Doing inotify on /dev is not a good idea because we will be woken up by
a lot of unrelated events.
There is a report of a performance regression on some IO benchmark
because of lock contention within the fsnotify subsystem due to this.
Instead, just watch for attribute changes on the /dev/videoX files
directly. We are only interested in attribute changes, udev should
notify us when the file is added or removed.
A driver node should use the target_duration and target_rate to adjust
the quantum and rate when the graph starts.
The camera nodes don't currently support any of this and simply enforce
a specific rate and duration for the graph clock. Mark this with a
FIXME. Otherwise, pipewire will complain that the node is ignoring the
configured graph rate.
We should really look at the graph target rate/quantum and only produce
a buffer when it is inside the current graph cycle. This would make it
possible to join audio and camera nodes and have them be in sync.
When we get buffers from the driver, check if we have at least as many
as we requested. If we have more, that's ok, we will simply not queue
them.
Previously we would ignore the input number of buffers and use the
allocated amount to fill up the buffer array, which might be too small
and then we crash.
It is valid for V4L2 devices to not implement any controls. QUERYCTRL
returns ENOTTY in these cases. Enumerating the controls must not fail in
these cases but return no controls.
The PropInfo either has a registered id (and then also a name from the
type-info) or a custom name as a string.
In all cases, the description contains a free form text that clarifies
the property.
Use the description in the stream controls name.
udev/kernel sometimes give "Video Capture N" as ID_V4L_PRODUCT, which is
bogus as a device product name. The ID_MODEL* field seem to always have
a sensible product name.
Get device.product.name always from ID_MODEL*, and use ID_V4L_PRODUCT
only as the last fallback.
Make the alignment parameter optional when negotiating buffers.
Default to a 16 bytes alignment and adjust for the max cpu
alignment.
Remove the useless align buffer parameter in plugins, we always
set it to 16 anyway.
When pipewire is accessed through the portal camera API then only the
camera node objects are visible for the client.
However, chromium wants to know the vendor and device ids to identify
cameras. And those properties are currently only added to the device
object.
Fix this by propagating the ids to the node object.
Fixes#1879
The inotify_event field has a variable sized type, so it's recommended to store
it at the end of the storage unit. Fixes gnu-variable-sized-type-not-at-end warning.
Mark some structures, arrays static/const at various places.
In some cases this prevents unnecessary initialization
when a function is entered.
All in all, the text segments across all shared
libraries are reduced by about 2 KiB. However,
the total size increases by about 2 KiB as well.
If the caller asks for MemFd, pass a DmaBuf because it is mostly
the same for v4l2.
Not very correct but it's not yet trivial to fall back to memfd.
And this way we have something to give to the clients that will work
when the client asks for MemFd or MemPtr.
When we add a new listener to an object, it will emit the full state
of the object. For this it temporarily sets the change_mask to all
changes. Restore the previous state after this or else we might not
emit the right change_mask for the next listener.
Consider the case where one there are two listeners on an object.
The object emits a change and the first listener wants to enumerate the
changed params. For this is adds a new listener and then triggers the
enumeration. If we set the change_mask to 0 after adding the listener,
the second listener would get a 0 change_mask and fail to update
its state.
