On Friday 08 July 2005 12:44 pm, JR Dalrymple wrote: > >On Thu, Jul 07, 2005 at 03:56:14PM -0800, JR Dalrymple wrote: > >>The clock seems to be running at about 1/4 speed (SMP related maybe?). > > <Off list reply> > > >If you also post the output of "sysctl kern.timecounter" and "vmstat > >-i" to the list, it may help diagnose the problem. > > genie# sysctl kern.timecounter > kern.timecounter.stepwarnings: 0 > kern.timecounter.nbinuptime: 357873388 > kern.timecounter.nnanouptime: 0 > kern.timecounter.nmicrouptime: 8296 > kern.timecounter.nbintime: 153797574 > kern.timecounter.nnanotime: 171942 > kern.timecounter.nmicrotime: 153621939 > kern.timecounter.ngetbinuptime: 656730 > kern.timecounter.ngetnanouptime: 40589 > kern.timecounter.ngetmicrouptime: 161872387 > kern.timecounter.ngetbintime: 0 > kern.timecounter.ngetnanotime: 0 > kern.timecounter.ngetmicrotime: 26596618 > kern.timecounter.nsetclock: 623 > kern.timecounter.hardware: i8254 > kern.timecounter.choice: TSC(-100) i8254(0) dummy(-1000000) > kern.timecounter.tick: 1 > kern.timecounter.smp_tsc: 0 > genie# vmstat -i > interrupt total rate > irq1: atkbd0 232 0 > irq6: fdc0 10 0 > irq10: fxp0 ahc1 5247043 73 > irq11: amr0 ahc0 3305544 46 > irq13: npx0 1 0 > lapic0: timer 532488625 7412 > lapic4: timer 532404584 7411 > lapic1: timer 532404584 7411 > lapic2: timer 532404584 7411 > Total 2138255207 29766 > > > > I can kind of make sense of what this is trying to help me, but it's > above my head as far as how to tune it. > > Thanks again for any help This should be fixed now as the i8254 timer was broken when used with APIC for a while. -- John Baldwin <jhb_at_FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve" = http://www.FreeBSD.orgReceived on Sat Jul 09 2005 - 11:37:40 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:38 UTC