GitLab

Part-of: <pulseaudio/pulseaudio!439>

Part-of: <pulseaudio/pulseaudio!439>

Edward Lee authored Oct 01, 2020 and Arun Raghavan committed Dec 16, 2020

Part-of: <pulseaudio/pulseaudio!360>

Edward Lee authored Oct 01, 2020 and Arun Raghavan committed Dec 16, 2020

Part-of: <pulseaudio/pulseaudio!360>

Edward Lee authored Oct 01, 2020 and Arun Raghavan committed Dec 16, 2020

Part-of: <pulseaudio/pulseaudio!360>

Edward Lee authored Oct 01, 2020 and Arun Raghavan committed Dec 16, 2020

This is a temporary addition, until <afunix.h> ships with the
MinGW toolchain.

Part-of: <pulseaudio/pulseaudio!360>

Edward Lee authored Oct 01, 2020 and Arun Raghavan committed Dec 16, 2020

Part-of: <pulseaudio/pulseaudio!360>

Edward Lee authored Oct 01, 2020 and Arun Raghavan committed Dec 16, 2020

Part-of: <!360>

Edward Lee authored Oct 01, 2020 and Arun Raghavan committed Dec 16, 2020

Part-of: <pulseaudio/pulseaudio!360>

Sebastian Krzyszkowiak authored Oct 29, 2020 and PulseAudio Marge Bot committed Dec 14, 2020

Some filters take parameters that effectively describe the hardware
they're being applied to (like echo-cancel allowing to specify the
mic array parameters for better noise filtering). This allows system
integrators to set default parameters for such modules per-device,
which will get used when the stream doesn't specify their own.

Part-of: <!400>

Tanu Kaskinen authored Sep 17, 2020 and PulseAudio Marge Bot committed Dec 14, 2020

The JACK sink and source priority was 0, but I believe the common case
is that when a JACK sink or source is loaded, the user wants to make it
the default device.

Fixes: pulseaudio/pulseaudio#983
Part-of: <pulseaudio/pulseaudio!369>

Igor Kovalenko authored Dec 14, 2020 and PulseAudio Marge Bot committed Dec 14, 2020

Part-of: <pulseaudio/pulseaudio!437>

Tanu Kaskinen authored Jun 06, 2020 and PulseAudio Marge Bot committed Dec 14, 2020

The old behaviour was such that if none of the normal mappings worked,
we would probe ALL fallbacks. I don't think that makes sense, and it
caused concrete issues: let's say we have a regular stereo mic device,
but there's no "front" PCM defined for it. In this situation we would
probe the stereo-fallback mapping (which uses "hw" instead of "front"),
and it would work, but then we'd also probe the "multichannel-input"
mapping, which would also work, so we end up with two mappings that
don't have any difference in behaviour.

I think it's better to simply pick the first working fallback and ignore
the rest.

Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/901
(issue is marked as confidential due to unreleased hardware)

Part-of: <!304>

SimonPilkington authored May 07, 2020 and PulseAudio Marge Bot committed Dec 14, 2020

Try $XDG_DATA_HOME, then $XDG_DATA_DIRS, and finally fall back to old
behaviour (prefix-defined directory).

core-util: Ignore non-absolute XDG base dirs

These are invalid per the spec.

Fixes: #862
Part-of: <!293>

Samuel Thibault authored Apr 26, 2020 and PulseAudio Marge Bot committed Dec 14, 2020

If we do not manage to rewind at all because there is nothing to rewind
any more, it means the latency is too small, and we let audio escape our
control. We should thus increase the watermark to fix this.

Fixes #871

Part-of: <!287>

Allow adding module arguments using udev PULSE_MODARGS environment variable and
fail module loading if there is a problem with PULSE_MODARGS

This helps setting e.g. 'tsched=0' for specific devices without a need to create
full load module entry in default.pa.

Part-of: <pulseaudio/pulseaudio!436>

Hui Wang authored Dec 10, 2020 and PulseAudio Marge Bot committed Dec 10, 2020

With the Auto-Mute enabled, if the headphone jack is plugged, the
alsa hda driver will mute the speaker and set pinctl of the speaker
to Hi-Z state, after this happens, even the pulseaudio unmute the
speaker, the speaker still couldn't output sound because the pinctl
is in Hi-Z state.

We found this issue on a Dell machine which has multi-function audio
jack, after the headphone is plugged in, the speaker's availability is
still unknown, users could select speaker from gnome-sound-setting,
but even the speaker is selected to be the active device, it couldn't
output sound.

The Auto-Mute is not useful if the pulseaudio is running since pa
could mute/unmute devices according to active port change, the ucm
for sof+hda already disabled the Auto-Mute, let us disable it for
hda audio if the machine has the internal speaker.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Part-of: <pulseaudio/pulseaudio!433>

nostrad authored Mar 26, 2020 and PulseAudio Marge Bot committed Dec 10, 2020

Since there is now support for specifying the index of an Element, add the
same config as is used for the output-mono variant, as they behave the same:
One volume control with no support for adjustments to the left and right
channels.

Part-of: <pulseaudio/pulseaudio!274>

Tanu Kaskinen authored Mar 14, 2020 and PulseAudio Marge Bot committed Dec 10, 2020

Some gaming sound cards have custom profiles with analog-game and
analog-chat mappings that exist simultaneously. The game sink should
have higher priority than the chat sink, but currently there's no way to
affect the sink priorities from the profile-set configuration (the
mapping priority is not propagated to the sink priority).

I first thought about adding the mapping priority to the sink priority,
but that could mess up the prioritization system in
pa_device_init_priority(). I ended up checking for the intended roles
property to reduce the chat sink priority. I also reduced the iec958
priority so that the chat and iec958 sinks don't end up with the same
priority.

Fixes: pulseaudio/pulseaudio#818
Part-of: <pulseaudio/pulseaudio!266>

Tanu Kaskinen authored Mar 14, 2020 and PulseAudio Marge Bot committed Dec 10, 2020

Despite the ratelimiting, this error gets logged too much.

I fixed a typo (Fallig -> Falling) while at it.

Related: pulseaudio/pulseaudio#824
Part-of: <pulseaudio/pulseaudio!265>

Patrick McLean authored Feb 26, 2020 and PulseAudio Marge Bot committed Dec 09, 2020

This same profile should also work for the GSX 1200, but I don't know the USB id
for that.

Part-of: <pulseaudio/pulseaudio!257>

Pierre Ossman (Work account) authored Jul 07, 2015 and Georg Chini committed Dec 09, 2020

Aids in tracking down exactly where in the chain underruns are
happening.
Signed-off-by: Pierre Ossman <ossman@cendio.se>

Tanu Kaskinen authored Feb 14, 2020 and Georg Chini committed Dec 09, 2020

In OpenEmbedded the PulseAudio recipe currently disables Valgrind
support by passing "ac_cv_header_valgrind_memcheck_h=no" to the
configure script (this was added to make it deterministic whether
Valgrdind support gets enabled or not). I'm converting the PulseAudio
recipe to use Meson, and I needed an option to disable Valgrind.

Tanu Kaskinen authored Feb 11, 2020 and Georg Chini committed Dec 09, 2020

Since there's no stable API for modules, all modules need to be compiled
together with the server. This version check tries to ensure that if
a version mismatch happens, there will be an informative error message
rather than a random crash.

Frédéric Danis authored May 28, 2019 and Georg Chini committed Dec 09, 2020

Since commit ad447d14 (in 2009) pa_read and pa_write take care of
handling EINTR error.
So, pa_read, pa_write, pa_iochannel_read and pa_iochannel_write can not
exit with errno set to EINTR, and testing it is useless.

Ben Buchwald authored Jan 06, 2020 and Georg Chini committed Dec 09, 2020

module-jackdbus-detect now accepts sink_name, sink_properties,
sink_client_name, sink_channel_map, source_name, source_properties,
source_client_name, and source_channel_map arguments that will be passed
through to module-jack-source and module-jack-sink (without the sink and
source prefixes, except where needed).

Jaechul Lee authored Dec 08, 2020 and PulseAudio Marge Bot committed Dec 09, 2020

This module uses 2 seconds for default max latency. If max_latency_msec
is set, max_latency value will be changed.

Part-of: <pulseaudio/pulseaudio!431>

Part-of: <pulseaudio/pulseaudio!432>

Dušan Kazik authored Nov 24, 2020 and Tanu Kaskinen committed Dec 08, 2020

Currently translated at 76.0% (400 of 526 strings)

Translation: pulseaudio/pulseaudio
Translate-URL: https://translate.fedoraproject.org/projects/pulseaudio/pulseaudio/sk/
Part-of: <pulseaudio/pulseaudio!432>

As long as modules use different keys, multiple loading of this module
should be safe. Even in case of same keys, either the pa_proplist_update
should correctly update the property list based on the mode or for the
volume writable case, the last rule should win.

Fixes: #942
Part-of: <!332>

Jaechul Lee authored Nov 16, 2020 and Tanu Kaskinen committed Dec 07, 2020

It takes much time when starting to capture because max latency is set
to 2 seconds as a initial value. null-source latency need to be set a
lower value than initial value to improve latency.

Igor Kovalenko authored Nov 30, 2020 and Tanu Kaskinen committed Dec 03, 2020

Part-of: <pulseaudio/pulseaudio!51>

Igor Kovalenko authored Nov 30, 2020 and Tanu Kaskinen committed Dec 03, 2020

Part-of: <pulseaudio/pulseaudio!51>

Part-of: <!51>

These tests aren't directly related to modargs. I suppose these were
first written for the modargs.h API and renaming was forgotten.

Part-of: <pulseaudio/pulseaudio!51>

Part-of: <pulseaudio/pulseaudio!51>

I believe nobody needs to pass octal numbers to PulseAudio, and if we
encounter integer strings starting with zeros, the intention is to use
them in base 10. Hexadecimal numbers are more common, and they can't be
interpreted in base 10 anyway, so they are still supported.

Part-of: <pulseaudio/pulseaudio!51>

Part-of: <pulseaudio/pulseaudio!51>