Re: [RFC PATCH 00/11] sched: CFS low-latency features

[Mathieu Desnoyers <mathieu.desnoyers@efficios.com>]

* Thomas Gleixner (tglx@linutronix.de) wrote:
> On Fri, 27 Aug 2010, Mathieu Desnoyers wrote:
> > * Paul E. McKenney (paulmck@linux.vnet.ibm.com) wrote:
> > > Why couldn't the timer_create() call record the start time, and then
> > > compute the sleeps from that time?  So if timer_create() executed at
> > > time t=100 and the period is 5, upon awakening and completing the first
> > > invocation of the function in question, the thread does a sleep calculated
> > > to wake at t=110.
> > 
> > Let's focus on the userspace thread execution, right between the samping of the
> > current time and the call to sleep:
> > 
> >   Thread A
> >   current_time = read current time();
> >   sleep(period_end - current_time);
> > 
> > If the thread is preempted between these two operations, then we end up sleeping
> > for longer than what is needed. This kind of imprecision will add up over time,
> > so that after e.g. one day, instead of having the expected number of timer
> > executions, we'll have less than that. This kind of accumulated drift is an
> > unwanted side-effect of using delays in lieue of real periodic timers.
> 
> Nonsense, that's why we provide clock_nanosleep(ABSTIME)

If we're using CLOCK_MONOTONIC, you're right, this could work. I was only
thinking of relative delays.

So do you think Paul's ideal would be a good candidate for the timer_create
SIGEV_THREAD glibc implementation then ?

Thanks,

Mathieu

> 
> Thanks,
> 
> 	tglx

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com