Attention is currently required from: Felix Singer, Tim Wawrzynczak, Angel Pons, Michael Niewöhner. Nico Huber has posted comments on this change. ( https://review.coreboot.org/c/libgfxinit/+/61211 )
Change subject: gma: Add `Tiger_Point` PCH ......................................................................
Patch Set 1:
(5 comments)
Commit Message:
https://review.coreboot.org/c/libgfxinit/+/61211/comment/9e00caf9_3e9e2b4d PS1, Line 12: - GMBUS (no idea, I'm even unable to verify BXT \o/)
linux: i915/display/intel_gmbus. […]
I'd hope they didn't change things this time. Usually, it would only be about clocking and maybe pin mapping.
https://review.coreboot.org/c/libgfxinit/+/61211/comment/28891d52_3bca2568 PS1, Line 13: - TBT / Type-C
I'll probably need some help here
If they did it right /o\ there'd be nothing to do. It should be handled by different hardware / layers. Something (AIUI often integrated into the PD controller) needs to decide to mux the lanes, DP-Aux channel and hot-plug signal to the display controller. And when that's done the display controller wouldn't have to know about it.
I didn't play with any actual implementations, though. So maybe I wish for too much sanity in the stack :)
File common/hw-gfx-gma-config.ads.template:
https://review.coreboot.org/c/libgfxinit/+/61211/comment/30263485_eb2fe96a PS1, Line 57: when Skylake => Cannon_Point); This would have to be changed back later when the CPU parts for Tiger Lake are added. Actually, I don't see the reason to change it in the first place?
https://review.coreboot.org/c/libgfxinit/+/61211/comment/1ae0c598_81cf5873 PS1, Line 168: Cannon_Point_On : <genbool> := Skylake_On and then PCH >= Cannon_Point; This definitely shouldn't change. It's supposed to mean `[From] Cannon Point On` iow. Cannon Point and later.
File common/hw-gfx-gma.ads:
https://review.coreboot.org/c/libgfxinit/+/61211/comment/1d4ab3e2_387d2111 PS1, Line 58: Cannon_Point comma? ;)