[U-Boot] [U-Boot, v4, 60/66] rockchip: Kconfig: preset TPL_LDSCRIPT via Kconfig for the RK3368

[Dr. Philipp Tomsich]

Masahiro & Tom,

the proposed fix is at
	https://patchwork.ozlabs.org/patch/799172/

Thanks,
Phil.

> On 07 Aug 2017, at 16:05, Tom Rini <trini at konsulko.com> wrote:
> 
> On Mon, Aug 07, 2017 at 10:48:12AM +0200, Dr. Philipp Tomsich wrote:
>> +Tom
>> 
>> Ok, so the problem is 'scripts/setlocalversion’, which does the following:
>>> if test -e include/config/auto.conf; then
>>>        . include/config/auto.conf
>>> else
>>>        echo "Error: kernelrelease not valid - run 'make prepare' to update it"
>>>        exit 1
>>> fi
>> in order to access the variables needed for:
>>> # CONFIG_LOCALVERSION and LOCALVERSION (if set)
>>> res="${res}${CONFIG_LOCALVERSION}${LOCALVERSION}"
>>> 
>>> # scm version string if not at a tagged commit
>>> if test "$CONFIG_LOCALVERSION_AUTO" = "y"; then
>>>        # full scm version string
>>>        res="$res$(scm_version)"
>>> else
>>>        # append a plus sign if the repository is not in a clean
>>>        # annotated or signed tagged state (as git describe only
>>>        # looks at signed or annotated tags - git tag -a/-s) and
>>>        # LOCALVERSION= is not specified
>>>        if test "${LOCALVERSION+set}" != "set"; then
>>>                scm=$(scm_version --short)
>>>                res="$res${scm:++}"
>>>        fi
>>> fi
>> 
>> So nothing wrong with the changes to Kconfig, but we’re triggering a weakness in our
>> build infrastructure here.
>> 
>> I guess we’ll have to implement a save version of scripts/localversion that does not
>> allow for the execution of arbitrary script code from auto.conf ...
> 
> Masahiro, do you have any ideas here?  Thanks!
> 
>> 
>>> On 07 Aug 2017, at 10:38, Andy Yan <andy.yan at rock-chips.com> wrote:
>>> 
>>> Hi Philipp:
>>> 
>>> 
>>> On 2017年08月07日 16:24, Dr. Philipp Tomsich wrote:
>>>> Andy,
>>>> 
>>>> Doing a non-buildman defconfig + make works fine.
>>>> I’ve seen this too (I did a full buildman for all arm, x86 and powerpc boards) and it seems as if a shell-script is trying to source the generated auto.config file.
>>>> I couldn’t spot where in buildman that happens, so I’ll need to leave this to someone with more insight into buildman to figure out…
>>> 
>>> 
>>>   I build with : make ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- firefly-rk3288_defconfig all
>>> Also can see the waring: inclue/config/auto.conf: line 209: ARCH: command not found.
>>>> 
>>>> Regards,
>>>> Phil.
>>>> 
>>>>> On 07 Aug 2017, at 03:19, Andy Yan <andy.yan at rock-chips.com> wrote:
>>>>> 
>>>>> Hi :
>>>>> 
>>>>>   When run buildman rockchip on the master branch of current u-boot-rockchip tree, I got warnings about 16 rockchip arm32 based boards:
>>>>> 
>>>>>  include/config/auto.config: line x: ARCH: command not found.
>>>>> 
>>>>>  the line x is the definition of the: SPL_LDSCRITP "arch/$(ARCH)/cpu/u-boot-spl.lds"
>>>>> 
>>>>>  not sure is there somethings wrong about this series change.
>>>>> 
>>>>> 
>>>>> On 2017年08月07日 01:18, Philipp Tomsich wrote:
>>>>>>> Set TPL_LDSCRIPT in Kconfig, so we don't have to pollute our
>>>>>>> header file.
>>>>>>> 
>>>>>>> Signed-off-by: Philipp Tomsich <philipp.tomsich at theobroma-systems.com>
>>>>>>> Reviewed-by: Simon Glass <sjg at chromium.org>
>>>>>>> ---
>>>>>>> 
>>>>>>> Changes in v4:
>>>>>>> - (added) sets TPL_LDSCRIPT via Kconfig
>>>>>>> 
>>>>>>> Changes in v3: None
>>>>>>> Changes in v2: None
>>>>>>> 
>>>>>>> arch/arm/mach-rockchip/Kconfig  | 7 +++++++
>>>>>>> include/configs/rk3368_common.h | 2 --
>>>>>>> 2 files changed, 7 insertions(+), 2 deletions(-)
>>>>>>> 
>>>>>> Applied to u-boot-rockchip, thanks!
>>>>>> 
>>>>>> 
>>>>>> 
>>>>> 
>>>> 
>>>> 
>>>> 
>>> 
>>> 
>> 
> 
> -- 
> Tom