Message-ID: <20180509103353.GA39838@e107564-lin.cambridge.arm.com>
Date:   Wed, 9 May 2018 11:33:54 +0100
From:   Brian Starkey <brian.starkey@....com>
To:     Ezequiel Garcia <ezequiel@...labora.com>
Cc:     linux-media@...r.kernel.org, kernel@...labora.com,
        Hans Verkuil <hverkuil@...all.nl>,
        Mauro Carvalho Chehab <mchehab@....samsung.com>,
        Shuah Khan <shuahkh@....samsung.com>,
        Pawel Osciak <pawel@...iak.com>,
        Alexandre Courbot <acourbot@...omium.org>,
        Sakari Ailus <sakari.ailus@....fi>,
        linux-kernel@...r.kernel.org,
        Gustavo Padovan <gustavo.padovan@...labora.com>
Subject: Re: [PATCH v9 10/15] vb2: add explicit fence user API

Hi Ezequiel,

On Fri, May 04, 2018 at 05:06:07PM -0300, Ezequiel Garcia wrote:
>From: Gustavo Padovan <gustavo.padovan@...labora.com>
>
>Turn the reserved2 field into fence_fd that we will use to send
>an in-fence to the kernel or return an out-fence from the kernel to
>userspace.
>
>Two new flags were added, V4L2_BUF_FLAG_IN_FENCE, that should be used
>when sending an in-fence to the kernel to be waited on, and
>V4L2_BUF_FLAG_OUT_FENCE, to ask the kernel to give back an out-fence.
>
>v7: minor fixes on the Documentation (Hans Verkuil)
>
>v6: big improvement on doc (Hans Verkuil)
>
>v5: - keep using reserved2 field for cpia2
>    - set fence_fd to 0 for now, for compat with userspace(Mauro)
>
>v4: make it a union with reserved2 and fence_fd (Hans Verkuil)
>
>v3: make the out_fence refer to the current buffer (Hans Verkuil)
>
>v2: add documentation
>
>Signed-off-by: Gustavo Padovan <gustavo.padovan@...labora.com>
>---
> Documentation/media/uapi/v4l/buffer.rst         | 45 +++++++++++++++++++++++--
> drivers/media/common/videobuf2/videobuf2-v4l2.c |  2 +-
> drivers/media/v4l2-core/v4l2-compat-ioctl32.c   |  4 +--
> include/uapi/linux/videodev2.h                  |  8 ++++-
> 4 files changed, 52 insertions(+), 7 deletions(-)
>
>diff --git a/Documentation/media/uapi/v4l/buffer.rst b/Documentation/media/uapi/v4l/buffer.rst
>index e2c85ddc990b..be9719cf5745 100644
>--- a/Documentation/media/uapi/v4l/buffer.rst
>+++ b/Documentation/media/uapi/v4l/buffer.rst
>@@ -301,10 +301,22 @@ struct v4l2_buffer
> 	elements in the ``planes`` array. The driver will fill in the
> 	actual number of valid elements in that array.
>     * - __u32
>-      - ``reserved2``
>+      - ``fence_fd``
>       -
>-      - A place holder for future extensions. Drivers and applications
>-	must set this to 0.
>+      - Used to communicate a fence file descriptors from userspace to kernel
>+	and vice-versa. On :ref:`VIDIOC_QBUF <VIDIOC_QBUF>` when sending
>+	an in-fence for V4L2 to wait on, the ``V4L2_BUF_FLAG_IN_FENCE`` flag must
>+	be used and this field set to the fence file descriptor of the in-fence.
>+	If the in-fence is not valid ` VIDIOC_QBUF`` returns an error.
>+
>+        To get an out-fence back from V4L2 the ``V4L2_BUF_FLAG_OUT_FENCE``
>+	must be set, the kernel will return the out-fence file descriptor in
>+	this field. If it fails to create the out-fence ``VIDIOC_QBUF` returns
>+        an error.
>+
>+	For all other ioctls V4L2 sets this field to -1 if
>+	``V4L2_BUF_FLAG_IN_FENCE`` and/or ``V4L2_BUF_FLAG_OUT_FENCE`` are set,
>+	otherwise this field is set to 0 for backward compatibility.
>     * - __u32
>       - ``reserved``
>       -
>@@ -648,6 +660,33 @@ Buffer Flags
>       - Start Of Exposure. The buffer timestamp has been taken when the
> 	exposure of the frame has begun. This is only valid for the
> 	``V4L2_BUF_TYPE_VIDEO_CAPTURE`` buffer type.
>+    * .. _`V4L2-BUF-FLAG-IN-FENCE`:
>+
>+      - ``V4L2_BUF_FLAG_IN_FENCE``
>+      - 0x00200000
>+      - Ask V4L2 to wait on the fence passed in the ``fence_fd`` field. The
>+	buffer won't be queued to the driver until the fence signals. The order
>+	in which buffers are queued is guaranteed to be preserved, so any
>+	buffers queued after this buffer will also be blocked until this fence
>+	signals. This flag must be set before calling ``VIDIOC_QBUF``. For
>+	other ioctls the driver just reports the value of the flag.
>+
>+        If the fence signals the flag is cleared and not reported anymore.
>+	If the fence is not valid ``VIDIOC_QBUF`` returns an error.
>+
>+
>+    * .. _`V4L2-BUF-FLAG-OUT-FENCE`:
>+
>+      - ``V4L2_BUF_FLAG_OUT_FENCE``
>+      - 0x00400000
>+      - Request for a fence to be attached to the buffer. The driver will fill
>+	in the out-fence fd in the ``fence_fd`` field when :ref:`VIDIOC_QBUF
>+	<VIDIOC_QBUF>` returns. This flag must be set before calling
>+	``VIDIOC_QBUF``. For other ioctls the driver just reports the value of
>+	the flag.
>+
>+        If the creation of the out-fence fails ``VIDIOC_QBUF`` returns an
>+	error.
>

I commented similarly on some of the old patch-sets, and it's a minor
thing, but I still think the ordering of this series is off. It's
strange/wrong to me document all this behaviour, and expose the flags
to userspace, when the functionality isn't implemented yet.

If I apply this patch to the kernel, then the kernel doesn't do what
the (newly added) kernel-doc says it will.

I'd use this sequence:

   * Remove all the stuff in patch 11 which checks/sets FLAG_IN_FENCE
     (i.e. the code added in videobuf2-v4l2.c, except for adding the
     extra argument to vb2_core_qbuf(), which will always be NULL).

   * Remove all the stuff in patch 12 which checks/sets FLAG_OUT_FENCE
     (i.e. the code added in videobuf2-v4l2.c, except for adding the
     extra argument to vb2_core_qbuf(), which will always be false).

   * Move this to be a "new" patch 13, and add back the code to
     videobuf2-v4l2.c, which hooks up the UAPI flags to the
     implementation.

There's also some ordering dependency with the V4L2_CAP_FENCES. I'd
put the V4L2_CAP_FENCES patch first, and it will always be unset.
Then, once everything is in-place, you can finally add the patches
which set it. That also allows you to check and document that the
FLAG_*_FENCE flags are only valid when that cap is set.

That way, the patches depend on each other in the right order (I can't
pick the patch exposing the userspace flags without first having the
patches implementing the behaviour), and userspace can never see
something inconsistent.

Cheers,
-Brian

>
>
>diff --git a/drivers/media/common/videobuf2/videobuf2-v4l2.c b/drivers/media/common/videobuf2/videobuf2-v4l2.c
>index 64503615d00b..b1c0fa2b0b88 100644
>--- a/drivers/media/common/videobuf2/videobuf2-v4l2.c
>+++ b/drivers/media/common/videobuf2/videobuf2-v4l2.c
>@@ -203,7 +203,7 @@ 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->reserved2 = 0;
>+	b->fence_fd = 0;
> 	b->reserved = 0;
>
> 	if (q->is_multiplanar) {
>diff --git a/drivers/media/v4l2-core/v4l2-compat-ioctl32.c b/drivers/media/v4l2-core/v4l2-compat-ioctl32.c
>index 4312935f1dfc..93c752459aec 100644
>--- a/drivers/media/v4l2-core/v4l2-compat-ioctl32.c
>+++ b/drivers/media/v4l2-core/v4l2-compat-ioctl32.c
>@@ -388,7 +388,7 @@ struct v4l2_buffer32 {
> 		__s32		fd;
> 	} m;
> 	__u32			length;
>-	__u32			reserved2;
>+	__s32			fence_fd;
> 	__u32			reserved;
> };
>
>@@ -606,7 +606,7 @@ static int put_v4l2_buffer32(struct v4l2_buffer __user *kp,
> 	    assign_in_user(&up->timestamp.tv_usec, &kp->timestamp.tv_usec) ||
> 	    copy_in_user(&up->timecode, &kp->timecode, sizeof(kp->timecode)) ||
> 	    assign_in_user(&up->sequence, &kp->sequence) ||
>-	    assign_in_user(&up->reserved2, &kp->reserved2) ||
>+	    assign_in_user(&up->fence_fd, &kp->fence_fd) ||
> 	    assign_in_user(&up->reserved, &kp->reserved) ||
> 	    get_user(length, &kp->length) ||
> 	    put_user(length, &up->length))
>diff --git a/include/uapi/linux/videodev2.h b/include/uapi/linux/videodev2.h
>index a8842a5ca636..1f18dc68ecab 100644
>--- a/include/uapi/linux/videodev2.h
>+++ b/include/uapi/linux/videodev2.h
>@@ -934,7 +934,10 @@ struct v4l2_buffer {
> 		__s32		fd;
> 	} m;
> 	__u32			length;
>-	__u32			reserved2;
>+	union {
>+		__s32		fence_fd;
>+		__u32		reserved2;
>+	};
> 	__u32			reserved;
> };
>
>@@ -971,6 +974,9 @@ struct v4l2_buffer {
> #define V4L2_BUF_FLAG_TSTAMP_SRC_SOE		0x00010000
> /* mem2mem encoder/decoder */
> #define V4L2_BUF_FLAG_LAST			0x00100000
>+/* Explicit synchronization */
>+#define V4L2_BUF_FLAG_IN_FENCE			0x00200000
>+#define V4L2_BUF_FLAG_OUT_FENCE			0x00400000
>
> /**
>  * struct v4l2_exportbuffer - export of video buffer as DMABUF file descriptor
>-- 
>2.16.3
>

Powered by blists - more mailing lists