| T42 | <Diego Lopes %lastname%> Hello people. I need some help here. My Sony Xperia X (F5121) just arrived, and I'm trying to install the Sailfish X on it. Someone have a step by step link to send me later? Thank you very much. | 03:06 |
|---|---|---|
| T42 | <birdzhang> Diego Lopes https://jolla.com/sailfishxinstall/ | 03:07 |
| T42 | <Diego Lopes %lastname%> @birdzhang [Diego Lopes https://jolla.com/sailfishxinstall …], Thanks. I am reading the page and I still don't understand the part 'Sony vendor binary image'. What is this for!? | 03:09 |
| T42 | <birdzhang> There is a link at the end of 5.1 part. BTW, this group is for port only, you can join SailfishOS fanclub https://t.me/joinchat/AWx9iUE-U9pZ_NeJlxXs0w | 03:12 |
| T42 | <adampigg> (Photo, 2560x1440) https://irc.thaodan.de/.imgstore/x7HTc8mev5.png | 14:25 |
| T42 | <adampigg> Kimmoli^^ also playing with electronics, and my prototype works! | 14:26 |
| mal | @adampigg but what does it do? | 14:26 |
| T42 | <adampigg> mal: opens my car doors :) | 14:27 |
| T42 | <adampigg> (the old renault 5) | 14:27 |
| mal | cool | 14:27 |
| T42 | <adampigg> the chip at the heart of it is a TEA5500, with 10^3 codes | 14:27 |
| T42 | <adampigg> there are docs online which say how to create the pair of transmitter/receiver | 14:27 |
| T42 | <adampigg> i have the receiver, but no transmitter | 14:28 |
| T42 | <adampigg> so, i used the oem circuit diagram and reversed the code from me receiver, and it blooming worked | 14:28 |
| rinigus | adampigg: does it also work on renault 5 of the neighbor? | 15:38 |
| T42 | <adampigg> Rinigus, renault 5s are rare now, but no, its coded to the receiver | 15:43 |
| T42 | <DylanVanAssche> Put it in a TOH for the PinePhone :D @adampigg | 16:00 |
| deathmist | mal: nfcd-binder-plugin MW build fails because nfcd-plugin is not present https://pastebin.com/T1mVN1dZ | 17:57 |
| mal | deathmist: build older version of that then | 17:58 |
| mal | deathmist: cd hybris/mw/nfcd-binder-plugin; git reset --hard 1.0.4; cd $ANDROID_ROOT | 17:59 |
| deathmist | yep, just did that and it worked. would this let me get NFC working since I've not worked on that yet myself? | 18:00 |
| mal | deathmist: that works only for android 8 or newer (lineage 15.1 or newer) | 18:02 |
| deathmist | my base is lineage-15.1 so I guess I'll look into that soon | 18:03 |
| mal | deathmist: you'll need that and jolla-settings-system-nfc in patterns for nfc | 18:05 |
| deathmist | alright, thanks! I'll see if I can test that on my next build already (and maybe FINALLY get call audio working too w/ hidl PA module v1.0 out of the box) | 18:09 |
| deathmist | btw how difficult should it be to migrate from 15.1 base to 16.0 when that is publicly available? | 18:15 |
| mal | difficult to say, depends on device | 18:30 |
| deathmist | mal: I added my own mass storage configs for testing in dcd (https://git.io/JeG05), now mic fails with "file /etc/usb-moded/mass-storage-jolla.ini conflicts between attempted installs of droid-config-cheeseburger-1-1.armv7hl and usb-moded-mass-storage-android-config-0.86.0+mer36-1.4.1.jolla.armv7hl" | 19:01 |
| deathmist | tried "sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -R -m sdk-install zypper rm usb-moded-mass-storage-android-config" already but it didn't help | 19:02 |
| rinigus | I am starting to look for the next SFOS device. As I am not thrilled with the current and upcoming official devices nor pro1 for different reasons, I wonder if there is some current(ish) higher end devices with good ports. | 19:07 |
| rinigus | Or considering some lower hanging ports, like xz2/xz2c/xz3 family. xz{2,3}'s seem to suffer currently from relatively poor camera, which is expected to improve with 4.14 kernel and its blobs (posts on https://forum.xda-developers.com/xperia-xz2/development/aosp-aosp-8-1-h8266-t3843269/page91). | 19:07 |
| rinigus | maybe someone is on the similar search or, ideally, looks for a new challenge and is ready-to-port? | 19:07 |
| mal | deathmist: is your device using which usb method, the old android one or configfs? | 19:09 |
| deathmist | mal: configfs, I'm looking into mass storage too because I've still not narrowed down the full problem of MTP on this device | 19:10 |
| deathmist | the MTP issue was iirc some "b.2" dir did not get created in /config even tho it definitely is right there in the init scripts | 19:11 |
| mal | deathmist: mass storage should not be used, it's old and had a lot of issues, not sure if it will even work with configfs | 19:16 |
| mal | deathmist: you can always create the configfs content yourself with some script? | 19:17 |
| deathmist | mal: ok, I'll revert everything related to mass-storage in that case. I'll look into that again, having MTP around is really convenient | 19:18 |
| mal | deathmist: also if you use configfs you should only have https://github.com/mer-hybris/droid-config-sony-ganges-pie/blob/master/patterns/jolla-hw-adaptation-kirin.yaml#L60 in patterns and then in spec https://github.com/mer-hybris/droid-config-sony-ganges-pie/blob/master/droid-config-common.inc#L12 | 19:18 |
| deathmist | mal: well, I didn't have either of those setup properly then. will removing usb-moded-connection-sharing-android-config from patterns disallow me from using USB network sharing? I use it (tho very rarely) | 19:25 |
| deathmist | actually it's not a big deal, since I have a dual-boot w/ LineageOS setup I'll just boot there if I ever need it | 19:25 |
| vknecht | not sure about higher end devices, but future really is with mainline-linux supported devices, isn't it ? so I'd guess, from lower to higher end, pinephone -> librem5 -> snapdragon845, bemol with sd845 being no garanteed support of telephony/usual smartphone features | 19:27 |
| vknecht | with some compromise, fairphone 3 maybe ? | 19:28 |
| rinigus | vknecht: for sfos use, maybe fp3 or some sd845 would be good compromise. in some respect sony devices have an appeal of being relatively well distributed, cover different form factors/price points, so it would have probably decent impact in terms of users.as far as I could see, moving within the same family of devices should be simple. | 19:33 |
| vknecht | ah yes, sonys are a pretty good choice from what I've seen, due to good kernel/device maintenance | 19:34 |
| vknecht | now with sfos the thing is: does the port base evolves with what sony/sodp supports, and does android support follows that | 19:37 |
| rinigus | vknecht: good question. looks like current xperia 10 android base is at https://github.com/mer-hybris/android/tree/sony-ganges-aosp-pie . I'd expect something similar for xz series would be needed. | 19:40 |
| rinigus | vknecht: from what I have been told, kernel 4.14 with, what's released as Q blobs, should work with the same blobs on P | 19:41 |
| rinigus | (android bases) | 19:41 |
| rinigus | vknecht: current development of android p is considered stable and is using 4.9 kernel. for better camera, 4.14 is needed, but that will be stabilized a bit later with development shifting to android q. for xz's, seems like strategy would be to port using android p as a base and later use 4.14 kernel. | 19:43 |
| rinigus | ... unless its impossible for any other reason. | 19:44 |
| vknecht | hmmm, recently saw on xda that aosp9 builds for X will soon be stopped in favor of 10 ; wondering if that version-speed-run is really beneficial to the user, not sure all aosp9 problems were ironed out | 19:46 |
| T42 | <elros34> @deathmist: not sure whether it works with configfs but you need this for mass storage https://build.merproject.org/package/binaries/nemo:testing:hw:motorola:moto_msm8960_jbbl/usb-moded?repository=sailfishos_3.1.0.12 | 19:48 |
| rinigus | vknecht: yes, same is going on with xz2/2c/3. but the fact that they still make those updates is already nice. | 19:50 |
| rinigus | ... but indeed, we cannot assume that aosp 9 will get much better later. unless there is a reason behind it, as with the camera drivers | 19:52 |
| mal | deathmist: I didn't know usb connection sharing works with configfs | 19:53 |
| vknecht | guess I'll have to upgrade my X aosp8 port to aosp9 to check 4.9 kernel makes things better (and I hope so), 4.14 why not but it's a bit fresh isn't it ? and will it be usable with aosp < 10 ? | 19:57 |
| vknecht | and iiuc there's a jolla announce to come about X ? | 19:57 |
| mal | vknecht: what problems do have with aosp8 base+ | 20:01 |
| rinigus | 4.14 was not recommended for general use at this stage. as for use with the earlier versions, I was told that for xz2 it should be ok (use with aosp9), but I haven't tested it. nor I have not heard of anyone else testing it | 20:02 |
| vknecht | mal, iirc 'bout same as you like bt problem, and especially high battery consumption (~180mW doing nothing in BatteryLog) | 20:05 |
| mal | vknecht: bt probably needs some kernel patch, just guessing, check rfkill to see what it lists, did you install bluebinder? | 20:17 |
| mal | vknecht: battery drain might need some things, I have some ideas for that | 20:18 |
| mal | vknecht: I think the device probably doesn't suspend at all, usually either ril or wlan is causing that, you can check in dmesg what is preventing suspend, if it's ofono then there are changes related to that in ofono after previous sfos release, in case it's wlan you could try https://github.com/mer-hybris/sailfish-connman-plugin-suspend | 20:23 |
| vknecht | wlan was no stranger to suspend problem, will try that connman thing | 20:24 |
| mal | dmesg usually tells if that is needed | 20:25 |
| mal | but if it's not suspending at all then it's more likely to be wlan than ril | 20:25 |
| mal | although ril can also be partially responsible | 20:26 |
| mal | vknecht: was bt failing completely or how did it behave? | 20:29 |
| vknecht | could say that, but it's been weeks sincmegae last looked at it, and I should provide logs | 20:47 |
| vknecht | https://mega.nz/#F!UbQU0IjY!5Ptb5RJYdTvBLlv_bVHKpA most recent I have, but will have to check again | 20:50 |
| mal | vknecht: can you check mcetool --get-suspend-stats | 20:58 |
| mal | vknecht: based on dmesg both ril and wlan could be problematic | 20:59 |
| vknecht | just powered the X back on with almost full battery, will let it run idly overnight and report ; now I'm uploading the 3.0.3.9 build in the same mega folder | 21:03 |
| vknecht | guess the mcetool will make more sense tomoroww | 21:03 |
| vknecht | *mcetool report | 21:03 |
| mal | vknecht: you can see quite easily if it suspends or not, just disconnect from USB and wait a moment and check | 21:05 |
| * vknecht removes himself from his couch | 21:05 | |
| * vknecht remembers usb developer mode connection is broken, logs over wlan... | 21:10 | |
| vknecht | [root@Sailfish nemo]# mcetool --get-suspend-stats | 21:13 |
| vknecht | uptime: 814.275 | 21:13 |
| vknecht | suspend_time: 399.841 | 21:13 |
| vknecht | f5121 build is uploaded, uploading f5321 one | 21:15 |
| mal | vknecht: ok, so it suspends at least somewhat | 21:20 |
| vknecht | not so utterly broken, then :-) but it's quite soon after a reboot, and without a SIM card ; anyhow I'll let it run for a while and provide logs | 21:28 |
| vknecht | the f5321/X Compact build is uploaded ; that's one benefit, kugo has its own build | 21:30 |
| vknecht | if you have any insight about J plans, feel free to share ;-) | 21:33 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
