| *** birdzhang1 is now known as birdzhang | 02:25 | |
| T42 | <nitin03 %lastname%> dmesg https://pastebin.com/T20jUS9P | 06:14 |
|---|---|---|
| T42 | <nitin03 %lastname%> systemctl list-unit-files --type=mount | 06:17 |
| T42 | <nitin03 %lastname%> returns this https://pastebin.com/RzF9d8ZM | 06:17 |
| T42 | <nitin03 %lastname%> https://pastebin.com/UZEkMxCx | 06:20 |
| T42 | <nitin03 %lastname%> https://pastebin.com/u3YDhgs8 | 06:21 |
| T42 | <nitin03 %lastname%> https://pastebin.com/n7JaGCXv | 06:25 |
| T42 | <nitin03 %lastname%> strace -f /sbin/droid-hal-init returns this...and I have masked the service https://pastebin.com/s2yY3Dka | 06:32 |
| deathmist | +@nitin03 well as you can see your ld.config.28.txt mount unit fails, that won't help you if you want DHI to load properly. do you even have a patched version of it taken from /system/etc/ and placed in /usr/libexec/droid-hybris/system/etc/ ? | 06:38 |
| deathmist | also please remove vendor-*.mount units, those are from my repo and as they're vendor related, they are highly device specific, I added them so I can fix cpusets without direct modification of the files and other more tiny things that are not important at all to get UI working | 06:39 |
| T42 | <nitin03 %lastname%> @deathmist [+@nitin03 well as you can see your ld.config.2 …], Yes I have it...and also the .mount file and its symlink | 06:40 |
| T42 | <nitin03 %lastname%> ld.config.28.txt https://pastebin.com/rv9qrzmn...also mount file https://pastebin.com/LnUsjcKZ | 06:45 |
| T42 | <nitin03 %lastname%> https://pastebin.com/rv9qrzmn | 06:45 |
| T42 | <birdzhang> nitin03 use /system/bin/strace if you have restart_syscall spam | 06:46 |
| deathmist | @nitin03 well that file is identical to my ld.config, show me YOUR copy of /system/etc/ld.config*.txt, you can't just copy paste my config files in your own repo necessarily | 06:48 |
| rinigus | r0kk3rz or mal: please create nemo:devel:hw:sony:h8216 at obs and set me as a maintainer. | 06:49 |
| deathmist | @nitin03 the file may not even be by the same name and that would explain your bind mount unit failing | 06:50 |
| deathmist | if that's not it full journalctl output after boot would be nice to see why it would be failing then | 06:53 |
| T42 | <nitin03 %lastname%> @birdzhang [nitin03 use /system/bin/strace if you have res …], strace https://pastebin.com/s2yY3Dka | 07:19 |
| T42 | <birdzhang> nitin03 as deathmist said, show your /system/etc/ld.config*.txt | 07:20 |
| T42 | <nitin03 %lastname%> @birdzhang [nitin03 as deathmist said, show your /system/ …], https://pastebin.com/rv9qrzmn | 07:38 |
| T42 | <nitin03 %lastname%> @deathmist [@nitin03 well that file is identical to my ld. …], https://pastebin.com/rv9qrzmn | 07:38 |
| T42 | <birdzhang> okay, then show ls -l /system/etc/ and ls -l /vendor/etc/ | 07:39 |
| T42 | <nitin03 %lastname%> https://pastebin.com/FThjDbUU | 07:42 |
| T42 | <nitin03 %lastname%> https://pastebin.com/5U95Ge0N | 07:44 |
| T42 | <birdzhang> it's /system/ect/ld.config.vndk_lite.txt on your device, not ld.config.28.txt | 07:47 |
| T42 | <nitin03 %lastname%> @birdzhang [it's /system/ect/ld.config.vndk_lite.txt on yo …], so need to change the name of file in sparse folder | 07:57 |
| T42 | <nitin03 %lastname%> right??and also the name of mount file and the symlink | 07:59 |
| T42 | <birdzhang> yup | 08:00 |
| T42 | <nitin03 %lastname%> @birdzhang [yup], thank u..i will check it out and report in the evening | 08:01 |
| T42 | <Thinhx2> @mal [@Thinhx2 you need to provide logs, journalctl …], https://del.dog/uqewuqefob | 09:34 |
| T42 | <Thinhx2> https://del.dog/pabegiqabi | 09:37 |
| r0kk3rz | rinigus: https://build.sailfishos.org/project/show/nemo:devel:hw:sony:h8216 | 11:35 |
| rinigus | r0kk3rz: thank you very much! | 11:37 |
| mal | r0kk3rz: what is h8216? why not use the device codename (which usually matches both single and dual-sim ones) | 11:38 |
| mal | or even platform name | 11:38 |
| mal | rinigus: ^ | 11:39 |
| r0kk3rz | its a bit inconsistent yeah | 11:39 |
| rinigus | mal: good questions. $DEVICE is h8216 and I was thinking on how to add support for other similar devices (dual sim, xz2c (single/dual), xz3). but I thought I will partition it into platform/implementation a touch later | 11:40 |
| rinigus | aka tama/akari | 11:41 |
| Mister_Magister | sony is so annoying with that | 11:41 |
| mal | rinigus: handling single vs dual sim is simple and can be checked in for example mer-hybris ganges repos | 11:41 |
| mal | rinigus: I would say using at least akari for that, you can see that handling those two different codenames is here https://github.com/mer-hybris/droid-config-sony-ganges-pie/blob/master/rpm/droid-config-i3113.spec#L1 | 11:44 |
| rinigus | mal: I did check it and it seems to me over-engineered (maybe not as you want to make it into separate images). I was thinking to make extra rpm (dual-sim or similar) that will just add that dual sim configs to the single sim device base. | 11:44 |
| rinigus | mal: OK, let's rename it into akari instead of h8216. later I was thinking to add nemo:devel:hw:sony:tama and make akari to use tama as a base. So, most of packages will be in tama. | 11:49 |
| rinigus | mal: now a bit more difficult aspect - what do I do with droid-system-sony-pie-template and vendor package? those are needed to make image, but I am not sure we can push them to obs for legal reasons | 11:51 |
| rinigus | ... there was a note regarding these packages in xperia 10 howto. although it was about github | 11:51 |
| Mister_Magister | Sooooooooooooo, mal sfos update on obs when | 13:39 |
| Mister_Magister | mal: if 3.2 enables selinux then do i have to enable selinux in kernel from now on or not yet? | 13:40 |
| T42 | <hacker12455> question, are the quirks applied correctly here? | 13:45 |
| T42 | <hacker12455> https://github.com/luka177/droid-hal-cedric/blob/master/droid-hal-cedric.spec | 13:45 |
| deathmist | Mister_Magister: I'm already building 3.2 images, I've made no modifications and everything seems to function like before even with SELinux still off :) | 13:50 |
| Mister_Magister | but is 3.2 on obs? | 13:51 |
| Mister_Magister | heh search on obs still on fire | 13:52 |
| Mister_Magister | there is no 3.2 on obs | 13:52 |
| deathmist | I'm a bit clueless still, what does that mean? I've just built local images with 3.2 sb2 target and RELEASE set | 13:53 |
| Mister_Magister | heh | 13:57 |
| mal | Mister_Magister: normally selinux should not be enabled yet on ported devices and things should be done as before | 14:22 |
| Mister_Magister | mal: normaly? | 14:37 |
| mal | well unless you want to port the needed patch for system/core etc then go ahead | 14:39 |
| Mister_Magister | ooh xd | 14:40 |
| T42 | <BusterBg_18> HI all, I started fresh once again and I´m stuck at this when creating rootfs "file /etc/ofono/ril_subscription.conf conflicts between attempted installs of droid-config-sirius-1-1.armv7hl and ofono-configs-mer-1.21+g | 15:24 |
| T42 | <BusterBg_18> it49.1-1.22.1.jolla.armv7hl" | 15:24 |
| T42 | <BusterBg_18> My bad, I forgot some lines on droid-config*.spec :p | 15:27 |
| T42 | <adampigg> Mister_magister if like me, you do it wrong and jimust set selinux to permissive instead of turned off, it wikl break in 3.2 when systemd turns it on | 15:56 |
| T42 | <DylanVanAssche> @adampigg What about mainline Linux? For libhybris I know you had/have to turn it off in some way. Wondering if the PineTab/PinePhone is affected by this? | 17:40 |
| T42 | <hacker12455> any idea why USB would not work at all? | 17:43 |
| T42 | <hacker12455> it only charges | 17:43 |
| T42 | <hacker12455> usb_moded in debug mode | 17:43 |
| T42 | <hacker12455> Fwd from hacker12455: https://bpaste.net/show/YTWAG | 17:43 |
| T42 | <hacker12455> logcat and journal don't show anything interesting | 17:44 |
| T42 | <adampigg> @DylanVanAssche [@adampigg What about mainline Linux? For libhy …], probably not, but we may still need selinux configs | 18:14 |
| Mister_Magister | @adampigg i have no idea what you just said | 18:45 |
| Thaodan | @DylanVanAssche the reason was that the Android SELinux rules could bug sfos this is why they were disabled at the kernel command line, as the pine devices don't have other SELinux rules there's no reason to disable it except maybe issues with the Sailfish OS base SELinux rules. | 18:55 |
| T42 | <Lukapanio> where can i get anbox rpms? | 19:16 |
| T42 | <Lukapanio> I built patched 3.18 kernel | 19:17 |
| rinigus | evening! I have issues with suspend. for starters - when on charger, I don't have any wakelock active and there is a bit of suspend. is it now expected on sfos? | 19:19 |
| T42 | <BusterBg_18> So what Im supposed to get when running strace -f /sbin/droid-hal-init? https://pastebin.com/wpNcbwPR | 19:19 |
| T42 | <elros34> @BusterBg_18 use /system/xbin/strace instead | 19:20 |
| vknecht | @LukaPanio: some old notes, ymmv... | 19:21 |
| vknecht | devel-su | 19:22 |
| vknecht | ssu ar home_birdzhang_anbox http://repo.merproject.org/obs/home:/birdzhang:/anbox/sailfishos_3.0.1.11_armv7hl/ | 19:22 |
| vknecht | pkcon refresh | 19:22 |
| vknecht | pkcon install anbox-sailfishos anbox-sailfishos-image-mixed-32-64-bit | 19:22 |
| vknecht | rm -f /var/lib/anbox/overlay/system/lib64/libGLESv* | 19:22 |
| vknecht | cp /vendor/lib/egl/libEGL_adreno.so /var/lib/anbox/overlay/vendor/lib/egl/ | 19:22 |
| T42 | <BusterBg_18> @elros34 Not found | 19:22 |
| vknecht | (eg. might want to use sailfish_latest_armv7hl/ instead of 3.0.1.11 build, maybe your device doesn't need mixed image | 19:23 |
| T42 | <elros34> @BusterBg_18 try to find it somwhere in system or vendor partition or download 64bit version somwhere from net. What you are trying to solve? | 19:23 |
| vknecht | and maybe the egl lib rm/cp isn't needed for you) | 19:23 |
| T42 | <BusterBg_18> @elros34 Trying to get a freaking gui on my 16.0 port | 19:24 |
| T42 | <Lukapanio> Hmm wifi is broken | 19:25 |
| T42 | <hacker12455> did you regenerate rootfs? | 19:25 |
| T42 | <hacker12455> if not this might be the issue | 19:25 |
| T42 | <nitin03 %lastname%> @BusterBg_18 [@elros34 Not found], try this /system/bin/strace -f /sbin/droid-hal-init | 19:26 |
| T42 | <BusterBg_18> nitin03 that worked | 19:29 |
| T42 | <BusterBg_18> dmesg https://pastebin.com/PgEBk0Sw | 19:32 |
| T42 | <nitin03 %lastname%> ld.config mounted | 19:34 |
| T42 | <nitin03 %lastname%> https://pastebin.com/BqUnPj3W | 19:34 |
| T42 | <BusterBg_18> nitin03 for what base are you porting? | 19:38 |
| T42 | <nitin03 %lastname%> dmesg https://pastebin.com/UVmX7FuY | 19:38 |
| T42 | <nitin03 %lastname%> strace https://pastebin.com/jW24Et5r | 19:38 |
| T42 | <nitin03 %lastname%> journalctl -a https://pastebin.com/xwMrZc6V | 19:38 |
| T42 | <nitin03 %lastname%> @BusterBg_18 [nitin03 for what base are you porting?], lineage-16.0 | 19:38 |
| T42 | <BusterBg_18> Do you have gui? | 19:42 |
| T42 | <nitin03 %lastname%> @BusterBg_18 [Do you have gui?], no | 19:42 |
| T42 | <BusterBg_18> Im having errors almost the same to yours | 19:44 |
| T42 | <BusterBg_18> Specially on strace | 19:45 |
| T42 | <BusterBg_18> mal: how is ld.config.vndk_lite.txt supposed to be modified? | 19:46 |
| T42 | <nitin03 %lastname%> @BusterBg_18 [Im having errors almost the same to yours], if you find any solution please do share | 19:47 |
| deathmist | @BusterBg_18 @nitin03 weren't you both doing 16.0 base and needed adapted ld.config.vndk_lite.txt files? can you share the file with me, I'll try to see if I can come up with something | 19:52 |
| T42 | <BusterBg_18> Sure thing | 19:53 |
| T42 | <BusterBg_18> https://github.com/BusterGR18/droid-config-sirius/blob/master/sparse/usr/libexec/droid-hybris/system/etc/ld.config.vndk_lite.txt | 19:53 |
| T42 | <BusterBg_18> deathmist: also, do you know whats the issue with the property area stuff? is it linked to vndk_lite.txt? | 19:58 |
| T42 | <BusterBg_18> Because I swear to god, debugging this port is getting on my nerves | 19:58 |
| deathmist | @BusterBg_18 nah, it's a "plat_property_contexts" file you're missing in SFOS /, for me a "ln -s /system/etc/selinux/plat_property_contexts /plat_property_contexts" was enough. check if the file exists first for yourself | 19:59 |
| deathmist | it could also be in your dev env out/ dir somewhere | 19:59 |
| deathmist | in which case you need to add it as a straggler file and rebuild droid-hal & make new image | 19:59 |
| T42 | <BusterBg_18> I already symlinked it tho :c | 20:00 |
| T42 | <nitin03 %lastname%> @deathmist [@BusterBg_18 @nitin03 weren't you both doing 1 …], https://pastebin.com/7UhhpJWN | 20:00 |
| piggz | vknecht: are all your PRs ready? | 20:01 |
| deathmist | looks like the files are identical, will be nice to figure out a way to fix stuff for vndk_lite ld.config files, @BusterBg_18 does the file exist? do you not get anything with "find $ANDROID_ROOT/out/ -name *property_contexts" ? for me that was enough, maybe you're missing something else too (like this ld.config stuff being resolved first) | 20:02 |
| T42 | <nitin03 %lastname%> @deathmist [@BusterBg_18 nah, it's a "plat_property_contex …], file exists | 20:02 |
| kimmoli | @adampigg http://kimmoli.com/tauhkaa/20191106_213634.jpg | 20:04 |
| piggz | kimmoli: whats that from? | 20:04 |
| T42 | <BusterBg_18> deathmist: the file exists on device itself on out is missing | 20:05 |
| deathmist | although for you both it seems the namespaces that I needed to modify aren't isolated, not sure what this means but maybe you don't need permitted paths stuff then? | 20:05 |
| kimmoli | piggz: CR-V rear | 20:05 |
| piggz | kimmoli: ah, no SUper 5 ;) | 20:05 |
| T42 | <nitin03 %lastname%> @deathmist [although for you both it seems the namespaces …], I was thinking that ...in the strace it gave a warning | 20:05 |
| T42 | <BusterBg_18> AFAIK isolated meant (in android terms) that the vendor is designed to only run that specific android version | 20:05 |
| T42 | <BusterBg_18> Instead of being able to boot everything | 20:06 |
| T42 | <BusterBg_18> Or something like that | 20:06 |
| T42 | <BusterBg_18> And I quote xda: "If the value is false, then your device can only use the GSI for the same on-device version of the OS" | 20:07 |
| T42 | <nitin03 %lastname%> I don't have property_contexts file in out directory | 20:07 |
| deathmist | @BusterBg_18 in that case we're looking in the wrong places right now, I didn't even get telnet before fixing mine so I can't exactly say what happened before that; how about you both remove/mask the ld.config* mount units and remove those related files in /usr/libexec/..., get journal after booting and I'll take a look | 20:08 |
| deathmist | it's ok if you don't have *property_contexts in out/, I didn't either and that symlink has fixed my property area issue on both 15.1 and 16.0 (I had to rename it from just "property_contexts" to "plat_property_contexts" during Pie rebase tho, you could try "property_contexts" as filename too) | 20:10 |
| deathmist | no need to make new image, just do it on the device. also make sure no stuff like droid-hal-init are masked so the boot can be attempted normally | 20:11 |
| T42 | <BusterBg_18> Sure thing just masked and deleted the file | 20:12 |
| T42 | <BusterBg_18> Jesus, the file is extremely big | 20:15 |
| vknecht | piggz, should be okay, but needs more external testing... | 20:15 |
| vknecht | 1 - snapshot: got a report from Mal that it didn't work for him on a device | 20:16 |
| vknecht | 2 - sdcard: have a bit of difficulties to test with usb-otg due to faulty otg/key here | 20:16 |
| vknecht | 3 - metadata: should be ok, but needs a bit of graphical/layout polish https://github.com/piggz/harbour-advanced-camera/pull/93#issuecomment-549447551 | 20:16 |
| piggz | vknecht: which order should i try? | 20:16 |
| vknecht | in this order I think | 20:16 |
| piggz | ok | 20:17 |
| vknecht | thanks for looking at it :-) | 20:19 |
| T42 | <BusterBg_18> deathimist: sorry for the lack of indentation https://paste.mozilla.org/PELFf1HE | 20:19 |
| vknecht | piggz, maybe sdcard is better at the end, since it changes FSOperations to be globally availabe so others would need rebase | 20:23 |
| deathmist | @BusterBg_18 property area error is way down in logs and DHI starts fine first time (as it should, it doesn't really work after that) so that's not the issue | 20:25 |
| T42 | <BusterBg_18> Hmmm.. so its actually related to vndk_lite isnt it? | 20:27 |
| T42 | <nitin03 %lastname%> @deathmist [@BusterBg_18 in that case we're looking in the …], https://pastebin.com/ALQsWC6i | 20:27 |
| deathmist | @nitin03 didn't I say something about this specifically? "droid-hal-init.service: Cannot add dependency job, ignoring: Unit droid-hal-init.service is masked." | 20:28 |
| T42 | <nitin03 %lastname%> @deathmist [@nitin03 didn't I say something about this spe …], oh sorry | 20:30 |
| deathmist | @BusterBg_18 does running /system/bin/surfaceflinger show bootanim? (not sure if it required "export EGL_PLATFORM=hwcomposer" to be run before too) | 20:32 |
| deathmist | not sure if your "ofonod: Unknown RIL transport: binder" issue can cause no graphics as you don't seem to have "ofono-ril-binder-plugin" in your patterns. I'll have to go now tho so I'll see you later | 20:43 |
| T42 | <BusterBg_18> No prob, I need to grab something to eat anyways :p | 20:46 |
| T42 | <BusterBg_18> Im going to add that to patterns, repackage and tell the dude that's helping me test to try it and also test if bootanim shows | 20:48 |
| T42 | <adampigg> (Photo, 1920x1080) https://irc.thaodan.de/.imgstore/XLfYbE3fvj.png | 21:09 |
| mal | nice | 21:26 |
| T42 | <BusterBg_18> mal: do you have an idea regarding ld.config.vndk_lite.txt modifications? | 22:51 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
