[coreboot-gerrit] Change in coreboot[master]: binaryPI: Fix UMA calculations

[Marshall Dawson (Code Review)]

Marshall Dawson has posted comments on this change. ( https://review.coreboot.org/19346 )

Change subject: binaryPI: Fix UMA calculations
......................................................................


Patch Set 2:

(2 comments)

https://review.coreboot.org/#/c/19346/2/src/northbridge/amd/pi/agesawrapper.c
File src/northbridge/amd/pi/agesawrapper.c:

Line 152: 	backup_top_of_ram(PostParams->MemConfig.Sub4GCacheTop);
> UMA if it is enabled to be located below 4GiB. If there is no memory above 
Why only Sub4GCacheTop and not a calculation that also uses UmaBase?  This is causing a problem currently.  It sets top of memory to be higher than the UMA address.

4GB example: Sub4GCacheTop=d0000000 and UmaBase=b0000000


https://review.coreboot.org/#/c/19346/2/src/northbridge/amd/pi/ramtop.c
File src/northbridge/amd/pi/ramtop.c:

PS2, Line 20: 0x78
> 32bit scratchpad. Unused in vendorcode as far as git grep can be trusted.
I have some reservations here.  This register hasn't been documented, even in NDA materials, since Family 14h.  What about D1F0x108 instead?


-- 
To view, visit https://review.coreboot.org/19346
To unsubscribe, visit https://review.coreboot.org/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I0c375e5da0dfef6cef0c50272356cd32a87b1ff6
Gerrit-PatchSet: 2
Gerrit-Project: coreboot
Gerrit-Branch: master
Gerrit-Owner: Kyösti Mälkki <kyosti.malkki at gmail.com>
Gerrit-Reviewer: Aaron Durbin <adurbin at chromium.org>
Gerrit-Reviewer: Kyösti Mälkki <kyosti.malkki at gmail.com>
Gerrit-Reviewer: Marc Jones <marc at marcjonesconsulting.com>
Gerrit-Reviewer: Marshall Dawson <marshalldawson3rd at gmail.com>
Gerrit-Reviewer: Paul Menzel <paulepanter at users.sourceforge.net>
Gerrit-Reviewer: build bot (Jenkins)
Gerrit-HasComments: Yes