| r0kk3rz | what are you trying to do? | 00:11 |
|---|---|---|
| T42 | <ankaos> @r0kk3rz [what are you trying to do?], I am trying to turn on the device. | 00:16 |
| r0kk3rz | okay... tell me what you have done so far | 00:18 |
| T42 | <ankaos> Open telnet. And send dmesg output | 00:20 |
| r0kk3rz | which telnet | 00:20 |
| r0kk3rz | 2323? | 00:20 |
| T42 | <ankaos> @r0kk3rz [2323?], Yes | 00:20 |
| T42 | <ankaos> Build mw | 00:21 |
| r0kk3rz | ok, print journalctl shortly after boot | 00:22 |
| *** Oksana_ is now known as Oksana | 00:56 | |
| T42 | <ankaos> @r0kk3rz [ok, print journalctl shortly after boot], https://paste.ubuntu.com/p/Q9y5FJVhgC/ | 00:56 |
| T42 | <ankaos> is it enough? | 00:57 |
| r0kk3rz | lets see | 00:57 |
| r0kk3rz | no, you need to paste the whole thing | 00:58 |
| r0kk3rz | it might be quite big | 00:58 |
| T42 | <ankaos> @r0kk3rz [it might be quite big], https://paste.ubuntu.com/p/nmXPWpYyBw/ | 01:03 |
| T42 | <ankaos> is it enough? | 01:04 |
| T42 | <birdzhang> @ankaos it that hybris 16.0? | 01:26 |
| T42 | <ankaos> @birdzhang [@ankaos it that hybris 16.0?], No. Hybris-14.1 | 01:28 |
| T42 | <birdzhang> https://github.com/mer-hybris/hadk-faq#hybris-14-1 | 01:28 |
| T42 | <birdzhang> many rc files under /usr/libexec/droid-hybris/system/etc/init/ | 01:28 |
| T42 | <Ogden %lastname%> (Photo, 1249x699) https://irc.thaodan.de/.imgstore/0K2ziX5gsf.png | 02:20 |
| Oksana | smatkovi: Would be glad to, but I don't think there is lately a build of apkenv on SailfishOS. I am trying to build it, see https://github.com/Wikiwide/apkenv and https://build.sailfishos.org/package/live_build_log/home:Wikiwide:apkenv/ApkEnv/sailfish_latest_armv7hl/armv8el | 06:47 |
| r0kk3rz | what do you hope to achieve with that? | 07:12 |
| T42 | <edp_17> Good morning! Can somebody help with this, please? mount and journalctl: https://paste.ubuntu.com/p/Hyg2rvjJyh/ | 07:15 |
| rinigus | mal: as requested, reminder regarding addition of packages in devel and testing common. it would make sense to add libncicore libnciplugin nfcd. Others needed for official-like ports: hw-ramdisk initrd-helpers yamui. | 07:17 |
| T42 | <neochapay> @edp_17 [Good morning! Can somebody help with this, ple …], show cat /proc/cmdline | 07:20 |
| T42 | <edp_17> cat /proc/cmdline: https://paste.ubuntu.com/p/PfKf5RFqrW/ | 07:21 |
| r0kk3rz | rinigus: looks like nfcd is already there | 07:23 |
| rinigus | r0kk3rz: not if I look in the monitor... where do you see it? | 07:24 |
| r0kk3rz | well, its somewhere | 07:24 |
| r0kk3rz | my port repo is building libncicore and libnciplugin without nfcd at least | 07:25 |
| r0kk3rz | and the pn54x plugin | 07:25 |
| T42 | <neochapay> @edp_17 [cat /proc/cmdline: https://paste.ubuntu.com/p/ …], You must disable selinux and audit. Add to cmdline "audit=0 selinux=0" | 07:25 |
| T42 | <edp_17> I have added BOARD_KERNEL_CMDLINE := selinux=0 into BoardCommonConfig.mk | 07:28 |
| T42 | <edp_17> Plus I have added : CONFIG_SECURITY_SELINUX_BOOTPARAM=y and | 07:29 |
| T42 | <edp_17> CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=0 into defconfig. | 07:29 |
| T42 | <elros34> selinux is disabled, droid-hal-init fails. Kernel is very noisy | 07:29 |
| T42 | <elros34> Do you use target 3.3.0 to build it? | 07:29 |
| T42 | <edp_17> Yes. I used the latest when I created the target and toolings | 07:30 |
| T42 | <elros34> did you add anything to straggler files? Do you have installed that package? Maybe you could decrease somehow log level for kernel in /proc/sys/kernel/printk | 07:32 |
| T42 | <edp_17> Nope, I have not added any straggler files and I have not installed that package either (not manually at least). | 07:34 |
| T42 | <edp_17> How can I decrease the log level in there? | 07:34 |
| T42 | <edp_17> Currently the content of that (/proc/sys/kernel/printk) is : 4 4 1 7 | 07:35 |
| T42 | <elros34> not sure, 4 is pretty low already | 07:37 |
| T42 | <elros34> Check out/*/root and compare it with your / in device | 07:39 |
| T42 | <edp_17> The 'systemctl list-units -t service --all': https://paste.ubuntu.com/p/cmMmRx2Trz/ | 07:39 |
| T42 | <elros34> Did you add hybris14 specific changes from faq? | 07:40 |
| T42 | <edp_17> They are different: https://paste.ubuntu.com/p/MZH7D26n6s/ | 07:42 |
| T42 | <elros34> are you sure that 'ssu re' in sdk and target returns 3.3.0.x and you don't have any straggler files in rpm/*spec? | 07:45 |
| T42 | <edp_17> In Platform: Device release is currently: 3.3.0.16 | 07:46 |
| T42 | <edp_17> How can I check this the target? | 07:47 |
| T42 | <elros34> use sb2 magic command | 07:47 |
| T42 | <edp_17> Thanks. You were rigth in there I got: Device release is currently: 3.2.1.20 | 07:48 |
| T42 | <edp_17> But how? I followed the hadk and downloaded the latest stuff. | 07:48 |
| T42 | <elros34> I always use versioned url, maybe lates wasn't updated when you use it | 07:49 |
| r0kk3rz | upgrading the targets is more important than the sdk | 07:49 |
| T42 | <edp_17> Well, this could explain why my N7000, S2 and hammerhead builds are not booting/starting. | 07:50 |
| T42 | <edp_17> Can I upgrade the target the normal way with zypper dup or I should delete and install the correct one from the releases.sailfishos.org? | 07:52 |
| r0kk3rz | zypper dup is fine | 07:52 |
| T42 | <edp_17> Ehh, I cannot because it needs root privilege in there. | 07:56 |
| T42 | <elros34> append -R switch to sb2 | 07:57 |
| T42 | <neochapay> @edp_17 [Plus I have added : CONFIG_SECURITY_SELINUX_BO …], In cmdline i not see selinux=0 | 07:58 |
| T42 | <elros34> @neochapay based on note in device repo, device's bootloader ignores abootimg cmdline. BOOTPARAM_VALUE=0 do same thing | 08:00 |
| T42 | <neochapay> @elros34 [@neochapay based on note in device repo, devic …], yea but where disabling audit ? | 08:01 |
| T42 | <elros34> it's not required to disable it | 08:01 |
| T42 | <neochapay> @elros34 [it's not required to disable it], I think core problem is "droid-hal-init: Failed to initialize property area" | 08:04 |
| T42 | <elros34> yes, he didn't upgrade sdk target to 3.3.0 which contains fix for for auto detecting unpackaged files. He probably miss *context files | 08:06 |
| T42 | <edp_17> The zypper dup gives me : https://paste.ubuntu.com/p/3dW5qNRWnq/ | 08:17 |
| T42 | <edp_17> Wouldn't be easier to remove target and tooling and re-install? | 08:18 |
| T42 | <Paul El %lastname%> Fwd from Paul: 🌐🌐🌐🌐🌐🌐🌐🌐🌐🌐🌐🌐I feel like the happiest person in the world since I get to know this new Crypto channel. I'm sharing my testimony because I want the whole world to know about this real channel follow his to see his good works👇 | 08:53 |
| T42 | <Paul El %lastname%> https://t.me/joinchat/AAAAAEfTGmRk7YlvP1doYw | 08:53 |
| T42 | <edp_17> I have decided to delete the existing ones and download the 3.3.0.16 target and tooling using the direct links. | 09:39 |
| T42 | <edp_17> Do I need to re-run the make hybris-hal as well or the build_packages.sh is enough? | 09:40 |
| T42 | <edp_17> When I run build_packages.sh, I got this : https://paste.ubuntu.com/p/NMRmG786hX/ | 09:52 |
| T42 | <edp_17> What else do I need to refresh? Sub-modules or something else? | 09:52 |
| T42 | <birdzhang> add to rpm/droid-hal-$DEVICE.spec straggler_files | 09:56 |
| rinigus | I am getting missing orientationadaptor when testing compass in csd: https://dpaste.org/K5ej | 09:57 |
| rinigus | device sony xz2 (tama), probably was before as well. is this expected or do I miss something in the port? compass seems to be off and shows wrong direction | 09:57 |
| T42 | <edp_17> @birdzhang : Thanks. | 09:58 |
| piggz_ | gm | 10:02 |
| T42 | <ankaos> @birdzhang [https://github.com/mer-hybris/hadk-faq#hybris- …], where is files? | 10:17 |
| T42 | <birdzhang> open the link, it's clear enough | 10:17 |
| piggz_ | jusa: ping ... a couple of questions for ya | 10:38 |
| T42 | <Verevka86> https://github.com/mer-hybris/droid-hal-device/issues/223 | 11:17 |
| T42 | <ankaos> @birdzhang [open the link, it's clear enough], sh-3.2# ls -lh /usr/libexec/droid-hybris/system/etc/init/ | 11:17 |
| T42 | <ankaos> total 8.0K | 11:17 |
| T42 | <ankaos> -rw-r--r— 1 root root 52 May 12 2020 hybris_extras.rc | 11:17 |
| T42 | <ankaos> -rw-r--r— 1 root root 893 May 12 2020 servicemanager.rc | 11:17 |
| T42 | <ankaos> script not work | 11:18 |
| T42 | <Verevka86> @RealDanct12 help | 11:18 |
| T42 | <Verevka86> @Verevka86 [https://github.com/mer-hybris/droid-hal-device …], @RealDanct12 help | 11:18 |
| T42 | <edp_17> Something is wrong: https://paste.ubuntu.com/p/DqHgZ8jbkY/ | 11:41 |
| T42 | <edp_17> The building_packages.sh fails with : droid-hal-kernel is needed by droid-hal-version-treltexx-0.0.1-202005121128.armv7hl | 11:42 |
| T42 | <edp_17> but when I try to install I get error too. | 11:42 |
| T42 | <edp_17> What am I doing wrong this time? | 11:42 |
| T42 | <elros34> here is workaround: http://collabedit.com/kr9xx | 11:42 |
| T42 | <ankaos> Elros use this link. http://paste.opensuse.org/40869869 | 11:45 |
| T42 | <ankaos> but only see; sh-3.2# ls -lh /usr/libexec/droid-hybris/system/etc/init/ | 11:45 |
| T42 | <ankaos> total 8.0K | 11:45 |
| T42 | <ankaos> -rw-r--r— 1 root root 52 May 12 2020 hybris_extras.rc | 11:45 |
| T42 | <ankaos> -rw-r--r— 1 root root 893 May 12 2020 servicemanager.rc | 11:45 |
| T42 | <edp_17> @elros34 : Thanks! Now as I read the solution I remember this was a discussion here not so while ago. | 11:55 |
| T42 | <edp_17> I have applied the fix into droid-hal-device.inc. Should I simple re-run the building_packages.sh or there is something else to do? | 11:56 |
| T42 | <elros34> rebuild package which you change so droid-hal | 11:57 |
| T42 | <edp_17> I rebuild all packages. 😊 (I know it is overkill but I let it run.) | 11:59 |
| T42 | <elros34> ... usually you only need to rebuild --mw once for release | 12:00 |
| T42 | <edp_17> Thanks for the advice. I just run the whole building_packages.sh without any parameter to re-build everything and create the image too. However I understand the benefit of re-building only the necessary package. (Do not need to re-build the ones that have been done over and over again.) | 12:04 |
| r0kk3rz | well, its quicker | 12:21 |
| Sellerie | Hi, I was referred from #sailfishos to ask here regarding the modifiability of the sfos kernel. I am theoretically interested in running wireguard on sailfish. Has anyone ver done that? | 13:19 |
| T42 | Sasank %lastname% was added by: Sasank %lastname% | 14:04 |
| T42 | <edp_17> Guys! You are AWESOME! 👏👏 After these fixes the image is booted into the initial settings on my Note 4!!!! Cool! 😄 | 14:04 |
| T42 | <ankaos> @edp_17 [Guys! You are AWESOME! 👏👏 After these fixe …], congratulations 👏👏👏 | 14:09 |
| T42 | <edp_17> Wifi works out of the box. I'll check the rest and let you know. | 14:12 |
| T42 | <ankaos> systemctl list-units -t service —all https://paste.ubuntu.com/p/MSYVGPDbjx/ | 14:35 |
| rinigus | 3.3.0.16 released on Tama devices as well. thank you all for help! | 18:41 |
| mal | rinigus: did you make sure you have vibrator fix on those? | 19:01 |
| mal | rinigus: btw, you don't need the droid-vibrator packages in tama project, you are using native-vibrator and that is available in jolla repos anyway, does that device really use hal sensors and not binderized sensors, probably both work though | 19:12 |
| mal | rinigus: also why nemo-qml-plugin-systemsettings ? | 19:13 |
| rinigus | mal: let's start with the last one - nemo-qml-plugin-systemsettings is needed to get sd card detected properly as well as internal storage. mmcblk issue as for some other devices | 19:19 |
| rinigus | let me find vibrator fix. I think I did, but it's better to check again | 19:22 |
| mal | rinigus: if you use the official hybris sony patches (and updated from git) then you have it | 19:24 |
| mal | rinigus: forgot about the systemsettings issue | 19:24 |
| rinigus | mal: let me still check it ... as for systemssettings, same here. we forget between releases | 19:25 |
| mal | that should be solved somehow in a proper way | 19:26 |
| mal | rinigus: you can see the commit here https://github.com/mer-hybris/hybris-patches/commits/hybris-sony-aosp-pie | 19:28 |
| rinigus | mal: yeah, systemssettings was a part of internal discussions that you had at Jolla. but no solution yet. | 19:29 |
| rinigus | mal: vibrator fix was pulled in via hybris-patches at https://github.com/sailfishos-sony-tama/droid-hal-sony-tama-pie . that allowed me to mess up whole tree and had to rebuild hadk from start (partially, as I have most of the port here anyway) | 19:33 |
| rinigus | ... mainly my own mistakes causing the resync / rebuild | 19:33 |
| mal | repo sync -l first and then apply the patches again | 19:35 |
| rinigus | mal: (btw, thanks a lot for review!) re droid-vibrator: I haven't enabled ffmemless vibrator, as far as I remember. any tips on how to check/use native-vibrator? | 19:37 |
| rinigus | when I was working on it, I had a feeling xperia 10 was using the same one. but I may have mixed things up | 19:38 |
| mal | rinigus: https://github.com/sailfishos-sony-tama/droid-config-sony-tama-pie/blob/master/jolla-configuration-tama.inc#L38 | 19:38 |
| mal | rinigus: there are 3 vibrator possibilities, ffmemless which is quite rare now, droid and native, most devices should use native | 19:38 |
| mal | unless ffmemless is supported | 19:38 |
| rinigus | mal: should native vibrator be enabled somewhere in config as well, in addition to the requirements | 19:40 |
| rinigus | ? | 19:40 |
| rinigus | it is in droid-hal-version-*.spec, there is vibration on pressing keys / pulling menus. but maybe not as strong as it should be, although I like it as it is... | 19:45 |
| mal | no, it's just which package is installed | 19:51 |
| mal | at least for droid and native packages | 19:51 |
| mal | ffmemless needs configutation but sonys don't support that | 19:52 |
| rinigus | mal: then, indeed, we use the native one and I should drop qt5-feedback-haptics-droid-vibrator from OBS | 19:55 |
| rinigus | sensors next... | 19:55 |
| mal | rinigus: which device is that tama? | 19:58 |
| mal | I mean is it the same year as xa2 or x10? | 19:58 |
| rinigus | mal: xperias xz2 xz2c xz3 | 19:58 |
| rinigus | x10 | 19:59 |
| rinigus | its a port based on aosp9 | 19:59 |
| mal | ok, x10 uses binder sensors | 20:00 |
| rinigus | sensors look to be using default config with plugins like `accelerometeradaptor = hybrisaccelerometeradaptor` | 20:00 |
| mal | everything is otherwise the same in hal and binder sensors except the low level communication | 20:00 |
| mal | so the configs don't change | 20:00 |
| mal | but if hal sensors work then it's fine | 20:01 |
| mal | but x10 has sensorfw-qt5-binder in OBS | 20:01 |
| mal | if you have android.hardware.sensors@1.0-service process running then binder sensors are possible | 20:02 |
| rinigus | mal:let me check | 20:02 |
| mal | rinigus: anyway not really important if all sensors work correctly | 20:03 |
| rinigus | mal: android.hardware.sensors@1.0-service is running | 20:03 |
| mal | rinigus: does compass work on that device? | 20:03 |
| mal | via for example the orienteering compass app | 20:04 |
| rinigus | mal: compass is not at its best. should work in android, not so good on sfos. shows wrong directions and doesn't really calibrate | 20:04 |
| rinigus | so, would be great to fix that | 20:04 |
| mal | hmm, does it first show something but then stop? | 20:05 |
| rinigus | mal: nope, it doesn't stop. mainly just wrong | 20:06 |
| mal | hmm, on x10 compass is not working | 20:07 |
| T42 | <adampigg> compass never works :D | 20:07 |
| mal | it gives couple of values and the just reports a constant value | 20:07 |
| mal | @adampigg on fp2 it works fine, and on many other devices | 20:08 |
| rinigus | mal: it shows values, updates them, but I will be very very lost if I would follow it. | 20:09 |
| mal | hmm | 20:10 |
| mal | I have a feeling the sony android 9 sensor support has some issues related to compass | 20:10 |
| rinigus | maybe. but what was about binderized sensors? how can I check those? | 20:12 |
| mal | rinigus: that android.hardware.sensors@1.0-service is the binder sensor interface | 20:12 |
| rinigus | mal: how do I use it? I presume I should change something in /etc/sensorfw/primaryuse.conf? and maybe install some extra package? | 20:14 |
| mal | you just replace sensorfw-qt5-hybris in OBS with sensorfw-qt5-binder (the package name only, everything else stays the same) | 20:18 |
| mal | the repo url is the same and everything | 20:18 |
| mal | you just use different spec to build it and that is determined by the package name | 20:18 |
| mal | but like I said if sensors work now there is no need to change anything | 20:19 |
| rinigus | mal: OK, then I will keep as it is. question regarding compass - is it magnetometer in sensorfw? | 20:20 |
| rinigus | or orientationadaptor? | 20:22 |
| mal | rinigus: compass is the orientationadaptor | 20:22 |
| rinigus | mal: in csd, compass test shows immediate "pass", no calibration needed. asked about it earlier, but nobody replied. corresponding journal log when running it is at https://dpaste.org/K5ej | 20:26 |
| mal | that looks odd | 20:29 |
| mal | it seems like it doesn't find the sensor at all | 20:30 |
| rinigus | mal: and at the same time, compass does show something. but the log does not feel happy at all | 20:31 |
| rinigus | mal: corresponding logcat from a fresh boot: https://dpaste.org/7jsO | 20:36 |
| mal | device_orientation != orientation | 20:37 |
| rinigus | hmm, that could be something else | 20:37 |
| T42 | <adampigg> mal: being fair, compass appears to work on the pro1 | 20:37 |
| mal | @adampigg does it use binder sensors? | 20:39 |
| rinigus | there is no just orientation though. but lots of funky sensors out there: https://dpaste.org/ifFC | 20:40 |
| T42 | <adampigg> mal: how do i know? | 20:40 |
| mal | what does test_sensors say? from libhybris-tests package | 20:40 |
| mal | @adampigg check the package name for sensors | 20:40 |
| T42 | <adampigg> hybris-libsensorfw-qt5-hal | 20:41 |
| T42 | <adampigg> nothing with binder | 20:41 |
| mal | so hal sensors | 20:41 |
| mal | isn't that android 9 also? | 20:41 |
| T42 | <adampigg> yes | 20:41 |
| T42 | <adampigg> if you are going to tell me im doing it wrong ... maybe we should make a list of expected packager per andoid base :D | 20:42 |
| mal | wrong is a wrong word in this case | 20:43 |
| T42 | <adampigg> s/wrong/less ideal ? | 20:43 |
| mal | well if hal works then it's fine | 20:45 |
| T42 | <ankaos> Fwd from ankaos: https://github.com/enesutku07/droid-hal-version-tissot https://github.com/enesutku07/droid-config-tissot https://github.com/enesutku07/droid-hal-tissot | 20:47 |
| T42 | <ankaos> Elros this good or bad. please look repos | 20:48 |
| T42 | <ankaos> https://paste.ubuntu.com/p/DJdHSqVhJ5/ @edp_17 | 21:03 |
| T42 | <ankaos> where problem? and how fix? | 21:03 |
| mal | @adampigg on x10 I get like 3 or 4 values from compass until it gets stuck at constant value, if I turn off compass and re-enable it and then turn the device it will show the correct direction but just won't update it | 21:13 |
| T42 | <adampigg> mal: which app? im using anchor | 21:14 |
| mal | I use messwerk | 21:14 |
| mal | anchor? | 21:14 |
| T42 | <adampigg> its a compass app in the store | 21:15 |
| T42 | <adampigg> lets see what messwerk says | 21:15 |
| mal | @adampigg same in anchor, it show the value and it seems correct but after coupl of values it won't update | 21:17 |
| mal | @adampigg the problem is in low level | 21:17 |
| mal | May 13 00:17:14 Xperia10-DualSIM sensorfwd[4033]: Hybris OrientationAdaptor processSample x: 60.7363 y: -1.15277 z: 18.4717 | 21:18 |
| mal | May 13 00:17:15 Xperia10-DualSIM sensorfwd[4033]: Hybris OrientationAdaptor processSample x: 61.4822 y: -1.15277 z: 18.4717 | 21:18 |
| mal | May 13 00:17:15 Xperia10-DualSIM sensorfwd[4033]: Hybris OrientationAdaptor processSample x: 61.6031 y: -1.1253 z: 18.4618 | 21:18 |
| mal | May 13 00:17:15 Xperia10-DualSIM sensorfwd[4033]: Hybris OrientationAdaptor processSample x: 61.6031 y: -1.1253 z: 18.4618 | 21:18 |
| mal | after that it will just repeat the last value at fixed interval | 21:18 |
| T42 | <adampigg> im just trying to confirm i def have a compass and not just a magnetometer ... anchor says compass is selected | 21:18 |
| T42 | <adampigg> where is it in messwerk? | 21:19 |
| mal | z axis of rotation | 21:19 |
| mal | I probably need to finish my sensor fusion code in sensorfw | 21:19 |
| T42 | <adampigg> yeah, works fine | 21:19 |
| mal | on x10 it's just crappy sony or qualcomm firmware/vendor blob | 21:20 |
| T42 | <adampigg> mal: i need to get back to BT routing ... still not working even tho PA config is now correct, i seem to have same issue as others, need to find some headphones to test | 21:20 |
| mal | that sensorfusion would manually use magnetometer, gyro and accelerometer to calculate compass heading | 21:21 |
| mal | using some matrix magic | 21:21 |
| T42 | <edp_17> Guys, what can be the reason if the device is up and running but not available neither with ssh nor telnet? | 22:17 |
| T42 | <edp_17> In settings/usb it says 'Currently not connected' however the usb cable is plugged in. | 22:17 |
| T42 | <edp_17> The dmesg | tail doesn't show the device. | 22:18 |
| r0kk3rz | so the device is booted to gui and responsive? | 22:19 |
| T42 | <edp_17> Yes. The initial set tup went through properly (except logging into the jolla account) and wifi also works. | 22:20 |
| T42 | <edp_17> The usb cable is charging the device. | 22:21 |
| r0kk3rz | ok | 22:22 |
| T42 | <edp_17> And with the same usb cable I can connect to my XA2 or Nexus 5 or i9100. | 22:22 |
| r0kk3rz | then its usbmoded doing something | 22:22 |
| T42 | <elros34> use ssh over wifi | 22:23 |
| T42 | <edp_17> @elros34 : Ohh, yes. Good point. | 22:25 |
| T42 | <edp_17> I need to type wifi password every time I reboot the phone. | 22:25 |
| r0kk3rz | mac issues | 22:26 |
| T42 | <edp_17> The usb_moded gives me: https://paste.ubuntu.com/p/bZsmXS7jZy/ | 22:28 |
| T42 | <edp_17> And I am still getting the 'KNOX: The mark is out of range' flooded into the logs. I should silence it but don't know how. | 22:33 |
| T42 | <edp_17> More precisely: Samsung kernel: KNOX: The mark is out of range | 22:33 |
| T42 | <edp_17> and Samsung kernel: [c0] KNOX: The mark is out of range | 22:33 |
| T42 | <elros34> grep it in kernel? | 22:36 |
| T42 | <edp_17> I have done it in device but now I did in kernel/ : https://paste.ubuntu.com/p/4J73hfD4JX/ | 22:39 |
| T42 | <elros34> wort to read what CONFIG_KNOX_KAP do | 22:41 |
| T42 | <elros34> maybe it's enabled | 22:42 |
| T42 | <edp_17> Yes it is enabled. I enabled because had the same issue earlier and thought the issue was that CONFIG_KNOX_KAP was unset. Seemingly setting this didn't solve the issue. | 22:43 |
| T42 | <elros34> clearly print is from xt_connmark.c, maybe some missing ifdef | 22:47 |
| r0kk3rz | samsung kernels often spew tonnes of garbage | 22:48 |
| T42 | <edp_17> I have quickly looked into xt_connmark.c What about if I just comment that out? Or set info only or something? | 22:53 |
| T42 | <elros34> I would try to comment out whole knoxvpn_uidpid function. Looks like instead using ifdef they add "START/END of KNOX_VPN" everywhere | 22:56 |
| T42 | <edp_17> What about if I comment out only this: https://github.com/edp17/android_kernel_samsung_exynos5433/blob/cm-14.1-treltexx/net/netfilter/xt_connmark.c#L149 | 22:57 |
| T42 | <edp_17> The knoxvpn_uidpid returns with -1 or 0. | 22:58 |
| T42 | <edp_17> Plus it sets the newmark which is causing the issue then. | 22:59 |
| T42 | <edp_17> I have commented out the function steps but kept the function itself. The make hybris-hal failed because it expected the function to return with something. Therefore I kept the return 0; too. | 23:07 |
| T42 | <edp_17> It is compiled. Now I try to flash it on device. Fingers crossed... | 23:08 |
| T42 | <edp_17> Yay! This has solved the issue. 😊 Now, those messages are gone from the logs and the device can be accessed via usb! 👍 | 23:19 |
| T42 | <ankaos> power key is working. but the screen is still in the mi logo. | 23:42 |
| T42 | <ankaos> read init debug in real rootfs not mounted | 23:43 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
