| T42 | <elros34> oh I thought I was losing my mind but found this line in updater-script: run_program("/sbin/sh", "-c", "busybox cp -R /system/xt897/* /system/")); | 00:15 |
|---|---|---|
| liux | cannot fetch https://releases.jolla.com/jolla-apps//armv7hl/repodata/repomd.xml since one day before. it's my network issue or jolla server issue? | 04:09 |
| mccreary | FYI, from my recovery: "# getprop ro.product.device | 05:29 |
| mccreary | asanti_c | 05:29 |
| mccreary | Historically 'asanti_c' was the CDMA build, but the separate builds were killed after cm-10 | 05:30 |
| mccreary | (ro.build.product is not set) | 05:32 |
| mccreary | Also, META-INF/com/google/android/updater-script from sailfishos-moto_msm8960_jbbl-release-3.0.2.8-alpha-9.zip doesn't have that extra run_program() call | 05:36 |
| mccreary | That is the installer ZIP linked from https://forum.xda-developers.com/photon-q-4g-lte/general/sailfishos-3-photon-q-t3870979 via https://is.gd/7Wrp4W | 05:38 |
| mccreary | NM, I see the line in updater-script from the cm-11 installer ZIP | 05:44 |
| mccreary | ifelse(is_substring("sprint", getprop("ro.boot.carrier")), run_program("/sbin/sh", "-c", "busybox cp -R /system/xt897/* /system/")); | 05:44 |
| mccreary | ro.boot.carrier is set to sprint in my recovery | 05:45 |
| mccreary | liux: I get a 404 error when trying to fetch https://releases.jolla.com/jolla-apps//armv7hl/repodata/repomd.xml (after it redirects to https://dvvwui2qyedk6.cloudfront.net/jolla-apps//armv7hl/repodata/repomd.xml) | 05:49 |
| vknecht | mccreary, seems to work if you insert a version number or "latest" between "//" | 06:56 |
| Mister_Magister | liux: u just got wrong url | 08:03 |
| liux | Mister_Magister, do you know how to make it redirects to correct url? it occurs when i execute "sudo mic create fs" to create image. and had no this error before yesterday. | 08:15 |
| vknecht | liux, probably you forgot the "RELEASE=3.0.3.9" line before mic (or whatever version) | 08:17 |
| liux | vknecht, yes you are right. I made a stupid mistake. just re-entered sdk after exited it, and did not re set RELEASE env | 08:21 |
| liux | vknecht, thank you very much | 08:22 |
| vknecht | you're welcome, it's an easy mistake to do... would be nice if mic warned/stopped when it's missing... | 08:24 |
| liux | may i have another question? I've built droid-config-$DEVICE-ssu-kickstarts-*.rpm on obs. how to use it to make image? should I unpack it and get the .ks for building? | 08:42 |
| vknecht | liux, yes | 08:44 |
| liux | vknecht, ok let me try. thx | 08:52 |
| vknecht | hmm, looks like all my sparse stuff don't end up in the rootfs archive, but can see them in target (comparing with other device, it shouldn't be the case it seems) | 09:02 |
| liux | the built droid-config-$DEVICE-ssu-kickstarts-*.rpm contains repo url of http://repo.merproject.org/obs/nemo but not my home repo | 09:11 |
| vknecht | you can correct those by hand I guess | 09:26 |
| liux | yes, i can correct it by hand. perhaps it'ok after i applied the nemo repo for my device | 09:27 |
| *** agander_ is now known as agander | 10:39 | |
| *** JustGerd02770 is now known as Gerd | 10:46 | |
| *** Gerd is now known as JustGerd | 10:46 | |
| T42 | Rockman_X was added by: Rockman_X | 11:20 |
| *** sicelo_ is now known as sicelo | 11:42 | |
| piggz | jusa, mal: is pulseaudio-modules-droid known to work on android 9? | 13:21 |
| mal | piggz: yes | 13:36 |
| mal | piggz: what error do you have? | 13:37 |
| mal | piggz: do you have /dsp symlink? | 13:42 |
| mal | piggz: also check /etc/acdbdata symlink validity | 13:43 |
| piggz | mal: lets check..... | 13:52 |
| piggz | mal: what should /dsp link to? | 13:53 |
| mal | probably /vendor/dsp | 13:53 |
| mal | do you have anything like that on the device | 13:54 |
| piggz | mal: i have a partition called dsp | 13:54 |
| mal | piggz: is it mounted? or even in fstab? | 13:55 |
| piggz | mal: no, but i also have a /vendor/dsp | 13:56 |
| mal | piggz: is it in fstab? | 13:56 |
| mal | piggz: how about that acdbdata symlink, maybe grab logcat | 13:56 |
| lilmayofuksu | hi guys. i just ported sailfishos to my device but when i try to flash it gives "realloc: Out of memory" and "tar: out of memory" errors | 13:57 |
| piggz | mal: and /etc/acdbdata doesnt link to anything...nothing like that in /system/etc | 13:57 |
| mal | piggz: how about in vendor? | 13:57 |
| mal | lilmayofuksu: that is a known issue on many devices, do you have older twrp for the device? | 13:58 |
| mal | lilmayofuksu: new twrp has too strict memory management | 13:58 |
| lilmayofuksu | i have a lg-g2 so it should have a older twrp | 13:58 |
| piggz | mal: yes..... | 13:59 |
| lilmayofuksu | what version i should look for | 13:59 |
| mal | lilmayofuksu: not sure really, I haven't had that problem myself | 13:59 |
| mal | lilmayofuksu: just try maybe a year old for example or something, probably just trial and error | 13:59 |
| mal | piggz: try fixing the symlink, also logcat :) | 14:00 |
| piggz | mal: aye im on it! | 14:00 |
| piggz | :D | 14:00 |
| lilmayofuksu | i hope it doesn't randomly crash cuz this phone literally has no screen | 14:01 |
| mal | lilmayofuksu: no screen? | 14:01 |
| lilmayofuksu | yea | 14:01 |
| mal | why? | 14:01 |
| vknecht | lilmayofuksu, what worked for me is to find busybox in out/ and copy it on device/sdcard, then use /sdcard1/busybox tar --numeric-owner ... like described in HADK 11.1 | 14:01 |
| vknecht | (then also flash/dd hybris-boot.img to boot partition manually) | 14:01 |
| mal | lilmayofuksu: try what vknecht said first | 14:01 |
| piggz | mal: might have something in logcat...gimme 5 | 14:04 |
| piggz | mal: ok, im got messages in logcat an journal at the same time with obvious error, let me post.... | 14:07 |
| T42 | <DSstill01> i fixed audio on fm radio, esketit | 14:07 |
| lilmayofuksu | busybox from out/ worked | 14:08 |
| lilmayofuksu | now time for see if i bricked this or not | 14:08 |
| vknecht | two good news :-) | 14:08 |
| piggz | mal: https://paste.gnome.org/pcqtrlngx see 15:03:48 | 14:08 |
| vknecht | congrats @DSstill01 | 14:08 |
| mal | piggz: is that lib there? | 14:09 |
| T42 | <DSstill01> :O | 14:09 |
| piggz | file exists | 14:09 |
| piggz | sh-3.2# file /vendor/lib/hw/audio.primary.msm8998.so | 14:09 |
| piggz | /vendor/lib/hw/audio.primary.msm8998.so: ELF 32-bit LSB shared object, ARM, EABI5 version 1 (SYSV), dynamically linked, BuildID[md5/uuid]=63c74db064744d879790c2cd7252443b, stripped | 14:09 |
| piggz | mal: yes, ofc id check that first :D | 14:09 |
| mal | piggz: just making sure :) | 14:09 |
| lilmayofuksu | i think i bricked it lol nothing on lsusb and no led | 14:10 |
| T42 | <DSstill01> is it suggested to modularize iris driver for fm instead of having it built.in? | 14:11 |
| T42 | <DSstill01> like many users did | 14:11 |
| mal | @DSstill01 depends on whether it works as builtin, on many devices it won't | 14:11 |
| T42 | <DSstill01> ok, then i wont even try it | 14:12 |
| mal | piggz: which linker fix are you using? | 14:14 |
| piggz | mal: overlay | 14:14 |
| piggz | so, i effectivley have no changes to linker | 14:15 |
| piggz | it just finds stuff in /system, and loads from /usr/libexec.... | 14:15 |
| piggz | mal: so, you think its the namespace error? | 14:16 |
| mal | piggz: probably not, but maybe try the x10 way https://github.com/mer-hybris/droid-config-sony-ganges-pie/commit/5d98fc2cd583860313648eeac9b7a7264a471ef3 | 14:17 |
| mal | just in case | 14:17 |
| liux | I'm facing with boot loop problem wit obs built rpms. it's ok for local build. I've compared the .packages of both, the versions of packages are same. it's strange. | 14:33 |
| liux | and i can not debug it with telnet because it reboots after seconds of telnet diconnecting | 14:34 |
| mal | give link to the project | 14:36 |
| liux | https://build.merproject.org/project/show/home:liux:nemo:devel:hw:xiaomi:aries | 14:37 |
| mal | liux: are the zip sizes the same for both? | 14:42 |
| liux | mal, not exactly for some packages, little size like 4kb larger for obs packages | 14:44 |
| mal | can you show the .ks file you use for OBS build | 14:46 |
| liux | mal, pasted here: https://paste.ubuntu.com/p/GddByHmt5r/ | 14:48 |
| mal | liux: https://build.merproject.org/project/show/nemo:devel:hw:xiaomi:aries | 14:54 |
| mal | Mister_Magister: https://build.merproject.org/project/show/nemo:devel:hw:asus:5z | 14:57 |
| mal | Mister_Magister: piggz https://build.merproject.org/project/show/nemo:devel:hw:fxtec:t5 | 14:59 |
| mal | liux: are you sure the config and droid-hal repos don't have anything missing? | 15:05 |
| mal | you might have some local changes you forgot to commit? | 15:05 |
| mal | liux: your config repo is a mess | 15:07 |
| mal | you should not modify submodule like that | 15:08 |
| liux | mal, I have no idea what's the correct package configuration so I just follows xiaomi | 15:08 |
| liux | follows xiaomi:vince's configuration | 15:08 |
| liux | I 've modified submodule for my local build | 15:09 |
| mal | you could have just used older commit and not use latest revision of submodule | 15:09 |
| liux | mal, do you think it's ok if i push a commit to make the submodule points to original revision? | 15:10 |
| liux | or i should re-create it on github? | 15:11 |
| mal | usually the adaptation repos use master branch, not something like cm-11.0, at least in the past there were issues if submodule had different branch, not sure if that is a problem anymore | 15:11 |
| mal | liux: no need to recreate, I would just change to submodule to point to mer-hybris and use commit 5758d400b50a4d391af268376d716f2456c1690d | 15:13 |
| mal | then also change the branch of the droid-config-aries to use master branch, simply by doing git branch master, git checkout master and git push origin master or something like that | 15:14 |
| liux | mal, got it. thank you. i'll try to modify it on github | 15:14 |
| mal | that assumes you don't already have a master branch, if so then just merge or reset master branch to cm-11.0 | 15:14 |
| mal | liux: also some idea for future, tag the repo when you do new "releases" of config repo, that makes package version nicer | 15:16 |
| mal | liux: like I have done https://github.com/mlehtima/droid-config-fp2-sibon/releases | 15:16 |
| mal | liux: also if you tag then you can setup webhooks and the package build will be automatically trigger when you push a new tag to github | 15:17 |
| mal | liux: btw, nice commit history, the [] at the beginning will end up in OBS changelog which is nice | 15:19 |
| mal | like here https://build.merproject.org/package/view_file/nemo:devel:hw:fairphone:fp2-sibon/droid-config-fp2-sibon/_service:tar_git:droid-config-fp2-sibon.changes?expand=1 | 15:20 |
| mal | that is generated from git log if there is [something] at the beginning of the commit message like you already have | 15:20 |
| liux | mal, that's good idea. I never worked on obs before porting so it could be good experience for me. thank you mal | 15:21 |
| T42 | <DSstill01> how to test nfc? | 15:45 |
| mal | find some simple text tag | 15:48 |
| T42 | <DSstill01> can't i test it without those? | 15:50 |
| mal | it only supports some simple tags | 15:53 |
| T42 | <DSstill01> Hmm, i tried test_nfc but it fails with a stupid error. If thats the only way to test nfc, then rip | 16:01 |
| mal | not sure if that even works | 16:11 |
| mal | you should build and install the sfos stuff for nfc | 16:12 |
| liux | mal: I've reset submodule to mer-hybris but met with unresolvable issue "nothing provides ssu-kickstart-configuration" on obs. could you help tell me what's the correct way to resolve it? | 16:14 |
| mal | liux: so you didn't reset it to the commit I said | 16:16 |
| mal | go to the submodule folder and run git reset --hard 5758d400b50a4d391af268376d716f2456c1690d | 16:17 |
| T42 | <DSstill01> Mal: i only have installed nfcd and the plugin. What do i have to build? | 16:22 |
| liux | mal: then commit in droid-config-$DEVICE then push to github? sorry i'm not familiar with submodule and not sure what's the correct operation for submodule | 16:23 |
| mal | liux: yes, just run that command I gave in submodule folder, then in the main folder of the config repo run git commit -a | 16:32 |
| liux | mal: it seems fine on github after pushed it on. thx | 16:37 |
| mal | looks ok now | 16:39 |
| deathmist | hey! making droidmedia fails on latest master with AudioPolicyPatch applied like this https://git.io/fjiJ9 make build log https://pastebin.com/XdrNLLeq | 16:41 |
| deathmist | I'm on 15.1 oreo branch | 16:43 |
| piggz | mal: thx for project | 16:46 |
| T42 | <DSstill01> Can nfc in sf be tested with another phone instead of a tag? | 16:46 |
| mal | afaik only simple text or url tags are supported | 16:47 |
| liux | deathmist, perhaps you need to find where defined AudioPolicyService in android. in my case i cannot use AudioPolicyPatch directly either. I'm using cm11.0 and the AudioPolicyService defined in different location against AudioPolicyPatch. perhaps it's similar for your newer version of android. | 16:47 |
| mal | @DSstill01 try the nfc logger app | 16:47 |
| T42 | <DSstill01> Mal: with text you mean a physical thing? | 16:47 |
| T42 | <DSstill01> Will try that | 16:48 |
| mal | yes, UI should support url or text tags | 16:48 |
| T42 | <DSstill01> I could buy a tag, but it would take ages to arrive | 16:49 |
| T42 | <DSstill01> What is nfc logger? I can't find it on store. Is it some -devel app? | 16:51 |
| mal | hmm, can't seem to find it now | 16:57 |
| mal | this one mentions nfc https://github.com/monich/harbour-logger | 16:58 |
| T42 | <DSstill01> I found this https://github.com/monich/harbour-logger , however on google i find only the ofono one which i have already installed :p | 16:58 |
| T42 | <DSstill01> Lol | 16:58 |
| mal | try building that | 16:59 |
| T42 | <DSstill01> Ok ^^ | 17:00 |
| T42 | <DSstill01> If i build the spec it just creates a sources rpm | 17:10 |
| piggz | mal: i appear to be having great trouble with the linker fix..... | 17:36 |
| piggz | according to systemctl, it ran the command | 17:36 |
| piggz | Process: 5544 ExecStartPre=/bin/sh /usr/bin/droid/droid-hal-early-init.sh (code=exited, status=0/SUCCESS) | 17:36 |
| piggz | but it hasnt... | 17:36 |
| piggz | sh-3.2# cat /system/etc/ld.config.28.txt | grep libexec | 17:36 |
| piggz | sh-3.2# | 17:36 |
| piggz | but if i run it myself, it does | 17:37 |
| piggz | sh-3.2# /usr/bin/droid/droid-hal-early-init.sh | 17:37 |
| piggz | sh-3.2# cat /system/etc/ld.config.28.txt | grep libexec | 17:37 |
| piggz | namespace.default.permitted.paths += /usr/libexec/droid-hybris/system/${LIB} | 17:37 |
| piggz | namespace.default.permitted.paths += /usr/libexec/droid-hybris/system/${LIB} | 17:37 |
| mal | piggz: hmm | 17:39 |
| mal | piggz: just to be sure, are the permission of that .sh file correct | 17:41 |
| Mister_Magister | mal: piggz: that script works for me | 17:57 |
| vknecht | @DSstill01: try over your wallet, or some clothes | 18:04 |
| vknecht | would be surprised if you don't have tagged stuff anywhere | 18:06 |
| vknecht | *somewhere | 18:06 |
| vknecht | id badge, etc. | 18:06 |
| piggz | mal: | 18:07 |
| piggz | sh-3.2# ls -lh /usr/bin/droid/droid-hal-early-init.sh | 18:07 |
| piggz | -rwxr-xr-x 1 root root 110 Jul 6 19:48 /usr/bin/droid/droid-hal-early-init.sh | 18:07 |
| mal | piggz: strange | 18:17 |
| T42 | <DSstill01> I dont have tagged stuff, i don't even care of nfc as i said a cpuple of days ago. It's just that it was so easy to fix in lineage-14.1 that now i want to port it to sf | 18:27 |
| T42 | <DSstill01> And as i said i would like buy it but it would take so much to arrive that it's better to take those logs | 18:28 |
| T42 | <DSstill01> Oh wait, i have a thing that generates pins to confirm payments, but i dunno if it works with nfc | 18:29 |
| T42 | <DSstill01> Or i can ask my sf users if they have one of those things, if i have some lol | 18:31 |
| T42 | <meierrom> Oneplus X? Can somebody tell me more about its sfos porter, a certain onyx? | 18:35 |
| mal | ? | 18:39 |
| T42 | <DSstill01> Isnt onyx the codename? Lol | 18:42 |
| mal | yes | 18:42 |
| T42 | <DSstill01> And the porter has the same name of the device? | 18:43 |
| mal | nope | 18:43 |
| mal | rinigus and kimmoli are porting oneplus x | 18:44 |
| T42 | <DSstill01> Ah | 18:44 |
| T42 | <meierrom> mal: Thanks! :) | 18:46 |
| piggz | mal: worked aroubd by using another service... | 18:48 |
| piggz | problem persists | 18:48 |
| T42 | <meierrom> mal: @DSstill01: yeah, onyx is the device name, my bad. :) | 18:57 |
| MDTech-us_MAN | Hi all, I'm trying to port sailfish to Moto Z2 (nash). I successfully built the image but when I try to flash it, I encounter an interesting situation. TWRP says "Updater process ended with ERROR 7". After opening up the updater script manually and try to run some of the commands manually it seems that the problem is that the tar command that extracts the image gets killed. What might cause this and what might solve | 18:58 |
| MDTech-us_MAN | it? | 18:58 |
| MDTech-us_MAN | after looking in dmesg I found "Out of memory: Kill process 1180 (tar) score 1554 or sacrifice child" | 18:59 |
| MDTech-us_MAN | so this is definitely the issue | 18:59 |
| MDTech-us_MAN | but what can I do about it? | 18:59 |
| T42 | <meierrom> kimmoli: rinigus: ping | 18:59 |
| KALUbE | MDTech-us_MAN: I solved this problem by downloading a newer version of busybox and replacing the default one in twrp | 19:00 |
| mal | MDTech-us_MAN: http://www.merproject.org/logs/%23sailfishos-porters/%23sailfishos-porters.2019-07-06.log.html#t2019-07-06T14:01:35 | 19:00 |
| MDTech-us_MAN | mal: it references "what worked for me is to find busybox in out/" what "out"? where would I find a newer busybox version? | 19:02 |
| KALUbE | MDTech-us_MAN: Download busybox for your architecture from https://busybox.net/downloads/binaries/1.31.0-defconfig-multiarch-musl/ | 19:02 |
| mal | MDTech-us_MAN: out in $ANDROID_ROOT of your build environment | 19:02 |
| MDTech-us_MAN | any reason why its not automatically included in the zip, if this is a known issue? | 19:04 |
| MDTech-us_MAN | since its built anyway | 19:05 |
| kimmoli | @meierrom pong | 19:06 |
| rinigus | meierrom: please ask, don't just ping. | 19:08 |
| T42 | <meierrom> kimmoli: Got myself a oneplus X. How is the current sfos port? Major issues? | 19:09 |
| kimmoli | all known issues listed here http://images.devaamo.fi/sfa/onyx/beta-13/ | 19:09 |
| T42 | <meierrom> kimmoli: daily driver for you? | 19:11 |
| kimmoli | there has been some problems with new oneplus bootloader and old twrp needed for sfos flashing | 19:11 |
| kimmoli | yes, i have one still at 2.1.0.11, and one on the latest | 19:11 |
| kimmoli | look at the XDA thread for the mentioned issues, iirc there is solution for it | 19:13 |
| T42 | <meierrom> kimmoli: so there may be challenged flashing it but it can be done? | 19:14 |
| T42 | <meierrom> rinigus: oneplus X: still using it with sfos? | 19:15 |
| piggz | mal: different subject from PA now....any thoughts on: | 19:16 |
| piggz | 07-06 20:15:45.392 4281 4435 W ServiceManagement: Waited one second for android.frameworks.sensorservice@1.0::ISensorManager/default | 19:16 |
| kimmoli | i personally didn't have any flashing issues, my devices are rather old and been on sfos for ages | 19:17 |
| kimmoli | about stability, some issues are picking up at 150 days uptime, but i think those are more sfos than device specific | 19:19 |
| T42 | <meierrom> kimmoli: I checked the link you gave me: wifi not working? | 19:19 |
| kimmoli | on first boot after flashing with twrp it gets wrong mac address. | 19:20 |
| kimmoli | so just dont connect to any network when it asks during first start. after you get pass the sailfish introduction, reboot | 19:21 |
| T42 | <meierrom> kimmoli: so later wifi is just fine, right? | 19:21 |
| kimmoli | and it will be fine | 19:21 |
| T42 | <meierrom> kimmoli: seems like a good port to me. Anything else I should know? | 19:24 |
| kimmoli | the battery charge percentage display sometimes stuck at ~100% | 19:24 |
| rinigus | meierrom: yes, I do. daily driver, no major issues for me | 19:26 |
| T42 | <meierrom> kimmoli: rinigus: thanks guys! I'm good to go then :) | 19:26 |
| ghosalmartin | Hello :) | 19:27 |
| MDTech-us_MAN | So, I managed to successfully flash the sailfish image, but when I start it up, I'm looking at a lineage os startup screen (is this normal?) | 19:47 |
| mal | often first boot fails, check host dmesg for usb messages | 19:48 |
| MDTech-us_MAN | I mean, is it supposed to be a lineage startup screen? | 19:48 |
| mal | and try telnet to 192.168.2.15 either port 23 or 2323 | 19:49 |
| mal | MDTech-us_MAN: hmm, did you flash sailfish kernel? | 19:49 |
| MDTech-us_MAN | yeah | 19:49 |
| MDTech-us_MAN | this is a tricky device though | 19:49 |
| MDTech-us_MAN | it has 2 system slots | 19:49 |
| MDTech-us_MAN | so I wiped both, set to slot A, installed lineage, then twrp, then sailfish, and rebooted to system | 19:50 |
| MDTech-us_MAN | should I try wiping and setting to slot B? | 19:51 |
| mal | usually sfos is installed to slot a | 19:52 |
| MDTech-us_MAN | what about lineage? | 19:52 |
| mal | it doesn't really matter as long as you set the partitions correctly in fixup-mountpoints | 19:52 |
| MDTech-us_MAN | should it be in the same one? | 19:52 |
| mal | you use same slot for both | 19:52 |
| MDTech-us_MAN | in fixup mountpoints I have both a and b for all the partitions | 19:54 |
| MDTech-us_MAN | but nothign for /system | 19:54 |
| MDTech-us_MAN | should there be? | 19:54 |
| mal | show what you added there | 19:55 |
| MDTech-us_MAN | https://github.com/MDTech-us-MAN/hybris-boot/commit/01da799b596e4f43b4cd7269dc3281629d234702 | 19:55 |
| MDTech-us_MAN | as the guide said, I went into my lineage os install and just used everythign I could find there | 19:56 |
| mal | MDTech-us_MAN: those should match what fstab has | 19:58 |
| mal | https://github.com/LineageOS/android_device_motorola_nash/blob/lineage-15.1/rootdir/etc/fstab.qcom | 19:58 |
| mal | so you only add the ones for slot a but without _a in the name | 19:59 |
| MDTech-us_MAN | gotcha | 19:59 |
| mal | MDTech-us_MAN: the script parses fstab and generates mount services of those and uses fixup to replace the devices in fstab with the real devices so the first part in fixup needs to match fstab | 20:00 |
| MDTech-us_MAN | ok, not I get it | 20:01 |
| MDTech-us_MAN | *now | 20:01 |
| mal | MDTech-us_MAN: interesting, that fstab doesn't have boot partition | 20:01 |
| MDTech-us_MAN | how far back do I need to rebuild? | 20:01 |
| mal | just rerun make hybris-hal after that and the build_packages.sh -d and the rebuild image, I assume you use hybris-15.1 base? | 20:02 |
| mal | if so check the faq linked in topic for 15.1 porting specific notes | 20:02 |
| MDTech-us_MAN | yea | 20:03 |
| MDTech-us_MAN | looking | 20:03 |
| MDTech-us_MAN | oof | 20:04 |
| MDTech-us_MAN | looks like I need to rebuild hal | 20:05 |
| MDTech-us_MAN | thanks for the suggestion. I didn't see those notes before | 20:05 |
| mal | it will only rebuild it partially since you only changed very small part of it | 20:05 |
| vknecht | nearing the root cause of the issue... mic just doesn't pull droid-config-$DEVICE-1.1 rpm (missing in /var/tmp/mic/cache/packages/adaptation-community-$DEVICE-3.0.3.9) so that's why the image misses almost all stuff in sparse/ ... any idea ? | 20:06 |
| vknecht | mic doesn't depend on git tags, does it ? | 20:07 |
| piggz | mal: sensorservice? is it supposed to be provded by droidmedia? | 20:07 |
| mal | piggz: yes | 20:08 |
| piggz | hmm, why isnt it being picked up then | 20:08 |
| vknecht | Mister_Magister, can you elaborate on this please ? http://www.merproject.org/logs/%23sailfishos-porters/%23sailfishos-porters.2018-02-07.log.html#t2018-02-07T00:28:35 | 20:12 |
| ghosalmartin | mal: any idea of a sane droidmedia version? | 20:13 |
| mal | ghosalmartin: for what? | 20:13 |
| Mister_Magister | vknecht: i copied ks file from other device and changed name | 20:13 |
| Mister_Magister | xd | 20:13 |
| ghosalmartin | DROIDMEDIA_VERSION= | 20:13 |
| Mister_Magister | this isn't really bad i don't even regenerate them | 20:13 |
| mal | ghosalmartin: ? | 20:13 |
| ghosalmartin | my git repo has all metadata stripped to save space | 20:13 |
| Mister_Magister | ghosalmartin: if you have only one commit and one branch getting droidmedia version won't work | 20:14 |
| ghosalmartin | so the command "DROIDMEDIA_VERSION=$(git --git-dir external/droidmedia/.git describe --tags | sed \ | 20:14 |
| ghosalmartin | -r "s/\-/\+/g")" breaks | 20:14 |
| Mister_Magister | remove folder | 20:14 |
| Mister_Magister | and sync with whole history | 20:14 |
| Mister_Magister | only the folder | 20:14 |
| mal | ghosalmartin: https://github.com/sailfishos/droidmedia/releases | 20:14 |
| ghosalmartin | mal: ta | 20:14 |
| vknecht | so it didn't pick the right repo ? | 20:14 |
| Mister_Magister | yeah | 20:14 |
| Mister_Magister | it didn't pick the version | 20:14 |
| Mister_Magister | vknecht: or u talking about copying ks | 20:14 |
| vknecht | thing is I get the effect you spoke about eg. I update/rebuild droid-configs, but mic simply doesn't pick it, ignoring the droid-config-$DEVICE.rpm, and caches/uses only droid-config-$DEVICE-flashing/policy/preinit-plugin/pulseaudio-setting | 20:17 |
| mccreary | piggz: I cleaned up my TWRP trees, and put them at https://gitlab.com/mccreary/android_kernel_fxtec_msm8998/ and https://gitlab.com/mccreary/android_device_fxtec_t5/ | 20:17 |
| Mister_Magister | vknecht: i didn't have that | 20:17 |
| piggz | mccreary: cool, ill take a look ... did you have any joy with bootctl? | 20:17 |
| piggz | what is the kernel based on? | 20:18 |
| Mister_Magister | ah wait u linked only last message now i see | 20:18 |
| vknecht | (the -sailfish and -bluez5 ones I have to comment in pattern cause "not installable provider) | 20:18 |
| MDTech-us_MAN | uhhh... I have a slightly less related issue: when I attempt to run sfossdk, I get "su: failed to execute /bin/bash: Resource temporarily unavailable" any hint as to what might cause it? | 20:18 |
| Mister_Magister | i copied ks file from other device and in ks there is file to local repo. i had old local repo pointing to different directory | 20:18 |
| vknecht | ok, don't think that's it in my case, I use the HA_REPO sed command | 20:19 |
| mccreary | piggz, bootctl is still broken with those TWRP trees | 20:21 |
| Mister_Magister | vknecht: did you regenerate patterns? | 20:22 |
| vknecht | yes | 20:22 |
| Mister_Magister | can u explain with more detail? | 20:22 |
| mccreary | piggz, the kernel is a fork of tdm's, which is CAF msm-4.4 tag LA.UM.7.4.r1-05000-8x98.0 with the F(x)Tec changes merged on top | 20:23 |
| ghosalmartin | mal: is wifi still done through the module or is there some glib binder way now? | 20:23 |
| mccreary | FYI, the lineage-16.0 branch is the same as tdm's | 20:24 |
| mal | ghosalmartin: wifi is native, no need to go binder | 20:27 |
| deathmist | I got further in build_packages by downgrading ofono-ril-binder-plugin to 1.0.7, now dvd build fails https://pastebin.com/fKG4wJzc and I remember making this change https://git.io/fjik8 what should I do as this is a fully clean build from scratch again? | 20:28 |
| mal | deathmist: sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -R -msdk-install zypper rm ofono-configs-mer | 20:31 |
| vknecht | can't really explain differently, but now I see droid-config-$DEVICE-patterns has files it shouldn't have ; must be something wrong with my delete_pattern_$DEVICE.list or not taken in account... | 20:33 |
| * vknecht goes reading droid-configs-device/droid-configs.inc, again... | 20:34 | |
| deathmist | mal: https://pastebin.com/WKBnuxFh solution 3 is the only one that does anything, is it safe and I'll just run build_packages again after? | 20:35 |
| mal | deathmist: sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -R -msdk-install zypper in droid-config-cheeseburger | 20:36 |
| mal | the ones in previous command were not sade | 20:37 |
| mal | *safe | 20:37 |
| * vknecht renames delete_patterns_$DEVICE.list to delete_pattern_$DEVICE.list ... | 20:37 | |
| deathmist | mal: yep that worked, thanks! I also possibly found out why my call audio never worked: I ran the make droidmedia command before with argument "armv7hlarm64" without a space; seems it never got copied from hadk pdf | 20:53 |
| * vknecht also adds droid-config-* to jolla-hw-adaptation ... | 20:53 | |
| deathmist | or audioflingerglue even* | 20:54 |
| mal | deathmist: ok | 21:10 |
| mal | deathmist: I didn't quite understand the mistake | 21:11 |
| ghosalmartin | mal: any idea why https://pastebin.com/mZMTqFmL would occur on startup? | 21:12 |
| ghosalmartin | feels like bad selinux files but dunno | 21:12 |
| ghosalmartin | crashes pulseaudio as well | 21:12 |
| ghosalmartin | and then I lose telnet access | 21:12 |
| mal | ghosalmartin: which android base | 21:12 |
| ghosalmartin | hybris-15.1 | 21:13 |
| deathmist | mal: in hadk 13.2 it has a part "$PORT_ARCH $(gettargetarch)", but if you copy the line it doesn't copy the space thus "armv7hl arm64" args becomes "armv7hlarm64" | 21:13 |
| mal | ok | 21:13 |
| ghosalmartin | hmm wonder if its due to "droid-hal-init: starting service 'surfaceflinger'..." | 21:13 |
| mal | ghosalmartin: surfaceflinger shouldn't be used anymore on 15.1 | 21:14 |
| mal | hwc2 ftw | 21:14 |
| ghosalmartin | yeah got that installed trying to get to UI now | 21:14 |
| ghosalmartin | finally managed to get a fresh build of everything, was running like 9 month old HAL packages :P | 21:14 |
| mal | ghosalmartin: in vase tou might need apply this patch to $ANDROID_ROOT/external/selinux_stubs https://pastebin.com/A6Ba6YBa and then rebuild hybris-hal and of course repackage it with "build_packages.sh -d" | 21:18 |
| mal | *case | 21:19 |
| ghosalmartin | mal: does that negate the need for the selinux files being present? | 21:19 |
| mal | no, different issue | 21:19 |
| mal | ghosalmartin: also have you added the files as said in faq | 21:19 |
| mal | the .rc files | 21:19 |
| ghosalmartin | i did a while back | 21:19 |
| ghosalmartin | but it seems my droid-config-bullhead didnt rebuild on obs | 21:21 |
| rinigus | meierrom: with onyx, would be wise to get sdcard and do twrp backup on it before flashing. that way you can easily return. sometimes, camera doesn't start (black viewfinder). reboot then and try again. if it works, it will keep working until you have to shutdown. then there is a small chance again to fail. other than that, I haven't seen much. good luck! | 21:24 |
| ghosalmartin | mal: is a symptom of needing that selinux patch above lipstick memory corruption? | 21:43 |
| mal | ghosalmartin: not sure | 21:44 |
| ghosalmartin | okays ill whack it in and find out | 21:44 |
| mal | ghosalmartin: if you have android side services failing then it might be needed | 21:44 |
| mal | depends on the erro | 21:44 |
| ghosalmartin | nah they seem fine | 21:44 |
| mal | *error | 21:44 |
| mal | hmm | 21:44 |
| mal | ghosalmartin: show full logcat | 21:45 |
| ghosalmartin | https://pastebin.com/Hgww6m7m | 21:45 |
| ghosalmartin | its interesting that I get some UI | 21:47 |
| ghosalmartin | but lots of crashes and i cant swipe from the sides | 21:47 |
| ghosalmartin | ah I already have that patch | 21:49 |
| ghosalmartin | mal: should test_hwc2 work? since test_hwcomposer works | 22:05 |
| mal | ghosalmartin: do you get and tombstones in /data/tombstones | 22:05 |
| ghosalmartin | 49 of them :P | 22:06 |
| mal | any of those from related processes | 22:06 |
| ghosalmartin | lemme clear it and reboot | 22:06 |
| ghosalmartin | its annoying because I did have this working with the surfaceflinger plugin, but cant remember particular fixes. i do this everytime, i forget a fix, dont write it down and spend weeks trying to replicate it | 22:07 |
| ghosalmartin | something is breaking mem https://pastebin.com/06zzA8Sn | 22:09 |
| T42 | <DSstill01> (Photo, 720x1280) https://irc.thaodan.de/.imgstore/oDxKwGkHa6.png After getting a lot of build errors i managed to build nfc logger. However there's not much info in that. Is thar enough to say if it works or not? | 22:09 |
| ghosalmartin | memfd_create to be precise | 22:09 |
| mal | ghosalmartin: memfd is not a problem | 22:13 |
| mal | I can give a fix to that tomorrow | 22:13 |
| ghosalmartin | it seems to be affecting everything, or is it just a distraction | 22:14 |
| mal | it should be only from pulseaudio afaik | 22:14 |
| mal | @DSstill01 not sure what that tells | 22:15 |
| ghosalmartin | now its happening from a lot of sfos like ngfd and QQmlThread and the maliit-server | 22:15 |
| mal | hmm | 22:15 |
| ghosalmartin | whats odd is that when i built my packages, it missed out the selinux files | 22:15 |
| T42 | <DSstill01> mal: almost nothing lol | 22:15 |
| ghosalmartin | which were required to boot otherwise droid-hal-init threw a hissy fit | 22:16 |
| mal | ghosalmartin: yes, hybris-hal build sometimes doesn't copy all needed files to out | 22:16 |
| ghosalmartin | now they were copied to there correctly | 22:16 |
| ghosalmartin | but not included by build_package -d | 22:16 |
| ghosalmartin | woah it decided to ask me for security code | 22:17 |
| ghosalmartin | wtf its booted | 22:17 |
| Mister_Magister | i love this kind of moments ghosalmartin | 22:18 |
| Mister_Magister | when you don't expect something to happen | 22:18 |
| ghosalmartin | but my pcs grapgics card is committing suicide i think | 22:18 |
| Mister_Magister | but it happens even if it shoudln't | 22:18 |
| ghosalmartin | lets hope a reboot works | 22:19 |
| ghosalmartin | brb | 22:19 |
| ghosalmartin | well my monitor has decided to die | 22:23 |
| ghosalmartin | good thing i was planning on running the box as a headless machine anywyas | 22:23 |
| ghosalmartin | hmm opening any sfos app causes the crash and reset | 22:26 |
| ghosalmartin | am going to bed anyways | 22:26 |
| ghosalmartin | cheers for the help mal | 22:26 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
