| T42 | <mintphin> you mean evdevtouch:/dev/input/event0? | 00:08 |
|---|---|---|
| T42 | <mintphin> oh wait | 00:08 |
| T42 | <mintphin> ok got it | 00:08 |
| T42 | <edp_17> @elros34 [sure using dev/block/*/by-name or /sys/block/m …], I've checked in the device's PIT: The boot partition's identifier is 9, so the boot partition device is mmcblk0p9. What I have in /var/lib/flash-partition/device-info is: PART_REAL_1="mmcblk0p9" | 00:14 |
| T42 | <edp_17> Should it be what's in there currently or PART_REAL_1="/dev/mmcblk0p9" ? | 00:15 |
| T42 | <elros34> without dev | 00:17 |
| T42 | <elros34> check also CPUCHECK_STRING | 00:17 |
| T42 | <edp_17> So, what's in the is correct. | 00:17 |
| T42 | <elros34> oh so that is the reason for failure. Completely empty? | 00:20 |
| T42 | <edp_17> Wait, no. It was only not readable by nemo: https://paste.ubuntu.com/p/rG3pN9SRzX/ | 00:23 |
| T42 | <edp_17> In /var/lib/flash-partition/device-info I have CPUCHECK_STRING="Exynos 5433 TRELTEXX". I think it should be "Samsung EXYNOS5433" as that is the Hardware in /proc/cpuinfo. | 00:25 |
| T42 | <edp_17> Back to the bluetooth problem in the OBS. I've double checked the .spec file in droid-config-treltexx and it has Requires: bluetooth-rfkill-event-hciattach and Requires: broadcom-bluetooth-bluez5 | 00:47 |
| T42 | <edp_17> So, it should build the correct .rpms. | 00:47 |
| T42 | <edp_17> Oh, what should be the name of the packages? Currently they are: bluetooth-rfkill-event-hciattach and broadcom-bluetooth-bluez5. Should they be bluetooth-rfkill-event and broadcom-bluetooth? | 00:49 |
| T42 | <elros34> name should match spec file you want to use | 00:49 |
| T42 | <edp_17> Does it cause a problem if the name doesn't match? It still generates the .rpm with the correct name, isn't it? | 00:51 |
| T42 | <elros34> both packages have 2 spec files, you need to choose which spec file you want to use for building. On local builds you use '-s rpm/some.spec'. On obs you select it by package name so broadcom-bluetooth-bluez5 will use broadcom-bluetooth-bluez5.spec file | 00:54 |
| T42 | <edp_17> I see. Okay, probably that was the reason why BT didn't work. I'll change the packages accordingly. | 00:56 |
| T42 | <edp_17> I guess, I should create those droid-hal* packages and upload them to OBS. Or is it enough to rebult droid-config-treltexx, bluetooth-rfkill-event-hciattach and broadcom-bluetooth-bluez5 packages in OBS? | 00:58 |
| T42 | <elros34> not sure why you want to rebuild droid-config/hal | 01:00 |
| T42 | <edp_17> I've changed some stuff in sparse (e.g. I added the /etc/bluez5/bluetooth/main.conf) | 01:01 |
| T42 | <edp_17> Yeah, I think don't need to rebuild droid-hal. | 01:02 |
| T42 | <elros34> I wonder why do you use rfkill-event-hciattch.spec instead | 01:05 |
| T42 | <elros34> rfkill-event.spec which requires broadcom-bluetooth | 01:05 |
| T42 | <elros34> but if that is what you used in local repo then ignore me | 01:07 |
| T42 | <edp_17> Yes, on my local repo I needed to use broadcom-bluetooth-bluez5.spec and bluetooth-rfkill-event-hciattach.spec | 01:08 |
| T42 | <elros34> ok | 01:09 |
| T42 | <edp_17> I am backing up my working/well configured sfos4 on device, thne restore a 3.3 and once OBS build finished, re-try OTA to 3.4 then 4 | 01:09 |
| T42 | <edp_17> Hope this time everything will work. 🤞 | 01:09 |
| T42 | <edp_17> How much do you know sensors, especially Light sensor? | 01:10 |
| T42 | <elros34> I don't have deep knowledge if that is waht you ask just some basic overview how it's handled in sfos. | 01:13 |
| T42 | <edp_17> I see. I ask anyway you might have some idea. On my n7000 build, if I enable light sensor in csd config file, the system becomes very sluggish. Basically nothing works properly. On the other hand on the android base all sensors (including light) do works and doesn't cause problem. | 01:18 |
| T42 | <elros34> maybe it probes value from sensor too often liek with 0 interval, check what process consume cpu and search for errors in sensorfwd | 01:20 |
| T42 | <edp_17> Okay, I'll do that, Thanks. | 01:21 |
| T42 | <elros34> I remember something similar for xperia mini pro but it was for accelerometer and only triggered via csd test app, in normal usage there was no problem but I didn't debug it further | 01:22 |
| T42 | <edp_17> Okay, I'll try what you recommended. | 01:23 |
| T42 | <edp_17> The other thing that drives me mad is the video recording. | 01:24 |
| T42 | <edp_17> On the very first build (which didn't even boot first but was everything fixed on device) the camera works. But on new builds the video recording doesn't work. I copied over the config files from the first build but didn't help. I cannot work out why. | 01:26 |
| T42 | <elros34> diff whole image/rootfs if you still have backup :) | 01:28 |
| T42 | <edp_17> Ouch, that would be massive, no? | 01:28 |
| T42 | <elros34> yes!! | 01:29 |
| T42 | <edp_17> Okay but not now. 😊 | 01:31 |
| T42 | <elros34> meld would be more useful then pur diff | 01:31 |
| T42 | <elros34> pure* | 01:31 |
| T42 | <edp_17> I need to extract the two rootfs from the twrp backup then compare. | 01:32 |
| T42 | <edp_17> Later, I'll try. | 01:33 |
| T42 | <edp_17> Now I try to OTA to the fresh OBS build. 😃 | 01:34 |
| T42 | <edp_17> @elros34 [@edp_17 hmm I guess kernel is already packaged …], Regarding this. I've checked what's in droid-hal-$DEVICE-kernel*.rpm. There are boot-initramfs.gz and kernel-3.0.101-lineage-edp17+. | 01:50 |
| T42 | <edp_17> I guess need to use this latter in var/lib/platform-updates-flash-bootimg.sh like: /usr/sbin/flash-partition boot /boot/kernel-3.0.101-lineage-edp17+ | 01:50 |
| T42 | <edp_17> Am I right? | 01:50 |
| T42 | <elros34> I think so. | 01:51 |
| T42 | <edp_17> The only problem is that I only have hybris-boot.img in /boot. I don't have that ernel-3.0.101-lineage-edp17+ file. | 01:52 |
| T42 | <elros34> did you install kernel package? hybris-boot.img is from img-boot package | 01:53 |
| T42 | <edp_17> No, only droid-hal-n7000-img-boot and droid-hal-n7000-kernel-modules are installed. droid-hal-n7000-kernel is not. | 01:56 |
| T42 | <elros34> I think you should not install img-boot as it will flash hybris-boot.img and make your device unbootable if your device expects zImage | 01:58 |
| T42 | <edp_17> Yes, but I don't know how to not install that. And I don't know how to install droid-hal-n7000-kernel instead. | 01:59 |
| T42 | <elros34> same as always, paterns/*.inc | 01:59 |
| T42 | <edp_17> I've manually installed droid-hal-n7000-kernel and now I have the kernel-3.0.101-lineage-edp17+ in /boot | 02:00 |
| T42 | <edp_17> Okay, thanks again! I cannot try OTA on this device yet as I only built 4.0. I am just preparing it to OTA. However, I'll try on my S2 once I put sfos back on it (I only need to build 4.0 for it first). | 02:02 |
| T42 | <elros34> I already mentioned it previously but if you didn't notice it remember that onshot flash is executed only after installing img-boot so you need make some changes to spec file if you want flashed kernel automatically after ota | 02:05 |
| T42 | <edp_17> I've missed this. What spec file do I need to change? | 02:06 |
| T42 | <elros34> somethig like that but for kernel instead img-boot: https://github.com/mer-hybris/droid-hal-device/blob/master/droid-hal-device.inc#L892-L894 | 02:13 |
| T42 | <edp_17> I've double checked droid-hal-device.inc on my ports and all has this. | 02:43 |
| T42 | <edp_17> No wonder as I updated submodules in all. 😃 | 02:44 |
| T42 | <edp_17> Oh, I see how you mean now! So, the original one img-boot works for devices where hybris-boot.img can be used (loke treltexx) on others (like S2 and n7000) I need to use kernel instead of img-boot. Got it. | 02:46 |
| T42 | <edp_17> I've done the OTA. Got the same issues. Kernel is not installed, /etc/bluez5/bluetooth/main.conf is not deployed (still had the channle =3 in there) and for making the bluetooth work I manually needed to flash the .rpms. I'll re-try later. Good night everyone. | 03:14 |
| T42 | <zinstack625> How can I debug loc_launcher? Strace revealed it seems to have some permission problems, but not with what. Just restarting interrupted syscall and no paths | 05:21 |
| T42 | <zinstack625> And somehow geolocation works without it? | 05:23 |
| T42 | Rock N Roll %lastname% was added by: Rock N Roll %lastname% | 05:29 |
| T42 | <mintphin> hey when i'm compiling the droid-hal-device packages this error shows up. any help? https://paste.ubuntu.com/p/hNxNmRvb42/ | 06:06 |
| T42 | <mintphin> i tried doing what the error said but this happens ```[mint@volt hadk]$ PORT_ARCH=armv7hl rpm/dhd/helpers/build_packages.sh --droid-hal | 06:07 |
| T42 | <mintphin> * Building rpm/droid-hal-laurel_sprout.spec | 06:07 |
| T42 | <mintphin> Fatal: 'xiaomi-laurel_sprout-armv7hl' is not a known build target | 06:07 |
| T42 | <mintphin> * Check /home/mint/hadk/droid-hal-laurel_sprout.log for full log. | 06:07 |
| T42 | <mintphin> !! building of package failed``` | 06:07 |
| T42 | <zinstack625> You probably have .hadk.env in your home directory, just source it in sdk | 06:12 |
| T42 | <zinstack625> And make sure you have setup your toolings and targets, it's in hadk pdf | 06:12 |
| T42 | <mintphin> i did the first one just now and nothing | 06:12 |
| T42 | <mintphin> and i have toolings already set up | 06:12 |
| T42 | <zinstack625> sdk-assistant list pls | 06:13 |
| T42 | <mintphin> hm i now see the problem | 06:13 |
| T42 | <mintphin> ```[mint@volt hadk]$ sdk-assistant list | 06:13 |
| T42 | <mintphin> SailfishOS-4.0.1.48 | 06:13 |
| T42 | <mintphin> |-SailfishOS-4.0.1.48-aarch64 | 06:13 |
| T42 | <mintphin> `-xiaomi-laurel_sprout-aarch64``` | 06:13 |
| T42 | <zinstack625> Yep | 06:15 |
| *** _sven is now known as IoT-Maker | 06:16 | |
| *** IoT-Maker is now known as IoT_Maker | 06:16 | |
| *** IoT_Maker is now known as _sven | 06:16 | |
| T42 | <mintphin> seems to be working now! | 06:16 |
| *** _sven is now known as IoT_Maker | 06:17 | |
| T42 | <mintphin> thank you | 06:17 |
| T42 | <mintphin> aw | 06:17 |
| T42 | <mintphin> it's not :8 | 06:17 |
| T42 | <mintphin> *:( | 06:17 |
| T42 | <mintphin> ok got around that error | 06:19 |
| T42 | <mintphin> but now it's insisting me to fix some kernel configs i've already fixed | 06:19 |
| T42 | <mintphin> like some config-related stuff | 06:19 |
| T42 | <zinstack625> Your kernel prob doesn't support those options | 06:21 |
| T42 | <mintphin> that's weird, i've already tried two kernels and had the same issue | 06:21 |
| T42 | <mintphin> i'm running 4.14 | 06:21 |
| T42 | <mintphin> *using | 06:22 |
| T42 | <mintphin> cuz that's the only kernel that's compatible with my device | 06:22 |
| T42 | <zinstack625> You might want to check those of the trouble, mine has a couple of those, but the options don't seem to be necessary | 06:22 |
| T42 | <zinstack625> You can hack mer-kernel-verify-config to make it compile | 06:23 |
| T42 | <mintphin> ah ok | 06:23 |
| T42 | <zinstack625> The thing is a huge pain anyways | 06:23 |
| T42 | <mintphin> it's saying CONFIG_DUMMY is assigned to y | 06:23 |
| T42 | <mintphin> even though i've disabled it | 06:23 |
| T42 | <zinstack625> Have you recompiled your kernel after that? | 06:23 |
| T42 | <mintphin> and i've enabled `CONFIG_NETFILTER_XT_MATCH_QTAGUID` | 06:23 |
| T42 | <mintphin> yup | 06:23 |
| T42 | <mintphin> i did | 06:23 |
| T42 | <mintphin> hmmm `cp: cannot stat './out/target/product/laurel_sprout/hybris-updater-script': No such file or directory` | 06:26 |
| T42 | <zinstack625> Are you sure the config in out/target/product/jasmine_sprout/obj/KERNEL_OBJ/.config is the one you configured? | 06:27 |
| T42 | <zinstack625> Otherwise your kernel might not be configured properly | 06:27 |
| T42 | <mintphin> oh yeah it's not that's weird | 06:27 |
| T42 | <zinstack625> And are you in hadk root directory? | 06:28 |
| T42 | <mintphin> yeah i am | 06:28 |
| T42 | <mintphin> i'm gonna have to check what defconfig this is using | 06:28 |
| T42 | <zinstack625> That should be in BoardConfig.mk in your device tree | 06:28 |
| T42 | <mintphin> i modified the right file | 06:29 |
| T42 | <mintphin> `# Kernel | 06:29 |
| T42 | <mintphin> TARGET_KERNEL_CONFIG := vendor/laurel_sprout-perf_defconfig` | 06:29 |
| T42 | <mintphin> ```# Kernel | 06:29 |
| T42 | <mintphin> TARGET_KERNEL_CONFIG := vendor/laurel_sprout-perf_defconfig``` | 06:29 |
| T42 | <mintphin> wait | 06:30 |
| T42 | <zinstack625> You haven't forgotten to run breakfast, have you? | 06:30 |
| T42 | <mintphin> nope | 06:30 |
| T42 | <mintphin> the issue i think lies to the fact i haven't cleaned the out folder before building | 06:31 |
| T42 | <mintphin> is that required? | 06:31 |
| T42 | <zinstack625> Should not, but run make clean just in case | 06:31 |
| T42 | <zinstack625> In ubu-chroot ofc | 06:32 |
| T42 | <mintphin> ye ik | 06:32 |
| T42 | <elros34> @mintphin did you add %define droid_target_aarch64 1 to droid-hal spec like little outdated error suggest you? | 08:09 |
| T42 | <mintphin> yeah | 14:27 |
| T42 | <mintphin> welp now i did | 14:27 |
| T42 | <mintphin> i'm still having issues however | 14:35 |
| T42 | <mintphin> `cp: cannot stat './out/target/product/laurel_sprout/hybris-updater-script': No such file or directory` | 14:35 |
| T42 | <mintphin> yes after recloning the same error occurs | 14:35 |
| T42 | <mintphin> ok i might build this on android 9 | 15:37 |
| T42 | <mintphin> it's more documented than android 10 | 15:37 |
| T42 | <mintphin> after all I have a manifest already ready for it | 15:38 |
| T42 | <mintphin> from which i'll just modify the kernel | 15:38 |
| mal | @mintphin did you apply the patches for android 10 build? | 15:42 |
| T42 | <mintphin> yup | 15:43 |
| T42 | <mintphin> if you mean hybris-patches, yes i have | 15:44 |
| mal | when did that error happen? | 16:34 |
| T42 | <zinstack625> I think loc_launcher crashes on my device because permissions, it runs fine as root. Any way to guess what's the problem? Strace shows nothing but the permissions being amiss | 16:52 |
| T42 | <elros34> maybe strace -f? It would be worth to check groups defined in service and logcat/dmesg | 17:10 |
| T42 | <mintphin> when running breakfast i'm getting these errors: `build/make/core/product_config.mk:234: error: Can not locate config makefile for product "lineage_laurel_sprout".` | 17:49 |
| T42 | <mintphin> any help? | 17:49 |
| T42 | <zinstack625> @elros34 tried that, perhaps it doesn't fork. Crashes like 0.2 sec after running "su -s /bin/bash -c "strace -f -o somefilename /system/vendor/bin/loc_launcher" gps" | 17:54 |
| T42 | <zinstack625> And I'm pretty sure it should run as user gps | 17:55 |
| T42 | <zinstack625> @mintphin typically the product name is defined in device/vendor/code/Android.mk, if I'm not wrong. It's either there or being pulled from there | 17:57 |
| T42 | <elros34> @zinstack625 are you sure it started as gps? Looking at some random rc files, it should be started as root and then change his uid to gps | 18:01 |
| T42 | <zinstack625> I believe devices are switching to running it as root, as a lot of trees suggest, mine's no exception | 18:07 |
| T42 | <zinstack625> If 97th line there is what I think it is, I'm correct: https://github.com/sailfish-motog7/android_device_motorola_sdm632-common/blob/hybris-16.0/config.fs#L97 | 18:08 |
| T42 | <zinstack625> If I could find that pesky rc file, I could make it run as root, but wouldn't that be a security risk? | 18:10 |
| T42 | <zinstack625> Running a binary blob, that could connect to internet in a privileged state | 18:11 |
| T42 | <Vena_92> I'm still surprised on how I made so much profit after seeing many people complain of being scammed, this is just amazing, am still surprised. thank you so much Mr Steven helms https://t.me/joinchat/AAAAAEwM6K8WD9aZrtSSiw | 18:27 |
| Thaodan | @admin ^ | 18:30 |
| T42 | <zinstack625> Telegram is overflowing with those, you gotta have a decent antibot to avoid those | 18:42 |
| T42 | <zinstack625> Rejext telegram, return to irc | 18:42 |
| Thaodan | zinkstack625: Well shieldy should handle those | 18:50 |
| T42 | Thaodan was added by: Thaodan | 18:51 |
| Thaodan | It is not invited | 18:51 |
| T42 | Thaodan was added by: Thaodan | 18:52 |
| Thaodan | Who is the owner of the group? | 18:53 |
| T42 | <zinstack625> Theodan: geolocation works without it, I just nuked it out of the rc file and it's fine | 18:57 |
| T42 | <zinstack625> Guess that solves it | 18:57 |
| T42 | <mintphin> i'm watching a video on aurora os and i don't know what the narrator is saying lol | 19:34 |
| T42 | <mintphin> it's on russian | 19:34 |
| T42 | <mintphin> *in | 19:34 |
| T42 | <zinstack625> @mintphin I'm russian, can rely on me | 19:35 |
| T42 | <zinstack625> Well that sounded weirder than I expected | 19:36 |
| T42 | <mintphin> only thing i got from the video is the privacy part | 19:36 |
| T42 | <mintphin> https://www.youtube.com/watch?v=o0ZMTCuou9U | 19:36 |
| T42 | <mintphin> i don't get the instagram part | 19:36 |
| T42 | <edp_17> I don't speak Russian, so don't understand any parts of it. | 19:40 |
| T42 | <mintphin> yeah me neither lol | 19:42 |
| T42 | <zinstack625> Just "busy busyness" propaganda, they have to get some market anyway | 19:45 |
| T42 | <zinstack625> But now to think about it, the way they phrase it, nobody using aurora os would be interested to expand sailfish os ecosystem, that's sad | 19:46 |
| T42 | <zinstack625> anyway, gmp-droid seems to be broken, anyone want to take a look? | 19:52 |
| T42 | <zinstack625> Jounal: https://pastebin.com/rP7ZvH5q, logcat: https://pastebin.com/3tHPjqfQ | 20:00 |
| mal | @zinstack62 does gst-droid work correctly in gallery for example? | 20:36 |
| T42 | <zinstack625> mal: seems like it | 20:54 |
| mal | which gmp-droid version do you use? | 21:00 |
| T42 | <zinstack625> mal: 0.1.11-1 | 21:03 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!