Add avx mixer to test and benchmark
Rework and unroll the avx mixer some more.
The SSE one is 10 times faster than the C one, The AVX is 20 times
faster. The SSE2 function is 5 times faster than the C one.
Let the mixer functions accumulate the intermediate results into a
larger size variable and then clamp to the final precission. This avoids
distortions because of intermediate clamping.
Although the access pattern of the reads are no longer sequential, the
writes are sequential and we don't need to read intermediate values.
Together with the avoided clamping this is probably faster overall.
Add a unit test for the various cases.
Add an EMPTY chunk flag to mark a piece of memory as 'empty'. For audio
this means silence.
Use the empty flag to avoid mixing 0 samples.
Set the empty flag in output buffers on audioconvert.
When there is no input, mix up to a quantum of data. Otherwise we might
send too much data to the next node and cause a delay if it does not
handle this.
pipewire will allocate buffers aligned to the max alignment required for
the CPU. Take this into account and don't expect larger alignment.
Fixes a warning in mixer-dsp when the CPU max alignment is 16 but the
plugin requires 32 bytes alignment for the AVX2 path (that would never
be chosen on the CPU).
See #2074
Parse the quantum_limit parameters and use this to scale the buffers so
that they can contain the maximum allowed samples instead of the
hardcoded 8192 value.
See #1931
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 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.
SPA_MEMBER is misleading, all we're doing here is pointer+offset and a
type-casting the result. Rename to SPA_PTROFF which is more expressive (and
has the same number of characters so we don't need to re-indent).
Use the DSP media subtype to describe DSP formats. DSP formats
don't include the rate, channels and channel position in the
format and must use the rate and duration from the position io. This
makes it possible to later change the samplerate dynamically without
having to renegotiate the graph.
The same goes for the video DSP format, which uses the io_video_size
from the io_position to get the size/stride. Set this up in the node
based on the defaults from the context.
Make it possible to define defaults in the daemon config file, such
as samplerate, quantum, video size and framerate. This is stored in
the context and used for the DSP formats.
This is more in line with wayland and it allows us to create new
interfaces in modules without having to add anything to the type
enum. It also removes some lookups to map type_id to readable
name in debug.
Move floatmix to the audiomixer plugin and change the name to
AUDIO_MIXER_DSP.
Add runtime selectable sse and sse2 optimizations.
Load the port mixer plugin dynamically based on the factory_name.
Add some more debug
Add a new PortConfig parameter to configure ports of elements that
are marked with the SPA_NODE_FLAG_*_PORT_CONFIG. This is used to
configure the operation of the audioconver/audioadapter nodes and
how it should convert the internal format. We want to use the
Profile parameter only for cases where there is an enumeration of
values, like with device configuration.
Add unit tests for audioconvert and adapter to check if they handle
PortConfig correctly.
Make the media session use the PortConfig to dynamically configure
the device nodes.
Remove audio-dsp, it is not used anymore and can/should be implemented
with a simple audioconvert spa node now and some PortConfig.
Define a set of standard factory names and document what they
contain. This makes it possible to change the implementation by
mapping the factory-name to a different shared library.
The interface struct has the type,version and methods of the
interface.
Make spa interfaces extend from spa_interface and make a
separate structure for the methods.
Pass a generic void* as the first argument of methods, like
we don in PipeWire.
Bundle the methods + implementation in a versioned inteface
and use that to invoke methods. This way we can do version
checks on the methods.
Make resource and proxy interfaces that we can can call. We
can then make the core interfaces independent on proxy/resource and
hide them in the lower layers.
Add add_listener method to methods of core interfaces, just
like SPA.
