vknecht | rinigus, hello... replicating tama OBS repo layout for loire: I really need to 1) Fork community-adaptation and 2) Add SSU feature like in https://github.com/sailfishos-sony-tama/droid-config-sony-tama-pie/commit/fa446d1bda858bc2c7dd6aa1c46a6ecf67317d9c ? | 12:20 |
---|---|---|
rinigus | vknecht: tbh, I don't remember. I guess you will need it if you have several devices (like we have xz2, xz2c, xz3) using the same port and if you wish to host aosp images as well. aosp is handy when you generate images for flashing using ci. | 12:43 |
rinigus | I can look later tonight, if needed, on what were those repos doing | 12:43 |
vknecht | rinigus, yes there are several (suzu, kugo)... I'm just not sure if both mods are really needed, guess I can remove later | 12:47 |
rinigus | vknecht: you could try without and see where it falls apart :) | 12:52 |
vknecht | indeed :-) | 12:54 |
vknecht | seems ok to just modify the kickstart file for locally building from OBS (will have to look into CI later) | 16:17 |
rinigus | vknecht: CI is setup for tama is somewhat different from the classical one. it downloads KS from OBS and builds all targets (xz2single, xz2dual, ...) in parallel. as a result, its easy to support all variants with the same script | 17:54 |
T42 | <adampigg> rinigus: do you build devel: and testing: images? | 18:21 |
rinigus | adampigg: not anymore, its testing only. devel was when we were adding new devices and making original port | 18:22 |
vknecht | rinigus, what about droid-system[-vendor] ? do you just build them locally then upload to nemo:testing:hw:sony:tama:extra-images, or are those produced by some CI ? | 18:33 |
rinigus | vknecht: locally, unfortunately. but now its done - until we move to aosp10 (or maybe just update the kernel), but that will come later with an eye of improving camera support | 18:34 |
vknecht | well, I think kernel update only require droid-hal-$HABUILD_DEVICE updates ; droid-system[-vendor] updates would only be needed if Sony's AOSP got security-patch update which unfortunately is still at 2019-08-01 | 18:39 |
vknecht | or further patch from hybris side (eg. I still need to update syspart for selinux patch) | 18:40 |
vknecht | recently updated from mer-hybris kernel 4.9.213 with no problem just by update droid-hal-loire | 18:41 |
vknecht | maybe you should check latest kernel, iirc there was some cam improvements | 18:43 |
rinigus | vknecht: will check when working on the next release. until then prefer to work on other projects | 19:08 |
vknecht | sure, you seem quite busy with flatpack and also map apps :-) | 19:09 |
vknecht | had to use gps some time ago and pure maps helped quite a lot | 19:11 |
vknecht | if anyone has an idea... in loire + 4.9 kernel there's a timing issue between hwcomposer and surfaceflinger services' start, and since surfaceflinger is not used in SFOS then hwcomposer service won't start | 19:13 |
vknecht | https://github.com/sonyxperiadev/device-sony-loire/commit/75811b2c2d85cc8147e1b4d6b29d22bfb5ac2c84 | 19:13 |
vknecht | at one point I naively tried to make minifs emit "init.svc.surfaceflinger=running" property but that didn't seem to help (maybe did it wrong) | 19:15 |
vknecht | any suggestion welcome | 19:15 |
vknecht | s/minifs/minisf/ | 19:15 |
vknecht | currently I use https://github.com/sailfish-on-loire/droid-config-sony-loire-pie/commit/aef307926e24d6f6f7852c5084553e5d6f6a5481 but it only works about once in five to ten boots | 19:17 |
piggz | mal: my kernel already has that patch for fuse | 19:19 |
piggz | and.... | 19:19 |
piggz | sh-3.2# mcetool --get-suspend-stats | 19:19 |
piggz | uptime: 41912.898 | 19:19 |
piggz | suspend_time: 0.000 | 19:19 |
mal | piggz: you mean you reverted it already? | 19:21 |
mal | piggz: the point was to revert a commit from fuse, do you have the sources somewhere for the device? | 19:22 |
piggz | mal: i mean it didnt need reverted as our kernel already has it as per your patch | 19:23 |
mal | hmm, that sounds odd | 19:23 |
piggz | mal: ignoring suspend for now ..... by default I cant SHS in becuase teh SSH daemon isnt started when using developer mode...is that known? | 20:00 |
piggz | then, if i start SSHD manually, the iptables rules dont let me have access | 20:02 |
piggz | so, I have to flush them etc | 20:02 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!