On 12/03/2022 08:02, Simon Glass wrote:
> On Thu, 10 Mar 2022 at 12:36, Alper Nebi Yasak <alpernebiya...@gmail.com>
> wrote:
>> On 06/03/2022 06:19, Simon Glass wrote:
>>> This boards uses SPL_FIT so does not need to support loading a raw image.
>>
>> This sounds OK to me, but...
>>
>>> Drop it to avoid binman trying to insert a symbol which has no value.
>>
>> I couldn't figure out how it leads to this in the code. I guess some
>> ifdefs or optimization steps dropping the binman symbols from the ELF file?
>
> Well the raw-image method uses a symbol in SPL which holds the
> image-pos of U-Boot. If we disable that, then the symbol is not used
> and we don't have to set it. See spl_ram_load_image().
Can you double-check this? Looks to me like the symbol isn't set both
before/after this patch only because there's no image using u-boot-spl
or u-boot-tpl entries. At u-boot-dm/fit-working, binman verbose output
shows symbols being set as a later patch adds such entries.
But, I did get u-boot-any symbols to disappear by disabling
CONFIG_{SPL,TPL}_BINMAN_SYMBOLS, and u-boot-spl symbols by editing
common/spl/spl.c to make that config's ifdef include those symbols.
