<samueldr>
which in turn has two links, one "Latest successful build from a finished evaluation", which is dated 2018-01-17
<samueldr>
so uh, channels advance since the job doesn't need aarch64 I guess
Dezgeg has quit [Ping timeout: 276 seconds]
<samueldr>
eh, I'm still confused by the interface, it looks like it should still be good, don't know why binutils would be rebuilding
<tomberek>
it would be nice to have an interface or a way to ascertain what has a cache. nix-build --dry-run -iA nixpkgs.hello is the best I've got so far
<samueldr>
it's possible that once aarch64 support is "in a reasonable state" it won't happen much
<samueldr>
(if from 18.03 aarch64 is supported, it maybe will block channels?)
orivej has joined #nixos-aarch64
orivej has quit [Ping timeout: 248 seconds]
Dezgeg has joined #nixos-aarch64
Dezgeg has quit [Client Quit]
Dezgeg has joined #nixos-aarch64
<gchristensen>
thats right samueldr
orivej has joined #nixos-aarch64
<tomberek>
I'm trying to set up a nix-serve of the arm7l builds (the cs.helsinki cache came back online) on a non-arm machine? What part needs the gcroot added? the drv?
<tomberek>
That does seem to prevent gc of the closure after nix-copy-closure
vcunat has joined #nixos-aarch64
<tomberek>
At the moment I'm using the arm machine to get the closures via nix-build from the cs.helsinki cache, the copying them to the x86 store. Can that just be done directly?
<vcunat>
you can nix-build on the x86 machine directly
<vcunat>
substitutes work cross-arch
<vcunat>
(I've already inspected some darwin paths from cache.nixos.org on my nixos machine.)
vcunat has quit [Quit: Leaving.]
Tasqa has quit [Ping timeout: 268 seconds]
<Dezgeg>
now we have chromium theoretically building... let's see if hydra manages under the 10 hour limit