[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [Xen-devel] ctx suddenly rises in HVM Xen Guest
> I have tested HVM XEN GUEST performance with LMbench and found a
> with the context switch time in guest. Following results are with
> scheduler, but SEDF scheduler has similar results.
> The context switch time will rise suddenly when the PROCESS NUMBER (in
> context switch microbenchmark) arrives a 'threshold', just like a
> function. Normally the 'threshold' of PROCESS NUMBER varies in the
> of 35-44 (in UP HVM XEN guest) and 52-62 (in SMP HVM XEN guest), which
> is relevant to UP/SMP and size of processes defined by LMbench. Before
> the 'threshold' context switch time keeps a relative stable value,
> threshold it also keeps a relative stable value. The native Linux has
> such problem.
Try adjusting the amount of shadow pagetable memory the guest has
allocated. You're probably suffering from shadow pagetable conflict
Xen-devel mailing list