Re: Incorrect behavior of timer_settime() for absolute dates in thepast
From: Andrew Morton
Date: Mon Nov 27 2006 - 21:21:02 EST
On Mon, 27 Nov 2006 15:32:21 +0100
John <me@xxxxxxxxxxx> wrote:
> John wrote:
> > I'm playing with the POSIX timers API. My platform is x86 running Linux
> > 126.96.36.199 patched with the high-resolution timer subsystem.
> > http://www.tglx.de/hrtimers.html
> > I'm seeing unexpected behavior from timer_settime().
> > int timer_settime(timer_t timerid, int flags,
> > const struct itimerspec *value, struct itimerspec *ovalue);
> > timer_settime() is used to arm a timer. If the TIMER_ABSTIME flag is
> > set, then the timer should fire when the appropriate clock reaches the
> > date specified by value. If that date is in the past, the timer should
> > fire immediately.
> > The Open Group Base Specifications Issue 6 states:
> > http://www.opengroup.org/onlinepubs/009695399/functions/timer_getoverrun.html
> > "If the flag TIMER_ABSTIME is set in the argument flags, timer_settime()
> > shall behave as if the time until next expiration is set to be equal to
> > the difference between the absolute time specified by the it_value
> > member of value and the current value of the clock associated with
> > timerid. That is, the timer shall expire when the clock reaches the
> > value specified by the it_value member of value. If the specified time
> > has already passed, the function shall succeed and the expiration
> > notification shall be made."
> > In my tests, when timer_settime() is called with an expiration date in
> > the past, the timer still takes some time to fire.
> > Here's a run-down of the code provided as an attachment:
> > I switch to a SCHED_RR scheduling policy. In other words, whenever my
> > process wants the CPU, it gets it. (No other SCHED_RR or SCHED_FIFO
> > processes on the system.) I mask the signal that will be delivered on
> > timer expiration. I then arm a timer with an expiration date in the
> > past, check whether the signal is pending, and block waiting for the
> > signal. I then print how long I've had to wait.
> > # ./a.out
> > RESOLUTION=1 ns
> > NOW=969.735545919
> > SLEEPING 1 SECOND...
> > NOW=970.735581398
> > NOW=970.735613525
> > NOW=970.735749017
> > nsdiff=135492 ns i.e. 135.5 µs
> > Any ideas?
> Is there a better forum to discuss this matter?
It hasn't been forgotten about.
This problem, plus the dynticks-makes-us-disable-the-TSC problem, plus
dynticks-breaks-teunis@xxxxxxxxxxxxxxx's-synaptics driver are all (IMO)
blocking a merge.
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/