PROBLEM: monotonic clock going backwards on ppc64

[Daniel Axtens]

Hi,

> I think I observed a potential problem, is this the correct place to report it? (CC me, not on list)

Yes

> [1.] One line summary: monotonic clock can be made to decrease on ppc64
> [2.] Full description:
> Setting the realtime clock can sometimes make the monotonic clock go back by over a hundred years.
> Decreasing the realtime clock across the y2k38 threshold is one reliable way to reproduce.
> Allegedly this can also happen just by running ntpd, I have not managed to reproduce that other
> than booting with rtc at >2038 and then running ntp.
> When this happens, anything with timers (e.g. openjdk) breaks rather badly.

I can confirm this on a physical 64bit book3e system and powerpc/next
with the provided reproducer.

> [X.] Other notes, patches, fixes, workarounds:
> The problem seems to be in vDSO code in arch/powerpc/kernel/vdso64/gettimeofday.S.
> (possibly because some values used in the calculation are only 32 bit?)
> Slightly silly workaround: 
> nuke the "cmpwi cr1,r3,CLOCK_MONOTONIC" in __kernel_clock_gettime
> Now it always goes through the syscall fallback which does not have the same problem.

I'll look at the asm but I'm no expert here so we might need to defer to
someone with real expertise. Falling back to the syscall sadly ruins
the performance gain of having a vDSO.

Regards,
Daniel