platform/upstream/pulseaudio.git
10 years agomodule: Fix crash in pa_module_unload_all() 96/8496/2 accepted/tizen/20130912.154532 accepted/tizen/20130912.190419 accepted/tizen/20130912.194640 submit/tizen/20130912.080303 submit/tizen/20130912.090605
Tanu Kaskinen [Thu, 11 Jul 2013 10:35:24 +0000 (13:35 +0300)]
module: Fix crash in pa_module_unload_all()

Previously, if there were no modules loaded when the daemon exited,
pa_module_unload_all() would crash due to giving zero count to
pa_xnew().

Thanks to Pierre Ossman for the patch.

Bug-Tizen: TZPC-4359

10 years agoset conflicting modules with murphy behind conditionals 31/8431/2 submit/tizen/20130820.023658
Jaska Uimonen [Mon, 19 Aug 2013 13:06:54 +0000 (16:06 +0300)]
set conflicting modules with murphy behind conditionals

these modules should not be loaded with module-murphy-ivi

10 years agomultilib support: re-add the baselibs.conf to packaging 41/8341/2
Patrick McCarty [Fri, 16 Aug 2013 17:42:11 +0000 (10:42 -0700)]
multilib support: re-add the baselibs.conf to packaging

The baselibs.conf is required to support multilib packages; currently,
pulseaudio-32bit and libpulse-32bit are required for the Steam
client and various Steam games.

Change-Id: Ia489433b3d08363320bf7546aebe229d96c5cfcc
Signed-off-by: Patrick McCarty <patrick.mccarty@linux.intel.com>
10 years agoremove pulseaudio.service which launch pa in system mode
vivian,zhang [Fri, 16 Aug 2013 04:09:07 +0000 (12:09 +0800)]
remove pulseaudio.service which launch pa in system mode

Change-Id: I48038cd35b05d395cfaf5f9a167ecb2261fe253c

10 years agocheck module-fileter-xx.so exists to avoid module load failure
vivian,zhang [Fri, 16 Aug 2013 04:07:29 +0000 (12:07 +0800)]
check module-fileter-xx.so exists to avoid module load failure

Change-Id: Id6384e358ed3bac62d14874e8fe22e8792a637d9

10 years agoRemove the toplevel manifest 40/8340/1
Patrick McCarty [Thu, 15 Aug 2013 17:39:55 +0000 (10:39 -0700)]
Remove the toplevel manifest

There is already a manifest in packaging/, so the toplevel one should be
removed.

Change-Id: I38603c91e195b93bae16a561193a4b8a6d2e026d
Signed-off-by: Patrick McCarty <patrick.mccarty@linux.intel.com>
10 years agoresetting manifest requested domain to floor submit/tizen/20130815.114042 submit/tizen/20130815.114433
vivian,zhang [Thu, 18 Jul 2013 10:51:40 +0000 (18:51 +0800)]
resetting manifest requested domain to floor

Change-Id: Idd9eed5f535b8cfad5610f9b4f39e0422d6e86e5

10 years agorelease: updated changelog.
Jaska Uimonen [Tue, 11 Jun 2013 14:18:16 +0000 (17:18 +0300)]
release: updated changelog.

10 years agorelease: added packaging for gerrit.
Jaska Uimonen [Tue, 11 Jun 2013 14:17:16 +0000 (17:17 +0300)]
release: added packaging for gerrit.

Change-Id: Iae038b2a9b8034d37a8bd88f71ff645d66445266

10 years agopackaging: added obs export script
Jaska Uimonen [Tue, 11 Jun 2013 14:05:01 +0000 (17:05 +0300)]
packaging: added obs export script

10 years agopackaging: adding pulseaudio manifest for smack
Jaska Uimonen [Thu, 8 Aug 2013 13:00:34 +0000 (16:00 +0300)]
packaging: adding pulseaudio manifest for smack

10 years agopackaging: added pulseaudio.spec.in
Jaska Uimonen [Tue, 11 Jun 2013 14:04:15 +0000 (17:04 +0300)]
packaging: added pulseaudio.spec.in

10 years agopackaging: added pulseaudio.service
Jaska Uimonen [Tue, 11 Jun 2013 14:03:27 +0000 (17:03 +0300)]
packaging: added pulseaudio.service

10 years agocombine: fix crash in output freeing - upstream patch
vivian,zhang [Thu, 18 Jul 2013 07:24:41 +0000 (15:24 +0800)]
combine: fix crash in output freeing - upstream patch

Change-Id: I036147804f985721fe1d06b2f8eb423f11b34848

10 years agoAdd bt_profile_set patch which fixed bt a2dp&hsp profile setting issues in mobile
vivian,zhang [Wed, 17 Jul 2013 03:17:40 +0000 (11:17 +0800)]
Add bt_profile_set patch which fixed bt a2dp&hsp profile setting issues in mobile

Change-Id: I3a83fa96cd5ed9d22dbb7630cc1cc79414a8845e

10 years agouse udev only for usb devices - samsung
Jaska Uimonen [Thu, 8 Aug 2013 08:28:39 +0000 (11:28 +0300)]
use udev only for usb devices - samsung

10 years agocope with possible infinite waiting in startup - samsung
Jaska Uimonen [Thu, 8 Aug 2013 08:27:44 +0000 (11:27 +0300)]
cope with possible infinite waiting in startup - samsung

10 years agoset alsa suspend timeout to zero - samsung
Jaska Uimonen [Thu, 8 Aug 2013 08:24:25 +0000 (11:24 +0300)]
set alsa suspend timeout to zero - samsung

10 years agocreate pa_ready file - samsung
Jaska Uimonen [Thu, 8 Aug 2013 08:23:38 +0000 (11:23 +0300)]
create pa_ready file - samsung

10 years agoadd bluetooth a2dp aptx codec support - samsung
vivian,zhang [Tue, 18 Jun 2013 08:23:45 +0000 (16:23 +0800)]
add bluetooth a2dp aptx codec support - samsung

Change-Id: I14ed8c1dfc1031afcd8428b8951aad63a132ea8f

10 years agoadd policy module - samsung
vivian,zhang [Tue, 18 Jun 2013 08:21:32 +0000 (16:21 +0800)]
add policy module - samsung

Change-Id: I50921a731805f180cff2feffff240c0808124685

10 years agoadd support for dlog - samsung
vivian,zhang [Tue, 18 Jun 2013 08:20:04 +0000 (16:20 +0800)]
add support for dlog - samsung

Change-Id: Id795c76941710aaf7c984f375f0f0e51297ada50

10 years agoEnhance for echo cancel - samsung
vivian,zhang [Tue, 18 Jun 2013 08:18:58 +0000 (16:18 +0800)]
Enhance for echo cancel - samsung

Change-Id: I625ca676228b58b5d53f72b9d14b3e811d212d1e

10 years agoAdd preload fileter for resample - samsung
vivian,zhang [Tue, 18 Jun 2013 08:12:20 +0000 (16:12 +0800)]
Add preload fileter for resample - samsung

Change-Id: Ia57b4eb1848d53bbfa15c04b25f4821554aff28f

10 years agoadd support for samsung power management - samsung
vivian,zhang [Tue, 18 Jun 2013 08:11:16 +0000 (16:11 +0800)]
add support for samsung power management - samsung

Change-Id: Id6efaab5a9888fc5eec2883c6be35090f6df375c

10 years agochanges to pa simple api - samsung
vivian,zhang [Tue, 18 Jun 2013 08:10:15 +0000 (16:10 +0800)]
changes to pa simple api - samsung

Change-Id: I997c02217a8dc14524480164aa0baeea901c7b4e

10 years agoadjust default bluetooth profile to off
Jaska Uimonen [Wed, 14 Aug 2013 12:00:52 +0000 (15:00 +0300)]
adjust default bluetooth profile to off

10 years agoadd conditional murphy module loading to default.pa.in
vivian,zhang [Thu, 18 Jul 2013 10:43:14 +0000 (18:43 +0800)]
add conditional murphy module loading to default.pa.in

Change-Id: I8199f4530ef5f10cf65488969b3a8998e05cb65c

10 years agofix draining not to lose prebuffered samples
Jaska Uimonen [Wed, 8 May 2013 12:31:07 +0000 (15:31 +0300)]
fix draining not to lose prebuffered samples

10 years agoadd internal corking state for sink-input
Jaska Uimonen [Thu, 7 Mar 2013 11:41:44 +0000 (13:41 +0200)]
add internal corking state for sink-input

10 years agonode-manager: adding node support for pactl
Jaska Uimonen [Wed, 5 Dec 2012 07:53:12 +0000 (09:53 +0200)]
node-manager: adding node support for pactl

10 years agonode manager: adding external node manager API
Jaska Uimonen [Tue, 4 Dec 2012 07:53:14 +0000 (09:53 +0200)]
node manager: adding external node manager API

10 years agomake pa_thread_mq_done() safe for subsequent calls
Janos Kovacs [Thu, 16 Aug 2012 00:47:48 +0000 (03:47 +0300)]
make pa_thread_mq_done() safe for subsequent calls

10 years agojack detection fix for wired headset
Jaska Uimonen [Sun, 10 Jun 2012 12:13:11 +0000 (15:13 +0300)]
jack detection fix for wired headset

10 years agobuild-sys: install files for a module development.
Jaska Uimonen [Tue, 5 Jun 2012 08:36:13 +0000 (11:36 +0300)]
build-sys: install files for a module development.

10 years agovolume ramp: add client api support for volume ramping
Jaska Uimonen [Wed, 8 Aug 2012 08:14:40 +0000 (11:14 +0300)]
volume ramp: add client api support for volume ramping

10 years agovolume ramp: adding volume ramping to sink-input
Jaska Uimonen [Wed, 8 Aug 2012 08:14:38 +0000 (11:14 +0300)]
volume ramp: adding volume ramping to sink-input

10 years agovolume ramp: add volume ramping to sink
Jaska Uimonen [Wed, 8 Aug 2012 08:14:39 +0000 (11:14 +0300)]
volume ramp: add volume ramping to sink

10 years agovolume ramp: additions to the low level infra
Jaska Uimonen [Wed, 8 Aug 2012 08:14:37 +0000 (11:14 +0300)]
volume ramp: additions to the low level infra

10 years agotests: Make loopback latency test more accurate master list upstream/4.0
Arun Raghavan [Mon, 27 May 2013 08:57:42 +0000 (14:27 +0530)]
tests: Make loopback latency test more accurate

This makes sure that we always take the timing at the point when we
write out the pulse, making the overall latency measurement more
accurate.

10 years agotests: Factor out loopback setup code
Arun Raghavan [Thu, 23 May 2013 09:57:40 +0000 (15:27 +0530)]
tests: Factor out loopback setup code

This moves over setup code for the loopback latency test into a private
library so that we can easily write more tests using the same framework.

10 years agotests: Adjust latency test calibration
Arun Raghavan [Wed, 22 May 2013 10:29:24 +0000 (15:59 +0530)]
tests: Adjust latency test calibration

This makes the test more robust by:

1. Decreasing the '1' threshold during calibration - the RMS value for
   the sine wave will be 0.5, so the previous code was making us take
   the ALSA mixer past 0dB.

2. Using the difference rather than absolute value for 0->1 transitions,
   so that we're somewhat independent noise in our calculations.

10 years agotests: Make loopback latency test run with rt priority
Arun Raghavan [Wed, 22 May 2013 06:33:47 +0000 (12:03 +0530)]
tests: Make loopback latency test run with rt priority

10 years agotests: Add a latency measurement test
Arun Raghavan [Tue, 21 May 2013 13:09:30 +0000 (18:39 +0530)]
tests: Add a latency measurement test

This test is intended to measure real latency by playing a sample to a
sink and capturing that over a loopback interface. The loopback can
either be physical (cable running from headphone out to line in) or
virtual (monitor source or module loopback).

Also included in this is calibration code to make sure that volumes are
sufficiently adjusted to be able to detect the played back signal (and
that there aren't false positives due to line noise).

One of the objectives of all this is to later factor out the setup code
to allow us to easily write more loopback tests for various
functionality (volumes, resampling, mixing, etc.).

10 years agobluetooth: Fix input port description
Tanu Kaskinen [Fri, 24 May 2013 12:17:13 +0000 (15:17 +0300)]
bluetooth: Fix input port description

10 years agozeroconf: Make Avahi usage in m-z-publish async
Arun Raghavan [Wed, 15 May 2013 04:11:19 +0000 (09:41 +0530)]
zeroconf: Make Avahi usage in m-z-publish async

This pushes all avahi-client code to a threaded mainloop from the PA
mainloop context. We need to do this because avahi-client makes blocking
D-Bus calls, and we don't want to block the mainloop for that long.

The only exception to this now that I don't see a workaround for is
during module unload time. However, this shouldn't be a huge problem
since in most cases, this will only happen at server shutdown time.

The bulk of the change is partitioning the data so that PA core objects
only (well, mostly) get accessed in the PA mainloop and Avahi calls
happen only in the Avahi threaded mainloop.

Bug: https://bugs.freedesktop.org/show_bug.cgi?id=58758

10 years agomainloop: Add API to set thread name for threaded mainloop
Arun Raghavan [Wed, 15 May 2013 04:07:45 +0000 (09:37 +0530)]
mainloop: Add API to set thread name for threaded mainloop

10 years agopulsecore: Fix assert in pa_msgobject creation
Arun Raghavan [Wed, 15 May 2013 04:06:17 +0000 (09:36 +0530)]
pulsecore: Fix assert in pa_msgobject creation

Allows for creation of derived types that don't have any other fields.

10 years agoshell-completion: Add list-cards to the pacmd completion.
poljar (Damir Jelić) [Wed, 22 May 2013 12:58:10 +0000 (14:58 +0200)]
shell-completion: Add list-cards to the pacmd completion.

The list-cards command was missing for both of the shell completions.
This fixes it.

10 years agorescue-streams: Fix wording of module description
Peter Meerwald [Wed, 15 May 2013 07:57:56 +0000 (09:57 +0200)]
rescue-streams: Fix wording of module description

Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
10 years agointended-roles: Fix typo in module description
Peter Meerwald [Wed, 15 May 2013 07:53:30 +0000 (09:53 +0200)]
intended-roles: Fix typo in module description

Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
10 years agopactl: Use colon to separate priority in profiles output
Peter Meerwald [Wed, 15 May 2013 07:39:26 +0000 (09:39 +0200)]
pactl: Use colon to separate priority in profiles output

for example:
  Profiles:
input:analog-stereo: Analog Stereo Input (sinks: 0, sources: 1, priority. 60)
output:analog-stereo: Analog Stereo Output (sinks: 1, sources: 0, priority. 6000)

it should be "priority: xxx", not "priority. xxx"

Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
10 years agothread-mainloop: Fix bug in example code
Mitchell Fang [Wed, 22 May 2013 08:23:27 +0000 (11:23 +0300)]
thread-mainloop: Fix bug in example code

Checking the operation state caused a deadlock, because the state
won't change before my_drain_callback() returns, and it doesn't
return before my_drain_stream_func() calls
pa_threaded_mainloop_accept().

10 years agocli: Fix plural in caption of source-outputs list
Peter Meerwald [Wed, 15 May 2013 15:49:42 +0000 (17:49 +0200)]
cli: Fix plural in caption of source-outputs list

$ pacmd list-source-outputs
>>> 4 source outputs(s) available.

outputs(s) makes no sense

Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
10 years agobluetooth: Support transport auto-release
Mikel Astiz [Fri, 10 May 2013 08:30:47 +0000 (10:30 +0200)]
bluetooth: Support transport auto-release

With BlueZ 5, if the remote device suspends the audio, the transport
state will change to "idle" and the endpoint is not required to release
the transport, since this could introduce race conditions. Therefore,
ignore the call to pa_bluetooth_transport_release() if the transport is
not acquired any more.

10 years agobluetooth: Update to new BlueZ 5 transport acquire/release API
Mikel Astiz [Fri, 10 May 2013 08:30:46 +0000 (10:30 +0200)]
bluetooth: Update to new BlueZ 5 transport acquire/release API

The new D-Bus API doesn't support access rights, which weren't used by
PulseAudio anyway, but it does solve a race condition: now optional
acquires can be implemented by bluetooth-util atomically using the D-Bus
TryAcquire() method.

10 years agobluetooth: Support media transport's State property
Mikel Astiz [Fri, 10 May 2013 08:30:45 +0000 (10:30 +0200)]
bluetooth: Support media transport's State property

BlueZ 5 exposes a 'State' property in the media transport interface.
With regard to PA, this replaces the profile-specific interfaces, since
they were being used to know if the audio was streaming or not.

10 years agobluetooth: Parse media transport's properties
Mikel Astiz [Fri, 10 May 2013 08:30:44 +0000 (10:30 +0200)]
bluetooth: Parse media transport's properties

Add the code to parse the properties of the media transport object when
a PropertiesChanged signal is received.

Note that the transport might have an owner other than BlueZ, and thus
the property changes would be emitted from arbitrary senders. For
performance reasons, the installed match considers the interface name
where the property has changed.

It could be possible to install and remove the D-Bus matches dynamically
when a new owner is registered/unregistered, but filtering based on the
interface name seems good enough already.

10 years agobluetooth: Support Properties.PropertiesChanged signal
Mikel Astiz [Fri, 10 May 2013 08:30:43 +0000 (10:30 +0200)]
bluetooth: Support Properties.PropertiesChanged signal

Install matches for signal Properties.PropertiesChanged and process the
properties corresponding to the tracked devices.

10 years agobluetooth: Support ObjectManager interface add/remove
Mikel Astiz [Fri, 10 May 2013 08:30:42 +0000 (10:30 +0200)]
bluetooth: Support ObjectManager interface add/remove

Install matches for signals ObjectManager.InterfacesAdded and
ObjectManager.InterfacesRemoved, and process the devices that are
registered and unregistered dynamically.

10 years agocore: Add an "internal" suspend cause
Arun Raghavan [Mon, 23 Jul 2012 08:50:05 +0000 (14:20 +0530)]
core: Add an "internal" suspend cause

This lets us suspend devices from within the core for short periods
without having to overload one of the existing suspend causes.

https://bugs.freedesktop.org/show_bug.cgi?id=64118

10 years agopactl: Flush stdout buffer when printing subscribe events.
Nikolay Amiantov [Sun, 5 May 2013 11:22:53 +0000 (15:22 +0400)]
pactl: Flush stdout buffer when printing subscribe events.

"pactl subscribe" is running continuously, and without flushing its output is
not usable for "process-on-arrival" per-line tasks, such as grepping. This
patch should fix this. For example, now:
pactl subscribe | grep 'server'
should print only server events as they arrive.

10 years agobluetooth: BlueZ 5 interface rename to org.bluez.MediaTransport1
Mikel Astiz [Mon, 29 Apr 2013 16:28:04 +0000 (18:28 +0200)]
bluetooth: BlueZ 5 interface rename to org.bluez.MediaTransport1

Use the new interface name if BlueZ 5 has been detected.

10 years agobluetooth: BlueZ 5 interface rename to org.bluez.Media1
Mikel Astiz [Mon, 29 Apr 2013 16:28:03 +0000 (18:28 +0200)]
bluetooth: BlueZ 5 interface rename to org.bluez.Media1

Use the new interface name if BlueZ 5 has been detected.

10 years agobluetooth: BlueZ 5 interface rename to org.bluez.MediaEndpoint1
Mikel Astiz [Mon, 29 Apr 2013 16:28:02 +0000 (18:28 +0200)]
bluetooth: BlueZ 5 interface rename to org.bluez.MediaEndpoint1

Use the new interface name if BlueZ 5 has been detected.

10 years agobluetooth: Fix error checking style
Tanu Kaskinen [Wed, 1 May 2013 10:39:36 +0000 (13:39 +0300)]
bluetooth: Fix error checking style

10 years agobluetooth: Parse the tree returned by ObjectManager
Mikel Astiz [Mon, 29 Apr 2013 16:27:59 +0000 (18:27 +0200)]
bluetooth: Parse the tree returned by ObjectManager

Parse the result of ObjectManager.GetManagedObjects(), which includes
all objects registered, their interfaces and the corresponding
properties per interface.

10 years agobluetooth: Detect BlueZ 5
Mikel Astiz [Mon, 29 Apr 2013 16:27:58 +0000 (18:27 +0200)]
bluetooth: Detect BlueZ 5

Check the existence of ObjectManager to detect the version of the
running daemon. If the interface exists, it should be BlueZ 5.

10 years agomodule: Assign the index before calling init()
Tanu Kaskinen [Mon, 29 Apr 2013 13:46:13 +0000 (16:46 +0300)]
module: Assign the index before calling init()

Any code that runs inside the init() callback sees an invalid module
index. Sometimes init() does things that cause hooks to be fired. This
means that any code that uses hooks may see an invalid module index.
Fix this by assigning the module index before init() is called.

There are no known issues in the upstream code base where an invalid
module index would be used, but an out-of-tree module
(module-murphy-ivi) had a problem with this.

BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=63923
10 years agobluetooth: Remove the 'bluez.name' property
João Paulo Rechi Vita [Sat, 27 Apr 2013 00:24:52 +0000 (21:24 -0300)]
bluetooth: Remove the 'bluez.name' property

The 'Name' property of the Device interface became optional in BlueZ 5
and may not be present anymore (that happens when testing against the
PTS 4.7.0), so it's better not to expose it to clients so they don't
rely on its existence.

10 years agobluetooth: Use 'Alias' value as the device description
João Paulo Rechi Vita [Sat, 27 Apr 2013 00:26:36 +0000 (21:26 -0300)]
bluetooth: Use 'Alias' value as the device description

The 'Alias' property should be preffered over the 'Name' property,
according to the BlueZ API documentation.

10 years agobluetooth: Add 'bluez.alias' property
João Paulo Rechi Vita [Fri, 26 Apr 2013 15:30:24 +0000 (12:30 -0300)]
bluetooth: Add 'bluez.alias' property

10 years ago.gitignore: Ignore src/*.gcno files created when building with --enable-gcov
Peter Meerwald [Wed, 17 Apr 2013 07:49:03 +0000 (09:49 +0200)]
.gitignore: Ignore src/*.gcno files created when building with --enable-gcov

Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
10 years agoswitch-on-port-available: Silence gcc warning.
poljar (Damir Jelić) [Fri, 19 Apr 2013 14:35:14 +0000 (16:35 +0200)]
switch-on-port-available: Silence gcc warning.

This silences this gcc warning:
module-switch-on-port-available.c:111:12: warning:
    'good' may be used uninitialized in this function

10 years agomodule: initialize module index to invalid value.
Ismo Puustinen [Fri, 12 Apr 2013 14:38:12 +0000 (17:38 +0300)]
module: initialize module index to invalid value.

m->init() was called while m->index was uninitialized, which was bad
style.

10 years agofdsem: Remember pa_write() type in pa_fdsem_post()
Peter Meerwald [Tue, 9 Apr 2013 14:10:16 +0000 (16:10 +0200)]
fdsem: Remember pa_write() type in pa_fdsem_post()

pa_write() knows two types of operation:
calling send() and calling write()

there is a flag (a pointer to an int) passed to pa_write()
which can remember which write type was successful

if the pointer is NULL or the int is 0, send() is tried first,
with a fallback to write() if send() resulted in ENOTSOCK

pa_fdsem_post() calls pa_write() with a NULL pointer;
unfortunately (at least with HAVE_SYS_EVENTFD_H #define'd) send()
always fails here and write() is called -- causing an extra syscall
quite frequently

strace:
send(17, "\1\0\0\0\0\0\0\0", 8, MSG_NOSIGNAL) = -1 ENOTSOCK (Socket operation on non-socket)
write(17, "\1\0\0\0\0\0\0\0", 8) = 8

the patch adds a write_type field to pa_fdsem to the successful
pa_write() type can be remembered and unnecessary send() calls are
avoided

Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
10 years agopasuspender: Resume before exiting in case of SIGINT or fork() failure.
Tanu Kaskinen [Wed, 13 Jun 2012 09:45:16 +0000 (12:45 +0300)]
pasuspender: Resume before exiting in case of SIGINT or fork() failure.

Pressing Ctrl-C in a terminal while pasuspender is running
causes the sinks and sources to stay suspended after
pasuspender has exited, which is very annoying. This patch
fixes that problem, and also a similar problem with fork()
failures.

10 years agoalsa: Don't use pa_strna() for port names
Tanu Kaskinen [Fri, 29 Mar 2013 15:17:14 +0000 (17:17 +0200)]
alsa: Don't use pa_strna() for port names

The name variable is never NULL, so there's no point in using
pa_strna().

10 years agocore-util: Don't accept random words in pa_parse_boolean()
Tanu Kaskinen [Wed, 13 Jun 2012 08:15:03 +0000 (11:15 +0300)]
core-util: Don't accept random words in pa_parse_boolean()

The old code accepted any word that started with "y", "Y",
"n", "N", "t", "T", "f" or "F". Fix this by having
a whitelist of full strings instead of checking just the
first letter.

10 years agobuild: make ARM NEON check in configure.ac more strict
Peter Meerwald [Thu, 28 Mar 2013 13:46:48 +0000 (14:46 +0100)]
build: make ARM NEON check in configure.ac more strict

the check for NEON so far only checked if -mfpu=neon is understood by the compiler,
however, this is not enough:

(i) #include <arm_neon.h> should be checked
(ii) -mfpu=neon must be passed before CFLAGS because eventually the per-library CFLAGS
for NEON code in src/Makefile.am are passed to the compiler before the global CFLAGS

in case the build environment passes CFLAGS to configure and we try to set conflicting
CFLAGS option, the former take precedence; CFLAGS cannot be overridden

this does not fix
http://lists.freedesktop.org/archives/pulseaudio-discuss/2012-December/015570.html
but at least makes the build fail in configure (and not while compiling stuff)
and gives better diagnostics

Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
10 years agodevice-port: Introduce pa_device_port_new_data
poljar (Damir Jelić) [Thu, 28 Mar 2013 11:05:11 +0000 (12:05 +0100)]
device-port: Introduce pa_device_port_new_data

Port creation is now slightly different. It is now similar to how
other objects are created (e.g. sinks/sources/cards).

This should become more useful in the future when we move more stuff to
the ports.

Functionally nothing has changed.

10 years ago.gitignore: Update m4/ ignore list
Tanu Kaskinen [Thu, 28 Mar 2013 08:49:27 +0000 (10:49 +0200)]
.gitignore: Update m4/ ignore list

10 years agobuild-sys: Use upstream gettext instead glib one
Javier Jardón [Wed, 27 Mar 2013 14:00:36 +0000 (14:00 +0000)]
build-sys: Use upstream gettext instead glib one

10 years agobluetooth: Improve code and log readability
João Paulo Rechi Vita [Wed, 27 Mar 2013 13:45:33 +0000 (10:45 -0300)]
bluetooth: Improve code and log readability

This commit makes the code cleaner, avoiding unnecessary line breaks. It
also changes the debug message elements order, to make it look more
natural ("path, interface, member" instead of "interface, path,
member").

10 years agobluetooth: Remove unnecessary braces
João Paulo Rechi Vita [Wed, 27 Mar 2013 13:45:32 +0000 (10:45 -0300)]
bluetooth: Remove unnecessary braces

10 years agodevice-port: Make it impossible to have dual-direction ports
Tanu Kaskinen [Fri, 29 Jun 2012 15:04:59 +0000 (18:04 +0300)]
device-port: Make it impossible to have dual-direction ports

10 years agoalsa-mixer: Remove the "name" option from the "General" section of path configuration...
Tanu Kaskinen [Fri, 29 Jun 2012 15:04:58 +0000 (18:04 +0300)]
alsa-mixer: Remove the "name" option from the "General" section of path configuration files

This means that the path names will always correspond to the
path configuration file names, so they will automatically be
unique (in the scope of one card).

10 years agoalsa-mixer: Introduce "description-key" option for paths
Tanu Kaskinen [Fri, 29 Jun 2012 15:04:57 +0000 (18:04 +0300)]
alsa-mixer: Introduce "description-key" option for paths

Previously the path description was looked up based on the
path name only. Since there can be multiple paths that use
the same description, it had to be possible to have multiple
paths with the same name.

Having the same name with multiple paths makes identifying
the paths more complex than necessary, so the plan is to
make it impossible to have paths with the same name. This
patch prepares for that by retaining the possibility to
still have the same description with multiple paths. Instead
of the path name, the path description is looked up by using
the "path description key" if it is set (path name is still
used as a fallback lookup key).

10 years agoswitch-on-port-available: Fix error reporting style
Tanu Kaskinen [Wed, 27 Mar 2013 14:06:04 +0000 (16:06 +0200)]
switch-on-port-available: Fix error reporting style

10 years agoswitch-on-port-available: Remove a redundant function argument
Tanu Kaskinen [Wed, 27 Mar 2013 14:05:26 +0000 (16:05 +0200)]
switch-on-port-available: Remove a redundant function argument

10 years agoswitch-on-port-available: Prepare for dual-direction ports going away
Tanu Kaskinen [Fri, 29 Jun 2012 15:04:56 +0000 (18:04 +0300)]
switch-on-port-available: Prepare for dual-direction ports going away

As an extra, I broke try_to_switch_profile() into smaller
functions, because the two levels of loops with continue
statements inside both were a bit hard to follow.

10 years agobuild-sys: Bump soname v4.0
Arun Raghavan [Mon, 3 Jun 2013 18:43:42 +0000 (00:13 +0530)]
build-sys: Bump soname

10 years agoAdd a NEWS file
Arun Raghavan [Mon, 3 Jun 2013 18:38:14 +0000 (00:08 +0530)]
Add a NEWS file

Short version based on Tanu's longer notes posted on list (and now on
the wiki, linked from this file.

10 years agoosx: Fix latency querying
Vadim Troshchinskiy [Wed, 29 May 2013 10:58:36 +0000 (13:58 +0300)]
osx: Fix latency querying

get_latency_us() used an uninitialized variable, and an incorrect
scope for some of the AudioObjectGetPropertyData() calls. As a result,
audio would randomly not work at all.

BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=65122
10 years agobuild-sys: Bump soname v3.99.2
Arun Raghavan [Thu, 23 May 2013 03:04:17 +0000 (08:34 +0530)]
build-sys: Bump soname

We dropped pa_format_info_free2 as an exposed symbol, but this was never
exposed via headers, so not counting this as ABI change.

10 years agortp: Reset timestamp with correct size if not found
David Henningsson [Wed, 22 May 2013 11:26:24 +0000 (13:26 +0200)]
rtp: Reset timestamp with correct size if not found

This fixes a compiler warning when building under Ubuntu.

Signed-off-by: David Henningsson <david.henningsson@canonical.com>
10 years agomodule-device-manager: Fix description restore
David Henningsson [Wed, 22 May 2013 11:21:24 +0000 (13:21 +0200)]
module-device-manager: Fix description restore

e->description is a pointer, not a fixed char array. Hence it
makes no sense to use strncmp.

This fixes a compiler warning when compiling under Ubuntu.

Signed-off-by: David Henningsson <david.henningsson@canonical.com>
10 years agocore-util: Add pa_safe_streq
David Henningsson [Wed, 22 May 2013 12:08:19 +0000 (14:08 +0200)]
core-util: Add pa_safe_streq

Like pa_streq, but does not blow up on NULL pointers.

Signed-off-by: David Henningsson <david.henningsson@canonical.com>
10 years agosink-input,source-output: Deal with FIX* flags and extended API
Arun Raghavan [Mon, 13 May 2013 16:03:22 +0000 (21:33 +0530)]
sink-input,source-output: Deal with FIX* flags and extended API

The sample spec fixup when FIX* flags are set was not being propagated
to the pa_format_info, causing the two to be out of sync when FIX* was
used.

10 years agobluetooth: Fix missing port->profile association
Mikel Astiz [Mon, 20 May 2013 09:48:33 +0000 (11:48 +0200)]
bluetooth: Fix missing port->profile association

Commit 17b3cb954b179392e80b0a46d8f2ba4693aec386 merged Bluetooth ports
into two ports (one for input, one for output) but the association
between ports and profiles was lost.

BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=64713