Patch Set 4:

Patch Set 4:

Patrick, Aaron, Nico: Can you share your opinions on my comments in patch set 4, wrt. using Kconfig symbols as commandline parameters vs. deriving necessary data from ELF headers. And in general, about the placement of amdcompress under util/cbfstool and not integrating it with amdfwtool. Thanks.

It was my understanding that Marshall is updating the patchset further. As for amdfwtool amdcompress I'm not familiar with the requirements currently. However, the compression requirements are orthogonal to just building the bootblock correctly. If the bootblock is linked correctly I would expect we could pull symbols out for driving utilities. However, we should be aware that we would need to mark this memory (where bootblock resides) as reserved. Typically we do that through cbmem and ensuring these programs live in cbmem. However, bootblock wouldn't have cbmem up yet so we need to ensure the use memory is reserved (both bootblock and romstage) in order to handle everything correctly for s3 transit.

My review was not about runtime operation at all, but complaints about the introduced redundancy, where values that are already present in ELF headers are regenerated in Makefile.inc from the Kconfig variables.

View Change

To view, visit change 37490. To unsubscribe, or for help writing mail filters, visit settings.