Commit graph

3649 commits

Author SHA1 Message Date
Andreas Rammhold fa6c5c3dc7
buildBazelPackage: prime fixed output derivations
All bazel fixed output derivations should be specific to the bazel
version that was used to generate them. There is not guarantee that the
build will still succeed or reproduces (without the cached fixed output)
if the fetch phase wasn't rerun with a different bazel version.

In the past bazel had been bumped but not all those packages that have
fixed outputs from bazel builds. This lead to compiling and somewhat
working TF versions that couldn't be reproduced without the cached fixed
outputs.
2020-03-03 01:12:26 +01:00
Michael Lingelbach 34296980d1 tensorflow: 1.15.0 -> 1.15.1
* Apply glibc 2.3 patch
* build tensorflow with bazel_1
* Bump openssl version to 1.1
2020-03-02 10:07:37 +01:00
Eelco Dolstra 4c0821461e
enable-coverage-instrumentation.sh: Fix unbound variable error
https://hydra.nixos.org/build/113299582
2020-02-28 12:31:50 +01:00
zimbatm b32a057425
srcOnly: remove the ellipsis (#80903)
The function doesn't make use of the extra arguments.
2020-02-24 23:12:37 +00:00
Samuel Dionne-Riel 972678ec74
Merge pull request #80680 from samueldr/fix/71157
runInLinuxVM: Ensure tools requiring /etc/passwd work
2020-02-23 16:05:21 -05:00
Samuel Dionne-Riel 687e2195d8 runInLinuxVM: Ensure tools requiring /etc/passwd work
This includes, but is not limited to:

 * whoami
 * nix >= 2.3.1

See

 * https://github.com/NixOS/nixpkgs/issues/71157
 * https://github.com/NixOS/nixops/issues/1216
 * https://github.com/nix-community/nixops-libvirtd/issues/5
2020-02-20 18:51:17 -05:00
Jan Tojnar 477ea066ca
Merge pull request #80091 from NixOS/npregit-dumpall
nix-prefetch-git: report deepClone & leaveDotGit
2020-02-19 07:37:11 +01:00
Profpatsch 493c357720 pkgs/build-support/writers: use runCommandLocal for scripts
If we just want to write a non-compiled script (e.g. writeDash), it’s
usually a lot faster just doing it locally. That’s what
`runCommandLocal` was introduced for, so let’s use it in `writers`.
2020-02-18 19:53:18 +01:00
(cdep)illabout 55e1529335 symlinkJoin: add more documentation 2020-02-18 16:35:28 +01:00
(cdep)illabout ae607de23d linkFarmFromDrvs: Add a trivial builder to create a linkFarm from a list of derivations 2020-02-18 16:35:28 +01:00
Andreas Rammhold 453589696b
Merge pull request #79816 from andir/buildRustCrate-no-override-dep
buildRustCrate: remove superfluous dependency overrides
2020-02-18 15:18:44 +01:00
Bas van Dijk 4e0a6bc4be symlinkJoin: fix "bash: Argument list too long" error
When the `paths` argument is too big `symlinkJoin` will fail with:

```
while setting up the build environment: executing '/nix/store/rm1hz1lybxangc8sdl7xvzs5dcvigvf7-bash-4.4-p23/bin/bash': Argument list too long
```

This is fixed by passing `paths` as a file instead of as an
environment variable.
2020-02-17 11:45:44 +01:00
Benjamin Hipple 6d881472ef rust: Fix for legacy fetch cargo
See inline comment and #79975 for details.
2020-02-15 17:41:35 -08:00
Jan Tojnar 28faf5bc86
nix-prefetch-git: Fix output mangling
`git repack` and `git gc` sometimes print “Nothing new to pack.”
to stdout, which breaks redirecting output to JSON file.

Let’s move the stdout of all git calls where it is not used to stderr
so that we still receive the info but it does not pollute our output.
2020-02-14 22:15:36 +01:00
Bjørn Forsman d394c82cde
build-fhs-userenv: runCommand -> runCommandLocal (#77253)
It's a trivial derivation, no need to build remotely (it's slower that
way).
2020-02-14 20:16:49 +02:00
lewo d49804cf45
Merge pull request #80102 from nlewo/fix-78744
dockerTools.buildLayeredImage: store all paths passed in final layer (fix 78744)
2020-02-14 09:48:52 +01:00
Mario Rodas 7696369bec
docker-slim: move expression to outside of build-support (#80078) 2020-02-14 09:42:18 +01:00
Antoine Eiche baa78de594 nixosTests.docker-tools: add bulk-layer test
A regression test for https://github.com/NixOS/nixpkgs/issues/78744.
2020-02-14 09:26:26 +01:00
Richard Wallace 3b65b3f6d6 dockerTools.buildLayeredImage: store all paths passed in final layer
Fixes #78744

My previous change broke when there are more packages than the maximum
number of layers. I had assumed that the `store-path-to-layer.sh` was
only ever passed a single store path, but that is not the case if
there are multiple packages going into the final layer. To fix this, we
loop through the paths going into the final layer, appending them to the
tar file and making sure they end up at the right path.
2020-02-14 09:26:26 +01:00
Benjamin Hipple eb11feaa0b treewide: change fetchCargoTarball default to opt-out
Changes the default fetcher in the Rust Platform to be the newer
`fetchCargoTarball`, and changes every application using the current default to
instead opt out.

This commit does not change any hashes or cause any rebuilds. Once integrated,
we will start deleting the opt-outs and recomputing hashes.

See #79975 for details.
2020-02-13 22:41:37 -08:00
Jan Tojnar f19a101a73
nix-prefetch-git: report deepClone & leaveDotGit
Just like in be30ba8e0e
we want `fetchgit (builtins.fromJSON (builtins.readFile ./src.json))` to work.
2020-02-14 06:05:43 +01:00
Benjamin Hipple c1542fc4cd rust: Replace migration README with GitHub issue
The readme was nice to discuss in the implementation PR, but now that this is
merged it's better to have an issue that can be linked against in PRs and
doesn't require further merges to update status.

Ported with a status update in #79975
2020-02-12 22:28:53 -05:00
Frederik Rietdijk 424697d512 Merge master into staging-next 2020-02-12 09:55:31 +01:00
Andreas Rammhold be5597fc9d
buildRustCrate: remove superfluous dependency overrides
By overriding each dependency on every level of the dependency tree we
are creating a lot of unnecessary instances of the same derivation

Looking at the output size of `nix-instantiate --trace-function-calls
-vvvv …` and the execution time I got about a 10x improvement after
applying this change.

It was probably good intentions that lead to these overrides but in
practice no tooling (that I know of) really needs this. `carnix` and
`crate2nix` are fine without those overrides. Furthermore I believe that
it is the job of the tooling around `buildRustCrate` to provide a
coherent set of overrides. By not enforcing all of the overrides, debug
flags, verbosity, … to be the same throughout the closure we also allow
consumers to override specific aspects of the crates. Some (older?)
crates might need different `crateOverrides` then newer crates with the
same name. Currently such situations can not (easily) be implemented
with the override in-place.
2020-02-11 11:48:45 +01:00
Benjamin Hipple 2115a2037c fetchcargo: use flat tar.gz file for vendored src instead of recursive hash dir
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.
2020-02-10 10:17:29 -05:00
Andreas Rammhold 56e11bc8df
buildRustCrate: remap the current build dir to / for (more) reproducible builds 2020-02-06 01:18:59 +01:00
Uri Baghin 10d1a04b02 bazel-deps: remove 2020-02-04 12:35:47 +11:00
Uri Baghin 3dc9c1b885 bazel-remote: 2019-01-12 -> 2020-01-29 2020-02-04 02:26:15 +01:00
worldofpeace 8ecbde9def
Merge pull request #78945 from jbedo/singularity-img
singularity-tools: fix buildImage
2020-02-03 15:19:47 -05:00
Justin Bedo 56c22350c0
singularity-tools: fix buildImage 2020-02-02 09:18:39 +11:00
Silvan Mosberger 0a351c3f65
dockerTools.*: Assertion against building for Darwin (#77952)
dockerTools.*: Assertion against building for Darwin
2020-01-31 21:17:40 +01:00
lewo 86f8732194
Merge pull request #78834 from nlewo/fix-two-layers-image
Fix dockerTools.buildLayerImage with 2 layers
2020-01-30 21:10:52 +01:00
Antoine Eiche 01a68479cc dockerTools.buildLayeredImage: assert maxLayers > 1
Since a layer is reserved for "customization", the image can not
contains less than 2 layers.

The user gets the following message at evaluation:

    nix-instantiate nixos/tests/docker-tools.nix
    trace: the maxLayers argument of dockerTools.buildLayeredImage function must be greather than 1 (current value: 1)
2020-01-30 21:09:52 +01:00
Antoine Eiche 283bcc1003 dockerTools.buildLayeredImage: fix image with only 2 layers
A test is also added to ensure an image with 2 layers can be built.
2020-01-30 21:09:44 +01:00
Andreas Rammhold a57d0fe0bb
buildRustCrate: fix #78412
`build.rs` files might create files. Those files are supposed to go into
`OUT_DIR` (envirionment variable) and not be overlayed onto the source
tree.
2020-01-28 14:07:58 +01:00
Andreas Rammhold 19698d15ce
buildRustCrateTests: add regression test for #74071 2020-01-28 14:07:58 +01:00
Frederik Rietdijk 7562c2112d
Merge pull request #77739 from NixOS/staging-next
Staging next
2020-01-28 14:03:29 +01:00
Andreas Rammhold cd27940df0
Merge pull request #78188 from andir/buildRustCrate-support-rlib
buildRustCrate: support `rlib` dependency type
2020-01-28 13:47:57 +01:00
Frederik Rietdijk dce0ca29d9 Merge master into staging-next 2020-01-28 10:46:13 +01:00
Mario Rodas c3ccac15eb
libredirect: fix build on darwin
Fix build failure on darwin due to absence of `O_TMPFILE`.
2020-01-23 20:20:20 -05:00
Andreas Rammhold 78faab1be0 buildRustCrateTests: add test case for rlib linking 2020-01-21 17:46:32 +01:00
Andreas Rammhold 406e0c9d51
buildRustCrateTests: fix some formatting issues 2020-01-21 17:32:48 +01:00
Andreas Rammhold d6a8b55fb0
buildRustCrate: treat rlib crates just like lib crates
Both version provide `rlib` files to link against. Previously we would
try to find a matching shared library in the `lib` output.
2020-01-21 17:22:59 +01:00
Andreas Rammhold 60fea726de
Merge pull request #77822 from andir/buildRustCrate-remove-custom-lib-path
buildRustCrate: remove custom lib path heuristic
2020-01-21 17:21:55 +01:00
Silvan Mosberger b6a9211bf4
dockerTools.*: Assertion against building for darwin
Building a docker image with darwin binaries just yields a confusing
error when ran:

  standard_init_linux.go:211: exec user process caused "exec format error"

This change prevents people from building such images in the first place
2020-01-20 19:14:12 +01:00
Jonas Schürmann 61870727de mirrors.nix: Use HTTPS in maven mirrors
Since 2020-01-15, the maven central repository does no longer support HTTP.
See https://blog.sonatype.com/central-repository-moving-to-https for details.
2020-01-20 09:54:14 +01:00
Mario Rodas 21d0b2067f
Merge pull request #76831 from mbrgm/docker-slim
docker-slim: init at 1.26.1
2020-01-18 07:51:03 -05:00
Andreas Rammhold 69c96adc53
buildRustCrateTests: use releaseTools.aggregate
Previously I did use `runCommand` to do the same. Using
releaseTools.aggregate seems a lot saner and we might get nicer hydra
output of the tests that are failing.
2020-01-16 13:24:15 +01:00
Andreas Rammhold 29a8575e3d
buildRustCrate: remove one of the odd library filename cases
It used to be the case (ref missing) that cargo did treat
`src/$libName.rs` as an alternative to `src/lib.rs` when the latter
wasn't present. Recently I failed to reproduce that with vanilla cargo
and it started to cause pain with some crates of the form:

some_crate/
 `- src
   `- main.rs
   `- some_crate.rs

We would build `src/some_crate.rs` and thing it is a library while that
might not be the actual case. This crate is a valid `bin` crate not a
`lib` crate as far as I can tell from the samples I took.

I removed support for the previously required heuristic and commented
out the test cases in case we will need them again. We could crawl in
the Git history but chances are that the next person looking into this
doesn't know about the history.
2020-01-16 13:24:13 +01:00
Anders Kaseorg 3cd8ce3bce treewide: Fix unsafe concatenation of $LD_LIBRARY_PATH
Naive concatenation of $LD_LIBRARY_PATH can result in an empty
colon-delimited segment; this tells glibc to load libraries from the
current directory, which is definitely wrong, and may be a security
vulnerability if the current directory is untrusted.  (See #67234, for
example.)  Fix this throughout the tree.

Signed-off-by: Anders Kaseorg <andersk@mit.edu>
2020-01-15 09:47:03 +01:00