mturney mturney has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/35896 )
Change subject: HACK sc7180: bootblock/verstage/romstage need to zero bss HACK ......................................................................
Patch Set 1:
(1 comment)
https://review.coreboot.org/c/coreboot/+/35896/1//COMMIT_MSG Commit Message:
https://review.coreboot.org/c/coreboot/+/35896/1//COMMIT_MSG@8 PS1, Line 8:
Could you please provide the reasoning here as to why this is needed?
As indicated this is a HACK patch and not expected to be reviewed. To your question, I have been trying to get the chroot build environment working for Trogdor and it behaves differently than the standalone environment we have been working with.
In this context, standalone refers to the "normal" coreboot build environment and chroot refers to the chromium chroot build environment. To get a headstart on trogdor development we have been building/testing code using standalone build environment.
Same patch-train builds in chroot environment but doesn't get out of bootblock. Symptom is bss sections are not initialized to 0, hence these patches. The changes were introduced one at a time, the bootblock.c change allowed code to reach verstage and go no further. The verstage.c change allowed code to reach romstage. Even with these HACK workarounds we can't get out of qclib, which is started by romstage, so the test stops here.
Even with these workarounds, there are other subtle issues, e.g. the console log messages in verstage and romstage are truncated, from what is seen with the standalone build.
I intend to create a buganizer ticket and attach console log output, etc. This patch was pushed upstream to share with Julius and other googlers who may have some insight into why the chroot build appears broken while the standalone build allows boot to kernel login prompt.