T42 | <smatkovi> I have created it afterwards, but then it complained about a missing file, which i then put into the manifest, but it was not found on the server, when i remember chiefly (re @elros34: so do you have that ...) | 06:09 |
---|---|---|
T42 | <smatkovi> You could be right^^ (re @Mister_Magister: man i doubt anything...) | 06:10 |
T42 | <smatkovi> No i just used these two lines, i can try that (re @SailfishFreenodeIRCBridgeBot: <mal>so my idea of u...) | 06:11 |
mal | @smatkovi quite sure I said there were more than two lines needed | 09:02 |
T42 | <smatkovi> Okay than it slipped through. I looked through the file and thought there were only lines from other devices. But i will comment these lines as you said now | 11:55 |
mal | it contained also some repos common to several devices and also overrides of some other repos | 12:55 |
T42 | <smatkovi> you were right, it worked, i am compiling :) | 12:56 |
T42 | <smatkovi> unfortunately at building soong it complains about undefined modules ctstestrunner-axt and cts-rollback-lib should i delete the previously downloaded sources and make a fresh repo sync? | 13:16 |
T42 | <elros34> did you apply hybris-patches? | 13:18 |
T42 | <smatkovi> no i forgot | 13:18 |
T42 | <elros34> clean out/ apply patches and try again | 13:18 |
T42 | <smatkovi> how can i clean out, with make clean? | 13:22 |
T42 | <elros34> probably or just remove directory | 13:22 |
T42 | <smatkovi> unfortunately it still says flickerlib depends on cts-wm-util | 13:29 |
T42 | <elros34> show whole output | 13:31 |
T42 | <smatkovi> https://pastebin.com/YjLyLxmv | 13:32 |
T42 | <elros34> so grep FlickerTests or flickerlib in device repositories. Is output really that short even after cleaning out/? | 13:34 |
T42 | <smatkovi> i cleaned and rerun again and now it complains about another lib in soong: https://pastebin.com/082dCPrX | 13:35 |
T42 | <smatkovi> okay i think i have to do the other things in the list, exporting disable path restrictions, not disable selinux and add android-version to device.spec, although i didn't find the spec file | 13:40 |
T42 | <elros34> that is different story. Did you grep for these keywords in device repos? Are you sure you have hybris-patches applied? | 13:42 |
T42 | <smatkovi> yes i applied them . for which keywords should i grep? | 13:44 |
T42 | <elros34> these I mentioned 10minutes ago or other you showed in your latest log or simple "test" word | 13:46 |
T42 | <smatkovi> is this the right command: grep -rnw -e "flicker" ? | 13:54 |
T42 | <elros34> I would add -i for sure or use gnome-search-tool like app | 13:55 |
T42 | <smatkovi> so this command gives nothing on device path, neither lower nor uppercase | 13:56 |
T42 | <smatkovi> in development and kernel i have matches | 13:57 |
T42 | <elros34> but it's not kernel, its probably some device repo which includes test module whatever name it has | 13:58 |
T42 | <elros34> cts/hostsidetests, /tests/FlickerTests | 13:59 |
T42 | <elros34> you need to find where it is and remove it from building | 13:59 |
T42 | <smatkovi> yes that did the trick :) | 15:53 |
T42 | <smatkovi> thanks | 15:53 |
T42 | <smatkovi> i just deleted those two folders and then removed the modules from building which were dependent on them. now it said some test dependencies failed building, also something with bluetooth and there was the option to bypass that, so i did that, but now at the next compilation step some direct dependencies are missing which seem to be SND_SOC_QDSP6V2, SND_SOC_APQ8084 and DEVFREQ_GOV_MSM_BW_HWMON | 16:49 |
T42 | <smatkovi> Oh, no, it seems to be that those three have direct missing dependencies | 18:29 |
mal | hmm, why are you getting those errors, aren't those kernel things? | 19:09 |
T42 | <neochapay> hi | 19:15 |
T42 | <neochapay> have some problem with porting to mediatec device | 19:15 |
T42 | <adampigg> thatcs cos MTK is funky :D | 19:16 |
T42 | <neochapay> ``` | 19:16 |
T42 | <neochapay> CANNOT LINK EXECUTABLE "/sbin/droid-hal-init": invalid ELF file "/system/lib64/bootstrap/libdl.so" load segment[4]: p_offset (0x3000) + p_filesz (0x0) ( = 0x3000) past end of file (0x28e0)``` | 19:16 |
T42 | <neochapay> Yea...but i think it's not mtk problem ^_^ | 19:16 |
T42 | <neochapay> full log is ``` | 19:18 |
T42 | <neochapay> Jun 23 22:27:48 MY6771 sh[1059]: CANNOT LINK EXECUTABLE "/sbin/droid-hal-init": invalid ELF file "/system/lib64/bootstrap/libdl.so" load segment[4]: p_offset (0x3000) + p_filesz (0x0) ( = 0x3000) past end of file (0x28e0) | 19:18 |
T42 | <neochapay> Jun 23 22:27:48 MY6771 sh[1059]: linker: CANNOT LINK EXECUTABLE "/sbin/droid-hal-init": invalid ELF file "/system/lib64/bootstrap/libdl.so" load segment[4]: p_offset (0x3000) + p_filesz (0x0) ( = 0x3000) past end of file (0x28e0) | 19:18 |
T42 | <neochapay> Jun 23 22:27:48 MY6771 audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=dbus-org.nemo.ssu comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success' | 19:18 |
T42 | <neochapay> Jun 23 22:27:48 MY6771 systemd[1]: droid-hal-init.service: Main process exited, code=exited, status=1/FAILURE | 19:18 |
T42 | <neochapay> Jun 23 22:27:48 MY6771 systemd[1]: droid-hal-init.service: Failed with result 'exit-code'. | 19:18 |
T42 | <neochapay> Jun 23 22:27:48 MY6771 systemd[1]: Failed to start droid-hal-init.``` | 19:18 |
T42 | <neochapay> Maybe i forgot disable some things into system/core/init ? | 19:18 |
T42 | <smatkovi> i don't know i deleted the hadk directory and try it again (re @SailfishFreenodeIRCBridgeBot: <mal>hmm, why are yo...) | 19:20 |
T42 | <neochapay> Looks so strange: ``` | 19:20 |
T42 | <neochapay> / # /system/bin/logcat | 19:20 |
T42 | <neochapay> /bin/sh: /system/bin/logcat: not found``` | 19:20 |
T42 | <neochapay> but ``` | 19:20 |
T42 | <neochapay> / # ls /system/bin/logcat -la | 19:20 |
T42 | <neochapay> -rwxr-xr-x 1 root root 166648 Jun 25 2023 /system/bin/logcat``` | 19:21 |
T42 | <TheVancedGamer> thats linker being the problem isn't it? | 19:21 |
T42 | <neochapay> Looks like that. Do you know where I can read about it? Or how fix it? (re @TheVancedGamer: thats linker being t...) | 19:22 |
T42 | <TheVancedGamer> `ls /system/bin/bootstrap` maybe | 19:22 |
T42 | <neochapay> ``` | 19:23 |
T42 | <neochapay> drwxr-xr-x 2 root root 4096 Jun 23 22:13 . | 19:23 |
T42 | <neochapay> drwxr-xr-x 4 root root 12288 Jun 23 22:13 .. | 19:23 |
T42 | <neochapay> -rwxr-xr-x 1 root root 1057144 Jun 25 2023 linker | 19:23 |
T42 | <neochapay> -rwxr-xr-x 1 root root 1445872 Jun 25 2023 linker64 | 19:23 |
T42 | <neochapay> lrwxrwxrwx 1 root root 6 Jun 23 22:13 linker_asan -> linker | 19:23 |
T42 | <neochapay> lrwxrwxrwx 1 root root 8 Jun 23 22:13 linker_asan64 -> linker64``` | 19:23 |
T42 | <neochapay> (re @TheVancedGamer: ls /system/bin/boots...) | 19:23 |
T42 | <TheVancedGamer> hm, idk now :P | 19:24 |
T42 | <TheVancedGamer> maybe wait for mal | 19:24 |
T42 | <elros34> @neochapay logcat will not work untill you fix droid-hal-init | 19:39 |
T42 | <neochapay> reasonable (re @elros34: @neochapay logcat wi...) | 19:39 |
mal | is apex failing? | 19:40 |
mal | *apexd | 19:40 |
T42 | <neochapay> ``` | 19:40 |
T42 | <neochapay> / # /system/bin/apexd | 19:40 |
T42 | <neochapay> CANNOT LINK EXECUTABLE "/system/bin/apexd": invalid ELF file "/system/lib64/bootstrap/libdl.so" load segment[4]: p_offset (0x3000) + p_filesz (0x0) ( = 0x3000) past end of file (0x28e0) | 19:40 |
T42 | <neochapay> linker: CANNOT LINK EXECUTABLE "/system/bin/apexd": invalid ELF file "/system/lib64/bootstrap/libdl.so" load segment[4]: p_offset (0x3000) + p_filesz (0x0) ( = 0x3000) past end of file (0x28e0)``` | 19:40 |
T42 | <neochapay> same | 19:40 |
mal | @smatkovi there is never need to remove the whole sources of hadk directory, that is just wasted time to do that | 19:41 |
T42 | <elros34> pure guess but do you handle somehow /linkerconfig/bootstrap -> /linkerconfig mounting? | 19:41 |
mal | I have never understood why people get the idea to remove everything if they have some problem | 19:42 |
T42 | <neochapay> /linkerconfig is empty (re @elros34: pure guess but do yo...) | 19:42 |
T42 | <elros34> yeah handling code is uncommented voa hybris-patches but maybe your device is first one which really needs it | 19:43 |
T42 | <neochapay> can you show which of path ? (re @elros34: yeah handling code i...) | 19:45 |
T42 | <elros34> do you even use hybris-18? | 19:54 |
T42 | <elros34> /bootstrap is commented out in init.rc but all these namespaces in system/core/init.cpp/mount_namespaces.cpp | 19:55 |
T42 | <neochapay> Yeap (re @elros34: do you even use hybr...) | 19:58 |
T42 | <elros34> something worth to read about it: https://android.googlesource.com/platform/system/apex/+/715e23df6e297462b6ef10e9abbb3ea752fdf6b1%5E%21/. Mind that I not an expert so this may be wrong | 20:00 |
T42 | <elros34> btw droid-hal-init.inc also copy some lib{dl,c,m}.so files, did you set android_version_major in spec file? | 20:07 |
T42 | <neochapay> to 10 (re @elros34: btw droid-hal-init.i...) | 20:07 |
T42 | <elros34> not 11? | 20:08 |
T42 | <neochapay> oops | 20:09 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!