T42 | <rodrisola> @elros34 its ok? | 03:55 |
---|---|---|
T42 | <rodrisola> https://www.toptal.com/developers/hastebin/upixarefim.apache | 03:56 |
Thaodan | rodrisola: Just use https://github.com/mer-hybris/hybris-initrd | 03:57 |
T42 | <rodrisola> hello Thaodan,, | 03:58 |
T42 | <rodrisola> I have to add something or modify something, I just build it as make j4 hybris-hal | 03:59 |
Thaodan | rodrisola: You can build the boot.img like this, its easier and cleaner if you modify the ramdisk from hybris-boot. Just adjust a spec like this for your device: https://github.com/mer-hybris/droid-hal-img-boot-sony-seine | 04:08 |
T42 | <rodrisola> thaodan: this clone in the folder hybris; is rigth? | 04:10 |
T42 | <rodrisola> ok thaodan | 04:12 |
Mister_Magister | Thaodan: i created PR for the spiiroin's branch to hopefully give it some traction https://github.com/sailfishos/mce/pull/7 | 04:30 |
Thaodan | Mister_Magister: sounds good, from what I understood it requires some work, if you put that it can be merged I think. | 04:32 |
Mister_Magister | Thaodan: what do you mean? | 04:34 |
Thaodan | Mister_Magister: the PR requires some work to merge it from what I understand | 04:36 |
Mister_Magister | yeah i can imagine that | 04:37 |
Mister_Magister | but its so much out of my scope | 04:37 |
T42 | <rodrisola> thaodan: Excuse my ignorance, I am new to building the sailfish rom, but I am trying to learn, I added droid-hal-img-boot to the hybris folder, and rebuilt the boot, with make j4 hybris-hal, hybris-boot, I am doing something wrong? | 05:00 |
Thaodan | @rodrisola: You have to build that package with ./rpm/dhd/helpers/build_package.sh. | 05:07 |
Thaodan | Its similar how described here for the Xperia 10 II | 05:08 |
Thaodan | https://docs.sailfishos.org/Develop/HW_Adaptation/Sailfish_X_Xperia_Android_10_Build_and_Flash/ | 05:08 |
T42 | <elros34> @rodrisola if it match original lineage boot.img then its ok. After all you just want to repack abootimg just to be sure you do it correctly, don't you? | 11:24 |
T42 | <rodrisola> yes | 15:15 |
T42 | <rodrisola> @elros34 ,repackage and normal power on lineage | 15:16 |
T42 | <elros34> so at least you have something working to compare | 15:19 |
T42 | <rodrisola> yes, but I have also compared two boot images, and with magiks boot, and it has a bigger initramfs; I imagine it is because of the super user; In other words, I imagine that the boot starts the initramfs, and I find out that my initramfs is damaged or does not start, because I can deduce that the magiks-boot starts its initramfs normally. | 15:24 |
T42 | <elros34> you can easily unpack initramfs also, hybris one is very simple and powerful: just busybox + script: https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L338 | 15:28 |
T42 | <elros34> just idea: when your device reboots: hold recovery button and so it reboots straight to it then you could try to get something from /proc/last_kmsg or /sys/fs/pstore/console-ramoops | 15:32 |
T42 | <rodrisola> look | 15:37 |
T42 | <rodrisola> https://www.toptal.com/developers/hastebin/ifozupumut.yaml | 15:37 |
T42 | <elros34> at what? There is nothing strange there: IIRC main script should be init file and somewhere there ther eis also busybox binary | 15:39 |
T42 | <rodrisola> @ elros34, where do I execute those commands, I don't understand on the host, why doesn't the phone restart, it stays on, and if I restart it, I enter twrp, or do I execute there? | 15:43 |
T42 | <rodrisola> if I try another initramfs, from another device, just to discard, would it work? | 15:44 |
T42 | <elros34> everybody use same, with one line change (data partition) so this will not help. If you ask about /proc/last_kmsg then you execute it in twrp (like cat /proc/last_kmsg) after you manage to reboot to it. | 15:46 |
T42 | <rodrisola> if i add this as init=/init cmdline | 15:48 |
T42 | <elros34> I would expect this is default value but nothing bad will happen if you add it | 15:50 |
T42 | <rodrisola> cat: /proc/last_kmsg: No such file or directory | 16:12 |
T42 | <Mister_Magister> /sys/fs/pstore/console_ramoops | 16:13 |
Mister_Magister | @elros34 /proc/last_kmsg is insanely legacy at this point | 16:13 |
Mister_Magister | it was last used on kernel 3.4 | 16:14 |
Mister_Magister | iirc | 16:14 |
Mister_Magister | haven't seen it in 3.10 | 16:14 |
T42 | <elros34> ok but I provide him ramops path also | 16:16 |
Mister_Magister | mkay | 16:16 |
Mister_Magister | just clarifying, haven't read context | 16:17 |
T42 | <rodrisola> cat: /sys/fs/pstore/console_ramoops: No such file or directory | 16:17 |
T42 | <elros34> so whole /sys/fs/pstore/ is empty? Are you sure your device reboots? | 16:19 |
T42 | <rodrisola> yes | 16:20 |
T42 | <elros34> so that will not help | 16:21 |
T42 | <rodrisola> no se reinicia se queda encendido, y no llega a apagarse, se queda en el logo de inicio, es este caso el logo de xiaomi | 16:21 |
T42 | <rodrisola> It does not restart, it stays on, and does not turn off, it stays in the startup logo, in this case the xiaomi logo | 16:22 |
T42 | <elros34> I was talking about restarts, if it stays on then ther eis no point in checking these files | 16:23 |
T42 | <TheVancedGamer> that much old hardware (re @SailfishFreenodeIRCBridgeBot: <Nokius>Hi! anyone i...) | 16:23 |
T42 | <TheVancedGamer> wow | 16:23 |
Mister_Magister | wdym old? | 16:25 |
T42 | <TheVancedGamer> well not old but you get the idea xd (re @TheVancedGamer: that much old hardwa...) | 16:25 |
T42 | <TheVancedGamer> Mister_Magister: not in my senses, battling with an OPi (re @SailfishFreenodeIRCBridgeBot: <Mister_Magister>wdy...) | 16:25 |
Mister_Magister | not sure what you just said but okay | 16:26 |
T42 | <TheVancedGamer> OPi = OrangePi | 16:26 |
T42 | <rodrisola> courbet:/ # ls /sys/fs/pstore/ | 16:26 |
T42 | <rodrisola> console-ramoops-0 pmsg-ramoops-0 | 16:26 |
T42 | <TheVancedGamer> setting up a build server for building sfos in the not-so-distant future | 16:26 |
T42 | <elros34> @rodrisola so now there are files? Dnn't you use tab completion when typing commands? | 16:27 |
Mister_Magister | or like, check the content for similar names | 16:28 |
T42 | <rodrisola> yes I use it, but restart again in twrp, to discard and now there are files | 16:28 |
Mister_Magister | @TheVancedGamer you want to use raspberrypi-similar device for building sfos? | 16:28 |
T42 | <elros34> so they are from twrp boot (useless) not from hybris initramfs | 16:29 |
T42 | <TheVancedGamer> Mister_Magister: maaaaybe :P | 16:29 |
Mister_Magister | are you okay? do you need medical assistance? | 16:29 |
T42 | <TheVancedGamer> no | 16:29 |
T42 | <TheVancedGamer> I know it's bad | 16:29 |
T42 | <TheVancedGamer> But it's a jumphost | 16:29 |
Mister_Magister | maybe you stubmled on the road and hit your head or smth | 16:30 |
T42 | <TheVancedGamer> I said it's a jumphost | 16:30 |
T42 | <TheVancedGamer> I won't actually build on the thing, it would explode | 16:30 |
Mister_Magister | you do know that android10 requires 16GB of ram | 16:30 |
Mister_Magister | or was it 11 | 16:30 |
T42 | <TheVancedGamer> How dare you say that to an Android dev who can compile android 12 with 8GB ram | 16:30 |
Mister_Magister | well | 16:31 |
Mister_Magister | "requires" | 16:31 |
Mister_Magister | if you know what you are doing thenā¦ | 16:31 |
Mister_Magister | limits are for peasants | 16:31 |
* Mister_Magister runs sfos on htc pico | 16:31 | |
T42 | <rodrisola> ok, so they are useless, because it does not start me as a rdnis device, | 16:31 |
T42 | <rodrisola> @elros34 ok, so they are useless, because it does not start me as a rdnis device, | 16:31 |
T42 | <elros34> not sure what do you mean, log is useless because it was created after you have reboot twrp. If it would be there when device reboots itself during hybris initramfs attempt then it would contains valid data | 16:36 |
T42 | <TheVancedGamer> At this point it would be better to add break=premount to cmdline | 16:36 |
T42 | <elros34> what reads this param, kernel or init script? If script then for sfos this would be just like flashing hybris-recovery | 16:42 |
T42 | <TheVancedGamer> @elros34 iirc init reads it | 16:44 |
T42 | <rodrisola> So what I deduce is that the initramfs is not starting, or that it does not activate the rndis in the kernel, that should be activated, right? | 17:04 |
T42 | <elros34> not sure what is the issue but hybris initramfs is cleaver enough to handle lack of rndis with controlled reboot: https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L272-L277 | 17:08 |
T42 | <rodrisola> Then the phone stays on in the startup logo, it is not charging the initramfs, otherwise it would turn off | 17:23 |
T42 | <rodrisola> The phone not reboot | 17:31 |
T42 | <rodrisola> I create kernel of Magik boot, ramfs from hybris | 17:33 |
T42 | <rodrisola> One last card, if you copied the init to the root and executed lineage with your boot? Would you derive from something? | 17:57 |
T42 | <rodrisola> This not solve! (re @SailfishFreenodeIRCBridgeBot: <Thaodan>rodrisola: ...) | 18:02 |
T42 | <elros34> any sign of life if you add 'reboot -f' here: https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L342 or 2 line lower? | 18:16 |
T42 | <rodrisola> Add sleep 15 and reboot -f in this line https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L337 and try | 18:36 |
T42 | <elros34> no, it will not work because busybox is not set up in this line that is why I suggest line 342 | 18:38 |
T42 | <rodrisola> @elros34 yes, initramfs is ok... reboot phone | 20:36 |
T42 | <elros34> so now put reboot few lines lower until you figure out where it hangs. Remeber DONE_SWITCH is "no" in your case | 20:38 |
T42 | <rodrisola> https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L347, this line | 21:00 |
T42 | <rodrisola> https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L346 | 21:01 |
T42 | <elros34> if that is really true then comment it out | 21:12 |
T42 | <rodrisola> in the this line | 21:14 |
T42 | <rodrisola> echo /sbin/mdev > /proc/sys/kernel/hotplug | 21:14 |
T42 | <rodrisola> https://github.com/mer-hybris/hybris-boot/blob/master/init-script#L130 | 21:14 |
T42 | <rodrisola> @elros34 https://www.toptal.com/developers/hastebin/azimexasis.yaml | 21:25 |
T42 | <elros34> interesting bug then | 21:26 |
T42 | <rodrisola> yeah, now my device is live! | 21:27 |
T42 | <foxclore> Hey, just a noob question, is it possible to build hybris17.1 on lineage-18.1 ? | 21:35 |
T42 | <rodrisola> no | 21:36 |
T42 | <rodrisola> @ elros34, ask, what do I do from there? since the system does not start me | 21:37 |
T42 | <foxclore> Okay thankieš | 21:37 |
T42 | <elros34> @rodrisola it's simple: follow hadk and telnet to device | 21:38 |
T42 | <rodrisola> @elros34, in the device sudo password is? | 21:46 |
T42 | <elros34> there is, you alread have root power | 21:46 |
T42 | <rodrisola> how to save change with vi in device | 21:53 |
T42 | <elros34> esc or ctrl + c then ':wq' | 21:54 |
Mister_Magister | ahh the most popular stack overflow question | 21:56 |
Mister_Magister | xd | 21:56 |
T42 | <rodrisola> Mister_Magister, no matter how much I press the correct keys, it does not save changes | 21:57 |
Mister_Magister | sure sure | 21:57 |
T42 | <elros34> depends on what you are trying to change, if initramfs files then it's pointless | 21:58 |
T42 | <rodrisola> I have access to the system, and change in the journal to automatic, but the phone is on the home screen, I do not get to have some life of the system. | 22:12 |
T42 | <rodrisola> The phone does not restart, it remains on, and I have access to telnet, I could change the journal.conf, but according to the guide it should already turn on or I am wrong | 22:31 |
T42 | <rodrisola> sorry, noob question! | 22:38 |
T42 | <rodrisola> system is not mounted! | 22:38 |
T42 | <rodrisola> in my case is system_ext | 22:39 |
T42 | <rodrisola> /help | 22:48 |
T42 | <rodrisola> .../ # systemd | 22:48 |
T42 | <rodrisola> .../bin/sh: systemd: not found | 22:48 |
T42 | <elros34> so does dmesg -w shows something if you run commented out hotplug code? | 23:41 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!