<angerman>
Sonarpulse I don’t understand your coreutils comment.
<angerman>
Sonarpulse there simply are no coreutils when cross compiling to mingw, as that package just can’t be built. If the expression is non-empty we try to build them.
pie__ has quit [Remote host closed the connection]
pie__ has joined #nixos-dev
pie_ has joined #nixos-dev
pie__ has quit [Read error: Connection reset by peer]
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-dev
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-dev
lassulus_ has joined #nixos-dev
lassulus_ is now known as Random_int
lassulus has quit [Ping timeout: 260 seconds]
Random_int is now known as lassulus
orivej has joined #nixos-dev
pie_ has quit [Remote host closed the connection]
pie_ has joined #nixos-dev
pie_ has quit [Read error: Connection reset by peer]
pie_ has joined #nixos-dev
pie_ has quit [Read error: Connection reset by peer]
pie_ has joined #nixos-dev
orivej has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #nixos-dev
orivej_ has joined #nixos-dev
orivej has quit [Ping timeout: 260 seconds]
pie__ has joined #nixos-dev
pie_ has quit [Read error: Connection reset by peer]
orivej_ has quit [Ping timeout: 255 seconds]
pie__ has quit [Read error: Connection reset by peer]
pie__ has joined #nixos-dev
pie__ has quit [Read error: Connection reset by peer]
pie___ has joined #nixos-dev
pie__ has joined #nixos-dev
pie___ has quit [Read error: Connection reset by peer]
pie__ has quit [Ping timeout: 245 seconds]
FRidh has joined #nixos-dev
sir_guy_carleton has quit [Quit: WeeChat 2.0]
sorear_ has joined #nixos-dev
gleber__ has joined #nixos-dev
angerman has quit [Ping timeout: 276 seconds]
sorear has quit [Ping timeout: 276 seconds]
sorear_ is now known as sorear
angerman has joined #nixos-dev
simpson has quit [Ping timeout: 276 seconds]
gleber_ has quit [Ping timeout: 276 seconds]
gleber__ is now known as gleber_
{^_^} has quit [Ping timeout: 276 seconds]
Cale has quit [Ping timeout: 276 seconds]
flokli has quit [Ping timeout: 276 seconds]
simpson has joined #nixos-dev
flokli has joined #nixos-dev
Cale has joined #nixos-dev
FRidh has quit [Quit: Konversation terminated!]
{^_^} has joined #nixos-dev
{^_^} has joined #nixos-dev
{^_^} has quit [Changing host]
goibhniu has joined #nixos-dev
orivej has joined #nixos-dev
__Sander__ has joined #nixos-dev
orivej has quit [Ping timeout: 240 seconds]
orivej has joined #nixos-dev
orivej has quit [Ping timeout: 256 seconds]
<gchristensen>
jtojnar: should we revert that PR?
<gchristensen>
I'm afraid we'll never know what all is possibly broken due to it unless we get the hash changes to 0
<gchristensen>
it means a fixed-output derivation changed its contents but not the output hash
<gchristensen>
so the `.drv` for the package changed, but the output path for the package did not change
<gchristensen>
so like an update to curl will cause all the fetchurl drvs to change but the sha256sum is the same, so the outpath will not change
<jtojnar>
gchristensen: but mbedtls (polarssl) is not a fixed output derivation
<gchristensen>
yeah, the effect cascades
<gchristensen>
the drv for a fixed-output input changed, causing the drv for the package to change... however the inputs still have the same outputs, causing the outputs for mbedtls to stay the same
<jtojnar>
but would not that mean one of its dependencies would be in the gist too?
<gchristensen>
I would think so :$
<jtojnar>
nope `nix-instantiate -A polarssl` still results in the same drv path
<gchristensen>
then I don't know, other than maybe your testing strategy isn't working as you'd expect
<gchristensen>
but I don't know, it could be something weird in ofborg that I've not seen before
<gchristensen>
their slack org is quite good, and btw I don't work for packet... I just really _like_ packet :)
<angerman>
gchristensen: it does look attractive.
<clever>
angerman: and using targetEnv = "none"; you can just point nixops at any machine you have root ssh on, and it will take over control of the box
<gchristensen>
I can't recommend it enough
<clever>
any nixos machine*
<gchristensen>
also their 2x bonded 10GB NICs drool.jpg
<angerman>
clever: we use aws for our ci, right?
<angerman>
gchristensen: if they had a SG location that would be neat.
<clever>
angerman: yeah, buildkite and hydra are all in aws currently
<bgamari>
Am I misunderstanding the semantics of passing an expression on the command line or is there really a bug here?
<{^_^}>
nix#2298 (by bgamari, open): `nix build -f` appears not to work when invoked with an expression
<clever>
bgamari: i dont think you can mix file and expression modes
<clever>
bgamari: nix-build and nix-shell also dont let you mix files and expressions
<bgamari>
clever, I guess I just don't see the reason why this should be the case
<bgamari>
Intuitively I would have thought that -f would define the context in which the expression is evaluated
<clever>
bgamari: -f tells it to just eval 'import <nixpkgs> {}'
<clever>
and to use whatever that returns
<clever>
and the expression, is just something it should eval, and use whatever it returned
<clever>
but it cant use both products at once
<bgamari>
right, if no expression is given then it seems reasonable to default to "the whole thing"
<clever>
you can also add -A to select an attribute in the resulting set
<bgamari>
in `nix build`?
<bgamari>
hmm, indeed
<bgamari>
but I guess that doesn't admit a general expression
<clever>
`nix-build '<nixpkgs>' -A hello` will build hello world
<clever>
`nix-build -E 'let pkgs = import <nixpkgs> {}; in { hello = pkgs.hello; }' -A hello` will read the hello attribute from the set defined in -E
<bgamari>
It seems to me like the semantics that I propose would be a backwards compatible change
<bgamari>
and would make -f and expressions interact far more intuitively
<clever>
and in the above case, only hello can be built
<clever>
also note, -A in nix-build/nix-shell behaves a lot more weirdly then you expect
<clever>
if you do -A foo.bar.baz, it will check if foo is a function, and call it with everything passed via --arg
ixxie has joined #nixos-dev
<clever>
then it will check for a .bar in the return value, and if its a function, repeat the same thing
<clever>
recursively
<bgamari>
right, this is why I'd rather just evaluate an expression
<clever>
i discovered that when trying to eval a nixops expression from another nix file
* bgamari
doesn't mind being explicit, but doing so is currently needlessly verbose
<bgamari>
clever, it's just quite surprising that `nix build -f <nixpkgs> hello` works, yet `nix build -f <nixpkgs> (hello)` does not
<bgamari>
On this note, when I type `nix build nixpkgs.hello`, where does the `nixpkgs` binding come from?
<bgamari>
Is this due to the NIX_PATH="nixpkgs=..." entry?
<clever>
`nix build` is a bit weird, in that it auto-detects if the string is an expression or an attribute path
<clever>
and the nixpkgs in your attr path is from NIX_PATH
<bgamari>
right
erickomoto has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
erickomoto has joined #nixos-dev
erickomoto has quit [Client Quit]
erickomoto has joined #nixos-dev
erickomoto has quit [Client Quit]
erickomoto has joined #nixos-dev
orivej has quit [Ping timeout: 276 seconds]
erickomoto has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
orivej has joined #nixos-dev
sir_guy_carleton has joined #nixos-dev
pie__ has joined #nixos-dev
erickomoto has joined #nixos-dev
pie__ has quit [Read error: Connection reset by peer]
pie__ has joined #nixos-dev
ckauhaus has joined #nixos-dev
Lisanna has quit [Quit: Lisanna]
ckauhaus has quit [Client Quit]
ckauhaus has joined #nixos-dev
Lisanna has joined #nixos-dev
Lisanna has quit [Client Quit]
pie__ has quit [Read error: Connection reset by peer]
pie__ has joined #nixos-dev
pie__ has quit [Read error: Connection reset by peer]
pie__ has joined #nixos-dev
ckauhaus has quit [Quit: WeeChat 2.0]
orivej has quit [Ping timeout: 245 seconds]
jtojnar has quit [Remote host closed the connection]
ixxie has quit [Ping timeout: 264 seconds]
<timokau[m]>
Can I somehow find out about which hardware the hydra runner `lucifer.ewi.tudelft.nl` is using?
<gchristensen>
what do you want to know?
<gchristensen>
timokau[m]: ^
<timokau[m]>
The ntl build is failing after I upgraded it. I suspect it has something to do with the tune parameter which I set to "x86" (if the system is x86). It fails with "illegal instruction"
<timokau[m]>
So it would be nice to have details about the CPU of that builder. It built fine locally and on ofBorg.
<gchristensen>
its definitely x86
<gchristensen>
iirc its an intel e5620
<timokau[m]>
Yeah I'm setting the `TUNE` parameter dependent on `stdenv.targetPlatform.isx86`, so as long as that works it has to be x86. The specifics would help reporting that to the author.
<timokau[m]>
Also is it possible to make a change to the build and test that on that builder? To make sure the TUNE parameter is responsible.
<timokau[m]>
Alright thanks! Are you just speaking from memory or is that documented somewhere?
<gchristensen>
just things I think I remember hearing :)
<gchristensen>
but it is without a doubt an x86 machine
<timokau[m]>
Okay thanks, I'll report that to the author. What about making sure a fix is tested on that build server? Or conversely run the current build on a different one?
<gchristensen>
we can't specify a fix be tested on that particular builder, unfortnately :/
FRidh has quit [Quit: Konversation terminated!]
<timokau[m]>
Alright thanks again. I've pinged the author about it (#43679), maybe he'll know more.
<timokau[m]>
And all the best for your upcoming surgery from some stranger on the internet :)
<gchristensen>
thanks!
erickomoto has quit [Ping timeout: 276 seconds]
jtojnar has joined #nixos-dev
pie_ has joined #nixos-dev
pie__ has quit [Read error: Connection reset by peer]
aminechikhaoui has quit [Quit: leaving]
aminechikhaoui has joined #nixos-dev
<domenkozar>
new day, new two solutions for broken macos builds :)
<domenkozar>
I wonder how far we can perfect this :P
sir_guy_carleton has quit [Quit: WeeChat 2.0]
pie__ has joined #nixos-dev
pie_ has quit [Read error: Connection reset by peer]