Re: PROBLEM: skew message does not handle negative ns skew

[Paul E. McKenney]

On Thu, Jun 08, 2023 at 10:41:51AM +0100, Chris Bainbridge wrote:
> On Wed, 7 Jun 2023 at 20:04, Paul E. McKenney <paulmck@xxxxxxxxxx> wrote:
> >
> > That is a bit user-unfriendly.  Does the following fix address this
> > issue at your end?
> 
> [    2.095149] clocksource: timekeeping watchdog on CPU3: Marking
> clocksource 'tsc' as unstable because the skew is too large:
> [    2.095152] clocksource:                       'hpet' wd_nsec:
> 515998611 wd_now: 1c29fb9 wd_last: 151e3b8 mask: ffffffff
> [    2.095154] clocksource:                       'tsc' cs_nsec:
> 515124524 cs_now: 8af4c89f9 cs_last: 874f8e80b mask: ffffffffffffffff
> [    2.095155] clocksource:                       Clocksource 'tsc'
> skewed -874087 ns (0 ms) over watchdog 'hpet' interval of 515998611 ns
> (515 ms)

Very good, thank you!

May I please apply your Tested-by?

							Thanx, Paul