lkml.org 
[lkml]   [2003]   [Mar]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: POSIX timer syscalls
Eric Piel wrote:
> george anzinger wrote:
>
>>By the way, I am seeing some reports from the clock_nanosleep test
>>about sleeping too long or too short. The too long appears to be just
>>not being able to preempt what ever else is running. The too short
>>(on the x86) is, I believe, due to the fact that more that 1/HZ is
>>clocked on the wall clock each jiffie.
>>
>>Try this:
>>
>>time sleep 60
>>
>>On the x86 it reports less than 60, NOT good.
>>
>
> I've run the test programs and they pass everything well (with my
> patchs) excepted the nanosleeps which seems to be finnished a bit too
> early. My system test is a 2.5.64 patched on a 4xItaniumII.
>
> My main question is to know if it's a problem even if the difference
> between the wakeup time and the requested time is smaller than the
> resolution of the clock, 976562ns ? I mean, at the resolution of the
> clock we could consider we woke up right at the good time, couldn't we?
>
> In addition time sleep 60 always gave me time over 1 minute, I guess
> it's a good point.
>
> Here is a part of the log of 'do_test':
>
> Testing behavor with clock seting...
> Retarding the clock
> Clock did not seem to move
> was: 1046969027s 703359000ns
> requested: 1046969023s 703359000ns
> now: 1046969022s 467072000ns
> diff is -1.236286998sec
> Cool clock_nanosleeptest.c,379:clock_nanosleep(clock, TIMER_ABSTIME,
> &ts, NULL)

Is it possible that a "clock_was_set()" call was missed? I.e in the
set_timeofday code?
>
> Testing signal behavor...
> handler1 entered: signal 31
> expected clock_nanosleeptest.c,227:clock_nanosleep(clock, 0, &ts, &rs):
> Interrupted system call
> Time remaining is 0s 989257306ns
> clock_nanosleeptest.c,245:slept too short!
> requested: 275s 207032000ns
> now: 275s 207030632ns
> diff is -0.000001368sec
>
> Testing undelivered signal behavor...
> Cool clock_nanosleeptest.c,267:clock_nanosleep(clock, 0, &ts, &rs)
> clock_nanosleeptest.c,283:slept too short!
> requested: 275s 223633000ns
> now: 275s 223632698ns
> diff is -0.000000302sec
>
>
> --Eric
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>

--
George Anzinger george@mvista.com
High-res-timers: http://sourceforge.net/projects/high-res-timers/
Preemption patch: http://www.kernel.org/pub/linux/kernel/people/rml

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

\
 
 \ /
  Last update: 2005-03-22 13:33    [W:0.068 / U:0.104 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site