diff mbox

tty/serial: fix config dependencies for samsung serial

Message ID 1409671320-7604-1-git-send-email-ch.naveen@samsung.com (mailing list archive)
State New, archived
Headers show

Commit Message

Naveen Krishna Chatradhi Sept. 2, 2014, 3:22 p.m. UTC 
From: Pankaj Dubey <pankaj.dubey@samsung.com>

Make the config symbols SERIAL_SAMSUNG_UARTS_4 and
SERIAL_SAMSUNG_UARTS depend on SERIAL_SAMSUNG rather than
PLAT_SAMSUNG.

Signed-off-by: Pankaj Dubey <pankaj.dubey@samsung.com>
Signed-off-by: Naveen Krishna Chatradhi <ch.naveen@samsung.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/tty/serial/Kconfig |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Comments

Arnd Bergmann Sept. 2, 2014, 3:46 p.m. UTC | #1 
On Tuesday 02 September 2014 20:52:00 Naveen Krishna Chatradhi wrote:
> Make the config symbols SERIAL_SAMSUNG_UARTS_4 and
> SERIAL_SAMSUNG_UARTS depend on SERIAL_SAMSUNG rather than
> PLAT_SAMSUNG.

Please always describe why you are doing a change. This patch
seems really pointless.

>  config SERIAL_SAMSUNG_UARTS_4
>         bool
> -       depends on PLAT_SAMSUNG
> +       depends on SERIAL_SAMSUNG
>         default y if !(CPU_S3C2410 || CPU_S3C2412 || CPU_S3C2440 || CPU_S3C2442)
>         help
>           Internal node for the common case of 4 Samsung compatible UARTs
>  
>  config SERIAL_SAMSUNG_UARTS
>         int
> -       depends on PLAT_SAMSUNG
> +       depends on SERIAL_SAMSUNG
>         default 4 if SERIAL_SAMSUNG_UARTS_4 || CPU_S3C2416
>         default 3
>         help
> 

Have you checked that it still builds on all samsung platforms when
SERIAL_SAMSUNG is disabled? We have had build errors in this area
in the past.

	Arnd
Naveen Krishna Ch Sept. 3, 2014, 8:21 a.m. UTC | #2 
Hi Arnd,

On 2 September 2014 21:16, Arnd Bergmann <arnd@arndb.de> wrote:
> On Tuesday 02 September 2014 20:52:00 Naveen Krishna Chatradhi wrote:
>> Make the config symbols SERIAL_SAMSUNG_UARTS_4 and
>> SERIAL_SAMSUNG_UARTS depend on SERIAL_SAMSUNG rather than
>> PLAT_SAMSUNG.
>
> Please always describe why you are doing a change. This patch
> seems really pointless.

The SERIAL_SAMSUNG_UARTS_4 and SERIAL_SAMSUNG_UARTS config options are
meaningful only if SERIAL_SAMSUNG is enabled. Hence the dependency
rules were changed. I will repost this patch with better description.

>
>>  config SERIAL_SAMSUNG_UARTS_4
>>         bool
>> -       depends on PLAT_SAMSUNG
>> +       depends on SERIAL_SAMSUNG
>>         default y if !(CPU_S3C2410 || CPU_S3C2412 || CPU_S3C2440 || CPU_S3C2442)
>>         help
>>           Internal node for the common case of 4 Samsung compatible UARTs
>>
>>  config SERIAL_SAMSUNG_UARTS
>>         int
>> -       depends on PLAT_SAMSUNG
>> +       depends on SERIAL_SAMSUNG
>>         default 4 if SERIAL_SAMSUNG_UARTS_4 || CPU_S3C2416
>>         default 3
>>         help
>>
>
> Have you checked that it still builds on all samsung platforms when
> SERIAL_SAMSUNG is disabled? We have had build errors in this area
> in the past.

Yes, it builds for other Samsung platforms.

>
>         Arnd

Thanks.
Arnd Bergmann Sept. 3, 2014, 10:56 a.m. UTC | #3 
On Wednesday 03 September 2014 13:51:56 Naveen Krishna Ch wrote:
> On 2 September 2014 21:16, Arnd Bergmann <arnd@arndb.de> wrote:
> > On Tuesday 02 September 2014 20:52:00 Naveen Krishna Chatradhi wrote:
> >> Make the config symbols SERIAL_SAMSUNG_UARTS_4 and
> >> SERIAL_SAMSUNG_UARTS depend on SERIAL_SAMSUNG rather than
> >> PLAT_SAMSUNG.
> >
> > Please always describe why you are doing a change. This patch
> > seems really pointless.
> 
> The SERIAL_SAMSUNG_UARTS_4 and SERIAL_SAMSUNG_UARTS config options are
> meaningful only if SERIAL_SAMSUNG is enabled. Hence the dependency
> rules were changed. I will repost this patch with better description.

My point is that the options are used by both the uart driver and
the platform code, e.g. for the purpose of PM debugging which can
be enabled even when the serial-samsung driver is turned off.

> >>  config SERIAL_SAMSUNG_UARTS_4
> >>         bool
> >> -       depends on PLAT_SAMSUNG
> >> +       depends on SERIAL_SAMSUNG
> >>         default y if !(CPU_S3C2410 || CPU_S3C2412 || CPU_S3C2440 || CPU_S3C2442)
> >>         help
> >>           Internal node for the common case of 4 Samsung compatible UARTs
> >>
> >>  config SERIAL_SAMSUNG_UARTS
> >>         int
> >> -       depends on PLAT_SAMSUNG
> >> +       depends on SERIAL_SAMSUNG
> >>         default 4 if SERIAL_SAMSUNG_UARTS_4 || CPU_S3C2416
> >>         default 3
> >>         help
> >>
> >
> > Have you checked that it still builds on all samsung platforms when
> > SERIAL_SAMSUNG is disabled? We have had build errors in this area
> > in the past.
> 
> Yes, it builds for other Samsung platforms.

Ok.

	Arnd
Abhilash Kesavan Sept. 23, 2014, 7:04 a.m. UTC | #4 
Hi Arnd,

I will be taking this forward as Naveen is no longer with Samsung.

On Wed, Sep 3, 2014 at 4:26 PM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Wednesday 03 September 2014 13:51:56 Naveen Krishna Ch wrote:
>> On 2 September 2014 21:16, Arnd Bergmann <arnd@arndb.de> wrote:
>> > On Tuesday 02 September 2014 20:52:00 Naveen Krishna Chatradhi wrote:
>> >> Make the config symbols SERIAL_SAMSUNG_UARTS_4 and
>> >> SERIAL_SAMSUNG_UARTS depend on SERIAL_SAMSUNG rather than
>> >> PLAT_SAMSUNG.
>> >
>> > Please always describe why you are doing a change. This patch
>> > seems really pointless.

I will re-post the patch with a better commit message once we sort the
below issue.

>>
>> The SERIAL_SAMSUNG_UARTS_4 and SERIAL_SAMSUNG_UARTS config options are
>> meaningful only if SERIAL_SAMSUNG is enabled. Hence the dependency
>> rules were changed. I will repost this patch with better description.
>
> My point is that the options are used by both the uart driver and
> the platform code, e.g. for the purpose of PM debugging which can
> be enabled even when the serial-samsung driver is turned off.

I looked through the code and could find two references to these
symbols in the platform-specific code: arch/arm/plat-samsung/init.c
and arch/arm/mach-s3c64xx/irq-pm.c. The first one is being used to
initialize the uart platform device in non-dt way while the second one
saves and restores the uart irq masks for all channels across a
suspend/resume cycle. Also, the save/restore will get bypassed in case
CONFIG_SERIAL_SAMSUNG_UARTS is not defined. It does not look like
either of these cases would cause any issue with the current patch,
can you please help me understand the problem ?

Regards,
Abhilash
>
>> >>  config SERIAL_SAMSUNG_UARTS_4
>> >>         bool
>> >> -       depends on PLAT_SAMSUNG
>> >> +       depends on SERIAL_SAMSUNG
>> >>         default y if !(CPU_S3C2410 || CPU_S3C2412 || CPU_S3C2440 || CPU_S3C2442)
>> >>         help
>> >>           Internal node for the common case of 4 Samsung compatible UARTs
>> >>
>> >>  config SERIAL_SAMSUNG_UARTS
>> >>         int
>> >> -       depends on PLAT_SAMSUNG
>> >> +       depends on SERIAL_SAMSUNG
>> >>         default 4 if SERIAL_SAMSUNG_UARTS_4 || CPU_S3C2416
>> >>         default 3
>> >>         help
>> >>
>> >
>> > Have you checked that it still builds on all samsung platforms when
>> > SERIAL_SAMSUNG is disabled? We have had build errors in this area
>> > in the past.
>>
>> Yes, it builds for other Samsung platforms.
>
> Ok.
>
>         Arnd
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
Arnd Bergmann Sept. 23, 2014, 8:07 a.m. UTC | #5 
On Tuesday 23 September 2014 12:34:26 Abhilash Kesavan wrote:
> >>
> >> The SERIAL_SAMSUNG_UARTS_4 and SERIAL_SAMSUNG_UARTS config options are
> >> meaningful only if SERIAL_SAMSUNG is enabled. Hence the dependency
> >> rules were changed. I will repost this patch with better description.
> >
> > My point is that the options are used by both the uart driver and
> > the platform code, e.g. for the purpose of PM debugging which can
> > be enabled even when the serial-samsung driver is turned off.
> 
> I looked through the code and could find two references to these
> symbols in the platform-specific code: arch/arm/plat-samsung/init.c
> and arch/arm/mach-s3c64xx/irq-pm.c. The first one is being used to
> initialize the uart platform device in non-dt way while the second one
> saves and restores the uart irq masks for all channels across a
> suspend/resume cycle. Also, the save/restore will get bypassed in case
> CONFIG_SERIAL_SAMSUNG_UARTS is not defined. It does not look like
> either of these cases would cause any issue with the current patch,
> can you please help me understand the problem ?

It's possible that it works fine now, I just pointed out that we
had bugs in this area in the past, so it would be good if you can
add to the changelog a statement that you have checked that the
new patch does not cause problems here.

	Arnd
diff mbox

Patch

diff --git a/drivers/tty/serial/Kconfig b/drivers/tty/serial/Kconfig
index 26cec64..4f27f0c 100644
--- a/drivers/tty/serial/Kconfig
+++ b/drivers/tty/serial/Kconfig
@@ -230,14 +230,14 @@  config SERIAL_SAMSUNG
 
 config SERIAL_SAMSUNG_UARTS_4
 	bool
-	depends on PLAT_SAMSUNG
+	depends on SERIAL_SAMSUNG
 	default y if !(CPU_S3C2410 || CPU_S3C2412 || CPU_S3C2440 || CPU_S3C2442)
 	help
 	  Internal node for the common case of 4 Samsung compatible UARTs
 
 config SERIAL_SAMSUNG_UARTS
 	int
-	depends on PLAT_SAMSUNG
+	depends on SERIAL_SAMSUNG
 	default 4 if SERIAL_SAMSUNG_UARTS_4 || CPU_S3C2416
 	default 3
 	help