Re: VirtIO vs Emulation Netperf benchmark results

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Oct 10, 2012 at 3:05 PM, Christoffer Dall
<c.dall@xxxxxxxxxxxxxxxxxxxxxx> wrote:
> On Wed, Oct 10, 2012 at 3:04 PM, Alexander Graf <agraf@xxxxxxx> wrote:
>>
>>
>> On 10.10.2012, at 21:02, Peter Maydell <peter.maydell@xxxxxxxxxx> wrote:
>>
>>> On 10 October 2012 19:50, Alexander Graf <agraf@xxxxxxx> wrote:
>>>> Network emulation in QEMU is more than just slow (expected).
>>>>
>>>>
>>>> Why is this expected? This performance drop is quite terrifying.
>>>
>>> Well, it's pretty poor (and suggests the exit path could
>>> be optimised) but the reason that it is insanely slow rather
>>> than merely not good is that the lan9118 requires a device
>>> register read or write for every word of data transferred to
>>> the network...
>>
>> Ouch. Any way we could fake a PCI controller and emulate some real hardware?
>>
> if we're doing anything else than the real hardware why not just use
> virtio then?
>
oh, you mean some real hardware that the guest may anyway have driver
support for but doesn't necessarily exist on the vexpress board?
_______________________________________________
kvmarm mailing list
kvmarm@xxxxxxxxxxxxxxxxxxxxx
https://lists.cs.columbia.edu/cucslists/listinfo/kvmarm


[Index of Archives]     [Linux KVM]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux