Re: gettimeofday() jumping into the future

[Andi Kleen <andi@firstfloor.org>]

Tim Ricketts <tr@earth.li> writes:

[adding right people to cc just in case it slipped past their filters,
keeping enough quote for context]

> On Thu, 23 Aug 2007, Michael Smith wrote:
> 
> > We've been seeing some strange behaviour on some of our applications
> > recently. I've tracked this down to gettimeofday() returning spurious
> > values occasionally.
> >
> > Specifically, gettimeofday() will suddenly, for a single call, return
> > a value about 4398 seconds (~1 hour 13 minutes) in the future. The
> > following call goes back to a normal value.
> 
> I have also seen this.
> 
> > This seems to be occurring when the clock source goes slightly
> > backwards for a single call. In
> > kernel/time/timekeeping.c:__get_nsec_offset(), we have this:
> > cycle_delta = (cycle_now - clock->cycle_last) & clock->mask;
> >
> > So a small decrease in time here will (this is all unsigned
> > arithmetic) give us a very large cycle_delta. cyc2ns() then multiplies
> > this by some value, then right shifts by 22. The resulting value (in
> > nanoseconds) is approximately 4398 seconds; this gets added on to the
> > xtime value, giving us our jump into the future. The next call to
> > gettimeofday() returns to normal as we don't have this huge nanosecond
> > offset.
> 
> Indeed.  I don't know where the suggestion of off by 2^32us came in
> later in this thread.  As you've already pointed out, it's off by
> 2^42ns.

[...]
> +static inline u64 __get_nsec_offset(void)
>   {
>   	cycle_t cycle_now, cycle_delta;
> -	s64 ns_offset;
> +	u64 ns_offset;
> 
>   	/* read clocksource: */
>   	cycle_now = clocksource_read(clock);
> 
> +	if (cycle_now < clock->cycle_last)
> +		return 0;

The old x86-64 pre-clocksource gettimeofday() implementation had a similar
check. It came from painful experience.

-Andi