The __x86_64__ macro identifies a CPU family, and is unfortunately not
enough to identify a concrete ABI.
The normal x86_64 ABI that is used by practical Linux distributions is
LP64 (i.e. 32-bit int, and 64-bit long and pointer), and defines
__x86_64__ and __LP64__.
x32 is a niche ILP32 ABI (i.e. 32-bit int, long and pointer) for x86_64
CPUs, which has different struct sizes due to sizeof(long) and
sizeof(void *) being smaller. It defines __x86_64__ and __ILP32__.
Signed-off-by: Simon McVittie <smcv@debian.org>
This is necessary on some 32-bit architectures that implement atomic
operations on 64-bit quantities as library calls, including Debian's
armel and mipsel ports.
Signed-off-by: Simon McVittie <smcv@debian.org>
Track the current route and the properties independently. Else we
might skip parsing the volumes/mute properties when the current
route didn't change.
Fixes#281
Use a global sync that both syncs the globals and completes the
operations.
In the case of a card profile change, first the nodes are removed
and added and then the Profile and Route info updated. We need to
be able to bind to the new node and get the device.profile.id
before we can find the active port of the node.
See #279
Continue processing the input data until we can't send anymore. When
we are following another driver, we need to process each buffer in
the cycle completely or we will start lagging behind.
Don't say that there is no runtime directory set when there is but
the server is not running.
Simplify things by first trying to connect to the user runtime directory
when set and then try the system directory. There is no need to manually
check the existance of the socket, connect wil that that and set the
appropriate error.
Implement load_module of the null sink
Keep track of sink_input/source_output linked sink/source in a
better way.
Associate the global with our streams and the streams with the
global.
Don't emit the sink_input/source_output before there is a linked device
and before our stream is READY.
Improve server info to make it more look like pulse.
The ConnectAudio method triggers Introspect and NewConnection calls from
hsphfpd to the backend, which will set the fd to use for the SCO data.
This imply to run the DBus loop while waiting for ConnectAudio reply to be
able to reply to those method calls.
MTU is hard-coded to 48. Using MTU from hsphfpd may prevent correct
sound operations.
Volume management is not yet implemented.
As hsphfpd is a prototype and its API subject to change, this backend is
disable by default.
This skeleton connects to hsphfpd daemon and list the managed devices.
Add /run/pipewire to the list of runtime directories so that clients can
connect to the system service without specifying runtime directory.
If the otherwise chosen directory (e.g. the user home directory) doesn't
contain any socket but /run/pipewire does, the latter will be used.
Add system service and socket in addition to the user service and
socket. The system service is started as pipewire user and group and
clients need to be in the pipewire group to be able to connect.
