Paul Mackerras <[email protected]> writes: > Benjamin Herrenschmidt writes: > >> Ok, does not using NTP fixes it ? > > Try this patch. With this the values from gettimeofday() or the VDSO > should stay exactly in sync with xtime even if NTP is adjusting the > clock. > > This patch still has quite a few debugging printks in it, so it's not > final by any means. I'll be interested to hear how it goes, and in > particular whether or not you see any "oops, time got ahead" messages. Without your patch, the clock works perfectly when NTP is not in use, but when NTP is in use I get a large amount of skew (3 min) after about half an hour. With your patch (tested against 2.6.16-rc3), there is no skew whether NTP is running or not, and the system has been up 90 mins so far. They two times appear to be the same. Regards, Roger -- Roger Leigh Printing on GNU/Linux? http://gutenprint.sourceforge.net/ Debian GNU/Linux http://www.debian.org/ GPG Public Key: 0x25BFB848. Please sign and encrypt your mail.
Attachment:
pgpItr6UYoONV.pgp
Description: PGP signature
- References:
- 2.6.16-rc2 powerpc timestamp skew
- From: Roger Leigh <[email protected]>
- Re: 2.6.16-rc2 powerpc timestamp skew
- From: Benjamin Herrenschmidt <[email protected]>
- Re: 2.6.16-rc2 powerpc timestamp skew
- From: Roger Leigh <[email protected]>
- Re: 2.6.16-rc2 powerpc timestamp skew
- From: Benjamin Herrenschmidt <[email protected]>
- Re: 2.6.16-rc2 powerpc timestamp skew
- From: Paul Mackerras <[email protected]>
- 2.6.16-rc2 powerpc timestamp skew
- Prev by Date: Re: [BUG] kernel 2.6.15.4: soft lockup detected on CPU#0!
- Next by Date: Re: Stuck creating sysfs hooks for a driver..
- Previous by thread: Re: 2.6.16-rc2 powerpc timestamp skew
- Next by thread: Re: 2.6.16-rc2 powerpc timestamp skew
- Index(es):