[M] Change in coreboot[master]: drivers/intel/gma: Hook up libgfxinit in romstage

[Jérémy Compostella (Code Review)]

Attention is currently required from: Bora Guvendik, Zhixing Ma, Anil Kumar K, Hannah Williams, Cliff Huang, Tarun Tuli, Nico Huber, Michał Żygowski, Subrata Banik, Paul Menzel, Angel Pons, Arthur Heymans.

Jérémy Compostella has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/70276 )

Change subject: drivers/intel/gma: Hook up libgfxinit in romstage
......................................................................

Patch Set 23:

(1 comment)

File src/soc/intel/common/block/graphics/graphics.c:

https://review.coreboot.org/c/coreboot/+/70276/comment/6deb6e49_3c446562 
PS23, Line 181: 	if (CONFIG(HWBASE_STATIC_MMIO)) {
> > Not really as the GOP driver uses its own BAR address anyway which is way more into the middle the 2GB memory hole.
> > [INFO]:[InitializeGfxMmioBase()]:[GfxMmio Base address: 0xAF000000]
>
> Is this even when we are overriding GtMmAdd UPD with 0xfa00_0000. if yes, then this is wrong. FSP/GFX PEIM should reuse the IGD base and avoid overriding.

You are correct, I looked at old logs without this UPD override and I assumed that since we are freeing this resources end of romstage it was gone.

But I am very curious, since we are clearing BAR 0 between romstage and ramstage, how does this address persist. Is there sone FSP-M UPD data persistence I am unaware of ? I can't see how this address is being saved. The FSP-M UPD is on stack in `do_memory_init`. Is is saved somewhere in memory by FSP-M for FSP-S to consume later ?

@Tarun, I don't think we should bind it to `RUN_FSP_GOP` flag but bind it to libgfxinit like instead:

```
if (CONFIG(MAINBOARD_USE_LIBGFXINIT) && CONFIG(HWBASE_STATIC_MMIO)) {
[...]
```
What do you think ?