Re: gettimeofday() jumping into the future

[James Courtier-Dutton]

> > On Thu, 23 Aug 2007, Michael Smith wrote:
> >
> >     
> [...]
>   
> > +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;
> >     


That does not work as cycle_now can go backwards as well as wrap around.
The only way to catch all edge cases is to be able to make an assumption
on the maximum acceptable difference between cycle_now and cycle_last,
where difference is the shortest distance between values if they were
pointers into a ring buffer. If this maximum difference value is less
than half of the max value cycle_now can have, one can take a very
simple approach.
Treat the cycle_now, cycle_last value as signed.
If they have same sign, do the comparison as unsigned.
If they have different signs, do the comparision as signed.

James


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/