Conversation
As pointed out on an irc channel, yet another example of kernel developers having to do crazy things to paper over hardware bugs: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=f076ef44a44d02ed91543f820c14c2c7dff53716
5
39
42

@gregkh a guestimate, how many if this kind of kernel fixes for hardware bugs do you think there is in the kernel? Would you say they are too "common"?

1
0
0

@gregkh though doesn't seem to future proof according the comment "will work up to 2381" ;)

2
0
0

@wagi @gregkh if this rockchip still runs in 2300s we have different problems.

0
0
0

@gregkh incredible

0
0
0

@wagi @gregkh
I wasn't sure if that's 2381 rockchip year or 2381 CE, as the difference will be longer than a year by then.

1
0
0

@gregkh They just decided to add an extra day to November (edit: instead of fixing what I assume the calendar is in firmware)? How common is this?

0
0
0

@grimthorpe @wagi @gregkh Good catch! I think 2381 will be the year when the difference is more than a year? 2016 + 365. So it looks like it's accurate up to the year where the years are still the same, and will be inaccurate at the point they differ.

1
0
0

@gregkh what an absolutely lovely commit message. I need to step up my game, this is gold 😄

0
0
0

@aspragg @grimthorpe @gregkh haven't look at the details, but 2381 can't clearly be right, there a few years between now and then which have more than 365 days. Printing used to be a hard problem. We solved this by ditching them. Maybe we should ditch calendars as well :)

0
0
0