calcru: runtime went backwards, RELENG_6, SMP

[Dmitry Morozovsky]

On Sat, 9 Jun 2007, Matthew Dillon wrote:

MD> :Well, I can of course shut the kernel up, but kernel time stability is still my 
MD> :concern. I run ntpd there and while sometimes it seems stable (well, sorta: 
MD> :drift are within several seconds...) there are cases of half-a-minute time 
MD> :steps. 
MD> :
MD> :Sincerely,
MD> :D.Marck                                     [DM5020, MCK-RIPE, DM3-RIPN]
MD> 
MD>     I think the only hope you have of getting the issue addressed is to
MD>     run FreeBSD current.  If you can reproduce the time slips under current
MD>     the developers should be able to track the problem down and fix it.  The
MD>     code is so different between those two releases that they are going to
MD>     have a hard time working the problem in FreeBSD-6.

Well, quick compile'n'install HEAD into the swap does not reveal time slips for 
now. However, this is with WITNESS, I'll try to turn off most debugging kernel 
options and try again.

MD>     If you don't want to do that, try forcing the timer to use the 8254
MD>     and see if that helps.  You may also have to reduce the system tick
MD>     to ~100-200 hz.

This will be my next test, and I'll follow up with the results.

Thanks!

Sincerely,
D.Marck                                     [DM5020, MCK-RIPE, DM3-RIPN]
------------------------------------------------------------------------
*** Dmitry Morozovsky --- D.Marck --- Wild Woozle --- marck at rinet.ru ***
------------------------------------------------------------------------