Re: [RFC v4 13/17] [media] vb2: add in-fence support to QBUF

[Gustavo Padovan <gustavo@padovan.org>]

2017-10-25 Brian Starkey <brian.starkey@arm.com>:

> Hi Gustavo,
> 
> On Fri, Oct 20, 2017 at 07:50:08PM -0200, Gustavo Padovan wrote:
> > From: Gustavo Padovan <gustavo.padovan@collabora.com>
> > 
> > Receive in-fence from userspace and add support for waiting on them
> > before queueing the buffer to the driver. Buffers are only queued
> > to the driver once they are ready. A buffer is ready when its
> > in-fence signals.
> > 
> > For queues that require vb2 to queue buffers to the v4l2 driver in same
> > order they are received from userspace we use fence_array to keep that
> > ordering. Basically we create a fence_array that contains both the current
> > fence and the fence from the previous buffer (which might be a fence array
> > as well). The base fence class for the fence_array becomes the new buffer
> > fence, waiting on that one guarantees that it won't be queued out of
> > order.
> 
> The API sounds/looks good to me, makes sense to let driver opt in/out
> via the ordered_in_vb2 thing. Thanks for implementing it!
> 
> > 
> > v5:	- use fence_array to keep buffers ordered in vb2 core when
> > 	needed (Brian Stark)
> > 	- keep backward compatibility on the reserved2 field (Brian Stark)
> > 	- protect fence callback removal with lock (Brian Stark)
> 
> Brian Starkey, but close ;-)
> 
> > 
> > v4:
> > 	- Add a comment about dma_fence_add_callback() not returning a
> > 	error (Hans)
> > 	- Call dma_fence_put(vb->in_fence) if fence signaled (Hans)
> > 	- select SYNC_FILE under config VIDEOBUF2_CORE (Hans)
> > 	- Move dma_fence_is_signaled() check to __enqueue_in_driver() (Hans)
> > 	- Remove list_for_each_entry() in __vb2_core_qbuf() (Hans)
> > 	-  Remove if (vb->state != VB2_BUF_STATE_QUEUED) from
> > 	vb2_start_streaming() (Hans)
> > 	- set IN_FENCE flags on __fill_v4l2_buffer (Hans)
> > 	- Queue buffers to the driver as soon as they are ready (Hans)
> > 	- call fill_user_buffer() after queuing the buffer (Hans)
> > 	- add err: label to clean up fence
> > 	- add dma_fence_wait() before calling vb2_start_streaming()
> > 
> > v3:	- document fence parameter
> > 	- remove ternary if at vb2_qbuf() return (Mauro)
> > 	- do not change if conditions behaviour (Mauro)
> > 
> > v2:
> > 	- fix vb2_queue_or_prepare_buf() ret check
> > 	- remove check for VB2_MEMORY_DMABUF only (Javier)
> > 	- check num of ready buffers to start streaming
> > 	- when queueing, start from the first ready buffer
> > 	- handle queue cancel
> > 
> > Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.com>
> > ---
> > drivers/media/v4l2-core/Kconfig          |   1 +
> > drivers/media/v4l2-core/videobuf2-core.c | 179 ++++++++++++++++++++++++++++---
> > drivers/media/v4l2-core/videobuf2-v4l2.c |  29 ++++-
> > include/media/videobuf2-core.h           |  17 ++-
> > 4 files changed, 208 insertions(+), 18 deletions(-)
> > 
> > diff --git a/drivers/media/v4l2-core/Kconfig b/drivers/media/v4l2-core/Kconfig
> > index a35c33686abf..3f988c407c80 100644
> > --- a/drivers/media/v4l2-core/Kconfig
> > +++ b/drivers/media/v4l2-core/Kconfig
> > @@ -83,6 +83,7 @@ config VIDEOBUF_DVB
> > # Used by drivers that need Videobuf2 modules
> > config VIDEOBUF2_CORE
> > 	select DMA_SHARED_BUFFER
> > +	select SYNC_FILE
> > 	tristate
> > 
> > config VIDEOBUF2_MEMOPS
> > diff --git a/drivers/media/v4l2-core/videobuf2-core.c b/drivers/media/v4l2-core/videobuf2-core.c
> > index 60f8b582396a..78f369dba3e3 100644
> > --- a/drivers/media/v4l2-core/videobuf2-core.c
> > +++ b/drivers/media/v4l2-core/videobuf2-core.c
> > @@ -346,6 +346,7 @@ static int __vb2_queue_alloc(struct vb2_queue *q, enum vb2_memory memory,
> > 		vb->index = q->num_buffers + buffer;
> > 		vb->type = q->type;
> > 		vb->memory = memory;
> > +		spin_lock_init(&vb->fence_cb_lock);
> > 		for (plane = 0; plane < num_planes; ++plane) {
> > 			vb->planes[plane].length = plane_sizes[plane];
> > 			vb->planes[plane].min_length = plane_sizes[plane];
> > @@ -1222,6 +1223,9 @@ static void __enqueue_in_driver(struct vb2_buffer *vb)
> > {
> > 	struct vb2_queue *q = vb->vb2_queue;
> > 
> > +	if (vb->in_fence && !dma_fence_is_signaled(vb->in_fence))
> > +		return;
> > +
> > 	vb->state = VB2_BUF_STATE_ACTIVE;
> > 	atomic_inc(&q->owned_by_drv_count);
> > 
> > @@ -1273,6 +1277,20 @@ static int __buf_prepare(struct vb2_buffer *vb, const void *pb)
> > 	return 0;
> > }
> > 
> > +static int __get_num_ready_buffers(struct vb2_queue *q)
> > +{
> > +	struct vb2_buffer *vb;
> > +	int ready_count = 0;
> > +
> > +	/* count num of buffers ready in front of the queued_list */
> > +	list_for_each_entry(vb, &q->queued_list, queued_entry) {
> > +		if (!vb->in_fence || dma_fence_is_signaled(vb->in_fence))
> > +			ready_count++;
> 
> I think there's still some races on vb->in_fence. Couldn't the
> callback on any of the buffers in the queued_list have their callback
> run at any moment?
> 
> That could make this loop do bad things, and similarly the loop in
> vb2_start_streaming() which calls __enqueue_in_driver().
> 
> > +	}
> > +
> > +	return ready_count;
> > +}
> > +
> > int vb2_core_prepare_buf(struct vb2_queue *q, unsigned int index, void *pb)
> > {
> > 	struct vb2_buffer *vb;
> > @@ -1361,7 +1379,74 @@ static int vb2_start_streaming(struct vb2_queue *q)
> > 	return ret;
> > }
> > 
> > -int vb2_core_qbuf(struct vb2_queue *q, unsigned int index, void *pb)
> > +static struct dma_fence *__set_in_fence(struct vb2_queue *q,
> > +					struct vb2_buffer *vb,
> > +					struct dma_fence *fence)
> > +{
> > +	if (q->last_fence && dma_fence_is_signaled(q->last_fence)) {
> > +		dma_fence_put(q->last_fence);
> > +		q->last_fence = NULL;
> > +	}
> > +
> > +	/* For drivers that require the ordering of buffers queued from
> > +	 * userspace to be the same it is queued to the driver - output
> > +	 * some m2m devices - we create a fence array with the fence from
> > +	 * the last queued buffer and this one, that way the fence for this
> > +	 * buffer can't signal before the last one.
> > +	 */
> > +	if (q->is_output || q->ordered_in_vb2) {
> 
> Just a stylistic thing - you could invert this conditional and return
> early, saving a level of indentation below.
> 
> > +		if (fence && q->last_fence) {
> > +			struct dma_fence **fences;
> > +			struct dma_fence_array *arr;
> > +
> > +			fences = kcalloc(2, sizeof(*fences), GFP_KERNEL);
> > +			if (!fences)
> > +				return ERR_PTR(-ENOMEM);
> > +
> > +			fences[0] = fence;
> > +			fences[1] = q->last_fence;
> > +
> > +			arr = dma_fence_array_create(2, fences,
> > +						     dma_fence_context_alloc(1),
> > +						     1, false);
> 
> I'm not sure if it's right to get a new fence_context every time here.
> The comment on dma_fence_context_counter sounds like the vb2_queue
> should have a single context of its own.
> 
> > +			if (!arr) {
> > +				kfree(fences);
> > +				return ERR_PTR(-ENOMEM);
> > +			}
> > +
> > +			fence = &arr->base;
> > +		} else if (!fence && q->last_fence) {
> > +			fence = dma_fence_get(q->last_fence);
> > +		}
> > +
> > +		q->last_fence = dma_fence_get(fence);
> 
> Seems like the refcounting isn't quite right here.
> 
> If you land in "else if (!fence && q->last_fence)", then "fence" gets
> a new reference to q->last_fence, and you take another one here -
> which looks like one too many.
> 
> I _think_ the right thing to do is just give the previous reference on
> q->last_fence to "fence", without taking a new one:
> 
> 	} else if (!fence && q->last_fence) {
> 		fence = q->last_fence;
> 	}
> 
> That's analogous to the other path where fence is not-NULL, and the
> fence-array takes ownership of the existing references on fence and
> q->last_fence, without taking new references.
> 
> Please double check my logic though!
> 
> > +	}
> > +
> > +	return fence;
> > +}
> > +
> > +static void vb2_qbuf_fence_cb(struct dma_fence *f, struct dma_fence_cb *cb)
> > +{
> > +	struct vb2_buffer *vb = container_of(cb, struct vb2_buffer, fence_cb);
> > +	struct vb2_queue *q = vb->vb2_queue;
> > +	unsigned long flags;
> > +
> > +	spin_lock_irqsave(&vb->fence_cb_lock, flags);
> > +	if (!vb->in_fence) {
> > +		spin_unlock_irqrestore(&vb->fence_cb_lock, flags);
> > +		return;
> > +	}
> > +
> > +	dma_fence_put(vb->in_fence);
> > +	vb->in_fence = NULL;
> > +	spin_unlock_irqrestore(&vb->fence_cb_lock, flags);
> > +
> > +	if (q->start_streaming_called)
> > +		__enqueue_in_driver(vb);
> > +}
> > +
> > +int vb2_core_qbuf(struct vb2_queue *q, unsigned int index, void *pb,
> > +		  struct dma_fence *fence)
> > {
> > 	struct vb2_buffer *vb;
> > 	int ret;
> > @@ -1372,16 +1457,18 @@ int vb2_core_qbuf(struct vb2_queue *q, unsigned int index, void *pb)
> > 	case VB2_BUF_STATE_DEQUEUED:
> > 		ret = __buf_prepare(vb, pb);
> > 		if (ret)
> > -			return ret;
> > +			goto err;
> > 		break;
> > 	case VB2_BUF_STATE_PREPARED:
> > 		break;
> > 	case VB2_BUF_STATE_PREPARING:
> > 		dprintk(1, "buffer still being prepared\n");
> > -		return -EINVAL;
> > +		ret = -EINVAL;
> > +		goto err;
> > 	default:
> > 		dprintk(1, "invalid buffer state %d\n", vb->state);
> > -		return -EINVAL;
> > +		ret = -EINVAL;
> > +		goto err;
> > 	}
> > 
> > 	/*
> > @@ -1398,30 +1485,75 @@ int vb2_core_qbuf(struct vb2_queue *q, unsigned int index, void *pb)
> > 
> > 	trace_vb2_qbuf(q, vb);
> > 
> > +	vb->in_fence = __set_in_fence(q, vb, fence);
> > +	if (IS_ERR(vb->in_fence)) {
> > +		ret = PTR_ERR(vb->in_fence);
> > +		goto err;
> 
> Between here, __set_in_fence() and vb2_qbuf() there's a reference
> leaked on "fence" in the case of error. I'm not sure what semantics
> you'd like, but it should be clear whose responsibility it is to clear
> up "fence" if __set_in_fence() fails (at the moment, looks like no-one
> does).
> 
> > +	}
> 
> ... and if __set_in_fence() succeeds, then vb->in_fence now owns the
> reference on "fence" that was passed in, so I guess you should set
> fence = NULL here.
> 
> > +
> > +	/*
> > +	 * If it is time to call vb2_start_streaming() wait for the fence
> > +	 * to signal first. Of course, this happens only once per streaming.
> > +	 * We want to run any step that might fail before we set the callback
> > +	 * to queue the fence when it signals.
> > +	 */
> > +	if (fence && !q->start_streaming_called &&
> 
> I think this should be vb->in_fence now, and the same for all further
> uses of "fence" in this function.
> 
> > +	    __get_num_ready_buffers(q) == q->min_buffers_needed - 1)
> > +		dma_fence_wait(fence, true);
> > +
> > 	/*
> > 	 * If streamon has been called, and we haven't yet called
> > 	 * start_streaming() since not enough buffers were queued, and
> > 	 * we now have reached the minimum number of queued buffers,
> > 	 * then we can finally call start_streaming().
> > -	 *
> > -	 * If already streaming, give the buffer to driver for processing.
> > -	 * If not, the buffer will be given to driver on next streamon.
> > 	 */
> > 	if (q->streaming && !q->start_streaming_called &&
> > -	    q->queued_count >= q->min_buffers_needed) {
> > +	    __get_num_ready_buffers(q) >= q->min_buffers_needed) {
> > 		ret = vb2_start_streaming(q);
> > 		if (ret)
> > -			return ret;
> > -	} else if (q->start_streaming_called) {
> > -		__enqueue_in_driver(vb);
> > +			goto err;
> > +	}
> > +
> > +	/*
> > +	 * For explicit synchronization: If the fence didn't signal
> > +	 * yet we setup a callback to queue the buffer once the fence
> > +	 * signals, and then, return successfully. But if the fence
> > +	 * already signaled we lose the reference we held and queue the
> > +	 * buffer to the driver.
> > +	 */
> > +	if (fence) {
> > +		ret = dma_fence_add_callback(fence, &vb->fence_cb,
> > +					     vb2_qbuf_fence_cb);
> > +		if (!ret)
> > +			goto fill;
> 
> Just spotted that you should probably handle the -EINVAL case here and
> jump to err.
> 
> > +
> > +		dma_fence_put(fence);
> > +		vb->in_fence = NULL;
> > 	}
> > 
> > +fill:
> > +	/*
> > +	 * If already streaming and there is no fence to wait on
> > +	 * give the buffer to driver for processing.
> > +	 */
> > +	if (q->start_streaming_called && !vb->in_fence)
> > +		__enqueue_in_driver(vb);
> 
> The callback could have run just before this, clearing vb->in_fence
> and ending up with __enqueue_in_driver() getting called twice.
> 
> > +
> > 	/* Fill buffer information for the userspace */
> > 	if (pb)
> > 		call_void_bufop(q, fill_user_buffer, vb, pb);
> > 
> > 	dprintk(2, "qbuf of buffer %d succeeded\n", vb->index);
> > 	return 0;
> > +
> > +err:
> > +	if (vb->in_fence) {
> > +		dma_fence_put(vb->in_fence);
> > +		vb->in_fence = NULL;
> > +	}
> > +
> > +	return ret;
> > +
> > }
> > EXPORT_SYMBOL_GPL(vb2_core_qbuf);
> > 
> > @@ -1632,6 +1764,8 @@ EXPORT_SYMBOL_GPL(vb2_core_dqbuf);
> > static void __vb2_queue_cancel(struct vb2_queue *q)
> > {
> > 	unsigned int i;
> > +	struct vb2_buffer *vb;
> > +	unsigned long flags;
> > 
> > 	/*
> > 	 * Tell driver to stop all transactions and release all queued
> > @@ -1659,6 +1793,21 @@ static void __vb2_queue_cancel(struct vb2_queue *q)
> > 	q->queued_count = 0;
> > 	q->error = 0;
> > 
> > +	list_for_each_entry(vb, &q->queued_list, queued_entry) {
> > +		spin_lock_irqsave(&vb->fence_cb_lock, flags);
> > +		if (vb->in_fence) {
> > +			dma_fence_remove_callback(vb->in_fence, &vb->fence_cb);
> > +			dma_fence_put(vb->in_fence);
> > +			vb->in_fence = NULL;
> > +		}
> > +		spin_unlock_irqrestore(&vb->fence_cb_lock, flags);
> > +	}
> > +
> > +	if (q->last_fence) {
> > +		dma_fence_put(q->last_fence);
> > +		q->last_fence = NULL;
> > +	}
> > +
> > 	/*
> > 	 * Remove all buffers from videobuf's list...
> > 	 */
> > @@ -1720,7 +1869,7 @@ int vb2_core_streamon(struct vb2_queue *q, unsigned int type)
> > 	 * Tell driver to start streaming provided sufficient buffers
> > 	 * are available.
> > 	 */
> > -	if (q->queued_count >= q->min_buffers_needed) {
> > +	if (__get_num_ready_buffers(q) >= q->min_buffers_needed) {
> > 		ret = v4l_vb2q_enable_media_source(q);
> > 		if (ret)
> > 			return ret;
> > @@ -2240,7 +2389,7 @@ static int __vb2_init_fileio(struct vb2_queue *q, int read)
> > 		 * Queue all buffers.
> > 		 */
> > 		for (i = 0; i < q->num_buffers; i++) {
> > -			ret = vb2_core_qbuf(q, i, NULL);
> > +			ret = vb2_core_qbuf(q, i, NULL, NULL);
> > 			if (ret)
> > 				goto err_reqbufs;
> > 			fileio->bufs[i].queued = 1;
> > @@ -2419,7 +2568,7 @@ static size_t __vb2_perform_fileio(struct vb2_queue *q, char __user *data, size_
> > 
> > 		if (copy_timestamp)
> > 			b->timestamp = ktime_get_ns();
> > -		ret = vb2_core_qbuf(q, index, NULL);
> > +		ret = vb2_core_qbuf(q, index, NULL, NULL);
> > 		dprintk(5, "vb2_dbuf result: %d\n", ret);
> > 		if (ret)
> > 			return ret;
> > @@ -2522,7 +2671,7 @@ static int vb2_thread(void *data)
> > 		if (copy_timestamp)
> > 			vb->timestamp = ktime_get_ns();;
> > 		if (!threadio->stop)
> > -			ret = vb2_core_qbuf(q, vb->index, NULL);
> > +			ret = vb2_core_qbuf(q, vb->index, NULL, NULL);
> > 		call_void_qop(q, wait_prepare, q);
> > 		if (ret || threadio->stop)
> > 			break;
> > diff --git a/drivers/media/v4l2-core/videobuf2-v4l2.c b/drivers/media/v4l2-core/videobuf2-v4l2.c
> > index 110fb45fef6f..4c09ea007d90 100644
> > --- a/drivers/media/v4l2-core/videobuf2-v4l2.c
> > +++ b/drivers/media/v4l2-core/videobuf2-v4l2.c
> > @@ -23,6 +23,7 @@
> > #include <linux/sched.h>
> > #include <linux/freezer.h>
> > #include <linux/kthread.h>
> > +#include <linux/sync_file.h>
> > 
> > #include <media/v4l2-dev.h>
> > #include <media/v4l2-fh.h>
> > @@ -178,6 +179,12 @@ static int vb2_queue_or_prepare_buf(struct vb2_queue *q, struct v4l2_buffer *b,
> > 		return -EINVAL;
> > 	}
> > 
> > +	if ((b->fence_fd != 0 && b->fence_fd != -1) &&
> > +	    !(b->flags & V4L2_BUF_FLAG_IN_FENCE)) {
> > +		dprintk(1, "%s: fence_fd set without IN_FENCE flag\n", opname);
> > +		return -EINVAL;
> > +	}
> > +
> > 	return __verify_planes_array(q->bufs[b->index], b);
> > }
> > 
> > @@ -203,9 +210,14 @@ static void __fill_v4l2_buffer(struct vb2_buffer *vb, void *pb)
> > 	b->timestamp = ns_to_timeval(vb->timestamp);
> > 	b->timecode = vbuf->timecode;
> > 	b->sequence = vbuf->sequence;
> > -	b->fence_fd = -1;
> > 	b->reserved = 0;
> > 
> > +	b->fence_fd = -1;
> > +	if (vb->in_fence)
> > +		b->flags |= V4L2_BUF_FLAG_IN_FENCE;
> > +	else
> > +		b->flags &= ~V4L2_BUF_FLAG_IN_FENCE;
> > +
> > 	if (q->is_multiplanar) {
> > 		/*
> > 		 * Fill in plane-related data if userspace provided an array
> > @@ -560,6 +572,7 @@ EXPORT_SYMBOL_GPL(vb2_create_bufs);
> > 
> > int vb2_qbuf(struct vb2_queue *q, struct v4l2_buffer *b)
> > {
> > +	struct dma_fence *fence = NULL;
> > 	int ret;
> > 
> > 	if (vb2_fileio_is_active(q)) {
> > @@ -568,7 +581,19 @@ int vb2_qbuf(struct vb2_queue *q, struct v4l2_buffer *b)
> > 	}
> > 
> > 	ret = vb2_queue_or_prepare_buf(q, b, "qbuf");
> > -	return ret ? ret : vb2_core_qbuf(q, b->index, b);
> > +	if (ret)
> > +		return ret;
> > +
> > +	if (b->flags & V4L2_BUF_FLAG_IN_FENCE) {
> > +		fence = sync_file_get_fence(b->fence_fd);
> 
> In DRM a fence_fd of -1 means "already signaled" - should that be
> allowed here too?

If it is already signaled we shouldn't even set the IN_FENCE
flag. There isn't much to gain from that. The current code already makes
sure to guarantee the ordering if a buffer don't have a fence.

Gustavo