Hi > 2011/5/24 François Revol <revol@xxxxxxx>: > >> Nope, still same problems. > >> And now that I checked the ticket, it seems that PIT mode 0 is > >> problematic on Xenserver (which is the one we were setting > > > already, > >> when entering the kernel), while mode 1 works (ticket #4500) > > > > Tried using some other timer source yet ? > > > > François. > > Yes, using LAPIC doesn't work, HPET works. Since the APIC timer is calibrated using the TSC which in turn is calibrated using the PIT it's not so surprising that both fail in this case. What you could try is to still use mode 0, but instead use a different channel, channel 2 (as channel one isn't usually implemented anymore) in the TSC calibration. That is the combination that Linux uses for their "fast TSC calibration using PIT" and I think also for the corresponding "slow" version. The dmesg output of Linux should tell you if and how it calibrated the TSC which should indicate what method is usable. If it turns out that using mode 0 on channel 2 is more widely compatible than using it on channel 0 we should simply switch that. Everything is in place for that anyway, just the channel variable needs to be flipped. Regards Michael