This is to workaround systems where we hang in playback because the buffer
does not report the space for whatever reason. The system will instead block
in PlayDevice, which always immediately follows WaitDevice in modern times
so this works out, and it seems to keep the device moving forward.
For a future revision, we are either going to clean this up more properly,
or attempt to move to PulseAudio's pa_stream_set_write_callback() API, but
this will do for SDL 2.0.18.
Reference #4387 for discussion and further information.
Case fallthrough warnings can be suppressed using the __fallthrough__
compiler attribute. Unfortunately, not all compilers have this
attribute, or even have __has_attribute to check if they have the
__fallthrough__ attribute. [[fallthrough]] is also available in C++17
and the next C2x, but not everyone uses C++17 or C2x.
So define the SDL_FALLTHROUGH macro to deal with those problems - if we
are using C++17 or C2x, it expands to [[fallthrough]]; else if the
compiler has __has_attribute and has the __fallthrough__ attribute, then
it expands to __attribute__((__fallthrough__)); else it expands to an
empty statement, with a /* fallthrough */ comment (it's a do {} while
(0) statement, because users of this macro need to use a semicolon,
because [[fallthrough]] and __attribute__((__fallthrough__)) require a
semicolon).
Clang before Clang 10 and GCC before GCC 7 have problems with using
__attribute__ as a sole statement and warn about a "declaration not
declaring anything", so fall back to using the /* fallthrough */ comment
if we are using those older compiler versions.
Applications using SDL are also free to use this macro (because it is
defined in begin_code.h).
All existing /* fallthrough */ comments have been replaced with this
macro. Some of them were unnecessary because they were the last case in
a switch; using SDL_FALLTHROUGH in those cases would result in a compile
error on compilers that support __fallthrough__, for having a
__attribute__((__fallthrough__)) statement that didn't immediately
precede a case label.
Case fallthrough warnings can be suppressed using the __fallthrough__
compiler attribute. Unfortunately, not all compilers have this
attribute, or even have __has_attribute to check if they have the
__fallthrough__ attribute. [[fallthrough]] is also available in C++17
and the next C2x, but not everyone uses C++17 or C2x.
So define the SDL_FALLTHROUGH macro to deal with those problems - if we
are using C++17 or C2x, it expands to [[fallthrough]]; else if the
compiler has __has_attribute and has the __fallthrough__ attribute, then
it expands to __attribute__((__fallthrough__)); else it expands to an
empty statement, with a /* fallthrough */ comment (it's a do {} while
(0) statement, because users of this macro need to use a semicolon,
because [[fallthrough]] and __attribute__((__fallthrough__)) require a
semicolon).
Applications using SDL are also free to use this macro (because it is
defined in begin_code.h).
All existing /* fallthrough */ comments have been replaced with this
macro. Some of them were unnecessary because they were the last case in
a switch; using SDL_FALLTHROUGH in those cases would result in a compile
error on compilers that support __fallthrough__, for having a
__attribute__((__fallthrough__)) statement that didn't immediately
precede a case label.
Even without the thread, it'll do an initial hardware detection at startup,
but there won't be any further hotplug events after that. But for many cases,
that is likely complete sufficient.
In either case, this cleaned up the code to no longer need a semaphore at
startup.
Fixes#4862.
The observed behavior is that any nonzero timeout value would hang until the device was paused and resumed. And a zero timeout value would always return 0 frames written even when audio fragments could be heard. Making a manual timeout system unworkable.
None of the straightforward systems imply that there's a detectable problem before the call to AAudioStream_write(). And the callback set within AAudioStreamBuilder_setErrorCallback() does not get called as we enter the hang state.
I've found that AAudioStream_getTimestamp() will report an error state from another thread. So this change codifies that behavior a bit until a better fix or more root cause can be found.
See SDL bug #4703. This implements two new hints:
- SDL_APP_NAME
- SDL_SCREENSAVER_INHIBIT_ACTIVITY_NAME
The former is the successor to SDL_HINT_AUDIO_DEVICE_APP_NAME, and acts
as a generic "application name" used both by audio drivers and DBUS
screensaver inhibition. If SDL_AUDIO_DEVICE_APP_NAME is set, it will
still take priority over SDL_APP_NAME.
The second allows the "activity name" used by
org.freedesktop.ScreenSavver's Inhibit method, which are often shown in
the UI as the reason the screensaver (and/or suspend/other
power-managment features) are disabled.
The recent change to make SDL_AUDIODRIVER support comma-separated lists
broke the previous behavior where an SDL_AUDIODRIVER that was empty
behaved the same as if it was not set at all. This old behavior was
necessary to paper over differences in platforms where SDL_setenv may
or may not actually delete the env var if an empty string is specified.
This patch just adds a simple check to ensure SDL_AUDIODRIVER is not
empty before using it, restoring the old interpretation of the empty
var.
Originally, SDL 1.2 used "pulse" as the name for its PulseAudio driver.
While it now supports "pulseaudio" as well for compatibility with SDL
2.0 [1], there are still scripts and distro packages which set
SDL_AUDIODRIVER=pulse [2]. While it's possible to remove this in most
circumstances or replace it with "pulseaudio" or a comma-separated list,
this may still conflict if the environment variable is set globally and
old binary builds of SDL 1.2 (e.g. packaged with older games) are being
used.
To fix this on SDL 2.0, add a hardcoded check for "pulse" as an audio
driver name, and replace it with "pulseaudio". This mimics what SDL 1.2
does (but in reverse). Note that setting driver_attempt{,_len} is safe
here as they're reset correctly based on driver_attempt_end on the next
loop.
[1] d951409784
[2] https://bugzilla.opensuse.org/show_bug.cgi?id=1189778
This might have changed at some point in the Pulse API, or this might have
always been wrong, but we didn't notice because the dynamic loading code
hides it by casting things to void *. The static path, where it
assigns the function pointer directly, puts out a clear compiler warning,
though.
Don't rely on checking __clang_major__ since it is not comparable
between different vendors. Don't use "#pragma clang attribute" since it
is only available in relatively recent versions, there's no obvious way
to check if it's supported, and just using __attribute__ directly (for
gcc as well) results in simpler code anyway.
Without this change, driver names don't get matched correctly;
for example "a" can get matched with "alsa" since it only checks
whether the string matches up to the length of the requested
driver name.
This is needed to support CHERI, and thus Arm's experimental Morello
prototype, where pointers are implemented using unforgeable capabilities
that include bounds and permissions metadata to provide fine-grained
spatial and referential memory safety, as well as revocation by sweeping
memory to provide heap temporal memory safety.
On most systems (anything with a flat memory hierarchy rather than using
segment-based addressing), size_t and uintptr_t are the same type.
However, on CHERI, size_t is just an integer offset, whereas uintptr_t
is still a capability as described above. Casting a pointer to size_t
will strip the metadata and validity tag, and casting from size_t to a
pointer will result in a null-derived capability whose validity tag is
not set, and thus cannot be dereferenced without faulting.
The audio and cursor casts were harmless as they intend to stuff an
integer into a pointer, but using uintptr_t is the idiomatic way to do
that and silences our compiler warnings (which our build tool makes
fatal by default as they often indicate real problems). The iconv and
egl casts were true positives as SDL_iconv_t and iconv_t are pointer
types, as is NativeDisplayType on most OSes, so this would have trapped
at run time when using the round-tripped pointers. The gles2 casts were
also harmless; the OpenGL API defines this argument to be a pointer type
(and uses the argument name "pointer"), but it in fact represents an
integer offset, so like audio and cursor the additional idiomatic cast
is needed to silence the warning.
WASAPI_WaitDevice is used for audio playback and capture, but needs to
behave slighty different.
For playback `GetCurrentPadding` returns the padding which is already
queued, so WaitDevice should return when buffer length falls below the
buffer threshold (`maxpadding`).
For capture `GetCurrentPadding` returns the available data which can be
read, so WaitDevice can return as soon as any data is available.
In the old implementation WaitDevice could suddenly hang. This is
because on many capture devices the buffer (`padding`) wasn't filled
fast enough to surpass `maxpadding`. But if at one point (due to unlucky
timing) more than maxpadding frames were available, WaitDevice would not
return anymore.
Issue #3234 is probably related to this.
On modern CPUs, there's no penalty for using the unaligned instruction on
aligned memory, but now it can vectorize unaligned data too, which even if
it's not optimal, is still going to be faster than the scalar fallback.
Fixes#4532.
While we should normally expect _something_ from the stream based on the
AudioStreamAvailable check, it's possible for a device change to flush the
stream at an inconvenient time, causing this function to return 0.
Thing is, this is harmless. Either data will be NULL and the result won't matter
anyway, or the data buffer will be zeroed out and the output will just be
silence for the brief moment that the device change is occurring. Both scenarios
work themselves out, and testing on Windows shows that this behavior is safe.
Some of the SDL_AudioDevice struct members aren't initialized until after returning from the OpenDevice function. Since Pipewire uses it's own processing threads, the callbacks can be entered before all members of SDL_AudioDevice are initialized, such as work_buffer, callbackspec and the processing stream, which creates a race condition. Don't use these members when in the paused state to avoid potentially using uninitialized values and memory.
This prevents the dsp target from stealing the audio subsystem but not
being able to produce sound, so other audio targets further down the list
can make an attempt instead.
Thanks to Frank Praznik who did a lot of the research on this problem!
A user reported that the mpv video player hangs after attempting to
set an unsupported number of channels with the SDL audio output,
because it thinks it's successfully opened the device. This makes
the failure graceful.
Removes the node nickname from sink/source nodes as it doesn't provide any useful information and names now match those used in Pulseaudio, so any stored configuration data will be compatible between the two audio backends.
Constify the min/max period variables, use a #define for the base clock rate used in the calculations and note that changing the upper limit can have dire side effects as it's a hard limit in Pipewire.
Replace "magic numbers" with #defines, explain the requirements when using the userdata pointer in the node_object struct and a few other minor code and comment cleanups.
Use the 'R' (rear) prefixed designations for the rear audio channels instead of 'S' (surround). Surround designated channels are only used in the 8 channel configuration.
Further refactor the device enumeration code to retrieve the default sink/source node IDs from the metadata node. Use the retrieved IDs to sort the device list so that the default devices are at the beginning and thus are the first reported to SDL.
The latency of source nodes can change depending on the overall latency of the processing graph. Incoming audio must therefore always be buffered to ensure uninterrupted delivery.
The SDL_AudioStream path was removed in the input callback as the only thing it was used for was buffering audio outside of Pipewire's min/max period sizes, and that case is now handled by the omnipresent buffer.
Extend device enumeration to retrieve the channel count and default sample rate for sink and source nodes. This required a fairly significant rework of the enumeration procedure as multiple callbacks are involved now. Sink/source nodes are tracked in a separate list during the enumeration process so they can be cleaned up if a device is removed before completion. These changes also simplify any future efforts that may be needed to retrieve additional configuration information from the nodes.
This uses the mechanism added in emscripten-core/emscripten#10843
which was applied to SDL1 and OpenAL. This adds the same for SDL2.
This also reverts commit 865eaddffed50dbd13e6564c3f73902472cf74e8
which did something similar, but the new mechanism is more effective.
The DJGPP compiler emits many warnings for conflicts between print
format specifiers and argument types. To fix the warnings, I added
`SDL_PRIx32` macros for use with `Sint32` and `Uint32` types. The macros
alias those found in <inttypes.h> or fallback to a reasonable default.
As an alternative, print arguments could be cast to plain old integers.
I opted slightly for the current solution as it felt more technically correct,
despite making the format strings more verbose.
Nia Alarie
The NetBSD kernel's audio resampling code is much simpler and lower quality than libsamplerate.
Presumably, if SDL always performs I/O on the audio device in its native frequency, we can avoid resampling audio in the kernel and let SDL do it with libsamplerate instead.
If we fail to connect to the the pa server, we have an assigned context
and mainloop that isn't connected. So, when PULSEAUDIO_pa_context_disconnect
is called, pa asserts and crashes the application.
Assertion 'pa_atomic_load(&(c)->_ref) >= 1' failed at pulse/context.c:1055, function pa_context_disconnect(). Aborting.
When converting audio from signed to unsigned values of vice-versa
the silence value chosen by SDL was the value of the device, not
of the stream that the data was being put into. After conversion
this would lead to a very high or low value, making the speaker
jump to a extreme positon, leading to an audible noise whenever
creating, destroying or playing scilence on a device that reqired
such conversion.
Original code assigned MCIMixSetup.ulSamplesPerSec value to it, but it
is just the freq... We now change spec->samples only either if it is 0
or we changed the frequency, by picking a default of ~46 ms at desired
frequency (code taken from SDL_audio.c:prepare_audiospec()).
With this, the crashes I have been experiencing are gone.
I _think_ this is a right thing to do; it fixes a .wav file I have here that
has blockalign==2 when channels==2 and bitspersample==16, which otherwise
would fail.
This is only supported on PulseAudio. You can set a description when opening
your audio device that will show up in pauvcontrol, which lets you set
per-stream volume levels.
Fixes Bugzilla #4801.
Anthony Pesch's notes on his patch:
"Currently, the WASAPI backend creates a stream in shared mode and sets the
device's callback size to be half of the shared stream's total buffer size.
This works, but doesn't coordinate will with the actual hardware. The hardware
will raise an interrupt after every period which in turn will signal the
object being waited on inside of WaitDevice. From my empirical testing, the
callback size was often larger than the period size and not a multiple of it,
which resulted in poor latency when trying to time an application based on the
audio callback. The reason for this looked something like:
* The device's callback would be called and and the audio buffer was filled.
* WaitDevice would be called.
* The hardware would raise an interrupt after one period.
* WaitDevice would resume, see that a a full callback had not been played and
then wait again.
* The hardware would raise an interrupt after another period.
* WaitDevice would resume, see that a full callback + some extra amount had
been played and then it would again call our callback and this process would
repeat.
The effect of this is that the pacing between subsequent callbacks is poor -
sometimes it's called very quickly, sometimes it's called very late.
By matching the callback's size to the stream's period size, the pacing of
calls to the user callback is improved substantially. I didn't write an actual
test for this, but my use case for this was my Dreamcast emulator
(https://redream.io) which uses the audio callback to help drive the emulation
speed. Without this change and with the default shared stream buffer (which
has a period of ~10ms) I would get frame times that were between ~3-30
milliseconds; after this change I get frame times of ~11-22 milliseconds.
Note, this patch also has a change that removes passing a duration to the
Initialize call. It seems that the default duration used (when 0 is passed)
does typically match up with the duration returned by GetDevicePeriod, however
the Initialize docs say:
> To set the buffer to the minimum size required by the engine thread, the
> client should call Initialize with the hnsBufferDuration parameter set to 0.
> Following the Initialize call, the client can get the size of the resulting
> buffer by calling IAudioClient::GetBufferSize.
This change isn't strictly required, but I made it to hopefully rule out
another source of unexpected latency."
Fixes Bugzilla #4592.
So if you go into System Preferences on a MacBook and toggle between a pair of
connected bluetooth headphones and built-in internal speakers, SDL will
switch the device it is playing sound through, to match this setting, on the
fly.
Likewise if the default output device is a USB thing and is unplugged; as the
default device changes at the system level, SDL will pick this up and carry
on with the new default. This is different from our unplug detection for
specific devices, as in those cases we want to send the app a disconnect
notification, instead of migrating transparently as we now do for default
devices.
Note that this should also work for capture devices; if the device changes,
SDL will start recording from the new default.
Fixes Bugzilla #4851.
Ryan C. Gordon
Sylvain
Ozkan Sezer
Cameron Gutman
Sam Lantinga
Misa
David Gow
Lee Salzman
Oschowa
Charlie Birks
Joshua Root
Sebastian Krzyszkowiak
Jessica Clarke
Brick
Splamy
Luis Cáceres
Joel Linn
Per Nyblom
Ivan Epifanov
Ethan Lee
Frank Praznik
nia
Cameron Cawley
Frank Praznik
Romain Roffé
Alon Zakai
Jay Petacat
Alistair Leslie-Hughes
Manuel V?gele
Ethan Lee