[coreboot-gerrit] [M] Change in coreboot[main]: treewide: convert to %#x hex prints

[Jon Murphy (Code Review)]

Attention is currently required from: Christian Walter, Erik van den Bogaert, Felix Held, Frans Hendriks, Fred Reitberger, Jan Samek, Jason Glenesk, Julius Werner, Karthik Ramasubramanian, Matt DeVillier, Nick Vaccaro, Subrata Banik, Tim Van Patten, Yu-Ping Wu.

Jon Murphy has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/78183?usp=email )

Change subject: treewide: convert to %#x hex prints
......................................................................

Patch Set 3:

(1 comment)

File src/drivers/i2c/tpm/tpm.c:

https://review.coreboot.org/c/coreboot/+/78183/comment/9e6e100d_f5c55f04 :
PS3, Line 537: %#x
> I don't know of the reason. I hope there are no downstream utilities parsing this log message. […]

The leading qualifier is '0x' so the only thing that would be capitalized is the printed hex digits. So the only thing that could be matching that since it's variable is a regex(that only matches upper case) or hardcoded check for a specific device ID from a printk which feels incredibly fragile,

If we were to break something as a result, I would contend that that's the broken code.  String sanitation should always be the first step in parsing.

I've searched for 'device-id 0x' in code search and I do see kernel drivers and depthcharge using the capitalized hex digits, but u-boot uses lower case.  All of the test data uses device IDs 0xB, 0x1A, and 0x1B, which are obviously capitalized.  At least for ChromeOS, I don't perceive this to be an issue.

I don't have a strong preference other than consistency, this is inconsistent(both across downstream code bases and within coreboot).  I'm concerned that there's so frequently push back on changes like this.  If we aren't willing to iterate and update the code and attempt to be more consistent and clean, I'm not sure how we can hope to improve.