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>
