T42 | <DarknessHiddenorg> HABUILD_SDK [ginkgo] HABUILD_SDK [ginkgo] tea@Asus-H61M-C-938:~/hadk$ hybris-patches/apply-patches.sh --mb | 12:53 |
---|---|---|
T42 | <DarknessHiddenorg> fatal: not a git repository (or any of the parent directories): .git | 12:53 |
T42 | <DarknessHiddenorg> how to apply the patch? | 12:54 |
T42 | <elros34> you know there is port for ginkgo? | 12:55 |
T42 | <DarknessHiddenorg> where (re @elros34: you know there is po...) | 13:04 |
T42 | <Umeaman> https://github.com/sailfish-on-ginkgo/ (re @DarknessHiddenorg: where) | 13:09 |
T42 | <elros34> yes, first result in google "sailfish ginkgo':) | 13:15 |
T42 | <DarknessHiddenorg> i think that for willow before (re @elros34: yes, first result in...) | 13:45 |
irulan[m] | is a port for the ppp availible? | 13:52 |
T42 | <Mister_Magister> man ppp sucks i prefered when everyone was talking about penises | 14:00 |
irulan[m] | I am not sure how a penis with sailfish works? | 14:01 |
T42 | <Mister_Magister> PP | 14:02 |
T42 | <Mister_Magister> it works decent | 14:02 |
irulan[m] | That I am awre of, but I do not have e pp obsolete | 14:03 |
T42 | <Mister_Magister> yee just joking | 14:03 |
irulan[m] | No problem | 14:03 |
irulan[m] | I was reading in the forum, that some people have done a port, where to find it? | 14:04 |
mal | @adampigg ^ | 14:05 |
T42 | <WrrrIamAnAngryCat> Hey I'm building sailfishos 4.5 for the oneplus 6t (fajita), and during the step 5.4 Building Relevant Bits of your Android base: | 14:25 |
T42 | <WrrrIamAnAngryCat> $ make -j$(nproc --all) hybris-hal droidmedia | 14:25 |
T42 | <WrrrIamAnAngryCat> | 14:25 |
T42 | <WrrrIamAnAngryCat> I receive the following error: | 14:25 |
T42 | <WrrrIamAnAngryCat> | 14:25 |
T42 | <WrrrIamAnAngryCat> ninja: error: '/home/user/hadk.fajita/out/host/linux-x86/bin/mkdtimg', needed by '/home/user/hadk.fajita/out/target/product/fajita/dtbo-pre.img', missing and no known rule to make it | 14:25 |
T42 | <WrrrIamAnAngryCat> | 14:25 |
T42 | <WrrrIamAnAngryCat> I've tried fixing it by doing `mka mkbootimg` (initially `mka mkdtimg` - but no target with that name was found) but it didn't help and I still get the same error message. | 14:25 |
T42 | <WrrrIamAnAngryCat> `( cd $ANDROID_ROOT && ln -s ../../../../prebuilts/misc/linux-x86/libufdt/mkdtimg out/host/linux-x86/bin/ )` fixed this issue | 14:29 |
T42 | <b100dian> make dtboimage? | 15:16 |
T42 | <b100dian> Although why is it called pre | 15:16 |
T42 | <adampigg> irulan[m]: https://github.com/sailfish-on-dontbeevil/documentation/wiki | 15:18 |
T42 | <WrrrIamAnAngryCat> still the same error (re @b100dian: make dtboimage?) | 16:09 |
T42 | <b100dian> Hmm grep *.mk files for dtbo-pre, maybe something shows up | 16:21 |
T42 | <WrrrIamAnAngryCat> nothing that interesting - I'm keeping the symlink as a fix for now (re @b100dian: Hmm grep *.mk files ...) | 16:28 |
T42 | <b100dian> oh, right, sorry I thought the dtbo-pre.img was not found, but I misread: it was the mkdtimg missing. Sounds like a such link is ok | 16:31 |
T42 | <WrrrIamAnAngryCat> I think that this part is now behind me - `#### build completed successfully (18 seconds) ####`, | 16:31 |
T42 | <WrrrIamAnAngryCat> | 16:31 |
T42 | <WrrrIamAnAngryCat> But now it fails at: | 16:31 |
T42 | <WrrrIamAnAngryCat> | 16:31 |
T42 | <WrrrIamAnAngryCat> ``` | 16:31 |
T42 | <WrrrIamAnAngryCat> $ rpm/dhd/helpers/build_packages.sh --droid-hal | 16:31 |
T42 | <WrrrIamAnAngryCat> [...] | 16:31 |
T42 | <WrrrIamAnAngryCat> + cp -a ./out/target/product/fajita/hybris-updater-script /home/user/hadk.fajita/installroot/boot | 16:31 |
T42 | <WrrrIamAnAngryCat> cp: cannot stat './out/target/product/fajita/hybris-updater-script': No such file or directory``` | 16:31 |
T42 | <WrrrIamAnAngryCat> I've searched the chat but only found 'nvm figured it out' kind of messages so I'm digging deeper | 16:32 |
T42 | <WrrrIamAnAngryCat> I've seen that it is may be missing due to badly implemented fixup-mountpoints (but it is correct, just checked the readlink on a real device). | 16:33 |
T42 | <WrrrIamAnAngryCat> lol - all I had to do was `make hybrid-updater-script` (re @SailfishFreenodeIRCBridgeBot: <Mister_Magister> ma...) | 16:40 |
T42 | <WrrrIamAnAngryCat> is it normal for `$ANDROID_ROOT/RPMS` to contain `armv7hl` packages instead of aarch64(PORT_ARCH)? | 16:51 |
T42 | <Walid> this script is no longer available https://susepaste.org/40869869 in hadk-faq | 17:06 |
T42 | <elros34> @WrrrIamAnAngryCat there must be something wrong if hybris-hal didn't build hybris-updater, check again whole make output. About arch rhen if you set aarch64 then you shouldn't have arm rpms in droid-local-repo. Not sure what temp directory is in ANDROID_ROOT/RPMS/ | 17:18 |
mal | @Walid what script should that have been? | 17:24 |
mal | ah, it was the symlink script | 17:25 |
T42 | <WrrrIamAnAngryCat> Okay - I've seen where was the armv7hl issue coming from, thanks (re @elros34: @WrrrIamAnAngryCat t...) | 17:31 |
T42 | <WrrrIamAnAngryCat> I'm also getting similar errors - and I'm trying to | 18:18 |
T42 | <WrrrIamAnAngryCat> | 18:18 |
T42 | <WrrrIamAnAngryCat> ``` | 18:18 |
T42 | <WrrrIamAnAngryCat> File /etc/bluez5/bluetooth/input.conf | 18:18 |
T42 | <WrrrIamAnAngryCat> from install of | 18:18 |
T42 | <WrrrIamAnAngryCat> droid-config-fajita-bluez5-1-202308281756.aarch64 (dir:/home/user/hadk.fajita/droid-local-repo/fajita) | 18:18 |
T42 | <WrrrIamAnAngryCat> conflicts with file | 18:18 |
T42 | <WrrrIamAnAngryCat> /etc/bluetooth/input.conf | 18:18 |
T42 | <WrrrIamAnAngryCat> from package | 18:18 |
T42 | <WrrrIamAnAngryCat> bluez5-configs-mer-5.65+git1-1.9.1.jolla.aarch64 (@System)``` | 18:18 |
T42 | <WrrrIamAnAngryCat> | 18:18 |
T42 | <WrrrIamAnAngryCat> I'm getting multiple errors like this - but I'm unable to find any information about which file (I've checked .inc files provides droid-config-fajita-bluez5) where should I add the obselote tag / how to identify which file is responsible for that config/ (re @TheVancedGamer: File /etc/bluez5/blu...) | 18:18 |
T42 | <WrrrIamAnAngryCat> I'm also getting similar errors - and I'm trying to | 18:18 |
T42 | <WrrrIamAnAngryCat> | 18:18 |
T42 | <WrrrIamAnAngryCat> ``` | 18:18 |
T42 | <WrrrIamAnAngryCat> File /etc/bluez5/bluetooth/input.conf | 18:18 |
T42 | <WrrrIamAnAngryCat> from install of | 18:18 |
T42 | <WrrrIamAnAngryCat> droid-config-fajita-bluez5-1-202308281756.aarch64 (dir:/home/user/hadk.fajita/droid-local-repo/fajita) | 18:18 |
T42 | <WrrrIamAnAngryCat> conflicts with file | 18:18 |
T42 | <WrrrIamAnAngryCat> /etc/bluetooth/input.conf | 18:18 |
T42 | <WrrrIamAnAngryCat> from package | 18:18 |
T42 | <WrrrIamAnAngryCat> bluez5-configs-mer-5.65+git1-1.9.1.jolla.aarch64 (@System) | 18:18 |
T42 | <WrrrIamAnAngryCat> ``` | 18:18 |
T42 | <WrrrIamAnAngryCat> I'm getting multiple errors like this - but I'm unable to find any information about which file (I've checked .inc files provides droid-config-fajita-bluez5) where should I add the obselote tag / how to identify which file is responsible for that config? (edited) (re @TheVancedGamer: File /etc/bluez5/blu...) | 18:18 |
T42 | <elros34> did you read https://etherpad.wikimedia.org/p/hadk-hot? | 18:22 |
T42 | <WrrrIamAnAngryCat> Didn't found my issue there but after trying adding Obsoletes: bluez5-configs-mer into many different places I think that I've figured it out... (re @elros34: did you read https:/...) | 18:25 |
T42 | <WrrrIamAnAngryCat> 🎉I've managed to boot sailfish os and I have working telnet access, but I get plenty of errors like this: | 19:58 |
T42 | <WrrrIamAnAngryCat> Feb 15 18:03:12 OnePlus6T sh[2611]: /bin/sh: can't open '/usr/bin/droid/droid-hal-early-init.sh': No such file or directory | 19:58 |
T42 | <WrrrIamAnAngryCat> in journal | 19:58 |
T42 | <WrrrIamAnAngryCat> ``` | 20:02 |
T42 | <WrrrIamAnAngryCat> Feb 15 18:12:06 OnePlus6T systemd[1]: Starting droid-hal-init... | 20:02 |
T42 | <WrrrIamAnAngryCat> Feb 15 18:12:06 OnePlus6T systemd[1]: droid-hal-init.service: Main process exited, code=exited, status=6/NOTCONFIGURED | 20:02 |
T42 | <WrrrIamAnAngryCat> Feb 15 18:12:06 OnePlus6T systemd[1]: droid-hal-init.service: Failed with result 'exit-code'.``` | 20:02 |
T42 | <WrrrIamAnAngryCat> | 20:02 |
T42 | <WrrrIamAnAngryCat> | 20:02 |
T42 | <WrrrIamAnAngryCat> after putting an empty file in place of droid-hal-early-init.sh this is what I get in logcat.. | 20:02 |
T42 | <TheMichalos> droid-hal-early-init is optional so you don't need to worry about it erroring | 20:05 |
T42 | <b100dian> Congrats. That is normal. Whatbdo you have in logcat after booting? | 20:06 |
T42 | <WrrrIamAnAngryCat> I cant run logcat - it says file not found :/ (re @b100dian: Congrats. That is no...) | 20:09 |
T42 | <b100dian> You need to run it as early as possible | 20:10 |
T42 | <b100dian> Or mask it and manually run it | 20:10 |
T42 | <WrrrIamAnAngryCat> Mask logcat? (re @b100dian: Or mask it and manua...) | 20:10 |
T42 | <b100dian> Maskthe systemd service that starts it | 20:11 |
T42 | <b100dian> User@10000 probably | 20:11 |
T42 | <b100dian> I am not at laptop.. | 20:11 |
T42 | <WrrrIamAnAngryCat> But this will solve the issue with logcat saying file not found? (re @b100dian: I am not at laptop..) | 20:16 |
T42 | <elros34> no, dont worry about logcat now, did you do what i tell you to about hybris-updater | 20:16 |
T42 | <b100dian> It will allow you to grab logs timely | 20:16 |
T42 | <elros34> logcat is last what you should worry | 20:17 |
T42 | <elros34> usually | 20:17 |
T42 | <WrrrIamAnAngryCat> This message, right? (re @elros34: @WrrrIamAnAngryCat t...) | 20:17 |
T42 | <WrrrIamAnAngryCat> I'll make clean and rebuild then | 20:17 |
T42 | <elros34> yes | 20:17 |
T42 | <elros34> no need to | 20:18 |
T42 | <b100dian> I think youre in good hands:) | 20:18 |
T42 | <elros34> you never need to run make clean, just make hybris-hal and read it carefully | 20:18 |
T42 | <WrrrIamAnAngryCat> there was something about `/boot lives on <empty string here>` first time I built it but now I don't see it | 20:19 |
T42 | <WrrrIamAnAngryCat> there are bunch of warnings - but they look like due to vendor stuff so probably not what is interesting for us | 20:19 |
T42 | <elros34> IIRC you should have boot lives to and data lives to messages | 20:20 |
T42 | <WrrrIamAnAngryCat> https://pastebin.pl/view/bbc66318 this is my entire output | 20:22 |
T42 | <elros34> change something in kernel and try again | 20:24 |
T42 | <WrrrIamAnAngryCat> I've added soem random FOO=BAR into config and it resulted in the same build | 20:26 |
T42 | <WrrrIamAnAngryCat> I've added soem random FOO=BAR into config and it resulted in the same log (edited) | 20:26 |
T42 | <elros34> random might be not enough, touch some source code | 20:26 |
T42 | <elros34> you should have rebuild hybris-boot.img | 20:27 |
T42 | <WrrrIamAnAngryCat> it rebuilds the hybris-boot.img (re @elros34: you should have rebu...) | 20:27 |
T42 | <elros34> but not in your log | 20:27 |
T42 | <WrrrIamAnAngryCat> places the new file and stuff - it just doesn't print the `/boot appears to live on` string | 20:27 |
T42 | <WrrrIamAnAngryCat> oh right | 20:28 |
T42 | <WrrrIamAnAngryCat> hybris-boot.img is old | 20:28 |
T42 | <WrrrIamAnAngryCat> I've done mka clean | 20:30 |
T42 | <WrrrIamAnAngryCat> ``` | 20:30 |
T42 | <WrrrIamAnAngryCat> [310/440] including hybris/hybris-boot/Android.mk ... | 20:30 |
T42 | <WrrrIamAnAngryCat> hybris/hybris-boot/Android.mk:71: warning: ********************* /boot appears to live on | 20:30 |
T42 | <WrrrIamAnAngryCat> hybris/hybris-boot/Android.mk:72: warning: ********************* /data appears to live on /dev/block/bootdevice/by-name/userdata | 20:30 |
T42 | <WrrrIamAnAngryCat> hybris/hybris-boot/Android.mk:306: warning: Skipping build of hybris-updater-script since HYBRIS_BOOT_PART is not specified``` | 20:30 |
T42 | <WrrrIamAnAngryCat> and indeed we have these messages | 20:30 |
T42 | <WrrrIamAnAngryCat> and it fails because of ninja: error: '/home/user/hadk.fajita/out/host/linux-x86/bin/mkdtimg', needed by '/home/user/hadk.fajita/out/target/product/fajita/dtbo-pre.img', missing and no known rule to make it | 20:30 |
T42 | <elros34> so this is your first issue "/boot appears to live on " | 20:30 |
T42 | <WrrrIamAnAngryCat> How can I fix it? (re @elros34: so this is your firs...) | 20:31 |
T42 | <WrrrIamAnAngryCat> My initial thought was that it is due to fixup-mountpoints but | 20:32 |
T42 | <WrrrIamAnAngryCat> 1) my device is upstreamed to hybris-boot | 20:32 |
T42 | <WrrrIamAnAngryCat> 2) it contains -e `'s block/bootdevice/by-name/boot sde11 '` - which is correct | 20:32 |
T42 | <elros34> probably you need to add /boot you android's fstab | 20:32 |
T42 | <WrrrIamAnAngryCat> I guess that `device/oneplus/sdm845-common/rootdir/etc/fstab.qcom` this one is the correct fstab | 20:35 |
T42 | <WrrrIamAnAngryCat> | 20:35 |
T42 | <WrrrIamAnAngryCat> and adding `/dev/block/bootdevice/by-name/boot /boot ext4 defaults defaults` is all I need? (re @elros34: probably you need to...) | 20:35 |
T42 | <WrrrIamAnAngryCat> `/boot appears to live on /dev/block/bootdevice/by-name/boot | 20:36 |
T42 | <WrrrIamAnAngryCat> ` | 20:36 |
T42 | <WrrrIamAnAngryCat> `/boot appears to live on /dev/block/bootdevice/by-name/boot` (edited) | 20:36 |
T42 | <WrrrIamAnAngryCat> hell yeah! | 20:36 |
T42 | <elros34> I would say emmc instead of ext4 | 20:36 |
T42 | <WrrrIamAnAngryCat> okay (re @elros34: I would say emmc ...) | 20:36 |
T42 | <WrrrIamAnAngryCat> also - one other issue I had was that flashable zip didn't work because my twrp doesn't have bunzip2 available.. I have just extracted it on host and manually installed on device.. but maybe there is some option to disable that compression or use some other thing that is supported by my recovery? | 20:37 |
T42 | <WrrrIamAnAngryCat> (and because twrp says that my device codename is `OnePlus6T` when it clearly is `fajita` - but I've just removed the line with assert()) | 20:38 |
T42 | <WrrrIamAnAngryCat> (and because twrp says that my device codename is `OnePlus6T` when it clearly is `fajita` - but I've just removed the line with assert(), it also fails) (edited) | 20:39 |
T42 | <elros34> yes bzip it's know issue, one (not the best) workaround can be: format data, copy sfos image, flash hybris-boot.img, telnet and unpack image there | 20:40 |
T42 | <WrrrIamAnAngryCat> so pretty much what I do except I unpack it on host... (re @elros34: yes bzip it's know i...) | 20:40 |
T42 | <elros34> but unpacking and copying is prone to permission errors | 20:41 |
T42 | <WrrrIamAnAngryCat> oh nono, i just get rid of the compression, and unpack tar using the exact command from installer script (re @elros34: but unpacking and co...) | 20:42 |
T42 | <elros34> you can switch to tar.gz but I do not have ready to use instruction, for sure it's somewhere in logs | 20:42 |
T42 | <WrrrIamAnAngryCat> alright I'll take care of it later then as it's not a priority at this stage (re @elros34: you can switch to ta...) | 20:42 |
T42 | <elros34> sure, based on your few lines of journal you should check mountpoints | 20:43 |
T42 | <elros34> on device | 20:43 |
T42 | <WrrrIamAnAngryCat> I'll wait for the updated build to finish building itself and check if that automagically fixed some issues | 20:44 |
T42 | <elros34> no it will not | 20:44 |
T42 | <WrrrIamAnAngryCat> don't take away my hope | 20:44 |
T42 | <elros34> whatever you like but it's waste of time | 20:44 |
T42 | <WrrrIamAnAngryCat> okay, | 20:45 |
T42 | <WrrrIamAnAngryCat> so I should boot it now and see the mountpoints? | 20:45 |
T42 | <elros34> yeah: "findmnt", check whether /system looks ok | 20:45 |
T42 | <WrrrIamAnAngryCat> there is no `/system` at all (re @elros34: yeah: "findmnt", che...) | 20:46 |
T42 | <elros34> check in journal why | 20:48 |
T42 | <WrrrIamAnAngryCat> ``` | 20:52 |
T42 | <WrrrIamAnAngryCat> Feb 15 18:55:46 OnePlus6T kernel: sda: sda1 sda2 sda3 sda4 sda5 sda6 sda7 sda8 sda9 sda10 sda11 sda12 sda13 sda14 sda15 sda16 sda17 | 20:52 |
T42 | <WrrrIamAnAngryCat> Feb 15 18:55:48 OnePlus6T systemd-udevd[1192]: Process '/bin/ln -s /dev/block/sda13 /dev/block/platform/soc/1d84000.ufshc/by-num/p13' failed with exit code 1.``` | 20:52 |
T42 | <WrrrIamAnAngryCat> | 20:52 |
T42 | <WrrrIamAnAngryCat> | 20:52 |
T42 | <WrrrIamAnAngryCat> Could this be the reason? (sda13 is block/bootdevice/by-name/system) (re @elros34: check in journal why) | 20:52 |
T42 | <elros34> no, according to hadk-hot search for "droid mount" | 20:53 |
T42 | <WrrrIamAnAngryCat> 0 results. (re @elros34: no, according to had...) | 20:54 |
T42 | <elros34> check if whether systemd mount units where created in $ANDROID_ROOT/droid-hal-device.log | 20:55 |
T42 | <WrrrIamAnAngryCat> file not found, will retry once full rebuild is done (re @elros34: check if whether sys...) | 20:56 |
T42 | <elros34> it should be there device is probably $DEVICE | 20:56 |
T42 | <WrrrIamAnAngryCat> oh right | 20:57 |
T42 | <WrrrIamAnAngryCat> `cp -a tmp/units/config.mount /home/user/hadk.fajita/installroot//usr/lib/systemd/system` I think that this may be it? | 20:59 |
T42 | <elros34> https://etherpad.wikimedia.org/p/hadk-hot#L125 | 20:59 |
T42 | <WrrrIamAnAngryCat> I have uevent but no signs of fstab (re @elros34: https://etherpad.wik...) | 21:04 |
T42 | <WrrrIamAnAngryCat> I have uevent but no signs of fstab, trying to figure it out (edited) (re @elros34: https://etherpad.wik...) | 21:04 |
T42 | <WrrrIamAnAngryCat> Also - should I add vendor to my fstab? -I'll have to copy it manually because there are no modules which do that sadly | 21:08 |
T42 | <WrrrIamAnAngryCat> for now I'm going to sleep. Poritng since 8 hours or so and I have to wake up early, thank you Elros for your help! I'll message you once I make progress, most likely tomorrow after work | 21:09 |
T42 | <elros34> if your device uses it then maybe you should add it | 21:13 |
T42 | <Walid> how can i build it? (re @elros34: so maybe suggested l...) | 21:30 |
T42 | <abhishek_0> Is the bridge working? | 21:45 |
T42 | <elros34> @walid buidl_packages.sh -b < path-to-sources> | 21:49 |
T42 | <elros34> https://github.com/mer-hybris/tls-padding | 21:52 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!