api.acp.hidden-ports and api.acp.hidden-profiles can be used to pass
a json array of ports and profiles to hide. They will not show and
will not be selectable.
Don't try to multiple the max_buffer_size with the frame scale or else
we might try to set a min_buffer_size larger than the max_buffer_size.
Instead, use the frame_scale only to scale the quantum_limit and then
clamp against the max_buffer size.
See #3000
When the period events are enabled, we need to wake up whenver there is
a period elapsed. Otherwise, we only wake up when there is avail_min.
Fixes#3676
Executable memfds can be a security issue.
The kernel warns about them like the following:
pipewire: memfd_create() called without MFD_EXEC or MFD_NOEXEC_SEAL set
Explicitly create all memfds a non-executable as they are not meant to
be executed, similar to the other possible backing filetypes.
As reported in #3217, PTP driver has to have the highest priority for packets to arrive in time
Everything connected to AES67 should be clocked by the Grandmaster clock
Convert capture and playback streams to any global channel position that
has been set. This can then be used to implement an upmix sink and add
such an example config.
When we get a Format update during a Format set_param, return > 0 to the
caller so that it knows we modified the param to something else.
This can be used to fixate or deny updates to Format params in the
stream and have the adapter adapt to this.
Keep the target and source fd in sync when we get the new fd from the
server in remote-node.
This makes it possible to refactor some things and only schedule nodes
by triggering the target.
Some JACK clients place OSC messages in MIDI buffers. Try to detect the
OSC messages and mark the control as OSC.
Also allow OSC control to be converted to JACK MIDI.
This should make things work better with native PipeWire clients that
handle MIDI and OSC.
We place the currently configured runtime settings in the properties but
clear the values when not negotiated. Don't do that but place the
defaults (when set) instead.
When we don't have the thread id yet, don't add the pollfds yet
but wait until we do our first wait operation.
Use flags for eventfd. We can use this to communicate between all kinds
of threads with read/write.
Use evl_init() in the init function, don't attach the main loop, just
the thread that dos the first poll.
Whe using pw_*_new_simple(), pass the properties to the automatically
created context as well so that we can configure it.
This makes it possible to, for example, load the client-rf.conf using the
simple API as well.
Some devices appear to set Supported_Max_Codec_Frames_Per_SDU == 1 while
claiming they support two channels per stream, which is then not
possible.
In this case, limit the number of channels by the number of frames per
SDU when selecting.
Also adjust PAC sorting.
If a client tries to update an ignored key, don't log a warning. Only
log a warning when one of the security properties changed.
This is what we used to do before commit
0dfa05117b
It fixes a warning when a stream disconnects and reconnects because
the serial.id changed.
This function is very much orthogonal in function to the usual
features provided by pipewire, and there does not seem to be any
3rd party users based on Debian Code Search and GitHub.
So let's remove it before any 3rd party users start depending on
pipewire for their valgrind detection.
Commit b5c21c1fbc ("pipewire: use dlclose unless PIPEWIRE_DLCLOSE is set to false")
introduced the PIPEWIRE_DLCLOSE environmental variable to
force/prevent the dlclose()ing of shared objects. However,
native pipewire modules were not adjusted to respect its value,
and instead have been still using `pw_in_valgrind()` to
determine whether or not they should be unloaded.
Fix that by adding a private `pw_should_dlclose()` function
and using that in impl-module.c:pw_impl_module_destroy().
Use (fixed-up) Doxygen manpage output for all program & module manpages.
This also allows formatting the manual pages properly in the HTML docs.
The Markdown pages work properly only with Doxygen >= 1.9.7, older
versions put them to wrong place in the HTML docs.
It doesn't make sense to hang these on the data loop, so let's have
these on the main loop instead. Also avoids a potential crash while
removing them (since removal happens on the main loop and the data loop
might be polling while we're doing the remove).
Ardour calls jack_port_get_buffer() from multiple threads. For audio
buffers this will result in mixing the input samples into the target
buffer and there is no window for having a corrupt buffer.
For MIDI, there is a problem because we need to convert from and to
PipeWire MIDI and while we do that from multiple threads, the midi
buffer can be incomplete or corrupt.
Fix this by building the intermediate POD to a thread-local scratch area
and then copy it to the target buffer. If this is done from multiple
threads there is no moment where incomplete data can be seen in the
target buffer.
Make the complete midi-scratch buffer thread_local so that we also avoid
a race when converting midi data from a foreign port.
When we write into the scratch buffer when mixing and converting input
midi data, we also avoid a race there.
Fixes#3632
Wim Taymans
Thomas Weißschuh
Pauli Virtanen
Frédéric Danis
Arun Raghavan
Dmitry Sharshakov
Aleksandr Melman
Barnabás Pőcze
Jonas Holmberg