Fixes#621723 (partially)
set_caps can fail if the video device is running, in that case
setting its format leads to EBUSY.
If set_caps fails then we will not have set up the buffer pool
(it will be NULL) which leads to a crash when we try to pull
buffers. If we fail the negotiate on set_caps failure, then we
won't go to playing state and won't crash.
This is a small improvement. Of course, a nicer fix would
be to make set_caps work in the case where the format is
unchanged. If the format has changed, failing is
probably correct because we need to close the device
(go to NULL state) in order to set caps.
Don't leak a string everytime get_uri() is called and a device
has been set. There's a limited number of devices, so just
intern the string instead of doing more elaborate housekeeping
and storing it in the instance struct or so.
For cameras/drivers that don't support e.g. VIDIOC_G_PARM we'd end up without
a framerate and would try to divide by 0, causing run-time warnings and all
frames to be timestamped with 0, which makes sinks that sync against the clock
drop them, causing 'hangs' (observed with the pwc driver and a Logitech QuickCam
Pro 4000). So if we do not know the framerate, simply don't adjust the
timestamps. Fixes#591451.
Clear format list and probed caps when going to NULL so if a new device
is set we'll probe the formats again instead of using previously
detected ones. Fixes bug #591747.
This also does the following changes:
(1) pull the bufferpool code out into gstv4l2bufferpool.c, and make a
bit more generic so it can be used both for v4l2src and v4l2sink
(2) move some of the device probing/configuration/caps stuff into
gstv4l2object.c so it does not have to be duplicated between
v4l2src and v4l2sink
Fixes bug #590280.
The v4l2 driver for USB webcams on OpenSolaris does not support select()
calls. Detect when select() fails, and skip polling the device afterward,
which restores the pre 0.10.14 behaviour on OpenSolaris.
Signed-off-by: Jan Schmidt <thaytan@noraisin.net>
Use GstPoll to wait for the fd of the video device to become readable before
trying to capture a frame. This speeds up stopping v4l2src a lot as it no
longer has to wait for the next frame, especially when capturing with low
framerates or when the video device just never generates a frame (which seems a
common issue for uvcvideo devices)
Fixes bug #563574.
Original commit message from CVS:
Patch by: Sascha Hauer <s dot hauer at pengutronix dot de>
Luotao Fu <l dot fu at pengutronix dot de>
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_v4l2fourcc_to_structure),
(gst_v4l2_get_caps_info):
Add support for grayscale v4l2 devices. Fixes bug #566616.
Original commit message from CVS:
* sys/v4l2/gstv4l2object.c:
* sys/v4l2/gstv4l2src.c:
* sys/v4l2/gstv4l2src.h:
* sys/v4l2/v4l2_calls.c:
* sys/v4l2/v4l2src_calls.c:
Fix memory leaks. Small code cleanups : No need for empty _init(). No
need to memset instance structures. Some more FIXME's.
Original commit message from CVS:
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_v4l2fourcc_to_structure),
(gst_v4l2_get_caps_info):
Add S910 and PWC formats with a low priority.
* sys/v4l2/v4l2src_calls.c: (gst_v4l2src_format_get_rank),
(gst_v4l2src_probe_caps_for_format):
Add more debugging.
Original commit message from CVS:
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_query):
Properly set the maximum latency value, in the same way it is done in
v4lsrc.
* sys/v4l2/v4l2src_calls.c:
Simplify fraction equality check, no need to use GValues for this.
Original commit message from CVS:
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_query):
Add warning messages stating exactly why the latency query failed.
* sys/v4l2/v4l2src_calls.c: (gst_v4l2src_set_capture):
In some cases, the negotiated framerate might be the default one which
is already set internally. But we still need to mark it down in fps_n
and fps_d so that the latency query can happen properly.
Original commit message from CVS:
Patch by: Sjoerd Simons <sjoerd at luon dot net>
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_negotiate):
Improve negotiation a bit more by picking the smallest possible
resolution that is larger than the resolution specified in the
first caps entry of the peer caps. Fixes bug #536994.
Original commit message from CVS:
Patch by: Sjoerd Simons <sjoerd at luon dot net>
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_class_init),
(gst_v4l2src_fixate), (gst_v4l2src_negotiate):
Provide a custom negotiation function to make sure to pick the highest
possible framerate and resolution. Fixes bug #536646.
Original commit message from CVS:
Patch by: William M. Brack <wbrack at mmm dot com dot hk>
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_v4l2fourcc_to_structure),
(gst_v4l2_get_caps_info):
Add NV12, NV21 and bayer support. See #520092.
Original commit message from CVS:
* sys/v4l2/gstv4l2src.c:
Don't include the gstv4l2xoverlay.h header as the XOverlay support
isn't implemented at all yet and this requires X headers to be
installed. Fixes bug #533264.
Original commit message from CVS:
Always copy buffers by default (handle safer with bugged drivers) and added a property to make it possible to use mmap effectively (no copy if possible) when application wants to. Fixes: #480557.
Original commit message from CVS:
* sys/v4l2/gstv4l2src.c:
* sys/v4l2/v4l2src_calls.c:
When probing the formats and sizes a camera supports, make
sure the best ones (highest resolution, prefered format)
end up at the beginning of the probed caps and the less
desirable ones at the end. This is important because the
order within the caps matters for things like fixation and
negotiation, ie. what format is chosen in the end.
With recent kernels, the current probing code will end up
querying the supported sizes from lowest resolution to
highest resolution, adding them to the probed caps in that
order, resulting to v4l2src fixating to the lowest possible
resolution if downstream does not express a size preference.
Also make up a somewhat random ranking of prefered output
formats for the same reason. Fixes#485828.
Original commit message from CVS:
* sys/v4l2/gstv4l2src.c:
* sys/v4l2/gstv4l2src.h:
* sys/v4l2/v4l2src_calls.c:
Implement LATENCY queries in the crudest way possible so I don't
have to use sync=false any longer when testing with videosinks.
Original commit message from CVS:
* sys/v4l2/gstv4l2src.c: (gst_v4l2_get_caps_info),
(gst_v4l2src_set_caps), (gst_v4l2src_get_mmap):
Restructure the setcaps function so that we can also compute the
expected GStreamer output size of the video frames.
Set frame_byte_size correctly so that read-based devices have a chance
of working correctly.
When grabbing a frame, discard frames that are not of the expected size.
Some cameras don't output the right framesize for the first buffer.
Try only a couple of times to get a valid frame, else error out.
* sys/v4l2/v4l2_calls.c: (gst_v4l2_get_capabilities),
(gst_v4l2_fill_lists), (gst_v4l2_get_input):
Add some more debug info when scanning the device.
* sys/v4l2/v4l2src_calls.c: (gst_v4l2_buffer_new),
(gst_v4l2_buffer_pool_new), (gst_v4l2_buffer_pool_activate),
(gst_v4l2src_fill_format_list), (gst_v4l2src_grab_frame),
(gst_v4l2src_set_capture), (gst_v4l2src_capture_init):
Add some more debug info when dequeing a frame.
Original commit message from CVS:
Patch by: Alexander Eichner <alexeichi@yahoo.de>
* sys/v4l2/gstv4l2src.c: (gst_v4l2src_init):
Use define here.
* sys/v4l2/gstv4l2tuner.c:
(gst_v4l2_tuner_set_frequency_and_notify):
Don't touch the property - its still disabled.
* sys/v4l2/v4l2src_calls.c: (gst_v4l2src_probe_caps_for_format),
(gst_v4l2src_grab_frame), (gst_v4l2src_get_size_limits):
* sys/v4l2/v4l2src_calls.h:
Improve fallback format negotionation. Fixes#451388
Original commit message from CVS:
2007-06-12 Andy Wingo <wingo@pobox.com>
* sys/v4l2/v4l2src_calls.c (gst_v4l2_buffer_finalize)
(gst_v4l2_buffer_class_init, gst_v4l2_buffer_get_type)
(gst_v4l2_buffer_new): Behave more like ximagesink's buffers, with
finalization and resuscitation. No longer public.
(gst_v4l2_buffer_pool_finalize, gst_v4l2_buffer_pool_init)
(gst_v4l2_buffer_pool_class_init, gst_v4l2_buffer_pool_get_type)
(gst_v4l2_buffer_pool_new, gst_v4l2_buffer_pool_activate)
(gst_v4l2_buffer_pool_destroy): Make the pool follow common
miniobject semantics, and be threadsafe.
(gst_v4l2src_queue_frame): Remove this function, as we just call
the ioctls directly in the two places where we queue buffers.
(gst_v4l2src_grab_frame): Return a flowreturn and fill the buffer
directly.
(gst_v4l2src_capture_init): Use the new buffer_pool_new function
to allocate the pool, which also preallocates the GstBuffers.
(gst_v4l2src_capture_start): Call buffer_pool_activate instead of
queueing the frames directly.
* sys/v4l2/gstv4l2src.h (struct _GstV4l2BufferPool): Make this a
real MiniObject instead of rolling our own refcounting and
finalizing. Give it a lock.
(struct _GstV4l2Buffer): Remove one intermediary object, having
the buffers hold the struct v4l2_buffer directly.
* sys/v4l2/gstv4l2src.c (gst_v4l2src_set_caps): Pass the caps to
capture_init so that it can set them on the buffers that it will
create.
(gst_v4l2src_get_read): For better or for worse, include the
timestamping and offsetting code here; really we should be using
bufferalloc though.
(gst_v4l2src_get_mmap): Just make grab_frame return one of our
preallocated, mmap'd buffers.
Original commit message from CVS:
2007-06-05 Andy Wingo <wingo@pobox.com>
* sys/v4l2/gstv4l2src.c (gst_v4l2src_start, gst_v4l2src_stop): Add
some useless comments.
* sys/v4l2/v4l2src_calls.c (gst_v4l2src_capture_init): Don't queue
frames before calling STREAMON, that might leave them in a state
where they can't be dequeued if we go back to NULL without calling
STREAMON, according to the docs.
(gst_v4l2src_capture_start): Enqueue buffers here instead, right
before we call STREAMON.
(gst_v4l2src_capture_deinit): Remove crack to work around dequeue
failures. (For me this code hung.) The pool refcounting is still
crack; added a note to that effect.
Original commit message from CVS:
2007-05-30 Andy Wingo <wingo@pobox.com>
* sys/v4l2/v4l2src_calls.h:
* sys/v4l2/v4l2src_calls.c (gst_v4l2src_fill_format_list): Store
the format list in the order that the driver gives it to us.
(gst_v4l2src_probe_caps_for_format_and_size)
(gst_v4l2src_probe_caps_for_format): New functions, fill GstCaps
based on the capabilities of the device.
(gst_v4l2src_grab_frame): Update for object variable renaming.
(gst_v4l2src_set_capture): Update to be strict in its parameters,
as in the set_caps below.
(gst_v4l2src_capture_init): Update for object variable renaming,
and reflow.
(gst_v4l2src_capture_start, gst_v4l2src_capture_stop)
(gst_v4l2src_capture_deinit): Update for object variable renaming.
(gst_v4l2src_update_fps, gst_v4l2src_set_fps)
(gst_v4l2src_get_fps): Remove; these functions don't have much
meaning outside of an atomic set_caps method.
(gst_v4l2src_buffer_new): Don't set buffer duration, it is not
known.
* sys/v4l2/gstv4l2tuner.c (gst_v4l2_tuner_set_channel): Remove
call to update_fps; not sure about this change.
(gst_v4l2_tuner_set_norm): Work around the fact that for the
moment we don't have an update_fps_func.
* sys/v4l2/gstv4l2src.h (struct _GstV4l2Src): Don't put v4l2
structures in the object, just store what we need. Do store the
probed caps of the device. Don't store the current frame rate.
* sys/v4l2/gstv4l2src.c (gst_v4l2src_init): Remove the
update_fps_function, for now. Update for new object variable
naming.
(gst_v4l2src_set_property, gst_v4l2src_get_property): Update for
new object variable naming.
(gst_v4l2src_v4l2fourcc_to_structure): Rename from ..._to_caps.
(gst_v4l2_structure_to_v4l2fourcc): Rename from ...caps_to_....
(gst_v4l2src_get_caps): Rework to probe the device for supported
frame sizes and frame rates.
(gst_v4l2src_set_caps): Rework to be strict in the given
parameters: if someone asks us to have a certain size and rate,
that is what we configure.
(gst_v4l2src_get_read): Update for object variable naming. Don't
leak buffers on short reads.
(gst_v4l2src_get_mmap): Update for object variable naming, and add
comments.
(gst_v4l2src_create): Update for object variable naming.
