[L] Change in coreboot[master]: drivers/intel/fsp2_0: Add API to convert BMP images to GOP BLT buffer

[Subrata Banik (Code Review)]

Attention is currently required from: Andrey Petrov, Kapil Porwal, Nick Vaccaro, Paul Menzel.

Subrata Banik has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/76921?usp=email )

Change subject: drivers/intel/fsp2_0: Add API to convert BMP images to GOP BLT buffer
......................................................................

Patch Set 4:

(2 comments)

Commit Message:

https://review.coreboot.org/c/coreboot/+/76921/comment/29139f9c_bec126ea :
PS3, Line 18: It also adds a
            : new config (BMP_LOGO_TO_GOP_BLT) to ensure backward compatibility with
            : older generation FSP.
> > UPDs are different between just passing BMP image to FSP vs GOP BLT to FSP.
>
> Can this be better modeled?

Subjected to the Intel's FSP design which I haven't seen changed in last several generations before introducing the BMP to BLT buffer change request in Alder Lake. Hence, in short, we don't have any control over it.
> > Are you saying pass the GOP BLT by default (if BMP_LOGO is enabled)?
>
> Almost. I mean, always built in the driver, but still run-time detect if BMP or BLT should be passed.

I guess that could have been done but the problem is we would be ended up compiling extra code "fsp_gop_blt.c" even for older platform where just passing the BMP is enough to get the splash screen. I have added the newer Kconfig to ensure older devices are not getting impacted with this new req of passing BLT buffer instead BMP directly.

I hope it clarifies why we need to maintain two different Kconfig option. The UPDs required to pass BLT buffer is not available on the older platform FSP hence, there wouldn’t be any value of calling that extra code.

File src/drivers/intel/fsp2_0/fsp_gop_blt.c:

PS3:
> I meant just a lot of the helper functions introduced here.

To make the code readable better. (thats the motivation i had with helper functions)
> Also, will “AMD FSP” ever use this?

AMD is using drivers/intel/fsp2_0 anyway hence, they should be able to use it if still using FSP. But as I said, this is SoC dependent code and not sure how cross-arch team will leverage that.