Re: Atom N450 + C3 + HPET == bad timer behaviour

2013-06-24 Thread Adrian Chadd
Hi, It's (transcribed): kern.timecounter.hardware: TSC kern.timecounter.choice: TSC(1000) ACPI-fast (900) HPET (950) i8254 (0) dummy(-1000) kern.timecounter.tick: 1 kern.timecounter.fast_gettime: 1 kern.timecounter.invariant_tsc: 1 kern.timecounter.smp_tsc: 1 kern.timecounter.smp_tsc_adjust: 0

Re: Atom N450 + C3 + HPET == bad timer behaviour

2013-06-21 Thread Alexander Motin
On 21.06.2013 04:02, Adrian Chadd wrote: On 20 June 2013 16:45, Adrian Chadd adr...@freebsd.org wrote: Hi, I'm having issues with HPET + C3 state on this Atom N450 based netbook. This is (shocking, I know!) running -HEAD (r251605.) If I use C2, HPET is fine. If I use RTC, i8254, LAPIC, C3 is

Atom N450 + C3 + HPET == bad timer behaviour

2013-06-20 Thread Adrian Chadd
Hi, I'm having issues with HPET + C3 state on this Atom N450 based netbook. This is (shocking, I know!) running -HEAD (r251605.) If I use C2, HPET is fine. If I use RTC, i8254, LAPIC, C3 is also fine. But C3 + HPET results in multi-second pauses where it should be 1 second. I've disabled

Re: Atom N450 + C3 + HPET == bad timer behaviour

2013-06-20 Thread Adrian Chadd
On 20 June 2013 16:45, Adrian Chadd adr...@freebsd.org wrote: Hi, I'm having issues with HPET + C3 state on this Atom N450 based netbook. This is (shocking, I know!) running -HEAD (r251605.) If I use C2, HPET is fine. If I use RTC, i8254, LAPIC, C3 is also fine. But C3 + HPET results in