**** BEGIN LOGGING AT Fri Nov 13 02:59:57 2020
Nov 13 06:12:03 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> > <NotKit> https://github.com/Sailfish-on-Cosmo-Communicator/android_device_planet_cosmocom + https://github.com/gemian/cosmo-linux-kernel-4.4/commits/ubports
Nov 13 06:12:03 <ArcusBot[m]> is there anything that needs to be done for hadk with the baseport? I see there's still mentions of volla/yggdrasil in the .sh scripts but not sure those are used by hadk (also it seems to look for .../planet/mt6771 instead of .../planet/cosmocom I guess that's from boardconfig.mk TARGET_KERNEL_SOURCE := kernel/planet/mt6771?)
Nov 13 07:21:01 <ArcusBot[m]> <@_telegram_101151514:hacklab.fi> yes
Nov 13 07:21:17 <ArcusBot[m]> <@_telegram_101151514:hacklab.fi> * yes, but mt6771 should be correct here
Nov 13 07:21:26 <ArcusBot[m]> <@_telegram_101151514:hacklab.fi> the scripts are not used
Nov 13 07:22:03 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> > <NotKit> yes, but mt6771 should be correct here
Nov 13 07:22:03 <ArcusBot[m]> hmm, I guess the local_manifest will need tweaking then
Nov 13 07:23:27 <ArcusBot[m]> <@_telegram_101151514:hacklab.fi> use ubports branch by the way
Nov 13 07:23:49 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> yeah i've used the ones you pasted here
Nov 13 07:24:52 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> so far got around 6% into hybris-hal building, hoping it's the path fix I skipped that fails there, just wanted to see if it will go through
Nov 13 07:25:19 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> droidmedia did build successfully
Nov 13 17:40:48 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Hey all. Trying to figure out the ARM target triple I need to cross-compile something for my Cosmo - its all prepared for repairs, so I don't want to flash Linux on it just yet, but I was wondering if anyone could have a look at /proc/cpuinfo (I _think_ that'll give me some idea) or elsewhere, and let me know what the target triple is? I.e, something *like* 'x86_64-unknown-linux-gnu`
Nov 13 17:40:49 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Cheers.
Nov 13 17:41:06 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> (More news on the thing soon...)
Nov 13 17:45:09 <ArcusBot[m]> <@_telegram_101151514:hacklab.fi> aarch64-linux-gnu probably
Nov 13 17:45:57 <ArcusBot[m]> <@_telegram_101151514:hacklab.fi> or basically whatever aarch64 target name is, it did not have incompatible ABIs on Linux as it is relatively new compared to ARM32
Nov 13 17:56:35 <mifritscher> on android: uname -a gets a "Linux localhost 4.4.146 #2 SMP PREEMPT Mon Jul 6 19:51:17 CST 2020 aarch6", cpuinfo gives "Processor	: AArch64 Processor rev 4 (aarch64)"
Nov 13 18:06:26 <ArcusBot[m]> <@_telegram_565620682:hacklab.fi> lol
Nov 13 18:06:52 <ArcusBot[m]> <@_telegram_565620682:hacklab.fi> double bot proxying
Nov 13 18:06:53 <ArcusBot[m]> <@_telegram_565620682:hacklab.fi> thats fun
Nov 13 18:07:37 <ArcusBot[m]> <@_telegram_565620682:hacklab.fi> probably an indicator that the proxy hub should be matrix... :)
Nov 13 18:13:51 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> @NotKit Cheers. I'll try and get that target triple configured then.
Nov 13 18:38:42 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Cool, that compiles fine. And on musl, too! :D
Nov 13 20:40:56 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Well, it didn't at first. Apparently I had to do some funky stuff with GCC and cross-compiling. Finally got it done in the end. To explain, I'm giving it a shot of writing the Linux CoDi interface in Rust...
Nov 13 20:41:13 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Which is why I wanted to use musl, and statically link.
Nov 13 20:42:22 <ArcusBot[m]> <@_telegram_101151514:hacklab.fi> have you seen https://github.com/gemian/codi-app?
Nov 13 20:44:10 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Yes, Davide was the one who, on my request, showed me that for the CoDi protocol.
Nov 13 20:44:17 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> I forked from his repo.
Nov 13 20:44:25 <ArcusBot[m]> <@_telegram_101151514:hacklab.fi> ah, ok
Nov 13 20:45:19 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> i need to commit the cross-compiling stuff, but essentially had to use Docker and rust-embedded/cross to compile for aarch64.
Nov 13 20:45:30 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> it was easier than messing with my host system (openSUSE).
Nov 13 20:47:37 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> I could do with some testing on a simple 'Hello, world!' musl aarch64 executable on: Sailfish, UBPorts, and Gemian at some point. But I need to commit first, and setup GitHub Actions.
Nov 13 20:48:02 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> There's also a aarch64-unknown-linux-gnu build, which will be done alongside musl on CI.
Nov 13 21:43:48 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> ugh, that's irritating. Setup GitHub Actions, and I can't download the produced binaries.. there's a issue about it, without publishing them.
Nov 13 21:44:29 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Not exactly ideal. Maybe I'll give GitLab a try.
Nov 13 22:19:40 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> > <NotKit> have you seen https://github.com/gemian/codi-app?
Nov 13 22:19:40 <ArcusBot[m]> Holy shit it works
Nov 13 22:20:14 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> Have time now displayed under sfos
Nov 13 22:22:48 <ArcusBot[m]> <@_telegram_565620682:hacklab.fi> lol
Nov 13 22:22:52 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> none of the toggles change anything but still
Nov 13 22:22:53 <ArcusBot[m]> <@_telegram_565620682:hacklab.fi> imjelly
Nov 13 22:23:49 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> should work straight away on debian too, just install a bunch of python thingies (setuptools and pip, then dbus, pyserial as needed what codiServer.py cries about)
Nov 13 22:24:18 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> oh evdev might need gcc and few others like python3 and devel but in the end shouuld just start up codi
Nov 13 22:24:22 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> didnt seem to work for me on gemian, the planet logo appears now when i close the device but i dont get a clock or anything
Nov 13 22:24:23 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> hmm
Nov 13 22:29:25 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> ahh
Nov 13 22:29:30 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> the only thing i can see missing from that repo is mouse support.
Nov 13 22:29:30 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> ran it in a terminal and it compained about lack of networkmanager
Nov 13 22:29:35 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> i'll probably email about that...
Nov 13 22:29:43 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> so switching back to NM makes it work again
Nov 13 22:29:49 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> weird.
Nov 13 22:29:53 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> let me check..
Nov 13 22:30:09 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> ah yeah, they're using nm's dbus
Nov 13 22:30:21 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> right, but i thought connman was the new thing? unless they got 4G working on nm?
Nov 13 22:30:31 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> i hope so, i prefer nm to connman.....
Nov 13 22:30:32 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> yeah connman needed for 4g data
Nov 13 22:30:43 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> as far as i can see anyhoo
Nov 13 22:30:44 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> right, but this makes it sound like it works with nm now
Nov 13 22:30:51 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> maybe for the upcoming update?
Nov 13 22:31:48 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> yeah looks like most functions aren't implemented yet module 'CodiFunctions' has no attribute 'SetFlightModeStatus'...module 'CodiFunctions' has no attribute 'SetDoNotDisturbStatus'...module 'CodiFunctions' has no attribute 'SetBatterySaverStatus'
Nov 13 22:32:18 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> yes, it is rather minimal right now
Nov 13 22:42:54 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> 4G works in networkmanager it seems
Nov 13 22:42:55 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> oh?
Nov 13 22:42:56 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> ccool
Nov 13 22:42:57 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> on gemian?
Nov 13 22:42:57 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> has there been an update?
Nov 13 22:42:58 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> hm, maybe i'll merge my branch to master.
Nov 13 22:42:59 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> /me shrugs
Nov 13 22:43:00 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> yepp
Nov 13 22:43:00 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> bugger
Nov 13 22:43:01 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> when you run the .py in a terminal
Nov 13 22:43:01 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> but ive no idea what that actually does
Nov 13 22:43:01 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> a line that says SetMouse does show up
Nov 13 22:43:01 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> oh no wait, it is there
Nov 13 22:43:02 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> looks like it uses xdotool
Nov 13 22:43:02 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> i didn't see that in my copy
Nov 13 22:43:02 <ArcusBot[m]> <@_telegram_45472932:hacklab.fi> it’s nice that the codi server code includes documentation for the packet formats at last
Nov 13 22:43:14 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> can't seem to exit the phone app and on sfos it has nothing (no contacts/history/no way to dial or exit) but ctrl c works
Nov 13 22:43:51 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> the phone app as in, the CoDi one? or the normal sailfish dialer app
Nov 13 22:44:38 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> the codi one
Nov 13 22:45:13 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> can't swipe out from it and after restart now stuck on planet logo
Nov 13 22:45:25 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> hehe yep same thing for me on gemian
Nov 13 22:46:10 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> opening and closing the lid seems to set it back to the clock face
Nov 13 22:47:15 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> i hope the mouse support comes, thats a major killer for me on debian, i cant stand this relative  pointer stuff
Nov 13 22:49:44 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> lid closing does nothing on sfos sadly
Nov 13 22:50:14 <ArcusBot[m]> <@_telegram_991509486:hacklab.fi> probably not plugged in as its kernel is from before may when gemian had esc tweaks and then the suspend
Nov 13 22:50:24 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> ahh
Nov 13 23:30:52 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> If anyone would be able to, as my Cosmo is not setup right now, could they please test either (or both!) the -gnu or -musl binaries generated by my project for the Linux side of CoDi? Thanks.
Nov 13 23:30:52 <ArcusBot[m]> You'll see here (https://github.com/shymega/cosmo-codi-linuxd/actions/runs/362489534), there's two artifacts. They'll be zipped, and normal files when extracted. You'll need to chmod +x them.
Nov 13 23:31:04 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Hopefully it'll work.. GitHub Actions was a bit of a pain.
Nov 13 23:32:10 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Although, weird that the annotations say it wasn't found.. the artifacts have the executable there. Probably a red herring.
Nov 13 23:33:10 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Ah, its the release job. That's only meant to build on release tags.
Nov 13 23:33:20 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> But yeah, I'd be grateful for any testing. Thanks!
Nov 13 23:47:45 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> both return a 'Hello, Cosmo!'
Nov 13 23:48:40 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> Excellent.
Nov 13 23:49:20 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> One is linked against glibc, but the other is statically linked against musl. They'er both from the same sources though.
Nov 13 23:49:28 <ArcusBot[m]> <@_telegram_701506086:hacklab.fi> fair
Nov 13 23:49:30 <ArcusBot[m]> <@_telegram_888165589:hacklab.fi> And configuration will probably be done via TOML.
**** ENDING LOGGING AT Sat Nov 14 02:59:56 2020