[U-Boot] [PATCH 1/2] cmd: ubifs: Add a dependency on CMD_UBI

[U-Boot] [PATCH 1/2] cmd: ubifs: Add a dependency on CMD_UBI

[Maxime Ripard]

CMD_UBIFS can't compile without CMD_UBI enabled. Make sure we can't end up
in that case.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
 cmd/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/cmd/Kconfig b/cmd/Kconfig
index 8e2a05de82ea..585a00c31f4e 100644
--- a/cmd/Kconfig
+++ b/cmd/Kconfig
@@ -858,6 +858,7 @@ config CMD_UBI
 
 config CMD_UBIFS
 	tristate "Enable UBIFS - Unsorted block images filesystem commands"
+	depends on CMD_UBI
 	select CRC32
 	select RBTREE if ARCH_SUNXI
 	select LZO if ARCH_SUNXI
-- 
2.11.0

[U-Boot] [PATCH 2/2] cmd: nand: Make the NAND options default to NAND_SUNXI

[Maxime Ripard]

If we depend on the ARCH_SUNXI configuration option, the boards that do not
have NAND support enabled (with the associated options) will not compile
anymore.

Depend on the NAND driver configuration option to make sure that is not the
case.

Reported-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
 cmd/Kconfig | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/cmd/Kconfig b/cmd/Kconfig
index 585a00c31f4e..91e3d6a46769 100644
--- a/cmd/Kconfig
+++ b/cmd/Kconfig
@@ -438,7 +438,7 @@ config CMD_MMC
 
 config CMD_NAND
 	bool "nand"
-	default y if ARCH_SUNXI
+	default y if NAND_SUNXI
 	help
 	  NAND support.
 
@@ -847,7 +847,7 @@ config CMD_UBI
 	tristate "Enable UBI - Unsorted block images commands"
 	select CRC32
 	select MTD_UBI
-	default y if ARCH_SUNXI
+	default y if NAND_SUNXI
 	help
 	  UBI is a software layer above MTD layer which admits use of LVM-like
 	  logical volumes on top of MTD devices, hides some complexities of
@@ -862,7 +862,7 @@ config CMD_UBIFS
 	select CRC32
 	select RBTREE if ARCH_SUNXI
 	select LZO if ARCH_SUNXI
-	default y if ARCH_SUNXI
+	default y if NAND_SUNXI
 	help
 	  UBIFS is a file system for flash devices which works on top of UBI.
 
-- 
2.11.0

[U-Boot] [PATCH 1/2] cmd: ubifs: Add a dependency on CMD_UBI

[Heiko Schocher]

Hello Maxime,

Am 03.03.2017 um 15:32 schrieb Maxime Ripard:
> CMD_UBIFS can't compile without CMD_UBI enabled. Make sure we can't end up
> in that case.
>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> ---
>   cmd/Kconfig | 1 +
>   1 file changed, 1 insertion(+)

Good catch! Thanks!

Acked-by: Heiko Schocher <hs@denx.de>

bye,
Heiko
>
> diff --git a/cmd/Kconfig b/cmd/Kconfig
> index 8e2a05de82ea..585a00c31f4e 100644
> --- a/cmd/Kconfig
> +++ b/cmd/Kconfig
> @@ -858,6 +858,7 @@ config CMD_UBI
>
>   config CMD_UBIFS
>   	tristate "Enable UBIFS - Unsorted block images filesystem commands"
> +	depends on CMD_UBI
>   	select CRC32
>   	select RBTREE if ARCH_SUNXI
>   	select LZO if ARCH_SUNXI
>

-- 
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

[U-Boot] [PATCH 2/2] cmd: nand: Make the NAND options default to NAND_SUNXI

[Chen-Yu Tsai]

On Fri, Mar 3, 2017 at 10:32 PM, Maxime Ripard
<maxime.ripard@free-electrons.com> wrote:
> If we depend on the ARCH_SUNXI configuration option, the boards that do not
> have NAND support enabled (with the associated options) will not compile
> anymore.
>
> Depend on the NAND driver configuration option to make sure that is not the
> case.
>
> Reported-by: Chen-Yu Tsai <wens@csie.org>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>

Tested-by: Chen-Yu Tsai <wens@csie.org>

> ---
>  cmd/Kconfig | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/cmd/Kconfig b/cmd/Kconfig
> index 585a00c31f4e..91e3d6a46769 100644
> --- a/cmd/Kconfig
> +++ b/cmd/Kconfig
> @@ -438,7 +438,7 @@ config CMD_MMC
>
>  config CMD_NAND
>         bool "nand"
> -       default y if ARCH_SUNXI
> +       default y if NAND_SUNXI

It would be nicer if it were a stronger constraint though. The user can still
enable CMD_NAND (such as by using a config from before this patch was applied)
and still break the build. I understand that this is hard to achieve since each
platform does things a bit differently.

ChenYu

>         help
>           NAND support.
>
> @@ -847,7 +847,7 @@ config CMD_UBI
>         tristate "Enable UBI - Unsorted block images commands"
>         select CRC32
>         select MTD_UBI
> -       default y if ARCH_SUNXI
> +       default y if NAND_SUNXI
>         help
>           UBI is a software layer above MTD layer which admits use of LVM-like
>           logical volumes on top of MTD devices, hides some complexities of
> @@ -862,7 +862,7 @@ config CMD_UBIFS
>         select CRC32
>         select RBTREE if ARCH_SUNXI
>         select LZO if ARCH_SUNXI
> -       default y if ARCH_SUNXI
> +       default y if NAND_SUNXI
>         help
>           UBIFS is a file system for flash devices which works on top of UBI.
>
> --
> 2.11.0
>

[U-Boot] [PATCH 2/2] cmd: nand: Make the NAND options default to NAND_SUNXI

[André Przywara]

On 03/03/17 14:32, Maxime Ripard wrote:
> If we depend on the ARCH_SUNXI configuration option, the boards that do not
> have NAND support enabled (with the associated options) will not compile
> anymore.
> 
> Depend on the NAND driver configuration option to make sure that is not the
> case.


> Reported-by: Chen-Yu Tsai <wens@csie.org>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>

As I came up with an identical patch today to fix sunxi/master:

Reviewed-by: Andre Przywara <andre.przywara@arm.com>


So this situation is a bit unfortunate: The sunxi/master branch was
updated apparently before even compile-testing it for more than one
sunxi board (it is broken for every sunxi board except Chip Pro at the
moment: buildman says 105/106). And even after reports and the
corresponding fixes appeared (thanks Chen-Yu and Maxime!) the branch is
_still_ broken as of today (10 days after the fix was posted).

So can we please have both
a) a better quality assurance before merging patches into this custodian
branch and
b) quicker merging of fixes?
To my understanding u-boot-sunxi should be always a good and working
candidate for anyone interested in Allwinner boards.

For a) it would already help to just simply build test sunxi:
$ tools/buildman/buildman sunxi
but probably extend this to cover at least all ARM boards.

Cheers,
Andre

> ---
>  cmd/Kconfig | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/cmd/Kconfig b/cmd/Kconfig
> index 585a00c31f4e..91e3d6a46769 100644
> --- a/cmd/Kconfig
> +++ b/cmd/Kconfig
> @@ -438,7 +438,7 @@ config CMD_MMC
>  
>  config CMD_NAND
>  	bool "nand"
> -	default y if ARCH_SUNXI
> +	default y if NAND_SUNXI
>  	help
>  	  NAND support.
>  
> @@ -847,7 +847,7 @@ config CMD_UBI
>  	tristate "Enable UBI - Unsorted block images commands"
>  	select CRC32
>  	select MTD_UBI
> -	default y if ARCH_SUNXI
> +	default y if NAND_SUNXI
>  	help
>  	  UBI is a software layer above MTD layer which admits use of LVM-like
>  	  logical volumes on top of MTD devices, hides some complexities of
> @@ -862,7 +862,7 @@ config CMD_UBIFS
>  	select CRC32
>  	select RBTREE if ARCH_SUNXI
>  	select LZO if ARCH_SUNXI
> -	default y if ARCH_SUNXI
> +	default y if NAND_SUNXI
>  	help
>  	  UBIFS is a file system for flash devices which works on top of UBI.
>  
> 

[U-Boot] [PATCH 2/2] cmd: nand: Make the NAND options default to NAND_SUNXI

[Maxime Ripard]

Hi,

On Mon, Mar 13, 2017 at 12:37:36AM +0000, André Przywara wrote:
> On 03/03/17 14:32, Maxime Ripard wrote:
> > If we depend on the ARCH_SUNXI configuration option, the boards that do not
> > have NAND support enabled (with the associated options) will not compile
> > anymore.
> > 
> > Depend on the NAND driver configuration option to make sure that is not the
> > case.
> 
> 
> > Reported-by: Chen-Yu Tsai <wens@csie.org>
> > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> 
> As I came up with an identical patch today to fix sunxi/master:
> 
> Reviewed-by: Andre Przywara <andre.przywara@arm.com>
> 
> 
> So this situation is a bit unfortunate: The sunxi/master branch was
> updated apparently before even compile-testing it for more than one
> sunxi board (it is broken for every sunxi board except Chip Pro at the
> moment: buildman says 105/106). And even after reports and the
> corresponding fixes appeared (thanks Chen-Yu and Maxime!) the branch is
> _still_ broken as of today (10 days after the fix was posted).
> 
> So can we please have both
> a) a better quality assurance before merging patches into this custodian
> branch and
> b) quicker merging of fixes?
> To my understanding u-boot-sunxi should be always a good and working
> candidate for anyone interested in Allwinner boards.
> 
> For a) it would already help to just simply build test sunxi:
> $ tools/buildman/buildman sunxi
> but probably extend this to cover at least all ARM boards.

Jagan usually takes the patches, but he seems to be off, and I was in
holidays last week.

I just pushed it, sorry for the breakage.

Maxime

-- 
Maxime Ripard, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20170320/260407cc/attachment.sig>

[U-Boot] [PATCH 2/2] cmd: nand: Make the NAND options default to NAND_SUNXI

[Jagan Teki]

On Tuesday 21 March 2017 12:22 AM, Maxime Ripard wrote:
> Hi,
>
> On Mon, Mar 13, 2017 at 12:37:36AM +0000, André Przywara wrote:
>> On 03/03/17 14:32, Maxime Ripard wrote:
>>> If we depend on the ARCH_SUNXI configuration option, the boards that do not
>>> have NAND support enabled (with the associated options) will not compile
>>> anymore.
>>>
>>> Depend on the NAND driver configuration option to make sure that is not the
>>> case.
>>
>>
>>> Reported-by: Chen-Yu Tsai <wens@csie.org>
>>> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
>>
>> As I came up with an identical patch today to fix sunxi/master:
>>
>> Reviewed-by: Andre Przywara <andre.przywara@arm.com>
>>
>>
>> So this situation is a bit unfortunate: The sunxi/master branch was
>> updated apparently before even compile-testing it for more than one
>> sunxi board (it is broken for every sunxi board except Chip Pro at the
>> moment: buildman says 105/106). And even after reports and the
>> corresponding fixes appeared (thanks Chen-Yu and Maxime!) the branch is
>> _still_ broken as of today (10 days after the fix was posted).
>>
>> So can we please have both
>> a) a better quality assurance before merging patches into this custodian
>> branch and
>> b) quicker merging of fixes?
>> To my understanding u-boot-sunxi should be always a good and working
>> candidate for anyone interested in Allwinner boards.
>>
>> For a) it would already help to just simply build test sunxi:
>> $ tools/buildman/buildman sunxi
>> but probably extend this to cover at least all ARM boards.
>
> Jagan usually takes the patches, but he seems to be off, and I was in
> holidays last week.

Sorry, I was off with holidays, will active and look for all.

thanks!
-- 
Jagan Teki
Free Software Engineer | www.openedev.com
U-Boot, Linux | Upstream Maintainer
Hyderabad, India.