330 Commits (80c0630e08fc0bbc6743bc2a20a13df01cf209c2)

Author SHA1 Message Date
  torben fb92210579 [netjack] fix MSG_CONFIRM 17 years ago
  paul b4c875b9fb correct memops mess (there were two instances of the header file, now there is just one, in jack/memops.h (not installed)) 17 years ago
  torben a166e4732d [netjack] add -D mode alsa_io friend, also change initial deadline to 1period 17 years ago
  paul 167ff61aac (1) configure-time detection of likely-to-collide existing JACK installs (2) do not link libjack against libraries needed only for netjack components 17 years ago
  torben 93cec3f0f7 switch to using jack_get_time instead of jack_get_microseconds 17 years ago
  torben 5d0523107b [netjack] fix warnings about unused variables without libsamplerate 17 years ago
  torben 643385081d [netjack] Fix number packet_cache for 0 payload case. 17 years ago
  torben 055826a2f2 [netjack] Fix Redundancy mode. 17 years ago
  torben 3f9efbad42 [netjack] Add -H option to turn off ntoh of payload data. reduces CPU load. 17 years ago
  torben eb44cb0f65 [netjack] Compensate transport also for codec latency. should be sample accurate over internet now. 17 years ago
  torben 9643ffbb40 [netjack] support building without libsamplerate 17 years ago
  torben cd4ee518dc [netjack] improve deadline handling, to recover from running too fast. 17 years ago
  torben 0573556d33 [netjack] Reduce dynamic deadline adjustement speed again. now it seems to really work. 17 years ago
  torben 157a656703 [netjack] fix master verification.... doh. 17 years ago
  paul dc837ebbd6 netjack-to-compile fixes for OS X 17 years ago
  torben 1ba12ae623 [netjack] Adjust deadline faster. 17 years ago
  paul 59263ef27a fix reversed naming of ALSA seq MIDI ports 17 years ago
  paul 606ab3652d torben's latest netjack code: dynamic deadline handling for packet loss; redundancy mode; fix memory corruption with celt; dont poll, when packet we want is already there; remove old framecnts from cache. works ... osort of :) 17 years ago
  paul 1878898b2b fixup removal of sysv SHM registry @ make install time (packagers take note); remove port delete/create messages from ALSA MIDI backend code 17 years ago
  paul b578ea2f0e fix dynamic SIMD for OS X ; add config.h to netjack driver code ; make a configure.ac comment about our inability to determine an -mFOO argument for gcc on OS X 17 years ago
  paul 9c62615486 "zero-latency" patch for netjack from torben 17 years ago
  paul 71a512249f more netjack driver compilation tweaks 17 years ago
  paul ac1f7aedcb compile and configuration fixes from OSX but useful everywhere 17 years ago
  paul ae0ea1f84a support OS X with new netjack code, plus a small netjack fix/improvement 17 years ago
  paul 66ec6fb249 support older versions of ALSA API in the ALSA MIDI backend 17 years ago
  paul de8dba1ff6 netjack changes from torben (packet loss handling leads to "dummy" backend-like behaviour; jitter buffer), along with smaller mods to alsa_{in,out} 17 years ago
  paul 0abc695b0e optional CELT support in netjack; fixes for a resampling issue in alsa_in and alsa_out; remove debug output from alsa_{in,out} 17 years ago
  paul e700090da7 cast float values to uint16_t before passing to htons() in netjack 17 years ago
  pieterpalmers 71674f0cc1 clean up the firewire backend a bit 17 years ago
  paul b43a97f961 simon jenkin's fix for float->16bit dithering 17 years ago
  pieterpalmers e6b3c05524 adapt the firewire driver for the thread prio changes from r3051 17 years ago
  paul 12dc1a6c66 shift all possible 24-in-32 values right, not just scaled ones 17 years ago
  paul 0c13c3f77c provide support for 24bit, not just 24-in-32 format 17 years ago
  paul bf5acc20a2 macro-ize the core conversions and fix mistakes in dithering cases 17 years ago
  paul ecd674e315 machete-technique at its finest: rather than think about merge functions, just remove them all because they are never used 17 years ago
  paul 2e10e088d0 finish what i started with earlier transformations - several functions were missed; fix up merge functions to avoid overflow; follow fons' advice and make min/max values symmetrical 17 years ago
  paul 898f36b317 next iteration of a cleanup of sample conversions. SUBJECT TO REVIEW. TEST. DO NOT USE FOR IMPORTANT INSTALLATIONS 17 years ago
  paul b5b34c4403 no more dither when converting from float to 24 bit values - i am not sure how this was added to JACK but it was a mistake 17 years ago
  paul cc0f5a221a use stdint types, not long long or int, for the intermediate holder of float->int sample type conversion 17 years ago
  paul 7a8979e059 remove unnecessary casts from return value of llrintf to long long 17 years ago
  paul 71c5bbc396 use llrintf instead of lrintf when rounding to long long; use llrintf when converting to 24bit, rather than a cast; document the structure of the names in memops.c since they are rather mysterious 17 years ago
  paul b912519147 more consistent naming of MIDI ports (vs. audio); drop use of ALSA seq client ID in names since it is not reliable or repeatable 17 years ago
  paul e78a19dafc alsa sequencer bridge work: correctly label only HW/PORT sequencer ports as Physical|Terminal (fixes renaming as system); change naming scheme of ports to show client names and not silly duplicate alsa-seq info; add port aliases so that connection via recognizable names is possible. Long term, a change to the client name would be appropriate - these are not alsa_pcm ports 17 years ago
  paul 69039cae1c fix export of pointers into JACK API headers, thus allowing a 64 bit JACK server to support 32 bit clients and vice versa (work done primarily by Torben, with a few cleanups from Paul) 17 years ago
  sletz 49e55b7223 Fix alsa driver, compiles again. 17 years ago
  sletz c17e21ee12 Florian Faber native 32 bits float patch for alsa driver. 17 years ago
  pieterpalmers b95cb85b80 remove leftover reference code chunk 18 years ago
  pieterpalmers 2cf4a791ee fix failure handling on NT driver start 18 years ago
  pieterpalmers 60532cf545 make dummy backend use clock_nanosleep when available. gives much better timing accuracy 18 years ago
  marco 15410b496a * Applied Mike Taht's fix for proper alsa xrun duration measurement 18 years ago