jasongrossman has quit [Remote host closed the connection]
jasongrossman has joined #nixos-chat
Synthetica has quit [Quit: Connection closed for inactivity]
drakonis_ has joined #nixos-chat
drakonis has joined #nixos-chat
drakonis_ has quit [Ping timeout: 252 seconds]
drakonis1 has joined #nixos-chat
aanderse has joined #nixos-chat
hedning_ has quit [Remote host closed the connection]
drakonis1 has quit [Quit: WeeChat 2.4]
drakonis_ has joined #nixos-chat
drakonis1 has joined #nixos-chat
drakonis has quit [Ping timeout: 258 seconds]
drakonis1 has quit [Quit: WeeChat 2.4]
pie__ has quit [Remote host closed the connection]
pie__ has joined #nixos-chat
endformationage has quit [Ping timeout: 258 seconds]
jasongrossman has quit [Ping timeout: 248 seconds]
Synthetica has joined #nixos-chat
lopsided98 has quit [Ping timeout: 248 seconds]
jasongrossman has joined #nixos-chat
jasongrossman has quit [Ping timeout: 244 seconds]
lopsided98 has joined #nixos-chat
emily has quit [Ping timeout: 250 seconds]
emily has joined #nixos-chat
jasongrossman has joined #nixos-chat
jD91mZM2 has joined #nixos-chat
__monty__ has joined #nixos-chat
<tilpner>
ar: What do you want to do?
<kraem>
tilpner: you ran nvidia right? i'm running nvidia prime in static mode but am getting screen tearing + the screen doesn't seem to refresh sometimes in terminals until i press a key on the keyboard - do you have any tips?
<tilpner>
No, I don't usually have the nvidia card powered
<tilpner>
And when I do, something about scaling is all wrong, and I don't even notice the tearing anymore
<tilpner>
primusrun has the fewest glitches so far, but still tearing
<kraem>
it's so strange that it's basically the same setup i ran in arch and everything was smooth, the tearing is much less in chromium than firefox though. What i can't live with is the terminals not updating until i press a button in vim and nvim - does it sound familiar?
<tilpner>
No, I don't rebuild-boot into an nvidia system to stay in terminals a lot. I remember there being weird blocky glitches around the cursor too
<manveru>
kraem: there was a setting for the tearing i think
<manveru>
i dunno, i just enabled everything that sounded like it's for optimus
<manveru>
no clue how it actually works... that's the beauty of nix ;P
<kraem>
hm ok :) what does `glxinfo | grep string` report for you?
<manveru>
can't check atm, only have ssh to it since i'm working
<kraem>
manveru: i understand - i'd be happy to hear more when you're at the machine :)
<manveru>
sure :)
drakonis has joined #nixos-chat
drakonis_ has quit [Ping timeout: 272 seconds]
Jackneill has quit [Ping timeout: 248 seconds]
Jackneill has joined #nixos-chat
<eyJhb>
Nixpkgs is the `same` as nix-channels, right?
<andi->
not exactly
<eyJhb>
Damn
<eyJhb>
What is the difference?
<eyJhb>
The files seems to be the same
<Taneb>
The channels have been tested a little more
<andi->
it isn't all the branches, it has different branches and doesn't have all the commits.. It only has those commits that have passed hydra tests
<Taneb>
And are updated automatically
<eyJhb>
But this is just regarding the files being the same, not wether or not they are tested more :) Trying to figure out what commit broke displaylink
<eyJhb>
So I guess I will have to use nixpkgs as a basis for this
<andi->
e.g. master will be "pushed" to the nixos-unstable branch whenever the nixos/trunk-combined jobset finishes
drakonis_ has joined #nixos-chat
drakonis has quit [Read error: Connection reset by peer]
<gchristensen>
"setting up tmpfiles" takes soooo long for me.
<gchristensen>
it is a humongous drag
<andi->
is that because of your rootfs hacks? :)
<gchristensen>
no :(
<gchristensen>
it is because r13y's --keep-failed and a tmpfiles.d rule to delete old /tmp/nix-build-* dirs
<andi->
all it does is call systemd-tmpfiles with --create --remove IIRC
<andi->
oh
<andi->
and it is too big to be an actual tmpfs?
<gchristensen>
probably :)
<gchristensen>
yeah. it is 33G right now, and "setting up tmpfiles" has been running for like 10min
* etu
migrated another system to tmpfs as rootfs today :p
* etu
is hooked on that sauce
<gchristensen>
nice
<eyJhb>
gchristensen: lets see if this bisect with displaylink works on dirty /etc etc. :p
<gchristensen>
:D
<eyJhb>
I really need a minimal config... Having to rebuild for virtualbox, latex, etc. each time, then I will NEVER finish...
<eyJhb>
Considering going to Uni for the sweet sweet 1 Gbps
<gchristensen>
:o
<eyJhb>
"only" have 100 Mbit/s at my apartment
Jackneill has quit [Ping timeout: 248 seconds]
<manveru>
kraem: so yeah, my setup was not working right, never really tested the 3d part, you're right
<manveru>
kraem: don't have much time to check atm, but optirun is basically broken if i use only bumblebee, haven't tried using only optimus_prime yet
<manveru>
the biggest thing i cared about is that the card stays off, since the notebook sounds like a fighter jet after a minute or two even if not doing shit
Jackneill has joined #nixos-chat
<eyJhb>
IS there a place in NixOS I can put specific iptables rules?
<eyJhb>
E.g. I have this - -A PREROUTING -d myip/32 -p tcp -m tcp --dport 6000 -j DNAT --to-destination 172.20.0.6:6000
<gchristensen>
networking.firewall.extraCommands
<gchristensen>
but warning, you have to be very careful in their construction or they won't be properly changed when doing a rebuild-switch or rollback
<eyJhb>
Thanks! Considering switching my `gate` (VPN server I forward trafic with), to NixOS
<eyJhb>
HOw so?
<gchristensen>
well with that prerouting rule it sounds like you're running a nat?
<eyJhb>
Basically yeah. FOrwarding traffic from my VPS, down a VPN, to my pfSense router, to my server at home, and then the whole way back
<eyJhb>
This bisect will take FOREVER.. Especially since my good commit is januar 2019, the bad is around yesterday. And the commit it is forcing me to test, is september 2018???
<gchristensen>
it'll cut it in half every time though
<eyJhb>
Ohh let me rephrase this
<eyJhb>
The first known good commit is february, the bad commit is june 05 (both 2019). The first to test was then junary.. That was marked good, the next was the september 2018.
<eyJhb>
Something is very wrong from the looks of it still? Should it be b etween the good and bad ?
<Taneb>
Not necessarily, git history doesn't need to correspond to actual history. It could have been rebased at a later date, for example
<eyJhb>
Hoping so..
<eyJhb>
Because this current rebuild is taking forever, Even with a minimal configuration
<eyJhb>
Does gcc really take that long to compile?
<eyJhb>
`After about 4.5 hours it finished.` not liking this
<kraem>
manveru: i see - i just did a fresh install and seems like everything is working. gonna do some more tests though. would you like me to ping you with my configuration if it is indeed working?
<manveru>
indeed, i'd appreciate that a lot :D
<kraem>
manveru: also; is the laptop a macbook? if that's the case I could point you to some settings to make the fans chill down :)
<manveru>
nope, it's some kind of noname thingy
<manveru>
doesn't even have a logo or model name
<manveru>
battery is dead too, and it weights a ton, but it works :D
SamuelMarks has joined #nixos-chat
<SamuelMarks>
hi
<SamuelMarks>
drakonis_: So decided to open-source everything under MIT/Apache-2.0. Released almost everything now! - The rest by the end of the month (waiting to finish my doc site generator, then will release WITH-docs ;])
<samueldr>
welp, found a typo in the single UNIX® specification :/
<samueldr>
or maybe not, not exactly obvious since they repeat it, but they reference Common Upper Layer Requirements as CURL instead of CULR
<SamuelMarks>
In the process of a major RIIR from Python to Rust though
<SamuelMarks>
You'll note that all those Python libraries have a classifier: 'Development Status :: 7 - Inactive',
<manveru>
yeah :)
<manveru>
how you like rust so far?
<SamuelMarks>
It's good, I'm—like the rest of the community—very much waiting for the Task, Executor, Async, Await syntax to be finished
<SamuelMarks>
E.g.: Currently using aria2c for downloads, but would much rather do it internally, but the thread pool, process pool, async pool, futures pool, events pool are all TBA
<manveru>
yeah, i keep hearing that
<SamuelMarks>
Also doing some C++ work, going to be offering major contributions to a bunch of top-level Apache projects
<SamuelMarks>
Maybe C++20 features will be out before Rust has finished its, then I'd be tempted to use C++ for some of the major parts that was going to write in Rust
<SamuelMarks>
You're the first person outside of offscale to review it =)
<eyJhb>
Anybody want to spare some computing power for compiling Nix stuff? ;) ;)
<sphalerite>
eyjhb: hydra, if you put it in nixpkgs ;)
<SamuelMarks>
Yeah, so I'm thinking that this is the kind of thing that could be used within a Docker image, natively on a developer machine, &etc &etc
<eyJhb>
sphalerite trying to pinpoint in nixpkgs when displaylink breaks... So I am basically on all kinds of commits, where it has no cache.. :(
<sphalerite>
oh
<sphalerite>
good luck I guess :p
<manveru>
SamuelMarks: yeah, the only issue i have is that it's still yaml... so people will eventually build some templating engine for it :P
<SamuelMarks>
The biggest drawbacks I can see from this are around caching (Travis, Docker, &etc.). So might need to codegenerate the relevant files (potentially bidirectional travis<->offsetup, Dockerfile<->offsetup)
<SamuelMarks>
manveru: Oh, you can just use a generic one, like spruce or spiff. JSON-ref isn't awful either
<eyJhb>
If this continues, I will prop spin up a Docker with NixOS on a server...
<tilpner>
eyjhb: Get a beefy VPS for a few hours and automate bisect
drakonis has joined #nixos-chat
<eyJhb>
tilpner: cannot automate this bisect... :/ REquires tons of reboots, and checking if the display works
<tilpner>
Oh no
<tilpner>
Maybe use the beefy VPS as a remote builder then? :/
<eyJhb>
Works now.. Always remembered /boot as constant full
<__monty__>
You've probably never generated as many initramfses as on nixos though.
<eyJhb>
Prop not.. Getting tired of rebooting...
drakonis_ has joined #nixos-chat
drakonis has quit [Ping timeout: 252 seconds]
drakonis_ has quit [Read error: Connection reset by peer]
drakonis_ has joined #nixos-chat
drakonis_ has quit [Ping timeout: 252 seconds]
drakonis_ has joined #nixos-chat
<joepie91[m]>
lol
<joepie91[m]>
"hmmm... there's some space in that ramfs..."
<joepie91[m]>
specifically, it's a read-only... 'submount'? whatever the hell those are called. basically a read-only remount of the /nix folder in / that appears as its own mounted volume
<eyJhb>
joepie91[m] but that is just me being stressed and not being able to retain a single thing currently!
* eyJhb
skipping bisects because it won't even build......
sir_guy_carleton has joined #nixos-chat
<samueldr>
that's fine and it happens
<joepie91[m]>
eyjhb: I also always forget it, so it's not you :)
<joepie91[m]>
or not just you anyway
<sphalerite>
joepie91[m]: re your "no bitNot function" papercut: I'm guessing it's because that would imply some specific integer size, unlike the other ops
<sphalerite>
I mean, what would you expect to come out of (bitNot 0)?
<__monty__>
-1?
<__monty__>
: >
drakonis has joined #nixos-chat
<kraem>
anyone running libinput-gestures? can't launch it in my i3config because it needs to be started with root. on arch i started it with `libinput-gestures-setup start` but the command is not available when installed in nixOS
<eyJhb>
I have no clue, why the rebuild is building aws-cp-sdk-s3..................................
<samueldr>
the s3 sdk is part of nix
<samueldr>
dependency to be able to push to s3
<eyJhb>
So just default dependency?
<samueldr>
yeah
<eyJhb>
Seems no fun.. Also seeing arduino stuff, which.. I am not sure makes sense
<manveru>
that did the trick, plus adding my user to "video" group
<manveru>
now optirun only temporarily starts the card, and i get 3d
sir_guy_carleton has quit [Quit: WeeChat 2.4]
__monty__ has quit [Quit: leaving]
tilpner has quit [Quit: WeeChat 2.4]
tilpner has joined #nixos-chat
<kraem>
manveru: ah cool - I did get everything working with the _nvidia prime static mode_ - I'm still having these weird glitches where the buffer sometimes not updates itself until i press another key when i'm scrolling in neovim for example - maybe i'll just try doing it through the intel drivers with bumblebee instead and see how it treats me, don't actually need the performance from the discrete card _but_
<kraem>
it's hardwired to my hdmi port :(
drakonis has quit [Read error: Connection reset by peer]
drakonis has joined #nixos-chat
drakonis_ has quit [Read error: Connection reset by peer]
drakonis_ has joined #nixos-chat
drakonis has quit [Ping timeout: 248 seconds]
drakonis_ has quit [Read error: Connection reset by peer]