour as previous kernel versions and avoid regressions, let's explicitly
>> select CONFIG_SPI.
>>
>> Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
>> ---
>> Shawn,
>>
>> If possible, please apply this as a bug fix for 3.18.
>
> Can you please mention the commit that introduced the regression in the
> commit log?

The problem is that 'git bisect' did not produce a valid result for
me. It says that:

commit cf377ad7d42c566356d79049536d9cb37499cb77
Merge: 212fe84 d8f0faa
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date:   Wed Oct 8 17:13:04 2014 -0400

    Merge tag 'soc-for-linus' of
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc


, is the first bad commit, but this does not look correct, so I was
not able to really identify the commit that caused this regression.

Here is my procedure (starting at 3.18-rc1):

git bisect start
git bisect bad
git bisect good v3.17
make imx_v6_v7_defconfig
cat .config | grep CONFIG_SPI=y

Then I add "git bisect good or bad" depending on the output of the
'cat' command.

Any ideas?

On Mon, Nov 03, 2014 at 10:45:20AM -0200, Fabio Estevam wrote:
> Hi Shawn,
> 
> On Mon, Nov 3, 2014 at 4:26 AM, Shawn Guo <shawn.guo@linaro.org> wrote:
> > On Wed, Oct 29, 2014 at 07:19:28PM -0200, Fabio Estevam wrote:
> >> From: Fabio Estevam <fabio.estevam@freescale.com>
> >>
> >> Since 3.18-rc1 CONFIG_SPI is no longer automatically selected for
> >> imx_v4_v5_defconfig and imx_v6_v7_defconfig, so in order to keep the same
> >> behaviour as previous kernel versions and avoid regressions, let's explicitly
> >> select CONFIG_SPI.
> >>
> >> Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
> >> ---
> >> Shawn,
> >>
> >> If possible, please apply this as a bug fix for 3.18.
> >
> > Can you please mention the commit that introduced the regression in the
> > commit log?
> 
> The problem is that 'git bisect' did not produce a valid result for
> me. It says that:
> 
> commit cf377ad7d42c566356d79049536d9cb37499cb77
> Merge: 212fe84 d8f0faa
> Author: Linus Torvalds <torvalds@linux-foundation.org>
> Date:   Wed Oct 8 17:13:04 2014 -0400
> 
>     Merge tag 'soc-for-linus' of
> git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
> 
> 
> , is the first bad commit, but this does not look correct, so I was
> not able to really identify the commit that caused this regression.
> 
> Here is my procedure (starting at 3.18-rc1):
> 
> git bisect start
> git bisect bad
> git bisect good v3.17
> make imx_v6_v7_defconfig
> cat .config | grep CONFIG_SPI=y
> 
> Then I add "git bisect good or bad" depending on the output of the
> 'cat' command.
> 
> Any ideas?

The CONFIG_SPI was removed by your commit 64546e9fe3a5 (ARM:
imx_v6_v7_defconfig updates), which cames from savedefconfig, I assume.
That said, in the code base that your patch was generated, option
CONFIG_SPI is redundant, But it's not the case on mainline.

Shawn

On Tue, Nov 4, 2014 at 6:16 AM, Shawn Guo <shawn.guo@linaro.org> wrote:

> The CONFIG_SPI was removed by your commit 64546e9fe3a5 (ARM:
> imx_v6_v7_defconfig updates), which cames from savedefconfig, I assume.
> That said, in the code base that your patch was generated, option
> CONFIG_SPI is redundant, But it's not the case on mainline.

Ops, sorry about that. That's correct. I have just sent v2 and pointed
out the faulty commits.

Thanks