| rinigus | mal: good morning! as it looks like I need audiopolicyserver for camera, I have made a small service with it. while doing so, I android compiled it as 64-bit - same as it is on AOSP on my device. while we missed few lib links to apex, those were easy to add | 09:06 |
|---|---|---|
| rinigus | missing links added in https://github.com/sailfishos-sony-tama/droid-config-sony-tama-pie/commit/6894adc8c4c24632641098d628bdc8d9e7a32c4c | 09:06 |
| rinigus | added service described in https://github.com/sailfishos-sony-tama/main/issues/115#issuecomment-787418732 | 09:07 |
| rinigus | result: camera works as it is expected, as far as I can tell | 09:07 |
| rinigus | I can switch between modes (video->photo->video), shoot them. | 09:08 |
| rinigus | mal: please look into this approach and let me know what you think. I now wonder, what was the purpose of older miniaf. | 09:10 |
| rinigus | mal: also, whether we should follow AOSP approach and split cameraserver and audioserver without pushing it all into one. that would allow to follow AOSP libs closer, at least in terms of 32- && 64-bit | 09:12 |
| piggz | rinigus: doesnt a fake audiopolicy server already exist? | 09:12 |
| piggz | mal: after PlatformSDK upgrade, build no longer has the crypt issues, but still fails at the config.status step (or just after) | 09:13 |
| rinigus | piggz: it does as a part of minimedia.cpp. now, on my device (and I bet on pro1), audioserver is 64 bit; cameraserver 32 bit. as a result, I don't have some 32bit libs that are required for fake audiopolicy server in /system/lib | 09:14 |
| piggz | ah, ok | 09:15 |
| rinigus | in addition, it turned out that on tama the order of ::instantiate was important. | 09:15 |
| rinigus | when having audiopolicy before camera, it got all blocked with no camera available. so, it makes sense to separate these services | 09:15 |
| rinigus | piggz: do you happen to know where are miniaf sources? | 09:18 |
| piggz | rinigus: https://github.com/mer-hybris/audioflingerglue/ ? | 09:21 |
| rinigus | piggz thanks! | 09:24 |
| swanux | Since yesterday I'm trying to solve my issue, and I tried another kernel config. The thing is, now I got this (https://pastebin.com/8tqq2zBB) build error. Does anyone have any ideas? | 09:57 |
| xmn | piggz Thanks for the PinePhone tip for sound. Although I could not find the pulseaudio dir. After the a update it started working! | 10:10 |
| xmn | ever inching close to a usable OS for the PP. Cheers | 10:11 |
| xmn | *closer | 10:11 |
| piggz | xmn: yeah, just volume to sort <cough>jusa_</cough> | 10:43 |
| piggz | xmn: i recently updated pulseadudio and the ucm files, so was probably that | 10:43 |
| xmn | well that was a nice surprise you gave me today :) | 10:48 |
| xmn | how hard is the audio volume thing? Seem strange that its a thing. But I guess audio is in flux these days with pulseaudio & pipewire mix around.in different distro | 10:49 |
| piggz | xmn: its the way sailfish does it ... there is a special pulseaudio pluging which manager the volume, and its that that doenst like our pulseaudio config for some reason | 10:55 |
| piggz | i suspect that if we had a volume app that controlled things individually like on a desktop, it would work | 10:56 |
| piggz | @eugenio_g7 ^^ was any special config needed on latte to get the alsa/UCM PA config to work with the meego volume plugin ? | 10:56 |
| piggz | i guess latte is a good reference .. and the new x86 port | 10:57 |
| xmn | hmm, sometime things that were thought to be smart turn out to cause issues later. This seems to be the case here. | 10:59 |
| xmn | thanks for sharing that info | 10:59 |
| piggz | xmn: the meego plugin does things like adjust the volume depending on the current use case, like call, media | 11:02 |
| piggz | the volume controls on the phone are routed to that plugin iirc | 11:03 |
| xmn | I see. I did see that it seem the volume button are contextual. So if your in a music app then you get volume adjustment and if not then it default to ringtone volume. | 11:05 |
| xmn | So piggz sailfish is using the meego plugin or something completely different | 11:05 |
| piggz | sailfish is using that plugin | 11:06 |
| xmn | I see | 11:06 |
| piggz | but,the plugin seems to get confused with our config | 11:07 |
| xmn | yeah | 11:07 |
| xmn | so its a matter of finding the right combo of configs? Or something more you think? | 11:08 |
| xmn | Clearly I'm not a programmer. :) just curious | 11:08 |
| T42 | <eugenio_g7> piggz: what's the issue on pp? Only ringtone volume gets changed? I think we had that on latte too | 11:48 |
| rinigus | @eugenio_g7: on which SFOS version? I have to set volume to mute to be able to mute volume on 4.0.1/aarch64 tama. when some video is playing, it is possible to regulate volume | 11:53 |
| T42 | <eugenio_g7> iirc it was pre 3.x on latte, but it was definitely an issue on our pulseaudio configs | 11:55 |
| rinigus | @eugenio_g7: OK, then it is something different | 11:57 |
| T42 | <adampigg> @eugenio_g7 [piggz: what's the issue on pp? Only ringtone v …], No volume is changed at all | 13:18 |
| T42 | Lucas %lastname% was added by: Lucas %lastname% | 14:27 |
| rinigus | when we make a photo using stock camera, are we supposed to hear photo "shot" sound? same goes for starting and stopping video | 14:50 |
| mal | rinigus: there should be some shutter sound | 14:56 |
| rinigus | mal: on my original (32 bit AOSP9) Tama port I don't have it. on AOSP10/aarch64 if I use MINIMEDIA_AUDIOPOLICYSERVICE_ENABLE - it is absent; if I use separate audiopolicy service, it is there | 14:58 |
| rinigus | is there a way to mute the sound if the user wishes to do so? | 14:59 |
| T42 | <adampigg> wasnt there a privacy thing that said cameras had to make sounds nowadays? | 14:59 |
| rinigus | @adampigg: I think so. but it was about default, I think. although, I am not sure | 15:00 |
| mal | rinigus: in some countries it's forbidden to have possibility to disable shutter sound | 15:15 |
| rinigus | mal: makes sense. then it is a bug in tama port (aosp9/arm32). | 15:17 |
| T42 | DarknessHiddenorg was added by: DarknessHiddenorg | 15:18 |
| piggz | grrrrr https://build.sailfishos.org/package/show/nemo:devel:hw:xiaomi:latte/xulrunner-qt5 | 18:21 |
| *** rubdos_ is now known as rubdos | 19:36 | |
| T42 | SteffenPrz was added by: SteffenPrz | 19:55 |
| rinigus | piggz: do you remember why we specify _target_cpu in FPD? it is set to armv7hl and, as a result, doesn't really get picked up for aarch64 :) | 20:10 |
| piggz | rinigus: just seen, and replied, mal ^^ | 20:11 |
| piggz | rinigus: im sure mal already has 64 bit patches for this | 20:11 |
| rinigus | piggz: nice! then will wait for mal reply. and then will try after 64 bit patches | 20:12 |
| mal | rinigus: yes, I have 64-bit patches for piggz's fpd code | 20:16 |
| mal | I just have forgotten to push those anywhere | 20:16 |
| piggz | mal: the usual behaviour is to send me a PR :D | 20:20 |
| piggz | then i can nitpick it | 20:20 |
| mal | :) | 20:21 |
| mal | piggz: good that someone else also nitpicks :D | 20:21 |
| piggz | it would give me great pleasure to find a unnescessary newline | 20:21 |
| mal | piggz: rinigus: https://github.com/piggz/sailfish-fpd-community/pull/35 | 20:26 |
| piggz | unfortunately i cant fault it | 20:55 |
| rinigus | piggz: this is a chance of a lifetime - make sure you check spaces as well | 21:38 |
| rinigus | mal: thank you for FPD PR. I'll compile it tomorrow night and hope to be able to test it on device as well | 22:07 |
| mal | rinigus: thanks | 22:07 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!