aboutsummaryrefslogtreecommitdiffstats
path: root/src/decode.c (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Start using song pointers in core data structuresEric Wong2008-04-121-34/+32
| | | | | | | Instead of copying URLs everywhere... [merged r7186 from branches/ew] git-svn-id: https://svn.musicpd.org/mpd/trunk@7244 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Drop metadata updates from HTTP for now (input HTTP, and shout)Eric Wong2008-04-121-60/+4
| | | | | | | | It is way more complicated than it should be; and locking it for thread-safety is too difficult. [merged r7183 from branches/ew] git-svn-id: https://svn.musicpd.org/mpd/trunk@7241 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Initial cut of fork() => pthreads() for decoder and playerEric Wong2008-04-121-142/+101
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I initially started to do a heavy rewrite that changed the way processes communicated, but that was too much to do at once. So this change only focuses on replacing the player and decode processes with threads and using condition variables instead of polling in loops; so the changeset itself is quiet small. * The shared output buffer variables will still need locking to guard against race conditions. So in this effect, we're probably just as buggy as before. The reduced context-switching overhead of using threads instead of processes may even make bugs show up more or less often... * Basic functionality appears to be working for playing local (and NFS) audio, including: play, pause, stop, seek, previous, next, and main playlist editing * I haven't tested HTTP streams yet, they should work. * I've only tested ALSA and Icecast. ALSA works fine, Icecast metadata seems to get screwy at times and breaks song advancement in the playlist at times. * state file loading works, too (after some last-minute hacks with non-blocking wakeup functions) * The non-blocking (*_nb) variants of the task management functions are probably overused. They're more lenient and easier to use because much of our code is still based on our previous polling-based system. * It currently segfaults on exit. I haven't paid much attention to the exit/signal-handling routines other than ensuring it compiles. At least the state file seems to work. We don't do any cleanups of the threads on exit, yet. * Update is still done in a child process and not in a thread. To do this in a thread, we'll need to ensure it does proper locking and communication with the main thread; but should require less memory in the end because we'll be updating the database "in-place" rather than updating a copy and then bulk-loading when done. * We're more sensitive to bugs in 3rd party libraries now. My plan is to eventually use a master process which forks() and restarts the child when it dies: locking and communication with the main thread; but should require less memory in the end because we'll be updating the database "in-place" rather than updating a copy and then bulk-loading when done. * We're more sensitive to bugs in 3rd party libraries now. My plan is to eventually use a master process which forks() and restarts the child when it dies: master - just does waitpid() + fork() in a loop \- main thread \- decoder thread \- player thread At the beginning of every song, the main thread will set a dirty flag and update the state file. This way, if we encounter a song that triggers a segfault killing the main thread, the master will start the replacement main on the next song. * The main thread still wakes up every second on select() to check for signals; which affects power management. [merged r7138 from branches/ew] git-svn-id: https://svn.musicpd.org/mpd/trunk@7240 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* whitespace cleanupMax Kellermann2008-04-121-1/+1
| | | | | Clean up some space indentations, replace with tabs. git-svn-id: https://svn.musicpd.org/mpd/trunk@7239 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* fix sign compare warningsMax Kellermann2008-04-121-7/+7
| | | | | | | Do explicit casts before comparing signed with unsigned. The one in log.c actually fixes another warning: in the expanded macro, there may be a check "logLevel>=0", which is always true. git-svn-id: https://svn.musicpd.org/mpd/trunk@7230 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* use unsigned integers in decoder.cMax Kellermann2008-04-121-5/+5
| | | | | | Use unsigned integers in decoderParent() for chunk numbers which cannot be negative. git-svn-id: https://svn.musicpd.org/mpd/trunk@7226 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* send notify signal after SIGCONTMax Kellermann2008-03-261-0/+2
| | | | | | | | | | When the decoder receives SIGCONT during waitNotify(), the kernel restarts the read() system call. This lets the decoder process block indefinitely, while the player process waits for it to react. This should probably be solved with a proper signal handler which aborts the read() system call, but for now, we just write to the pipe to make it wake up. git-svn-id: https://svn.musicpd.org/mpd/trunk@7216 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* notify the decoder instead of polling 100hzMax Kellermann2008-03-261-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the decoder process is faster than the player process, all decodedd buffers are full at some point in time. The decoder has to wait for buffers to become free (finished playing). It used to do this by polling the buffer status 100 times a second. This generates a lot of unnecessary CPU wakeups. This patch adds a way for the player process to notify the decoder process that it may continue its work. We could use pthread_cond for that, unfortunately inter-process mutexes/conds are not supported by some kernels (Linux), so we cannot use this light-weight method until mpd moves to using threads instead of processes. The other method would be semaphores, which historically are global resources with a unique name; this historic API is cumbersome, and I wanted to avoid it. I came up with a quite naive solution for now: I create an anonymous pipe with pipe(), and the decoder process reads on that pipe. Until the player process sends data on it as a signal, the decoder process blocks. This can be optimized in a number of ways: - if the decoder process is still working (instead of waiting for buffers), we could save the write() system call, since there is nobody waiting for the notification. [ew: I tried this using a counter in shared memory, didn't help] - the pipe buffer will be full at some point, when the decoder thread is too slow. For this reason, the writer side of the pipe is non-blocking, and mpd can ignore the resulting EWOULDBLOCK. - since we have shared memory, we could check whether somebody is actually waiting without a context switch, and we could just not write the notification byte. [ew: tried same method/result as first point above] - if there is already a notification in the pipe, we could also not write another one. [ew: tried same method/result as first/third points above] - the decoder will only consume 64 bytes at a time. If the pipe buffer is full, this will result in a lot of read() invocations. This does not hurt badly, but on a heavily loaded system, this might add a little bit more load. The preceding optimizations however are able eliminate the this. - finally, we should use another method for inter process notifications - maybe kill() or just make mpd use threads, finally. In spite of all these possibilities to optimize this code further, this pipe notification trick is faster than the 100 Hz poll. On my machine, it reduced the number of wakeups to less than 30%. git-svn-id: https://svn.musicpd.org/mpd/trunk@7215 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* fix -Wconst warningsMax Kellermann2008-02-051-2/+2
| | | | | | [ew: cleaned up the dirty union hack a bit] Signed-off-by: Eric Wong <normalperson@yhbt.net> git-svn-id: https://svn.musicpd.org/mpd/trunk@7180 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Cleanup #includes of standard system headers and put them in one placeEric Wong2008-01-031-8/+1
| | | | | | | | | | | | | This will make refactoring features easier, especially now that pthreads support and larger refactorings are on the horizon. Hopefully, this will make porting to other platforms (even non-UNIX-like ones for masochists) easier, too. os_compat.h will house all the #includes for system headers considered to be the "core" of MPD. Headers for optional features will be left to individual source files. git-svn-id: https://svn.musicpd.org/mpd/trunk@7130 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* silence is constant, as is the buffer we pass to playAudioEric Wong2008-01-011-3/+1
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@7123 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Simplify decode cleanup logic a bitEric Wong2008-01-011-26/+15
| | | | | | | | | | | DECODE_STATE_STOP is always set as dc->state, and dc->stop is always cleared. So handle it in decodeStart once rather than doing it in every plugin. While we're at it, fix a long-standing (but difficult to trigger) bug in mpc_decode where we failed to return if mpc_decoder_initialize() fails. git-svn-id: https://svn.musicpd.org/mpd/trunk@7122 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Merge branches/ew r7104Eric Wong2007-12-281-18/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | thread-safety work in preparation for rewrite to use pthreads Expect no regressions against trunk (r7078), possibly minor performance improvements in update (due to fewer heap allocations), but increased stack usage. Applied the following patches: * maxpath_str for reentrancy (temporary fix, reverted) * path: start working on thread-safe variants of these methods * Re-entrancy work on path/character-set conversions * directory.c: exploreDirectory() use reentrant functions here * directory/update: more use of reentrant functions + cleanups * string_toupper: a strdup-less version of strDupToUpper * get_song_url: a static-variable-free version of getSongUrl() * Use reentrant/thread-safe get_song_url everywhere * replace rmp2amp with the reentrant version, rmp2amp_r * Get rid of the non-reentrant/non-thread-safe rpp2app, too. * buffer2array: assert strdup() returns a usable value in unit tests * replace utf8ToFsCharset and fsCharsetToUtf8 with thread-safe variants * fix storing playlists w/o absolute paths * parent_path(), a reentrant version of parentPath() * parentPath => parent_path for reentrancy and thread-safety * allow "make test" to automatically run embedded unit tests * remove convStrDup() and maxpath_str() * use MPD_PATH_MAX everywhere instead of MAXPATHLEN * path: get rid of appendSlash, pfx_path and just use pfx_dir * get_song_url: fix the ability to play songs in the top-level music_directory git-svn-id: https://svn.musicpd.org/mpd/trunk@7106 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* decode: fix seek when pausedEric Wong2007-09-071-0/+1
| | | | | | We need to SIGCONT the decoder process to allow for seeking while paused. git-svn-id: https://svn.musicpd.org/mpd/trunk@6864 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Once again get rid of busy waiting and use SIGSTOP for pauseEric Wong2007-09-061-4/+10
| | | | | | | | | | | | | | The problems I had were related to the OSS driver and USB device I was using. The problems existed even with the old busy-waiting scheme enabled. OSS - Bithead USB => bad ALSA - Bithead USB => OK OSS - Onboard i8x0 => OK ALSA - Onboard i8x0 => OK bad - slow shutdown, pauses, dropped audio after pause/resume git-svn-id: https://svn.musicpd.org/mpd/trunk@6861 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* return to busy-waiting on pause for now..Eric Wong2007-09-061-10/+4
| | | | | | | | Until we can fix it properly (or replace it with a cleaner event system), I don't want this in trunk. Currently there are strange pauses when queueing and during shutdown that I can't seem to figure out right away. git-svn-id: https://svn.musicpd.org/mpd/trunk@6860 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* removing debug messages from signal handlersJ. Alexander Treuman2007-08-271-3/+5
| | | | | | | | | As unfortunate as it is to remove such useful debugging messages, it's necessary to fix a potential deadlock with signal handling. A bunch of functions the debug functions call aren't safe to call from a signal handler. There are some alternate solutions, but they're neither pretty nor simple. So just remove them entirely for now. git-svn-id: https://svn.musicpd.org/mpd/trunk@6828 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* send SIGSTOP to player and decoder processes on pause, tooEric Wong2007-08-271-4/+10
| | | | | | | | | | | | | | | | | | | | | | | | as with the stop command, this will cause the player and decoder to suspend and not wake up hundreds of times a second to poll a variable for wakeup. This will reduce power consumption on some CPUs while mpd is paused and not playing. tests: pause && unpause => OK pause && stop && play => OK pause && exit && restart w/statefile && unpause => OK pause && block sound device && \ unpause => failed to open sound device \ => still paused and suspended => unblock sound device && unpause => OK (playing) In all cases, the player process releases the audio device when paused before going into the suspended state. git-svn-id: https://svn.musicpd.org/mpd/trunk@6822 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* decode: close audio device after initial open if pausedJ. Alexander Treuman2007-08-121-1/+5
| | | | | | | | | | Currently, if we start decoding while the pause flag is set, we open the audio device and leave it opened, blocking other apps from using it. The obvious thing to do is to not open the audio device if the pause flag is set, but the open call also sets the audio format. Therefore I'm leaving the open call in, and just closing it immediately afterwards if the pause flag is set. git-svn-id: https://svn.musicpd.org/mpd/trunk@6745 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* decode: reformatting for better readabilityJ. Alexander Treuman2007-08-121-6/+9
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@6742 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* decode: prefer fileDecodeFunc over streamDecodeFunc for filesJ. Alexander Treuman2007-06-251-9/+10
| | | | | | | | | Only wavpack implements both fileDecodeFunc and streamDecodeFunc, and it's fileDecodeFunc provides more functionality. So try using that first. This commit also fixes a bug where the plugin test loop wouldn't break once a suitable plugin was found if it used fileDecodeFunc. git-svn-id: https://svn.musicpd.org/mpd/trunk@6655 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Remove a useless if (quit) block. It's enclosed in two while loops: theJ. Alexander Treuman2007-06-091-4/+0
| | | | | | | top depending on !quit, which doesn't set it anywhere before the if (quit) block is reached, and the inner one which doesn't set quit at all. Since it's a local variable and can't be modified externally, it'll never be hit. git-svn-id: https://svn.musicpd.org/mpd/trunk@6524 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Send SIGCONT to the decode process before sending it SIGTERM.J. Alexander Treuman2007-06-071-0/+1
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@6503 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Don't send the main process SIGUSR1 until we've sent the decode processJ. Alexander Treuman2007-06-041-1/+2
| | | | | SIGSTOP. git-svn-id: https://svn.musicpd.org/mpd/trunk@6486 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Don't kill the player process (and effectively the decode process) whenJ. Alexander Treuman2007-06-041-4/+16
| | | | | | completely stopped. Instead, send them SIGSTOP to pause the process until they're needed again. Then send them SIGCONT instead of re-spawning them. git-svn-id: https://svn.musicpd.org/mpd/trunk@6485 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Cleanup some formatting in decode.c.J. Alexander Treuman2007-06-041-10/+4
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@6480 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* The massive copyright updateAvuton Olrich2007-04-051-1/+1
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@5834 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Don't stop decoding when unpausing fails because no audio device can beJ. Alexander Treuman2007-04-031-5/+5
| | | | | opened. Instead just remain in paused state. git-svn-id: https://svn.musicpd.org/mpd/trunk@5820 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Reformatting some code.J. Alexander Treuman2007-04-031-5/+5
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@5819 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* #1) fix a few potential deadlock conditons in decode.c when crossfading is ↵Warren Dukes2007-02-111-5/+6
| | | | | | | enabled #2) fix a deadlock condition when attempting to seek if the decoder quit and returned to playerInit() git-svn-id: https://svn.musicpd.org/mpd/trunk@5325 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* on second thought, comment out the debug statement so that users aren't ↵Warren Dukes2007-01-241-1/+1
| | | | | potentially flooded git-svn-id: https://svn.musicpd.org/mpd/trunk@5270 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* add debug statement for bug #1380Warren Dukes2007-01-241-0/+1
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@5269 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Don't initialize globals to zero (or NULL)Eric Wong2007-01-141-1/+1
| | | | | | | | Some compilers and linkers aren't smart enough to optimize this, as global variables are implictly initialized to zero. As a result, binaries are a bit smaller as more goes in the .bss and less in the text section. git-svn-id: https://svn.musicpd.org/mpd/trunk@5254 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Do not set metadataState to PLAYER_METADATA_STATE_READ when decoding stops.J. Alexander Treuman2006-12-241-1/+0
| | | | | | | | This fixes a bug where streams that won't play somehow appear with the metadata of a previously played stream. As far as I can tell, the only reason this is done is to sync any buffered metadata with the displayed metadata when decoding stops, so there should be no other adverse effects. git-svn-id: https://svn.musicpd.org/mpd/trunk@5161 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Adding support for seeking HTTP streams.J. Alexander Treuman2006-12-231-1/+3
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@5159 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Replace strdup and {c,re,m}alloc with x* variants to check for OOM errorsEric Wong2006-08-261-2/+2
| | | | | | | | | | | | | | | | | | | I'm checking for zero-size allocations and assert()-ing them, so we can more easily get backtraces and debug problems, but we'll also allow -DNDEBUG people to live on the edge if they wish. We do not rely on errno when checking for OOM errors because some implementations of malloc do not set it, and malloc is commonly overridden by userspace wrappers. I've spent some time looking through the source and didn't find any obvious places where we would explicitly allocate 0 bytes, so we shouldn't trip any of those assertions. We also avoid allocating zero bytes because C libraries don't handle this consistently (some return NULL, some not); and it's dangerous either way. git-svn-id: https://svn.musicpd.org/mpd/trunk@4690 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Fix warnings for -Wmissing-prototypesAvuton Olrich2006-08-201-1/+1
| | | | | | Add -Wmissing-prototypes if compiling with gcc Static where possible git-svn-id: https://svn.musicpd.org/mpd/trunk@4657 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Revert leaks from r4311, and also the leak fixes as a result of thatEric Wong2006-08-141-7/+3
| | | | | | | | | | | | | | | | | | | | | | utf8ToFsCharset() and fsCharsetToUtf8() got very broken in r4311, and resulted in several commits to fix those leaks. Unfortunately, not all of those newly introduced leaks were fixed, nor was the result pretty. Also, fixed a double-free in lsPlaylists(). This is very hard to trigger (and therefore exploit) at the moment because we check printDirectoryInfo() beforehand. Intended behavior for utf8ToFsCharset() and fsCharsetToUtf8() as God^H^H^Hshank originally intended is now documented in path.h to prevent future errors like this. mpd could still use some good valgrind testing before the 0.12.0 release. <plug>In addition to reducing heap fragmentation, malloc reductions from mpd-ke greatly reduces the chance of leaks from happening due to programming errors.</plug> git-svn-id: https://svn.musicpd.org/mpd/trunk@4639 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* more sparse cleanupsEric Wong2006-08-081-1/+1
| | | | | | | | | * less-commonly compiled things like ao/mvp outputs * Adding -Wno-transparent-union to SPARSE_FLAGS makes it check inside decode.c, directory.c, player.c, and sig_handlers.c * remove unused variables leftover from the master process in sig_handlers.c git-svn-id: https://svn.musicpd.org/mpd/trunk@4598 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* logging cleanupsEric Wong2006-08-011-2/+2
| | | | | | | | | | | | | | * Moved all logging-related stuff into log.c (and not myfprintf.c) * ISO C90-compliant strftime usage: %e and %R replaced with %d and %H:%M respectively * Got rid of variadic macros since some old-school compilers don't like them * compiling with -DNDEBUG disables the DEBUG() macro git-svn-id: https://svn.musicpd.org/mpd/trunk@4512 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* remove clumsy strncpy useEric Wong2006-07-301-18/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | strncpy isn't really safe because it doesn't guarantee null termination, and we have had to work around it in several places. strlcpy (from OpenBSD) isn't great, either because it often leaves errors going unchecked (by truncating strings). So we'll add the pathcpy_trunc() function with is basically strlcpy with a hardcoded MAXPATHLEN as the limit, and we'll acknowledge truncation since we only work on paths and MAXPATHLEN should be set correctly by the system headers[1]. file-specific notes: inputStream_http: eyeballing the changes here, it seems to look alright but I haven't actually tested it myself. ls: don't even bother printing a file if the filename is too long (and when is it ever?) since we won't be able to read it anyways. metadataChunk: it's only metadata, and it's only for showin the user, so truncating it here souldn't be a big issue. memset to zero in init is unecessary, so lets not waste cycles [1] - If the system headers are screwed up, then we're majorly screwed regardless of what we do :x git-svn-id: https://svn.musicpd.org/mpd/trunk@4491 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Add mpd-indent.shAvuton Olrich2006-07-201-258/+253
| | | | | | Indent the entire tree, hopefully we can keep it indented. git-svn-id: https://svn.musicpd.org/mpd/trunk@4410 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Fixing a whopping 4 memory leaksJ. Alexander Treuman2006-07-161-1/+6
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@4371 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* [CLEANUP] Cleanup spaces->tabsAvuton Olrich2006-07-141-87/+87
| | | | | Make functions static where it makes sense git-svn-id: https://svn.musicpd.org/mpd/trunk@4335 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Change shank's email addressJ. Alexander Treuman2006-07-141-1/+1
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@4333 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Remove test on a local variable that won't have been modified yetJ. Alexander Treuman2006-07-141-1/+0
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@4324 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Remove unused playSilenceOrSleep macroJ. Alexander Treuman2006-07-141-6/+0
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@4323 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Remove some misleading comments, fix shanks terrible spellingJ. Alexander Treuman2006-07-141-6/+1
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@4322 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* Huge header update, update the copyright and addAvuton Olrich2006-07-131-1/+1
| | | | | the GPL header where necessary git-svn-id: https://svn.musicpd.org/mpd/trunk@4317 09075e82-0dd4-0310-85a5-a0d7c8717e4f
* merge with mpd/trunk up to r3925Eric Wong2006-03-161-21/+60
| | | git-svn-id: https://svn.musicpd.org/mpd/trunk@3926 09075e82-0dd4-0310-85a5-a0d7c8717e4f