These power supply issues are quite fun, like other things. The mistake would be to say that, because there is noise, coreboot is wrong and vendor firmware is right. It can be far more subtle than that. It could even be the case that coreboot is right, the vendor code is wrong, and the correction is eliciting noise because coreboot has revealed some other problem.
I can't tell you how many times we've looked at, e.g., u-boot for some ARM, realized we had no idea why the code in question (vendor code) ever worked, and fixed it. And found some other issue, somewhere else, because of the fix.
ron