This is the mail archive of the
libc-alpha@sourceware.cygnus.com
mailing list for the glibc project.
Re: Linux nanosleep investigation.
- To: Kaz Kylheku <kaz at ashi dot footprints dot net>
- Subject: Re: Linux nanosleep investigation.
- From: Andrea Arcangeli <andrea at suse dot de>
- Date: Wed, 16 Feb 2000 21:37:02 +0100 (CET)
- cc: Andreas Jaeger <aj at suse dot de>, "Kevin B. Hendricks" <khendricks at ivey dot uwo dot ca>, Ulrich Drepper <drepper at cygnus dot com>, libc-alpha Mailinglist <libc-alpha at sourceware dot cygnus dot com>
On Tue, 15 Feb 2000, Kaz Kylheku wrote:
>timespec when the thread should wake up; but nanosleep consistently causes it
>to wake up one jiffy later than that time. With pthread_cond_timedwait,
That's wrong. It causes it to wakeup in mean half jiffy. Sometime zero
jiffy after, some time 1 jiffy after.
>[..] this
>program will wake up every delta period in a predictable way.
What's the problem if it doesn't wakeup with a nanosecond
precision? That's my whole point.
>This won't work properly if the sleep arbitrarily extends the delta by a whole
>clock tick!
What won't work properly? Are you going to burn a CDROM, your videocard,
your nic, or whatever?
Andrea