T42_ | <vilez0> i was building the droid-hal but it gives invalid for CONFIG_SUNRPC, CONFIG_UEVENT_HELPER_PATH, and CONFIG_LOCKD although they are defined in the defconfig it does not build the package | 10:25 |
---|---|---|
T42_ | <elros34> sunrpc/lockd for sure it's warning/optional and i would say even more not recommended as other optional network filesystems | 11:01 |
T42_ | <elros34> what excatly you got for ueven_helper_path? | 11:02 |
T42_ | <vilez0> @elros34 it gives unset | 11:20 |
T42_ | <elros34> did you set it to "" or =y/n? | 11:21 |
T42_ | <vilez0> "" | 11:21 |
T42_ | <vilez0> empty string | 11:21 |
T42_ | <elros34> show your whole defconfig then (use some pastebin service) | 11:22 |
T42_ | <elros34> did you run make hybris-hal after changing defconfig/ | 11:23 |
T42_ | <vilez0> https://github.com/Vilez0/kernel_xiaomi_sweet_k6a/commit/3104a26eb94898ded8fa2b2a561360e8ad49b380 (re @elros34: show your whole defc...) | 11:26 |
T42_ | <vilez0> @elros34 i executed make hybris-boot and make hybris-recovery after changing defconfig | 11:27 |
T42_ | <elros34> remove # from defconfig lines | 11:31 |
T42_ | <elros34> CONFIG_UEVENT_HELPER is not even enabled. I woudn't change that then | 11:33 |
T42_ | <vilez0> lemme try | 11:36 |
T42_ | <vilez0> @elros34 https://paste.medip.dev/kHViUS` `uevent solved but it still doesnt build | 11:58 |
T42_ | <elros34> there is no error in this log see second to last line for full log with reall error | 12:00 |
T42_ | <vilez0> that file doesnt exists | 12:01 |
T42_ | <elros34> it really must be there, run 'cat <path to that file>' in platform sdk | 12:03 |
T42_ | <elros34> must be in your $ANDROID_ROOT/ | 12:04 |
T42_ | <vilez0> @elros34 https://paste.medip.dev/SgIeh5 | 12:15 |
T42_ | <vilez0> as you see, the file doesnt exists | 12:15 |
T42_ | <elros34> strange, nothing similar: droid-hal*.log? | 12:16 |
T42_ | <vilez0> in $ANROID_ROOT there is nothing similar | 12:17 |
T42_ | <elros34> try adding 'bash -x ' when you run build_packages.sh script | 12:18 |
T42_ | <vilez0> okay | 12:18 |
T42_ | <vilez0> @elros34 i did a reboot for the server, after the reboot i re executed the command then the file appeared 😳 | 12:31 |
T42_ | <vilez0> ok i solved the errors | 13:13 |
T42_ | <vilez0> now we have another build error | 13:13 |
T42_ | <vilez0> `RPM build errors: | 13:13 |
T42_ | <vilez0> absolute symlink: /etc/udev/rules.d/60-persistent-v4l.rules -> /dev/null | 13:13 |
T42_ | <vilez0> absolute symlink: /usr/libexec/droid-hybris/system/bin/linker -> /apex/com.android.runtime/bin/linker | 13:13 |
T42_ | <vilez0> absolute symlink: /usr/libexec/droid-hybris/system/bin/linker64 -> /apex/com.android.runtime/bin/linker64 | 13:13 |
T42_ | <vilez0> absolute symlink: /usr/libexec/droid-hybris/system/bin/linker_asan -> /apex/com.android.runtime/bin/linker | 13:13 |
T42_ | <vilez0> absolute symlink: /usr/libexec/droid-hybris/system/bin/linker_asan64 -> /apex/com.android.runtime/bin/linker64 | 13:13 |
T42_ | <vilez0> File listed twice: /usr/libexec/droid-hybris/system/etc/init/apexd.rc | 13:13 |
T42_ | <vilez0> Installed (but unpackaged) file(s) found: | 13:13 |
T42_ | <vilez0> /bugreports | 13:13 |
T42_ | <vilez0> /d | 13:13 |
T42_ | <vilez0> /sdcard | 13:13 |
T42_ | <vilez0> ` | 13:13 |
T42_ | <vilez0> full log: https://paste.medip.dev/p/nm6oEt | 13:13 |
mal | read about straggler files | 13:14 |
mal | an example https://github.com/mlehtima/droid-hal-fp2-sibon/blob/master/droid-hal-fp2-sibon.spec#L18 you of course only need the file that are listed in that "Installed (but unpackaged) file(s) found" | 13:15 |
T42_ | <vilez0> okay | 13:19 |
T42_ | <vilez0> `File /etc/ofono/binder.conf | 13:51 |
T42_ | <vilez0> from install of | 13:51 |
T42_ | <vilez0> droid-config-sweet_k6a-1-202407061327.aarch64 (dir:/home/edip/hadk/hybris/droid-hal-version-sweet_k6a/.mb2/filtered-output-dir) | 13:51 |
T42_ | <vilez0> conflicts with file from package | 13:51 |
T42_ | <vilez0> ofono-configs-binder-1.1.11-1.5.1.jolla.aarch64 (@System)` | 13:51 |
T42_ | <vilez0> file conflicts | 13:51 |
T42_ | <elros34> https://sailfishos.wiki/link/20#bkmrk-build_packages.sh-fa-0 | 13:55 |
T42_ | <vilez0> i guess the link that present in the wiki (https://github.com/mer-hybris/droid-config-sony-nile/blob/master/droid-config-common.inc#L11-L12) indicates wrong lines (re @elros34: https://sailfishos.w...) | 14:27 |
T42_ | <elros34> yeah last commit added few lines, obviously you need these related to ofono | 14:28 |
T42_ | <vilez0> 👍 | 14:29 |
T42_ | <vilez0> it stuck in bootloader (no bootloop, just waiting on redmi logo) | 15:17 |
T42_ | <vilez0> *rebooted now | 15:18 |
mal | any usb interface visible on host? | 15:18 |
T42_ | <vilez0> no | 15:19 |
T42_ | <vilez0> while i was porting postmarketos (mainline kernel), it was necessary to erase dtbo before boot (because bootloader will append dtb in the end of the kernel and corrupt the kernel) | 15:23 |
T42_ | <vilez0> is there something similar there? | 15:23 |
T42_ | <elros34> you got reboot after 1 minute? | 15:24 |
T42_ | <vilez0> after ~1-2 minutes | 15:24 |
T42_ | <elros34> quiet significant difference 1minute or 1-2 minutes | 15:24 |
T42_ | <elros34> did you check init.log? | 15:26 |
mal | check the usb kernel configs | 15:26 |
T42_ | <vilez0> its exactly 1 minute and 15 second | 15:27 |
mal | so could be that usb interface cannot be enabled | 15:27 |
T42_ | <vilez0> lemme check | 15:28 |
T42_ | <vilez0> where is it located? (re @elros34: did you check init.l...) | 15:29 |
T42_ | <elros34> check hadk-hot, you can find it via recovery | 15:30 |
T42_ | <vilez0> there is no init.log :D | 15:37 |
T42_ | <vilez0> so it doesnt even start | 15:38 |
T42_ | <vilez0> *this means | 15:38 |
T42_ | <elros34> are you sure you have all of these set correctly in generated .config (not defconfig) https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L344-L352? | 15:39 |
T42_ | <vilez0> yes, https://paste.medip.dev/Aiwc4L | 15:42 |
T42_ | <vilez0> CONFIG_SYSFS_DEPRECATED and CONFIG_DUMMY are not set | 15:43 |
T42_ | <elros34> then I guess is mount issue? Anyway did you enabled CONFIG_USB_CONFIGFS_*RNDIS? | 15:46 |
T42_ | <vilez0> `$ cat out/target/product/sweet_k6a/obj/KERNEL_OBJ/.config | grep -i CONFIG_USB_CONFIGFS_*RNDIS | 15:47 |
T42_ | <vilez0> # CONFIG_USB_CONFIGFS_RNDIS is not set | 15:47 |
T42_ | <vilez0> | 15:47 |
T42_ | <vilez0> `no its not enabled | 15:47 |
T42_ | <vilez0> i will recheck the partitions again (re @elros34: then I guess is moun...) | 15:47 |
T42_ | <vilez0> from fixup-mountpoints | 15:48 |
T42_ | <vilez0> btw there was "failed to mount /dev/block/bootdevice/by-name/userdata at /data: device or resource busy" error while flashing sailfishos | 15:50 |
T42_ | <vilez0> could it be the problem? | 15:50 |
T42_ | <vilez0> /bootdevice | 15:50 |
T42_ | <elros34> no because /data was already mounted | 15:50 |
T42_ | <elros34> enable rndis and try top get telnet, it will be faster | 15:51 |
T42_ | <vilez0> okay | 15:51 |
T42_ | <vilez0> @elros34 i did it but still doesnt boot (no log file) | 17:14 |
T42_ | <vilez0> the device isnt shown in lsusb also/no telnet connection | 17:14 |
T42_ | <vilez0> do hybris use pstore like all other android do? | 17:15 |
T42_ | <elros34> everything hybris-boot.img uses is in init-script in hybris-boot repo | 17:16 |
T42_ | <elros34> did you format data partition before flashing hybris-boot.img? | 17:17 |
T42_ | <vilez0> yes, i did | 17:17 |
T42_ | <vilez0> i checked it 2 times also, there is no problem with data part | 17:17 |
T42_ | <vilez0> im looking into pstore and looks like there is a kernel panic : https://irc.thaodan.de/.imgstore/661320ad/file_6336.jpg | 17:19 |
T42_ | <vilez0> oops, sent screenshot by mistake | 17:19 |
T42_ | <vilez0> i was looking into pstore and looks like there is a kernel panic | 17:19 |
T42_ | <vilez0> https://paste.medip.dev/vRczwN | 17:19 |
T42_ | <elros34> hard to tell from this few lines | 17:21 |
T42_ | <vilez0> its corrupted mostly | 17:21 |
T42_ | <elros34> anything change if you increase/decrease 'sleep 60' in init-script? | 17:22 |
T42_ | <vilez0> lemme try | 17:22 |
T42_ | <elros34> I hope you are booting hybris-boot not hybris-recovery | 17:23 |
T42_ | <vilez0> i flashed the zip (generated by build_script) from recovery | 17:24 |
T42_ | <vilez0> i flashed the zip (generated by build_script) from ofox recovery (edited) | 17:24 |
T42_ | <elros34> but after you enabled rndis? | 17:25 |
T42_ | <vilez0> yes, i enabled rndis | 17:26 |
T42_ | <vilez0> then executed make hybris-boot (which will rebuild the kernel also) | 17:26 |
T42_ | <vilez0> then repackaged droid-hal and recreated the .zip via build_script --mic | 17:26 |
T42_ | <vilez0> and then reflashed | 17:28 |
T42_ | <elros34> to much work, you should flash directly hybris-boot.img and wathc host dmesg and boot time at first boot. That way you also can avoid switching slots or some other isssues | 17:28 |
T42_ | <vilez0> okay | 17:30 |
T42_ | <elros34> add 'DBG_REASON=""' before 'if ! [ "$DBG_REASON" = "" ] ; then' in init script | 17:30 |
T42_ | <vilez0> ok | 17:31 |
T42_ | <elros34> then you may have init.log | 17:31 |
T42_ | <b100dian> SDEMAGPIE, curious what phone was this? | 17:34 |
T42_ | <vilez0> *SDMMAGPIE, | 17:35 |
T42_ | <vilez0> its redmi note 12 pro 4g (re @b100dian: SDEMAGPIE, curious w...) | 17:35 |
T42_ | <vilez0> sweet_k6a | 17:35 |
T42_ | <vilez0> there is only 1 sdmmagpie device(xiaomi davinci, redmi k20) in libhybris wiki | 17:36 |
T42_ | <vilez0> which seems its unmaintained | 17:36 |
T42_ | <b100dian> I think the same codename was for Mi Note 10 (tucana) but the chipset is slightly different https://m.gsmarena.com/compare.php3?idPhone1=12209&idPhone2=9936 | 17:39 |
T42_ | <vilez0> yes, there is also poco x3 (nfc), pixel 4a, redmi note 10 pro 4g, mi 11 lite 4g | 17:41 |
T42_ | <b100dian> Are you using the xiaomi or the lineage kernel out of curiosity? | 17:42 |
T42_ | <vilez0> lineage kernel (re @b100dian: Are you using the xi...) | 17:47 |
T42_ | <vilez0> https://github.com/MiDoNaSR545/kernel_xiaomi_sweet_k6a | 17:47 |
T42_ | <vilez0> this one | 17:47 |
T42_ | <elros34> this is lineage 21 kernel, di you test it on lineage 20? | 18:08 |
T42_ | <vilez0> not me directly, but other developers did (re @elros34: this is lineage 21 k...) | 18:13 |
T42_ | <vilez0> changing this made it appears in lsusb | 18:24 |
T42_ | <vilez0> `Bus 002 Device 046: ID 18d1:d001 Google Inc. Nexus 4 (fastboot)` (re @elros34: add 'DBG_REASON=""' ...) | 18:24 |
T42_ | <elros34> check init.log | 18:25 |
T42_ | <elros34> and create init_enter_debug2 file in sfos rootfs | 18:25 |
T42_ | <vilez0> there is not init.log | 18:28 |
T42_ | <vilez0> data and system are mounted from recovery | 18:28 |
T42_ | <vilez0> the folder exists but there is no such file | 18:28 |
T42_ | <vilez0> *there is no | 18:28 |
T42_ | <elros34> did you check correct place? /data/.stowaways/sailfishos/init.log? | 18:30 |
T42_ | <vilez0> yes | 18:30 |
T42_ | <elros34> did you check this sleep 60 thing we were talking about? | 18:31 |
T42_ | <vilez0> tbh, i did it with DBG_REASON="" | 18:34 |
T42_ | <vilez0> i changed it but i forgot to test | 18:34 |
T42_ | <vilez0> .d | 18:34 |
T42_ | <elros34> can you touch some debug file here https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L174 and it appears in recovery? This way you can at least create some debug info | 18:34 |
T42_ | <vilez0> ok | 18:35 |
T42_ | <elros34> comment out this one too https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L130 | 18:35 |
T42_ | <elros34> or whole function | 18:36 |
T42_ | <vilez0> @elros34 it created my file | 19:20 |
T42_ | <vilez0> https://paste.medip.dev/LiGy8P | 19:20 |
T42_ | <vilez0> i still didnt understand why it doesnt create init.log | 19:32 |
T42_ | <elros34> replace this line https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L409 with 'cp /init.log /target/initramfs.log; sync'. Not sure it will work like that. You can also redirect dmesg to file. Do you get enaything in DBG_REASON before you clear it up? | 19:38 |
T42_ | <vilez0> https://paste.medip.dev/p/i6sK99 @elros34 | 20:02 |
T42_ | <vilez0> DBG_REASON: Refusing to boot. See /diagnosis.log | 20:02 |
T42_ | <vilez0> `+ echo 'CONFIG_DUMMY=y found in /proc/config.gz, must be disabled' | 20:03 |
T42_ | <vilez0> `wtf | 20:03 |
T42_ | <elros34> I was pretty sure it must be in correct | 20:03 |
T42_ | <vilez0> https://paste.medip.dev/AAkQe5 | 20:04 |
T42_ | <vilez0> CONFIG_DUMMY is unset | 20:04 |
T42_ | <vilez0> i guess now i should copy the diagnosis.log :d (re @vilez0: DBG_REASON: Refusing...) | 20:05 |
T42_ | <elros34> not really | 20:05 |
T42_ | <vilez0> btw https://paste.medip.dev/p/Rj69nr | 20:06 |
T42_ | <vilez0> dme`sg` | 20:06 |
T42_ | <vilez0> btw https://paste.medip.dev/p/Rj69nr | 20:06 |
T42_ | <vilez0> dmesg (edited) | 20:06 |
T42_ | <elros34> zcat /proc/config.gz > /target/kernel_config | 20:07 |
T42_ | <elros34> then compare with your .config, for sure it's different | 20:08 |
T42_ | <vilez0> but, why? | 20:10 |
T42_ | <elros34> that is good question. I can see you have disabled CONFIG_AUTOFS4_FS according to initramfs.log. Is it enabled in generated .config file? | 20:11 |
T42_ | <elros34> you are booting 4.14.336-KewL kernel is it same you are building? | 20:12 |
T42_ | <vilez0> CONFIG_AUTOFS4_FS is enabled in my .config file | 20:22 |
T42_ | <vilez0> and yes, the kernel that im building is kewl kernel | 20:22 |
T42_ | <vilez0> is there a way to force rebuild the kernel | 20:24 |
T42_ | <vilez0> maybe it can solve it | 20:24 |
T42_ | <vilez0> *force rebuild entire kernel | 20:24 |
T42_ | <elros34> make some random change and run make hybris-hal, see whether it finish without error | 20:25 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!