<samueldr> clever: on my few attemps it didn't work
<samueldr> with or without
<samueldr> resize wasn't a recognized command!
<samueldr> so I left both of the methods here
<clever> samueldr: resize is a binary in the xterm and busybox packages
monk has joined #nixos-aarch64
<samueldr> yeah, busybox's resize, which existed, told me it wasn't a recognized command!!
<clever> heh
<clever> i only tested the xterm resize
<samueldr> note: recognized command
<samueldr> different error than a missing command
<samueldr> just like of you called `busybox foobarbaz`
<samueldr> if you*
<clever> samueldr: i dont see that error msg in the src
<samueldr> (maybe I misremember the exact phrasing, but it was a different error iirc)
<clever> [clever@amd-nixos:~/iohk/explorer-app-extended]$ nix-build '<nixpkgs>' -A busybox -o busybox
<clever> [clever@amd-nixos:~/iohk/explorer-app-extended]$ busybox/bin/resize
<clever> COLUMNS=279;LINES=37;export COLUMNS LINES;
<clever> samueldr: seems just fine to me
<samueldr> odd
<samueldr> I was trying it against the qemu serial terminal in the gtk app
<samueldr> but didn't investigate
<samueldr> abandoned the moment I saw it didn't work
<clever> i was just using it in qemu with virtualisation.graphics = false;
<clever> samueldr: did you see my recent adventures on the rpi?
<samueldr> too many sprites at once?
<samueldr> bit banging vga?
<samueldr> oh ntsc
<samueldr> oh, composite?
<clever> yep
<samueldr> with the composite output
<clever> without any aid from the firmware
<samueldr> cool!
<clever> and yeah, too many sprites on the same line
<clever> samueldr: in theory, i could add this code into rpi-open-firmware, and then get a working /dev/fb0 on linux
<clever> due to security features i cant turn off, you wont have the full hw accel or vsync in linux
<samueldr> "security"
<clever> my best guess, is that the arm cant configure the HVS by default, to protect crypto keys
<samueldr> probably something along the line
<clever> allowing that would basically defeat things like secureboot or trustzone
<samueldr> "security" but not for thee
<clever> the official firmware turns that off at some point during boot
<clever> but i dont know how
<clever> samueldr: i also had a crazy idea recently, could i write a VPU hypervisor?...
<samueldr> could you?
<clever> in theory, it would let me run the official firmware under my hypervisor
<clever> and then log everything it did
monk has left #nixos-aarch64 ["Error from remote client"]
monk has joined #nixos-aarch64
apache8080 has quit [Ping timeout: 260 seconds]
apache8080 has joined #nixos-aarch64
rajivr has joined #nixos-aarch64
orivej has quit [Ping timeout: 264 seconds]
apache8080 has quit [Ping timeout: 245 seconds]
cole-h has quit [Quit: Goodbye]
cole-h has joined #nixos-aarch64
patagonicus7 has joined #nixos-aarch64
patagonicus has quit [Ping timeout: 246 seconds]
patagonicus7 is now known as patagonicus
h0m1 has quit [Ping timeout: 264 seconds]
h0m1 has joined #nixos-aarch64
andoriyu_ has joined #nixos-aarch64
andoriyu has quit [Read error: Connection reset by peer]
ajs124 has quit [Quit: Bridge terminating on SIGTERM]
ajs124 has joined #nixos-aarch64
orivej has joined #nixos-aarch64
orivej has quit [Ping timeout: 256 seconds]
<davidak[m]> Pinephone pre-orders have gone live!
<davidak[m]> and i have just ordered :)
monk has left #nixos-aarch64 ["Error from remote client"]
monk has joined #nixos-aarch64
<samueldr> nice
<samueldr> too bad for people missing out (if any)
monk has left #nixos-aarch64 ["Error from remote client"]
<ashkitten> ooh pinephone :D
<CodeKiwi> what is convergence
<CodeKiwi> oh
<CodeKiwi> The convergence package features a Beta Edition PinePhone with a more RAM memory (3GB LPDDR3) and eMMC storage (32GB) as well as a compatible USB-C docking bar capable of charging, digital video output, dual USB 2.0 type-A host ports for peripherals (such as keyboard and mouse) and fast Ethernet.
<ashkitten> it's referring to desktop convergence, as in it has a dock so you can plug in a monitor and keyboard and stuff
<ashkitten> it makes me sad that "dock" nowadays refers to a brick you plug in, not a thing you actually stand your device on
<CodeKiwi> fast ethernet? gigabit or bust, next time, pinephone
monk has joined #nixos-aarch64
<ashkitten> yeah the ethernet in the dock is 100mbit, because the pinephone only has usb 2.0
<CodeKiwi> if i can't attach a phone to a helios64 and get full speeds i don't want it
<CodeKiwi> i don't have one of those yet either :(
<ashkitten> 2.5gbit ethernet?
<ashkitten> i don't know if there is a phone with that
<CodeKiwi> anyone have nix on it yet
<CodeKiwi> phone to gigabit computer to 2.5
<CodeKiwi> i didn't know 2.5 was a thing until today
<ashkitten> my home network is 10gbit fiber :D
<ashkitten> (internally, not internet)
<davidak[m]> nice
<ashkitten> i ordered a pine
<davidak[m]> there are probably phones with fast wifi
<ashkitten> yeah, there probably are phones with wifi 6 already
<CodeKiwi> wi...fi...6?
Gaelan_ is now known as Gaelan
<ashkitten> new name for 802.11ax
<CodeKiwi> i didn't know there was an x
<CodeKiwi> learning all kinds of things today
<samueldr> the "dock" is a good standard type-c eDP thingy with a well-supported realtek chip for ethernet
<samueldr> at $25 it's not a bad price for a known good dongle
<samueldr> the PD power input is also (imo) at a good location
<Ke> yes known good is magic
<Ke> though as mentioned, I also consider pinephone not maybe perfect with their supply chains
<Ke> so the known good can be something else
<Ke> like they shipped wrong usb serials and nvme adapters
<Ke> still I guess pine64 is above average in indie category
<samueldr> the nvme adapter was fine, the cable was "wrong"... except mine was "wrong" and it's fine
<samueldr> you just had to fold it over
<samueldr> (not ideal, but at least it doesn't need to be returned)
<samueldr> but yeah, the serial thing isn't great
<Ke> I think one of the nasty parts is that they told that people have been shipped a wrong product and no action needs to be taken, so I actually waited for replacement
<samueldr> :/
<Ke> not sure how complicated it would have been to send explicit personal confirmation to all involved
<samueldr> yeah, consumer relations isn't on their strong points (yet)
<samueldr> that's one thing people should be aware, going in, that there will be "no software support"
<Ke> in general I do like that people involved can speak more freely compared to most companies though
<Ke> like the controlled output from puri.sm makes them seem even more shady
<Ke> when you know things are wrong, but all you see is their pr
<Ke> so people are eg. reverse engineering shipping rates etc.
<samueldr> are people (not reviewers or influencers) receiving their librem5?
<Ke> yes, but rate is unclear
<samueldr> I'm glad it was too expensive for me to get one at the time
<Ke> I personally know one
evils has joined #nixos-aarch64
<Ke> my assumption is that their estimate of reaching shipping parity or whatever in february did not materialize
<samueldr> (they said shipping party... they are having a shipping party) j/k
<Ke> I actually got a refund, because I got emotionally frustrated with their messaging and delays
<Ke> I would probably have received mine, if they shipped in order
<Ke> I was within first 200 orders
<samueldr> I got somewhat annoyed not even being involved at all other than "mobile non-android linux dev"
<samueldr> I wouldn't imagine if I had backed it... and saw what is basically an almost equal product being shipped much quicker
<samueldr> let's not make ourselves any illusions, the i.MxM platform on the librem5 is not miles ahead of the pinephone
<Ke> it will remain unclear how much bearing the mali400 will have
<samueldr> exactly
<samueldr> the main difference where it matters is the GPU
<Ke> otherwise not, sure
<samueldr> in the end it might be that the pinephone modem will be even more liberated than the librem modems!
<samueldr> (because of community efforts)
<Ke> I don't care about games, but I want to watch videos and I don't want to use special player or special compositor that supports the mali
<Ke> or to patch everything to support the old interfaces
<samueldr> I don't really know whether vivante would be better
<samueldr> it's claimed that the upstream driver already exists and all
<samueldr> but I don't _know_ :)
<Ke> well I believe the hw capabilities are more versatile?
<samueldr> my knowledge stopped mainly at their name and mainline driver status
<Ke> driver maybe not, but if librem-5 ships enough, drivers probably will happen
<samueldr> though the A64 and its Mali-400 ships in more than just the pinephone
<Ke> also if pinephone ships, maybe compositors will really reach out to work on what the mali400 can work with
<samueldr> so that's kind of an advantage
<samueldr> well, people involved with phone-first wayland compositors _are_ working with the pinephone (too)
<Ke> sure
<samueldr> (I'm about to get more into stage-2 things on the pinephone, so I'll be able to tell more soon)
<ashkitten> i'll probably be running postmarketos on my pinephone for a bit at least
<ashkitten> since nixos isn't exactly ready
<Ke> I'd love a opt-in version of nixos-mobile modules, but that probably would take a lot more effort, I have almost 0 time now with the baby and work
<ashkitten> an opt-in version?
<Ke> I guess that's a thing that must happen for nixos inclusion
<Ke> like not enable all the modules at once
<ashkitten> well like
<Ke> it's probably possible right now, but would need to look at it
<ashkitten> mobile-nixos is mostly just kernel and bootloader stuff for devices that can't natively boot nixos with grub or such
<ashkitten> anything userspace would be in nixos proper
<Ke> but if you look at it, there's much more, like adb and usb things etc.
<Ke> and also a build wrapper that I explicitly do not want
<ashkitten> i guess, but it's all stuff that would not be merged into nixos proper
<ashkitten> or probably shouldn't be merged
<Ke> would be cool to have all that stuff as completely separate modules
<ashkitten> how would that work, anyways?
<Ke> but anyway that's a lot of work on top of the insane amount of work already done
<Ke> like all other modules do, apart maybe from the early stages, that somehow need to replace the current ones
<Ke> there is probably quite a lot of work there
<ashkitten> well like, it pretty much needs to build an image with the bootloader and flash it where the phone expects the bootloader to be
<Ke> and also I believe the current form is there to ease the development
<Ke> I would not say it's a must, I have done that stuff manually for all my other devices
<Ke> I use nixos only to generate a bunch of files and then run package them myself
steveeJ has quit [Ping timeout: 240 seconds]
steveeJ has joined #nixos-aarch64
sorki has joined #nixos-aarch64
srk has quit [Ping timeout: 268 seconds]
sorki is now known as srk
adamzivcak has joined #nixos-aarch64
zupo has joined #nixos-aarch64
zupo has quit [Ping timeout: 244 seconds]
makefu has quit [Quit: WeeChat 2.9]
makefu has joined #nixos-aarch64
zupo has joined #nixos-aarch64
cole-h has quit [Ping timeout: 264 seconds]
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
zupo has joined #nixos-aarch64
monk has left #nixos-aarch64 ["Error from remote client"]
monk has joined #nixos-aarch64
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
zupo has joined #nixos-aarch64
FRidh has joined #nixos-aarch64
superherointj has joined #nixos-aarch64
alpernebbi has joined #nixos-aarch64
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
monk has left #nixos-aarch64 ["Error from remote client"]
monk has joined #nixos-aarch64
adamzivcak has left #nixos-aarch64 [#nixos-aarch64]
superherointj has quit [Read error: Connection reset by peer]
superherointj_ has joined #nixos-aarch64
Cadey has quit [Quit: WeeChat 3.0]
superherointj_ has quit [Quit: Leaving]
orivej has joined #nixos-aarch64
jumper149 has joined #nixos-aarch64
jumper149 has quit [Client Quit]
zupo has joined #nixos-aarch64
monk has left #nixos-aarch64 ["Error from remote client"]
monk has joined #nixos-aarch64
orivej has quit [Ping timeout: 256 seconds]
bbigras has quit [Ping timeout: 268 seconds]
LinuxHackerman has quit [Ping timeout: 268 seconds]
AlesHuzik[m] has quit [Ping timeout: 268 seconds]
mica[m] has quit [Ping timeout: 268 seconds]
unclechu has quit [Ping timeout: 268 seconds]
DavHau[m] has quit [Ping timeout: 260 seconds]
LinuxHackerman has joined #nixos-aarch64
Fuseteam has quit [Ping timeout: 268 seconds]
Ke has quit [Ping timeout: 268 seconds]
railroadmanualji has quit [Ping timeout: 268 seconds]
craige[m]1 has quit [Ping timeout: 268 seconds]
dev_mohe has joined #nixos-aarch64
AlesHuzik[m] has joined #nixos-aarch64
bbigras has joined #nixos-aarch64
unclechu has joined #nixos-aarch64
mica[m] has joined #nixos-aarch64
DavHau[m] has joined #nixos-aarch64
Fuseteam has joined #nixos-aarch64
railroadmanualji has joined #nixos-aarch64
Ke has joined #nixos-aarch64
craige[m]1 has joined #nixos-aarch64
dev_mohe has quit [Quit: dev_mohe]
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<samueldr> I don't understand... they are separate modules, with their own enable options
<samueldr> maybe one day you'll be able to show us what you mean :/
<Ke> ah ok
<samueldr> the build wrappers are also opt-in
<samueldr> (e.g. the main default.nix)
<Ke> mean by what?
<samueldr> by what you asked for
<Ke> maybe I am just wrong
<samueldr> by "opt-in version of [Mobile NixOS] modules"
<Ke> if there is enable already, it's probably what I mean
<samueldr> if there's a module that is not "enabled" or "chosen", it's a bug imo
rajivr has quit [Quit: Connection closed for inactivity]
<samueldr> ("chosen" here means something like the system type, where you select by name)
monk has left #nixos-aarch64 ["Error from remote client"]
dev_mohe has joined #nixos-aarch64
<Ke> let's say I have my own nixos-config based on nixpkgs, where I want to make modules available, but not enabled at all, is that possible?
<samueldr> I haven't verified, but I'm pretty sure the defaults enable nothing
<samueldr> there are somethings that are "added" to the build, but as long as you don't evaluate them are (I think) mostly no-ops because of lazyness
<samueldr> (added as in new build outputs to the system.build attr)
<Ke> the mobile stages?
<samueldr> yeah
<samueldr> ah, maybe I'm somewhat wrong, I'd need to check and fix
<Ke> nice
<samueldr> *maybe* it does some disabling for the conflicting NixOS things
<Ke> I wondered, how that could have been possible
<samueldr> without an enable/disable option
<Ke> but yes that absolutely makes sense
<samueldr> but it should still be possible to do with an enable/disable toggle
<samueldr> so I guess just importing the entry point will work... but I haven't verified if importing the entry point works without the "device description" https://github.com/NixOS/mobile-nixos/blob/master/modules/module-list.nix
<samueldr> I would assume things like `mobile.hardware.soc` end up maybe being required?
<samueldr> though... thinking about it... maybe not?
<samueldr> I'd have to check what the state is "no-op" wise with adding the config to a system
<Ke> well I don't mind device description, if it does not enable anything
<samueldr> I think they all change metadata used by mobile nixos, and *maybe* the default mobile nixos setup automatically assumes the kernel described there is desired
<samueldr> they are one of the oldest thing in the whole config
monk has joined #nixos-aarch64
<samueldr> though, if anything is "rude" default-config wise, like if it tramples needlessly some configs when important and unused, that would be a bug that needs fixing imo
<Ke> yes, I only assumed we are not there yet
<Ke> not sure why
<samueldr> I ended up delaying it for later, but I actually want to run all my systems "with" mobile-nixos
<samueldr> either dogfooding the stage-1 fun
<samueldr> or simply ambiant in the config, without side-effects
<samueldr> (inclusive or)
<Ke> "it"
<samueldr> "it" was explained in the second part of the sentence
<samueldr> "it" being "dogfooding Mobile NixOS config in my common config"
<Ke> ah
monk has left #nixos-aarch64 ["Error from remote client"]
ib07 has quit [Ping timeout: 245 seconds]
ib07 has joined #nixos-aarch64
monk has joined #nixos-aarch64
zupo has joined #nixos-aarch64
dev_mohe has quit [Quit: dev_mohe]
cole-h has joined #nixos-aarch64
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
zupo has joined #nixos-aarch64
FRidh has quit [Quit: Konversation terminated!]
orivej has joined #nixos-aarch64
alpernebbi has quit [Quit: alpernebbi]
dev_mohe has joined #nixos-aarch64
dev_mohe has quit [Client Quit]
ib07 has quit [Ping timeout: 245 seconds]
andoriyu_ is now known as Andoriyu
monk has left #nixos-aarch64 ["Error from remote client"]
monk has joined #nixos-aarch64
apache8080 has joined #nixos-aarch64
monk has left #nixos-aarch64 ["Error from remote client"]
zupo has quit [Ping timeout: 256 seconds]
zupo has joined #nixos-aarch64
monk has joined #nixos-aarch64
apache8080 has quit [Ping timeout: 260 seconds]
srk has quit [Ping timeout: 268 seconds]
srk has joined #nixos-aarch64
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
apache8080 has joined #nixos-aarch64
evils has quit [Ping timeout: 256 seconds]
<apache8080> samueldr: have you ever split where the /boot and the rest of the rootfs reside?
<samueldr> yes
<samueldr> but that doesn't mean much without knowing the "boot chain"
<samueldr> which SoC, which bootloader combination
<apache8080> I am trying a setup where /boot is on a separate disk and the rest of the rootfs is on something else, but I am kinda confused as how to do this with sd-images
<samueldr> basically, the bootloader has to be able to ue the location where /boot is
<samueldr> be able to use*
<apache8080> so I have an nvidia jetson that has u-boot which can boot off of the internal eMMC
<apache8080> I want to put /boot on the internal eMMC
<apache8080> and then I want the rest of the rootfs on a SATA drive
<samueldr> it should be doable... but you have to *somehow* get *a* NixOS booted
<samueldr> to then do the setup
<samueldr> you could nixos-install from that NixOS
<apache8080> so we have a system for building sd-images that boot fine
<samueldr> nixos-install is not forbidden from the sd image... the sd image *conveniently* allows installing in-place, but it's not the only way
<apache8080> I was just thinking of taking those sd-image and pulling out the /boot part and putting it on the eMMC then putting the rest of the image on the sata
<apache8080> and in the configuration.nix I was going to point fileSystems."/" to the sata drive and fileSystems."/boot" to the eMMC
<samueldr> there is no boot partition... there is a FAT32 partition which is used solely for the raspberry pi
<samueldr> that should work
<samueldr> assuming indeed the bootloader will go for the stuff in the eMMC
<apache8080> as long as there is an extlinux.conf the bootloader will go for it I believe
<samueldr> I really can't give more details because of the "fun" platform :)
<apache8080> yeah I understand
<samueldr> but your plan should be doable
<apache8080> just wanted to make sure I wasn't doing something completely illogical
<samueldr> I had a raspberry pi with a not-so-dissimilar setup
<apache8080> appreciate the help
<samueldr> sd card held /boot
<samueldr> usb drive held rootfs
<samueldr> and it's just that... fileSystems configured appropriately
<apache8080> oh nice, do you have sources online for that?
<samueldr> but "just" that is hard to get to if you can't boot *something* first
<samueldr> I don't think I have the sources offline either
<apache8080> ah ok
<samueldr> it was an experiment I destroyed quickly because it didn't work well
<samueldr> (power issues with the external usb drive)
<samueldr> though, I *do* think I booted to a temporary system (from an sd card image), and did a nixos-install
<samueldr> hint: on your x86_64 nixos system, you should (also) have nixos-install all the time, available
<samueldr> it's just... always there with NixOS :)
<apache8080> ok
apache8080 has quit [Ping timeout: 256 seconds]
zupo has joined #nixos-aarch64
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]