Add some more mentions that modules can not be loaded in a remote
instance in PipeWire.
Also mention the special internal remote name to connect to the local
pw-cli instance.
Fixes#2988
JACK does not automatically generate aliases, so don't do that either.
The problem is that no new ports can be created if one already exists
with the same name or any of the aliases. This causes problems when a
port is renamed and a new port is created with the old name because the
alias might still be the old name and port creation fails.
Add a jack.fill-aliases for this purpose.
Fixes#3154
When a jack client is opened, the thread_utils from the context are
stored globaly. Replace this with a generic thread_utils when we close
the client again to avoid referencing freed memory.
When the first parsed part of the value does not look like a container,
dump the complete length of the value.
If a value would contain 'Tunnel sink', it would previously only
serialize the 'Tunnel' part and ignore the rest.
See #3212
Previously, a client disconnecting while a sample was playing could
lead to issues. For example, if a client disconnected before the
"ready" signal of the sample-play arrives, `operation_new_cb()`
would be called and that would try to use the client's pw_manager,
however, that has previously been destroyed in `client_disconnect()`.
If the client disconnected after the "ready" signal but before the reply
has been sent, then `sample_play_ready_reply()` would never be called
since operations are completed via the client's pw_manager which
would already be destroyed at that point.
Fix this by installing a listener on the client, and properly
cancelling the operation and making sure that the pending_sample
is correctly destroyed.
Before this patch, every link between node A and B would increment the
activation counter of B and add a new target to A. If there are N links
between A and B, resuming node A would then do N decrements of the
activation counter of B (by iterating the target_list) before finally
activating B.
This is not optimal, we can share the same activation count for all the
links between A and B.
Add a new pw_node_peer struct to keep track of links between A and B.
Activating a link between A and B activates the single activation of B,
deactivating all links deactivates B again. Waking up B after A finished
now no longer depends on the number of links between A and B.
This is particularly important for remote nodes because before the patch
they would get the activation memory and the eventfd of the peer __for
each link__. With huge amounts of links (like in stress tests) this
would result in too many fds. filtering out the fds for the same peer
was not easily possible because the server would still increment the
counters for each link and sharing the eventfd would require refcounting
it and closing duplicates.
After this patch the remote node receives 1 activation memory and eventfd
for each peer node, independent of the number of links between them. Even
for stereo streams this saves half of the memory and fds.
In jack_port_unregister(), don't free the port completely but mark it
as removing. This will then do the portregistration callback before
freeing the port.
Add some more debug.
