I find it extremely unlikely we'll ever package two different patch
versions of the same minor Rust version. New patch versions should
generally be compatible, so we shouldn't give people the expectation
they'll be able to pin to one. And by including the patch version in
the attribute name, we'd have to change the attribute every time a
patch version was released, which would unnecessarily create diff
noise and maintenance headaches for what should be a seamless upgrade.
The patch is included in rust master[1], but neither that, nor the
QuiltOS version we were using previously, apply to 1.42.0, so I've
included the fixed version here.
[1]: 4f15867faf
The Hydra build [1] failed because it was unable to link to `LLVM9`; add
`llvmShared` to `passthru` in order to stay up to date with required
LLVM versions. Also quote the homepage URLs, since that's preferred.
[1] https://hydra.nixos.org/build/112989779/nixlog/1
I thought about doing a seperate output for these, but they're tiny
compared to the size of the binary, so there's no point.
(cherry picked from commit 0489c1b4b2)
Cargo uses git-rs which is made to be built against the bundled libgit2
version that hasn't been part of a stable release yet. Using our libgit2
instead of the master version fails during runtime as they are not
compatible anymore.
After the next libgit2 update we can try again but it is likely that
there will also be yet another cargo release at that point in time…
This has several advantages:
1. It takes up less space on disk in-between builds in the nix store.
2. It uses less space in the binary cache for vendor derivation packages.
3. It uses less network traffic downloading from the binary cache.
4. It plays nicely with hashed mirrors like tarballs.nixos.org, which only
substitute --flat hashes on single files (not recursive directory hashes).
5. It's consistent with how simple `fetchurl` src derivations work.
6. It provides a stronger abstraction between input src-package and output
package, e.g., it's harder to accidentally depend on the src derivation at
runtime by referencing something like `${src}/etc/index.html`. Likewise, in
the store it's harder to get confused with something that is just there as a
build-time dependency vs. a runtime dependency, since the build-time
src dependencies are tarred up.
Disadvantages are:
1. It takes slightly longer to untar at the start of a build.
As currently implemented, this attaches the compacted vendor.tar.gz feature as a
rider on `verifyCargoDeps`, since both of them are relatively newly implemented
behavior that change the `cargoSha256`.
If this PR is accepted, I will push forward the remaining rust packages with a
series of treewide PRs to update the `cargoSha256`s.
This was only introduced in 1.40.0 and doesn't work on older versions.
thread 'main' panicked at 'Error: no rules matched rustc-dev.', src/bootstrap/builder.rs:231:21
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace.
failed to run: /build/rustc-1.38.0-src/build/bootstrap/debug/bootstrap dist rustc-dev
Tests have been disabled since over a year and now the
code starts to bit-rot. As it seems unlikely that they
will come back in near future, let's just remove it.
There ver very many conflicts, basically all due to
name -> pname+version. Fortunately, almost everything was auto-resolved
by kdiff3, and for now I just fixed up a couple evaluation problems,
as verified by the tarball job. There might be some fallback to these
conflicts, but I believe it should be minimal.
Hydra nixpkgs: ?compare=1538299
