Re: Synchronization protocol around TODR / time-related variables

[Andrew Doran <ad%netbsd.org@localhost>]

On Wed, Jan 01, 2020 at 09:05:43AM -0800, Jason Thorpe wrote:

> It looks to me like in our kern_tc.c, timebasebin has actually replaced
> the historical function of the boottime variable, though I have to admit I
> haven't fully wrapped my head around kern_tc.c.  Anyway, if that's true,
> we can nuke the old boottime completely, and have getboottime() return the
> converted-to-ts value of timebasebin using the same lockless strategy.
> 
> Agree?

kern_tc.c is one of those alright.  Yes it looks that way to me too.

Andrew