*** Newami has joined #automotive | 01:12 | |
*** Newami has quit IRC | 01:19 | |
*** fredw has joined #automotive | 07:31 | |
*** khem has quit IRC | 07:50 | |
*** jacobo has joined #automotive | 07:51 | |
*** khem has joined #automotive | 07:54 | |
*** kbroulik has joined #automotive | 08:21 | |
*** vrubiolo has joined #automotive | 08:29 | |
*** jobol has joined #automotive | 08:32 | |
*** skz81 has joined #automotive | 08:37 | |
*** jaillet has joined #automotive | 08:41 | |
*** leon-anavi has joined #automotive | 09:01 | |
gkiagia | update on halibut: I rebuilt everything last night and I have a working image now. everything seems to work except for the mixer... I'm checking that now | 09:04 |
---|---|---|
*** walzert has joined #automotive | 09:10 | |
gkiagia | hmm, mixer works on second boot... wonder what's wrong there | 09:14 |
*** skz81 has quit IRC | 09:17 | |
walzert | good morning, i made a better version of the video I showed 2 weeks ago in the developer meeting. will post the link in the wiki too https://www.youtube.com/watch?v=eED8VNr5Mio&list=UU3YkYlPvipLx4AnLzxKy2Fg | 09:19 |
*** skz81 has joined #automotive | 09:24 | |
gkiagia | dl9pf: bluetooth doesn't block the ui on halibut either (h3 + kf) | 09:25 |
gkiagia | I've noticed that the audiomixer takes a while to initialize... and during that time, the homescreen and the mixer app throw out a lot of connection errors... it seems to happen before the backend is initialized, so I suspect it waits for the signal composer API for a while | 09:28 |
gkiagia | also happens on master | 09:28 |
*** fredw has quit IRC | 09:43 | |
*** fredw has joined #automotive | 09:44 | |
*** fredw has quit IRC | 09:48 | |
*** kbroulik has quit IRC | 10:35 | |
*** jaillet has quit IRC | 11:06 | |
*** vrubiolo has quit IRC | 11:13 | |
*** vrubiolo has joined #automotive | 11:16 | |
*** tgamblin has quit IRC | 11:46 | |
*** Newami has joined #automotive | 11:52 | |
*** tgamblin has joined #automotive | 12:38 | |
*** walzert has quit IRC | 12:50 | |
smurray | gkiagia: signal composer does seem to be a bit slow to start. I may try setting it as auto-start to try to parallelize things | 13:22 |
smurray | gkiagia: things do look better, I have some changes for the radio and mediaplayer bindings to make corking tracking more robust. AFAICT you always have to switch to PAUSED state on REQUEST_STATE in order to get the follow up request for PLAYING, so I added the logic to mediaplayer. Otherwise, it thinks it's corked forever after the first time. | 13:25 |
gkiagia | alright, good to know | 13:26 |
smurray | gkiagia: it takes a bunch more logic to handle corking than simply stopping mediaplayer <-> radio, I'm starting to think it's more trouble than it's worth | 13:30 |
smurray | gkiagia: but that's probably a post-CES discussion ;) | 13:31 |
gkiagia | I guess it depends on your UX expectations | 13:33 |
gkiagia | I'd love to have this discussion | 13:33 |
smurray | gkiagia: also, to fix the alsasrc issue, I resorted to taking the pipeline to NULL instead of READY. I definitely think the current policy behavior wrt endpoints will receive heavy pushback from desktop folks, I sure wouldn't want to do this kind of hacking on more than the two or three apps we need | 13:36 |
*** kbroulik has joined #automotive | 14:21 | |
smurray | gkiagia: I'm guessing the answer is no, but can you think of a way currently to tell if you're being corked because of something on another role, or on the same role? | 15:37 |
gkiagia | there is no way | 15:38 |
smurray | gkiagia: okay, that's what I figured | 15:47 |
gkiagia | smurray: dl9pf I just pushed 2 changes that fix bluetooth audio in the media player: you can now switch back from bluetooth to local audio without issue and you can interrupt media from the phone with alexa or navigation messages | 16:07 |
smurray | gkiagia: sweet, I'll rig up a build to do some tests | 16:07 |
smurray | gkiagia: I guess the phone ringtone playing is the only thing we know of that still needs tweaking? | 16:08 |
gkiagia | calls should also work, in theory, but I had bad luck with my test just now (I only got 1 direction audio - from the mic to the phone - and after the call ended, my phone completely disconnected the HFP profile) | 16:09 |
smurray | okay, good to know | 16:10 |
gkiagia | the ringtone still needs tweaking, yes | 16:10 |
smurray | I'll take a look in the next day or two | 16:10 |
gkiagia | I'm sure the media pipeline from the phone to the speakers is working properly (since we get audio in the media player with a2dp), so I'm suspecting my android might not be behaving correctly (there are some known issues with bt in this version of lineage os, so...) | 16:12 |
smurray | gkiagia: I've got an iphone here, will try when I get a chance | 16:12 |
gkiagia | FYI: I pushed the commits for master, but I actually tested on halibut with the backport (because that's what I had most ready at hand) | 16:14 |
smurray | gkiagia: okay, that's how I've been testing as well (patched version of dl9pf 's sandbox) | 16:15 |
gkiagia | next in my list for tomorrow is documentation for the config, unless you have something you want me to look at | 16:17 |
smurray | gkiagia: I'll do some testing today and see if anything shakes out | 16:18 |
*** leon-anavi has quit IRC | 16:24 | |
*** jacobo has quit IRC | 16:33 | |
*** kbroulik has quit IRC | 16:36 | |
*** psilva has quit IRC | 17:02 | |
*** jobol has quit IRC | 17:27 | |
*** skz81 has quit IRC | 17:48 | |
*** vrubiolo has quit IRC | 20:13 | |
*** tgamblin has quit IRC | 20:51 | |
*** Newami has joined #automotive | 22:53 | |
*** Newami has quit IRC | 23:03 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!