summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* If we cannot open an ALSA device with SND_PCM_NO_AUTO_FORMAT retry withoutLennart Poettering2008-12-231-4/+24
| | | | | This should fix compatibility with some sound cards which only support 24 bit packed samples.
* make sure we don't hit an assert when we issue two rewind requests in aLennart Poettering2008-12-211-17/+25
| | | | | | single iteration Closes rhbz 472757.
* Implement PA_SOURCE_MESSAGE_GET_LATENCYLennart Poettering2008-12-211-0/+30
| | | | | | | The pipe source needs to implement PA_SOURCE_MESSAGE_GET_LATENCY since we'll otherwise hit an assert. Closes #424.
* Use FIONREAD instead of TIOCINQLennart Poettering2008-12-211-2/+2
| | | | | | | On Linux the two ioctls FIONREAD and TIOCINQ share the same number. However TIOCINQ is used for terminals while FIONREAD is used (among others) for sockets and pipes. Hence use FIONREAD here since it seems more appropriate.
* Allow access("/dev/dsp", W_OK) succeedLennart Poettering2008-12-181-5/+5
| | | | | | | Some stupid apps (such as X-lite Softphone) check for W_OK on /dev/dsp. This is of course crazy but easy for us to support. Closes rhbz #474313
* Make sure we drop CAP_NICE if RT is not allowedLennart Poettering2008-12-181-7/+39
| | | | | but make sure we still allow RT if RLIMIT_RTPRIO is properly set when PA is called.
* downgrade a few warningsLennart Poettering2008-12-171-4/+4
|
* it's better to always use the index of a module instead of the pa_module*Lennart Poettering2008-12-171-13/+15
|
* unload tunnel modules from a new stack frameLennart Poettering2008-12-171-2/+2
|
* actually unload the modules from a new stack frameLennart Poettering2008-12-171-1/+1
|
* Don't store pointer to pa_moduleLennart Poettering2008-12-171-3/+3
| | | | | pa_module pointers might become invalid at any time, so we use the stable uin32_t index of the module for identifying or modules instead.
* introduce new function pa_module_unload_request_by_indexLennart Poettering2008-12-172-2/+14
|
* Fix bug walking on module list.Luiz Augusto von Dentz2008-12-171-1/+1
|
* Don't hit an assert when checking for idlenessLennart Poettering2008-12-174-5/+18
| | | | Closes #398
* Load module-volume-restore and module-device-restore before all other modulesLennart Poettering2008-12-171-4/+4
| | | | | | Since m-v-r and m-d-r become active in the FIXATE hooks for stream/device creation we need to make sure the modules are loaded before the first devices/streams are created.
* Pass GDBM_NOLOCK to gdbmLennart Poettering2008-12-172-2/+2
| | | | | | | | | | gdbm still uses BSD lockf for locking which is incompatible with NFS. Since we don't need the locking here since the db files in question should never be accessed by more than one process since they are per-home-dir and per-machine we can disable locking without any ill results. This should fix rhbz #471279.
* Don't mix front-center into rear channelsLennart Poettering2008-12-171-3/+93
| | | | | | | | If there's a center channel on input that is not available on output make sure we mix front-center only into front-left/right and rear-center into rear-left/right. Closes #400
* fix implementation of bind now ltdl loader for libtool 2.2Lennart Poettering2008-12-161-26/+31
|
* Fix return value of --dump-resample-methodsLennart Poettering2008-12-161-0/+1
| | | | | | | | We should return 0 when --dump-resample-methods is called. Problem identified by "hchen59". Closes bug #407.
* libtool 2.2 updatesLennart Poettering2008-12-162-5/+6
|
* Check if environ is actually set before we use itLennart Poettering2008-12-161-14/+21
| | | | | | | | Some applications seem to set **environ to NULL for various reasons. Before we iterate through it we thus need to make sure it is actually set to make sure we don't segfault. Closes rhbz #473080
* Merge commit 'flameeyes/libtool-2.2'Lennart Poettering2008-12-163-81/+5
|\
| * Update code to use libtool 2.2. Use convenience recursive libltdl.Diego 'Flameeyes' Pettenò2008-08-083-81/+5
| | | | | | | | Also remove the code for older libtool from the daemon.
| * Create an m4 directory for common macros and use it.Diego 'Flameeyes' Pettenò2008-08-088-42/+49
| | | | | | | | | | | | | | Split acinclude.m4 in multiple macro files. Let it be known to autoconf and aclocal to use the m4 directory. Ignore macro files copied or linked by libtool and intltool.
| * Remove gettext macros from configure.ac, intltool is used.Diego 'Flameeyes' Pettenò2008-08-081-2/+0
| | | | | | | | | | When using intltool there is no need to _also_ use gettext, so remove gettext macros so that autoreconf does not copy them over.
| * Bump autoconf requirement to 2.62 (latest released version).Diego 'Flameeyes' Pettenò2008-08-071-1/+1
| |
* | make shm marker architecture independant, patch from michich, closes #401Lennart Poettering2008-11-011-4/+4
| |
* | Make sure libpulse never gets unloadedLennart Poettering2008-11-011-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When an .so is unloaded during runtime all TLS keys it has registered need to be freed because the destructor callbacks of the TLS key might otherwise point to invalid code. Hence it would appear sensible to destruct the TLS keys from a function marked as __attribute__ ((destructor)). However functions marked like that are also called when an application terminates, on exit(). If a thread continues to run until the very exit it might still want to access that TLS data. The destructor functions are called while all other threads are still running. If __attribute ((destructor)) is used to destruct TLS keys for such threads this might hence cause a crash when the application shuts down. To circumvent this problem we'll now compile libpulse with -z nodelete, to make it unnecessary to delete the TLS data ever and thus avoiding the problem. It's suboptimal, but for now I see no better solution.
* | a bit of pa_bool_t'izationLennart Poettering2008-11-012-6/+7
| |
* | Implement new flags DONT_INHIBIT_AUTO_SUSPEND and START_UNMUTEDLennart Poettering2008-10-2613-24/+156
| |
* | bump micro versionLennart Poettering2008-10-251-1/+1
| |
* | Merge branch 'new-world-order'Lennart Poettering2008-10-256-545/+299
|\ \ | | | | | | | | | | | | Conflicts: src/Makefile.am
| * | a lot more build system updatesLennart Poettering2008-10-225-128/+126
| | |
| * | make new build logic actually workLennart Poettering2008-10-222-60/+65
| | |
| * | update map fileLennart Poettering2008-10-221-1/+3
| | |
| * | temporary commit to allow flameeyes a lookLennart Poettering2008-10-212-511/+259
| | |
| * | reorder setting of AM_CFLAGS a bitLennart Poettering2008-10-211-20/+21
| | |
* | | make sure to use 64bit rounding even on 32bit machines when converting to ↵Lennart Poettering2008-10-241-3/+3
| | | | | | | | | | | | pa_usec_t
* | | warn if ALSA wakes us up and there is actually nothing to doLennart Poettering2008-10-223-22/+66
| | |
* | | don't set the volume of pacat unless it is explicitly setLennart Poettering2008-10-221-1/+3
| | |
* | | update map fileLennart Poettering2008-10-221-1/+3
| | |
* | | reorder setting of AM_CFLAGS a bitLennart Poettering2008-10-221-20/+21
|/ /
* | make the debug trap macro a proper macro in macro.hLennart Poettering2008-10-211-0/+6
| |
* | Try to catch certain driver errorsLennart Poettering2008-10-214-6/+69
| | | | | | | | | | ... by verifying return values of snd_pcm_avail_update() and snd_pcm_begin_mmap() for their sanenness.
* | make log meta, time, backtrace configurable using command lineLennart Poettering2008-10-211-0/+27
| |
* | Make log meta, time, backtrace configurable using config fileLennart Poettering2008-10-214-74/+70
| |
* | support changing logging parameters during runtime using the CLILennart Poettering2008-10-211-0/+104
| |
* | Support showing a backtrace on log messagesLennart Poettering2008-10-213-16/+125
| |
* | include log.h near the end so that macro.h can be included in log.h and ↵Lennart Poettering2008-10-211-1/+3
| | | | | | | | defines pa_bool_t properly
* | add pa_config_parse_unsigned()Lennart Poettering2008-10-212-0/+19
| |