*** ahjolinna_ is now known as ahjolinna | 01:18 | |
Danct12 | https://github.com/mer-hybris/droid-hal-device/commit/070ae8cecf9424aaa6c66f0be0c60e5953be3971 | 03:30 |
---|---|---|
Danct12 | huh, looks like the binder is compiled only if it's android 9 | 03:30 |
enigma9o7 | I'd also like to understand that forum post about how to chroot ubuntu chromium? | 03:56 |
Umeaboy | Hi! | 10:38 |
usvi | hola | 10:41 |
Umeaboy | I ran into a little mishap following the Xperia X guide. | 10:42 |
Umeaboy | Here's my result: https://pastebin.com/m2FnhyUM | 10:42 |
Umeaboy | I don't know why I did wrong. I followed the guide step by step. | 10:42 |
Umeaboy | Everything worked perfectly until then. | 10:43 |
Umeaboy | I have to go. I'll leave this session active until I get back. | 10:43 |
Umeaboy | Here's the guide I'm following: https://sailfishos.org/wiki/Sailfish_X_Xperia_Android_9_Build_and_Flash#Sailfish_OS_Hardware_Adaptation_Development_Kit_for_Sony_Xperia_10 | 10:43 |
mal | Umeaboy: it seems it doesn't like some character in your git email or something | 10:45 |
mal | r0kk3rz: yes, we can probably remove the requirement for registered nick for now but I don't have ops on this channel so I cannot do it | 10:58 |
Umeaboy | mal: If you take a look at the line with my emailadress, have you seen any illegal characters in it? | 11:47 |
Umeaboy | Would it be sufficable to do git config --global user.email "anotheremailadresshere@operator.com" in the HABUILD SDK or do I have to start over? | 11:49 |
mal | well I only told what the error suggests, try setting the email again, it's easy to add hidden characters by accident | 11:50 |
mal | or maybe it combines name and email somewhere and that causes issues, not sure really | 11:51 |
Umeaboy | It still gives me the same problem. | 11:51 |
Umeaboy | I'll change the emailadress | 11:52 |
Umeaboy | Nope. That didn't work as well. | 11:52 |
Umeaboy | Seems like the python script for the repo command is not functioning correctly. | 11:53 |
Umeaboy | Anything else I can try? | 11:54 |
Umeaboy | Is it related to my locale settings in the HABUILD SDK? | 11:57 |
Umeaboy | I found this page mentioning it together with my error: https://stackoverflow.com/questions/9942594/unicodeencodeerror-ascii-codec-cant-encode-character-u-xa0-in-position-20 | 11:58 |
Umeaboy | Fixed it. | 12:02 |
Umeaboy | Repo init doesn't allow non english characters in the user.name setting. | 12:03 |
Umeaboy | Can this be added to that guide for furture reference? | 12:03 |
Umeaboy | future | 12:03 |
Umeaboy | I can't seem to login to edit the guide. | 12:06 |
Umeaboy | Can any kind person do that for me, please? | 12:06 |
Umeaboy | It would be much appreciated. | 12:06 |
Umeaboy | mal: Can you change it? | 12:11 |
Umeaboy | Gotta go. | 12:58 |
electro | hi all, how are you ? | 16:38 |
electro | elros34, how are you ? | 16:43 |
electro | hi mal, i can continue to debug my devuce startup | 17:03 |
electro | device | 17:03 |
electro | disable selinux, disable droid-hal-init, disable Mem, try echo V watchdog | 17:06 |
electro | What can i do to continue debugging | 17:07 |
mal | did any of those change anything? even in /proc/last_kmsg? | 17:08 |
electro | no | 17:09 |
electro | i can send you my last_log | 17:09 |
electro575 | mal, https://dpaste.de/nFBp#L1,159 | 17:11 |
electro575 | https://dpaste.de/6VTg | 17:11 |
electro575 | https://dpaste.de/0087#L75,686,718,923,932 | 17:11 |
electro575 | mal, the .zip is not signed, it's maybe that | 17:18 |
mal | electro575: this doesn't look good "Emergency Remount R/O", when did you last time reflash rootfs? maybe it got corrupted during all the failures? reflash it, then add the systemd service disabling before you boot to sfos for the first time, so at least droid-hal-init and user@100000 should be disabled | 17:18 |
mal | electro575: doesn't matter, twrp can install unsigned zips without issues | 17:18 |
electro575 | ok | 17:18 |
mal | electro575: not sure if there were device that wouldn't boot unsigned kernels | 17:18 |
electro575 | can you give me command to disable droid-hal-init and user@100000 | 17:19 |
mal | electro575: do you know how to disable those systemd service in recovery, use adb shell and there /data/.stowaways/sailfishos is the sfos root location | 17:19 |
electro575 | ok | 17:20 |
electro575 | yes but i don't have the command | 17:20 |
mal | in adb shell, cd /data/.stowaways/sailfishos, then "ln -s /dev/null etc/system/system/droid-hal-init.service" and "ln -s /dev/null etc/system/system/user@100000.service" | 17:20 |
mal | then install adb to your computer first | 17:21 |
electro575 | i have adb | 17:21 |
electro575 | thanks ^^ | 17:21 |
mal | oops those commands I gave are wrong | 17:21 |
mal | correct ones: "cd /data/.stowaways/sailfishos", "ln -s /dev/null etc/systemd/system/droid-hal-init.service" and "ln -s /dev/null etc/systemd/system/user@100000.service" | 17:21 |
mal | make sure each command works | 17:22 |
electro575 | ok thanks | 17:22 |
electro575 | ok good, i reboot | 17:25 |
electro575 | what is the test cmd for : Emergency Remount R/O | 17:26 |
mal | ? | 17:27 |
electro575 | last_kmsg | 17:28 |
electro575 | what can i send to you as log ? | 17:29 |
mal | if the device rebooted again then last_kmsg | 17:30 |
electro575 | ok | 17:30 |
mal | you of course need to make the boot stop in that case | 17:30 |
electro575 | yes | 17:30 |
electro575 | again this | 17:31 |
electro575 | dispite of disable droid-hal-init.service | 17:31 |
electro575 | Emergency Remount R/O | 17:31 |
electro575 | and user@100000.service | 17:31 |
mal | show the log | 17:32 |
mal | how does /proc/cmdline look like now? | 17:33 |
electro575 | ok | 17:33 |
electro575 | my device, reboot and i can do telnet | 17:33 |
electro575 | i have just the battery picture because my device is plug into my computer | 17:34 |
electro575 | https://dpaste.de/gfME | 17:34 |
electro575 | this is strange too : [TIMA trusted boot]: SEAndroid MAGIC failure (boot.img) | 17:35 |
mal | battery picture? | 17:35 |
electro575 | https://dpaste.de/OiLn | 17:36 |
electro575 | yes | 17:36 |
electro575 | in the device screen | 17:36 |
electro575 | charging battery picture | 17:37 |
electro575 | if i do this command : echo "continue" >/init-ctl/stdin | 17:37 |
electro575 | my device reboot and i can do telnet, not bootloop because of /data/.stowaways/sailfishos/init_enter_debug2 presence | 17:38 |
electro575 | this comment "Emergency Remount R/O" disappear | 17:39 |
vknecht | probably you should unplug usb when the device reboots, so it doesn't go straight to charging/actdead mode | 17:41 |
electro575 | ok, i try | 17:41 |
mal | electro575: what kind of battery image does the display show? | 17:44 |
electro575 | check battery level | 17:44 |
electro575 | i have reboot without the usb cable pluging | 17:45 |
mal | ? | 17:45 |
electro575 | last_dkms : https://dpaste.de/Mumc | 17:45 |
mal | it's really relevant to know how exactly the display looks | 17:46 |
electro575 | busy to determine the battery level | 17:46 |
electro575 | now, i have Samsung GALAXY Note II GT-N7105 display text | 17:46 |
electro575 | audit: initializing netlink socket (disabled) | 17:48 |
mal | electro575: are you sure you have the changes elros34 mentioned in use? | 17:49 |
mal | based on last_kmsg it fails in preinit of sailfish | 17:49 |
electro575 | elros changes, yes | 17:49 |
electro575 | but if you know say it again to me | 17:49 |
mal | 100% sure? did you add the repo to .ks file when building the image? | 17:50 |
electro575 | i link it to you | 17:50 |
mal | check packages file to be sure correct versions have been installed to the image | 17:50 |
electro575 | how ? in the manifest.xml ? | 17:51 |
mal | in the folder where the sailfish zip is | 17:51 |
electro575 | ok | 17:51 |
mal | @elros34 it seems the boot of device electro575 ports reboots almost at the beginning of preinit of sfos | 17:51 |
electro575 | list of packages : https://dpaste.de/LTxx | 17:53 |
mal | well I have no idea what repo you added so I cannot check if those are correct | 17:54 |
electro575 | i give it to you | 17:54 |
electro575 | https://dpaste.de/QXVa | 17:56 |
mal | electro575: sfos repo, not android repo | 17:57 |
electro575 | yes | 17:57 |
mal | something elros34 gave you | 17:57 |
electro575 | repo --name=glibc-kernel-3.0 --baseurl=http://repo.merproject.org/obs/home:/elros34:/glibc/sailfishos_3.0.3.9/ | 17:58 |
electro575 | humm ok, it's 3.0.3.9 and me 3.0.3.9 | 17:58 |
electro575 | 3.0.3.10 | 17:58 |
electro575 | https://dpaste.de/Vz9X | 17:59 |
mal | elros34 probably knows more of the issues you might have | 18:00 |
electro575 | elros34 is not connect | 18:00 |
electro575 | mal, where the file must be locate ? | 18:01 |
electro575 | /home/sailfish-dev/hadk/Jolla-@RELEASE@-t0lte-@ARCH@.ks | 18:01 |
electro575 | or here ? /home/sailfish-dev/hadk/hybris/droid-configs/installroot/usr/share/kickstarts/Jolla-@RELEASE@-t0lte-@ARCH@.ks | 18:01 |
mal | first one | 18:02 |
electro575 | the repo of elros34 is not present in this file : /hadk/hybris/droid-configs/installroot/usr/share/kickstarts/Jolla-@RELEASE@-t0lte-@ARCH@.ks | 18:02 |
mal | the versions in your packages file seem to match the repo | 18:04 |
electro575 | ok, so strengh ! | 18:05 |
electro575 | so it's the preinit how fail | 18:06 |
electro575 | who fail | 18:06 |
electro575 | i don't know what to do anymore | 18:06 |
electro575 | mal, https://dpaste.de/srD5 | 18:11 |
mal | didn't @elros34 have some device that needed some changes, maybe check what kernel option he changed | 18:13 |
electro575 | where can i see this changed ? | 18:14 |
mal | show the .config file you have somewhere in out/ | 18:16 |
electro575 | location of this ? in the device ? | 18:17 |
mal | in $ANDROID_ROOT/out it's under there somewhere, use "find $ANDROID_ROOT/out -name .config" | 18:17 |
electro575 | this file ? /home/sailfish-dev/hadk/out/target/product/t0lte/obj/KERNEL_OBJ/.config | 18:18 |
electro575 | or this ? obj/busybox/full/.config | 18:18 |
mal | first one obviously | 18:19 |
electro575 | https://dpaste.de/o4o3 | 18:21 |
mal | electro575: elros34 gave some config earlier and it had quite strange set of changes, not sure if those should be used | 18:24 |
electro575 | humm, i don't know | 18:25 |
electro575 | where elros34 is ? ^^ | 18:25 |
electro575 | 107 guys on this channel, nobody have do changes for other devices ? | 18:26 |
electro575 | changes is for hybris-boot or sailfishOS ? | 18:27 |
electro575 | if it's hybris-boot, can we use topic of cm11 or last ? | 18:28 |
mal | electro575: most device have newer kernel than you have | 18:29 |
electro575 | lineageOS14.1 or 16 work correctly on t0lte, so ^^ i suppose it's possible but | 18:29 |
electro575 | okey | 18:29 |
electro575 | what kind of changes is specificaly ? it's about hardware ? | 18:30 |
electro575 | where it's possible to see what kind of changes can i do ? | 18:30 |
mal | electro575: this is the link I found in log https://pastebin.com/uH6Ew01B you might have to do some other changes if you would try those, to both init-script and kernel checker | 18:31 |
electro575 | init-script and kernel checker ? | 18:32 |
electro575 | this changes are to do with defconfig no ? | 18:32 |
electro575 | if i change defconfig | 18:32 |
electro575 | i just build hybris-boot ? | 18:32 |
mal | yes, but to be able to use those you need to make sure nothing else breaks, init-script verifies kernel configs and won't continue if those are not "correct" | 18:33 |
mal | https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L326 | 18:33 |
electro575 | ok | 18:34 |
electro575 | i go to try this | 18:34 |
mal | you can see there that some of the configs set to =n in the link I gave are required to be =y for boot to continue so you need to edit init-script in hybris/hybris-boot to make sure those do not prevent booting | 18:34 |
mal | also kernel checker needs to be fix so it won't complain about those if it tries to prevent build | 18:34 |
electro575 | i don't understand all | 18:35 |
electro575 | simple sentence if you can ^^ | 18:35 |
electro575 | ok | 18:36 |
electro575 | i understand | 18:36 |
electro575 | but why : you need to edit init-script in hybris/hybris-boot | 18:37 |
electro575 | mal, do all changes of the pastebin ? | 18:38 |
electro575 | on the defconfig | 18:38 |
electro575 | but why edit init-script ? | 18:38 |
electro575 | to avoid booting | 18:39 |
mal | make a backup of your defconfig, revert to original version, then add only those changes to the end and then make sure init-script accepts those settings meaning check the lines I linked againsta the kernel changes, then build | 18:40 |
electro575 | ok thanks | 18:41 |
Danct12 | good night | 18:42 |
Danct12 | anyway, back to porting | 18:42 |
electro575 | where can i find original version of lineageos_t0lte_defconfig | 18:42 |
electro575 | good night Danct12 | 18:43 |
Danct12 | calls doesn't work, the phone does make calls, the other phone i have ringed, but i can't hear anything on my phone | 18:43 |
mal | Danct12: using the old image or new? | 18:44 |
Danct12 | i just updated the target and recreated the rootfs hours ago | 18:44 |
mal | electro575: from the original android repo or you can just reset sources you have already | 18:44 |
Danct12 | so it's new | 18:45 |
electro575 | reset sources ? how | 18:46 |
electro575 | repo reset ? | 18:46 |
electro575 | mal, do you know how reset source ? | 18:52 |
Umeaboy | I read the warning box about renaming the tooling and that worked, but as soon as I did sdk-assistant create $VENDOR-$DEVICE-$PORT_ARCH http://releases.sailfishos.org/sdk/targets/Sailfish_OS-latest-Sailfish_SDK_Target-armv7hl.tar.7z it said: FAIL: unknown option -- | 18:56 |
Umeaboy | What am I missing? | 18:56 |
electro575 | sdk-assistant create SailfishOS-3.0.0 http://releases.sailfishos.org/sdk/targets/Sailfish_OS-latest-Sailfish_SDK_Tooling-i486.tar.7z | 18:57 |
electro575 | humm ok | 18:57 |
electro575 | sdk-assistant create $VENDOR-$DEVICE-$PORT_ARCH http://releases.sailfishos.org/sdk/targets/Sailfish_OS-latest-Sailfish_SDK_Target-armv7hl.tar.7z | 18:58 |
electro575 | try this | 18:58 |
mal | electro575: can't you really search anything online yourself? git reset --hard HEAD in kernel sources, just make sure you take backup of the changes you have done | 18:58 |
mal | Umeaboy: you haven't set the .hadk.env correctly or otherwise it's not being used | 18:59 |
mal | Umeaboy: check the changes related to that | 18:59 |
Umeaboy | source .hadk.env ? | 18:59 |
Umeaboy | Would that work or mnake it worse? | 18:59 |
mal | just check all changes related to that | 19:00 |
mal | there were those other files created or edited in the same section of hadk pdf | 19:00 |
Umeaboy | mal: Here's a screenshot of my .hadk.env: https://ibb.co/2KfYVb0 | 19:03 |
Umeaboy | I'm no expert, but it looks OK to me. | 19:03 |
Umeaboy | I'll be off for an hour or two. | 19:05 |
mal | I said look at the other files also | 19:06 |
Danct12 | anyway, here's logs for calls doesn't work/silent calls | 19:07 |
piggz | mal: given that ive so far not managed to get t5 paternised ... whats the change of latest updating to 3.1.0.11? | 19:07 |
Danct12 | journalctl https://paste.ubuntu.com/p/BsGrztPpsr/ | 19:07 |
Danct12 | logcat https://paste.ubuntu.com/p/PDxTywsNd4/ | 19:07 |
mal | Danct12: did you check the old image for any pulseaudio related changes? | 19:08 |
Danct12 | i haven't, | 19:10 |
Danct12 | i already knew about this issue for a day now or two, but wanted to get other stuffs working before going to get calls working | 19:11 |
mal | Danct12: the "unknown ril command 130" in logs might not be good, monich can tell more about that probably | 19:12 |
electro575 | when you said : then build -> just make hybris-boot ? or make all | 19:14 |
mal | make hybris-boot should be enough | 19:15 |
mal | hopefully | 19:15 |
electro575 | yes, but when you say see init-script | 19:16 |
mal | electro575: did you make the changes to that? | 19:17 |
electro575 | init-script accepts those settings meaning check the lines | 19:17 |
electro575 | what do you want to say ? | 19:17 |
electro575 | if the line in init-script is correct -> what's up | 19:19 |
electro575 | # writes to /diagnosis.log if there's a problem | 19:19 |
electro575 | mal, all settings must be Y for this log | 19:20 |
electro575 | /diagnosis.log | 19:20 |
electro575 | it's that ? | 19:20 |
mal | what did you change in the init-script | 19:21 |
mal | go to hybris/hybris-boot in sources and run git diff | 19:21 |
electro575 | no changes on init-script | 19:22 |
mal | electro575: I told you to fix it | 19:23 |
electro575 | https://dpaste.de/x6wP | 19:23 |
mal | FIX IT | 19:23 |
electro575 | yes | 19:24 |
mal | I told you what to do, make sure none of the things you manually set to =n in defconfig are in the =y list of init-script and same for the other list | 19:24 |
electro575 | for the other list ? in init-script | 19:25 |
electro575 | to pass all init | 19:26 |
mal | how can a simple thing be so difficult to understand | 19:26 |
electro575 | ok | 19:27 |
electro575 | i understand | 19:27 |
electro575 | with this, init will be okey so | 19:27 |
electro575 | if i fix it | 19:27 |
mal | show what changes you did | 19:29 |
mal | using git diff like I told | 19:29 |
electro575 | what changes you did -> in my defconfig | 19:33 |
electro575 | using git diff in hybris/hybris-boot -> sure ? | 19:33 |
mal | I will soon give up, this is going nowhere | 19:34 |
electro575 | i have just fix this CONFIG_... for this line in init-script | 19:37 |
electro575 | for x in CONFIG_CGROUPS CONFIG_AUTOFS4_FS CONFIG_DEVTMPFS_MOUNT CONFIG_DEVTMPFS CONFIG_UNIX CONFIG_INOTIFY_USER CONFIG_SYSVIPC CONFIG_NET CONFIG_PROC_FS CONFIG_SIGNALFD CONFIG_SYSFS CONFIG_TMPFS_POSIX_ACL CONFIG_VT | 19:38 |
electro575 | ok, this is good for changes | 19:42 |
electro575 | i git diff | 19:42 |
electro575 | mal, https://dpaste.de/HjGZ | 19:43 |
electro575 | this for defconfig change : https://dpaste.de/htn6 | 19:44 |
electro575 | mal, with what log have you seen the init-script is not correctly launched ? | 19:59 |
mal | electro575: so you didn't fix init-script? | 20:30 |
electro575 | nearly | 20:31 |
electro575 | i compile all | 20:32 |
mal | electro575: based on the changes you gave you did nothing to init-script | 20:32 |
electro575 | because of just flash hybris-boot is not enought | 20:32 |
mal | you don't seem to understand anything I say | 20:32 |
electro575 | maybe, but if the log appear | 20:33 |
mal | this is enough for today as this is really not going anywhere because you don't do the changes I ask | 20:33 |
electro575 | init-script | 20:33 |
electro575 | ok | 20:33 |
electro575 | thanks | 20:34 |
mal | electro575: you don't have to care about logs, the changes to init-script are purely based on defconfig changes, nothing else | 20:34 |
mal | I told you what to do | 20:34 |
electro575 | yes | 20:34 |
mal | and you didn't do it | 20:34 |
T42 | <Harsh18262> @Mal was irc chats active from morning? | 20:35 |
mal | what do you mean? | 20:35 |
T42 | <Harsh18262> There was no msgs on telegram from 9:26 | 20:36 |
T42 | <Harsh18262> Might be possible it was just me | 20:36 |
mal | there was no talk on this channel all morning | 20:36 |
electro575 | mal, sorry but my english is soso | 20:37 |
T42 | <Harsh18262> Oh ok then | 20:37 |
electro575 | i fix it in defconfig | 20:37 |
mal | electro575: NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!1 | 20:38 |
mal | electro575: you fix init-script so it accept the values you have in defconfig | 20:38 |
electro575 | ok, now i understand | 20:38 |
mal | I said many times to fix init-script | 20:39 |
electro575 | possible yes, but so long sentences | 20:39 |
mal | electro575: also the defconfig changes you just linked do not match the link I gave | 20:39 |
electro575 | so i take my old_defconfig | 20:40 |
mal | NO!!! | 20:40 |
T42 | <elros34> Harshwardhan: telegram bot was broken due to server mainntenance | 20:40 |
electro575 | and do changes with the link | 20:40 |
mal | you take original defconfig from lineage, then add the lines in the link I gave (https://pastebin.com/uH6Ew01B) to the end of the original file | 20:41 |
electro575 | ok | 20:41 |
mal | I'm going offline now | 20:42 |
electro575 | and fix init-script | 20:42 |
electro575 | with ? the defconfig | 20:42 |
electro575 | sorry mal | 20:42 |
electro575 | make sure none of the things you manually set to =n in defconfig are in the =y list of init-script and same for the other list | 20:44 |
mal | do you understand that or not? | 20:45 |
electro575 | yes | 20:46 |
electro575 | but if the kernel checker give ERROR | 20:47 |
mal | then you fix checker like I told | 20:47 |
electro575 | ok, i will try yes | 20:47 |
mal | you do anything you need to use the changes I gave in the link | 20:48 |
electro575 | ok | 20:48 |
electro575 | have a nice night | 20:48 |
electro575 | and after check what on device ? | 20:48 |
T42 | <elros34> I will simplify it for you, you already used that defconfig changes and it has all needed changes by init-debug. I created it when we had no telnet connection, bootloop and misleading last_kmsg. | 20:49 |
mal | @elros34 try to make him do what is needed | 20:51 |
electro575 | yes | 20:53 |
T42 | <elros34> I am not sure what is needed right now, this changes was only too get telnet. Looks like selinux=0 is ignored by kernel and selinux is initialized anyway so SECURITY_SELINUX=n or AUDIT=n might be good idea. Also based on last_kmsg it stops printing right after "Welcome to Sailfish*". Next line in /sbin/preinit writes to watchdog so I | 20:57 |
T42 | advised him to disable watchdog at all for now. | 20:57 |
electro575 | i must go sleep, sorry | 20:59 |
electro575 | bye | 20:59 |
T42 | <elros34> It would be useful to know when excatly it reboots so commenting out "exec /lib/systemd/system*" in /sbin/preinit could confirm where it fails | 20:59 |
mal | @elros34 last thing in last_kmsg was the first print from preinit | 21:03 |
mal | ah you noticed that already | 21:04 |
T42 | <elros34> yes and it should print info about BOOTSTATE before starting systemd | 21:05 |
vknecht | preinit plugins ftw | 21:44 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!