Re: [Qemu-devel] [PATCH] virtio-net: do not start queues that are not enabled by the guest

[Jason Wang]

On 2019/2/22 下午12:22, Michael S. Tsirkin wrote:
> On Thu, Feb 21, 2019 at 10:10:08PM -0500, Michael S. Tsirkin wrote:
> > On Fri, Feb 22, 2019 at 11:04:05AM +0800, Jason Wang wrote:
> > > On 2019/2/22 上午9:35, Michael S. Tsirkin wrote:
> > > > On Thu, Feb 21, 2019 at 05:40:22PM +0800, Jason Wang wrote:
> > > > > On 2019/2/21 下午4:18, Yuri Benditovich wrote:
> > > > >
> > > > >           For 1.0 device, we can fix the queue_enable, but for 0.9x device how 
> > > > > do
> > > > >           you enable one specific queue in this case? (setting status?)
> > > > >
> > > > >
> > > > >       Do I understand correctly that for 0.9 device in some cases the device 
> > > > > will
> > > > >       receive feature _MQ set, but will not receive 
> > > > > VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET?
> > > > >       Or the problem is different?
> > > > >
> > > > >
> > > > > Let me clarify, VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET is used to control the the
> > > > > number of queue pairs used by device for doing transmission and reception. It
> > > > > was not used to enable or disable a virtqueue.
> > > > >
> > > > > For 1.0 device, we should use queue_enable in pci cfg to enable and disable
> > > > > queue:
> > > > >
> > > > >
> > > > > We could do:
> > > > >
> > > > > 1) allocate memory and set queue_enable for vq0
> > > > >
> > > > > 2) allocate memory and set queue_enable for vq1
> > > > >
> > > > > 3) Set vq paris to 1
> > > > >
> > > > > 4) allocate memory and set queue_enable for vq2
> > > > >
> > > > > 5) allocate memory and set queue_enable for vq3
> > > > >
> > > > > 6) set vq pairs to 2
> > > > I do not think spec allows this.
> > > >
> > > >
> > > > The driver MUST follow this sequence to initialize a device:
> > > > 1. Reset the device.
> > > > 2. Set the ACKNOWLEDGE status bit: the guest OS has noticed the device.
> > > > 3. Set the DRIVER status bit: the guest OS knows how to drive the device.
> > > > 4. Read device feature bits, and write the subset of feature bits understood by 
> > > > the OS and driver to the
> > > > device. During this step the driver MAY read (but MUST NOT write) the 
> > > > device-specific configuration
> > > > fields to check that it can support the device before accepting it.
> > > > 5. Set the FEATURES_OK status bit. The driver MUST NOT accept new feature bits 
> > > > after this step.
> > > > 6. Re-read device status to ensure the FEATURES_OK bit is still set: otherwise, 
> > > > the device does not
> > > > support our subset of features and the device is unusable.
> > > > 7. Perform device-specific setup, including discovery of virtqueues for the 
> > > > device, optional per-bus setup,
> > > > reading and possibly writing the device’s virtio configuration space, and 
> > > > population of virtqueues.
> > > > 8. Set the DRIVER_OK status bit. At this point the device is “live”.
> > > >
> > > >
> > > > Thus vqs are setup at step 7.
> > > >
> > > > # of vq pairs are set up through a command which is a special
> > > > buffer, and spec says:
> > > >
> > > > The driver MUST NOT send any buffer available notifications to the device 
> > > > before setting DRIVER_OK.
> > >
> > > So you meant write to queue_enable is forbidden after DRIVER_OK (though it's
> > > not very clear to me from the  spec). And if a driver want to enable new
> > > queues, it must reset the device?
> >
> > That's my reading.  What do you think?


Looks like I can infer this from the spec, maybe it's better to clarify.


> Btw some legacy drivers might violate this by addig buffers
> before driver_ok.


Yes, but it's probably too late to fix them.

Thanks.


> > > > > But this requires a proper implementation for queue_enable for vhost which is
> > > > > missed in qemu and probably what you really want to do.
> > > > >
> > > > > but for 0.9x device, there's no such way to do this. That's the issue.
> > > > 0.9x there's no queue enable, assumption is PA!=0 means VQ has
> > > > been enabled.
> > > >
> > > >
> > > > > So
> > > > > driver must allocate all queBes before starting the device, otherwise there's
> > > > > no way to enable it afterwards.
> > > > As per spec queues must be allocated before DRIVER_OK.
> > > >
> > > > That is universal.
> > >
> > > If I understand correctly, this is not what is done by current windows
> > > drivers.
> > >
> > > Thanks
> > >
> > >
> > > > > There're tricks to make it work like what is
> > > > > done in your patch, but it depends on a specific implementation like qemu which
> > > > > is sub-optimal.
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >           A fundamental question is what prevents you from just initialization 
> > > > > all
> > > > >           queues during driver start? It looks to me this save lots of efforts
> > > > >           than allocating queue dynamically.
> > > > >
> > > > >
> > > > >       This is not so trivial in Windows driver, as it does not have objects for 
> > > > > queues
> > > > >       that it does not use. Linux driver first of all allocates all the
> > > > >       queues and then
> > > > >       adds Rx/Tx to those it will use. Windows driver first decides how many 
> > > > > queues
> > > > >       it will use then allocates objects for them and initializes them from 
> > > > > zero to
> > > > >       fully functional state.
> > > > >
> > > > >
> > > > > Well, you just need to allocate some memory for the virtqueue, there's no need
> > > > > to make it visible to the rest until it was enabled.
> > > > >
> > > > > Thanks