pennae

would definitely be interesting to have another look at the performance impact now. not having gone over it again completely this doesn't *look* like it should cause performance regressions, so...

we've run this PR through our perf benchmarks, and while we do see an increase in instructions executed the real-world impact on walltime is in the noise. turning the string_view...

it's important to note that both of these are just syntactic sugar, they're merged to `{ a = rec { b = 5; c = b; }; }` during parsing.

yeah, that is a potential problem :confused: there are a few nixos modules that use floats, but those shouldn't be impacted too much because upgrading nix would most likely rebuild...

this also affect `toString` in a different way. `toString` always produces six fractional digits, rounding small floats like 1e-9 to 0.

https://github.com/NixOS/nix/blob/master/src/libexpr/primops.cc#L2511 is probably the culprit here

looking at other builtins: most of them do not set positions for attributes they create. this isn't currently fixable without a lot of `const_cast`ing (which might make merging a fix...

it would be good to have positions on attributes from builtins though, since *every* instance of `recursiveUpdate` now doesn't have positions on its attrs any more. there may be more...

we're open to moving the jobserver prototype we made in https://github.com/NixOS/nixpkgs/pull/143820 into nix itself. unfortunately the gnu jobserver protocol isn't universally supported, ghc for example uses semaphores instead of pipes....

affected and can confirm. then AMD T14 Gen2 spins the fan up very briefly even when turning it off. this patch is sufficient to stop the 2-minutely spinup while preventing...