Hi Heiko,
On Wed, 31 Jul 2013 10:31:12 +0200, Heiko Schocher hs@denx.de wrote:
Hello Albert,
Am 31.07.2013 10:16, schrieb Albert ARIBAUD:
Hi Heiko,
On Wed, 31 Jul 2013 09:36:19 +0200, Heiko Schocherhs@denx.de wrote:
On the other hand, it may be hard to immediately know what functions throughout U-boot are safe to call from within board_init_f(); maybe we should start thinking about checking and marking these, the simplest way being to suffix them with "_f" once we have made sure they are safe to call from within board_init_f().
Hmmm... Maybe instead we should think (also in thinking common bring up for all boards) about:
getting rid of board_init_f in u-boot code, instead use for all boards spl code to init needed things and copy and relocate u-boot to ram in spl code ... so we have in u-boot no longer such restictions ... but thats just an idea which whirs in my head ... without thinking to deep in it.
But this approach would have some advantages ...
Well, the original SPL was basically board_init_f() plus some code to copy U-Boot from wherever it was to DDR, so it was tightly linked to board_init_f(). But... first, SPL has evolved into a "U-Boot lite" where much can happen beyond board_init_f() -- think Falcon mode, for instance -- and second, there are boards which do not have SPL at all, and their board_init_f() can thus not be "moved to SPL".
Hmm... all boards use board_init_f ... and spl do pieces from board_init_f So why should it not be possible to do all init things in spl code? Code beyond board_init_f is optional ...
It is, in the original "SPL is just board_init_f plus some copying" view. In the current "SPL is U-boot only not full-featured", it becomes false.
And yes, there are a lot of boards, which have no spl, but they can execute spl code (thinking of the lof of powerpc boards which booting from nor flash ... spl code can also run from nor ... and copy the u-boot piece of the image to ram, relocate it ...)
And yes, a side effect could be, that all boards can use Falcon boot mode.
;-)
So no, I don't think we can move U-Boot's design from "_f/_r" to "SPL/U-Boot".
I am not sure ...
I see this approach of likening SPL to _f and U-boot to _r as forcing a dual-binary model onto all boards whereas not all boards require it. I prefer a model where _f can exist, _r can exist, and for each target, the maintainer decides which binaries are built and for each one, whether _f and/or _r is present and what _r does.
Amicalement,