Re: [kvm-devel] [PATCH 2/4] Revert virtio tap hack

2008-05-04 Thread Avi Kivity
Anthony Liguori wrote: > Anthony Liguori wrote: >> While it has served us well, it is long overdue that we eliminate the >> virtio-net tap hack. It turns out that zero-copy has very little >> impact on >> performance. The tap hack was gaining such a significant performance >> boost >> not becau

Re: [kvm-devel] [PATCH 2/4] Revert virtio tap hack

2008-05-03 Thread Anthony Liguori
Anthony Liguori wrote: While it has served us well, it is long overdue that we eliminate the virtio-net tap hack. It turns out that zero-copy has very little impact on performance. The tap hack was gaining such a significant performance boost not because of zero-copy, but because it avoided dro

[kvm-devel] [PATCH 2/4] Revert virtio tap hack

2008-05-03 Thread Anthony Liguori
While it has served us well, it is long overdue that we eliminate the virtio-net tap hack. It turns out that zero-copy has very little impact on performance. The tap hack was gaining such a significant performance boost not because of zero-copy, but because it avoided dropping packets on receive