Re: Timekeeping hosed by factor 3, high lapic[01] interrupt rates

From: Jens Schweikhardt <schweikh_at_schweikhardt.net>
Date: Thu, 23 Jun 2005 22:58:54 +0200
John et al,

On Wed, Jun 22, 2005 at 10:57:47AM -0400, John Baldwin wrote:
# On Tuesday 21 June 2005 03:51 pm, Jens Schweikhardt wrote:
# > John,
# >
# > # > See
# > http://lists.freebsd.org/pipermail/freebsd-current/2005-May/050210.html # >
# > for the whole thread.
# > #
# > # I did glance over it, but I don't really have time to read every e-mail
# > on # current_at_ in detail. :(  Do you have verbose dmesg's for both cases? 
# > If so, # can you get the 'lapic: Divisor' lines from each boot?
# >
# > Sure, they are the same:
# > lapic: Divisor 2, Frequency 99829738 hz
# >
# > It appears Chris Gilbert has the same problem, see "Timekeeping gone
# > whack in 6-CURRENT" on this list.
# 
# Ok, this is weird.  Does this happen if you disable SMP via 
# 'kern.smp.disabled=1' from the loader?

No, with "set kern.smp.disabled=1" at the loader prompt,
timekeeping is ok and vmstat -i displays

interrupt                          total       rate
irq1: atkbd0                         684          3
irq13: npx0                            1          0
irq14: ata0                           63          0
irq15: ata1                          109          0
irq18: em0                            27          0
irq24: ahd0                        23192        106
irq25: ahc0                           16          0
lapic0: timer                     430979       1986
Total                             455071       2097

Regards,

	Jens
-- 
Jens Schweikhardt http://www.schweikhardt.net/
SIGSIG -- signature too long (core dumped)
Received on Thu Jun 23 2005 - 18:59:21 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:37 UTC