Avi Kivity wrote:
Anthony Liguori wrote:
In modern KVM, the IO thread is capable of interrupting the CPU
whenever it needs to process IO. Therefore this "problem" no longer
exists.
It would still be good to verify that the problem no longer exists.
This is not a cosmetic change; some testing is needed to verify it
doesn't introduce new latencies.
N.B. dynticks is the preferred timer in QEMU on Linux. To even hit this
code path, you'd have to use an explicit -clock hpet or -clock rtc. I
don't have an hpet on my laptop and -clock rtc boots just as fast as it
did before.
Do we really care about optimizing latency with -clock rtc though?
--
Regards,
Anthony Liguori
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html