Commit Graph
106 Commits
Author SHA1 Message Date
Mike Brady bdc2dca3ec 3.3.7rc0 2020-07-06 17:09:15 +01:00
Mike Brady 02224666da Squashed commit of the following:
Don't use strndup as old versions of OpenWrt don't seem to have it
    Update FREEBSD.md
    Fix comparison between old and new metadata strings by checking they have equal lengths as well.
    Remove a redundant definition to prevent use on an uninitialised copy, clean up some of the less-used backends
    Remove some unwanted and unused variable declarations from audio_dummy.c and audio_soundio.c
    Add the -fno-common flag to compilation.
    Treat the "mper" attribute as the 64-bit item that it is rather than a 32-bit item as hithereto. Output it as a hexadecimal number to correspond with the format of the track id obtained from AppleScript
    Make changes to that it compiles under gcc-10 with -fno-common
    Add a SIGCHLD handler to reap zombie processes generated after script invocations where wait_for_completion is set to "no".
    Update and rename CONTRIBUTING.md to REPORTING ISSUES.md
    Add a few more commands and clarify some text.
    Change from talking about a server to talking about a client. Technically, e.g. iTunes is a client of Shairport Sync.
    Store the UserAgent so as to recognise forked-daapd so as to always send a revision_number of 1 when asking for playerstatus of forked-daapd.
    Add or update some of the copyright notices
    Add the ability to set the volume directly to the D-Bus RemoteControl interface and to the MPRIS interface. Both use the recently-discovered ability to set the device_volume.
    Merge in Pieter De Gendt's work on resampling on the jack backend.
    Merge pull request #939 from pdgendt/feature/jack-soxr-resampling
    Add SetAirplayVolume to the D-Bus RemoteControl interface.
    Add SetVolume to the MPRIS interface.
    Hook up the Volume property in the MPRIS interface.
    Modify RemoteCommand in the D-Bus interface to return the HTTP status and response.
    Change the type of airplay_volume from int to double in the metadata hub.
    Add a few sample commands in the D-Bus document.
2020-02-20 21:32:35 +00:00
Mike Brady 686fc2a5c1 Add missing frame status field, fix a bug with the first missing frame variable, expose the settings for resending, remove some commented out code. 2019-10-10 09:23:20 +01:00
Mike Brady e2294decb1 Squashed commit of the following:
commit 3dda7928b6df6ba7febc20f22faff5c8be024ef4
Author: Mike Brady <mikebrady@eircom.net>
Date:   Sun Sep 22 20:59:50 2019 +0100

    Update RELEASENOTES-DEVELOPMENT.md

commit 153eb84cff292bd127c516e6486ba9c1f9a3ce16
Author: Mike Brady <mikebrady@eircom.net>
Date:   Sat Sep 21 19:30:26 2019 +0100

    Update RELEASENOTES-DEVELOPMENT.md

commit b574d49b37f874cc7b91163a81fcb9dc6c760ef4
Author: Mike Brady <mikebrady@eircom.net>
Date:   Sat Sep 21 19:03:13 2019 +0100

    Update RELEASENOTES-DEVELOPMENT.md

commit bce0613283e233ddfa24d080adff3450b4ddd909
Author: Mike Brady <mikebrady@eircom.net>
Date:   Sat Sep 21 18:59:18 2019 +0100

    Update configure.ac

commit 7dc145361186806a560495d8171a387f5856f2f0
Author: Mike Brady <mikebrady@eircom.net>
Replace the code for determining when resends should be sent for missing audio packets.

Multiple sequential resend requests are now rolled up into one.

Should address Issue #865.
2019-09-22 21:04:07 +01:00
Chris Boot b9cf91b2bb More typo fixes across the tree
Previously I have made spelling changes as caught by Debian's lintian
tool, which has a short list of common misspellings that it tests for.
This time I decided to run codespell against the source and fix all the
obvious problems it came up with. It also uses a list of misspellings
but it's much larger than lintian's, but also has some false positives
so I went over the list by hand.

The command I used to do check the source was:
git ls-tree -rz --name-only HEAD | xargs -0 codespell -L minimise,errorstring
2019-07-27 00:05:38 -03:00
Mike Brady 1f45b8c006 allow calculation of nominal source frame rates even when the connection is error prone. 2019-04-02 15:37:38 +01:00
Mike Brady 5a7b7c011b Add the ability to receive uncompressed 16 bit big-endian (?) / 44,100 fps / interleaved stereo PCM. 2019-03-18 11:21:27 +00:00
Mike Brady 46d117b69a add a volume control mutex to serialise multiple attempts to change the volume 2019-02-01 21:07:35 +00:00
Mike Brady c5ff0dfed0 Rebuild the volume control calculations. Implement software mute. Try to delay alsa initialisation as long as possible. 2019-01-30 09:24:10 +00:00
Mike Brady 3434761517 Add code for calculating and recognising the interpolation threshold. 2019-01-04 17:10:57 +00:00
Mike Brady 4046c0e73d Fix dither, add dither to silences during lead-in, interpolatied frames and keeping out of standby, only if the output device doesn't use a mixer. Check configuration at initialisation. 2018-12-14 15:07:00 +00:00
Mike Brady 9d5ff22f2d Connect up the stall response to the external command calling -- add a single argument to the unfixable error call. 2018-12-10 22:28:52 +00:00
Mike Brady 277d401d24 Ensure watchdog always starts with the starting time of the thread. Add watchdog bark count. On the third bark, execute an unfixable error script if present. 2018-12-04 17:25:15 +00:00
Mike Brady 78da7c453c Make the stop requester volatile, duh. 2018-12-03 18:01:54 +00:00
Mike Brady 4bb29bc198 Make sure that shared watchdog count is correctly updated in the rtsp_conn_info structure! 2018-12-03 16:52:32 +00:00
Mike Brady 76ee6d5e43 Add new watchdog thread for the player -- should be more reliable than the existing code. 2018-11-26 19:33:22 +00:00
Mike Brady 3cb359c747 Tighten up integer promotion checks to calculate large delays properly. Fix input rate calculation restarting. 2018-10-23 10:26:08 +00:00
Mike Brady e0a4e5cde8 Format tidy-up 2018-10-21 17:10:23 +01:00
Mike Brady e5259e8764 Remove a few compiler warnings. 2018-10-21 17:00:56 +01:00
Mike Brady d2dba7cdba Seems to work for higher rates 2018-10-19 22:14:23 +01:00
Mike Brady de071aeffc Try to fix modulo calculations 2018-10-18 17:44:50 +01:00
Mike Brady c9b3d2a2f4 Many changes to compilation and linking flags. Stop using HAVE_* flags except where necessary, use CONFIG_* for optional stuff, use HAS_* for immediate definitions, use USE_* for AM definitions. Probably introduced bgs, sigh. 2018-09-09 20:41:31 +01:00
Mike Brady 1ee806a53a Fix display of drift -- the sign is correct now. All timing is relative to the local clock. Add a rough estimate of the correction necessary. Use the most recent 128 timing samples. 2018-08-22 13:02:42 +01:00
Mike Brady 5141e2f5dc Format tidy-up 2018-08-14 12:14:49 +01:00
Mike Brady 519135734f Calculate actual and nominal input frame rates more accurately, allowing for flush and resume events. 2018-08-14 12:13:02 +01:00
Mike Brady 2f58856268 Add source rate calculation based on source frame and timing information, calculated from start of play session. 2018-08-13 14:44:22 +01:00
Mike Brady a8cfc9915f display the number of samples used to calculcate drift, clean up the settings for the number of samples to skip at the start and for the minimum number of samples to use. Store 64 samples. 2018-08-13 14:06:40 +01:00
Mike Brady 3870195cf9 Begin to use a unified and more accurate scheme for translating between local time and rtp timestamp, taking clock drift and actualt source rate into account if available. Not dealing with highes output rates yet. 2018-08-05 17:23:07 +01:00
Mike Brady f6b3c420c7 Enable input frame rate calculation to handle a pause/resume correctly. 2018-07-30 14:07:39 +01:00
Mike Brady b75c82d14a Add input frame rate calculation. 2018-07-30 13:39:38 +01:00
Mike Brady 76c5dd9285 Stop using SIGUSR1 for cancelling threads, use pthread_cancel and friends instead, fix some memory leaks, add accurate ouput rate calculation to statistics, add "quit" verb to MPRIS and native d-bus interfaces. Probably still buggy. 2018-07-30 10:07:11 +01:00
Mike Brady c85e7e05d1 Add a frames per second output. This should be exact if the machine is connected up to, say, an NTP time service. It uses CLOCK_MONOTONIC time. 2018-07-28 19:02:02 +01:00
Mike Brady c8c70b6016 Add quit commands to the MPRIS and native D-Bus command set. Not tested ecologically. 2018-07-18 15:32:07 +01:00
Mike Brady a4eaace76b Begin to try to get Shairport Sync to quit cleanly, relinquishing memory and ports etc. properly so that valgrind can be more useful. 2018-07-17 10:28:06 +01:00
Mike Brady 9a8a863825 Clean out player_+thread_please_stop references. 2018-07-15 16:25:56 +01:00
Mike Brady 01f2a374b9 Use pthread_cancel to stop a player thread rather than SIGUSR1 and pthread_kill. 2018-07-15 13:54:33 +01:00
Mike Brady 768dc7759a Tidy up log output and add timers to inform statements. Move more stuff to the player thread pop function. Warn about cancellation points in debug versions of mutex lock and unlock. Ensure they are not cancellation points when debug not in use. Warn about Mac OS X cancellation points. 2018-07-15 11:54:48 +01:00
Mike Brady d3b79b965d Make more of the code amenable to pthread cancel 2018-07-14 19:34:31 +01:00
Mike Brady c2e3fa5aa7 Move ab_mutex, flow control CV and flush mutex initialisation and teardown out of the player thread. Thus remove need for player theread lock. Format some of the surce files. 2018-07-14 13:38:29 +01:00
Mike Brady d224f47e37 Remove unneeded volume control mutex, adjust some timings, add wait time to slow mutex locks. 2018-06-01 11:21:16 +01:00
Mike Brady 6d08725539 If the player thread is locked, don't block player_put_packet. Instead, discard the packet. 2018-05-17 16:15:17 +01:00
Mike Brady 7d0eb37958 Move to using pthread_cancel in place of pthread_kill -- due to a problem in CYGWIN, but the move makes the code simpler and more rugged. 2018-05-17 10:16:44 +01:00
Mike Brady b9d3a036b1 Revise resend interval calculation to spread it out over a long latency. 2018-05-15 12:08:01 +01:00
Mike Brady 418b1ba7c7 Add a read-write mutex to protect a player thread when it's being created or deleted; make player_flush get a read lock to get access. 2018-05-12 15:10:27 +01:00
Mike Brady 21e1e12c76 If it's an AirPlay user agent up to and including version 363 (iOS 11.1.2) use the old offset calculation by adding 11,205 frames to the offset. 2018-05-04 18:25:52 +01:00
Mike Brady 3823f337d0 incorporate 3.2d30 stuff -- lots of important bug fixes -- and include the extra diagnostic information from RC2 2018-04-14 14:06:09 +01:00
Mike Brady 49d5486df6 Keep timestamp info from the incoming audio packet 2018-04-14 13:56:44 +01:00
Mike Brady e76a636668 Fix serious and long-standing bugs in the control, audio and timing threads that could be responsible for low resilience on noisy and unrealiable networks. Add ability to simiulate UDP receive and send failures on timing, control and audio ports. Significantly improve resend requesting and handling. 2018-04-14 13:01:42 +01:00
Mike Brady ad450d745b Add an extra field to a buffer for checking purposes. 2018-04-12 08:52:47 +01:00
Mike Brady 665a53a8b5 Allow sps to deal with duplicate RTSP messages like RECORD and SETUP. Add some diagnostics to the player. 2018-04-10 17:35:39 +01:00