[U-Boot] [PATCH v2 2/9] dm: core: Move dev_get_addr() etc. into a separate file

Simon Glass sjg at chromium.org
Wed May 17 10:09:22 UTC 2017


Hi Masahiro,

On 16 May 2017 at 04:04, Masahiro Yamada <yamada.masahiro at socionext.com> wrote:
> Hi Simon,
>
>
> 2017-05-02 0:18 GMT+09:00 Simon Glass <sjg at chromium.org>:
>> Move this group of address-related functions into a new file. These use
>> the flat device tree. Future work will provide new versions of these which
>> can support the live tree.
>>
>> Signed-off-by: Simon Glass <sjg at chromium.org>
>
>
>
>
>> diff --git a/include/dm.h b/include/dm.h
>> index a179c8a6e3..84f789d807 100644
>> --- a/include/dm.h
>> +++ b/include/dm.h
>> @@ -8,6 +8,7 @@
>>  #define _DM_H_
>>
>>  #include <dm/device.h>
>> +#include <dm/fdtaddr.h>
>>  #include <dm/platdata.h>
>>  #include <dm/uclass.h>
>
>
> I guess your motivation for 1/9 is this.

Yes, plus the fact that later is changes again.

>
> If you force all drivers to include these 4 headers,
> what is the point of splitting?

I can change this inclusion later. The main point is to reduce the
work required to introduce livetree to something I can bear.

> No, please do not do this.
>
> Please drop 1/9 and,
> add #include <dm/fdtaddr.h> where it is necessary.
>
> I know this is painful, but onetime pain.

It is two-time pain so far as I know, and I suspect three. This is
just the wrong place to be including a header that is only needed if
you are not using live tree. These sorts of details will drive
everyone mad.

>
> Perhaps you can use coccinelle for the conversion?

I have not used that tool, but it looks interesting. Have you used it?

Perhaps we can just let this be and come back to it when everything is stable.

Regards,
Simon


More information about the U-Boot mailing list