The Alphabet (or, the sorting of Kconfig symbols)

The Alphabet (or, the sorting of Kconfig symbols)

[Russell King - ARM Linux]

For those who don't know it, the order of letters in the alphabet is:

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

and we use this order to determine the position of "select" entries in
Kconfig files.  We do that so that we can reduce conflicts.  Conflicts
such as these kinds of rubbish:

diff --cc arch/arm/mach-shmobile/Kconfig
index 3b8c87461d67,5249ff0511a8..000000000000
--- a/arch/arm/mach-shmobile/Kconfig
+++ b/arch/arm/mach-shmobile/Kconfig
@@@ -133,7 -116,7 +126,11 @@@ config ARCH_R8A779
        select MIGHT_HAVE_PCI
        select SH_CLK_CPG
        select RENESAS_IRQC
++<<<<<<< HEAD
 +      select SYS_SUPPORTS_SH_CMT
++=======
+       select ARCH_DMA_ADDR_T_64BIT if ARM_LPAE
++>>>>>>> arm-soc/for-next
...
diff --cc arch/arm/mach-zynq/Kconfig
index f03e75bd0b2b,cd2d55ab834a..000000000000
--- a/arch/arm/mach-zynq/Kconfig
+++ b/arch/arm/mach-zynq/Kconfig
@@@ -2,19 -2,11 +2,23 @@@ config ARCH_ZYN
        select ARM_AMBA
        select ARM_GIC
++<<<<<<< HEAD
 +      select ARCH_HAS_CPUFREQ
 +      select ARCH_HAS_OPP
 +      select COMMON_CLK
 +      select CPU_V7
 +      select GENERIC_CLOCKEVENTS
++=======
++>>>>>>> arm-soc/for-next
        select HAVE_ARM_SCU if SMP
        select HAVE_ARM_TWD if SMP
        select ICST
-       select MIGHT_HAVE_CACHE_L2X0
-       select USE_OF
-       select HAVE_SMP
-       select SPARSE_IRQ
        select CADENCE_TTC_TIMER
++<<<<<<< HEAD
 +      select ARM_GLOBAL_TIMER if !CPU_FREQ
++=======
+       select ARM_GLOBAL_TIMER
+       select MFD_SYSCON
++>>>>>>> arm-soc/for-next

Please note down the above sequence of letters for future reference. :)

-- 
FTTC broadband for 0.8mile line: now at 9.7Mbps down 460kbps up... slowly
improving, and getting towards what was expected from it.

The Alphabet (or, the sorting of Kconfig symbols)

[Russell King - ARM Linux]

I think I'm going to give up trying to merge arm-soc for my build tree,
the conflicts are far too numerous to properly fix up without spending
a /lot/ of time on this.  So from tonight, my builds will *not* include
arm-soc in any shape or form.

You can all help to avoid that by managing your files better so that
these kinds of conflicts don't happen.

On Fri, Apr 04, 2014 at 12:16:47PM +0100, Russell King - ARM Linux wrote:
> For those who don't know it, the order of letters in the alphabet is:
> 
> A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
> 
> and we use this order to determine the position of "select" entries in
> Kconfig files.  We do that so that we can reduce conflicts.  Conflicts
> such as these kinds of rubbish:
> 
> diff --cc arch/arm/mach-shmobile/Kconfig
> index 3b8c87461d67,5249ff0511a8..000000000000
> --- a/arch/arm/mach-shmobile/Kconfig
> +++ b/arch/arm/mach-shmobile/Kconfig
> @@@ -133,7 -116,7 +126,11 @@@ config ARCH_R8A779
>         select MIGHT_HAVE_PCI
>         select SH_CLK_CPG
>         select RENESAS_IRQC
> ++<<<<<<< HEAD
>  +      select SYS_SUPPORTS_SH_CMT
> ++=======
> +       select ARCH_DMA_ADDR_T_64BIT if ARM_LPAE
> ++>>>>>>> arm-soc/for-next
> ...
> diff --cc arch/arm/mach-zynq/Kconfig
> index f03e75bd0b2b,cd2d55ab834a..000000000000
> --- a/arch/arm/mach-zynq/Kconfig
> +++ b/arch/arm/mach-zynq/Kconfig
> @@@ -2,19 -2,11 +2,23 @@@ config ARCH_ZYN
>         select ARM_AMBA
>         select ARM_GIC
> ++<<<<<<< HEAD
>  +      select ARCH_HAS_CPUFREQ
>  +      select ARCH_HAS_OPP
>  +      select COMMON_CLK
>  +      select CPU_V7
>  +      select GENERIC_CLOCKEVENTS
> ++=======
> ++>>>>>>> arm-soc/for-next
>         select HAVE_ARM_SCU if SMP
>         select HAVE_ARM_TWD if SMP
>         select ICST
> -       select MIGHT_HAVE_CACHE_L2X0
> -       select USE_OF
> -       select HAVE_SMP
> -       select SPARSE_IRQ
>         select CADENCE_TTC_TIMER
> ++<<<<<<< HEAD
>  +      select ARM_GLOBAL_TIMER if !CPU_FREQ
> ++=======
> +       select ARM_GLOBAL_TIMER
> +       select MFD_SYSCON
> ++>>>>>>> arm-soc/for-next
> 
> Please note down the above sequence of letters for future reference. :)
> 
> -- 
> FTTC broadband for 0.8mile line: now at 9.7Mbps down 460kbps up... slowly
> improving, and getting towards what was expected from it.
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

-- 
FTTC broadband for 0.8mile line: now at 9.7Mbps down 460kbps up... slowly
improving, and getting towards what was expected from it.

The Alphabet (or, the sorting of Kconfig symbols)

[Grant Likely]

On Fri, Apr 4, 2014 at 4:16 AM, Russell King - ARM Linux
<linux@arm.linux.org.uk> wrote:
> For those who don't know it, the order of letters in the alphabet is:
>
> A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
>
> and we use this order to determine the position of "select" entries in
> Kconfig files.  We do that so that we can reduce conflicts.  Conflicts
> such as these kinds of rubbish:

Equally applies to #include ordering.

g.

The Alphabet (or, the sorting of Kconfig symbols)

[Richard Cochran]

On Fri, Apr 04, 2014 at 04:24:27AM -0700, Grant Likely wrote:
> 
> Equally applies to #include ordering.

Yes.

Can we add this rule into Documentation/CodingStyle?

Thanks,
Richard

The Alphabet (or, the sorting of Kconfig symbols)

[Christopher Covington]

On 04/04/2014 08:11 AM, Richard Cochran wrote:
> On Fri, Apr 04, 2014 at 04:24:27AM -0700, Grant Likely wrote:
>>
>> Equally applies to #include ordering.
> 
> Yes.
> 
> Can we add this rule into Documentation/CodingStyle?

Perhaps warnings (or errors?) from scripts/checkpatch.pl would be helpful as well.

Christopher

-- 
Employee of Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by the Linux Foundation.

The Alphabet (or, the sorting of Kconfig symbols)

[Laurent Pinchart]

On Friday 04 April 2014 14:11:10 Richard Cochran wrote:
> On Fri, Apr 04, 2014 at 04:24:27AM -0700, Grant Likely wrote:
> > Equally applies to #include ordering.
> 
> Yes.
> 
> Can we add this rule into Documentation/CodingStyle?

If you care to submit a patch I'll ack it (after review of course :-)).

-- 
Regards,

Laurent Pinchart

The Alphabet (or, the sorting of Kconfig symbols)

[Uwe Kleine-König]

On Fri, Apr 04, 2014 at 04:24:27AM -0700, Grant Likely wrote:
> On Fri, Apr 4, 2014 at 4:16 AM, Russell King - ARM Linux
> <linux@arm.linux.org.uk> wrote:
> > For those who don't know it, the order of letters in the alphabet is:
> >
> > A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
> >
> > and we use this order to determine the position of "select" entries in
> > Kconfig files.  We do that so that we can reduce conflicts.  Conflicts
> > such as these kinds of rubbish:
> 
> Equally applies to #include ordering.
With #includes there might be exceptions though because of dependencies
between headers.
I don't know how common these are today, I didn't have had dependency
problems for some time now, but I wouldn't want checkpatch to treat the
order of #includes as error, while for Kconfig it's fine.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-K?nig            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

The Alphabet (or, the sorting of Kconfig symbols)

[Richard Cochran]

On Tue, Apr 08, 2014 at 08:52:25PM +0200, Uwe Kleine-K?nig wrote:
> With #includes there might be exceptions though because of dependencies
> between headers.

That is surely a sign of broken code.

Thanks,
Richard

The Alphabet (or, the sorting of Kconfig symbols)

[Sergei Shtylyov]

Hello.

On 04/09/2014 12:21 AM, Richard Cochran wrote:

>> With #includes there might be exceptions though because of dependencies
>> between headers.

> That is surely a sign of broken code.

    Yeah, such dependencies should normally be resolved in the headers 
themselves, not the including code. Even the USB maintainer have finally 
agreed with that. :-)

WBR, Sergei