The current implementation rejects all input with bytes > 0x7E, which includes
all multibyte UTF-8 sequences. According to ECMA-404, Section 9, only double
quotation marks, backslashes, and characters 0x00 - 0x1F must be escaped
in JSON strings, so non-ascii bytes can just be passed without escaping.
This also mirrors what the decoder does above.
Of course this allows invalid UTF-8 characters to be encoded. Checks for this
could be added as well, but at least the decoder does not seem to do that.
And from what I can tell from a quick glance, the text output path does not
check that either.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/1310
Previously, using the -f json or --format=json flags did not return JSON for the following commands:
- get-sink-volume
- get-source-volume
- get-sink-mute
- get-source-mute
This change adds proper JSON output for these commands.
The CI for merge requests is failing because the meson script is unable
to parse the version. With some print debugging I determined that the
version string being generated on the CI is empty because it has no
git tags. I've added a command to fetch the tags before the build.
This issue was found by enabling ubsan. For me it consistently triggered
after about 28 seconds running a simple example that plays a sine wave
via the mainloop api.
I added a log and confirmed that before the ubsan is triggered the
index variable j is indeed 32 which is out-of-bounds.
Co-authored-by: Arun Raghavan <arun@asymptotic.io>
pa_module_load API's return value is integer which is
enum pa_error_code_t with minus such as -PA_ERR_IO
if the module loading is failed.
pa_cli_command_load gets a return value of pa_module_load
as pa_error_code_t which is wrong.
Minus integer value could not covert to enum which is defined
equal or larger than 0 so that pa_cli_command_load would
recognize the return value as larger than 0 if pa_module_load
return value (integer) is minus.
To fix this issue, I modified return value check logic
of pa_module_load API.
As same as pa_module_load's return type, integer would be used
to check if module load is failed in pa_cli_command_load
and the return value would be compared with minus.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/3801
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/814>
My log files get completely clobbered by this; thousands of lines of:
Jan 18 18:14:44 pulseaudio[29946]: [pulseaudio] backend-native.c: Dock Status: undocked
Jan 18 18:15:39 pulseaudio[29946]: [pulseaudio] backend-native.c: Battery Level: 50%
Jan 18 18:15:39 pulseaudio[29946]: [pulseaudio] backend-native.c: Dock Status: undocked
Jan 18 18:16:34 pulseaudio[29946]: [pulseaudio] backend-native.c: Battery Level: 50%
Jan 18 18:16:34 pulseaudio[29946]: [pulseaudio] backend-native.c: Dock Status: undocked
Jan 18 18:17:29 pulseaudio[29946]: [pulseaudio] backend-native.c: Battery Level: 50%
Jan 18 18:17:29 pulseaudio[29946]: [pulseaudio] backend-native.c: Dock Status: undocked
Jan 18 18:18:25 pulseaudio[29946]: [pulseaudio] backend-native.c: Battery Level: 50%
Jan 18 18:18:25 pulseaudio[29946]: [pulseaudio] backend-native.c: Dock Status: undocked
Jan 18 18:19:20 pulseaudio[29946]: [pulseaudio] backend-native.c: Battery Level: 50%
Jan 18 18:19:20 pulseaudio[29946]: [pulseaudio] backend-native.c: Dock Status: undocked
Jan 18 18:20:15 pulseaudio[29946]: [pulseaudio] backend-native.c: Battery Level: 50%
This seems like it should be a debug log, not a notice.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/810>
The pa_alsa_ucm_set_port() function is passed both a mapping context and
a device port, and both of these refer to their respective UCM device.
While switching over to having one port per mapping per UCM device, I
expected both of these to be the same device struct, so added an assert
checking so.
This assertion gets triggered when we have multiple UCM verbs declaring
the same UCM device name. The root cause here is that the ports' UCM
device references are set once while creating the ports for the card, so
they happen to be those of a specific verb and may not match those from
a different UCM verb's profiles' mappings.
Solving the root cause necessitates a larger refactor. What we actually
assume here is that name of the UCM device is same for both the port and
the UCM context, which ends up always true in practice. For now, replace
the assert with a check and error.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/802>
Some versions of the ALSA libraries run into a segmentation fault when
we query a UCM device/modifier status without first setting a UCM verb.
It's not a reasonable thing to do anyway, so check for this case and
return an error. Also do the check in other helpers.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/801>
When we connect Lenovo XT99 bt headset in the Ubuntu 22.04, this
headset could only work in A2DP profile, couldn't work in HFP profile
with a high chance.
This headset supports mSBC, after pulseaudio replies "+BCS:2" to
headset, we expect to receive a "AT+BCS=2\r" from the headset, but
with a high chance, it will receive 2 AT commands in a buffer like
this "AT+CHLD=?\rAT+BCS=2\r", and we also observed other 2 AT commands
in a buffer like this "AT+NREC=0\rAT+CGMI?\r".
Here we don't suppose there is only one AT command in a buffer, we
will find each command by the delimiter "\r" and handle each command
by sequence.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/804>
When `pwd.h` header is not available (i.e. not using glibc) and environment
variables are not set (e.g. running via `env --ignore-environment`) client
library would crash due to uninitialized variable in `pa_get_home_dir()`.
Add missing initialization to fix that.
Fixes: #3792
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/800>
If pa_memblockq_push needs to write into the middle of a chunk, target chunk
is split into head and tail sharing the same memblock. Size of head and
tail chunks is adjusted correctly, head chunk pointer into memblock remains
unchanged from target chunk.
The problem is with tail chunk offset into memblock which should be advanced
past write region of memblock, but currently it is left as 0.
This is causing an issue where seeking a few frames back into the middle of
memblock and writing a frame there ends up with tail chunk referencing frames
from very beginning of memblock causing corrupted output from memblockq.
Fix this by adjusting tail chunk offset into memblock past write region and
add a test case.
Fixes#3789
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/798>
ref: https://lore.kernel.org/lkml/20221207154939.2532830-4-jeffxu@google.com/
The new MFD_NOEXEC_SEAL and MFD_EXEC flags allows application to
set executable bit at creation time (memfd_create).
When MFD_NOEXEC_SEAL is set, memfd is created without executable bit
(mode:0666), and sealed with F_SEAL_EXEC, so it can't be chmod to
be executable (mode: 0777) after creation.
when MFD_EXEC flag is set, memfd is created with executable bit
(mode:0777), this is the same as the old behavior of memfd_create.
Signed-off-by: Rudi Heitbaum <rudi@heitbaum.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/792>
From https://bugs.debian.org/1006631:
> dbus supports policy files in both `/usr/share/dbus-1/system.d` and
> `/etc/dbus-1/systemd`. [The] recently released dbus 1.14.0, officially
> deprecates installing packages' default policies into `/etc/dbus-1/systemd`,
> instead reserving it for the sysadmin. This is the same idea as the
> difference between `/usr/lib/udev/rules.d` and `/etc/udev/rules.d`.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/783>
Currently there is no way to unset the default sink or source once it was
configured manually by the user.
This patch introduces the special name @NONE@, which can be used with the pacmd
or pactl set-default-sink and set-default-source commands to unset the user
configured default. When the default is unset, pulseaudio will return to the
standard default sink or source selection mechanism based on priority.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/785>
The code that removes the mixer path if probing fails can be called in
the path that sets a non-off device profile on hotplug *before*
card->active_profile is updated, which results in spuriously removing
the mixer path. By this point, context->ucm->active_verb would be set
to the same as the profile name, so we can use that instead to avoid
the issue.
On Apple Silicon machines with the UCM profiles in the Asahi Linux repo,
this manifests as the headphones jack having hardware volume controls
*only* if PA is started with headphones connected and until they are
disconnected. Hotplugs end up triggering the bad codepath, and it falls
back to software volume (which is particularly a problem when the
hardware volume happens to be very low or 0 at that point).
Fixes: a9cc1373e2 ("alsa: ucm - update the mixer path also after volume probe")
Signed-off-by: Hector Martin <marcan@marcan.st>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/752>
The ucm_get_device_property() function adds to each UCM device's
playback_volumes (or capture_volumes) hash map an associated volume
mixer keyed with the UCM verb. These key-value pairs are then iterated
over in various places which assume the key is a profile name. This
assumption is no longer true since we can generate multiple profiles to
use conflicting devices.
A previous commit 4527890416 ("alsa-ucm: Stop conflating profile name
with UCM verb name") fixes some instances of this assumption, but misses
the relation explained above. Fix more instances of misleading
"profile"s where the UCM verb name is actually meant.
Fixes: 4527890416 ("alsa-ucm: Stop conflating profile name with UCM verb name")
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/771>
Although it's a valid UCM configuration to have multiple devices using
the same PlaybackPCM or CapturePCM, it's unclear how PulseAudio should
handle the cases where multiple of these devices are enabled. Some
options I can think of are:
- Merge all devices sharing the same PCM into the same mapping, open
only one PCM substream for this mapping, and add 'combination ports'
that enable combinations of the devices. This has been the case until
recently, although the combination port logic was broken. A problem
with this is that we can't independently control device volumes. We
most likely cannot use hardware volumes either.
- Have one mapping for each device in the same profile, and open one PCM
substream for each mapping. This is the current state, and it fails
when there are fewer substreams than devices. Otherwise it works, but
it's still confusing, as sound directed to a device-specific mapping
might end up playing at multiple devices.
- Make multiple profiles each with combinations of upto-substream-count
devices, and have one mapping/substream per device. This still causes
the confusion mentioned above. And it's likely that the substream
count will almost always be one, where this case degenerates into the
last one.
- Have one mapping for each device in the same profile, but open only
one PCM substream. I assume this is possible with software mixing, but
it is still confusing like the above, and probably less performant.
- Generate multiple profiles each with one of the shared-PCM devices,
again with one mapping/substream for that one device. The trade-off
with this is that we can't use multiple of these devices at the same
time. However, this doesn't have the output device confusion,
combination port's volume problems, or the substream count limitation.
This patch takes a short-cut to achieve the last option, by considering
shared-PCM devices implicitly conflicting with each other.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/743>
While switching profiles of the same UCM verb, existing code first
disables devices that are only on the first profile to avoid conflicts.
However, it only disables devices, not modifiers. Even worse, modifiers
which have PlaybackPCM/CapturePCM are incorrectly treated as devices and
result in a segmentation fault.
Check what we are disabling, and call the appropriate disable function
for both devices and modifiers. Modifiers are disabled before devices,
because _dismod calls fail when the modifier's supported devices are
disabled.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/742>
Currently module-switch-on-connect overwrites the default sink or source that
the user has configured. This means that when the overwritten default sink or
source becomes unavailable, the new default will be chosen based on priority
and the default will not return to the originally configured value.
This patch solves the issue by introducing new core variables for the sink
or source chosen by the policy module which have higher priority than the
user configured defaults.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/784>
This helps to export correct APIs for compiler toolchain which
does not support version script file. For example, mingw clang.
The APIs in libpulse.def are similar with map-file except those
are in pulse-simple and pulse-mainloop-glib. Those are exported
in different shared library in Windows platform.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/761>
There is no requirement for chunk index to be aligned, we only need chunk length
to be multiple of sample frame size.
Fixes: 6434853b0 ("memblockq: Do not allow non-frame indices in the memblock queue")
Fixes: 22827a5e1 ("protocol-native: Fail if trying to push unaligned memblock into queue")
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/779>
`pa_pstream_send_memblock()` would split incoming memblock into parts not
exceeding maximum pool block size.
To make sure split parts of memblock are still frame-aligned add new `align` arg
to `pa_pstream_send_memblock`, find out required alignment from stream sample
format and pass it there. Bump default alignment to 256 which is good up to
32bit 64ch frames.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/780>
Looks like sbc_decode() would seldom access more than specified input length
bytes from input buffer if input length is less than expected frame size.
Fix potential access past allocated memory by checking if input contains
complete frame before calling sbc_decode()
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/781>
The assumption that the format enum is ordered by size is not valid for quite
some time, since 24bit formats were appended to format enum later than 32bit
formats. This causes resampler to produce properly aligned memblock of size
larger than maximum mempool block size if input format is 24bit and output
format is 32bit.
Oversized block is getting split by `pa_pstream_send_memblock()` into parts of
size not exceeding maximum mempool block size. This usually works well but for
32ch 32bit 48000Hz stream the frame alignment is 128 bytes and maximum mempool
block size value is multiple of 64 but not 128 bytes, therefore resulting parts
are misaligned.
On receiving side this causes extra allocation of 128 byte chunk while `mcalign`
helper reassembles properly aligned frame out of second block of misaligned
size. While first and second properly aligned frames are retrieved successfully
from `mcalign` helper, third retrieved frame would end up with properly aligned
size but misaligned memblock index (in this example, that would be 64 bytes.)
Attempt to push a chunk with misaligned memblock index causes assertion failure
Assertion 'uchunk->index % bq->base == 0' failed at memblockq.c:289,
function pa_memblockq_push(). Aborting.
Fix oversized block issue by checking proper size of format instead of enum
value.
Fixes: a67c21f09 ("merge 'lennart' branch back into trunk.")
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/778>
When tunnel-sink-new was used in combination with module-combine-sink, PA
would hang because the main thread was blocked waiting for the execution
of the latency snapshot message. The message would never be processed
because the rtpoll associated with the control_inq of module-combine-sink
was never run.
This patch fixes the problem by running the rtpoll in the thread function
to process incoming messages. Though there are no users of the rtpoll for
module-tunnel-source-new, the same change is applied there.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/773>
When bluetooth transport has both both sink and source, pulseaudio would
synchronize writing out frames with reading frames from peer to make fair
schedule of reads and writes. Pulseaudio allows two blocks of data to be sent to
peer before synchronizing writes with reads just in case that peer implements
similar write schedule.
It could happen that first blocks are still missed by peer, which would cause
pulseaudio writes to stall waiting for first frames from peer.
Fix this by allowing more data frames out until data from peer is actually
received.
Closes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/1424
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/777>
If the same role is named in trigger_roles and cork_roles, a stream with that
role will crash PA. This patch fixes the crash and re-introduces the old
behavior, so that for example specifying trigger_roles=alarm, phone and
cork_roles=alarm, multimedia means that a phone stream will cork alarm and
multimedia streams while an alarm stream will only cork multimedia streams.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/767>
For module-role-ducking, trigger and ducking groups were introduced some years
ago. This patch extends the functionality to module-role-cork, so that trigger
and cork roles may now contain "/" separated groups.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/767>
Incoming RFCOMM string has extra end-of-command terminating character which
breaks both AT+BIA= and AT+BAC= parsers which only expect a comma.
This leads to error parsing last element of response in both cases and could
prevent detecting mSBC availability if mSBC codec id comes last, e.g. AT+BIA=1,2
Fix this by additionally checking for delimiters in both parsers.
Fixes: 3c63f8e6d ("backend-native: Fix stack corruption reading RFCOMM AT+BIA= response")
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/768>
This fixes the rare case of resume_time being bigger than time_stamp. Which
happens sometimes when a gstreamer client is quickly seeking through a
media file. The resulting integer underflow then causes a huge value in
current_time which will break the playback.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/764>
* Enable macOS specific modules (module-bonjour-publish,
module-coreaudio-detect and module-coreaudio-device)
* Correctly set `PA_SOEXT` (.so, .dylib and .dll)
* Build `poll-posix.c` and `semaphore-osx.c`
* Drop linker flag `-Wl,-z,nodelete` on Darwin
* Drop linker flag `-Wl,--no-undefined` on Darwin
* Prefer to `clock_gettime` over compat impl for old Darwin
* Disable SCM credential on Darwin
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/746>
Rate adjustment timer is set up when combine sink is resumed and relased when
combine sink is suspended. Do not create this timer again while module is loaded
to prevent duplicate effort causing assertion in time_callback.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/756>
Faststream backchannel decoder does not know whether incoming stream is mono or
stereo before first packet is decoded, and some devices return stereo stream.
As it is not easy to change source sample spec after source is created, use
stereo sample spec always and perform conversion if mono stream is found.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/628>
When a stream is started but has not yet called smoother_2_put(), pa_smoother_2_get()
returns the time since the start of the stream even if the stream was started paused.
When the stream is started paused, pa_smoother_2_get() should return 0 instead. This
patch fixes the problem.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/745>
When `indicator` is initialized to `1`:
- it always succeeds the `indicator == CIND_CALL_INDICATOR` check;
- hence always calls `continue`;
- hence never reaches the end of the `while` loop where `indicator++` is
called;
- hence `indicator` never contains any other value than `1` meaning
`cind_enabled_indicators` is ever updated.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/740>
This reverts commit b05e34e092.
Now that backend-native uses a different way to get to its own
`native_backend` instance - without going through
`pa_bluetooth_discovery` - this patch can be reverted again, as nothing
outside bluez5-util is supposed to know the internals of this struct.
That's what the many functions are for which all take pointers to this
(at that point) opaque struct instead.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/740>
This removes the inverse/recursive dependency of backend-native on the
`pa_bluetooth_discovery` struct, which is supposed to be opaque outside
of `bluez5-util` in favour of the many accessor functions defined in
`bluez5-util.h`.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/740>
Fix the following build failure without C++:
../output-1/build/pulseaudio-16.1/meson.build:1:0: ERROR: Unknown compiler(s): [['/home/autobuild/autobuild/instance-1/output-1/per-package/pulseaudio/host/bin/powerpc64-buildroot-linux-gnu-g++']]
The following exception(s) were encountered:
Running "/home/autobuild/autobuild/instance-1/output-1/per-package/pulseaudio/host/bin/powerpc64-buildroot-linux-gnu-g++ --version" gave "[Errno 2] No such file or directory: '/home/autobuild/autobuild/instance-1/output-1/per-package/pulseaudio/host/bin/powerpc64-buildroot-linux-gnu-g++'"
Fixes:
- http://autobuild.buildroot.org/results/6526a21bd4da3b8458188f27c1ec04c381e4b673
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/737>
AT+BIA is used to enable/disable CIND indicators by Bluetooth HFP spec.
By default, all indicators are enabled on connection.
AT+BIA will configure which indicators should be disabled then,
the disabled indicators may be enabled later on again with AT+BIA.
When the connection is lost and recovered, all indicators are enabled
again. The HF will reconfigure the indicators again with an AT+BIA
command.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/631>
Add libatomic_ops dependencies to libraries/modules that showed a
failure on an arch that does not have native atomic operations support.
Not all optional dependencies were tested, so it is possible that
some optional modules are still missing libatomic_ops dependencies.
Signed-off-by: Nicolas Cavallari <nicolas.cavallari@green-communications.fr>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/732>
Attempting to use atomics operations on an architecture that does not
support them generally results in a link error:
ld: /tmp/ccjYcMPP.o: in function `func':
testfile.c:(.text+0x1c): undefined reference to `__sync_bool_compare_and_swap_4'
The current build system uses cc.compiles() to check if atomic ops are
supported, but cc.compiles() does not attempt to link, so the test fails
to enable libatomics_opts.
Fix this by using cc.links() instead of cc.compiles().
Signed-off-by: Nicolas Cavallari <nicolas.cavallari@green-communications.fr>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/732>
The full identifier check must be executed for the new melem
creation, otherwise the duplicate control element code check
is reached.
Example (using the snd-aloop driver):
numid=56,iface=PCM,name='PCM Notify',device=1,subdevice=1
numid=62,iface=PCM,name='PCM Notify',device=1,subdevice=2
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/730>
The new helem must be tracked and old helem must be cleared
to make the code work properly. Introduce the pointer to helem
as the private value for melem and add the necessary code.
Also, add a check for the duplicate mixer elements. The duplicate
mixer element invokes the abort check in alsa-lib. Print a warning
instead and handle the exit gracefully.
Fixes: def8eb074 ("alsa-mixer: allow to re-attach the mixer control element")
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/730>
Right now we try to add all UCM devices of a verb to a single profile.
But if some devices using different PCMs are configured as conflicting
with one another, we will only be able to utilize one of them, chosen
seemingly based on the order in the UCM config file.
This is not a problem with conflicting devices sharing a PCM, as they
are assigned to the same mapping and the ports mechanism only enables
one of them to be active at a time.
To utilize all devices in a UCM verb even when there are conflicting
devices using different PCMs, calculate subsets of devices which
can be simultaneously used and create a profile for each such set.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
While switching profiles, it was enough to switch UCM verbs since that
disables all enabled UCM devices and every profile had a distinct verb.
However, switching to the current verb does not disable any devices.
To support multiple profiles for a verb we need to explicitly disable
the old profile's devices, since they might be conflicting with the new
profile's devices and will prevent them from being enabled. Compare both
profiles' mappings, and disable the devices not in the new mappings.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
After previous patches, we should be generating no combination ports, so
we don't need to store multiple modifiers per mapping. Simplify the code
based on this.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
After previous patches, we should be generating no combination ports, so
we don't need to store multiple devices per mapping. Simplify the code
based on this.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
After previous patches, we should be generating no combination ports, so
we don't need to store multiple devices per port. Simplify the code
based on this.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
A previous commit makes mapping names depend on the UCM device name.
Since UCM device names are unique, this means a mapping will at most
have one port and thus no combination ports can be generated.
This removes the dead code in the pa_alsa_ucm_add_ports_combination()
function, unrolls the remaining code in its helper functions that it
used, and renames it to pa_alsa_ucm_add_port() to signal that it no
longer generates combinations.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
PulseAudio combines UCM devices that have the same PlaybackPCM or
CapturePCM value into a single mapping with multiple ports. It also
creates ports in the same mapping for each valid combination of those
UCM devices.
Since mappings are the things we put in profiles, we can put in a
profile either all devices of a joint mapping or none of them. This
causes some complications with device conflicts. For example, a
different UCM device might be marked as conflicting with some (but not
all) of the devices in a joint mapping. In this case we can do one of
three things:
- Include all devices in one profile, and hope the conflicting device
isn't chosen as the mapping's active port. We shouldn't do this as it
puts conflicting devices in the same profile.
- Make one profile with the joint group, and one with the other device.
This is somewhat acceptable as we have no conflicts, but we sacrifice
some compatible combinations of devices.
- Do not group the devices into the same mapping, and make one profile
for each compatible combination of devices. This appears to be the
best option, one where we can always have the maximum number of
working devices.
This patch chooses the third option and makes one input and/or output
mapping per UCM device, by using UCM device names instead of PCM device
strings in the mapping names.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
Combination port logic calculates some useful properties for device
groups that we could reuse while generating multiple profiles to support
conflicting devices. Split them into their own functions.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
Right now this check is rejecting devices whose UCM config specifies
neither a conflicting device nor a supported device list, and accepting
devices which specify both. However, a device without neither list is
actually unrestricted, and a device with both lists is a configuration
error. Fix the check to accept the former.
Furthermore, this is missing another case where an already selected
device might have a supported devices list that doesn't have the
candidate device. Make this function also check against that, and also
make it accept devices already in the set.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
The existing code meant to generate device groups for combination ports
is tightly coupled to port creation. Similar functionality would be
useful to generate nonconflicting device groups for multiple profiles as
well, so this tries to rewrite it into a more reusable state.
Several things (e.g devices, mapping contexts) use idxsets to store a
device selection. This also switches this conformance check and device
group generation to using idxsets to make it easier to work with those,
with the eventual aim to unify device group representations.
Also try to adjust users of these functions to use idxsets these will
need/return, without causing too much interference.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
To support having multiple profiles per UCM verb, split the profile
creation into two parts based on whether they should run once for each
verb or for each profile (maybe multiple times per verb).
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
The ucm_create_mapping() function is not idempotent. It looks like it
was meant to be called once per device for the devices of a UCM verb
and takes a profile argument simply because a verb has generated a
single profile so far.
Make sure creating mappings per device and adding those mappings to the
profiles happens as separate steps to make it easier to split UCM verbs
and profiles as concepts.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
So far each profile had the exact name as their associated UCM verb,
which caused the one to be used where the other should have been.
Explicitly get and use the verb name where that was intended, and make
sure things about profiles aren't named after verbs.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
Currently each UCM verb generates one profile named the same as the
verb, meaning it's trivial to know which verb the profile belongs to.
This will be slightly harder to do when we generate multiple profiles
per UCM verb (e.g. to make use of conflicting devices).
It would still be possible to parse the profile name to get the UCM
verb, but instead let's keep track of the struct instance representing
the profile's associated verb. This also lets us remove a block of code
searching for the verb by its name.
Co-authored-by: Jaroslav Kysela <perex@perex.cz>
[Alper: Reused Jaroslav's UCM profile context changes for UCM verb
instead of combined devices.]
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
While switching profiles, it's possible that we will want to do more
work besides switching UCM verbs. The alsa-card module already has our
profiles as structs, but passes in only the names instead of the entire
struct. Make things work with the struct instead, so we can add other
things (like a UCM context) to it and use those here.
Co-authored-by: Tanu Kaskinen <tanuk@iki.fi>
[Alper: Split into its own commit and integrated Tanu's snippet.]
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
Right now manipulating device status is done inline once while setting a
port. However, we will need to reuse this code to disable conflicting
devices of a device we want to enable. Split it into enable and disable
helper functions.
There is another issue with the device enable logic, where trying to
disabling an already disabled device sometimes fails. To avoid that,
implement a status helper and check if the device we want to enable is
already enabled/disabled before trying to do so.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
Modifiers currently keep their conflicting and supported devices's
names, and these names are resolved to devices every time we need to use
them. Instead, resolve these device names while creating the modifier
struct and keep track of the resulting device structs in idxsets, same
as how device structs keep track of their support relations.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
This is intended to make the current and upcoming code a bit clearer, as
we won't need to constantly check for the existence of these idxsets
before using or operating on them.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
Add complementary functions to the existing idxset iterate(),
steal_first(), first(), next() functions that work in the reverse
direction: reverse_iterate(), steal_last(), last() and previous().
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
This is functionally equivalent to get_by_data(s, p, NULL) == p, but
with a more obvious name and form because some existing code is instead
manually iterating through idxsets to check for existence of an item.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/596>
It may be possible that the ALSA control element appears
again. Allow this combination by checking, if the pulseaudio
mixer element already exists. Do not create the duplicate
mixer element in this case.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/729>
PulseAudio v5.99 or later hits assertion at alsa-lib mixer API due to
wrong handling of removal event for mixer element.
pulseaudio: mixer.c:149: hctl_elem_event_handler: Assertion `bag_empty(bag)' failed.
The removal event is defined as '~0U', thus it's not distinguished from
the other type of event just by bitwise operator.
At the removal event, class implementator for mixer API should detach
mixer element from hcontrol element in callback handler since alsa-lib
has assertion to check the list of mixer elements for a hcontrol element
is empty or not after calling all of handlers. In detail, please refer to
MR to alsa-lib:
* https://github.com/alsa-project/alsa-lib/pull/244
This commit fixes the above two issues. The issue can be regenerated by
`samples/ctl` Python 3 script of alsa-gobject.
* https://github.com/alsa-project/alsa-gobject/
It adds some user-defined elements into sound card 0. When terminated by
SIGINT signal, it removes the elements. Then PulseAudio dies due to the
assertion.
Fixes: 1fd8848e64 ("alsa-util: Add functions for accessing mixer elements through mixer class")
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/728>
GST_TYPE_BITMASK is 64-bit bit mask while corresponding channel_mask in
pulseaudio is int therefore usually 32-bit. Switch to uint64_t instead
to match internal representation in gstreamer.
Fixes pulseaudio crash on ARM 32-bit when pulseaudio is compiled with
gstreamer and either LDAC or aptX support is available.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/723>
A recent commit added i->origin sink for the sink inputs of the combine sinks.
Therefore pa_sink_process_input_underruns() treated the combine sink like
filter sinks. pa_sink_process_input_underruns() calls itself with the
origin sink, which is only correct for filter sinks because they run in the
thread context of the origin sink. The combine sink however has its own
thread context, so pa_sink_process_input_underruns() was executed in the
wrong context.
This patch fixes the issue by skipping the section for module-combine-sink.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/722>
Default build configuration would fail to run on a system without systemd-logind
(or elogind) and without ConsoleKit daemon responding on dbus interface. Here,
module-console-kit would fail to initialize, preventing daemon from starting.
Make module-console-kit an optional build feature to allow opt-out.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/719>
On FreeBSD (and probably other BSDs as well), the FIONREAD ioctl
on UDP sockets does not return the size of the next datagram (like
it does on Linux), but returns the size of the output buffer: this
count contain multiple datagrams and also contains the headers.
We fixed this by taking the result of the FIONREAD as lower bound
for the size, adding an upper bound and then removing the check
that the sizes should be exactly the same.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/718>
Turned out that pa_sdp_info::enable_opus is never initialized, which seldom
makes module-rtp-recv believe it will be playing OPUS-encoded stream even though
discovered SDP record does not indicate OPUS codec in metadata.
Fix this by adding missing initializer.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/720>
pulseaudio crash occurred when I play a file using pacmd play-file command.
The file is not aligned with its frame size and the last rendering size
is also not aligned. Thus, an assertion was generated at the end of the
file as the following.
memblockq.c: Assertion 'uchunk->length % bq->base == 0' failed at
../src/pulsecore/memblockq.c:288, function pa_memblockq_push(). Aborting.
When I play the file using paplay, it works good. So, I changed to
pa_memblockq_push_align instead of pa_memblockq_push to prevent the
assertion.
Signed-off-by: Jaechul Lee <jcsing.lee@samsung.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/717>
The logic for detecting which type of volume was given incorrectly interpreted
any value with a decimal as a VOL_LINEAR. It also could set multiple flags,
which would put the flags variable into an indeterminate state. Additionally,
the flags stack variable was uninitialized which could also lead to an
indeterminate flag state.
Percentages are now prioritized over all other types, and only one type flag
can be set.
RFC 4566 states that SDP record is terminated with CRLF, and parsers should be
able to accept records terminated with just LF. Pulseaudio only accepts LF here.
Fix this by accepting both CRLF and LF terminators.
The combine sink used the current time and counter when calculating
the latency if smoother_2 was enabled. This lead to wrong latency
reports. This patch fixes the problem by using the snapshot time
and counter instead.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/711>
Bluetooth transport layer already allows for packets larger than mSBC frame, and
there are up to 1 + MTU / (mSBC packet size) complete frames to be decoded from
each incoming SCO packet.
Now decoder fails when there is more than one complete frame available, which
could happen if MTU size is larger than 1.5 * (mSBC packet size) = 90
Fix this by adding a loop over avialable frames, and adjust decoded buffer size
to allow decoding up to 1 + MTU / (mSBC packet size) frames at once.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/706>
When the module is loaded and avahi_client_new() fails because the client cannot
connect, a shutdown of the module is scheduled. In parallel, the client_callback
is called with AVAHI_ERR_DISCONNECTED and another connection attempt is made
which also fails and triggers a second unload of the module. This crashes PA,
because there is already an unload in progress.
This patch fixes the problem by checking if an unload is already scheduled.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/710>
The timestamp used for updating the smoother was taken at the wrong time.
It may take some time until an async message is executed (measured up to
2ms), therefore the timestamp used to update the smoother must be taken
before the message is executed and not inside the message.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/705>
When the tunnel modules had no connection and a re-init was pending, the module
could be unloaded without cancelling the pending re-init. When the timer expired
in that situation, this lead to a crash. This patch fixes the problem by keeping
a reference when the module is scheduled to be re-initialized.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/705>
The AVRCP service is known to not be connected before the A2DP transport
is, resulting in PulseAudio asking BlueZ for an initial 'Volume' value
but not getting it because the property doesn't exist.
To prevent end-users from conjecturing this to be the source of whatever
issue they're observing, demote it to a warning.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/707>
When configured, reinitialize the module instead of exiting. This
allows a restart/reconnect, but the module to appear to always be alive
when the user does: "pactl list modules". (The sink will still not
exist until the tcp connection is established.)
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/688>
The io thread, after connection, sends a message asking for a sink to be
created. After the ctl thread is done with creation, it sends a message
back to the io thread so it can continue. This ensures that the sink
only exists when it's connected to something.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/688>
When the --format json parameter is given on the command line, we
attempt to produce a JSON output for most commands.
Our implementation of the JSON serialization uses vsnprintf to output
numbers. Unfortunately, vsnprintf is affected by the locale and more
specifically the LC_NUMERIC variable.
When LC_NUMERIC is set to, for instance, fr_FR.UTF-8, floating-point
numbers are output with a comma as the decimal separator, which is then
considered invalid JSON.
$ LC_NUMERIC=fr_FR.UTF-8 pactl --format json list sinks | jq .
parse error: Objects must consist of key:value pairs at line 1, column 435
This is the token which failed to parse:
}},"balance":0,00,"base_volume":{
Fixed by overriding the LC_NUMERIC value when we request JSON output.
Signed-off-by: Olivier Gayot <olivier.gayot@sigexec.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/702>
When monitor source becomes idle it may happen that monitored sink has no
uncorked inputs anymore and can now be suspended. To allow this, detect if state
is changed for monitor source and check state of monitored sink instead.
This change allows pulseaudio to suspend devices when pavucontrol volume meters
are disabled and corresponding peaks resampled streams are corked.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/697>
Turned out that SelectConfiguration is only used for outgoing connections, and
incoming connection from bluetooth headset using SBC codec ends up with a
bitpool as large as declared by headset. When resulting bitpool is so large that
SBC frame size plus RTP header size exceeds write MTU size, number of frames per
packet becomes zero causing crash dividing by zero in update_sink_buffer_size()
Fix this by limiting available bitpool value exposed for SBC endpoints.
Fixes: 89082cbfa ("bluetooth: a2dp dual channel SBC XQ codec configurations")
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/695>
Commit c6d6ca541 ("bluetooth/gst: Replace buffer accumulation in adapter
with direct pull") removed the `timestamp` parameter from GStreamer
transcoders due to being unused, but these should instead be propagated
to the GStreamer encoding buffers.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/494>
Bluetooth codecs should always have fixed in/output and are hence able
to have their results directly read from the codec, instead of
accumulating in a buffer asynchronously that is subsequently only read
in the transcode callback. The Bluetooth backends calling encode/decode
also expect these fixed buffer sizes.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/494>
Handling multiple threads does not come without overhead, especially
when the end-goal is to ping-pong them making the whole system run
serially. This patch rips out all that thread handling and instead
"chains" buffers to be encoded/decoded directly into the pipeline,
making them execute their work on the current thread. The resulting
buffer can be pulled out from appsink immediately without require extra
locking and signalling. While the overhead on modern systems is found
to be negligible or unnoticable, code complexity of such locking and
signalling systems is prevalent making it the main drive behind this
refactor.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/494>
Drop rtpldacpay and payload the LDAC encoded output manually in the
RTP header.
The RTP payload seems to be required as it carries the frame count
information. Right now, rtpldacpay does not add this so construct
the RTP header and payload manually.
Strangely some devices like Shanling MP4 and Sony XM3 would still
work without this while some like the Sony XM4 does not.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/689>
If UCM defines the private alsa-lib configuration, the ELD controls
are expected to use this device configuration too.
With this change:
I: [pulseaudio] alsa-util.c: Successfully attached to mixer '_ucm0009.hw:Loopback'
Without:
I: [pulseaudio] alsa-util.c: Successfully attached to mixer '_ucm0009.hw:Loopback'
I: [pulseaudio] alsa-util.c: Successfully attached to mixer 'hw:4'
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/673>
The hw: device can be addressed using the card index (hw:0)
or the card identifier (ASCII string - hw:Loopback). Both
mixers are equal.
The previous code was fine for the mixers without the UCM
private prefixes (_ucmXXXX). Make code more robust, create
two aliased mixer structures in the mixers array.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/673>
This makes it possible to define multiple sinks/sources on detection
of the jack server. This allows one to for example create a separate
sink for conferencing software and route that in jack to another
channel on their audio interface.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/669>
Even though the file name is currently behringer-umc22.conf, the USB ID
actually belongs to Texas Instruments PCM2902, which is a generic chip
used in multiple products. Some products have true mono input unlike
Behringer UMC22, which has two mono inputs combined into one stereo PCM
device.
This patch removes the "mono,mono" mapping from Behringer UMC22, which
hopefully won't be missed too much (there are still "mono,aux1" and
"aux1,mono" mappings available for mono recording).
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/667>
If the preferred ports are not set in this function, the
entrys_equal() always returns false in the card_put_hook_callback().
This will make the entry be written into the metadata and the
preferred ports will be cleaned by a mistake.
And we met a hdmi audio bug which has sth to do with this issue, on
the machines with the legacy HDA audio driver, the hdmi port has lower
priority than speaker, users need to manually select the hdmi to be
active output port, then the preferred output port is hdmi for this
sound card, after reboot, the card_put_hook_callback() in the
module-card-restore.c will be called and the preferred ports are
cleaned by a mistake, then the hdmi output port or hdmi sink couldn't
switch to be active after reboot or resume automatically. That is
because the preferred ports are cleaned and hdmi port has lower
priority than speaker, the profile_good_for_output() in the
module-switch-on-port-available.c always returns false.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Change d7f95170a1 added a dependency on device
adapter pointer being valid while checking if bluetooth profile is supported by
device.
When adapter object is released, each device holding pointer to adapter being
released is notified to reset that to NULL. Since adapter objects are released
first when discovery object is unreferenced, each device will have adapter
pointer reset before the time device objects are released.
Fix observed crash by examining device adapter pointer. If it is NULL report
that device does not support any bluetooth profile instead of looking at UUIDs
supported by adapter.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/646>
Add a log_interval parameter to control the amount of logging. Default is
no logging. Like for adjust_time, the parameter is a double to allow values
below 1s.
If the log interval is too small, logging will occur on every iteration.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/56>
The adjust_time parameter is changed to double to allow better granularity
and adjust times below 1s. This may be useful for a better latency control,
although with alsa devices and the current smoother code no significant
improvement could be found for values below 500ms.
This patch also changes the default adjust time to 1s, the old value of 10s
does not allow a tight control of the end to end latency and would lead to
unnecessary jitter.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/56>
The previous patch slows down initial convergence. Therefore do not use
the controller weight until we can assume that we reached an equilibrium.
Because it takes some time before the reported latency values are reliable,
assume that a steady state is reached when the target latency has been
crossed twice.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/56>
In many situations, the P-controller is too sensitive and therefore exhibits rate hunting.
To avoid rate hunting, the sensibility of the controller is set by the new parameter
adjust_threshold_usec. The parameter value is the deviation from the target latency in usec
which is needed to produce a 1 Hz deviation from the optimum sample rate.
The default is set to 250 usec, which should be sufficient in most cases. If the accuracy
of the latency reports is bad and rate hunting is observed, the parameter must be increased,
while it can be lowered to achieve less latency jitter if the latency reports are accurate.
More details at
https://www.freedesktop.org/software/pulseaudio/misc/rate_estimator.odt
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/56>
The current loopback controller can produce a rate jump of up to 1% at startup,
which may be audible. To prevent large initial jumps, a second controller is
introduced, which produces a rate, that is not more than 2‰ away from the last
rate. Only during the startup phase, the rates produced by this controller will
be nearer to the base rate than those produced by the original controller.
Therefore choosing the rate which is nearer to the base rate will ensure that
the secondary controller only moderates the startup phase and has no influence
during continued operation.
The maximum step size of the original controller after the initial jump is
limited to 2.01‰ of the base rate, see documentation at
https://www.freedesktop.org/software/pulseaudio/misc/rate_estimator.odt
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/56>
Currently module-loopback detects underruns even if sink_input_pop_cb()
was not yet called twice and initial latency adjustments are active.
This leads to unnecessary rewind requests.
This patch delays detecting underruns until the initial adjustments
are done.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/56>
Taking addresses of fields in a packed struct are not guaranteed to be
aligned, resulting in warnings such as:
../src/pulsecore/shm.c: In function 'sharedmem_create':
../src/pulsecore/shm.c:198:25: error: taking address of packed member of 'struct shm_marker' may result in an unaligned pointer value [-Werror=address-of-packed-member]
198 | pa_atomic_store(&marker->pid, (int) getpid());
| ^~~~~~~~~~~~
The struct already has its fields and types laid out in such a way that
the desired packing (without padding) is guaranteed - enforce this with
a `static_assert` to get rid of the unaligned pointer warning.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/653>
GCC warns on all these `fail_unless` calls:
warning: too many arguments for format [-Wformat-extra-args]
`fail_unless` only takes an expression and optionally a string literal
as message with formatting args. Passing NULL for this message should
not be necessary as indicated by all the other tests not passing it
either.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/653>
e04f14eb/dc9dc70f introduced preprocessor warnings to deny the use of
any `alsa-lib` older than `1.2.1`, and with a future patch disallowing
warnings entirely through `-Werror` we now need a distribution that
serves a new enough `alsa-lib`.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/653>
This patch adds an alternative time smoother implementation based on the theory
found at https://www.freedesktop.org/software/pulseaudio/misc/rate_estimator.odt.
The functions were written to replace the current smoother functions nearly on
a one-to-one basis, though there are a few differences:
- The smoother_2_put() function takes a byte count instead of a sound card
time as argument. This was changed because in most places a sample count
was converted to a time before passing it to the smoother.
- The smoother needs to know sample rate and frame size to convert byte
counts to time.
- A smoother_2_get_delay() function was added to directly retrieve the stream
delay from the smoother.
- A hack for USB devices was added which works around an issue in the alsa
latency reports for USB devices.
The smoother delivers much better precision than the current implementation.
For results, see the document referenced above.
The new functions are still unused. The following patches will convert all
callers of the smoother functions so that they can use both smoother
implementations, depending on a configure option.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/55>
This patch adds a test program that generates a square wave of a given frequency,
length and sample rate. This is then resampled to another rate, rewound and the
rewound part is run through the resampler again. After that, the results of the
first and second resampler pass are compared.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/120>
This patch is in preparation of allowing virtual sinks to specify their own
max_rewind limit.
Currently pa_sink_set_max_rewind_within_thread() simply sets the value of
max_rewind and informs the sink inputs about the new value. Virtual sinks
may however provide their own limit on max_rewind.
This patch allows to query the active sink inputs for the max_rewind value
they support and sets max_rewind to the minimum supported value. This way,
the max_rewind value from the virtual sinks can be communicated to the master
sink.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/120>
This patch is in preparation of allowing virtual sinks to specify their own
max_rewind limit.
Currently, virtual sinks cannot specify their max_rewind limit, but just copy
the value from the master sink. This may not be correct, if the DSP code of the
virtual sink has limited (or no) rewinding capability.
Because the DSP code of the virtual sink is rewound in the process_rewind()
callback of the sink input, it must be ensured, that rewinding a sink input
to the master of a virtual sink is limited similar to rewinding a sink.
There are two remaining exceptions:
1) If an underrun is detected. In that case, the filter should be reset anyway.
2) When the sink input of the filter is moved and attached to the destination
sink.
The move case is handled without involvement of the implementer, so the implementer
can only receive a rewind larger than max_rewind when the filter should be reset
anyway.
All existing virtual sinks do not distinguish between reset and rewind of the
filter.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/120>
If the output implements a process_rewind() callback, the resampler delay is
not taken into account. This leads to glitches during volume changes when
source and source output rates differ.
This patch fixes the problem.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/120>
The introduction of the history queue makes it possible to implement moving
of streams without involving the implementer. Instead of dropping all data
from the render memblockq and requesting the implementer to rewrite the
data, the render memblockq is now reconstructed from the history queue.
Additionally, the render queue will be filled with silence matching the
amount of audio that is left playing on the old sink to avoid playing
the same audio twice.
This patch slightly breaks moving for virtual sinks because they do not
yet include the resampler delay in their latency reports. This will be
fixed in a different patch set.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/120>
This patch uses the two previous patches to implemnt pseudo-rewinding for the
resamplers by feeding some old data into the resampler after a reset. This is
necessary because PA is using external resamplers that do not implement
rewinding.
To get exactly the same output data from the resampler after a rewind if possible,
the matching period is calculated. This is the number of input samples that produces
an integral number of output samples. After the matching period, the resampler state
repeats. If the matching period is not too large, feeding history into the resampler
will start at a point that is a multiple of the matching period back in time. Then
the resampler will produce exactly the same samples.
The PA_RESAMPLER_MAX_HISTORY value has been replaced by PA_RESAMPLER_MAX_DELAY_USEC
and the required number of history samples is calculated from the sink input sample
rate. The number of history samples can be as large as about 12500.
This fixes glitches during volume changes when the sink runs on a rate different
from the sink input rate.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/120>
The pa_resampler_get_delay() function allows to retrieve the current resampler
delay in input samples for all supported resamplers. The return value is a double
to maintain precision when using variable rate resamplers. Because in many places
the delay is needed in usec, pa_resampler_get_delay_usec() was also supplied.
The speex resampler now skips leading zero samples to provide meaningful delay values.
In the next patch, the pa_resampler_prepare() function will be used to train the
resampler after a rewind. It takes data from a history memblockq and runs it through
the resampler. The output data is discarded.
To make this logic possible, the soxr resampler had to be converted to use variable
rate. The fixed rate version has a variable delay, therefore the logic above could
not be applied. Additionally, with fixed rate, the delay is larger than 150ms in
some situations, while with variable rate the delay is fixed and comparable to the
other resamplers.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/120>
A new memblockq is added to the sink input code to keep some history of the
input data. The queue is kept in sync with the render memblockq. The old input
data will be used to prepare the resampler after a rewind.
pa_resampler_request() and pa_resampler_result() have been changed to round
as good as possible to avoid loosing or duplicating samples during rewinds.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/120>
When sink is suspended for reconfiguration changing sample spec, upon resume
internal thread_info max_request and max_rewind are out of date and possibly
not aligned to frame size anymore.
Recalculate thread max_request and max_rewind before resuming sink.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/658>
When ICE I/O error occurs ICE connection is closed via IceCloseConnection.
This causes crash while releasing session manager connection later because
this ICE connection was initiated and is managed by session manager, and it will
attempt to close this ICE connection again.
Fix this by closing session manager connection instead.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/650>
Commit 7fd89e491 ("bluetooth: Try to reconnect SCO") introduces a call
to pa_msleep but failed to include the header, resulting in a:
../pulseaudio/src/modules/bluetooth/backend-native.c: In function ‘sco_acquire_cb’:
../pulseaudio/src/modules/bluetooth/backend-native.c:336:17: warning: implicit declaration of function ‘pa_msleep’ [-Wimplicit-function-declaration]
336 | pa_msleep(300);
| ^~~~~~~~~
(Un)fortunately this implicit declaration gets resolved at link-time,
otherwise the issue would have been caught sooner.
Fixes: 7fd89e491 ("bluetooth: Try to reconnect SCO")
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/651>
Just like the manpage these written out log levels should correspond to
the numerical values listed before, intead of being in the opposite
order and provoking thoughts of the relation being the wrong way around
where 0=debug and 4=error.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/647>
Requiring user to invoke send-message with correctly quoted parameters string
is not good for usability. Wrap parameters string into JSON string and try
again. If that works, log a warning use wrapped JSON string with parameters.
As an example these two commands will now invoke the same action:
pactl send-message /card/bluez_card... switch-codec '"CODECNAME"'
pactl send-message /card/bluez_card... switch-codec CODECNAME
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/648>
Check whether a Bluetooth profile is supported both by the remote device
and the local host before creating a card profile for it.
This is useful when some of the media profiles have not been registered
with bluetoothd because ex., oFono is not running and the headset
backend is not available.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/638>
The remote device list of UUIDs reflects which profiles are supported by
the remote device alone. We currently rely solely on this list to decide
if a certain card profile is supported, and thus should be created and
get connected.
This used to be accurate when the Bluetooth modules were first written,
but now BlueZ is more dynamic and local profile support can be added or
removed during runtime. The adapter's list of UUIDs is an accurate
representation of the profiles supported by the local host at a certain
moment.
This commit combines the list of UUIDs supported by remote device and
the list of UUIDs supported by the local host to determined whether a
Bluetooth profile is actually supported or not, and whether it should be
expected to get connected during device connection.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/638>
When the SCO connection is in use, if you disconnect first and then connect,
the SCO connection will occasionally fail, and the Bluetooth error code is 42
(0x2A in hexadecimal). This is usually because an error occurred when the SCO
connection was initiated, we need to try to reconnect to optimize the handling
of this problem. The log returned by the kernel is as follows:
Bluetooth: sco_connect_cfm: hcon 0000000003328902 bdaddr 40:ef:4c:0c:11:f0 status 42
Bluetooth: sco_sock_connect status is -38
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/622>
When module-combine-sink is used to create virtual surround card it is expected
that each slave channel receives data for exactly one combined sink channel.
Currently this is not the case because module-combine-sink would follow
core->disable_remixing setting. Usually this means that multiple channels of
combined sink are getting remixed into slave channel, and there is no option to
disable this behavior just for combined sink.
Improve this by implementing "remix" modarg for module-combine-sink,
default to original behavior (follow core->disable_remixing setting).
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/627>
Mainloop test uses io callback for PA_IO_EVENT_INPUT on stdin.
With glib enabled PA_IO_EVENT_INPUT translates to glib G_IO_IN event which also
matches descriptor in EOF state. While io callback does not check for EOF after
reading from file descriptor this is causing mainloop-test to repeatedly read 0
bytes once EOF is reached, rearm defer callback and spam test log.
Fix this by disarming io callback when EOF is reached in test run.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/625>
The word variables used in the _pactl and _pacmd functions are
unlocalized. Thus, the variable appears in the user's environment when
tab-completing with pactl or pacmd. This may clobber another variable
of the same name, which is undesirable.
Localize the word variable to fix this issue.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/623>
When packaging a new version for OpenEmbedded, I use the
buildhistory-diff tool to check what changed between the versions. The
version number in the module directory means that I get tons of diff
output due to changes in file paths. There are many removed and added
files and it's hard to see if something else than just the version
number changed.
That motivated me to write this patch. Removing the version number has
the downside that it makes it easier to have version mismatches between
the daemon and the modules, but
https://gitlab.freedesktop.org/pulseaudio/pulseaudio/merge_requests/249
will make the handling of such situations better.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/230>
Currently, module-tunnel uses the default fixed latency of 250ms as fixed
latency.
There is no reason for such a large latency. This patch adds a parameter
latency_msec to the module to set the fixed latency at load time of the
module. The parameter can range from 5 to 500 milliseconds. With this
patch, I was able to run a tunnel sink at 7ms latency without problems.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/53>
Currently module-tunnel uses only a rough estimate of the current stream
latency and reports wrong latencies in certain situations. This leads to
very inexact and unstable latency reports for the virtual sink.
This patch fixes the issue by introducing latency snapshots like they
are used in module-loopback. Because the latency reports are now correct,
the update interval for latency re-calculations can be reduced to 1s.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/53>
Currently the combine-sink uses the trivial resampler by default.
This patch changes the default to the configured resampler.
Also the default update time is changed from 10s to 1s to achieve
faster convergence and higher precision.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/53>
Currently, it takes one adjust time before the smoother is updated after an
unsuspend. Before the first update, the smoother will not be aware of the
slave sink latencies, leading to incorrect latency reports.
This patch moves the first smoother update to one latency time after the
sink was unsuspended, thereby improving initial latency reports. This
only partially resolves the problem because the smoother takes multiple
updates to adapt to the slave sink latencies.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/53>
This patch adds a rate controller similar to the one used in module-loopback
to limit step size and maximum deviation from the base rate. Rate changes
are handled more smoothly by the controller. The patch has not much impact
on the behavior of the module, except that there is less rate hunting.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/53>
Currently module-combine-sink uses only a rough estimate of the current
slave sink latencies to calculate the rate for the various sink inputs.
This leads to very inexact and unstable latency reports for the virtual
sink.
This patch fixes the issue by introducing latency snapshots like they
are used in module-loopback. It also changes the definition of the
target latency to ensure that there is always one sink which uses the
base rate.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/53>
Commit f89d64b98e fixed a crash
when disabling adapters. However, now if any device is removed
ofono card is removed, even if it belongs to different device.
Add a check for the device being unlinked to our callback to fix.
Signed-off-by: Juho Hämäläinen <juho.hamalainen@jolla.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/624>
With USB Alternate Setting 3 size of HCI payload is 72 bytes which is already
larger than mSBC frame size. Largest known size of HCI payload is with USB
Alternate Setting 5 (144 bytes), make it the default SCO socket MTU.
Reserve additional space in bluetooth encoder buffer to cover this case.
Since mSBC encoder and decoder will now work with larger packet sizes, drop
assertions about MTU larger than mSBC frame size.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/617>
While the threading model for combine is different from other filters
(which expect to just piggy-back on the I/O thread of the most
downstream sink), it might still be valuable to set this field to
indicate that this sink input is intended to behave as a filter stream
rather than a conventional stream.
At the very least, routing behaviour and cycle detection should act on
these streams as with any other filter.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/399>
With clang compiler including cpuid.h will produce error if architecture is not
x86-based, and cheching if cpuid.h exists via Meson has_header() is not enough.
Fix this by creating a list of headers checked to be usable via Meson
check_header() function, and move cpuid.h to that list.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/614>
This reverts commit 3ac73598c6.
Systemd v249 has new entries of hwdb for node and unit in IEEE 1394 bus
(hwdb.d/80-ieee1394-unit-function.hwdb). It can obsolete my workaround
added by commit 3ac73598c6 ("udev: use ID_MODEL/ID_VENDOR to give
friendly name for FireWire devices"). The hwdb entry is handy prepared.
When user finds missing entry, it's preferable to file issue or merge
request in systemd project site.
IEEE 1394 bus is enough legacy and it's easy to expect that few developer
can evaluate the change. For reviewers, I describe the original issues and
the integration of hwdb in systemd side.
In systemd, udev rule for sound card (rules.d/78-sound-card.rules) has
below line to assign information in hwdb to instance for sound card.
```
IMPORT{builtin}="hwdb"
```
In the case, the udev hwdb builtin finds information according to
modalias by following nodes in device topology tree toward root. For
sound card associated to unit in node in IEEE 1394 bus, it's inconvenient
since hwdb had no entry for the unit. The instance for node in IEEE 1394
bus doesn't have modalias. As a result, the builtin reaches 1394 OHCI
controller in PCI Express bus which maintains the IEEE 1394 bus, then the
value for ID_VENDOR_FROM_DATABASE and ID_MODEL_FROM_DATABASE properties
from hwdb of pci device (hwdb.d/20-pci-vendor-model.hwdb) for the sound
card.
For example, when two nodes are in IEEE 1394 bus and one of them has
unit instance for audio and music functions, the topology of the bus is
depicted in following diagram:
```
* 1394 OHCI controller (pci*, modalias)
* node A - (pci*/fw0, /dev/fw0)
* node B - (pci*/fw1, /dev/fw1)
* unit B-1 - (pci*/fw1/fw1.0, modalias)
* sound card 0 - (pci*/fw1/fw1.0/sound/card0, card0)
```
In the case, the udev hwdb builtin picks up from hwdb of pci device for
the sound card:
```
$ udevadm test-builtin hwdb /sys/class/sound/card2
Load module index
Parsed configuration file /usr/lib/systemd/network/99-default.link
Parsed configuration file /usr/lib/systemd/network/73-usb-net-by-mac.link
Created link configuration context.
ID_PCI_CLASS_FROM_DATABASE=Serial bus controller
ID_PCI_SUBCLASS_FROM_DATABASE=FireWire (IEEE 1394)
ID_PCI_INTERFACE_FROM_DATABASE=OHCI
ID_VENDOR_FROM_DATABASE=Texas Instruments
ID_MODEL_FROM_DATABASE=XIO2213A/B/XIO2221 IEEE-1394b OHCI Controller [Cheetah Express]
Unload module index
Unloaded link configuration context.
```
The aim of my workaround is to avoid using ID_VENDOR_FROM_DATABASE and
ID_MODEL_FROM_DATABASE for sound card associated to unit in IEEE 1394
bus. Instead, ID_VENDOR and ID_MODEL properties are used.
However, it has another issue. For the properties, the udev rule for
sound card has the other lines for sound card associated to unit in
IEEE 1394 bus, below:
```
SUBSYSTEMS=="firewire", ATTRS{guid}=="?*", \
ENV{ID_BUS}="firewire", ENV{ID_SERIAL}="$attr{guid}", ENV{ID_SERIAL_SHORT}="$attr{guid}", \
ENV{ID_VENDOR_ID}="$attr{vendor}", ENV{ID_MODEL_ID}="$attr{model}", \
ENV{ID_VENDOR}="$attr{vendor_name}", ENV{ID_MODEL}="$attr{model_name}"
SUBSYSTEMS=="firewire", GOTO="skip_pci"
```
The values of ID_VENDOR and ID_MODEL properties come from vendor_name and
model_name attributes in parent instance of the sound card, therefore
they come from audio and music units in IEEE 1394 bus. Unfortunately
these attributes are not available always.
All of nodes in IEEE 1394 bus should have configuration ROM in place
according to IEEE 1212 and Linux FireWire subsystem parses the content of
ROM to detect units in the node. At the same time, the subsystem manages
to detect information about vendor and model according to standard layout
defined by 1394 Trading Association[1].
When the content of ROM is against the standard, the subsystem is
discouraged the name detection. In the case, vendor_name and model_name
attributes are not available, and supplemental information should be from
software implementation.
The new hwdb (hwdb.d/80-ieee1394-unit-function.hwdb) added to systemd v249
can solve the above issues. The prepared names for vendor and model in
hwdb are assigned to both node and unit. The udev hwdb builtin can find
the vendor and model names for the unit according to modalias before
arriving at pci-device. Regardless of standard or non-standard
configuration ROM, the hwdb gives prepared names of vendor and model.
This is an example of Mark of the Unicorn (MOTU) Traveler. The search
finishes at instance for unit in IEEE 1394 bus expectedly:
```
$ udevadm test-builtin hwdb /sys/class/sound/card2
Load module index
Parsed configuration file /usr/lib/systemd/network/99-default.link
Parsed configuration file /usr/lib/systemd/network/73-usb-net-by-mac.link
Created link configuration context.
ID_MODEL_FROM_DATABASE=Traveler
ID_VENDOR_FROM_DATABASE=MOTU
IEEE1394_UNIT_FUNCTION_AUDIO=1
IEEE1394_UNIT_FUNCTION_MIDI=1
Unload module index
Unloaded link configuration context.
```
[1] Configuration ROM for AV/C Devices 1.0 (Dec. 12, 2000, 1394 Trading
Association, TA Document 1999027)
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/610>
Try to register profile support again after RegisterProfile fails, when
BlueZ indicates no one else is implementing the profiles we are
interested in.
Ideally this would rely on a list of UUIDs supported by the profile
manager instead of the adapter, but BlueZ has no such API.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/593>
Create pa_bluetooth_profile_status_t to represent all stages an external
Bluetooth profile can go through:
0. Inactive: Initial state, no D-Bus object has been registered for
this profile yet.
1. Active: an object implementing the org.bluez.Profile1 interface has
been registered on the system bus.
2. Registering: RegisterProfile has been called.
3. Registered: RegisterProfile succeeded.
This will be useful to handle RegisterProfile failures, as well as
dynamically register and un-register a profile based on the current
active seat.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/593>
HF indicator 2 (see [assigned-numbers], Hands-Free Profile) is able to
report battery percentage at 1% intervals (in range [0, 100]), contrary
to the `+XAPL` `+IPHONEACCEV` extension which only supports 10%
increments. This does not guarantee increased granularity however, as
peers may still be limited to imprecise battery measurements internally
or round to coarser percentages.
Supporting both additionally broadens the range of devices for which PA
can report its battery level.
[assigned-numbers]: https://www.bluetooth.com/specifications/assigned-numbers/
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/482>
Whenever a device disconnects the device is not removed from BlueZ, only
the profiles that had an active connection are disconnected. Since we
were providing this battery level based on AT commands received through
HSP/HFP these services should be responsible for deregistering it again.
Deregister the interface to signal BlueZ (And UPower in return) that the
battery level won't be accurate/updated anymore.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/482>
The peer will wait some time and eventually time out the connection if
no reply is sent back. When sending `ERROR` the peer can decide to break
the RFCOMM connection immediately or continue when a command is not
critical.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/482>
Devices for Apple's iOS uses a few extra HFP AT commands to
inform the iPhone about the headphone's battery status.
Apple documented the AT commands in the following document:
https://developer.apple.com/hardwaredrivers/BluetoothDesignGuidelines.pdf
The patch has been tested with a Bose QC35, which results
in the following communication:
D: [pulseaudio] backend-native.c: RFCOMM << AT+VGS=14
D: [pulseaudio] backend-native.c: RFCOMM >> OK
D: [pulseaudio] backend-native.c: RFCOMM << AT+XAPL=009E-400C-0129,3
D: [pulseaudio] backend-native.c: RFCOMM >> +XAPL=iPhone,2
D: [pulseaudio] backend-native.c: RFCOMM >> OK
D: [pulseaudio] backend-native.c: RFCOMM << AT+XEVENT=Bose SoundLink,158
D: [pulseaudio] backend-native.c: RFCOMM >> OK
D: [pulseaudio] backend-native.c: RFCOMM << AT+IPHONEACCEV=2,1,4,2,0
N: [pulseaudio] backend-native.c: Battery Level: 50%
N: [pulseaudio] backend-native.c: Dock Status: undocked
D: [pulseaudio] backend-native.c: RFCOMM >> OK
[Marijn: Adapt for recent HSP/HFP code changes]
Co-authored-by: Marijn Suijten <marijns95@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/482>
Alsa UCM device string can contain private configuration prefix required to make
correct device open call. Private prefix is dynamically generated by UCM manager
depending on internal state. Since pulseaudio sink/source port names currently
contain device string, these may change between runs breaking volume database
and module arguments referring to sink/source.
Fix this by skipping UCM private prefix available via `_alibpref` key while
creating UCM mapping name. Mapping object will still contain unmodified
device string for device open call.
See also https://github.com/alsa-project/alsa-ucm-conf/issues/104
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/598>
When child `gsettings-helper` terminates prematurely, unconditionally reading
from child pipe fails in a busy loop until child process is reaped.
Fix this by terminating module upon PA_IO_EVENT_HANGUP or PA_IO_EVENT_ERROR.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/600>
These two log messages are most likely intended for the path that was
just tried, but they are mistakenly printing the name of the port's
current path. Fix them.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/594>
Recently we found an issue of output volume on speaker and headphone,
they should have their own volume but in practice they share one
output volume.
This issue happens on the laptops which use the ucm2 sof-hda-dsp,
originally the speaker has output volume A while the headphone has the
output volume B, suppose the speaker is the active port at the moment
and the output volume is A, users plug a headphone to the jack and the
headphone becomes the active port, in this process, ucm_set_port()
calls _disdev/_enadev which triggers the io_mixer_callback(), in the
meanwhile, the module_device_restore will restore the headphone's
volume to B, it will call set_volume_cb() to set the volume to B, but
this value is not written to hw immediately, during the time of
waiting for the B to be written to the hw, the io_mixer_callback()
calls get_volume_cb(), it reads hw volume and gets the volume A, then
it overrides the output volume to A, this results in the headphone
gets the volume A instead of B.
If a machine doesn't use the ucm, this issue will not happen since the
set_port_cb() will not trigger the io_mixer_callback(). If the ports
don't belong to the same sink/source, this issue also doesn't happen.
BugLink: http://bugs.launchpad.net/bugs/1930188
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/577>
* Minimal implementation of --system on win32.
* Wrap main with a Windows Service on win32 (with a fallback to
running it directly).
* Update PA_SYSTEM_{RUNTIME,STATE,CONFIG}_PATH and HOME dynamically
on Windows (overrides the build config, similar to the existing
config path replacement logic).
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/549>
Having G_MESSAGES_DEBUG=all set in the environment (a normal thing to do
when debugging Gnome troubles) causes gsettings-helper to emit a bunch
of helpful gnome debug logs (which is good), but before this change they
were printed on stdout rather than stderr (which was bad!). Rather than
going somewhere the user could see, these log messages were being sent
to the pulesaudio server and interpreted as the src/modules/stdin-util.c
protocol. pulseadio waits to see a '!' message from gsettings-helper
before continuing startup. With the log messages mixed in messing up
the stdin-util protocol, pulseaudio never saw the '!' message, and so
never completed startup.
This simple fix relies on a recent glib > 2.68 (Mar 2021), so builds
against old versions of glib will still have this problem! We consider
this good enough until some complains otherwise.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/1222
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/579>
New card database entry version 5 for card profile is sticky flag.
New messaging API handlers set-profile-sticky and get-profile-sticky.
When card profile is sticky, always restore it even if it is unavailable,
and prevent switching from it when ports become unavailable.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/568>
It seems that in sound context environment variable is not available for
match expression.
This commit utilizes walkthrough to refer to attributes in fw node. The
combination of vendor, model, units is enough to match the node since
the attributes of fw unit doesn't have vendor.
Fix: 37358e42c4 ("alsa: Suppress udev detection of sound card for some units on IEEE 1394 bus")
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/566>
The Volume property on org.bluez.MediaTransport1 is required to utilize
Absolute Volume, but it will only become availabe if the peer device
supports the feature. This happens asynchronously somewhere after the
transport itself has been acquired, after which the callbacks are
attached and software volume is reset.
To prevent race conditions availability of the property is also checked
on startup through a "Get" call.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/239>
Write the current volume to the `Volume` DBus property to keep the
volume on the remote in sync. Without this the remote device shows the
wrong volume, and any attempts to change it will cause an unexpected
jump when the local volume has also been adjusted.
Thanks to prior investments to improve volume synchronization, setting
up callbacks and sending initial volume to the peer for HFP/HSP
implementing this feature is as easy as unconditionally assigning a
valid function to `set_source_volume`. `source_setup_volume_callback`
is already responsible for attaching a `SOURCE_VOLUME_CHANGED` hook and
sending initial (restored) volume to the peer (signifying support for
Absolute Volume - if not derived from the presence of FEATURE_CATEGORY_2
on the profile yet).
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/239>
Like the previous commit this handles `Volume` property changes but
applies them to an A2DP sink instead of source stream. As mentioned in
the AVRCP spec v1.6.2 §5.8 the rendering device (A2DP sink) is
responsible for performing volume attenuation meaning PulseAudio should
pass through audio as-is without performing any attenuation in SW.
Setting a valid pointer to `set_sink_volume` and returning `true` from
`should_attenuate_volume` attaches a hardware callback to `pa_sink` such
that no volume attenuation is performed anymore.
In addition to receiving volume change notifications it is also possible
to control remote volume by writing a new value to the DBus property.
This is especially useful when playing back to in-ear audio devices
which usually lack physical buttons to adjust the final volume on the
sink.
While software volume (used before this patch) is generally fine it is
annoying to crank it up all the way to 100% when a previous connection
to a different device left saved volume on the peer at a low volume.
Providing this bidirectional synchronization is most natural to users
who wish to use physical controls on their headphones, are used to this
from their smartphone, or aforementioned volume mismatches where both PA
as source and the peer as sink/rendering device are performing
attenutation.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/239>
The A2DP spec mandates that the audio rendering device - the device
receiving audio, in our case a `pa_source` - is responsible for
performing attenuation:
AVRCP v1.6.2, §5.8:
The SetAbsoluteVolume command is used to set an absolute volume to be used by the rendering device.
BlueZ models this call as a change of the `Volume` property on the
`org.bluez.MediaTransport1` interface. Supporting Absolute Volume is
optional but BlueZ unconditionally reports feature category 2 in its
profile, mandating support. Hence remote devices (ie. a phone) playing
back audio to a machine running PulseAudio assume volume is to be
changed through SetAbsoluteVolume, without performing any local
attenuation.
Future changes will implement this feature the other way around: setting
an initial value for the `Volume` property as well as propagating
`pa_source` volume changes back to the peer.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/239>
In the case, where the latency is larger than the maximum block size,
module-null-sink will request multiples of the maximum block size from
the sink input instead of limiting the requested amount of data to the
the configured latency.
This patch fixes the problem.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/554>
This reverts commit 96369919e5.
The commit was originally for the issue of Headphone can't output
sound, that was because the Headphone and Lineout share the 1st alsa
mixer and DAC, but this commit introduced a new issue of the speaker
is not muted after switching to headphone.
A recent merged kernel commit (f48652bbe3ae@linux) could fix the 1st
issue, so we could revert the fix of the 1st issue from PA, then the
2nd issue is fixed automatically.
BugLink: https://gitlab.freedesktop.org/pipewire/pipewire/-/issues/747
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/553>
The variable card_name in sink_input_preferred_sink_changed_cb and
source_output_preferred_source_changed_cb could be used uninitialized,
which leads to invalid database entries.
This patch fixes the problem.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/543>
The loopback message may be called after the sink input is already destroyed which causes
a crash. Also memory is leaked because the message object is not correctly freed.
This patch fixes the problems by adding a "dead" flag to the message structure and freeing
the message object on exit.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/541>
When fallback mapping is selected all subsequent profile sets containing
selected mapping are ignored. When there are only e.g. fallback input mappings
available, admitted profile set will only contain one profile with selected
first input fallback mapping and no outputs, and rest of profiles will only
contain outputs and no inputs. When there are only fallback input and output
mappings, there will be no profiles admitted at all.
Fix this by making sure that selected first fallback input or output mapping
is actually allowed to exist in all probed profile sets.
Note while this change allows selected fallback mappings to be found in duplex
configuraitons, probing fallbacks still can fail if there is more than one input
fallback and first one (selected) does not work in duplex configurations.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/539>
This patch fixes the following error:
$ pacat --file-format=ogg -r test.ogg
Failed to open audio file.
$ parecord sep.flac
Failed to open audio file.
libsndfile errors out if a WAV or OGG file is set to have anything but
SF_ENDIAN_FILE:
f4d1646e5c
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/526>
A few headsets have issues if HFP HF profile connection is attempted before
HSP HS profile connection is closed. Looks like this could happen because
bluez bluetoothd alows to make simultaneous HSP HS and HFP HF peer connections.
One of affected headsets is WH-1000XM2
Until we find out how to prevent simultaneous HSP HS and HFP HF connections,
when native backend has HFP HF profile enabled (this is the default) do disable
HSP HS completely unless user explicitly request it via discovery modarg.
Do this by adding module-bluetooth-discover arg enable_native_hsp_hs,
default to inverse of enable_native_hfp_hf.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/538>
For mSBC to work correctly the following must be set correctly
- codec object
- transport write method
- transport setsockopt method
Use helper method to set all three simultaneously.
Static configuration structure may be cleaner solution.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/507>
HFP Audio Connection SCO configuration is negotiated symmetrically in both
directions, and USB HCI SCO packet framing is also symmetric in both directions.
This means that packet size will be the same for reads and writes over HFP SCO
socket.
HFP profile specification states that valid speech data shall exist on the
Synchronous Connection in both directions after the Audio Connection is
established.
This guarantees that an incoming packet will arrive shortly after SCO connection
is established. Use it's size to fix write MTU in case kernel value is wrong.
Discussion here https://lore.kernel.org/patchwork/patch/1303411/
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/507>
The HFP protocol supports the ability to negotiate codecs if that is
supported by both AG and HF. This patch adds advertising of codec
negotiation support and the ability to negotiate a codec change. The
only currently supported extra codec (as of HF 1.7.1) is mSBC. mSBC
requires that the transmission be done over an eSCO link with
Transparent Data. The linux kernel ensures the former, but we have to
manually set the socket to transparent data.
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/507>
Adding processing support for the mSBC codec is somewhat problematic,
because, although it is a SBC codec, the a2dp handling can't simply be
reused because the codec is used on an eSCO link with transparent
data, meaning the transmission unit has to be 48 bytes (fragmenting
the codec packets) and reassembly and boundary detection is required
to be done by the implementation. Therefore we have to implement
separate render and push routines for msbc that do this fragmentation.
Fragmentation is done by emulating circular buffers. The receive
(push) buffer is easy, since the mSBC packet size is 60, simply have a
buffer of this size in the sbc_info area where the fragments are
reassembled. Once we have a full 60 bytes, decode and restart from
zero. The send (render) buffer is more problematic, since the
transmit must be done from contiguous memory. This means that the
buffer must be the lowest common multiple of the transmission unit and
the packet size. This value is 240 since 240/48 == 5 and 240/60 == 4.
So the buffer pointers are reset at 240 which is a whole number of
both rendered packets and eSCO transmission units.
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/507>
When an application sets a device for a newly created stream, we treat
that as a temporary setting, and don't save it as the preferred device
for future streams. The handling for this was broken, however: if the
stream already had a preferred device saved in the stream-restore
database, that was unset.
This was a regression introduced in
bc0e728320 and
70bbbcdc84. These commits tried to detect
in subscribe_callback() when the preferred device is cleared, but as a
side effect the preferred device started to get cleared from the
database also when a stream was created with a device set by the
application.
There's no way for subscribe_callback() to distinguish the different
cases of the preferred device being NULL. This problem is solved by
using the PREFERRED_SINK/SOURCE_CHANGED hooks. The hooks are only called
when the preferred device really changes.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/1063
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/535>
The hooks are fired when the preferred device changes. This is useful
for module-stream-restore.
I added new set_preferred_sink/source() functions for firing the hooks.
The functions also log the preferred device changes.
There was already pa_sink_input_set_preferred_sink(), but that had a
side effect of moving the stream, so I needed a new function. Since it
can be confusing when the two similarly named functions should be
called, I added a comment for pa_sink_input_set_preferred_sink() that
explains the different situations.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/535>
Bluez prepends newly registered profile to a list of supported profiles,
and new peer profile connections are attempted in reverse order of profile
registration.
Currently native backend would register HFP AG profile before HSP AG profile.
When peer supports both HFP HF and HSP HS profiles, this registration order
causes extra HSP HS connection attempt before native backend would reject it
to make sure peer is reconnected with HFP HF profile.
Reorder HSP AG profile registration before HFP AG to make sure peer supporting
both profiles connects with HFP HF profile first.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/534>
Setting these callbacks adds the HW_{VOLUME,MUTE}_CTRL flag even when
PulseAudio is solely responsible for performing attenuation whilst only
keeping the peer posted on changes. For this case the hardware callback
is not registered at all but instead a hook is attached to catch
PA_CORE_HOOK_{SINK,SOURCE}_VOLUME_CHANGED. Only when the peer performs
attenuation (the peer is in HeadSet/HandsFree role) are the callbacks
used, without touching PA software volume at all. A future change could
potentially use software volume to compensate for the extremely coarse
16 steps of volume control in HSP and HFP, and to allow volume over
100%.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/519>
Since commit cb91d7a1 the watermark is increased when there is nothing to rewind.
This is also done in the case when there was actually no rewind requested at all,
so the watermark is increased needlessly.
This patch fixes the issue by skipping the rewind if none is requested.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/530>
Originally written for A2DP this rework of that patch enables late-bound
hardware volume control on HFP and HSP. As per the specification the
headphones (where gain control for both speaker and microphone could
happen in hardware on the peer) are supposed to send initial values for
these before the SCO connection is created; these `AT+VG[MS]` commands
are also used to determine support for it. PA uses this information in
`add_{sink,source}` to attach hardware volume callbacks, _if_ it is
supported. Otherwise PA performs the attenuation in software.
Unfortunately headphones like the WH-1000XM3's connect to A2DP
initially and only send `AT+VGS` (microphone hardware gain is not
supported) _during_ SCO connection when the user switches to the HFP
profile afterwards; the callbacks set up dynamically in
`rfcomm_io_callback` are written after the sink and source have been
created (`add_{sink,source}`), leaving them without hardware volume
callbacks and with software volume when adjusted on the PA side. (The
headphones can still send volume updates resulting in abrupt changes if
software and peer volume differ. Furthermore the same attenuation is
applied twice - once in PA software, once on the peer).
To solve this problem we simply check whether the callbacks have been
attached whenever the peer sends a volume change, and if not attach the
callbacks to the sink/source and reset software volume.
Fixes: d510ddc7f ("bluetooth: Perform software attenuation until HF/HS reports gain control")
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/528>
HF/HS hardware attenuation is optional on HFP: the peer indicates
support with the AT+BRSF command, when bit 4 is set. That does not
explicitly mandate speaker or microphone gain control; either is
dynamically detected as soon as `AT+VG[MS]=` is received. Otherwise
software attenuation is performed.
It is also optional on HSP but nothing is mentioned about feature
detection, assume it is the same as HFP: perform software attenuation
until the HF/HS peer sends an `AT+VG[MS]=` command.
When PA is a HS/HF (and the peer the AG) we attenuate both channels in
software and unconditionally keep the peer up to date with
`AT+VGM/AT+VGS` commands.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/521>
Generalize the distinction between local and peer-attenuated volumes
into a function, paving the way for future changes where this needs to
be checked in more places and when A2DP Absolute Volume support is
added.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/521>
Sink and source naming is more generic when dealing with audio that is
directional in the sense that it either goes to or comes from the other
device, but not necessarily a microphone or speaker. A concrete example
is the swapped meaning when the current device is in the HeadSet
profile. The incoming audio can come from any source, not necessarily a
microphone. Likewise, audio captured by the microphone of the headset is
not necessarily played back by a speaker on the AG, it is merely acting
as a sink for the data: further handling is irrelevant to the naming.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/521>
For the upcoming A2DP AVRCP Absolute Volume feature the code in BlueZ5
has to be generic to be reusable. Move this conversion so that it
becomes possible to implement A2DP volume - which uses different values
- on top without duplicating existing callback functionality.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/521>
The format of COMMAND line sent from HS to AG is COMMAND<cr>
The format of RESPONSE line sent from AG to HS is <cr><lf>RESPONSE<cr><lf>
Split rfcomm_write into rfcomm_write_command and rfcomm_write_response to handle
line formatting correctly.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/520>
Commit 4868fcf5f3 ("daemon: Rely on
systemd unit file for X11 plugin initialization") added a new systemd
unit file, pulseaudio-x11.service, generated from a respective .in file.
Unfortunately, this was only hooked up to meson, and is not currently
installed by autotools. Among other breakage, "make dist" produces a
tarball that meson is then unable to build (because a file is missing).
Signed-off-by: Faidon Liambotis <paravoid@debian.org>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/514>
On cpu-volume-test, cpu_info is initialized only on i386/amd64 systems,
and otherwise passed on to pa_cpu_init_orc() uninitialized.
If one was unlucky enough, they could end up with cpu_info.cpu_type ==
PA_CPU_X86 on a non-x86 system, and use and test the Orc codepath
without that being functional, and thus with the test failing.
This has been observed in the wild on the ppc64el Debian buildds. See
Debian bug #982740 for more context.
Define cpu_info here in the same way as in other tests.
Signed-off-by: Faidon Liambotis <paravoid@debian.org>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/511>
Intel TGL HDMI/DP codec provides 9 pins (Linux kernel, 9a11ba7388f16:
ALSA: hda: hdmi - add Tigerlake support), and with the DP MST enabled,
the linux kernel will build 11 output devices (3, 7, 8, 9, 10, 11, 12,
13, 14, 15, 16), and the alsa-lib will map 11 PCM devices from HDMI:0
to HDMI:10, but current pulseaudio only supports 8 HDMI/DP devices,
if users plug the HDMI/DP monitor to the last 3 ports, the users will
not see the output device from pulseaudio or gnome.
We have experienced this issue on a dell TGL machine with a dock, we
plugged 2 HDMP/DP monitors on the dock, but we could only see 1
HDMI/DP output device from pulseaudio or gnome, through investigation,
we found one monitor is plugged in the 2nd port from last.
Here we add 3 HDMI/DP output devices.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/495>
When HFP HF support is enabled in native backend, peer HFP HF profile connection
is preferred over same peer HSP HS profile connection if peer supports both
profiles.
Enforce the preference by rejecting HSP HS profile connections from such peer.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/491>
Native backend implements HFP AG but not HFP HF yet, therefore headset=auto
functionality is still needed if HFP HF is required.
To make headset=auto work again, drop both HFP AG and HSP AG roles while
performing handover from native backend when oFono is detected running.
While at it, restore profile description to Headset Head Unit (HSP/HFP)
to note that HFP may be still provided via oFono backend.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/491>
Change default backend from 'auto' to 'native' so that in the usual
install pulseaudio uses the native backend with HFP_HF handling.
set default to false unless the backend is the native one, in which
case the default becomes true.
Additionally set default value of enable_native_hfp_hf to false unless
the backend is the native one, in which case the default becomes
true. so that we only bind the HFP_HF end point in the native case
(leaving it free for ofono in the ofono backend or auto case)
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/491>
HFP 1.6 requires a stateful negotiation of AT commands. The prior
version got away with initialising HFP simply by replying 'OK' to
every negotiation attempt. This one actually tries to parse the state
and make sure the negotiation occurs correctly
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
v4:
- Update for PA 11.0
- Finally sort out CIND negotiaton for complex headsets
v3:
- remove internal debugging
- added comment for t->config being not null for hfp
- removed unused returns from hfp_rfcomm_handle()
- remove rfcomm comment
- use pa_startswith
- simplify negotiation
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/491>
When all headsets supported both HSP and HFP, life was good and we
only needed to implement HSP in the native backend. Unfortunately
some headsets have started supporting HFP only. Unfortuantely, we
can't simply switch to HFP only because that might break older HSP
only headsets meaning we need to support both HSP and HFP separately.
This patch separates them from a joint profile to being two separate
ones. The older one retains the headset_head_unit name, meaning any
saved parameters will still select this (keeping us backward
compatible). It also introduces a new headset_handsfree.
For headsets that support both HSP and HFP, the two profiles will
become separately visible and selectable. This will only matter once
we start adding features to HFP that HSP can't support (like wideband
audio).
Signed-off-by: <James.Bottomley@HansenPartnership.com>
---
v6:
- merge profile switching fixes patch from Rodrigo Araujo
v5:
- rename option to enable_native_hfp_hf
- don't call profile_done for HFP_HF unless it was initialised
v3:
- Update for PA 11.0
v2:
- fold in review feedback
- add global disable option for not registering HFP
v3:
- change parameter to enable_profile_hfp
- update device_supports_profile to be aware of hfp/hsp exclusivity
- change parameter to enable_profile_hfp_hf
bluetooth: separate HSP and HFP (to me merged with this patch)
Hi.
First, just to say that your patches are going great. Finally I can use
the microphone of my HFP only headset (a version of a Bluedio T2+).
So far, I've only encontered one problem: the auto_switch option of
module_bluetooth_policy stops working. Dug through the code and I think
you missed a few spots were you have to hangle the new headset_handsfree
profile in module_bluetooth_policy.c
Applying the following after applying your v5 patches fixed the issue
for me, now when I start making a VOIP call the profile switches to
headset_handsfree and the mic works automatically, and when the call
finishes it reverts back to a2dp.
Thanks and best regards.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/491>
The PA_BLUETOOTH_PROFILE names should mirror the PA_BLUETOOTH_UUID
names using profile_function instead of randomly made up names. Fix
this with the transformation:
PA_BLUETOOTH_PROFILE_HEADSET_HEAD_UNIT -> PA_BLUETOOTH_PROFILE_HSP_HS
PA_BLUETOOTH_PROFILE_HEADSET_AUDIO_GATEWAY -> PA_BLUETOOTH_PROFILE_HFP_AG
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
v4: update for PA 11.0
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/491>
When the source-output of a virtual source with volume sharing disabled is moved,
the source output volume is reset to 100%. This patch fixes the problem by
applying the virtual source volume to the source-output after the move.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/509>
The encoding and decoding pipeline are essentially identical: both push
data in via an appsrc, route it through a codec-specific (opaque)
element, and finally pull data out of an appsink. The code already makes
it impossible to have an encoding and decoding pipeline simultaneously
set up in `gst_info`, and converting `bool for_encoding` to a tri-state
(encode, decode, or both) would be messy; particularly when encoding and
decoding could possibly differ in format.
This change removes a swath of code and removes the possibility of
misusing `enc_` or `dec_` in the wrong place (ie. after copying a bit of
code and forgetting to rename one or two). When bidirectional codecs
come online a second codec instance (`gst_info`) can simply be created
and controlled independently.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/487>
LDAC encoder already supports S16, S24, S32 and F32LE. Using FLOAT32LE
for the sample format would avoid the additional call for conversion to
pa_sconv_s32le_from_float32ne. perf tool shows this as being the function
called frequently after encode. So, just avoid this by using sample format
as F32LE.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/486>
Make the code ever so slightly more generic by not using appsrc and
appsink in codec-specific logic when assigning caps specific to the raw
(PCM) format provided by or returned to PA.
Note that caps have to be set (= event) after starting, can't send
events in flushing state.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/484>
SBC codec decrements bitpool value by fixed amount each time it is asked to
reduce output bitrate. This results in reduced audio quality with SBC codec.
Implement increase_encoder_bitrate for SBC codec by adding 1 to bitpool value
each time encoder bitrate needs to be increased to restore SBC audio quality.
While at it, remove bitpool decrement limit to use connection agreed value
instead as we will be able to restore quality later.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/474>
Bluetooth thread may ask encoder to reduce bitrate if writing is not keeping up
with inputs or writing to bluetooth socket takes too much time.
Assuming conditions leading to reduced bitrate are intermittent, allow periodic
attempts to increase encoder bitrate, by default at most twice per second.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/481>
Similar to the situation/comment in `endpoint_release` BlueZ does not
request any reply to `ClearConfiguration()` either; sending one results
in the same "0 matched rules" warning from dbus-daemon:
dbus-daemon[1309]: [system] Rejected send message, 0 matched rules; type="method_return", sender=":1.71" (uid=1000 pid=87548 comm="../build/src/daemon/pulseaudio -vvvv -n -F ../buil") interface="(unset)" member="(unset)" error name="(unset)" requested_reply="0" destination=":1.3" (uid=0 pid=1308 comm="/usr/lib/bluetooth/bluetoothd -d ")
Solve this by only creating a return message when an (othwise empty)
reply is solicited for, just like in `endpoint_release`.
Unfortunately we also have to make sure to not send any error back if no
reply is requested, but fortunately an argument parsing error here is
extremely unlikely.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/472>
We move the codec specific bits to their own respective files and now
make the codec specific initialisation use a GstBin, which the generic
GStreamer module now uses in the pipeline.
It is job of the codec specific function to add elements in the GstBin
and link the added elements in the bin. It should also set up the ghost
pads as a GstBin has no pads of it's own and without which the bin
cannot be linked to the appsrc/appsink.
Also, we now only initialise either the encoding or the decoding
pipeline and not both. The codec init API already gets passed the
for_encoding flag. We pass and use the same to codec specific init
functions.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
As we now support codecs other than SBC, we might have codec which does
not have an encode or a decode capability. Specifically, in the case of
LDAC there isn't a known decoder implementation available. For such a
case, we should not register the corresponding endpoint.
In case of LDAC, as decoding cannot be supported, we should not register
a sink endpoint or vice versa in the other scenario.
To do this, we check if encode_buffer or decode_buffer entry for a codec
has been set in pa_a2dp_codec and accordingly prevent or allow it's
registration.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
When it comes to codecs provided via GStreamer, we register all codecs
if GStreamer option is enabled for bluez5 via meson. However, the
GStreamer plugin required for the codec might not be present on the
system. This results in the codec being available for registration with
the bluez stack or selection by the user, but, trying to use the said
codec then fails.
To prevent the above, we now use the can_be_supported codec API to check
if the codec is usable and if not, we do not register the said codec and
also prevent users from switching to it.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
This API internally checks if a requested codec can be supported on the
system. This is especially required for codecs supported via GStreamer
where the availability of a plugin decides if the said codec can be
supported.
This will be used to prevent registration of a codec which the remote
endpoint device might be able to support, but, PulseAudio can't as the
codec is not available on the system due to the absence of a plugin.
We can also prevent listing or switching to an unavailable codec.
Note that the codec negotiation happens with the bluez stack even before
a device is connected. Because of this, we need to make sure that gst_init
is called before checking for the availability of a plugin. Since
module-bluez5-device gets loaded only after a connection to the device
has been established, doing the gst_init in that or one of the bluetooth
modules is not feasible.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
For example, using the following on the command line will return the
current codec for a bluetooth device
pacmd send-message /card/bluez_card.4C_BC_98_80_01_9B/bluez get-codec
where 4C_BC_98_80_01_9B is the bluetooth device.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
This exposes the currently active codec on the source or sink via the
proplist and can be seen in output of pacmd list-sinks/list-sources.
Also set it on the card. In case of a bi-directional codec, the codec
for the sink and source could be different. For example, for aptX-LL,
the codec name on card, sink and source would be aptx-ll, aptx and sbc
respectively.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
For example, using the following on the command line will return the
list of possible codecs for a bluetooth device
pacmd send-message /card/bluez_card.4C_BC_98_80_01_9B/bluez list-codecs
where 4C_BC_98_80_01_9B is the bluetooth device.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
This adds a generic gstreamer codec module based on which other
bluetooth codecs viz. aptX, aptX-HD, LDAC and AAC can be supported.
The GStreamer codec plugins used here themselves depend on the native
codec implementation.
aptX/aptX-HD -> libopenaptx
LDAC -> libldac
AAC -> Fraunhofer FDK AAC
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
This uses the messaging API to initiate a codec switch.
While a particular codec might be applicable only for a particular
profile, for eg. aptX can only be applicable for A2DP sink or source
and not for let's say HSP, the codec switching logic has not been
tied to the logic for switching profiles.
Codec can be switched by running the following on the command line.
pacmd send-message /card/bluez_card.XX_XX_XX_XX_XX_XX/bluez switch-codec{"ldac_hq"}
pacmd send-message /card/bluez_card.XX_XX_XX_XX_XX_XX/bluez switch-codec {"ldac_mq"}
pacmd send-message /card/bluez_card.XX_XX_XX_XX_XX_XX/bluez switch-codec {"ldac_sq"}
pacmd send-message /card/bluez_card.XX_XX_XX_XX_XX_XX/bluez switch-codec {"aptx_hd"}
pacmd send-message /card/bluez_card.XX_XX_XX_XX_XX_XX/bluez switch-codec {"aptx"}
pacmd send-message /card/bluez_card.XX_XX_XX_XX_XX_XX/bluez switch-codec {"sbc"}
Codec name passed above is matched against pa_a2dp_codec->name. Note that
the match is case sensitive. XX_XX_XX_XX_XX_XX needs to be substituted with
the actual bluetooth device id.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
Instead of letting a codec with higher index have higher priority,
just use a lower index for high priority. This allows the for loop
iterating over the codecs to be written in a straightforward manner
and not have to iterate from the end. FWIW Pipewire does the same.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/440>
If the current active profile is off, it has no sinks and sources, and
if users plug a headset to the audio port, the profile including this
audio port becomes available and should be selected as active profile.
But with the current design, the profile_good_for_output() will return
false because the sources in off profile and target profile doesn't
match.
For example:
(Before users plug headset)
Profiles:
HiFi (Speaker): Default (sinks: 1, sources: 1, priority: 8100, available: no)
HiFi (Headphones): Default (sinks: 1, sources: 1, priority: 8200, available: no)
off: Off (sinks: 0, sources: 0, priority: 0, available: yes)
Active Profile: off
(After users plug headset)
Profiles:
HiFi (Speaker): Default (sinks: 1, sources: 1, priority: 8100, available: yes)
HiFi (Headphones): Default (sinks: 1, sources: 1, priority: 8200, available: yes)
off: Off (sinks: 0, sources: 0, priority: 0, available: yes)
Active Profile: off
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/354>
Volume scaling in dB mode is broken if max dB is negative.
I have a Nobsound USB amplifier (1908:2220) that reports a dB range
of -127.07 dB to -128 dB in Alsa.
While this is likely a driver/device bug, in my naive imagination
userspace wouldn't bother too much with the absolute values and just set
out_dB(percent) = min_dB + (max_dB - min_dB) * percent
However, this is not what PulseAudio is doing, instead max_dB is used
as base_volume with which the desired software volume is multiplied
while min_dB does not seem to be taken into account.
The result is that with this device only a tiny portion of the volume
slider is usable.
Setting it to 97% already reaches min_dB which effectively turns any
(software) audio knob to an on/off switch.
To work around this, simply set the has_dB flag to false if max_dB is
negative.
This falls back to using raw Alsa values (ranging from 0 - 255), now
the settings in pavucontrol perfectly mirror those in alsamixer.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/447>
For GNOME X11 sessions, avoid relying on xdg autostart desktop files
to initialize the X11 plugins. This is now handled via a systemd unit
file.
The xdg autostart is still installed, but has been made to instruct
GNOME to skip it with X-GNOME-HiddenUnderSystemd. This is still the
primary way to initialize X11 plugins for other DEs.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/467>
The scripts in this directory are loaded (in GNOME sessions thus far)
at the time of starting Xwayland for X11 clients (may happen on session
start, or on demand whenever X11 clients are started).
This will ensure the relevant X11 modules are loaded as long as there's
a Xwayland instance, thus X11 clients that might make use of them.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/467>
After applying the commit 0d50e787 ("alsa-card: improve the profile
availability logic"), we met an new issue. when system selects the
initial profile, the profile off is selected instead of a profile with
a valid output device on it. That is the issue we met:
Profiles:
HiFi: Default (sinks: 2, sources: 2, priority: 8000, available: no)
off: Off (sinks: 0, sources: 0, priority: 0, available: yes)
Active Profile: off
Ports:
[Out] Headphones: Headphones (priority: 300, latency offset: 0 usec, not available)
Part of profile(s): HiFi
[Out] Speaker: Speaker (priority: 100, latency offset: 0 usec)
Part of profile(s): HiFi
...
I know the commit 0d50e787 really fixed something, but we still need
to fix the new issue, to do so, this patch introduces a priority bonus
for alsa profiles and separate the alsa profiles to 3 groups:
group a (will be granted priority bonus dynamically):
a profile has only output ports and at least one port is not unavailable
a profile has only input ports and at least one port is not unavailable
a profile has both input and output ports, and at least one output and
one input ports are not unavailable
group b (will be marked unavailable)
a profile has only output ports and all ports are unavailable
a profile has only input ports and all ports are unavailable
a profile has both output and input ports, and all ports are unavailable
group c
the rest profiles, their priority and availability is not changed.
With this change, the profile HiFi will become avaialbe:yes, and will
not be granted priority bonus if no input port is plugged.
The priority bonus provides a higher priority base to profiles, this
guarantees this patch doesn't break the fix of 0d50e787.
https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/927
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/355>
WSAStartup was not being called for pacat and pactl built with meson,
causing them to fail in pa_mainloop_new with "cannot create wakeup
pipe". This issue also affects other applications linking to libpulse
other than the pulseaudio daemon, which calls WSAStartup itself.
When built with autotools, WSAStartup would have been called in
DllMain, which is recommended against by the documentation [1].
To fix these issues, the WSAStartup/WSACleanup calls can be moved
into pa_mainloop_new/pa_mainloop_free.
[1] https://docs.microsoft.com/en-us/windows/win32/api/winsock/nf-winsock-wsastartup
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/456>
State database binary file format may depend on system architecture,
for instance gdbm binary format depends on architecture word size,
making x86 and x64 gdbm files incompatible.
If this is the case, it is handled by adding system architecture name to
database file name using automatically configured CANONICAL_HOST string.
Meson build define CANONICAL_HOST to be system architecture name, while
autotools build extends this with vendor and and operating system components.
Switch autotools build to use host_cpu for CANONICAL_HOST to match Meson
configuration. For backwards compatibility always use existing database file
matching CANONICAL_HOST prefix if it exists.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/425>
When an alsa source with fixed latency is used, the actual latency of the source
will only be one fragment size. This is not taken into account when the required
sink latency is calculated.
This patch fixes the issue.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/451>
Document some things that should be helpful to at least new
contributors. Since we don't have a way to show this when people are
creating MRs, also copy over the next to a merge request template so
that creates a dropdown that folks might look at when creating an MR.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/444>
While module-ladspa-sink is still being loaded and before pa_sink_put() is
called there may be an attempt to reconfigure master sink when avoid-resampling
is true. This breaks attempting to suspend ladspa-sink which is still in INIT
state.
Fix this by skipping pa_sink_suspend if PA_SINK_IS_LINKED is false.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/445>
The array read functions need the state pointer as an additional argument because the
array may be in the middle of a parameter list and the state pointer must be advanced
to the element after the array.
Additionally fixes some compiler warnings.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/450>
This is seen at least on HP EliteDesk 800 DM and HP EliteDesk 800 SFF.
This is used by the analog-output-headphones-2 path, but all other paths
on the same sink need to handle the element too. The existing
configuration is inconsistent between files regarding whether headphone
outputs should be muted or not when not using them. I chose to be
consistent within files, which means that Headphone,1 handling is
inconsistent between files in the same way that the existing Headphone
and Headphone2 handling is. (My opinion is that unused paths should be
always muted, but I didn't want to do that policy change in this patch.)
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/272>
Previously both paths had description "Headphones", which I assume can
cause confusion with users who see two ports with identical names. I
don't have this kind of hardware myself nor have I heard complaints from
users, this is just something I noticed while reading the configuration
files.
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/272>
On some Dell AIO machines, there is no internal mic, only a multi
function audio jack, so the only input devices are headphone-mic and
headset-mic, and they share the Jack with headphone.
When there is no headset plugged in that Jack, the headphone-mic
and headset-mic are off. And since there is no available port under
the analog input source, this source is unlinked (if there is
internal mic, the source will not be unlinked). so the only pa-source
left in the PA is analog-stereo-monitor.
After the headset is plugged, we need to let switch_to_port() handle
headset-mic and headphone-mic conditionally, this will guarantee the
source will be created if it is unlinked before plugging, and then the
input profile could be selected correctly.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/390>
We have at least one USB hardware which supports the 8
channels in one mixer element:
https://github.com/alsa-project/alsa-ucm-conf/pull/25
POSITION_MASK_CHANNELS define was added for the future extensions.
The override_map variable was changed from bool to mask (unsigned int).
The channel map override settings is handled for channels up to eight now.
Also added missing override-map.3 .. override-map.8 to the configuration
parser array.
The driver channel position was added to the override mask arguments
(syntax is driver:pulseaudio like left:all-left). If ommited, the ALSA's
channel positions are guessed by index.
Link: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/292
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/389>
Use safe values for the min_dB and max_dB fields when the position mask
is unset to avoid breakage for the upper levels.
If the range is incorrect, the volume range shown in pavucontrol shows
strange values.
(Thanks to Wim Taymans for the idea.)
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/389>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/400>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/304>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/436>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/433>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/274>
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: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/818
Part-of: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/266>
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.
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.
Since commit ad447d1468 (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.
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).
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.
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/51>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/51>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/51>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/51>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/51>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/51>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/51>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/406>
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: <https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/merge_requests/240>
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
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: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/909
Signed-off-by: Sanchayan Maity <sanchayan@asymptotic.io>
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: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/978
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: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/89
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))) {
|
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);
^
Commits 323195e305 ("switch-on-port-available: Switch to headphones on
unknown availability") and d83ad6990e ("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: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/1043
Since not all users will have environments that asks what they plugged
in when their hardware supports TRRS inputs but don't have impedance
sensing, let's emulate our previous default behaviour of enabling the
headphone port at least.
This can likely be improved so users can configure the module to select
for the device they are most likely to plug in (so an option to enable
just the microphone, or headphones+headset-mic ports).
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/1028
HDMI jacks are configured like this:
[Jack HDMI/DP]
append-pcm-to-name = yes
The pa_alsa_jack.name field is then "HDMI/DP" and pa_alsa_jack.alsa_name
is set to "HDMI/DP,pcm=3 Jack" or similar. If we compare the name fields
of HDMI paths, they appear to use the same jack element even though they
are different in reality, so all HDMI ports got incorrectly assigned to
the same availability group.
Previously they were set once per mapping, which caused the numbering to
restart from 1 for every mapping, so ports were incorrectly assigned to
the same group.
Almost all reports from users, I have seen in last years, were not valid.
The report is also printed when the system scheduler does not wake
the pulseaudio thread in the right time. Users are not able to distinguish
between slow machine and the real problem.
Move the log level from 'error' to 'debug' for those messages.
The right fix should be to measure the time between the call invocation and
return to determine (and skip) the scheduling problems, but it is another
extra code just to debug things.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Packaging shouldn't be using the automatic setting anyway, and let's
disable by default for one release and mark this as experimental so we
can flush out any corner cases.
Since the RTP timestamp is converted to time units and back, a small
error can creep up, which then results in a single frame error in where
we place the buffer in the output memblockq. This results in minor
glitches, so we check for and eliminate the error.
With GStreamer 1.18, the old behaviour of storing the capture time in
DTS is gone (which is reasonable, since the semantics really don't
match). So instead, we get a capture timestamp when the buffer is being
pushed from udpsrc. This should eventually move into udpsrc, and the
timestamp should come from the cmsg instead of the clock.
We still fallback to the DTS if the meta isn't available, as the meta
might be dropped in older versions of rtpL16pay due to a bug.
Hashmap loaded_device_paths contain objects holding keys to entries, and
these objects must be alive while map is emptied.
Reorder freeing this hashmap before destroying device objects to fix
crash on exit.
If port becomes unavailable then PA_CORE_HOOK_PORT_AVAILABLE_CHANGED
callbacks may eventually destroy related source or sink object. Call
this hook after stream is moved to prevent crash reading from freed
memory.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/1008
If write_entry fails to store new entry in database, next time we can try creating new entry again.
With DBUS enabled this will create another dbus entry for same name leading to crash inserting duplicate into dbus_entries map.
Fix this by checking if dbus entry exists in dbus_entries map before creating it.
Fixes: #974
Although the hdmi-output is in well_known_descriptions[] table,
the hdmi device names are indexed (hdmi-output-0), thus there
is no match to assign the proper type automatically.
This patch puts the correct hdmi type to all relevant hdmi
configuration files.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Add missing import of util.h. This fixes a build failure with the
Xcode 12 command line tools which manifests as follows:
error: implicit declaration of function 'pa_thread_make_realtime'
is invalid in C99 [-Werror,-Wimplicit-function-declaration]
Ref https://trac.macports.org/ticket/61107
The current implementation for RTP send isn't optimised for sending MTU
bytes of data like rtp-native. For eg. if MTU is 1280 bytes and we have
to send 1276 bytes, two packets are send out one of 1268 bytes and other
of 8 bytes. Sending out a packet of 8 bytes has a significant overhead
and we should be sending MTU bytes of data.
Fix this by accumulating MTU bytes of data and sending data only on
accumulation of MTU worth of data.
Needed for distcheck to work with a checked-in pulseaudio.pot. Not
updating to the very latest (0.21) to not force the dependency bump to
something "too new" on distros.
We have a requirement to "hide" some hardware drivers, because
other (main) UCM configuration will refer them.
This patch use special error codes to notify the upper layers
to skip the module loading.
BugLink: https://github.com/alsa-project/alsa-ucm-conf/issues/30
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
pa_core_check_idle() uses pa_core.exit_idle_time, which is set after the
pa_core_new() call, so pa_core_check_idle() needs to be called later.
This patch preserves the fact that core state is set to PA_CORE_RUNNING
after checking idle (now in main). It doesn't seem to matter anyway and
main(pa_core_new(state:PA_CORE_STARTUP)->...->state:PA_CORE_RUNNING)
seems right as well.
There were three bugs:
1) j->state_plugged was set to PA_AVAILABLE_UNKNOWN too early. It must
be set only after we have found that the jack is shared by two ports.
The result of setting it too early was that no jack ever could have
the PA_AVAILABLE_YES status.
2) The inner jack loop iterated through p->jacks instead of p2->jacks,
so the code didn't compare jacks between two ports at all. As a result
all ports were put in the same availability group.
3) The inner jack loop checked j->state_plugged instead of
j2->state_plugged. The result was that the speaker port, which uses the
Headphone jack to toggle between unknown and unavailable, was put in the
same group with the headphone port.
This change prepares for adding a doxygen target to the Meson build
system. The DOXYGEN_OUTPUT_DIRECTORY substitution variable is needed so
that the output will go to the build directory. I also replaced @srcdir@
with @top_srcdir@. I think it looks cleaner, since the ".." parent
directory traversal is avoided. It also happened to make writing the
Meson rules easier.
In the current scenario of reading samples from the appsink, it is
observed that we do not actually read all the data available in the
appsink to read. This results in a choppy sound or heard as gaps in
the playback.
The underlying reason for this happening is as follows. Let's say
the appsink new sample callback is called 2-3 times, but, with the
underlying fdsem post machinery when pa_rtp_recv eventually gets
called, there would be those 2-3 samples to read. However, we were
only reading one sample in the current implementation.
Fix this by reading all samples from the appsink in a loop, coalescing
them and then writing to the memchunk.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/889
Signed-off-by: Sanchayan Maity <sanchayan@asymptotic.io>
If the profile is generated from UCM, the priority won't be set so it
stays as 0.
Assume a card has two available profiles, when the selected one becomes
unavailable, module-switch-on-port-available's find_best_profile()
should pick the next available one. However, since the priority is 0,
the "off" profile was chosen instead of the available one.
So let's set the priority to 1 to make profile that is available has
higher priority than "off" profile.
Pipewire has started shipping copies of PulseAudio's ALSA card profiles.
It would be useful if both projects could share the same profiles and
this patch is a step toward that.
UAC v2 and v3 support insertion control (jack detection), and the
created jack mixers have "- Input" suffix and "- Output" suffix for
input jack and output jack, respectively.
Add these jacks so we don't always need to rely on UCM or PulseAudio
profile-set.
When compiled with ASAN: -O1 -fsanitize=address -fno-omit-frame-pointer,
the following issues are seen:
==17217==ERROR: LeakSanitizer: detected memory leaks
Direct leak of 32 byte(s) in 1 object(s) allocated from:
#0 0x7fceba414b40 in __interceptor_malloc (/usr/lib/x86_64-linux-gnu/libasan.so.4+0xdeb40)
#1 0x7fceb9b3eac9 in pa_xmalloc pulse/xmalloc.c:63
#2 0x7fceb9b3ed22 in pa_xmemdup pulse/xmalloc.c:94
#3 0x7fceb9e1eed5 in _pa_xnewdup_internal pulse/xmalloc.h:86
#4 0x7fceb9e1eed5 in init_remap_c pulsecore/remap.c:580
#5 0x7fceb9e1efe5 in pa_init_remap_func pulsecore/remap.c:608
#6 0x5574e72422b7 in remap_init2_test_channels tests/cpu-remap-test.c:303
#7 0x5574e7242420 in rearrange_special_test tests/cpu-remap-test.c:345
#8 0x5574e7245ce5 in srunner_run (/home/eenurkka/pulse/pulseaudio/src/.libs/cpu-remap-test+0x9ce5)
...
SUMMARY: AddressSanitizer: 192 byte(s) leaked in 6 allocation(s).
Fix those issues by freeing the allocated resources properly.
Signed-off-by: Eero Nurkkala <eero.nurkkala@offcode.fi>
When compiled with ASAN: -O1 -fsanitize=address -fno-omit-frame-pointer,
the following issue is seen:
==14272==ERROR: LeakSanitizer: detected memory leaks
Direct leak of 1072 byte(s) in 1 object(s) allocated from:
#0 0x7f0180966d28 in __interceptor_calloc (/usr/lib/x86_64-linux-gnu/libasan.so.4+0xded28)
#1 0x7f018039f043 in pa_xmalloc0 pulse/xmalloc.c:74
#2 0x7f01803c5cc8 in pa_hashmap_new_full pulsecore/hashmap.c:61
#3 0x7f01803c5df9 in pa_hashmap_new pulsecore/hashmap.c:76
#4 0x556ee75ff7f4 in remove_all_test tests/hashmap-test.c:96
#5 0x556ee7602965 in srunner_run (/home/eenurkka/pulse/pulseaudio/src/.libs/hashmap-test+0x6965)
SUMMARY: AddressSanitizer: 1072 byte(s) leaked in 1 allocation(s).
Fix it by freeing the resource properly.
Signed-off-by: Eero Nurkkala <eero.nurkkala@offcode.fi>
When the Thumb instructions set is used and frame pointers are enabled
(-fno-omit-frame-pointer), r7 can't be used, because it's used for the
frame pointer. Trying to use r7 caused the compilation to fail.
Thanks to Andre McCurdy for suggesting[1] this fix, all I had to do was to
test that it works. The code builds now, and cpu-remap-test also
succeeds.
[1] https://lists.openembedded.org/g/openembedded-core/message/136786
When playing music for a period of time, the Shared Memory is
frequently accessed, and occasionally read/write errors occur,
which causes the program to crash
[Current thread is 1 (Thread 0xffff86823010 (LWP 2841))]
(gdb) bt
0 0x0000ffff8702a714 in __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
1 0x0000ffff870188e8 in __GI_abort () at abort.c:79
2 0x0000ffff873b5728 in do_read (p=p@entry=0x3673a170, re=re@entry=0x3673a338) at pulsecore/pstream.c:856
3 0x0000ffff873b7fd8 in do_pstream_read_write (p=0x3673a170) at pulsecore/pstream.c:248
4 0x0000ffff873b8368 in srb_callback (srb=<optimized out>, userdata=0x3673a170) at pulsecore/pstream.c:287
5 0x0000ffff873b8bec in srbchannel_rwloop (sr=0x36766ae0) at pulsecore/srbchannel.c:190
6 0x0000ffff87339c70 in dispatch_pollfds (m=0x36670db0) at pulse/mainloop.c:655
7 0x0000ffff87339c70 in pa_mainloop_dispatch (m=m@entry=0x36670db0) at pulse/mainloop.c:898
8 0x0000ffff8733a01c in pa_mainloop_iterate (m=0x36670db0, block=<optimized out>, retval=0xffffd9683030) at pulse/mainloop.c:929
9 0x0000ffff8733a0d8 in pa_mainloop_run (m=m@entry=0x36670db0, retval=retval@entry=0xffffd9683030) at pulse/mainloop.c:945
10 0x0000000000406894 in main (argc=<optimized out>, argv=<optimized out>) at daemon/main.c:1144
Signed-off-by: zhaochengyi <zhaochengyi@uniontech.com>
On the machines with the ucm used, the different input/output devices
often have different pcm stream, so they often belong to different
sources and sinks, this is greatly different from the design of all
devices connected to a codec (without ucm).
For example, on a machine with ucm2 used:
the internal dmic is on source#0
the external mic is on the source#1
the internal spk is on sink#0
the external headphone is on sink#1
Users expect that after plugging the external device, it will become
the active device automatically. The switch-on-port-available could
make it to be the active_port on its own source/sink, but can't make
source/sink to be default_source/sink since the sources/sinks belong
to the same profile (HiFi usually).
If we adjust the source/sink priority according to ucm ports priority,
the device_port.c could handle the default_source/sink changing then.
Usually we set higher priority for external device than internal
device in the ucm.
In order to bring the lowest side effect on the source/sink priority,
I change the ucm priority to units digit first, then add it to the
original priority.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
We met a weird situation on a couple of Lenovo machines and at least
on one Dell machine. First we open the gnome-sound-setting, then
suspend and resume the system, after the system resume back, the audio
devices change to dummy, the audio doesn't work anymore. And pacmd
list-cards shows no available sound card.
Through debugging I found after resume, the alsa receives POLLERR
events and it will call unsuspend to recover the pcm, but at that
moment, the device nodes in /dev/snd/ is not accessible, so the
snd_pcm_open() fails and the pulseaudio unload the module-alsa-card.
Here I add retry and pa_msleep if snd_pcm_open fails, I tested it on
all machines which have this problem, pa_msleep(25) is ok for most of
them, there is only one machine which needs to call pa_msleep(25)
twice, so for safety reason, I set the max retry times to 4.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
It's possible for mdev to be NULL. In this case, an assert is taken
in pa_alsa_open_mixer_by_name() with debug builds, and a crash with
release builds. However, it's possible to bypass this trouble by taking
the error path if mdev is NULL.
Reported-by: Jarkko Sankala <jarkko.sankala@offcode.fi>
Signed-off-by: Eero Nurkkala <eero.nurkkala@offcode.fi>
In single profile mode (headphone and speaker use different PCMs),
when headphone is plugged in, pa_device_port_set_available() will call
pa_core_update_default_sink/source() before posting
PA_SUBSCRIPTION_EVENT_CARD|PA_SUBSCRIPTION_EVENT_CHANGE to the gnome.
And pa_core_update_default_sink/source() will post
PA_SUBSCRIPTION_EVENT_SERVER | PA_SUBSCRIPTION_EVENT_CHANGE to the gnome.
So the original event sequence is:
1. PA_SUBSCRIPTION_EVENT_SERVER | PA_SUBSCRIPTION_EVENT_CHANGE
2. PA_SUBSCRIPTION_EVENT_CARD | PA_SUBSCRIPTION_EVENT_CHANGE
In gnome-control-center:
When it receives PA_SUBSCRIPTION_EVENT_SERVER, it will call
req_update_server_info () to update the panel;
When it receives PA_SUBSCRIPTION_EVENT_CARD, it will update
the card information, for example, when the headphone is connected,
it will call gtk_list_store_append() to append the headphone.
Let's use an example to clarify the correct sequence.
Assume we plug in headphone. PA will set the default sink to headphone
from speaker, and hope gnome sound setting "Output Deivce" to highlight to
"headphone". PA should send PA_SUBSCRIPTION_EVENT_CARD firstly to notify
gnome-control-center "headphone" is plugged in. And then it sends
PA_SUBSCRIPTION_EVENT_SERVER to trigger sound setting to highlight
to "headphone".
Signed-off-by: Libin Yang <libin.yang@intel.com>
Previously avoid_resampling was always false unless the sink or source
implementation explicitly configured the variable. The null sink doesn't
explicitly configure it, so it didn't switch the sample rate as
expected when avoid_resampling was enabled.
This change means that also sinks that don't support rate switching can
have avoid_resampling set to true, but I think that's fine, because
pa_sink_reconfigure() doesn't try to do anything if the reconfigure()
callback isn't set.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/923
I spent a little time working through the implementation of
pa_hashmap, and wrote a test suite while doing so. It tests a few
basic edge cases, like saturating all buckets of the hashtable.
In some cases, the I/O connector functionality can be shared
and we cannot determine the proper purpose automatically.
We just know that something was inserted to the jack.
Introduce a group identifier (a simple string - unique
per group) which helps to determine the proper ports
for the application. The user interface may be used
to set the wanted behaviour.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
fac_table[] lacks of PA_SUBSCRIPTION_EVENT_CARD item. This will cause
pulseaudio crash when it tries to dump the PA_SUBSCRIPTION_EVENT_CARD
event when DEBUG is defined.
Signed-off-by: Libin Yang <libin.yang@intel.com>
There is some case that multiple ucm devices share an amixer Jack
like "Headphones", "Headset" and "Mic2" share the "Headphone Mic Jack",
When the Jack state is changed, the module-switch-on-port-available
will process them in the order they are in the jack->ucm_devices, and
the last device will decide the final setting.
But usually users put priority for those devices and expect the
final setting is based on the highest priority device if there is no
other policies like manual selection. So here do some change to store
the ucm_devices according to their priority (from low to high).
For example, we have ucm devices definition like below (ucm2):
SectionDevice."Mic2" {
Comment "Headphones Stereo Microphone"
...
Value {
CapturePriority 200
...
}
SectionDevice."Headset" {
Comment "Headset Mono Microphone"
...
Value {
CapturePriority 300
...
}
}
Without this patch, the final setting is based on Mic2, after applying
this patch, the final setting is based on the Headset (with higher
priority than Mic2).
Signed-off-by: Hui Wang <hui.wang@canonical.com>
There were three maybe-uninitialized warnings when building with
Autotools (for some reason I don't see these with Meson):
modules/raop/raop-sink.c: In function ‘thread_func’:
modules/raop/raop-sink.c:543:16: warning: ‘intvl’ may be used uninitialized in this function [-Wmaybe-uninitialized]
if (intvl < now + u->block_usec) {
^
In file included from ./pulsecore/macro.h:270,
from ./pulsecore/cpu-x86.h:25,
from ./pulsecore/cpu.h:23,
from ./pulsecore/core.h:26,
from modules/raop/raop-sink.c:48:
./pulsecore/log.h:129:28: warning: ‘check_timing_count’ may be used uninitialized in this function [-Wmaybe-uninitialized]
#define pa_log_warn(...) pa_log_level_meta(PA_LOG_WARN, __FILE__, __LINE__, __func__, __VA_ARGS__)
^~~~~~~~~~~~~~~~~
modules/raop/raop-sink.c:404:14: note: ‘check_timing_count’ was declared here
uint32_t check_timing_count;
^~~~~~~~~~~~~~~~~~
modules/raop/raop-sink.c:500:27: warning: ‘last_timing’ may be used uninitialized in this function [-Wmaybe-uninitialized]
pa_usec_t since = now - last_timing;
^~~~~
I moved the intvl variable initialization out of the for loop, because
it looked like the variable value is supposed to be remembered between
the iterations. I don't know if the variable declaration (without
initialization) in the beginning of the loop caused the compiler to
touch the variable between iterations, probably not, but I'm pretty sure
that's undefined behaviour.
Other than that maybe-undefined behaviour, these compiler warnings may
be false positives, since the variables are initialized when u->first is
true.
I initialized the three variables in to the same value as what is used
when resetting them when u->first is true. I didn't test these changes,
but they look safe to me.
Prior to commits f899d5f466 and
f62a49b8cf, GNOME's sound settings
overwrote the routing for all entries in the stream-restore database
when selecting a device. Now we prevent that from happening (see the
aforementioned commits), but the old overwritten settings can still be in
the database after updating to PulseAudio 14.0, and they can cause
problems, as documented here:
https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/832
We can't distinguish between devices set by GNOME's sound settings
and devices set by the user, so this patch discards all old device
settings, even though that is going to cause PulseAudio to forget routing
settings for many users. This is less bad than keeping the incorrect
routing settings in the database, because it's difficult for users to
figure out how to fix the situation when e.g. speaker test tones go to
the internal speakers no matter what device is selected as the default,
whereas old manual configuration can be restored restored by doing the
manual configuration again. Also, it's probably more common to have at
some point changed the default device in GNOME's sound settings than it
is to have any manual per-stream routing settings.
This is disabled by default, because this causes data loss, but
distributions that use GNOME are recommended to enable this with
the --enable-stream-restore-clear-old-devices (Autotools) or
-Dstream-restore-clear-old-devices=true (Meson) build option.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/832
pa_namereg_is_valid_name() will hit an assertion if the name string is
NULL. Maybe it would make sense to change pa_namereg_is_valid_name() so
that it would return false on NULL, but I didn't want to change the
function semantics at this time.
e->device and e->card can be NULL even when device_valid and card_valid
are set to true if the database contains bad data.
I ran into this crash while developing new code, I haven't seen the
crash in the wild.
Storing the version in the entry struct is pointless. We should always
write entries using the current version. When we encounter older
versions when reading, those need to be converted to the current version
anyway, because all code that uses the entry struct assumes that the
data is stored according to the current version semantics.
We're currently at the first version of the database entries, so
currently there's no version conversion happening. I have a patch that
will increment the entry version, so this is preparation for that.
In case the local UDP port is blocked by a firewall by default, send
an initial timing packet so the connection tracking will accept the
response packages.
Otherwise, the connection will fail with an 'RTSP/1.0 500 Internal
Server Error' after some timeout.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/840
libpulsecore is not supposed to depend on the client library.
Removing the dependency caused build failures, which are fixed by adding
more stuff to libpulsecommon.
- remove duplicate mixer initialization in sink
- use the similar mixer initialization for source like for sink
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
The UCM device must be activated using the pa_alsa_ucm_set_port()
call on boot (the sink creation time). In case when the
mixer controls are not defined for the application in the
UCM configuration, the mixer_handle is NULL.
Call the pa_alsa_ucm_set_port() before the mixer_handle check.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Since the .pot file can be automatically generated, it hasn't been
included in the git repository so far. However, we're planning[1] to
start using Fedora's Weblate translation service, and it requires the
.pot file to be in the repository.
[1] https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/785
The Zanata version is more recent, and there haven't been other changes
to upstream than the fixes in a15cde4179.
I modified the Zanata version to include those fixes.
The Zanata version is more recent, and there don't seem to be many
changes that would be lost in our upstream version since 2012 (based on
the file header, the Zanata version was probably in sync with master
some time in 2012).
Merged with this command, which takes new translations from the Zanata
version, but doesn't take changed translations, except when they clear
the fuzzy flag from a translation:
msgcat --use-first po/ca.po po/ca.po.zanata -o po/ca.po
I also manually copied new translator names to the header comment. I
didn't update the header key/value section, but here are the interesting
bits from the Zanata version:
"POT-Creation-Date: 2015-10-06 16:57+0200\n"
"PO-Revision-Date: 2019-07-25 10:42+0000\n"
"Last-Translator: Robert Antoni Buj Gelonch <rbuj@fedoraproject.org>\n"
"Plural-Forms: nplurals=2; plural=(n != 1)\n"
Support for multiple codecs needs to use a new Bluez API which pulseaudio
does not implement yet.
So register explicitly only SBC codec which is provided by pulseaudio A2DP
codec API.
It's completely normal to not have explicit channel configuration for
stereo devices. In fact, the ALSA developers actively avoid configuring
the channels for stereo devices.
I also dropped the word "duplex" from the messages, because "stereo
duplex" implies bidirectionality, but most devices use one direction
only.
When running under systemd from its `.service` file, the daemon is
started with `--daemonize=no`. This means that the default logging
target is `stderr` (see the documentation for `--log-target` in
`pulse-daemon.conf(5)`). That works fine, but results in all the
structured logging data from the `pa_log()` calls being thrown away and
not making it into the journal.
In order to preserve structured logging data, and hence make the
messages in the journal a little more useful (for example, allowing the
user to filter by message priority), explicitly pass
`--log-target=journal` in the `.service` file. This should always be
appropriate because the journal should always be used with systemd.
Signed-off-by: Philip Withnall <withnall@endlessm.com>
This updates things based on changes in the templates that we use. Also
pins the ref in the template repo so that our build does not break when
the template parameters change.
This does mean that we should likely periodically check the ci-templates
repo, but this seems to be better than the build breaking unexpectedly.
gcc10 can effectively emit single precision registers if right
operand modifier constraint is not in use
This results in assembler rejecting the code
/tmp/ccEG4QpI.s:646: Error: VFP/Neon double precision register expected -- `vtbl.8 d3,{d0,d1},s8'
/tmp/ccEG4QpI.s:678: Error: invalid instruction shape -- `vmul.f32 d0,d0,s8'
Therefore add %P qualifier to request double registers sinece 'w' could
mean variable could be stored in s0..s14 and GCC defaults to printing out s0..s14.
Note those registers map to d0..d7 also.
Output generated is exactly same with gcc9, and it also now compiles
with gcc10
Its not documented well in gcc docs and there is a ticket for that
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84343
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Currently pa_{sink,source}_move_streams_to_default_{sink,source}() check the
availability of the old sink or source. The sink or source is only marked as
unavailable if the active port of a sink or source is not available.
Therefore sinks or sources without port are always considered available,
even if they are in the process of being unlinked and streams are not
rescued.
This patch removes the availability check because it is unnecessary. The
functions are only called if the sink or source becomes unavailable or if
the default sink or source changes, therefore the default_sink_changed or
default_source_changed argument can be used as an indicator if the old
sink or source is still present. In the case that the old default sink or
source becomes unavailable, the function will be called twice, once when
the default sink or source changes and once when the old sink or source
is unlinked.
pactl has these commands, too. Use the same order as the
man page, except the undocumented 'help'. Note that the commands are
sorted alphabetically when completed anyway, though
that can be disabled since Bash 4.4
(https://unix.stackexchange.com/q/215937/111181)
The zsh completions already support set-default-sink and
set-default-source.
On certain types of filesystem (especially NFS appliances which support
multiple operating systems), the user's home directory may report as
being owned by root rather than the user, yet still permit the user to
create and modify files normally (which will be owned by them).
Our users have home directories hosted on a NetApp storage appliance
which uses mixed-mode ACLs but where the home directory is set up with
NTFS ACLs at the top level. This means they have the expected effective
permissions, but the ownership reports as root. This could also be the
case if the filesystem were using NFS4 ACLs or similar.
Currently, when the master of a virtual source is moved, the change of the
asyncmsgq is not propagated to other attached virtual sources. This leads
to a crash when the original master source is no longer available.
This patch fixes the issue by modifying the moving callback to propagate the
change to attached virtual sources.
Virtual sinks show a similar bug but that will be fixed in a different patch
series.
The memblockq stores data in the virtual sink format, not in the master
sink format. The wrong sample spec caused a crash when the virtual sink
rendered data whose length was not divisible by the sink input frame
size.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/786
The request_rewind() callback of the uplink sink called
pa_sink_process_rewind(), which is not allowed. Things are supposed to
happen so that first a rewind is requested, and then during the next
rtpoll loop the sink will process the request. Calling
pa_sink_process_rewind() during the request phase caused a crash.
Having a request_rewind() callback is completely unnecessary, because
it's only useful for forwarding the request to a downstream sink. In
this case there is no downstream sink.
I also set max_rewind to 0, because the sink doesn't support rewinding.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/775
If a channels argument is passed module-jackdbus-detect, it is passed to both
module-jack-sink and module-jack-source when those are started. This is a
problem if you want a different number of input channels from output channels.
In particular, if you want more of one than you physically have of the other,
it will fail. This commit adds separate source_channels and sink_channels
arguments to be able to specify the channels arguments to module-jack-source
and module-jack-sink separately. The combined channels argument is kept for
backwards compatibility and will be used as a default for source_channels and
sink_channels if either of them is omitted.
module-jackdbus-detect documents the channels argument as optional and "if
omitted, the sink wil use the number of physical output port and the source
will use the number of physical input ports registered in the JACK server."
However, although it would correctly omit the channels argument to
module-jack-sink and module-jack-source if its channel argument was omitted,
its argument validation was broken to consider omitting channels an error.
This commit properly validates the channels argument so omitting it is
accepted.
As the comment says, switching to HDMI automatically often causes
problems. Commit bae8c16bfa
("switch-on-connect: Do not ignore HDMI sinks") enabled switching to
HDMI earlier. It was known already then that HDMI monitors don't
necessarily have speakers on them, but I accepted the patch on the
basis that module-switch-on-connect acts only if the card profile is
switched to HDMI, so if switching to HDMI is wrong, then already the
profile switch should cause problems. I didn't think of the case where
the default sink is on some other card, in which case switching the
profile on the HDMI card doesn't cause problems by itself.
I don't want to revert bae8c16bfa, because João needs to be able to
configure their systems to automatically switch to HDMI. Instead, this
patch utilizes the new blacklisting feature in module-switch-on-connect
to blacklist HDMI sinks by default. Switching to HDMI can be enabled by
setting the blacklist modarg to an empty string or something that
doesn't match HDMI sinks.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/749
The ALSA mixer can be opened multiple times (especially for UCM
in the probe). This adds a simple mixer cache to prevent
multiple open calls.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
This patch adds the autoreconnect feature to the raop module.
This is mainly to be used in a server context, but can be used
also in a desktop usage context.
With autoreconnect feature, the raop module behaves like this:
- At initialisation or in case of the RTSP TCP connection lost, it
tries to reconnect every 5 seconds
- In case of any fatal error, it tries to reconnect every 5 seconds
- In UDP mode, if no timing packets received anymore for a long time,
RTSP connection is closed, then it tries to reconnect..
- After reconnection, once RTSP session has been established again,
playing is resumed automatically.
- When the connection is not established yet (or loss), the sink
behaves like a null sink. In the source code I called it "autonull",
even if autonull is set to autoreconnect param value, it could be
split into two different params.
The old wording could be understood so that the default sink/source
would always be used, but sometimes a policy module does a different
decision (for example module-stream-restore).
Related: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/767
This is just invalid. It results to an error in almost all cases.
The hw:<number> scheme is sufficient to get the right card mixer.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Remove the implicit rule. It is perfectly ok to have the jack with
the same name for another I/O in the driver. Trust only the
value obtained from UCM.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
The UCM library is used to get the fallback values from the verbs
and the defaults section. There is no reason to duplicate this code
inside application.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
When users select an input device from gnome-control-center UI, the
source of this input device will be set to the
configured_default_source and the default_source, these actions are
expected, but after these actions, the gnome-control-center will call
extension_cb() to modify the entries in the database, let all stream
entries to bind the source users select, this is not correct since the
source is default_source now.
This is a temp fix for this issue, after gnome-control-center fixes
this problem, this patch should be reverted.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When a source is unlinked, all streams of this source are moved to
default_source, this action is implemented in the core rather than
modules now.
And after this change, the module-rescue-streams is not needed, but
for backward compatibility, we keep it as a dummy module.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When the active port of a source becomes unavailable, all streams from
that source should be moved to the default source.
When the active port of a source changes state from unavailable, all
streams that have their preferred_source set to this source should be
moved to this source.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When a new source appears, all streams that have their
preferred_source set to the new source should be moved to the new
source.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When the default source changes, the streams from the old default
source should be moved to the new default source, unless the
preferred_source string is set to the old default source and the
active port of the old default source is not unavailable
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When the user moves a stream to the current default source, the
preferred_source should be set to NULL and module-stream-restore
should clear the routing for that stream in the stream database. From
that point on the stream will be always routed to the default source.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
If the source here is NULL, that means users want to clear the
preferred_source and move the source-output to the default_source,
otherwise set the preferred_source to the source->name and move the
source-output to the source. After that fire the source_output_change
event.
After adding this API, we can use this API to simplify the entry_apply
in the module-stream-restore.c.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
The finial objective is to store the preferred source name in the
source-output struct, and use module-stream-restore to save and
restore it.
This patch just replaces the save_source with preferred_source, and
tries to keep the original logic.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Hardwares without SNDRV_PCM_INFO_RESUME capability, like USB Audio,
don't support snd_pcm_resume() when it's in suspended state.
Let's use snd_pcm_hw_params_can_resume() to check hardware's capability
before snd_pcm_resume() attempt. If it doesn't support resume, just go
to snd_pcm_drop() to leave suspended state directly.
remixing-produce-lfe controls upmixing, and remixing-consume-lfe
controls downmixing. The motivation is that a user might want to
synthesize LFE while playing stereo audio on his/her 5.1 speakers,
but at the same time follow the industry recommendation to omit
the LFE channel when producting a stereo downmix (e.g. for headphones)
from 5.1 content. Or the other way round.
Fixes: #753.
The mixer identifiers should be used for snd_mixer_selem API.
Use them as first, then try to fallback to the raw control
identifiers.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
- sync mixer logic added
- mixer path creation, empty set in mapping creation, paths added in path creation
- path creation moved inside new port creation as it might be called twice otherwise
- some comments added
This allows us to support the PlaybackVolume and CaptureVolume commands
in UCM, specifying a mixer control to use for hardware volume control.
This only works with ports corresponding to single devices at the
moment, and doesn't support stacking controls for combination ports.
The configuration is intended to provide a control (like Headphone
Playback Volume), but we try to resolve to a simple mixer control
(Headphone) to reuse existing volume paths.
On the UCM side, this also requires that when disabling the device for
the port, the volume should be reset to some default.
When enabling/disabling combination devices, things are a bit iffy since
we have no way to reset the volume before switching to a combination
device. It would be nice to have a combination-transition-sequence
command in UCM to handle this and other similar cases.
PlaybackSwitch and CaptureSwitch are yet to be implemented.
When users select an output device from gnome-control-center UI, the
sink of this output device will be set to the configured_default_sink
and the default_sink, these actions are expected, but after these
actions, the gnome-control-center will call extension_cb() to modify
the entries in the database, let all stream entries to bind the sink
users select, this is not correct since the sink is default_sink now.
This is a temp fix for this issue, after gnome-control-center fixes
this problem, this patch should be reverted.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When a sink is unlinked, all streams of this sink are moved to
default_sink, this action is implemented in the core rather than
modules now.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When the active port of a sink becomes unavailable, all streams from
that sink should be moved to the default sink.
When the active port of a sink changes state from unavailable, all
streams that have their preferred_sink set to this sink should be moved
to this sink.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When a new sink appears, all streams that have their preferred_sink
set to the new sink should be moved to the new sink.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When the default sink changes, the streams from the old default sink
should be moved to the new default sink, unless the preferred_sink
string is set to the old default sink and the active port of the old
default sink is not unavailable
Signed-off-by: Hui Wang <hui.wang@canonical.com>
When the user moves a stream to the current default sink, the
preferred_sink should be set to NULL and module-stream-restore
should clear the routing for that stream in the stream database. From
that point on the stream will be always routed to the default sink.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
If the sink here is NULL, that means users want to clear the
preferred_sink and move the sink-input to the default_sink, otherwise
set the preferred_sink to the sink->name and move the sink-input to
the sink. After that fire the sink_input_change event.
After adding this API, we can use this API to simplify the entry_apply
in the module-stream-restore.c.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
The finial objective is to store the preferred sink name in the
sink-input struct, and use module-stream-restore to save and restore
it.
This patch just replaces the save_sink with preferred_sink, and tries
to keep the original logic.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
pa_modargs_get_value() returns a const string -- instead of discarding
the const qualifier, let's just duplicate the string and free it
explicitly in the failure case.
Add a new module argument, blacklist, which is a regular expression.
If the sink/source name matches the provided blacklist regex, don't
automatically switch to it. By default, no devices are blacklisted.
Add a new function to check whenever a regex pattern is valid, plus
extra NULL asserts in pa_match.
Add an xauthority parameter and use it in the startup script.
Apparently on some systems the X authentication cookie is not stored in
~/.Xauthority but in some dynamic location pointed to by the XAUTHORITY
environment variable. The environment variable therefore needs to be set
in the PulseAudio daemon environment in order to have access to the X
server from the PulseAudio daemon, but the variable is not necessarily
set when starting PulseAudio. For example, systemd starts PulseAudio
outside the X session. The start-pulseaudio-x11 script is run in the
X session, so it has the environment variable available, and can pass it
to the X modules, which then can set the variable in the daemon
environment.
RedHat bug: https://bugzilla.redhat.com/show_bug.cgi?id=1723065
Debian bug: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=593746
Based on patch by Alexander Kurtz <kurtz.alex@googlemail.com>
Currently, the version check snippet uses a 'bash' extension which
arithemtically evaluates variables prior to expansion. This approach
does not nesseceraly work on other shells which may complain with
'5: Illegal numer' error. Expand the arithmetic expression before
evaluation to avoid such an error.
Signed-off-by: Vasilis Tsiligiannis <acinonyx@openwrt.gr>
This adds a GStreamer-based RTP implementation to replace our own. The
original implementation is retained for cases where it is not possible
to include GStreamer as a dependency.
The idea with this is to be able to start supporting more advanced RTP
features such as RTCP, non-PCM audio, and potentially synchronised
playback.
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
It is possible that we might want to have a separate userdata to be used
for these callbacks, so let's split them out.
This is particularly needed when using an pa_rtpoll_item around pa_fdsem
since that uses its own before/after callback but will essentially have
whatever is using the fdsem set up the work callback appropriately (and
thus at least the work callback's userdata needs to be separated from
the before/after callback -- we might as well then just separate all
three).
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
This moves RTP implementation-specific information out of
module-rtp-send/recv. This is basically done by making the
pa_rtp_context structure opaque from the perspective of these modules.
We can then potentially replace the underlying RTP implementation with
something else transparently.
One RTP detail that does "leak" is the RTP timestamp. We provide this to
module-rtp-recv so that it can perform rate adjustments to match the
sender rate.
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
module-rtp-send itself doesn't really need to handle this, the
implementation can keep track (and make sure sending happens in MTU
sized chunks).
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
There doesn't seem much value in supporting streaming U8/mulaw/alaw on
the network, and it's unlikely these get any testing. Makes more sense
to drop these formats and just convert to L16 if we're dealing with
source media in that format.
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
In commit f727cd9a `int error` member in `struct pa_context` was changed
to `pa_context_error *error`. The memory that is allocated with
`pa_xnew0` in src/pulse/context.c:142 is never freed, and it causes
a leak of 4 bytes. The leak can be easily detected with leak-sanitizer.
Almost all distributions patch the configuration to disable
flat-volumes, because users tend to find the concept confusing (and it
also causes nasty surprises when some application pushes the volume to
100%). Let's remove the need for patching and disable the feature by
default.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/691
Silences these warnings:
[509/574] Compiling C object 'src/tests/a4ccf2d@@alsa-mixer-path-test@exe/alsa-mixer-path-test.c.o'.
../src/tests/alsa-mixer-path-test.c:24:20: warning: ‘load_makefile’ defined but not used [-Wunused-function]
static pa_strlist *load_makefile() {
^~~~~~~~~~~~~
../src/tests/alsa-mixer-path-test.c:17:20: warning: ‘get_default_paths_dir’ defined but not used [-Wunused-function]
static const char *get_default_paths_dir(void) {
^~~~~~~~~~~~~~~~~~~~~
Reading properties from the X11 root window is meant to provide 1:1 results
with reading the configuration directly in the local case. This configuration
is essentially different only in remote cases.
Add an extra check for the SSH_CONNECTION envvar, so we don't even need
opening a X11 display connection for IPC in the most usual case.
Subdirectories add to the top-level cdata (specifically, the SIMD
detection happens in the pulsecore meson.build), so we were missing
HAVE_MMX/SSE2/NEON defines without this fix.
pulseaudio does not link against libbluetooth, as it's only talking to the
bluez daemon over dbus. So the build dependency on libbluetooth is overly
restrictive, as some embedded systems choose to ship without libbluetooth
but still have bluez daemon support.
This syncs the meson to the autotools configuration behavior by changing
the bluez option to a default on boolean.
I don't know why these options were being passed to configure
(--sysconfdir has been there from the very beginning, --localstatedir
got added when the system mode was added). Overwriting system files by
default is not good, so let's not set these options.
The various software volume implementations were being built as part of
libpulsecommon for some reason. These should only ever be used in the
daemon, so they should be in libpulsecore.
Removes a warning from HAVE_GCONF not being set, and fixes generation of
a large section that depends on OS_IS_WIN32 being explicitly set to 0.
We can't set OS_IS_WIN32 to 0 by default since a bunch of code uses it
via an ifdef rather than by value.
This was being done automatically by autotools, now we need to manually
specify this for each executable/library with a dependency in a
non-standard directory.
For ease of maintaining both build systems, use the same version info
sequences as configure.ac. This should be simplified after Autotools has
been dropped.
- Rename "pulsedspdir" to the same "padsplibdir" that Autotools uses.
- Add a new option "pulsedsp-location" that is only used for padsp.in,
just like Autotools' --with-pulsedsp-location.
- Use 'set' instead of 'set_quoted' to avoid PULSEDSP_LOCATION getting
quoted twice.
Rename struct rtp_payload to rtp_sbc_payload as it is specific for SBC
codec payload.
Add proper checks for endianity in rtp.h header and use uint8_t type
where appropriated.
Field frame_count is only 4 bit number, so add checks to prevent overflow.
And because is_fragmented field is not parsed by decoder there is no
support for decoding fragmented SBC frames. So throw an error in this case.
Add explanation why minimal bitpool value is used in SBC codec as initial
bitpool value for A2DP source.
Set buffer size for reading/writing from/to A2DP socket to exact link MTU
value. This would ensure that A2DP codec does not produce larger packet as
maximal possible size which can be sent.
Because A2DP socket is of SOCK_SEQPACKET type, it is guaranteed that
we do not read two packets via one read/recvmsg call.
Properly check for all return values of encode/encode methods of A2DP codec
functions. They may fail at different levels. Also encode or decode API
method may return zero length buffer (e.g. because of algorithmic delay of
codec), so do not fail in this case.
conditions this may lead to massive slowdown of floating point operations
when underflows or denormals are encountered. In particular, this problem
was observed with the soxr resampler after applying
https://gitlab.freedesktop.org/pulseaudio/pulseaudio/merge_requests/120
Therefore this patch adds -ffast-math to the link flags of the pulseaudio
daemon. Linking with -ffast-math adds a procedure set_fast_math() to the
startup code of the daemon. On x86, the procedure sets bit 6 and 15 of the
mxcsr register. When these bits are set, denormals and results of
underflowing operations are truncated to 0.
For all our MMX/SSE code, we use a temporary channel variable, assigned
to the DI register, which is zero'ed as the very first operation in the
inline assembly code, before any other code is run.
With GCC 9.1, while using -O2, the DI register is also used for the
input operand. This is perfectly legal, but causes our code to become
incorrect because the output operand that is assigned to DI is not
explicitly marked as being clobbered before inputs are read.
This change fixes the problem by adding an earlyclobber annotation (&)
to the DI output argument.
We've added new API and updated an enum. A bunch of function parameters
have been marked as const, but this probably shouldn't count as a change
anyway.
The original code that was written was trying to detect what hypervisor
we were running under, rather than testing the presence bit first. We
don't really need the former, so let's use the more comprehensive latter
instead.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/684
If process_rewind() is called with nbytes = 0, process_rewind() will
nevertheless request a rewrite of the render memblockq.
This patch fixes the problem by adding the render memblockq length to the
rewrite amount only if nbytes > 0.
Currently the rewind logic for the source output is broken if the output
does not implement a process_rewind() callback. In that case, the read
index of the delay memblockq is rewound. This is wrong, because the data
that is going to be re-written was not yet read. Instead the write index
should be rewound and the read index left untouched. This is the reason
for the rewind glitches of monitor sources.
This crash occurs when PA is connected to a phone through the oFono
backend.
When disabling the Bluetooth adapter, pa_bluetooth_device is removed before
hf_audio_card. Both keep refs on pa_bluetooth_transport. Those removal will
call pa_bluetooth_transport_free() from device_free() (bluez5-util.c) and
hf_audio_card_free() (backend-ofono.c).
In the end, the call to pa_bluetooth_transport_free() calls
pa_hasmap_remove() through pa_bluetooth_transport_unlink(), but since
memory has already been freed, the second try results in a segfault.
Triggering hf_audio_card removal during pa_bluetooth_device removal allows
hf_audio_card to be freed at the right time.
setup_stream() crashes when calling set_nonblock() with an invalid
stream_fd.
On a new call, the ofono backend gets notified of a new connection.
The ofono backend sets the transport state to playing, and that triggers
a profile change, which sets up the stream for the first time.
Then module-bluetooth-policy sets up the loopbacks. The loopbacks get
fully initialized before the crash.
After module-bluetooth-policy has done its things, the execution
continues in the transport state change hook. The next hook user is
module-bluez5-device, whose handle_transport_state_change() function
gets called. It will then set up the stream again even though it's
already set up. I'm not sure if that's a some kind of a bug.
setup_stream() can handle the case where it's unnecessarily called,
though, so this second setup is not a big problem.
The crash happens, because the connection died due to POLLHUP in the IO
thread before the second setup_stream() call.
Without this, meson on Solaris exits with:
src/daemon/meson.build:138:15: ERROR: Unknown variable "systemd_dep".
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
The warnings:
modules/bluetooth/a2dp-codec-sbc.c: In function ‘default_bitpool’:
modules/bluetooth/a2dp-codec-sbc.c:161:13: warning: this statement may fall through [-Wimplicit-fallthrough=]
switch (mode) {
^~~~~~
modules/bluetooth/a2dp-codec-sbc.c:169:9: note: here
case SBC_SAMPLING_FREQ_44100:
^~~~
modules/bluetooth/a2dp-codec-sbc.c:170:13: warning: this statement may fall through [-Wimplicit-fallthrough=]
switch (mode) {
^~~~~~
modules/bluetooth/a2dp-codec-sbc.c:180:9: note: here
case SBC_SAMPLING_FREQ_48000:
^~~~
These were valid warnings in that an invalid channel mode would result
in unintended fallthroughs, but the end result would anyway been a crash
in the pa_assert_not_reached() at the end of the function, so
functionally there's no change.
The original atomic implementation in pulseaudio based on
libatomic stated that the intent was to use full memory barriers.
According to [1], the load and store implementation based on
gcc builtins matches sequential consistent (i.e. full memory barrier)
load and store ordering only for x86.
I observed random crashes in client applications using memfd srbchannel
transport on an armv8-aarch64 platform (cortex-a57).
In all those crashes the first read on the pstream descriptor
(the size field) was wrong and looked like it contained old data.
I boiled the relevant parts of the srbchannel implementation down to
a simple test case and could observe random test failures.
So I figured that the atomic implementation was broken for armv8
with respect to cross-cpu memory access ordering consistency.
In order to come up with a minimal fix, I used the newer
__atomic_load_n/__atomic_store_n builtins from gcc.
With
aarch64-linux-gnu-gcc (Linaro GCC 7.3-2018.05) 7.3.1 20180425
they compile to
ldar and stlxr on arm64, which is correct according to [1] and [2].
The other atomic operations based on __sync builtins don't need
to be touched since they already are of the full memory barrier
variety.
[1] https://www.cl.cam.ac.uk/~pes20/cpp/cpp0xmappings.html
[2] <https://community.arm.com/developer/ip-products/processors
/b/processors-ip-blog/posts/armv8-a-architecture-2016-additions>
The function calculates the correct timeout (in microseconds) to assign
in the `u` variable, but then assigns `m->prepared_timeout` the value
of the `timeout` argument (in milliseconds).
We met two problems recently, one happened on a Lenovo machine with
dual analogue codecs, the other happened on a Dell machine with
a digital mic directly connected to PCH. The two problems are
basically same, there is an internal mic and an external mic, the
internal mic always shows up in the gnome-control-center, the external
mic only shows up when it is plugged. After the external mic is
plugged and users select it from gnome-control-center, the
gnome-control-center will read all saved streams through extension_cb,
and bind the source of external mic to all streams, after that the
apps only record sound via the source of external mic, after the
external mic is unplugged, the internal mic will automatically be
selected since it is the only left input device in the
gnome-control-center, since users don't select it, all streams are
still bond the source of external mic. When users record sound via
apps, they can't record any sound even the default_source is the
source of internal mic and the internal mic is selected in the UI.
It is very common that a machine has internal mic and external mic,
but this problem didn't expose before, that is because both internal
mic and external mic belong to one source, but for those two
machines, the internal mic belongs to one source, while the external
mic belongs to another source (they are in differnt codecs or one is
in the codec and the other is from PCH),
To fix it with a mininal change, we just check if the active_port is
PA_AVAILABLE_NO or not when building a new stream, if it is, don't
restore the device to the new built stream, let pa_source_output_new()
decide the source device for this stream.
And we also do the same change to sink_input.
This change only affects the new built streams, it will not change
the database, so the users' preference is still saved in the database,
after the active_port is not PA_AVAILABLE_NO, the new streams will
still restore to the preferred device.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
This test relies on parsing the generated Makefile. A meson equivalent
requires to re-write all the parser.
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
For now esound is not supported with the meson build, although it
wouldn't be that hard to support it.
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
There's already a hook that modifies the search path when run from the
build tree.
if (pa_run_from_build_tree()) {
pa_log_notice("Detected that we are run from the build tree, fixing search path.");
#ifdef MESON_BUILD
c->dl_search_path = pa_xstrdup(PA_BUILDDIR PA_PATH_SEP "src" PA_PATH_SEP "modules");
#else
c->dl_search_path = pa_xstrdup(PA_BUILDDIR);
#endif
} else
I'm not sure how it works behind the hood, but by setting
--dl-search-path, we get errors in the logs when running `make
check-daemon`:
E: [pulseaudio][daemon/ltdl-bind-now.c:75 bind_now_open()] Failed to open module /home/arno/proj/pulse/src/pa.up/src/.libs/.libs/module-native-protocol-unix.so:
/home/arno/proj/pulse/src/pa.up/src/.libs/.libs/module-native-protocol-unix.so: cannot open shared object file: No such file or directory
I: [pulseaudio][pulsecore/module.c:197 pa_module_load()] Loaded "module-native-protocol-unix" (index: #3; argument: "").
So basically, PA tries two paths, fails the first time (obviously we can
see the path is not correct), then tries again with another path (where
does it gets it?) and succeeds. So there's no obvious error if you don't
look at the log.
This commit removes the useless `--dl-search-path`, which has the effect
to remove the errors in the logs.
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
It was omitted. This patch fixes unexpected behavior that avoid-
resampling does not work in some cases.
Signed-off-by: Sangchul Lee <sangchul1011@gmail.com>
Brings things in line with the autotools build, and adds ALSA mixer
paths and profile-sets into the meson build system as well.
The module installation path is also now customisable.
This flag results in calls to (at least) isfinite() and isnan() becoming
skipped, and a constant false returned. This caused volume-test to fail
on Debian: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=916504
Since PulseAudio deals with negative infinities with volume dB values,
this is not a problem only in volume-test. We shouldn't use -ffast-math
at all.
This patch adds a new feature to the core which allows to send messages
to objects. An object can register/unregister a message handler with
pa_message_handler_{register, unregister}() while a message can be sent
to the handler using the pa_message_handler_send_message() function.
A message has 4 arguments (apart from passing the core):
object_path: The path identifying the object that will receive the message
message: message command
message_parameters: A string containing additional parameters
response: Pointer to a response string that will be filled by the
message handler. The caller is responsible to free the string.
The patch is a precondition for the following patches that allow clients
to send messages to pulseaudio objects.
There is no restriction on object names, except that an object path
always starts with a "/". The intention is to use a path-like syntax,
for example /core/sink_1 for a sink or /name/instances/index for modules.
The exact naming convention still needs to be agreed.
The current code uses a pa_strbuf to construct the escaped string. This
will generate a linked list member for each character which may be very
inefficient.
This patch avoids the use of pa_strbuf by allocating a sufficiently large
string which can be filled with the output data.
These events were missing, because the
pa_core_update_default_sink/source() calls were assumed to send the
subscription events when necessary. Often that indeed is the case, but
if the current configured default sink doesn't exist, and then the
current default sink is set as the configured default sink, the
configured default sink changes but the default sink doesn't, and in
this case pa_core_update_default_sink() doesn't send the change event.
module-default-device-restore relies on getting a notification whenever
the configured default sink changes, and the missing event meant that
the files containing the configured sink and source weren't updated in
some cases.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/648
The recent change in ALSA upstream stripped -I$include/alsa path from
pkgconfig. We already fixed for this change in some places but still
the code for UCM was overlooked, and this resulted in the unresolved
symbols in alsa card module. Fix them as well.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Pulseaudio SBC codec defines that audio samples are in PA_SAMPLE_S16LE
format which is little endian. But libsbc library expects audio samples by
default in host endianity which is big endian on big endian system. So SBC
support on big endian system is broken. To fix this problem tell libsbc
library that audio samples are in little endian to match PA_SIMPLE_S16LE
sample format.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=91359
Remove dead code and replace numeric bitpool values by macro definitions.
Maximal bitpool value in fill_capabilities() was reduced from 64 to 53
(SBC_BITPOOL_HQ_JOINT_STEREO_44100) because default_bitpool() already set
maximal value to 53.
This patch does not change SBC behavior as maximal bitpool was already
limited to 53. So it is just clean up.
This patch introduce new modular API for bluetooth A2DP codecs. Its
benefits are:
* bluez5-util and module-bluez5-device does not contain any codec specific
code, they are codec independent.
* For adding new A2DP codec it is needed just to adjust one table in
a2dp-codec-util.c file. All codec specific functions are in separate
codec file.
* Support for backchannel (microphone voice). Some A2DP codecs (like
FastStream or aptX Low Latency) are bi-directional and can be used for
both music playback and audio call.
* Support for more configurations per codec. This allows to implement low
quality mode of some codec together with high quality.
Current SBC codec implementation was moved from bluez5-util and
module-bluez5-device to its own file and converted to this new A2DP API.
This adds API to allow clients to schedule a callback in the mainloop
thread without the mainloop lock being held. This is meant for a case
where the client might be dealing with locking its own objects in
addition to the mainloop thread itself. In this case, it might need ton
control the locking order of the two, to match the order in other
threads, as it might not always be able to allow for its objects to be
locked after the mainloop thread lock.
The current null-source implementation has several bugs:
1) The latency reported is the negative of the correct latency.
2) The memchunk passed to pa_source_post() is not initialized
with silence.
3) In PA_SOURCE_MESSAGE_SET_STATE the timestamp is always set
when the source transitions to RUNNING state. This should only
happen when the source transitions from SUSPENDED to RUNNING
but also if it changes from SUSPENDED to IDLE.
4) The timing of the thread function is incorrect. It always
uses u->latency_time, regardless of the specified source
latency.
5) The latency_time argument seems pointless because the source
is defined with dynamic latency.
This patch fixes the issues by
1) inverting the sign of the reported latency,
2) initializing the memchunk with silence,
3) changing the logic in PA_SOURCE_MESSAGE_SET_STATE so that
the timestamp is set when needed,
4) using u->block_usec instead of u->latency_time for setting
the rtpoll timer and checking if the timer has elapsed,
5) removing the latency_time option.
So far PulseAudio only supported two different work formats: S16NE if
it's sufficient to represent the input and output formats without loss
of precision and FLOAT32NE in all other cases. For systems that use
S32NE exclusively, this results in unnecessary conversions from S32NE to
FLOAT32NE and back again.
Add S32NE remap operations and make use of them (for the COPY and
TRIVIAL resamplers) if both input and output format are S32NE. This
avoids the back and forth conversions between S32NE and FLOAT32NE,
significantly improving performance for those cases.
pa_init_remap_func() takes care to initialise pa_remap_t.do_remap to
NULL before calling init_remap_func (the CPU-specific remap init
function) and invokes init_remap_c if init_remap_func did not set
pa_remap_t.do_remap to non-NULL. remap_init_test_channels() calls
init_remap_func() directly so it must make sure pa_remap_t.do_remap is
set to NULL. Otherwise we'll end up with a random value in
pa_remap_t.do_remap if there is no CPU-optimised remap function for the
current operation.
Consumers are expected to use <alsa/asoundlib.h> instead of
<asoundlib.h>.
This is in preparation of an change to pkgconfig(alsa) to
not pollute CFLAGS with -I/usr/include/alsa anymore.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
This is added to keep backward compatibility. The default value of
this new argument is false. Therefore, triggering by source-output
will be activated only if it is set to true explicitly.
Signed-off-by: Sangchul Lee <sc11.lee@samsung.com>
Previously, media.role property of only sink-input is used to
determine to trigger and apply ducking or cork to sink-inputs.
On the other hand, some use cases require that source-output
also need to trigger the effect to sink-inputs. Therefore this
patch adds logic to retrieve source-ouputs to find trigger role
by checking media.role property and apply ducking/cork to sink-
inputs that meet conditions.
Signed-off-by: Sangchul Lee <sc11.lee@samsung.com>
Currently, the ladspa-sink is not suspended when the master sink is suspended.
With this patch, the ladspa-sink will be suspended with suspend cause
PA_SUSPEND_UNAVAILABLE when the master sink is suspended for other reasons
than PA_SUSPEND_IDLE. This fixes issue #15.
Currently, virtual sinks and sources are not suspended when the master sink
or source is suspended. To implement this, the slave must be able to track
the suspend cause of the master.
With this patch, the sink input suspend callback will not only be called
when the sink or source is changing state, but also when the suspend cause
changes. Similar to the set_state_in_*_thread_cb() functions, the suspend
callback receives a state and a suspend cause as additional arguments.
Because the new state and suspend cause of the sink or source have already
been set, the old values are passed to the callback.
Currently, when a system is waking up from suspend, the resume process of the
ALSA sink and source is unstable. Sometimes the device needs to be restarted
multiple times and when the system was suspended between snd_pcm_mmap_begin()
and snd_pcm_mmap_commit(), pulseaudio crashes on resume.
Additionally, variables are not reset after the resume, so that sink/source
report wrong latencies.
This patch fixes the issues by closing and re-opening the PCM if recovery
from an error condition is not possible. Additionally, the variables are
reset, so that latencies are reported correctly.
After a suspend/resume cycle of a system, it may be possible that module-loopback
accumulates several seconds of audio in the memblockq before the alsa sink becomes
active again. Also it may be possible for other reasons that the actual loopback
latency is too different from the target latency to be adjusted in a reasonable
time by the normal rate controller.
This patch adds the option fast_adjust_threshold_msec to module-loopback. If set,
the latency will be forcefully adjusted to the target latency by dropping or
inserting samples if the actual latency differs more than fast_adjust_threshold_msec
from the target latency.
Also the calculation of the real adjust time would fail when the system was
suspended because that case was not considered. Now the real adjust time
calculation is skipped if the time passed between two calls of adjust_rates()
appears significantly too long.
pa_split_in_place() and pa_split_spaces_in_place() are modifed
to use size_t type instead of integer type.
alsa-ucm.c is revised according to this change.
Signed-off-by: Sangchul Lee <sc11.lee@samsung.com>
The old minimum version was set in commit 57e3ccaf51 based on what the
commit author happened to have installed at the time. Russell Treleaven
now confirmed that Debian 8's gettext version, 0.19.3, works fine too,
or at least PulseAudio builds without errors. There might be room to
lower the required version even further, but that requires someone to
test older gettext versions.
In a former commit 37358e42c4 ("alsa: Suppress udev detection of sound
card for some units on IEEE 1394 bus"), PulseAudio has udev rules to
suppress handling some units on IEEE 1394 bus for a below issue:
Bug 199365 - repeating bus resets on Firewire bus with Focusrite Saffaire 26/io
https://bugzilla.kernel.org/show_bug.cgi?id=199365
However, I found that the rules match another model; Focusrite Liquid
Saffire 56. For detail, refer to below patch for Linux sound subsystem:
[alsa-devel] [PATCH] ALSA: bebob: use more identical mod_alias for
Saffire Pro 10 I/O against Liquid Saffire 56
https://mailman.alsa-project.org/pipermail/alsa-devel/2019-February/146003.html
For PulseAudio, the udev rule should be improved, because Liquid Saffire 56
(an application of TCAT TCD2200 ASIC, a.k.a Dice Jr.) can be handled by
pulseaudio without the issue.
This commit changes udev rule with model name instead of model_id from
configuration ROM. Below is data on udevd for Liquid Saffire 56, for
your information:
$ udevadm info -q all -p /sys/bus/firewire/devices/fw1.0/sound/card2/
P: /devices/pci0000:00/0000:00:01.2/0000:03:00.2/0000:04:07.0/0000:0a:00.0/0000:0b:00.0/fw1/fw1.0/sound/card2
E: DEVPATH=/devices/pci0000:00/0000:00:01.2/0000:03:00.2/0000:04:07.0/0000:0a:00.0/0000:0b:00.0/fw1/fw1.0/sound/card2
E: ID_BUS=firewire
E: ID_FOR_SEAT=sound-pci-0000_0b_00_0
E: ID_ID=firewire-0x00130e04018001e9
E: ID_MODEL=LIQUID_SAFFIRE_56
E: ID_MODEL_FROM_DATABASE=XIO2213A/B/XIO2221 IEEE-1394b OHCI Controller [Cheetah Express]
E: ID_MODEL_ID=0x000006
E: ID_PATH=pci-0000:0b:00.0
E: ID_PATH_TAG=pci-0000_0b_00_0
E: ID_PCI_CLASS_FROM_DATABASE=Serial bus controller
E: ID_PCI_INTERFACE_FROM_DATABASE=OHCI
E: ID_PCI_SUBCLASS_FROM_DATABASE=FireWire (IEEE 1394)
E: ID_SERIAL=0x00130e04018001e9
E: ID_SERIAL_SHORT=0x00130e04018001e9
E: ID_VENDOR=Focusrite
E: ID_VENDOR_FROM_DATABASE=Texas Instruments
E: ID_VENDOR_ID=0x00130e
E: SOUND_INITIALIZED=1
E: SUBSYSTEM=sound
E: SYSTEMD_WANTS=sound.target
E: TAGS=💺systemd:
E: USEC_INITIALIZED=9802422583
Fixes: 37358e42c4 ("alsa: Suppress udev detection of sound card for some units on IEEE 1394 bus")
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Similar to module-tunnel-sink-new, module-virtual-source did not create
a rtpoll for the uplink sink. This lead to a crash when the uplink sink
was used by module loopback, because module-loopback relies on the sink
to provide a rtpoll. Additionally, the sink was not unlinked when the
module was unloaded.
This patch fixes both issues. The rtpoll created is never run by the sink,
so the patch is no real fix but just a workaround to make module-loopback
happy.
pa_card_profile_set_available needs to check if the card is linked
before firing PA_CORE_HOOK_CARD_PROFILE_AVAILABLE_CHANGED, so callbacks
connected to it receive a fully initialized card object.
This fixes a crash introduced by commit 30a551bbc
"switch-on-port-available: Check if we need to change the active
profile".
If one device tries to use PulseAudio to send audio over A2DP to another
device with bluez-alsa, that doesn't work because PulseAudio uses an
incorrect RTP payload type and bluez-alsa checks that the RTP payload
type is correct. According to the A2DP spec, the payload type should be
set to a number between 96 and 127.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/591
We split out some of the check-daemon tests that take a long time to
run, and also reduce how long we wait for the daemon to start up. This
should make the CI process quicker.
This allows us to disable automatically updating build system files in
case things change. This is desirable in the common case, but not
necessarily for CI, where we want the ability to take a build directory
as an artifact from one stage to the next (i.e. into a fresh checkout).
I can't promise that the logic is *exactly* the same as the logic
currently in use with the autotools, but it seems correct to me.
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
If `x11-xcb` is found, then let's force other X11 dependencies to be
there as well. That makes things a bit easier, and that's also what is
done in the autotools build system.
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
This is to avoid using the construct 'join_paths(prefix, get_option(...))'
everywhere in the meson files. It's better to settle the paths question
once and for all at the beginning.
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
Before this commit ucm_port_contains() was using a strncmp to compare
UCM-device-names without first checking that the part of the port_name
being compared and the device-name have the same length, this was causing
it to return true for both "InternalMic-IN1" and "InternalMic-IN12" when
port_name contained "InternalMic-IN1".
We hit this with the bytcr_rt5651 UCM profile which has "InternalMic-IN1",
"InternalMic-IN2" and "InternalMic-IN12" devices, for devices with their
internal mic connected to IN1, or IN2, or using stereo internal mics
connected to both. This problem resulted in various problems including
the RECMIXL? BST2 switch getting turned on when selecting only
"InternalMic-IN1", as well as confusing the gnome-control-center sound
panel, which could not figure out which device is selected in this case.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
This is because the meson build requires meson 0.47, which is not
available in the current Ubuntu LTS (18.04).
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
The previous commit introduces logic in module-switch-on-port-available
that may change a card's active profile when its availability changes to
PA_AVAILABLE_NO. To choose the new active profile, it needs a consistent
view of the new availability of all profiles, so this commit changes the
order which the ALSA driver updates all profiles' availability to ensure
the active profile is last.
This is not generic enough to cover cases were we may want to take an
action on availability changes of profiles other than the active one
that also need a consistent view of all profiles' availability. But we
don't have any callbacks implementing such action at the moment.
When a port becomes unavailble its profile may also become unavailable.
If that profile is the card's active profile, we need to switch the
card's active profile to a different one.
If we don't do that a card may get stuck on a profile without available
ports, but its sink and source will still exist, preventing
module-rescue-streams to move the streams to a different card with
available ports.
The relation between port availability and profile availability is
defined by the driver, and for the ALSA driver a profile is considered
available if there is at least one (available || unknown) port for each
direction implemented by the profile. Because of that we can only check
the profile's availability and priority when looking for the best
profile and don't need to look at port's priorities.
https://phabricator.endlessm.com/T24904
It is helpful to improve reproducibility build [1] since
PA_SRCDIR/PA_BUILDDIR contains build path,
--disable-running-from-build-tree could drop these macros at
precompilation.
[1] https://reproducible-builds.org/
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
HDMI ports are normally present on cards connected to an internal bus,
and module-switch-on-connect should switch to them when a HDMI monitor
is plugged.
This is specially relevant on setups where the HDMI port of a machine is
connected to a different audio card then the analog outputs, which is
the case for machines with AMD graphics cards.
When reviewing another change in rsa_encrypt(), Felipe Sateler pointed
out some deficiencies in error handling. This patch adds error handling
for all openssl calls in rsa_encrypt().
This patch doesn't propagate the error all the way up to the
pa_rtsp_client owner, because there's no mechanism for doing that. I
could implement such mechanism myself, but I think it's better I don't
make such complex changes to the RAOP code, because I don't have any
RAOP hardware to test the changes. The result is that module-raop-sink
will just sit around without doing anything. I think this is still
better than having no error handling at all.
Sample format(e.g. 16 bit, 24 bit) was not considered even if the
avoid-resampling option is set or the passthrough mode is used.
This patch checks both sample format and rate of a stream to
determine whether to avoid resampling in case of the option is set.
In other word, it is possble to use the stream's original sample
format and rate without resampling as long as these are supported
by the device.
pa_sink_input_update_rate() and pa_source_output_update_rate() are
renamed to pa_sink_input_update_resampler() and pa_source_output
_update_resampler() respectively.
functions are added as below.
pa_sink_set_sample_format(), pa_sink_set_sample_rate(),
pa_source_set_sample_format(), pa_source_set_sample_rate()
Signed-off-by: Sangchul Lee <sc11.lee@samsung.com>
mpv and vlc play "normal" 5.1 AC3 and DTS files as if they had a
"5.1 (Side)" layout. Which is fine and consistent with ITU
recommendations if the user has a proper 7.1 system. But if the user
actually has a 5.1 system, PulseAudio will try to remap, poorly, between
the "5.1 (Side)" and "5.1" layouts, sending either an average between
front and rear channels, or an attenuated version of that average,
depending on the remixing-use-all-sink-channels setting.
This is not desired, the "Side" channels should be sent to "Rear", it is
only an unfortunate nomenclature confusion.
This patch does not fix 5.1 <-> 7.1 remixing.
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
Headphones should have higher priority than lineout. Many people have
speakers always connected to lineout, and when plugging in headphones,
the audio should move to the headphones, which requires headphones
to have higher priority than lineout.
Previously this was handled by marking lineout unavailable when plugging
in headphones, but we don't do that any more.
This reverts commit 66f97c35bd. The commit
message was:
alsa-mixer: Disable line-out if headphone jack is plugged
Line-out gets muted when headphones are plugged in on HDA cards, encode
this in the line-out path so pulse can match that state.
I don't think the mentioned auto-muting happens any more by default,
and some users want to use lineout while having headphones plugged in.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/583
The bool was inverted for some reason - maybe because the next line
prints enable-remixing that needs to be inverted from disable_remixing,
and somehow this logic was accidentally copied to the avoid-resampling
handling.
Fixes: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/568
This adds some basic infrastructure to test passthrough support. Right
now, it just creates a passthrough stream and makes sure negotiation
works. We'll add in more tests as we go along.
This sync the meson version detection to match what we do in the
autotools build, which is to use git-version-gen, which in turn looks
for <topdir>/.tarball-version, or generates the version based on a git
tag or environment variable.
webrtc.cc:202:19: warning: comparison of integer expressions of different signedness:
'int' and 'std::vector<webrtc::CartesianPoint<float> >::size_type' {aka 'long unsigned int'}
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
Regarding the module:
This is unlike the autotools where liboss-util is built as a library,
here we build everything in the oss module, as apparently there's no
other consumer for liboss-util.
Regarding padsp:
Setting the install mode for padsp requires meson 0.47, so instead we
set padsp.in as executable in the git repository (which is what glib
does for gdbus-codegen btw).
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
Please notice that the bluez5 version seems wrong in the dependency
declaration: `>= 4.x`, while we're talking about version 5.
The ofono part will need to be made optional when we start to work on
the meson_options file.
I follow the current configure.ac to define 'HAVE_BLUEZ', but it looks
like this part would benefit from a bit of rework. Setting HAVE_BLUEZ
when we have dbus+sbc sounds weird, there's probably a better name for
this variable.
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
This is unlike the autotools where we check that a header exist, here we
use pkgconfig because upstream ships a pkgconfig. I don't know from
which version though...
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
This flag will make the loader fail if symbols are not resolved. It
seems to be our best bet to uncover every missing module dependencies.
For more details, I recommend to read:
<http://www.kaizou.org/2015/01/linux-libraries/>
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
This is based on the initial protocol_native library that is already
defined, and then by looking at the Makefile.am to work out the
dependencies.
It's not clear whether we really need database_c_args, maybe there's
things that can be simplified.
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
This is to be consistent. In pa currently, as built by the autotools,
libalsa-util is a shared library. Moreover, all the libraries for the
modules, as defined in `src/meson.build`, are also shared libraries.
So let's stick to shared libraries everywhere for now, for simplicity.
We can rework that later on.
Signed-off-by: Arnaud Rebillout <arnaud.rebillout@collabora.com>
Split a big conditional into separate checks and use pa_safe_streq
instead of checking if a pointer is valid and calling pa_streq inside a
conditional.
This adds a Dockerfile to generate a Docker image with the required
dependencies on top of the standard Ubuntu 18.04 image. The Gitlab CI
then runs the PulseAudio build within this image.
A bug was filed to bugzilla.kernel.org for a quirk of some models which
ALSA BeBoB driver supports.
Bug 199365 - repeating bus resets on Firewire bus with Focusrite Saffaire 26/io
https://bugzilla.kernel.org/show_bug.cgi?id=199365
Some models (two models as long as I know) have a quirk to disappear from
IEEE 1394 bus at disconnections of packet streaming. Corresponding
character devices are removed according to 'remove' callbacks of relevant
drivers from Linux dd core. Then the models re-appear on the bus by
generating bus resets and corresponding character devices are added
according to 'probe' callbacks from Linux dd core.
In a view of ALSA applications, this looks that plug-out/plug-in occur in
a sequential order for the models when they stop playback/capture substream.
For most applications, this doesn't cause large issue. However, this quirk
is not good for combination of below modules in PulseAudio. PulseAudio
enters endless loop to detect the models and start/stop PCM substream.
- module-udev-detect
- module-alsa-card
- module-suspend-on-idle
In detail, please read my comment no.6:
https://bugzilla.kernel.org/show_bug.cgi?id=199365#c6
This commit suppressed udev detection of sound card for the issued models.
For the models, 'PULSE_IGNORE' flag is added to udev rules, then
module-udev-detect don't handle the models and PulseAudio never uses the
models automatically. In a scenario for users to load
module-alsa-card/module-alsa-sink/module-alsa-source by hand, although
these modules can still stop PCM substreams with module-suspend-on-idle,
PulseAudio never enters the endless loop because udev detection doesn't
work for the models. In this case, as long as special files for ALSA
character devices for these models are the same, corresponding sinks and
sources are available even if the voluntary plug-out/plug-in occur.
(Focusrite Saffire Pro 10 i/o with systemd 237)
$ udevadm info -q all -p /devices/pci0000:00/0000:00:07.0/fw1/fw1.0/sound/card1
P: /devices/pci0000:00/0000:00:07.0/fw1/fw1.0/sound/card1
E: DEVPATH=/devices/pci0000:00/0000:00:07.0/fw1/fw1.0/sound/card1
E: ID_BUS=firewire
E: ID_FOR_SEAT=sound-pci-0000_00_07_0
E: ID_ID=firewire-0x00130e01000606e0
E: ID_MODEL=Pro10IO
E: ID_MODEL_FROM_DATABASE=XIO2213A/B/XIO2221 IEEE-1394b OHCI Controller [Cheetah Express]
E: ID_MODEL_ID=0x000006
E: ID_PATH=pci-0000:00:07.0
E: ID_PATH_TAG=pci-0000_00_07_0
E: ID_PCI_CLASS_FROM_DATABASE=Serial bus controller
E: ID_PCI_INTERFACE_FROM_DATABASE=OHCI
E: ID_PCI_SUBCLASS_FROM_DATABASE=FireWire (IEEE 1394)
E: ID_SERIAL=0x00130e01000606e0
E: ID_SERIAL_SHORT=0x00130e01000606e0
E: ID_VENDOR=Focusrite
E: ID_VENDOR_FROM_DATABASE=Texas Instruments
E: ID_VENDOR_ID=0x00130e
E: SOUND_INITIALIZED=1
E: SUBSYSTEM=sound
E: SYSTEMD_WANTS=sound.target
E: TAGS=:systemd:seat:
E: USEC_INITIALIZED=957089064
(Focusrite Saffire Pro 26 i/o with systemd 237)
$ udevadm info -q all -p /devices/pci0000:00/0000:00:07.0/fw1/fw1.0/sound/card1
P: /devices/pci0000:00/0000:00:07.0/fw1/fw1.0/sound/card1
E: DEVPATH=/devices/pci0000:00/0000:00:07.0/fw1/fw1.0/sound/card1
E: ID_BUS=firewire
E: ID_FOR_SEAT=sound-pci-0000_00_07_0
E: ID_ID=firewire-0x00130e0100030cdd
E: ID_MODEL=Pro26IO
E: ID_MODEL_FROM_DATABASE=XIO2213A/B/XIO2221 IEEE-1394b OHCI Controller [Cheetah Express]
E: ID_MODEL_ID=0x000003
E: ID_PATH=pci-0000:00:07.0
E: ID_PATH_TAG=pci-0000_00_07_0
E: ID_PCI_CLASS_FROM_DATABASE=Serial bus controller
E: ID_PCI_INTERFACE_FROM_DATABASE=OHCI
E: ID_PCI_SUBCLASS_FROM_DATABASE=FireWire (IEEE 1394)
E: ID_SERIAL=0x00130e0100030cdd
E: ID_SERIAL_SHORT=0x00130e0100030cdd
E: ID_VENDOR=Focusrite
E: ID_VENDOR_FROM_DATABASE=Texas Instruments
E: ID_VENDOR_ID=0x00130e
E: SOUND_INITIALIZED=1
E: SUBSYSTEM=sound
E: SYSTEMD_WANTS=sound.target
E: TAGS=:systemd:seat:
E: USEC_INITIALIZED=1071026684
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
I looked for outdated links related to the GitLab migration. These are
the only ones I found. There were also some links to various bug
reports in the old Bugzilla, but those don't really need updating, since
Bugzilla should stay readable for a long time.
We move over helper functions to get rate, channels, channel map and
sample format (if PCM) in the public API, so users of the extended API
are more easily able to pull out these values from pa_format_info.
We can provide a better overall user experience with Bluetooth cards by
always choosing the higher audio quality profile (A2DP) by default and
updating the profile selection dynamically according to which streams
are active at a certain moment. The default initial selection has been
addressed by "85daab272 bluetooth: set better priorities for profiles"
and the dynamic profile selection is covered by module-bluetooth-policy.
In addition, module-card-restore's database entries for Bluetooth devices
are retained after a device is removed from the system, leading to the
previously selected profile being restored after a new pairing with the
same device, with no way for the user to erase this memory and reset the
default profile except manually fiddling with module-card-restore's
database.
This commit adds a module argument to have module-card-restore ignore
Bluetooth profiles and this behavior is set as default.
The internal operation_set_state function already returns early if the
new state is the same as the existing state. The attached patch extends
this to return early if already in a finalised (done/cancelled) state,
i.e. blocks attempts to re-finalise into a different state.
This helps avoid unlinking more than once (or crashing on ref count
assertion).
I was not certain whether an assertion would be a better alternative -
with such a crash helping highlight usage problems...
The situation that lead to this was the thought of someone stupidly
trying to pa_operation_cancel() a callback within the callback
execution itself, while designing a solution for a memory leak related
to cancellation within my Rust binding. While no-one should do such a
thing, if they did, they'd either trip up a ref count assertion, or the
operation would be unlinked twice, which would be bad. It's a simple
thing to catch and mitigate, and could prove to be a useful
bulletproofing measure for this function in general.
We recently changed the umask of the daemon from 022 to 077, which broke
module-pipe-sink in the system mode, because nobody was allowed to read
from the pipe.
module-pipe-source in the system mode was probably always broken,
because the old umask of 022 should prevent anyone from writing to the
pipe.
This patch uses chmod() after the file creation to set the permissions
to 0666, which is what the fkfifo() call tried to set.
Bug link: https://bugs.freedesktop.org/show_bug.cgi?id=107070
Having a single level macro for stringizing LADSPA_PATH doesn't work,
because the '#' preprocessor operator doesn't expand any macros in its
parameter. As a result, we used the string "LADSPA_PATH" as the search
path, and obviously no plugins were ever found.
This adds a two-level macro in macro.h and uses that to expand and
stringize LADSPA_PATH.
Bug link: https://bugs.freedesktop.org/show_bug.cgi?id=107078
Add configuration option 'stream_name' for stream/session name so user
will see it on receiver side as RTP Strean ($stream_name)
ex: load-module module-rtp-send source=rtp.monitor stream_name=MyServerMedia
There has been a function to get supported sample rates from alsa and
an array for it in userdata of each module-alsa-sink/source. Similarly,
this patch adds a function to get supported sample formats(bit depth)
from alsa and an array for it to each userdata of the modules.
Signed-off-by: Sangchul Lee <sc11.lee@samsung.com>
pa_sink_get_state() and pa_source_get_state() just return the state
variable. We can as well access the state variable directly.
There are no behaviour changes, except that module-virtual-source
accessed the main thread's sink state variable from its push() callback.
I fixed the module so that it uses the thread_info.state variable
instead. Also, the compiler started to complain about comparing a sink
state variable to a source state enum value in protocol-esound.c. The
underlying bug was that a source pointer was assigned to a variable
whose type was a sink pointer (somehow using the pa_source_get_state()
macro confused the compiler enough so that it didn't complain before).
I fixed the variable type.
pa_sink_input_get_state() and pa_source_output_get_state() just return
the state variable. We can as well access the state variable directly.
There are no behaviour changes, except that some filter sources accessed
the main thread's state variable from their push() callbacks. I fixed
them so that they use the thread_info.state variable instead.
The only thing that the drained state was being used for was "pacmd
list-sink-inputs". In all other cases the drained and running states
were treated as equivalent. IMHO, this usage doesn't justify the
complexity that the additional state brings.
This patch was inspired by a bug report[1] that pointed out an error in
an if condition in pa_sink_input_set_state_within_thread(). The buggy
code is now removed altogether.
[1] https://bugs.freedesktop.org/show_bug.cgi?id=106982
When the user manually switches the profile of a bluetooth headset from
"off" to "a2dp_sink", the port availability changes from "unknown" to
"yes", which triggered a recursive profile change in
module-switch-on-port-available. Such recursivity isn't (and possibly
can't) be handled well (that is, PulseAudio crashed), so let's avoid
doing bluetooth profile changes from module-switch-on-port-available
(they're useless anyway).
Bug link: https://bugs.freedesktop.org/show_bug.cgi?id=107044
Now that both backend-native and backend-ofono can coexist and
backend-ofono is always loaded, even on systems without oFono, failing
to register with org.ofono is not necessarily an error.
This lowers the failure message log level from error to info.
This allows constifying public API functions that report their errors
via the context error but don't modify the context in any other way.
Philosophical arguments could be made why this is wrong, but I believe
in practice this is a net positive change.
Paves the way towards more of the API using const pointers.
Some pa_context_* functions return their errors by setting the context
error, even when there's no other change in the context state. This
prevented constifying the pa_context arguments of such functions. This
patch puts the error in its own struct behind a pointer, so that setting
the error doesn't any more count as modifying the pa_context object.
A bit hacky approach, but it allows to preserve LFE output position
even in reduced output modes 2.1 and 4.1.
Signed-off-by: Nazar Mokrynskyi <nazar@mokrynskyi.com>
Current code does not check whether pa_play_file call failed. Hence no error is
reported in the cli interface if playback failed because e.g. file isn't
readable by the daemon.
If a sound card doesn't have the "front" device defined for it, we have
to use the "hw" device for stereo. Not so long ago, the analog-stereo
mapping had "hw:%f" in its device-strings and everything worked great,
except that it caused trouble with the Intel HDMI LPE driver that uses
the first "hw" device for HDMI, and we were incorrectly detecting it as
an analog device. That problem was fixed in commit ea3ebd09, which
removed "hw:%f" from analog-stereo and added a new stereo fallback
mapping for "hw".
Now the problem is that if a sound card doesn't have the "front" device
defined for it, and it supports both mono and stereo, only the mono
mapping is used, because the stereo mapping is only a fallback. This
patch makes the mono mapping a fallback too, so the mono mapping is used
only if there's absolutely nothing else that works.
This can cause trouble at least in theory. Maybe someone actually wants
to use mono output on a card that supports both mono and stereo. But
that seems quite unlikely.
If the given proplist is NULL, the function creates a new (empty)
proplist. That caused a compiler warning after the constification, which
is why the new proplist is now created using a separate variable.
Existing documentation was unclear about which property list would be the
one changed (merged into), making it seem (along with the non-const
proplist pointer param, which needs changing seperately), that the proplist
object for which a pointer is given will be the one merged into, instead of
the internal cached entry's proplist.
Previously, the "avoid-resampling" option of daemon.conf is to make the
daemon try to use the stream sample rate if possible for all sinks or
sources.
This patch applies this option to module-udev-detect and module-alsa-card
as a module argument in order to override the default value of daemon.conf.
As a result, user can use this argument for more fine-grained control.
e.g.) set it false in daemon.conf and set it true for module-udev-detect
or a particular module-alsa-card in default.pa.(or vice versa)
To set it, use "avoid_resampling=true or false" as the module argument.
Signed-off-by: Sangchul Lee <sc11.lee@samsung.com>
There are only stereo and 5.1 output modes supported natively on this
sound card, but with this config more modes like 2.1, 4.0, 4.1 and 5.0
are now exposed. Also profiles list is cleaner now with all profiles
explicitly specified.
Last thing is removed support for microphone on Linux kernels older than
4.3-rc1, which shouldn't be an issue with future version of PulseAudio
likely be installed on newer kernels anyway.
Signed-off-by: Nazar Mokrynskyi <nazar@mokrynskyi.com>
This should make it easier for clients to elevate their audio threads to
real time priority without having to dig through much through specific
system internals.
BlueZ 4 is no longer supported by BlueZ community for a long long time,
also by moving to BlueZ 5 it should make it even more clearer that
BlueZ 4 is no longer an option.
Attempt to use Acquire method if available since it directly returns
the fd in the reply or an error if that the connection could not be
created while Connect offer neither of these and depend on
NewConnection to deliver the fd.
When a new card shows up (during pulseaudio startup or hotplugged),
pulseaudio needs to pick the initial profile for the card. Unavailable
profiles shouldn't be picked, but module-alsa-card sometimes marked
unavailable profiles as available, causing bad initial profile choices.
This patch changes module-alsa-card so that it marks all profiles
unavailable whose all output ports or all input ports are unavailable.
Previously only those profiles were marked as unavailable whose all
ports were unavailable. For example, if a profile contains one sink and
one source, and the sink is unavailable and the source is available,
previously such profile was marked as available, but now it's marked as
unavailable.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=102902
Currently the loopback module uses sample spec and channel map of the
sink by default. It leads to double resample if source and sink sample
specs are different and no rate/format specified in arguments. This
patch causes the source sample spec and channel map to be used by
default.
set_nonblock() will always set the file descriptor to non-blocking,
regardless of the nonblock argument.
This patch fixes the issue by passing the correct argument to the
fcntl() call. The bug had no impact because there is only one caller
of pa_make_fd_block() in poll-win32.c
This is a working implementation of a build with meson. The server,
utils, and most modules build with this, and it is possible to run from
a build tree and play/capture audio on ALSA devices.
There are a number of FIXMEs, of course, and a number of features that
need to be enabled (modules, dependencies, installation, etc.), but this
should provide everything we need to get there relatively quickly.
To use this, install meson (distro package, or mesonbuild.com) and run:
$ cd <pulseaudio src dir>
$ meson <builddir>
$ ninja -C <builddir>
The iec958 output uses device 2 and the iec958 input uses device 0. The
USB configuration in alsa doesn't set up the device numbers correctly,
which is why we need custom configuration in PulseAudio. Ideally this
would be fixed in alsa, but trying to get help for that wasn't
successful.
The volume_map variable was initialized only for PCM streams, but the
variable was passed to pa_cvolume_remap() also for non-PCM streams. The
volume remapping is never necessary for passthrough streams (PCM or
not), because no volume will be applied anyway, so let's skip the
pa_cvolume_remap() call for all passthrough streams.
jack->melem can be null if the jack disappears between probing the card
and the init_jacks() call. I don't know if jacks actually ever disappear
like that (seems unlikely), but this check is in any case needed as long
as init_jacks() has proper handling for the jack disappearing case
(rather than just an assert).
There was a crash report[1] that indicated that card_suspend_changed()
called report_jack_state() with a null melem. I don't know if that was
because the jack actually disappeared, or is there some other bug too.
[1] https://bugs.freedesktop.org/show_bug.cgi?id=104385
The commit "alsa-util: Set ALSA report_delay flag in pa_alsa_safe_delay()"
broke the build on ALSA versions below 1.1.0 because the time stamp
configuration function was introduced in 1.1.0.
This patch makes the usage of snd_pcm_status_set_audio_htstamp_config()
dependent on ALSA version.
dist_gsettingsdataconvert_DATA was set only if GSettings was enabled. If
the developer that generates the tarball doesn't have GSettings enabled,
pulseaudio.convert wouldn't get included in the tarball.
The schema file was not being added to the tarball even if GSettings was
enabled.
The current code does not call snd_pcm_status_set_audio_htstamp_config()
to configure the way timestamps are updated in ALSA. In kernel 4.14 and
above a bug in ALSA has been fixed which changes timmestamp behavior.
This leads to inconsistencies in the delay reporting because the time
stamp no longer reflects the time when the delay was updated if the
ALSA report_delay flag is not set. Therefore latencies are not calculated
correctly.
This patch uses snd_pcm_status_set_audio_htstamp_config() to set the
ALSA report_delay flag to 1 before the call to snd_pcm_status(). With
this, time stamps are updated as expected.
sco_process_render does not unref the memblock when it encounters an error.
This patch fixes the issue. It also changes the return value to 1 in the case
of EAGAIN. Because the data was already rendered and cannot be re-sent, we
have to discard the block.
Because the modified EAGAIN handling prevents the log message about EAGAIN
after POLLOUT from being printed, the log message was moved to
a2dp/sco_process_render().
The rewrite of the thread function does not change functionality much,
most of it is only cleanup, minor bug fixing and documentation work.
This patch also changes the send buffer size for a2dp sink to avoid lags
after temporary connection drops, following the proof-of-concept patch
posted by Dmitry Kalyanov.
Bug-Link: https://bugs.freedesktop.org/show_bug.cgi?id=58746
Additionally the patch changes the fixed latency for HSP playback from 125
to 25 ms. Tests showed that this produces better audio sync, which is
expected as HSP should have smaller latency than A2DP.
module-allow-passthrough has a (necessary) hack to replicate the default
sink selection and format negotiation from sink-input.c. One thing that
got missed in this replication is the possibility that the sink input is
not compatible with the default sink. When this happen, we now exit
gracefully.
Reconfiguration callback should also be set in case of avoiding resampling
option. This patch set the callback for every case because the callback
has already conditions to leave if it is not needed.
Also unnecessary codes of setting alternate sample rate to 0 are removed.
Signed-off-by: Sangchul Lee <sc11.lee@samsung.com>
This was originally planned to be done by paprefs when it starts, but
since the schema is now fully controlled by pulseaudio, it makes sense
to run the conversion from pulseaudio instead.
A new paprefs release is expected soon, and it will only support
GSettings. In order to have the default configuration work with the new
paprefs version, we need to enable GSettings by default.
If both module-gconf and module-gsettings are enabled when building
PulseAudio, both modules will be part of the default configuration. That
can cause trouble, because when the GConf database is migrated to
GSettings, the old configuration in GConf is not removed, so both
module-gconf and module-gsettings will try to load modules.
Generally it's not necessary to have both modules enabled even at build
time, so let's default to having only one of them enabled.
g_settings_get_child() returns a new GSettings object that needs to be
freed when it's not used any more. This patch collects all the childern
to a GPtrArray and frees them at the end of main(). They can't be freed
earlier, because that would prevent the "changed" signals from being
delivered.
The removed g_signal_connect() call didn't make sense. The callback
expects to be called when individual module groups are changed, not when
the top level object is changed. Also, module_group_callback() expects
user_data to be non-NULL, but here it was set to NULL.
According to the documentation of g_settings_list_children(), the listed
children may be removed at any time, so g_settings_get_child() may
return NULL. This is probably very unlikely to happen in practice, but
it's good to check anyway.
It is confusing if there's a thing named "module" which defines up to 10
modules to load. Calling the thing a "module group" instead should make
it easier to understand.
Originally the idea was to provide the "modules" schema with paprefs,
but since module-gsettings refers to the "modules" schema in its code,
that would make module-gsettings depend on paprefs, which is not good.
Now all schemas are provided by module-gsettings, so the paprefs
dependency is avoided. Unfortunately this means that if paprefs is
modified to load some new modules, the schema in pulseaudio needs to be
updated as well.
This also makes the module-gconf section conditional on HAVE_GCONF,
because if only gsettings support is built, the gconf section in the
configuration file would be redundant and confusing.
GConf is deprecated, and distributions are removing it. paprefs depends
on GConf, so in order to avoid paprefs getting removed as well, paprefs
has to be changed to use something else than GConf. GSettings is the
easiest alternative to migrate to, although it has the same problems
that GConf had: no support for system mode or networking.
This patch takes the non-GConf specific code from module-gconf and puts
it in stdin-util.[ch], which is then reused by module-gsettings.
module-gsettings is designed to be very similar to module-gconf.
Migration is expected to happen as follows: Distributions update
PulseAudio and paprefs at the same time, or first PulseAudio and then
paprefs. paprefs depends on module-gsettings, and module-gsettings
conflicts with module-gconf. Therefore module-gconf gets automatically
removed during the paprefs update. After the update an old PulseAudio is
likely to be running with module-gconf loaded. If the user tries to use
paprefs during this period, whatever the user does in paprefs won't have
any effect until PulseAudio is restarted (probably by a reboot or
relogin). This is not ideal, but will have to do.
When module-gsettings is loaded, it runs gsettings-data-convert
(implemented in a later patch). That will copy the settings from GConf
to GSettings. If gsettings-data-convert is not available (it's part of
GConf, so it may have already been uninstalled), then any previous
paprefs settings are lost.
The suspend-sink and suspend-source documentation for pacmd was quite
terse, so I copied the more complete documentation from pactl. I
couldn't resist doing some other minor edits along the way too.
Bug-link: https://bugs.freedesktop.org/show_bug.cgi?id=105907
The alsa sink calls pa_sink_suspend() from the set_port() callback.
pa_sink_suspend() can only be called from the main thread, but the
set_port() callback was often called from the IO thread. That caused an
assertion to be hit in pa_sink_suspend() when switching ports.
Another issue was that pa_sink_suspend() called the set_port() callback,
and if the callback calls pa_sink_suspend() again recursively, nothing
good can be expected from that, so the thread mismatch was not the only
problem.
This patch moves the mixer syncing logic out of pa_sink/source_suspend()
to be handled internally by the alsa sink/source. This removes the
recursive pa_sink_suspend() call. This also removes the need to have the
mixer_dirty flag in pa_sink/source, so the flag and the
pa_sink/source_set_mixer_dirty() functions can be removed.
This patch also changes the threading rules of set_port(). Previously it
was called sometimes from the main thread and sometimes from the IO
thread. Now it's always called from the main thread. When deferred
volumes are used, the alsa sink and source still have to update the
mixer from the IO thread when switching ports, but the thread
synchronization is now handled internally by the alsa sink and source.
The SET_PORT messages are not needed any more and can be removed.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=104761
The suspend cause isn't yet used by any of the callbacks. The alsa sink
and source will use it to sync the mixer when the SESSION suspend cause
is removed. Currently the syncing is done in pa_sink/source_suspend(),
and I want to change that, because pa_sink/source_suspend() shouldn't
have any alsa specific code.
There are no behaviour changes, the code from almost all the SET_STATE
handlers is moved with minimal changes to the newly introduced
set_state_in_io_thread() callback. The only exception is module-tunnel,
which has to call pa_sink_render() after pa_sink.thread_info.state has
been updated. The set_state_in_io_thread() callback is called before
updating that variable, so moving the SET_STATE handler code to the
callback isn't possible.
The purpose of this change is to make it easier to get state change
handling right in modules. Hooking to the SET_STATE messages in modules
required care in calling pa_sink/source_process_msg() at the right time
(or not calling it at all, as was the case on resume failures), and
there were a few bugs (fixed before this patch). Now the core takes care
of ordering things correctly.
Another motivation for this change is that there was some talk about
adding a suspend_cause variable to pa_sink/source.thread_info. The
variable would be updated in the core SET_STATE handler, but that would
not work with the old design, because in case of resume failures modules
didn't call the core message handler.
The previous code made the SET_STATE message fail if trigger() failed.
However, trigger() was called after pa_sink/source_process_msg(), which
meant that the main thread that sent the SET_STATE thought that resuming
failed, but nothing was undone in the IO thread, so in the IO thread
things seemed as if the sink/source was successfully resumed. (I don't
use OSS myself, so I don't know what kind of practical problems this
could cause).
Unless some complex undo logic is implemented, I believe it's best to
ignore all failures in trigger(). Most error cases were already ignored,
and the only one that wasn't ignored doesn't seem too serious.
I also moved trigger() to happen before pa_sink/source_process_msg(),
which made it necessary to add new state parameters to trigger(). The
reason for this move is that I want to move the SET_STATE handler code
into a separate callback, and if things are done both before and after
pa_sink/source_process_msg(), that makes things more complicated.
The previous code checked the return value of
pa_sink/source_process_msg() before calling trigger(), but that was
unnecessary, since pa_sink/source_process_msg() never fails when
processing the SET_STATE messages.
When resuming a sink or source, pa_sink/source_process_msg() should be
called only if resuming is successful. pa_sink/source_process_msg()
updates thread_info.state and notifies streams about the new state, but
if resuming fails, there's no state change.
pa_sink_get_state() is supposed to be used from the main thread. In this
case it doesn't really matter, because the SET_STATE handler is executed
while the main thread is waiting, but since the state is available also
in thread_info, let's use that. All other modules use thread_info.state
too, so at least this change improves consistency.
Also, we can use the PA_SINK_IS_OPENED macro to simplify the code a bit.
build_pollfd() isn't likely to fail, but if it does, pa_sink/source_put()
will crash on an assertion failure. I haven't seen such crash happening,
this is just something that I noticed while studying the state change
code.
When the sink is unlinked, there's no need to update the monitor suspend
state. In fact, trying to do that causes an assertion failure, because
pa_source_sync_suspend() wasn't written to handle the case where the
sink is unlinked.
The suspend cause isn't yet used by any of the callbacks. The alsa sink
and source will use it to sync the mixer when the SESSION suspend cause
is removed. Currently the syncing is done in pa_sink/source_suspend(),
and I want to change that, because pa_sink/source_suspend() shouldn't
have any alsa specific code.
This adds a pa_suspend_cause_t parameter to the sink/source_set_state()
functions, and moves part of the work that pa_sink/source_suspend() does
to sink/source_set_state(). The reason for this code shuffling is that I
plan to make all suspend cause changes available to modules through the
state change callbacks. This is the first step towards that.
Additionally, pa_source_sync_suspend() is changed to also update the
suspend cause of the monitor source when the suspend cause of the
monitored sink changes. That probably doesn't have much effect on
anything, but I think it makes sense to mirror the sink suspend cause in
the monitor source.
pa_source_sync_suspend() has also a bug fix: previously it was probably
possible that a sink might get suspended while in the passthrough mode.
When the sink then resumed (while still in the passthrough mode),
pa_source_sync_suspend() would resume also the monitor source, even
though the monitor source should be kept suspended when the sink is in
the passthrough mode. Now the monitor source won't be resumed in this
situation.
This removes the need to hardcode the ELD device index in the path
configuration. The hardcoded values don't work with the Intel HDMI LPE
driver.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=100488
We have so far assumed that HDMI always uses device indexes 3, 7, 8, 9,
10, 11, 12 and 13. These values are hardcoded in the path configuration.
The Intel HDMI LPE driver, however, uses different device numbering
scheme. Since the indexes aren't always the same, we need to query the
hw device index from ALSA.
Later patches will use the queried index for HDMI jack detection and ELD
information reading.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=100488
When module-filter-apply tries to find a matching source-output for
a given sink-input and a stream within the same group exists on the
monitor source of the filter, module-filter apply falsely assumes
that the source belongs to another instance of the filter and tries
to access source->output_from_master->source, which leads to a
segmentation fault.
This patch fixes the issue by ignoring the stream if the source is
the monitor source of the filter.
glibc 2.27 is to be released soon, and it will provide memfd_create().
If glibc provides the function, we must not define it ourselves,
otherwise building fails due to conflict between the two implementations
of the same function.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=104733
Usually PulseAudio is built with a linker that supports the -z,now
option, and that option should have the same effect (i.e. the dynamic
linker resolves all symbols when the program is started) as re-execing
with the LD_BIND_NOW environment variable set, so usually the re-execing
is redundant.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=104789
It was reported that PulseAudio causes error messages in syslog from
dbus-daemon:
Jan 14 04:51:32 gentoo dbus-daemon[2492]: [system] Rejected send message, 2 matched rules; type="error", sender=":1.15" (uid=1000 pid=2864 comm="/usr/bin/pulseaudio --start --log-target=syslog ") interface="(unset)" member="(unset)" error name="org.bluez.MediaEndpoint1.Error.NotImplemented" requested_reply="0" destination=":1.1" (uid=0 pid=2670 comm="/usr/libexec/bluetooth/bluetoothd ")
The default policy on the system bus is to not let any method call
replies through if they have not been requested, and apparently
bluetoothd doesn't want replies to the Release() call.
This also changes the reply type from error to normal reply. The "not
implemented" error didn't make sense to me. We don't do any cleanup in
the Release() handler, probably because there's nothing to do. If there
is some cleanup that we should do, then it's a serious bug not to do it.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=104646
Allow usage of an already existing fifo (named pipe) within the
pipe-sink module. Also, the used fifo is going to be removed upon
module unload only if the fifo is created by the same module.
This is based on a patch by Rolo <rolo@wildfish.com> that replaced the
old ID with the new one. I deemed it better to leave the old ID in use
(I can't verify if the old ID was correct or not).
The original commit message:
Every time I reinstall or update Ubuntu I have to make this change
to get it to recognise my Native Instruments Traktor Audio 6
external soundcard.
Each time I remember the change by hunting down this forum post in
German,
https://forum.ubuntuusers.de/topic/traktor-audio-6-erkannt-aber-nicht-anwaehlbar/3/#post-8759808
(I don't speak German).
I'm not sure if the ID is just incorrect or if perhaps the hardware
identifies itself differently on slightly different models, so
perhaps we need to duplicate the line - I'm well outside of my
comfort zone here and I know barely anything about how hardware
works on Linux but figured if it helps me it would help others so I
should put it forward.
Thanks!
Previously the suspend cause was logged as a hexadecimal number, now
it's logged as a human-friendly string.
Also, the command line interface handled only a subset of causes when
printing them, now all suspend causes are printed.
The Intel HDMI LPE driver works in a peculiar way when the HDMI cable is
not plugged in: any written audio is immediately discarded and underrun
is reported. That resulted in an infinite loop, because PulseAudio tried
to keep the buffer filled, which was futile since the written audio was
immediately consumed/discarded.
This patch adds special handling for the LPE driver: if the active port
of the sink is unavailable, the sink suspends itself. A new suspend
cause is added: PA_SUSPEND_UNAVAILABLE.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=100488
This removes the symdef header generation m4 magic in favour of a
simpler macro method, allowing us to skip one unnecessary build step
while moving to meson, and removing an 11 year old todo!
The function is declared in pulse/format.h and it has Doxygen
documentation, which tells me that the intention was to make the
function available to clients.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=103806
In proces(), the do_move() function calls pa_{sink_input,source_output}_set_property().
This triggers a call to {sink_input,source_output}_proplist_cb() which called process()
a second time.
This patch avoids the duplicate and nested call to process() by checking if
PA_PROP_FILTER_APPLY_MOVING is set in {sink_input,source_output}_proplist_cb().
This patch adds a sink_input_properties argument to module-ladspa-sink,
which can be helpful for customizing the appearance of the sink input in
various volume control applications, or to differentiate between
multiple instances of the module.
The get_cpuid() function in cpu-x86.c was buggy on x86-64. When building
without optimizations, the homegrown assembly code overwrote the
beginning of the function argument list on the stack. That happened to
work fine on regular x86-64, but caused crashing with the x32 ABI.
At least GCC and clang provide cpuid.h, which has the __get_cpuid()
function that can be used instead of the homegrown assembly.
The PA_REG_* constants can be removed as well, because they're not used
any more.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=103656
When a sink input was unlinked between the calls to pa_sink_move_all_start() and
pa_sink_move_all_finish(), pa_sink_move_all_finish() tried to finish the move
of the already unlinked sink input, which lead to an assertion in
pa_sink_input_finish_move(). The same applies for the source side.
This patch fixes the problem by checking the state of the sink input or
source output in pa_*_move_all_finish().
Bug report: https://bugs.freedesktop.org/show_bug.cgi?id=103752
module-switch-on-connect would switch to any new sink, even if the sink
was a filter or a null-sink.
This patch adds a command line option ignore_virtual to the module, which
lets module-switch-on-connect ignore virtual sinks and sources. The flag
is true by default because the purpose of the module is to switch to new
hardware when it becomes available.
module-device-manager doesn't change the routing of those streams that
have been explicitly routed by the user, which is good. Similarly, it
should leave those streams alone whose routing was decided by the
application that created the stream. This patch implements that.
BugLink: https://github.com/wwmm/pulseeffects/issues/99
When a stream is created, and the stream creator specifies which device
should be used, that can affect automatic routing policies.
Specifically, module-device-manager shouldn't apply its priority list
routing when a stream has been routed by the application that created
the stream.
A stream that was initially routed by the application may be moved for
some valid reason (e.g. user requesting a move, or the original device
disappearing). When the stream is moved away from its initial device,
the "device requested by application" flag isn't relevant any more, so
it's set to false and never reset to true again.
The change in module-device-manager's routing logic will be done in the
following patch.
Fixes the following compiler errors:
./pulsecore/sconv-s16be.h:41:6: warning: type of 'pa_sconv_s24_32be_from_float32ne' does not match original declaration [-Wlto-type-mismatch]
void pa_sconv_s24_32be_from_float32ne(unsigned n, const float *a, uint8_t *b);
^
pulsecore/sconv-s16le.c:413:6: note: 'pa_sconv_s24_32be_from_float32ne' was previously declared here
void pa_sconv_s24_32le_from_float32ne(unsigned n, const float *a, uint32_t *b) {
^
pulsecore/sconv-s16le.c:413:6: note: code may be misoptimized unless -fno-strict-aliasing is used
./pulsecore/sconv-s16be.h:40:6: warning: type of 'pa_sconv_s24_32be_to_float32ne' does not match original declaration [-Wlto-type-mismatch]
void pa_sconv_s24_32be_to_float32ne(unsigned n, const uint8_t *a, float *b);
^
pulsecore/sconv-s16le.c:388:6: note: 'pa_sconv_s24_32be_to_float32ne' was previously declared here
void pa_sconv_s24_32le_to_float32ne(unsigned n, const uint32_t *a, float *b) {
^
pulsecore/sconv-s16le.c:388:6: note: code may be misoptimized unless -fno-strict-aliasing is used
./pulsecore/sconv-s16be.h:56:6: warning: type of 'pa_sconv_s24_32be_from_s16ne' does not match original declaration [-Wlto-type-mismatch]
void pa_sconv_s24_32be_from_s16ne(unsigned n, const int16_t *a, uint8_t *b);
^
pulsecore/sconv-s16le.c:365:6: note: 'pa_sconv_s24_32be_from_s16ne' was previously declared here
void pa_sconv_s24_32le_from_s16ne(unsigned n, const int16_t *a, uint32_t *b) {
^
pulsecore/sconv-s16le.c:365:6: note: code may be misoptimized unless -fno-strict-aliasing is used
./pulsecore/sconv-s16be.h:55:6: warning: type of 'pa_sconv_s24_32be_to_s16ne' does not match original declaration [-Wlto-type-mismatch]
void pa_sconv_s24_32be_to_s16ne(unsigned n, const uint8_t *a, int16_t *b);
^
pulsecore/sconv-s16le.c:342:6: note: 'pa_sconv_s24_32be_to_s16ne' was previously declared here
void pa_sconv_s24_32le_to_s16ne(unsigned n, const uint32_t *a, int16_t *b) {
^
pulsecore/sconv-s16le.c:342:6: note: code may be misoptimized unless -fno-strict-aliasing is used
Signed-off-by: Constantine Kharlamov <Hi-Angel@yandex.ru>
The filter sources should have the same max_rewind as the master source,
but these modules didn't update max_rewind when the master max_rewind
changed.
Previously max_rewind was always set to the full hw buffer size, but
the actual maximum rewind amount is limited to the part of the hw buffer
that is in use.
The rewind request that was done when lowering the sink latency had to
be moved to happen before updating max_rewind.
The practical benefit of this change: When using a filter source on a
monitor source, the filter source latency is increased by max_rewind.
Without this change the max_rewind of an alsa sink is often
unnecessarily high, which leads to unnecessarily high latency with
filter sources.
Monitor sources themselves don't suffer from the latency issue, because
they use the current sink latency instead of max_rewind for the extra
buffer that they keep to deal with rewinds.
Coverity complained about data->sink being possibly NULL when it's
dereferenced later. It was difficult for me to figure out whether that
was a false positive or not. Hopefully the comments make it a bit
easier to reason about the code in the future.
CID: 1323591
This rejigs the update_rate() logic to encompass changes to the sample
spec as a whole, as well as passthrough status. As a result,
sinks/sources provide a reconfigure() method which allows
reconfiguration as required.
The behaviour itself is currently unchanged -- alsa-sink/-source do not
actually implement anything other than rate updates for now (nor are
they ever requested to). This can be modified in the future, to allow,
for example 24-bit output when incoming media supports it, as well as
channel count changes for passthrough sinks.
Another related change is that passthrough status is now part of
sink/source reconfiguration, and we can stop doing a suspend/unsuspend
when entering/leaving passthrough state. So that part is now divided
in two -- pa_sink_reconfigure() sets the sink in passthrough mode if
required, and pa_sink_enter_passthrough() sets up everything else
(this currently means only volumes, but could disable other processing)
for passthrough mode.
The configured adjust time does not match exactly the real adjust time. Also
the adjust time varies. To improve latency estimation use an average of the
measured adjust times instead of the configured value in all calculations.
Changes:
- Mention that source outputs have volume too.
- Don't claim that most distributions have flat volumes enabled.
- Volumes use a cubic scale, not logarithmic.
- Reword the warning about using the conversion functions on hardware
volumes. The old wording gave the incorrect impression that hardware
volumes could never be converted to dB or linear scale.
Since HSP had higher priority than A2DP, the default profile when
connecting a new headset was HSP. To me it makes more sense to default
to high-quality output. We already have some automatic policies to
switch to HSP when it's needed.
I also made the A2DP source and HSP/HFP gateway profiles have lower
priority than the A2DP sink and HSP headset profiles. The A2DP source
and HSP/HFP gateway profiles should only be activated if the remote
device initiates audio streaming, so it makes sense to have lower
priority for those profiles.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=103058
This avoids the following autoconf warning:
configure.ac:89: warning: AC_COMPILE_IFELSE was called before AC_USE_SYSTEM_EXTENSIONS
../../lib/autoconf/specific.m4:368: AC_USE_SYSTEM_EXTENSIONS is expanded from...
configure.ac:89: the top level
This reverts commit ca63fbc1d8.
I applied the patch too hastily. force-speaker.conf is supposed to be
used only when the alsa mixer doesn't contain any elements that would
indicate the existence of a speaker port, but the reverted patch is a
workaround for a different problem. On the two affected EeePC machines
the Headphone element needs to be unmuted when using speakers. The
analog-output-speaker-always path happens to do that, but that's
unintentional. analog-output-speaker was changed[1] to mute the
headphone output when using the speaker port, and
analog-output-speaker-always should have been changed too, but that was
forgotten.
The kernel driver is buggy if it has a Headphone mixer element that
mutes both headphones and speakers, so this should be fixed in alsa. If
we end up having a workaround in PulseAudio for the broken driver, it
should be implemented with a new profile set and path configuration
files.
[1] https://cgit.freedesktop.org/pulseaudio/pulseaudio/commit/?id=22aac4e9fdb3786178f7815a0cb2150f588b1582
This adds a port, card and profile to RAOP sinks to make it
possible to change the latency at runtime (and have it persist)
using pavucontrol or pactl set-port-latency-offset.
Also move the IP:port part of the sink name to the port name.
Correct spelling of 'through' in a comment helps to fix a warning :)
also drop some unrelated comments
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
the comment /* Fall through. */ fixes the warning, but gcc-7 seems to be more
picky about the position in the code
pulsecore/sink-input.c: In function ‘pa_sink_input_update_proplist’:
pulsecore/sink-input.c:1531:13: warning: this statement may fall through [-Wimplicit-fallthrough=]
for (state = NULL; (key = pa_proplist_iterate(i->proplist, &state));) {
^~~
pulsecore/sink-input.c:1539:9: note: here
case PA_UPDATE_REPLACE: {
^~~~
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
EAGAIN is used allover the code rather than EWOULDBLOCK
POSIX allows EAGAIN and EWOULDBLOCK to have the same value (and in fact it is)
don't check for EWOULDBLOCK
modules/raop/raop-client.c: In function ‘send_udp_audio_packet’:
modules/raop/raop-client.c:473:41: warning: logical ‘or’ of equal expressions [-Wlogical-op]
if (written < 0 && (errno == EAGAIN || errno == EWOULDBLOCK)) {
^~
modules/raop/raop-client.c: In function ‘resend_udp_audio_packets’:
modules/raop/raop-client.c:528:45: warning: logical ‘or’ of equal expressions [-Wlogical-op]
if (written < 0 && (errno == EAGAIN || errno == EWOULDBLOCK)) {
^~
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
Use pa_assert_se() to check return value (pro forma) like everywhere else
Coverity ID: #154313
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
make install was failing on Windows because there cp is
used to replace ln -s commands but cp doesn't resolve its
first argument as relative to the second one.
This patch changes the install-bashcompletion-aliases rules
to chdir to the target dir so that cp works correctly. This
is the solution recomended in the automake documentation.
The macro LADSPA_PATH was defined as a list of directories quoted but
without taking into account that the directory names, specially on
Windows, can contain backslashes that need escaping.
This patch removes the quoted from the macro and uses the C preprocessor
to quote it properly using a helper macro.
Under MinGW, LC_MESSAGES is defined in libint.h which is not
included when pulseaudio is configured with nls disabled.
LC_MESSAGES is referenced when setting PA_PROP_APPLICATION_LANGUAGE.
This patch just disables setting that property when ENABLE_NLS
is not defined.
It seems that the intention was to create create write_thread_event on
thread_mainloop instead of main_mainloop (the first parameter of
io_new() is thread_mainloop, io_free() is called on thread_mainloop
etc.).
As long as both mainloops are implemented with pa_mainloop, this bug has
no effect on behaviour, because the io_new() implementation is the same.
And indeed, with the current code base both mainloops are always
pa_mainloops. However, when the tunnel-new modules switches to pa_rtpoll
as the pa_mainloop_api provider, this bug would cause problems.
Pulseaudio tries to pick the best profile (on startup or
hotplugged), the best profile is the profile with the highest
priority which isn't unavailable.
Due to the facts that iec958 ports available status always (?)
is unknown, and that it is generally more likely that a user use
hdmi than iec958, lets prioritze hdmi over iec958.
This patch shift the analog-* mappings +5 and hdmi-* mappings +5.
Use predefined values depending on the server, and make it configurable.
AirPlay is supposed to have 2s of latency. With my hardware, this is
more 2.352 seconds after numerous tests.
Switch from pausing/resuming the smoother to resetting it because the
smoother got stuck returning the same value after an idle/running cycle,
making latency calculation wrong.
This breaks a lot of headsets, so disabling by default. Can be
re-enabled in configuration for specific hardware where it is deemed
necessary.
Also added some debug logging to be able to examine what MTU size is
reported by the device.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=102660
Some sound cards don't have any alsa-lib configuration, but they used to
work well enough up to PulseAudio 10. PulseAudio 11 stopped using "hw:0"
for the analog-stereo mapping, and instead defined it as a fallback
mapping without any mixer handling. As a result, switching between
headphones and speakers stopped working without changing the mixer
settings manually at least on Toshiba Chromebook 2. This patch adds the
mixer handling back to the fallback mapping.
I also renamed "unknown-stereo" to "stereo-fallback", because I like
that name more.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=102560
We're supposed to prioritize USB sound cards over PCI sound cards, but
the priority bonus for the "internal" form factor prevents this from
happening. Not all (if any) USB sound cards have the form factor
property set, whereas at least on my laptop the on-board sound card has
the form factor set to "internal".
The order of the pa_sink_input_put() and pa_sink_put() calls in filter
modules was swapped in commit edc465da77 ("virtual sources and sinks:
Don't double attach a sink input or source output on filter load").
If flat volumes and volume sharing is enabled, the pa_sink_input_put()
call will update volumes of the whole tree of virtual sinks that are
connected to the root sink. The recursive updating procedure tried to
also update the volume of the new sink for which pa_sink_put() had not
yet been called, causing an assertion failure.
This patch tries to make sure that the volume of not-yet-linked sinks
is never changed. pa_sink_put() will set the sink volume correctly, so
it's fine to skip the not-yet-linked sinks during pa_sink_input_put().
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=102549
The autospawn mechanism already had a root-autospawn protection mechanism. When
using systemd that was lost. Systemd 234 has a mechanism to conditionalize unit starting
on the running user, so lets do that to protect against root autospawning.
A race condition prevents the AES non-audio bit from being set
when enabling IEC61937 passthrough on resume with no sink-input
connected (pa_sink_is_passthrough returns false). The non-audio
bit should really be set when opening the sink.
Force the sink to suspend/resume when actually entering passthrough
mode, and likewise force a suspend-resume on leaving passthrough mode.
Tested with E-AC3 streams which do need the AES bit set for my
Onkyon receiver to detect the format instead of playing it as
PCM.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
This is basically a copy of module-always-sink but doing the same for
sources. Whenever no source is available, a module-null-source is loaded
and whenever a new source is available again, module-null-source is
unloaded.
By this, anything using a source will automatically be switched to the
null source when the actual source disappears, and back to the actual
source if it appears again.
There are actually two HSP HS UUIDs. My theory is that the second one
was added, because someone was not happy with the old UUID being used
for identifying two different things (the HSP profile as a whole, and
the HS role within the HSP profile). Some headsets only use the new
UUID, and those headsets won't work if we don't recognize the new UUID.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=93898
to_alsa_dB() returns a result rounded to two decimal places (instead of
using integer truncation) to avoid small errors when converting between
dB and volume.
Consider playback at -22 dB (which is supported by ALSA) but results in
the higher level of -21 dB plus software attenuation.
pa_sw_volume_from_dB(-22) = 28172
pa_sw_volume_to_dB(28172) = -21.9997351
to_alsa_dB(-21.9997351) = -2199
ALSA value 106 = -2200
ALSA value 107 = -2100
...
rounding = +1 /* "accurate or first above" */
snd_mixer_selem_ask_playback_dB_vol(me, -2199, rounding, &alsa_val)
alsa_val = -2100
Signed-off-by: Ian Ray <ian.ray@ge.com>
Some modules may only be loaded once, and trying to load them
twice from default.pa makes PulseAudio startup fail. While that could
be considered a user error, it's nicer to not be so strict. It's not
necessarily easy to figure what went wrong, if for example the user
plays with RAOP and adds module-raop-discover to default.pa, which first
works fine, but suddenly stops working when the user at some point
enables RAOP support in paprefs. Enabling RAOP in paprefs makes
module-gconf load the module too, so the module gets loaded twice.
This patch adds a way to differentiate module load errors, and
make cli-command ignore the error when the module is already
loaded.
It was reported that PulseAudio weakens the umask to 022 if it's
initially set to 077. That's not as big problem as it might seem,
but it's still a problem. The umask affects the permissions of the state
files, and those aren't readable by other users anyway in the per-user
mode, because PulseAudio puts them in directories that aren't
accessible to other users. In the system mode the state files will be
readable by everyone, though, even by those users that don't otherwise
have access to PulseAudio. The state files are slightly
privacy-sensitive, because they contain e.g. history of applications
that have used PulseAudio.
I can't think of any use cases where access to the state files by other
users would be necessary, either in the per-user mode or in the system
mode, so let's use umask 077. This doesn't prevent access to any
sockets in the system mode, because all directories that PulseAudio
creates in the system mode will have permissions 755 regardless of the
umask, and the sockets themselves always have permissions 777.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=102060
Users may want to change the parameters of some load-once modules in
~/.config/pulse/default.pa. That should be possible by including
/etc/pulse/default.pa from the per-user configuration file, and then
unloading a module and reloading it with different parameters. However,
that doesn't work, because the unload-module command will not unload the
module immediately, so the subsequent load-module command will fail when
the module can be loaded only once.
This patch makes the module unloading synchronous. "pacmd unload-module
module-cli-protocol-unix" is something that might not like this change,
since the command will unload the code that is processing the command,
but I tested it and it works fine. When pa_module_unload() is called,
that won't yet remove the module code from memory, the lt_dlclose() call
is postponed until it's safe to remove the code from memory.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=102205
module-switch-on-port-available didn't do anything when a port changes
its status if the card didn't have any sinks or sources. This was to
avoid bad things during card initialization, but the if condition also
prevented any profile switches away from the "off" profile, because the
card has no sinks or sources when the "off" profile is active.
pa_card nowadays has the "linked" flag that
module-switch-on-port-available could have checked instead, but since it
doesn't make sense to emit port status change events before the card has
been initialized, I added the check in pa_device_port_set_available()
instead.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=101794
It was reported that on a certain USB card, identified as
"0d8c:0102 C-Media Electronics, Inc. CM106 Like Sound Device",
the "PCM Capture Source" element had to be set to "IEC958 In" before
the iec958 input would work.
The iec958-stereo-input.conf file didn't exist before, although the path
was referenced in the default.conf profile configuration file.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=101973
When connecting a headset via the native backend, the transport state was
not updated correctly.
This patch sets the state to PLAYING in transport_acquire() if necessary.
If the description is not updated when moving, the old automatically
generated description will refer to the old master sink after the move,
which is not nice.
Setting the allow_negative flag of pa_{source,sink}_get_latency_within_thread() to true
leads to improved end to end latency estimation and to correct handling of negative port
latency offsets.
There are one headset jack on the front panel of TB16, through this
jack, we have one stereo headphone output (hw:%f,0,0) and one mono
headset-mic input (hw:%f,0,0); and there is one speaker output jack
(hw:%f,1,0) on the rear panel of TB16.
The detail information of the Dell dock TB16:
http://www.dell.com/support/article/sg/en/sgbsdt1/SLN301105
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Currently, if a stream is manually moved to a filter sink or source managed by
module-filter-apply, the stream will be silently re-routed to the master sink
or source, because the filter.apply property is not set on that stream. We can
assume, that the users intention however was to have the stream filtered.
Therefore this patch changes the logic, so that the stream will not be moved
to the master but remains on the filter sink or source. To handle the change
of a property correctly, the filter.apply property must be set temporarily.
An additional property filter.apply.set_by_mfa was introduced to mark those
streams, so that filter.apply can be removed again when the stream moves away
from the filter.
When a phone is connected via bluetooth and switches to HFP, the sinks
and sources will have higher priority than the built-in devices.
Therefore they are chosen as default and module-bluetooth-policy will
incorrectly insert loopback modules that loop the phone back to itself.
This patch fixes the problem by lowering the priority of sink and source
if PulseAudio is in the headset role. The priority is also lowered if the
device is an a2dp source. In both cases it does not make sense to make the
source or sink default unless there is no other sound device available.
Sources should probably be added to pa_core.sources in pa_source_put(),
but currently they're added in pa_source_new(). A lot of stuff can
happen between the pa_source_new() and pa_source_put() calls, and
it has happened that the default source was updated during this time.
Therefore, pa_core_update_default_source() needs to take it into account
that not every source is necessarily linked.
Currently pulseaudio crashes with an assertion in pa_rtpoll_item_new_asyncmsgq_read()
or pa_rtpoll_item_new_asyncmsgq_write() if a loopback is applied to a tunnel-new
sink or source, because tunnel-{sink,source}-new do not set thread_info.rtpoll.
The same applies to module-combine-sink and module-rtp-recv.
This patch is not a complete fix for the problem but provides a temporary band-aid
by initializing thread_info.rtpoll properly. The rtpoll created is never run, but
loopback and combine-sink nevertheless work, see comments in the code.
This patch does not work for module-rtp-recv, but using rtp-recv with a remote
sink does not seem to make a lot of sense anyway.
Bug link: https://bugs.freedesktop.org/show_bug.cgi?id=73429
PA builds fine on MinGW except for the use of the scandir function in
pulsecore/conf-parser.c, so I provided a Win32 implementation. With this
patch the latest code builds on Win32 without problems.
If only part of the buffer is written into stdout by stdout_callback,
the buffer_index variable is increased by the number of written bytes,
buffer_length variable is decreased while the allocated buffer size
remains the same. That suggests that the current allocated size is
calculated as (buffer_index + buffer_length). However the current
stream_read_callback implementation writes new data to the start of the
buffer and allocates too little space, so that (buffer + buffer_index +
buffer_length - 1) could actully point outside of the allocated buffer.
pa_usec_t is an unsigned type, but there were calculations that used it
as if it were a signed type.
If the latency is negative, pa_simple_get_latency() now reports 0.
Added some comments too.
The on_the_fly_snapshot variable contains the amount of bytes that has
been sent from the source IO thread to the main thread, but not yet
pushed to the stream memblockq. The data is in the stream format, but
the bytes-to-usec conversion used the source format, which caused random
latency reporting errors.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=81075
This allows us to restore the default device properly when a
hotpluggable device (e.g. a USB sound card) is set as the default, but
unplugged temporarily. Previously we would forget that the unplugged
device was ever set as the default, because we had to set
configured_default_sink to NULL to avoid having a stale pa_sink pointer,
and also because module-default-device-restore couldn't resolve the name
of a currently-unplugged device to a pa_sink pointer.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=89934
This reverts commit 69c212f8c1.
Reasons:
The original reason for the patch was to work around some issue
regarding the profile not connecting immediately (sorry, I don't really
know the details), but that issue was fixed later by commit 998dfdf4cc,
so the original reason doesn't apply any more.
Automatically changing the profile when the transport state changes to
PLAYING has traditionally been handled by module-bluetooth-policy, and
as far as I can tell, there's no reason to change that.
The assertion is unsafe. It's not guaranteed that the profile change
will always succeed (at least pa_thread_mq_init() can fail due to
reaching the maximum file descriptor limit).
There are two reasons for this change:
1. If it is a Dell desktop machine with the realtek codec, and there
is no internal microphone on it, there is one physical audio jack
which can support headphone, headset and microphone, but this audio
jack does not have hardware capability to distinguish what is plugged
in, after users plug in a headphone and select headphone from UI
program, the headphone can't output any sound. There are many reasons
for this issue, one of them is the active_port of pa_source is set
to headphone-mic, that means the kernel audio driver will configure
this audio jack to be a microphone jack instead of headphone jack.
If we make the priority of headset-mic a bit higher than headphone-mic,
the headset-mic will be the active_port of pa_source unless users
select the headphone-mic on purpose, then this issue will be fixed.
2. Nowadays, the headset is more popular than traditional microphone,
It is highly possible that users plug in a headset instead of
microphone, it makes sense to make the headset-mic's priority higher
than headphone-mic's.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
module-stream-restore primarily uses the role of a stream for restoring. The sink-inputs
and source-outputs of filters all have role "filter", therefore currently all filters are
treated equally and are restored to the same device and volume.
This patch lets module-stream-restore ignore the streams that connect the filter to the
master.
Bug link: https://bugs.freedesktop.org/show_bug.cgi?id=100065
These changes are from fedora.zanata.org. The authors are
Sam Friedmann <sfriedma@redhat.com>
Wim Taymans <wim.taymans@gmail.com>
Edouard Duliege <edouard.duliege@gmail.com>
When a filter sink is moving, it's not connected to any master sink, and
therefore it's not connected to any IO thread either. In this situation
trying to move a stream that is connected to the filter sink is likely
to result in crashing, because starting the move involves sending a
message to the IO thread. Sometimes this works by accident (the
asyncmsgq of the filter sink still points to the old master sink's
asyncmsgq), but we really should never attempt it. This patch blocks all
moves where the moving stream is connected to a filter sink that itself
is in the middle of a move.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=100277
In sink_put() and source_put(), pa_core_update_default_{sink,source}() was called
before the PA_CORE_HOOK_{SINK,SOURCE}_PUT hook. Therefore module-switch-on-connect
could not correctly determine the old default sink/source if no user default was
set and a sink/source with higher priority than any other sink/source turned up.
This patch corrects the problem by swapping the order of the hook call and the
pa_core_update_default_sink() call.
Additionally it corrects a problem in module-switch-on-connect. If, after the
change above, the new sink/source was the first sink/source to appear, pulseaudio
would crash because module-switch-on-connect assumed that the default sink/source
was not NULL. The patch checks if the default sink/source is NULL and only sets
the new default sink/source in that case.
When a filter is loaded and module-switch-on-connect is present, switch-on-connect
will make the filter the default sink or source and move streams from the old
default to the filter. This is done from the sink/source put hook, therefore streams
are moved to the filter before the module init function of the filter calls
sink_input_put() or source_output_put(). The move succeeds because the asyncmsq
already points to the queue of the master sink or source. When the master sink or
source is attached to the sink input or source output, the attach callback will call
pa_{sink,source}_attach_within_thread(). These functions assume that all streams
are detached. Because streams were already moved to the filter by switch-on-connect,
this assumption leads to an assertion in pa_{sink_input,source_output}_attach().
This patch fixes the problem by reverting the order of the pa_{sink,source}_put()
calls and the pa_{sink_input,source_output}_put calls and creating the sink input
or source output corked. The initial rewind that is done for the master sink is
moved to the sink message handler. The order of the unlink calls is swapped as well
to prevent that the filter appears to be moving during module unload.
The patch also seems to improve user experience, the move of a stream to the filter
sink is now done without any audible interruption on my system.
The patch is only tested for module-echo-cancel.
Bug-Link: https://bugs.freedesktop.org/show_bug.cgi?id=100065
When sinks are compared during the default sink selection, the active
port's availability is inspected. Therefore, the default sink should be
updated when the active port changes, because the new port may have
different availability status than the old port.
For example, let's say that a laptop has an analog sink with a speaker
and a headphone port, and headphones are initially plugged in, so both
ports can be used[1]. The headphone port is initially the active port.
There's also a null sink in the system. When the headphones are
unplugged, the headphone port becomes unavailable, and the null sink
becomes the new default sink. Then module-switch-on-port-available
changes the analog sink port to speakers. Now the default sink should
change back to the analog sink, but that doesn't happen without this
patch.
[1] Actually we currently mark speakers as unavailable when headphones
are plugged in, but that's not strictly necessary. My example relies on
both ports being available initially, so the bug can't be reproduced
with the current mixer configuration.
When the ofono backend released a tranport during suspend of sink or source, the
transport state was not changed to IDLE. Therefore pa_bluetooth_transport_set_state()
would return immediately when trying to resume. Even though the transport was acquired
correctly, setup_stream() would never be called and the resume failed.
This patch sets the transport state to IDLE when the transport is released. On resume,
the first call to transport_acquire() will be done from the message handler of the
*_SET_STATE message when source or sink are set to RUNNING. This call will only request
the setup of the connection, so setup_stream() cannot be called.
When the transport changes the state to PLAYING in hf_audio_agent_new_connection(),
handle_transport_state_change() is called. Because the sink or source state is already
RUNNING, the pa_{source,sink}_suspend() call will not lead to a state change message
and the I/O thread must be signaled explicitely to setup the stream.
The first setup of the device would also fail, which was only visible when the profile
was restored after connecting the headset. When trying to restore the headset_head_unit
profile, the profile was shortly set to off, so the headset always returned to a2dp.
This patch allows a delayed setup for the headset_head_unit profile, so that the profile
can successfully be restored.
When suspending due to idle timeout the transport will not change its
state, also in case the fd is closed due to POLLERR/POLLHUP events
the release shall check if the fd is still set otherwise it will fail
to be acquired again.
This means something went wrong, which in case of ofono backend it is
probably due to the profile not connecting immediately, but it can be
safely restored in that case the transport is playing which means the
profile has recovered connectivity.
The compiler warned about number_of_frames being possibly used
uninitialized, and on closer inspection I found that it was indeed not
initialized if saved_frame_time_valid is false.
In commit fe70b9e11a "source/sink: Allow pa_{source,
sink}_get_latency_within_thread() to return negative values" the
number_of_frames variable was added as an unsigned version of the l
variable, and number_of_frames partially replaced the l variable. The
replacement should have gone all the way, however. This patch removes
the remaining uses of the l variable and substitutes number_of_frames
on its place, and as a result, number_of_frames is now always
initialized.
It doesn't make sense to use a sink or source whose active port is
unavailable, so let's take this into account when choosing the default
sink and source.
Currently the default sink policy is simple: either the user has
configured it explicitly, in which case we always use that as the
default, or we pick the sink with the highest priority. The sink
priorities are currently static, so there's no need to worry about
updating the default sink when sink priorities change.
I intend to make things a bit more complex: if the active port of a sink
is unavailable, the sink should not be the default sink, and I also want
to make sink priorities dependent on the active port, so changing the
port should cause re-evaluation of which sink to choose as the default.
Currently the default sink choice is done only when someone calls
pa_namereg_get_default_sink(), and change notifications are only sent
when a sink is created or destroyed. That makes it hard to add new rules
to the default sink selection policy.
This patch moves the default sink selection to
pa_core_update_default_sink(), which is called whenever something
happens that can affect the default sink choice. That function needs to
know the previous choice in order to send change notifications as
appropriate, but previously pa_core.default_sink was only set when the
user had configured it explicitly. Now pa_core.default_sink is always
set (unless there are no sinks at all), so pa_core_update_default_sink()
can use that to get the previous choice. The user configuration is saved
in a new variable, pa_core.configured_default_sink.
pa_namereg_get_default_sink() is now unnecessary, because
pa_core.default_sink can be used directly to get the
currently-considered-best sink. pa_namereg_set_default_sink() is
replaced by pa_core_set_configured_default_sink().
I haven't confirmed it, but I expect that this patch will fix problems
in the D-Bus protocol related to default sink handling. The D-Bus
protocol used to get confused when the current default sink gets
removed. It would incorrectly think that if there's no explicitly
configured default sink, then there's no default sink at all. Even
worse, when the D-Bus thinks that there's no default sink, it concludes
that there are no sinks at all, which made it impossible to configure
the default sink via the D-Bus interface. Now that pa_core.default_sink
is always set, except when there really aren't any sinks, the D-Bus
protocol should behave correctly.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=99425
Previously, if front:x didn't work, we would try to use hw:x for analog
stereo output. There's no guarantee that hw:x is an analog output,
however. For example, the Intel HDMI LPE driver uses hw:x for HDMI
output, and PulseAudio incorrectly created analog profiles for that
card, because front:x doesn't work but hw:x does.
This patch changes things so that the analog stereo mapping doesn't any
more use hw:x as a fallback. A separate "unknown stereo" fallback
mapping is added to handle the rare case where hw:x is the only PCM
device that works.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=100488
The percent calculation could overflow in the pa_*volume_snprint*() functions.
For large volumes, volume * 100 can exceed UINT32_MAX.
This patch adds appropriate type casts.
When the volume exceeds PA_VOLUME_MAX in pa_sw_volume_multiply() or
pa_sw_volume_divide(), volume settings are insanely high and the
user should be notified about it.
This patch adds volume clamping to pa_sw_volume_divide() and prints
a warning when the volume is clipped in both functions.
In pa_{source,sink}_new() and pa_{source,sink}_put() the current hardware
volume was miscalculated:
hw volume (dB) = real volume (dB) + soft volume (dB)
was used instead of
hw volume (dB) = real volume (dB) - soft volume (dB)
This lead to a crash in pa_alsa_path_set_volume() if high volumes were
set and the port was changed.
This patch fixes the calculation. Thanks to Tanu for pointing out
the correct solution.
Bug link: https://bugs.freedesktop.org/show_bug.cgi?id=65520
Several virtual sources and sinks apart from module-echo-cancel also query the master
sink or source to estimate the current latency. Those modules might potentially show
the bug that is described for module-echo-cancel in bug 100277.
This patch checks in the message handlers for the PA_{SINK,SOURCE}_MESSAGE_GET_LATENCY
if the master sink or source is valid and returns 0 as latency if not. This is however
not yet sufficient to solve the issue. Additional patches will follow.
When module-echo-cancel is loaded and there is only one sound card, then during a
profile switch, all sinks and sources can become temporarily unavailable. If
module-always sink is loaded, it will load a null-sink in that situation. If
also module-switch-on-connect is loaded, it will try to move the sink-inputs to
the new null-sink. If a sink-input was connected to the echo-cancel sink,
pa_sink_input_start_move() will send a PA_SINK_GET_LATENCY message to the
echo-cancel sink. The message handler will then in turn call
pa_sink_get_latency_within_thread() for the invalid master sink of module-echo-cancel.
This lead to a segfault.
This patch checks in the message handler if the master sink (or source) is valid and
returns 0 if not.
If a HFP audio gateway was connected via the ofono backend, pulse would
segfault during shutdown of the daemon. pa_bluetooth_discovery_unref()
removed the devices and transports before the ofono backend was freed.
Because the ofono backend keeps its own list of transports, transport_free()
was then called during termination of the ofono backend with an invalid
transport. Bug reported by Andrew Hlynskyi.
This patch moves the termination of the ofono and native backends before
freeing the devices.
The 'portable' form factor was currently missing meaning it is not
getting any form-factor priority at all and it would therefore always
be ranked lower then internal devices (which receive 400 form factor
priority). The priority 450 is smaller then 'speaker', based on the
idea that a portable device might have less quality then a dedicated
'speaker' device (some Yamaha amplifiers announce themselves as such).
https://bugs.freedesktop.org/show_bug.cgi?id=100579
The reported latency of source or sink is based on measured initial conditions.
If the conditions contain an error, the estimated latency values may become negative.
This does not indicate that the latency is indeed negative but can be considered
merely an offset error. The current get_latency_in_thread() calls and the
implementations of the PA_{SINK,SOURCE}_MESSAGE_GET_LATENCY messages truncate negative
latencies because they do not make sense from a physical point of view. In fact,
the values are truncated twice, once in the message handler and a second time in
the pa_{source,sink}_get_latency_within_thread() call itself.
This leads to two problems for the latency controller within module-loopback:
- Truncating leads to discontinuities in the latency reports which then trigger
unwanted end to end latency corrections.
- If a large negative port latency offsets is set, the reported latency is always 0,
making it impossible to control the end to end latency at all.
This patch is a pre-condition for solving these problems.
It adds a new flag to pa_{sink,source}_get_latency_within_thread() to allow
negative return values. Truncating is also removed in all implementations of the
PA_{SINK,SOURCE}_MESSAGE_GET_LATENCY message handlers. The allow_negative flag
is set to false for all calls of pa_{sink,source}_get_latency_within_thread()
except when used within PA_{SINK,SOURCE}_MESSAGE_GET_LATENCY. This means that the
original behavior is not altered in most cases. Only if a positive latency offset
is set and the message returns a negative value, the reported latency is smaller
because the values are not truncated twice.
Additionally let PA_SOURCE_MESSAGE_GET_LATENCY return -pa_sink_get_latency_within_thread()
for monitor sources because the source gets the data before it is played.
The old code worked incorrectly in several situations. For example,
trying to use the "master" argument wouldn't work, because if
"sink_master" wasn't specified, pa_namereg_get() would pick the default
sink as the master sink.
The latency controller will try to adjust to the configured latency regardless
of underruns. If the configured latency is set too small, it will lead to
periodically occuring underruns. Therefore an underrun protection is implemented
which will increase the target latency if too many underruns are detected.
Underruns are tracked and if more than 3 underruns occur, the target latency
is increased in increments of 5 ms. One underrun per hour is accepted.
The protection ensures, that independent from the configured latency the
module will converge to a stable latency if the configured latency is too
small.
The print_msg argument to update_minimum_latency() had to be re-introduced,
because there is one place where the message should not be logged.
Currently passing parameters to a filter loaded by module-filter-apply is
not possible.
To enable passing parameters to a filter this patch uses an additional property
filter.apply.{MODULE_NAME}.parameters. This way, filters like virtual-surround-sink
and ladspa-sink are fully supported. For example:
paplay file.wav --property=filter.apply=ladspa-sink \
--property=filter.apply.ladspa-sink.parameters="plugin=ladspa \
label=ladspa_stereo control=0"
Currently, module-filter-apply cannot load module-ladspa-sink because filter-apply
provides the argument "sink_master" but ladspa-sink expects "master" instead.
Therefore this patch adds the sink_master argument to module-ladspa-sink.
Additionally, the autoloaded argument was also added.
Currently, module-filter-apply cannot load module-virtual-surround-sink because filter-apply
provides the argument "sink_master" but virtual-surround-sink expects "master" instead.
Therefore this patch adds the sink_master argument to module-virtual-surround-sink.
Additionally, the autoloaded argument was also added.
fltr->name should be freed before freeing fltr. Because filter_free()
can never be called from other places without f set, the pa_assert()
can be removed and filter_free() can be used in process() as well.
When the specified pid no longer exists as a child of the process (since
it was already reaped by the SIGCHLD handler), errno is set to ECHILD, not
to ESRCH.
If source or sink are changed, the current sink input rate may be different
from the default rate. Switch sink input rate back to default to avoid the
influence of the previous combination of source and sink.
During a move sink_input->sink is not valid. This leads to a crash when
sink_input_set_rate() is called from the moving() callback. This patch
fixes the problem.
The previous patch assumed constant port latency offsets. The offsets can
however be changed by the user, therefore these changes need to be tracked
as well. This patch adds the necessary hooks.
Also the print_msg argument was removed from update_minimum_latency() and
update_latency_boundaries() because the message should always be logged.
With the current code, the user can request any end-to-end latency. Because there
is no protection against underruns, setting the latency too small will result in
repetitive underruns.
This patch tries to mitigate the problem by calculating the minimum possible latency
for the current combination of source and sink. The actual calculation has been put
in a separate function so it can easily be changed. To keep the values up to date,
changes in the latency ranges have to be tracked.
The calculated minimum latency is used to limit the configured latency.
The minimum latency is only a "best guess", so the actual minimum may be much
larger (for example for USB devices) or much smaller than the calculated value.
Changes of the port latency offsets are not yet handled, this will be done in a
separate patch.
The old code makes no sense to me. Why would multiple references mean
that a previously read-only memblock is suddenly writable? I'm pretty
sure that the original intention was to treat multi-referenced blocks
as read-only. I don't have any examples where the old code would have
caused bad behaviour, however.
The old pa_sink_set_fixed_latency() call didn't take into account that
other places use pa_frame_align() on the pa_pipe_buf() result, so the
configured latency could be sometimes slightly too high.
Adding a buffer_size variable in userdata makes it a bit easier to keep
all places that deal with the buffer size in sync.
Users may configure the device alias to have characters outside the
ASCII range, so our name cleanup routine was too aggressive. Let's just
make sure that the device description is a valid UTF-8 string.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=98160
Currently internal > speaker > headphone and pci > usb > bluetooth.
Invert both of these sets, with the reasoning that a headphone and
speakers are something that a user has actively attached and should
therefore get a higher priority. The same reasoning is applied for
the bus type, i.e. bluetooth and usb should be higher than pci,
because they most likely have been actively attached be a user.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=99222
The function can only fail if there's not enough memory available, and
if that happens, the convention in PulseAudio is to abort.
CID: 1353106, 1353108, 1353140
We don't always know whether the in-flight memory chunks will be
rendered or skipped (if the source is not in RUNNING). This can cause us
to have an erroneous estimate of drift, particularly when the canceller
starts.
To avoid this, we explicitly flush out the send and receive sides of the
message queue of audio chunks going from the sink to the source before
trying to perform a resync.
When moving from a user suspended source or sink to an idle suspended source or sink
the sink input or source output would not be uncorked because we did not check for
the suspend cause.
Uncorking also would not be possible in that situation because the state change callback
of the source output or sink input is called before the new source or sink is attached,
leading to a crash of pulseaudio due to a cork() call without valid source or sink.
The previous patch fixes this problem, therefore sink input or source output can now also
be uncorked when the destination is idle suspended.
If pa_sink_input_cork() or pa_source_output_cork() were called without a sink
or source attached, the calls would crash pulseaudio.
This patch fixes the problem, so that a source output or sink input can still
be corked or uncorked while source or sink are invalid. This is needed to
correct the corking logic in module-loopback.
There were two bugs in the old logic. The first one:
If a device has two profiles, the old code would start the wait timer
when the first profile connects, but when the second profile connects,
the timer would not get stopped and the CONNECTION_CHANGED hook would
not get fired, because the code for that was inside an if block that
only gets executed when the first profile connects. As a result,
module-bluez5-device loading would always be delayed until the wait
timeout expires.
The second bug:
A crash was observed in device_start_waiting_for_profiles(). That
function is called whenever the connected profile count changes from 0
to 1. The function also has an assertion that checks that the timer is
not running when the function is called. That assertion crashed in the
following scenario with a headset that supports HSP and A2DP:
1. First HSP gets connected. The timer is started.
2. Then HSP gets disconnected for some reason. The timer is still
running.
3. Then A2DP gets connected. device_start_waiting_for_profiles() is
called, because the connected profile count changed from 0 to 1 again.
The timer is already running, so the assertion fails.
First I thought I'd remove the assertion from
device_start_waiting_for_profiles() and just restart the timer on the
second call, but then I figured that when the device returns to the
"everything disconnected" state in step 2, it would be better to stop
the timer. The purpose of the timer is to delay the notification of the
device becoming connected, but if the device becomes disconnected during
the waiting period, the notification doesn't make sense any more, and
therefore the timer doesn't make sense either.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=100237
The webrtc canceller seems to have changed to require that the
set_stream_drift_samples() method be called before every call of
ProcessStream().
So we now call ec->set_stream_drift_samples() before calling
ec->record() by:
1. Always calling do_push_drift_comp() instead of only when the sink is
running
2. Calling set_stream_drift_samples() in the loop with record() instead
of outside
We do kind of leak this quirk of the webrtc canceller into the generic
bits of module-echo-cancel, but this should not be harmful in the
general case either.
The auto_switch argument was added in PulseAudio 10.0. In that release
the argument type was boolean. The type was changed to integer in commit
3397127f00. This patch adds backwards compatibility so that old
configuration files won't break when upgrading PulseAudio to 11.0.
With headset=auto it is possible that AG devices are connected and handled
via the native backend when ofono is started. Because the HS role will then
be disabled in the native backend, AG devices must be disconnected and any
future connections will be handled by ofono.
This patch changes the behavior of the headset=auto switch for module-bluez5-discover.
With headset=auto now both backends will be active at the same time for the AG role and
the switching between the backends is only done for the HS role.
headset=ofono and headset=native remain unchanged.
This allows to use old HSP only headsets while running ofono and to have headset support
via pulseaudio if ofono is started with the --noplugin=hfp_ag_bluez5 option.
document behaviour of pa_shared_remove() in case name does not exist
Coverity ID: #1380672
thanks to Georg Chini for suggesting to swap patch title and commit message
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
The "profile->card != c->card" check always evaluated to false, so the
CardProfileUpdated signal was never sent. The reason: call_data was
assigned to a pa_card_profile pointer, but the correct type is a pa_card
pointer.
readdir_r() was supposed to be a thread-safe version of readdir(), but
the interface turned out to be problematic. Due to the problems and the
fact that readdir() is safe enough on modern libc implementations, glibc
deprecated readdir_r() in version 2.24.
The man page contains more information about what's wrong with
readdir_r(): http://man7.org/linux/man-pages/man3/readdir_r.3.html
On systems with constrained CPUs, we might run into a situation where
the master source/sink is configured to have too high a latency.
On the source side, this would cause us to wake up with a large chunk of
data to process, which might cause us to exhust our RT limit and thus be
killed.
So it makes sense to limit the overall latency that we request from the
source (and correspondingly, the sink, so we don't starve for playback
data on the source side).
The 10 blocks maximum is somewhat arbitrary (I'm assuming the system has
enough headroom to process 10 chunks through the canceller without
getting close to the RT limit). This might make sense to make tunable in
the future.
If the ALSA device supports granular pointer reporting, we end up in a
situation where we write out a bunch of data, iterate, and then find a
small amount of data available in the buffer (consumed while we were
writing data into the available buffer space). We do this 10 times
before quitting the write loop.
This is inefficient in itself, but can also have wider consequences. For
example, with module-combine-sink, this will end up pushing the same
small chunks to all other devices too.
Given both of these, it just makes sense to not try to write out data
unless a minimum threshold is available. This could potentially be a
fragment, but it's likely most robust to just work with a fraction of
the total available buffer size.
don't ignore server port parsing errors as suggested by Hajime Fujita
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
Cc: Hajime Fujita <crisp.fujita@nifty.com>
wath may be NULL, as suggested by Hajime Fujita
Coverity ID: #1398156
setting val = NULL is not needed
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
Cc: Hajime Fujita <crisp.fujita@nifty.com>
for example, in case HAVE_MEMFD is #undef, checking with #if HAVE_MEMFD
gives a warning (gcc 5.4.1, Ubuntu)
pulsecore/shm.c: In function 'sharedmem_create':
pulsecore/shm.c:208:5: warning: "HAVE_MEMFD" is not defined [-Wundef]
#if HAVE_MEMFD
use #ifdef or #if defined() to check for presence of a #define
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
Both input and output device were chosen with the same device number.
This is problematic as those numbers don't have to correspond.
Additionally the input device was named after the output device. This
commit adresses both issues by providing specific parameters for each
type.
This is a rebase of Wim Taymans patch to support the HSP headset role that has
somehow been forgotten. Original patch can be found at
https://lists.freedesktop.org/archives/pulseaudio-discuss/2015-February/023242.html
Rebase and minor changes by Georg Chini.
In addition to the HSP Audio Gateway, also add support for the headset
role in the native bluetooth backend. In this role, pulseaudio is used as
headset.
In the headset role, we create source and sink to receive and send the samples
from the gateway, respectively. Module-bluetooth-policy will automatically load
loopback modules to link these to a sink and source for playback. Because this
makes the source the speaker and the sink the microphone, we need to reverse the
roles of source and sink compared to the gateway role.
In the gateway role, adjusting the sink volume generates a +VGS command to set
the volume on the headset. Likewise, receiving AT+VGS updates the sink volume.
In the headset role, receiving a +VGS should set the source volume and any
source volume changes should be reported back to the gateway with AT+VGS.
Clang didn't like the variable length array:
pulsecore/iochannel.c:358:17: error: fields must have a constant size:
'variable length array in structure' extension will never be supported
uint8_t data[CMSG_SPACE(sizeof(int) * nfd)];
^
Commit 451d1d6762 introduced the variable length array in order to have
the correct value in msg_controllen. This patch reverts that commit and
uses a different way to achieve the same goal.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=99458
do...while not reachable, loop should try different ports in case EADDRINUSE is returned
Coverity ID: #1398161
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
the modeling file help to avoid false positives and increase scanning
accuracy by explaining code Coverity can't see (out of tree libraries);
the model file must be uploaded by an admin to:
https://scan.coverity.com/projects/pulseaudio?tab=analysis_settings
the pa_assert_se() macro needs to be rewritten for Coverity so that
the assignment is not declared a side-effect
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
The previous commit, "loopback: Initialize latency at startup and during
source/sink changes", was an old version of the patch that got
accidentally pushed instead of the last version. This commit does the
changes that were omitted when applying the old patch.
The current code does not make any attempt to initialize the end-to-end latency
to a value near the desired latency. This leads to underruns at startup because
the memblockq is initially empty and to very long adjustment times for long
latencies because the end-to-end latency at startup is significantly shorter
than the desired value.
This patch initializes the memblockq at startup and during source or sink changes
so that the end-to-end latency will be near the configured value. It also ensures
that there are no underruns if the source is slow to start and that the latency
does not grow too much when the sink is slow to start by adjusting the length of
the memblockq until the source has called push for the first time and the sink
has called pop for the second time. Waiting for the second pop is necessary
because the sink has not been started when the first pop is called.
For clarity, variables have been separated into input, output and main thread
variables.
FlatCarbon was the flattened Carbon version used in Mac OS Classic
(i.e., pre Mac OS X.)
It was shipped as legacy software until 10.8, then dropped completely.
Using CoreServices is good enough, manually including FlatCarbon headers
only lead to build failures for users who had old versions of Xcode
lingering around their machines.
v2: don't accidentally drop the OS X semaphore implementation.
A recent patch changed the MTU size from the default value of 48 to the value
returned by getsockopt(). This breaks HSP for some setups. To circumvent the
problem, this patch introduces a boolean parameter "autodetect_mtu" for
module-bluetooth-discover, module-bluez5-discover and module-bluez5-device to
make this use of getsockopt() configurable.
This serves to explicitly document the various cases we deal with in
pa_sink_update_rate()/pa_source_update_rate() rather than have some of
them hidden behind the initialisation of desired_rate.
This adds an "avoid-resampling" option to daemon.conf that makes the
daemon try to use the stream sample rate if possible (the device needs
to support it, which currently only ALSA does), and there should not be
any other stream connected).
This should enable some of the "audiophile" use-cases where users wish
to play high sample rate audio files without resampling.
We still will do conversion if sample formats don't match, though. This
means that if you want to play 96 kHz/24 bit audio without any
modification the default format will need to be set to be 24-bit as
well. This will force all streams to be upconverted, which, other than
the wasted resources, should be relatively harmless.
The RTSP client is not waiting anymore a new header after the
previous one (which can never occurs if RAOP is disconnected)
but after sending a command.
This patch fixes Issue #36.
https://github.com/hfujita/pulseaudio-raop2/issues/36
This patch is based on a similar idea as the previous one -- disabling
the flag right after the session is getting closed, rather than waiting
for a response from the server.
This patch fixes the issue #31.
https://github.com/hfujita/pulseaudio-raop2/issues/31
This patch sets c->is_recording = false when the RTSP FLUSH command
is issued. This avoids a race between the server response and
the record activation in some cases.
Regression introduced in commit 8c6407f:
raop: Merge TCP and UDP code paths + refactoring
Anyway, we need to determine if initial volume has to be setup before
sending RECORD or after:
- Setting it up *before* shouldn't be a problem: sink.c waits for
CONNECT state, set the volume and client.c triggers RECORD only once
he's got the SET_PARAMETER reply from server.
- Setting it up *after* seems to be more difficult if we try not to
send any audio before receiving the SET_PARAMETER reply form server. A
solution may be to send SET_PARAMETER just after the RECORD server
response is received and hope that it get processed by server during the
2sec latency/buffering time...
Attached patch implement that last solution. Works for me, but I cannot
guaranty it will with your hardware...
Some time one device announces multiple addresses (e.g. IPv4 one
and IPv6 one). Or some user may own multiple RAOP devices with
the same model name.
This patch adds device port to device description so that users
can distinguish appropriate RAOP sink by its address.
This patch switch the packet-buffer to use core memory pool instead of
manually allocating the room required for storing TCP/UDP packets. Packets
are now stored using pa_memchunk instead of internal struct. Quite a few
malloc saved compare to previous design.
ALAC encoding is to be prefered simply because ALAC audio packet reverse-
engineering and implementation is in better shape than raw PCM. Sending ALAC
audio does not mean compressing audio and thus linking an external library to
do so. ALAC packets has the ability to carry uncompressed PCM frames, and
that's what is implemented at the time.
TCP and UDP implementation are following two diffrent code path while code
logic is quite the same. This patch merges both code path into a unique one
and, thus, leads to a big refactoring. Major changes include:
- moving sink implementation to a separate file (raop-sink.c)
- move raop-sink.c protocol specific code to raop-client.c
- modernise RTSP session handling in TCP mode
- reduce code duplications between TCP and UDP modes
- introduce authentication support
- TCP mode does not constantly send silent audio anymore
About authentication: OPTIONS is now issued when the sink is preliminary
loaded. Client authentication appends at that time and credential is kept
for the whole sink lifetime. Later RTSP connection will thus look like this:
ANNOUNCE > 200 OK > SETUP > 200 OK > RECORD > 200 OK (no more OPTIONS). This
behaviour is similar to iTunes one.
Also this patch includes file name changes to match Pulseaudio naming
rules, as most of pulseaudio source code files seem to be using '-'
instead of '_' as a word separator.
RAOP authentication is using standard HTTP challenge-response authentication
scheme. This patch adds two helper functions that generate the proper hash
(for both techniques) given a username, a password and session related tokens.
MD5 hashing will be needed during the authentication process.
Original patch by Martin Blanchard. Patch splitted by
Hajime Fujita <crisp.fujita@nifty.com>.
Base64 implementation is now in a common file called raop_util.c.
Old Base64 files are removed but copyright is preserved.
Original patch by Martin Blanchard, patch splitted by
Hajime Fujita <crisp.fujita@nifty.com>.
When playback stops, a FLUSH command is send to the server and the sink
goes to IDLE. If playback resumes quickly, sink goes back to RUNNING
(without being SUSPENDED) and the sink should just start streaming again.
This patch implements this behaviour.
This patch adds an RTP audio packet retransmission support and a
circular buffer implementation for it.
This patch was originally written by Matthias Wabersich [1] and
later debugged and integrated into the latest tree by Hajime Fujita
[1]: https://bugs.freedesktop.org/show_bug.cgi?id=42804#c44
During the discovery phase, raop servers send their capabilities
(supported encryption, audio codec...). These should be passed to the
raop sink via module's arguments.
Original patch written by Martin Blanchard, then modified by Hajime
Fujita <crisp.fujita@nifty.com> based on review comments by
Anton Lundin <glance@acc.umu.se>.
Now resolver_cb always dtrdup()s string blocks given by Avahi,
to make the code easier to maintain.
There are two versions in the RAOP protocol; one uses TCP and the
other uses UDP. Current raop implementation only supports TCP
version.
This patch adds an initial UDP protocol support for RAOP.
It is based on Martin Blanchard's work
(http://repo.or.cz/w/pulseaudio-raopUDP.git/shortlog/refs/heads/raop)
which is inspired by Christophe Fergeau's work
(https://github.com/zx2c4/pulseaudio-raop2).
Matrin's modifications were edited by Hajime Fujita, so that it
would support both TCP and UDP protocol in a single module.
Also this patch includes a fix that was found thanks to Matthias,
who reported that his ALAC
codec support fixed the issue.
https://bugs.freedesktop.org/show_bug.cgi?id=42804#c30
This macro compares if the given two strings, with the maximum length
of n, are equal. Useful for strings that are not NULL-terminated.
Reviewed-by: Anton Lundin <glance@acc.umu.se>
Constants should be declared simply with "const". With struct members,
"static" means that all struct instances share the same variable, i.e.
all instances always see the same value. That's of course already
implied in the concept of "constant". Newer Vala versions don't allow
mixing "const" and "static".
In alsa-lib, snd_pcm_hw_params() internally calls snd_pcm_prepare(), thus
user space applications have no need to call snd_pcm_prepare() after calls
of snd_pcm_hw_params(). An explicit calls of snd_pcm_prepare() is expected
in a case to recover PCM substreams.
Current implementation of PulseAudio modules for ALSA playbacking/capturing
results in double calls of snd_pcm_prepare(). The second call for hw plugin
of alsa-lib executes ioctl(2) with SNDRV_PCM_IOCTL_PREPARE command in state
of SNDRV_PCM_STATE_PREPARED for the PCM substream. This has no effects to
the PCM substream as long as corresponding drivers are implemented
correctly.
This commit removes the second call for the reason.
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Issue: When HFP/HSP profile is used with certain BT chipsets, the
audio sounds heavily distorted, with very slow playback full of noise.
During recording, the samples are dropped and it distorts the recorded
audio samples.
The root cause of both the issues are related to the fixed MTU sizes
in the PA stack, which is 48 bytes. Here, the BT chipset CC256x had
180 bytes MTU and it was being under-utilized and the rate at which
the samples were being accepted where not matching the expected rate,
and hence the distortion.
Solution: The appropriate solution to this problem is by reading the
MTU size of the SCO socket using getsockopts dynamically.
BugLink: http://bit.ly/2gDpGPv
BugLink: http://bit.ly/2hQsARK
The current build script hardcodes the $pkglibdir in the padsp command.
This works and is a reasonable default. However, distributions that
know where they install, can override this path and thus make padsp
work for any architecture that has the library installed by using the
following configure argument:
--with-pulsedsp-location='/usr/\\$$LIB/pulseaudio'
This works because ld.so considers $LIB a variable that will expand to
several location paths, depending on the architecture of the binary
being executed.
In debian, for example, this would work for libpulsedsp.so installed in
/usr/lib/x86_64-linux-gnu/ for amd64 and /usr/lib/i386-linux-gnu/ for
i386, with a single padsp command.
The autoreconf invocation below will already pick up any overrides the
user might have made to their LIBTOOLIZE variable. Overriding it here
will break on Darwin systems where libtoolize is not called glibtoolize,
and is not necessary, so just remove it.
X11 has its own bell volume setting, controlled with the "xset b"
command. If we use that volume, then the "System Sounds" slider in
pavucontrol doesn't affect the x11-bell sample volume, which in my
opinion is a bad thing. Ignoring the volume suggestion from X11 allows
module-stream-restore to apply the "event" role volume.
Any compiler flags should be set before asking the compiler to check for
thread-local storage with AX_TLS, since compiler flags (in this case
-mmacosx-version-min=10.5) can influence the outcome of that check.
* remove suggestion of '-C' due to incorrect ordering of options for
_arguments
* avoid suggesting multiple options
* add suggestion of "--", followed by executable programs
* after "--server=<hostname>" or "-s <hostname>" suggest "--"
* after "-- <program>" continue standard tab completion
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=98639
Not all VOIP applications (specially those which use alsa) set media.role to
phone. This means we need some heuristic to determinate if we want to switch
from a2dp to hsp profile based on number and types of source output (recording)
streams.
And also some people want to use their bluetooth headset (with microphone) as
their default recording device but some do not want to because of low quality.
This patch implements optional heuristic which is disabled by default. It is
disabled by default to not break experience of current pulseaudio users because
heuristic cannot be optimal. Heuristic is implemented in module-bluetooth-policy
module and decide if pulseaudio should switch to a hsp profile or not. It checks
if there is some source output with pass all these conditions:
* does not have set media.role
* does not use peak resample method (which is used by desktop volume programs)
* has assigned client/application (non virtual stream)
* does not record from monitor of sink
And if yes it switch to hsp profile.
By default this heuristic is disabled and can be enabled when loading module
module-bluetooth-policy with specifying parameter auto_switch=2
Because it is disabled by default nobody will be affected by this change unless
manually change auto_switch parameter.
Signed-off-by: Pali Rohár <pali.rohar@gmail.com>
In the current RTSP implementation, there is a vulnerable window
between the RTSP object creation and the URL initialization.
If any RTSP command is issued during this period, it will lead to
crash by assertion violation.
This patch introduces pa_rtsp_exec_ready(), which returns if it is
safe to issue RTSP commands.
Reviewed-by: Anton Lundin <glance@acc.umu.se>
Add a function performing a call to the OPTIONS request; also,
in some special cases, tuning transport parameters is required (default:
"RTP/AVP/TCP;unicast;interleaved=0-1;mode=record") ! The RAOP client for
example needs to overwrite them.
Reviewed-by: Anton Lundin <glance@acc.umu.se>
pa_ioline_close does not free the ioline structure itself, so we
have to unref the structure if we want to free it.
Reviewed-by: Anton Lundin <glance@acc.umu.se>
pa_socket_client_new_string() did not work as expected when an IPv6
address string like "2001:db8::1" is passed as the "name" parameter.
This is because the name parameter is then passed to pa_parse_address(),
which thinks the last colon as a separator between hostname (or address)
and a port number. To prevent pa_parse_address() from doing this, an IPv6
address must be bracketed with "[]" (e.g. "[2001:db8::1]"). [1]
This patch fixes pa_socket_client_new_string() so that it internally
adds brackets to an IPv6 address. This decision is based on a
discussion at [2].
[1]: http://lists.freedesktop.org/archives/pulseaudio-discuss/2014-October/022010.html
[2]: http://lists.freedesktop.org/archives/pulseaudio-discuss/2014-November/022401.html
Reviewed-by: Anton Lundin <glance@acc.umu.se>
The intuitive meaning of "missing" would be the difference between
tlength and the current queue length, and that's how memblockq-test
assumed pa_memblockq_pop_missing() to define the term "missing", but
that was an incorrect assumption, causing the last
pa_memblockq_pop_missing() return value assertion to fail.
This patch fixes the failing assertion and adds some comments about how
the "missing" and "requested" variables in memblockq work.
The function isn't used anywhere else than memblockq-test. Also, the
function is confusing, because it defines "missing" differently than
pa_memblockq_pop_missing(). pa_memblockq_missing() calculated the
missing amount like this:
missing = tlength - length,
where "length" is the current queue length. pa_memblockq_pop_missing(),
on the other hand, calculates the missing amount like this:
missing = tlength - length - requested,
where "requested" is an internal variable that keeps track of how much
the server has requested data from the client and how much of the
requests are yet to be fulfilled by the client.
memblockq-test is broken at the moment, because it assumes that
pa_memblockq_pop_missing() calculates "missing" the same way that
pa_memblockq_missing() used to calculate it. A patch for fixing that
will follow.
This reverts commit 74251f0786.
The reverted commit was not intended to make any behavioral changes, but
it broke at least the case where a client writes more data than the
server has requested.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=99211
Current pacat code reads whatever available from STDIN and writes
it directly to the playback stream. A minimal buffer is created
for each read operation; no further reads are then allowed unless
earlier read buffer has been fully consumed by a stream write.
While quite simple, this model breaks upon the new requirements of
writing only frame-aligned data to the stream (commits #1 and #2).
The kernel read syscall can return a length much smaller than the
frame-aligned size requested, leading to invalid unaligned writes.
This can easily be reproduced by choosing a starved STDIN backend:
pacat /dev/random pa_stream_write() failed: EINVAL
echo 1234 | pacat pa_stream_write() failed: EINVAL
or by playing an incomplete WAV file in raw, non-paplay, mode.
So guard against such incomplete kernel reads by writing only in
frame-aligned sizes, while caching any trailing partial frame for
subsequent writes.
Other operation modes are not affected. Non-raw paplay playback is
handled by libsndfile, ensuring complete reads, and recording mode
just writes to the STDOUT fd without any special needs.
CommitReference #1: 22827a5e1e
CommitReference #2: 150ace90f3
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=98475
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=77595
Suggested-by: David Henningsson <diwic@ubuntu.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Streams are detached when they are removed or moved away from a device,
or when a filter device that they're connected to is removed or moved.
If these cases overlap, a crash will happen due to "double-detaching".
This can happen if a filter sink is removed, and a stream connected to
that filter sink removes itself when its sink goes away.
Here are the steps in more detail: When a filter sink is unloaded, first
it will unlink its own sink input. This will cause the filter sink's
input to be detached. The filter sink propagates the detachment to all
inputs connected to it using pa_sink_detach_within_thread(). After the
filter sink is done unlinking its own sink input, it will unlink the
sink. This will cause at least module-combine-sink to remove its sink
input if it had one connected to the removed filter sink. When the
combine sink removes its sink input, that input will get detached again,
and a crash follows.
We can relax the assertions a bit, and skip the detach() call if the
sink input is already detached.
I think a better fix would be to unlink the sink before the sink input
when unloading a filter sink - that way we could avoid the
double-detaching - but that would be a much more complicated change. I
decided to go with this simple fix for now.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=98617
The functions that call attach()/detach() for all streams on a sink or
source didn't update the "attached" flag accordingly. Since the flag is
only used in assertions, this omission didn't cause any harm in normal
use.
The "attached" flag is only used for asserting that the stream is in the
expected state when attaching or detaching.
Sometimes the flag was checked and updated before calling the attach or
detach callback, and sometimes after. I think it makes more sense to
always check it before calling the callback.
At the time the "unlink post" hook is fired, the stream is not any more
connected to its old device, so it makes sense to reset the sink/source
pointer to NULL before firing the hook. If this is not done, the pointer
may become stale during the "unlink post" hook, because
module-bluetooth-policy does a card profile change in its "unlink post"
callback, so even if the pointer is valid when module-bluetooth-policy's
callback is called, it will be invalid in subsequent callbacks.
In the "unlink post" hook it's not guaranteed that the stream's old
device exists any more, so let's use the "unlink" hook that is safer.
For example, module-bluetooth-policy does a card profile change in the
source-output "unlink post" hook, which invalidates the source-output's
source pointer.
When the "unlink" hook is fired, the stream is still linked to its
device, which affects the return values of the check_suspend()
functions. The unlinked streams should be ignored by the check_suspend()
functions, so I had to add extra parameters to those functions.
This fixes a crash that happens if the bluetooth headset is the only
non-monitor source in the system and the last "phone" stream dies.
When the stream dies, the native protocol calls pa_source_output_unlink()
and would call pa_source_output_unref() next, but without this patch,
things happen during the unlinking, and the unreffing ends up being
performed on a stream that is already freed.
pa_source_output_unlink() fires the "unlink" hook before doing anything
else. module-bluetooth-policy then switches the headset profile from HSP
to A2DP within that hook. The HSP source gets removed, and at this point
the dying stream is still connected to it, and needs to be rescued.
Rescuing fails, because there are no other sources in the system, so the
stream gets killed. The native protocol has a kill callback, which again
calls pa_source_output_unlink() and pa_source_output_unref(). This is
the point where the native protocol drops its own reference to the
stream, but another unref call is waiting to be executed once we return
from the original unlink call.
I first tried to avoid the double unreffing by making it safe to do
unlinking recursively, but I found out that there's code that assumes
that once unlink() returns, unlinking has actually occurred (a
reasonable assumption), and at least with my implementation this was not
guaranteed. I now think that we must avoid situations where unlinking
happens recursively. It's just too hairy to deal with. This patch moves
the bluetooth profile switch to happen at a time when the dead stream
isn't any more connected to the source, so it doesn't have to be
rescued or killed.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=97906
Bug 96741 shows a case where an assertion is hit, because
pa_asyncq_new() failed due to running out of file descriptors.
pa_asyncq_new() is used in only one place (not counting the call in
asyncq-test): pa_asyncmsgq_new(). Now pa_asyncmsgq_new() can fail too,
which requires error handling in many places. One of those places is
pa_thread_mq_init(), which can now fail too, and that needs additional
error handling in many more places. Luckily there weren't any places
where adding better error handling wouldn't have been easy, so there are
many changes in this patch, but they are not complicated.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=96741
pa_memblockq_push() expects all memchunks to be aligned to PCM frame
boundaries, and that means both the index and length fields of
pa_memchunk. pa_rtp_recv(), however, used a memblock for storing both
the RTP packet metadata and the actual audio data. The metadata was
"removed" from the audio data by setting the memchunk index
appropriately, so the metadata stayed in the memblock, but it was not
played back. The metadata length is not necessarily divisible by the PCM
frame size, which caused pa_memblock_push() to crash in an assertion
with some sample specs, because the memchunk index was not properly
aligned. In my tests the metadata length was 12, so it was compatible
with many configurations, but eight-channel audio didn't work.
This patch adds a separate buffer for receiving the RTP packets. As a
result, an extra memcpy is needed for moving the audio data from the
receive buffer to the memblock buffer.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=96612
When unloading a module, lt_dlclose() may remove the module from memory.
If a module unloads itself, it's not safe to call lt_dlclose()
synchronously from pa_module_unload(), because the execution may return
to the module code that was removed from memory. To avoid this
situation, let's postpone lt_dlclose() until it's safe to call it.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=96831
The module doesn't build any more[1], and when starting to investigate
the build failure, I asked the module author if he'd know something
about the breakage. He said that he didn't know about backward
compatibility problems with libxen, but more importantly, he said that
the module probably doesn't have any users[2]. It doesn't make sense to
keep maintaining a module that doesn't have users, so let's drop it.
[1] https://bugs.freedesktop.org/show_bug.cgi?id=98793
[2] https://lists.freedesktop.org/archives/pulseaudio-discuss/2016-November/027172.html
SW: Pulseaudio 8.0 / BlueZ 5.39
Symptoms:
While disconnecting/reconnecting a paired bluetooth headset (LG HBS750)
audio fails roughly on every other connection.
On a failed connection "pactl list cards" shows the bluetooth device's
card but "Active Profile: off". Issuing "pacmd set-card-profile X
a2dp_sink" makes audio work immediately.
I realized that when this happened, the previous disconnection did not
remove the card, instead it was only configured for "Active Profile:
off" but otherwise left in place.
Upon looking at PA debug logs I saw that the transport for the a2dp_sink
was first set into disconnected state and then into idle state. In
"device_connection_changed_cb()" this causes the
"pa_bluetooth_device_any_transport_connected()" return true and the
module-bluez5-device is not unloaded.
Further investigation shows that this is caused by a race of
module-bluez5-device.c:thread_func() and
MediaPoint1::ClearConfiguration().
When the FD in thread_func() is closed (POLLHUP) an
BLUETOOTH_MESSAGE_STREAM_FD_HUP message is sent into the main thread.
The handler of this message unconditionally sets the transport into IDLE
state. This is a problem if it has already been set into DISCONNECTED
state.
Executing below command will not produce any audio:
pacat --channels=3 /dev/urandom
Turns out that pa_stream_write() breaks large audio buffers into
segments of the maximum memblock size available -- a value which
is not necessarily frame aligned.
Meanwhile the server discards any non-aligned client audio, as a
security measure, due to some earlier reported daemon crashes.
Thus divide sent audio to the expected aligned form.
CommitReference-1: 22827a5e1e
CommitReference-2: 150ace90f3
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=98475
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=77595
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Users reported pacat crashes when playing certain multi-channel
audio. For example:
pacat --channels=2 /dev/zero works
pacat --channels=3 /dev/zero pa_stream_write() failed: EINVAL
pacat --channels=4 /dev/zero works
pacat --channels=5 /dev/zero pa_stream_write() failed: EINVAL
pacat audio playback is pipe-like, from STDIN to PA write stream.
Meanwhile STDIN "ready to read" events got regularly triggered
before the write stream was even created, or at moments where the
stream could not accept any more audio.
In these out-of-sync cases, the write stream could not report the
appropriate buffer lengths it accepts, thus a default of 4K bytes
was chosen -- compatible by luck with some channel counts and
incompatible with others.
Instead of choosing a faulty default in these scenarios, mute the
the STDIN events until the write stream is available & queriable.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=98475
Reported-by: Tanu Kaskinen <tanuk@iki.fi>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Although such 9.0 clients support memfd transport, they have an
iochannel bug that would break memfd audio if they're run in 32
bit mode over a 64-bit kernel. Influence them to use the POSIX
shared memory model instead.
Also bump the protocol version to exclusively mark such v9.0
libraries. Check commit 451d1d6762 for further details.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=97769
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Users reported audio breakage for 32-bit pulse clients connected
to a 64-bit server over memfds. Investigating the issue further,
the problem is twofold:
1. iochannel's file-descriptor passing code is liberal in what it
issues: produced ancillary data object's "data" section exceeds
length field. How such an extra space is handled is a grey area
in the POSIX.1g spec, the IETF RFC #2292 "Advanced Sockets API
for IPv6" memo, and the cmsg(3) manpage.
2. A 64-bit kernel handling of such extra space differs by whether
the app is 64-bit or 32-bit. For 64-bit apps, the kernel
smartly ducks the issue. For 32-bit apps, an -EINVAL is
directly returned; that's due to a kernel CMSG header traversal
bug in the networking stack "32-bit sockets emulation layer".
Compare Linux Kernel's socket.h cmsg_nxthdr() code and the
32-bit emulation layer version of it at net/compat.c
cmsg_compat_nxthdr() for further info. Notice how the former
graciously ignores incomplete CMSGs while the latter _directly_
complains about them -- as of kernel version 4.9-rc5.
(A kernel patch is to be submitted)
Details:
iochannel typically uses sendmsg() for passing FDs & credentials.
>From RFC 2292, sendmsg() control data is just a heterogeneous
array of embedded ancillary objects that can differ in length.
Linguistically, a "control message" is an ancillary data object.
For example, below is a sendmsg() "msg_control" containing two
ancillary objects:
|<---------------------- msg_controllen---------------------->|
| |
|<--- ancillary data object -->|<----- ancillary data object->|
|<------- CMSG_SPACE() ------->|<------- CMSG_SPACE() ------->|
| | |
|<-------- cmsg_len ------->| |<-------- cmsg_len ------->| |
|<------- CMSG_LEN() ------>| |<------- CMSG_LEN() ------>| |
| | | | |
+-----+-----+-----+--+------+--+-----+-----+-----+--+------+--+
|cmsg_|cmsg_|cmsg_|XX|cmsg_ |XX|cmsg_|cmsg_|cmsg_|XX|cmsg_ |XX|
|len |level|type |XX|data[]|XX|len |level|type |XX|data[]|XX|
+-----+-----+-----+--+------+--+-----+-----+-----+--+----+-+--+
^^^^^^^ Ancil Object #1 ^^^^^^^ Ancil Object #2
(control message) (control message)
^
|
+--- sendmsg() "msg_control" points here
Problem is, while passing FDs, iochannel's code try to avoid
variable-length arrays by creating a single cmsg object that can
fit as much FDs as possible:
union {
struct cmsghdr hdr;
uint8_t data[CMSG_SPACE(sizeof(int) * MAX_ANCIL_DATA_FDS)];
} cmsg; ^^^^^^^^^^^^^^^^^^
Most of the time though the number of FDs to be passed is less
than the maximum above, thus "cmsg_len" is set to the _actual_ FD
array size:
cmsg.hdr.cmsg_len = CMSG_LEN(sizeof(int) * nfd);
^^^
This inconsistency tricks the kernel into thinking that we have 2
ancillay data objects instead of one! First cmsg is valid as
intended, but the second is instantly _corrupt_ since it has a
cmsg_len size of 0 -- thus failing kernel's CMSG_OK() tests.
For 32-bit apps on a 32-bit kernel, and 64-bit apps over a 64-bit
one, the kernel's own CMSG header traversal macros just ignore the
second "incomplete" cmsg. For 32-bit apps over a 64-bit kernel
though, the kernel 32-bit socket emulation macros does not forgive
such incompleteness and directly complains of invalid args (due to
a subtle bug).
Avoid this ugly problem, which can also bite us in a pure 64-bit
environment if MAX_ANCIL_DATA_FDS got extended to 5 FDs, by
setting "cmsg_data[]" array size to "cmsg_len".
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=97769
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Openssl 1.1.0 made all structs opaque, which caused a build failure in
raop_client.c. The struct member assignments are now replaced with a
call to RSA_set0_key().
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=96726
Reviewed-by: Felipe Sateler <fsateler@debian.org>
KTH is a Swedish institution of higher education, and in its full name
spelled Kungliga Tekniska högskolan.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Bluez5 uses different profile names as bluez4, so we need to check for
a2dp_sink and headset_head_unit too for bluez5 support.
Signed-off-by: Pali Rohár <pali.rohar@gmail.com>
module-xenpv-sink.c - In pa__init(...), memory for pa_modargs *ma is not released before returning from function.
Signed-off-by: Deepak Srivastava <srivastava.d@samsung.com>
systemd-hostnamed provides an icon for the machine it is running on.
If it is running, module-zeroconf-publish uses this icon for the
'icon-name' attribute in the Avahi properties. module-zeroconf-discover
passes this icon to module-tunnel using the module parameter
{sink/source}_properties.
This allows to display a portable, desktop or phone instead of
the generic sound card icon.
passing an invalid sample_spec to
pa_sample_size_of_format(),
pa_frame_size(),
pa_bytes_per_second(),
pa_bytes_to_usec(),
pa_usec_to_bytes()
currently gives a result of 0
this is problematic as
(a) it leads to many potential divide-by-zero issues flagged by Coverity,
(b) pa_sample_spec_valid() is called often and the mostly unnecessary validation
of the sample_spec cannot be optimized away due to pa_return_val_if_fail()
(c) nobody checks the result for 0 and the behaviour is not documented
this patch replaces pa_return_val_if_fail() with pa_assert()
note that this commit changes the API!
note that pa_return_val_if_fail() strangely logs an assertion, but then happily
continues...
fixes numerious CIDs
PA_PAGE_SIZE using sysconf() may return a negative number
CID 1137925, CID 1137926, CID 1138485
instead of calling sysconf() directly, add function pa_page_size()
which uses the guestimate 4096 in case sysconf(_SC_PAGE_SIZE) fails
using PA_ONCE to only evaluate sysconf() once
CID 1353122
this is a false-positive because
if (dbus_message_has_interface(p->message, "org.bluez.Manager") ||
dbus_message_has_interface(p->message, "org.bluez.Adapter"))
d = NULL;
else if (!(d = pa_hashmap_get(y->devices, dbus_message_get_path(p->message)))) {
pa_log_warn("Received GetProperties() reply from unknown device: %s (device removed?)",
dbus_message_get_path(p->message));
goto finish2;
}
d can be NULL only if p->message interface is org.bluez.Manager or
org.bluez.Adapter. If
dbus_message_is_method_call(p->message, "org.bluez.Device", "GetProperties")
returns true, we know that the interface is org.bluez.Device.
thanks, Tanu!
pa_ncpu() is supposed to report the number of processors available on
the system. For that, it currently calls sysconf(_SC_NPROCESSORS_CONF).
However, since the operating system can disable individual processors,
we should call sysconf(_SC_NPROCESSORS_ONLN) to determine the number
of processors currently available [1]. Consequently, the once-test will
fail since pthread_setaffinity_np() is called with CPUs that are
currently not available.
It might also be advisable to change the code in the future to use CPU
sets on Linux as even the suggested change is not 100% safe but at least
it improves over the existing code. If PulseAudio was to be run in a CPU
set [2], the number of processors available to PulseAudio could be even
less than the number of CPUs currently online (_SC_NPROCESSORS_CONF).
[1] https://www.gnu.org/software/libc/manual/html_node/Processor-Resources.html
[2] http://man7.org/linux/man-pages/man7/cpuset.7.html
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=96809
Signed-off-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
The pipe buffer is likely to be a power of 2 (e.g. 4096 bytes). This
works nicely for 16 bit stereo samples but breaks when using 24 bit
samples.
This patch aligns the buffer using pa_frame_align().
create_card_profile() used to get called separately for HSP and HFP,
so if a headset supports both profiles, a profile named
"headset_head_unit" would get created twice. The second instance would
get immediately freed, so that wasn't a particularly serious problem.
However, I think it makes more sense to create the profile only once.
This patch makes things so that before a profile is created, we check
what name that profile would have, and if a profile with that name
already exists, we don't create the profile.
A couple of Yocto releases (jethro and krogoth) have non-upstream
patches that suffer from this double creation. The patches add
associations between profiles and ports, and those associations use
the profile name as the key. When the second profile gets freed, the
associations between the profile and its ports get removed, and since
the profile name is used as the key, this erroneously affects the
first profile too. Crashing ensues.
BugLink: https://bugzilla.yoctoproject.org/show_bug.cgi?id=10018
Add transport_set_state() that encapsulates changing the variable,
logging and firing the change hook.
I also made a cosmetic change to the corresponding BlueZ 5 log
message so that both messages have the format that I like.
A hashmap is more convenient than a linked list for storing the UUIDs,
so change the BlueZ 4 code accordingly.
Rename the BlueZ 4 UUID constants to match the BlueZ 5 naming.
The only changes to the BlueZ 5 code are the addition of one comment
and making another comment a bit clearer.
The properties_received flag affects whether the device should be
considered valid, so let's update the valid flag after setting the
properties_received flag.
There's a call to device_update_valid() anyway later when setting
the device adapters, so this change isn't strictly necessary, but
this makes it more obvious that the code is correct (and less
fragile).
module-card-restore should only restore the initial state of new
cards, but profile_available_changed_callback() changed the profile
whenever the saved profile became available. That caused interference
with module-bluetooth-policy, which also sets card profiles based on
the availability changes.
The original reason for having this code was to work around the
problem that bluetooth cards used to be created with only one profile
available, and other profiles would become available soon after the
card creation. Now the bluetooth card creation is delayed until all
profiles are available, so this bad workaround can be removed.
Discussion:
https://lists.freedesktop.org/archives/pulseaudio-discuss/2016-August/026575.html
The CONNECTION_CHANGED hook is used to notify the discovery module
about new and removed devices. When a bluetooth device connects, the
hook used to be called immediately when the first profile connected.
That meant that only one profile was marked as available during the
card creation, other profiles would get marked as available later.
That makes it hard for module-card-restore to restore the saved
profile, if the saved profile becomes available with some delay.
module-card-restore has a workaround for this problem, but that turned
out to interfere with module-bluetooth-policy, so the workaround will
be removed in the next patch.
The BlueZ 4 code doesn't need changes, because we use the
org.bluez.Audio interface to get a notification when all profiles are
connected.
Doesn't really affect logic, but Coverity reports this as dead-code, and
I figure it makes sense to be consistent about our use of HAVE_MEMFD.
CID: 1352045
tests/core-util-test.c uses ck_assert_int_lt() which was introduced
in check 0.9.10
make this dependency (with --enable-tests) explicit in configure.ac
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
on oldish Ubuntu 12.04:
tests/core-util-test.c: In function ‘main’:
tests/core-util-test.c:269:66: error: ‘SIGABRT’ undeclared (first use in this function)
tcase_add_test_raise_signal(tc, modargs_test_replace_fail_1, SIGABRT);
Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
With this patch module-bluetooth-policy automatically switch from a2dp profile
to hsp profile if some VOIP application with media.role=phone wants to start
recording audio.
By default a2dp profile is used for listening music, but for VOIP calls is
needed profile with microphone support (hsp). So this patch will switch to
hsp profile if some application want to use microphone (and specify it in
media.role as "phone). After recording is stopped profile is switched back
to a2dp. So this patch allows to use bluetooth microphone for VOIP applications
with media.role=phone automatically without need of user interaction.
Signed-off-by: Pali Rohár <pali.rohar@gmail.com>
It was a very confusing state variable that required a lot of
fiddling. It was also redundant in that it can be computed from
the other variables, removing any risk of it getting out of sync.
In the same spirit, make sure "requested" also always contains a
sane value, even though it may not be used by every caller.
This involves in particular pa_memblockq_missing() and
pa_memblockq_pop_missing(). The test demonstrates that the latter
doesn't work as expected. It should report whenever queue level is
drained below target level. Instead, it reports any case that the queue
level is drained, even when it is still above target level.
- Set the loglevel once in the main entry code instead of in each test function.
- Check pool allocation succeeded.
- Reduce code by using utility function to allocate chunks.
- Improve coverage by using utility function to validate queue invariants.
In particular, the relations between base, minreq, tlength, length,
missing, maxlength follow certain rules. On change, these invariants can
be violated, which requires additional code to restore them. Setting one
value can thus cause a cascade of changes. This utility function can
assert those invariants after changing something.
Having it handled in the callers proved to be a poor fit as it
became difficult to handle a shrinking minreq sanely. It could end
up in a state where the request was never sent downstream to the
client.
The reason for depending on the socket unit is rather unobvious, so
let's add a comment to help people reading the service unit file. Felipe
Sateler explained the rationale well in the commit message of
7cb524a77b, so I just copied the same text into the comment.
This commit fixes two problems:
1. Because there are no implicit dependencies between sockets and services,
the socket, as set up by systemd will race with the socket, as set up
by the pulseaudio daemon. This can cause the pulseaudio.socket unit to
fail (even though the pulseaudio service started just fine), which can
confuse users.
2. While it is possible to use the service without the socket, it is not
clear why it would be desirable. And a user installing pulseaudio and
doing `systemctl --user start pulseaudio` will not get the socket
started, which might be confusing and problematic if the server is to
be restarted later on, as the client autospawn feature might kick in.
We do support system mode for the cases where it makes sense, so it's
really not sensible to be unconditionally snarky at our users for doing
it.
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
Replace the current latency controller with a modified P-controller. For
better readability separate the controller function. For small latency
differences, the controller forms a classical P-controller while it saturates
at 1% deviation from the base rate for large latency differences.
After switching source or sink, call adjust_rates after a third of a second
instead of waiting one full adjust time. This will ensure that latency regulation
starts as soon as possible.
Restaring the timer and obtaining the latency snapshots belong to the timer callback.
To maintain an adjust time as near as possible to the configured value, the timer is
now restarted immediately at the beginning of the timer callback.
To improve the overall latency estimation, the delay between the two snapshots
is taken into account. To minimize the snapshot delay, the order of the snapshots
is reverted. Additionally the latency at the base rate is calculated. It will be
used later as the input to the latency controller.
The delay and render memblockq are using the source and sink sample specs,
so using pa_bytes_to_usec() will produce better estimates of the delays than
using pa_resmpler_result(). Because the delays are considered to be part of
the sink or source latency, they are added to them. source_output_buffer
becomes obsolete.
The behaviour is to leave the value unchanged. The idea is to init the value
with a default before the call and not treat a missing value as error. That
way, only parsing errors or validating errors actually return error codes.
The alsa card hasn't so far set any availability for profiles. That
caused an issue with some HDMI hardware: the sound card has two HDMI
outputs, but only the second of them is actually usable. The
unavailable port is marked as unavailable and the available port is
marked as available, but this information isn't propagated to the
profile availability. Without profile availability information, the
initial profile policy picks the unavailable one, since it has a
higher priority value.
This patch adds simple logic for marking some profiles unavailable:
if the profile only contains unavailable ports, the profile is
unavailable too. This can be improved in the future so that if a
profile contains sinks or sources that only contain unavailable ports,
the profile should be marked as unavailable. Implementing that
requires adding more information about the sinks and sources to
pa_card_profile, however.
BugLink: https://bugzilla.yoctoproject.org/show_bug.cgi?id=8448
I want module-alsa-card to set the availability of unavailable
profiles before the initial card profile gets selected, so that the
selection logic can use correct availability information.
module-alsa-card initializes the jack state after calling
pa_card_new(), however, and the profile selection happens in
pa_card_new(). This patch solves that by moving parts of pa_card_new()
to pa_card_choose_initial_profile() and pa_card_put().
pa_card_choose_initial_profile() applies the profile selection policy,
so module-alsa-card can first call pa_card_new(), then initialize the
jack state, and then call pa_card_choose_initial_profile(). After that
module-alsa-card can still override the profile selection policy, in
case module-alsa-card was loaded with the "profile" argument. Finally,
pa_card_put() finalizes the card creation.
An alternative solution would have been to move the jack
initialization to happen before pa_card_new() and use pa_card_new_data
instead of pa_card in the jack initialization code, but I disliked
that idea (I want to get rid of the "new data" pattern eventually).
The order in which the initial profile policy is applied is reversed
in this patch. Previously the first one to set it won, now the last
one to set it wins. I think this is better, because if you have N
parties that want to set the profile, we avoid checking N times
whether someone else has already set the profile.
There is currently no use for allowing modules to cancel card creation,
and I don't see need for that in the future either. Let's simplify
things by removing the failure handling code.
This allows us to parse an extra set of modargs to tack on to an
existing set. Duplicates in the second set are ignored, since this fits
our use best. In the future, this could be extended to support different
merge modes (ignore dupes vs. replace with dupes), but I've left this
out since there isn't a clear need and it would be dead code for now.
Renamed all variables pertaining to latency offsets of sinks and sources,
calling them "port_latency_offset" or similar instead. All of these variables
refer to latency offsets inherited from ports, rather than being unique to
the sinks or sources themselves.
This change is to pave the way for additional functionality for setting
latency offsets on sources and sinks independenly from the value they inherit
from their port. In order to implement them we first need this rename so that
the two latency offsets can be stored individually and summed when reporting
the total latency of the source or sink.
The renames made are:
pa_sink_set_latency_offset() -> pa_sink_set_port_latency_offset()
pa_source_set_latency_offset() -> pa_source_set_port_latency_offset()
sink->latency_offset -> sink->port_latency_offset
sink->thread_info.latency_offset -> sink->thread_info.port_latency_offset
source->latency_offset -> source->port_latency_offset
source->thread_info.latency_offset -> source->thread_info.port_latency_offset
PA_SINK_MESSAGE_SET_LATENCY_OFFSET -> PA_SINK_MESSAGE_SET_PORT_LATENCY_OFFSET
PA_SOURCE_MESSAGE_SET_LATENCY_OFFSET -> PA_SOURCE_MESSAGE_SET_PORT_LATENCY_OFFSET
Unlike pa_sink_set_port(), which calls pa_sink_set_latency_offset() to update
the latency offset of the sink to match that of its newly set port,
pa_source_set_port() did not do so. This patch adds the appropriate call to
pa_source_set_latency_offset() in pa_source_set_port() to fix this.
json-c has a symbol clash (json_object_get_type) with json-glib (which
at least a number of our GNOME clients use). This patch moves to our own
JSON parser so that we can avoid this kind of situation altogether.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=95135
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
Sink(-input) and source(-output) called unlink function when reference
count dropped to zero. This would result in unlink hooks being called
with an object having a reference count of zero, and this is not a
situation we want modules to have to deal with. It is better to just
remove the redundant unlinking code from sink(-input) and
source(-output) and assert on reference count in unlink functions as well.
It is expected that in well behaving code the owner of an object will
always unlink the object before unreferencing.
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
Setting the font-size CSS property on a widget overrides the system
font-size, and since qpaeq provides no mechanism for setting the
application's font-size, we should not do this.
This commit also removes the font-size property from commented-out calls to
setStyleSheet() so that if these are ever reinstated, this behaviour is
not reintroduced.
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
LADSPA allows float format only, but module-ladspa-sink possibly
could be loaded with ***any*** 'format' parameter. Therefore noisy
sound heard. This patch avoids to be configured as invalid format.
Signed-off-by: KimJeongYeon <jeongyeon.kim@samsung.com>
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
While investigating bug 95352, I noticed that
pa_pstream_set_revoke_callback() and pa_pstream_set_release_callback()
were identical - both set the release callback.
pa_pstream_set_revoke_callback() was obviously broken - it was setting
the wrong callback.
The only place where set_revoke_callback() is called is in
protocol-native.c. The code there looks like this:
pa_pstream_set_revoke_callback(c->pstream, pstream_revoke_callback, c);
pa_pstream_set_release_callback(c->pstream, pstream_release_callback, c);
Since set_release_callback() is called last, the release callback gets
set correctly. The only problem is that the revoke callback stays
unset. What are the consequences of that? The code that calls the
revoke callback looks like this:
if (p->revoke_callback)
p->revoke_callback(p, block_id, p->revoke_callback_userdata);
else
pa_pstream_send_revoke(p, block_id);
So the intended callback is replaced with a pa_pstream_send_revoke()
call. What does the intended callback, that doesn't get called, do?
if (!(q = pa_thread_mq_get()))
pa_pstream_send_revoke(p, block_id);
else
pa_asyncmsgq_post(q->outq, PA_MSGOBJECT(userdata), CONNECTION_MESSAGE_REVOKE, PA_UINT_TO_PTR(block_id), 0, NULL, NULL);
So the native protocol's revoke callback is anyway going to call
pa_pstream_send_revoke() when called from the main thread. If the
revoking is done from an IO thread, an asynchronous message is sent to
the main thread instead, and the message handler will then call
pa_pstream_send_revoke().
In conclusion, the only effect of this bug was that
pa_pstream_send_revoke() was sometimes being called from an IO thread
when it should have been called from the main thread. I don't know if
this caused the crash in bug 95352. Probably not.
If a card has been hot-plugged after pulseaudio start, alsa-lib still has
old configuration in memory, which doesn't have PCM definitions for the
new card. Thus, this error appears, and the device doesn't work:
I: [pulseaudio] (alsa-lib)confmisc.c: Unable to find definition 'cards.USB-Audio.pcm.front.0:CARD=0'
I: [pulseaudio] (alsa-lib)conf.c: function snd_func_refer returned error: No such file or directory
I: [pulseaudio] (alsa-lib)conf.c: Evaluate error: No such file or directory
I: [pulseaudio] (alsa-lib)pcm.c: Unknown PCM front:0
I: [pulseaudio] alsa-util.c: Error opening PCM device front:0: No such file or directory
The snd_config_update_free_global() function makes alsa-lib forget any
cached configuration and reparse all PCM definitions from scratch next
time it is told to open anything.
The trick has been copied from Phonon.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=54029
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
For various use-cases a passthrough stream should have priority over all
other streams and get exclusive access to the sink regardless of whether
any other streams are playing.
An example use-case is ensuring Kodi can successfully start video
playback (with passthrough) even if an external notification sound
happened to be playing at the same time.
Signed-off-by: Arun Raghavan <git@arunraghavan.net>
As reported by valrgrind
==30002== Conditional jump or move depends on uninitialised value(s)
==30002== at 0x5CB883C: pa_cmsg_ancil_data_close_fds (pstream.c:193)
==30002== by 0x5CBB161: do_write (pstream.c:759)
==30002== by 0x5CB8B51: do_pstream_read_write (pstream.c:233)
==30002== by 0x5CB8EE8: io_callback (pstream.c:279)
...
The pa_cmsg_ancil_data structure has two main guards:
'creds_valid', which implies that it holds credentials
information, and 'nfd', which implies it holds file descriptors.
When code paths create a credentials ancillary data structure,
they just set the 'nfd' guard to zero. Typically, the rest of
pa_cmsg_ancil_data fields related to fds are _all_ left
_uninitialized_.
pa_cmsg_ancil_data_close_fds() has broken the above contract:
it accesses the new 'close_fds_on_cleanup' flag, which is related
to file descriptors, without checking the 'nfd == 0' guard first.
Fix this inconsistency.
Reported-by: Alexander E. Patrakov <patrakov@gmail.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
As shown by valgrind
==10615== Conditional jump or move depends on uninitialised value(s)
==10615== at 0x5CC0483: shm_marker_size (shm.c:97)
==10615== by 0x5CC1685: shm_attach (shm.c:381)
==10615== by 0x5CC1990: pa_shm_cleanup (shm.c:453)
==10615== by 0x5CC068E: sharedmem_create (shm.c:150)
...
Solution is to fix the shm_marker_size() signature itself: At
certain code paths like shm_attach(), we don't want to initialize
_any_ field in the passed SHM segment descriptor except after
making sure all error exit conditions have been passed.
Reported-by: Alexander E. Patrakov <patrakov@gmail.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
We ended up dealing with it once in module init, and once more in the
new module callback. Avoiding it in the second case by name seems to be
the cleanest solution (else, we need to store the module index somewhere
in pa_dbusiface_core, which seems about as bad).
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
Commit ae415b07a0 ("dbus: Use hooks for
module new and removed events") changed the new module monitoring from
the asynchronous subscription system. Previously handle_load_module()
created the new pa_dbusiface_module object before we got
a notification of the loading of the module, but now we get the
notification already within the pa_module_load() call. That resulted
in a crash, because the module_new_cb() created the
pa_dbusiface_module object before pa_module_load() returned, and then
handle_load_module() would create another pa_dbusiface_module object
for the same module.
This patch removes the pa_dbusiface_module_new() call from
handle_load_module(). module_new_cb() is now responsible for all
pa_dbusiface_module object creations, except the ones that are created
during the initialization of module-dbus-protocol.
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
Fix memory leak in pa_resampler_new() in resampler.c, Deallocating
memory of r->lfe_filter in case of fail.
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
The current LFE crossover filter removes low frequencies from the main
channels and puts them into the LFE channel with the wrong amplitude.
It is not known for sure what is the correct relative amplitude (acoustic
measurements are required with real hardware), and changing that might
introduce a new bug, "it clips the LFE channel".
So just disable the feature by default until a better understanding
emerges how it should work. This, essentially, returns the defaults
to their state as of PulseAudio 6.0.
Some more observations:
- Most of available active analog speakers on the market do the
necessary crossover filtering already, and HDMI receivers can be
configured to do that, too, so a crossover filter in PulseAudio is
harmful in these use cases.
- The "laptop with a builtin subwoofer" use case requires manual
configuration anyway because the default crossover frequency (120 Hz) is
wrong for laptop speakers.
- Finally, Windows 10 with a built-in USB audio driver does not synthesize
the LFE channel given a 5.1 card and a stereo audio stream by default.
Hides: https://bugs.freedesktop.org/show_bug.cgi?id=95021
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
ffmpeg_data was not freed properly before return due to error.
It is now freed properly.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=95347
Signed-off-by: Sachin Kumar Chauhan <sachin.kc@samsung.com>
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
This is needed so we don't keep stale jack availability information
while the card is suspended.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=93259
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
Dynamic memory allocated to 'module_name' and 'fltr' was being leaked.
Its now freed properly before return.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=95293
Signed-off-by: Sachin Kumar Chauhan <sachin.kc@samsung.com>
Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
Rather than entirely ignore streams for which we have automatically
loaded a filter, this makes module-device-manager only avoid rerouting
such streams within their existing filter hierarchy.
If, for example, m-d-m decided to move a stream which is currently
routed to speakers/mic which we requested echo cancellation for, to a
USB headset, the previous logic would disallow such a move even though
it was legitimate.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=93443
Signed-off-by: Arun Raghavan <git@arunraghavan.net>
This adds an ignore mechanism to module-device-manager and uses that
from within module-filter-apply, rather than having m-d-m have knowledge
of anything related to m-f-a.
Signed-off-by: Arun Raghavan <git@arunraghavan.net>
Let's assume that there are two output ports, and they are on
different profiles:
* Integrated speakers (priority: 10000, available)
* HDMI (priority: 5900, not available)
Then the user plugs in an HDMI monitor with speakers. Since the HDMI
priority is lower than the speaker priority, we don't route to HDMI by
default. However, the user manually switches the profile to use the
HDMI output.
Then the user plugs out the monitor, so we switch back to speakers.
When the monitor is plugged back in, the user needs to manually switch
the audio output again. That should be improved: if the user preferred
to the HDMI output over the speakers, we should remember that and
automatically switch to HDMI whenever it becomes available.
The lack of automatic switching is even worse when the monitor goes to
a sleep mode after some period of inactivity. The monitor audio may
become unavailable, and PulseAudio can't distinguish that from the
case where the monitor is physically unplugged. Even worse, the
monitor may become unavailable for a short while when adjusting the
display parameters (for example, media center software may adjust the
display parameters to match the media that is being played back). In
these cases we clearly should switch automatically back to HDMI when
it becomes available again.
This patch fixes the problem by setting pa_card.preferred_input_port
and pa_card.preferred_output_port when the user changes the card
profile or a port, and switching to the preferred port when it becomes
available.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=93946
I will modify module-switch-on-port-available so that it will keep
track of which input and output port the user prefers on the card,
based on the user's profile and port switches. The preference needs
to be saved on disk, for which I will use module-card-restore.
To facilitate communication between the two modules, this patch adds
preferred_input_port and preferred_output_port fields to pa_card, and
a hook for monitoring the variable changes. It would be nice if the
two modules would communicate directly with each other, but
implementing that would be somewhat complicated, so I chose this time
for adding the functionality to the core. In theory some other routing
module might want to manage the new variables instead of
module-switch-on-port-available, but admittedly that's not very likely
to happen...
If first part of test is false and e->device is NULL pa_streq will
segfault. Fix by using pa_safe_streq, which checks strings for NULL
before doing strcmp.
This reverts commit 12a202c510.
This is needed for now to avoid a clash in clients using json-glib. The
commit added a call to json_object_get_type() in client code that didn't
exist before, and this breaks some apps like Rhythmbox and Totem. This
will be fixed in the future by possibly dropping json-c as a dep.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=95135
Now that all layers in the stack support memfd blocks, add memfd
support for the daemon's global core mempool. Also introduce
"enable-memfd=" daemon argument and configuration option.
For now, memfd support is an opt-in feature to be activated only
when daemon's enable-memfd= is set to yes.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Now that all layers in the stack support memfd blocks, add memfd
pools support for client context and audio playback data.
Use such memfd pools by default only if the server signals memfd
support in its connection negotiations.
Also add ability for clients to force-disable memfd transport
through the `enable-memfd=' client configuration option.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
This saves some proplist allocations and a couple of code lines. Also,
logging is better, because the set_property() functions work with
string values, while the update_proplist() functions assume opaque
binary data, and therefore can't log the property values.
pa_sink_input_update_proplist() is inconvenient in many cases, because
it requires allocating a new proplist, even if the goal is to just set
one property. pa_sink_input_update_properties also can't properly log
property changes, because it has to assume that all values are
arbitrary binary data.
This patch adds pa_sink_input_set_property() for setting a string
value for a single property, and pa_sink_input_set_property_arbitrary()
for setting a binary value for a single property.
pa_sink_input_update_properties() is reimplemented as a wrapper around
pa_sink_input_set_property_arbitrary() to centralize logging and
sending change notifications.
(The above mentions only sink input functions for brevity, but the
same changes are implemented for source outputs too.)
device-manager reroutes all streams whenever a new device appears.
When filter-apply has loaded a filter for some stream, the filter
device may not be what device-manager considers the best device for
the stream, which means that when an unrelated device appears,
device-manager may break the filtering that filter-apply had set up.
This patch changes filter-apply so that it saves the filter device
name to the stream proplist when it sets up a filter. device-manager
can then check the proplist when it does rerouting, and skip the
rerouting for streams that have a filter applied to them.
The proplist isn't cleaned up when the stream moves away from the
filter device, so before doing any decisions based on the
filter_device property, it should be checked that the stream is
currently routed to the filter device. It seemed simpler to do it this
way compared to setting up stream move monitoring in filter-apply and
removing the property when the stream moves away from the filter
device.
Before a device is unlinked, the unlink hook is fired, and it's
possible that a routing module tries to move streams to the unlinked
device in that hook, because it doesn't know that the device is being
unlinked. Of course, the unlinking is obvious when the code is in an
unlink hook callback, but it's possible that some other module does
something in the unlink hook that in turn triggers some other hook,
and it's this second hook where the routing module may get confused.
This patch adds an "unlink_requested" flag that is set before the
unlink hook is fired, and moving streams to a device with that flag
set is prevented.
This patch is motivated by seeing module-device-manager moving a
stream to a sink that was being unlinked. It was a complex case where
an alsa card was changing its profile, while an echo-cancel sink was
connected to the old alsa sink. module-always-sink loaded a null sink
in the middle of the profile change, and after a stream had been
rescued to the null sink, module-device-manager decided to move it
back to the old alsa sink that was being unlinked. That move made no
sense, so I came up with this patch.
When autoloaded, module-echo-cancel doesn't support moving the sink
input and source output that it creates, but the move prevention was
implemented by manually requesting module unloading in the middle of
the stream move procedure, rather than by just setting the DONT_MOVE
flags. This patch removes the module unloading code from the moving()
callbacks and adds the DONT_MOVE flags. In addition to saving some
code, this also prevents problems related to trying to move streams
connected to the echo cancel sink or source while the echo cancel sink
or source is in the middle of a move too (a crash will happen in such
situation, as demonstrated in
https://bugs.freedesktop.org/show_bug.cgi?id=93443).
srbchannel needs fd passing. Otherwise we get the following error
for systems without SCM_CREDENTIALS support:
Code should not be reached at pulsecore/pstream-util.c:95,
function pa_pstream_send_tagstruct_with_fds(). Aborting.
[[ The root cause is that we define HAVE_CREDS only if
SCM_CREDENTIALS is defined, but SCM_CREDENTIALS is a Linux-specific
symbol. Thus HAVE_CREDS is always disabled on Solaris.
And since pulse couples the non-portable creds passing support
with the portable fd passing one, through _35_ places where
HAVE_CREDS is used, a real fix needs a PA redesign -- assuming that
latency on Solaris is something people care about. ]]
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=94339
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Per glibc feature_test_macros(7), setting compiler flags to
-std=c11 (or any c* variant like c99) enforces strict ANSI
mode.
Enforcing strict ANSI makes all declarations under _GNU_SOURCE
unavailable. This leads to build warnings in the form of:
warning: implicit declaration of function ‘syscall’
Thus replace -std=c11 with -std=gnu11
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
In case of invalid argument for volume, the crash occurs in pa_stream_interaction_done().
pa_xnew() is replaced with pa_xnew0() to fix it.
Signed-off-by: Sangchul Lee <sc11.lee@samsung.com>
Now, trigger_roles, ducking_roles and volume can be divided into several groups by slash.
That means each group can be affected by its own volume policy.
If we need to apply ducking volume level differently that is triggered from
each trigger role(s), this feature would be useful for this purpose.
For example, let's assume that tts should take music and video's volume down to 40%
whereas voice_recognition should take those and tts's volume down to 20%.
In this case, the configuration can be written as below.
trigger_roles=tts/voice_recognition ducking_roles=music,video/music,video,tts volume=40%/20%
If one of ducking role is affected by more than two trigger roles simultaneously,
volume of the ducking role will be applied by method of multiplication.
And it works in the same way as before without any slash.
Signed-off-by: Sangchul Lee <sc11.lee@samsung.com>
`Mic` is now detected as `Mic-In/Mic Array` (there are 2 microphones physically, nice to se this being understood).
`Line` is now detected as `Line In`.
Removed all output modes except officially supported stereo, 5.1 and stereo S/PDIF.
Also microphone/line in now might be used simultaneously with either output mode, yay!
The code was mixing sink and sink input domain rate updates, and that
only works if the rate of the RTP stream is the same as the rate of the
sink. This changes all the calcuations to be on the sink-input rate,
since that's the rate we are trying to guess (and resample for).
Now that we have the necessary infrastructure to memexport and
mempimport a memfd memblock, extend that support higher up in the
chain with pstreams.
A PA endpoint can now _transparently_ send a memfd memblock to the
other end by simply calling pa_pstream_send_memblock() – provided
the block's memfd pool was earlier registered with the pstream.
If the pipe does not support memfd transfers, we fall back to
sending the block's full data instead of just its reference.
** Further details:
A single pstream connection usually transfers blocks from multiple
pools including the server's srbchannel mempool, the client's
audio data mempool, and the server's global core mempool.
If these mempools are memfd-backed, we now require registering
them with the pstream before sending any blocks they cover. This
is done to minimize fd passing overhead and avoid fd leaks.
Moreover, to support all these pools without hard-coding their
number or nature in the Pulse communication protocol itself, a new
REGISTER_MEMFD_SHMID command is introduced. That command can be
sent _anytime_ during the pstream's lifetime and is used for
creating on demand SHM ID to memfd mappings.
Suggested-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Color global mempools with a special mark. This special marking
is needed for handling memfd-backed pools.
To avoid fd leaks, memfd pools are registered with the connection
pstream to create an ID<->memfd mapping on both PA endpoints.
Such memory regions are then always referenced by their IDs and
never by their fds, and so their fds can be safely closed later.
Unfortunately this scheme cannot work with global pools since the
registration ID<->memfd mechanism needs to happen for each newly
connected client, and thus the need for a more special handling.
That is, for the pool's fd to be always open :-(
Almost all mempools are now created on a per-client basis. The
only exception is the pa_core's mempool which is still shared
between all clients of the system.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
To transfer memfd-backed blocks without passing their fd every time,
thus minimizing overhead and avoiding fd leaks, a command is sent
with the memfd fd as ancil data very early on.
This command has an ID that uniquely identifies the memfd region.
Further memfd block references are then exclusively done using this
ID.
This commit implements the details of such 'permanent' mappings on
the receiving end, using memimport segments.
Suggested-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Memfd is a simple memory sharing mechanism, added by the systemd/kdbus
developers, to share pages between processes in an anonymous, no global
registry needed, no mount-point required, relatively secure, manner.
This patch introduces the necessary building blocks for using memfd
shared memory transfers in PulseAudio.
Memfd support shall also help us in laying out the necessary (but not
yet sufficient) groundwork for application sandboxing, protecting PA
from its clients, and protecting clients data from each other.
We plan to exclusively use memfds, instead of POSIX SHM, on the way
forward.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
pa_shm_create_rw() is responsible for creating two types of memory:
POSIX shared memory and regular malloc()-ed ones.
A third memory type, memfds, will be added later. Thus to add this
extra shared memory type in a sane manner, refactor private memory
allocations into their own static methods.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Soon we're going to have three types of memory pools: POSIX shm_open()
pools, memfd memfd_create() ones, and privately malloc()-ed pools.
Thus introduce annotations for the memory types supported and change
pa_mempool_new() into a factory method based on required memory.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
The PA daemon currently uses a single SHM file for all clients
sending and receiving commands over the low-latency srbchannel
mechanism.
To avoid leaks between clients in that case, and to provide the
necessary ground work later for sandboxing and memfds, create the
srbchannel SHM files on a per-client basis.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
In future commits, server-wide SHMs will be replaced with per-client
ones that will be dynamically created and freed according to clients
connections open and close.
Meanwhile, current PA design does not guarantee that the per-client
mempool blocks are referenced only by client-specific objects.
Thus reference count the pools and let each memblock inside the pool
itself, or just attached to it, increment the pool's refcount upon
allocation. This way, per-client mempools will only be freed when no
further component in the system holds any references to its blocks.
DiscussionLink: https://goo.gl/qesVMV
Suggested-by: Tanu Kaskinen <tanuk@iki.fi>
Suggested-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
The daemon `shm-size-bytes' configuration value was read, and then
directly used, for creating the initial server-wide SHM files.
This is fine for now, but soon, such server-wide SHMs will be replaced
with per-client SHM files that will be dynamically created and deleted
according to clients open and close. Thus, appropriately cache this
configuration value.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
sd_journal_send() implicitly add fields for the source file,
function name and code line from where it's invoked. As code
location fields CODE_FILE, CODE_LINE and CODE_FUNC are handled
by PA's log module, we do not want the automatic values
supplied by the sd_journal API.
Without suppressing these, both the actual log event source
and the call to sd_journal_send() will be logged:
$ journalctl -b -f -o json-pretty
[...]
CODE_FILE : [ pulsecore/log.c, pulsecore/module.c ],
CODE_LINE : [ 505, 181 ],
MESSAGE : Failed to load module module-gconf
CODE_FUNC : [ pa_log_levelv_meta, pa_module_load ],
[...]
(Commit log adapted from abrt libreport commit d1eaae97f0287f)
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
This patch deals with the case that applications start new streams corked.
In case of module-role-cork it will only mute the stream because corking is
removed later by the application.
When a trigger stream changes mute or cork state, the cork streams should
react to this. The same applies if a stream changes its role to or from the
trigger role.
When corking do not ignore streams without media.role. Instead treat
them as if media.role="no_role", so that you can specify "no_role" as
trigger or cork role.
If u->save_time_event is non-NULL when the module is being unloaded,
it means that there are some changes to the database that haven't
yet been flushed to the disk.
Acked-by: David Henningsson <david.henningsson@canonical.com>
By refactoring volume probing into its own function, we can reduce
indentation a lot. Also, if an error occurs during the volume probe,
that volume element is now always skipped (instead of taking down
the entire path with it).
Also, a bug for elements with more than two channels is fixed, as
previously, the volume parsing code was continuing, potentially
referencing somewhere outside the array (which has max two channels).
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If you have headphones plugged in and plug in HDMI; you want sound
to stay on headphones.
If you have HDMI plugged in and you plug in headphones; you want sound
to switch to headphones.
Hence we need to take priority into account as well when determining
whether to switch to a new profile or not.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=93903
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
It is expected that the underlying AGC mechanism will likely provide a
single volume for the source rather than a per-channel volume. Dealing
with per-channel volumes just adds complexity with regards to the
actual volume setting (depending on whether volume sharing is enabled or
not, we would set the volume on the source output of the virtual source,
and their sample specs may be different).
Using a single volume allows us to sidestep this problem entirely.
This is required to have unequal channel counts on capture in and out
streams, which is needed for beamforming to work. The deinterleaved API
only works with floating point samples.
The calculations around how many samples were sent to the canceller
engine was not updated when we started supporting different channel
counts for playback and capture.
This is needed for building with anonymous unions. A bunch of calls to
fail() that used to mysteriously work need fixing -- fail() is a macro
that takes a printf-style message as an argument. Not passing this
somehow worked with the previous compiler flags, but breaks with
-std=c11.
The AGC code no longer seems to honour the analog volume limits we set,
and internally uses 0-255 as the volume range. So we switch to use that
(keeping the old API usage as is in case this gets fixed upstream).
This allows us to inherit the sample spec parameters from the sink and
source master (rather than forcing 32 kHz / mono). It is still possible
to override some of the parameters for the source side with modargs.
My original testing showed that these parameters provided a decent
perf/quality trade-off on lower end hardware (which I no longer have
access to). I figure it makes sense to continue with that for now, and
in the future this can be relaxed (use_master_format=yes could be the
default, and resource-constrained systems can disable it).
In the refactoring, I'm expressing the constraints in what I see to be a
more natural way -- rec_ss expresses what we're feeding the canceller,
so it makes sense to apply the constraints on what the canceller accepts
there. This then propagates to the output spec.
This also exposes the range of sample rates that the library actually
supports (8, 16, 32 and 48 kHz).
The original intention was to configure low enough parameters to keep
CPU consumption down. Prior to this change, we assumed that the EC
backend would override the sink parameters based on the source
parameters to achieve this goal, and with this change we remove that
assumption by forcing the default parameters for the sink to be low
enough.
It's not possible to enable the intelligibility enhancer at the
moment, because the feature would require modifying the audio that we
play to speakers, which we don't do currently. All audio processing is
done at the source side, and it's not easy to change that.
This patch is based on Arun Raghavan's code, I just reordered things
a bit and reworded the FIXME comment.
This creates a longer filter that is more complex and less sensitive to
incorrect delay reporting from the hardware. There is also a
delay-agnostic mode that can eventually be enabled if required.
In some very quick testing, not enabling this seems to provide better
results during double-talk.
In file included from pulse/timeval.c:32:0:
pulse/timeval.c: In function 'pa_timeval_add':
./pulsecore/macro.h:303:28: warning: left shift of negative value [-Wshift-negative-value]
? ~(~(type) 0 << (8*sizeof(type)-1))
reported by Ubuntu gcc-6
gcc-6 adds -Wshift-negative-value (enabled by -Wextra) which warns
about left shifting a negative value. Such shifts are undefined
because they depend on the representation of negative values.
also works with -Wshift-overflow=2
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
modules/module-stream-restore.c: In function 'clean_up_db':
modules/module-stream-restore.c:2344:74: warning: comparison of constant '0' with boolean expression is always true [-Wbool-compare]
pa_assert_se(entry_write(u, item->entry_name, item->entry, true) >= 0);
reported by Ubuntu gcc-6
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Loading X stuff from default.pa is a bad idea, since it doesn't work
if there are multiple X sessions, or PulseAudio is started outside the
X session. Since it's a bad idea, don't encourage it by including
examples that do so.
I also removed the sample loading examples. I don't think the examples
are particularly useful, since nothing uses the loaded samples once
module-x11-bell is removed from the configuration.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=93109
Previously a missing key would cause this kind of log output:
D: [pulseaudio] module-device-manager.c: Database contains invalid data for key: sink:auto_null (probably pre-v1.0 data)
D: [pulseaudio] module-device-manager.c: Attempting to load legacy (pre-v1.0) data for key: sink:auto_null
D: [pulseaudio] module-device-manager.c: Size does not match.
D: [pulseaudio] module-device-manager.c: Unable to load legacy (pre-v1.0) data for key: sink:auto_null. Ignoring.
That is now replaced with
D: [pulseaudio] module-device-manager.c: Database contains no data for key: sink:auto_null
This gets rid of an error message from the debug log. If
volume_factor_source would actually be used somewhere, this bug would
have caused more severe problems.
volume_factor_source should have the source's channel map. When moving
the stream, the volume needs to be remapped from the old source's
channel map to the new source's map. However, when the stream is being
moved, there is a period where the old source has already been
forgotten and the new source isn't yet known, so the remapping can't
be done directly between the two channel maps. Instead, the volume is
remapped from the old source's map to the stream's own map when the
move starts, and again remapped from the stream's map to the new
source's map when the move finishes.
The first remapping was missing, causing the second remapping fail and
print an error to the log.
(I checked the sink input code as well. It didn't have this bug.)
Applying the volume after resampling means mismatch between the volume
channel map and the data channel map.
volume_factor_source is not currently used anywhere, so this bug
hasn't been causing any problems. I noticed it while reading the code.
This will likely be needed in the future when we start supporting high
bitrate passthrough, and there actually seem to be people 352/384 kHz
out there (potentially as an intermediate production step).
If 'pa_modargs_new' returns a NULL, we need to be careful to not call
'pa_modargs_free' in the failure path since it requires that we pass it
a non-null argument. Also updates 'module-bluetooth-policy.c:pa__init'
to follow the standard "goto fail" pattern used everywhere else.
Signed-off-by: Jason Gerecke <killertofu@gmail.com>
There is no way to check CPU type in a portable way across ABIs.
Assume if pointers are 64-bit that CPU is capable to perform fast
64-bit operations. Add an extra check to handle x32-ABI.
PulseAudio by default builds with -Wundef. If we add -Werror=undef this
missing define is fatal. By default build log is full of entries like:
In file included from ./pulsecore/core.h:47:0,
from ./pulsecore/module.h:31,
from ./pulsecore/sink-input.h:31,
from pulsecore/sound-file-stream.c:36:
./pulsecore/sample-util.h: In function 'pa_mult_s16_volume':
./pulsecore/sample-util.h:58:5: warning: "__WORDSIZE" is not defined [-Wundef]
#if __WORDSIZE == 64 || ((ULONG_MAX) > (UINT_MAX))
^
(NetBSD-7.99.21 with default GCC 4.8.5)
This change fixes build issues on NetBSD.
This also address a bug reported by Shawn Walker from Oracle (possibly Solaris):
Bug 90880 - builds can fail due to non-portable glibc-specific internal macro usage
This is an OpenPGP/MIME signed message (RFC 4880 and 3156)
The old code fetched the channel name via AudioObjectGetPropertyData()
and accessed the "returned" data as a plain char buffer.
This may or may not have worked at some point according to the Apple
CFString documentation, which warns that the actual data layout is an
implementation detail and subject to change at any time.
On recent OS X versions, this behavior led to "random data" channel
names like >H��{, H��{<.
We need to actually let AudioObjectGetPropertyData() populate a CFString
struct and convert this into a plain char buffer.
The conversion function will not free the CFString, so do that in the
caller.
Signed-off-by: Mihai Moldovan <ionic@ionic.de>
This isn't a great fix, but we need ALSA API to do this right. In the
mean time, USB devices work fine with timer-based scheduling, so there's
no reason to force a large minimum latency by disabling tsched on them.
We might be compiled without eventfd support, or something else
might go wrong. And it's fully possible to continue using the old
channel rather than just disconnecting.
Signed-off-by: Pierre Ossman <ossman@cendio.se>
This forces the canceller engine to be invoked even if playback is not
currently active. We need to do this for cases where the engine provides
additional processing that is independent of playback, such as noise
suppression and AGC.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=83557
I want to enable client.conf.d, because in OpenEmbedded-core we have
a graphical environment called Sato that runs as root. Sato needs to
set allow-autospawn-for-root=true in client.conf, but the default
configuration in OpenEmbedded-core should not set that option. With
this patch, I can create a Sato-specific package that simply installs
50-sato.conf in /etc/pulse/client.conf.d without conflicting with the
main client.conf coming from a different package.
daemon.conf.d is enabled just because it would be strange to not
support it while client.conf.d is supported.
This allows a configuration scheme where after loading configuration
from "somefile", the parser loads configuration from files in
directory "somefile.d". This feature needs to be enabled on a per-file
basis, though, and this patch doesn't yet enable the feature for any
files.
The relationship between sinks, sources, cards, profiles, and ports
is becoming ever more intertwined, to the point that if you try to
include one file from the other, you're likely to end up with some
weird error somewhere else.
Work around this by creating a new typedefs.h, which does not depend
on anything else, and just creates a few typedefs.
(Can be expanded with more typedefs in the future if the need arises.)
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
When synthesized alsa path is freed there is an assert from NULL
proplist. Create empty proplist for the path to fix.
Signed-off-by: Juho Hämäläinen <juho.hamalainen@nomovok.com>
From the NetBSD manual:
The first argument of these functions is of type int, but only a very
restricted subset of values are actually valid. The argument must either
be the value of the macro EOF (which has a negative value), or must be a
non-negative value within the range representable as unsigned char.
Passing invalid values leads to undefined behavior.
-- ctype(3)
On some systems (at least Arch) DATADIRNAME is not defined. This
caused PULSE_LOCALEDIR to point to a wrong directory. This seemed like
an issue introduced in 7.0, but probably something else was updated in
Arch at the same time, causing DATADIRNAME to become undefined,
because there were no changes between 6.0 and 7.0 that could have
caused this.
After noticing that localedir is a standard variable, my first idea
was to use pulselocaledir='${localedir}' in configure.ac, but Jan
Steffens pointed out that it causes the final PULSE_LOCALEDIR to
become "${prefix}/share/locale", that is, the variables weren't fully
expanded. I then found a FAQ item in Autoconf's manual[1], which
recommends not to define any absolute installation directories in
configure.ac, because the installation directories should be possible
to change when running make. The recommended solution is to define the
constant in AM_CPPFLAGS instead, so that's what this patch does.
[1] https://www.gnu.org/software/autoconf/manual/autoconf-2.69/html_node/Defining-Directories.html
pa_module_unload() takes two pointers: pa_module and pa_core.
The pa_core pointer is also available via the pa_module object,
so the pa_core argument is redundant
[David Henningsson: Rebased to git HEAD]
It doesn't work currently (fails and falls back to PCM), due to channel
count mismatch between the sink sample spec and the sample spec required
by IEC61937.
To be reverted when someone implements changing channel count without
switching profiles. This would also be required for HBR passthrough over
HDMI.
Reported-by: Xamindar <junkxamindar@gmail.com>
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
In case pa_card_set_profile is called with save=false, then probably
it makes more sense not to update the port's preferred profile as well.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If we always write entries of the latest version, we can simplify
code a little by only handling old versions in the "entry_read"
function and assume we have the latest version everywhere else.
Suggested-by: Tanu Kaskinen <tanuk@iki.fi>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This makes the routing slightly more aggressive:
* It will try to route to another profile, if such a profile
is preferred by the port.
* It will allow changing profiles on transitions both to
PA_AVAILABLE_YES and PA_AVAILABLE_NO
To accommodate there is also some refactoring.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
In case input or output names are filled in, we can use this to
get a better match in the profile_good_for_input/output functions
instead of guessing based on number of sources and channels.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
It can be useful for routing modules to know a profile's input
and output parts, in order to e g change output profile
while keeping the input profile unchanged.
For now filling in these fields is optional and a routing module
must be able to handle NULL in these fields.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Adding AGC broke this test, so we hard-disable the volume code in test
mode. This is probably okay for now, since at least with analog AGC, the
source volume changes and the data we get is going to be with AGC
applied, but digital gain won't be encapsulated here.
Long term, we might need to figure out how to deal with this properly.
Without this, we hit an assert because the channel count in
new_reference (which was inherited from the master) is lower than the
channel count of the filter.
packet.h defines:
typedef struct pa_packet pa_packet;
and packet.c defines:
typedef struct pa_packet {
...
} pa_packet;
With old versions of gcc (such as gcc 4.5) this causes a redefinition
error at compile time:
pulsecore/packet.c:43:3: error: redefinition of typedef 'pa_packet'
pulsecore/packet.h:26:26: note: previous declaration of 'pa_packet' was here
In order to fix this, this commit changes the definition in packet.c
to just:
struct pa_packet {
...
};
This way, the contents of the structure remain opaque to users of
pa_packet outside packet.c, and the 'pa_packet' type remains usable.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=91334
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The drain reporting improvements that were added to alsa-sink were only
being applied to directly connected sink inputs. This patch makes the
same logic also recurse down the filter hierarchy, so drains are
acknowledged more accurately (and not late) even if there is a filter
sink in between.
Also does some minor reorganisation of the code and sprinkles in some
comments as documentation.
We encountered an alsa plugin a while ago (not sure if the source
can be shared) which had mixer controls, but no descriptors to
poll for changes.
Quit early to avoid latter assertion failures.
BugLink: https://bugs.launchpad.net/bugs/1092377
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The combination "Front Headphone" + "Headset Mic Phantom"
was found on one the machines we enable. Without this patch,
the headset mic appeared plugged in when nothing was plugged
into the jack.
BugLink: https://bugs.launchpad.net/bugs/1513384
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
sync-playback just had a much longer timeout than it should have, and
extended-test was using the default. We set the expected amount of time,
so the test is more correct (if it takes longer than this, something
probably actually broke).
In rtp.c:
if (sscanf(t+9, "%i %64c", &_payload, c) == 2)
the string c seems to be non-null terminated. It is later used as
following:
c[strcspn(c, "\n")] = 0;
The same piece of code is responsible for the inability of pulseaudio
on OpenWRT to handle RTP stream at the rate 48000 from another
machine:
[pulseaudio] sdp.c: Failed to parse SDP data: missing data.
It turns out that uClibc does not agree with glibc about "%64c", see
http://git.uclibc.org/uClibc/tree/docs/Glibc_vs_uClibc_Differences.txt
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=92568
The 0.1.2 version of libsoxr fixes soxr_process() crash after soxr_clear() is used, so check the library version at compile time and use soxr_clear() if possible.
Now that we have memory usage benchmarks collected at our disposal,
introduce a gnuplot script to plot the newest version.
To avoid scaling issues, memory is plotted in a "double y axis" form,
with VM usage on the left, and dirty RSS memory usage on the right.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
Add shell script to sample PulseAudio memory usage while increasing
the number of connected 'paplay' clients over time.
Linux kernel /proc/$PID/smaps Private and Shared_Dirty fields are used
to accurately measure the total size of used dirty pages over time.
This shall be useful for benchmarking the PA daemon's memory while
introducing new features like per-client SHM access and memfds.
Also add an empty benchmarks-collection directory 'benchmarks/'. All
output from the benchmarking tools shall be saved in this place, with
timestamps and symbolic links to the newest versions.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
The gnome/unity-control-center UIs have a master volume slider, and
three sub-sliders: balance, fade, and subwoofer. Balance and fade
use PA's set_balance and set_fade APIs accordingly, but the subwoofer
slider sometimes does unintuitive things.
In order to make that slider behave better, let's add a LFE balance
API that these volume control UIs can use instead. With this API,
the UI can balance between "no subwoofer" and "only subwoofer" with
"equal balance" in the middle, which would make it more consistent
with the behaviour of the other sliders.
BugLink: https://bugzilla.gnome.org/show_bug.cgi?id=753847
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
We currently only support one and two channels for volumes, and
bail out otherwise. This makes Xonar users unhappy because they
have a volume with eight channels, and bailing out means they
don't have a path/port at all.
This way they will at least have a port, which will in turn make
the gnome/unity UI behave better.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=84983
BugLink: https://bugzilla.gnome.org/show_bug.cgi?id=745017
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Seccomp-BPF uses SIGSYS signal to trigger
the trap handler attached to sys_open.
If the signal is blocked then the kernel kills
the process whenever pulse audio calls 'open'.
The result backtrace is terminating in sys_open.
That's why it is required to keep SIGSYS unblocked
if it is currently unblocked and trapped.
This patch allows to have pulse audio working
in the Chromium sandbox.
Signed-off-by: Julien Isorce <j.isorce@samsung.com>
Signed-off-by: Arun Raghavan <git@arunraghavan.net>
In case a tarball-version file is present, use that and quit.
Otherwise git will continue looking for directories, potentially
finding .git directories which are dirty and mark the version as such.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=90936
We need to guard the pstream with an extra ref to ensure
it is not destroyed at the time we check whether or not the
srbchannel is destroyed.
Reported-by: Takashi Iwai <tiwai@suse.de>
BugLink: http://bugzilla.opensuse.org/show_bug.cgi?id=950487
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
In case the same jack causes one port to become available and another
one unavailable, the available should be reported first.
This is to avoid unnecessary changes: e g, consider a 'Headphone Jack'
making 'Headphone' available and 'Speaker' unavailable. In case the
unavailable change triggers first, and there is also a currently available
third port (e g 'Digital out'), the routing system might choose to route
to this port because neither of the 'Speaker' and 'Headphone' ports are
available.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
With the exception of when trying to clean up shm files,
it's useful to warn if opening them fails, regardless of reason.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If bashcompletiondir was empty, the check didn't catch that. As
a result, the symlinks that were supposed to be generated in the
completion directory were created in the root directory.
I'm not sure how much they are needed nowadays with the latest
changes to the subset elimination (I found this while
researching a bug on an older PA version), but I guess they could
be added for consistency at least.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Commit 262bdae0330e used symbols which are only available if systemd
support was compiled in. Fix by using the appropriate #ifdef guards.
Also document the resulting PULSE_LOG_JOURNAL environment variable
behavior if systemd journal support was not compiled in.
[Diwic: changed wording slightly.]
Reported-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
By introducing such an environment variable, applications using the
PA client libraries can configure these libraries to send their logs
directly to the journal.
While client libraries journal logging can be indirectly achieved
using PULSE_LOG_SYSLOG, this pollutes the journal. Meta data gets
replicated twice: once in the journal meta fields and once in the
syslog(3) plain-text message itself.
For attaching any backtraces, also introduce the PA-specific journal
meta field PULSE_BACKTRACE. This is the recommend journal practice
instead of appending any furuther data to the logging message itself.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
The nodelete flag indicates that we don't want our libraries to be
unloaded. It's only relevant on libraries, so let's not use it for
executables. Trying to use it on executables breaks things on some
platforms.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=90878
Pulseaudio fails to build on the Alpha architecture due to a failure
in the volume-test of the test suite. I had reported this to the
Debian bug tracker [1] but the maintainer has asked that I forward the
patch to this mail list. The failure in volume-test occurs because it
is compiled with -ffast-math which implies -ffinite-math-only of which
the gcc manual states that it optimizes for floating-point arithmetic
with the assumption that arguments and results are not NaNs or
+/-infinity, and futher notes that it may result in incorrect output.
On the Alpha platform that is somewhat an understatement as the use of
non-finite floating-point arithmetic with -ffinite-math-only results in
a floating-point exception and the termination of the program.
The volume-test converts volumes into decibels (so a zero volume
becomes a negative infinity) and proceeds to add two volumes (in
decibels), thus does arithmetic with non-finite floating point numbers
despite being compiled with -ffast-math!
I attach a patch that protects against the arithmetic with non-finite
numbers for your consideration. With that patch the test-suite passes
on Alpha.
Cheers
Michael.
[1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=798248
It is possible that we get a zero-length memchunk to work with.
Specifically, this happens the resampler (which is called before the
lfe-filter) consumes all the input data, but does not (yet) produce any
output data.
Reproduced using:
pulseaudio --resample-method=soxr-mq
pactl load-module module-null-sink sink_name=lfe_test channels=3 channel_map=front-left,front-right,lfe
paplay --raw /dev/zero --rate=48000 -d lfe_test
Thanks to the original reporter for the backtrace:
Bug: https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/1496577
Originally pointed out by Georg Chini.
Calculating buffer = buffer + (send_counter - recv_counter)
in one branch and buffer = 2 * buffer - (recv_counter - send_counter)
looks very obviously wrong. In other words, before the patch, the
contribution from the previous lines was double-counted.
Libraries from modlibexec_LTLIBRARIES list require
not only libpulsecommon but also libpulse and
libpulsecore from lib_LTLIBRARIES list.
This patch fix race in 'make -j X install' (with X is 2 and more)
when building/installing inside chroot placed on RAM-disk(tmpfs).
Signed-off-by: Zavadovsky Yan <zavadovsky.yan@gmail.com>
pa_module_free is called from more than one place, not all of
these places correctly removed the module from the
modules_pending_unload array, potentially causing a dangling pointer
in that array.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Based on some googling, strtod_l() is defined in xlocale.h on BSD.
Glibc seems to define it in stdlib.h, but only if GNU extensions are
enabled (otherwise the function won't be available). So, this patch
should fix the use of strtod_l() on BSDs, but on other systems things
may or may not be still broken.
The original patch author is Jakob Fink <jfink@gmx.at>. He sent this
patch to the freebsd-gnome mailing list:
http://lists.freebsd.org/pipermail/freebsd-gnome/2015-April/032138.html
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=90285
We have no strtof_l calls in the code, so it doesn't make sense to
check that function's availability. We have one strtod_l call, so
let's check that instead.
I don't know if this change makes any practical difference. I just
wondered why we had HAVE_STRTOF_L ifdefs in core-util.c for code that
didn't use strtof_l.
In this release cycle, libpulse had some internal code changes, but no ABI
changes.
libpulse-simple and libpulse-mainloop-glib had no changes whatsoever.
Previously the UCM code created one jack object per device name (which
is not the same thing as creating one jack object per device, because
the UCM device namespace is scoped on per-verb basis, so devices in
different verbs may have the same name). I think it's conceptually
cleaner to create one jack object per alsa kcontrol. I plan to do
similar refactoring on the traditional mixer code later.
Previously module-alsa-card assigned to pa_alsa_jack.plugged_in
directly, and then did the port availability updating manually. The
idea of pa_alsa_jack_set_plugged_in() is to move the availability
updating to the mixer infrastructure, where it really belongs.
Similarly, pa_alsa_jack.has_control was previously modified directly
from several places. The has_control field affects the port
availability, and pa_alsa_jack_set_has_control() takes care of
updating the availability.
For now, pa_alsa_jack_set_plugged_in() and
pa_alsa_jack_set_has_control() only update the port availability
when using UCM. My plan is to adapt the traditional mixer code later.
"Front Line Out" was found in the wild on one of the machines we enable.
I figured I could just as well add "Rear Line Out" too, because that's
just as likely to show up.
As a reminder, "Front Line Out" means "a line out jack physically located
on the front side", where as "Line Out Front" means "a line out jack
playing back front left and front right channels in a channel map".
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This helps figuring out why bootstrap.sh is failing...
Directing the error message to /dev/null was very strange. I don't
know what the original motivation might have been. My guess is that
it was added unintentionally.
These mapping names are used in sb-omni-surround-5.1.conf, which needs
to use separate mappings for input and output, since they are
associated with different alsa devices.
Document how to modify the client libraries logging behvaior
using any of the PA-specific environment variables.
Using the PULSE_LOG_* environment variables makes debugging
and tracing PA applications quite easy, thus the need for an
official documentation text.
Signed-off-by: Ahmed S. Darwish <darwish.07@gmail.com>
<EP-E358F00C1D9A449EAE69225B9D2530F8>
Updated the error string for mute commands in pactl. If someone forget to specify
the parameter on the commandline the resulting error message isn't quite right.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=90117
Signed-off-by: Deepak Srivastava <srivastava.d@samsung.com>
This fixes a crash. sink_input_pop_cb() drains the message queue that receives
memchunks from the combine sink thread to avoid requesting more audio too soon.
The same message queue received also SET_REQUESTED_LATENCY messages, which
generate rewind requests. Rewind requests shouldn't be issued in the pop()
callback, doing so results in an assertion error. Therefore, it was not safe to
drain the message queue in the pop() callback, but usually the queue is empty,
so this bug was not immediately detected.
This patch splits the message queue into two queues: audio_inq and control_inq.
audio_inq receives only messages containing memchunks, and control_inq receives
only the SET_REQUESTED_LATENCY messages. The pop() callback only drains the
audio queue, which avoids the rewind requests in the pop() callback.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=90489
We have crashes related to modules loaded after unload. This added
warning can provide some information about what that module is,
which in turn can help us solve the crashes, hopefully.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=90108
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
When we the underlying sink/source goes away, there is an intermediate
state where the asyncmsgqs that we were using for the sink-input and
source-output go away. This is usually okay if the sink-input and
source-output are moved to another device, but can be problematic if we
don't support moving (which is the case when the filter is autoloaded).
This becomes a problem because of the following chain of events:
* The underlying sink goes away
* Moving the filter sink-input fails (because it is autloaded)
* At this point the sink-input has no underlying sink, and thus
no underlying asyncmsgq
* This also applies to all sink-inputs connected to the echo-cancel
module
* The sink-input is killed, triggering a module unload
* On unlink, module-rescue-streams tries to move sink-inputs to
another sink, starting with a START_MOVE message
* There is no asyncmsgq for the message, so we crash
* We can't just perform a NULL check for the asyncmsgq, since there
are state changes we need to effect during the move
To fix this, we pretend to allow the move to the new sink, and then
unlink ourselves *after* the move is complete. This ensures that we
never find ourselves in a position where we need the underlying
sink/asyncmsgq to be present when it is not.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=90416
In case the sample spec is not known, as can be the case when
pa_stream_new_extended is used, we cannot satisfy the PULSE_LATENCY_MSEC
request.
As a workaround disable being able to use PULSE_LATENCY_MSEC in this case.
Reported-by: Fritsch <fritsch@xbmc.org>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
...otherwise this code will fail on big-endian architectures.
Cc: Hui Wang <hui.wang@canonical.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Modern versions of MinGW and Visual Studio provide socket errno
defines that make no sense (no API sets them). Make sure we
continue to use the old WSAE ones that are actually returned by
Windows' socket API.
Signed-off-by: Pierre Ossman <ossman@cendio.se>
pulsecore/sink.c: In function 'pa_sink_put':
pulsecore/sink.c:648:53: warning: logical not is only applied to the left hand side of comparison [-Wlogical-not-parentheses]
pa_assert(!(s->flags & PA_SINK_DYNAMIC_LATENCY) == (s->thread_info.fixed_latency != 0));
^
pulsecore/source.c: In function 'pa_source_put':
pulsecore/source.c:599:55: warning: logical not is only applied to the left hand side of comparison [-Wlogical-not-parentheses]
pa_assert(!(s->flags & PA_SOURCE_DYNAMIC_LATENCY) == (s->thread_info.fixed_latency != 0));
^
rewrite expression to suppress warning:
!(x & MASK) == (y != 0)
<->
!(x & MASK) == !(y == 0)
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
warnings emited by gcc 5.1:
utils/padsp.c: In function 'dsp_trigger':
utils/padsp.c:1902:39: warning: logical not is only applied to the left hand side of comparison [-Wlogical-not-parentheses]
while (!pa_operation_get_state(o) != PA_OPERATION_DONE) {
^
utils/padsp.c: In function 'dsp_cork':
utils/padsp.c:1937:39: warning: logical not is only applied to the left hand side of comparison [-Wlogical-not-parentheses]
while (!pa_operation_get_state(o) != PA_OPERATION_DONE) {
^
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Fall back to the previous /etc/bash_completion.d dir on failures
(either old bash completion or not installed).
changes over Ville Skyttä's patch:
define PKG_CHECK_VAR macro which became available only in pkg-config 0.28
see https://bugs.freedesktop.org/show_bug.cgi?id=88782 and
https://bugs.freedesktop.org/show_bug.cgi?id=89540
Signed-off-by: Ville Skyttä <ville.skytta@iki.fi>
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
pulsecore/filter/biquad.c: In function 'biquad_lowpass':
pulsecore/filter/biquad.c:52:10: warning: declaration of 'gamma' shadows a global declaration [-Wshadow]
pulsecore/filter/biquad.c: In function 'biquad_highpass':
pulsecore/filter/biquad.c:86:10: warning: declaration of 'gamma' shadows a global declaration [-Wshadow]
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
see https://bugs.freedesktop.org/show_bug.cgi?id=86818
an exit code of 1 makes systemd believe that the service failed;
better return 0 to denote that PA sucessfully stopped on the user's
request
sidenote: systemd's SuccessExitStatus= could be used to turn code 1 into a
code denoting success
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Cc: jan.steffens@gmail.com
On 32bits OS, this test case fails. The reason is when rewinding to
the middle of a block, some of float parameters in the saved_state
are stored in the memory from FPU registers, and those parameters will
be used for next time to process data with lfe. Here if FPU register
is over 32bits, the storing from FPU register to memory will introduce
some variation, and this small variation will introduce small
variation to the rewinding result.
So adding the tolerant variation for comparing the rewind result, make
this test case can work on both 64bits OS and 32bits OS.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
This fixes buffer attr calculation so that we set the source latency to
the requested latency. This makes sense because the intermediate
delay_memblockq is just a mechanism to send data to the client. It
should not actually add to the total latency over what the source
already provides.
With this, the meaning of fragsize and maxlength become more
meaningful/accurate with regards to ADJUST_LATENCY mode -- fragsize
becomes the latency the source is configured for (which is then
approximately the total latency until the buffer reaches the client).
Maxlength, as before, continues to be the maximum amount of data we
might hold for the client before overrunning.
Fix the following warnings.
CC pulsecore/filter/libpulsecore_6.0_la-lfe-filter.lo
pulsecore/filter/lfe-filter.c: In function 'pa_lfe_filter_rewind':
pulsecore/filter/lfe-filter.c:179:9: warning: format '%lu' expects argument of type 'long unsigned int', but argument 6 has type 'size_t' [-Wformat=]
pa_log_debug("Rewinding LFE filter %lu samples to position %lli. No saved state found", samples, (long long) f->index);
^
pulsecore/filter/lfe-filter.c:183:5: warning: format '%lu' expects argument of type 'long unsigned int', but argument 6 has type 'size_t' [-Wformat=]
pa_log_debug("Rewinding LFE filter %lu samples to position %lli. Found saved state at position %lli",
^
CC pulsecore/filter/libpulsecore_6.0_la-biquad.lo CC pulsecore/filter/libpulsecore_6.0_la-lfe-filter.lo
pulsecore/filter/lfe-filter.c: In function 'pa_lfe_filter_rewind':
pulsecore/filter/lfe-filter.c:179:9: warning: format '%lu' expects argument of type 'long unsigned int', but argument 6 has type 'size_t' [-Wformat=]
pa_log_debug("Rewinding LFE filter %lu samples to position %lli. No saved state found", samples, (long long) f->index);
^
pulsecore/filter/lfe-filter.c:183:5: warning: format '%lu' expects argument of type 'long unsigned int', but argument 6 has type 'size_t' [-Wformat=]
pa_log_debug("Rewinding LFE filter %lu samples to position %lli. Found saved state at position %lli",
^
Installs all the build dependencies, and runs make check and
check-daemon.
V1: Based on Arun Raghavan's travis file. Added trusty repositories to get
newer libs.
V2: Explicitly list all dependencies instead of relying on the Ubuntu
package Build-Dependencies. Send notifications to pulseaudio-discuss
V3: Install libsystemd-daemon-dev, libsystemd-id128-dev,
libsystemd-journal-dev, and libsystemd-login-dev. Send notifications to
pulseaudio-commits. Drop libjson0-dev, libjson-c-dev is the package to
depend on.
Previously pa_parse_volume() clamped the value to fit in the valid
range, but I think it's better to reject values outside the valid
range.
This also changes the percentage parsing to allow non-integer values.
We currently use pa_yes_no to write module arguments, so they can not be
localised. Instead add a new pa_yes_no_localised function and use it in pactl
(and thus, revert all other places to use the non-localised version).
BugLink: https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/1445358
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Fixes a compiler warning:
../../src/modules/udev-util.c: In function 'pa_udev_get_info':
../../src/modules/udev-util.c:228:443: warning: 'bus' may be used uninitialized in this function [-Wmaybe-uninitialized]
if (!pa_streq(bus, "firewire") && (v = udev_device_get_property_value(card, "ID_MODEL_FROM_DATABASE")) && *v)
This makes the GUIs (e g gnome/unity-control-center) look more consistent
with other inputs/outputs that also have ports.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This works around bug 80850: a mapping can only have one channel map,
and in case of a 6-out 10-in device, the mapping will be adjusted to
have both 10 and 6 channels, which does not work.
Reported-by: Benjamin Tegge <benjaminosm@googlemail.com>
Suggested-by: Raymond Yau <superquad.vortex2@gmail.com>
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=80850
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
For recently supported FireWire sound devices, udev's database assign
the name of IEEE 1394 Phy/Link chipset to ID_XXX_FROM_DATABASE. This is
not friently names to users.
This commit applies a workaround to skip ID_XXX_FROM_DATABASE for any
FireWire devices.
[Fixed up by David Henningsson <david.henningsson@canonical.com>]
Reported-by: Andras Muranyi <muranyia@gmail.com>
Reference: https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/1381475
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
We don't and probably never will have any pa_atod() callers that would
require "NaN" to be accepted, so let's filter those out in pa_atod(),
instead of requiring the callers to handle not-a-numbers appropriately
(which they generally forget to do).
This small helper will simplify code in many modules.
The hooks added through pa_module_hook_connect will be freed just
before pa__done is called (so trying to add hooks during pa__done
will result in assertion failure).
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
CoreAudio routines that return an error status do so with the
OSStatus type, which is not a UInt32: typical OS X errors are
negative numbers.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
When a sink or source is freed, there may be pending volume changes that
didn't get applied before the IO thread got torn down. Those pending
changes need to be freed.
The memory leak was reported here:
http://thread.gmane.org/gmane.comp.audio.pulseaudio.general/23162/focus=23169
Reported-by: Alexander E. Patrakov <patrakov@gmail.com>
When crossover_freq is set to 0, this restores the old behaviour
of letting the LFE channel be the average of the source channels,
without additional processing. This can be useful e g in case the
user already has a hardware crossover.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
so far, this test only includes rewind test, it works as below:
let lfe-filter process 2 blocks mono lfe channel audio samples, the
sample format is PA_SAMPLE_S16LE, save the processed data to the temp
buffer, then rewind the lfe-filter back 1 block and 1.5 blocks
respectively, reprocess the audio samples from the rewind position,
then comparing the output data with previously saved data.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
The resampler framework just forwards the request to the lfe filter.
There are no resampler impl that can rewind yet, so just reset the
resampler impl instead of properly rewinding yet.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Store current filter state at every normal block process.
When a rewind happens, rewind back to the nearest saved state,
then calculate forward to the actual sample position.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
To avoid the macro trap: I call pa_memblock_new_malloced with
"pa_xmemdup" as data parameter, and that would expand to *two*
calls to pa_xmemdup in case that remains a macro, which is clearly
not intended.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Add a user defined parameter lfe-crossover-freq for the lfe-filter,
to pass this parameter to the lfe-filter, we need to change the
pa_resampler_new() API as well.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
- Remove imported dead code
- Fix compiler warnings
- Fix non-GCC compiler compilation (use more portable macros)
- Change lr4 struct to include a biquad struct
Thanks to Alexander Patrakov for suggesting many of these changes.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
When enable-lfe-remixing is set, an LFE channel is present in the
resampler's destination channel map but not in the source channel map,
we insert a low-pass filter instead of just averaging the channels.
Other channels will get a high-pass filter.
In this patch, the crossover frequency is hardcoded to 120Hz (to be fixed
in later patches).
Note that in current state the LFE filter is
- not very optimised
- not rewind friendly (rewinding can cause audible artifacts)
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The chrome OS audio server has some already existing code, which
has been made available under a BSD-style license, which should be
safe to import by us.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Flushing the asyncmsgq can cause arbitrarily callbacks to run, potentially
causing recursion into pa_thread_mq_done again. Because of this; rtpoll which
is cleared in the second iteration is tried to free once again by the first
iteration leading to PA crash.
While investigating bug 89672 it was found that pa_thread_mq_done
was called recursively. Regardless of whether the recursion should
be stopped by other means, it seems to make sense to make
pa_thread_mq_done more robust so that it can be called twice
(and even recursively) without harm.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=89672
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
pa_atou(), pa_atol() and pa_atod() are stricter than the libc
counterparts (the PA functions reject strings that have trailing extra
stuff in them). I have been under the impression that the PA functions
only accept "obviously valid numbers", that is, I have assumed that
these would be rejected: " 42" (leading whitespace), "" (empty
string) and "-18446744073709551615" in case of pa_atou().
I noticed that empty strings are accepted, however, and on closer
inspection I found that leading whitespace is accepted too, and even
that pa_atou() thinks that "-18446744073709551615" is the same thing
as "1"! This patch makes the parsing functions more strict, so that
they indeed only accept "obviously valid numbers". I decided to also
disallow leading plus signs, just because I don't like them.
In src/pulsecore/core-util.c:set_nice() we currently use a temporary
dbus-connection to set the nice-level via rtkit. However, we never
close that connection. This is fine, as the connection is shared and
dbus-core will manage it. But no other part of pulseaudio (except
set_scheduler()) uses the libdbus1 managed connections. Therefore,
we effectively end up with an unused dbus-connection that is not
integrated into any main-loop. dbus-daemon will send bus-notifications
to the connection (as libdbus1 installs matches for those by default
(it has to!)) until the outgoing queue is full. Thus, we waste several
KBs (or MBs? I didn't look it up) of memory for a message queue that
is never dispatched.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
In case SHM is full or disabled, audio data is sent through the
io/srbchannel. When this channel in turn gets full, memblocks
could previously be split up. This could lead to crashes in case
the split was on non-frame boundaries (in combination with full
memblock queues).
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=88452
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
a separate free-list is used to recycle memory of fixed-sized packets
with up to MAX_APPENDED_SIZE of data
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
if length exceeds maximum appended size, create a packet of
type dynamic instead of type appended
this is a preparation to use a separate free-list for packets
document semantics of pa_packet_new_*() functions
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
v2: (thanks David Henningson)
* fix double assignment of data in pa_tagstruct_new_fixed(), two statements on one line
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
add 128 bytes of storage in each tagstruct that will initially
be used; if this storage is exceeded the type changes to _DYNAMIC
v3: (thanks David Henningson)
* add comments explaining how memory is handled by different tagstruct types
v2: (thanks Alexander Patrakov)
* replace constant 100 with GROW_TAG_SIZE (the increment in with a dynamic tagstruct grows when extend()ed)
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
pa_tagstruct_free_data() is used in only one place
to pass data from a tagstruct to a packet
this patch is a temporary solution which introduces an extra
malloc(); will be resolved shortly...
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
... in order to prepare for a new type _APPENDED
remove the assert() for dynamic in pa_tagstruct_data() as
the function makes sense for all tagstruct types (and the returned pointer
is const)
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
pa_tagstruct_new() is called either with no data, i.e. (NULL, 0)
to create a dynamic tagstruct or with a pointer to fixed data
introduce a new function pa_tagstruct_new_fixed() for the latter case
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
This change doesn't affect behaviour, because accessing boolean fields
in the new data was safe even after the done() call, but it was still
bad style.
While adding functions for writing and reading pa_bvolume structs, I
found myself wondering if I could make it simpler to write and read
the basic types that a pa_bvolume consists of, without having to worry
about network byte ordering, remembering to call extend() and getting
the length and read index adjustments just right. This is what I came
up with.
There is a functional change too: previously the
pa_tagstruct_get_foo() functions didn't modify the read index in case
of errors, but now, due to read_tag() modifying the read index at an
early stage, the read index gets modified also in case of errors. I
have checked the call sites, and I believe there's no code that would
rely on the "no read index modification on error" property of the old
functions. If reading anything from a tagstruct fails, the whole
tagstruct is considered invalid (typically resulting in a protocol
error and client connection teardown).
Message id 0 is PA_SOURCE_OUTPUT_MESSAGE_GET_LATENCY. So, every time PulseAudio
sent PA_SOURCE_OUTPUT_MESSAGE_GET_LATENCY message to the loopback source output,
it actually hit the SOURCE_OUTPUT_MESSAGE_LATENCY_SNAPSHOT handler instead. As a
result, the SOURCE_OUTPUT_MESSAGE_LATENCY_SNAPSHOT handler was called when not
intended, the default PA_SOURCE_OUTPUT_MESSAGE_GET_LATENCY handler was not called
at all, and the latency was thus evaluated incorrectly.
Reported-by: Georg Chini <georg@chini.tk>
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
...because we will later try with plug:* which will probably succeed,
so this is not an error.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Added ID and names for the resampler presets and also updated the working sample rate deduction to take the new resampler into account. The initial libsoxr backend version does not variable rate resampling, so it is disabled in this case.
see https://bugs.freedesktop.org/show_bug.cgi?id=68135
pacat and paplay man pages both claim to describe the paplay program
(which is actually a symlink to pacat) -- this is inconsistent and
redundant, so drop the paplay man page
a follow-up patch will add man page symlink for all programs
implemented by pacat, not just paplay
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
see
https://bugs.freedesktop.org/show_bug.cgi?id=85011
in case NOAUTOSPAWN is set and no server has been specified, PA starts listening on DBUS
for a new server, and the state is PA_CONTEXT_CONNECTING, but pa_context_connect()
returns -1; it should return 0.
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Also, remove the talk about "fast" variants of functions that remove
entries from an array. Currently there's no need for order-preserving
functions, so all functions are "fast".
While at it, also remove SOCKET_SERVER_GENERIC, because it is always
being overwritten with a specific socket type.
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
An assertion was already used in pa_socket_server_new_unix(), this
makes the TCP variants consistent with that.
Even if pa_socket_server_new() could fail, the error handling wasn't
good, because there was no "goto fail", meaning that the fd would have
been leaked.
This fixes a crash that occurred when trying to access non-existent
port data. Doing this:
pa_alsa_port_data *data = PA_DEVICE_PORT_DATA(port);
is not a good idea when using UCM, because in the UCM mode ports don't
have any data, so the data pointer points to some random memory.
During my work on module-loopback I found a bug that sometimes crashes pulse when
module-loopback is loaded due to pushing a zero-length block into the memblockq.
As there is a one-line fix I thought you might want it for 6.0.
ALSA mutes speaker when Line Out is plugged in by default, so
we should follow that convention.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Recent testing has shown some srbchannel related bugs that
indicates that the srbchannel feature is not ready to be enabled
by default.
Therefore, temporary disable it for the 6.0 release and re-enable
it in git master once 6.0 is released.
Bugs:
https://bugs.freedesktop.org/show_bug.cgi?id=88452https://bugs.freedesktop.org/show_bug.cgi?id=88167
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
FSF addresses used in PA sources are no longer valid and rpmlint
generates numerous warnings during packaging because of this.
This patch changes all FSF addresses to FSF web page according to
the GPL how-to: https://www.gnu.org/licenses/gpl-howto.en.html
Done automatically by sed-ing through sources.
When line out path is active, we want to mute speakers for obvious
reasons, and headphones to avoid volume spikes.
Reported-by: TienFu Chen <tienfu.chen@canonical.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
pa_source_state_t can have value PA_SOURCE_INVALID_STATE, not
PA_SINK_INVALID_STATE. It happens to be the same here, but it can break
sometimes.
Issue detected by PVS Studio.
In some cases, depending on the instruction that performs the load, orc
ignores the size of the parameter when loading it for the first time.
Explicitly load the parameter into a temp to make sure it is loaded
correctly, like we do for the 2ch case.
See https://bugzilla.gnome.org/show_bug.cgi?id=742271
Since the srb memblock and the audio data were coming from separate
pools, and the base index was per pool, they could actually still
collide.
This patch changes the base index to be global and atomically
incremented.
Reported-by: Arun Raghavan <arun@accosted.net>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
When pactl is invoked with any options or the -- specifier, optind will
be > 1. Therefore using a static 3 value is wrong. Use optind+2 as both
offset and count difference.
Bug-Debian: http://bugs.debian.org/774810
I noticed that when resuming the tunnel sink, there was a small amount
of previously played audio before the new audio started to play.
Normally that probably wouldn't be noticeable, because there would be
a few seconds of silence played before suspending the sink due to
inactivity, so the unwanted old audio would be just silence, but in my
configuration sinks are suspended immediately when there's nothing
playing to them, so the glitch becomes audible.
Pavel Machek reported in his blog that our message about the system mode
has a dead link in it. And this link is also present in translations.
So, I replaced it in the source and fixed all translations using a script:
for a in po/*.po ; do msgcat --no-wrap $a | sed
's@http://pulseaudio.org/wiki/WhatIsWrongWithSystemMod
@http://pulseaudio.org/wiki/WhatIsWrongWithSystemMode @g' | sed
's@http://pulseaudio.org/wiki/WhatIsWrongWithSystemMode@http://www.freedesktop.org/wiki/Software/PulseAudio/Documentation/User/WhatIsWrongWithSystemWide/@g'
| sed 's@/\.@/ .@g' | sed 's@/,@/ ,@g' | msgcat - > $a.new
git add -i # to filter out formatting changes
The "/." and "/," replacements are needed so that various terminal
emulators don't include the trailing "." or "," into the clickable URL.
The resulting patch is attached, just in case, in order to avoid
damaging non-ASCII characters.
--
Alexander E. Patrakov
>From 7dcd197571840e467d688f0f7354253730bbcc15 Mon Sep 17 00:00:00 2001
From: "Alexander E. Patrakov" <patrakov@gmail.com>
Date: Sat, 29 Nov 2014 20:56:27 +0500
Subject: [PATCH] Fix the WhatIsWrongWithSystemWide URL
Reported by Pavel Machek in http://pavelmachek.livejournal.com/126190.html
All translations were also fixed using a script.
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
This reverts commit 7276faca72.
Using the new systemd socket activation for PulseAudio will cause PulseAudio to not
have any connection with D-Bus, breaking device reservation protocol, module-jackdbus-detect
and module-dbus-protocol. Therefore, autospawn is now still enabled by default even if you
build with systemd daemon headers.
The bluetooth card is created when the first profile becomes
available, which means that the card may have profiles that are not
available when the card is initialized. If module-card-restore tries
to restore such profile, that will fail, and the card will be
initialized with the "off" profile active.
This patch modifies module-card-restore so that if follows the profile
availability status, and when the saved profile becomes available, it
is activated. Additionally, module-card-restore is modified so that it
doesn't even try to restore unavailable profiles, when the necessary
information is available. In practice there are two existing places
where the profile is restored, and only one of those contexts has the
necessary information available. Unfortunately, it's the more
important context (card creation) where the information is not
available. This means that module-card-restore will set the initial
profile of a new card even if the profile is unavailable, and this
will cause an ugly warning in the log, even though there's nothing
abnormal happening.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=87081
In case a transport is currently disconnected and transitions to
idle, that should not count as a "remote hang up" event.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This fixes a "use of uninitialised value" error in previous memblock commit.
Reported-by: Alexander Patrakov <patrakov@gmail.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
In case PA_MEMPOOL_DISABLE is set, pa_memblock_new_pool can return
NULL. It does not make sense to set up a srbchannel without a shared
memory pool, so just fail in this case.
Reported-by: Alexander Patrakov <patrakov@gmail.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Every new memexport object now gets an ever increasing base index,
that prevents block ID collisions between different memexport
objects on the same pstream.
In particular, this prevents block ID collision between the srb memblock
(which has its own memexport object) and audio data blocks.
Reported-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This fixes an issue when requesting module unload for
module-bluetooth-discover. When unloading the module, it also unloads
module-bluez4-discover and/or module-bluez5-discover, and that
invalidated the state variable that was used for iterating through the
modules idxset.
The pa_module.unload_requested flag could now otherwise be removed,
but it's still being (ab)used in the bluetooth modules.
mingw32 does not have "getuid", so ifdef it properly.
Reported-by: Michael DePaulo <mikedep333@gmail.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Because the adapters reference the devices hashmap on free, we mush
free the adapters hashmap first and then the devices hashmap.
Reported-by: Alexander Patrakov <patrakov@gmail.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Because debian does not run with the freebsd libc, but rather uses the
GNU one, it chose to not define __FreeBSD__, but rather __FreeBSD_kernel__.
Use the alternative when the functionality tested is for kernel
features, and keep the __FreeBSD__ one when using freebsd libc
headers.
If this patch is applied, debian could drop all the current patches when
importing 6.0 :)
If the libbluetooth headers aren't available, we shouldn't treat that
as an error unless --enable-bluez5-native-headset has been explicitly
given to configure.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=86582
I don't know if it can cause any problems if HAVE_BLUEZ_4,
HAVE_BLUEZ_5, HAVE_BLUEZ, HAVE_BLUEZ_5_OFONO_HEADSET or
HAVE_BLUEZ_5_NATIVE_HEADSET are undefined when the corresponding
features are not enabled, but it certainly won't hurt to define the
variables also when the features are not enabled.
Send the right command to set the speaker and microphone gain.
Note that setting the volume on the Headset should use the unsolicited
result code. Receiving the volume from the Headset uses the AT
command.
get rid of the following warning when compiling with NDEBUG:
modules/alsa/alsa-mixer.c: In function 'element_is_subset':
modules/alsa/alsa-mixer.c:3125:18: warning: 'a_limit' may be used uninitialized in this function [-Wmaybe-uninitialized]
long a_limit;
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
supresses a warning when compiling with NDEBUG:
pulsecore/aupdate.c: In function 'pa_aupdate_read_end':
pulsecore/aupdate.c:82:14: warning: variable 'n' set but not used [-Wunused-but-set-variable]
unsigned n;
pulsecore/sink-input.c: In function 'pa_sink_input_unlink':
pulsecore/sink-input.c:648:27: warning: variable 'p' set but not used [-Wunused-but-set-variable]
pa_source_output *o, *p = NULL;
pulsecore/sink-input.c: In function 'find_filter_sink_input':
pulsecore/sink-input.c:1523:14: warning: unused variable 'i' [-Wunused-variable]
unsigned i = 0;
pulsecore/sink-input.c: In function 'pa_sink_input_start_move':
pulsecore/sink-input.c:1569:27: warning: variable 'p' set but not used [-Wunused-but-set-variable]
pa_source_output *o, *p = NULL;
CC pulsecore/libpulsecore_5.0_la-sink.lo
pulsecore/sink.c: In function 'pa_sink_unlink':
pulsecore/sink.c:673:24: warning: variable 'j' set but not used [-Wunused-but-set-variable]
pa_sink_input *i, *j = NULL;
pulsecore/source-output.c: In function 'find_filter_source_output':
pulsecore/source-output.c:1179:9: warning: unused variable 'i' [-Wunused-variable]
int i = 0;
CC pulsecore/libpulsecore_5.0_la-source.lo
pulsecore/source.c: In function 'pa_source_unlink':
pulsecore/source.c:616:27: warning: variable 'j' set but not used [-Wunused-but-set-variable]
pa_source_output *o, *j = NULL;
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
the macro PA_UNUSED may be used to suppress a warning when a variable
is not used, or assigned and never used; this typically happens
when the only use of the variable is within an assert() that can
be optimized away (i.e. with NDEBUG set)
has an effect with GCC only
v2: (thanks to Alexander Patrakov)
* fix patch subject/description
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
fixes many warnings when compiling with NDEBUG, such as
CC pulse/libpulse_la-channelmap.lo
pulse/channelmap.c: In function 'pa_channel_map_init_auto':
pulse/channelmap.c:397:1: warning: control reaches end of non-void function [-Wreturn-type]
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
output DEPRECATED warnings for libsamplerate in configure and
PA daemon's log
libsamplerate offers no particular advantage over the speex
resampler and is distributed under GPL; support for it will be removed
in one of the next releases
v2: (thanks Arun Raghavan)
* log a warning (instead of info)
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Fixes warning: 'new_active' may be used uninitialized in this function,
and could potentially cause erronous behaviour in case an invalid port
name was specified.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
In case there are two independent jacks for one port (e.g. Dock
Headphone Jack and Headphone Jack), the availability ends up being
incorrect if the first one was _NO (not plugged) and the second gets
_YES (plugged). Also pulse complains about the state being inconsistent
which isn't true.
Fix this by preferring more precise states (yes/no) over unknown and yes
over others. However in case a plugged jack makes the port unavailable
let that overrule everything else.
The old code tried to look up the port object by using an object path,
but the ports hashmap uses port names as keys, so the method failed
always.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=85369
This patch adds a module argument "headset=ofono|native|auto" to
module-bluetooth-discover and module-bluez5-discover.
To make Arun's happy, the default is 'native' if compiled in, otherwise
'ofono'. 'Auto' will try to autoswitch depending on whether ofono is
running or not.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This implements some autodetect if both headset backends are compiled in:
First we try to contact the oFono service, if that's not working,
then we start the native backend instead.
Likewise if the oFono service is going offline/online, we load/unload
the native backend accordingly.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Enable both ofono and native backends to be built into the same
libbluez5-util. Never build the null backend.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Seems that after commit 467b4b9be systemd usage has been added into
src/daemon/main.c but there is no link for the corresponding
library in the final pulseaudio binary.
This might be missed in some build systemd due to overlinking,
but it's correct to add this in here explicitly
Signed-off-by: Colin Guthrie <colin@mageia.org>
Commit fa092af59c removed an argument to pa_rtpoll_run, but
forgot to remove that argument for all callers to pa_rtpoll_run.
This commit removes the remaining ones.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
frames_per_block is the mempool's maximum block size in frames
v2 (thanks David Henningson)
* rename max_frames to frames_per_block
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
These two control names are currently being added to the HDA driver,
so let's support them in PulseAudio as well.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
the stat command should only output statistics, not info
behaviour was deprecated anno 2011 in 8ace9185 "pactl: Make stat backwards
compatible" -- fix this now
v2: (thanks Tanu Kaskinen):
* adjust shell completion
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
When enabled, this method is prefered over pulseaudio's built in
systems so we should try our best to ensure that it cannot be spawned
outside of the mechanisms desired.
Packagers should call 'systemctl --global enable pulseaudio.socket' to
enable the socket for all users, or alternatively ship an enabling
symlink in /usr/lib/systemd/user/sockets.target.wants/ folder. It may
also make sense for distributions to add in a ConditionNNN= line to the
socket unit if they have a downstream mechanism for enabling or
disabling pulseaudio.
If individual users wish to opt out of this vendor (or administrator)
decision, they can call 'systemctl --user mask pulseaudio.socket'
This --start is patched out in several downstreams to allow users to easily
disable PA by simply disabling autospawn.
If autospawn is enabled, then the first pactl command will start it and if not
it will fail and the script will exit.
When switching to systemd socket activation, we very much do not want to
start PA manually here. We could replace it with a
systemctl --user start pulseaudio
but really it just makes sense to rely on the socket activation as this
should apply equally to non-systemd setups which use PA's own autospawn.
This adds support to module-native-protocol-unix to take over already
listening sockets passed in via socket activation (e.g. from systemd)
Most of the code is isolated to socket-server but some cleanup code also
had to be tweaked to ensure we do not overzealously close open fds.
In newer versions of systemd some libraries were combined for the sake of
general simplicity.
This change checks against the newer name first and avoids separate pkgconfig
checks if it's found. We probably want to keep support for the older library
names for some time. systemd does allow for the shipping of compatibility
pkgconfig files to not break downstream code like ourselves which is why this
likely hasn't been "fixed" until now.
With this change we no longer rely on systemd having been built with those
compatibility pkgconfig files.
We currently use the term SYSTEMD when referring to libsystemd-login
and JOURNAL when referring to libsystemd-journal.
I will be shortly adding support for libsystemd-daemon and in
preparation I figured it would be a good idea to clarify the names
used currently before adding another!
The analog-output path should be suppressed when there are more
specific paths available. Currently that usually doesn't happen. The
suppression can be done with the path subset detection, and this patch
makes that work (another approach would be to mark the elements as
required-absent, like analog-input does, but I like the subset
suppression more, because it requires less stuff in the configuration
files). The problem with listing the now-removed elements in
analog-output.conf was that if the sound card had e.g. a Speaker
element, then the switch behaviour was different between analog-output
and analog-output-speakers, so analog-output was not considered a
subset of analog-output-speakers.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=74609
We weren't writing out one character from the "OK" response, and the
"AT" part of the "+VGS" and "+VGM" commands was missing. Also, the spec
says that the command is terminated by only a CR and not an LF (probably
doesn't hurt, but let's adhere to the spec for now).
Parse the gain changed AT commands from the headset and fire 2 new
hooks as a result. The device will connect to those hooks and change the
source/sink volumes.
When the source/sink volume changes, set the gain on the microphone or
speaker respectively. Make sure we do nothing if the transport can not
handle the gain changes.
Add a simple native headset backend that implements support for the
blutooth HSP profile.
This allows pulseaudio to output audio to a Headset using the HSP profile.
Make the native backend the default.
Bash-completion 1.90 introduced support for on-demand loading
of completions. Install the completion file as 'pulseaudio' to match
the main command, and create symlinks as aliases for other supported
commands in order to support the new system.
There is no use in trying to load data in legacy format, if we
already know that there is no data at all.
Also clarify in the debug message whether there is invalid data
or no data at all.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
So far the command name has been figured out by looking one or two
items back in the $words array, but I needed a way to figure out the
command given an arbitrary number of parameters. I was implementing
a command for removing devices from the device-manager database, and
the command would take a list of devices. Since the number of devices
that need to be completed can be arbitrarily large, the previous "look
one or two words back" approach didn't work.
This new approach is more verbose, but I think it's also easier to
follow. There's some duplication that would be easy to avoid by
merging some of the commands, but I decided to not do that, to make
it more obvious what the code does.
Apparently "WIBBLE" is just a test, and maybe the test was "How
long does it take until somebody notices a strange row in configure.ac
and tries to remove it", if so, the test result is "a little over
three years". :-)
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
We will just ignore the memblock if this happens. We already have
a check for this in the client library, so this one is just for
security reasons.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If somebody tries to push a non-frame-aligned memblock onto the
memblockq, then we should fail the write. Otherwise the daemon will
crash, see https://bugs.freedesktop.org/show_bug.cgi?id=77595
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Since we don't allow lengths that are not frame aligned,
it does not make sense to allow indices that are not frame aligned
either.
Also, allowing such a thing to be added causes the daemon to crash
later instead (see https://bugs.freedesktop.org/show_bug.cgi?id=77595 ).
Also drop _se from assert (there is no side effect).
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This fixes a bug in latency configuration. The wrong type in the cast
caused UINT64_MAX being not treated as special, so the configured
latency was set to UINT64_MAX usecs, which of course is absurdly huge
latency.
This should not have any effect on behaviour. The goal is to align
with the pattern that I think we should follow:
Object initialization:
- put() is the place to create references from other objects to the
newly created object. In this case, adding the transport to
discovery->transports was moved from new() to put, and adding the
transport to device->transports was moved from set_state() to
put().
Object destruction:
- unlink() undoes put() and removes all references from other objects
to the object being unlinked. In this case setting the
device->transports pointer to NULL was moved from set_state() to
unlink(), and setting the discovery->transports pointer to NULL was
moved from free() to unlink().
- free() undoes new(), but also calls unlink() so that object owners
don't need to remember to call unlink() before free().
This fixes getting the binary name in the Hurd, or any other port using
the GNU C library, but only in the case where the library is directly
linked to. Opening with dlopen will not work.
Change in v3: reorder header includes and definitions
Change in v2: use a weak reference to main, so that we
don't crash when main cannot be found.
This test broke when PA_MAX_INPUTS_PER_SINK was increased from 32 to 256.
Because we currently don't have time to figure out why, let's just set
NSTREAMS to 20 in the meantime.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The default maximum latency is 10 seconds, which is not good,
especially since the tunnel sink doesn't support rewinding. Due to the
lack of rewinding, e.g. volume changes take a long time with large
latencies.
It seems at some point the code migrated to use the entry_write calls,
but fill_db is still using the old syntax, causing the entry to be
invalid.
The crash happens when clean_up_db gets called, which then calls
entry_read, causing the crash.
Signed-off-by: Ricardo Salveti de Araujo <ricardo.salveti@canonical.com>
If the keyboard is unplugged, it looks like the kernel is reporting
back -ENODEV when trying to close the fd. This is probably a kernel
error, but still, it's better to complain than to crash.
Buglink: https://bugs.freedesktop.org/show_bug.cgi?id=80867
Reported-by: Stelios Bounanos
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
It is possible that the chosen active_port doesn't equal
new_data->active_port, using p->name is more accurate.
Please refer to sink_new_hook_callback()
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Now a2dp and hsp sinks and sources will have different names which means that
applications and other modules can use sink/source to distinguish selected
profile.
Module module-device-restore uses sink/source name and port name as identifier,
so if different profiles have different names module-device-restore can store
volume settings for each profile.
So with this patch it is possible to configure different volume settings for
a2dp and hsp profiles.
This patch does not change port names so gnome applications will be happy.
Note that similar patch is needed also for bluez5, but I'm not using bluez5
so I cannot write or test it.
Signed-off-by: Pali Rohár <pali.rohar@gmail.com>
On a machine without fixed connecting audio devices like internal
microphone or internal speaker, and when there is no external audio
devices plugging in, the default source/sink is alsa_input/alsa_output
and there is no input devices/output devices listed in the gnome
sound-setting.
Under this situation, if we connect a bluetooth headset, the gnome
sound-setting will list bluez input/output devices, but they are not
active devices by default. This looks very weird that sound-setting
lists only one input device and one output device, but they are not
active. To change this situation, we add an argument, the policy is
if a new source/sink is connected and current default source/sink's
active_port is AVAILABLE_NO, we let the new added one switch to
default one.
BugLink: http://bugs.launchpad.net/bugs/1369476
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Calling the callback while setting it up can make things
complicated for clients, as the callback can do arbitrarily
things.
In this case, a protocol error caused the srbchannel to be
owned by both the pstream and the native connection.
Now the read callback is deferred, making sure the callback
is called from a cleaner context where errors are handled
appropriately.
Reported-by: Tanu Kaskinen <tanu.kaskinen@linux.intel.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
As a third parameter, add the number of samples to read/write in
every iteration. This will help slow CPUs.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This allow 'off' profile to be choosen when no other profile is available
which is considered better since it requires less resources than other
profiles.
Now that we have switched to using the mixer handle only,
there is no use for sending hctl handles around.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Use the new mixer API to get callbacks, instead of using the hctl
API. Using the hctl API caused a memory leak, because alsa-lib itself
used the hctl callbacks, which we were previously overriding.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Instead of using the hctl interface, we can find controls belonging
to other iface types than "mixer". We do this by introducing a new
mixer class "SND_MIXER_ELEM_PULSEAUDIO" and create snd_mixer_elem's
for all PCM and CARD iface types (as Jacks are of the CARD type and
ELD controls are of the PCM type).
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Register as a HandsfreeAudioAgent with oFono during backend
initialization and unregiter during backend finalization. This commit
also adds a check when receiving method calls or signals to make sure
the sender matches with the D-Bus service we're registered with.
The remapper and channel mixing code have (faster) specialized and (slower)
generic code certain code path. The flag force_generic_code can be set to
force the generic code path which is useful for testing. Code duplication
(such as in mix-special-test) can be avoided, cleanup patches follow.
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
New function allows to pass data pointer that is a member
of the outer structure that need to be freed too when data
is not needed anymore.
Signed-off-by: Lukasz Marek <lukasz.m.luki2@gmail.com>
Recognize the Dock headphone jack in the same way the normal & front
headphone jacks are detected.
Reviewed-by: David Henningsson <david.henningsson@canonical.com>
Usually, PA will use the PULSE_SERVER X11 property instead of using XDG_RUNTIME_DIR,
so this environment variable does not matter.
If this property is not available, or if one is using the pacmd cli protocol,
the client will go ahead and call pa_make_secure_dir on XDG_RUNTIME_DIR/pulse.
This will either fail (if you're another regular user), or succeed (if you're root).
Both scenarios are bad - failing will cause the connection to fail, and succeeding
is even worse, as it can cause *other* connections to fail (as the directory
ownership has changed).
Instead fail and complain loudly.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=83007
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Valgrind is not correctly handling ALSA TLV syscalls, which leads
to false warnings, looking like this:
"Conditional jump or move depends on uninitialised value(s)"
Unfortunately, alsa-lib itself also uses these values which valgrind
falsely believe are uninitialized, so not all warnings are removed,
but this is what we can do from PA until the valgrind bug is fixed.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
In the (theoretical) case that no other elements exists but
"Line HP Swap", the presence of that element signals that there are
headphone and line-out outputs, otherwise there would be nothing to
swap.
building PA with -O0 leads to test failure in mix-test on i386
issue reported by Felipe, see
http://lists.freedesktop.org/archives/pulseaudio-discuss/2014-August/021406.html
the problem is the value 0xbeffbd7f: when byte-swapped it becomes 0x7fbdffbe and according
to IEEE-754 represents a signalling NaN (starting with s111 1111 10, see http://en.wikipedia.org/wiki/NaN)
when this value is assigned to a floating point register, it becomes 0x7ffdffbe, representing
a quiet NaN (starting with s111 1111 11) -- a signalling NaN is turned into a quiet NaN!
so PA_FLOAT32_SWAP(PA_FLOAT32_SWAP(x)) != x for certain values, uhuh!
the following test code can be used; due to volatile, it will always demonstrate the issue;
without volatile, it depends on the optimization level (i386, 32-bit, gcc 4.9):
// snip
static inline float PA_FLOAT32_SWAP(float x) {
union {
float f;
uint32_t u;
} t;
t.f = x;
t.u = bswap_32(t.u);
return t.f;
}
int main() {
unsigned x = 0xbeffbd7f;
volatile float f = PA_FLOAT32_SWAP(*(float *)&x);
printf("%08x %08x %08x %f\n", 0xbeffbd7f, *(unsigned *)&f, bswap_32(*(unsigned *)&f), f);
}
// snip
the problem goes away with optimization when no temporary floating point registers are used
the proposed solution is to avoid passing swapped floating point data in a
float; this is done with new functions PA_READ_FLOAT32RE() and PA_WRITE_FLOAT32RE()
which use uint32_t to dereference a pointer and byte-swap the data, hence no temporary
float variable is used
also delete PA_FLOAT32_TO_LE()/_BE(), not used
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Reported-by: Felipe Sateler <fsateler@debian.org>
Debian/kFreeBSD 9.2 comes with sys/capability.h but it is not usable; work around it
the patch does several things:
* it makes the comment point to the correct bugtracker issue: https://bugs.freedesktop.org/show_bug.cgi?id=72580
* it handles Debian/kFreeBSD the same way as FreeBSD
* it logs a warning that capabilities are actually NOT dropped
daemon/caps.c: In function ‘pa_drop_caps’:
daemon/caps.c:93:2: error: #error "Don't know how to do capabilities on your system. Please send a patch."
#error "Don't know how to do capabilities on your system. Please send a patch."
^
Makefile:9575: recipe for target 'daemon/pulseaudio-caps.o' failed
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Since we don't have "limited" clients, a client that authenticates
correctly is automatically authorized. However, it's the authentication
that can go wrong, rather than the authorization.
Buglink: https://bugs.freedesktop.org/show_bug.cgi?id=78566
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Debug and info messages are primarily meant for developers,
rather than end users. Let's save translators' time,
and leave them untranslated.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Module-device-restore sets reference_volume, but soft_volume remains at
zero dB, so if a device only has soft_volume (i e no hw volume controls),
its volume was not restored correctly.
Reported-by: Richardo Salveti de Araujo <ricardo.salveti@canonical.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If the transport for the profile doesn't exist, the old behaviour was
to leave cp->available at the default value, which is
PA_AVAILABLE_UNKNOWN, but if there's no transport, the profile should
be marked as unavailable.
Since the RAOP sink supports only some formats and channel counts, we
shouldn't blindly use pa_core.default_sample_spec. This patch changes
things so that we default to PA_SAMPLE_S16NE and 2 channels, and only
take the sample rate from pa_core.default_sample_spec.
With the new multichannel profile, we can remove this one and
handle the four channel input as a generic multichannel fallback.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
json_tokener_parse() simply returns NULL on error these days
latest json-c (post 0.12) doesn't automatically include json-c/bits.h anymore
causing compilation errors
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
drop support for json 0.9 and require json-c 0.11 (this will also avoids confusion
which json package is needed due to the upstream rename)
json 0.9 lacks json_object_object_get_ex()
json-c 0.11 was released 20130402
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
pulse/format.c: In function 'pa_format_info_get_prop_type':
pulse/format.c:252:5: warning: implicit declaration of function 'is_error' [-Wimplicit-function-declaration]
pulse/format.c:287:13: warning: 'json_object_object_get' is deprecated (declared at /usr/local/include/json-c/json_object.h:290) [-Wdeprecated-declarations]
pulse/format.c:293:13: warning: 'json_object_object_get' is deprecated (declared at /usr/local/include/json-c/json_object.h:290) [-Wdeprecated-declarations]
pulse/format.c: In function 'pa_format_info_get_prop_int_range':
pulse/format.c:364:5: warning: 'json_object_object_get' is deprecated (declared at /usr/local/include/json-c/json_object.h:290) [-Wdeprecated-declarations]
pulse/format.c:369:5: warning: 'json_object_object_get' is deprecated (declared at /usr/local/include/json-c/json_object.h:290) [-Wdeprecated-declarations]
pulse/format.c: In function 'pa_format_info_prop_compatible':
pulse/format.c:676:9: warning: 'json_object_object_get' is deprecated (declared at /usr/local/include/json-c/json_object.h:290) [-Wdeprecated-declarations]
pulse/format.c:680:9: warning: 'json_object_object_get' is deprecated (declared at /usr/local/include/json-c/json_object.h:290) [-Wdeprecated-declarations]
json-c 0.10 (released 20120530) added json_object_object_get_ex()
json-c 0.12 (released 20140410) deprecated json_object_object_get()
PulseAudio depends on json 0.9 or json-c 0.11, drop support for json 0.9
in a subsequent patch and require json-c 0.11 (this will also avoids confusion
which json package is needed due to the upstream rename)
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
json-c documentation states that "No reference counts will be changed.
There is no need to manually adjust reference counts through the
json_object_put/json_object_get methods unless..."
hence fix pa_format_info_get_prop_type() and pa_format_info_get_prop_int_range();
note that pa_format_info_prop_compatible() is OK
the json_object_array_get_idx() bug reported by Arun, thanks!
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Cc: Arun Raghavan <arun@accosted.net>
This commit adds basic support for devices implementing HSP Headset
Unit, HSP Audio Gateway, HFP Handsfree Unit, HFP Audio Gateway to the
BlueZ 5 bluetooth audio devices driver module (module-bluez5-device).
handle_srbchannel_memblock() should return when memblock sanity checks fail
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Cc: David Henningsson <david.henningsson@canonical.com>
return from setup_srbchannel() when pa_srbchannel_new() fails
pa_srbchannel_new() depends on HAVE_SYS_EVENTFD_H, e.g. Debian/kFreeBSD doesn't
have it
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Acked-by: David Henningsson <david.henningsson@canonical.com>
pa_fdsem_open_shm() returns NULL when HAVE_SYS_EVENTFD_H is #undefined
pa_srbchannel_new() and pa_srbchannel_new_from_template() depend on
pa_fdsem_open_shm() and shall properly cleanup stuff, and return NULL as well;
otherwise, function pa_fdsem_get() will assert:
Assertion 'f' failed at pulsecore/fdsem.c:284, function pa_fdsem_get(). Aborting.
Debian/kFreeBSD doesn't HAVE_SYS_EVENTFD_H
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Cc: David Henningsson <david.henningsson@canonical.com>
handle both signals on Debian/kFreeBSD, otherwise sigbus-test fails:
Running suite(s): Sig Bus
Let's see if this worked: This is a test that should work fine.
And memtrap says it is good: yes
tests/sigbus-test.c:59:E:sigbus:sigbus_test:0: (after this point) Received signal 11 (Segmentation fault)
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
on systems lacking #defines HAVE_ACCEPT4, HAVE_PIPE2, SOCK_CLOEXEC
pulsecore/core-util.c: In function 'pa_open_cloexec':
pulsecore/core-util.c:3348:1: warning: label 'finish' defined but not used [-Wunused-label]
pulsecore/core-util.c: In function 'pa_socket_cloexec':
pulsecore/core-util.c:3370:1: warning: label 'finish' defined but not used [-Wunused-label]
pulsecore/core-util.c: In function 'pa_pipe_cloexec':
pulsecore/core-util.c:3393:1: warning: label 'finish' defined but not used [-Wunused-label]
pulsecore/core-util.c: In function 'pa_accept_cloexec':
pulsecore/core-util.c:3415:1: warning: label 'finish' defined but not used [-Wunused-label]
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
to print a pa_usec_t, the format specifier to use is "%" PRIu64
modules/module-combine-sink.c: In function 'update_latency_range':
modules/module-combine-sink.c:750:5: warning: format '%lu' expects argument of type 'long unsigned int', but argument 6 has type 'pa_usec_t' [-Wformat]
modules/module-combine-sink.c:750:5: warning: format '%lu' expects argument of type 'long unsigned int', but argument 7 has type 'pa_usec_t' [-Wformat]
to print a size_t, use %zu
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
The old logic assumed that if path A was a subset of path B, the
element list in B would have all elements of A in the beginning of
B's list, in the same order as A. This assumption was invalid, causing
some subset cases to not get detected. We need to search through the
full element list of B every time before we can conclude that B
doesn't have the element that we're inspecting.
The IS_ACTIVE() macro does a pa_sink/source_get_state() on our sink and
source, which does not work in the state change callback, since the
state is not actually committed at that point.
There was no code that included files from other directories using
the #include "..." style before.
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
-S makes the option parser to not try parsing arguments as options
after "--" has appeared in the command line.
-A "-*" makes the option parser to not try parsing arguments as
options after the first non-option argument. The "-*" pattern means
that if there are unrecognized parameters that look like options
(i.e. start with a dash), those should not terminate the option
parsing.
The options were divided to multiple sets to prevent (or at least try
to prevent) completing e.g. --server after -s was already given. This,
however, caused problems, because after the user had written
"pactl --server foo", further completions stopped to work. The
"server" option set didn't contain any other options, so once Zsh
detected that the "server" option set was in use, it thought that no
other options were valid.
The special casing for "-s", "-n", "--server" and "--client-*" at the
end of _pactl_completion() was probably an attempt to deal with this
problem. Those special cases are unnecessary now that the option
specification given to _arguments is more correct.
_set_remote() is supposed to find out if a remote server has been
specified on the command line, but previously it only checked for -s
and ignored --server, causing the completion code to connect to the
local server instead when it should have connected to the remote
server to get the data for the completions.
This makes the Zsh completions work out-of-the-box. I also moved
pulseaudio-zsh-completion.zsh to zsh/_pulseaudio to be in line with
the common naming convention of Zsh completion files.
This makes it easy to log a message every time the reference ratio
changes. I also need to add a hook for reference ratio changes, but
that need will go away if the stream relative volume controls will be
created by the core in the future.
PA_MAYBE_INT16_SWAP() should call PA_INT16_SWAP(), not PA_INT32_SWAP
PA_MAYBE_INT16_SWAP() is not used (yet), so no big deal :)
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
for example, the conversion function for
convert_from_float32ne(PA_SAMPLE_S16LE) can also be used for
convert_to_s16ne(PA_SAMPLE_FLOAT32LE)
v2: ARM can potentially be big- or little endian; only apply
optimization on LE based on WORDS_BIGENDIAN #define (thanks, Tanu)
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
In case all other profiles fail, try this fallback mapping as well.
It allows the device to specify the channel count, so it can be used
for devices that only supports being opened in multichannel mode.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
A fallback mapping or profile will only be considered for probing
if all non-fallback profiles fail.
If auto-profiles are used, a profile made up of one non-fallback
mapping and one fallback mapping will be considered a fallback profile.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Allow a mapping to relax the exact channel restriction:
exact-channels = yes | no # If no, and the exact number of channels is not supported,
# allow device to be opened with another channel count
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Remove extra-hdmi.conf, as the performance reasons behind it are invalid
Add 7.1 profiles
Add extra HDMI devices, for a total of 8
Add DTS-encoded profiles (they need dcaenc from git)
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
tunnel-new handled a corked stream conditional in the thread_func to be
sure the stream isn't corked. Un/Corking is now handled in the
state change callback.
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
The stream is now corked when the sink or source becomes suspended and
uncorked when it's back idle/ready.
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Since rtkit v11, the top limit for rttime is 200 ms (previously it
was wrongly limited to 2 seconds).
Reviewed-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Ricardo Salveti de Araujo <ricardo.salveti@canonical.com>
This comment can potentially save a lot of debugging effort and fixing
an ABI break, even though I don't think it's particularly likely that
anyone will ever extend pa_ext_device_manager_role_priority_info.
A recent patch broke the build on FreeBSD, which does not have
HAVE_CREDS defined. Also, make sure any attempts to enable the
srbchannel on such architectures fail.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=80642
Reported-by: Ryan Lortie
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Runs four tests:
1) Small packets, iochannel
2) Big packets, iochannel
3) Small packets, srbchannel
4) Big packets, srbchannel
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The srbchannel is enabled if protocol version >= 30 and
SHM is available. There is also a module parameter
srbchannel=false that can be used for disabling the srbchannel.
The setup is done in these steps:
1) Server receives authentication (like today)
2) Server sends enable_srbchannel to client
3) Server sends memblock to client
4) Client receives enable_srbchannel
5) Client receives memblock
6) Client sends enable_srbchannel back to server
7) Client switches over
8) Server receives enable_srbchannel and switches over
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This increments protocol version to v30 and adds two new commands
to enable and disable an shm ringbuffer, as well as client side
implementation.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
For writing, we prefer writing through the srbchannel if one is available,
and we have no ancil data to send.
For reading, we support reading from both in parallel. This meant replicating
a struct used for reading, so a lot of this patch is just a search/replace in
do_read to use the appropriate channel for reading.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
To keep the data and the ringbuffer separate, let's add another
mempool just for the ringbuffer(s). That way, the client can open
the ringbuffer shm file in rw mode and keep the data in ro mode.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
patch 'memblock, pstream: Allow send/receive of remote writable memblocks'
adds an extra parameter to pa_memimport_get()
change test program accordingly
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Cc: David Henningsson <david.henningsson@canonical.com>
The shared ringbuffer memblock must be writable by both sides.
This makes it possible to send such a memblock over a pstream without
the "both sides writable" information getting lost.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This is a preparation for the shm ringbuffer, which needs to be able
to be writable by both sides, because there are atomic variables they
both need to modify.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
An shm ringbuffer that is used for low overhead server-client communication.
Signalling is done through eventfd semaphores - it's based on pa_fdsem to avoid
syscalls if nothing is waiting on the other side.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This patch adds support to iochannel, pstream and pstream-util
to send file descriptors over a unix pipe.
Currently we don't support writing both creds and fds in the same
packet, it's either one or the other (or neither).
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The file descriptors are read from the iochannel just like the creds are.
So instead of passing just creds (and creds_valid), we now pass the
entire pa_ancil struct.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
To support later patches that add sending/receiving file descriptors,
let's add this struct.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
As a way to highlight warnings and errors in GCC output
This will be available in GCC 4.9, but some distros backported
the feature to lower versions
http://gcc.gnu.org/gcc-4.9/changes.html
As the automake documentation says:
AM_CPPFLAGS: The contents of this variable are passed to every compilation
that invokes the C preprocessor; it is a list of arguments to the preprocessor.
For instance, -I and -D options should be listed here
AM_CFLAGS: This is the variable the Makefile.am author can use to pass in
additional C compiler flags.
http://www.gnu.org/software/automake/manual/html_node/Program-Variables.html
The code in the "io_fail" section was only used for HUP handling, but
there were jumps to there also from places where reading or writing
failed, because the read/write failure could have been caused by HUP.
This patch simplifies things by checking for HUP condition before
trying to read or write. Now if reading or writing fails, we will
jump to "fail" directly instead of going via the "io_fail" label. As
a result, the "io_fail" label isn't needed any more.
Previously relative cookie paths were searched from the home
directory, now they are searched from the config home directory. This
fixes the problem that XDG_CONFIG_HOME didn't have effect on cookie
paths.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=75006
If a relative path is passed to pa_authkey_load(), it will interpret
the path as relative to the home directory. This is wrong, because
relative paths should be interpreted to be relative to the config home
directory. Before fixing pa_authkey_load(), this patch prepares for
the change by using absolute paths when the file actually needs to be
in the home directory (i.e. the fallback cookie path for the native
protocol and the default cookie path for the esound protocol).
The only place where pa_authkey_load() was called was in
pa_authkey_load_auto(), and the only functionality that
pa_authkey_load() was to log a warning if load() fails. That log
message is now in pa_authkey_load_auto(), so pa_authkey_load() has no
use any more.
pa_context already ignored the return value of pa_client_conf_load(),
so the only places where the return value was not ignored were the
D-Bus server lookup thing and pax11publish. I don't think those cases
are negatively affected if they ignore errors in opening or parsing
client.conf.
pa_client_conf_env() never failed anyway, so returning int was
obviously redundant.
block_usec should be determined by the sink max latency, not the other
way around. This change doesn't cause any change in behaviour, but
makes the code more logical. Further updates to block_usec are already
done correctly, so this is the only place that needs modification.
Mark the sink as DYNAMIC_LATENCY and implement update_sink_latency_range
on its sink-input to collect the combined latency range of all sinks.
Implement update_requested_latency on the sink to configure the final
latency by combining the sink-input requested latencies. This makes us
honour the client latency request.
Also add more debug log.
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=47899
- Make sure "no evacuation sink/source found" is not printed when
fallback source/sink is selected
- Restore previous behaviour of fallback sink/source selected
(first one instead of last one)
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Recently met a problem: when I disconnect the bluetooth headset, the
pulseaudio automatically switch the sound to sink of HDMI output
instead of the sink of internal speaker even though there is no HDMI
cable connected.
To fix this problem, I want to change the rule of selecting the target
sink if the default sink is not available. (same rules apply to the
source selecting):
construct a new hashmap with all ports (of all relevant sinks) and
then call find_best on the new hashmap to find the best port, finally
find the corresponding sink using the best port.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Having an extra variable that tracks the wakeup status introduces a
race where the variable is set but the data has yet to propagate from
the write end of the pipe to the read end. When this happens the
system goes into a tight loop as select() always returns immediately.
There are several intertwined changes that I couldn't separate into
nicer commits. This is mostly just refactoring, but this also fixes
a bug: the old code set the device valid in parse_device_properties()
even if the device's adapter was invalid (had NULL address).
To improve the clarity of the code, I split the device_info_valid
variable into two booleans: properties_received and valid.
I added function device_update_valid() that checks all conditions that
affect the device validity. The function can then be called from any
place where something changes that potentially affects the device
validity. However, currently the only validity-affecting thing that
can change is the device adapter, so device_update_valid() is only
called from set_device_adapter().
I added the aforementioned set_device_adapter() function so that
whenever the adapter is set, the device validity gets updated
automatically.
The new properties_received variable allowed me to remove the
is_property_update function parameters.
This is a cosmetic change. There are a couple of places where we check
whether the adapter object is valid, and while checking whether the
address property is set works just fine, I find it nicer to have a
dedicated flag for the object validity. This improves maintainability
too, because if there will ever be more adapter properties that affect
the adapter validity, the places that check if the adapter is valid
don't need to be updated.
On FIONREAD returning 0 bytes, we cannot return success, as the caller
(rtpoll_work_cb in module-rtp-recv.c) would then try to
pa_memblock_unref(chunk.memblock) and, because memblock is NULL, trigger
an assertion.
Also we have to read out the possible empty packet from the socket, so
that the kernel doesn't tell us again and again about it.
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
This fixes assertion failures that manifest themselves with cards that
support only weird rates such as 37286Hz. Tested with snd-pcsp.
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=48109
Add the output from its sink-input attached callback and remove it
again from the detach callback. This simplifies some output_enable
and we can also avoid posting 2 messages for the sink.
Surround 2.1 is one of the more common surround profiles these days,
so it's about time we support it.
The "surround21" was added to alsa-lib a few months ago, and there
hasn't yet been an alsa-lib release since, but I doubt it will change.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This name is more acurate with regards of what role we're currently
playing and we've already been using it in
pa_bluetooth_profile_to_string() since 449d6cb.
As it is implemented, the early request mode can in some cases be counter-productive. The mode is designed to give the client a steady request/report rate of small-ish chunks (A somewhat silly client requirement but at least Flash and Firefox break horribly when you break this.).
Unfortunately PulseAudio does not have any mechanism for telling a sink/source how often it should request/report data. So a more blunt hack was applied where the entire latency is restricted to the fragment size.
So far so good, but where the current code breaks down is when the sink cannot satisfy this tiny latency request. We then "report" to the client what we can guarantee by setting the fragment size to the sink's/source's full buffer size/latency.
This severely changes the resulting buffer attributes from what the client requested, and in practice breaks applications. The most prominent user of this feature is the ALSA plugin, and it doesn't even have a mechanism of adapting to the server giving back something different than what was requested.
So long term, the whole early request mode needs to be implemented in a better way. Either the sink's/source's need to grow the ability to control request/report rate. Or we put some form of timer based emulation in front of them on behalf of these clients.
Short term, we should change the behaviour of what happens when we cannot guarantee a fragment rate. Instead of giving the client really shitty buffering parameters as a result, we should just keep the requested attributes and do things on a best-effort basic. Basically how things would behave if the client didn't have the early request bit at all.
The attached patch does just that, as well as expand on the comment about how the early request thing is implemented.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=66962
speex_resample_float() does not work with speex compiled with
--enable-fixed-point, because speex expects its float input
to be normalized to ±32768 instead of the more usual ±1.
It is possible to fix speex_resample_float(), as demonstrated at
http://lists.freedesktop.org/archives/pulseaudio-discuss/2014-May/020617.html
However, a better idea is to avoid using the speex-float resampler and
the associated s16 <-> float conversions that speex will immediately undo
internally if it is known that speex has been compiled with FIXED_POINT.
So, transparently change speex-float-* to speex-fixed-* in that case.
Signed-off-by: Alexander E. Patrakov <patrakov@gmail.com>
Reported-by: Fahad Arslan <fahad_arslan@mentor.com>
Cc: Damir Jelić <poljarinho@gmail.com>
Cc: Peter Meerwald <pmeerw@pmeerw.net>
FIXED_POINT detection is based on code by Peter Meerwald.
The warnings were produced because the command-line flag redefined the
value of _FORTIFY_SOURCE coming from the specs on some distributions,
including Gentoo. So, undefine this macro before defining it.
Forcing all mute changes to go through set_mute() makes it easier to
check where the muted field is changed, and it also allows us to have
only one place where notifications for changed mute are sent.
This refactoring reduces duplication, as mute_changed() used to do the
same things as set_mute(). Other benefits are improved logging
(set_mute() logs the mute change, mute_changed() used to not do that)
and the soft mute state is kept up to date, because set_mute() sends
the SET_MUTE message to the IO thread.
The set_mute_in_progress flag is an extra precaution for preventing
recursion in case a sink/source implementation's set_mute() callback
causes mute_changed() to be called. Currently there are no such
implementations, but I think that would be a valid thing to do, so
some day there might be such implementation.
The callback just called pa_source_output_get_mute(), which doesn't
have any side effects, and the return value wasn't used either, so
the callback was essentially a no-op.
Currently the alsa sink and source write directly to s->muted during
initialization, but I think it's better to avoid direct writes, and
use the set_mute() function instead, because that makes it easier to
figure out where s->muted is modified. This patch prevents the
set_mute() call from crashing in the state assertion.
Forcing all volume changes to go through set_volume_direct() makes
it easier to check where the stream volume is changed, and it also
allows us to have only one place where notifications for changed
volume are sent.
Forcing all reference volume changes to go through
set_reference_volume_direct() makes it easier to check where the
reference volume is changed, and it also allows us to have only one
place where notifications for changed reference volume are sent.
State can be used by remap function implementations to
speed up the remapping, e.g. by precomputing things or
even by generating specialized code for a specific channel
remapping task
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Initialization of the remap structure now happens in one place
Rename calc_map_table() to setup_remap(), copy sample format and
channel specs; the remap structure is initialized when we know the
work sample format of the resampler
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
pa_init_remap_func() only sets the appropriate remapping function, it
does not initialize the pa_remap struct
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
"i->save_muted = i->save_muted || mute" makes no sense. The intention
was most likely to use "save" instead of "mute" in the assignment.
This line originates from reverting the volume ramping code, commit
8401572fd5.
The idea of "i->save_muted |= save" is that even if the mute state
doesn't change, save_muted should still be updated, but only if the
transition is from "don't save" to "save".
Changing "!i->muted == !mute" to "mute == i->muted" is cosmetic only.
The rationale behind the old form was probably that when we still had
pa_bool_t, booleans could in theory be defined as int, so comparing
the values without the ! operator was not entirely safe. That's
unnecessary now that we use the standard bool type, which can only
have values 0 or 1.
A value of 0 for adjust_time should disable rate adjustment.
Fix a bug where a 0 value causes rate adjustment to be called
continuously instead after an unsuspend event.
Initially (in commit ef422fa4ae),
pa_make_secure_dir followed a simple principle: "make a directory, or,
if it exists, check that it is suitable". Later this evolved into "make
a directory, or, if it exists, ensure that it is suitable". But the
check remained.
The check is now neither sufficient nor necessary. On POSIX-compliant
systems, the fstat results being checked are actually post-conditions of
fchmod and fchown. And on systems implementing POSIX ACLs, fstat only
reflects a part of the information relevant to the security of the
directory permissions, so PulseAudio could accept an existing insecure
directory anyway.
Also, the check still fires on non-POSIX-compliant filesystems like CIFS.
As a user cannot do anything to fix it, just accept insecure permissions
in this case.
This parameter was never assigned, so just remove it.
Note that the only current user of this function is shmasyncq.c,
which is unused - we don't even build it. But I fixed it up anyway.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Source ports hashmap is created without value freeing function, which
results in (hashmap values) device ports not being freed when source
ports are removed or module is unloaded. This results in memory leak
during normal operation and during daemon shutdown dbus_protocol shared
object isn't unreferenced correctly, leaving dbus_protocol object in
core->shared, which causes assert when shared hashmap is checked for
isempty() before freeing.
This generates a list of deprecated things, which is accessible from
the table of contents frame. The list, however, isn't the important
thing here. The important thing is that this also prevents doxygen
from stripping all documentation for the deprecated things.
Previous discussion:
http://thread.gmane.org/gmane.comp.audio.pulseaudio.general/18794
This patch also adds a description how the heuristic works and mentions that
there is a scaling factor that can be adjusted if there is audible clipping.
VOL_RELATIVE if a bit flag (1 << 4), hence we can simply do
if (vol_flags & VOL_RELATIVE) ...
instead of
if ((vol_flags & VOL_RELATIVE) == VOL_RELATIVE) ...
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Must use one way to specify volumes consistently, e.g.
+3dB +3dB, mixing different ways is not allowed, such as
40% 1000
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Example: pactl set-sink-volume "sink_name" 32000 40000
If the number of volumes provided is different than the number of channels
(excluding the case where a single volume is provided), an error message
is displayed explaining why the volumes could not be set.
patch proposed by Parin Porecha
code refactoring and commit message slightly edited by Peter Meerwald
fix bug
https://bugs.freedesktop.org/show_bug.cgi?id=77108
see getopt(3):
""By default, getopt() permutes the contents of argv as it scans, so that
eventually all the nonoptions are at the end. Two other modes are also
implemented. If the first character of optstring is '+' or the envi‐
ronment variable POSIXLY_CORRECT is set, then option processing stops
as soon as a nonoption argument is encountered. If the first character
of optstring is '-', then each nonoption argv-element is handled as if
it were the argument of an option with character code 1. (This is used
by programs that were written to expect options and other argv-elements
in any order and that care about the ordering of the two.) The special
argument "--" forces an end of option-scanning regardless of the scan‐
ning mode.""
prepend optstring with '+' to use POSIXLY_CORRECT mode
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
On Haswell hardware, there are multiple HDMI outputs capable of
digital sound output. As they were identically named, KDE's control
center was unable to distinguish them, restored the wrong profile and
thus routed sound to the wrong HDMI monitor.
Also, having identically-named menu items in other mixer applications
looks like a bug.
Now that we have a generic function in device-port.h, we can use
it instead of the custom one.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
We now have a port->card pointer, we can use it instead of iterating
over cards to find the correct one.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Also, initialize userdata with zeros to avoid invalid pointers in
client_free().
This fixes a crash when client_free() is called before
create_client(). The whole issue could be avoided by using some other
mechanism than defer events for running the two functions, but I'll
do that change later (I have also other cleanups planned for
zeroconf-publish).
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=76184
The old code loaded cookies at the time of loading the client
configuration, which could lead to creation of multiple cookie files.
For example, when pa_client_conf_load() was called, the default cookie
file was created, and then if PULSE_COOKIE was set,
pa_client_conf_env() would create another cookie file.
This patch moves the loading of the cookie to a separate function,
which pa_context calls just before needing the cookie, so the cookie
won't be loaded from the default file if PULSE_COOKIE is set. This
patch also splits the single cookie and cookie_file fields in
pa_client_conf into multiple fields, one for each possible cookie
source. That change allows falling back to another cookie source if
the primary source doesn't work.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=75006
Avoid unpredictable behaviour in case e.g. the HOME environment
variable is incorrectly set up for whatever reason.
I haven't seen non-absolute HOME anywhere, but this feels like a good
sanity check anyway.
Servers older than 0.9.15 don't know anything about cards, and card
operations will return a NULL pa_operation object when connected to
that old server. We must check the pa_operation pointer before passing
it to pa_operation_unref(), otherwise a NULL operation will result in
a crash.
In case a port has not yet been saved, which is e g often the case
if a sink/source has only one port, reading volume/mute will be done
without port, whereas writing volume/mute will be done with port.
Work around this by setting a default port before the fixate hook,
so module-device-restore can read volume/mute for the correct port.
BugLink: https://bugs.launchpad.net/bugs/1289515
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
hwparams_copy needs to be reset (as it is also reset for the third and
fourth try) before the second try.
If the reset is not done and the first try fails:
D: [lt-pulseaudio] alsa-util.c: Maximum hw buffer size is 743 ms
I: [lt-pulseaudio] alsa-util.c: snd_pcm_hw_params_set_buffer_size_near() failed: Invalid argument
I: [lt-pulseaudio] alsa-util.c: snd_pcm_hw_params_set_period_size_near() failed: Invalid argument
I: [lt-pulseaudio] alsa-util.c: snd_pcm_hw_params_set_buffer_size_near() failed: Invalid argument
D: [lt-pulseaudio] alsa-util.c: Set only period size (to 1102 samples).
We have three failures and finally the fourth (only period size) succeed.
With this patch:
D: [lt-pulseaudio] alsa-util.c: Maximum hw buffer size is 743 ms
I: [lt-pulseaudio] alsa-util.c: snd_pcm_hw_params_set_buffer_size_near() failed: Invalid argument
D: [lt-pulseaudio] alsa-util.c: Set period size first (to 1102 samples), buffer size second (to 4408 samples).
We only fail with the first try, the second (period followed by buffer) is
fine.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
This fixes a case where pa__done() is called while
AVAHI_MESSAGE_PUBLISH_ALL is waiting for processing. The
pa_asyncmsgq_wait_for(AVAHI_MESSAGE_SHUTDOWN_COMPLETE) call will
process all pending messages, and processing AVAHI_MESSAGE_PUBLISH_ALL
causes publish_all_services(), and that in turn accesses u->services,
which has been already freed at this point. If we are shutting down,
we shouldn't react to any of the messages that the Avahi thread is
sending to the main thread.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=76184
When given an explicit device.description in card_properties, prefer
this information over other default prefixes (e.g. 'Built-in Audio')
when constructing sink/source descriptions.
For example, if I manually configure the card description to be
"FooBar", I then expect that the sinks and created by the card also
have "FooBar" in their description instead of generic "Built-in
Audio".
In some cases, "Analog Input" could show up as well as
"Headset Mic" (or "Headphone Mic"), because I forgot to add the
relevant "required-absent" lines when I added the headset mic path.
As a result, both "Analog Input" and "Headset Mic" showed up on the
Logitech USB 530 Headset.
Reported-by: Steve Magoun <steve.magoun@canonical.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
CC modules/module_tunnel_sink_la-module-tunnel.lo
modules/module-tunnel-source-new.c: In function 'read_new_samples':
modules/module-tunnel-source-new.c:145:16: warning: declaration of 'read' shadows a global declaration [-Wshadow]
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
If mixer_handle is not NULL, then hctl_handle won't be NULL either.
The redundant check was confusing, because it looked like we would
leak the mixer_handle if mixer_handle is non-NULL and hctl_handle is
NULL.
Currently the latency information is being updated based on the encoded
SBC data instead of the decoded PCM data. Fixing this required moving
the timing update to be after the packet has been decoded.
The Nokia E7 running Symbian Belle Refresh seems to generate invalid SBC
packets every few minutes. This causes pulseaudio to disconnect the
stream and log "SBC decoding error (-3)".
If a single packet is bad, pulseaudio should keep playing the stream.
Some people want module-rtp-send to send silence when the sink that is
monitored goes idle, and some people want module-rtp-send to pause the
RTP stream to avoid unnecessary bandwidth consumption.
If a stream is started corked and remains corked, the sink/source
remained idle without being properly suspended. This patch fixes
that issue.
BugLink: https://bugs.launchpad.net/bugs/1284415
Tested-by: Ricardo Salveti <ricardo.salveti@canonical.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Steps to reproduce:
1) Leave LFE remixing disabled (the default)
2) Start playback of stereo material on e g 5.1 surround, notice nothing in LFE
3) Now change profile to e g 4.0 surround and then back to 5.1 surround
4) Notice that LFE channel is now remixed
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This makes sure that there is no window between pa_sink/source_new()
and _put() where enumerating sinks/sources causes an assert (several
calls in sink/source_get_info need a linked sink or source).
A segfault was reported on this line:
pa_return_val_if_fail(PA_SINK_IS_LINKED(pa_sink_get_state(data->sink)), -PA_ERR_BADSTATE);
After expanding the pa_sink_get_state() macro, the line looks like
this:
pa_return_val_if_fail(PA_SINK_IS_LINKED(data->sink->state), -PA_ERR_BADSTATE);
So data->sink was apparently NULL. That could happen if we try to fall
back to the default sink, but format negotiation fails.
This bug was introduced in commit
71816ecb7f.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=74646
This fixes a build error with mingw32:
pulsecore/.libs/libpulsecommon_4.99_la-lock-autospawn.o: In function `unref':
/home/abuild/rpmbuild/BUILD/pulseaudio-4.99.2/src/pulsecore/lock-autospawn.c:123: undefined reference to `pa_thread_free_nojoin'
collect2: error: ld returned 1 exit status
pa_thread_free_nojoin() was initially only implemented for the pthread
based pa_thread backend, because it was incorrectly assumed that
autospawning (the only user of pa_thread_free_nojoin()) is not used on
Windows.
Reported-By: Michael DePaulo <mikedep333@gmail.com>
Reintroduces a cleaned-up version of commit 30ce3a14e5 which
was reverted by 1ce71cbd8206d1be59ac62274ad83cdbe693a96a; for more information see
http://thread.gmane.org/gmane.comp.audio.pulseaudio.general/17479/focus=17487
The patch intends to reduce computational load when resampling AND remapping. The PA
resampler performs the following steps:
sample format conversion -> remapping -> resampling -> sample format conversion
In case the number of output channels is higher than the number of input channels, the
resampler has to be run more often than necessary. E.g. in case of mono to 4-channel remapping,
the resampler runs on 4 channels separately.
To ímprove this, the PA resampler pipeline is made adaptive:
if out-channels <= in-channels:
sample format conversion -> remapping -> resampling -> sample format conversion
if out-channels > in-channels:
sample format conversion -> resampling -> remapping -> sample format conversion
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
the patch changes the interface of the (internal) fit_buf() function:
fit_buf() manages the memblock of the buf chunk, it reallocates the memblock
if the requested number of bytes ('len') if larger than the memblock's size ('size')
and optionally preserves 'copy' bytes
the code should be in line with the comment now
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
rindex() appears to be "non-standard" to an extent, and it caused a
build failure on mingw32.
From the man page of rindex(): "POSIX.1-2008 removes the
specifications of index() and rindex(), recommending strchr(3) and
strrchr(3) instead."
This fixes an assertion crash:
[pulseaudio] source.c: Assertion 'PA_SOURCE_IS_LINKED(s->state)' failed at pulsecore/source.c:734, function pa_source_update_status(). Aborting.
The crash happened when a Bluetooth headset profile was changed from
a2dp to hsp. During the profile change three devices are created:
a sink, a monitor source for the sink, and a regular source. First
pa_sink/source_new() are called for each device, and that puts the
devices to u->core->sinks/sources. Then, pa_sink_put() is called for
the sink, and that in turn calls pa_source_put() for the source. At
that point module-device-manager decides to reroute all source
outputs. The non-monitor source that the Bluetooth card created hasn't
been linked yet at this stage, because it will only be linked after
the sink and the monitor source have been linked. So,
module-device-manager should take into account during the rerouting
that not all sinks and sources are necessarily linked. This patch does
that.
Reported-By: Iskren Hadzhinedev <i.hadzhinedev@gmail.com>
Not having ORC_SOURCE defined results in different tarballs depending on
whether the dev issuing 'make dist' has orc support enabled or disabled.
Specifying ORC_SOURCE unconditionally addresses that, without causing
negative effects on users not having orc in the end.
The bigger than usual bump in libpulse-simple was warranted by the
change in pa_simple_flush() that allows also record streams to be
flushed. There are no changes to the function signature, but it's in
practice a change in the ABI anyway, because new clients using the new
possibility won't work with older versions of the library.
libpulse-mainloop-glib got a bug fix in commit
68156d3f79.
A crash was observed that was caused by pa_memblockq_peek() returning
a NULL memblock in sink_input_pop_cb(). The scenario where this was
happening was
1. Delete 2 rtp-recv's connected to a ladspa-sink
2. Delete ladspa-sink
3. Delete alsa-sink
4. Create alsa-sink
5. Create ladspa-sink
6. Create 2 rtp-recv's connected to the ladspa-sink
The crash was probably caused by a rewind that made the read index go
negative while the write index was at least zero, causing there to be
a gap in the memblockq. The problematic rewind might have been caused
by adding the rtp-recv stream to the ladspa-sink. That has not been
proven, but this looks very similar to the bug that was fixed in
module-virtual-sink in commit 6bd34156b1.
pulsecore/core-format.c was recently added to libpulsecommon, and
core-format.c depends on functions in libpulse, which libpulsecommon
doesn't link to. That broke building with --as-needed. This patch adds
pulse/format.c to libpulsecommon, so that core-format.c doesn't need
to depend on libpulse any more. format.c pulls in also the dependency
to json-c.
Reported-By: Jan Steffens <jan.steffens@gmail.com>
The modargs are in both cases (a succesfull as well as a failed module
initialization) freed already in pa__done().
To avoid leaking modargs memory before they are assigned to u->modargs, the
code is reorganized to first allocate userdata, and then allocate the modargs.
Local variable ma is not needed anymore.
discussion here
http://lists.freedesktop.org/archives/pulseaudio-discuss/2013-December/019661.html
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Reported-by: poljar (Damir Jelić) <poljarinho@gmail.com>
The "fix flags" (PA_SINK_INPUT_FIX_FORMAT etc.) don't work properly
with the pa_stream_new_extended() interface. This patch fixes it so
that the same effect can be achieved by leaving some of the PCM
parameters unspecified in format info objects. Also, when converting
a sample spec to a format info when using the old pa_stream_new()
interface, the "fix flags" are taken into account in that conversion.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=68952
The check is done for clients that use pa_stream_new() but not for
clients that use pa_stream_new_extended(). This is inconsistent. We
could check that the volume channels match the channels set in the
format info struct that is passed to pa_stream_new_extended(), but
that doesn't work if the format info doesn't contain the channel
information (that can happen if the client wants the server to choose
the channel count for the stream). And it should also be possible to
pass a mono volume for a multi-channel stream. The check could be
extended to handle all these cases, but I don't see much point in
wasting time on that. The server will anyway validate the stream
parameters, it's not particularly important to fail already when the
stream is being created at the client side.
The function will be used in pa_sink_input_new() and
pa_source_output_new() to convert the sample spec given by the client
to a format info object. The set_format, set_rate and set_channels
will be set according to the stream flags (PA_SINK_INPUT_FIX_FORMAT
etc.).
The function will be used in pa_sink_input_new() and
pa_source_output_new(). The fallback parameters are used to merge the
data in the format info with the sink/source sample spec and channel
map, when the format info is lacking some information.
This also fixes an issue in pa_format_info_to_sample_spec(): it did
no validation for the channels value. Now the validation is taken care
of in pa_format_info_get_channels().
This also fixes an issue in pa_format_info_to_sample_spec(): it did
no validation for the rate value. Now the validation is taken care of
in pa_format_info_get_rate().
I will need to use the function from outside libpulse.
I added the channel map argument, because the function will be called
from another function that is expected to initialize the channel map.
I don't know if it's in practice necessary, but it shouldn't do any
harm either.
Quoting Ryan Lortie from [1]:
I assumed from my reading of the Linux code ("cap_clear()...") that it
was clearing all capabilities of the process when in fact it is only
clearing the "special to root" capabilities.
The FreeBSD version of the code indeed clears _all_ capabilities beyond
ones that the process already has (ie: cannot open any new files, create
sockets, etc.)
This has a pretty obvious adverse effect on pulseaudio's ability to do
what it needs to do -- indeed, it bombs out pretty quickly due to an
inability to read its own config file.
[1] https://bugs.freedesktop.org/show_bug.cgi?id=72580#c11
Hello.
Over time, I became aware of several instances of tempting but
semantically incorrect usage of PulseAudio API (one from my own bad
proposal of "improving" Wine, one from Parole media player and one
from Webkit-GTK). I want to document these gotchas so that other
developers don't fall for that. See the attached patch.
I have checked that the rendered HTML is correct, but need someone to
confirm the factual accuracy of the proposed changes and, possibly, to
improve the wording.
--
Alexander E. Patrakov
I don't like the expression "this Pulseaudio" (even though that's
originally written by me), just "PulseAudio" is enough. Also, on
FreeBSD there's no libcap, so let's refer only to "capabilities".
cap_init() and friends are Linux-specific, so only use them if we're on
Linux.
Add support for FreeBSD capabilities if we find <sys/capability.h> to be
available there.
Add an #else (not Linux or FreeBSD) case with an #error requesting
contributions for other platforms.
This patch keeps the cap_init check in configure.ac but removes the
error if it fails. This will ensure we link to -lcap if needed, but
won't fail for the case that capabilities are part of the core system
(as on FreeBSD).
We do however, modify the header check to ensure we fail if there is no
<sys/capability.h> at all and we are on a system where it could be
installed. The logic here is that it is better to give the user the
chance to install it than it is to proceed silently with a disabled
security feature on a system where it could easily be supported.
--without-caps remains an option if the user wants to force it.
https://bugs.freedesktop.org/show_bug.cgi?id=72580
The journal is a component of systemd, that captures Syslog messages,
Kernel log messages, initial RAM disk and early boot messages as well
as messages written to STDOUT/STDERR of all services, indexes them and
makes this available to the user.
It can be used in parallel, or in place of a traditional syslog daemon,
such as rsyslog or syslog-ng.
The journal offers a couple of improvements over traditional logging
facilities (e.g. advanced filtering capabilities).
This patch adds support for logging directly to the journal using its
native API.
while ((r = pa_mainloop_iterate(m, 1, retval)) >= 0)
;
if (r == -2)
return 1;
else if (r < 0)
return -1;
else
return 0;
the last else is never reached, discovered by coverity
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Generating docs for file mainloop-api.h...
/home/pmeerw/src/pa-missing/src/pulse/mainloop-api.h:118: warning: Found unknown command `\pa_threaded_mainloop'
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
I think this makes the code a bit nicer to read and write. This also
reduces the chances of off-by-one errors when checking the bounds of
channel count values.
I think this makes the code a bit nicer to read and write. This also
reduces the chances of off-by-one errors when checking the bounds of
sample rate values.
I think this makes the code a bit nicer to read and write. This also
reduces the chances of off-by-one errors when checking the bounds of
the sample format value.
set_scheduler() assumes that if sys/resource.h was found then we will
find RLIMIT_RTTIME there, but this is a non-POSIX extension on Linux.
Change the check to ensure that RLIMIT_RTTIME is actually defined.
Linux indeed defines this as a macro, and POSIX specifies that the other
RLIMIT_ constants must be macros, so having this as an #ifdef seems
correct.
bootstrap.sh uses some non-POSIX features of bash, so we can't use
/bin/sh. Unlike /bin/sh, bash can be installed anywhere in the path, so
we should use /usr/bin/env to find it.
This helps systems that have bash in /usr/local/bin, for example.
PCM Devices which have the BATCH flag set update the PCM pointer only with
period size granularity. Using timer based scheduling does not have any
advantage in this mode. For one devices which have that flag set usually update
the position pointer in software after getting the period interrupt. So
disabling the period interrupt is not possible for this kind of devices.
Furthermore writing to or reading from the buffer slice for the current period
is not possible since the position inside the buffer is not known. On the other
hand the tsched algorithm seems to get easily confused for this kind of
hardware, which results in garbled audio output. This typically means that timer
based scheduling needs to be manually disabled on systems with such devices.
Auto disabling tsched in this case allows these systems to run with the default
configuration.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
the main intent is to make testing different sample rate resampling
implementations easier; so far there is only global control via
resample-method (command line argument and /etc/pulse/daemon.conf)
module-remap-*'s only purpose is resampling (comprising format conversion,
channel remapping, sample rate adjustment), it can easily be introduced
into any audio pipeline
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
by using pa_modargs_get_sample_rate() we avoid inconsistant validity
checking of the sample rate in various places
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
buf in struct ffmpeg_data is reset() initially and freed, but never
actually used
when a new block is allocated ffmpeg_data->buf[c].length is used
(which is always 0) to compute the new block size
so, drop buf
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
When setting attribute foo, or in this case the card profile, in my
opinion the thing passed to the set_foo() function should be of the
type of foo, not a string identifier that can be used to search for
the actual foo in set_foo().
This is mostly a question of taste, but there's at least some small
benefit from passing the actual object: often the profile object is
already available when calling pa_card_set_profile(), so passing the
card name would cause unnecessary searching when pa_card_set_profile()
needs to look up the profile from the hashmap.
Don't call pthread_join() to join a thread from a different
process than the thread was created in. Doing so can lead to
undefined behaviour.
On OpenBSD, the symptom was a pulseaudio process with a single
thread waiting forever for other threads to join. Since that
process also held the autospawn lock, starting new pulseaudio
processes with --start kept failing. The problem was analyzed
with help from Philip Guenther.
This patch adds a pa_thread_free_nojoin() function which can
be used to free resources for a thread without a join, as
suggested by Tanu Kaskinen.
See https://bugs.freedesktop.org/show_bug.cgi?id=71738
f434087e42 introduced the potential to not select a card profile if
all the profiles were marked as unavailable.
While this is very unlikely, it's a theoretical posibility, so if the
initial choice of a profile fails, try harder.
When parsing device properties, missing adapter will result in
device_info_valid being set to -1. It is then logical that if the
adapter goes missing at a later point, device_info_valid gets set to
-1 also in that situation.
The function did two things: set device_info_valid to -1 and called
device_free() for each device in the hashmap. Setting
device_info_valid to -1 was unnecessary. The main purpose of that was
to fire DEVICE_CONNECTION_CHANGED as a side effect, but that hook is
fired anyway in device_free(), as a side effect of removing all
transports. Calling device_free() can be delegated to pa_hashmap, when
freeing or emptying it.
Normally DEVICE_CONNECTION_CHANGED is fired when the first transport
becomes connected, but it may happen that the first transport becomes
connected already before the device properties have been received. In
that case the hook should be fired at the time the device properties
are received. This patch makes the hook to be fired at the right time.
At this point this doesn't make any other practical difference than
making the code more logical, but in the next patch I'll fire the
DEVICE_CONNECTION_CHANGED hook in set_device_info_valid(), and at that
point it's important that the device isn't marked valid too early,
because otherwise external code would see "valid" devices that however
don't have the adapter set.
This reverts commit c327850d9e as
the workaround in that commit is no longer needed after the real
bug has been fixed.
Conflicts:
src/pulsecore/core-util.c
Commit 7e344b5 hade the side effect of forcing every socket to
be non-blocking on Windows. This is because of a (documented)
side effect of WSAEventSelect(). So we need to make sure to restore
blocking behaviour afterwards for relevant sockets.
When connecting to a remote server your local generated authentication
cookie is used. If remote server's cookie is different from your local
one you aren't allowed to connect. You can use the cookie argument
or define a wider acl in remote server configuration for
module-native-protocol.
It's bad form to assume in free() that any member of the struct has
been initialized. I ran into problems with this when I reordered
things in pa_sink_input_new() and pa_source_output_new().
leftover_buf points to the output buffer of a stage containing leftover
data; similar for leftover_buf_size and have_leftover
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
some resampler implementations (e.g. libsamplerate and ffmpeg) do not consume
the entire input buffer; the impl_resample() function now has a return value
returning the number of frames in the input buffer not processed
these frames must be saved in appropriate buffer and presented together with
new input data
also change the parameter names from in_samples, out_samples to in_n_frames,
out_n_frames, respectively (n_frames = samples / channels)
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
When a card is being created and no profile has been assigned
pa_card_new will attempt to select one from the list but it does that
without checking the available flag which can lead to select profiles
not available.
The size of pa_card_profile_info cannot change even if it just a field
appended to end because each entry is appended to a contiguous memory
and accessed by offset this may lead clients to access invalid data.
To fix a new struct called pa_card_profile_info2 is introduced and shall
be used for now on while pa_card_profile_info shall be considered
deprecated but it is still mantained for backward compatibility.
A new field called profiles2 is introduced to pa_card_info, this new field
is an array of pointers to pa_card_profile_info2 so it should be possible
to append new fields to the end of the pa_card_profile_info2 without
breaking binary compatibility as the entries are not accessed by offset.
These kcontrol names have started to show up lately, in
combination with surround internal speakers.
BugLink: https://bugs.launchpad.net/bugs/1236965
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The code got removed by accident during the cleanup in commit 9c438bcac6. So
this patch is needed to bring it back and make things work like documented.
There was a question in IRC about whether pa_simple_read() blocks or
not. It's already documented on the simple API overview page, but it's
good to say it also in the function reference. As a bonus, I added
some additional details to the documentation too.
The function was redundant, because all it did was call adapter_free()
for each adapter in the hashmap, and that can be delegated to
pa_hashmap when freeing or emptying it.
The pacat completion didn't complete the right devices for the --record
and --playback flags.
This patch fixes this and makes the device completion for pacat easily
expandable.
This fixes a bug where calling time_restart can leave the current event
in the cache, even though the restart scheduled the event in the future.
This would cause the event to get executed more frequently than it should.
Previously module-bluez5-discover and module-bluez4-discover were being
tracked using their pa_module pointer. But during daemon shutdown these
modules are unloaded before module-bluetooth-discover, leaving stale
pointers in module-bluetooth-discover's userdata. To avoid this problem
this commit makes module-bluetooth-discover keep track of
module-bluez5-discover and module-bluez4-discovery by their indexes.
PA_SAMPLE_24NE generated in pa_sndfile_read_sample_spec is not
handled in pa_sndfile_readf and writef function. paplay/parecord
used to get aborted for 24bit depth wav files
module-alsa-{sink,source}.c call pa_alsa_{sink,source}_new with
mapping set to NULL. Guard against this, like the rest of the
function does.
module-alsa-card does not use NULL, so this went unnoticed so far.
When the creation of u->thread fails, then pa_thread_mq_done() in
pa__done() will crash, because pa_thread_mq_init() was never called.
Allocating the thread_mq object separately, instead of embedding it
in the userdata struct, allows pa__done() to call pa_thread_mq_done()
only when necessary.
This is a cleaner solution, because it also removes paths that are
being removed because they are subsets of other paths.
Otherwise, the lingering paths could cause jack detection related
assertion failures.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=69676
Reported-and-tested-by: Kalev Lember <kalevlember@gmail.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The author of this module, Tanu Kaskinen, has said that this module
"is not suitable for general use". Also, it is still causing crashes
on card removal (see bug 69871).
Qpaeq, and possibly other tools, use this module - but they can load
the module manually if they still wish to use it.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Create a wrapper module called module-bluetooth-discover to avoid
breaking backward-compatibility of default.pa. This wrapper may
eventually be dropped altoghether with BlueZ 4 support.
For quite some time now the device driver module doesn't work well
without the discovery module, so for the BlueZ 5 support we'll prevent
the device driver module to be loaded if the discovery module is not
loaded.
Create the thread function, the render and push functions for A2DP, the
process message function for communication between the I/O thread and
the main thread, and other helper functions related to them.
Get the remote device information stored in pa_bluetooth_discovery. This
also creates the mandatory parameter 'path' for module-bluez5-device,
which is used to inform the object path of the remote device in BlueZ on
the module load.
bluetoothd always send the GetManagedObjects() reply messages with the
objects array argument following an in-depth order starting from the
root. That means parents will always be known at the time their children
objects are parsed, if clients parse the objects in the same order they
appear in the array, as we do in PulseAudio.
This commit tries to protect PulseAudio in the case bluetoothd changes
that behavior for some reason. It hasn't been tested, since this
situation never occurs.
Create the pa_bluetooth_transport structure to store information about
the bluetooth transport and utility functions to manipulate this
structure. The acquire() and release() operations are function pointers
in the pa_bluetooth_transport structure to make possible for different
transport backends to provide different implementations of these
operations. Thre is also a userdata field for the transport backend
provide data for the acquire/release functions.
This commit also creates a new function
pa_bluetooth_device_any_transport_connected() to check if there is any
audio connection between the host and a remote device.
pa_bluetooth_discovery is the struct that holds information about known
Bluetooth audio devices and other information about the Bluetooth stack.
This commit also creates bluez5-util.[ch], which will hold a lot of
utility functions to help with the BlueZ 5 support.
module-bluetooth-proximity has not worked for quite a while, since it
uses pre-BlueZ4 APIs. Nobody complained since then, which is a good
indication that it doesn't have much users. Even the original commit
message refers to it more as a toy than as something of great use: "add
new fun module that automatically mutes your audio devices when you
leave with your bluetooth phone, and unmutes when you come back"
Removing it we completely remove the dependency on libbluetooth.
We need diferent symbol prefixing for the current BlueZ 4 support and
the new BlueZ 5 support which is about to enter the codebase, to avoid
symbol clashing and crashing the daemon in the case both modules are
loaded at the same time.
This commit replaces all pa_bluetooth_ and pa_bt_ (for consistency)
prefixes with pa_bluez4_, for both lower-case and upper-case, what was
done with the following sed commands:
$ sed -i s/pa_bluetooth_/pa_bluez4_/g src/modules/bluetooth/*bluez4*
$ sed -i s/PA_BLUETOOTH_/PA_BLUEZ4_/g src/modules/bluetooth/*bluez4*
$ sed -i s/pa_bt_/pa_bluez4_/g src/modules/bluetooth/*bluez4*
$ sed -i s/PA_BT_/PA_BLUEZ4_/g src/modules/bluetooth/*bluez4*
The current set of bluetooth modules only support up to BlueZ 4. Since
the BlueZ API when through a big change with the release of BlueZ 5 the
modules will be forked into a new set for BlueZ 5.
This commit also fixes the spelling of Bluetooth (it's a trademark which
should always be spelled with capital B) and the spelling of my name,
and also update the copyright note dates throughout the Bluetooth
modules.
This reverts commit 2247b18739.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit c4bd51a345.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit d9ed42c40f.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit d22ea7ff76.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit 61e8fd8854.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit cfb96b2530.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit 114edb0696.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit 235611a7d1.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit 2f79fb580a.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit 6fdf2b05b8.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit 9615def4b9.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This reverts commit 0e4c16e120.
This is part of the reversion of BlueZ 5 support so it can be added back
in a separate set of modules. This makes the code easier to maintain and
decrease PulseAudio's binary size.
This patch fixes a small mistake where we actually log that we are
reverting to the auto resampler if we can't use the 'copy' resampler but
never do the revert.
This would lead to a crash if the user chooses the 'copy' resampler and
then tries to play something that needs to be resampled.
According to coding style, one should have one assertion per line
and not combine assertions.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
A recent feature addition added a dependency on X11, but this
dependency was not specified in Makefile.am, leading to linker
errors.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This patch updates the ax_pthread autoconf macro to the latest version
shipped with autoconf-archive: 2013.06.09.13
This also silences multiple warnings on autoconf 2.68+:
configure.ac:471: warning: AC_LANG_CONFTEST: no AC_LANG_SOURCE call detected in body
This updates the acx_libwrap.m4 macro for autoconf 2.68 and fixes
warnings like:
configure.ac:471: warning: AC_LANG_CONFTEST: no AC_LANG_SOURCE call detected in body
At the moment, port names combined from multiple devices are generated
based on the order that the devices are specified in config. This makes
programmatic use of thsee ports a bit painful, so let's make them be
combined in alphabetical order.
Add new PlaybackRate/CaptureRate values for UCM that can be used to
specify custom rates for devices. This value can either be set on the
verb, which makes it apply to all devices, or on the device to override
the verb setting.
This allows mappings to override some or all of the sample_spec used to
open the ALSA device. The intention, to start with, is to use this for
devices in UCM that need to be opened at a specific rate (like modem
devices). This can be extended to allow overrides in profile-sets as
well.
Since the hashmap stores a pointer to the key provided at pa_hashmap_put()
time, it make sense to allow the hashmap to be given ownership of the key and
have it free it at pa_hashmap_remove/free time.
To do this cleanly, we now provide the key and value free functions at hashmap
creation time with a pa_hashmap_new_full. With this, we do away with the free
function that was provided at remove/free time for freeing the value.
Make the PulseAudio tunnel behave the same way as a client
when it comes to figuring out how to connect to the current
PulseAudio daemon. This can be useful if you start a second
PulseAudio instance for e.g. network access.
Sometimes it would be nice to disable module-suspend-on-idle for
specific devices. For me the use case is to keep a HDMI sink running
all the time to avoid loss of audio when starting to play a stream to
the device (the HDMI receiver eats a bit from the beginning of the
stream when the device is opened). This is arguably a hacky solution
to the problem, but on the other hand, I think it's very sensible to
interpret negative timeout in the module-suspend-on-idle.timeout
property as disabling the suspending altogher. This is also how the
exit-idle-time configuration option behaves (negative value disables
automatic exiting).
I moved the property parsing from the timer restart function to the
function that creates the device_info objects, because if the timeout
is negative, we don't need to create the device_info object at all.
The old tunnel module duplicates functionality that is in libpulse,
due to implementing the native protocol, and the protocol code in
the old tunnel module tends to get broken every now and then, because
people forget to update the tunnel module protocol implementation
when changing the native protocol. module-tunnel-source-new avoids this
problem by using libpulse to communicate with the remote server.
With very low input sample rates the memory pool max block size may
not be big enough, in which case we should return the size of one
frame. Returning zero caused crashing.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=68616
This patch adds the ability to restore profiles if they are added after
card creation.
Adding profiles after card creation mainly happens for bluetooth cards.
Buglink: https://bugs.freedesktop.org/show_bug.cgi?id=65349
There is no function to load the authentication cookie for a context.
You can only set environment variables. This patch adds
pa_context_load_cookie_from_file().
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
If the sink rate is not updated, then the monitor source will appear
to have a different rate than the sink, but in reality there's never
any resampling done when moving data from the sink to the monitor
source, so it's a lie that the monitor source has a different rate.
The result of lying is that clients that capture from the monitor
source will have streams that run too fast or slow.
When a sink changes its sample rate, also the monitor source rate
needs to be changed. In order to determine whether a source supports
rate changing, the code checks if the update_rate() callback is set,
but monitor sources don't have that callback set, so the old code
always failed to change the monitor source rate.
This patch fixes the monitor source rate changing by handling monitor
sources as a special case in pa_source_update_rate(): if the source is
a monitor source, then the update_rate() callback is not required.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=66424
This adds asserts to check if the implementation has an update rate
function defined for the unlikely event that some implementation forgets
to assign a update rate function we can simply bail.
It is expected from the resampling implementations to have such a
function even if the state of the resampler is completely reset.
This patch fixes this assertion:
Assertion 'r->i_ss.rate >= r->o_ss.rate' failed at ../../src/pulsecore/resampler.c:1744, function peaks_init(). Aborting.
The pa_resampler struct contains many implementation specific
structures. These create overhead and don't belong there anyways.
This patch moves the implementation specific structures out of the
pa_resampler structure.
If a capture stream captures from a single sink input (so the capture
stream is a so called "direct on input" stream), then it needs to
connect to the monitor source of the sink to which the sink input is
connected. Previously the correct source was not figured out
automatically, causing the capture stream creation to fail.
The old tunnel module duplicates functionality that is in libpulse,
due to implementing the native protocol, and the protocol code in
the old tunnel module tends to get broken every now and then, because
people forget to update the tunnel module protocol implementation
when changing the native protocol. module-tunnel-sink-new avoids this
problem by using libpulse to communicate with the remote server.
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
The Bash shell completion for pacat --device combines the name of the
last sink and the name of the first source. This patch fixes that by
adding a whitespace separator in the list of devices.
Buglink: https://bugs.freedesktop.org/show_bug.cgi?id=68106
This makes it easier for users of this API to add/updated a volume
factor by doing a _remove_volume_factor() followed by an
add_volume_factor(), rather than having to either remember whether this
is the first set operation or have an API to query whether a factor has
already been set.
This is needed by the tunnel module rewrite, which runs pa_mainloop in
the IO thread instead of pa_rtpoll.
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
interactive sessions are initiated with a hello message in order to
receive a welcome message from the PA daemon and a command prompt
interactive sessions have a terminal connected to stdin
non-interactive sessions execute commands given on the command line
or received via stdin; non-interactive sessions have neither welcome
message nor command prompt
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Some HD-audio codecs (at least ALC269VB and ALC283) become quite noisy on
high Mic Boost levels. So e g, if there is a "Mic Boost" and a "Capture"
control, both ranging from 0 dB to +30 dB, you get better quality if
"Mic Boost" is 0 dB and "Capture" is +30 dB, than the other way around.
By changing the order in the configuration files, this patch makes us prefer
leaving "Mic Boost" low and "Capture" high if the user selects a medium gain.
(This is based on limited experience, and there is no guarantee that there are
no sound cards that work the other way around, and therefore this patch could
potentially regress quality on those machines. Hopefully those are fewer, so
this is what we should default to.)
BugLink: https://bugs.launchpad.net/1085402
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Currently the biggest possible sink latency is 10 seconds. The total
latency of the loopback is divided evenly for the source, an
intermediate buffer and the sink, so if I want to test 10 s sink
latency, the total needs to be three times that, i.e. 30 seconds.
Usually, you want to use one input or output at a time: e g,
you expect your speaker to mute when you plug in headphones.
Therefore, the headphones+speaker port should have lower priority
and both headphones and speaker.
A practical formula to do this is 1/x = 1/xa + 1/xb + .. + 1/xn.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
We document the default values in daemon.conf, but this was not
updated when we changed the default from speex-float-3 to speex-float-1.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The log message didn't match the code, so one of them was wrong. It's
entirely possible that the code is wrong, but I didn't have the
motivation to study the code enough to understand what the code is
supposed to do.
Dependant in British English is a person who is financially supported by
someone else. To express software dependency relations "dependent"
should be used instead, which is correct for both British and US
English.
u->sink->state is not yet updated, so the state must be read from
u->sink->thread_info.state. This makes pausing and resuming of the
smoother happen at the right time.
Thanks to Pierre Ossman for the patch.
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.
The reference ratio should always be kept up-to-date. If the reference
ratio is not updated when the input volume changes, the stale
reference ratio ends up being used as the new input volume when the
input is moved.
All pa_cvolume_snprint(), pa_volume_snprint(),
pa_sw_cvolume_snprint_dB() and pa_sw_volume_snprint_dB() calls have
been replaced with pa_cvolume_snprint_verbose() and
pa_volume_snprint_verbose() calls, making the log output more
informative and the code sometimes simpler.
The source output and sink inputs should be corked if the corresponding
sink/source is suspended, as handled during module initialization. This
also needs to be handled during stream move, because the suspend state
of the destination sink/source might be different to the previous one.
This fixes the issue with an infinite number of "Requesting rewind due
to end of underrun" traces after a stream move.
A dynamic array is a nice simple container, but the old interface
wasn't quite what I wanted it to be. I like GLib's way of providing
the free callback at the container creation time, because that way
the free callback doesn't have to be given every time something is
removed from the array.
The allocation pattern was changed too: instead of increasing the
array size always by 25 when the array gets full, the size gets
doubled now (the lowest non-zero size is still 25).
The array can't store NULL pointers anymore, and pa_dynarray_get() was
changed so that it's forbidden to try to access elements outside the
valid range.
The set of supported operations may seem a bit arbitrary. The
operation set is by no means complete at this point. I have included
only those operations that are required by the current code and some
unpublished code of mine.
It's easier to work with the port description if it can be assumed
that it's always non-NULL. I have checked that the current code base
always ensures a non-NULL description.
With the new behaviour, you will not always get a callback after a
successful write. Make sure the callers can properly handle this.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This patch removes all occurrences of double and triple
newlines.
Command used for this:
find . -type d \( -name ffmpeg \) -prune -o \
-regex '\(.*\.[hc]\|.*\.cc\)' \
-a -not -name 'adrian-aec.*' -a -not \
-name reserve.c -a -not -name 'rtkit.*' \
-exec sed -i -e '/^$/{N;s/^\n$//}' {} \;
Two passes were needed to remove triple newlines.
The excluded files are mirrored files from external sources.
This patch replaces every occurrence of ')\n{' with ') {'.
Command used for this:
find . -type d \( -name ffmpeg \) -prune -o \
-regex '\(.*\.[hc]\|.*\.cc\)' \
-a -not -name core-util.c -a -not \
-name adrian-aec.c -a -not -name g711.c \
-exec sed -i -e '/)$/{N;s/)\n{$/) {/}' {} \;
The excluded files are mirrored files from external sources.
This patch replaces every occurrence of '){' with ') {'.
The ffmpeg source tree was excluded since it will disappear anyways.
Command used for this:
find . -type d \( -name ffmpeg \) -prune -o \
-regex '\(.*\.[hc]\|.*\.cc\)' \
-exec sed -i -e 's/){/) {/' {} \;
This patch replaces every occurrence of 'if(' with 'if ('.
The ffmpeg source tree was excluded since it will disappear anyways.
Command used for this:
find . -type d \( -name ffmpeg \) -prune -o \
-regex '\(.*\.[hc]\|.*\.cc\)' \
-exec sed -i -e 's/ if(/ if (/' {} \;
This patch removes all tabs hidden inside the source tree and replaces
them with 4 spaces.
Command used for this:
find . -type d \( -name bluetooth \) -prune -o
-regex '\(.*\.[hc]\|.*\.cc\)' -a -not -name 'reserve*.[ch]'
-a -not -name 'gnt*.h' -a -not -name 'adrian*'
-exec sed -i -e 's/\t/ /g' {} \;
The excluded files are mirrored files from external sources containing
tabs.
This code is from heftig, but the mistake that I'm fixing here is my
own. Before applying heftig's patch, I downgraded the level of one of
the log messages. I managed to downgrade a different message than what
I intended, so now I'm undoing that mistake.
file_path contains the last tried file name, including the suffix, so
the error message was wrong:
Tried to open target file '/tmp/test.log.99', '/tmp/test.log.99.1',
'/tmp/test.log.99.2' ... '/tmp/test.log.99.99', but all failed.
The outputs are removed from the idxset before output_free() is
called. Trying to remove them again in output_free(), and asserting
that it should succeed caused crashing whenever outputs were freed.
This bug was introduced in commit
061878b5a4.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=65901
We need the mainloop lock to be taken around pa_mainloop_api_once() to
prevent an assert due to the defer event creation and setting of the
destroy callback not being performed atomically.
This needs us to expose a bit of implementation detail, but this seems
to be the cleanest way without an API change.
The specific problem is that pa_mainloop_api_once() needs to first
create a defer event and then set its destroy callback. If the defer
event is completed before the callback is set, an assert will be
trigerred.
Now that we don't *always* get a callback after having written
something, make sure we can continue writing as long as it fully
succeeds.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
To save some CPU (in low latency scenarios), don't re-enable the
"writable" event after it has succeeded. It is very likely the next
write will succeed right away too.
This means that we always need to handle EAGAIN/EWOULDBLOCK as a
successful write of 0 bytes, so I also verified that all callers to
pa_iochannel_write handled this correctly.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The tsched_watermark is in bytes, not in usecs. Fix this by introducing
a new variable, and also use that variable in some places for optimisation.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If there is a "Line Out" jack present, then add this path. The fallback
analog-output will be a subset of this path and removed.
I only use the "Line Out Jack" or "Line Out Front Jack" for actual jack
detection - without anything connected to the front jack, it makes little
sense to enable the port.
(Another option could perhaps be to use different paths for stereo line out
and surround line outs, but that could be a possible future improvement.)
Assume that the headphone port volume is lower than the speaker volume.
When plugging in headphones, if the path is active, while the jack is
being inserted and before it is actually detected as being plugged in,
it will still receive the signal being played (which is at a higher
volume than it will be when plugged in completely). The volume
difference manifests as a volume spike when the headphones are plugged
in, before the final volume is set.
This patch is required to prevent such a volume spike when plugging in
headphones. The problem is not fixed completely, but the spike is
shortened. To be fixed completely, we need to apply the port volume
before unmuting the new path.
In the default configuration, PulseAudio's rlimit-rttime is set to
1000000 (100%), which is higher than what RealtimeKit requires from
its clients (200000, 20%).
Make an attempt to still get realtime scheduling by clamping the
current RLIMIT_RTTIME to what RealtimeKit accepts. Warn about doing
this.
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.
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.).
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
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>
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().
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.
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.
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.
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.
Install matches for signals ObjectManager.InterfacesAdded and
ObjectManager.InterfacesRemoved, and process the devices that are
registered and unregistered dynamically.
"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.
Parse the result of ObjectManager.GetManagedObjects(), which includes
all objects registered, their interfaces and the corresponding
properties per interface.
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
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.
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>
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.
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.
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>
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.
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").
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).
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).
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.
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
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>
This fixes a later assertion failure in module-stream-restore.
Buglink: https://bugs.launchpad.net/bugs/896602
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
A stationary computer usually has headphone jack(s) and line out jacks.
In some cases analog-output.conf will be a subset of
analog-output-headphones.conf, causing line outs to be unusable (because
headphones are unplugged).
This late in the cycle, this was the safest way I could think of to try
to fix this for a particular computer. In later versions of PulseAudio
we could consider making a dedicated line out path instead, and have
proper jack detection there.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
send_counter/recv_counter relate to the bytes (play stream) passed
through the queue, hence the same sample spec must be used
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
Acked-by: Stefan Huber <shuber@sthu.org>
As far as I can see, having a mono path in a stereo mapping doesn't
make any sense. It also causes breakage: if the Master Mono mixer
element has two volume channels, the analog-output path gets removed
due to being a subset of analog-output-mono, and that in turn causes
the Master element getting muted. Users generally don't like that.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=54673
"any word starting with the letters" parts; this does not hold
any longer with commit 0e29e7365907ffbe90df768a4dea277dba40d495
core-util: Don't accept random words in pa_parse_boolean()
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
When a "Headphone Mic" jack becomes available, we do not know if
a headphone or a mic has been plugged in. Therefore, setting both
paths to "unknown" is, in theory, the correct thing to do.
However, in practice, people are more likely to plug in a headphone
rather than a mic. Therefore, allow autoswitch to the headphone port
when the jack is plugged in.
A more advanced implementation would consider what was plugged in last
time depending on what port was selected on the input side at that
time, and set availability accordingly. However, such an implementation
will have to wait (probably at least until we have our fancy routing
system implementation).
Buglink: https://bugs.launchpad.net/bugs/1169143
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
I recently came across a device without any ALSA-level mixer controls,
everything was physical knobs on the hardware.
This patch enables that device to get a port too ("Analog Input").
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The hdmi_eld_changed callback is called by alsa-lib at shutdown.
In that case, just exit instead of trying to access something with
already closed handles.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
We need to pick the right port as early as possible, before the
first volume is picked up. Hence this module needs to be loaded
before the sound card modules are loaded.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If a card is hot-plugged (which all cards will be when we load
this module before module-*-detect), make sure we don't start up
a sink with an unavailable port selected.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This way port availability has been filled in when we create the
sink, which will later enable us to pick the right port directly.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
It appears that, libltdl will find the .la file in the builddir and
figure out where the real .so is.
This also requires .ifexists to be fixed up to correspondingly search in
<dlsearchpath>/.libs.
We know we always serve up LPCM, and exposing this via D-Bus lets Rygel
set the appropriate metadata while presenting the raw (i.e.
non-transcoded) stream to clients.
same for e.g. versus e.g.\ and e.g. versus E.g.
this is ueber-nitpicking: will anybody ever notice?
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
A recent patch changed the path files from PA_BUILDDIR to PA_SRCDIR.
Do the same to the profile-set files for consistency (and to fix
out of tree builds).
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The mixer paths are not available in ${builddir} - we need to look in
${srcdir}. This should fix running an in-tree build without make install
as well as alsa-mixer-path-test in make distcheck.
Since the most straightforward way to define PA_SRCDIR was in
Makefile.am, I'm moving PA_BUILDDIR there as well for consistency.
On some machines which has a headset jack, the headset mic does not have its own
jack detection. Then we can look at the headphone jack to get some indication:
We know that if the headphone is unplugged, so is the headset mic. The opposite
is not guaranteed since the user might have plugged in a headphone, not a headset.
Also, there exist multi-function jacks which support both Headphone, Mic in headphone jack
and Headset Mic. In this case the jack name will be "Headphone Mic", not "Headphone", so
we need to include this name too.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The old assumption seemed to be that if a sink or source has the
DYNAMIC_LATENCY flag set, it can never change, so the fixed latency
will always be zero. This assumption doesn't hold with filter sinks
and sources that are moved around.
This fixes a crash with two module-virtual-sink instances on top of
each other, when the bottom one is moved from a sink without dynamic
latency to a sink with dynamic latency. What happened was that first
the bottom virtual sink "updated" (due to this bug nothing was
actually updated) its fixed latency to match the master sink (zero
fixed latency), and then the top virtual sink updated its fixed
latency to match the master sink. The master sink was the bottom
virtual sink, whose fixed latency should have been set to zero, but it
was not, so the pa_sink_set_fixed_latency_within_thread() failed in
the assertion "latency == 0".
An example: let's say that there's an alsa sink and two filter sinks
on top of each other:
alsa-sink <- filter1 <- filter2
With the old code, if filter1 gets moved to another sink, and the
new sink doesn't have the LATENCY and DYNAMIC_LATENCY flags set
(unlike alsa-sink), filter1's flags are updated fine in the moving()
callback, but filter2 is not notified at all about the flag changes.
With this patch, the flag changes are propagated to filter2 too.
Forcing the shm file to be read-only makes shm_unlink() fail on OS X.
Thanks to Albert Zeyer for reporting the bug and investigating the
root cause.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=62988
During a stream, most packets sent are either memblocks (with SHM info),
or requests for more data. These are only slightly bigger than the
header.
This patch makes it possible to write these packages in one write
instead of two: a memcpy of just a few bytes is worth saving extra
syscalls for write and poll.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
I don't know if it matters a lot, but most certainly it must be
the new channel that's supposed to be made low-delay, not the existing
listening socket, right?
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Found on Logitech B530 USB Headset / kernel 3.8. Because we don't
have different path for headset and headphone today, just add
Headset to the existing headphone path.
BugLink: https://bugs.launchpad.net/bugs/1159687
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
the README suggests to set the following dl-search-path: -p $(pwd)/src/.libs/
N: [lt-pulseaudio] daemon-conf.c: Detected that we are run from the build tree, fixing search path.
E: [lt-pulseaudio] ltdl-bind-now.c: Failed to open module /redacted/pulseaudio/src/.libs/.libs/module-device-restore.so: /home/pmeerw/src/pulseaudio/src/.libs/.libs/module-device-restore.so: cannot open shared object file: No such file or directory
the last part seems superfluous, so -p $(pwd)/src/
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
Due to a misconfiguration on my side my hdmi card didn't load with
snd-hda-codec-hdmi but through the fallback mechanism. Pulseaudio
would crash during early because hctl_handle was null, so skip
init_eld_ctls when hctl_handle is null to prevent a crash.
Thanks to David Henningsson for helping me find the underlying issue.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
the specialized code path just duplicate samples, so are only
applicable if the volume in map_table is == 1.0 (or == 0x10000);
don't use them for volumes >= 1.0
compare the integer version of the volume stored in map_table;
comparing floats is ugly (als leads to compiler warnings)
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
This consumes less power, has low (no?) perceivable difference, and
allows the default configuration to work out of the box on low-end
systems (such as netbooks).
It's valid for a path to have zero elements, e g if it contains
a single jack only. Earlier, this would cause an assertion failure
in pa_path_condense.
Also convert pa_bool_t to bool.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This reverts commit a9c3f2fb0f.
It has been recently agreed that ports should somehow have some physical
meaning, leading to the port merge in module-bluetooth-device.
With this assumption in mind, it is very unlikely that a card would
add or remove ports dynamically. Therefore, the core can be simplified
by removing the support for this.
The revert affects the code added to module-card-restore in commit
a1a0ad1af2, which can now be partially
removed.
Conflicts:
src/pulsecore/card.c
src/pulsecore/core.h
As the automake documentation says:
AM_CPPFLAGS: The contents of this variable are passed to every compilation
that invokes the C preprocessor; it is a list of arguments to the preprocessor.
For instance, -I and -D options should be listed here
AM_CFLAGS: This is the variable the Makefile.am author can use to pass in
additional C compiler flags.
http://www.gnu.org/software/automake/manual/html_node/Program-Variables.html
Let's officially support that people use maxlength to put an upper
bound on playback latency.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If minreq is not explicitly specified, it was always initialized to
20 ms (DEFAULT_PROCESS_MSEC). However when the total latency is not
much higher than 20 ms, this is way too high. Instead use
tlength/4 as a measure: this will give a decent sink_usec in all
modes (both traditional, adjust latency and early request modes).
This greatly improves PulseAudio's ability to ask for data in time
in low-latency scenarios.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Tlength should never be set higher than maxlength. While this is
corrected by memblockq later, we still need a correct tlength for
the subsequent calculations.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
On a multi-homed system, the user may wish RTP to be used only on
specific interfaces. The default binding of 0.0.0.0 for the source
address causes SAP multicast on all interfaces, which is not ideal.
Introduce a new module argument, that allows selection of the source IP,
and thus interface.
(changes in v2: s/srcip/source_ip)
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
The module argument 'source' already has special meaning as the
pa_source, however, the argument 'destination' expects an IP address.
Prior to introducing a source IP modarg for the source IP address,
rename the 'destination' argument to 'destination_ip'. Include
compatibility support for old RTP users so they don't need to change
their module usage immediately.
(changes in v2: minor formatting fixes, s/dstip/destination_ip)
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Before introducing new functionality, clarify the variable names
dest -> dst_addr
sa[46] -> dst_sa[46]
sap_sa[46] -> dst_sap_sa[46]
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
It checks all files in the mixer/paths directory and checks
- that the file can be parsed without errors
- that the file is actually shipped in the makefile
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
It's fairly uncommon, but it happens that jack detection is enabled
for some reason, e g hardware design. In that case, we cannot use
jack detection, but we can still use the hint to pick up that there
is a path.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This reverts commit 6733caf114.
Apparently, the EOF bit gets set only after there has been an attempt
to read more data than the file contains, so just reading the last
byte isn't sufficient.
fgets() returns NULL in case there's an error or f is at EOF. The
while condition just checked that f is not at EOF, therefore an error
must have happened.
u->asyncmsg is accessed from two IO threads. teardown() shouldn't
flush the queue from the main thread while both IO threads are still
potentially using the queue. This patch fixes that error by flushing
the queue from the sink input thread when the sink input is being
unlinked.
Flushing the queue in teardown() caused this assertion in
pa_asyncmsgq_get() to crash sometimes: pa_assert(!a->current)
process() may be called with a stream that doesn't have its sink/source set.
This can happen if the proplist change callback is called when the stream is
moving.
The sink input may_move_to() callbacks can be called while the source
output is not connected to any source (i.e. is currently moving too),
and vice versa.
Thanks to Frédéric Dalleau for reporting this bug.
ipacl-test fails if there is no SSH server running on your machine.
Since it is not a PulseAudio error not to have an SSH server running,
this test should not be run as part of the "make check" test suite.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Previously, a drain request was acknowledged up to two hw buffers
too late, causing unnecessary delays.
This implements a new chain of events called process_underrun
which triggers exactly when the sink input has finished playing,
so the drain can be acknowledged quicker.
It could later be improved to give better underrun reporting to
clients too.
Tested-by: Dmitri Paduchikh <dpaduchikh@gmail.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This patch adds support for completion of remote PulseAudio server
arguments it also suppresses error messages when unable to connect to
PulseAudio (only for the completion function).
The previous volume handling could cause ear damage: by default the
ladspa sink volume was 100%, and with flat volumes that would cause
the master sink volume to jump to 100% too.
The previous AAC pass-through patch (commit: 53807e4a) introduced
a new encoding format type: PA_ENCODING_MPEG2_AAC_IEC61937,
which is mostly used in pa_format_info, but forgot to increment the
protocol version number. The version needs to be incremented, because
clients need some way of checking whether the server supports the new
encoding.
If we find a microphone output port, it is probably something else
than a microphone. Therefore label it "Bluetooth output" instead of
"Microphone".
Same goes for Headphones and Speakers, but in the other direction.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
There was a recent thread on Linux Audio Users mailinglist about
whether to do so or not, and it looks like most people would prefer
having a stereo default (but even better would have been a
module-jack-card where you can easily set channels/profiles on the fly).
Reference:
http://lists.linuxaudio.org/pipermail/linux-audio-user/2013-February/091068.html
Reported-by: Kaj Ailomaa <zequence@mousike.me>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If there is a proper monitor name, we expose this as a device.product.name
property on the port. This can be useful for UIs who might want to show
this name.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The alsa mixer kcontrol has "device index" 3, 7, 8, and 9.
We need to configure this properly.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Currently, this function only reads the monitor name, but could
be extended to read e g supported formats as well.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If BlueZ crashes exactly while PulseAudio waits for the GetProperties reply, the
device has already been removed from the hashmap and therefore an assertion
failure is experienced.
The solution consists of ignoring the reply in these cases.
The problem can be observed in the following traces:
D: [pulseaudio] bluetooth-util.c: Bluetooth daemon appeared.
D: [pulseaudio] bluetooth-util.c: dbus: interface=org.bluez.Manager, path=/, member=AdapterAdded
D: [pulseaudio] bluetooth-util.c: Adapter /org/bluez/497/hci1 created
D: [pulseaudio] bluetooth-util.c: Registering /MediaEndpoint/HFPAG on adapter /org/bluez/497/hci1.
D: [pulseaudio] bluetooth-util.c: Registering /MediaEndpoint/HFPHS on adapter /org/bluez/497/hci1.
D: [pulseaudio] bluetooth-util.c: Registering /MediaEndpoint/A2DPSource on adapter /org/bluez/497/hci1.
D: [pulseaudio] bluetooth-util.c: Registering /MediaEndpoint/A2DPSink on adapter /org/bluez/497/hci1.
D: [pulseaudio] bluetooth-util.c: dbus: interface=org.bluez.Adapter, path=/org/bluez/497/hci1, member=DeviceCreated
D: [pulseaudio] bluetooth-util.c: Device /org/bluez/497/hci1/dev_90_84_0D_B2_C7_04 created
D: [pulseaudio] bluetooth-util.c: dbus: interface=org.freedesktop.DBus, path=/org/freedesktop/DBus, member=NameOwnerChanged
D: [pulseaudio] bluetooth-util.c: Bluetooth daemon disappeared.
E: [pulseaudio] bluetooth-util.c: Assertion 'p->call_data == d' failed at modules/bluetooth/bluetooth-util.c:685, function get_properties_reply(). Aborting.
Enable advanced AEC methods to use different specs (i.e., number of
channels) for rec and out stream. A typical application is beam forming
resp. multi-channel AEC, which takes multiple record channels to produce
an echo-canceled output stream.
This commit alters the EC API as follows: the EC's init() used to get
source and sink's sample spec/channel map. The new interface renamed
source to rec and sink to play and additionally passes sample spec and
channel map of the out stream. The new parameter names of init()
{rec,play,out}_{ss,map} are more intuitive and also resemble to the
parameter names known from run(). Both rec_{ss,map} and out_{ss,map} are
initialized as we knew it from source_{ss,map} before being passed to
init(). The previous EC implementations only require trivial changes,
i.e., setting rec_{ss,map} to out_{ss,map} at the end of init() in case
that out_{ss,map} is modified in init().
The card profile availability flag already provides all the necessary
information and therefore all Bluetooth ports can be merged, leaving
the two generic ones only: "bluetooth-input" and "bluetooth-output". The
availability of these port now represents whether the device is
streaming audio, with the following mapping:
- PA_AVAILABLE_UNKNOWN: some profile connected but not streaming
- PA_AVAILABLE_NO: no profiles connected
- PA_AVAILABLE_YES: some profile streaming (regardless of which)
Each port's flag represents the profiles with the corresponding I/O
capabilities (pa_direction_t).
Use the card profile availability flag instead of port availability in
order to automatically switch profiles, for example when a paired phone
starts streaming A2DP audio.
Use the transport's state to not only update the ports availability, but
also to update the card profile availability flag. The interpretation is
as follows:
- PA_AVAILABLE_UNKNOWN: BT profile is connected but no audio streaming
- PA_AVAILABLE_NO: BT profile disconnected
- PA_AVAILABLE_YES: BT profile connected and audio streaming
Some cards are capable to announce if a specific profile is available or
not, effectively predicting whether a profile switch would fail or would
likely succeed. This can for example be useful for a UI that would gray
out any unavailable profile.
In addition, this information can be useful for internal modules
implementing automatic profile-switching policies, such as
module-switch-on-port-available or module-bluetooth-policy.
In particular, this information is essential when a port is associated
to multiple card profiles and therefore the port availability flag does
not provide enough information. The port "bluetooth-output" falls into
this category, for example, since it doesn't distinguish HSP/HFP from
A2DP.
Generalize the availability flag in order to be used beyond the scope of
ports.
However, pa_port_availability_t is left unchanged to avoid modifying the
protocol and the client API. This should be replaced by pa_available_t
after a validation phase of this new generic enum type.
When the play stream from the EC sink has not enough data available then
the EC implementation is currently bypassed by directly forwarding the
record bytes to the EC source. Since EC implementations maintain their
own buffers and cause certain latencies, a bypass leads to glitches as
the out stream stream jumps forth and back in time. Furthermore, some
EC implementations may also apply noise reduction or other sound
enhancing techniques, which are therefore bypassed, too.
Fix this by passing silence bytes to the EC implementation if the play
stream runs empty. Hence, this patch keeps the EC implementation running
even if the play stream has no data available.
The echo canceller module can pass arguments to the EC implementation
via the module parameter aec_args. However, the echo-cancel-test passes
EC arguments via a separate argv[] option, which is inconsistent. Fix
this.
Unloading modules in the reverse order is the "more logical" thing
to do, and speeds up shutdown somewhat, e g by not loading
module-null-sink at shutdown.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
thread_mq.outq may contain some unprocessed messages, which should be
dispatched before unreffing the sink and source. If the sink and
source are unreffed before all messages to them have been dispatched,
the unreffing won't free the sink and source, and that in turn will
likely cause problems with things getting freed in a wrong order.
A transport should be considered connected only after the connection
procedure is complete, as expressed in audio_state_to_transport_state().
module-bluetooth-device should be loaded only after at least one
transport is not only created (during configuration), but also
connected.
This fixes the issue of premature acquire attempts sometimes experienced
when a headset is connected (issue not present in v3.0 though).
The previous patch removed module-gconf's dependency on the userdata
pointer of the free callback, and that was the only place where the
userdata pointer of pa_free2_cb_t was used, so now there's no need for
pa_free2_cb_t in pa_hashmap_free(). Using pa_free_cb_t instead allows
removing a significant amount of repetitive code.
similar to volume functions, simplifies leftover samples handling
for SIMD'd code path
use concrete pointer type (e.g. int16_t*) instead of void*,
saves several casts
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
move code to function pa_mult_s16_volume() in sample-util.h
use 64 bit integers on 64 bit platforms (it's faster)
on i5, 2.5GHz (64-bit)
Running suite(s): Mult-s16
32 bit mult: 1272300 usec (avg: 12723, min = 12533, max = 18749, stddev = 620.48).
64 bit mult: 852241 usec (avg: 8522.41, min = 8420, max = 9148, stddev = 109.388).
100%: Checks: 1, Failures: 0, Errors: 0
on Pentium D, 3.4GHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 2228504 usec (avg: 22285, min = 18775, max = 29648, stddev = 3865.59).
64 bit mult: 5546861 usec (avg: 55468.6, min = 55028, max = 64924, stddev = 978.981).
100%: Checks: 1, Failures: 0, Errors: 0
on TI DM3730, Cortex-A8, 800MHz (32-bit)
Running suite(s): Mult-s16
32 bit mult: 23708900 usec (avg: 237089, min = 191864, max = 557312, stddev = 77503.6).
64 bit mult: 22190039 usec (avg: 221900, min = 177978, max = 480469, stddev = 68520.5).
100%: Checks: 1, Failures: 0, Errors: 0
there is a test program called mult-s16-test which checks that the functions compute the
same results, and compares runtime
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
have individual function for mixing stream with different sample format instead
of huge case block in pa_mix()
shorter functions, prepare for optimized code path
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
idea is to allow optimized code path (similar to volume code)
and rework/specialize mixing cases to enable runtime performance improvements
no functionality changes in this patch
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
The patch intends to reduce computational load when resampling AND remapping. The PA
resampler performs the following steps:
sample format conversion -> remapping -> resampling -> sample format conversion
In case the number of output channels is higher than the number of input channels, the
resampler has to be run more often than necessary. E.g. in case of mono to 4-channel remapping,
the resampler runs on 4 channels separately.
To ímprove this, the PA resampler pipeline is made adaptive:
if out-channels <= in-channels:
sample format conversion -> remapping -> resampling -> sample format conversion
if out-channels > in-channels:
sample format conversion -> resampling -> remapping -> sample format conversion
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
Initialize the variable to zero by using pa_xnew0() instead of
pa_xnew(). This also allows us to remove a bunch of other zero
initialization statements.
Reported-by: Peter Meerwald <p.meerwald@bct-electronic.com>
bug probably caused by alignment requirement; sizeof(a->w) is a pointer, sizeof(a->w_arr) is an array
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
computes EC block size in frames (rounded down to nearest power-of-2) based
on sample rate and milliseconds
move code from speex AEC implementation to module-echo-cancel such that
functionality can be reused by other AEC implementations
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
prevents
CC module_ladspa_sink_la-module-ladspa-sink.lo
modules/module-ladspa-sink.c:1332:5: warning: "HAVE_DBUS" is not defined
modules/module-ladspa-sink.c:1370:5: warning: "HAVE_DBUS" is not defined
in case HAVE_DBUS is not available
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
alsa/use-case.h in needed
require at least version 1.0.24 in configure.ac
prevents the following error at compile time:
CC libalsa_util_la-alsa-util.lo
In file included from modules/alsa/alsa-mixer.h:51,
from modules/alsa/alsa-util.h:36,
from modules/alsa/alsa-util.c:46:
modules/alsa/alsa-ucm.h:27:22: error: use-case.h: No such file or directory
In file included from modules/alsa/alsa-mixer.h:51,
from modules/alsa/alsa-util.h:36,
from modules/alsa/alsa-util.c:46:
modules/alsa/alsa-ucm.h:89: error: expected ‘)’ before ‘*’ token
modules/alsa/alsa-ucm.h:169: error: expected specifier-qualifier-list before ‘snd_use_case_mgr_t’
make[3]: *** [libalsa_util_la-alsa-util.lo] Error 1
Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Remixing one channel map to another is (except for special cases) done
via a linear mapping between channels, whose corresponding matrix is
computed by calc_map_table(). The k-th row in this matrix corresponds to
the coefficients of the linear combination of the input channels that
result in the k-th output channel. In order to avoid clipping of samples
we require that the sum of these coefficients is (at most) 1. This
commit ensures this.
Prior to this commit tests/remix-test.c gives 52 of 132 matrices that
violate this property. For example:
'front-left,front-right,front-center,lfe' -> 'front-left,front-right'
prior this commit after this commit
I00 I01 I02 I03 I00 I01 I02 I03
+------------------------ +------------------------
O00 | 0.750 0.000 0.375 0.375 O00 | 0.533 0.000 0.267 0.200
O01 | 0.000 0.750 0.375 0.375 O01 | 0.000 0.533 0.267 0.200
Building the matrix is done in several steps. However, only insufficient
measures are taken in order to preserve a row-sum of 1.0 (or leaves it
at 0.0) after each step. The current patch adds a post-processing step
in order check for each row whether the sum exceeds 1.0 and, if
necessary, normalizes this row. This allows for further simplifactions:
- The insufficient normalizations after some steps are removed. Gains
are adapted to (partially) resemble the old matrices.
- Handling unconnected input channls becomes a lot simpler.
- Separate the cases with PA_RESAMPLER_NO_REMAP or PA_RESAMPLER_NO_REMIX
set and remove redundant if-conditions.
- Fix C90 compiler warning due to mixing code and variable declaration.
- Do not repeatedly count number of left, right and center channels in
the input channel map.
The logic of calc_map_table() remains unaltered.
Capability dropping when changing the user in the system
mode was previously implemented by calling
prctl(PR_SET_KEEPCAPS, 0), but that doesn't necessarily
work. It's possible that the KEEPCAPS flag is locked to 1,
in which case the prctl() call fails with EPERM (this
happens at least on Harmattan). This patch implements
explicit capability dropping after changing the user.
Earlier, -1 was returned if the memchunk size was not a multiple of the frame
size. Now, it is verified unconditionally through an assertion. Error code -1
is still returned when the memblock queue is full.
In those few cases where the return value of pa_memblockq_push() is checked,
an overflow is assumed to be the reason in case an error code is returned.
If 'PlaybackChannels' and 'CaptureChannels' are absent in the UCM
file for a device, assume the device is stereo duplex.
Reported-by: Luke Yelavich <luke.yelavich@canonical.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Now you can actually see *which* sink/source that sends a specific
message to the log, which is quite useful if you have more than
one sound card.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
pa_silence_memory() pulls sample-util as a dependency, so it had to
be moved from libpulsecore to libpulsecommon. sample-util in turn
pulls some more stuff.
use (1<<15) instead of 0x7fff as a factor when converting from s16 to float32
use (1<<31) instead of 0x7fffffff as a factor when converting from s32 to float32
the change is motivated by the following desireable properties:
* s16_from_f32(f32_from_s16(x)) == x for all possible s16 values
* x / (1.0f << 15) == x * (1.0f / (1 << 15)) for all x in s16
above changes enable easier optimization while guaranteeing bit-exact results
further, other audio sample conversion code (libavresample) does it the same way
v3 (comments Tanu):
* fix saturation in pa_sconv_s16le_from_f32ne_neon(), use vqrshrn
v2 (comments Tanu):
* fix comments in ARM NEON code
* use llrintf() in pa_sconv_s32le_from_float32ne()
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
Cc: Tanu Kaskinen <tanuk@iki.fi>
I went to implement the possibility to use the default sink/source
but found that it was already working. So I figured I'd update
the help text instead.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Problem: s16 to s32 conversion is performed as s16->float->s32 (via work
format float) for resamplers TRIVIAL, COPY, PEAKS.
Precision and efficiency suffers: e.g. 0x9fff results in 0x9ffe4001 (instead
of 0x9fff0000) and there are two sample format conversions instead of one
conversion.
Solution: If input or output format is s16, then choose the work format
to be s16 as well.
If remapping is to be performed, we could stick to work format float32ne for
precision reseans. This is debateable.
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
v2 (comments by Paul Menzel):
* generate test samples from -1..1, -0x8000..0x7fff
* check all output samples (not just half of them)
the idea is to compare the output of the C (reference) implementation
against the output of the optimized code path; currently, there are MMX
and SSE implementation for the mono-to-stereo remapper for s16 and float
sample formats
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
Cc: Paul Menzel <paulepanter@users.sourceforge.net>
SSE sconv was not tested before, only SSE2 was (on CPUs supporting both
instruction sets)
now both code path are tested on CPUs supporting both
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
nsamples should be forced to be a multiple of channels; do so correctly
and don't make nsamples larger than it actually is
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
This adds two new commands to pactl:
set-default-sink
set-default-source
This command has been part of the native protocol for a long time,
no reason not to expose it in pactl.
The use of the pseudo-blocking D-Bus calls leads to the problem that
NameLost signals are received after the reply to ReleaseName().
The problem with this is that a later acquisition of the same audio
device can potentially receive the NameLost signal corresponding to
the previous instance, due to the fact that the signal hasn't been
popped from the D-Bus message queue.
The simplest approach to solve this problem is to poll the actual name
owner from the D-Bus daemon, in order to make sure that we did really
lose the name.
The proposal uses a blocking call to GetNameOwner to avoid incosistent
states in the internal APIs: it would otherwise be possible to have a
"busy" device before the reservation has been lost, in the unlikely
case if some other process acquires the name before we got the
confirmation that the NameLost was actually true.
The function is interesting for both rd_device and rd_monitor so make
it part of the rd_device public API to avoid duplicated code.
The decision to move the function to reserve.c is motivated by the fact
that other projects (i.e. jack) use reserve.c only. Therefore, adding a
reserve->reserve-monitor dependency should be avoided.
This patch adds bash completion for pulseaudio and all of the utilities.
Channel maps and properties are not yet completed.
This should make mostly pactl/pacmd more useful for bash users.
Thanks to Denis Kasak for the awk magic (fetching ports and profiles
from the card info).
While reading from the SCO socket, there is no guarantee regarding the
resulting packet size. In some rare cases, it might not even match the
alignment expected in pa_source_post(), resulting in an assertion
failure inside pa_volume_memchunk():
I: [alsa-sink] module-loopback.c: Could not peek into queue
I: [alsa-sink] module-loopback.c: Could not peek into queue
I: [alsa-sink] module-loopback.c: Could not peek into queue
E: [bluetooth] sample-util.c: Assertion 'pa_frame_aligned(c->length, spec)' failed at pulsecore/sample-util.c:725, function pa_volume_memchunk(). Aborting.
Program received signal SIGABRT, Aborted.
[Switching to Thread 0x7fffda98f700 (LWP 8058)]
0x00007ffff6177935 in raise () from /lib64/libc.so.6
Missing separate debuginfos, use: debuginfo-install alsa-lib-1.0.26-1.fc17.x86_64 dbus-libs-1.4.10-7.fc17.x86_64 flac-1.2.1-9.fc17.x86_64 glibc-2.15-58.fc17.x86_64 gsm-1.0.13-6.fc17.x86_64 json-c-0.10-2.fc17.x86_64 libICE-1.0.8-1.fc17.x86_64 libSM-1.2.1-1.fc17.x86_64 libX11-1.5.0-2.fc17.x86_64 libXau-1.0.6-3.fc17.x86_64 libXext-1.3.1-1.fc17.x86_64 libXi-1.6.1-1.fc17.x86_64 libXtst-1.2.0-3.fc17.x86_64 libogg-1.3.0-1.fc17.x86_64 libsndfile-1.0.25-2.fc17.x86_64 libtool-ltdl-2.4.2-3.1.fc17.x86_64 libudev-182-3.fc17.x86_64 libuuid-2.21.2-3.fc17.x86_64 libvorbis-1.3.3-1.fc17.x86_64 libxcb-1.9-1.fc17.x86_64 speex-1.2-0.14.rc1.fc17.x86_64
This patch adds zsh completion for pulseaudio and all of the utilities.
Channel maps and properties are not yet completed.
This should make mostly pactl/pacmd more usefull for zsh users.
This patch adds the ability to toggle mute for sink/sources and
sink-inputs and source outputs.
All mute commands now accept 1|0|toggle as an argument.
Make sure the reply to SetConfiguration() is sent before the internal
hook is fired. This is important because the hook could have side
effects including D-Bus interfactions (i.e. transport Acquire() being
called during module startup).
The assertion in hsp_process_render() assumes that, if a memory block is
already set by the time the function is reached, its size matches
write_block_size.
This can however fail if a transport has been released and acquired
back, in the unlikely case where the MTU has changed in the meantime,
assuming the memory block wasn't released.
The u->channels <= 0 check failed if the channels argument was not
given at all, making the whole module loading fail. I don't think the
check is necessary at all - negative values are not possible, and if
someone gives 0 as the argument, it's probably ok if we act as if
there was no channels argument at all.
Calling change_cb() whenever anything happens in the ownership of the
bus name caused trouble in PulseAudio in this scenario:
1. PulseAudio is using a device and owns the corresponding service
name.
2. Another application requests device release.
3. PulseAudio releases the device.
4. Change in the bus name ownership: PulseAudio gives up the
ownership, and nobody owns the name.
5. reserve-monitor notices that, and notifies PulseAudio.
6. Since reserve-monitor reports the device as "not busy", PulseAudio
decides to reserve the bus name immediately back to itself and
opens the device again.
The other application will forcibly take the bus name to itself, as
it should according to the protocol, but the other application may
have trouble opening the device if it tries to do that before
PulseAudio has had time to react to the NameLost signal.
This can be solved by not calling change_cb() if there are no changes
in the device busy status. In this scenario the device is considered
"not busy" while PulseAudio is owning the bus name, so PulseAudio gets
no notification when the ownership changes from PulseAudio to nobody.
[The original commit message didn't have any explanation why this
change is made, so I'll add that information here myself.
--Tanu Kaskinen]
This change is from the developers of a Haskell binding[1]. According
to them, this change isn't strictly necessary, but their code gets
significantly cleaner if they can register an operation callback that
is called when the operation is cancelled due to the context getting
disconnected.
[1] https://github.com/favonia/pulse
As pointed out by Tanu, checking both error conditions is redundant and
raises the question whether it's possible that one of the conditions is
true while the other is false.
Therefore, simplify the condition by just checking one part of the
disjunction.
The function was used to check whether the basic properties of the
Bluetooth device have been received. This can be simplified by just
checking d->device_info_valid, since the state of the audio interface
is only relevant inside pa_bluetooth_device_any_audio_connected(), which
is used to trigger the discovery callback.
While checking device_info_valid, special care must be taken with all
three possible values: when set to -1, it means some error was triggered
while getting the device properties. Therefore, these devices can also
be ignored outside bluetooth-util.
Besides that, the patch slightly modifies the behavior of the internal
API affecting pa_bluetooth_discovery_get_by_address() and
pa_bluetooth_discovery_get_by_path(), since they will return the device
no matter the state of the audio interface. This however makes sense and
should have no influence in the current codebase given that the modules
make use of devices only after the discovery hook has been triggered.
The function is used to make sure some basic information has already
been gathered before the device is being used. At this point profile
states can be ignored, since their initial value will be
PA_BT_AUDIO_STATE_INVALID and thus effectively similar to
PA_BT_AUDIO_STATE_DISCONNECTED due to audio_state_to_transport_state().
The change should make no difference given that the behavior of
pa_bluetooth_device_any_audio_connected() doesn't change: by the time
TRUE is returned, a transport needs to exist. This means a profile
will exist in CONNECTING or CONNECTED state and thus the old
implementation of device_audio_is_ready() would also have returned TRUE.
Trivially fix some style issues affecting line wrap (128 chars max with
the exception of multi-line comments, which are limited to 80),
indentation and unnecessary parentheses.
Allow configuration of number of channels when using module-jackdbus-detect
to load jack-sink and jack-source. This is useful when the default channel
count doesn't match the logical channel count desired, e.g. with multi-
channel audio interfaces.
Signed-off-by: Peter Nelson <peter@fuzzle.org>
pa_bluetooth_discovery_sync() waited until all pending method calls
had completed. I don't understand what the benefit of that could be,
so I removed the function. We should avoid blocking as much as
possible, and the code that used pa_bluetooth_discovery_sync() didn't
look like it really needed to wait for anything.
The new null implementation works with arbitrary sample specs for source
and sink. In particular, it handles a different number of channels for
source and sink.
Signed-off-by: Stefan Huber <s.huber@bct-electronic.com>
Acked-by: Peter Meerwald <p.meerwald@bct-electronic.com>
In order to support different blocksizes for source and sink (e.g, for
4-to-1 beamforming/echo canceling which involves 4 record channels and 1
playback channel) the AEC API is altered:
The blocksize for source and sink may differ (due to different sample
specs) but the number of frames that are processed in one invokation of
the AEC implementation's run() function is the same for the playback and
the record stream. Consequently, the AEC implementation's init()
function initalizes 'nframes' instead of 'blocksize' and the source's
and sink's blocksizes are derived from 'nframes'. The old API also
caused code duplication in each AEC implementation's init function for
the compution of the blocksize, which is eliminated by the new API.
Signed-off-by: Stefan Huber <s.huber@bct-electronic.com>
Acked-by: Peter Meerwald <p.meerwald@bct-electronic.com>
In case that source and sink use different sample specs (e.g., different
number of channels) the computation of the latency difference fails.
To fix this, we obtain the corresponding latencies in terms of time using
the respective sample specs instead of buffer sizes.
Signed-off-by: Stefan Huber <s.huber@bct-electronic.com>
Acked-by: Peter Meerwald <p.meerwald@bct-electronic.com>
In main() of echo-cancel-test it is wrongly assumed that the EC
implementation's init() function properly initializes sink_ss. In
contrast, pa__init() sets sink_ss by default to
sink_master->sample_spec. Fix this by setting sink_ss to default
parameters and let EC implementation's init() override these settings.
Signed-off-by: Stefan Huber <s.huber@bct-electronic.com>
Acked-by: Peter Meerwald <p.meerwald@bct-electronic.com>
Argument argv[5] is accessed when argc>4, which leads to an invalid
access for argc==5. Fix this.
Signed-off-by: Stefan Huber <s.huber@bct-electronic.com>
Acked-by: Peter Meerwald <p.meerwald@bct-electronic.com>
In addition to moving the freeing a bit later, unnecessary checks for
t->device are removed. t->device is initialized to a non-NULL value
when the transport is created, and it's never changed.
it's useless to get the same SF_FORMAT_INFO three times, just compare the
name/extention in the same loop.
Signed-off-by: Wang Xingchao <xingchao.wang@intel.com>
I noticed that the doxygen API (http://freedesktop.org/software/pulseaudio/doxygen)
does not include ext-device-manager.h. The following patch adds ext-device-manager.h
and ext-device-restore.h to the list of files processed by doxygen.
It doesn't matter if the function fails (I'm not sure if
it's even possible), because the read data isn't used for
anything and the daemon will terminate in any case. The
void cast should get rid of a Coverity warning.
Removing the whole pa_read() call should be ok too, but I
guess it's nice to clean up the pipe before terminating...
Coverity warned about an ignored return value. I'm not sure
if there's something that should be done if writing fails;
at least I couldn't think of anything. Would logging an
error be acceptable here?
pa__done() calls stop_thread(), and stop_thread() already
frees the smoother. The duplicate freeing is not strictly
a bug, but static analyzers (in this case Coverity) may
complain about double-freeing, because when pa__done()
"frees" the smoother (which doesn't actually ever happen),
the pointer is not nulled. pa__done() then calls
bt_transport_release(), which will also free the smoother
if it's not NULL.
The analyzer complaint could be silenced also by nulling
the pointer in pa__done(), but since this is clearly
redundant code, I chose to remove it.
This fixes bug 38728 [1]. When equalizer features are unavailable in running
pulseaudio daemon, try to load relevant module. If this fails, following error
is printed on stderr instead of a confusing traceback:
It seems that running pulseaudio does not support equalizer features and
loading module-equalizer-sink module failed. Exiting...
[1] https://bugs.freedesktop.org/show_bug.cgi?id=38728
Signed-off-by: Matěj Laitl <matej@laitl.cz>
Make the internal function bt_transport_acquire() consistent with the
API in bluetooth-util by replacing the old 'start' parameter with
exactly the opposite: 'optional'.
Therefore, all calls to the function need to negate the second
parameter.
Note also that the name is more accurate now that setup_stream() is not
called inside bt_transport_acquire().
Do not call setup_stream() automatically inside bt_transport_acquire().
Instead, the caller is responsible to use both functions as necessary.
As a first trivial step, setup_stream() is now called manually after
all calls to bt_transport_acquire(u, TRUE), with the exception of
setup_transport() where the thread is still about to start and thus
setup_stream() will be called later on from thread_func().
All D-Bus infrastructure is now unused after bluetooth-util has covered
the pieces that were pending. Therefore, all D-Bus related code in
module-bluetooth-device can be safely removed.
The transport state also reflects the state of the audio interface. The
state redundancy can thus be minimized by always using the first one,
and avoiding the use of profile-specific states with the exception of
finding out the initial state of a transport.
The state of this interface is needed for one single reason: we need to
wait until all profiles have been connected (or more precisely, until
are connection attempts are finished). This can be made more explicit in
the code by just checking the CONNECTING state (and not loading
module-bluetooth-device during that state), but otherwise treating all
transport types equally.
Ideally, audio_state should be completely removed but it's left there to
avoid an issue with module-card-restore, as documented in the source
code's comments.
Transports can be acquired with different access rights, but in practice
"rw" was always used inside module-bluetooth-device. In addition, this
feature is removed in BlueZ 5.0 and therefore it is convenient to
abstract all this inside bluetooth-util.
Use transport state to calculate the corresponding port availability,
and while doing so use bluetooth-util to receive profile state updates
instead of directly parsing D-Bus PropertyChanged signals.
Move the function to the utility library where the enum is defined. At
same time avoid using the default clause in order to make sure the
compiler will complain if the enum type gets extended.
Similarly to the microphone gain, the speaker gain can be abstracted
inside the transport object, even though the actual D-Bus interface in
BlueZ differs.
The microphone gain represents the volume of the incoming audio stream
from the headset. This can be nicely abstracted inside the transport
object in bluetooth-util, so the modules don't have to take care about
the D-Bus details.
memcpy() of the null implementation's run() copied data for only one
channel. Set the number of channels to 1 in init() in order to guarantee
this.
Signed-off-by: Stefan Huber <s.huber@bct-electronic.com>
Acked-by: Peter Meerwald <p.meerwald@bct-electronic.com>
apply_diff_time() fails when dropping bytes from the playback stream
and the sample spec of sink and source differ as source's sample spec is
used. Fix this by using sink's sample spec.
Signed-off-by: Stefan Huber <s.huber@bct-electronic.com>
Acked-by: Peter Meerwald <p.meerwald@bct-electronic.com>
This module works pretty similar to the module-role-cork.
It should be used as an alternative to that module. Basically
it decreases the volume of the streams specified in ducking_roles
in the presence of at least one stream specified in trigger_roles.
Also, it's possible to choice the volume that will be used in the
ducking streams and if it should operates in all devices or not.
For basic reference: http://en.wikipedia.org/wiki/Ducking
Move the connection of sink/source-related hooks to module
initialization and shutdown, to group all of them together. There is
no need to connect them every time the card profile is changed.
The hook is now deprecated so avoid using it and instead use the
recently introduced PA_BLUETOOTH_HOOK_TRANSPORT_STATE_CHANGED which also
reports the disconnection event.
Add the transport-handling hooks to the centralized list of hooks in
pa_bluetooth_hook_t. These are intended to replace the now deprecated
transport-specific hook list in pa_bluetooth_transport_hook_t.
Transport objects have an associated state even though it's not
explicitly exposed in BlueZ's D-Bus API (prior to 5.0). Instead, the
state is implicitly represented in the profile-specific D-Bus interface
(i.e. org.bluez.Headset, org.bluez.AudioSink, etc.) but it can be
convenient that bluetooth-util would abstract this separation.
The old implementation is limited to parsing the profile state, but
the D-Bus API actually exposes many more properties that are currently
not being considered, specially within org.bluez.Headset.
Centralize the Bluetooth hooks in one single place, starting with
the device hooks, while removing the duplicated ones (in this case
PA_BLUETOOTH_DEVICE_HOOK_REMOVED).
The hook PA_BLUETOOTH_HOOK_DEVICE_CONNECTION_CHANGED gets fired also
when a device is being removed, so there is actually no need to have
this duplicated hook.
Devices will have zero or one transports per profile, and besides the
typical lookup is also profile-based. Therefore, replace the old hashmap
(which used the transport path as key) with a simple array which holds
a transport pointer per profile.
Path-based transport lookups are required in a discovery basis, before
the associated device is known. Therefore, it makes more sense to
maintain a hashmap in the discovery structure itself, instead of
iterating all devices.
The code can be simplified since it's just trying to round the result of
the division. Note that the resulting behavior is slightly different,
specially when the volume is 0. In this case, it will remain at 0,
instead of being set to 1.
Transports always have an associated device, so add the pointer as a
member to the structure, and remove the discovery pointer since it
already exists in the device object.
d->hfgw_state is just another profile that should be considered exactly
as the rest inside device_audio_is_ready(), which is being used to
decide if the discovery hook gets triggered.
Therefore, there seems to be no reason to make an exception for this
profile and skip checking if the condition d->audio_state !=
PA_BT_AUDIO_STATE_INVALID holds true.
This change makes no practical difference but delaying the load of the
module also for HFGW until Audio.State is received. The benefit is
that the behavior and the code are more consistent across profiles.
I was looking at a log that showed that a suspend happened (at
a strange time), but the log didn't tell me why the suspend was done.
This patch tries to make sure that that won't happen again.
With BlueZ 5 it is possible to have profile registered by a third party
process which does not share the same bus id as bluetoothd so it is
necessary to store the sender of the transport to be able to talk to it.
Note that this is backward compatible.
In some cases (typically during pairing) UUIDs might be reported by
BlueZ incrementally, that is, as soon as they have been discovered. At
this point module-bluetooth-device might already be loaded, so the late
UUID announcements need to be handled and additional card profiles
might need to be created accordingly.
A left over "required-any" made this path useless for most people.
While we're at it, also add "Front Headphone" like for the normal
speaker path.
Tested-by: Colin Guthrie <gmane@colin.guthr.ie>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If we expose this information, UIs can use this to make better
decisions about what icon to display.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Added Dell Inspiron 3420, 3520 and Vostro 2420, 2520.
Note that this is only necessary for kernels 3.3 to 3.5, as 3.6
has phantom jack support.
BugLink: https://bugs.launchpad.net/bugs/1076840
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
We inadvertantly stopped supporting non-standard rates when the
passthrough work was done. This makes sure that if no standard rates are
supported, we try to fallback to whatever ALSA gives us.
The sink can be resumed while the source is still in PA_SOURCE_INIT.
This is the case if a module such as module-stream-restore routes the
audio to the sink during pa_sink_put(), leading to an inconsistent
state: the sink stays RUNNING but the transport is not actually
acquired.
When a headset is having a profile switch, we can either leave the
SCO state unmodified (as it was before this patch) or we can
alternatively request it (as older versions of PA).
This patch tries to avoid a potential regression in case a module
such as module-suspend-on-idle is not present, due to the provided
resume-on-running policy. Without this patch, and without such a policy,
the sink and sources would stay suspended until the user manually
performed another profile switch (i.e. hsp->off->hsp).
There are several other solutions currently being discussed as a longer
term solution, some of which require extendind the core. This patch is
therefore proposed as a short-term workaround to avoid the regression.
bt_transport_acquire() might get called from the main thread, in case
the IO thread hasn't been started yet. In this case, we should not call
setup_stream() since this is going to be called in the beginning of
thread_func().
If the transport is already acquired and the stream needs to be started,
call setup_stream() directly instead of bt_transport_acquire(u, TRUE).
Both calls are identical in these conditions, with the exception of the
log trace which has now been moved to setup_stream().
Given that headsets have just one single port exposing whether the
audio is streaming (playing) or not, it's not possible that
module-bluetooth-policy would distinguish A2DP/HSP cases, and thus
the automatic selection of the card profile is not deterministic.
For this reason, disable the policy entirely for headsets and focus
only on HFGW and A2DP source profiles.
Merge the former "hsp-output" and "a2dp-output" ports into one single
port, in order to fix the regression of having several independent
entries in the UI.
Since commit e32a408b3c, we silence the
input memblock in order to give the resampler enough input samples, if
necessary.
But if there is no need to resample the hrir, the resampled memblock is
actually the same as the input memblock. Thus, we have to make sure that
we do not silence it in this case.
Without this patch, device modules will be left around after the
device has been disconnected and when they are reconnected, the
discovery module will load duplicate device module instances.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=57239
This is a minor optimization too, but the main benefit is that it's
makes the code easier to understand (I hope), since run_callback()
won't be called at times when it's not needed.
The new helper function makes it easier to check whether any audio
profiles are connected. That information is needed by the discovery
module for deciding whether a new device module should be loaded. The
device module should use this information too to unload itself at the
right time, but that's currently not implemented.
Use a more accurate name for the function since it doesn't just check
if it is an audio device (which can be detected quite early), but it
also checks if the most relevant properties (device info, etc.) have
been received.
Besides, add the const qualifier to the pointer since it's not going to
be modified.
The Device.Connected was only used for tracking whether a device module
should be loaded, but that information is already included in the
individual profile state properties. The property can therefore be
completely ignored without any loss in functionality.
Stream-restore DBus API method argument list is missing last boolean
argument apply_immediately, causing assert to fail in AddEntry handling.
Signed-off-by: Juho Hämäläinen <jusa@hilvi.org>
This makes sure we don't try to plug in a passthrough stream if the
final sink/source sample spec doesn't match what we want. In the future,
we might want to change rate updates to try a full sample spec update
for passthrough streams.
https://bugs.freedesktop.org/show_bug.cgi?id=50951
This drops ChangeLog generation from git. It does not make sense to
distribute 500 kB of ChangeLog given how easy it is these days to browse
history with git.
We might replace this with a release-annoucement-esque high-level
summary of changes.
Since some devices can be chatty with regards to how often they return
from poll(), this adds a PA_UNLIKELY() to all the the rewind_requested
checks in our sink modules to make the general case (no rewind was
requested) the fast path.
When a rewind is requested on a sink input, the request parameters are
stored in the pa_sink_input struct. The parameters are reset during
rewind processing, and if the sink decides to ignore the rewind
request due to being suspended, stale parameters are left in
pa_sink_input. It's particularly problematic if the rewrite_bytes
parameter is left at -1, because that will prevent all future rewind
processing on that sink input. So, in order to avoid stale parameters,
every rewind request needs to be processed, even if the sink is
suspended.
Reported-by: Uoti Urpala
...over "Digital Input Source:Analog Input". It makes life a little
easier for users of Dell xps m1330.
Just an old Ubuntu delta I never upstreamed until now.
The patch was originally written by Daniel T Chen <crimsun@ubuntu.com>.
BugLink: https://bugs.launchpad.net/bugs/453966
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If the mainloop is just about to enter polling, but m->state
is not POLLING yet when some other thread calls
pa_mainloop_wakeup(), the mainloop will not be woken up.
It's safe to write to the wakeup pipe at any time, so let's
just remove the check.
Based on feedback in the bug below (comments 128, 129, 131).
BugLink: https://bugs.launchpad.net/bugs/946232
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Previously, if there was a hole in a recording stream,
pa_stream_peek() would crash. Holes could be handled silently inside
pa_stream_peek() by generating silence (wouldn't work for compressed
streams, though) or by skipping any holes. However, I think it's
better to let the caller decide how the holes should be handled, so
in case of holes, pa_stream_peek() will return NULL data pointer and
the length of the hole in the nbytes argument.
This change is technically an interface break, because previously the
documentation didn't mention the possibility of holes that need
special handling. However, since holes caused crashing anyway in the
past, it's not a regression if applications keep misbehaving due to
not handing holes properly.
Some words about when holes can appear in recording streams: I think
it would be reasonable behavior if overruns due to the application
reading data too slowly would cause holes. Currently that's not the
case - overruns will just cause audio to be skipped. But the point is
that this might change some day. I'm not sure how holes can occur
with the current code, but as the linked bug shows, they can happen.
It's most likely due to recording from a monitor source where the
thing being monitored has holes in its playback stream.
BugLink: http://bugs.launchpad.net/bugs/1058200
Refactor code to fetch avail, delay and timestamp values
in a single call to snd_pcm_status().
The information reported is exactly the same as before,
however it is extracted in a more atomic manner to
improve timer-based scheduling.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
This reverts commit 1569601864.
Rethinking this, it makes more sense to not add this to the check
framework. This is mostly useful for exposing ALSA driver issues, and
it's handy to be able to build this as a standalone executable.
To reproduce, add resampler-method = ffmpeg in daemon.conf
then start PA, and load module-loopback
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0xb2f1db40 (LWP 23047)]
0x00000000 in ?? ()
(gdb) bt
0 0x00000000 in ?? ()
1 0xb7c463cb in pa_resampler_set_input_rate (r=0x80e9438, rate=44011) at pulsecore/resampler.c:365
2 0xb7c6321d in pa_sink_input_process_msg (o=0x80e87a0, code=3, userdata=0xabeb, offset=0, chunk=0x0)
at pulsecore/sink-input.c:1833
3 0xb7e9840b in sink_input_process_msg_cb (obj=0x80e87a0, code=3, data=0xabeb, offset=0, chunk=0x0)
at modules/module-loopback.c:538
4 0xb7c2709b in pa_asyncmsgq_dispatch (object=0x80e87a0, code=3, userdata=0xabeb, offset=0, memchunk=0xb2f1d17c)
at pulsecore/asyncmsgq.c:322
5 0xb7c4c6e3 in asyncmsgq_read_work (i=0x80dd580) at pulsecore/rtpoll.c:564
6 0xb7c4b34a in pa_rtpoll_run (p=0x80fb7e0, wait_op=true) at pulsecore/rtpoll.c:238
7 0xb7dd90af in thread_func (userdata=0x80afe88) at modules/alsa/alsa-sink.c:1785
8 0xb7bf3291 in internal_thread_func (userdata=0x8095d08) at pulsecore/thread-posix.c:83
9 0xb7ab9d4c in start_thread (arg=0xb2f1db40) at pthread_create.c:308
10 0xb79f3ace in clone () at ../sysdeps/unix/sysv/linux/i386/clone.S:130
As a packager, I want to be able to do "git format-patch
v3.0..origin/master" and I want the generated patches to apply cleanly
against the 3.0 tarball. The patches don't currently apply cleanly if
there are any changes to the .gitignore files, because the .gitignore
files are not shipped in the tarball. This patch fixes that problem.
A rewind may erase data that sink_input counted in playing_for or
underrun_for earlier. Add code adjusting those values after a rewind.
One visible symptom of this bug was problems recovering from an
underrun. When a client calls pa_stream_write() with a large block of
memory, the function can split that into smaller pieces before sending
it to the server. When receiving new data for a stream that had
silence queued due to underrun, the server would do a rewind to
replace the queued-but-not-played silence with the new data. Because
of the bug, this rewind itself would not change underrun_for. It's
possible for multiple rewinds to be done without filling the sink
buffer in between (which is what would eventually reset underrun_for).
In this case, the server rapidly processing the split packets would
rewind the stream for _each_ of them (as underrun_for would stay set),
erasing valid audio as a result.
As Peter Meerwald <p.meerwald@bct-electronic.com> discovered, our ARM
svolume code performance is quite terrible when the incoming samples are
not word-aligned. This can very easily be the case, since the
architecture only requires that the samples be 16-bit aligned, and we
might end up running the innermost loop after processing modulo-4
samples. The performance degradation was ~50x on a Cortex A9
(Pandaboard).
This reworks the svolume logic to first consume enough samples to make
sure the rest is word aligned, and reordering the processing to work
with 4 samples at a time first, and then finally deal with the
remainder.
With this, performance is comparable for arbitrary alignments (~3x
faster than the C code).
The callback relies on the sample spec being finalized, which is not
true with the NEW hook.
In case you're wondering about the "hook EARLY - 1, to match before
stream-restore" comment that was not changed even though the code that
the comment concerned was changed: the comment was apparently written
at a time when module-stream-restore used the NEW hook too, and later
stream-restore has been changed to use the FIXATE hook. So, the
comment was wrong/nonsensical before this patch. Since these two
modules now use the same hook again, the comment makes sense again.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=55135
Previously thread_func() used PA_SINK_IS_OPENED() to check whether
some data should be rendered. process_render_null() used a different
check: it would return immediately if the sink was not in the RUNNING
state. This caused a busy loop when the sink was in the IDLE state,
because process_render_null() didn't update the timestamp, and
thread_func() still kept the timer active using the old timestamp.
pa_rtpoll_run() would return immediately because of the old timestamp.
This is fixed by using the same check in both thread_func() and
process_render_null(). Since the checks are the same, it's actually
redundant to have the check in process_render_null(), so it is now an
assertion.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=54779
The sink has different frame size than the sink input, so
the max_request and max_rewind values of the sink input need
to be converted when setting the sink max_request and
max_rewind values.
The conversion is already done correctly in
sink_input_update_max_request_cb() and
sink_input_update_max_rewind_cb().
Once the sink input has been routed in pa_sink_input_new(),
the sample spec and channel map have already become fixed.
The sink input and source output must use the same stream
format, because the data is copied as-is.
When module-loopback is loaded without arguments, the ss and
map variables are initialized with dummy values. This caused
a problem, because also pa_memblockq_new() was called with
the dummy values, making it work incorrectly. The base was
set to 1 instead of the real frame size, which in turn
caused alignment related crashes.
This fixes at least one crash that has been observed. The
multiplication in trivial_resample() overflowed when
resampling from 96 kHz to 48 kHz, causing an assertion
error:
Assertion 'o_index * fz < pa_memblock_get_length(output->memblock)' failed at pulsecore/resampler.c:1521, function trivial_resample(). Aborting.
Without the assertion, the memcpy() after the assertion
would have overwritten some random heap memory.
When compiling without HAVE_SYMLINK the runtime dir is a real directory,
which is attempted to be created. In the case it already exists we shouldn't
error out. The HAVE_SYMLINK-enabled code already does this.
Currently, Windows versions of pacat and friends fail because the current
poll emulation is not sufficient (it only works for socket fds).
Luckily Gnulib has a much better emulation that seems to work good enough.
The implementation has been largely copied (except a few bug fix
regarding timeout handling, to be pushed upstream) and works on pipes
and files as well. The copy has been obtained through their gnulib-tool utility,
which gives a LGPLv2.1+ licensed file.
This fixes the "Assertion (!e->dead) failed" error coming and lets pacat
and friends stream happily to/from a server (I didn't actually test parec).
With some optimised sconv implementations (read NEON), rounding
inaccuracy might lead to a difference of 1 with the reference
implementation. The inaccuracy is worth the performance gain.
Also increases floating-point accuracy while printing errors to make
errors easier to analyse.
Rounding with 0.5 causes us to always round up for any value of the form
x.5. IEEE754 specifies round-to-nearest-even as the behaviour in this
case. This might not always be possible with NEON code, but this change
gets us much closer to it.
final:
* includes some minor style fixes and build-time changes to allow
building a single binary for neon and non-neon systems
v4:
* fix for sample length < 4
v3:
* convert from intrinsics to inline assembly
v2:
* load and store data with vld1/vld1q and vst1/vst1q, resp., to work
around alignment issues of compiler-generated vldmia instruction
* remove redundant check for NEON flags
Ubuntu/Linaro gcc 4.6.3
arm-linux-gnueabi-gcc -O2 -mcpu=cortex-a8 -mfloat-abi=softfp -mfpu=neon
runtime on beagle-xm:
D: [pulseaudio] sconv_neon.c: checking NEON sconv_s16le_from_float
I: [pulseaudio] sconv_neon.c: NEON: 3754 usec.
I: [pulseaudio] sconv_neon.c: ref: 58594 usec.
D: [pulseaudio] sconv_neon.c: checking NEON sconv_s16le_to_float
I: [pulseaudio] sconv_neon.c: NEON: 1831 usec.
I: [pulseaudio] sconv_neon.c: ref: 10528 usec.
I: [pulseaudio] sconv_neon.c: Initialising ARM NEON optimized conversions.
conversion may be off by one for some samples due to rounding issues
Commit dd31d652a ("utils: Adding a function to get volume from string")
uses pa_sw_volume_from_dB(), which is part of libpulse, in libpulsecore.
This breaks as-needed builds. We fix this by also building the volume
code in libpulsecommon.
UUIDs might be announced at any time, so a hook is needed to notify any
interested module. In practice, the UUIDs are quite stable with the
exception of the pairing procedure, where the UUIDs are reported by
BlueZ as soon as they are discovered.
This allows us to test the sconv code with the incoming samples at
various byte alignments. The test is also now split into correctness and
performance checks.
The IDT/Sigmatel codec driver often creates a "Mic Jack Mode" for
every mic jack, so it can change functionality between Mic and Line In.
However, as the "Mic Jack" is the standard naming, our current solution
does not make the Line In port unavailable when nothing is plugged in.
This patch makes the "Line In" port not to be created just because there
is a "Mic Jack Mode" that could be set to "Line". This makes the behaviour
consistent with e g "Dock Mic Jack Mode", "Front Mic Jack Mode" etc, where
we don't create a "Dock Line" or "Dock Mic" port either.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
modules/rtp/module-rtp-recv.c:462:8: warning: 'r' may be used uninitialized in this function [-Wuninitialized]
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
This factors out the basic measurement code for each test into a
separate block so that each test can be broken down into a basic
correctness test, and a performance comparison with minimum effort.
Am 23.10.2012 08:25, schrieb Arun Raghavan:
> On Tue, 2012-08-21 at 13:32 +0200, Thomas Martitz wrote:
>> Am 21.08.2012 08:51, schrieb Rémi Denis-Courmont:
>>> Le mardi 21 août 2012 00:50:34 Thomas Martitz, vous avez écrit :
>>>> There are tons of warnings, most of them because the function is not
>>>> recognized as printf-like.
>>> Removing checks looks very fishy.
>>>
>>> To use C99 and/or GNU format specifiers on MingW, you need to use the
>>> gnuprintf attribute instead of printf. With printf, the format string is
>>> validated according to the antiquated MSVC rules.
>>>
>> Interesting, I didn't know about gnuprintf. FWIW, what are those
>> antiquated MSVC rules? I assumed the return value which isn't int for
>> some affected functions?
> Is this one going to be respun?
>
Yes, here you go.
>From c5f15eec69bf95c9a1261e0d82abbd039156e75e Mon Sep 17 00:00:00 2001
From: Thomas Martitz <kuge@rockbox.org>
Date: Wed, 8 Aug 2012 17:38:04 +0200
Subject: [PATCH 1/3] gccmacro: Work around warnings due to printf redirection
by libintl.
Libintl defines printf as libintl_printf, which breaks the format
attribue. Unfortunately the workaround around provided by libintl
is only enabled for cygwin, but not for mingw builds. Therefore
install the workaround manually.
CXX libwebrtc_util_la-webrtc.lo
modules/echo-cancel/webrtc.cc: In function 'pa_bool_t pa_webrtc_ec_init(pa_core*, pa_echo_canceller*, pa_sample_spec*, pa_channel_map*, pa_sample_spec*, pa_channel_map*, uint32_t*, const char*)':
modules/echo-cancel/webrtc.cc:196:9: warning: 'rm' may be used uninitialized in this function [-Wuninitialized]
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
CC libpulsecore_2.98_la-svolume_arm.lo
pulsecore/svolume_arm.c: In function 'pa_volume_s16ne_arm':
pulsecore/svolume_arm.c:50:8: warning: assignment discards 'const' qualifier from pointer target type [enabled by default]
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
On Windows, strerror can actually return "Unknown Error"
(e.g. for large errnums). The code assumes the return value to be helpful.
Make it slightly more helpful by catching the message and appending the
errnum.
Calling pa_read() and pa_write() seems more appropriate since they deal better
with platform specific issues. This doesn't actually fix any open issue since
only stdio is affected but it seems more future proof.
On Windows, fdsem.c:flush() fails because sockets are set to non-blocking
mode, since pa_read() returns -1 (and errno == EWOULDBLOCK). I guess pa_read()
is expected to block in this case so make it actually block by calling poll().
Without this reading from stdin will eventually end with EOF (if there happens
to be a newline sign in the stream), because read() returns 0.
This patch fixes raw data input and piping to pacat on Windows.
Instead of repeatedly asking the discovery API to find a device given
our device path, let's hold a pointer to the device and make sure we
remove the reference when the hook is fired reporting that the device
has been removed. This makes the code easier to follow and slightly
more efficient.
The internal API in bluetooth-util should not use the const qualifier
for operations involving a device object. After all, the structure
contains many pointers and thus the const qualifier provides no real
protection.
Sometimes the kernel does not schedule us in due time, thus causing
an underrun. Adding a detection and a debug message will be a helpful
step in determining the cause of an underrun.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
CC pulseaudio-dumpmodules.o
daemon/dumpmodules.c:93:27: warning: redundant redeclaration of ‘lt__PROGRAM__LTX_preloaded_symbols’ [-Wredundant-decls]
/usr/include/ltdl.h:106:36: note: previous declaration of ‘lt__PROGRAM__LTX_preloaded_symbols’ was here
the declaration is provided by ltld.h of libtool since version 2.4, require the 2.4 instead of 2.2
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
Instead of repeatedly asking the discovery API to find a transport given
our transport path, let's hold a pointer to the transport and make sure
we remove the reference when the hook is fired reporting that the
transport has been removed. This makes the code easier to follow and
slightly more efficient.
The recently added hook can be used to detect that the transport being
used has been removed. In this case, the profile needs to be set to off.
Additionally, the change fixes a significant problem: without this
transition, the transport could be destroyed while the hook slots (i.e.
nrec_changed_slot) were still set. This led to a double free of these
objects in stop_thread().
The internal API in bluetooth-util should not use the const qualifier
for operations modifying the transport object. This is specially useful
in order to use the available hooks.
If profile could not be successfully initialized, the card should be
set to PROFILE_OFF automatically. If sinks or sources exist, they need
to be destroyed, therefore stop_thread() is called.
Remove stream moving policies from module-bluetooth-device. It is not
clear if such policies are needed at all and in case yes, they should be
implemented in module-bluetooth-policy.
These days we don't set core->default_sink/source as soon as somebody
asks for it. To retain consistent behaviour (i e the asterisk),
we need to call pa_namereg_get_default_sink/source.
Reported-by: Daniel Manrique <daniel.manrique@canonical.com>
Reported-by: Brendan Donegan <brendan.donegan@canonical.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If the card is being set to off profile, it is not necessary to check
if the device exists. This could potentially happen during shutdown,
immediately before the module is unloaded.
It might happen that a PropertyChanged signal is received but the
corresponding card profile has not been created, leading to an assertion
failure in filter_cb() due to inexistent ports. This can happen if BlueZ
misbehaves, or also if the UUIDs are reported later on (i.e. during
pairing discovery). In any case, the signal should just be ignored.
Handling the signal DisconnectRequested should be unnecessary since the
profile-specific interfaces will be later disconnected, leading to
module unload.
Additionally, the signal is problematic: if an interface (i.e.
A2DP AudioSource) is playing at the time DisconnectRequested is
signaled, the following sequence can occur:
1. AudioSource is playing
2. DisconnectRequested is received
3. Module is unloaded due to DisconnectRequested
4. AudioSource state changes from playing to connected
5. module-bluetooth-discover loads the module
6. AudioSource state changes from connected to disconnected
Therefore the module is unnecessarily loaded, to be unloaded immediately
afterwards. This can easily be reproduced if a device is unpaired while
the audio is streaming.
The simplest solution to this consists of removing step 3, by just
ignoring the DisconnectRequested signal. This reverts commit
8169a6a6c9.
Upstreamed from Debian: "Although in principle Ac '97 hardware has a
separate mono LFE pin nothing seems to use it. To make matters worse
it does confuse PulseAudio's port selection slightly which causes
audio in virtualbox not to work out of the box."
Credit: Sjoerd Simons <sjoerd@debian.org>
Credit: Martin-Éric Racine <martin-eric.racine@iki.fi>
BugLink: https://bugs.launchpad.net/bugs/1016969
BugLink: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=673847
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
I was hoping this would improve bootup speed, but it doesn't seem
to do so here, at least not much. But at least it reduces the logs
a little.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The GNOME-OSTree build system currently creates chroots and bind
mounts to the source tree outside the root; this means that we can't
necessarily run git inside the root, because in the case of git
submodules, the .git repository will point to outside root.
Also, error out fatally if we fail to determine the version; it
makes no sense to put UNKNOWN in e.g. the pkg-config files
since this will just cause errors later on.
Handle availability changes in Bluetooth ports inside
module-bluetooth-policy. The implemented behavior is similar to how
module-switch-on-port-available behaves, but the conditions are more
relaxed and thus more profile changes are triggered.
When PA is doing gateway role, let module-suspend-on-idle resume the
audio stream automatically. This will work until the user (or the remote
side, which we also consider user-initiated) suspend the stream
manually.
Card profile hfgw should be no different from the rest, and thus no
internal policy inside module-bluetooth-device should decide to switch
to its profile automatically.
This should be handled by policy modules.
Handle the Playing->Connected transition gracefully by releasing the
transport and setting the sink and sources as suspended. This is
necessary since the IO thread might not encounter a HUP always.
Until today, setting the card to some profile resulted in a transport
acquisition, leading to audio stream setup. This is generally not very
interesting and even undesireable for HFGW use-cases, where the
Gateway role (the remote end) would typically request the SCO link.
Nevertheless, there is no safe way to implement such check without race
conditions, since the BlueZ's state can change between the state report
and the call to Acquire(). The chances for this to reproduce are quite
low though, since interface state changes are relatively slow.
This race condition requires that BlueZ's API is extended in order to
perform the operation atomically, which has already been discussed and
ack-ed in the BlueZ mailing list.
Note that this patch does not introduce a new race condition, since it
already existed before (the PropertyChanged->Acquire race condition,
affecting HFGW use-cases). It is just more explicit now.
If the acquisition of the transport fails, the profile should still be
set. In this case the audio is not actually streaming, so the sink and
source will be created but left suspended.
If the transport needs to be acquired later, for example because the
user wants to route the audio the remote device, the suspend flag should
have to be changed.
Use the port availability flag to expose whether a certain profile is
connected and whether it's doing actual audio streaming.
The proposed mapping is the following:
- Profile disconnected: port is unavailable
- Profile is connected (but not streaming/playing): availability unknown
- Profile is streaming/playing: port is available
The availability-unknown is specially interesting: it involves that if
the sink/source exists (corresponding card profile set), it is currently
in suspended state.
For example, for SCO cases (HFGW or HSP), this means the SCO is down. A
policy module would typically not change this, unless someone is really
trying to use the sink/source. This situation would be nicely handled by
module-suspend-on-idle, which would automatically connect SCO.
On the other hand, if the user wants to control the status of the SCO,
it will still be possible by resuming the sink or source (suspend=0).
This works out-of-the-box since most UIs would show to the user ports
whose availability is unknown.
The configuration of the transport that depends on the MTU should be
performed every time the transport has been acquired, since the
parameters depend on what the Media API provides. This requires to
update the parameters of the sinks and sources as well.
This patch moves this code into a new function that will be called
when the stream is starting (setup_stream), from the IO thread.
This makes the code more robust, since the existing multiple calls to
bt_transport_acquire() do not rely on setup_bt() being able to acquire
the transport.
There should be one port per sink/source so a dummy set_port callback
will be enough.
Adding this callback avoid the "operation not implemented" error
message and additionally makes the module work nicely with
module-switch-on-port-available.
The transport might have disapeared exactly before acquiring, so we
should avoid an assertion failure, in this case inside the function
pa_bluetooth_discovery_get_by_path().
The HFGW source should be consistent with the sink by not setting the
"phone" intended role.
Even though setting this role seems to make sense strictly speaking, the
rest of the codebase doesn't handle this well. Therefore, the audio
coming from a Bluetooth phone can be routed back to the same device.
Make code more readable by introducing the helper function
bt_transport_is_acquired(). This also adds assertions to check whether
the internal state is consistent.
Previously, sink_input_kill_cb would cleanup u->sink an then unload the
module. However, during module unload, both save_state and dbus_done
tried to use u->sink, causing a segfault or assertion failure.
The segfault is easy to reproduce: Load module-equalizer-sink and then
press ctrl-C to terminate pulseaudio.
This commit removes the u->sink cleanup in sink_input_kill_cb, since
u->sink will be cleaned up by the module's pa__done as well (after it
has been used).
Signed-off-by: Matthijs Kooijman <matthijs@stdin.nl>
Nowadays, we are using more hashmaps and other things, than we did
before. Therefore, I often get the "flist is full (don't worry)"
message. This change should avoid that message. I was unable to find
any significance in increase of memory footprint from this change.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Interestingly, the name is properly matched even though there
is no paths/iec958-stereo-input.conf file.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This check was valid before we introduced per-source-output volumes, so
dropping it now. Thanks to Alban Browaeys <prahal@yahoo.com> for
catching this.
During initialization, the approach avoids having a needless short
period of corked state in case the sink is suspended, by always creating
the source-output corked and uncorking it immediately afterwards when
the sink is not suspended.
During initialization, the approach avoids having a needless short
period of corked state in case the source is suspended, by always
creating the sink-input corked and uncorking it immediately afterwards
when the source is not suspended.
Keeping the SBC check separate means we can keep the SBC_LIBS/CFLAGS
separate, which is cleaner. Thanks to Jan Steffens for pointing out that
this was broken (SBC_* wasn't actually changed to match the configure
change).
This reverts commit da5078e5c7.
Let device module figure out the priority based on the state of the
profiles.
Note that most likely all profiles will be in PA_BT_AUDIO_STATE_CONNECTED
state so 'Off' will be the initial profile then it is up to the policy
module to switch to the most suitable profile.
pa_sink_input_seek() calculates output lenth (slength) and
corresponding input length (ilength). During an underrun, the function
generates slength bytes of silence and adds ilength to the
underrun_for value. However, the ilength value may be shortened to
match resampler limits, and there's no corresponding adjustment to
slength. Thus, the length of the generated silence is longer than
resampler output would have been, and underrun_for should be increased
by more than the limited ilength. This error makes the user-visible
since_underrun field in struct pa_timing_info too small. Fix by using
the original value calculated before limiting in this case.
The problem that the comment mentions doesn't actually
exist, because when the sink latency is changed to a smaller
value, the sink implementor will request the required
rewind.
Add support for hfgw card profile in module-bluetooth-policy, just like
a2dp_source is handled.
In this case also the sink needs to be connected using module-loopback.
Instead of focusing on a2dp_source only, prepare the module to support
several profiles. It will be possible to enable/disable each of them
using module arguments.
Property bluetooth.protocol did make a distinction between A2DP sink and
source roles but on the contrary did not separate HFP roles (headset vs
gateway). For consistency, they should both behave similarly.
This automatically fixes another incosistency: the HFGW (or HSP) sink
was set to bluetooth.protocol="sco", while the source was set to "hsp".
There is no use for this distinction, since the protocol (including the
role) is the same.
On ARM, pa_object has less strict alignment requirements
than e.g. pa_sink and pa_source, so when pa_object is cast
to pa_sink, the compiler thinks that it's unsafe. In this
case, however, the pointer given to pa_sink_ref() was a
pa_sink pointer to begin with, so casting it first to
pa_object and then back to pa_sink is entirely safe.
This particular source of warnings is extremely annoying,
because this message is printed for any compilation unit
that includes sink.h, source.h or any other header that
defines a class, and the message tends to get printed
multiple times for one compilation unit:
In file included from ./pulsecore/source-output.h:37:0,
from ./pulsecore/source.h:49,
from ./pulsecore/sink.h:40,
from ./pulsecore/core.h:50,
from daemon/daemon-conf.h:31,
from daemon/cmdline.h:25,
from daemon/cmdline.c:38:
./pulsecore/sink-input.h: In function 'pa_sink_input_ref':
./pulsecore/sink-input.h:245:1: warning: cast increases required alignment of target type [-Wcast-align]
Profile a2dp_source, just like any other card profile, should have
state guards when the profile is being changed. If the BlueZ interface
is not connected, the profile should be set to "off".
Fixes this:
/usr/bin/ld: utf8_test-utf8-test.o: undefined reference to symbol 'pa_log_set_level'
/usr/bin/ld: note: 'pa_log_set_level' is defined in DSO /home/takaskin/dev/pulseaudio/src/.libs/libpulsecommon-2.98.so so try adding it to the linker command line
/home/takaskin/dev/pulseaudio/src/.libs/libpulsecommon-2.98.so: could not read symbols: Invalid operation
With this fix, `check-daemon` doesn't need a system-wide running pulseaudio
anymore.
The method to use is to invoke `make check-daemon` under `src/` and it just
works! :)
The return value of dbus_message_iter_next() doesn't need to be checked
since the while condition will be false anyway (arg type will be
DBUS_TYPE_INVALID).
The method ListDevices() in org.bluez.Adapter was deprecated in BlueZ
4.61, and is going to be removed in future releases. Instead, a property
was introduced for this purpose in BlueZ 4.7.
The method ListAdapters() in org.bluez.Manager was deprecated in BlueZ
4.61, and is going to be removed in future releases. Instead, a property
was introduced for this purpose in BlueZ 4.22.
The module module-bluetooth-device should never be given parameter
'profile' twice, even if both HFGW and A2DP are playing. This patch
proposed to consider HFGW first.
This simplifies the code a lot, in favour of the D-Bus Media interface
in BlueZ. The old socket-based IPC mechanism has been deprecated and is
about to be removed soon.
Without -avoid-version, libtool creates a libwebrtc-util.so.0 and
libwebrtc-util.so.0.0.0 which are not cleaned up by make uninstall,
which in turn causes make distcheck to fail.
In UCM basic functions, we only assign intended roles from modifier
to sink/source, but we don't have a chance to set the ucm modifiers.
Here we amend the functions so that when roled stream starts or
stops, we have the following results:
1. stream will be routed to sink/source specified in modifier by
module-intended-roles
2. After that, modifier will be enabled or disabled.
3. when multiple streams with matched roles of modifier start, only
the first one will enable the modifier, and when they end, the
last one will disable the modifier.
Signed-off-by: Feng Wei <wei.feng@freescale.com>
Signed-off-by: Arun Raghavan <arun.raghavan@collabora.co.uk>
This moves out the webrtc bits into a small helper library to shield the
rest of module-echo-cancel from being linked with a C++ linker. This is
required because automake will _always_ link module-echo-cancel in C++
mode if any of its deps (even conditional ones) are in C++.
Without this file, pulseaudio aborts with:
E: [pulseaudio] alsa-mixer.c: Assertion 'eb = p2->elements' failed at
modules/alsa/alsa-mixer.c:3179, function path_set_condense(). Aborting.
I know that pa_streq() is defined with an extra parenthesis, but everytime I
look at it, it makes me nervous. :D
So it's better to add the parenthesis here.
Modifiers may have their own PlaybackPCM/CapturePCM and for these, we
create separate sinks/sources. These are marked with the
device.intended_roles property to let role-based routing take care if
streams are tagged appropriately.
The proplist isn't used by the conventional alsa-mixer code path, but
can be used by UCM to transfer properties from UCM data to the
sinks/sources corresponding to a mapping. These properties could be used
later in policy, etc.
The specific use for which I'm writing this now is for UCM modifiers
that have their own PlaybackPCM/CapturePCM field. These will be
translated to a separate sink/source corresponding to the modifier by
adding an additional mapping per sink/source. These mappings' proplist
will be populated with the name of the modifier and corresponding
"device.intended_role" property. The latter will be used in the usual
routing-by-role way, and the former will be used during sink/source
activation and deactivation to know what UCM modifier is to be enabled
or disabled.
Jack in UCM is decided by UCM device name, although in fact
not all UCM devices have "jacks". Because port is also mapped
to UCM device, we can always find target port when some jack
event happens.
Signed-off-by: Feng Wei <wei.feng@freescale.com>
UCM basic functions will provide another way to handle the alsa mixer
and controls. That means alsa card module will make use of alsa ucm
configurations provided by various audio systems instead of mixer and
paths configurations provided by PA. PA profiles come from UCM verb, PA
sinks/sources and ports come from UCM devices.
In case the proper UCM configurations are found, ucm branches are
activated, or we will still fall through to the original way.
Signed-off-by: Feng Wei <wei.feng@freescale.com>
Some ASUS netbooks, such as the 1015 CX, have only one 3.5 mm jack,
but it can be used either as a headphone or as a mic (but not both
simultaneously).
This patch adds support for the "Headphone Mic" path that is used
on these devices, so that we can use the jack as an external mic, and
doing so without muting the speaker.
BugLink: https://bugs.launchpad.net/bugs/1018262
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Since commit 12af302a last month, cards always have at least one
profile, so there is also always an active_profile (this makes the code
more simple). However, module-coreaudio-device did not create a profile
yet, causing a crash of PulseAudio when used on OS X. This patch fixes
this crash, by adding a single "On" profile. I've also added a TODO for
adding an "Off" profile which removes all sinks and sources -- I can
work on resolving this TODO later on.
As these functions are called together and are related, we might merge
them and call setting_select from pa_alsa_path_select by passing
optional pa_alsa_setting argument.
Make also the setting_select static as it is not called outside of
alsa-mixer.c after this change.
[Additional note from Tanu Kaskinen: this change improves the
mute-during-activation feature, because now the mixer changes related
to selecting the setting happen while the hw is muted.]
Move pa_alsa_setting_select call just after the pa_alsa_path_select in
[sink | source]_set_port_cb functions as there is no dependency to volume
calculations that are done between these two calls. Idea here is to make
possible to merge these two functions since they are called together from
other places too.
It's assumed in a couple of places that entry_read()
initializes entry->profile to a non-NULL string. This patch
makes those assumptions hold.
Tested-by: Mikel Astiz <mikel.astiz.oss@gmail.com>
module-card-restore now saves the latency offsets.
This change includes a entry version bump.
The entry now consists of a port count and a port name and offset for
every port that belongs to the relevant card.
This is a basic module for enabling loopback as soon as a new bluetooth A2DP
source is created. The module is given a source and a media role using command
line. This allows module-intended-roles or module-device-manager to choose a
target sink for the stream.
At module-loopback load, if no sink is given, the default sink is used. If the
stream has a media.role property, the property cannot be used because a the
source or sink is forced to default. Both module-intended-roles and
module-device-manager are affected. The same apply to sources.
With this patch, if sink or source is missing, routing modules can be used.
I forgot half of the front headphone patch, i e, to hide the
speaker output when the front headphone is connected. Thanks to
Shih-Yuan Lee for noticing.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This makes updating of permissions on existing directories optional with
pa_make_secure_dir() and pa_make_secure_parent_dir(). This makes sure
that the recursive directory creation doesn't end up modifying existing
directories, and also fixes a problem where creating an auth cookie
(specifically ~/.esd_auth) would end up modifying permissions on ~.
Thanks to Frédéric Danis for reporting this.
Many desktops have headphone on the front and line outs on the back.
Sometimes this means that the headphone is labelled "Front Headphone Jack",
but the volume controls are only "Headphone Playback Volume", i e,
without the "Front" prefix.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Some devices have inverted right channel, so when you add left and right,
the result is silence, or very faint sound. In recent kernels (3.5,
perhaps also 3.4) these are starting to be marked with a special
"Inverted Internal Mic" capture switch.
While we might want to add some reverse summing mechanism in the
future, for now, we just turn the thing off to avoid the problem of
recording silence.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
For kernel 3.6, "phantom jack" kctls have been added. They serve as
a marker that a particular port exist. They were made so we can detect
that there actually are speakers and internal mic on a laptop, even if
there are no other indications (volume controls etc).
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Several laptops have speaker ports, and/or internal mic ports, but we have
no way of detecting that. So we make the port(s) always show up for these
devices.
BugLink: https://bugs.launchpad.net/bugs/946232
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This includes updating the native protocol and the client API.
A new command was added to allow setting the latency offset.
Also the card list command now shows the latency offset if there
are ports available.
Update protocol to 27.
Set the state variable immediately to zero so if we fail to open the
configuration file we don't check an uninitialized pointer and free an
nonexistent proplist.
In practice there is always at least one profile, and I
don't think there will ever be cards without profiles.
Therefore, I added assertions to pa_card_new() stating that
the card new data must always contain at least one profile.
Now a lot of code can be simplified, because it's guaranteed
that the profiles hashmap and the active_profile field are
always non-NULL.
In my opinion, pa_card_set_profile() should assert that name
is not NULL, and it would be the job of the client interface
to filter out NULLs from the client input, but this is done
this way also when setting sink and source ports, so for
consistency I'll do this this way for now.
I don't like long function parameter lists, and I plan to
add some more state data to the parser which would make the
parameter lists even longer without this refactoring.
module-tunnel doesn't care about the proplist contents, so
pa_tagstruct_get_proplist() is only used for removing the
data from the tagstruct buffer. In that case it's more
convenient to just pass NULL as the proplist argument.
This function is now marked as deprecated. It is functionally identical
to g_get_current_time(), so we use that instead. The GLib API docs
suggest g_source_get_time(), but that does not provide wallclock time
(which is what the pa_time_event API expects), so we don't use it.
The latency offset type should be signed (int64_t) so we can also add
a negative latency offset.
This also includes changing the type of the sink/source
offsets and updating pacmd so it handles negative numbers.
pacmd was extended so it can handle the new latency offset.
A new function was added so we can set the latency also the list
commands were extended to print the latency offset on the ports.
A latency offset variable was added to the sink/source struct.
Also a function was introduced to update the latency offset of the
sink/source and a new message type was introduced so we can send the latency
offset to the IO thread.
The latency offset is automatically populated with the latency from the
currently active port.
A latency offset variable was added to the port struct and a function to
set the latency offset.
The latency offset does nothing for now, but it will be later added to
the sink/source latency.
realpath() on OS X behaves GNUish and accepts NULL for resolved_name
only on 10.6 and higher. Older versions will crash, if resolved_name is
NULL.
All versions define PATH_MAX, though. Better play it safe and use the
generic PATH_MAX version of pa_realpath on Mac OS X systems.
Signed-off-by: Mihai Moldovan <ionic@ionic.de>
PropertyChanged signal of org.BlueZ.MediaTransport is processed in
pa_bluetooth_transport_parse_property() which updates t->nrec.
This is called by :
- First by filter_cb() of bluetooth-util.c
- Then by filter_cb() of module-bluetooth-device.c which retrieve value
of t->nrec before calling parse function, then it checks if t->nrec
has changed before updating bluetooth.nrec property.
As t->nrec has alreday been changed during first process, property
update is never performed.
This patch creates a new hook in pa_bluetooth_transport called
PA_BLUETOOTH_TRANSPORT_HOOK_NREC_CHANGED.
The hook is fired by bluetooth-util.c when the transport's NREC
property changes.
module-bluetooth-device.c won't listen the PropertyChanged signal of
MediaTransport anymore. Instead, it will use the hook in
pa_bluetooth_transport to get a notification when the NREC property
changes, and update the sink or source proplist accordingly.
const qualifier for returned pointer of
pa_bluetooth_discovery_get_transport() is removed.
This makes pa_make_secure_dir() create any missing parent directories in
the given path as well. This is useful, for example, on a pristine
system with a clean $HOME that needs ~/.config/pulse/ to be created when
~/.config does not exist.
Specifying the volume when creating a new stream is not an
equivalent act as setting the volume with a volume control
application. When creating a new stream, stream-restore
shouldn't save the volume, but when changing the volume,
then saving it is ok. For example, when I say
"paplay --volume=10000 somefile.wav", I mean that I want the
new stream to have volume 10000. I don't mean that also
future paplay invocations (without the --volume option)
should have that same volume.
This patch effectively reverts
546bcf3f2f.
The bluetooth device should have ports so we can attach a latency to
the ports.
Every profile (a2dp, hsp...) has his own set of ports depending on the
number of sinks and sources it provides.
I doesn't make sense to require all callers of
pa_device_port_new() to create the hashmap themselves. There
are and there will be no cases where a port without any
profiles would be desired.
pacmd should allow unloading modules by name.
The command_unload() function was expanded to handle names while
unloading modules.
If there are multiple modules with the same name all
of them will be unloaded.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=48289
Compilation with -DDEBUG_TIMING fails due to a missing header:
modules/alsa/alsa-source.c: In function 'check_left_to_record':
modules/alsa/alsa-source.c:426:9: warning: implicit declaration of function 'raise' [-Wimplicit-function-declaration]
modules/alsa/alsa-source.c:426:9: error: 'SIGTRAP' undeclared (first use in this function)
Signed-off-by: Eero Nurkkala <eero.nurkkala@offcode.fi>
Compilation with -DDEBUG_TIMING fails due to a missing header:
modules/alsa/alsa-sink.c: In function 'check_left_to_play':
modules/alsa/alsa-sink.c:453:9: warning: implicit declaration of function 'raise' [-Wimplicit-function-declaration]
modules/alsa/alsa-sink.c:453:9: error: 'SIGTRAP' undeclared (first use in this function)
Signed-off-by: Eero Nurkkala <eero.nurkkala@offcode.fi>
[These symbols were removed in libudev.so.1.0.0. Replace them with
hardcoded strings. -- heftig]
Signed-off-by: Marc-Antoine Perennou <Marc-Antoine@Perennou.com>
When a Bluetooth headset is connected only to HFP profile (not connected
to A2DP) and host streams to it, a crash occurs if host disconnects.
When HFP disconnects, audio thread will fail on POLLHUP then generate
a message to set PA profile to Off before ending.
If this message is managed before PA unload bluetooth device module,
all works fine.
But, if this message is managed during module unload, this finish by
re-entrance in release code (stop_thread) and a crash.
This fix prevents to process profile change when module is unloading.
If ~/.pulse/ already exists we stick to it in order not to lose
configuration and so that pulse configuration may still be shared across
the network with different PA versions.
XDG_SESSION_COOKIE was supposed to be used for access control to the
session and should not be exposed to other users.
Replace usage of XDG_SESSION_COOKIE by XDG_SESSION_ID which is the right
choice here since it is actually a proper session ID that may be
published.
pa_bool_t and dbus_bool_t cannot be used interchangably since their size
might (and do) vary. This caused a crash on some systems which was
reported and root caused by: Aidan Thornton <makosoft@googlemail.com>.
Ref: https://bugs.gentoo.org/show_bug.cgi?id=398097
Changes in v2:
- Call the mapping a generic 4-channel input mapping
instead of a 4-channel mic array mapping. The mapping
will be used also by sound cards that have two stereo
input jacks, so in those cases talking about mic arrays
is wrong.
- Added a comment about using the "hw" device name.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=45813
libsamplerate_resample() assumed that src_process() would
always consume the whole input buffer. That was an invalid
assumption leading to crashes.
This patch adds a leftover memchunk for storing any
non-consumed input. When pa_resampler_run() is called next
time, the leftover is prepended to the new input.
Changes in v3:
- Make the calculations in pa_resampler_result() and
pa_resampler_max_block_size() more readable and more
correct.
- Rework the leftover storing: instead of using a dedicated
buffer for it, store it in the beginning of remap_buf.
This can avoid some memory copying. (The idea was
suggested by Wang Xingchao.)
- Use a generic save_leftover() function instead of doing
the leftover copying in the resampler implementation.
- Use the leftover logic also with the speex and ffmpeg
resamplers.
[ed: dropped the speex bit since the API guarantees that
it will consume everything -- Arun]
Changes in v2:
- If add_leftover() is called with zero-length input while
the leftover length is non-zero, we don't try to acquire
the input memblock.
- Instead of taking a reference to the original input in
libsamplerate_resample(), we copy the leftover data to a
new memblock. This is done, because otherwise, if the
input is one of the internal buffers, the data can get
overwritten before reading it in add_leftover().
- Store add_leftover_buf size in bytes instead of samples
(more convenient, but less consistent with other code).
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=47156
All utilities should have the --help and --version command line options.
These two were added to pacmd, also the goto label was changed from fail
to quit like in the other utilities.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=45030
Due to the formatting of `README` the only occurence of PulseAudio in a
text is capitalized. People reading `README` should at least see the
correct spelling once.
In contrast to reformat just add a spelling section as an easy solution.
Since it is not that important the last place in the ordering might
suffice.
State which thread calls set_port in which scenario.
Reported-by: Jyri Sasha <oku@iki.fi>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Log in as user A, fast user switch to user B, let user B change
port, volume or mute status, then switch back to user A.
At this point we must make sure that the ALSA and PA volumes are
synchronised by writing to the ALSA mixer when the ALSA device
becomes available.
BugLink: https://bugs.launchpad.net/bugs/915035
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If deferred volumes were activated, set_volume does not really set
the volume, and is probably only meant to be called from the main
thread.
As we're currently really setting the port and the mute here (i e
modifying ALSA), we should really modify the volume as well.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Make sure we can't be called into by remaining references to
sink-inputs and source-outputs after we have unloaded, as
that will likely lead to segfaults.
Thanks to Tanu for providing valuable input on this patch.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Coverity thinks that expected_method_sig can be NULL when
it's dereferenced by pa_streq(). Adding assertions doesn't
hurt here (in my opinion), and that should get rid of the
warnings.
Coverity thinks that device_name can be NULL when it's
dereferenced by strcmp. Adding an assertion doesn't hurt
here (in my opinion), and that should get rid of the
warning.
Coverity thinks that sample can be NULL when it's
dereferenced after this line. Adding an assertion doesn't
hurt here (in my opinion), and that should get rid of the
warning.
Valid channel id range is from 0 to SND_MIXER_SCHN_LAST,
inclusive, so the size of the masks array in pa_alsa_element
has to be SND_MIXER_SCHN_LAST + 1. Similar "too small"
arrays were also in alsa-sink's and alsa-source's userdata,
but actually those arrays were not used at all so they were
removed.
element_is_subset() in alsa-mixer.c skipped the last channel
id when iterating the element masks array; that's now fixed
as well.
Thanks to David Henningsson for spotting the too small
arrays in alsa-sink and alsa-source and the
element_is_subset() problem.
Add also an assertion for the sample spec validity. The
existing code already does crash in case of an invalid
sample spec, but the error would not be as obvious: the
crash would happen due to a divide-by-zero operation in
pa_frame_aligned().
Misbehaving clients can try to set a filter sink to output to
itself, leading to crashes later on. This patch protects us from that.
Thanks to Roman Beslik for testing and finding an error in the first
version of this patch.
Tested-by: Roman Beslik <rabeslik@gmail.com>
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=44397
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Changes since v1:
Use max value of jack_port_get_latency_range to calculate the latency
and squash compiler warnings cased by using jack_port_get_total_latency
Modifying latency only works inside a callback, and for hardware the
latency is generally fixed on jack, so just take the max value.
Signed-off-by: Maarten Lankhorst <m.b.lankhorst@gmail.com>
This causes problems with 24kHz audio (results in echoing)
when upscaling to 44.1kHz or 48kHz.
It can be reapplied when the optimisation works for all cases.
This reverts commit 8539fe9765.
When the runtime path gets long (which can happen on some NFS
mounts where $HOME is not just /home/$USER), it can grow
longer the 108 char limit imposed by sockaddr_un.sun_path.
This just calls realpath which should ultimately point into
/tmp in most cases and result in a much smaller path.
Only do this when we are adding on a name component to the
runtime path so creating the actual symlink will still get
the original, long name, but this shouldn't be a problem
as it never goes into the sockaddr_un.sun_path.
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=44680
This fixes pa_sample_spec init to use the correct API. Not doing so
triggers a valgrind warning as we call pa_sample_spec_valid() on this
later on, which checks the rate and channels fields. Thanks to Rémi
Denis-Courmont for reporting this.
If somebody writes the line "required-any=ignore", that shouldn't
mean a required-any element needs to be present for the path to
succeed probing.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Some older cards do not have jack detection. This patch makes the
port still show up.
An implementation detail: the "required = ignore" line has in itself
no effect, but we have to write *something* there, or else the entire
jack detection section will be ignored by the parser.
BugLink: https://bugs.launchpad.net/bugs/961286
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
modules/alsa/alsa-mixer.c:3110:21:
warning: format '%lx' expects argument of type 'long unsigned int', but argument 7 has type 'pa_channel_position_mask_t' [-Wformat]
modules/alsa/alsa-mixer.c:3110:21:
warning: format '%lx' expects argument of type 'long unsigned int', but argument 8 has type 'pa_channel_position_mask_t' [-Wformat]
pa_channel_position_mask_t is type defined to uint64_t, and to display
uint64_t, it's better to use PRIx64 primitives.
Signed-off-by: Deng Zhenrong <dzrongg@gmail.com>
Now that the client API exposes availability information for ports
on cards, we can make output consistent with "list sinks" and
"list sources".
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
For switching profiles, we are a little more cautious, only switch
from an unavailable port to an available one. Profile switching is
mainly used for HDMI/DisplayPort, and this is to avoid switching from
analog to HDMI/DP when it becomes available.
See http://lists.freedesktop.org/archives/pulseaudio-discuss/2012-March/012991.html
and replies for more information.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Calling adjust_rates after teardown results in segfault, and
judging from the Ubuntu bug report, this can happen.
Actively prevent this by destroying the time event, and by
setting adjust_time to 0, we also prevent this routine being
called on max request update.
BugLink: https://bugs.launchpad.net/bugs/946400
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
As David points out, the previous commit made a couple of asserts
redundant (the XOR covers all cases that were previous tested for).
Remove these redundant commits now.
Commit 54cddc6ddf removed an assert that
looked redundant but was not. This commit reinstates it in a slightly
modified form. It is not stated as (a ^ b) instead of (!a || !b) in
order to make the condition more obvious.
For some reason, a badly behaving client was trying to set a NULL
port, which caused PulseAudio to crash. Add safeguards on two levels
just to be protected. (Also remove a redundant check.)
BugLink: https://bugs.launchpad.net/bugs/951273
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
ConsoleKit has been deprecated and replaced by systemd's logind daemon,
hence provide the same functionality of module-console-kit in
module-systemd-login. This also makes sure that the CK module becomes a
NOP if the system is booted with systemd, resp. that the systemd module
becomes a NOP if the system is booted without systemd, thus being nice
to OSes such as Debian which want to support multiple init systems.
In most cases, we use dbus from more than one thread, as we
e.g. enable real-time scheduling from the ALSA threads.
Therefore set dbus to thread-safe mode by default, as recommended
in https://bugs.freedesktop.org/show_bug.cgi?id=47060#c5
This fixes a bug where PulseAudio could crash in two parallel
calls to pa_make_realtime.
BugLink: https://bugs.launchpad.net/bugs/937933
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This reverts commit 24ff719675.
If these files aren't compiled in both libpulse and
libpulsecommon, some things will try use non-public
functions from libpulse. Therefore those internal functions
have to be included directly in libpulsecommon.
This problem appears to be only visible with --as-needed,
which is why the problem wasn't noticed immediately. The
problem has existed also earlier, and it was fixed by
Maarten Bosmans in commit 2de2c735. The commit that is
now reverted basically reverted Maarten's commit (I didn't
know that when I wrote the bad patch).
A part of Xen's paravirtualized audio driver has been developed as a
pulseaudio module. This module acts as a tunnel over Xen's shared memory
mechanism and allows a domU guest to send audio data to a dom0 backend.
Reference: https://bugs.freedesktop.org/show_bug.cgi?id=43503
While developing the new UI we had to ask ourselves the question of whether
"speakers" should be considered available when headphones are plugged in.
In most cases, they are not available and therefore we should list them
as such.
OTOH, we don't want unplugging the headphones to be considered an act of
wanting to use the speakers (the user might prefer HDMI), and there might
be line-outs that keeps the speakers from unmuting anyway. So, at this point,
I think the most reasonable would be to make the speakers have
PA_PORT_AVAILABLE_NO when headphones are plugged in and
PA_PORT_AVAILABLE_UNKNOWN when they are not. But we might want to revisit
this decision once we have the priority lists up and running.
The same reasoning applies for "Internal Mic", which should become unavailable
when any other mic is plugged in.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
For volume control UIs to be able to show ports in inactive profiles,
expose all ports together with the card info. This includes updating
the protocol and the client API to show the connection between ports
and for which profiles the ports are relevant.
Update protocol to 26.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This module tries to switch to a port when availability changes to
"YES", and tries to switch away when availability changes to "NO".
Once there is a priority list infrastructure in place and ready,
this functionality might be redundant, but this will do as an
interim solution.
For Nvidia and Intel, support probing of up to four HDMI devices.
Also add port information to all HDMI profiles.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Support the new jack detection interface implemented in Linux 3.3
(and Ubuntu's 3.2 kernel).
Jacks are probed and detected using the snd_hctl_* commands, which
means we need to listen to them using fdlists. As this detection
needs to be active even if there is currently no sink for the jack,
so this polling is done on the card level.
Also add configuration support in paths, like this:
[Jack Headphone]
required-any = any
...where 'Jack Headphone' should match 'Headphone Jack' as given by
ALSA (as seen in e g 'amixer controls').
"Required", "required-any" and "required-absent" is supported. Using
required-any, one can have several ports even though there is no
other indication in the mixer that this path exists.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
It provides a virtual surround sound effect.
v2: Normalize hrir to avoid clipping, some cleanups
v3: use fabs, not abs
v4: implement changes proposed by Tanu Kaskinen
v5: likewise
v6: use channel map from hrir file
v7: remove hrir_ss and hrir_map form userdata
v8: update naming of sink
Our flist implementation suffers from the ABA problem
(see http://en.wikipedia.org/wiki/ABA_problem), causing PulseAudio
to crash very rarely, usually inside memblock operations.
By turning stored pointers into stored table indices, we have some
extra bits that we can use to store tag bits, which is a known
workaround for the ABA problem.
Buglink: https://bugs.launchpad.net/bugs/924416
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This allows clients to get a "fake" sample space for compressed formats
that we can support. This should make size/time conversion for things
like calculating buffer attributes simpler.
Since a given property can be single-valued, an array or (in the case of
ints) a range, clients need an API to figure out what type of value a
property holds. This adds such an API. The actual property type
enumeration is kept in the PA_PROP_* namespace and not the
PA_FORMAT_INFO* namespace so that it can later be reused for properties
generically if required.
This adds integer range/array and string array property getters to the
pa_format_info API. Corresponding tests added as well to ensure the code
is valgrind-clean.
The corresponding functions are added to map-file manually for now.
These utility functions could be handy to clients.
pa_format_info_to_sample_spec_fake() isn't made public, but the return
value is changed to keep in sync with pa_format_info_to_sample_spec().
This makes handling errors in getter functions more graceful, rather
than triggering warnings/asserts. Better to be less trigger-happy about
these things since this is now public-facing API.
We currently only have setters and clients need to be able to query
these values as well. The return types for these functions needed to be
changed to int since this is public API now.
Hi. Could you please apply the attached trivial patch so that I could drop
the corresponding instructions from dcaenc's README file in the future? It
adds a profile for on-the-fly DTS encoding, similar to the existing AC3
profile.
--
Alexander E. Patrakov
>From 22310a1c28385acc7ce883e020b9eb2e5b0813b7 Mon Sep 17 00:00:00 2001
From: "Alexander E. Patrakov" <patrakov@gmail.com>
Date: Sun, 12 Feb 2012 17:19:48 +0600
Subject: [PATCH] alsa: add DTS profile
This requires dcaenc from http://aepatrakov.narod.ru/dcaenc/
libtools causing relinking on installation, to make this succeed
libpulsecommon needs to be installed before the other libraries and the
padsp libraries needs to be installed afterwards.
Unfortunately autotools doesn't consider dependencies when running the
install target, thus we have to enforce the ordering ourselves
libpulsecommon is a private library only for use within pulseaudio, so
lets move it into a private directory
[ed: from discussion on IRC, while it looks redundant to have modules in
lib/pulse-$VER and private libraries in $pkglibdir, this is actually
desirable. For example, a multi-architecture libpulse would have a
$pkgdir per architecture (/usr/lib/<arch>/pulseaudio) whereas you'd
still want the server-specific modules in /usr/lib/pulse...) -- Arun]
The tunnel source has been broken since protocol v22 (PA 1.0),
and connecting fails with a protocol error. Fix.
BugLink: https://bugs.launchpad.net/bugs/923661
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Fix the documentation: the protocol file was not reflecting the code
properly for version 22.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If alsa settings/options were used, the string to construct the
name was freed, leading to dangling pointers to strings inside
the hashmap.
BugLink: https://bugs.launchpad.net/bugs/932804
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
assuming RAND_MAX is around 1<<31, rand() >> 1 generates large numbers as
random volume data; these likely causes saturated sample values after
applying the volume function -- not a good test
For muted channels, we forgot to increment a pointer, so if one
channel was muted but not the other, sound became distorted in a
Darth Vader like way. To test the difference, start two input
streams and pan one of them hard left (or right).
And hey, if you didn't think it sounded like Darth Vader, it's
your imagination that's broken, not mine! ;-)
BugLink: https://bugs.launchpad.net/bugs/928757
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If a *_UNLINK_POST hook causes a sink-input/source-output's sink/source
to go away, the subsequent attempt to update the sink/source status will
cause an assert. We deal with this by checking the sink/source status
before trying to update it.
The practical problem is that some users were left with only one
"LFE on Mono" port, as analog-output was considered a subset of
analog-output-lfe-on-mono. Which was not what they wanted.
BugLink: https://bugs.freedesktop.org/show_bug.cgi?id=40910
BugLink: https://bugs.launchpad.net/bugs/922656
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Flush the message queue before tearing down, and dest==NULL is valid in case moving failed.
With this my module-loopback finally no longer causes frequent crashes.
* If we don't have "Digital Speakers", we should say "Speakers"
instead of "Analog Speakers", and similar for other ports.
* Change "IEC958" to "S/PDIF" (more well known name)
* Add new ports and mappings for HDMI
* Change "Internal" to "Built-in" for the card name
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
"Front Speaker", "Surround Speaker" seems to be a common enough name
to make it into alsa-utils, so we should probably care about it as
well. In this case, there was a macbook pro whose speakers didn't work
without these controls.
BugLink: http://bugs.launchpad.net/bugs/551441
Reported-by: Jeroen T. Vermeulen <jtv@canonical.com>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The original intention of this code was probably that if
adding filter1 succeeded but adding filter2 failed, then
filter1 should be removed so that either both or none of the
filters get added.
We didn't do anything anyway in case of failures. When we
give NULL as the error, dbus_bus_remove_match() can act
asynchronously, so it becomes faster. Also, the bus daemon
can avoid sending any replies, which reduces the amount of
traffic.
The recent change to turn off the IEC958 element for analog paths
exposed a bug in AC3 profiles. These were inheriting the analog output
path instead of explicitly selecting the iec958 path.
Thanks to David Henningsson for pointing this out.
This is needed for the Creative Audigy CA0106 to work. Also makes sure
that the LED for optical out is shut down in analog modes on MacBooks
(these share a port for analog and digital output).
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=44741
Just like we turn off the "Front Mic" element when we select "Rear Mic",
we should also turn off the "Front Mic Boost" element. And the same for
the other inputs.
Reported-by: Len Owens <len@ovenwerks.net>
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
--exit-idle-time and --scache-idle-time were marked as having an
optional argument when the argument is actually mandatory. This causes a
crash when using this argument.
Thanks to Matthijs Kooijman (blathijs on IRC) for pointing this out.
We can't call IT_PROG_INTLTOOL because it requires intltool even with
--disable-nls, but it provides USE_NLS which is required for the po/
build. This might take a while to get fixed upstream, so working around
this in our build system for now.
configure doesn't provide an abs_top_srcdir early enough, so we just use
srcdir for that. There was also a problem with the path we were setting
in VERSIONING_LDFLAGS that is fixed now.
From d8b81d5393df36085009bf9f69d41fa85e2ae58a Mon Sep 17 00:00:00 2001
From: Nitin A Kamble <nitin.a.kamble@intel.com>
Date: Sat, 10 Dec 2011 09:09:06 +0100
Make assembly syntax compatible to the X32 toolchain and fix the
following kind of compilations errors with X32 gcc.
| pulsecore/svolume_mmx.c: Assembler messages:
| pulsecore/svolume_mmx.c:107: Error: `(%esi,%rdi,4)' is not a valid base/index expression
| pulsecore/svolume_mmx.c:135: Error: `(%esi,%rdi,4)' is not a valid base/index expression
| pulsecore/svolume_mmx.c:161: Error: `(%esi,%rdi,4)' is not a valid base/index expression
| pulsecore/svolume_mmx.c:162: Error: `8(%esi,%rdi,4)' is not a valid base/index expression
| pulsecore/svolume_mmx.c:180: Error: `(%esi,%rdi,4)' is not a valid base/index expression
| pulsecore/svolume_mmx.c:210: Error: `(%esi,%rdi,4)' is not a valid base/index expression
| pulsecore/svolume_mmx.c:244: Error: `(%esi,%rdi,4)' is not a valid base/index expression
| pulsecore/svolume_mmx.c:245: Error: `8(%esi,%rdi,4)' is not a valid base/index expression
| make[3]: *** [libpulsecore_1.1_la-svolume_mmx.lo] Error 1
Originally these assembly lines were written for x86_64 ABI, now they
are also compatible with X32 ABI [3][4].
The patch was submitted to the OpenEmbedded-Core list [1][2].
[1] http://lists.linuxtogo.org/pipermail/openembedded-core/2011-December/014189.html
[2] http://git.yoctoproject.org/cgit.cgi/poky-contrib/commit/?h=nitin/x32&id=2d8eec54f755c51f2eff600390f5a4b3cc2a7662
[3] https://wiki.yoctoproject.org/wiki/X32_abi
[4] http://en.wikipedia.org/wiki/X32_ABI
This is actually implemented in module-protocol-stub as
'auth-group-enable'. An unfortunate typo because the other argument is
spelt as 'enabled', but it's better to be slightly inconsistent than to
change the public interface.
makes the Adrian echo canceller implementation optional at compile time
this patch supersedes an earlier patch proposal and addresses the following
comments:
* separate patch from speex dependency rework (Arun)
* check that at least one EC implementation is available (Arun)
* properly align yes/no in configure summary for Adrian (Frederic)
make speex library dependency optional, this affects the resampler
and the echo canceller module
this patch supersedes an earlier patch proposal and addresses the following
comments:
* fix order of pa_echo_canceller_method_t enum and ec_table (Frederic)
* the default resampler is speex if available as before, otherwise ffmpeg (Arun)
* does not touch the Adrian EC implementation (see separate patch) (Arun)
* If mapping_probe_paths() fails to open the mixer, all paths are now
removed from the mapping's path sets.
* pa_path_set.probed isn't really used for anything (removed).
* If profile probing is configured to be skipped, mapping_paths_probe()
should still be called.
Thanks to Tanu for spotting.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
To be able to add ports to all profiles, we need to probe all
profiles at startup. To speed this up, we now have a cache of
probes paths which is owned by the profile set. Since paths
are now owned by the profile set, the path set must now have
a hashmap of paths instead of a linked list.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The thinkpad ACPI driver sometimes creates a virtual sound card,
which at best exposes a volume control. Save some startup time, and
unnecessary error messages in the log, by ignoring it.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The algorithm had been implemented the same way as the trivial resampler. But
an important difference between the two is that the trivial resampler can write
an output as soon as the first corresponding input sample is seen, whereas the
peaks resampler must have read all input samples before writing an output
sample.
With this rework, the peaks resampler now outputs samples correctly when the
input data is spanning multiple memblocks.
Instead of using PA_SCACHE_ENTRY_SIZE_MAX, the size for FRAME_SIZE_MAX_ALLOW is
set directly to the same value. This removes the need for the core-scache.h
include, which caused an unwanted dependency of libpulsecommon on libpulsecore.
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=41539
Without this fix, errors about previous definitions are generated in
numerous locations.
Example of error:
CC libpulsecore_1.98_la-auth-cookie.lo
In file included from ../../src/pulsecore/source.h:46:0,
from ../../src/pulsecore/sink.h:40,
from ../../src/pulsecore/core.h:50,
from ../../src/pulsecore/shared.h:25,
from ../../src/pulsecore/auth-cookie.c:33:
../../src/pulsecore/device-port.h:40:24: error: redefinition of typedef
'pa_core'
../../src/pulsecore/core.h:29:24: note: previous declaration of
'pa_core' was here
make[3]: *** [libpulsecore_1.98_la-auth-cookie.lo] Error 1
Overall it would be nicer if we could avoid this kind of fix, but it
would require further reorganisation that I'm not prepared to undertake
right now.
The recommended way of setting available status is to call
pa_device_port_set_available, which will send a subscription event
to the relevant card. It will also fire a hook.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This forms the base for being able to expose all ports of all
profiles (even inactive ones) to clients.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Since both cards and sinks can hold references to a port, it makes
sense to reference count them. Although no current implementation
actually has sinks with ports but without a card, it felt wrong
to make it harder to make such an implementation in the future.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This makes handling of echo-cancel streams (and potentially other
filters that need to work with multiple streams) cleaner. We do this by
adding the ability to apply filter on a group of streams rather than
always treating streams individually.
This is required, for example, when changing the input source for an
echo-cancel source output. When this happens, we want to change the
module-echo-cancel instance to actually cancel between the new source
and old sink.
To do this when one stream in a group moves, we create a new filter
instance between the current master sink/source pair and move the
relevant streams to the virtual sink/source from this new instance.
This moves out code from module-stream-restore and makes an internal API
out of it to get a "stream group" for a given sink input or source output.
This is factored out for reuse in module-filter-*.
The stream group basically provides some means of attaching a logical
identification to the stream (by role, application id, etc.).
When autoloaded, it is expected that module-filter-apply (or whatever is
loading us) will take care of applying the filter on the correct
sink/source master. Instead of adding complexity by tracking what is
currently being filtered, we just disallow filtering anything except the
original master sink/source and let module-filter-apply or whatever is
loading us deal with dynamic sink/source changes.
This makes what devices are being cancelled clearer in the UI (at the
cost of being somewhat less clear when multiple devices of the same name
are plugged, but at least that's a much smaller set than everyone).
In pa_create_stream_callback, a stream is inserted into
s->context->record_streams only if it's a record stream. Otherwise it's
inserted into s->context->playback_streams. However, in stream_unlink the
stream is removed from s->context->playback_streams only if it's a playback
stream and otherwise it's removed from s->context->record_streams.
Thus, if the stream is an upload stream, we first insert it into
s->context->playback_streams in pa_create_stream_callback and then try to
remove it unsuccessfully from s->context->record_streams in stream_unlink. This
means that we are leaking hashmap entries until the context is freed,
constantly consuming more memory with applications that upload and unload a
large number of samples through one context.
Of course, this begs the question whether upload streams even belong in either
of those hashmaps. I don't want to mess around with the code too much at this
point though, so this patch should be a sufficient improvement.
This adds some infrastructure for canceller implementations to also
perform acoustic gain control. Cancellers now have a couple of new API
calls that allow them to get/set capture volume.
This is made slightly complex by the fact that cancellation happens in
thread context while most volume mangling needs to be done in main
context. To deal with this, while getting the volume we save source
volume updates as they are propagated to thread context and use this
cached value for queries. To set the volume, we send an async message to
main context and let that set the source volume.
This updates corked streams' resamplers when switching sample rates on a
sink/source, which means the restriction of allowing sample rate updates
only when no streams are attached to a sink/source is now relaxed to
preventing updates only when there is a running stream attached.
The purpose of this command is to print all the internal volume
variables for sinks/sources and all corresponding
sink-inputs/source-outputs to make debugging and reasoning about
volume-related issues easier.
This adds a boolean module parameter to disable automatic dynamic
latency readjustments on underruns, but leaves automatic dynamic
watermark readjustments untouched.
Allow module-bluetooth-device to listens to HandsfreeGateway state
changes using DBUS signals. When an handsfree connects, module-bluetooth-device
is loaded and goes to playing state. When the handsfree disconnect audio,
the card profile is set to "off". If the headset connects audio again after
that, the card profile should switch to "hfgw" again to match state of audio
connection.
If card profile is set to "off", the audio stream should be released.
Current implementation releases the stream when the card profile
is changed to "hsp" or "hfgw" again and immediatly reconnects after that.
This happens in the following scenario :
An HandsfreeGateway connects RFCOMM and then SCO. A card appears in
PA and can be used. If for some reason, SCO is disconnected,
module-bluetooth-device is unloaded. The card will disappear, even
if RFCOMM is still connected. After that, it is not possible to
connect SCO again from PA.
This patch will add the necessary quirks so that pulseaudio can register
an endpoint on the /MediaEndpoint/HFPHS path. This endpoint is to be
used for HFP Handsfree profile.
Some tests (remix-test, sig2str) only display information, so they are not
useful for automated testing. Others (interpol-test, once-test, thread-test)
do return an error on failure, so should be included in TESTS.
Instead of spilling thousands of lines of output, make check now runs the
test-suite in about 100 lines or so. If running under make check, the output of
tests is reduced. The MAKE_CHECK environment variable is used for this, so that
when running the test manually, the full output is still shown. Furthermore,
pa_log is used consistently instead of printf, so that all test output goes to
stderr by default. Colored output from make check goes to stdout.
When a test program exits with a nonzero return value (or an assert is hit),
the test is regarded as a FAIL.
This makes `make check` a little more useful.
This dumps out an additional file with each line having a command of the
form:
p <number of playback samples processed>
c <number of capture samples processed>
d <drift as passed to set_drift()>
The test program can be provided this file to "replay" the data exactly
as when it was run live.
The non-drift-compensation path is retained as-is since it is much
simpler.
This adds the ability for echo cancellers to provide their own drift
compensation, and hooks in the appropriate bits to implement this in the
WebRTC canceller.
We do this by introducing an alternative model for the canceller. So
far, the core engine just provided a run() method which was given
blocksize-sized chunks of playback and record samples. The new model has
the engine provide play() and record() methods that can (in theory) be
called by the playback and capture threads. The latter would actually do
the processing required.
In addition to this a set_drift() method may be provided by the
implementation. PA will provide periodic samples of the drift to the
engine. These values need to be aggregated and processed over some time,
since the point values vary quite a bit (but generally fit a linear
regression reasonably accurately). At some point of time, we might move
the actual drift calculation into PA and change the semantics of this
function.
NOTE: This needs further testing before being deemed ready for wider use.
The "(((audio_sample << 1) | 1) << frame->scale_factor[ch][sb])"
part of expression
"frame->sb_sample[blk][ch][sb] =
(((audio_sample << 1) | 1) << frame->scale_factor[ch][sb]) /
levels[ch][sb] - (1 << frame->scale_factor[ch][sb])"
in "sbc_unpack_frame" function can sometimes overflow 32-bit signed int.
This problem can be reproduced by first using bitpool 128 and encoding
some random noise data, and then feeding it to sbc decoder. The obvious
thing to do would be to change "audio_sample" variable type to uint32_t.
However the problem is a little bit more complicated. According
to the section "12.6.2 Scale Factors" of A2DP spec:
scalefactor[ch][sb] = pow(2.0, (scale_factor[ch][sb] + 1))
And according to "12.6.4 Reconstruction of the Subband Samples":
sb_sample[blk][ch][sb] = scalefactor[ch][sb] *
((audio_sample[blk][ch][sb]*2.0+1.0) / levels[ch][sb]-1.0);
Hence the current code for calculating "sb_sample[blk][ch][sb]" is
not quite correct, because it loses one least significant bit of
sample data and passes twice smaller sample values to the synthesis
filter (the filter also deviates from the spec to compensate this).
This all has quite a noticeable impact on audio quality. Moreover,
it makes sense to keep a few extra bits of precision here in order
to minimize rounding errors. So the proposed patch introduces a new
SBCDEC_FIXED_EXTRA_BITS constant and uses uint64_t data type
for intermediate calculations in order to safeguard against
overflows. This patch intentionally addresses only the quality
issue, but performance can be also improved later (like replacing
division with multiplication by reciprocal).
Test for the difference of sbc encoding/decoding roundtrip vs.
the original audio file for joint stereo, bitpool 128, 8 subbands
and http://media.xiph.org/sintel/sintel-master-st.flac sample
demonstrates some quality improvement:
=== before ===
--- comparing original / sbc_encoder.exe + sbcdec ---
stddev: 4.64 PSNR: 82.97 bytes:170495708/170496000
=== after ===
--- comparing original / sbc_encoder.exe + sbcdec ---
stddev: 1.95 PSNR: 90.50 bytes:170495708/170496000
The header files with constants and structures for audio specific
interaction with Pulseaudio are suppose to be under LGPL license.
For some odd reason a2dp-codecs.h ended up being under GPL license
which is against the intention of this being shared and re-used by
non-GPL programs. Fix this now to avoid any future confusion.
This adds a pa_str_in_list() to check for a given string in a
space-separated list of strings. For now, this is merely present to
avoid duplication of role matching code (intended roles can be a
space-separate list) across modules.
The documentation says we expect a comma-separate list of intended
roles, but the code splits the string on whitespaces, so this corrects
the documentation to match the implementation.
The error message for snd_pcm_hw_params_set_period_wakeup was
printing "ret", but "ret" wasn't being set.
Signed-off-by: Dylan Reid <dgreid@chromium.org>
Allow a module argument to specify that we should act globally
rather than just within a given sink.
The default value is to not opporate globally thus retaining the
current behaviour.
Operate on a list of 'trigger roles' and 'cork roles'. i.e.
react to any stream with a role in the trigger list and apply a
cork to any stream with the a role in the cork list.
The trigger roles default to 'phone' and the cork roles default
to both 'music' and 'video' thus achieving the same functionality
as currently when called without any arguments.
When starting via a console login, PA will likely not have a session DBus
to play with. As there is no X11 environment, libdbus will be unable
to launch a session DBus for us and thus the module will fail to load
which in turn prevents PA from loading.
If the user subsequently logs into X11 this it will still not be possible
to load the module as the server will be ignorant of the X11 and DBus
environment variables so a longer term solution for handling this should
be found.
Some of the license wording was less than clear. Try to clarify the
different GPL 'downgrade' scenarios but also be generic to ensure that
those packagers where GPL is a problem check thoroughly before they ship.
Inspired by comments from Brian Cameron @ Oracle via fdo#41822
I'd rather not have to do this, as I don't really see the point in
duplicating what is done in pkgconfig, but this is likely the
easiest way to avoid nasty hacks.
pa_sink/source_used_by() ignores corked/monitor streams, but we need to
make sure there aren't any of these while updating rate (at least for
now -- this is a restriction that would be nice to get rid of).
Basically adds code to handle passthrough sources. This isn't a tested
path at the moment, but in the future, when we do wish to support these,
it'll save us the trouble of having to sync all the code again.
This is where the actual changes happen.
Some additional checks would be required to make sure the
rate is actually supported
Tested with both PCM and passthrough streams
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Avoid resampling or use integer resampling when supported by the
sinks/sources
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
This adds the WebRTC echo canceller as another module-echo-cancel
backend. We're exposing both the full echo canceller as well as the
mobile echo control version as modargs.
Pending items:
1. The mobile canceller doesn't seem to work at the moment.
2. We still need to add bits to hook in drift compensation (to support
sink and source from different devices).
The most controversial part of this patch would probably be the
mandatory build-time dependency on a C++ compiler. If the optional
--enable-webrtc-aec is set, then there's also a dependency on libstdc++.
The new module argument can be used to provide a custom
directory for loading alsa path configuration files. This is
useful for testing: no need to be root to create test
configuration files.
module-stream-restore and modile-filter-apply can get into an infinite
loop if m-s-r is called before m-f-a (m-s-r rescues a stream and
attaches it to a sink/source, which then triggers m-f-a to move it back
to the filter sink/source, and so on). The purpose of the m-f-a hooks is
to beat m-s-r, so moving them to be run first.
This removes the nasty side-effect that a call to
pa_namereg_get_default_{source,sink}() will also *set* the default
source/sink.
This is a more complete fix for commit 766dbc68 ("conf: Make sure
module-dbus-protocol is loaded after module-default-device-restore")
https://bugs.freedesktop.org/show_bug.cgi?id=40897
This makes sure that we only perform any processing (resync or actual
cancellation) after the source provides enough data to actuall run the
canceller.
When a source-output isn't connected to our virtual source, we skip echo
cancellation altogether. This makes sense in general, and makes sure
that we don't end up adjusting for delay/drift when nothing is
connected. This should make convergence faster when the canceller
actually starts being used.
This increase the threshold for difference between the playback and
capture stream before samples are dropped from 1ms to 5ms (the
cancellers are generally robust to this much and higher). Also, we make
this a module parameter to allow easier experimentation with different
values.
Watermark level and latency values are not restored when
resuming, the values used prior to suspending are reused.
This leads to side effects when underruns happen and buffer
sizes are updated, PulseAudio can never meet lower latency
requirements.
Solution: keep track of watermark and latency values on sink or
source creation, and reapply them on resume to start with
a clean slate.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
This problem was found when tracing down a crash coming from the
esound protocol, which does not set a channel map.
BugLink: http://bugs.launchpad.net/bugs/864071
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
N.B.: As Colin notes, this is because commit 117c7145 was incomplete
("format: Fix channel map handling")
If module-jackdbus-detect failed in the later part of initialization,
the ma variable was freed twice.
BugLink: http://bugs.launchpad.net/bugs/867444
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Instead of relying on the snd_mixer_* functions failing, we check for
POLLERR and POLLNVAL first. After this, any errors in handling the mixer
events are deemed fatal (that is we cause the ALSA source/sink thread to
terminate).
The case where POLLERR is set but POLLNVAL is not does not actually
occur, but we're making this a soft failure (stop polling the mixer, but
don't kill the I/O thread). If other conditions where POLLERR occurs
turn up, we need to handle them explicitly.
Thanks to Linus Torvalds for helping get this right.
Loading between a sink and its monitor causes a deadlock (while sending
messages for latency snapshots). It isn't a case that has any real
conceivable use, so let's just disallow it.
This improves the error handling in the mixer rtpoll callback. It avoids
a crash if an error occurs (the rtpoll_item is freed but still
referenced), and specifically makes sure we don't continue trying to
poll the device if the card is disconnected.
This makes set_formats() put PCM formats lower down the list than
compressed formats since we negotiate by picking the first format in
this list that is also in the client-provided list of possible formats
during sink input creation.
This will be incorrect if we ever decide to do encoding in PA (for
things like AC3/DTS encoding for multichannel output over S/PDIF).
This patch was already added earlier with commit ID 2f86ba4f, but the
changes got reverted by commit 3adc43b ("win32: Make once-test work").
However, this still doesn't work on OSX as here, pthread is in general
available, but the barrier APIs aren't.
The purpose of this patch is to make it possible to configure stream volumes
before pulseaudio is run for the first time. This is useful, for example, in
embedded products where the default volumes have to be sensible already in
the first boot.
These are not used for anything at this point, but this
makes it easy to add ad-hoc debug prints that show the
memblockq name and to convert between bytes and usecs.
For both the headers and the library we should provide clean, three part
strings as this has been what we've previously done in the past
and some external systems apparently rely on this format. While it's not
something we've officially commented on before, there is no real advantage
to us to change it so let's not try to tidy things up too much
considering some third party apps (e.g. Skype) seem to dislike a two
part version string.
pa_{sink,source}_volume_change_apply were being called by the ALSA I/O
thread on every iteration, causing a pa_rtclock_now() call, which can
sometimes be heavy. We avoid this call by making sure there actually are
changes to apply before proceeding into the function.
While we're at it, also dropping a redundant check on s->write_volume.
Once in a million or so, this typo causes a crash when two threads
simultaneously try to call "pa_asynqmsgq_write_poll".
BugLink: http://bugs.launchpad.net/bugs/853560
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
If module-dbus-protocol fails to start, pa__done is still called,
which falsified the assumption that u->connections was always set.
BugLink: http://bugs.launchpad.net/bugs/855729
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This makes sure that when we're traversing the device chain for sources
and sinks with shared volume, we handle the case that a sink-input or
source-output of one of these might be unlinked (while unloading a
module, for example).
module-dbus-protocol gets the default sink, which sets the default sink
if not already set. This is turn makes module-default-device-restore do
nothing.
To solve the problem, make sure module-default-device-restore is loaded
before module-dbus-protocol and not the other way around.
BugLink: http://bugs.launchpad.net/bugs/843780
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Sometimes the ALSA mixer can be modified during a point at shutdown
which causes a race condition trying to update the volume of an
unlinked sink.
Includes typo fix by our Chief Typo Spotter, Colin, and a clarifying
comment by me.
BugLink: http://bugs.launchpad.net/bugs/841968
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Uses the shared volume infrastructure by default with an option to
fallback on the old pretend-volume-sharing-that-kind-of-works if someone
wants it that way.
Users who keep left != right (or any sort of unbalanced channel volumes)
will likely want to disable shared volumes since it will cause their
master sink/source volume to be balanced.
This really isn't a very pleasant scenario since users would need to
manually set up echo cancellation in their config for this (until we
have a way to store module configuration). That said, the majority case
benefits from the volume sharing, so let's not wait for the
configuration infrastructure to be ready to use this.
Uses the shared volume infrastructure by default with an option to
fallback on the old pretend-volume-sharing-that-kind-of-works if someone
wants it that way.
This handles the case where a virtual sink/source and it's master have
different channel counts. The solution is not ideal because if the
former has fewer channels and the master has channel volumes that are
not all at the same level, it will lose this information and have all
channels at the same level.
This is not just a theoretical problem, since module-echo-cancel
prefers a mono virtual source/sink and will usually be sitting on top of
a stereo ALSA source/sink.
That said, I don't really see a good solution to this problem, so the
idea is to make volume sharing optional (on by default) in
module-echo-cancel, so that the few people who care can then disable it
if they so desire.
This just covers Lennart's concern over the terminology used.
The majority of this change is simply the following command:
grep -rli sync[-_]volume . | xargs sed -i 's/sync_volume/deferred_volume/g;s/PA_SINK_SYNC_VOLUME/PA_SINK_DEFERRED_VOLUME/g;s/PA_SOURCE_SYNC_VOLUME/PA_SOURCE_DEFERRED_VOLUME/g;s/sync-volume/deferred-volume/g'
Some minor tweaks were added on top to tidy up formatting and
a couple of phrases were clarified too.
When dealing with proplists passed as modargs, we need the unescaped form
in order to properly deal with quotes (ticks + double quotes). As the previous
code always called pa_unescape() before adding it into the modarg hashmap, this
was impossible.
This modification simply stores two proplists. If the unescaped value
is different from the raw value, we also keep the raw form.
When parsing proplist arguments, we use this raw form and do the unescaping
ourselves when processing it.
This changes the current behaviour which required you to double escape
proplists arguments. This double escape mechanism did allow you to mix
and match what types of quotes you used to delimit the individial
proplist values, but it made the actual data much harder to pass in.
This approach has the drawback that you cannot mix and match the quotes
you use, but this is a very minor issue and IMO pales in comparison to
the general clarity gained.
See the discussion on the mailing list for more background:
http://lists.freedesktop.org/archives/pulseaudio-discuss/2011-September/011220.html
The Kinect shows up as a UAC device after the firmware has been loaded,
but in order to be detected by pulseaudio a 4-channels input only
mapping is needed. Provide a new profile for that and set it with a udev
rule.
fdo#39664
pa_core_maybe_vacuum now vacuums if there are either no streams or all devices are suspended.
The mempool_vacuum argument to module-suspend-on-idle is gone and defaults to true now.
We used to support older DBus versions but 1.3.0 is two years old
now and by requiring it we cut down of deviated code paths at
runtime and thus have less support issues.
fdo#40635
The Apple TV for example uses a non-default port, but we previously ignored
this. We now correctly parse the server string but in so doing, we end up
parsing the address twice. As we need a pure IP/hostname of the device itself
to use in our requests, this is somewhat unavoidable.
Sadly there are still other problems with Apple TVs, but this is still
one step closer.
Fixes part of #950
All of these functions are not actually defined in format.h
(they are defined in internal.h) and thus should really be
included only in libpulsecommon and implemented in a separate
source file.
However if that approach was taken, and these functions were
included in libpulsecommon, then they would have a link time
dependancy on libpulse (as these four functions use other
pa_format_info_* functions). As the opposite is already true
(libpulse depends on libpulsecommon), this is not possible as
it creates a circular dependancy.
Thus the only option is to just to include these four functions
in the map-file, but not actually export any public headers for
them. Of course users could use this implementation by defining
them in their own headers, but the only practical problem
with this approach is naming conflicts which is unlikely to happen.
fdo#40616
Without this change any applications calling e.g. pa_format_info_new()
and friends will be explicitly linked against libpulsecommon-$MAJORMINOR.so
which is something we specifically avoid as it may contain ABI/API unstable
functions.
Also ensure we export pa_format_info_from_string() for external use.
fdo#40616
Rather than write all the keys out for each port, simply write a 'null'
port entry and modify the read code to 'fallback' to this when it cannot
find a key. This is needed as the code used when writing the key may not
actually have the sink ports available at the time it uses them,
and thus can cause a segv. This approach adds some degree of overhead
but it's relatively minimal and it can be mitigated by compiling
without support for legacy database formats if so desired.
Thanks to David Henningsson for pointing out the problem.
I've seen more than one system where the volume control named
"Front" is a part of audio path for headphones. This is somewhat
of a compromise: While we don't merge it into the path, as that
would be regressing machines where "Front" isn't a part of the
audio path, it would still enable sound on these machines.
BugLink: http://bugs.launchpad.net/bugs/804178
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Unlinked streams can not be moved, so don't try.
BugLink: http://bugs.launchpad.net/bugs/837386
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Since none of our types is derived from GType, explicitly signal this.
The default value for this attribute changed at some point, breaking our
bindings.
When unloading, some module may end up trin to move a sink-input or
source-output back onto our virtual sink/source, causing an infinite
loop of us moving the stream away and having it moved back.
We prevent this from happening by preventing any stream from being
attached during unload.
Pacat remembers the number of cork requests, and then cork/uncork the stream
accordingly.
With this change, it makes below test script work correctly:
pacat -p --property=media.role="music" <long-sound> &
sleep 2
pacat -p --property=media.role="phone" <short-sound>
wait
Initial idea by Lu Guanqun, but modified by Colin Guthrie (so blame
me if it's broken)
The priorities should be set in the approximately likeliness that someone
would like to use them. This is hard to guess of course, but this is
a better guess compared to the existing configuration.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This will allow for volumes to be saved separately for e.g. Headphones vs. Speakers.
At present it is possible that no volume will be saved for the device prior to the port
switch. In this case the volume will not change from the value set under the other port.
In an ideal world we would save the volume before switching port, but that would require
a new hook.
The device restore saves both the current port for a device, and the volume/mute/formats.
The latter three are really related to the device+port, not the device on it's own.
Thus we store just the port for a device in one key and the volume/mute/formats
under a different key that includes both the device and the port.
I initially included put the Speex preprocessing assuming that we'd want
to use the digital gain control and noise suppression from Speex for all
echo cancelling implementations. In practice, we're probably going to
get entire implementations all processing in one package (WebRTC, custom
modules from various vendors, etc.).
This moves out this preprocessing and related knobs into the speex
implementation, which serves to clean out all implementation-specific
details from the module-echo-cancel core.
This is for bits of code or API where the concepts and constructs are
extensibly the same for sinks and sources. To avoid duplication we can
simply define our structure/API and then use this enum to differentiate
the two.
This reverts commit ba163b8b23.
After discussion with Colin, decided that we don't want to do this
piecemeal -- need some discussion on the mailing-list and then we should
do this across the tree in one fell swoop post 1.0.
Conflicts:
src/modules/module-device-restore.c
This is required for E-AC3 streams, as well as to let receivers we're
sending non-PCM data (which avoids playing noise if the data is
incorrect for some reason).
I don't know the exact cause for someone to submit a bug report for
this error message: if someone is truly offended by it (if so it is
a reminder that some people are more sensitive than others, and I do
want those people to feel welcome as well), or if it's a system's
check (if this goes through, it shows the system works, and the person
might put more work into his/her next patch), or if it's just a bug
(after all, it's not that weird to run two instances of PulseAudio?).
Either one could be reason enough to apply IMO.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
The "Line HP Swap" element needs to be set correctly for some Dove
boards to work correctly. Thanks to Daniel T Chen for the patch.
BugLink: http://bugs.launchpad.net/bugs/451635
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This switch needs to be off for analog output to work on several
Audigy cards.
Thanks to Daniel T Chen for the patch.
BugLink: http://bugs.launchpad.net/bugs/408370
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
As various modules can subscribe to unlink callbacks unloading some modules
may trigger hooks in other modules.
The callbacks associated with these hooks could in turn need to use the core
in some capacity (e.g. perhaps they are module loading modules
(e.g. *-discover, filter-apply or gconf etc. and need to use the core to
unload modules they've loaded).
This change simply ensures that all modules and cached samples are unloaded
before freeing the core.
The callback should also be reset in reset_calbacks().
The extra check in _volume_change_apply() is needed because when the sink is unlinked the callbacks are reset,
but there still may be pending volume changes.
This adds a PA_SINK_SET_FORMATS flag to the pa_sink_flags enum,
signalling that a sink allows the set of supported formats to be set
externally. The idea is for clients to be able to know what sinks
support this ability and adapt their UI appropriately.
This loads module-filter-heuristics and module-filter-apply by default
so that applications can request filters via properties. Not adding this
to system.pa -- the assumption is that people running system mode would
want more fine-grained control over such options.
Removing the bit that automatically loads module-echo-cancel for phone
streams. Clients need to specifically opt in for this now with
filter.want, until we are reasonably certain this won't break other apps
(Skype for one, possibly others).
The order of freeing the hashmaps is important here, because otherwise a string used as key is freed before the hashmap
is freed.
Valgrind reports this as:
Invalid read of size 1
at 0x4107042: pa_idxset_string_hash_func (idxset.c:67)
by 0x4106026: remove_entry (hashmap.c:93)
by 0x41061BF: pa_hashmap_free (hashmap.c:110)
by 0x71DD143: pa_dbusiface_core_free (iface-core.c:2105)
by 0x71F2169: module_dbus_protocol_LTX_pa__done (module-dbus-protocol.c:595)
by 0x406DC51: pa_module_free (module.c:162)
by 0x406E01D: pa_module_unload_all (module.c:210)
by 0x4068842: core_free (core.c:169)
by 0x406FD5D: pa_object_unref (object.c:64)
by 0x805224D: pa_core_unref (core.h:184)
by 0x805560B: main (main.c:1159)
Address 0x4d099c0 is 0 bytes inside a block of size 100 free'd
at 0x4025BF0: free (vg_replace_malloc.c:366)
by 0x40F128C: pa_xfree (xmalloc.c:131)
by 0x71E4CEB: pa_dbusiface_device_free (iface-device.c:1293)
by 0x71DCD7E: free_device_cb (iface-core.c:2062)
by 0x41061D7: pa_hashmap_free (hashmap.c:113)
by 0x71DD125: pa_dbusiface_core_free (iface-core.c:2104)
by 0x71F2169: module_dbus_protocol_LTX_pa__done (module-dbus-protocol.c:595)
by 0x406DC51: pa_module_free (module.c:162)
by 0x406E01D: pa_module_unload_all (module.c:210)
by 0x4068842: core_free (core.c:169)
by 0x406FD5D: pa_object_unref (object.c:64)
by 0x805224D: pa_core_unref (core.h:184)
The commit 7ebc5033 resulted in segfaults, because format->plist was not allocated.
The solution is not to allocate pa_format_info on the stack, but to properly use pa_format_info_new().
Also a typo regarding pa_tagstruct_putu8 is corrected.
I was looking at a log, and noticed the following lines:
I [pulseaudio] svolume_mmx.c: Initialising MMX optimized functions.
I [pulseaudio] remap_mmx.c: Initialising MMX optimized remappers.
I [pulseaudio] svolume_sse.c: Initialising SSE2 optimized functions.
I [pulseaudio] remap_sse.c: Initialising SSE2 optimized remappers.
I [pulseaudio] sconv_sse.c: Initialising SSE2 optimized conversions.
It seemed odd that some messages were somewhat precise in
what functionality was initialized, while the svolume
messages told me that they had initialized just "functions".
So I made the svolume log messages more precise to match the
sconv and remap messages.
Removes the comma as the proplist separator since that makes
pa_proplist_from_string() break and prints only the encoding if there
are no properties (instead of "<encoding>, (no properties)").
This makes the pa_bool_t members of userdata not be a single bit field
since pa_bool_t can be an int, potentially causing signedness issues in
comparisons.
This implements the sink get_formats() and set_formats() API in
alsa-sink. Modules can use this to allow users to specify what formats
their receivers support.
This adds API to let external sources specify what formats a sink
supports. Sinks must opt-in to allow this, and can perform some
validation if required.
Note in protocol-dbus.c specifically, method_signatures needs to be freed
before method_handlers, because otherwise h->method_name is freed while it is
still in use as a key in the method_signatures hashmap.
revents marked as POLLOUT|POLLERR|POLLWRNORM in "underrun" case that will
trigger unexpected log "ALSA woke us up to write new data to the device, but
there was acturally nothing to write...".
This patch avoids this log message.
This is not ideal but in order to aid people using it in scripts
etc, we will maintain backwards compatibiliy here.
Also add a 'short' mode and mention in the man page that this
will ultimately become the default at some point in the future.
This patch displays thread name in the log, that would be more descriptive.
It improves Xingchao (xingchao.wang@intel.com)'s patch which shows thread id.
Signed-off-by: Lu Guanqun <guanqun.lu@intel.com>
* Fix extension API function export list.
* Ensure we trigger a subscription event when things change.
* Send the index with our subscription events.
* Clear out any existing formats when saving.
* Call the correct extension command for subscriptions.
As reported in http://kpaste.net/04f1f3f
it is possible to call enumeration_is_subset with null pointers.
Handle that case instead of crashing. (It is also possible that
Tanuk's pending element_is_subset patch solves the issue, but this
nevertheless gives some extra security.)
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
This patch introduces some extra protocol information, so protocol
version is bumped. This functionality is primarily needed to solve
a long standing issue in alsa-plugins, which should ignore underruns
if and only if it is obsolete, i e, if more data has been written to
the pipe in the meantime (which will automatically end the underrun).
BugLink: http://bugs.launchpad.net/bugs/805940
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
We want to set the volume callbacks only if volume sharing
is not used. When volume sharing is used, we don't want to
mess with the stream volumes.
This was broken in 6c6b50
Now that subset mixer paths are removed, this workaround is no longer needed.
This effectively reverts 1c38b5d478 but due
to me forgetting to add files and adding a couple extra workarounds after,
it's easier to just do this manually rather than run git-revert.
In order to try and avoid 'spamming' the user with port choices,
attempt to detect and remove any pointless paths in a path set. That is
any paths which are subsets of other paths.
This should solve a problem case with some USB Headsets which result in
two paths both involving the 'Speaker' element. When no 'Master' element
exists (which is quite common on head/handsets), then the first path
(analog-output) will contain the 'Speaker' in a way that completely fits
with in the use of the 'Speaker' element in the other path
(analog-output-speaker).
Unification is really just a 'lowest common denominator' system. If any
paths do not support volume, mute or decibels, then mark them all as not
having them.
This was originally done this way because the flags set on sinks that
dictate if it supports h/w volume, mute etc. could not be changed after
the sink was created.
The fact that these flags could not change has now been change in the
previous commits, and thus there is now no need to use this 'lowest
common denominator' approach as we can fully support the various
different combinations, even if they change after initial creation
of the sinks/source.
Some sink flags are really just a product of what callbacks
are set on the device. We still enforce a degree of sanity
that the flags match the callbacks set, but we also set the
flags automatically in our callback setter functions to
help ensure that a) people use them and b) flags & callbacks
are kept in sync.
This is not currently useful but future commits will make further
changes concerning automatic setting of flags and event delivery
that makes this structure necessary.
When receiving an eol, the object will be NULL and we're not handling
this correctly. Thanks to Drew Ogle <dantealiegri@gmail.com> for
reporting this and providing a patch.
For source and sinks, we allow both the index and name to be invalid
while introspecting (the default source/sink will be used). This
modifies the assert appropriately. Thanks to Drew Ogle
<dantealiegri@gmail.com> for reporting and providing a patch.
This picks sane defaults for the sample spec used (32 kHz, mono) and
preprocessing (on by default). This should make it unncessary to provide
additional parameters in the default desktop case.
The main exception would be decreasing the sample rate for hardware with
limited processing power (can bring it down to 16 or 8 kHz).
Those macros cover almost all functionality of attributes.m4 that was
used, so that file can be removed.
The CC_NOUNDEFINED macro is used directly in configure.ac.
This is a workaround - these usb headsets have one output volume
control only, labeled "Speaker". This causes the default profile
set to not control the volume at all, which is a bug. Workaround
that by creating a separate profile set.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Cross-compiling for win32 failed after the previous #include removal.
Somehow when building for Linux the struct timeval definition got picked up elsewhere.
When a TCP socket is created the size of the send buffer (SO_SNDBUF) used is
determined by the OS, using the net.ipv4.tcp_wmem sysctl parameter. Previously
a call to setsockopt set the buffer size to a value that was too small, and
that in some cases could result in underruns and choppy playback. This
setsockopt call has now been removed so that the value determined by the OS
is used unchanged.
Note that the value used for the send buffer size is the 2nd value in
net.ipv4.tcp_wmem, e.g. if this is set to "4096 65536 8388608" the send buffer
size is set to 65536.
This adds code to specifically support legacy entries.
I kept this code in a separate commit so that it can be (relatively)
easily removed at some point in the future.
This simply exposes the formats that a device supports
via a simple protocol extension that will allow clients
to setup what a connected receiver supports format wise.
This has the advantage of allowing versioned updates in the future,
thus allowing us to be more user friendly going forward (as opposed
to just ignoring entries from old versions).
The primary motivation for this, however, is to allow variable length
storage in each entry which will be needed for upcoming work.
At present this commit will ignore any legacy entries but support
for reading and subsequently converting legacy entries will be added
shortly.
After the rework to the add pa_sink_input_new_data_set_sink() (and
the source equiv) calling with a NULL sink object will hit an assert.
This caused crashes with the esd protocol and there was the potential
(albeit unlikely) for a crash when creating a sink input without any
sinks available (module-always-sink mitigates this risk but it's still
a potential crasher).
This was added to ensure symmetry between playback and recording streams
code, but in reality this makes little sense practically speaking and thus
it is removed.
The previous logic in ade0a6f884
does not work with for input volumes.
This was discussed on the mailing list:
https://tango.0pointer.de/pipermail/pulseaudio-discuss/2011-May/010091.html
This approach can introduce a problem when setting the volumes
for sources. What follows is Tanu Kaskinen's analysis:
[quote]
I'll quote the log:
D: protocol-native.c: Client pavucontrol changes volume of source alsa_input.pci-0000_00_1b.0.analog-stereo.
D: alsa-source.c: Requested volume: 0: 45% 1: 45%
D: alsa-source.c: in dB: 0: -20.71 dB 1: -20.71 dB
D: alsa-source.c: Got hardware volume: 0: 45% 1: 45%
D: alsa-source.c: in dB: 0: -21.00 dB 1: -21.00 dB
D: alsa-source.c: Calculated software volume: 0: 101% 1: 101% (accurate-enough=no)
D: alsa-source.c: in dB: 0: 0.29 dB 1: 0.29 dB
D: source.c: Volume going up to 29273 at 270475970821
D: source.c: Volume change to 29273 at 270475970821 was written 34 usec late
D: alsa-source.c: Written HW volume did not match with the request: 0: 45% 1: 45% (request) != 0: 42% 1: 42%
D: alsa-source.c: in dB: 0: -21.00 dB 1: -21.00 dB (request) != 0: -22.50 dB 1: -22.50 dB
Looking at the last line, the requested volume seems to hit exactly the
right step (-21.00dB), but for some reason Alsa decides to choose
something else. I'm pretty sure that this happens because of rounding
errors. In the first phase we ask Alsa what dB value we should set, and
it returns -21.00 dB. The value is given as a long int, but we convert
that to pa_cvolume. Then when we set the volume, we convert the
pa_cvolume value back to a long integer. At this point I believe it gets
converted to -2101. This is not visible in the debug message for some
reason - the rounding algorithm must be different from what was used
with the pa_cvolume -> long conversion.
[/quote]
The commit after this contains a patch that addresses this issue.
This piggy backs onto the previous changes for protocol 22 and
thus does not bump the version. This and the previous commits should be
seen as mostly atomic. Apologies for any bisecting issues this causes
(although I would expect these to be minimal)
The "rm" basm constraint doesn't work with my version of gcc (4.5.2),
not even in a simple example. Since we usually only have 5 registers
available on i386, force it to be memory on that architecture.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Passing a NULL-terminated array of pa_format_info pointers is a bit
unwieldy for clients. Instead of this, let's pass in an array of
pointers and the number of elements in the array.
We were calculating new latency based on the latency set on the old
sink/source, rather than the actual latency requested by the client.
Over a series of moves, this will lead the latency being ~halved each
time, resulting in an eventual rewind flood from a latency that cannot
be handled.
We were using the block size in bytes instead of samples, which meant
preprocessing was broken. This fix makes a large-ish difference in the
quality of echo-cancellation with speex.
The smoother is paused on initialization and resumed when the sink
state is set to running. Otherwise, early latency estimates are
too low since there is some delay between module initialization and
entering the running state.
After the smoother is initially resumed, it is paused when the sink
state is not running. The previous behavior was to pause only when
the sink enters suspended state, however, this would lead to large
errors in latency estimates after the sink has been idle for some
time.
Add a variable to track whether the actual volume is set or not.
Suppose this:
min volume: -126 max volume: 0
then when user wants to set some constant volume to -10, it would fail.
While the alsa values are typically positive, some values are "funky"
and have negative values. It is desirable to fix this at the alsa
level so that the numbers are positive, but it's not technically
invalid, and thus we have to support it.
Discussed here:
http://thread.gmane.org/gmane.comp.audio.pulseaudio.general/9832
and
http://thread.gmane.org/gmane.linux.alsa.devel/85459
If module initialisation fails, the speex done() function might try to
free a value that's not been allocated yet. Adding protection for this
condition.
The echo suppress attenuation value was being incorrectly modified.
Fixed and added 2 arguments to change the attenuation of the residual
echo filter. Default values of the speex preprocessor will be used when
omitted.
This allows the selective enabling of speex' preprocessing algorithms
before running the echo-canceller -- for now this includes automatic
gain control, noise suppression and echo suppression. It's all off by
default for now, though at some point in the near future we might want
to enable at least denoising by default.
The denoising works pretty well, though we might want to add a way to
tweak the noise-suppression knob that libspeex provides.
The AGC option is just a stop-gap -- we need a real AGC mechanism that
tweaks the source volume rather than doing this in software.
The speex documentation mentions VAD and dereverb, but it appears that
these are not complete yet.
We don't do all this in a separate module from module-echo-cancel to
avoid the overhead of adding another virtual source. It makes more sense
to make a separate virtual source module that can be used for cases
where preprocessing is useful but AEC is not (for e.g. noise suppression
for fan noise in a recording application).
Another reason to keep this integrated with the AEC module is that the
echo suppression bits use the speex echo canceller state. This does leak
some information about the AEC implementation into module-echo-cancel,
but this is unavoidable.
Linking libpulse with gold or when using ld --no-add-needed fails
as libpulse uses dbus methods directly but isn't explicitly linked to it.
So link to it when needed :)
The proplist used may never be freed if an error condition was found with
CHECK_VALIDITY macro and the formats idxset was never freed regardless
of error state.
This change fixes adds a new maco CHECK_VALIDITY_GOTO() which allows
for cleanup to be done before returning.
In a setup with one or more filter sinks or sources there is always at
least one stream existing. In such a situation normal mempool
vacuuming never happens. This patch causes suspend-on-idle module to
vacuum memory when ever it notices that all sinks and sources are
suspended. The behavior can be enabled with a module parameter.
This quite is an old patch. It was added to N900 to avoid unnecessary
wake-ups when the phone is in power save mode (= blank screen and
no user interaction). In this situation if the user had a browser
window with flash animation open pulseaudio kept waking up every
10 seconds, causing a severe hit to use times.
Anyway I do not see any reason to send timing updates if the sink or
source where the stream is connected to is suspended.
This has been broken since c376ac5920 when run
without any arguments. Passing in -v (verbose) caused the test to work fine.
I think this oversight is just a thinko in the original work but it obviously
broke 'make check' and thus distcheck.
Also fix a couple compiler warnings.
This updates the tunnel module for protocol version >= 19.
module-tunnel-sink does not proxy server-side passthrough support (yet).
This would require a few more changes, namely keeping track of what
formats are available and if any other sink inputs are connected on the
server-side.
This replaces the simple string used by pa_format_info's proplist with a
JSON string (accessed via new API only). This allows us to express lists
and ranges more cleanly, and embed type information for future
extensibility.
We use json-c for JSON parsing. This is a lightweight depdency (32 KB on
my system) and avoids the hassle of having to reinvent a JSON parser.
Also included is a test which verifies functionality and is
valgrind-clean.
When the sink format changes and we kill the stream, clients need a way
to know (a) what device they should reconnect to, and (b) what the
stream running time was when the stream got killed (pa_stream_get_time()
won't work after the stream has been killed). This adds these two bits
of information in the event callback's proplist parameter.
IEC61937-encapsulated E-AC3 frames contain 6 audio blocks per substream,
which corresponds to 1536 samples contained a 24576-byte frame. To cope
with this, we maintain the s16le stereo sample spec, but quadruple the
sample rate so that the conversion remains accurate.
This event is emitted if the sink-input could not be moved to a new sink
because it doesn't support the format of the sink-input. Clients can
reconnect their stream with a different format if they wish or
gracefully exit.
The assertion message is misleading, since the passthrough connection
can fail for reasons the client has no control over (like other sink
inputs being connected).
This was just introduced for debugging and should not have been in the
final commit. Won't make a difference at the moment since this function
is used as a pointer, but removing this in case we change this in the
future.
PA_ALIGNED can't always guarantee that the alignment we want (the GCC
man page suggests that the linker might not be able to meet the
alignment requirements we desire). Instead, we now allocate some extra
memory and guaratee that the alignment we require is met.
(Based on Colin's review) We mark modules as being autoloaded so that
they can handle this as a special case if needed (which is required by
module-echo-cancel for now). This inverts how things were done and makes
using these modules manually less error-prone.
Since we currently have two mechanisms to signal a passthrough
connection (non-PCM format or PA_SINK_INPUT_PASSTHROUGH flag), we move
all the related checks into functions and use those everywhere.
This makes things more consistent, and should we decide to get rid of
the flag, we only need to change pa_sink_input_*_is_passthrough()
accordingly.
When a passthrough sink-input is added, we need to reconfigure the
sink's sample rate since no resampling occurs. We revert to the original
rate when the passthrough sink-input is removed.
In most cases it is expected that clients cannot consume compressed
data from monitor sources, so we suspend the monitor source when the
sink goes into passthrough mode.
Eventually, when the extended API includes client notifications for
changed formats, we should emit a notification on the monitor so that
clients can decide what they want to do when this happens (disconnect or
consume the data anyway).
This will eventually be replaced by a hook to let clients know that the
stream has moved so that they can gracefully reconnect and renegotiate a
supported format.
We frequently need to free an idxset containing pa_format_infos, so
define an internal free function that can be used directly with this
(instead of defining it once-per-file).
This is just sync-playback.c modified to use the extended API. We need
something more sophisticated for properly testing the compressed
formats, but that is a non-trivial task in itself.
This removes the passthrough flag from sinks since we will drop
exclusively passthrough sinks in favour of providing a list of formats
supported by each sink. We can still determine whether a sink is in
passthrough mode by checking if any non-PCM streams are attached to it.
This is the beginning of work to support compressed formats natively in
PulseAudio. This adds a pa_stream_new_extended() that takes a format
structure, sends it to the server (=> protocol extension) and has the
server negotiate with the appropropriate sink to figure out what format
it should use.
This is work in progress, and works only with PCM streams. Actual
compressed format support in some sink needs to be implemented, and
extensive testing is required.
More details on how this is supposed to work is available at:
http://pulseaudio.org/wiki/PassthroughSupport
This adds a get_formats() vfunc for sinks to provide a list of formats
they can support. pa_sink_check_formats() can be used during or after
routing to determine what formats from a stream the sink can support.
(Based on Colin's review) We mark modules as being autoloaded so that
they can handle this as a special case if needed (which is required by
module-echo-cancel for now). This inverts how things were done and makes
using these modules manually less error-prone.
When using transport configured via Media API sample spec needs to be
updated since codec configuration may affect it when e.g. headset
configure a different frequency or number of channels from default.
Commit 65ef80b fixed building with xcb-util >= 0.3.8, but the reply is never
checked (possible SIGSEGV if the reply is NULL) nor freed (memory leak at each
call of the functions).
Also, remove include and dependencies on xcb-atom, as it was only meaningful
for xcb_atom_get() and STRING, and depend instead on xcb >= 1.6 for
XCB_ATOM_STRING.
If u->connections isn't empty when module-dbus-protocol is
unloaded, then connection_free() is called for the
remaining connections when the idxset is freed.
connection_free() tries to remove the connection from the
idxset, but that fails, because the item has already been
removed from the idxset in this scenario.
The problem is solved by not trying to remove the connection
from the idxset in connection_free(). Instead, whoever wants
to delete connections, has to remove the connection from the
idxset in addition to calling connection_free().
This patch adds a new update mode specifier that can be optionally
given in match rules after the regexp. Property list updates triggered
by the rule will honour the given mode. The two allowed modes are 'merge'
and 'replace', corresponding to PA_UPDATE_MERGE and PA_UPDATE_REPLACE
respectively. If omitted, the mode defaults to PA_UPDATE_MERGE, ie. to
the original behavior.
For example, to force 'media.role' to be overwritten with 'bar' for
streams matching foo you can use an entry like this:
foo replace "bar"
This will really overwrite media.role to bar even if it has already been
set to something else by the application.
Thanks to Krisztian Litkey for the original patch and the description
above. In addition to implementing the new feature, this patch fixes
a number of bugs in the parsing code.
It's possible that the memblockq of a sink input is rewound to a negative read
index if the sink input is moved between sinks shortly after its creation. When
this happens, pa_memblockq_peek() returns a memchunk whose 'memblock' field is
NULL and whose 'length' field indicates the length of the gap caused by the
negative read index. This will trigger an assert in play-memblockq.c.
If the memblockq had a silence memchunk, pa_memblockq_peek() would return
silence for the duration of the gap and the assert would be avoided. However,
this approach would prevent the sink input from being drained and is thus not
possible. Instead, we handle the aforementioned situation by dropping the gap
indicated by the 'length' field of the memchunk and by peeking the actual data
that comes after the gap.
This scenario seems to be quite rare in everyday use, but it causes a severe
bug in the handheld world. The assert can be triggered e.g. by loading two null
sinks, playing a sample from the cache to one of them and then moving the
created sink input between the two sinks. The rewinds done by the null sinks
seem to be quite long (I don't know if this is normal behaviour or something
fishy in module-null-sink).
See also:
6bd34156b1
virtual-sink: Fix a crash when moving the sink to a new master right after setup.
https://tango.0pointer.de/pipermail/pulseaudio-discuss/2011-February/009105.html
Reproduce:
This problem can be reproduced with the following script:
SAMPLE_PATH="/usr/share/sounds/alsa/"
SAMPLE="Front_Left"
pactl remove-sample $SAMPLE 2> /dev/null
pactl upload-sample $SAMPLE_PATH$SAMPLE.wav
mod1=`pactl load-module module-null-sink sink_name=null1`
mod2=`pactl load-module module-null-sink sink_name=null2`
pactl play-sample $SAMPLE null1
input=`pactl list | grep "Sink Input #" | tail -n 1 | cut -d# -f2`
echo "Sample $SAMPLE playing as Sink Input #$input"
pactl move-sink-input $input null2
pactl move-sink-input $input null1
pactl unload-module $mod1
pactl unload-module $mod2
With automaticl filter loading by module-filter-apply, setting the
virtual sink/source to have the "phone" intended role will break routing
when you first connect a phone stream to an ALSA device and then turn on
your Bluetooth headset. This happens because module-intended-roles
doesn't move a stream if it is already on a device that provides the
required role.
This patch introduces a "manual_load" parameter that is meant to be used
when not using module-filter-apply for loading the AEC module. If this
parameter is set, the virtual devices are given the "phone" role, else
we count on module-filter-heuristics to do the right thing.
This makes the core code in the filter-* modules generic enough to be
used on sources or sinks. We need special handling for modules that
introduce more than one sink (for now echo-cancel only).
This makes sure that we handle streams moving between sinks properly. To
do this, we change the way the filter.* properties are handled a little
bit.
Firstly, this splits up the "filter.apply" property into two properties
- "filter.want" and "filter.apply". "filter.apply" acts as before - it
bypasses module-filter-heuristics and directly tells module-filter-apply
what filters are to be applied.
"filter.want" is used to tell module-filter-heuristics what filters the
client wants. The module then decides whether to actually apply the
filter or not (for now, this makes sure we don't apply echo-cancellation
even if requested on phone sinks (where it is assumed AEC is taken care
of or is not required).
Next, we also make sure that we track whether the client set
"filter.apply" or module-filter-heuristics did - and in the latter case,
we recalculate "filter.apply" and then have module-filter-apply apply
the filter if required. This introduces some evil in the form of causing
the move_finish callback to possibly trigger another move, but we
protect for this case (with a property) to be doubly sure of not causing
an infinite loop.
This makes sure that we don't apply AEC on sinks that are already
connected to a "phone" device, the assumptiong being that anything
marked as such either doesn't have need it, or handles it itself.
When running two connect-stress tests at the same time the liklihood of >32 streams
per sink increases. All it takes is for an event sound to fire to trigger an abort of
the test.
This leaves just a little bit of wriggle room for a couple external streams.
Of course the overall problem is still there but this just makes it
slightly less likely without really affecting the test itself.
There were several memory leaks. In addition to those,
pa_dbus_protocol_add_interface() used a string from the
caller as a key to a hashmap, instead of a copy of the
string. This caused trouble when the caller freed the
string while the key was still in use in the hashmap.
Defining this macro on a global level is disadvantageous for other APIs,
and as we need it for clock_gettime() only on Mac OS X, define it
locally in pulsecore/core-rtclock.c only.
This value is passed on to the instances of module-coreaudio-device that
are loaded upon device detection. The value is purely optional, as the
device module will fall back to to its default if it's not given.
Adds an autoclean option (defaults to TRUE) that controls whether
module-filter-apply cleans up unused modules or not. This is useful in
cases where you know that a filter will be used often and thus can avoid
overhead from repeated module load/unload.
This makes the volume tests run in two loops and print the minimum,
maximum and standard deviation of readings from the inner loop. This
makes it easier to reason out performance drops (i.e. algorithmic
problems vs. other system issues such as processor contention).
For systems which have a fcntl() implementation, we can simplify the
code which determines whether a file selector is valid in pa_poll().
The old code, which is harder to read and more expensive, stays around
for all platforms we need to emulate poll() for using select(), and
which don't provide fcntl(). IOW, for Windows.
On Mac OS X, however, the detection for bad fds via more select() calls
doesn't work, resulting in hung main loops, so the patch fixes a real
bug there.
This brings more uniformity to arguments to match module-echo-cancel
(which needs both sink and source masters, hence the disambiguation).
This will allow other modules to load filters in a more uniform way
in the future without kludges to deal with variation in arguments.
When volume changes in bluetooth device PulseAudio volume is rounded
one too low, so if bluetooth headset changes volume and that volume
is immediately set again for bluetooth device, bluetooth step drifts
lower all the time. Volume is incremented by one in the conversion so
that we get right bluetooth step when re-applying volume.
Signed-off-by: Juho Hämäläinen <ext-juho.hamalainen@nokia.com>
This module implements a simply policy decision that any newly plugged
in devices should be used.
This is a reasonable approach and paprefs will be updated to allow for
this option to be turned on or off.
This is more or less a stop-gap solution. When priority lists are
implemented in the core, then policy modules may ultimately be
re-engineered to adjust the priority lists rather than doing any of
their own routing per-se.
Previously the userdata for the volume callbacks was saved to
pa_core.shared only once when loading module-bluetooth-device, and only when
the SCO over PCM feature was used. That breaks volume handling in cases where
the HSP profile is used without the SCO over PCM setup. Now the userdata is
set always when a sink or source is created, and removed when a sink or source
is removed.
The current implementation is totally bogus, it cast the over_sink
userdata to the bluetooth-device userdata... It was failing nicely
because the previous code had a gentle safe-guard in u->profile ==
PROFILE_HSP, and u->profile was just random.
There is no easy way to associate additional data to a sink or
source. Two solutions seems possible: looking up loaded modules and
check which one was handling the sink/source, or using pa_shared. I
went for the second solution.
This pulls a2dp-codecs.h from BlueZ which contains the capabilities
structures for SBC and MPEG. We currently have these manually added to
ipc.h, so pulling this header makes our files identical to upstream.
The check is inspired by a driver that returned higher dB limit from
snd_mixer_selem_get_playback_dB_range() than what _ask_playback_vol_dB()
returned at maximum integer volume.
This is pretty cosmetic change; there's no actual functionality added.
Previously the volume_writable information was available through the
pa_sink_input_is_volume_writable() function, but I find it cleaner to have a
real variable.
The sink input introspection variable name was also changed from
read_only_volume to volume_writable for consistency.
This update pulls in commit c495077c [1] to fix a build error.
commit c495077cf8a8c37afd90875ec5a5b16b294be15e
Author: Siarhei Siamashka <siarhei.siamashka@nokia.com>
Date: Tue Mar 29 01:57:39 2011 +0300
sbc: better compatibility with ARM thumb/thumb2
ARM assembly optimizations fail to compile in thumb mode, but are fine
for thumb2. Update ifdefs in the code to make use of ARM assembly only
when it is safe and also make sure that no optimizations are missed
when compiling for thumb2.
The problem was reported by Paul Menzel:
https://tango.0pointer.de/pipermail/pulseaudio-discuss/2011-February/009022.html
This patch is tested with OpenEmbedded using `minimal-uclibc` for `MACHINE = "at91sam9260ek"`.
Note that changes to ipc.h from 8f3ef04b had to be manually reapplied.
[1] http://git.kernel.org/?p=bluetooth/bluez.git;a=commit;h=c495077cf8a8c37afd90875ec5a5b16b294be15e
Just picking up a crash report from Ubuntu, here's the result.
--
David Henningsson, Canonical Ltd.
http://launchpad.net/~diwic
From 934c52c79bb6faed56a64d6e15f9b285f687afee Mon Sep 17 00:00:00 2001
From: David Henningsson <david.henningsson@canonical.com>
Date: Mon, 28 Mar 2011 14:30:44 +0200
Subject: [PATCH] module-jack-sink/source: protect against null return in jack_get_ports
According to jack_get_ports documentation, it seems like returning NULL
is valid, and that it should be freed using jack_free.
Reported-by: Grayson Peddie
BugLink: http://bugs.launchpad.net/bugs/733424
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
On 64-bit systems LONG_MAX is greater than the largest possible value of a
uint32_t variable, which caused the compiler to warn about a comparison that is
always false. On 32-bit systems pa_atou() can return a value that will overflow
when assigned to e->volume_limit, which has type long, so the comparison was
necessary.
This dilemma is resolved by using pa_atol() instead of pa_atou().
Vala uses the name of the *.vapi file to determine the libraries to link
against. Since the pa_glib_mainloop_*() functions are in a separate
library (libpulse-mainloop-glib.so) the corresponding objects in the
Vala bindings have to be in a separate *.vapi file.
If you are compiling an app without the GLib integration you could use:
$ valac --pkg=libpulse test.vala
but if you do use GLib you can use:
$ valac --pkg=libpulse-mainloop-glib test.vala
(libpulse is a dep of the libpulse-mainloop-glib so no need to specify
it explicitly)
We should not call pa_core_ref() anywhere in the code. Doing so
will prevent proper daemon shutdown as the only call (in daemon/main.c)
to pa_core_unref() should always call free_core() and perform a normal
shutdown (i.e. unload all modules gracefully).
With Tanu's patch, the server no longer starts when a server is configured.
While this is sensible in most circumstances there is a corner case where
we still want to start.
In a typical X11 login, module-x11-publish will be loaded and will thus
set the PULSE_SERVER X11 property on the root window. This then hits the
check introduced in f1d1447e and exits. If PA had previously crashed
(thus leaving behind it's X11 properties) then this means that we will not
autospawn nor even allow ourselves to be started manually until
pax11publish -r is run to clear out the X11 properties. This is obviously
not desirable.
This patch introduces a more in-depth check of the server. If it looks like
a local unix domain socket, then we do not exit straight away and instead
probe further. This should not pose any problems with e.g. remote SSH
usage as the DBus Machine ID is used in the server string.
As spotted by Tanu Kaskinen:
The first process: daemon_pipe is not closed if the first fork() call
fails. Even if it doesn't fail, the first process never closes
daemon_pipe[0].
The second process: daemon_pipe[1] is not closed if anything fails
between the first and the second fork() call. Also, if the second fork
fails, then the finish section writes to daemon_pipe2[1], even though
only the third process should do that. Also, if anything fails between
the first and the second fork, then the second process never writes
anything to daemon_pipe[1]. I don't know what happens in the first
process in this case - does it get an error or does pa_loop_read() get
stuck.
The third process: No problems :)
0.4.9 errors out at compile time, and might as well bump to 0.4.11 since
that's the version being tested with and has been around for a while
now. Thanks to Paul Menzel <paulepanter@users.sourceforge.net> for
pointing this out.
1. PA uses Vala's "Posix" package (see line 23 of libpulse.vapi).
These dependencies have to be declared in the *.deps file.
2. Fix obvious copy/paste error.
3. Rename the parameter to match the C function. This simplifies
understanding what this parameter means.
4. According to the official documentation the "dev" parameter
for playback/capture streams on connection may be NULL and this
is the default. Change the method definition accordingly.
The previous commit intoduced a double fork which caused a more or less immediate
successful return prior to the hard work of actually starting a daemon.
This patch simply used pipe() to only signal our father when the daemon really
has finished starting.
It's perfectly normal for BlueZ to disable some endpoint types, so printing a
log message at error level isn't a good idea.
For facilitating an informative message in case some endpoint type is disabled,
the send_and_add_to_pending() function interface is also changed to be more
generic (the pa_bluetooth_device pointer is replaced with a void pointer).
Fix mistakes in:
commit 36e7f8a92e
Author: Colin Guthrie <cguthrie@mandriva.org>
Date: Sun Mar 20 14:09:14 2011 +0000
tunnel: Fix tunnel streams with recent servers
Prompted by bug #750. Thanks to Emil Renner Berthing for the original
patch.
This change makes it possible to configure an arbitrary constant volume for a
volume element in the path configuration, which is applied when the path is
selected. Note: this is only useful when the exact hardware and driver are
known beforehand.
Mostly warnings about unused stuff.
Furthermore, the first hunk is a fix for the change in 177948a6.
Finally, comment in AEC_dtd was translated and the code simplified slightly.
CC module_bluetooth_device_la-module-bluetooth-device.lo
modules/bluetooth/module-bluetooth-device.c: In function ‘a2dp_process_render’:
modules/bluetooth/module-bluetooth-device.c:1335:30: warning: pointer targets in passing argument 6 of ‘sbc_encode’
differ in signedness [-Wpointer-sign]
../src/modules/bluetooth/sbc/sbc.h:92:9: note: expected ‘ssize_t *’ but argument is of type ‘size_t *’
CC module_rygel_media_server_la-module-rygel-media-server.lo
modules/module-rygel-media-server.c:383:13: warning: ‘append_property_dict_entry_object_array’ defined but not used [-Wunused-function]
CC module_echo_cancel_la-adrian-aec.lo
modules/echo-cancel/adrian-aec.h:360:15: warning: ‘AEC_getambient’ defined but not used [-Wunused-function]
modules/echo-cancel/adrian-aec.h:368:14: warning: ‘AEC_setgain’ defined but not used [-Wunused-function]
modules/echo-cancel/adrian-aec.h:374:14: warning: ‘AEC_setaes’ defined but not used [-Wunused-function]
modules/echo-cancel/adrian-aec.h:377:16: warning: ‘AEC_max_dotp_xf_xf’ declared ‘static’ but never defined [-Wunused-function]
CC module_echo_cancel_la-module-echo-cancel.lo
modules/echo-cancel/module-echo-cancel.c: In function ‘time_callback’:
modules/echo-cancel/module-echo-cancel.c:266:12: warning: variable ‘fs’ set but not used [-Wunused-but-set-variable]
CC module-virtual-sink.lo
modules/module-virtual-sink.c: In function ‘sink_input_pop_cb’:
modules/module-virtual-sink.c:206:15: warning: variable ‘current_latency’ set but not used [-Wunused-but-set-variable]
pulsecore/core-util.c: In function ‘pa_hexstr’:
pulsecore/core-util.c:1858: warning: cannot optimize loop, the loop counter may overflow [-Wunsafe-loop-optimizations]
modules/alsa/alsa-mixer.c: In function ‘pa_alsa_decibel_fix_dump’:
modules/alsa/alsa-mixer.c:3678: warning: cannot optimize possibly infinite loops [-Wunsafe-loop-optimizations]
modules/alsa/alsa-mixer.c: In function ‘pa_alsa_path_set_new’:
modules/alsa/alsa-mixer.c:2640: warning: cannot optimize loop, the loop counter may overflow [-Wunsafe-loop-optimizations]
modules/module-default-device-restore.c: In function ‘load’:
modules/module-default-device-restore.c:67: warning: ignoring return value of ‘fgets’,
declared with attribute warn_unused_result [-Wunused-result]
modules/module-default-device-restore.c:88: warning: ignoring return value of ‘fgets’,
declared with attribute warn_unused_result [-Wunused-result]
pulsecore/authkey.c: In function ‘generate’:
pulsecore/authkey.c:58: warning: ignoring return value of ‘ftruncate’,
declared with attribute warn_unused_result [-Wunused-result]
pulsecore/core-util.c: In function ‘pa_make_secure_dir’:
pulsecore/core-util.c:261: warning: ignoring return value of ‘fchown’,
declared with attribute warn_unused_result [-Wunused-result]
The HAVE_CLOCK_GETTIME macro protects timespec and related functions, nothing of which is used in
pa_rtclock_from_wallclock. And silently just not converting was not the proper solution anyway.
Also add an assert in pulse/mainloop.c to report the integer overflow that was triggered by the wrong
pa_rtclock_from_wallclock. Without the assert, debugging was painful.
This patch enables logging of text debug messages (pa_log feature) into a file or a device driver.
Example : pulseaudio --log-target=file:./mylog.txt
(Minor tweaks by Colin + Arun)
This just makes some of the bluetooth stuff a little clearer and
standardises on two-tab indents as this seems most common.
Also added two headers to the bluetooth sbc source that were missing.
This fixes the checking of supported compiler flags and the following error message for svolume_mmx:
pulsecore/svolume_mmx.c:157:76: error: invalid use of a cast in a inline asm context requiring an l-value:
remove the cast or build with -fheinous-gnu-extensions
: "+r" (samples), "+r" (volumes), "+r" (length), "=D" ((pa_reg_x86)channel), "=&r" (temp)
~~~~~~~~~~~~^~~~~~~
The optimized filter gets enabled when the code is compiled
with -mcpu=/-march options set to target the processors which
support ARMv6 instructions. This code is also disabled when
NEON is used (which is a lot better alternative). For additional
safety ARM EABI is required and thumb mode should not be used.
Benchmarks from ARM11:
== 8 subbands ==
$ time ./sbcenc -b53 -s8 -j test.au > /dev/null
real 0m 35.65s
user 0m 34.17s
sys 0m 1.28s
$ time ./sbcenc.armv6 -b53 -s8 -j test.au > /dev/null
real 0m 17.29s
user 0m 15.47s
sys 0m 0.67s
== 4 subbands ==
$ time ./sbcenc -b53 -s4 -j test.au > /dev/null
real 0m 25.28s
user 0m 23.76s
sys 0m 1.32s
$ time ./sbcenc.armv6 -b53 -s4 -j test.au > /dev/null
real 0m 18.64s
user 0m 15.78s
sys 0m 2.22s
In the case of scale factors calculation optimizations, the inline
assembly code has instructions which update flags register, but
"cc" was not mentioned in the clobber list. When optimizing code,
gcc theoretically is allowed to do a comparison before the inline
assembly block, and a conditional branch after it which would lead
to a problem if the flags register gets clobbered. While this is
apparently not happening in practice with the current versions of
gcc, the clobber list needs to be corrected.
Regarding the other inline assembly blocks. While most likely it
is actually unnecessary based on quick review, "cc" is also added
there to the clobber list because it should have no impact on
performance in practice. It's kind of cargo cult, but relieves
us from the need to track the potential updates of flags register
in all these places.
By using SBC_ALWAYS_INLINE trick, the implementation of 'sbc_calculate_bits'
function is split into two branches, each having 'subband' variable value
known at compile time. It helps the compiler to generate more optimal code
by saving at least one extra register, and also provides more obvious
opportunities for loops unrolling.
Benchmarked on ARM Cortex-A8:
== Before: ==
$ time ./sbcenc -b53 -s8 -j test.au > /dev/null
real 0m3.989s
user 0m3.602s
sys 0m0.391s
samples % image name symbol name
26057 32.6128 sbcenc sbc_pack_frame
20003 25.0357 sbcenc sbc_analyze_4b_8s_neon
14220 17.7977 sbcenc sbc_calculate_bits
8498 10.6361 no-vmlinux /no-vmlinux
5300 6.6335 sbcenc sbc_calc_scalefactors_j_neon
3235 4.0489 sbcenc sbc_enc_process_input_8s_be_neon
2172 2.7185 sbcenc sbc_encode
== After: ==
$ time ./sbcenc -b53 -s8 -j test.au > /dev/null
real 0m3.652s
user 0m3.195s
sys 0m0.445s
samples % image name symbol name
26207 36.0095 sbcenc sbc_pack_frame
19820 27.2335 sbcenc sbc_analyze_4b_8s_neon
8629 11.8566 no-vmlinux /no-vmlinux
6988 9.6018 sbcenc sbc_calculate_bits
5094 6.9994 sbcenc sbc_calc_scalefactors_j_neon
3351 4.6044 sbcenc sbc_enc_process_input_8s_be_neon
2182 2.9982 sbcenc sbc_encode
Previous variant was basically derived from C and MMX implementations.
Now new variant makes use of 'vmax' instruction, which is available in
NEON and can do this job faster. The same method for calculating scale
factors is also used in 'sbc_calc_scalefactors_j_neon'.
Benchmarked without joint stereo on ARM Cortex-A8:
== Before: ==
$ time ./sbcenc -b53 -s8 test.au > /dev/null
real 0m3.851s
user 0m3.375s
sys 0m0.469s
samples % image name symbol name
26260 34.2672 sbcenc sbc_pack_frame
20013 26.1154 sbcenc sbc_analyze_4b_8s_neon
13796 18.0027 sbcenc sbc_calculate_bits
8388 10.9457 no-vmlinux /no-vmlinux
3229 4.2136 sbcenc sbc_enc_process_input_8s_be_neon
2408 3.1422 sbcenc sbc_calc_scalefactors_neon
2093 2.7312 sbcenc sbc_encode
== After: ==
$ time ./sbcenc -b53 -s8 test.au > /dev/null
real 0m3.796s
user 0m3.344s
sys 0m0.438s
samples % image name symbol name
26582 34.8726 sbcenc sbc_pack_frame
20032 26.2797 sbcenc sbc_analyze_4b_8s_neon
13808 18.1146 sbcenc sbc_calculate_bits
8374 10.9858 no-vmlinux /no-vmlinux
3187 4.1810 sbcenc sbc_enc_process_input_8s_be_neon
2027 2.6592 sbcenc sbc_encode
1766 2.3168 sbcenc sbc_calc_scalefactors_neon
The code for scale factors calculation with joint stereo support has
been moved to a separate function. It can get platform-specific
SIMD optimizations later for best possible performance.
But even this change in C code improves performance because of the
use of __builtin_clz() instead of loops similar to what was done
to sbc_calc_scalefactors earlier. Also technically it does loop
unrolling by processing two channels at once, which might be either
good or bad for performance (if the registers pressure is increased
and more data is spilled to memory). But the benchmark from 32-bit
x86 system (pentium-m) shows that it got clearly faster:
$ time ./sbcenc.old -b53 -s8 -j test.au > /dev/null
real 0m1.868s
user 0m1.808s
sys 0m0.048s
$ time ./sbcenc.new -b53 -s8 -j test.au > /dev/null
real 0m1.742s
user 0m1.668s
sys 0m0.064s
This prevents overflows and audible artefacts for the audio files which
originally had loudness maximized. Music from audio CD disks is an
example of such files, see http://en.wikipedia.org/wiki/Loudness_war
Buffer position in X array was not always 16-bytes aligned.
Strict 16-byte alignment is strictly required for powerpc altivec
simd optimizations because altivec does not have support for
unaligned vector loads at all.
NREC stands for Noise Reduction and Echo Cancelation, it can be changed
at any point by the headset.
When set to "1" indicates that those algorithms shall be enabled by
default and "0" means the headset probably have them active so they
should be disabled in PA side.
Use minimum bitpool configured to get the maximum block_size possible,
also remove checks for how much has been written when decoding sbc frames
since the block size may change due to bitpool changes.
When audio skips it could be that there is some bandwidth limitation in
the link e.g. headset doesn't support EDR (< 2.0), and by reducing
the bitpool it may find a better rate that either prevent the skips
completely or at least reduce them.
Acquire now return input and output MTU of the file descriptor so it is
no longer necessary to get those after acquiring the fd, which less round
trips and faster response time when switching profiles.
The ORCC stage does this anyway (and this was buggy anyway as it had an extra
'src/' prefix so never worked properly when it was introduced in d6cdd80).
Thanks to Maarten Bosmans for pointing it out.
Although by "cork" I really mean "cork+mute" as that's what the module
does.
If e.g. Rhythmbox is paused when a phone call comes in, the current
stream state will be corked and thus we should not track it for future
uncorking when the phone call ends.
Likewise if the stream is just muted (manually) we will not take any
action either when the phone stream is seen first, nor when it
disappears.
Also add some additional debug messages.
Without this, p->max_dB could never be less than 0 dB, because the loop at the
end of pa_alsa_path_probe() would reset p->max_dB to 0 as soon as the loop
encountered a channel that wasn't touched by any element.
There was a similar issue for p->min_dB too (it could never be more than 0 dB),
which is also fixed by this patch.
This feature is mainly useful in embedded systems that have built-in speakers.
In such situations the full audio path is known beforehand, so it's possible to
know what is the maximum sensible volume, and any higher volume can be
disabled.
The volume limit is set in path configuration files in the [Element] section,
using option "volume-limit". The value is the desired maximum volume step of
the volume element.
This adds volume scaling for 1- and 2-channel software volume scaling
using Orc. While testing the MMX and SSE backends on a Core2, I see an
~2x performance benefit over the hand-rolled MMX and SSE code. Since I
haven't been able to test on other architectures, the Orc code is only
used when MMX/SSE* is present. This can be changed in the future after
testing on AMD and ARM machines.
Somewhere in the history of the MMX tests, the number of channels was
changed from 1 to 2, but the number of samples was not increased to make
it even (multiple of the frame size).
It seems git managed to mess up a git-am with a patch from
David which moved where this function was called element_probe
to within itself (recursive which could normally lead to an
infinite loop, but as it was now never called from anywhere else,
this didn't happen).
Thank you to Maarten for spotting and following up the issue.
This value is not a technical upper limit, it's just a 'sensible'
value that is not crazy high, but also allows software amplification
above 0dB (aka 100%) for very quiet audio sources.
We recommend that a comprehensive volume control UI should allow
users to set volumes up to this limit, although of course should
deal gracefully if the user has set the volume even higher than this
without resulting in a feedback loop that effectively limits the
upper volume.
The value chosen is +11dB. This was selected somewhat subjectively
and is very similar to the current 150% that gnome-volume-control
uses (which is ~+10.57dB).
On the plus side, we now recommend that everyone allows
'Volumes up to 11' which is pretty awesome.
http://en.wikipedia.org/wiki/Up_to_elevenhttps://tango.0pointer.de/pipermail/pulseaudio-discuss/2010-April/006945.htmlhttps://tango.0pointer.de/pipermail/pulseaudio-discuss/2010-April/006950.html
This is not set by pax11publish, but module-x11-publish does so this tool
should tidy that up. It is only removed when passing -r and is
ignored when actually setting up it's own properties from the conf
files/guesswork.
This reverts commit 95a98fe6f2.
Conflicts:
src/Makefile.am
src/pulsecore/envelope.c
src/pulsecore/envelope.h
src/tests/envelope-test.c
(part of a patch series removing all ramping code)
If the virtual sink is moved to a new master right after it has been created,
then the virtual sink input's memblockq can be rewound to a negative read
index. The data written prior to the move starts from index zero, so after the
rewind there's a bit of silence. If the memblockq doesn't have a silence
memchunk set, then pa_memblockq_peek() will return zero in such case, and the
returned memchunk's memblock pointer will be NULL.
That scenario wasn't taken into account in the implementation of
sink_input_pop_cb. Setting a silence memchunk for the memblockq solves this
problem, because pa_memblock_peek() will now return a valid memblock if the
read index happens to point to a hole in the memblockq.
I believe this isn't the best possible solution, though. It doesn't really make
sense to rewind the sink input's memblockq beyond index 0 in the first place,
because now when the stream starts to play to the new master sink, there's some
unnecessary silence before the actual data starts. This is a small problem,
though, and I don't grok the rewinding system well enough to know how to fix
this issue properly.
I went through all files that call pa_memblockq_peek() to see if there are more
similar bugs. play-memblockq.c was the only one that looked to me like it might
be broken in the same way. I didn't try reproducing the bug with
play-memblockq.c, though, so I just added a FIXME comment there.
When we have a filter sink that does some processing, currently the
benefits of the flat volume feature are not really available. That's
because if you have a music player that is connected to the filter sink,
the hardware sink doesn't have any idea of the music player's stream
volume.
This problem is solved by this "volume sharing" feature. The volume
sharing feature works so that the filter sinks that want to avoid the
previously described problem declare that they don't want to have
independent volume, but they follow the master sink volume instead.
The PA_SINK_SHARE_VOLUME_WITH_MASTER sink flag is used for that
declaration. Then the volume logic is changed so that the hardware
sink calculates its real volume using also the streams connected to the
filter sink in addition to the streams that are connected directly to
the hardware sink. Basically we're trying to create an illusion that
from volume point of view all streams are connected directly to the
hardware sink.
For that illusion to work, the volumes of the filter sinks and their
virtual streams have to be managed carefully according to a set of
rules:
If a filter sink follows the hardware sink volume, then the filter sink's
* reference_volume always equals the hw sink's reference_volume
* real_volume always equals the hw sink's real_volume
* soft_volume is always 0dB (ie. no soft volume)
If a filter sink doesn't follow the hardware sink volume, then the filter
sink's
* reference_volume can be whatever (completely independent from the hw sink)
* real_volume always equals reference_volume
* soft_volume always equals real_volume (and reference_volume)
If a filter sink follows the hardware sink volume, and the hardware sink
supports flat volume, then the filter sink's virtual stream's
* volume always equals the hw sink's real_volume
* reference_ratio is calculated normally from the stream volume and the hw
sink's reference_volume
* real_ratio always equals 0dB (follows from the first point)
* soft_volume always equals volume_factor (follows from the previous point)
If a filter sink follows the hardware sink volume, and the hardware sink
doesn't support flat volume, then the filter sink's virtual stream's
* volume is always 0dB
* reference_ratio is always 0dB
* real_ratio is always 0dB
* soft_volume always equals volume_factor
If a filter sink doesn't follow the hardware sink volume, then the filter
sink's virtual stream is handled as a regular stream.
Since the volumes of the virtual streams are controlled by a set of rules,
the user is not allowed to change the virtual streams' volumes. It would
probably also make sense to forbid changing the filter sinks' volume, but
that's not strictly necessary, and currently changing a filter sink's volume
changes actually the hardware sink's volume, and from there it propagates to
all filter sinks ("funny" effects are expected when adjusting a single
channel in cases where all sinks don't have the same channel maps).
This patch is based on the work of Marc-André Lureau, who did the
initial implementation for Pulseaudio 0.9.15.
Make sure that mic and line (with common names) use the specific
path instead of the analog-input one.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Now you can add required-any to elements in a path and the path
will be valid as long as at least one of the elements are present.
Also you can have required, required-any and required-absent in
element options, causing a path to be unsupported if an option is
(not) present (simplified example: to skip line in path if
"Capture source" doesn't have a "Line In" option).
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Add front mic, rear mic, and docking line-in. These are likely to be
present on modern hda chips, for reference see
linux-2.6/sound/pci/hda/hda_codec.c:hda_get_input_pin_label
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
When creating synthesized paths, pa_alsa_path_set_new() created duplicate
elements for each path, and one of the duplicate elements would be marked as
required absent. That made path probing fail. While debugging this, I noticed
also that pa_alsa_path_synthesize() didn't initialize p->last_element properly.
This will allow modules to know when a card profile has changed
and take appropriate action. This might prove useful when developing
UCM so that the appropriate verb can be set.
This allows modules to know when certain ports are changed.
This will allow e.g. a filter module (or LADSAP) to only load
when a certain port is used on the device (e.g. to only filter
headphones and not normal speakers).
(Comment from Colin Guthrie: This may also have use in UCM)
There are two known cases where read-only or non-existing sink input volume is
relevant: passthrough streams and the planned volume sharing logic.
Passthrough streams don't have volume at all, and the volume sharing logic
requires read-only sink input volume. This commit is primarily working towards
the volume sharing feature, but support for non-existing sink input volume is
also added, because it is so closely related to read-only volume.
Some unrelated refactoring in iface-stream.c creeped into this commit too (new
function: stream_to_string()).
This change doesn't add any functionality in itself, but it will be useful in
the future for operating on chains of sinks or sources that are piggy-backing
on each other.
For example, the PA_PROP_DEVICE_MASTER_DEVICE property could
be handled in the core so that each virtual device doesn't have to maintain it
separately. By using the origin_sink and destination_source pointers the core
is able to see at stream creation time that the stream is created by a virtual
device, and then update that device's property list using the name of the
master device that the stream is being connected to. The same thing can be done
also when the stream is being moved from a device to another, in which case the
_MASTER_DEVICE property needs updating.
Using uClibc
AM_LIBADD = $(PTHREAD_LIBS) $(INTLLIBS)
is not empty because `$(INTLLIBS)` is set to `-lintl`. This uncovered a missing space in commit ef0cc745 which resulted in the following build error.
[…]
CC libbluetooth_ipc_la-ipc.lo
modules/bluetooth/ipc.c: In function 'bt_audio_service_open':
modules/bluetooth/ipc.c:65:26: warning: dereferencing type-punned pointer might break strict-aliasing rules [-Wstrict-aliasing]
modules/bluetooth/ipc.c: In function 'bt_audio_service_get_data_fd':
modules/bluetooth/ipc.c:110:13: warning: cast increases required alignment of target type [-Wcast-align]
modules/bluetooth/ipc.c:110:4: warning: dereferencing type-punned pointer will break strict-aliasing rules [-Wstrict-aliasing]
make[3]: *** No rule to make target `-lintl', needed by `libbluetooth-ipc.la'. Stop.
make[3]: Leaving directory `/oe/build-minimal-uclibc/minimal-uclibc-dev/work/armv7a-oe-linux-uclibceabi/pulseaudio-0.9.22-r11.0+gitr0+051d82133f0ae6a57bf66fd200bc8e3591a7d5ca/git/src'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/oe/build-minimal-uclibc/minimal-uclibc-dev/work/armv7a-oe-linux-uclibceabi/pulseaudio-0.9.22-r11.0+gitr0+051d82133f0ae6a57bf66fd200bc8e3591a7d5ca/git/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/oe/build-minimal-uclibc/minimal-uclibc-dev/work/armv7a-oe-linux-uclibceabi/pulseaudio-0.9.22-r11.0+gitr0+051d82133f0ae6a57bf66fd200bc8e3591a7d5ca/git'
make: *** [all] Error 2
This patch is taken from OpenEmbedded where it has been present since 2009 for PulseAudio 0.9.15 and greater [1].
[1] http://git.0pointer.de/?p=pulseaudio.git;a=commit;h=ef0cc74567b3bb98378c17f6a523bf18ba132ed7
[2] http://cgit.openembedded.org/cgit.cgi/openembedded/commit/?id=7c33f2e906a20e139d53b4f2d8fbc2773a4725b3
Signed-off-by: Henning Heinold <heinold@inf.fu-berlin.de>
Signed-off-by: Paul Menzel <paulepanter@users.sourceforge.net>
CC: Henning Heinold <heinold@inf.fu-berlin.de>
The check whether POSIX socket.h or WIN32 winsock2.h must be included can be
made centrally. The downside is that some functionality of e.g. arpa/inet.h is
also implemented in winsock.h, so that some files that don't use socket
functions, but do use inet.h functions, must also include pulsecore/socket.h.
(as well as arpa/inet.h)
Instead <pulsecore/poll.h> should be included. That file includes poll.h on
platform where it is appropriate. Also remove some unnecessary <ioctl.h>
includes.
In theory putenv could be used to handle freeing of strings yourself, but this
was not done in PulseAudio. That leaves no advantages in using putenv. With
setenv you're at the mercy of the implementation whether the strings leak, but
at least that is better then a certain leak, as it was before.
This patch reflects a new capability that Lennart was wishing
for. Wish granted...
Re-submitting it now that alsa-lib 1.0.24
provides additional entry points to disable period
wakeups in timer-scheduling mode if hardware can
work without it (HDAudio, oxygen and Intel SST).
Example with standard playback on HDAudio output
Before change:
Top causes for wakeups:
3.8% ( 5.4) [hda_intel] <interrupt>
2.8% ( 4.0) alsa-sink
After change:
Top causes for wakeups:
2.3% ( 3.0) alsa-sink
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@intel.com>
This greatly simplifies the Orc-related make rules. The old system of
distributing generated files is gone, which means that anyone who wants
to build with Orc support enabled needs to have the orcc compiler
installed (presumably the orc 'devel' package in most distros).
The CFLAGS and LIBS of libsamplerate, libspeex and libsndfile are added to the pulse libraries depending on those libraries.
Also src/pulse/version.h is generated by configure, so it does not need to go in BUILT_SOURCES.
(Slightly tweaked by Colin Guthrie to fix a merge problem relating to BUILT_SOURCES)
This will make it possible to remove the empty Makefile.am files.
- module-...-symdef.h files are all generated in src/modules, instead of in the subdir of the module.
- The default inclusion of src/modules subdirs in AM_CFLAGS can be removed, where necessary (raop) the subdir is
included in the specific CFLAGS.
- The src/daemon and src/modules directories are always created on make, to facilitate out of tree builds.
- AM silent rules are used for the generation of symdef files by m4.
- For echo-cancel, keep the build dir include for now (and mkdir it) although limit it to just the echo-cancel
module's CFLAGS (Colin Guthrie)
If many small blocks are in queue, handle_seek is being called
for every one of them, sometimes causing a rewind. Delay the
call until all blocks are handled, then call handle_seek only
once.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Allow a message in the queue to perform both a seek and a post data.
For clients that do not use PA_SEEK_RELATIVE (e g gstreamer), this
cuts the message count - and sometimes even the rewinds - in half.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
When logging a suppression message do so on the same log level as the
suppressed messages.
Cherry picked by Colin Guthrie from ec5a785712
with a couple of additional changes due to extra limiting in master
that was not present in stable-queue.
The same logic is applied to the sample rate adjustments in module-rtp-recv,
module-loopback and module-combine:
- Each time an adjustment is made, the new rate can differ at most 2‰ from the
old rate. Such a step is equal to 3.5 cents (a cent is 1/100th of a
semitone) and as 5 cents is generally considered the smallest observable
difference in pitch, this results in inaudible adjustments.
- The sample rate of the stream can only differ from the rate of the
corresponding sink by 25%. As these adjustments are meant to account for
very small clock drifts, any large deviation from the base rate suggests
something is seriously wrong.
- If the calculated rate is within 20Hz of the base rate, set it to the base
rate. This saves CPU because no resampling is necessary.
This prevents the smoother attached to the stream clock from being
updated while the stream is corked, which in turn ensures that once
corking is completed, pa_stream_get_time() always returns the same value
until the stream is uncorked - i.e., the clock does not advance when the
client believes that it will not.
The actual call to pa_smoother_put() happens on things like stream
suspend/unsuspend, which trigger timing updates. This changes the
smoother coefficients, which means that a call to pa_smoother_get() for
the same value of 'x' can return different values before and after a
timing update.
Previously this argument passed to configure only worked if --disable-hal and
--disable-bluez was also passed which wasn't immediately obvious to the
untrained compiler.
This change simply makes --disable-dbus disable the other two as well
and errors out of specific, incompatible --enable/--disable flags
are provided.
The summary table is also adjusted and intended to try and show the
dependency relationship a little.
Users have often come to me when their build doesn't work and typically
this is when they do not have dev headers for DBUS and udev installed
when building. Put in some specific warnings about these optional,
but critical, elements.
This will likely display this message on platforms where they are not
available, so patches welcome to hide them in a semi-intelligent way.
Therefore, we must reflect this in the default.pa. Several users
have reported an error with consolekit when starting a self-built PA
due to the default config not working properly. This works around the
issue but we should include a warning on configure about the lack of
DBUS and udev support as this is a common mistake.
These functions are used in OSS programs where the "flags" parameter for
open() is not a build-time constant and the build has _FORTIFY_SOURCE
enabled.
The arguments sink_dont_move and source_dont_move have been added to toggle
module automatic unloading when the sink or source were no longer
available, rather than just moving them to the next available
sink/source (via rescue streams).
Reviewed and tweaked by Colin Guthrie.
A2DP spec allow bitpool changes midstream which is why sbc configuration
has a range of values for bitpool that the encoder can use and decoder
must support.
Bitpool changes do not affect the state of encoder/decoder so they don't
need to be reinitialize when this happens, so the impact is fairly small,
what it does change is the frame length so encoders may change the
bitpool to use the link more efficiently.
Based on comments from Arun Raghavan, make clear that we're referring
to the libspeex API documentation for the speex- options. Also correct
the quality range (was incorrectly 0..9 but is 0..10 now) and clean up
the line length.
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
In the file src/pulsecore/random.c
I found that the log information "Failed to get proper entropy. Falling
back to seeding with current time." would never be printed.
This change corrects the issue.
A Debian user seemed confused that the man page for pulse-daemon.conf
does not refer to the libspeex API Resampler for use in PulseAudio's
*-src. Fix it by explicitly mentioning speex.
(http://bugs.debian.org/541206)
Signed-off-by: Daniel T Chen <crimsun@ubuntu.com>
As per the comments in git-version-gen, the .version file that is
typically generated automatically during build should be included
in the tarball, so generate it in the dist-hook.
This fixes a problem when running make distcheck which uses a read-only
source folder and thus cannot generate the $top_srcdir/.version as
part of it's build process (it uses a _build folder which is r/w)
and thus cannot be built.
This is needed to better support out of tree builds (including
distcheck) and to ensure the necessary folders are created in the
build tree on configure and also works around an intl-tools bug
(https://bugs.launchpad.net/intltool/+bug/605826)
The Makefile.am's used are minimal (and in some cases completely
blank). At present they do not include anything interesting
with the majority of the real work still done by the monolitic
src/Makefile.am
It may make sense to start splitting out src/Makefile.am into
smaller chunks but this commit makes the minimum changes to address
the issues that result from using make distcheck and other out of
tree builds.
Note: This 'breaks' the ability to type make in e.g. the src/modules
folder and have all of PA rebuilt accordingly (this is because the
static Makefiles previously present just did a "make -C ..") which
was purportedly for use in emacs. But I'm sure there will be a better
and more robust way to configure emacs to do your builds properly if
this behaviour is still desirable.
When we are dealing with a tag against a commit with other tags applied,
then favour the first tag, not the last one. This allows us to tag a release
as e.g. v1.0, v1.1-dev and v2.0-dev and get a version of 1.0 rather than the
newer tags.
This also removes support for git versions that do not accept
the --match argument as we also now make use of the --contains arg
which was added at the same time (I think)
Additionally, the v variable is reset just incase it is already exported
in the shell for some reason.
Due to how our branching worked out, these new features will
debut in v1.0 and not v0.9.22 which has already been released
from the stable-queue branch
This is required to make sure that the source output between
module-echo-cancel and ALSA can't get plugged to the virtual source or
monitor of the virtual sink that we expose. This could be triggered by
changing the profile of the underlying ALSA device.
There is a call to this function where 'skip' variable is NULL. Looks
like this code doesn't get hit very often, probably because a suitable
default sink can be found to move the stream to. However, if we can't
move to the default sink and skip is NULL, there will be a segfault.
Currently if sink base volume differs from 0dB and sync-volume is used,
wrong volume values are written to hw. This patch fixes that.
Signed-off-by: Juho Hämäläinen <ext-juho.hamalainen@nokia.com>
Wrong order of libraries causes libtool to link primarily from lirc
library location and not from $(libdir) location in install
phase. This is a problem if you have an older version pulseaudio libs
installed under same directory as liblirc and you are building a
newer version with some other directory prefix.
Build appeared to fails when a new function is added to
libpulsecommon. This is because build uses libpulsecommon that is
installed into system instead of the one that is being built. Explicit
linking to libpulsecommon fixes the problem.
<optdesc><p>Specify the symbolic name of the sink/source to play/record this stream on/from.</p></optdesc>
<optdesc><p>Specify the symbolic name of the sink/source to play/record this stream on/from. The special names @DEFAULT_SINK@, @DEFAULT_SOURCE@ and @DEFAULT_MONITOR@ can be used to specify the default sink, source and monitor respectively.</p></optdesc>
<optdesc><p>Move the specified playback stream (identified by its numerical index) to the specified sink (identified by its symbolic name or numerical index).</p></optdesc>
<optdesc><p>Move the specified recording stream (identified by its numerical index) to the specified source (identified by its symbolic name or numerical index).</p></optdesc>
<optdesc><p>Load the specified module with the specified arguments into the running sound server. Prints the numeric index of the module just loaded to STDOUT. You can use it to unload the module later.</p></optdesc>
<optdesc><p>Set the specified card (identified by its symbolic name or numerical index) to the specified profile (identified by its symbolic name).</p></optdesc>
<optdesc><p>Set the specified sink (identified by its symbolic name or numerical index) to the specified port (identified by its symbolic name).</p></optdesc>
</option>
<option>
<p><opt>get-default-source</opt></p>
<optdesc><p>Returns the symbolic name of the default source.</p></optdesc>
<optdesc><p>Set the specified source (identified by its symbolic name or numerical index) to the specified port (identified by its symbolic name).</p></optdesc>
<optdesc><p>Set a latency offset to a specified port (identified by its symbolic name) that belongs to a card (identified by its symbolic name or numerical index).
<arg>OFFSET</arg> is a number which represents the latency offset in microseconds</p></optdesc>
</option>
<option>
<p><opt>get-sink-volume</opt><arg>SINK</arg></p>
<optdesc><p>Get the volume of the specified sink (identified by its symbolic name or numerical index) displayed in the same format as the `info` command.</p></optdesc>
<optdesc><p>Get the volume of the specified source (identified by its symbolic name or numerical index) displayed in the same format as the `info` command.</p></optdesc>
for PulseAudio client applications. If the version in the user's
home directory does not exist the global configuration file is
loaded. See <manrefname="pulse-client.conf"section="5"/> for
@ -355,12 +415,6 @@ USA.
<sectionname="UNIX Groups and users">
<p>Group <arg>pulse-rt</arg>: if the PulseAudio binary is marked
SUID root, then membership of the calling user in this group
decides whether real-time and/or high-priority scheduling is
enabled. Please note that enabling real-time scheduling is a
security risk (see below).</p>
<p>Group <arg>pulse-access</arg>: if PulseAudio is running as a system
daemon (see <opt>--system</opt> above) access is granted to
members of this group when they connect via AF_UNIX sockets. If
@ -369,7 +423,7 @@ USA.
<p>User <arg>pulse</arg>, group <arg>pulse</arg>: if PulseAudio is running as a system
daemon (see <opt>--system</opt> above) and is started as root the
daemon will drop priviliges and become a normal user process using
daemon will drop privileges and become a normal user process using
this user and group. If PulseAudio is running as a user daemon
this user and group has no meaning.</p>
</section>
@ -381,43 +435,20 @@ USA.
latency of the PulseAudio daemon from the system load and is thus
the best way to make sure that PulseAudio always gets CPU time
when it needs it to refill the hardware playback
buffers. Unfortunately this is a security risk on most systems,
buffers. Unfortunately this can be a security risk on some systems,
since PulseAudio runs as user process, and giving realtime
scheduling priviliges to a user process always comes with the risk
scheduling privileges to a user always comes with the risk
that the user misuses it to lock up the system -- which is
possible since making a process real-time effectively disables
preemption.</p>
preemption. To solve this problem, PulseAudio uses rtkit to safely
acquire real-time scheduling when available.</p>
<p>To minimize the risk PulseAudio by default does not enable
real-time scheduling. It is however recommended to enable it
on trusted systems. To do that start PulseAudio with
<opt>--realtime</opt> (see above) or enabled the appropriate option in
<file>daemon.conf</file>. Since acquiring realtime scheduling is a
priviliged operation on most systems, some special changes to the
system configuration need to be made to allow them to the calling
user. Two options are available:</p>
<p>On newer Linux systems the system resource limit RLIMIT_RTPRIO
(see <manrefname="setrlimit"section="2"/> for more information)
can be used to allow specific users to acquire real-time
scheduling. This can be configured in
<file>/etc/security/limits.conf</file>, a resource limit of 9 is recommended.</p>
<p>Alternatively, the SUID root bit can be set for the PulseAudio
binary. Then, the daemon will drop root priviliges immediately on
startup, however retain the CAP_NICE capability (on systems that
support it), but only if the calling user is a member of the
<arg>pulse-rt</arg> group (see above). For all other users all
capababilities are dropped immediately. The advantage of this
solution is that the real-time priviliges are only granted to the
PulseAudio daemon -- not to all the user's processes.</p>
<p>Alternatively, if the risk of locking up the machine is
<p>If the risk of locking up the machine is
considered too big to enable real-time scheduling, high-priority
scheduling can be enabled instead (i.e. negative nice level). This
can be enabled by passing <opt>--high-priority</opt> (see above)
when starting PulseAudio and may also be enabled with the
approriate option in <file>daemon.conf</file>. Negative nice
appropriate option in <file>daemon.conf</file>. Negative nice
levels can only be enabled when the appropriate resource limit
RLIMIT_NICE is set (see <manrefname="setrlimit"section="2"/> for
more information), possibly configured in
@ -427,10 +458,18 @@ USA.
<sectionname="Environment variables">
<p>The PulseAudio client libraries check for the existance of the
<p>The PulseAudio client libraries check for the existence of the
following environment variables and change their local configuration accordingly:</p>
<p><arg>$PULSE_SERVER</arg>: the server string specifying the server to connect to when a client asks for a sound server connection and doesn't explicitly ask for a specific server.</p>
<p><arg>$PULSE_SERVER</arg>: the server string specifying the server
to connect to when a client asks for a sound server connection and doesn't
explicitly ask for a specific server. The server string is a list of
server addresses separated by whitespace which are tried in turn. A server
address consists of an optional address type specifier (unix:, tcp:, tcp4:,
tcp6:), followed by a path or host address. A host address may include an
optional port number. A server address may be prefixed by a string enclosed
in {}. In this case the following server address is ignored unless the prefix
string equals the local hostname or the machine id (/etc/machine-id).</p>
<p><arg>$PULSE_SINK</arg>: the symbolic name of the sink to connect to when a client creates a playback stream and doesn't explicitly ask for a specific sink.</p>
@ -440,6 +479,9 @@ USA.
<p><arg>$PULSE_CLIENTCONFIG</arg>: path of file that shall be read instead of <file>client.conf</file> (see above) for client configuration.</p>
<p><arg>$PULSE_COOKIE</arg>: path of file that contains the PulseAudio
authentication cookie. Defaults to <file>~/.config/pulse/cookie</file>.</p>
<p>These environment settings take precedence -- if set -- over the configuration settings from <file>client.conf</file> (see above).</p>
description:'Enable building and installation of man pages')
option('tests',
type:'feature',
description:'Enable unit tests')
option('system_user',
type:'string',value:'pulse',
description:'User for running the PulseAudio daemon as a system-wide instance (pulse)')
option('system_group',
type:'string',value:'pulse',
description:'Group for running the PulseAudio daemon as a system-wide instance (pulse)')
option('access_group',
type:'string',value:'pulse-access',
description:'Group which is allowed access to a system-wide PulseAudio daemon (pulse-access)')
option('database',
type:'combo',value:'tdb',
choices:['gdbm','tdb','simple'],
description:'Database backend')
option('legacy-database-entry-format',
type:'boolean',
description:'Try to load legacy (< 1.0) database files (card, device and volume restore)')
option('stream-restore-clear-old-devices',
type:'boolean',value:false,
description:'Forget per-stream routing settings that have been set before version 14.0. Recommended when using GNOME. See https://gitlab.freedesktop.org/pulseaudio/pulseaudio/-/issues/832')
option('running-from-build-tree',
type:'boolean',
description:'Enable running from build tree')
option('atomic-arm-linux-helpers',
type:'boolean',value:true,
description:'Use inline asm or libatomic_ops instead')
option('atomic-arm-memory-barrier',
type:'boolean',value:false,
description:'Enable memory barriers (only really needed in SMP arm systems)')
# Paths
option('padsplibdir',
type:'string',
description:'Specify location where OSS wrapper library will be installed')
# This one gets inserted into a shell double-quoted string, so needs to be escaped (\$LIB). Meson
# removes a layer of escapes when parsing the description string, so we need to double the
# backslashes for "meson configure" to display the right text.
option('pulsedsp-location',
type:'string',
description:'Specify location where OSS wrapper library will be loaded from; can use \\$LIB')
option('modlibexecdir',
type:'string',
description:'Specify location where modules will be installed')
option('alsadatadir',
type:'string',
description:'Directory for ALSA card profiles (defaults to ${datadir}/pulseaudio/alsa-mixer)')
option('systemduserunitdir',
type:'string',
description:'Directory for systemd user service files')
option('udevrulesdir',
type:'string',
description:'Directory for udev rules (defaults to /lib/udev/rules.d)')
option('bashcompletiondir',
type:'string',
description:'Directory for bash completion scripts ["no" disables]')
option('zshcompletiondir',
type:'string',
description:'Directory for zsh completion scripts ["no" disables]')
# Optional features
option('alsa',
type:'feature',value:'auto',
description:'Optional ALSA support')
option('asyncns',
type:'feature',value:'auto',
description:'Optional Async DNS support')
option('avahi',
type:'feature',value:'auto',
description:'Optional Avahi support')
option('bluez5',
type:'feature',value:'auto',
description:'Optional BlueZ 5 support')
option('bluez5-gstreamer',
type:'feature',value:'auto',
description:'Optional BlueZ 5 GStreamer support')
option('bluez5-native-headset',
type:'boolean',
description:'Optional native headset backend support (BlueZ 5)')
option('bluez5-ofono-headset',
type:'boolean',
description:'Optional oFono headset backend support (BlueZ 5)')
option('consolekit',
type:'feature',value:'auto',
description:'Optional ConsoleKit support')
option('dbus',
type:'feature',value:'auto',
description:'Optional D-Bus support')
option('elogind',
type:'feature',value:'auto',
description:'Optional elogind support')
option('fftw',
type:'feature',value:'auto',
description:'Optional FFTW support')
option('glib',
type:'feature',value:'auto',
description:'Optional GLib 2 support')
option('gsettings',
type:'feature',value:'auto',
description:'Optional GSettings support')
option('gstreamer',
type:'feature',value:'auto',
description:'Optional GStreamer dependency for media-related functionality')
option('gtk',
type:'feature',value:'auto',
description:'Optional Gtk+ 3 support')
option('hal-compat',
type:'boolean',
description:'Optional HAL->udev transition compatibility support (needs udev)')
option('ipv6',
type:'boolean',
description:'Optional IPv6 support')
option('jack',
type:'feature',value:'auto',
description:'Optional JACK support')
option('lirc',
type:'feature',value:'auto',
description:'Optional LIRC support')
option('openssl',
type:'feature',value:'auto',
description:'Optional OpenSSL support (used for Airtunes/RAOP)')
