CheRRy2fly21 has quit [Quit: -a- IRC for Android 2.1.59]
CheRRy2fly21 has joined #nixos-aarch64
cole-h has quit [Ping timeout: 240 seconds]
<colemickens>
I can't seem to scp/rsync over, any other tips?
<colemickens>
oops, theres probably a stage-1 extra packages or something
<colemickens>
of course, extraUtils :)
justanotheruser has joined #nixos-aarch64
<colemickens>
samueldr: modeset-vsync works, modeset-atomic doesnt, modset-double-buffered is slow, but works.
<colemickens>
samueldr: plain old modeset seemed to hang it ?
<samueldr>
plain old modeset doesn't swap buffers :)
<samueldr>
it "works", but works outside of view
<colemickens>
oh okay
<colemickens>
is it a good sign that modeset-vsync works?
<samueldr>
yes
<samueldr>
I think
<samueldr>
it must be notifying the driver in some ways that, in turn, does the right thing
<colemickens>
I imagine the jedi hand wave, "ways".
<andi->
My OnePlus 3T is finally no longer in use so I can start tinkering with mobile nixos \o/
<samueldr>
andi-: great!
<andi->
don't expect too much :D
<samueldr>
I don't :)
<samueldr>
but it's great nonetheless
<colemickens>
samueldr: should I try to run some more things, like Sway ( I think sway wants logind, so idk if that will work) ?
<andi->
It isn't like I find a blackhole to hide my reality in there while I go down that path
<samueldr>
I think I don't have a 3T, but a 3, so maybe verify first, but it should be brick-safe since the programmers are out there
<samueldr>
colemickens: definitely
<andi->
I did boot it a while ago already when I had to do the last wipe
<samueldr>
try anything in stage-2 you can
<samueldr>
colemickens: I literally did the minimum effort of configuring x11 for drm and seeing it didn't work
<samueldr>
nothing more
<colemickens>
I'm sort of confused if I'm actually in stage 2. The NIXOS_SYSTEM partition is mounted at /mnt ?
<samueldr>
you're still in stage-1 then
<samueldr>
you know about stage-1/stage-2 in NixOS (non-mobile), right?
<samueldr>
initramfs is stage-1, then we switch root into the full system
<colemickens>
yes
<samueldr>
that full system is stage-2
<samueldr>
good
<samueldr>
just checking :)
<colemickens>
yeah I'll dig into that and see why it's not fully switching then. I assume when I get stage-2 proper then I'll get a nix daemon, etc?
<samueldr>
yes
<samueldr>
you get a bog standard nixos system
<samueldr>
there is literally nothing "mobile" about it :)
<samueldr>
¨ mobile.boot.stage-1.shell.enable
<samueldr>
this will stop the boot
<samueldr>
so if you're using it with the serial console, just exit out of it to continue the boot
<samueldr>
(exit as in the exit built-in)
<samueldr>
andi-: yeah, I remembered you booted it, and it didn't surprise me that it worked considering what I read about the differences between 3 and 3T
<samueldr>
I don't think there is anything challenging with it
<samueldr>
specific to it
<andi->
I don't even know what to do with it once it boots NixOS...
<samueldr>
there is all the work to get icnss wifi working, then all the work for gpu accel, sound, modems, etc
<samueldr>
so if the goal is not to improve the support, probably there's not much for you right now
<andi->
I guess I can give it a try once I'm bored :D
<samueldr>
once at least wifi is on, you can make it a toy to look at phone environment alternatives
<samueldr>
maybe even useful!
<andi->
true
<samueldr>
oh, there might be some mainline work
<andi->
any idea what it takes to get the GSM part working? I've been thinking of adding a backup path into my home-network via LTE.. Could of course use a regular LTE stick but that would be trivial
<samueldr>
not yet
<samueldr>
still a completely open avenue to me
<samueldr>
though I assume hybris support is required
<samueldr>
which is too for sounds and probably GPU accel
<samueldr>
though before I get to work with hybris to understand it, I have a couple things to do
<samueldr>
I guess I should really explore the mainline options soon
<samueldr>
I wonder how modem will work on mainline though
CheRRy2fly21 has quit [Ping timeout: 265 seconds]
h0m1 has quit [Ping timeout: 272 seconds]
h0m1 has joined #nixos-aarch64
knerten2 has joined #nixos-aarch64
knerten1 has quit [Ping timeout: 240 seconds]
patagonicus1388 has joined #nixos-aarch64
patagonicus1388 is now known as patagonicus138
rajivr has joined #nixos-aarch64
knerten2 has quit [Ping timeout: 240 seconds]
noonien has quit [Quit: Connection closed for inactivity]
<colemickens>
I saw the push from the bot in #nixos, ha.
<samueldr>
yeah
<samueldr>
you can't just hide anything from it :)
orivej has joined #nixos-aarch64
justanotheruser has quit [Ping timeout: 240 seconds]
ninjin_ has joined #nixos-aarch64
cole-h has joined #nixos-aarch64
alp has joined #nixos-aarch64
<colemickens>
what should my expectations be around rndis after stage-2 succeeds?
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<colemickens>
(btw, boot selection blocks boot if FBterm is broken ;))
<samueldr>
fbterm?
<samueldr>
I don't follow
<colemickens>
sorry i meant the framebuffer
<samueldr>
and yeah, boot selection menu assumes there is an interface
<samueldr>
BUT
<colemickens>
it fails when showing the boot generation selection
<samueldr>
volume keys do work!
<colemickens>
I just short circuited it for now
<samueldr>
and power button too!
<samueldr>
colemickens: rndis might should can work in stage-2, but there is nothing configuring it... if it was already configured in stage-1 it will work in stage-2
<samueldr>
that's how I use it
<samueldr>
since it's configured in stage-1, it ends up just working
<colemickens>
hm, that's what I was hoping.
zupo_ has joined #nixos-aarch64
<colemickens>
I don't have to manually enable dhcpd in stage-2?
<colemickens>
It seems like I'm getting a usb0 device but I'm not getting a lease
<samueldr>
I enable rndis in stage-1 and use the unmodified rootfs from hydra
<colemickens>
same
<samueldr>
though I guess sometimes it doesn't have the time to lease?
orivej has quit [Ping timeout: 272 seconds]
<colemickens>
do process (like udhcpd) get killed during pivot?
<colemickens>
or stage-1 -> stage-2, I should be more speciifc
<samueldr>
yes
<samueldr>
uh
<samueldr>
no
<samueldr>
it's hard to explain
<samueldr>
I think what will happen is that udhcpcd isn't able to process since files are gone, all mounts are gone
<samueldr>
but its process is not culled since we don't do any kind of cleanup in mobile nixos yet
<colemickens>
I see.
<colemickens>
I can probably just set static ip on my dev machine to unblock then
<samueldr>
oh
<samueldr>
right
<samueldr>
I do that maybe
<samueldr>
I don't actually remember
<samueldr>
(have I ever said that I don't like networking?)
zupo_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
ib07 has quit [Ping timeout: 240 seconds]
patagonicus1381 has joined #nixos-aarch64
patagonicus138 has quit [Ping timeout: 272 seconds]
patagonicus1381 is now known as patagonicus138
<sphalerite>
samueldr: I heard you're good at networking, could you help me with this problem I'm having?
<samueldr>
>:|
<samueldr>
ifconfig eth0 down
<samueldr>
or whatever
<sphalerite>
re the above, doesn't systemd kill everything from the initramfs? Or am I confusing things?
<samueldr>
its own switch_root does
<samueldr>
which it's not using
<sphalerite>
aaah
<sphalerite>
which we're not even using in nixos :D
<sphalerite>
regular nixos
<samueldr>
also yeah
<samueldr>
though I think there's culling of non-@ processes
<samueldr>
>> It does not provide a complete skeleton yet. Mainly for now it’s missing the kernel derivation. Though we might be able to provide likely source locations depending on some factors.
<sphalerite>
"msm-3.18 upstream kernel" being what pmos uses AFAIU
<samueldr>
I just wanted to see if it would work, and yeah, seems like the logic for lineageos is right for this device too
<sphalerite>
so lineageos is the first one to try?
<samueldr>
that's what I'd do
<samueldr>
I don't have a strong opinion
<samueldr>
though well-known repositories are better than random forks
<sphalerite>
hm, lineageos dropped support for the device in 17.x
<sphalerite>
ah well, I'll try it anyway
<samueldr>
that doesn't really matter for us
<samueldr>
it's not like z00t and addison have 17.x support
<sphalerite>
well, older unmaintained kernel… though I guess that's not _actually_ different on devices that are still supported x)
<samueldr>
exactly
<sphalerite>
hm, what's the current state on the kernelPatches arg? Maybe that needs to be removed from autoport :)
<samueldr>
oh
<samueldr>
yes
<sphalerite>
so I just take it out of the kernel callPackage?
<samueldr>
yep
<sphalerite>
perfect
<sphalerite>
now building the first image attempt \o/
<samueldr>
(speaking of, I should really use ERB for the skeleton files, as currently it's a mess)
<sphalerite>
that was pretty fast.
<sphalerite>
oh, hm, where does the kernel config come from in autoport? I have some doubts that that will match up with the kernel sources :)
<samueldr>
from the OEM kernel build
<samueldr>
though it's a stepping stone
<samueldr>
what I would do: (1) commit the from-autoport config (2) run normalize with whatever sources you want and commit _that_ (3) start configuring
<samueldr>
alternatively you could download the latest linageos for santoni, extract its boot.img, extract its kernel, extract its kernel configuration
<samueldr>
(which at some point will be handled for you in autoport)
<sphalerite>
alrighty
<samueldr>
if you have doubts about new configs, or failure to build, compare with the santoni defconfig from the lineageos kernel repo I guess
<sphalerite>
hm, do you prefer enableRemovingWerror over removeWerror so that all the boolean things start with enable?
<samueldr>
>> What's the difference between is* and enable* parameters? is are for intrinsic "properties" of the build. enable*, while enable* serves to enable a feature. The distinction becomes a big hazy with things like enableCompilerGcc6Quirk and enableRemovingWerror, but the same applies. The gcc6 quirk or removing Werror is not a property of the build, but a feature needed for the current implementation to work.
<samueldr>
hmmm, so that detection isn't exactly foolproof :/
<sphalerite>
right… but I think it might be ok? The main problem here is that it misled me into enabling some msm8937 kernel config options, which stopped the santoni camera code from being used.
<sphalerite>
OOH I HAVE A BOOT.IMG
<samueldr>
yeah
<samueldr>
you probably don't want to change what's enabled in those options compared to OEM/LineageOS
zupo has quit [Ping timeout: 260 seconds]
<sphalerite>
welp, it seems to be trying to boot, but failing
<samueldr>
failing how?
<samueldr>
if it reboot immediately after a fastboot boot, try another gcc
<samueldr>
if it reboots after a small while from fastboot, maybe changing the display backlight level a bit, the kernel started trying to do something
<samueldr>
there is a flashable zip that includes only the "vendorey" bits
<sphalerite>
that domain looks totally trustworthy :D
zupo_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<samueldr>
it kind of is due to signatures
<sphalerite>
nice
<samueldr>
though it is totally verifiable if you want
<samueldr>
download the firmware from xiaomi and look at the files
<samueldr>
same hashes = okay
<samueldr>
you'll have a different install script though
<samueldr>
which can be observed
<samueldr>
but yeah, that looks sketchy at first glance
<sphalerite>
hm, vendor for santoni isn't available, only firmware
<sphalerite>
oh boy 30KiB/s download
knerten2 has joined #nixos-aarch64
<sphalerite>
holy CRAP the phone connects to wifi again (with lineageos)
<sphalerite>
this was an eternal problem with it, and the reason I have it
<sphalerite>
welp, since the firmware download will take another 20min, I'm going to bed
<sphalerite>
gnight!
<sphalerite>
thanks very much for your help again samueldr
<sphalerite>
samueldr++
<{^_^}>
samueldr's karma got increased to 278
knerten1 has quit [Ping timeout: 264 seconds]
alp has joined #nixos-aarch64
<samueldr>
you're welcome
<samueldr>
hopefully we can get through the issues soon
<samueldr>
I would assume starting from lineageos' config would help
knerten1 has joined #nixos-aarch64
knerten2 has quit [Ping timeout: 260 seconds]
Darkmatter66 has quit [Quit: ZNC 1.7.5 - https://znc.in]
Darkmatter66 has joined #nixos-aarch64
knerten1 has quit [Remote host closed the connection]
knerten1 has joined #nixos-aarch64
V has joined #nixos-aarch64
fps has joined #nixos-aarch64
Dezgeg has joined #nixos-aarch64
knerten2 has joined #nixos-aarch64
knerten1 has quit [Ping timeout: 272 seconds]
<colemickens>
I goofed. I locked myself out, then tried to "start over". Now everytime it tries to pivot to stage-2 it complains that NIXOS_SYSTEM is corrupt. Even though I'm flashing to userdata like I was before? idk