Wednesday, 2021-05-19

deathmist_Umeaboy: 17.1 is cutting edge in hybris currently, 18.1 based adaptations aren't possible yet. also have you even tried to build without the patches? they're mandatory for 16.0+00:06
Umeaboydeathmist_, That's why I asked if they're mandatory.00:07
malUmeaboy: it is, check the PR in faq repo00:09
malI mean for hybris-17.1 things00:09
deathmist_oh, I never %define android_version_major 10 to as no one mentioned it anywhere back then, welp that was probably why it was always broken for me then00:12
Umeaboymal: hybris-16.0 and hybris-17.1 have the same structure?00:12
deathmist_I'll take a crack at it later I suppose; the space requirements for hybris-17.1 are insane though00:12
UmeaboyOK. I have to change my spec then, deathmist_00:12
deathmist_Umeaboy: as per mal's earlier message00:13
riniguspiggz: version release component is kept in one of the files in build. Didn't know about such jolla store limitation03:35
mighty17<T42 "<elros34> @mighty17 you are miss"> Oh right derp xD, now to figure out how to get wifi working, usb internet doesn't work as well :(04:31
riniguspiggz: I will add an option to disable release mangling for a project. Sounds like it will make sense for store projects04:43
T42<Zainal %lastname%> Any info porting sailfish 4 for oneplus X ? Thanks05:34
mighty17Use the hadk guide for porting :D06:03
piggz_rinigus: does tbuilder do the mb2 prepare step?08:45
riniguspiggz_: it should be done by mb2 automatically if you don't use shadow builds. if I remember it right10:07
rinigusbut I can check the sources tonight10:08
T42<adampigg> Rinigus,.might try and  build xulrunner as a stress test!10:21
rinigus@adampigg: but have you seen that prep step was not applied?10:25
T42<adampigg> No, just wondering10:54
rinigus@adampigg: let me know how it goes11:04
rinigusjust checked - if not shadow build then `-p` is added to `mb2` command. so, should be fine11:08
rinigus@adampigg ^11:08
*** Nico[m] is now known as Nico-old-defunct12:15
mighty17hello, i need some help for sfos my device has broadcom bluetooth and acc to hadk-faq i need to do something like this commit but mine is bcm4330 and the commit has bcm4345c0 with `bcm4345c0.conf` but bcm4330 doesnt have a `bcm4330.conf` file so what am i supposed to do13:05
calebccffI wonder why the FaQ isn't just a github wiki, those allow anyone to edit... Rather than the PR process which is poorly suited to documentation13:06
poetasterthat's a good question. I had elaborate help that clarified what was only anecdotal in the FAQ13:07
mighty17calebccff: are u caleb? the op6 caleb?13:08
calebccffmighty17: Yeah that's me :>13:08
mighty17you are everywhere xD13:08
calebccffSFOS is where I started aha13:09
mighty17oh thats cool!13:09
mighty17well the hadk-faq just says `add configs:` i mean thats very vague and i have no clue on what to do for that :/13:15
mighty17@edp_17 it seems our devices share same module bcm4330 mind sharing how did you do it xD13:18
poetasterI'v joined #sailfishos and -porter on and posted to the forum.14:17
poetasterfor those who haven't heard yet.
T42hungarianboi was added by: hungarianboi14:48
mighty17<poetaster "I'v joined #sailfishos and -port"> Well I'm on matrix do I need to change?14:59
poetaster@mighty17, I'm just being proactive. but I'm not going to stick to freenode long.15:03
mighty17Yeah that is a serious issue about freenode, thanks for raising awareness15:07
mighty17I found it here didn't know about that15:07
poetasterI know that none of us has time for this but I started a thread:
T42<adampigg> maybe get mal to move onto telegram :D15:10
mal@adampigg :)15:13
mal@adampigg still relevant
T42<adampigg> yup :)15:17
rinigusohhh, it is damn tempting to ping those T 4 215:19
malsome channels have been moving to oftc15:21
rinigusif/when we move, please make sure that there is matrix bridge15:23
malrinigus: oftc has a matrix bridge at least, not sure abotu libera15:24
riniguslet me check15:24
rinigusno good hits yet; didn't find in their web either. probably too new, but I guess that will come as well15:27
poetaster@adampigg you might be onto something. it's war at libera riht now. the freenode bots are spamming.15:27
poetasteradam, I'm already running bridges between irc/xmpp/matrix and mattermost so. ... telegram?15:28
poetasterOk. so maybe I'll : set up a matrix server with an irc bridge for us.15:30
poetasterload average: 0.02 0.02 0.0215:31
T42<adampigg> its probably not wise to just jump into something immediately, best to get a decent community consensus over the following week or so ?15:33
Mister_Magister@adampigg sfos meeting would do i think15:35
T42<adampigg> yup15:36
poetaster@adampigg ... a hostile power has my personal data... I don't know about waiting for consensus15:37
poetasterOk, then I'll go dark till the powers that be decide to do nothing.15:37
poetasterthey (freenode which is no longer freenode) are being vERY hostile15:38
poetasterah, wait, maybe a great firewall has risen.15:39
malthere is the sailfish community meeting tomorrow so maybe that is a good place to talk15:42
T42<adampigg> mal: can we get you on slack? :D15:42
T42<adampigg> or, ms teams??15:42
poetaster9 AM tomorrow? I have homeschooling. sigh.15:44
poetasterit's too late for tomorrows meeting (well, at least 3 days before meeting is in bold).15:49
malwell there is the free discussion part of it15:49
poetasterah, that's true. damn. I'm just sick and tired of plunking my child in front of a screen.15:53
poetastersigh. thank god he's playing outside.15:53
*** Keij0_ is now known as Keij015:58
poetasternico (kuroNeko) put it  on the agenda16:11
Nico-old-defunctI need to register a new nick for the meeting...16:14
poetasteroh, damn. I just changed my pw and mail.16:20
poetastertoo late anyway. sigh.16:20
pketosure the topic can be discussed in tomorrows meeting even though it didn't meet the 3 days deadline16:35
Nico-old-defunctThere may not be an official answer tomorrow, but since there are no other topics anyway, I'm pretty sure it would have been discussed in any case16:37
pketoit's not a hard deadline, "Please _try_ and have your topics ready at least 3 days before the meeting"16:37
riniguspiggz_: I have added an option to tbuilder allowing you to keep release as is in spec. see README and example config.yaml under "project" subfolder of tbuilder19:06
piggzrinigus: i also thought of another nice-to-have...19:23
piggzthe option of appending the build target in the name/version?19:23
riniguspiggz: not sure. usually we have separate repos for different targets. that allows users to switch between SFOS versions and get them linked with the correct repo.19:25
rinigusif you have target in the name, amazfish-sfos-4.0.1 will disappear when you switch to 4.119:25
piggzrinigus: which is what i need ... both amazfish and advanced-camera break between 4.0 and 4.1 :)19:26
piggzi need to host multiple versions, but dont have a way to do that atm19:26
rinigusseparate repositories?19:27
piggzbecuase 4.1 isnt obs, and all the ea users are piping up19:27
piggzrinigus: yeah, just a temprary priblem19:27
riniguspiggz: not sure how temporary it is19:27
piggzrinigus: is that implying something?19:28
piggzie, 4.1 may not be on obs?19:28
piggzactually, it will be a double pain, as aarch64 doesnt work anyway19:29
rinigusno, just observation that it is with every release these days.19:29
piggzand already somone released an advanced-cam 64bit version ahead of me ... tho, without the 4.1 patches19:29
riniguswould be nice to have separate some place where packages for different sfos versions can be pushed. openrepos, ideally19:29
rinigusfor port, I can put repos into the cloud and just config to get them from there. for apps, not so nice19:30
piggzi could hack it on openrepos iwth seperate user accounts :D19:30
piggzbut that would become unmanagemable very quickly!19:31
malrinigus: about your tama actdead issue, the log suggests charger state is unknown or off and triggers reboot19:38
riniguspiggz: ideally, we need something where we can push whole repos. and which users can easily find and configure19:39
rinigusmal: thanks for looking into it!19:39
piggzrinigus: umm, like obs? :D19:39
malrinigus: that line "DSME[2720]: state: Actdead (charger: off(?), alarm: not set)"19:39
rinigusmal: before forcing reboot with actdead, I usually unplug the charger. let me try without unplugging and see if the same line will be there19:41
malrinigus: also the usb audio issue, remember to make the udev rule for ignoring built-in audio card if you enable module-udev-detect in pulseaudio19:41
malrinigus: I though the log meant it automatically rebooted?19:42
rinigusmal: it did at least once. but that was when I plugged to PC port. (although, not sure it was relevant). frequently it is stuck with lid black screen19:43
malrinigus: do you have this
rinigusmal: yes, I do19:43
piggzrinigus: did you have the hadk somewhere in text format?19:45
rinigusthat's for aosp10 based and using tbuilder19:50
rinigusmal: this time it was hanging on lid screen. log at
rinigusDSME later found the charger - DSME[2648]: state: Actdead (charger: on, alarm: not set)20:00
rinigusagain, had to force reboot20:01
malrinigus: is that the whole log?20:14
piggzmal: starting a fresh port, how to you minimise a full lineage 17.1 repo?20:41
malpiggz: what do you mean? the hybris-17.1 is somewhat minimized already20:44
piggzmal: oh, i was talking about the los android.git .. i couldnt see a hybris-17.120:45
piggzmal: my bad ... it was in the stale list!20:46
piggzmal: python version error?20:49
piggz  File "/parentroot/parentroot/data/piggz/mer/android/droid.gs290/.repo/repo/", line 7920:49
piggz    file=sys.stderr)20:49
piggz        ^20:49
piggzSyntaxError: invalid syntax20:49
piggzthis should help
piggzmal: does the ubuntu trusty rootfs have python 3.6, which is required for repo?20:54
T42<ruspartisan> Last time I checked, chroot had very old Ubuntu (14.04 afair) that was too old for python required for repo20:55
piggzyeah, im adding a ppa20:56
malpiggz: I think it has if you install it manually21:20
malpiggz: was it 3.5 or 3.6 in there21:20
piggzmal: it has 3.5, but 3.6 is required21:20
piggzgoogling shows approprate ppa21:20
malpiggz: hmm, I thought the 3.5 is the hard minimum requirement for repo, 3.6 is the soft requirement21:26
T42<b100dian> mal: got minimer working! The composer stack corruption was because system overlaying of droid-hybris/system (mixing libs)21:27
mal@b100dian ok, good21:28
piggzABUILD_SDK [GS290] piggz@linux-f1uu:/mer/android/droid.gs290$ repo init -u git:// -b hybris-17.121:28
piggzrepo: warning: your Python 3 version is no longer supported; Please upgrade to Python 3.6+.21:28
malpiggz: yes, but it doesn't exit at that point afaik
malpiggz: older than 3.5 will cause exit21:29
deathmist_piggz: I've started building droid HAL on my host since that issue appeared, I know it can be fixed but doing it on host is also slightly faster for me21:30
malI have built all newer android bases directly in ubuntu 20.04 (and 18.04 before that) for a long time21:31
piggzdeathmist_: yeah, i imagine it may be faster ... my host is opensuse tho, nothing like a ubuntou setup21:31
malpiggz: so the requirements seem wrong then21:31
malpiggz: run repo on host, everything in HABUILD21:32
mal*everything else21:32
piggzyeah, that would work too21:32
piggzbut, i now have python3.6 in habuild anyway21:32
UmeaboyI have tried to pull my device vendor blobs, but fail to do so. I did extract the vendor.img and mounted the image as rw to a random folder and the couldn't find the majority of the files even though I used the right zip described in the file for Lineage. Need help.22:26
UmeaboyI extracted the vendor.img.lz4 and ran lz4 -d vendor.img.lz4 vendor.img and then I ran simg2img vendor.img vendor_unpacked.img and that went fine. I then mounted the vendor_unpacked.img to a random folder that I created and that went fine. I even remounted the folder as rw.22:28
UmeaboyThere must be some step I'm missing.22:28
UmeaboyHere's the result:
T42<b100dian> I think extracting vendor files is for creating your own image. I am using the one on device for example, not customized22:42
UmeaboyI think I made a boo boo.22:42
UmeaboyI have to follow this page:
malUmeaboy: what device is that?22:44
Umeaboymal: The beyond2lte. I realized that I did the wrong thing when I tried to pull the blobs.22:45
malwhat vendor?22:45
UmeaboyAnd now it found the blobs. :(22:47
UmeaboySo no harm, no foul. :)22:47
T42<b100dian> " droid-hal-init: Could not find '' for ctl.interface_start"22:52
T42<b100dian> I don't have the 2.1 rc file on my device, only the 2.3 file22:52
T42<b100dian> Do you know where is 2.1 referenced from?22:53
Umeaboyb100dian: A previous update zip perhaps?22:56
T42<b100dian> a couple of seconds above it says `droid-hal-init: Parsing file /vendor/etc/init/`23:10
T42<b100dian> Umeaboy: what
Umeaboyb100dian: Which Lineage base and what vendor?23:11
T42<b100dian> 17.1, xiaomi23:12
Umeaboyb100dian: Which modelname?23:16
T42<b100dian> tucana23:16
Umeaboyb100dian: What the build date on your current running Lineage system?23:19
Umeaboy20200907 ?23:19
UmeaboySettings -> About phone23:20
T42<b100dian> Hmm I just build it yesterday I think, The vendor has Oct 25 22:15:30 CST 202023:21
T42<b100dian> *built23:21
UmeaboyCheck if you can find it here:
UmeaboyLook here otherwise:
UmeaboyAny of the previous zips.23:27
T42<b100dian> Oh, you mean I _should_ have a 2.1 service23:29
T42<b100dian> I'll be looking, thanks23:29
T42<b100dian> I'll try to define it first to point to the 2.3 one:)23:30
T42<b100dian> but the question is how to determine which file was parsed that init jumped to the conclusion that it should start it in the version 2.123:30
UmeaboyAsk the person that made the device repo.23:31
UmeaboyAnd the vendor repo of course.23:31

Generated by 2.17.1 by Marius Gedminas - find it at!