| *** Guest53300 is now known as bshah[m] | 01:57 | |
| ohyash | I wonder if anyone is awake at this time | 06:32 |
|---|---|---|
| ohyash | how do I connect to the internet? | 06:55 |
| ohyash | while operating blind on sailfish | 06:56 |
| ohyash | over telnet | 06:56 |
| ohyash | or whatever | 06:56 |
| ohyash | really no body is there? | 07:05 |
| octonezd | Finally replaced HDD, trying to port Sailfish onto my LeEco Le 2, got error error: in `sync -c -j8 --fetch-submodules`: revision stable/cm-14.1 in LineageOS/android_device_leeco_s2 not found` after repo sync | 07:31 |
| octonezd | Tried placing cm-14.1 instead of stable/cm-14.1 in local_manifests, got same error but with revision cm-14.1 instead | 07:33 |
| octonezd | What I am doing wrong? | 07:33 |
| ohyash | octonezd, wrong timing. All moderators are probably sleeping. Try 2-3 hours later maybe | 07:35 |
| octonezd | Okay :( | 07:36 |
| octonezd | Upd: I fixed it by placing last commit long hash | 07:37 |
| octonezd | https://hastebin.com/amoluyiqet.makefile | 08:12 |
| octonezd | What I am doing wrong? | 08:12 |
| wiktorek140 | octonezd: are you sure you have downloaded correct kernel repository? | 08:15 |
| octonezd | I did. I checked, and found out I should have called folder the msm8976, instead of s2 | 08:16 |
| octonezd | So I am waiting for repo sync rn | 08:16 |
| wiktorek140 | is something changed in sailfish or ubuntu/linux mint that i cannot telent into device? | 08:37 |
| ohyash | wiktorek140, Maybe Ubuntu/Mint side error. | 08:38 |
| wiktorek140 | device is showed in lsusb, in ifconfing i have address | 08:38 |
| ohyash | I can telnet into my device in a zip that I built yesterday | 08:38 |
| wiktorek140 | i builded zip yesterday too | 08:39 |
| wiktorek140 | but... | 08:39 |
| wiktorek140 | its in hybris-13.0 so maybe this is faulty | 08:39 |
| ohyash | maybe some error in the mid way of the porting process. Maybe your build isn't able to boot kernel | 08:40 |
| ohyash | wiktorek140, Not sure of that tho. I'm on hybris-14.1 | 08:41 |
| wiktorek140 | i have gui, everything work (audio dont but its my fault) | 08:41 |
| ohyash | oh | 08:41 |
| ohyash | then I have no idea | 08:41 |
| ohyash | I got no GUI :( | 08:41 |
| ohyash | just telnet | 08:41 |
| wiktorek140 | me too, ant thats the point | 08:41 |
| wiktorek140 | what device? | 08:41 |
| ohyash | YU YUREKA(tomato) | 08:42 |
| wiktorek140 | vandor is YU? | 08:42 |
| ohyash | yes | 08:42 |
| wiktorek140 | first time to hear | 08:42 |
| ohyash | its Indian. and sells only in India | 08:42 |
| wiktorek140 | did u have dmesg and journal? | 08:43 |
| ohyash | thats why probably | 08:43 |
| ohyash | I got those | 08:43 |
| ohyash | don't know how to find exct culprits in those huge ass logs | 08:43 |
| ohyash | exact* | 08:43 |
| wiktorek140 | send it, i will look ;p | 08:43 |
| ohyash | wiktorek140, did you get GUI in your first build? | 08:44 |
| wiktorek140 | ohyash: second, missing symlink to vendor files | 08:44 |
| ohyash | dmesg output : https://pastebin.com/2Gc0K5iN | journalctl : https://pastebin.com/TTjJ3sjK | 08:44 |
| wiktorek140 | ohyash: try to add in kernel config CONFIG_AUTOFS4_FS=y | 08:47 |
| ohyash | wiktorek140, done | 08:52 |
| wiktorek140 | if you have strace command try to "strace /sbin/droid-hal-init" | 08:53 |
| ohyash | I think strace would have to be installed. I can't get my device to connect to internet :/ | 08:54 |
| wiktorek140 | you mean connect with pc or internet? | 08:55 |
| ohyash | internet. telnet works. I can't do zypper in strace because there's no internet in device | 08:56 |
| wiktorek140 | hmm | 08:56 |
| ohyash | the connecting tutorial in hadk is somewhat confusing to me | 08:57 |
| wiktorek140 | where you lost? ;p | 09:02 |
| wiktorek140 | you can have other names than wlan0 if u have wifi | 09:03 |
| ohyash | In the get connected section (chapter 9.3.3). I perform the HOST$ part as it is (with wlo1 for my wifi) but I dont know how to get usb0 IP for TARGET$ sizde | 09:05 |
| wiktorek140 | ifconfig | 09:05 |
| wiktorek140 | and it will be show under something like enp0s20u2 | 09:06 |
| ohyash | yeah, for the phone. | 09:07 |
| ohyash | and what do I put in the nameserver line? | 09:07 |
| ohyash | the broadcast ID? | 09:07 |
| wiktorek140 | i supose just let it be as is | 09:07 |
| ohyash | ohk | 09:07 |
| ohyash | cool | 09:09 |
| ohyash | it works | 09:09 |
| ohyash | I got internet | 09:10 |
| wiktorek140 | ohyash: great work :) | 09:11 |
| ohyash | so I ran strace /sbin/droid-hal-init | 09:11 |
| ohyash | it gave me "execve("/sbin/droid-hal-init", ["/sbin/droid-hal-init"], [/* 10 vars */]) = 0 | 09:12 |
| ohyash | create_module("", 0) = 0xa79 | 09:12 |
| ohyash | " | 09:12 |
| ohyash | and then Process 2681 detached | 09:12 |
| wiktorek140 | and thats it? | 09:13 |
| ohyash | yes | 09:13 |
| ohyash | nothing else showed up | 09:13 |
| wiktorek140 | and systemctl status droid-hal-init | 09:14 |
| ohyash_ | how long does this test_hwcomposer go tho? | 09:14 |
| wiktorek140 | i dont remeber | 09:15 |
| wiktorek140 | ohyash: did u add to your pattern detrius? | 09:15 |
| ohyash_ | wiktorek140, yes | 09:16 |
| ohyash_ | hw_composer isnt giving any segfault. the faq has more commands to try if I get segfault | 09:17 |
| wiktorek140 | try other | 09:19 |
| ohyash_ | other? | 09:20 |
| wiktorek140 | i mean other command from faq | 09:20 |
| wiktorek140 | try run /system/bin/surfaceflinger | 09:21 |
| ohyash_ | arent any of these programs supposed to give outputs? The test_hwcomposer was silent for a long time so I ctrl^C it | 09:23 |
| ohyash_ | now this surfaceflinger is alse silent | 09:23 |
| ohyash_ | now this surfaceflinger is also silent | 09:23 |
| wiktorek140 | did the screen blink during running it? | 09:24 |
| ohyash_ | nope | 09:24 |
| ohyash_ | all I see the vendor's logo spash | 09:25 |
| ohyash_ | constant | 09:25 |
| ohyash_ | splash* | 09:25 |
| wiktorek140 | im out of ideas what u can do more | 09:27 |
| ohyash_ | yeah its weird | 09:27 |
| ohyash_ | is the dmesg, journal normal? | 09:27 |
| ohyash_ | I mean do even working sfos give out similar output/errors/warnings | 09:28 |
| ohyash_ | Nokius_, you here? | 09:28 |
| wiktorek140 | ohyash_: try this " remove the word "exec" from the last line of /usr/bin/droid/droid-hal-startup.sh, to make this permanent add a modified copy of the file to $ANDROID_ROOT/hybris/droid-configs/sparse/usr/bin/droid/droid-hal-startup.sh (this has to be fixed properly at some point)" | 09:29 |
| ohyash_ | wiktorek140, I was gonna. But mal told me its old and fixed already. Thats why the line starts with "not needed anymore" | 09:30 |
| wiktorek140 | but always can try it. sometimes old fixes fix troubles :) | 09:31 |
| ohyash_ | wiktorek140, I can't. I dont even have the bin/ dir in my $ANDROID_ROOT/usr | 09:32 |
| wiktorek140 | but this file is in hybris | 09:33 |
| wiktorek140 | and u can try on device edit it | 09:33 |
| wiktorek140 | via vim or nano | 09:33 |
| ohyash_ | done | 09:35 |
| ohyash_ | how do I "test" it | 09:35 |
| wiktorek140 | reboot device | 09:35 |
| wiktorek140 | or force reboot by button combination (usual pover and vol-down) | 09:36 |
| ohyash_ | nope, no help | 09:37 |
| ohyash_ | still no GUI | 09:37 |
| ohyash_ | :( | 09:37 |
| wiktorek140 | still splashscreen or maybe black screen? | 09:37 |
| ohyash_ | the logo splashscreen | 09:38 |
| ohyash_ | when do the other moderators wake up. I'm not who to even ask further before quitting | 09:39 |
| wiktorek140 | at least u can just ask and maybe they will reply when wake up :) | 09:40 |
| elros | moderators? post latest logs, maybe someone will read it | 09:40 |
| ohyash_ | Oh, BTW, do I need to clear warnings from kernel_defconf? | 09:40 |
| elros | if they are important | 09:41 |
| wiktorek140 | ohyash_: some of it is better to set up | 09:41 |
| ohyash_ | I cleared errors. warnings looked to be optional but some were systemd things. Possible to be the reason? | 09:43 |
| ohyash_ | the CGGROUP warnings are not important right? | 09:52 |
| elros | they are | 09:53 |
| ohyash_ | elros, should I set them to n? | 09:54 |
| ohyash_ | that link recommends so I guess | 09:54 |
| elros | how do I know what warning you get? | 09:54 |
| ohyash_ | elros, these https://hastebin.com/repucuduwi.sql | 09:55 |
| elros | read comment under CONFIG_AUDIT, I think these CONFIG_CGROUP_MEM* were also needed | 09:59 |
| mal | wiktorek140: the exec removal was obsoleted by this https://github.com/mer-hybris/droid-hal-configs/commit/77968c5d285ba770b6ae6e4eb73634218cda8a81 | 10:01 |
| mal | ohyash_: what device is that? is it qcom or something else? | 10:03 |
| ohyash_ | mal, YU Yureka. | 10:10 |
| ohyash_ | qcom | 10:10 |
| ohyash_ | SD615 | 10:11 |
| ohyash_ | elros, alright turned all CGROUP_MEM_ to y | 10:14 |
| ohyash_ | and CONFIG_AUDIT to n | 10:15 |
| octonezd | https://hastebin.com/suhalonage.sql why sdk_assistant can't create targets? | 10:16 |
| octonezd | nevermind, I am just stupid | 10:21 |
| mal | ohyash_: those flag names depend on kernel version 3.4 had those names, some newer kernels have new flags | 10:23 |
| ohyash_ | wont mer_verify_kernel_connfig report updated flags? | 10:24 |
| mal | no | 10:24 |
| mal | because currently it doesn't check kernel version | 10:24 |
| mal | so how about you check what kernel version your device is using | 10:25 |
| ohyash_ | I just set some of the flags as one of the allowed values. | 10:25 |
| ohyash_ | shouuld I remove those lines? | 10:25 |
| mal | stop overthinking and do what I said | 10:25 |
| ohyash_ | ok | 10:25 |
| ohyash_ | so what do I do now? | 10:27 |
| mal | 13:25 < mal> so how about you check what kernel version your device is using | 10:27 |
| mal | soon I will stop helping you if this continues | 10:27 |
| octonezd | sb2 -t $VENDOR-$DEVICE-$PORT_ARCH gcc main.c -o test | 10:29 |
| octonezd | sb2: Error: Invalid target specified, aborting. | 10:29 |
| octonezd | What I am doing wrong? | 10:29 |
| ohyash_ | kernel 3.10.107 | 10:30 |
| octonezd | I have tooling named leeco-s2-armv7hl | 10:30 |
| ohyash_ | octonezd, did you do this? sdk-assistant create $VENDOR-$DEVICE-$PORT_ARCH http://releases.sailfishos.org/sdk/latest/Jolla-latest-Sailfish_SDK_Target-armv7hl.tar.bz2? | 10:32 |
| mal | ohyash_: the flags in that kernel are CONFIG_MEMCG, CONFIG_MEMCG_SWAP, CONFIG_MEMCG_KMEM | 10:35 |
| mal | ohyash_: push your adaptation repos to github with all current changes so I can check those | 10:35 |
| wiktorek140 | mal: thanks for info ;) | 10:38 |
| wiktorek140 | mal: are come specific kernel flag required for hybris-14.1? excluded this one u write there and from kernel check? | 10:45 |
| OhYash | mal, that(git) I think would take a little time. On it. | 10:46 |
| mal | wiktorek140: no special changes for 14.1 in kernel, at least I can't remember any | 10:52 |
| wiktorek140 | mal: so in theory mer config from hybris-13.0 should work with hybris-14.1 at this same version of kernel? | 10:54 |
| mal | OhYash: your dmesg from earlier today showed that you didn't run the symlink script I said to run yesterday | 10:54 |
| mal | wiktorek140: why are you doing it like that, just fix the flags in 14.1 kernel | 10:55 |
| mal | OhYash: also you didn't add the one package to patterns | 10:55 |
| wiktorek140 | im just curious | 10:55 |
| OhYash | mal: That is weird. Cuz I did both. Maybe you checked the defconf and journalctl of previous build. Was it the one where I put pastebin links for both in a single message? | 10:57 |
| mal | then you didn't build some that was needed | 10:59 |
| OhYash | yes maybe | 11:02 |
| OhYash | I have my kernel config up in git | 11:02 |
| OhYash | here : https://github.com/OhYash/android_kernel_cyanogen_msm8916/blob/hybris-14.1/arch/arm64/configs/lineageos_tomato_defconfig | 11:02 |
| OhYash | droid_hal_tomato spec : https://github.com/OhYash/droid-hal-tomato | 11:03 |
| mal | didn't you say you needed straggler_files? | 11:04 |
| elros | what is that CONFIG_MEMCG=! ? :P | 11:05 |
| OhYash | mal: yes, I had the same error as in 7.2.2 | 11:05 |
| OhYash | oh wait | 11:06 |
| OhYash | they'r emissing | 11:06 |
| mal | OhYash: did you commit your changes? | 11:07 |
| mal | elros: it's the same as CONFIG_CGROUP was in 3.4 kernel but for newer kernels | 11:08 |
| OhYash | mal : haha oops. Thats whats missing | 11:08 |
| octonezd | Is error in rpm/droid-hal-s2.spec build related to my device source code (log: https://hastebin.com/ejebuhuduc.coffeescript)? | 11:08 |
| elros | mal: yes I know I mean "!" instead "y" in his defconfig | 11:08 |
| mal | octonezd: add this to your rpm/dhd https://github.com/mer-hybris/droid-hal-device/pull/187/files | 11:09 |
| OhYash | updated with stragggler | 11:09 |
| mal | elros: well that is just wrong | 11:09 |
| mal | OhYash: where did you get the idea to do CONFIG_MEMCG=! in your defconfig, instead of =y | 11:10 |
| octonezd | Thanks, @mal! It worked | 11:10 |
| OhYash | that link in the comment suggested setting all these CGROUPS to no for performance and the allowed valuew here were y, ! | 11:11 |
| OhYash | changing | 11:11 |
| octonezd | Can I make build/packages use more CPUs, other than just one? | 11:15 |
| octonezd | Like in make -j8 | 11:15 |
| mal | OhYash: once again overthinking | 11:17 |
| mal | OhYash: there are only 3 options in kernel defconfig N, Y or M | 11:18 |
| octonezd | I am getting dependency error when building rpm/droid-config-s2.spec: No provider of pkgconfig(android-headers) found. | 11:18 |
| OhYash | mal : but... https://hastebin.com/repucuduwi.sql | 11:19 |
| OhYash | droid_hal_version_tomato : https://github.com/OhYash/droid-hal-version-tomato | 11:19 |
| mal | OhYash: you are the first person to misunderstand that, congratulations | 11:20 |
| OhYash | droid_config_tomato : https://github.com/OhYash/droid-config-tomato | 11:20 |
| OhYash | Ow shet | 11:20 |
| mal | OhYash: so did you commit the changes to config repo? doesn't seem like it | 11:21 |
| elros | OhYash: check again your defconfig, now you need to fix your fix | 11:21 |
| OhYash | I did change it to y from ! | 11:22 |
| OhYash | and its updated | 11:22 |
| OhYash | https://github.com/OhYash/android_kernel_cyanogen_msm8916/blob/hybris-14.1/arch/arm64/configs/lineageos_tomato_defconfig | 11:22 |
| mal | OhYash: you have same flag twice and one missing | 11:22 |
| OhYash | oh oh yep | 11:22 |
| mal | could you please focus a little | 11:22 |
| OhYash | I think I've done all the git upload work. Now am with calm mind | 11:24 |
| octonezd | Any idea what is causing missing dependency of pkgconfig(android-headers)? | 11:24 |
| octonezd | I checked IRC logs, found something in 2015, but it seems there were no answer. Also found issue in mail lists, but it was on-device | 11:25 |
| elros | when you get this error | 11:26 |
| octonezd | Building of rpm/droid-config-s2.spec | 11:27 |
| octonezd | https://hastebin.com/lehefixaju.coffeescript here is output | 11:27 |
| mal | OhYash: if you say config repo is updated then you didn't do what I said | 11:28 |
| mal | octonezd: did droid-hal rpms build successfully? | 11:28 |
| elros | octonezd: it was related to missing droid-hal-$DEVICE-devel | 11:29 |
| octonezd | droid-hal-s2 build successfully | 11:29 |
| OhYash | oh, pattern commit | 11:30 |
| octonezd | So I should build droid-hal-s2-devel? | 11:31 |
| OhYash | got the pattern | 11:31 |
| octonezd | I tried doing build_packages.sh --droid-hal and got this: File '/repodata/repomd.xml' not found on medium 'http://repo.merproject.org/obs/nemo:/devel:/hw:/common/sailfish_latest_i486/' | 11:34 |
| mal | octonezd: ? you are building for i486 device? | 11:38 |
| mal | octonezd: what is your PORT_ARCH | 11:38 |
| octonezd | PORT_ARCH="armv7hl" | 11:39 |
| mal | then how does that path contain sailfish_latest_i486 | 11:39 |
| mal | octonezd: which build target did you install? | 11:39 |
| octonezd | both armv7hl, I think | 11:40 |
| mal | it's amazing in how many different ways people can mess up builds | 11:40 |
| octonezd | I guess my hands are growing from wrong place then :P | 11:41 |
| octonezd | So, how do I check which arch target uses? | 11:41 |
| octonezd | Upd: checked bash history and now I feel stupid, cuz I put i486 as target... | 11:41 |
| OhYash | Today doesn't seem to be a good start of work for mal | 11:43 |
| mal | octonezd: show output of sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -R -m sdk-install ssu lr | 11:51 |
| mal | in platform sdk | 11:51 |
| mal | octonezd: oh, you already found your mistake, anyway that command I gave shows the repos inside the target | 11:52 |
| octonezd | Where did I go again: `ImportError: No module named rpm`. It happend during mic create | 12:05 |
| OhYash | mal: does the detritus affect libhybris and other plugins or should I just rebuild droid-hal* and droid-config*? | 12:07 |
| octonezd | I should just `python2 -m pip install --user rpm`? | 12:09 |
| mal | OhYash: just rebuild config package | 12:09 |
| mal | octonezd: zypper in rpm-python | 12:09 |
| mal | and sudo if needed | 12:10 |
| octonezd | Thanks! Now I got another error:Error <creator>: URLGrabber error: https://releases.jolla.com/jolla-apps/v2.2.0.29/armv7hl/repodata/repomd.xml - e.errno: 14 | 12:11 |
| mal | octonezd: RELEASE=2.2.0.29 not v2.2.0.29 | 12:12 |
| octonezd | Yeah, just noticed I did such a mistake | 12:13 |
| mal | OhYash: you are still missing the symlinks from your repo, are those there or did you just forget to commit those (you need to do git add for those before they can be commited) | 12:19 |
| OhYash | The script from faq right? done. Here : https://github.com/OhYash/droid-config-tomato/tree/master/sparse/usr/libexec/droid-hybris/system/etc/init | 12:30 |
| mal | OhYash: ok, so now try to rebuild the image and see how it behaves | 12:35 |
| octonezd | So, I managed to install it, but all I see is leeco logo | 12:44 |
| octonezd | Last thing that /init.log says is beggining inject loop | 12:46 |
| octonezd | Also CUSTOMPRODUCT in log says Failed to boot init in real rootfs | 12:47 |
| octonezd | Here is /init.log https://hastebin.com/nekizonibe.md | 12:49 |
| mal | octonezd: which telnet port? | 12:51 |
| octonezd | 23 | 12:52 |
| mal | octonezd: did you install that normally? without any hacks? | 12:53 |
| mal | octonezd: do you see the sailfish rootfs in /target folder in the telnet? | 12:55 |
| octonezd | yes, it installed fine in twrp | 12:55 |
| octonezd | Only one error appeared, however: no such command [[] | 12:55 |
| mal | where? | 12:56 |
| octonezd | In TWRP | 12:56 |
| octonezd | lemme flash it again and I will give you screenshot | 12:56 |
| mal | I don't care about twrp output | 12:57 |
| mal | octonezd: just recheck the rootfs in telnet | 12:57 |
| octonezd | Okay, waiting for it to complete flash | 12:57 |
| octonezd | Is it fine for mount errors to be in TWRP during install? | 12:58 |
| mal | it's normal to have some mount related error | 12:58 |
| mal | depends on what the error is | 12:58 |
| mal | and whether you actually defined the userdata partition correctly | 12:58 |
| octonezd | failed to mount userdata at /data | 12:58 |
| mal | if the installation claims to succeed then it should be fine | 13:01 |
| octonezd | For some reason it doesn't show device IP in dmesg/lsusb -v | 13:01 |
| mal | ifconfig is where you check that | 13:04 |
| octonezd | Also gnome settings say there is two wired usb connections | 13:05 |
| octonezd | ifconfig shows only my local IP in that interface, it seems, and that ip is 192.168.2.1 | 13:05 |
| mal | stop overthinking, you already got in to it via telnet so | 13:06 |
| octonezd | But I can no longer get into telnet! | 13:06 |
| octonezd | That is the problem | 13:06 |
| mal | then what did you do differently now? | 13:06 |
| octonezd | Nothing, just rebooted it. | 13:06 |
| mal | dmesg should so the boot state assuming it's even in sailfish side | 13:07 |
| octonezd | If I do sudo lsusb -v | grep iSerial I see my phone serial number | 13:10 |
| octonezd | I guessed telnet IP and port. It was 192.168.2.15 and port was 2323. | 13:11 |
| mal | so you go to that now? | 13:11 |
| octonezd | folder /target doesn't exist | 13:11 |
| octonezd | ls: cannot access /target: No such file or directory | 13:12 |
| mal | you are confusing me | 13:12 |
| octonezd | Yes, I got into phone's telnet | 13:12 |
| mal | which telnet port are you using now? | 13:12 |
| octonezd | 2323 | 13:12 |
| mal | the /target is relevant only for 23 | 13:12 |
| mal | 2323 is actual sailfish side, 23 is the kernel init | 13:12 |
| octonezd | what should I check now? | 13:13 |
| mal | logs | 13:13 |
| mal | dmesg, journalctl, /usr/libexec/droid-hybris/system/bin/logcat as usual | 13:13 |
| octonezd | https://hastebin.com/uhugesikuv.md - dmesg, https://hastebin.com/uyiquguheg.cs - journalctl, logcat read failure | 13:15 |
| mal | which android base? | 13:16 |
| octonezd | lineage-14.1 | 13:16 |
| octonezd | cm-14.1* | 13:16 |
| mal | I assume you didn't run the symlink script mentioned in faq (linked in topic) in section about 14.1 porting? | 13:16 |
| octonezd | No | 13:17 |
| mal | that is needed for 14.1 | 13:17 |
| octonezd | Oh, ok. | 13:17 |
| mal | from the faq: Run this script in $ANDROID_ROOT http://paste.opensuse.org/40869869 | 13:18 |
| octonezd | Uh, the script is N/A, it seems. | 13:18 |
| octonezd | Disallowed Key Characters | 13:18 |
| mal | ? | 13:18 |
| mal | that link works fine for me | 13:18 |
| octonezd | Seems like firefox screwing something up then | 13:19 |
| mal | I also use firefox | 13:19 |
| mal | octonezd: I assume you downloaded the script from there first | 13:19 |
| octonezd | I didn't download that script | 13:20 |
| mal | then how did you try to run it? | 13:20 |
| octonezd | I were not able to open the link | 13:21 |
| octonezd | Guess some extension screwed it up, cause incognito opened it just fine | 13:22 |
| octonezd | So, I got the script and ran it, should I started build again? | 13:22 |
| mal | build config rpms and then image | 13:24 |
| mal | or optionally you can create symlinks on your device for testing (obviously you need to edit the script for that) | 13:25 |
| octonezd | Okay,I did build_packages.sh -c Waiting now for mic to create stuff | 13:27 |
| octonezd | Alright, screen works now | 13:45 |
| mal | so you got to UI now? | 13:46 |
| mal | does touchscreen work? | 13:46 |
| octonezd | yep, touch works | 13:47 |
| mal | ok good | 13:47 |
| octonezd | Mobile data seems to work, WiFi seems to not | 13:47 |
| octonezd | In setup, at least. | 13:47 |
| mal | many qcom devices need wifi driver as module, check faq | 13:48 |
| mal | octonezd: I suggest you go through chapters 13.2 and 13.3 soon (make sure you have the latest hadk pdf 2.2.0.1) | 13:52 |
| octonezd | Does sailfish os have support for h/w keys? | 14:21 |
| octonezd | Or it is completely gesture-driven | 14:21 |
| mal | octonezd: gestures mostly, the normal three android buttons at the bottom of the screen are not used | 14:25 |
| octonezd | And what about fingerprint support? | 14:33 |
| mal | not yet available for ports afaik | 14:33 |
| octonezd | Wlan doesn't work even after installing service... | 14:38 |
| octonezd | systemctl status wlan-module-load says that it is active (exited) | 14:38 |
| mal | then check logs | 14:39 |
| mal | dmesg for example | 14:39 |
| mal | is the module name correct? | 14:40 |
| mal | test in commandline | 14:40 |
| octonezd | https://hastebin.com/oxiyedawev.sql | 14:40 |
| octonezd | modprobe wlan outputs nothing | 14:41 |
| mal | what does lsmod say? | 14:41 |
| octonezd | it is empty | 14:42 |
| mal | try modprobe pronto_wlan | 14:42 |
| octonezd | There is no wlan module in lib/modules/3.10.versionstuff/ | 14:42 |
| mal | are modules even enabled for your kernel? | 14:42 |
| mal | check in defconfig | 14:43 |
| octonezd | Yes, PRONTO_WIFI is set to m | 14:43 |
| mal | octonezd: I mean CONFIG_MODULES or something like that | 14:43 |
| mal | some kernels have modules disabled completely | 14:43 |
| octonezd | CONFIG_MODULES=y | 14:43 |
| mal | octonezd: show your kernel sources | 14:44 |
| octonezd | https://github.com/OctoNezd/android_kernel_leeco_msm8976 | 14:44 |
| octonezd | I didn't commited & pushed the defconfig with wifi | 14:44 |
| mal | so why isn't the module there? did you rebuild kernel using make hybris-boot in HABUILD_SDK and then repackage droid-hal rpms using build_packages.sh -d? | 14:46 |
| mal | (or make hybris-hal) | 14:46 |
| octonezd | I didn't rebuild kernel in HABUILD, only repackaged. Will rebuild | 14:47 |
| mal | well there is the problem, of course kernel needs to be rebuilt if you change configs | 14:47 |
| mal | *change defconfig | 14:47 |
| octonezd | I thought build_packages will also rebuild kernel | 14:49 |
| mal | nope, all android side build is done in HABUILD_SDK, including kernel | 14:49 |
| mal | octonezd: so if you edit something outside hybris or rpm folders in $ANDROID_ROOT you need to rebuild something in HABUILD_SDK | 14:51 |
| mal | in most cases only device repos or kernel need changing in android side, some exceptions are known though | 14:52 |
| *** Keij0_ is now known as Keij0 | 15:51 | |
| *** Keij0_ is now known as Keij0 | 16:30 | |
| ohyash_ | hey | 16:55 |
| ohyash_ | mal : I reflashed with all the code changes you mentioned | 16:56 |
| ohyash_ | now I get a blue screen after vendor splash | 16:56 |
| ohyash_ | full blue screen of death | 16:56 |
| ohyash_ | and I can't telnet in | 17:01 |
| mal | ohyash_: which device was that again? | 17:05 |
| ohyash_ | YU YUreka | 17:05 |
| ohyash_ | tomato | 17:05 |
| mal | ohyash_: does it show up in dmesg? | 17:07 |
| mal | or ifconfig | 17:07 |
| ohyash_ | dmesg :: enp0s19f2u1: link is not ready | 17:08 |
| ohyash_ | and doesnt show up in ifconfig | 17:08 |
| mal | ohyash_: but did it show before that the usual Mer messages | 17:09 |
| ohyash_ | after | 17:10 |
| mal | what? | 17:10 |
| ohyash_ | I think the first time I connected it, it probably detected it | 17:10 |
| ohyash_ | SerialNumber: Mer Debug setting up (DONE_SWITCH=yes) | 17:10 |
| ohyash_ | this is only when I connected it for the first time | 17:10 |
| mal | I'm lost again | 17:10 |
| ohyash_ | then I restarted it and reconnected I couldn't telnet | 17:10 |
| mal | I have no idea what you are trying to tell | 17:11 |
| ohyash_ | Uh | 17:11 |
| ohyash_ | I flashed the zip | 17:11 |
| ohyash_ | It booted into blue screen | 17:11 |
| ohyash_ | then I took it off | 17:11 |
| ohyash_ | reconnected and started, again it gave blue screen | 17:12 |
| ohyash_ | that DONE_SWITCH=yes only showed up on the first connect | 17:12 |
| ohyash_ | not afterwatrds again | 17:12 |
| mal | so the second boot shows no sign of Sailfish? | 17:12 |
| ohyash_ | Yep, not even Mer | 17:12 |
| ohyash_ | message | 17:13 |
| mal | so nothing about usb? | 17:13 |
| mal | in dmesg | 17:13 |
| ohyash_ | Detects device as a Storage device. | 17:13 |
| ohyash_ | but Make and model number are there | 17:13 |
| ohyash_ | I'm going to reflash it now | 17:14 |
| ohyash_ | and am going to telnet in the first boot | 17:14 |
| ohyash_ | interesting. Now my laptop is not detecting phone even in the fastboot mode | 17:16 |
| ohyash_ | [39892.392428] usb 2-2: device descriptor read/64, error -110 | 17:16 |
| ohyash_ | more of scary then interesting now :o | 17:17 |
| ohyash_ | than* | 17:17 |
| ohyash_ | woof | 17:18 |
| ohyash_ | I got telnet | 17:18 |
| ohyash_ | weird | 17:18 |
| mal | very odd indeed | 17:18 |
| ohyash_ | not able to fastboot now tho | 17:18 |
| ohyash_ | I got telnet. What can I do | 17:19 |
| ohyash_ | dmesg https://hastebin.com/uyezabijig.pl | 17:19 |
| mal | EGL_PLATFORM=hwcomposer test_hwcomposer | 17:21 |
| ohyash_ | journalctl : https://hastebin.com/jasamugitu.bash | 17:23 |
| ohyash_ | test_hwcomposer finally gives some output | 17:24 |
| ohyash_ | width: 1080 height: 1920 | 17:24 |
| ohyash_ | OpenGL ES 2.0 (OpenGL ES 3.2 V@145.0 (GIT@I38fdb1353f)) | 17:24 |
| ohyash_ | and is running | 17:24 |
| ohyash_ | mal: still have bluescreen | 17:24 |
| mal | ohyash_: I think I might have an idea, check end of faq | 17:32 |
| mal | ohyash_: so add the lines similar way you added straggler_files | 17:33 |
| ohyash_ | the one just above anbox info? | 17:34 |
| mal | below it | 17:35 |
| mal | I just added some new instructions | 17:35 |
| ohyash_ | just noticed it was offline and reloaded. Now I see it | 17:36 |
| mal | ohyash_: so the problem might be that your device has that certain define in device/cyanogen not in device/yu | 17:36 |
| mal | so the automatic detection failed | 17:36 |
| ohyash_ | mal: so after this should I proceed with building zip or are there any more possible changes? | 17:38 |
| mal | ohyash_: not 100% sure which packages should be rebuilt so better run build_packages.sh completely after adding the new lines | 17:38 |
| ohyash_ | ok | 17:39 |
| mal | hopefully it will detect the changed header | 17:39 |
| ohyash_ | I just hope it detects fastboot | 18:00 |
| ohyash_ | -- still at build_packages tho | 18:00 |
| mal | just curious where do you need fastboot? don't you install the zip via recovery? | 18:02 |
| ohyash_ | I never flashed recovery. I just fastboot boot TWRP-blahblah.img whenever I flash a rom | 18:04 |
| mal | ok | 18:05 |
| ohyash_ | meanwhile I was curious about this NameTelegram users here | 18:05 |
| ohyash_ | whats the telegram group? | 18:05 |
| ohyash_ | funny thing is I see mine too :"YashYadavTelegra". Maybe I have fiddled with matrix in the past | 18:06 |
| mal | no idea, I have never used telegram | 18:06 |
| ohyash_ | ok | 18:06 |
| ohyash_ | flashing zip build-4 | 18:42 |
| mal | did fastboot work now? | 18:43 |
| branek | hi all... any idea why I'm getting different MAC address on wlan0 each time I reboot Sailfish? | 18:45 |
| r0kk3rz | probably because you're not reading the real one in | 18:46 |
| branek | And I need to set WLAN password again after each reset... | 18:46 |
| branek | r0kk3rz: how/where to check. Device dumpling, Lineage 15.1... | 18:46 |
| ohyash_ | OH MY F! | 18:47 |
| ohyash_ | It booted! | 18:47 |
| r0kk3rz | yeah wlan settings are related to mac address | 18:47 |
| ohyash_ | Ahoy!! | 18:47 |
| ohyash_ | I can finally see!! | 18:48 |
| mal | magic | 18:48 |
| ohyash_ | wifi isnt working | 18:48 |
| ohyash_ | mal: To me, yes! xD | 18:49 |
| mal | ohyash_: read what I wrote here earlier today about building wifi as module | 18:49 |
| mal | branek: do you have built wlan driver as module? | 18:49 |
| mal | branek: which wlan chip, qcom or broadcom? | 18:50 |
| ohyash_ | alright. soon after I'm done with startup steps | 18:50 |
| mal | ohyash_: a hint, you can skip tutorial by pressing display corners in clockwise order starting from top left corner | 18:50 |
| branek | mal: not sure about chip neither if wlan are built as modules, wlan worked immediately after boot-up... | 18:52 |
| branek | mal: will check now... | 18:52 |
| mal | branek: broadcom chips often need some magic mac address service and qcom chips behave better as modules usually | 18:53 |
| ohyash_ | how do I screenshot tho? | 18:53 |
| mal | ohyash_: I think it was both volume buttons at the same time for a moment | 18:54 |
| branek | mal: This was set in kernel: Qualcomm core WLAN driver for QCA6174 chipset - so I guess it's qcom | 18:56 |
| branek | mal: any special handling of modules, or I just enable it as module, and everything should be on right place after build? | 18:57 |
| mal | branek: check faq | 18:57 |
| mal | branek: show your kernel sources | 18:58 |
| branek | mal: https://github.com/branek/android_kernel_oneplus_msm8998/tree/hybris-15.1 | 18:59 |
| mal | branek: hmm, I know a device with same chip and it didn't need wlan as module, not any other hacks, just a simple init service to start the wlan chip | 19:03 |
| mal | at least same driver I mean | 19:03 |
| ohyash_ | So far, wifi, bluetooth, camera dont work. Only single sim is detected. Gallery is not showing images in full screen mode | 19:04 |
| branek | mal: to be honest, I also think this was not the case before... I just rebuilt everything after adding ADRENO_QUIRKS. | 19:05 |
| branek | mal: butnot sure 100%. | 19:05 |
| ohyash_ | but I think I'll tackle them all slowly. Later. Gotta sleep | 19:05 |
| ohyash_ | Thank you all who have helped me. | 19:05 |
| ohyash_ | and mal, I love you | 19:05 |
| louisdk | I was wondering if the bq aquaris m10 was easier to port due to it's ubuntu toch libhybris implementation? | 19:06 |
| mal | ohyash_: for black pictures you might need https://github.com/mer-hybris/droid-config-sony-nile/commit/ff891dd4a0b35671956647f5c1cbd98529394fd4 | 19:06 |
| ohyash_ | browser crashed | 19:07 |
| ohyash_ | and doesnt show chars as I type | 19:07 |
| ohyash_ | :/ | 19:07 |
| ohyash_ | any other way to test internet? | 19:07 |
| mal | ohyash_: for browser crash you probably need WANT_ADRENO_QUIRKS | 19:07 |
| branek | mal: do you think that not wiping data/cache before flashing could be a problem? I just want to switch kernel on boot partition to be able to get android back on device easeily. | 19:08 |
| branek | *easily | 19:08 |
| mal | ohyash_: so to the list where you added QCOM_BSP just a moment ago add "#define WANT_ADRENO_QUIRKS 1\" | 19:08 |
| ohyash_ | mal: Easy, Done :) | 19:09 |
| ohyash_ | mobile internet seems to be working. curl | 19:10 |
| mal | ohyash_: you seem to have wlan as module already so have you added the service as mentioned in faq | 19:10 |
| mal | ohyash_: for camera and phone calls etc do chapters 13.2 adn 13.3 from hadk pdf, make sure you have hadk pdf version 2.2.0.1 | 19:12 |
| mal | branek: cache or data shouldn't be a problem | 19:12 |
| mal | branek: at least not if you boot to same android version you used as base for sailfish | 19:13 |
| ohyash_ | mal: should I set all the configs with WLAN in name to m? | 19:16 |
| mal | ohyash_: no, you already have the needed flags as module https://github.com/OhYash/android_kernel_cyanogen_msm8916/blob/hybris-14.1/arch/arm64/configs/lineageos_tomato_defconfig#L551 | 19:16 |
| mal | ohyash_: you only need to add the service to configs | 19:16 |
| ohyash_ | Doesn't detect incoming calls. Also doesnt detect when the person on the other side picks up a call. | 19:29 |
| mal | ohyash_: I told you already how to fix calls | 19:34 |
| branek | mal: wiping data/cache and reflashing fixed issue with different MAC... maybe it's related with PIN set-up on ANDROID/encription, whatever... | 19:53 |
| sailfishos-porte | I've flashed Sailfish on my device. And it's not starts. I saw only MI logo. I've got a log of Sailfish. This error is repeating every 30 seconds https://pastebin.com/m9N9rHh0 | 19:56 |
| mal | branek: hmm, ok | 19:58 |
| mal | branek: it would be interesting to know which one was required | 19:59 |
| branek | mal: ok, one of next reflasing i will try to figure out... | 20:00 |
| mal | sailfishos-porte: do not just take random parts of logs, get the whole logs, dmesg, journalctl and /usr/libexec/droid-hybris/system/bin/logcat also which device and android base? | 20:00 |
| mal | branek: wondering if it's some permission issue? | 20:00 |
| branek | mal: interesting is that first part of MAC was changed to C0:EE:FB when it was failing... now it's OUI, 94:65:2D - both are OUI from OnePlus... | 20:04 |
| r0kk3rz[m] | Whats the mac supposed to be? | 20:09 |
| branek | r0kk3rz: well, I suppose this one which is working: 94:65:2D, but will check what is in android when I restore it... | 20:14 |
| branek | ok, next thing is cellular: device is dual modem, but I didn't set it as dual in droid-configs yet... | 20:18 |
| branek | rild is started | 20:18 |
| branek | two channels, and ofonod is in process list... | 20:18 |
| branek | this is also output of logcat -b radio: https://pastebin.com/xAtAadDq | 20:18 |
| branek | it's complaining about /dev/socket/settingsd | 20:20 |
| mal | branek: assuming you have 2.2.0.29 sailfish running then ofono supports multple config files, you can enable second sim slot like this https://github.com/mer-hybris/droid-config-f5121/blob/master/sparse/etc/ofono/ril_subscription.d/f5122.conf | 20:27 |
| mal | branek: of course you also need to enable the correct packages in patterns | 20:27 |
| branek | mal: do you have anything to add here: https://github.com/branek/droid-config-dumpling/blob/hybris-15.1/patterns/jolla-configuration-dumpling.yaml | 20:31 |
| mal | branek: depends on the device, the multisim package and the mapplaucher at the end | 20:32 |
| mal | branek: with multisim package you should I the config I linked | 20:33 |
| branek | mal: ok, I understand, will do that now... anything about logcat output I copied above? | 20:35 |
| mal | branek: does the first sim work? | 20:38 |
| branek | mal: nope... | 20:38 |
| mal | branek: btw, depending on how the ril is in that device it might not work yet in hybris-15.1 | 20:39 |
| mal | branek: we don't have support for the new android side way yet | 20:39 |
| *** Nokius_ is now known as Nokius | 20:39 | |
| branek | mal: got it, I was scared about that... | 20:40 |
| mal | branek: I think all devices have moved to the new communication method | 20:40 |
| branek | mal: what that means? I'll have to wait until support is done properly? | 20:43 |
| mal | branek: yep, android moved to binderized communication with ril and that is still being worked on, parts of it have already been done, not sure how much is missing | 20:44 |
| mal | branek: does ofono say anything? | 20:46 |
| monich | mal: in a few weeks ofono should get properly binderized | 20:49 |
| branek | mal: output of journalctl | grep ofono https://pastebin.com/dnAZSm19 | 20:50 |
| mal | monich: good to know, I saw the early work already being added to some places | 20:50 |
| mal | branek: ok, so you just need to wait | 20:50 |
| monich | most likely binderization will eventually involve a) installing (pulling in) ofono plugin and b) modifying ofono config | 20:50 |
| monich | so that binder dependency doesn't get pulled in on the devices that don't need it | 20:51 |
| mal | sounds a like good idea | 20:52 |
| * branek is waiting patiently | 20:52 | |
| mal | branek: meanwhile you can try to fix other other issues you might have | 20:52 |
| branek | mal: of course... next are sensors... I'm just scared not to get the same answer... :) | 20:54 |
| branek | mal: but, I will continue that tomorrow... | 20:55 |
| mal | branek: there are two options for sensors, do you have 32-bit sensors.something.so | 20:55 |
| mal | if you only have 64-bit then there could be a problem | 20:55 |
| branek | mal: no | 20:55 |
| branek | no 32 bit... | 20:55 |
| mal | then there is an ugly hack I made which was never used but it might work, or not | 20:56 |
| mal | I haven't tested if that would work with 64-bit stuff at all | 20:57 |
| monich | binderized 32-bit ofono should be able to support both 32- and 64-bit androids! | 20:58 |
| branek | mal: could you please take a look here: https://pastebin.com/FwKULs7w | 20:58 |
| mal | monich: we were already discussing other things, sensors | 20:58 |
| branek | there is: ./system/vendor/lib/sensors.ssc.so | 20:59 |
| mal | monich: which reminds me, other things that have been binderized is gps support in android 8, I made one way to fix it but wondering if there is simpler way like that for ofono | 20:59 |
| monich | yes, I noticed I was just praising myself :) | 20:59 |
| monich | mal: yes, if it works nicely with ofono, a similar approach can be applied to other components | 21:01 |
| monich | the low level binder stuff is a separate library, no ofono dependencies there | 21:01 |
| r0kk3rz[m] | Ofono needs binder now :/ | 21:02 |
| monich | I know! I'm working on that! | 21:02 |
| r0kk3rz[m] | I mean thats lame... | 21:03 |
| r0kk3rz[m] | But its amazing its worked for so long | 21:04 |
| pseudodev | I keep on getting this issue after the packages are installed duing mic | 21:19 |
| pseudodev | qemu: uncaught target signal 11 (Segmentation fault) - core dumped | 21:19 |
| pseudodev | /bin/find: tracker terminated by signal 11 | 21:19 |
| pseudodev | It keeps on saying this and it is been 30 mins now. Any ideas why? | 21:20 |
| mal | branek: have you checked with strace which sensor lib it tries to load | 21:44 |
| wiktorek140 | just strange, kernel in theory launch but it doesnt start telnet :/ | 21:46 |
| wiktorek140 | mal: possible that fix in hybris-boot will fix it in hybris-14.1? | 22:01 |
| mal | wiktorek140: what fix? depends on what the problem is, you can try, some devices have issues with partial configfs support and full android gadget support at the same time | 22:04 |
| wiktorek140 | the main and for now only one problem ;p kernel boot, but serial via usb dont ;p | 22:05 |
| wiktorek140 | mal: on hybris-13.0 work everything correct, excluding that i cannot reboot device | 22:06 |
| mal | wiktorek140: try the latest hybris-boot and see what happens | 22:06 |
| wiktorek140 | mal: i building right now, and i will test it | 22:08 |
| wiktorek140 | mal: it fix it, stuck before rootfs but something can be done now ;) | 22:13 |
| mal | wiktorek140: good | 22:15 |
| wiktorek140 | mal: better than nothing but now i need to find out what stop int from booting | 22:16 |
| mal | wiktorek140: did you run the symlink script from faq? | 22:17 |
| mal | wiktorek140: oh, you said it's before it gets to sailfish side | 22:17 |
| wiktorek140 | yes, i will check it exist on device | 22:17 |
| mal | so you are now in port 23? | 22:17 |
| wiktorek140 | yes | 22:17 |
| wiktorek140 | not 2323 | 22:18 |
| mal | the errors are usually either incorrect defconfig or something wrong with fixup or something like that | 22:18 |
| wiktorek140 | missing mountpoints can cause that too? | 22:19 |
| mal | I think so, it can fail to mount userdata | 22:20 |
| wiktorek140 | mal: i supose it was my fault because i change by accident fixup-mountpoint and it didnt mount anything xD | 22:21 |
| mal | not usually missing fixup but broken fixup, meaning incorrect path for the source | 22:21 |
| mal | the error should be quite clear in port 23 log | 22:22 |
| wiktorek140 | mal: i will test it with fixed mountpoints | 22:24 |
| wiktorek140 | mal: its got better, device only blink in system and disapear | 22:28 |
| mal | wiktorek140: what did you rebuild? do remember that fixups are used also in sailfish side (droid-hal) | 22:30 |
| mal | so just flashing kernel is not enough | 22:30 |
| wiktorek140 | mal: thats can be the case. only boot | 22:31 |
| wiktorek140 | mal: need to rebuild libhybris and other? | 22:33 |
| mal | just droid-hal and then the final image | 22:34 |
| mal | so build_packages.sh -d and image build | 22:34 |
| wiktorek140 | building final image, it will take a moment | 22:36 |
| branek | mal: I created link ln -s /system/vendor/lib/sensors.ssc.so /odm/hw/lib/sensors.qcom.so and now test-sensors returns with around 50 sensor list... so I guess that is ok. | 22:42 |
| branek | mal: tried to follow HADK, and results from IRC search, and now I am little bit lost, which file to change... because CSD does not work for any sensor... | 22:43 |
| wiktorek140 | mal: fixed mount points = boot to gui in first try ;p | 23:04 |
| mal | branek: does harbour-messwerk show any working sensors? | 23:04 |
| mal | wiktorek140: good | 23:04 |
| wiktorek140 | so it need to be polished but main part work ;) | 23:07 |
| branek | mal: rotation - working, compas (magnetometer) - working | 23:11 |
| branek | other not | 23:11 |
| branek | *others | 23:11 |
| mal | branek: ah, have you configured csd? | 23:12 |
| mal | branek: something like this https://github.com/mer-hybris/droid-config-sony-nile/blob/master/sparse/usr/share/csd/settings.d/hw-settings.ini | 23:12 |
| branek | mal: this: https://github.com/mlehtima/droid-config-fp2-sibon/blob/master/sparse/usr/share/csd/settings.d/hw-settings.ini | 23:12 |
| mal | yep | 23:12 |
| mal | that is needed also for many sensors to work | 23:13 |
| branek | no.... will try now... | 23:13 |
| mal | if that is not enough you can then try something like this in addition to the csd config https://github.com/piggz/droid-config-mido/blob/master/sparse/etc/sensorfw/sensord.conf.d/60-sensor-selection-overrides.conf | 23:17 |
| branek | mal: sensors are responsive now, with hw-settings.ini, all except compas... | 23:23 |
| branek | I will find out what is wrong with compass later.... thank you mal, lot of help from you. Good night to everyone in CET TZ. | 23:25 |
| mal | branek: sometimes compass only works on the second start of the compass app | 23:28 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!