| freemangordon | ugh, got it working, the only missing piece was stream format | 08:04 |
|---|---|---|
| freemangordon | Wizzup: any idea why we have capture and playback streams on voice? what are they used for? | 08:08 |
| arno11 | freemangordon: cool, btw how did you add the stream format ? directly in the dtsi ? | 10:58 |
| arno11 | (just out of curiosity) | 10:58 |
| uvos | freemangordon: you mean the streams in pa? | 10:59 |
| uvos | because pa has no concept of configurations not being assoicated with a stream | 10:59 |
| uvos | so you can not create any usecase configuration not associated with a stream | 10:59 |
| uvos | essentally the same problem as dapm just in userspace | 11:00 |
| fmg_d4 | stream format can't be set iiuc | 11:10 |
| fmg_d4 | i hacked the port formatin the code | 11:10 |
| fmg_d4 | but will try to find the proper way to set it through dt | 11:11 |
| uvos | so by stream you ment the hw streams ie the voide dai in cpcap not pa's streams? | 11:12 |
| fmg_d4 | yes | 11:12 |
| fmg_d4 | hi! | 11:12 |
| fmg_d4 | ater on I will need your help to make that work prperly | 11:13 |
| fmg_d4 | but, i have calls workin on 6.6 | 11:14 |
| fmg_d4 | with all hacks removed | 11:14 |
| fmg_d4 | besides i2s i set on voice dai | 11:15 |
| fmg_d4 | argh, my d4 kbd is next to useless | 11:15 |
| uvos | whats wrong with it? | 11:15 |
| fmg_d4 | it misses 10% of the presses | 11:16 |
| uvos | nice, its pretty unlikely i will be of any use given my knowlage of alsa but i will try. Hw wise the voice dai essentally just needs us to enable it and clock it | 11:16 |
| uvos | otherwise its inidpendant from cpu | 11:16 |
| uvos | ofc i dont know how to do this properly in alsa otherwise we would not be here | 11:16 |
| uvos | ill try and catch up on the backscroll later | 11:16 |
| fmg_d4 | you will, as I have no idea why do we need voice capture dai stram, for example | 11:17 |
| fmg_d4 | thanks | 11:17 |
| uvos | fmg_d4: regarding broken d4 keyboard | 11:17 |
| uvos | i had one of those too | 11:17 |
| fmg_d4 | will ping you when I am back home from work | 11:17 |
| uvos | i carefully pealed it back and cleaned the contacts | 11:17 |
| fmg_d4 | so it is hw issue? | 11:18 |
| uvos | never had usual units miss keypresses (besides there was a kernel bug at some point) | 11:18 |
| uvos | but the kernel bug was solved long ago | 11:19 |
| uvos | never had problems after that | 11:19 |
| fmg_d4 | i was thinking it is design issue | 11:19 |
| fmg_d4 | i see | 11:19 |
| uvos | hmm no something is wrong with that unit :P | 11:19 |
| fmg_d4 | ok | 11:19 |
| uvos | d4s keyboard is usually really good | 11:19 |
| fmg_d4 | will clean it up | 11:20 |
| fmg_d4 | or, it might self-heal now i am going to use it regularly :) | 11:20 |
| uvos | helps to masage it | 11:20 |
| uvos | if you leave a d4 for a long time sometims some keys dont work at all | 11:21 |
| fmg_d4 | yeah, makes snse | 11:21 |
| uvos | in this case just pressing down hard and wiggeling on eatch key helps | 11:21 |
| inky | folks, so i wrote code which detects the screen rotation. | 15:31 |
| inky | but when i press my program, maemo would rotate the screen back. | 15:32 |
| inky | so i need also to tell maemo to not switch back to landspace with my app. | 15:32 |
| sicelo | hildon application? | 15:40 |
| Wizzup | freemangordon: amazing @ working | 15:40 |
| inky | well, no, just pascal/lazarus app. i can detect that the screen got rotated and which are the new dimensions by xrandr library. | 15:40 |
| inky | but then i need to tell hildon to not rotate my app back to landscape. | 15:41 |
| inky | when i switch to it. | 15:41 |
| sicelo | one option is to hildonize it | 15:44 |
| Wizzup | you need to set the atom yup | 15:44 |
| Wizzup | if you use gtk2, there is a way to do it from libhildon, if you use qt, you can use setproperty | 15:44 |
| inky | yes i think i will try to call x11 functions to set xatom. | 15:45 |
| Wizzup | pretty sure lazarus uses gtk2 but ok :D | 15:46 |
| inky | yes it uses gtk2. | 15:46 |
| inky | it has gtk3 but i use gtk2, and it is default. | 15:46 |
| inky | so i need to make a wrapper to libhildon i guess. which function? | 15:47 |
| inky | btw Wizzup i use my comics-daily app every day, and i installed it as debian package, because i am able to build debian package. as soon as jenkins has it i will build it there. | 15:47 |
| Wizzup | I think fpc could maybe just generate these for you for libhildon | 15:48 |
| Wizzup | ok, well, I am in the middle of a renovation so I don't think the build server works atm | 15:48 |
| inky | there is a tool, yes. i'll try to use it. | 15:48 |
| Wizzup | but I do plan o fix it tonight or tomorrow | 15:48 |
| Wizzup | h2pas iirc | 15:48 |
| inky | aha. | 15:48 |
| inky | ty! | 15:48 |
| freemangordon | Wizzup: uvos: any idea what https://github.com/NotKit/android_kernel_motorola_omap4-common/blob/hybris-11.0/sound/soc/codecs/cpcap.c#L2478 is vs https://github.com/NotKit/android_kernel_motorola_omap4-common/blob/hybris-11.0/sound/soc/codecs/cpcap.c#L2478 | 16:35 |
| freemangordon | what is Voice Playback vs Voice Capture vs incall-playback vs incall-capture | 16:36 |
| freemangordon | Voice PB/CAP is for microphone or? | 16:37 |
| arno11 | freemangordon: playback is for earpiece/speaker and capture for microphone (at least in PA, alsa, ucm, cmtspeech). | 19:02 |
| arno11 | (but maybe your question is different or more tricky) | 19:03 |
| Wizzup | freemangordon: sorry, still doing renovation things, but home on wednesday, do you still need an answer to the question | 21:09 |
| freemangordon | Wizzup: yes, I need all teh info I can get | 21:32 |
| freemangordon | arno11: ok, but does this mean that both streams are active during call? | 21:33 |
| Wizzup | I will have to check with what we currently have, but I think voice capture is for cpu recording | 21:38 |
| freemangordon | ok. do you know what all those https://github.com/NotKit/android_kernel_motorola_omap4-common/blob/hybris-11.0/sound/soc/codecs/cpcap.c#L2430 are about? | 21:40 |
| Wizzup | no, that seems to be from android, I don't know quite exactly, but with the names I can guess what some of them are for | 21:41 |
| Wizzup | heh fmaudio playback | 21:42 |
| freemangordon | yes, it is andoid | 21:43 |
| freemangordon | but I can;t guess some of them | 21:43 |
| freemangordon | like, what is BPVoice? | 21:44 |
| Wizzup | not sure to be honest | 21:47 |
| freemangordon | as I think we want those too | 21:48 |
| Wizzup | yeah, ultimately having bluetooth work would be nice | 21:48 |
| Wizzup | assuming it is bluetooth | 21:48 |
| freemangordon | no, bluetooth is BT | 21:48 |
| freemangordon | there is "cpcap bt-call" and "cpcap bt" | 21:49 |
| Wizzup | ok | 21:49 |
| freemangordon | Wizzup: do you know where internal mic is connected? to baseband or to cpcap? | 21:50 |
| Wizzup | sorry, I don't really have the answer, I can only guess | 21:52 |
| Wizzup | I think it might be both, but I would guess cpcap | 21:52 |
| freemangordon | lets see if uvos has more info | 21:53 |
| Wizzup | but this is really great progress still :) | 21:54 |
| freemangordon | right | 21:54 |
| freemangordon | hmm, BPVoice could be BasebandProcessorVoice | 21:58 |
| freemangordon | whatever is that supposed to mesan | 21:58 |
| Wizzup | maybe direct routing or something | 22:05 |
| Wizzup | or maybe it's lte? | 22:05 |
| Wizzup | the droid4 has a lte modem for in the us | 22:05 |
| freemangordon | ah, right | 22:05 |
| freemangordon | yeah, that would make sens | 22:05 |
| freemangordon | e | 22:05 |
| arno11 | freemangordon: yes both streams are probably active during call (even for voice recording) | 22:14 |
| freemangordon | hmm, 6.6 does not seem to boot with flat battery | 22:38 |
| Wizzup | how flat and what charger? is that different from 6.1? | 22:39 |
| Wizzup | in my experience you need to replug charger (maybe this is a thing of the past) when booting to get it to see it | 22:39 |
| freemangordon | yes, the same charger/cable | 22:39 |
| Wizzup | from kexec I mean | 22:39 |
| freemangordon | does not help | 22:39 |
| freemangordon | it just hang in very early stage | 22:40 |
| freemangordon | oh, now it booted, weird | 22:40 |
| freemangordon | on 5th attempt perhaps | 22:40 |
| uvos | freemangordon: the lte dose not and can not do voice | 23:40 |
| uvos | its not whats up here | 23:40 |
| Wizzup | hmm | 23:40 |
| uvos | so there are two things, the configuration of the cpcap side of the voice dai which needs to be configured by the cpu for modem operation | 23:41 |
| uvos | and the path used to record audio from the modem to the cpu | 23:42 |
| uvos | going by the names Voice Playback/Capture should be the voice dai on cpcap connected to the modem (see the cpcap datasheet) | 23:43 |
| uvos | and incall-playback/incall-capture is the modem <-> cpu interface used for call recording or possibly also audio injection into the call stream, atho android dosent do the latter, and i never got any of this to work on mainline despite trying some | 23:45 |
| uvos | https://apps.fcc.gov/eas/GetApplicationAttachment.html?id=1580564 | 23:56 |
| uvos | ntw | 23:56 |
| uvos | btw | 23:56 |
| uvos | LTE Band 13, VOIP only | 23:56 |
| uvos | no suprise | 23:56 |
| uvos | but yes the modem dose NOT support volte and its voice operation is used purly via the cpu doing voip | 23:57 |
| uvos | also lol https://apps.fcc.gov/eas/GetApplicationAttachment.html?id=1580522 | 23:58 |
| uvos | thats not a droid4 inductive charging door | 23:58 |
| uvos | thats a bionic door | 23:58 |
| uvos | upps | 23:58 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!