rinigus | @b100dian: asked SODP folks regarding status of nagara, no reply yet. but I wonder if you tried flashing AOSP on your xperia5iv and checked how well does it work. as that's the starting point for the port. | 07:00 |
---|---|---|
rinigus | (reply from SODP will probably come on monday) | 07:01 |
T42 | <b100dian> rinugus: flashing means building, right? | 08:41 |
T42 | <b100dian> I've unlocked bootloader yesterday without thinking much, it is at build 64.1.A.0.913 which is A13 now, not sure if there were any updates still on A13 | 08:42 |
T42 | <b100dian> rinigus: misspelled the ping earlier | 08:42 |
T42 | <b100dian> I see AOSP14 is also available for nagara. Will try an android A13 build this evening, my "build system" in undergoing power maintainance now | 08:47 |
rinigus | @b100dian: nice!!! that's great to hear. just be careful and mind https://github.com/sonyxperiadev/bug_tracker/issues/824 | 09:14 |
rinigus | @b100dian: aosp14 doesn't have blobs released yet. | 09:15 |
rinigus | as for unlocking, I do wonder whether it would make sense to update to the latest stock as well. | 09:16 |
T42 | <b100dian> Sounds familiar:) | 09:16 |
rinigus | and also it was recommended before to try to establish volte connection on stock. probably to get your corresponding files/settings up to date | 09:17 |
rinigus | as blobs are for aosp13 and are on revision v4a, they could be more stable. | 09:18 |
rinigus | @b100dian: looking forward for your test results :) | 09:19 |
rinigus | @b100dian: re flashing - yes, I think you have to build it then. from xda I get an impression that custom roms are frequently based on stock for xperias | 09:26 |
T42 | <b100dian> I'll use sony instructions | 09:27 |
rinigus | great! | 09:32 |
rinigus | @b100dian: maybe of help - https://github.com/sailfishos-sony-tama/main/blob/hybris-10/hadk-sony-xz2.md | 14:03 |
rinigus | I wonder if Jolla made a similar doc for current xperia 10v | 14:04 |
rinigus | but start is still from aosp and I do wonder what your tests will show. | 14:04 |
T42 | <b100dian> Yes, running repo command from sony's intructions for now | 14:06 |
T42 | <b100dian> I wanted to see how the modem-config looks before and after inserting a sim card with the stock fw, but there's no /vendor/oem on this one (or it will appear later :hmm:) | 15:12 |
T42 | <TheVancedGamer> anybody here that has worked on ofono-sailfish before? :D | 17:09 |
T42 | <TheVancedGamer> I think mal? | 17:09 |
T42 | <TheVancedGamer> I'm not sure if that pinged you | 17:13 |
rinigus | @TheVancedGamer: just in case - mal , you have been asked for help ^ . but I think it pinged earlier as well | 17:15 |
T42 | <TheVancedGamer> anyway for when you're active, | 17:44 |
T42 | <TheVancedGamer> mal: I've been looking at why the USimAuthentication stuff doesn't work on ofono, and it seems nobody bothered to plug it in? https://github.com/sailfishos/ofono/blob/32ab6b3c785b5497dd6fdbddd715cff67826e36e/ofono/doc/sim-auth-api.txt#L44 | 17:44 |
T42 | <TheVancedGamer> ofono seems to have the APIs | 17:44 |
mal | @TheVancedGamer yes, I have been working on ofono stuff | 17:53 |
mal | what is that USimAuthentication used for? | 17:54 |
T42 | <TheVancedGamer> mal: I'm working on software based IMS, and I need to send some challenges to the sim card to connect to the IPsec tunnel | 18:04 |
T42 | <TheVancedGamer> radio HAL has a function called requestIccSimAuthentication, but it doesn't seem to be used by ofono plugins | 18:05 |
T42 | <TheVancedGamer> that function should be synonymous to the USimAuthentication function | 18:06 |
T42 | <b100dian> sony's repo sync locked 99% in the way.. hey, let's restart it, sounds like checking if up-to-date takes as much.. Maybe I shoudn't have used a remote VM but my impression is that it downloads more than a regular Lineage manifest would.. | 20:19 |
T42 | <b100dian> rinigus: building AOSP is probably for tomorrow | 20:19 |
T42 | <edp_17> mal: could you please help me to fix audio? I have no clue where to see and what. I've removed audiosystem-passthrough as it was failing. Now there is no error in pulseaudio: https://paste.ubuntu.com/p/Cz6rqgrPdq/ | 22:44 |
T42 | <edp_17> Also, in journal looks like vendor.audio-hal-2-0 is started: https://paste.ubuntu.com/p/CZN7svfkcx/ | 22:44 |
T42 | <edp_17> And in logcat, the audio lib (/vendor/lib64/hw/audio.primary.msm8953.so) is loaded: https://paste.ubuntu.com/p/ZX82wQCxcC/ | 22:45 |
T42 | <edp_17> But when I play a sound, nothing comes out. :( | 22:45 |
T42 | <edp_17> I've added AUDIOSYSTEM_PASSTHROUGH_TYPE=hw2_0 into /etc/sysconfig/pulseaudio and installed audiosystem-passthrough back. | 23:10 |
T42 | <edp_17> This way the "[audiosystem-passthrough] No interfaces could be configured!" error gone from journal. | 23:12 |
T42 | <edp_17> Well, according to this https://github.com/mer-hybris/audiosystem-passthrough/blob/master/README.md | 23:17 |
T42 | <edp_17> As I do have "Registered vendor.qti.hardware.radio.am@1.0::IQcRilAudio/slot1 (start delay of 5957ms)" in logcat | 23:18 |
mal | which quirks have you tried? | 23:18 |
T42 | <edp_17> I've changed /etc/sysconfig/pulseaudio to AUDIOSYSTEM_PASSTHROUGH_TYPE=qti and installed audiosystem-passthrough-af. | 23:19 |
T42 | <edp_17> mal: do you mean in /etc/pulse/arm_droid_card_custom.pa ? | 23:19 |
mal | yes | 23:19 |
mal | which android base is that? | 23:20 |
T42 | <edp_17> Nothing as I don't have that file. | 23:20 |
T42 | <edp_17> Los17.1 | 23:20 |
mal | so how did audiosystem-passthrough fail? | 23:21 |
T42 | <edp_17> mal: I've tried this but didn't work: https://github.com/mlehtima/droid-config-fp4/blob/devel/sparse/etc/pulse/arm_droid_card_custom.pa | 23:21 |
mal | make sure sink format is correct | 23:21 |
T42 | <edp_17> I don't know what is the correct format. | 23:22 |
mal | usually it should not be set unless audio policy configuration shows something else than 16-bit | 23:22 |
mal | check the android side audio policy configuration xml and find primary output in there | 23:22 |
mal | since you have that vendor.qti.hardware.radio.am@1.0::IQcRilAudio you should have pulseaudio-modules-droid-hidl and audiosystem-passthrough | 23:24 |
T42 | <edp_17> audio_policy_configuration.xml: https://paste.ubuntu.com/p/jK5jGm3cCz/ | 23:24 |
mal | you can see it's 16-bit as usual, so do not set format in .pa file | 23:24 |
T42 | <edp_17> I have them both pulseaudio-modules-droid-hidl and audiosystem-passthrough: https://paste.ubuntu.com/p/VsCG9RxZ8p/ | 23:25 |
mal | ok | 23:25 |
T42 | <edp_17> Should I use arm_droid_card_custom.pa or not? | 23:25 |
mal | if you want to try some quirks, also only use load-module module-udev-detect if you have the other needed config for that | 23:27 |
mal | that udev detect needs something like this https://github.com/mlehtima/droid-config-fp4/blob/devel/sparse/lib/udev/rules.d/99-pulseaudio-ignore-builtin-card.rules where the id attr is something you need to find in sysfs | 23:28 |
T42 | <edp_17> I've only set "load-module module-droid-card rate=48000 quirks=-output_deep_buffer" in /etc/pulse/arm_droid_card_custom.pa. | 23:32 |
T42 | <edp_17> Still no sound: https://paste.ubuntu.com/p/9Nvvn458xV/ | 23:32 |
mal | where is the start of playback in that | 23:33 |
mal | anything in logcat when you try to play audio? | 23:34 |
T42 | <edp_17> Play music started at line 1173 : https://paste.ubuntu.com/p/q5FHtFGXJK/ | 23:35 |
mal | do you really need that AUDIOSYSTEM_PASSTHROUGH_TYPE ? also is qti_0 even valid value? | 23:35 |
T42 | <edp_17> Music started plaing at line 6905 (logcat): https://paste.ubuntu.com/p/6k8jrMZ32W/ | 23:37 |
T42 | <edp_17> I read that AUDIOSYSTEM_PASSTHROUGH_TYPE value from here https://github.com/mer-hybris/audiosystem-passthrough/blob/master/README.md | 23:38 |
T42 | <edp_17> I am not sure I need it, but I do have "vendor.qti.hardware.radio.am@1" in my logs, so i thought needed. | 23:38 |
mal | the list is qti, af or hw2_0, no such thing as qti_0 | 23:38 |
mal | and usually that is not set | 23:39 |
T42 | <edp_17> Yeah, that's a typo. | 23:39 |
mal | are you using normal audio module or that compat module | 23:39 |
T42 | <edp_17> hidl_compat | 23:40 |
mal | and you are sure it gets loaded by pulseaudio? | 23:40 |
T42 | <edp_17> pulseaudio is masked. | 23:40 |
mal | I mean that it loads hidl_compat | 23:41 |
T42 | <edp_17> How should I check whether that is loaded by pulseaudio? | 23:41 |
mal | strace for example, how did you mount it? | 23:41 |
mal | there is usually some bind mount magic needed | 23:41 |
T42 | <edp_17> I have this: https://paste.ubuntu.com/p/RwM488rxcQ/ | 23:43 |
mal | and it loads that? I mean it might not get to load default one if there is something else that matches | 23:44 |
T42 | <edp_17> I see this in journal: https://paste.ubuntu.com/p/jWHxJHzRJH/ | 23:45 |
mal | hmm, is your vendor.audio-hal-2-0 32-bit or 64-bit? | 23:46 |
T42 | <edp_17> The device is 64 bit. | 23:46 |
mal | that's not what I asked | 23:46 |
T42 | <edp_17> I have no idea whether vendor.audio-hal-2-0 is 32 or 64 bit. How can I check? | 23:47 |
mal | file some.so | 23:47 |
mal | or actually file what_ever_program | 23:47 |
mal | so check ps aux or the audio hal .rc file to see which program that hal is | 23:48 |
T42 | <edp_17> Is it? https://paste.ubuntu.com/p/xXzxhvZTD2/ | 23:48 |
mal | I said the program, not the lib | 23:49 |
T42 | <edp_17> ps aux: https://paste.ubuntu.com/p/Y6zVFKyFK8/ | 23:49 |
mal | probably /vendor/bin/hw/android.hardware.audio@2.0-service | 23:50 |
T42 | <edp_17> https://paste.ubuntu.com/p/ZvBxStqvKg/ | 23:51 |
T42 | <edp_17> Looks like 32 bit | 23:51 |
mal | which should have been obvious, you could have just checked for anything audio related in there first before pasting it | 23:51 |
T42 | <edp_17> And why is it important whether 32 or 64 bit? | 23:55 |
T42 | <edp_17> I am bind-mounting both audio.hidl_compat (32 and 64 bit) | 23:56 |
mal | well compat only works if that process is 32-bit | 23:56 |
T42 | <edp_17> In droid-hal-early-init.sh . Or isn't that relevant? | 23:56 |
T42 | <edp_17> I see. But it looks like 32 bit, no? | 23:57 |
mal | yes | 23:57 |
T42 | <edp_17> So, I am puzzled why don't have sound. | 23:57 |
mal | so what bind mount do you really have? that file you showed had default one but the other paste showed different one | 23:58 |
T42 | <edp_17> This is df -h: https://paste.ubuntu.com/p/NWcyD5MdwS/ | 23:58 |
T42 | <edp_17> Or how can I check? | 23:59 |
mal | strace of pulseaudio would tell | 23:59 |
T42 | <edp_17> mounts : https://paste.ubuntu.com/p/63nZ9Qrnwd/ | 23:59 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!