[U-Boot] [PATCH] u-boot: adjust config_cmd_default.h not to raise ton of warnings

[U-Boot] [PATCH] u-boot: adjust config_cmd_default.h not to raise ton of warnings

[Pavel Machek]

If there's duplicty between config system and config_cmd_default, a
ton of warnings is raised, because one uses plain defines, and other
defines it to 1. Adjust config_cmd_default.h not to provoke the
warnings.

Signed-off-by: Pavel Machek <pavel@denx.de>

diff --git a/include/config_cmd_default.h b/include/config_cmd_default.h
index e79a13b..2f60d63 100644
--- a/include/config_cmd_default.h
+++ b/include/config_cmd_default.h
@@ -16,29 +16,29 @@
  * hardware, not fully tested, etc.).
  */
 
-#define CONFIG_CMD_BDI		/* bdinfo			*/
-#define CONFIG_CMD_BOOTD	/* bootd			*/
-#define CONFIG_CMD_CONSOLE	/* coninfo			*/
-#define CONFIG_CMD_ECHO		/* echo arguments		*/
-#define CONFIG_CMD_EDITENV	/* editenv			*/
-#define CONFIG_CMD_ENV_EXISTS	/* query whether env variables exists */
-#define CONFIG_CMD_FPGA		/* FPGA configuration Support	*/
-#define CONFIG_CMD_IMI		/* iminfo			*/
-#define CONFIG_CMD_ITEST	/* Integer (and string) test	*/
+#define CONFIG_CMD_BDI 1	/* bdinfo			*/
+#define CONFIG_CMD_BOOTD 1	/* bootd			*/
+#define CONFIG_CMD_CONSOLE 1	/* coninfo			*/
+#define CONFIG_CMD_ECHO	1	/* echo arguments		*/
+#define CONFIG_CMD_EDITENV 1	/* editenv			*/
+#define CONFIG_CMD_ENV_EXISTS 1	/* query whether env variables exists */
+#define CONFIG_CMD_FPGA	1	/* FPGA configuration Support	*/
+#define CONFIG_CMD_IMI 1	/* iminfo			*/
+#define CONFIG_CMD_ITEST 1	/* Integer (and string) test	*/
 #ifndef CONFIG_SYS_NO_FLASH
-#define CONFIG_CMD_FLASH	/* flinfo, erase, protect	*/
-#define CONFIG_CMD_IMLS		/* List all found images	*/
+#define CONFIG_CMD_FLASH 1	/* flinfo, erase, protect	*/
+#define CONFIG_CMD_IMLS	1	/* List all found images	*/
 #endif
-#define CONFIG_CMD_LOADB	/* loadb			*/
-#define CONFIG_CMD_LOADS	/* loads			*/
-#define CONFIG_CMD_MEMORY	/* md mm nm mw cp cmp crc base loop */
-#define CONFIG_CMD_MISC		/* Misc functions like sleep etc*/
-#define CONFIG_CMD_NET		/* bootp, tftpboot, rarpboot	*/
-#define CONFIG_CMD_NFS		/* NFS support			*/
-#define CONFIG_CMD_RUN		/* run command in env variable	*/
-#define CONFIG_CMD_SAVEENV	/* saveenv			*/
-#define CONFIG_CMD_SETGETDCR	/* DCR support on 4xx		*/
-#define CONFIG_CMD_SOURCE	/* "source" command support	*/
-#define CONFIG_CMD_XIMG		/* Load part of Multi Image	*/
+#define CONFIG_CMD_LOADB 1	/* loadb			*/
+#define CONFIG_CMD_LOADS 1	/* loads			*/
+#define CONFIG_CMD_MEMORY 1	/* md mm nm mw cp cmp crc base loop */
+#define CONFIG_CMD_MISC	1	/* Misc functions like sleep etc*/
+#define CONFIG_CMD_NET 1	/* bootp, tftpboot, rarpboot	*/
+#define CONFIG_CMD_NFS 1	/* NFS support			*/
+#define CONFIG_CMD_RUN 1	/* run command in env variable	*/
+#define CONFIG_CMD_SAVEENV 1	/* saveenv			*/
+#define CONFIG_CMD_SETGETDCR 1	/* DCR support on 4xx		*/
+#define CONFIG_CMD_SOURCE 1	/* "source" command support	*/
+#define CONFIG_CMD_XIMG 1	/* Load part of Multi Image	*/
 
 #endif	/* _CONFIG_CMD_DEFAULT_H */

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[U-Boot] [PATCH] u-boot: adjust config_cmd_default.h not to raise ton of warnings

[Stephen Warren]

On 05/01/2015 03:14 AM, Pavel Machek wrote:
>
> If there's duplicty between config system and config_cmd_default, a
> ton of warnings is raised, because one uses plain defines, and other
> defines it to 1. Adjust config_cmd_default.h not to provoke the
> warnings.

I believe the correct way to fix this is to fix the board config files 
that cause the defines to be duplicated.

[U-Boot] [PATCH] u-boot: adjust config_cmd_default.h not to raise ton of warnings

[Pavel Machek]

On Fri 2015-05-01 09:24:02, Stephen Warren wrote:
> On 05/01/2015 03:14 AM, Pavel Machek wrote:
> >
> >If there's duplicty between config system and config_cmd_default, a
> >ton of warnings is raised, because one uses plain defines, and other
> >defines it to 1. Adjust config_cmd_default.h not to provoke the
> >warnings.
> 
> I believe the correct way to fix this is to fix the board config files that
> cause the defines to be duplicated.

Yes, I'm doing that, too (patch is waiting for Marek to be applied),
but that does not really fix the problem -- it really should be
consistent; otherwise we'll hit subtle bugs sooner or later.
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[U-Boot] [PATCH] u-boot: adjust config_cmd_default.h not to raise ton of warnings

[Stephen Warren]

On 05/03/2015 04:01 AM, Pavel Machek wrote:
> On Fri 2015-05-01 09:24:02, Stephen Warren wrote:
>> On 05/01/2015 03:14 AM, Pavel Machek wrote:
>>>
>>> If there's duplicty between config system and config_cmd_default, a
>>> ton of warnings is raised, because one uses plain defines, and other
>>> defines it to 1. Adjust config_cmd_default.h not to provoke the
>>> warnings.
>>
>> I believe the correct way to fix this is to fix the board config files that
>> cause the defines to be duplicated.
> 
> Yes, I'm doing that, too (patch is waiting for Marek to be applied),
> but that does not really fix the problem -- it really should be
> consistent; otherwise we'll hit subtle bugs sooner or later.

I don't believe consistency requires complicating the defaults file.

[U-Boot] [PATCH] u-boot: adjust config_cmd_default.h not to raise ton of warnings

[Pavel Machek]

On Sun 2015-05-03 10:40:40, Stephen Warren wrote:
> On 05/03/2015 04:01 AM, Pavel Machek wrote:
> > On Fri 2015-05-01 09:24:02, Stephen Warren wrote:
> >> On 05/01/2015 03:14 AM, Pavel Machek wrote:
> >>>
> >>> If there's duplicty between config system and config_cmd_default, a
> >>> ton of warnings is raised, because one uses plain defines, and other
> >>> defines it to 1. Adjust config_cmd_default.h not to provoke the
> >>> warnings.
> >>
> >> I believe the correct way to fix this is to fix the board config files that
> >> cause the defines to be duplicated.
> > 
> > Yes, I'm doing that, too (patch is waiting for Marek to be applied),
> > but that does not really fix the problem -- it really should be
> > consistent; otherwise we'll hit subtle bugs sooner or later.
> 
> I don't believe consistency requires complicating the defaults file.

I do. It already causes problems.
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[U-Boot] [PATCH] u-boot: adjust config_cmd_default.h not to raise ton of warnings

[Pavel Machek]

On Fri 2015-05-01 11:14:11, Pavel Machek wrote:
> 
> If there's duplicty between config system and config_cmd_default, a
> ton of warnings is raised, because one uses plain defines, and other
> defines it to 1. Adjust config_cmd_default.h not to provoke the
> warnings.
> 
> Signed-off-by: Pavel Machek <pavel@denx.de>

Ping? Yamada, can you apply this? Tom?

								Pavel

> diff --git a/include/config_cmd_default.h b/include/config_cmd_default.h
> index e79a13b..2f60d63 100644
> --- a/include/config_cmd_default.h
> +++ b/include/config_cmd_default.h
> @@ -16,29 +16,29 @@
>   * hardware, not fully tested, etc.).
>   */
>  
> -#define CONFIG_CMD_BDI		/* bdinfo			*/
> -#define CONFIG_CMD_BOOTD	/* bootd			*/
> -#define CONFIG_CMD_CONSOLE	/* coninfo			*/
> -#define CONFIG_CMD_ECHO		/* echo arguments		*/
> -#define CONFIG_CMD_EDITENV	/* editenv			*/
> -#define CONFIG_CMD_ENV_EXISTS	/* query whether env variables exists */
> -#define CONFIG_CMD_FPGA		/* FPGA configuration Support	*/
> -#define CONFIG_CMD_IMI		/* iminfo			*/
> -#define CONFIG_CMD_ITEST	/* Integer (and string) test	*/
> +#define CONFIG_CMD_BDI 1	/* bdinfo			*/
> +#define CONFIG_CMD_BOOTD 1	/* bootd			*/
> +#define CONFIG_CMD_CONSOLE 1	/* coninfo			*/
> +#define CONFIG_CMD_ECHO	1	/* echo arguments		*/
> +#define CONFIG_CMD_EDITENV 1	/* editenv			*/
> +#define CONFIG_CMD_ENV_EXISTS 1	/* query whether env variables exists */
> +#define CONFIG_CMD_FPGA	1	/* FPGA configuration Support	*/
> +#define CONFIG_CMD_IMI 1	/* iminfo			*/
> +#define CONFIG_CMD_ITEST 1	/* Integer (and string) test	*/
>  #ifndef CONFIG_SYS_NO_FLASH
> -#define CONFIG_CMD_FLASH	/* flinfo, erase, protect	*/
> -#define CONFIG_CMD_IMLS		/* List all found images	*/
> +#define CONFIG_CMD_FLASH 1	/* flinfo, erase, protect	*/
> +#define CONFIG_CMD_IMLS	1	/* List all found images	*/
>  #endif
> -#define CONFIG_CMD_LOADB	/* loadb			*/
> -#define CONFIG_CMD_LOADS	/* loads			*/
> -#define CONFIG_CMD_MEMORY	/* md mm nm mw cp cmp crc base loop */
> -#define CONFIG_CMD_MISC		/* Misc functions like sleep etc*/
> -#define CONFIG_CMD_NET		/* bootp, tftpboot, rarpboot	*/
> -#define CONFIG_CMD_NFS		/* NFS support			*/
> -#define CONFIG_CMD_RUN		/* run command in env variable	*/
> -#define CONFIG_CMD_SAVEENV	/* saveenv			*/
> -#define CONFIG_CMD_SETGETDCR	/* DCR support on 4xx		*/
> -#define CONFIG_CMD_SOURCE	/* "source" command support	*/
> -#define CONFIG_CMD_XIMG		/* Load part of Multi Image	*/
> +#define CONFIG_CMD_LOADB 1	/* loadb			*/
> +#define CONFIG_CMD_LOADS 1	/* loads			*/
> +#define CONFIG_CMD_MEMORY 1	/* md mm nm mw cp cmp crc base loop */
> +#define CONFIG_CMD_MISC	1	/* Misc functions like sleep etc*/
> +#define CONFIG_CMD_NET 1	/* bootp, tftpboot, rarpboot	*/
> +#define CONFIG_CMD_NFS 1	/* NFS support			*/
> +#define CONFIG_CMD_RUN 1	/* run command in env variable	*/
> +#define CONFIG_CMD_SAVEENV 1	/* saveenv			*/
> +#define CONFIG_CMD_SETGETDCR 1	/* DCR support on 4xx		*/
> +#define CONFIG_CMD_SOURCE 1	/* "source" command support	*/
> +#define CONFIG_CMD_XIMG 1	/* Load part of Multi Image	*/
>  
>  #endif	/* _CONFIG_CMD_DEFAULT_H */
> 

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[U-Boot] [PATCH] u-boot: adjust config_cmd_default.h not to raise ton of warnings

[Masahiro Yamada]

Hi Pavel,

2015-05-15 17:36 GMT+09:00 Pavel Machek <pavel@denx.de>:
> On Fri 2015-05-01 11:14:11, Pavel Machek wrote:
>>
>> If there's duplicty between config system and config_cmd_default, a
>> ton of warnings is raised, because one uses plain defines, and other
>> defines it to 1. Adjust config_cmd_default.h not to provoke the
>> warnings.
>>
>> Signed-off-by: Pavel Machek <pavel@denx.de>
>
> Ping? Yamada, can you apply this? Tom?


I did not know this patch was delegated to me.
At least, I see it is marked with trini.

I am OK with this patch, but more inclined to agree with Stephen's comment:
"I believe the correct way to fix this is to fix the board config
files that cause the defines to be duplicated"


Once you start to enable CONFIG_CMD_* in Kconfig, you should not include
config_cmd_default.h.



-- 
Best Regards
Masahiro Yamada