Commit graph

1148 commits

Author SHA1 Message Date
Tuomas Tynkkynen c909f1b18e stdenv: Add updateAutoconfGnuConfigScriptsHook for aarch64
This is required for Aarch64 since a lot of source tarballs ship with
outdated configure scripts that don't recognize aarch64. Simply
replacing the config.guess and config.sub with new versions from
upstream makes them build again.

This same approach is used by at least Buildroot and Fedora. In
principle this could be enabled for all architectures but
conditionalizing this on aarch64 avoids a mass rebuild on x86.
2017-01-25 00:01:52 +02:00
Tuomas Tynkkynen 7c8a060c09 stdenv: Bringup aarch64 architecture support 2017-01-25 00:01:51 +02:00
John Ericson bfb147b6a8 top-level: Only splice as needed for performance 2017-01-24 11:37:56 -05:00
John Ericson a1a798f017 top-level: crossSystem is no longer exposed to packages. Use *Platform. 2017-01-24 11:37:56 -05:00
John Ericson 1c0365bd88 cross-stdenv: Inline useless bindings and reindent
Semantics should be unchanged
2017-01-24 11:37:56 -05:00
John Ericson 92edcb7ebb top-level: Lay the groundwork for {build,host,target}Platform
The long term goal is a big replace:
  { inherit system platform; } => buildPlatform
  crossSystem => hostPlatform
  stdenv.cross => targetPlatform
And additionally making sure each is defined even when not cross compiling.

This commit refactors the bootstrapping code along that vision, but leaves
the old identifiers with their null semantics in place so packages can be
modernized incrementally.
2017-01-24 11:37:56 -05:00
John Ericson bf17d6dacf top-level: Introduce buildPackages for resolving build-time deps
[N.B., this package also applies to the commits that follow it in the same
PR.]

In most cases, buildPackages = pkgs so things work just as before. For
cross compiling, however, buildPackages is resolved as the previous
bootstrapping stage. This allows us to avoid the mkDerivation hacks cross
compiling currently uses today.

To avoid a massive refactor, callPackage will splice together both package
sets. Again to avoid churn, it uses the old `nativeDrv` vs `crossDrv` to do
so. So now, whether cross compiling or not, packages with get a `nativeDrv`
and `crossDrv`---in the non-cross-compiling case they are simply the same
derivation. This is good because it reduces the divergence between the
cross and non-cross dataflow. See `pkgs/top-level/splice.nix` for a comment
along the lines of the preceding paragraph, and the code that does this
splicing.

Also, `forceNativeDrv` is replaced with `forceNativePackages`. The latter
resolves `pkgs` unless the host platform is different from the build
platform, in which case it resolves to `buildPackages`. Note that the
target platform is not important here---it will not prevent
`forcedNativePackages` from resolving to `pkgs`.

--------

Temporarily, we make preserve some dubious decisions in the name of preserving
hashes:

Most importantly, we don't distinguish between "host" and "target" in the
autoconf sense. This leads to the proliferation of *Cross derivations
currently used. What we ought to is resolve native deps of the cross "build
packages" (build = host != target) package set against the "vanilla
packages" (build = host = target) package set. Instead, "build packages"
uses itself, with (informally) target != build in all cases.

This is wrong because it violates the "sliding window" principle of
bootstrapping stages that shifting the platform triple of one stage to the
left coincides with the next stage's platform triple. Only because we don't
explicitly distinguish between "host" and "target" does it appear that the
"sliding window" principle is preserved--indeed it is over the reductionary
"platform double" of just "build" and "host/target".

Additionally, we build libc, libgcc, etc in the same stage as the compilers
themselves, which is wrong because they are used at runtime, not build
time. Fixing this is somewhat subtle, and the solution and problem will be
better explained in the commit that does fix it.

Commits after this will solve both these issues, at the expense of breaking
cross hashes. Native hashes won't be broken, thankfully.

--------

Did the temporary ugliness pan out? Of the packages that currently build in
`release-cross.nix`, the only ones that have their hash changed are
`*.gcc.crossDrv` and `bootstrapTools.*.coreutilsMinimal`. In both cases I
think it doesn't matter.

 1. GCC when doing a `build = host = target = foreign` build (maximally
    cross), still defines environment variables like `CPATH`[1] with
    packages.  This seems assuredly wrong because whether gcc dynamically
    links those, or the programs built by gcc dynamically link those---I
    have no idea which case is reality---they should be foreign. Therefore,
    in all likelihood, I just made the gcc less broken.

 2. Coreutils (ab)used the old cross-compiling infrastructure to depend on
    a native version of itself. When coreutils was overwritten to be built
    with fewer features, the native version it used would also be
    overwritten because the binding was tight. Now it uses the much looser
    `BuildPackages.coreutils` which is just fine as a richer build dep
    doesn't cause any problems and avoids a rebuild.

So, in conclusion I'd say the conservatism payed off. Onward to actually
raking the muck in the next PR!

[1]: https://gcc.gnu.org/onlinedocs/gcc/Environment-Variables.html
2017-01-24 11:37:56 -05:00
Nicolas B. Pierron da8cf2662a Fix missing overlays argument in stdenv/linux/default.nix 2017-01-16 01:17:33 +01:00
Nicolas B. Pierron f5dfe78a1e Add overlays mechanism to Nixpkgs.
This patch add a new argument to Nixpkgs default expression named "overlays".

By default, the value of the argument is either taken from the environment variable `NIXPKGS_OVERLAYS`,
or from the directory `~/.nixpkgs/overlays/`.  If the environment variable does not name a valid directory
then this mechanism would fallback on the home directory.  If the home directory does not exists it will
fallback on an empty list of overlays.

The overlays directory should contain the list of extra Nixpkgs stages which would be used to extend the
content of Nixpkgs, with additional set of packages.  The overlays, i-e directory, files, symbolic links
are used in alphabetical order.

The simplest overlay which extends Nixpkgs with nothing looks like:

```nix
self: super: {
}
```

More refined overlays can use `super` as the basis for building new packages, and `self` as a way to query
the final result of the fix-point.

An example of overlay which extends Nixpkgs with a small set of packages can be found at:
  https://github.com/nbp/nixpkgs-mozilla/blob/nixpkgs-overlay/moz-overlay.nix

To use this file, checkout the repository and add a symbolic link to
the `moz-overlay.nix` file in `~/.nixpkgs/overlays` directory.
2017-01-16 01:17:33 +01:00
John Ericson abaf790ea9 stdenv/booter.nix: Add longer note explaining indexing 2017-01-13 13:47:17 -05:00
John Ericson ff35560460 linux stdenv: Inline stage funs to conform to new convention
Code is just moved around
2017-01-13 13:23:26 -05:00
John Ericson b10f415c21 linux stdenv: Remove stray use of stage0 to bootstrap more elegantly 2017-01-13 13:23:26 -05:00
John Ericson 6a45e911c4 linux stdenv: Utilize overrides and prevStage better
`gcc-unwrapped` basically replaces `gccPlain`. It may seem like an ugly
polution to stick it in all-packages, but a future PR will enshrine this
`*-unwrapped` pattern. In any event, the long term goal is stdenvs might
need to tweak how compilers are booted and wrapped, but the code to build
the unwrapped compilers themselves should be generic.
2017-01-13 13:23:25 -05:00
David Grayson 0f33b9f7f1 top-level: Do stdenvOverrides in stage.nix even if crossSystem exists.
Instead, the cross stdenv will patch up the override field -- the complexity
is now confined to the one place it matters.
2017-01-13 13:23:25 -05:00
John Ericson 3e197f7d81 top-level: Normalize stdenv booting
Introduce new abstraction, `stdenv/booter.nix` for composing bootstraping
stages, and use it everywhere for consistency. See that file for more doc.

Stdenvs besides Linux and Darwin are completely refactored to utilize this.
Those two, due to their size and complexity, are minimally edited for
easier reviewing.

No hashes should be changed.
2017-01-13 13:23:23 -05:00
John Ericson 0ef8b69d12 top-level: Modernize stdenv.overrides giving it self and super
Document breaking change in 17.03 release notes
2017-01-13 10:36:11 -05:00
Tuomas Tynkkynen bde8632106 coreutils: Build with libattr to support xattrs
Fixes #21649
2017-01-05 00:55:35 +02:00
John Ericson 19fbe80c3d top-level: avoid another assert false while we're at it 2016-12-15 17:09:46 -05:00
John Ericson 7960a1b1b8 linux stdenv: Avoid assert false
On one hand, don't want to pass garbage that affects hash, on the other
hand footguns are bad.

Now, factored out the derivation so only need to pass in what is used.
2016-12-15 17:09:29 -05:00
John Ericson 670256dc39 linux stdenv: Rename the bootstrap directory to bootstrap-files 2016-12-15 17:05:39 -05:00
John Ericson a94af71da7 linux stdenv: Fix assert that broke tests...and hydra 2016-12-15 16:45:08 -05:00
John Ericson 5c6234a7d3 top-level: Allow manually specifying a stdenv, and fix stdenv tests
- The darwin test can now force the use of the freshly-booted darwin stdenv
 - The linux test now passes enough dummy arguments

This may make debugging harder, if so, check out #20889
2016-12-03 17:21:07 -08:00
John Ericson 4751d9e5ad top-level: turn the screw
- Non-cross stdenvs are honest and assert that `crossSystem` is null

 - `crossSystem` is a mandatory argument to top-level/stage.nix, just like
   `system` and `platform`

 - Broken default arguments on stdenvs for testing are gone.

 - All stdenvs (but little-used stdenvNix) take the same arguments for easy
   testing.
2016-12-01 11:24:33 -05:00
John Ericson 39753f5360 top-level: Close over fewer arguments for stdenv stages
This makes the flow of data easier to understand. There's little downside
because the args in question are already inspected by the stdenvs.

cross-compiling in particular is simpler because we don't need to worry
about overriding the config closed over by `allPackages`.
2016-11-30 19:11:03 -05:00
John Ericson d240a0da1a top-level: Remove cycles: stdenv calls in top-level but not vice versa
This commit changes the dependencies of stdenv, and clean-up the stdenv
story by removing the `defaultStdenv` attribute as well as the `bootStdenv`
parameter.

Before, the final bootstrapping stage's stdenv was provided by
all-packages, which was iterating multiple times over the
top-level/default.nix expression, and non-final bootstrapping stages'
stdenvs were explicitly specified with the `bootStdenv` parameter.

Now, all stages' stdenvs are specified with the `stdenv` parameter.
For non-final bootstrapping stages, this is a small change---basically just
rename the parameter.
For the final stage, top-level/default.nix takes the chosen stdenv and
makes the final stage with it.

`allPackages` is used to make all bootstrapping stages, final and
non-final alike. It's basically the expression of `stage.nix` (along with a
few partially-applied default arguments)

Note, the make-bootstrap-tools scripts are temporarily broken
2016-11-30 19:10:59 -05:00
John Ericson 2df990967b Move up inherit binding for consistency 2016-11-30 19:03:22 -05:00
John Ericson ddeb0d2d6b top-level: Stop exposing all stdenvs 2016-11-30 19:03:01 -05:00
Domen Kožar 45f579b9e7 allowUnfree: mention the solution that works for nix-shell as well 2016-11-16 15:14:19 +01:00
Shea Levy 7df3d7446f Add initial basic support for cross-compiling to iOS 2016-11-15 16:31:55 -05:00
John Ericson f68e16f023 top-level: Make cross compiling slightly saner
Removes the weird stdenv cycle used to match the old infrastructure.
It turns out that matching it so precisely is not needed.
2016-11-12 20:20:44 +01:00
John Ericson 6bfe04277f top-level: Make config-overriden stdenv bootstrap more normally 2016-11-06 21:28:38 -08:00
John Ericson e22346c35e top-level: Make stdenvCross which appears at first glance normal...
...but actually is weird just like the original
2016-11-06 21:27:38 -08:00
Domen Kožar 62edf873aa Merge pull request #18660 from aneeshusa/add-override-attrs
mkDerivation: add overrideAttrs function
2016-10-30 11:32:15 +01:00
John Ericson d9bc6eb7f0 top-level: Make stdenv/default.nix more concise with inherit (expr) id; syntax 2016-10-27 21:47:46 -07:00
Frederik Rietdijk 7077a270bf Merge remote-tracking branch 'upstream/master' into HEAD 2016-10-26 13:06:43 +02:00
Joachim F 3d5630fac9 Merge pull request #19769 from groxxda/license
stdenv.hasLicense: ? supports nested lookup
2016-10-24 15:19:12 +02:00
Alexander Ried a0ac2ae35e stdenv: throwEvalHelp performance (#19779) 2016-10-22 20:24:56 +02:00
Frederik Rietdijk e56832d730 Merge remote-tracking branch 'upstream/master' into HEAD 2016-10-22 17:23:24 +02:00
Alexander Ried 43ce115ca9 stdenv.hasLicense: ? supports nested lookup
this avoids one copy of the attrset
2016-10-22 02:43:13 +02:00
Dan Peebles e4dba74e8a darwin.stdenv: update MACOSX_DEPLOYMENT_TARGET
No point in claiming we're compatible with a version we don't try to
support, and this will probably help with a few other things elsewhere.
2016-10-16 20:47:43 -04:00
Dan Peebles 6a380c20e0 darwin.make-bootstrap-tools: Fix test target
I'll probably get rid of this eventually since the test-pkgs stuff is
more realistic, but this gives quick feedback for now.
2016-10-15 21:36:51 -04:00
Vladimír Čunát 027efec879 Merge staging without python splitting for now
The split needs more time to finish rebuilding,
but the rest seems OK and there are security fixes.
2016-10-14 09:24:21 +02:00
Vladimír Čunát e0210d7cbf tested job: fixup evaluation after 5dadc3a
I believe hardcoding system to builtins.currentSystem isn't very good,
as e.g. Hydra surely evaluates on Linux only...
2016-10-14 09:21:38 +02:00
Dan Peebles 5dadc3a30c darwin.make-bootstrap-tools: fix for new stdenv setup 2016-10-13 22:32:19 -04:00
Vladimír Čunát 727fc259d6 Merge branch 'master' into staging 2016-10-13 09:53:16 +02:00
Vladimír Čunát af38c05587 stdenv stripHash(): fixup after #19324 2016-10-12 23:45:30 +02:00
Dan Peebles d8a0307a5d [darwin.stdenv] Fix to work on Sierra
This reinstates the libSystem selective symbol export machinery we used
to have, but locks it to the symbols that were present in 10.11 and skips
the actual compiled code we put into that library in favor of the system
initialization code. That should make it more stable and less likely to
do weird stuff than the last time we did this.
2016-10-12 00:08:13 -04:00
Profpatsch bef6bef0d2
stdenv/stripHash: print to stdout, not to variable
`stripHash` documentation states that it prints out the stripped name to
the stdout, but the function stored the value in `strippedName`
instead.

Basically all usages did something like
`$(stripHash $foo | echo $strippedName)` which is just braindamaged.
Fixed the implementation and all invocations.
2016-10-11 18:34:36 +02:00
Aneesh Agrawal 39b64b52ed mkDerivation: add overrideAttrs function
This is similar to `overrideDerivation`, but overrides the arguments to
`mkDerivation` instead of the underlying `derivation` call.

Also update `makeOverridable` so that uses of `overrideAttrs` can be
followed by `override` and `overrideDerivation`, i.e. they can be
mix-and-matched.
2016-10-02 11:08:34 -04:00
Eelco Dolstra 7a4209c356 Merge remote-tracking branch 'origin/master' into staging 2016-09-20 17:46:09 +02:00