| Umeaboy | deathmist_: How would you solve this? https://ghostbin.co/paste/afwj7 | 00:11 |
|---|---|---|
| Umeaboy | Everything has worked somewhat fine until that. | 00:12 |
| T42 | <elros34> @leha155 custom vendor.mount shouldn't be needed. Systemd will make sure that /vendor, /vendor/firmware_mnt and so on are mounted in correct order automatically | 00:28 |
| Umeaboy | I can't seem to find community-adaptation-localbuild.spec in the rpm directory | 00:59 |
| Umeaboy | Even though it gets created according to the output. | 01:00 |
| Umeaboy | Found it....... It wasn't in $ANDROID_ROOT/rpm/ | 01:03 |
| Umeaboy | It was in /home/kristoffer/hadk/hybris/mw/community-adaptation/rpm/community-adaptation-localbuild.spec | 01:04 |
| Umeaboy | @elros34: I know I'm missing alot, but I can't find that error in the hot-hadk or in the HADK pdf. You have any idea on what to do? | 01:34 |
| T42 | <elros34> Umeaboy have you tried using correct sdk target (4.0.1 instead 3.4.0)? | 02:00 |
| Umeaboy | @elros34: Why does it say 4.0.1.48 here then? | 02:08 |
| Umeaboy | https://ghostbin.co/paste/huso | 02:08 |
| T42 | <elros34> it's only name of target you set when preparing sdk, you can call it sfos-5.0 | 02:09 |
| Umeaboy | Still the same result after updting: https://ghostbin.co/paste/sx4px | 02:13 |
| Umeaboy | So, what's next? | 02:15 |
| Umeaboy | @elros34: Is there something I've missed to do? | 02:19 |
| T42 | <elros34> clearly that update did nothing. IMHO the best way is remove old target and create new using versioned urls. 'Latest' url yesterday was pointing to 3.4.0, today it's 4.1.0 | 02:22 |
| Umeaboy | OK. | 02:24 |
| Umeaboy | @elros34: It looks alright now, right? https://ghostbin.co/paste/wrsty5 | 02:33 |
| Umeaboy | !! Could not determine if community-adaptation package is available, exiting. | 02:35 |
| Umeaboy | After running rpm/dhd/helpers/build_packages.sh --configs | 02:35 |
| T42 | <elros34> no, wrong target name: $VENDOR instead SailfishOS: I need to go sleep, good luck | 02:41 |
| Umeaboy | OK. | 02:45 |
| rinigus | Thaodan I don't see massive difference between symlinks and bind mounts. But in this case, it may make more sense to patch Android source that had lib path hard coded (mal found it earlier). | 04:52 |
| rinigus | I'll start with the transition to 4.1 and then see which bus we have | 04:53 |
| T42 | <A_T_R> here i am attaching hybris-patches > log https://del.dog/keckestani.txt | 08:51 |
| T42 | <A_T_R> is there anything wrong in this ? | 08:51 |
| hello-everyone | hello, I am trying to port sailfishos. The device in question already runs postmarketos mainline kernel but the porting docs use android kernel. Is there a guide (or overview steps) to port to device already running mainline ? | 10:12 |
| T42 | <Akatsu39> Ask Caleb maybe? | 10:29 |
| calebccff | what's the device | 10:29 |
| calebccff | you could basically just copy the PinePhone port, fix up a few things aaaand be good to go | 10:30 |
| T42 | <ankaos> hı. how fix this error? https://paste.ubuntu.com/p/2pvQYyx8fn/ | 11:56 |
| T42 | <Verevka86> @ankaos [hı. how fix this error? https://paste.ubuntu.c …], Use 4.0.1.48 😁 | 11:58 |
| T42 | <ankaos> @Verevka86 [Use 4.0.1.48 😁], sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R ssu release 4.0.1.48 XD | 11:59 |
| T42 | <ankaos> @Verevka86 [Use 4.0.1.48 😁], or: PlatformSDK [enes@enes hadk]$ sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R ssu release 4.0.1.48 | 11:59 |
| T42 | <ankaos> Changing release from 4.0.1.48 to 4.0.1.48 | 11:59 |
| T42 | <ankaos> Your device is now in release mode! | 11:59 |
| T42 | <ankaos> DBus unavailable, falling back to libssu | 11:59 |
| T42 | <elros34> remove /var/lib/rpm/__db.00* | 11:59 |
| T42 | <elros34> in target | 12:00 |
| T42 | <ankaos> @elros34 [in target], this is answer :) | 12:00 |
| T42 | <ankaos> @elros34 [remove /var/lib/rpm/__db.00*], ı dont have this file | 12:01 |
| T42 | <elros34> so maybe that is the issue, there is similar error about database mismatch. Have you tried in target or in sdk? | 12:03 |
| T42 | <ankaos> PlatformSDK [enes@enes hadk]$ sudo rm -rf /srv/mer/targets/samsung-hero2lte-armv7hl/var/lib/rpm/__db.001 | 12:03 |
| T42 | <ankaos> PlatformSDK [enes@enes hadk]$ sudo rm -rf /srv/mer/targets/samsung-hero2lte-armv7hl/var/lib/rpm/__db.002 | 12:03 |
| T42 | <ankaos> PlatformSDK [enes@enes hadk]$ sudo rm -rf /srv/mer/targets/samsung-hero2lte-armv7hl/var/lib/rpm/__db.003 | 12:03 |
| T42 | <ankaos> PlatformSDK [enes@enes hadk]$ sudo rm -rf /srv/mer/targets/samsung-hero2lte-armv7hl/var/lib/rpm/__db.004 | 12:03 |
| T42 | <ankaos> fix :) | 12:03 |
| piggz | rinigus: mal: incoming bug for fpd where any user can unlock the sessions | 12:07 |
| piggz | mal: also, did you get to look for vsync changes? | 12:07 |
| rinigus | piggz: lovely! | 12:07 |
| rinigus | great bug btw | 12:07 |
| piggz | yeah ... a volla user just reported it to me :) | 12:08 |
| rinigus | never make an account for someone on your phone and all is fine. but yeah, would have to fix it | 12:08 |
| mal | piggz: bug in which part? | 12:09 |
| piggz | mal: not sure, presumably, in the daemon, we look for a good fingerprint, and unlock undonditionally, but should only be the logged in user? | 12:09 |
| piggz | ive never created multiple users | 12:10 |
| rinigus | pretty sure we have a single DB and we don't have a link between uid<->FP. | 12:12 |
| rinigus | ... as it was done before multiuser support | 12:12 |
| piggz | yup | 12:13 |
| mal | piggz: official fingerprint support at least handles users | 12:17 |
| piggz | mal: cool, can you leak the code? :D | 12:18 |
| mal | :) | 12:19 |
| Danct12 | piggz, i can get a copy of the official fingerprint for you :) | 12:20 |
| Danct12 | just kidding | 12:20 |
| piggz | mal: anyway, lipstick vsync ;) | 12:20 |
| Mister_Magister | piggz: (psst lipstick hdr) | 12:50 |
| T42 | <Verevka86> https://paste.ubuntu.com/p/8D6TBMbb2Q/ | 12:59 |
| T42 | <Verevka86> Help, libgbinder not build 😔 | 13:00 |
| rinigus | @Verevka86: I have seen something similar. just disabled tests. after some time, enabled tests back and all worked. so, rather strange error and never got to the proper reason | 14:19 |
| T42 | <Verevka86> @rinigus [@Verevka86: I have seen something similar. jus …], I also periodically see it, how can I properly disable the tests? | 14:21 |
| rinigus | "properly" is not the word I would use :) . but in .spec, just comment them out | 14:26 |
| rinigus | @Verevka86 ^ | 14:28 |
| T42 | <Verevka86> Compile, thanks 🤝 | 14:34 |
| T42 | <b100dian> Anyone here that saw a similar stack corruption in the composer android sevice? https://pastebin.ubuntu.com/p/qtcspZKXTw/ | 16:54 |
| T42 | 阎莫幄 %lastname% was added by: 阎莫幄 %lastname% | 17:32 |
| T42 | <Verevka86> @Verevka86 [Compile, thanks 🤝], 4.1.0.23 hybris-17.1- works well on my xiaomi mi6, doesn't work- camera, flashlight and screenshot | 18:11 |
| rinigus | Just tested with 4.1.0.23 with tama aarch/aosp10, and video issue with not always resuming is still there | 19:18 |
| rinigus | interestingly, I managed to hit the same issue with aosp9/arm32bit port as well. it is just rare and harder to trigger. | 19:19 |
| rinigus | so, looks to be bug that was there from earlier days... | 19:20 |
| Umeaboy | Hi again! | 19:30 |
| Umeaboy | @elros34: I got the same result again. | 19:30 |
| Umeaboy | So, what's the next step now? | 19:45 |
| T42 | <elros34> you meant that pythone error when building droid-config? Put that 1 line error here and maybe somebody will recognize it. In the mean time you run git status in mw/droid-config and confirm that your repo is clean and same as on github | 19:51 |
| Umeaboy | There's no mw directory in $ANDROID_ROOT | 19:57 |
| Umeaboy | @elros34: ^^ | 19:57 |
| Umeaboy | There is one in $ANDROID_ROOT/hybris though. | 19:58 |
| T42 | <elros34> right it's hybris/ | 19:58 |
| Umeaboy | However there's no droid-config in it. | 19:58 |
| Umeaboy | Only community-adaptation community-adaptation.log | 19:58 |
| Umeaboy | droid-configs is in $ANDROID_ROOT/hybris/ | 20:00 |
| Umeaboy | I suspect that I have to start over from scratch excluding the making of the repos. | 20:03 |
| Umeaboy | And keep the full output from the terminal in a log file. | 20:03 |
| Umeaboy | So that everyone can see if I made a mistake somewhere. | 20:03 |
| Umeaboy | Unless someone else here has a better quciker idea. | 20:04 |
| T42 | <elros34> yeah that will change nothing, please show that git status | 20:04 |
| Umeaboy | @elros34: https://ghostbin.co/paste/rgkzamj | 20:05 |
| T42 | <elros34> so what git diff, and log shows | 20:06 |
| T42 | <elros34> from where did you get that jolla-hw-adaptation-beyond2lte.yaml? | 20:07 |
| Umeaboy | https://ghostbin.co/paste/oqv8btj | 20:08 |
| Umeaboy | Not entirely sure, but I think it was generated. | 20:09 |
| Umeaboy | Noooo.... Perhaps it was me that created it manually because it was supposed to exist, but didn't. | 20:09 |
| Umeaboy | I could be wrong though. | 20:10 |
| T42 | <elros34> so use droi-config from your github because that one in your local tree is differen | 20:11 |
| T42 | <elros34> t | 20:11 |
| T42 | <elros34> in patterns/ dir your only need to have *.inc files, everything else can be removed | 20:11 |
| T42 | <elros34> next time if you have some issues make sure your local droid-config/hal/kernel match that one in github | 20:13 |
| Umeaboy | OK. | 20:19 |
| piggz | rinigus: how much space does your hybris17 port take? | 20:51 |
| mal | piggz: my source tree without out folder is 99 GB | 20:59 |
| Umeaboy | Wow! Deleting the other stuf from the droid-configs worked. :) | 21:00 |
| piggz | mal: ah crap, i need more space | 21:00 |
| piggz | i have an old meamo sdk that takes 16gb, maybe that can go! | 21:00 |
| lpotter | archive it to an external harddrive... ya never know when you are going to need it! :D | 21:04 |
| piggz | yeah, suppose i can send it to the nas over night | 21:10 |
| piggz | mal: lpotter: operation free-space is underway .. will be able to start a new port tomorrow! | 21:35 |
| lpotter | nice | 21:38 |
| Umeaboy | I reach to building libgbinder and now it stopped: https://ghostbin.co/paste/mydw77 | 22:04 |
| Umeaboy | mal: Any idea? | 22:05 |
| deathmist_ | Umeaboy: always check logs first things first https://piggz.co.uk/sailfishos-porters-archive | 22:07 |
| Umeaboy | Ok. | 22:09 |
| Umeaboy | Skipping that package then. | 22:09 |
| mal | Umeaboy: sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper in libglibutil-devel libglibutil | 22:09 |
| mal | you can't just skip that, libgbinder is critical package for newer android bases | 22:10 |
| Umeaboy | OK. | 22:11 |
| Umeaboy | Number 1? | 22:11 |
| Umeaboy | Solution 1: install libglibutil-1.0.53-0.aarch64 (with vendor change) | 22:11 |
| Umeaboy | ? | 22:11 |
| mal | of course | 22:11 |
| Umeaboy | OK. | 22:11 |
| Umeaboy | Done. No errors. | 22:12 |
| Umeaboy | Does that mean I can skip to --gg ? | 22:12 |
| mal | how is that related to this issue? | 22:12 |
| Umeaboy | I was wondering if I should run rpm/dhd/helpers/build_packages.sh --mw or rpm/dhd/helpers/build_packages.sh --gg | 22:14 |
| Umeaboy | From this point. | 22:14 |
| mal | well you need to finish build in mw at some point | 22:14 |
| mal | you can skip the package you already built | 22:15 |
| Umeaboy | OK. | 22:15 |
| Umeaboy | mal: What about libgbinder-radio ? | 22:23 |
| Umeaboy | I chose n as it's related to the meego libgbinder version. | 22:35 |
| mal | Umeaboy: that is also required if you plan to have working modem | 22:37 |
| mal | Umeaboy: if you get any issues with vendor change then just install the packages manually like I mentioned earlier | 22:37 |
| Umeaboy | mal: sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper in libgbinder-radio didn't work. | 22:47 |
| mal | why? | 22:47 |
| *** Mister_Magister_ is now known as Mister_Magister | 22:48 | |
| mal | is that really the problem was the problem installation of the previously built package | 22:48 |
| Umeaboy | https://ghostbin.co/paste/o3enxk | 22:48 |
| mal | show the error you got earlier when you had issues building that | 22:49 |
| Umeaboy | mal: Didn't you see it here? | 23:06 |
| Umeaboy | https://ghostbin.co/paste/mydw77 | 23:06 |
| mal | didn't you install that libglibutil and the rerun build_packages.sh --mw and skip things before libgbinder and build that and everything after it | 23:07 |
| Umeaboy | mal: https://ghostbin.co/paste/n6vf5f | 23:13 |
| Umeaboy | That's where I am right now. | 23:13 |
| Umeaboy | I choose n for libgbinder | 23:14 |
| Umeaboy | Now I get: Build libgbinder-radio? [Y/n/all] | 23:15 |
| mal | I told you to build libgbinder | 23:20 |
| T42 | <leha155> @Umeaboy [deathmist_: Rome wasn't built in a day. :)], Hallow, after rebuild droid-hal, I have working droid-hal-init. But I have "ServiceManagement: Waited for hwservicemanager.ready for a second, waiting another..." only in logcat. I add ld.config.29.txt, but that didn't solve the problem, can you halp me? | 23:24 |
| T42 | <leha155> mount output https://pastebin.com/cAMV6GTX | 23:24 |
| Umeaboy | leha155: Check the hot-hadk? | 23:39 |
| Umeaboy | http://collabedit.com/kr9xx | 23:39 |
| Umeaboy | Your issue should be there as well as a solution. | 23:40 |
| mal | @leha155 is selinux enabled or disabled? is that hybris-17.1 based build? | 23:41 |
| Umeaboy | mal: If I recall correctly it is hybris-17.1 that leha155 is trying to build. | 23:42 |
| T42 | <leha155> selinux enabled, audit enabled, it is hybris-17.1. | 23:46 |
| mal | @leha155 is hwservicemanager running? do you have this line in droid-config spec https://github.com/mlehtima/droid-config-fp3/blob/master/rpm/droid-config-fp3.spec#L18 | 23:47 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!