- 48,000 frames per second operation,
- 48k lossless stereo, 5.1 and 7.1 surround sound
- Multichannel and multi rate operation on ALSA, PipeWire, PulseAudio, FreeBSD, stdout and unix pipe output backends.
Automatic, flexible and controllable output format (rate, sample format and channel count) selection.
Full FFmpeg integration to support transcoding, resampling, and new audio formats.
Better operation on lower powered devices down to e.g. Raspberry Pi B.
Reduced Docker image sizes with a slimmed-down FFmpeg library.
Enhanced timestamp handling for better synchronization.
Improved the sync error calculation.
A new "vernier" resampling and interpolation method for low-power CPUs.
Bug fixes and minor enhancements.
Note -- there are many breaking changes from previous versions of Shairport Sync!
Add three parameters to the backend play() function call -- (1) a flag indicating whether the samples are timed or not. If timed, (2) the timestamp and (3) the local time at which the first frame should be heard.
Update missing pipewire library message.
Makefile.am fix.
Update the SHM version and fix the SHM name so that it works with FreeBSD.
Remove redundant (?) AC_HEADER_STDC check.
Fix compilation and installation under FreeBSD -- changes to allow building in a separate directory broke the FreeBSD build process.
Add AC_CHECK_LIB for gcrypt in case PKG_CHECK_MODULES fails to find it.
Find gcrypt using pkg-config
Add code to detect when the RTSP channel goes idle for a period, and use it to reset clocks and timings, etc. so that SPS resumes correctly where the source has gone to sleep while playing a realtime stream and has subsequently woken up.
Don't exit if a UDP Clock Control packet is empty. Also check minimum packet size.
Remove some very experimental code which may cause memory and port leaks.
Strip file path from the filename used in debug messages, information messages, warnings and fatal error messages.
BB fix to Makefile.am -- overwriting the DBUS stuff with MPRIS definitions, duh.
Fix some problems building on FreeBSD and tidy up the use of the "sed" editor.
Fix some problems building on picore.
Fix an uninitialised variable
start using the "nqptp" SMI
Clean up and improve the way lead-in sillence is used to give a more accurate sync time to the first packet of audio.
Allow very good sync to occur with 0.3 seconds of lead-in silence.
Improve the use of the buffer offset.
Format some other files.
Many threads have a main loop that checks some exit variable in the loop condition. To stop a thread, another thread sets that exit variable, sends a SIGUSR1 to that thread to get it out of blocking calls like read(), write() or select(), and finally joins it.
There was a race condition though: if the signal arrived after the loop condition was checked but before the blocking system function was entered, the target thread essentially ignored the signal, made the blocking call anyway, and could get stuck forever there.
The only way around this race condition is to block SIGUSR1 in the destination thread during the critical time interval. This requires that the blocking system call and the call to unblock SIGUSR1 be an atomic unit. There is only one system call that makes this possible: pselect(), which takes a signal mask to replace the current signal mask with, but only while pselect() runs. Other functions like read(), recv(), write() and sendto() may only be called if we can be sure that they don't block. For this purpose, the file descriptors must be set to non-blocking mode and we must prepend the call to them with a call to pselect() to block until the socket becomes readable/writable.
mDNS is now handled more or less in the same way audio is.
Multiple mDNS backends are available, shairport tries them all until one
works.
This makes it, IMHO, cleaner, and easier to add other mDNS backends in
the future, such as using tinysvcmdns or dns-sd library.
Also it will also allow to define mDNS backends as plugins, in a similar
way it is done in PR #230.
Mike Brady
Mike Brady
Mike Brady
Belbo
Paul LietarandJames Laird