Change in coreboot[master]: soc/intel/common/block/notify: Implement coreboot notify native driver

[Subrata Banik (Code Review)]

Attention is currently required from: Arthur Heymans, Felix Singer, Tim Wawrzynczak, Angel Pons, Sridhar Siricilla, Lean Sheng Tan, Werner Zeh, Patrick Rudolph, EricR Lai.
Subrata Banik has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/60405 )

Change subject: soc/intel/common/block/notify: Implement coreboot notify native driver
......................................................................

Patch Set 19:

(1 comment)

Commit Message:

https://review.coreboot.org/c/coreboot/+/60405/comment/5cd8bf8c_e7820006 
PS19, Line 7: soc/intel/common/block/notify: Implement coreboot notify native driver
> And getting fixes in coreboot code is generally faster than getting fixes in FSP ^^

Here is my motivation apart from what Arthur, Angel and Sheng had mentioned.

1. Boot time impact, sending EOP from FSP takes ~35ms on all Intel platforms (I have worked between SKL till ADL) until we had to refactor the code into coreboot and implement the logic wisely understanding the timing impact, it took ~3ms.
2. Reduce the closed source binaries and specifically the optional one. Do you know one can boot to an OS even after skip calling FSP-S and onwards all APIs (including FSP APIs)? What is the downside of this design? Missing SoC recommended chipset programming. These patch sets would like to bring those recommended programming (as documented in EDS/FAS) in open-source coreboot so we know what is the downside.
3. Maintenance on ADL, after UPD refactoring, FSP missed sending EOP (as default was sending EOP from DXE). Who will debug such an issue and tap on FSP UPD. Having code in coreboot would help more eyes to catch such issues and reduce YoY maintenance.
4. Leveraging more common code, all code being implemented as part of these patch sets are just leveraging existing IA common code libraries effectively without crossing any border between FSP close source and coreboot open source.

We were enough motivated in past to make the FSP-T optional and we would like to empower coreboot in more possible ways.

If your concern is redundancy then there are always ways to fix that, isn’t it.