[coreboot-gerrit] [M] Change in coreboot[master]: util/sconfig: Include optional maiboard chip header

[Tim Van Patten (Code Review)]

Attention is currently required from: Raul Rangel, Martin L Roth, Jon Murphy, Arthur Heymans, Angel Pons, Karthik Ramasubramanian, Felix Held.

Tim Van Patten has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/74736 )

Change subject: util/sconfig: Include optional maiboard chip header
......................................................................

Patch Set 3:

(2 comments)

File util/sconfig/main.c:

https://review.coreboot.org/c/coreboot/+/74736/comment/70f0f5ed_d8ba7ad0 
PS3, Line 1402: &chip_h[strlen("src/")]
> Mainboard path is got from devicetree path which in turn always contains "src/"

Contains or starts with "src/"?

If it's just 'contains', then this is (essentially) starting at a random location within `mainboard_path`, since the path doesn't start with "src/". If it's 'starts', then please add a comment stating why that's guaranteed to be the case, so we always get a good path starting 4 chars into the string.

https://review.coreboot.org/c/coreboot/+/74736/comment/80f0d4aa_7908ac97 
PS3, Line 1405: 	free(mainboard_path);
> I think the memory location immediately preceding the allocated/freed address will have some metadat […]

Everything I'm reading states that double-freeing is undefined behavior (at best) and corrupts the memory allocation structures at worst.

Where do you see that it's safe to double-free memory?