Timekeeping hosed by factor 3, high lapic[01] interrupt rates
John Baldwin
jhb at FreeBSD.org
Mon Jun 20 18:23:37 GMT 2005
On Saturday 18 June 2005 06:59 am, Jens Schweikhardt wrote:
> On Fri, Jun 17, 2005 at 02:50:42PM -0400, John Baldwin wrote:
> ...
> # Is this the patch you are running with?
>
> Yes, with if (1) instead, which we know is the same.
Ok.
> # >Index: clock.c
> # ===================================================================
> # RCS file: /usr/cvs/src/sys/i386/isa/clock.c,v
> # retrieving revision 1.220
> # diff -u -r1.220 clock.c
> # --- clock.c 14 May 2005 09:10:01 -0000 1.220
> # +++ clock.c 27 May 2005 19:42:54 -0000
> # @@ -784,7 +784,7 @@
> # * clocks, setup the interrupt handler for the 8254 timer 0 so
> # * that it can drive hardclock().
> # */
> # - if (!using_lapic_timer) {
> # + if (!using_lapic_timer || 1) {
> # intr_add_handler("clk", 0, (driver_intr_t *)clkintr,
> NULL, # INTR_TYPE_CLK | INTR_FAST, NULL);
> # i8254_intsrc = intr_lookup_source(0);
> #
> # Also, can you get the output of 'sysctl kern.clockrate' for both cases?
>
> As reported upthread, it's the same for both cases,
>
> $ sysctl -a | grep hz
> kern.clockrate: { hz = 1000, tick = 1000, profhz = 666, stathz = 133 }
> debug.psm.hz: 20
>
> 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@ in detail. :( Do you have verbose dmesg's for both cases? If so,
can you get the 'lapic: Divisor' lines from each boot?
--
John Baldwin <jhb at FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve" = http://www.FreeBSD.org
More information about the freebsd-current
mailing list