Commit Graph

47 Commits

Author SHA1 Message Date
John Ericson
f0b634c7e8 Merge pull request #24610 from Ericson2314/platform-normalization
Platform normalization
2017-04-17 17:28:01 -04:00
John Ericson
8c99aab3ea lib: Fix system parsing, and use for doubles lists
The old hard-coded lists are now used to test system parsing.

In the process, make an `assertTrue` in release lib for eval tests; also
use it in release-cross
2017-04-17 17:13:01 -04:00
John Ericson
b477851f34 top-level: Less indirection for lib in release*.nix 2017-04-17 17:13:01 -04:00
Bjørn Forsman
d45df036c3 release(-lib).nix: add nixpkgsArgs parameter
This allows customizing the nixpkgs arguments by the caller. My use case
is creating a personal nixpkgs channel containing some unfree packages.

The default is still to not build unfree packages, so for nixpkgs this
is no functional change.
2017-04-11 19:43:34 +02:00
Franz Pletz
e3b84d71a6
nixpkgs: add aarch64-linux to release-lib
cc #23638
2017-03-08 17:13:34 +01:00
John Ericson
c869fe022e top-level: no more need to expose splicedPackages
This was just done temporarily on the last cross-overhauling PR for
testing purposes.
2017-01-25 09:24:55 -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
John Ericson
916012121c release-cross: Factor out system filtering 2016-12-27 15:41:07 -08:00
John Ericson
db0dbd0b81 release-cross: Use the same genAttrs logic for testOnCross as testOn
Eventually we'll want to test cross-compiling *from* various platforms. For
now, its good to be consistent.
2016-12-27 15:41:07 -08:00
Domen Kožar
5c750e4940 Build all derivations at least for 64bit Linux 2016-11-28 15:19:47 +01:00
Tuomas Tynkkynen
628c86c51a release-lib: Remove unused allowTexliveBuilds option
This is dead code since the old monolithic TeXLive was dropped in
da421bc75f.
2016-09-11 01:13:49 +03:00
John Ericson
f073df60d6 Replace ./../* with ../* in Nix expressions (#16414) 2016-06-22 10:39:50 +02:00
Nicolas B. Pierron
6313a5698a Replace references to all-packages.nix, by references to the top-level of nixpkgs repository. 2016-03-13 18:25:52 +00:00
Eelco Dolstra
7cd3d502bb copy-tarballs.pl: Revive extracting all tarballs from release.nix 2015-12-16 15:38:59 +01:00
Peter Simons
a5de21c230 release-lib.nix: set inHydra anyways, it might be useful some day 2015-07-17 15:33:49 +02:00
Peter Simons
32a81524fe release-lib: rename config option "inHydra" to "allowTexliveBuilds"
Also, take the value of that attribute as an argument to the module so
that Hydra maintainers who don't mind building TexLive have a chance to
do so.
2015-07-17 14:13:40 +02:00
Eelco Dolstra
7f54f99656 Don't build texLive in Hydra
It's way too big (texlive-core-2014 alone is > 1.5 GB).
2015-06-25 17:34:08 +02:00
Florian Friesdorf
88e58a4fa2 cygwin: stdenv 2015-05-28 10:53:34 +02:00
Eelco Dolstra
0461f35894 Rename scrubDrv -> hydraJob and make it more effective
It now strictly evaluates all remaining attributes, preventing
unevaluated thunks that cannot be garbage-collected. It's also applied
to all jobs in Nixpkgs' release.nix.

This reduces hydra-eval-jobs' memory consumption on the 14.12
release-combined jobset from 5.1 GB to 2.0 GB.
2015-03-20 19:46:18 +01:00
Eelco Dolstra
738cf42639 release-lib: Simplify 2015-03-20 19:37:55 +01:00
Eelco Dolstra
b981eb3f7b release-lib.nix: Style cleanup 2015-03-20 19:33:08 +01:00
Eelco Dolstra
fddc347689 Tweak 2015-01-06 11:55:47 +01:00
Eelco Dolstra
ab10131663 Remove scheduling priorities
This was only used for stdenv and is pretty obsolete now.
2015-01-06 11:55:47 +01:00
Nikolay Amiantov
aa14fb6389 release-lib: add option to use another package set 2014-11-14 00:34:51 +03:00
Luca Bruno
46b78bdef4 release: do not process broken packages, we induce they don't have platforms at all 2014-09-03 13:02:45 +02:00
Eelco Dolstra
a2cbc77e4f Only show/build a package on the platforms listed in meta.platforms
The function ‘mkDerivation’ now checks whether the current platform
type is included in a package's meta.platform field.  If not, it
throws an exception:

  $ nix-build -A linux --argstr system x86_64-darwin
  error: user-thrown exception: the package ‘linux-3.10.15’ is not supported on ‘x86_64-darwin’

These packages also no longer show up in ‘nix-env -qa’ output.  This
means, for instance, that the number of packages shown on
x86_64-freebsd has dropped from 9268 to 4764.

Since meta.platforms was also used to prevent Hydra from building some
packages, there now is a new attribute meta.hydraPlatforms listing the
platforms on which Hydra should build the package (which defaults to
meta.platforms).
2013-11-05 00:06:10 +01:00
Eelco Dolstra
e3c641b39a mercurial: Update to 2.6.1
Also, set a default for web.cacerts so that the system certificates on
NixOS are used.
2013-05-15 13:15:53 +02:00
Eelco Dolstra
c5f45036e6 release-lib.nix: Make the set of supported platforms an argument 2013-03-26 13:12:25 +01:00
Eelco Dolstra
bcc14c4d3c Add an "unstable" aggregate to replace the "unstable" view
Views are obsolete, aggregates are the declarative replacement.
2013-03-26 11:59:00 +01:00
Eelco Dolstra
78de3c912d Only return Hydra jobs for supported platforms 2013-03-26 11:59:00 +01:00
Eelco Dolstra
cffc126e14 Make Nixpkgs jobs unique
That is, there are now distinct jobs like ‘coreutils.x86_64-linux’ and
‘coreutils.x86_64-darwin’, rather than a single job ‘coreutils’ with
multiple builds.  This means that testing a job is simpler:

  $ nix-build pkgs/top-level/release.nix -A coreutils.x86_64-linux

See https://github.com/NixOS/hydra/issues/60 for the motivation.
2013-03-26 11:58:59 +01:00
Rickard Nilsson
1886d1db6a Merge remote-tracking branch 'upstream/master' into stdenv-updates 2013-01-20 16:53:21 +01:00
Eelco Dolstra
d8e722985b Add config option ‘allowUnfree’
If set to false, mkDerivation will throw an exception if a package has
an unfree license.  ‘release-lib.nix’ uses this to enforce that we
don't build unfree packages as part of the Nixpkgs channel.  Since
this is set through Nixpkgs' ‘config’ argument, it's more finegrained
than $HYDRA_DISALLOW_UNFREE.
2013-01-17 23:41:37 +01:00
Eelco Dolstra
5be0a9acd7 Rename hostDrv -> crossDrv, buildDrv -> nativeDrv
This is for consistency with terminology in stdenv (and the terms
"hostDrv" and "buildDrv" are not very intuitive, even if they're
consistent with GNU terminology).
2012-12-28 19:08:19 +01:00
Eelco Dolstra
72684fb42f Remove support for the obsolete powerpc-darwin and i686-darwin platforms 2012-11-29 14:10:49 +01:00
Eelco Dolstra
7cd2bdaa58 * Add x86_64-freebsd to release-lib.nix
svn path=/nixpkgs/trunk/; revision=32886
2012-03-08 12:52:00 +00:00
Eelco Dolstra
8778f0b368 * Change the priority of trunk builds back.
svn path=/nixpkgs/trunk/; revision=32823
2012-03-06 12:21:33 +00:00
Lluís Batlle i Rossell
46082f25f9 Updating from trunk
svn path=/nixpkgs/branches/stdenv-updates/; revision=23027
2010-08-07 20:14:04 +00:00
Yury G. Kudryashov
3c4d5457a8 Add recurseForRelease flag
I don't want kde-4.5 to be visible for nix-env, but I want hydra to build it

svn path=/nixpkgs/trunk/; revision=22930
2010-08-03 14:02:42 +00:00
Lluís Batlle i Rossell
fa1a414aee Merging from trunk - I had to resolve conflicts on coreutils; I hope I resolved that
well


svn path=/nixpkgs/branches/stdenv-updates/; revision=21975
2010-05-25 20:38:18 +00:00
Eelco Dolstra
719f023f33 * A quick hack to make release.nix evaluate in a reasonable amount of
time and space on the new (non-ATerm) Nix expression evaluator.  It
  turns out that release.nix relied rather heavily on maximal laziness
  for efficiency: every job calls `allPackages { inherit system; }'
  for each platform.  This causes the dependencies of the job to be
  reevaluated for every job/platform combination.  This is very slow
  and (because the evaluator doesn't have a garbage collector yet)
  eventually causes the evaluator to run out of memory and be killed.

  As a workaround, I've replaced the calls to `allPackages' with a
  quasi-memoised `pkgsFor' function.  It "caches" the result by going
  through a variable such as `pkgs_x86_64_linux', which is evaluated
  only once.  Evaluation now only takes 4.4s and 545 MiB on my
  machine.

  A cleaner solution may be to move the `system' argument outwards so
  that entire set of jobs is called only once for each value of
  `system'.

svn path=/nixpkgs/trunk/; revision=21966
2010-05-25 10:35:14 +00:00
Lluís Batlle i Rossell
70b2dd393c Updating from trunk (to get uclibc changes...)
I want to cross-build utillinux and coreutils, and that play will have to go in stdenv-updates I think.


svn path=/nixpkgs/branches/stdenv-updates/; revision=20506
2010-03-09 16:31:36 +00:00
Eelco Dolstra
d9b48d35fb svn path=/nixpkgs/branches/stdenv-updates/; revision=20504 2010-03-09 16:07:00 +00:00
Lluís Batlle i Rossell
1522caa556 Enabling cross-builds with uclibc again (I had that too much abandoned).
Hydra now should even test it.


svn path=/nixpkgs/trunk/; revision=20500
2010-03-09 15:48:25 +00:00
Lluís Batlle i Rossell
570759c288 Getting the hydra cross-trunk builds have proper low priority (that of toJob)
svn path=/nixpkgs/trunk/; revision=20495
2010-03-09 14:49:30 +00:00
Lluís Batlle i Rossell
da8c9de5a8 Updating the hydra builds for release-cross.nix
Now the attributes listed as Jobs will look even uglier - very long names, too much repeating.


svn path=/nixpkgs/trunk/; revision=20492
2010-03-09 14:14:30 +00:00
Rob Vermaas
b44af59afb separate cross builds in separate release expression, will add extra jobset
svn path=/nixpkgs/trunk/; revision=20484
2010-03-09 10:33:31 +00:00