On 18.08.2012 22:56, Paolo Bonzini wrote:
> Il 12/08/2012 12:08, Michael Tokarev ha scritto:
Commit b1f416aa8d870fab71030abc9401cfc77b948e8e breaks vhost_net
because it always registers the virtio_pci_host_notifier_read() handler
function on the ioeventfd, even when vhost_net.ko is u
Il 12/08/2012 12:08, Michael Tokarev ha scritto:
>> > Commit b1f416aa8d870fab71030abc9401cfc77b948e8e breaks vhost_net
>> > because it always registers the virtio_pci_host_notifier_read() handler
>> > function on the ioeventfd, even when vhost_net.ko is using the ioeventfd.
>> > The result is both
On 06.08.2012 18:56, Paolo Bonzini wrote:
> Commit b1f416aa8d870fab71030abc9401cfc77b948e8e breaks vhost_net
> because it always registers the virtio_pci_host_notifier_read() handler
> function on the ioeventfd, even when vhost_net.ko is using the ioeventfd.
> The result is both QEMU and vhost_net.
On Mon, Aug 06, 2012 at 04:56:10PM +0200, Paolo Bonzini wrote:
> Commit b1f416aa8d870fab71030abc9401cfc77b948e8e breaks vhost_net
> because it always registers the virtio_pci_host_notifier_read() handler
> function on the ioeventfd, even when vhost_net.ko is using the ioeventfd.
> The result is bot
Paolo Bonzini writes:
> Commit b1f416aa8d870fab71030abc9401cfc77b948e8e breaks vhost_net
> because it always registers the virtio_pci_host_notifier_read() handler
> function on the ioeventfd, even when vhost_net.ko is using the ioeventfd.
> The result is both QEMU and vhost_net.ko polling on the
On Mon, Aug 6, 2012 at 3:56 PM, Paolo Bonzini wrote:
> Commit b1f416aa8d870fab71030abc9401cfc77b948e8e breaks vhost_net
> because it always registers the virtio_pci_host_notifier_read() handler
> function on the ioeventfd, even when vhost_net.ko is using the ioeventfd.
> The result is both QEMU an
Commit b1f416aa8d870fab71030abc9401cfc77b948e8e breaks vhost_net
because it always registers the virtio_pci_host_notifier_read() handler
function on the ioeventfd, even when vhost_net.ko is using the ioeventfd.
The result is both QEMU and vhost_net.ko polling on the same eventfd
and the virtio_net.
Il 06/08/2012 15:39, Stefan Hajnoczi ha scritto:
>> > hw/virtio-pci.c | 14 +++---
>> > hw/virtio.c | 7 +--
>> > hw/virtio.h | 3 ++-
>> > 3 file modificati, 14 inserzioni(+), 10 rimozioni(-)
> Internationalized diff stat, I never noticed... :)
Must be new with the latest u
On Mon, Aug 6, 2012 at 2:26 PM, Paolo Bonzini wrote:
> Commit b1f416aa8d870fab71030abc9401cfc77b948e8e breaks vhost_net
> because it always registers the virtio_pci_host_notifier_read() handler
> function on the ioeventfd, even when vhost_net.ko is using the ioeventfd.
> The result is both QEMU an
Commit b1f416aa8d870fab71030abc9401cfc77b948e8e breaks vhost_net
because it always registers the virtio_pci_host_notifier_read() handler
function on the ioeventfd, even when vhost_net.ko is using the ioeventfd.
The result is both QEMU and vhost_net.ko polling on the same eventfd
and the virtio_net.
10 matches
Mail list logo