dtzWill has quit [(Ping timeout: 260 seconds)]
orivej has quit [(Ping timeout: 276 seconds)]
<gchristensen> anyone have opinions about things? I think we should merge https://github.com/NixOS/nixpkgs/pull/31150
<gchristensen> (rust / carnix)
<gchristensen> it may not be perfect but it is a nice big step
regnat[m] has quit [(Remote host closed the connection)]
florianjacob has joined joined #nixos-dev
phreedom has quit [(Remote host closed the connection)]
Sonarpulse has quit [(Ping timeout: 255 seconds)]
phreedom has joined joined #nixos-dev
hedning[m] has joined joined #nixos-dev
nocent has joined joined #nixos-dev
pstn has joined joined #nixos-dev
hl has joined joined #nixos-dev
olejorgenb[m] has joined joined #nixos-dev
stites[m] has joined joined #nixos-dev
sphalerite has joined joined #nixos-dev
regnat[m] has joined joined #nixos-dev
copumpkin has joined joined #nixos-dev
grahamc has joined joined #nixos-dev
rycee has joined joined #nixos-dev
adisbladis[m] has joined joined #nixos-dev
peterhoeg has joined joined #nixos-dev
FRidh[m] has joined joined #nixos-dev
moredread[m] has joined joined #nixos-dev
<bgamari> LnL, are you sure that is in fact the package?
<bgamari> LnL, it doesn't actually appear to be exposed in haskellPackages
mbrgm has quit [(Ping timeout: 240 seconds)]
mbrgm has joined joined #nixos-dev
mbrgm has quit [(Ping timeout: 240 seconds)]
mbrgm has joined joined #nixos-dev
ris has quit [(Ping timeout: 264 seconds)]
gleber_ has quit [(Ping timeout: 246 seconds)]
cbarrett has quit [(Ping timeout: 246 seconds)]
FRidh[m] has quit [(Ping timeout: 246 seconds)]
pstn has quit [(Ping timeout: 246 seconds)]
stites[m] has quit [(Ping timeout: 240 seconds)]
peterhoeg has quit [(Ping timeout: 255 seconds)]
nocent has quit [(Ping timeout: 252 seconds)]
grahamc has quit [(Ping timeout: 252 seconds)]
copumpkin has quit [(Ping timeout: 240 seconds)]
sphalerite has quit [(Ping timeout: 240 seconds)]
adisbladis[m] has quit [(Ping timeout: 248 seconds)]
regnat[m] has quit [(Ping timeout: 276 seconds)]
hl has quit [(Ping timeout: 240 seconds)]
florianjacob has quit [(Ping timeout: 255 seconds)]
rycee has quit [(Ping timeout: 250 seconds)]
moredread[m] has quit [(Ping timeout: 276 seconds)]
hedning[m] has quit [(Ping timeout: 276 seconds)]
olejorgenb[m] has quit [(Ping timeout: 276 seconds)]
gleber_ has joined joined #nixos-dev
orivej has joined joined #nixos-dev
FRidh has joined joined #nixos-dev
vcunat has joined joined #nixos-dev
stqism has joined joined #nixos-dev
<stqism> I know it's only partially related, but how do I go about getting a nix package that's part or Linux built with macos by hydra?
<vcunat> stqism: hydra.nixos.org with nixpkgs? Or your nix/hydra?
<stqism> vcunat like nixpkgs hydra
<stqism> The package is in the actual repo, it just doesn't build for os x
<vcunat> stqism: by meta.platforms
<vcunat> depending what makes sense, it can be e.g. = platforms.unix
<vcunat> or = platforms.linux + platforms.darwin;
<stqism> Okay, wait. All I have to do is test it locally and submit a pr with it set to unix?
<vcunat> s/+/++/
<stqism> That doesn't sound bad at all tbh
goibhniu has joined joined #nixos-dev
<stqism> Hype, I can finally move from brew to nix 100%
<adisbladis> <3
cbarrett has joined joined #nixos-dev
<vcunat> well, yes, it's simple
<vcunat> Mostly. Sometimes the package needs a fix for the different platform.
<vcunat> I must admit that x86_64-linux is in a bit better shape than x86_64-darwin.
<stqism> Yeah, that's a lot of why I want to help there
<stqism> Brew is great
<stqism> At fucking you over with garbage sprinkled everywhere
<vcunat> Nice.
<vcunat> I think contributors are the main missing part for darwin in nixpkgs.
<vcunat> (in comparison to linux)
<stqism> Yeah, the general difference in contributors from like brew to nix is like 10 fold
<stqism> So brew gets tons of love, but nix is cooler
<stqism> I had to fork archey to add nix support tbh
vcunat has quit [(Ping timeout: 246 seconds)]
vcunat has joined joined #nixos-dev
florianjacob has joined joined #nixos-dev
__Sander__ has joined joined #nixos-dev
hedning[m] has joined joined #nixos-dev
pstn has joined joined #nixos-dev
hl has joined joined #nixos-dev
olejorgenb[m] has joined joined #nixos-dev
stites[m] has joined joined #nixos-dev
sphalerite has joined joined #nixos-dev
nocent has joined joined #nixos-dev
regnat[m] has joined joined #nixos-dev
grahamc has joined joined #nixos-dev
copumpkin has joined joined #nixos-dev
adisbladis[m] has joined joined #nixos-dev
rycee has joined joined #nixos-dev
peterhoeg has joined joined #nixos-dev
FRidh[m] has joined joined #nixos-dev
moredread[m] has joined joined #nixos-dev
FRidh has quit [(Remote host closed the connection)]
jtojnar has quit [(Quit: jtojnar)]
<gchristensen> vcunat: do you think the platform check changing from abort to throw is disputable
<vcunat> no reason in particular
<vcunat> maybe I just miss some reason why people used abort in that many cases
<vcunat> And I meant it (also) as comparison to the typo changes where it's clear.
<gchristensen> right
<vcunat> gchristensen: you merged the typo fixes to master but haven't backported to 17.09, right?
<gchristensen> maybe rebase https://github.com/NixOS/nixpkgs/pull/32365 against master, which already has the checkMeta fixes and the fixup for it
<gchristensen> well I have both commits backported, trying to decide if I push the backport of the abort -> throw :)
<vcunat> oh, that part is in master as well?
<gchristensen> no it isn't
<gchristensen> so I guess that is my answer
<vcunat> :-)
<vcunat> master first in any case
<vcunat> abort -> throw can probably wait for the rest of the PR
<vcunat> it doesn't seem to get more attention, so I'll probably finish and merge it by the end of this week
<gchristensen> being able to run borg locally would be nice
vcunat has quit [(Ping timeout: 246 seconds)]
<gchristensen> vcunat: I think you'll need to rebase p/check-meta for the diff to stop showing the meta changes in the PR
<gchristensen> and should probably delete treewide: checkMeta fixes for x86_64-linux / fixup! treewide: checkMeta fixes for x86_64-linux
<gchristensen> niksnut: I'm coming up with a whole laundry list of patches for you :$
makefu has quit [(Ping timeout: 250 seconds)]
vcunat has joined joined #nixos-dev
makefu has joined joined #nixos-dev
ma27 has joined joined #nixos-dev
JosW has joined joined #nixos-dev
ma27 has quit [(Quit: WeeChat 1.9.1)]
ckauhaus has joined joined #nixos-dev
ma27 has joined joined #nixos-dev
ma27 has quit [(Client Quit)]
vcunat1 has joined joined #nixos-dev
<vcunat1> sphalerite: you use some kernel override, right?
vcunat has quit [(Ping timeout: 246 seconds)]
<vcunat1> on 281e979da I can build linuxPackages_4_14.nvidia_x11
<vcunat1> /nix/store/lv5hlapivfdwcg2fb65jkh3z47j3dm33-nvidia-x11-384.98-4.14.3
vcunat1 is now known as vcunat
vcunat has quit [(Client Quit)]
vcunat has joined joined #nixos-dev
<sphalerite> i don't think so…
<vcunat> is the derivation the same?
<vcunat> /nix/store/gws9lk3j9m1289xc6bpdabx3cwrs51vj-nvidia-x11-384.98-4.14.3.drv
* vcunat lost the log from before
<vcunat> (the IRC log)
dtzWill has joined joined #nixos-dev
<sphalerite> nixpkgs 17.09.2281.b4a0c011e81; build log http://sprunge.us/EFQg; derivation is /nix/store/4v1axb2dv9c0vbri9h8jhypvwxd99c7p-nvidia-x11-384.90-4.14.3.drv
<sphalerite> vcunat: ^
<vcunat> sphalerite: I see
<vcunat> I tested bad commit
<vcunat> but the thing has been fixed by me on 17.09
<sphalerite> yeah I was just about to say, I don't have tht commit in my nixpkgs repo
<vcunat> nixos-17.09-small contains it already
<sphalerite> I just fetched from the nixpkgs-channels repo and still don't have it…
<sphalerite> nixos-17.09-small is still on 68a8b335c860fe72a8b401a7d40cba147fa0ac2b as far as I can see
<vcunat> the first 17.09 commit with the fixes is 00b92751f6
<vcunat> sphalerite: ^^
<sphalerite> but yeah release-17.09 is on the commit you mentioned
<sphalerite> ah ok, thanks!
<aminechikhaoui> niksnut: nix bump ? :)
<aminechikhaoui> in 17.09 if possible
ma27 has joined joined #nixos-dev
<niksnut> aminechikhaoui: ok
ma27 has quit [(Ping timeout: 255 seconds)]
ma27 has joined joined #nixos-dev
ma27 has quit [(Ping timeout: 258 seconds)]
ma27 has joined joined #nixos-dev
<orivej> vcunat: could you try restarting the aborted jobs in https://hydra.nixos.org/eval/1416829? (I don't know why they are aborted)
ma27 has quit [(Ping timeout: 246 seconds)]
ma27 has joined joined #nixos-dev
taktoa has quit [(Remote host closed the connection)]
ma27 has quit [(Ping timeout: 246 seconds)]
ma27 has joined joined #nixos-dev
goibhniu has quit [(Ping timeout: 246 seconds)]
FRidh has joined joined #nixos-dev
<FRidh> how come certain users get a stack overflow and others not? https://github.com/NixOS/nixpkgs/commit/859dc02fa3f42bca9b7770eb96f4fd8224cf1576
orivej has quit [(Ping timeout: 246 seconds)]
<gchristensen> when using the nixos test driver, where all does it store state?
<vcunat> FRidh: there's something strange around this on my machine
<vcunat> (apparently)
<vcunat> now I'm also getting the same error when I put nix.package = nixUnstable in my configuration.nix
<vcunat> (and it disappears when I delete that override)
ma27 has quit [(Ping timeout: 246 seconds)]
ma27 has joined joined #nixos-dev
<FRidh> vcunat: cache of nix?
<vcunat> FRidh: what do you mean?
<FRidh> vcunat: If I am correct Nix caches results of an evaluation. Could it be related to that?
<vcunat> cache from previous runs?
<vcunat> I don't think nix does that for *evaluation*
<gchristensen> I think not
<gchristensen> otherwise it wouldn't use 19,000,000 GB of RAM to eval multiple machines
<vcunat> so many zeros?
<vcunat> Where do you get a machine with that much RAM?
<FRidh> What else could be causing the kind of "state" you were experiencing?
<vcunat> allowUnfree = true; ?
<vcunat> I'll try
<vcunat> no, disabling that didn't help
makefu has quit [(Quit: WeeChat 1.9.1)]
<vcunat> nuking whole config.nix didn't help either
__Sander__ has quit [(Quit: Konversation terminated!)]
<FRidh> I suppose the following line is going to need some optimization then: https://github.com/NixOS/nixpkgs/blob/master/pkgs/top-level/python-packages.nix#L93
makefu has joined joined #nixos-dev
<stqism> 9:06 AM <vcunat> haveUnlimitedRam = true; ?
<vcunat> How could I've missed that option :-)
<gchristensen> import (fetchurl { url = "https://downloadmoreram.com })
<vcunat> gchristensen: it doesn't work
<vcunat> :-(
<vcunat> > The uploader has not made this video available in your country.
* gchristensen didn't even bother to close the quote
<vcunat> I really hate geo-sensitive stuff.
<vcunat> Maybe it's related to trying to implement EDNS client subnet...
<domenkozar> gchristensen: what's the problem with two projects pinning nixpkgs?
<gchristensen> well it is a bit of a nightmare
<gchristensen> I should have probably solved this ages ago with overlays
<domenkozar> if you use pinning solution as per nixconf, each project just imports local nixpkgs
<gchristensen> yeah but so
<gchristensen> heh
<gchristensen> I'm using one project to add dependencies and things to the NixOS test framework code used in the other, and in some cases override dependencies in the other one. it is a real pickle and not elegant
<domenkozar> seems like a separation of concerns issue
<gchristensen> there are lots of concerning issues tbh
<domenkozar> but hard to point fingers without seeing code :)
<domenkozar> gchristensen: like any software that is used :)
<gchristensen> :D
<gchristensen> I suspect biting the bullet and moving to overlays will be the right thing
ma27 has quit [(Ping timeout: 240 seconds)]
ma27 has joined joined #nixos-dev
ma27 has quit [(Client Quit)]
ma27 has joined joined #nixos-dev
jtojnar has joined joined #nixos-dev
orivej has joined joined #nixos-dev
goibhniu has joined joined #nixos-dev
Sonarpulse has joined joined #nixos-dev
FRidh has quit [(Quit: Konversation terminated!)]
ris has joined joined #nixos-dev
ris has quit [()]
romildo has joined joined #nixos-dev
romildo has quit [(Quit: Leaving)]
<jtojnar> why would I get a notification about https://hydra.nixos.org/build/65215116 today?
JosW has quit [(Quit: Konversation terminated!)]
<vcunat> lately I sometimes receieved notification about several months old builds
<vcunat> jtojnar: no idea why that happens
vcunat has quit [(Quit: Leaving.)]
makefu has quit [(Ping timeout: 250 seconds)]
goibhniu has quit [(Ping timeout: 268 seconds)]
makefu has joined joined #nixos-dev
ckauhaus has quit [(Quit: Leaving.)]
<bgamari> Sonarpulse, Do you see how this could happen?
<Sonarpulse> I hope I didn't fuck up my _FOR_BUILD pr
<bgamari> Sonarpulse, you mean 43e113739786ed51de7e829637493a58273cdf02
<bgamari> looks reasonable to me
<Sonarpulse> yeah
<Sonarpulse> what is your rebased branch?
<Sonarpulse> bgamari: ^
<bgamari> just force pushed to ben-cross
<Sonarpulse> ok
<bgamari> Sonarpulse, ^]
<Sonarpulse> hahah
<Sonarpulse> bgamari: oh you got rid of cross-elegant
<Sonarpulse> yeah it's gonna be fucked up
<Sonarpulse> without that
<bgamari> ahh
<Sonarpulse> btw
<Sonarpulse> I want to move the goalpost on my GCC pr :D
* bgamari still somehow hasn't quite developed a sense for the dependency tree of these PRs
<Sonarpulse> hehe yeah it sucks sorry
<bgamari> naw, just hard
<Sonarpulse> left to right, bottom to top is a valid toposort
<bgamari> so I guess the question is where does the line between "essential" and "nice to have" fall?
<bgamari> up to and including The Big PR?
<Sonarpulse> bintools wrapper actually might be more nice to have
<Sonarpulse> but I've rebased so many things on it
<Sonarpulse> that one there ^
<Sonarpulse> I know its "safe" in that it keeps the good commit working
<Sonarpulse> but it doesn't fix the bad commit either
<Sonarpulse> I *think* it is all good changes, the remaining badness of the bad commit is unrelated stuff like --rpath vs --rpath-link
<Sonarpulse> and mingw stuff
<Sonarpulse> but I feel bad merging a mass rebuild that only *seems* like an improvement
* bgamari doesn't mind depending upon a mass rebuild
<bgamari> since it's just me
<Sonarpulse> well i have no qualms rebasing my own stuff
<Sonarpulse> since i already did
<Sonarpulse> ...but i like merging things
<Sonarpulse> so I don't need to think about rebasing them
<bgamari> understandable
<Sonarpulse> and I'd give it like a 70% chance my builder.sh changes there
<Sonarpulse> actually are sufficient
<bgamari> those are good odds :)
<Sonarpulse> --rpath fix and whatnot being orthogonal
<Sonarpulse> I'll merge :)
<Sonarpulse> to hell with it
<bgamari> so I guess I should give rebasing cross-elegant another shot?
<bgamari> and then rebase on top of that?
<bgamari> Sonarpulse, note that there's no need to hurry on my account; I'm just wondering whether there is a way I can make progress on my cross branch
<Sonarpulse> bgamari: I did do a rebase
<bgamari> since I am rapidly approaching the point in the project where I will be blocked on it
<Sonarpulse> of much of this
<Sonarpulse> including my old rotted rebased-to-death ben-cross
<bgamari> Sonarpulse, you mean you rebased cross-elegant?
<Sonarpulse> yes!
<Sonarpulse> over my old gcc change
<bgamari> hmm, odd
<bgamari> GitHub seems to think the that the last commits were added on Aug 28
<bgamari> ahh
<bgamari> so you did
<bgamari> it even rebased onto recent master clealy
<bgamari> cleanly*
<gchristensen> niksnut: is there a reason you didn't merge my backport of the H.S. fixes to nix 1.1?
<bgamari> Sonarpulse, rebased and building
<bgamari> let's see how this goes
<Sonarpulse> bgamari: oh rebased cross-elegant?
<Sonarpulse> I just did that lol
<bgamari> and my branch on top
<Sonarpulse> bgamari: well just rebased everything on my end
<Sonarpulse> and push
<Sonarpulse> *pushed
<Sonarpulse> heh I wonder if we got the same thing
<bgamari> Sonarpulse, hmmm, with just plain rebased cross-elegant I get an assertion failure from
<bgamari> gccCrossStageStatic = assert targetPlatform != buildPlatform; let
<bgamari> in all-packages.nix
<Sonarpulse> hmm
<bgamari> strangely enough with my branch on top I don't get this assertion
<Sonarpulse> plain rebased cross-elevant being what I just pushed?
<bgamari> but it fails for other reasons and I don't know what I would have done to fix this
<Sonarpulse> not the attribute for me to instantiate?
<Sonarpulse> note at niksnut's request I removed some __ pefixes
<Sonarpulse> so not more __targetPackages
<Sonarpulse> or __deps*
<Sonarpulse> just targetPackages
<Sonarpulse> and deps*
clever has joined joined #nixos-dev
<bgamari> Sonarpulse, I get the same assertion failure with your cross-elegant branch
<Sonarpulse> bgamari: sorry meant "got" not "not"
<Sonarpulse> i.e. how can I reproduce?
<bgamari> hmm
<bgamari> let me try to extract something
<Sonarpulse> bgamari: ok
<Sonarpulse> bgamari: https://hydra.nixos.org/jobset/nixpkgs/ericson2314-cross-trunk evaluates mostly fine with cross-elegant
<Sonarpulse> got to add mips LDEMULATION
<Sonarpulse> that's the only issue
<bgamari> Sonarpulse, git@github.com:bgamari/nix-cross-test.git
<Sonarpulse> ok
<bgamari> nix-build . -A knownGoodCrossed
<bgamari> after checking out nix-pkgs in the parent directory
<Sonarpulse> you can get from the lib
<bgamari> or change `nix` to taste
<bgamari> ahh
<Sonarpulse> bgamari: weird
<Sonarpulse> nix-instantiate --arg crossSystem '(import ./lib).systems.examples.armv7l-hf-multiplatform' -A shadow
<Sonarpulse> fails
<Sonarpulse> nix-instantiate --arg crossSystem '(import ./lib).systems.examples.armv7l-hf-multiplatform' -A glibcCross.bin succeeds
<bgamari> hmm
<bgamari> that is indeed odd
<Sonarpulse> maybe my splicing is fucked up?
* bgamari is trying to work out how we are supposedly ensuring that gccCrossStageStatic isn't called during cross compilation
<Sonarpulse> bgamari: gccCrossStageStatic
<bgamari> afterall crossLibcStdenv relies on it
<Sonarpulse> should be rarely refered to at all
* Sonarpulse checks
<bgamari> right above gccCrossStageStatic
<bgamari> glibcCross indeed relies on crossLibcStdenv
<Sonarpulse> indeed
<Sonarpulse> bgamari: but shadow has a run-time dep on glibcCross
<bgamari> ahh
<bgamari> yes
<Sonarpulse> so if it uses glibcCross when build != host
<bgamari> that might be why my branch works around this
<Sonarpulse> then the assert should succeed
<Sonarpulse> hmm
<bgamari> oh
<bgamari> I see
ma27 has quit [(Ping timeout: 246 seconds)]
<bgamari> IIRC I removed the dependency of shadow on glibc
<Sonarpulse> yeah
<Sonarpulse> bgamari: I did in a repl
<Sonarpulse> `callPackage ({ glibcCross }: glibcCross) {}`
<Sonarpulse> which fails
<Sonarpulse> so there's some weird splicing thing happenign
<bgamari> ahh
<bgamari> that will be exciting to track down ;)
<Sonarpulse> indeed, sigh