summaryrefslogtreecommitdiffstats
path: root/src/pulsecore
Commit message (Collapse)AuthorAgeFilesLines
* memblock: make it easy to disable mempool usage with $PULSE_MEMPOOL_DISABLELennart Poettering2009-09-171-0/+7
|
* core: add an additional volume factor that is applied after resampling took ↵Lennart Poettering2009-09-112-10/+56
| | | | place
* native: send PA_COMMAND_PLAYBACK_BUFFER_ATTR_CHANGED messages only to ↵Lennart Poettering2009-09-101-13/+15
| | | | clients that understand it
* simd: be more precise which SIMD optimizations we activateLennart Poettering2009-09-095-16/+28
|
* sample-util: add pa_convert_size() call for converting sizes between two ↵Lennart Poettering2009-09-092-0/+12
| | | | sample specs
* ratelimit: allow non-static ratelimit structsLennart Poettering2009-09-091-4/+14
|
* protocol-native: log explicitly when someone asks us to quitLennart Poettering2009-09-081-0/+2
|
* core-util: properly fill in exception array for pa_reset_sigs() ↵Lennart Poettering2009-09-081-1/+1
| | | | (llvm-clang-analyzer)
* conf-parser: properly initialize variable we free() later (llvm-clang-analyzer)Lennart Poettering2009-09-081-1/+1
|
* cli-command: don't necessarily ovveride failure code of files ↵Lennart Poettering2009-09-081-2/+0
| | | | (llvm-clang-analyzer)
* cli-command: modernizationsLennart Poettering2009-09-081-16/+16
|
* llvm-clang-analyzer: drop a few unnecessary assignments and other trivial fixesLennart Poettering2009-09-089-29/+30
|
* Merge remote branch 'wtay/optimize'Lennart Poettering2009-09-073-9/+12
|\
| * x86: also call see init for SSE2Wim Taymans2009-09-071-1/+1
| |
| * x86: only install some functions when SSE2Wim Taymans2009-09-072-8/+11
| | | | | | | | | | Remap and volume functions use SSE2 instructions so only install them when SSE2 is present.
* | vector: don't try to build vector stuff on altivecLennart Poettering2009-09-071-1/+2
| |
* | volume: fix definition of PA_VOLUME_MAX and introduce PA_VOLUME_INVALID and ↵Lennart Poettering2009-09-071-2/+2
| | | | | | | | use it wherever applicable
* | libpulse: introduce PA_BYTES_SNPRINT_MAX and make use of it wherever applicableLennart Poettering2009-09-062-7/+7
|/
* remap: build sse code only on x86v0.9.16-test7Lennart Poettering2009-09-031-0/+2
|
* core: drop unnecessary variable initializationLennart Poettering2009-09-031-1/+1
|
* remap_sse: fix inner loop increment on SSELennart Poettering2009-09-031-2/+2
| | | | | In each iteration we can process 2^4 S16NE samples and 2^5 F32NE samples, that's twice as much as in MMX, hence correct the increments.
* remap_sse: reindent macro so that diff to MMX is nicerLennart Poettering2009-09-031-19/+19
| | | | | Completely useless, but diff -u remap_mmx.c remap_sse.c is much nicer this way.
* sconv_sse: fix leftover counterWim Taymans2009-09-021-2/+2
|
* Merge branch 'master' of ssh://rootserver/home/lennart/git/public/pulseaudioLennart Poettering2009-09-021-0/+10
|\
| * core-util: call dbus_connection_set_exit_on_disconnect() on shared busses to ↵Lennart Poettering2009-09-021-0/+10
| | | | | | | | make sure dbus_shutdown() isn't fatal
* | core-util: don't leak memory in pa_unset_env_recorded()Lennart Poettering2009-09-021-2/+4
|/
* sink: simplify pa_sink_render_full() by replacing it by a pa_sink_render() ↵Lennart Poettering2009-09-021-76/+14
| | | | plus a couple of pa_sink_render_full()
* daemon: clean up environment when forking off childrenLennart Poettering2009-09-021-0/+5
|
* core-util: add api for setting env vars and record them so that we can undo ↵Lennart Poettering2009-09-022-0/+32
| | | | them n fork
* core: fill up memblock with pa_sink_render_into_full() in ↵Lennart Poettering2009-09-011-11/+5
| | | | pa_sink_render_full() instead of doing our own loop
* core: handle suspended state in pa_sink_render_full() similar to the other ↵Lennart Poettering2009-09-011-1/+10
| | | | render functions
* core: add missing sink_unref()Lennart Poettering2009-09-011-0/+3
|
* remap: fix counters for mmx and sse remapWim Taymans2009-08-312-26/+42
| | | | | Take the size of the sample into account when calculating the amount of samples we process in parallel.
* cli: make sure 'dump' uses pa_cvolume_max() to deduce a single-channel ↵Lennart Poettering2009-08-311-2/+2
| | | | volume from a multi-channel volume
* cli: apply single-channel volume changes equally to all channelsLennart Poettering2009-08-311-3/+3
|
* core: always allow volume setting with single-channel pa_cvolumeLennart Poettering2009-08-314-10/+33
|
* native: make sure clients cannot trigger an assert by sending us invalid ↵Lennart Poettering2009-08-291-1/+6
| | | | volume info
* core: initialize sink/source priorities automatically based on their proplistsLennart Poettering2009-08-293-0/+50
|
* namereg: select default sink by priorityLennart Poettering2009-08-291-8/+23
|
* protocol-native: replace use of pa_namereg_is_valid_name() by ↵Lennart Poettering2009-08-291-10/+15
| | | | pa_namereg_is_valid_name_or_wildcard() where applicable to allow use of @@ wildcards
* namereg: add new pa_namereg_is_valid_name_or_wildcard() callLennart Poettering2009-08-292-1/+22
|
* core: add priority field to pa_sink/pa_sourceLennart Poettering2009-08-285-0/+10
|
* sconv: quieten gcc a bitLennart Poettering2009-08-281-3/+1
|
* core: move 'flags' field into 'pa_sink_input_new_data' structure so that ↵Lennart Poettering2009-08-2810-35/+37
| | | | hooks can access it
* Fix checking for NULL after usageAndy Shevchenko2009-08-283-6/+3
| | | | | | The pa_xmalloc calls oom() in case of NULL pointer returned by malloc() on one hand and dereferencing of pointer is happen early than actual check on other hand. Thus, just remove useless checks.
* core-util: Fix logic of pa_make_path_absolute()Andy Shevchenko2009-08-281-6/+6
| | | | | Make it works as described in commentary (when fn is NULL, pa_assert(fn) could break logic).
* convert: add sse/sse2 s16 to float32ne conversionsWim Taymans2009-08-273-0/+238
|
* macro: add macro to align variablesWim Taymans2009-08-271-0/+6
|
* remap: cleanup assembler a littleWim Taymans2009-08-272-29/+26
|
* sample-util: avoid stack overrunWim Taymans2009-08-271-2/+2
| | | | | The linear volume array needs enough padding when preparing the volumes for mixing.