On Sun, Feb 14, 2016 at 06:17:03PM +0100, Paolo Bonzini wrote:
> Currently, dataplane threads are shut down during migration because
> vring.c is not able to track dirty memory. However, all the relevant
> parts of QEMU have been made thread-safe now, so we can drop vring.c
> completely. With the
On 19/02/2016 08:42, Michael S. Tsirkin wrote:
> On Sun, Feb 14, 2016 at 06:17:03PM +0100, Paolo Bonzini wrote:
> > Currently, dataplane threads are shut down during migration because
> > vring.c is not able to track dirty memory. However, all the relevant
> > parts of QEMU have been made thread
On Sun, Feb 14, 2016 at 06:17:03PM +0100, Paolo Bonzini wrote:
> Currently, dataplane threads are shut down during migration because
> vring.c is not able to track dirty memory. However, all the relevant
> parts of QEMU have been made thread-safe now, so we can drop vring.c
> completely. With the
On 16/02/2016 09:57, Christian Borntraeger wrote:
> > Currently, dataplane threads are shut down during migration because
> > vring.c is not able to track dirty memory. However, all the relevant
> > parts of QEMU have been made thread-safe now, so we can drop vring.c
> > completely. With these
On 02/14/2016 06:17 PM, Paolo Bonzini wrote:
> Currently, dataplane threads are shut down during migration because
> vring.c is not able to track dirty memory. However, all the relevant
> parts of QEMU have been made thread-safe now, so we can drop vring.c
> completely. With these patches, virtio
Currently, dataplane threads are shut down during migration because
vring.c is not able to track dirty memory. However, all the relevant
parts of QEMU have been made thread-safe now, so we can drop vring.c
completely. With these patches, virtio-dataplane is now simply "virtio
with ioeventfd in a
