[PATCH 0/2] 9p/trans_virtio: handle request cancellation

[PATCH 0/2] 9p/trans_virtio: handle request cancellation

[Greg Kurz]

The 9p protocol mostly relies on a request/reply dialog between the
client and the server. A notable exception to this rule is request
cancellation (ie, flush in 9p wording): when the client requests an
in-flight request to be flushed, the server should only reply to the
flush request and not to the flushed in-flight request (otherwise it
considers the in-flight request to have completed just like it has
never been flushed).

It is up to the client to inform the transport that the in-flight request
has been flushed and won't receive a reply. This is achieved with the
'cancelled' operation of the struct p9_trans_module.

This operation isn't currently implemented with the virtio transport.
As a consequence, flushed requests leave buffers in the used list
forever and the virtqueue ends up in being able to process only one
request at a time.

This issue never popped up because the 9p server in QEMU had a bug
and would always reply to flushed requests. But this will be fixed
soon, so it is time to implement the 'cancelled' operation in the
9p virtio transport.

--
Greg

---

Greg Kurz (2):
      virtio: allow to detach a buffer from the virtqueue
      9p/trans_virtio: implement cancelled callback


 drivers/virtio/virtio_ring.c |   28 ++++++++++++++++++++++++++++
 include/linux/virtio.h       |    1 +
 net/9p/trans_virtio.c        |   18 ++++++++++++++++++
 3 files changed, 47 insertions(+)

[PATCH 1/2] virtio: allow to detach a buffer from the virtqueue

[Greg Kurz]

It is possible for a device to stop using buffers without pushing them
back to the driver. This is the case for example with the 9p virtio
device: if the driver flushes an in-flight request, the 9p specification
specification [*] mandates the server to "to purge the pending response".
The reply to the flush request indicates that the 9p server has stopped
using the buffers of the flushed in-flight request. But since the server
doesn't push back the associated buffers, they don't go back to the free
list. This leads the virtqueue to end up with a single slot to handle all
the dialog with the device, ie, serialize all I/Os.

This patch hence gives the possibility for device specific code to
explicitly detach a given buffer from the used ring and put it back
to the free list.

[*] http://man.cat-v.org/plan_9/5/flush

Signed-off-by: Greg Kurz <groug@kaod.org>
---
 drivers/virtio/virtio_ring.c |   28 ++++++++++++++++++++++++++++
 include/linux/virtio.h       |    1 +
 2 files changed, 29 insertions(+)

diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index eb30f3e09a47..886e9d054de3 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -936,6 +936,34 @@ void *virtqueue_detach_unused_buf(struct virtqueue *_vq)
 }
 EXPORT_SYMBOL_GPL(virtqueue_detach_unused_buf);
 
+/**
+ * virtqueue_detach_buf - detach specific buffer
+ * @vq: the struct virtqueue we're talking about.
+ *
+ * Returns NULL or the "data" token handed to virtqueue_add_*().
+ * This should only be used if the driver really knows the buffer
+ * isn't needed anymore by the device.
+ */
+void *virtqueue_detach_buf(struct virtqueue *_vq, void *buf)
+{
+	struct vring_virtqueue *vq = to_vvq(_vq);
+	unsigned int i;
+
+	START_USE(vq);
+
+	for (i = 0; i < vq->vring.num; i++) {
+		if (vq->desc_state[i].data != buf)
+			continue;
+		detach_buf(vq, i, NULL);
+		END_USE(vq);
+		return buf;
+	}
+
+	END_USE(vq);
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(virtqueue_detach_buf);
+
 irqreturn_t vring_interrupt(int irq, void *_vq)
 {
 	struct vring_virtqueue *vq = to_vvq(_vq);
diff --git a/include/linux/virtio.h b/include/linux/virtio.h
index 988c7355bc22..850158518ce5 100644
--- a/include/linux/virtio.h
+++ b/include/linux/virtio.h
@@ -80,6 +80,7 @@ bool virtqueue_poll(struct virtqueue *vq, unsigned);
 bool virtqueue_enable_cb_delayed(struct virtqueue *vq);
 
 void *virtqueue_detach_unused_buf(struct virtqueue *vq);
+void *virtqueue_detach_buf(struct virtqueue *vq, void *buf);
 
 unsigned int virtqueue_get_vring_size(struct virtqueue *vq);
 

Re: [PATCH 1/2] virtio: allow to detach a buffer from the virtqueue

[Michael S. Tsirkin]

On Wed, Dec 20, 2017 at 06:46:42PM +0100, Greg Kurz wrote:
> It is possible for a device to stop using buffers without pushing them
> back to the driver. This is the case for example with the 9p virtio
> device: if the driver flushes an in-flight request, the 9p specification
> specification [*] mandates the server to "to purge the pending response".
> The reply to the flush request indicates that the 9p server has stopped
> using the buffers of the flushed in-flight request. But since the server
> doesn't push back the associated buffers, they don't go back to the free
> list. This leads the virtqueue to end up with a single slot to handle all
> the dialog with the device, ie, serialize all I/Os.
> 
> This patch hence gives the possibility for device specific code to
> explicitly detach a given buffer from the used ring and put it back
> to the free list.
> 
> [*] http://man.cat-v.org/plan_9/5/flush
> 
> Signed-off-by: Greg Kurz <groug@kaod.org>

It would be better to just change the server to mark all flushed
requests as used. Why isn't that an option?

> ---
>  drivers/virtio/virtio_ring.c |   28 ++++++++++++++++++++++++++++
>  include/linux/virtio.h       |    1 +
>  2 files changed, 29 insertions(+)
> 
> diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
> index eb30f3e09a47..886e9d054de3 100644
> --- a/drivers/virtio/virtio_ring.c
> +++ b/drivers/virtio/virtio_ring.c
> @@ -936,6 +936,34 @@ void *virtqueue_detach_unused_buf(struct virtqueue *_vq)
>  }
>  EXPORT_SYMBOL_GPL(virtqueue_detach_unused_buf);
>  
> +/**
> + * virtqueue_detach_buf - detach specific buffer
> + * @vq: the struct virtqueue we're talking about.
> + *
> + * Returns NULL or the "data" token handed to virtqueue_add_*().
> + * This should only be used if the driver really knows the buffer
> + * isn't needed anymore by the device.
> + */
> +void *virtqueue_detach_buf(struct virtqueue *_vq, void *buf)
> +{
> +	struct vring_virtqueue *vq = to_vvq(_vq);
> +	unsigned int i;
> +
> +	START_USE(vq);
> +
> +	for (i = 0; i < vq->vring.num; i++) {
> +		if (vq->desc_state[i].data != buf)
> +			continue;
> +		detach_buf(vq, i, NULL);
> +		END_USE(vq);
> +		return buf;
> +	}
> +
> +	END_USE(vq);
> +	return NULL;
> +}
> +EXPORT_SYMBOL_GPL(virtqueue_detach_buf);
> +
>  irqreturn_t vring_interrupt(int irq, void *_vq)
>  {
>  	struct vring_virtqueue *vq = to_vvq(_vq);

Unfortunately the used index gets out of sync with the available index
then.

E.g. you are breaking the invariant that used == avail means ring empty.

Any chance to preserve this invariant?



> diff --git a/include/linux/virtio.h b/include/linux/virtio.h
> index 988c7355bc22..850158518ce5 100644
> --- a/include/linux/virtio.h
> +++ b/include/linux/virtio.h
> @@ -80,6 +80,7 @@ bool virtqueue_poll(struct virtqueue *vq, unsigned);
>  bool virtqueue_enable_cb_delayed(struct virtqueue *vq);
>  
>  void *virtqueue_detach_unused_buf(struct virtqueue *vq);
> +void *virtqueue_detach_buf(struct virtqueue *vq, void *buf);
>  
>  unsigned int virtqueue_get_vring_size(struct virtqueue *vq);
>  

Re: [PATCH 1/2] virtio: allow to detach a buffer from the virtqueue

[Greg Kurz]

On Fri, 19 Jan 2018 21:49:38 +0200
"Michael S. Tsirkin" <mst@redhat.com> wrote:

> On Wed, Dec 20, 2017 at 06:46:42PM +0100, Greg Kurz wrote:
> > It is possible for a device to stop using buffers without pushing them
> > back to the driver. This is the case for example with the 9p virtio
> > device: if the driver flushes an in-flight request, the 9p specification
> > specification [*] mandates the server to "to purge the pending response".
> > The reply to the flush request indicates that the 9p server has stopped
> > using the buffers of the flushed in-flight request. But since the server
> > doesn't push back the associated buffers, they don't go back to the free
> > list. This leads the virtqueue to end up with a single slot to handle all
> > the dialog with the device, ie, serialize all I/Os.
> > 
> > This patch hence gives the possibility for device specific code to
> > explicitly detach a given buffer from the used ring and put it back
> > to the free list.
> > 
> > [*] http://man.cat-v.org/plan_9/5/flush
> > 
> > Signed-off-by: Greg Kurz <groug@kaod.org>  
> 
> It would be better to just change the server to mark all flushed
> requests as used. Why isn't that an option?
> 

It is just because I started to look at this from a 9p client code
perspective. It supports several other transports than virtio, and
has a set of transport hooks. One of this hook is called when the
server is supposed to have cancelled a request. My first thought
was to rely on this 'cancelled' hook, like it is done for the RDMA
and fd-based transport.

But now, I realize I should be able to come up with something simpler
in the virtio case, if I do like you suggest. It would require a change
in the client though, as the current code assumes anything pushed by the
other end contains a 9p message, which would be wrong if QEMU does
something like virtqueue_push(vq, elem, 0).

I'll give a try.

Thanks!

> > ---
> >  drivers/virtio/virtio_ring.c |   28 ++++++++++++++++++++++++++++
> >  include/linux/virtio.h       |    1 +
> >  2 files changed, 29 insertions(+)
> > 
> > diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
> > index eb30f3e09a47..886e9d054de3 100644
> > --- a/drivers/virtio/virtio_ring.c
> > +++ b/drivers/virtio/virtio_ring.c
> > @@ -936,6 +936,34 @@ void *virtqueue_detach_unused_buf(struct virtqueue *_vq)
> >  }
> >  EXPORT_SYMBOL_GPL(virtqueue_detach_unused_buf);
> >  
> > +/**
> > + * virtqueue_detach_buf - detach specific buffer
> > + * @vq: the struct virtqueue we're talking about.
> > + *
> > + * Returns NULL or the "data" token handed to virtqueue_add_*().
> > + * This should only be used if the driver really knows the buffer
> > + * isn't needed anymore by the device.
> > + */
> > +void *virtqueue_detach_buf(struct virtqueue *_vq, void *buf)
> > +{
> > +	struct vring_virtqueue *vq = to_vvq(_vq);
> > +	unsigned int i;
> > +
> > +	START_USE(vq);
> > +
> > +	for (i = 0; i < vq->vring.num; i++) {
> > +		if (vq->desc_state[i].data != buf)
> > +			continue;
> > +		detach_buf(vq, i, NULL);
> > +		END_USE(vq);
> > +		return buf;
> > +	}
> > +
> > +	END_USE(vq);
> > +	return NULL;
> > +}
> > +EXPORT_SYMBOL_GPL(virtqueue_detach_buf);
> > +
> >  irqreturn_t vring_interrupt(int irq, void *_vq)
> >  {
> >  	struct vring_virtqueue *vq = to_vvq(_vq);  
> 
> Unfortunately the used index gets out of sync with the available index
> then.
> 
> E.g. you are breaking the invariant that used == avail means ring empty.
> 
> Any chance to preserve this invariant?
> 
> 
> 
> > diff --git a/include/linux/virtio.h b/include/linux/virtio.h
> > index 988c7355bc22..850158518ce5 100644
> > --- a/include/linux/virtio.h
> > +++ b/include/linux/virtio.h
> > @@ -80,6 +80,7 @@ bool virtqueue_poll(struct virtqueue *vq, unsigned);
> >  bool virtqueue_enable_cb_delayed(struct virtqueue *vq);
> >  
> >  void *virtqueue_detach_unused_buf(struct virtqueue *vq);
> > +void *virtqueue_detach_buf(struct virtqueue *vq, void *buf);
> >  
> >  unsigned int virtqueue_get_vring_size(struct virtqueue *vq);
> >    

[PATCH 2/2] 9p/trans_virtio: implement cancelled callback

[Greg Kurz]

When 9p requests are successfully flushed, we must manually move the
associated buffers to the virtqueue freelist, since the server doesn't
send a reply.

Signed-off-by: Greg Kurz <groug@kaod.org>
---
 net/9p/trans_virtio.c |   18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c
index f3a4efcf1456..d3216af124c4 100644
--- a/net/9p/trans_virtio.c
+++ b/net/9p/trans_virtio.c
@@ -206,6 +206,23 @@ static int p9_virtio_cancel(struct p9_client *client, struct p9_req_t *req)
 	return 1;
 }
 
+static int p9_virtio_cancelled(struct p9_client *client, struct p9_req_t *req)
+{
+	struct virtio_chan *chan = client->trans;
+	unsigned long flags;
+
+	spin_lock_irqsave(&chan->lock, flags);
+	if (virtqueue_detach_buf(chan->vq, req) != req) {
+		spin_unlock_irqrestore(&chan->lock, flags);
+		return 0;
+	}
+	chan->ring_bufs_avail = 1;
+	spin_unlock_irqrestore(&chan->lock, flags);
+	/* Wakeup if anyone waiting for VirtIO ring space. */
+	wake_up(chan->vc_wq);
+	return 0;
+}
+
 /**
  * pack_sg_list_p - Just like pack_sg_list. Instead of taking a buffer,
  * this takes a list of pages.
@@ -737,6 +754,7 @@ static struct p9_trans_module p9_virtio_trans = {
 	.request = p9_virtio_request,
 	.zc_request = p9_virtio_zc_request,
 	.cancel = p9_virtio_cancel,
+	.cancelled = p9_virtio_cancelled,
 	/*
 	 * We leave one entry for input and one entry for response
 	 * headers. We also skip one more entry to accomodate, address

Re: [PATCH 0/2] 9p/trans_virtio: handle request cancellation

[Greg Kurz]

Hi Michael,

I wish you a happy new year, and I really need some feedback on this
series because it holds down some patches on the QEMU side.

Thanks,

--
Greg

On Wed, 20 Dec 2017 18:46:27 +0100
Greg Kurz <groug@kaod.org> wrote:

> The 9p protocol mostly relies on a request/reply dialog between the
> client and the server. A notable exception to this rule is request
> cancellation (ie, flush in 9p wording): when the client requests an
> in-flight request to be flushed, the server should only reply to the
> flush request and not to the flushed in-flight request (otherwise it
> considers the in-flight request to have completed just like it has
> never been flushed).
> 
> It is up to the client to inform the transport that the in-flight request
> has been flushed and won't receive a reply. This is achieved with the
> 'cancelled' operation of the struct p9_trans_module.
> 
> This operation isn't currently implemented with the virtio transport.
> As a consequence, flushed requests leave buffers in the used list
> forever and the virtqueue ends up in being able to process only one
> request at a time.
> 
> This issue never popped up because the 9p server in QEMU had a bug
> and would always reply to flushed requests. But this will be fixed
> soon, so it is time to implement the 'cancelled' operation in the
> 9p virtio transport.
> 
> --
> Greg
> 
> ---
> 
> Greg Kurz (2):
>       virtio: allow to detach a buffer from the virtqueue
>       9p/trans_virtio: implement cancelled callback
> 
> 
>  drivers/virtio/virtio_ring.c |   28 ++++++++++++++++++++++++++++
>  include/linux/virtio.h       |    1 +
>  net/9p/trans_virtio.c        |   18 ++++++++++++++++++
>  3 files changed, 47 insertions(+)
> 
>