Friday, 2015-10-30

*** uncle-j_j has quit IRC00:09
*** drobilla has quit IRC00:42
*** NickSB2 has joined #lv201:37
*** son0p has quit IRC02:14
*** son0p has joined #lv202:16
*** artfwo has quit IRC02:28
*** drobilla has joined #lv203:15
*** son0p has quit IRC03:58
*** son0p has joined #lv203:59
*** edogawa has joined #lv206:16
*** falktx has joined #lv207:27
*** falktx has quit IRC07:50
*** falktx has joined #lv208:47
*** sigma6 has joined #lv209:04
*** artfwo has joined #lv212:49
*** NickSB2 has quit IRC12:52
*** aombk2 has joined #lv212:59
*** aombk3 has quit IRC13:02
*** ricardocrudo has joined #lv213:21
*** artfwo has quit IRC13:22
*** artfwo has joined #lv213:49
*** artfwo has quit IRC14:41
*** deva has joined #lv216:26
*** sigma6 has quit IRC16:34
*** sigma6 has joined #lv216:37
*** son0p has quit IRC16:41
*** son0p has joined #lv216:43
*** sigma6 has quit IRC16:46
drobillafalktx: I avoided needed a callback notification thing in Lilv after unload by checking if the plugin is valid with lilv_plugin_validate17:05
drobillafalktx: Maybe not ideal, but no new API, and if the bundle has been unloaded the plugin won't be valid anymore, so it works17:06
drobillahttp://dev.drobilla.net/changeset/580817:06
falktxoh nice17:08
falktxdrobilla: just a small question about this. if I do a lilv_world_unload_bundle that contains a plugin, will that plugin still appear when I do lilv_world_get_all_plugins() ?17:09
drobillafalktx: Yes17:09
drobillafalktx: Unfortunately there's no way around that since deleting it might leave dangling pointers in the app17:10
drobillafalktx: Something should still be done about that bit17:10
drobillafalktx: I was just getting it working in Ingen and realized I didn't *need* it, so went head with what worked for now17:10
falktxas I see it, ideally the pointers will be kept valid but the plugin would no longer be part of the get_all_plugins()17:11
drobilla(API at the end of http://drobilla.net/docs/ingen/protocol.html if you still care)17:11
drobillaProbably a lilv_world_reload_plugins() or some such, where the app is responsible for making sure it has no plugin pointers if it calls it.  Maybe also a "zombie" flag17:11
drobillafalktx: I suppose I could still keep them around but in a separate internal list for eventual destruction17:12
drobillaGoing out for a bit, bbl17:12
falktxyes17:12
falktxok, thanks17:12
rgareusdid you guys find out what the idle-load in ingen was about?  the 5% CPU usage on ARM?17:13
*** rncbc has joined #lv217:16
drobillaI trimmed some fat but I don't know about the ARM situation17:16
drobillaIIRC it wasn't an *empty* graph though, so maybe not the same situation even ignoring architecture17:16
drobillaI can't tell any significant difference between it and a trivial Jack client anyway17:17
rgareuson x86 I don't have a measurable idle-load, either.17:17
drobillaIt's odd because client-per-node should have *more* overhead if anything, but I'm not super familiar with how internal clients work so perhaps not17:18
falktxI need to try and load ingen later17:26
falktxdrobilla: something is not quite right with this cpu or compiler, some plugins are consuming heavy amounts of cpu17:26
*** NickSB2 has joined #lv217:39
*** uncle-j_j has joined #lv218:41
*** falktx has quit IRC19:09
*** HarryHaaren has joined #lv219:45
*** son0p has quit IRC19:48
*** son0p has joined #lv219:50
*** falktx has joined #lv220:00
*** deva has quit IRC20:58
*** artfwo has joined #lv221:05
*** drobilla has quit IRC21:17
*** rgareus is now known as rgareus|afk21:26
*** uncle-j_j has quit IRC22:18
*** son0p has quit IRC23:04
*** son0p has joined #lv223:06
*** HarryHaaren has quit IRC23:27
*** edogawa has quit IRC23:56

Generated by irclog2html.py 2.13.0 by Marius Gedminas - find it at mg.pov.lt!