On 2015/4/3 0:26, Fam Zheng wrote:
> On Thu, 04/02 17:21, Paolo Bonzini wrote:
>>
>>
>> On 02/04/2015 17:16, Fam Zheng wrote:
> After qemu_iovec_destroy, the QEMUIOVector's size is zeroed and
> the zero size ultimately is used to compute virtqueue_push's len
> argument. The
On 04/03/2015 12:26 AM, Fam Zheng wrote:
> On Thu, 04/02 17:21, Paolo Bonzini wrote:
>>
>>
>> On 02/04/2015 17:16, Fam Zheng wrote:
> After qemu_iovec_destroy, the QEMUIOVector's size is zeroed and
> the zero size ultimately is used to compute virtqueue_push's len
> argument
On 02/04/2015 18:26, Fam Zheng wrote:
> There is another problem with your patch - read_size is not initialized in
> non-RW paths like scsi and flush.
Right, but...
> I think the optimization for write is a separate thing, though. Shouldn't
> below
> patch already fix the migration issue?
...
On Thu, 04/02 17:21, Paolo Bonzini wrote:
>
>
> On 02/04/2015 17:16, Fam Zheng wrote:
> > >> After qemu_iovec_destroy, the QEMUIOVector's size is zeroed and
> > >> the zero size ultimately is used to compute virtqueue_push's len
> > >> argument. Therefore, reads from virtio-blk de
On 02/04/2015 17:16, Fam Zheng wrote:
> >> After qemu_iovec_destroy, the QEMUIOVector's size is zeroed and
> >> the zero size ultimately is used to compute virtqueue_push's len
> >> argument. Therefore, reads from virtio-blk devices did not
> >> migrate their results correctl
On Thu, 04/02 16:51, Paolo Bonzini wrote:
>
>
> On 02/04/2015 16:39, Fam Zheng wrote:
> > On Thu, 04/02 15:37, Paolo Bonzini wrote:
> >> After qemu_iovec_destroy, the QEMUIOVector's size is zeroed and
> >> the zero size ultimately is used to compute virtqueue_push's len
> >> argument. Therefore,
On 02/04/2015 16:39, Fam Zheng wrote:
> On Thu, 04/02 15:37, Paolo Bonzini wrote:
>> After qemu_iovec_destroy, the QEMUIOVector's size is zeroed and
>> the zero size ultimately is used to compute virtqueue_push's len
>> argument. Therefore, reads from virtio-blk devices did not
>> migrate their
On Thu, 04/02 15:37, Paolo Bonzini wrote:
> After qemu_iovec_destroy, the QEMUIOVector's size is zeroed and
> the zero size ultimately is used to compute virtqueue_push's len
> argument. Therefore, reads from virtio-blk devices did not
> migrate their results correctly. (Writes were okay).
Can't
After qemu_iovec_destroy, the QEMUIOVector's size is zeroed and
the zero size ultimately is used to compute virtqueue_push's len
argument. Therefore, reads from virtio-blk devices did not
migrate their results correctly. (Writes were okay).
Save the size in submit_requests, and use it when the r
