Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

[Doug Siebert]

"Richard B. Johnson" wrote:
>
>I think it's a math problem in the test code. Try this:
>
[code deleted]
>
>Note that two subsequent calls to gettimeofday() must not return the 
>same time even if your CPU runs infinitely fast. I haven't seen any 
>kernel in the past few years that fails this test.


I find that claim to be pretty ridiculous, I have never seen such a
thing in any standard.  I was writing code allowing the for '='
condition five years ago, because I was assuming that it might live
long enough for this sort of thing to start happening.  Simple defensive
programming (probably smart even if POSIX was to declare this to be
the case)

So then I was I was curious if I could find any systems fast enough to
violate this.  I didn't have to look far, my new laptop with a 600MHz
Pentium III (running kernel 2.2.16, not that it matters) hits the "break"
in your program (i.e., same time from the two gettimeofday() calls) every
single time I run it.  If I add another identical call to gettimeofday()
immediately after the second one, that makes the result of the (now)
third call 1us greater so the code loops as you intended.

What you claim may have been true due to the inability of CPUs to execute
two system calls within a microsecond, but that horse has now left the
barn.  You will need to request a getnanotimeofday() be created if you
want to allow two consecutive calls to always return different values
(modulo SMP systems and ~13 more years of Moore's Law)

-- 
Douglas Siebert
douglas-siebert@uiowa.edu

A computer without Microsoft software is like chocolate cake without ketchup.

Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

[Alan Cox]

> barn.  You will need to request a getnanotimeofday() be created if you
> want to allow two consecutive calls to always return different values
> (modulo SMP systems and ~13 more years of Moore's Law)

Or you use rdtsc instructions for x86. There intel do guarantee that no two
rdtsc's execute in parallel on the same cpu and it counts in cpu clocks.

Unfortunately rdtsc is only on newer x86 cpus and not useful in some smp
configurations

Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

[dean gaudet]

On Fri, 2 Mar 2001, Richard B. Johnson wrote:

> Note that two subsequent calls to gettimeofday() must not return the
> same time even if your CPU runs infinitely fast. I haven't seen any
> kernel in the past few years that fails this test.

i don't see any requirement for this in SuS.

http://www.opengroup.org/onlinepubs/007908799/xsh/gettimeofday.html

it'd be a pretty disappointing requirement.

-dean

Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

[Mike Galbraith]

On Fri, 2 Mar 2001, Richard B. Johnson wrote:

> Yes and no. It takes microseconds to call the kernel for anything (time
> getpid() ), so it seldom loops. All the kernel has to do is remember

Hi,

c0109286  system_call +<22/40> (0.21) pid(4265)
c011c7e7  sys_gettimeofday +<13/a8> (0.27) pid(4265)
c010e3b2  do_gettimeofday +<e/88> (0.48) pid(4265)
c01092aa  ret_from_sys_call +<6/21> (0.76) pid(4265)
c0109286  system_call +<22/40> (0.19) pid(4265)
c0120b45  sys_getpid +<d/1c> (0.18) pid(4265)
c01092aa  ret_from_sys_call +<6/21> (0.77) pid(4265)
                                     ^^^^ time in usecs

This is a 500Mhz PIII.  It wouldn't take much more cpu/memory speed
to get under a usec.  The overhead of calling the kernel on this box
is almost exactly 1 usec.

	-Mike

Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

[george anzinger]

"Richard B. Johnson" wrote:
> 
> On Fri, 2 Mar 2001, george anzinger wrote:
> 
> > "Richard B. Johnson" wrote:
> 
~snip~

> > > Note that two subsequent calls to gettimeofday() must not return the
> > > same time even if your CPU runs infinitely fast. I haven't seen any
> > > kernel in the past few years that fails this test.
> >
> > Oh!  With only micro second resolution how is this avoided?  The only
> > "legal" thing to do to avoid this is for the fast boxes to loop until
> > the requirement is satisfied.  Is this really done?
> >
> > George
> >
> 
> Yes and no. It takes microseconds to call the kernel for anything (time
> getpid() ), so it seldom loops. All the kernel has to do is remember
> the last value returned. If the time isn't past that time yet, bump
> that value and return it instead of waiting.
> 
Well, "has to do" and "does" are two different animals.  My reading of
the code shows that it does not.  I have a bit of code that does
gettimeofday() calls as fast as possible and on some boxes (ix86) have
seen the difference as low as 1 micro second.  It is not beyond
imagination that a box might return the same time two times in a row
given the processors performance increases we are seeing.  I, for one,
don't find this objectionable.  I WILL take exception to time running
backward, however.  (I don't see how this is avoided on the leap second
delete, but I have just started looking at this issue.)  As to returning
a time in the future as you suggest, I think this is a bad policy.  If
the box can actually do two gettimeofdays in one micro second or less,
it SHOULD return the same time (given the resolution can not resolve the
difference).  If this becomes an issue, and it will, those that care
should use the clock_gettime() call which should return time in nano
seconds.  This is part of the POSIX standard code for which we are
working on at:


http://sourceforge.net/projects/high-res-timers/

George

Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

[Richard B. Johnson]

On Fri, 2 Mar 2001, george anzinger wrote:

> "Richard B. Johnson" wrote:
> > 
> > On Fri, 2 Mar 2001, Christopher Friesen wrote:
> > 
> > > John Being wrote:
> > >
> > > > gives following result on box in question
> > > > root@******:# ./clo
> > > > Leap found: -1687 msec
> > > > and prints nothing on all other  my boxes.
> > > > This gives me bunch of troubles with occasional hang ups and I found nothing
> > > > in kernel archives at
> > > > http://www.uwsg.indiana.edu/hypermail/linux/kernel/index.html
> > > > just some notes about smth like this for SMP boxes with ntp. Is this issue
> > > > known, and how can I fix it?
> > >
> > > I've run into non-monotonic gettimeofday() on a PPC system with 2.2.17, but it
> > > always seemed to be almost exactly a jiffy out, as though it was getting
> > > hundredths of a second from the old tick, and microseconds from the new tick.
> > > Your leap seems to be more unusual, and the first one I've seen on an x86 box.
> > >
> > > Have you considered storing the results to see what happens on the next call?
> > > Does it make up the difference, or do you just lose that time?
> > >
> > > Chris
> > 
> > I think it's a math problem in the test code. Try this:
> > 
> > #include <stdio.h>
> > #include <sys/time.h>
> > 
> > #define DEB(f)
> > 
> > int main()
> > {
> >     struct timeval t;
> >     double start_us;
> >     double stop_us;
> >     for(;;)
> >     {
> >         gettimeofday(&t, NULL);
> >         start_us  = (double) t.tv_sec * 1e6;
> >         start_us += (double) t.tv_usec;
> >         gettimeofday(&t, NULL);
> >         stop_us  = (double) t.tv_sec * 1e6;
> >         stop_us += (double) t.tv_usec;
> >         if(stop_us <= start_us)
> >             break;
> >         DEB(fprintf(stdout, "Start = %f, Stop = %f\n", start_us, stop_us));
> >     }
> >     fprintf(stderr, "Start = %f, Stop = %f\n", start_us, stop_us);
> >     return 0;
> > }
> > 
> > Note that two subsequent calls to gettimeofday() must not return the
> > same time even if your CPU runs infinitely fast. I haven't seen any
> > kernel in the past few years that fails this test.
> 
> Oh!  With only micro second resolution how is this avoided?  The only
> "legal" thing to do to avoid this is for the fast boxes to loop until
> the requirement is satisfied.  Is this really done?
> 
> George
> 

Yes and no. It takes microseconds to call the kernel for anything (time
getpid() ), so it seldom loops. All the kernel has to do is remember
the last value returned. If the time isn't past that time yet, bump
that value and return it instead of waiting.


Cheers,
Dick Johnson

Penguin : Linux version 2.4.1 on an i686 machine (799.53 BogoMips).

"Memory is like gasoline. You use it up when you are running. Of
course you get it all back when you reboot..."; Actual explanation
obtained from the Micro$oft help desk.

Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

[george anzinger]

"Richard B. Johnson" wrote:
> 
> On Fri, 2 Mar 2001, Christopher Friesen wrote:
> 
> > John Being wrote:
> >
> > > gives following result on box in question
> > > root@******:# ./clo
> > > Leap found: -1687 msec
> > > and prints nothing on all other  my boxes.
> > > This gives me bunch of troubles with occasional hang ups and I found nothing
> > > in kernel archives at
> > > http://www.uwsg.indiana.edu/hypermail/linux/kernel/index.html
> > > just some notes about smth like this for SMP boxes with ntp. Is this issue
> > > known, and how can I fix it?
> >
> > I've run into non-monotonic gettimeofday() on a PPC system with 2.2.17, but it
> > always seemed to be almost exactly a jiffy out, as though it was getting
> > hundredths of a second from the old tick, and microseconds from the new tick.
> > Your leap seems to be more unusual, and the first one I've seen on an x86 box.
> >
> > Have you considered storing the results to see what happens on the next call?
> > Does it make up the difference, or do you just lose that time?
> >
> > Chris
> 
> I think it's a math problem in the test code. Try this:
> 
> #include <stdio.h>
> #include <sys/time.h>
> 
> #define DEB(f)
> 
> int main()
> {
>     struct timeval t;
>     double start_us;
>     double stop_us;
>     for(;;)
>     {
>         gettimeofday(&t, NULL);
>         start_us  = (double) t.tv_sec * 1e6;
>         start_us += (double) t.tv_usec;
>         gettimeofday(&t, NULL);
>         stop_us  = (double) t.tv_sec * 1e6;
>         stop_us += (double) t.tv_usec;
>         if(stop_us <= start_us)
>             break;
>         DEB(fprintf(stdout, "Start = %f, Stop = %f\n", start_us, stop_us));
>     }
>     fprintf(stderr, "Start = %f, Stop = %f\n", start_us, stop_us);
>     return 0;
> }
> 
> Note that two subsequent calls to gettimeofday() must not return the
> same time even if your CPU runs infinitely fast. I haven't seen any
> kernel in the past few years that fails this test.

Oh!  With only micro second resolution how is this avoided?  The only
"legal" thing to do to avoid this is for the fast boxes to loop until
the requirement is satisfied.  Is this really done?

George

Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

[Richard B. Johnson]

On Fri, 2 Mar 2001, Christopher Friesen wrote:

> John Being wrote:
> 
> > gives following result on box in question
> > root@******:# ./clo
> > Leap found: -1687 msec
> > and prints nothing on all other  my boxes.
> > This gives me bunch of troubles with occasional hang ups and I found nothing
> > in kernel archives at
> > http://www.uwsg.indiana.edu/hypermail/linux/kernel/index.html
> > just some notes about smth like this for SMP boxes with ntp. Is this issue
> > known, and how can I fix it?
> 
> I've run into non-monotonic gettimeofday() on a PPC system with 2.2.17, but it
> always seemed to be almost exactly a jiffy out, as though it was getting
> hundredths of a second from the old tick, and microseconds from the new tick. 
> Your leap seems to be more unusual, and the first one I've seen on an x86 box.
> 
> Have you considered storing the results to see what happens on the next call? 
> Does it make up the difference, or do you just lose that time?
> 
> Chris

I think it's a math problem in the test code. Try this:

#include <stdio.h>
#include <sys/time.h>

#define DEB(f)

int main()
{
    struct timeval t;
    double start_us;
    double stop_us;
    for(;;)
    {
        gettimeofday(&t, NULL);
        start_us  = (double) t.tv_sec * 1e6;
        start_us += (double) t.tv_usec;
        gettimeofday(&t, NULL);
        stop_us  = (double) t.tv_sec * 1e6;
        stop_us += (double) t.tv_usec;
        if(stop_us <= start_us)
            break;
        DEB(fprintf(stdout, "Start = %f, Stop = %f\n", start_us, stop_us));
    }
    fprintf(stderr, "Start = %f, Stop = %f\n", start_us, stop_us);
    return 0;
}

Note that two subsequent calls to gettimeofday() must not return the
same time even if your CPU runs infinitely fast. I haven't seen any
kernel in the past few years that fails this test. 


Cheers,
Dick Johnson

Penguin : Linux version 2.4.1 on an i686 machine (799.53 BogoMips).

"Memory is like gasoline. You use it up when you are running. Of
course you get it all back when you reboot..."; Actual explanation
obtained from the Micro$oft help desk.

Re: strange nonmonotonic behavior of gettimeoftheday -- seen similar problem on PPC

[Christopher Friesen]

John Being wrote:

> gives following result on box in question
> root@******:# ./clo
> Leap found: -1687 msec
> and prints nothing on all other  my boxes.
> This gives me bunch of troubles with occasional hang ups and I found nothing
> in kernel archives at
> http://www.uwsg.indiana.edu/hypermail/linux/kernel/index.html
> just some notes about smth like this for SMP boxes with ntp. Is this issue
> known, and how can I fix it?

I've run into non-monotonic gettimeofday() on a PPC system with 2.2.17, but it
always seemed to be almost exactly a jiffy out, as though it was getting
hundredths of a second from the old tick, and microseconds from the new tick. 
Your leap seems to be more unusual, and the first one I've seen on an x86 box.

Have you considered storing the results to see what happens on the next call? 
Does it make up the difference, or do you just lose that time?

Chris


-- 
Chris Friesen                    | MailStop: 043/33/F10  
Nortel Networks                  | work: (613) 765-0557
3500 Carling Avenue              | fax:  (613) 765-2986
Nepean, ON K2H 8E9 Canada        | email: cfriesen@nortelnetworks.com