| T42_ | <adampigg> what are the rules for if audiosystem-passthrough-dummy-af is needed | 14:47 |
|---|---|---|
| T42_ | <Mister_Magister> if audioflinger is needed | 14:47 |
| T42_ | <Mister_Magister> you'll see logcat yelling at you that it wants audioflinger | 14:49 |
| T42_ | <adampigg> ta, will check .. currnetly it fails, but might not be needed ... audio is the next task. wlan, bt, sensors are all working | 14:50 |
| Mister_Magister | you might need the… | 14:50 |
| Mister_Magister | https://github.com/VerdandiTeam/droid-config-pipa/commit/20f47aca91fe38758847d8e1f08e2ec246ac1657 | 14:51 |
| Mister_Magister | its common issue | 14:51 |
| T42_ | <adampigg> ta | 14:57 |
| T42_ | <adampigg> @Mister_Magister looks like the also usual issue of the config parser not understanding the later format xml | 15:45 |
| T42_ | <Mister_Magister> dunno what issue you're referring to | 15:47 |
| mal | @adampigg I have some local changes for pulse droid module | 16:01 |
| mal | some missing AUDIO_OUTPUT_FLAG_* stuff in case that is the issue for you | 16:01 |
| T42_ | <adampigg> mal: no, the issue is that the parse seems to expect , seperator for eg formats field, but android now uses a space | 16:11 |
| T42_ | <adampigg> last few ports ive hand-edited the xml and bind-mounted them over in /vendor | 16:11 |
| T42_ | <adampigg> mal: looks like there are some commits in pa modules droid to allow a space sep, but maybe not everywhere? | 16:18 |
| T42_ | <adampigg> conversion.c: [/vendor/etc/bluetooth_audio_policy_configuration.xml:9] Bad sample rate value 24000 16000 | 16:18 |
| T42_ | <adampigg> mal: looking at the code, it should work, need to check version | 16:46 |
| mal | @adampigg correct repo i.e. droid vs droid-jb2q? | 16:49 |
| mal | rinigus: please push that suspend service patch to hybris-patches if it can exists there | 16:51 |
| rinigus | mal: absolutely. will do tonight | 16:53 |
| rinigus | mal: and if there is an interest to checkout MCE patch, please do. but I thought I will close it as it is not going to be needed right now. so, it makes sense not to bother you with it | 16:54 |
| rinigus | mal: PR submitted | 18:00 |
| T42_ | <adampigg> mal: this version i | pulseaudio-modules-droid | package | 14.2.102.2-1.1.6.bso | aarch64 | adaptation-halium | 18:10 |
| mal | rinigus: where is that mUseSuspendCounter set to true? it comes from constructor but what calls it | 18:10 |
| rinigus | mal: just a sec, I think it was a control service. let me check | 18:12 |
| T42_ | <adampigg> ah, 14.2.102 is quite old! | 18:13 |
| rinigus | mal: line 153 for me in main.cpp in the same folder | 18:13 |
| mal | @adampigg which halium project? | 18:14 |
| T42_ | <adampigg> itll be 12 .. updating now | 18:14 |
| mal | the fork is behind | 18:14 |
| mal | and missing the space changes | 18:15 |
| T42_ | <adampigg> yeah | 18:16 |
| mal | rinigus: just wondering how that would work is you just change that to false, since the else branches use wakelockfd | 18:16 |
| mal | but of course that is a bit different otherwise | 18:17 |
| rinigus | mal: not very well, I suspect. based on my debugging using extra logs, I have seen that the same wakelock could be acquired multiple times. for example by qcril. if you swap that var, it will not be supported properly | 18:18 |
| T42_ | <adampigg> it was actually 13, but same ver | 18:18 |
| mal | rinigus: ok | 18:18 |
| T42_ | <adampigg> mal: sound working | 18:25 |
| mal | @adampigg what is the reason for the fork? | 18:32 |
| T42_ | <adampigg> mal: see the patch in the repo | 18:32 |
| T42_ | <adampigg> its form UT | 18:32 |
| T42_ | <adampigg> *from | 18:32 |
| rinigus | with nagara we have an annoying bug where camera could basically lockup the device. that happens if you have viewfinder active and switch off the screen. | 18:57 |
| rinigus | on tama, I could see that in such case some HAL errors appear and all is stopped. I suspect that it doesn't happen in nagara and camera tries to battle through suspend windows. | 18:57 |
| rinigus | all will be fine if I open a wakelock. | 18:57 |
| rinigus | but I wonder, if it is now responsibility of apps to close camera stream on switching screen off, droidmedia, or something else. | 18:57 |
| *** Mister_Magister_ is now known as Mister_Magister | 19:37 | |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!