T42 | <lackingcheesepuffs> @elros34 [@lackingcheesepuffs add these to patterns: htt …], it built but it is unable to boot | 05:55 |
---|---|---|
rinigus | mal: ohm update made the calls work. thank you! | 08:56 |
rinigus | I have major echo (for the one calling me) and seems like no noise cancellation during the call. | 08:57 |
rinigus | grabbed the logs from hybris10 (with echo) and hybris9 (all works); logs at https://github.com/sailfishos-sony-tama/main/issues/125 | 08:57 |
rinigus | will have to check how it is in android as well. | 08:58 |
rinigus | mal: what about phone calls on seine? do you guys have echo or calls are fine? | 08:58 |
rinigus | what's interesting is that with the echo, looks like noise cancelling microphone is still used (voice-dmic-ef) | 08:59 |
T42 | <kiam001> Give it a new hadk, the current dosent Work with 17.1 | 12:52 |
rinigus | @kiam001: please wait and let mal finish his work on preparing 17.1 repos. | 12:54 |
mal | @kiam001 I need to write some instructions to faq about 17.1 once I have clean repos to use as example | 12:56 |
T42 | <kiam001> OK thy | 12:58 |
T42 | <kiam001> OK thx* | 12:58 |
mal | rinigus: I was wondering if could do a bit similar thing that is done for dummy audioflinger in audiosystem-passthrough but for audio policy | 17:53 |
rinigus | mal: https://github.com/sailfishos-sony-tama/miniaudiopolicy | 17:54 |
rinigus | mal: that I did for fixing camera issue with the video for tama. I wonder whether it went under your radar when I reported regarding it | 17:55 |
rinigus | it does sound to me as a way to go. works nicely on tama/aarch64 and avoids lockups as I had when mediaservice activated dummy audio policy before camera service | 17:56 |
mal | rinigus: I remember you talked about it but I didn't know you had code for it already | 17:57 |
mal | I mean code in github | 17:57 |
rinigus | mal: maybe forgot to publish a link... | 17:59 |
mal | I'll try that on fp3 | 17:59 |
mal | could be that I missed it | 17:59 |
rinigus | sounds like you have the same issues then | 17:59 |
mal | rinigus: looks like you did post the link but I didn't notice it | 18:00 |
mal | rinigus: I have some microphone issues during calls on my latest build, need to debug that | 18:01 |
rinigus | mal: no wonder. I suspect if I grep for 'mal' in the channel archive, I will get 75% of the log. 25% is probably people leaving and joining channel | 18:01 |
rinigus | interesting, as I have some mic issues as well. mainly echo and not noise cancelling during a call. but will probably look into it after I push 4.0.1.48 update for aosp9 based port | 18:02 |
rinigus | ... tama | 18:03 |
mal | rinigus: yes, might be similar, not fully sure because I had the test device very close | 18:08 |
mal | both devices I used for the test call were like 20-30 cm apart so nice loops sometimes when it caught the audio from the other device | 18:09 |
rinigus | mal: I asked to be assisted - 20-30cm is probably not enough | 18:16 |
rinigus | note that I am using that audio policy service, so I am not sure it is from there. although, the code in that service is rather trivial | 18:17 |
mal | well that is pretty much what droidmedia did for it | 18:19 |
rinigus | indeed | 18:21 |
mal | so should be fine, we'll see once I have built my new | 18:21 |
mal | *new version | 18:21 |
rinigus | mal: note that I needed few ICU links in sparse for 64bit audiopolicy server to work: https://github.com/sailfishos-sony-tama/droid-config-sony-tama-pie/commit/6894adc8c4c24632641098d628bdc8d9e7a32c4c | 18:25 |
mal | rinigus: I have a better fix | 18:27 |
mal | rinigus: try adding this line instead https://github.com/mlehtima/droid-config-fp3/blob/master/sparse/usr/libexec/droid-hybris/system/etc/ld.config.29.txt#L11 | 18:29 |
mal | not sure if using system namespace is good or if vendor would be better | 18:29 |
rinigus | mal: I will try. don't understand why we have bin in ldconfig, but that is surely described somewhere. | 18:42 |
rinigus | will write it down and try when I get back to the aarch64 version of the port | 18:42 |
mal | rinigus: those are used to separate what search path and other things binaries location in different based use | 18:43 |
mal | *different places | 18:44 |
rinigus | mal: and if it is not mentioned then some smaller default lib path is used, right? | 18:44 |
mal | yes which caused the issue that it couldn't find libraries from apex path | 18:45 |
mal | without having it there only odm, system and vendor paths were used | 18:45 |
rinigus | mal: that's a good fix then :) | 18:47 |
T42 | <Verevka86> Xiaomi MI6 hybris17.1 (armv7hl)- work- ril, wifi, sensors, bluetooth, nfc, not work- audio, camera 😔 | 21:50 |
mal | @Verevka86 do you see a message about libaudioroute.so in journal when you start pulseaudiò? | 21:54 |
T42 | <Verevka86> @mal [@Verevka86 do you see a message about libaudio …], audio fix 🤝 | 22:01 |
mal | @Verevka86 I fixed it like this https://github.com/mlehtima/droid-config-fp3/tree/master/sparse/usr/libexec/droid-hybris/system/lib64 | 22:05 |
mal | just a symlink to the correct place in that folder | 22:06 |
mal | you seem to have 32-bit build so might be a bit different | 22:06 |
T42 | <Verevka86> @mal [@Verevka86 I fixed it like this https://github …], only I had to do it in lib not lib64 | 22:07 |
mal | ok | 22:07 |
mal | do you have mini*service processes running on the device? | 22:07 |
mal | related to camera issue | 22:07 |
T42 | <elros34> In 4.0.1 some jolla apps use now generic booster instead silica-qt5. Is there a fix or workaround for it available? | 22:10 |
mal | what problem does that cause? | 22:20 |
T42 | <elros34> not really a problem but silica-qt5 speeds up applications start, generic booster is slow | 22:22 |
mal | yeah, there are some issues with sandboxing still | 22:24 |
T42 | <elros34> is it expected that csd is not installed by default anymore? It seems mic doesn't install Recommended packages | 23:59 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!