Change in coreboot[master]: [RFC] device: Introduce new method for setting device states

[Nico Huber (Code Review)]

Attention is currently required from:  Felix Singer, Raul Rangel, Furquan Shaikh, Matt DeVillier, Angel Pons, Michael Niewöhner, Patrick Rudolph, Felix Held.
Nico Huber has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/52493 )

Change subject: [RFC] device: Introduce new method for setting device states
......................................................................

Patch Set 17:

(2 comments)

File src/soc/intel/skylake/chip.c:

https://review.coreboot.org/c/coreboot/+/52493/comment/a14e55ed_3b35eb25 
PS13, Line 461: if (!xdci_can_enable())
              : 		params->XdciEnable = 0;
> Agreed, I simply missed that we set it before. […]

Oh, no. Now you triggered the software engineer in me...

const struct device_enable_map devmap[] = {
            { &params->XdciEnable,  PCH_DEVFN_USBOTG, .filter = xdci_can_enable },
            { &params->SomeDisable, PCH_DEVFN_STH,    .filter = devmap_invert },
    };

i.e. we'd actually only need a single abstract post-processing step.

File src/soc/intel/skylake/fspdevmap.c:

https://review.coreboot.org/c/coreboot/+/52493/comment/e800e28a_ffae8f35 
PS13, Line 15: devmap[i].inverted ^ dev->enabled
> `works like` != `is the same` ;) I don't really care... […]

Please don't squabble. I merely wanted to point out that there's
also an argument for the current order `inverted ^ enabled`. I
don't think there is a single correct answer. This is something
every author should decide for themself.