[U-Boot] [PATCH v4] POST cleanup.

[Michael Zaidman]

On Mon, May 10, 2010 at 2:17 PM, Detlev Zundel <dzu at denx.de> wrote:
> Hi Michael,
>
>> Combine previously submitted POST cleanup patch series
>> into single patch. Fix plain POST support introduced recently
>> for mpc812x arch.
>>
>> - Revives POST for blackfin arch;
>> - Removes redundant code:
>>      arch/blackfin/lib/post.c
>>      board/ssv/common/post.c
>>      arch/powerpc/cpu/ppc4xx/commproc.c
>>      arch/powerpc/cpu/mpc512x/common.c
>> - fixes up the post_word_{load|store} usage;
>
> Thanks for the big work.
>
> It would have been really nice if you put the maintainers on CC (simply
> include a CC: ... in the patch git-send-email does the rest) of the
> boards that you change config files.  This way they knew that there was
> a change which they should test and ack.
>
> For the powerpc parts, you have my
>
> Acked-by: Detlev Zundel <dzu at denx.de>
>
Ok, will be done.

>> diff --git a/board/barco/barco.c b/board/barco/barco.c
>> index c5fe8c4..2d01f1b 100644
>> --- a/board/barco/barco.c
>> +++ b/board/barco/barco.c
>> @@ -348,12 +348,3 @@ int serial_tstc (void)
>>  {
>>       return 0;
>>  }
>> -
>> -unsigned long post_word_load (void)
>> -{
>> -     return 0l;
>> -}
>> -void post_word_store (unsigned long val)
>> -{
>> -     return;
>> -}
>
> This never worked :)  The config file has no trace of any POST, so the
> change you do is pretty useless:

These changes are necessary for CONFIG_LOGBUFFER which barco board
defines that in turn requires the post_word_load and post_word_store
support.

>
>> diff --git a/include/configs/barco.h b/include/configs/barco.h
>> index e00f84a..b1af701 100644
>> --- a/include/configs/barco.h
>> +++ b/include/configs/barco.h
>> @@ -136,6 +136,8 @@
>>  #define CONFIG_LOGBUFFER
>>  #ifdef       CONFIG_LOGBUFFER
>>  #define CONFIG_SYS_STDOUT_ADDR               0x1FFC000
>> +#define CONFIG_SYS_POST_WORD_ADDR    \
>> +             (CONFIG_SYS_SDRAM_BASE + CONFIG_SYS_MAX_RAM_SIZE - 4)
>>  #else
>>  #define CONFIG_SYS_STDOUT_ADDR               0x2B9000
>>  #endif
>
> But then again, it doesn't matter.

This matters for the CONFIG_LOGBUFFER support as I explained above.


Thanks,
Michael