[Nix-dev] Re: [Nix-commits] SVN commit: nix - 22282 - MarcWeber - nixpkgs/branches/stdenv-updates/pkgs/stdenv/generic

[Lluís Batlle i Rossell]

On Thu, Jun 17, 2010 at 01:31:42AM +0200, Marc Weber wrote:
> Can we shorten this discussion?
> 
> Somehow I have to decide whether I should keep this in a private branch
> or not. It is important to me so I will keep and maintain it anyway.

After talking a bit with Ludovic, we achieved a consensus at least between he
and me, that I'll try to summarize:
1. Hydra does not need the in-build parallelization, she has enough with the
inter-build parallelization
2. There is no way for nix to manage the balance between in-build and
inter-build parallelization (make -j, and nix max-jobs).
3. Nix developers, wanting a quick build for testing, can add the "make -j X"
temporarily before commiting the expression.
4. The benefits of your change come at the expense of some clarity in stdenv,
and some impurities, and in the nix world many don't like impurities much.
5. By default, generic builder expressions should not be built parallelizing at
all.

We think that the in-build parallelization should come only once linked properly
with the inter-build parallelization, maybe with help of some 'derivation'
arguments and some decided-by-nix builder environment variables. Then,
particular nixpkgs jobs could get benefit from that parallelization.
A simple policy could be "give in-build parallelization if nix cannot give
inter-build parallelization".

> 
> If I have to keep it in a separate branch I can longer use the buildfarm
> (or I have to build / setup my own..)
> 
> We have 2 options:
> a) revert
> b) improve
I'm for the revert, considering what I wrote above. I prefer a
nix/generic-builder clean, as I see little benefits in the current
implementation.

Regards,
Lluís.