Attention is currently required from: Maulik V Vaghela, Paul Menzel, Mario Scheithauer, Subrata Banik, Arthur Heymans, Lean Sheng Tan, Werner Zeh, Patrick Rudolph. Angel Pons has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/55367 )
Change subject: soc/intel/elkhartlake: Introduce Intel PSE ......................................................................
Patch Set 43:
(1 comment)
File src/soc/intel/elkhartlake/fsp_params.c:
https://review.coreboot.org/c/coreboot/+/55367/comment/02768196_636116e8 PS43, Line 373: CONFIG_PSE_BUFFER_SIZE
This is a good idea. […]
When possible, build-time checks are preferred over run-time checks.
From the user's perspective, it's much better to get a "this won't work" error when building than to notice something isn't working after flashing coreboot. Especially when externally reflashing isn't easy (e.g. hard-to-flash board, no suitable programmer readily available...). Even if externally reflashing is possible, getting coreboot logs out of a non-booting machine is tedious (flashconsole exists, but can cause issues), and then one needs to find the error message in the log. When this happens, users contact the coreboot community (most often via IRC) and typically get help from developers. Build-time checks avoid all the pain and misery of troubleshooting, so helping users pretty much boils down to deciphering the error message.
TL;DR: Using build-time checks instead of run-time checks where possible makes helping users easier and avoids wasting developers' time troubleshooting run-time issues caused by things known to be wrong at build-time.
-
Angel Pons (Code Review)