| sledges | PSA: Sailfish OS community virtual meeting at #mer-meeting in just under 15mins | 07:46 |
|---|---|---|
| NeoChapay | https://twitter.com/neochapay/status/1174599143907610624 hehe | 08:46 |
| Mister_Magister | hehe | 08:59 |
| mal | NeoChapay: did you make some additional things to qmimodem ofono support so sfos sees the modem status, other than voice call support you had in the PR | 09:41 |
| NeoChapay | mal: i backported it from upstream https://git.merproject.org/mer-core/ofono/merge_requests/233/diffs | 09:43 |
| mal | NeoChapay: I meant the sailfish manager support which is missing from that qmimodem support | 09:45 |
| NeoChapay | mal: many dbus methods who heeded by sfos | 11:36 |
| mal | NeoChapay: I have plans to try to add support for that sailfish manager to qmimodem | 11:41 |
| NeoChapay | mal: i can check full functionality becouse i broke ts wire | 11:44 |
| mal | NeoChapay: can or can't? | 11:45 |
| NeoChapay | can't :) | 11:46 |
| T42 | <srLuxint> Hey guys. Do you by any chance know how exactly voice data is transported within ofono framework during voice call? I mean where can I get the raw data samples or whatever is transported over the radio. | 15:10 |
| monich | srLuxint: that's not ofono's business, ofono only manages modem's control channels | 15:13 |
| monich | the same with data calls, ofono can only activate/deactivate those - to capture the actual data you would have to use methods that are not related to ofono at all | 15:15 |
| T42 | <srLuxint> @monich [the same with data calls, ofono can only activ …], Oh. Seems to be the reason why I can't find anything to mention about voice data. So do you then know where the roots of the voice data path come from? At least any clue? | 15:19 |
| monich | no, not really :/ | 15:20 |
| monich | my guess would be as good as yours | 15:20 |
| T42 | <srLuxint> :( Sad. Ok. Let me ask then one more question. Probably some one aware of the issue with this audioflingerglue thing? Can the voice part of the sound subsystem work without it? The deal is I have PA working well and creating a sink-input for voice stream and printing same logs on the target device as on the working one except of some | 15:26 |
| T42 | afglue plugin callback trigger-messages which seem to relate to voice stream, e.g. call state or call type. | 15:26 |
| T42 | <srLuxint> https://elinux.org/images/6/61/Elc_telephony_piirainen_0.pdf page 8. | 15:43 |
| T42 | <srLuxint> Oh. It seems like my investigation has come to an end at some point. "Cellular audio and Bluetooth audio streams stay in the SoC side ...". Which makes me dive into PA droid-* plugins I guess. | 15:43 |
| mal | @srLuxint what device are you doing? | 16:03 |
| T42 | <srLuxint> mal: It's a custom Qualcom based device. So no useful 'what device' type of info I'm afraid. Thus I'm fine with some high-level insights type of help. :D | 16:11 |
| mal | @srLuxint which android base? | 16:12 |
| T42 | <srLuxint> 8.0. | 16:13 |
| mal | @srLuxint does the device have vendor.qti.hardware.radio.am hidl service? | 16:14 |
| mal | not exactly sure how to check that | 16:15 |
| T42 | <srLuxint> Not any running at least. Neither the vendor hw folder has any binary too. | 16:17 |
| mal | not quite sure what provides that, it's not running directly, could be some extension somewhere | 16:18 |
| mal | the full hidl interface is vendor.qti.hardware.radio.am@1.0::IQcRilAudio | 16:18 |
| T42 | <srLuxint> Oh. So this is the one that is responsible for cellular audio data? | 16:19 |
| mal | that is similar to afglue but for hidl based devices | 16:19 |
| mal | at least of those | 16:19 |
| mal | yes, that is part of it but just a small part | 16:20 |
| mal | what exactly are you trying to do? | 16:20 |
| T42 | <srLuxint> Oh, yeah. I've forgot to mention the very issue. :D Voice data is neither played on device nor send from device. Yet the sound subsystem (x-maemo role stuff at least) works fine. Also the DTMF seems to work which points me to some routing (I guess?) problem. | 16:23 |
| mal | and you have properly built audioflingerglue and pa afglue plugin, if so and it still fails then maybe try https://github.com/mer-hybris/pulseaudio-modules-droid-hidl | 16:25 |
| mal | that is replacement for afglue | 16:25 |
| mal | needed at least on some android 9 devices, not sure about android 8 | 16:26 |
| mal | if you try that remove the pa afglue plugin first | 16:26 |
| T42 | <srLuxint> Thanks! Much appreciated! I'll try it. Also is there any info on how exactly does it work? I mean I have no idea on what and in fact how it happens with the miniafservice and how does it relate to PA droid-glue module. | 16:29 |
| mal | @srLuxint droid-glue is loaded automatically, same way as that hidl plugin | 16:37 |
| mal | it's loaded from pa droid plugin using this https://github.com/mer-hybris/droid-hal-configs/commit/edfbf7cb81e0f609eddb12ccefec83a6a0756765 | 16:37 |
| T42 | <srLuxint> Yeah, I'm totally familiar with PA config. I've been watching through it for hours. :D My question is more about high-level concepts that stay behind this exact (and related ones) piece of software. | 16:46 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!