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

Part-of: <!51>

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

Part-of: <!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: <!51>

Part-of: <!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: <!51>

Part-of: <!51>

If an application calls the function when the server doesn't support the
feature, the result should be just an error from the function. Without
the check the whole connection gets terminated due to protocol error.

Part-of: <pulseaudio/pulseaudio!51>

It wasn't immediately obvious to me what these checks are supposed to
do. Explicitly checking against the min/max values should make the code
easier to understand.

Part-of: <!51>

Georg Chini authored Jan 14, 2020 and Tanu Kaskinen committed Dec 03, 2020

The following new functions have been added:

pa_message_params_read_double_array() - read an array of double from list
pa_message_params_read_int64_array() - read an array of int64 from list
pa_message_params_read_uint64_array() - read an array of uint64 from list
pa_message_params_read_string_array() - read an array of strings from list

Part-of: <!51>

Georg Chini authored Jan 14, 2020 and Tanu Kaskinen committed Dec 03, 2020

The following functions have been added:

pa_message_params_write_double() - writes a double to a pa_message_params structure
pa_message_params_write_int64() - writes an integer to a pa_message_params structure
pa_message_params_write_uint64() - writes an unsigned to a pa_message_params structure
pa_message_params_write_bool() - writes a boolean to a pa_message_params structure

pa_message_params_read_double() - read a double from a parameter list
pa_message_params_read_int64() - read an integer from a parameter list
pa_message_params_read_uint64() - read an unsigned from a parameter list
pa_message_params_read_bool() - read a boolean from a parameter list

The patch also improves the doxygen documentation im message-params.h

Part-of: <!51>

Georg Chini authored Jan 14, 2020 and Tanu Kaskinen committed Dec 03, 2020

Part-of: <!51>

Georg Chini authored Jan 14, 2020 and Tanu Kaskinen committed Dec 03, 2020

The patch adds the possibility to escape curly braces within parameter strings
and introduces several new functions that can be used for writing parameters.

For writing, the structure pa_message_params, which is a wrapper for pa_strbuf
has been created. Following new write functions are available:

pa_message_params_new() - creates a new pa_message_params structure
pa_message_params_free() - frees a pa_message_params structure
pa_message_param_to_string_free() - converts a pa_message_param to string and
frees the structure
pa_message_params_begin_list() - starts a list
pa_message_params_end_list() - ends a list
pa_message_params_write_string() - writes a string to a pa_message_params structure
pa_message_params_write_raw() - writes a raw string to a pa_message_params structure

For string parameters that contain curly braces or backslashes, those characters
will be escaped when using pa_message_params_write_string(), while write_raw() will
put the string into the buffer without any changes.

For reading, pa_message_params_read_string() reverts the changes that
pa_message_params_write_string() might have introduced.

The patch also adds more restrictions on the object path name. Now only
alphanumeric characters and one of "_", ".", "-" and "/" are allowed.
The path name may not end with a / or contain a double slash. If the user
specifies a trailing / when sending a message, it will be silently removed.

Part-of: <!51>

Georg Chini authored Jan 14, 2020 and Tanu Kaskinen committed Dec 03, 2020

For better readability, "pactl list message-handlers" is introduced which
prints a formatted output of "pactl send-message /core list-handlers".

The patch also adds the functions pa_message_params_read_raw() and
pa_message_params_read_string() for easy parsing of the message response
string. Because the functions need to modify the parameter string,
the message handler and the pa_context_string_callback function now
receive a char* instead of a const char* as parameter argument.

Part-of: <pulseaudio/pulseaudio!51>

Georg Chini authored Jan 14, 2020 and Tanu Kaskinen committed Dec 03, 2020

This patch adds a small message handler to the core which enables
clients to list available handlers via the list-handlers message.
Command: pacmd send-message /core list-handlers
pactl can be used with the same parameters.

The patch also introduces a convention for the return string.
It consists of a list of elements where curly braces are used
to separate elements. Each element can itself contain further
elements. For example consider a message that returns multiple
elements which each contain an integer and an array of float.
A response string would look like that:
{{Integer} {{1st float} {2nd float} ...}}{...}

Part-of: <!51>

Georg Chini authored Jan 14, 2020 and Tanu Kaskinen committed Dec 03, 2020

Part-of: <!51>

Georg Chini authored Jan 14, 2020 and Tanu Kaskinen committed Dec 03, 2020

This patch adds the PA_COMMAND_SEND_OBJECT_MESSAGE command to protocol-native
so that clients can use the messaging feature introduced in the previous patch.

Sending messages can in effect replace the extension system for modules. The
approach is more flexible than the extension interface because a generic string
format is used to exchange information. Furthermore the messaging system can be
used for any object, not only for modules, and is easier to implement than
extensions.

Part-of: <!51>

target_machine provides information about the machine on which the compiled
binary's output will run.

cpu_family() returns CPU family name (such as x86_64, not more specific amd64)

Part-of: <pulseaudio/pulseaudio!426>

Restore an option to disable OSS output available with autotools.

Jaechul Lee authored Nov 06, 2020 and PulseAudio Marge Bot committed Nov 27, 2020

playing sound through null sink takes almost 2 seconds at first time
playback when norewinds is set. Because block_usec is set 2 seconds at
initializing time. The value will be changed 50 msec after calling
update_request_latency callback.

Part-of: <pulseaudio/pulseaudio!406>

At least Gentoo and OpenSUSE install alsa-info.sh in /usr/sbin, look there too.

Part-of: <pulseaudio/pulseaudio!422>

Laurent Bigonville authored Aug 27, 2020 and PulseAudio Marge Bot committed Nov 25, 2020

Debian kfreebsd uses the GNU libc that uses cpu_set_t instead of cpuset_t

Also do not include unnecessary headers on this platform

Fixes: #851
Part-of: <pulseaudio/pulseaudio!356>

Christopher Snowhill authored Jan 23, 2020 and PulseAudio Marge Bot committed Nov 24, 2020

This replaces the original virtual surround sink with a total
rewrite, aiming to implement any number of hrir use cases,
including asymmetrical impulses as two separate left and right
output files. It uses FFTW3 FFT convolution, using the overlap-
save method, with full rewind support. It operates in steps
equal to the resampled length of the hrir, and overlaps input
blocks in increments equal to the size of the FFT block. If
using paired hrirs, it requires matched sample spec and sample
rates and channel maps. For best results, the input files should
have speaker maps, rather than expecting the sample loader to
auto detect the mapping.

Part-of: <pulseaudio/pulseaudio!240>

Arun Raghavan authored Nov 23, 2020 and Arun Raghavan committed Nov 23, 2020

This comment was missed in 757eb264.

Kai-Heng Feng authored Sep 02, 2020 and Arun Raghavan committed Nov 23, 2020

The HP Thunderbolt Dock [1] has two separate USB cards, a headset jack
and an optional module which is a speakerphone.

This patch adds new description for them, and mark the intended-roles as
phone for the speakerphone module.

[1] https://store.hp.com/us/en/pdp/hp-thunderbolt-dock-120w-g2-with-audio

Sanchayan Maity authored Sep 02, 2020 and Arun Raghavan committed Nov 23, 2020

The .include meta command already supports specifying a directory and
when including a directory, all files with the extension '.pa' in that
directory will be parsed in alphabetical order.

This feature can be used to add support for default.pa.d directory, so
that packages for other applications or users can just drop in a file
for configuration without changing the default.pa which is shipped.

We use the PA_DEFAULT_CONFIG_DIR for this, however, since meson quotes
this build variable, introduce an unquoted version for this purpose and
use it with .include.

Fixes: pulseaudio/pulseaudio#909



Signed-off-by: Sanchayan Maity <sanchayan@asymptotic.io>

Tanu Kaskinen authored Sep 03, 2020 and Arun Raghavan committed Nov 23, 2020

We already supported the CLFE element, which should be semantically
equivalent, so I just copied all the CLFE element definitions.

The Center/LFE element is seen on Creative X-Fi with 20K1 chipset cards.

Fixes: #978

Tanu Kaskinen authored Sep 22, 2020 and Arun Raghavan committed Nov 23, 2020

The webrtc backend of module-echo-cancel uses sscanf() to parse floating
point numbers from module arguments, which didn't work when the locale
used a comma for the decimal point. Setting the LC_NUMERIC locale
variable to C makes the pulseaudio process use a period as the decimal
point regardless of the user's locale configuration.

Fixes: pulseaudio/pulseaudio#89

Tanu Kaskinen authored Sep 24, 2020 and Arun Raghavan committed Nov 23, 2020

This call has been seen failing on FreeBSD, and it's difficult to debug
without seeing which error the function returned.

Fixes: pulseaudio/pulseaudio#988

Tanu Kaskinen authored Oct 11, 2020 and Arun Raghavan committed Nov 23, 2020

If there are RAOP devices in the network, it doesn't necessarily mean
that the user wants to play music to them.

Fixes: pulseaudio/pulseaudio#993

Arun Raghavan authored Jul 23, 2020 and Arun Raghavan committed Nov 23, 2020

Newer GCC warns us that the channel_map and volume in legacy entries are
accessed via pointers, and these might be unaligned as the legacy entry
is a packed structure. For this reason, we read out those values into
local variables before accessing them as pointers.

The warnings are:

[146/433] Compiling C object src/modules/module-device-restore.so.p/module-device-restore.c.o
../src/modules/module-device-restore.c: In function ‘legacy_entry_read’:
../src/modules/module-device-restore.c:554:51: warning: taking address of packed member of ‘struct legacy_entry’ may result in an unaligned pointer value [-Waddress-of-packed-member]
  554 |     if (le->volume_valid && !pa_channel_map_valid(&le->channel_map)) {
      |                                                   ^~~~~~~~~~~~~~~~
../src/modules/module-device-restore.c:559:48: warning: taking address of packed member of ‘struct legacy_entry’ may result in an unaligned pointer value [-Waddress-of-packed-member]
  559 |     if (le->volume_valid && (!pa_cvolume_valid(&le->volume) || !pa_cvolume_compatible_with_channel_map(&le->volume, &le->channel_map))) {
      |                                                ^~~~~~~~~~~
../src/modules/module-device-restore.c:559:104: warning: taking address of packed member of ‘struct legacy_entry’ may result in an unaligned pointer value [-Waddress-of-packed-member]
  559 |     if (le->volume_valid && (!pa_cvolume_valid(&le->volume) || !pa_cvolume_compatible_with_channel_map(&le->volume, &le->channel_map))) {
      |                                                                                                        ^~~~~~~~~~~
../src/modules/module-device-restore.c:559:117: warning: taking address of packed member of ‘struct legacy_entry’ may result in an unaligned pointer value [-Waddress-of-packed-member]
  559 |     if (le->volume_valid && (!pa_cvolume_valid(&le->volume) || !pa_cvolume_compatible_with_channel_map(&le->volume, &le->channel_map))) {
      |                                                                                                                     ^~~~~~~~~~~~~~~~
[211/433] Compiling C object src/modules/module-stream-restore.so.p/module-stream-restore.c.o
../src/modules/module-stream-restore.c: In function ‘legacy_entry_read’:
../src/modules/module-stream-restore.c:1076:51: warning: taking address of packed member of ‘struct legacy_entry’ may result in an unaligned pointer value [-Waddress-of-packed-member]
 1076 |     if (le->volume_valid && !pa_channel_map_valid(&le->channel_map)) {
      |                                                   ^~~~~~~~~~~~~~~~
../src/modules/module-stream-restore.c:1081:48: warning: taking address of packed member of ‘struct legacy_entry’ may result in an unaligned pointer value [-Waddress-of-packed-member]
 1081 |     if (le->volume_valid && (!pa_cvolume_valid(&le->volume) || !pa_cvolume_compatible_with_channel_map(&le->volume, &le->channel_map))) {
      |                                                ^~~~~~~~~~~
../src/modules/module-stream-restore.c:1081:104: warning: taking address of packed member of ‘struct legacy_entry’ may result in an unaligned pointer value [-Waddress-of-packed-member]
 1081 |     if (le->volume_valid && (!pa_cvolume_valid(&le->volume) || !pa_cvolume_compatible_with_channel_map(&le->volume, &le->channel_map))) {
      |                                                                                                        ^~~~~~~~~~~
../src/modules/module-stream-restore.c:1081:117: warning: taking address of packed member of ‘struct legacy_entry’ may result in an unaligned pointer value [-Waddress-of-packed-member]
 1081 |     if (le->volume_valid && (!pa_cvolume_valid(&le->volume) || !pa_cvolume_compatible_with_channel_map(&le->volume, &le->channel_map))) {
      |

Arun Raghavan authored Jul 23, 2020 and Arun Raghavan committed Nov 23, 2020

This uses the year 2038-safe g_get_real_time() as recommended instead.
Bumps GLib dependency to 2.28 as a result.

Arun Raghavan authored Jul 23, 2020 and Arun Raghavan committed Nov 23, 2020

The preference in glib is now to just use the const qualifier directly.

This is because we reverted the previous API breaking change to the
meaning of the pa_mainloop_prepare() timeout argument.

Arun Raghavan authored Nov 23, 2020 and Arun Raghavan committed Nov 23, 2020

This reverts commit 6b1719d0, as it
inadvertently broke the semantics of timeout in the API.

Fixes: pulseaudio/pulseaudio#1039

With Meson GStreamer is already disabled by default, let's match that
with Autotools.

pali authored Nov 21, 2020 and Tanu Kaskinen committed Nov 23, 2020

modules/alsa/alsa-sink.c: In function ‘pa_alsa_sink_new’:
modules/alsa/alsa-sink.c:2603:15: warning: declaration of ‘state’ shadows a previous local [-Wshadow]
         void *state;
               ^~~~~
modules/alsa/alsa-sink.c:2270:11: note: shadowed declaration is here
     void *state = NULL;
           ^~~~~
  CC       modules/alsa/module_alsa_sink_la-module-alsa-sink.lo
modules/alsa/alsa-source.c: In function ‘pa_alsa_source_new’:
modules/alsa/alsa-source.c:2289:15: warning: declaration of ‘state’ shadows a previous local [-Wshadow]
         void *state;
               ^~~~~
modules/alsa/alsa-source.c:1975:11: note: shadowed declaration is here
     void *state = NULL;
           ^~~~~
modules/alsa/module-alsa-card.c: In function ‘prune_singleton_availability_groups’:
modules/alsa/module-alsa-card.c:691:71: warning: pointer of type ‘void *’ used in arithmetic [-Wpointer-arith]
             pa_hashmap_put(group_counts, p->availability_group, count + 1);
                                                                       ^

Tanu Kaskinen authored Nov 21, 2020 and Arun Raghavan committed Nov 23, 2020

Commits 323195e3 ("switch-on-port-available: Switch to headphones on
unknown availability") and d83ad699 ("module-alsa-card: Drop
availability groups with only one port") broke switching from headphones
to speakers when headphones are unplugged. switch_from_port() selects
speakers, whose availability is unknown and availability group is unset,
and then calls switch_to_port(). The new logic in switch_on_port()
unintentionally blocked that switch.

This patch moves the problematic logic from switch_to_port() to
port_available_hook_callback() where it doesn't interfere with
switch_from_port().

Fixes: pulseaudio/pulseaudio#1043