Commit Graph

91 Commits

Author SHA1 Message Date
John Ericson
8f54d931cd
Merge pull request from obsidiansystems/no-stdenv-cross-adapter
stdenv cross adapter: Get rid if `selfNativeBuildInput`
2018-03-06 20:17:13 -05:00
John Ericson
8d58008039 stdenv cross adapter: Get rid if selfNativeBuildInput
It is finally no longer used anywhere

Fixes 
2018-03-06 20:05:13 -05:00
Daniel Barlow
9c50ae6898 lib, treewide: Add missing MIPS arches, and fix existing usage
Existing "mips64el" should be "mipsel".

This is just the barest minimum so that nixpkgs can recognize them as
systems - although required for building individual derivations onto
MIPS boards, it is not sufficient if you want to actually build nixos on
those targets
2018-02-23 20:43:42 -05:00
Will Dietz
5dcf3cdf37 stdenv: automatically update config.sub for musl 2018-02-13 09:44:36 -06:00
John Ericson
1cde06acf6 stdenv cross adapter: Get rid of extra propagatation of buildInputs
Binutils is patched so we don't rely on `--rpath-link`.
2017-12-29 17:32:28 -05:00
John Ericson
7a733ceda4 cross stdenv: No more native-fishing hack.
The comment says it was put there for pkgconfig, but now pkgconfig has
been dealt with.
2017-09-21 15:49:54 -04:00
John Ericson
3af3d6efc2 stdenv cross adapater: Remove old extraBuildInputs
They, unlike their native counterparts, run on the wrong platform and
are therefore invalid.
2017-09-21 14:52:42 -04:00
John Ericson
05ef1034c7 stdenv cross adaptor: Add optional overrides parameter
By default, all previous overrides are discarded as before, as they
would only apply to the old host platform. But sometimes it is useful to
add some new ones, and this optional parameter allows that.
2017-09-21 14:52:07 -04:00
John Ericson
738bb4777c stdenv: Update autotools scripts on all Aarch64, not just Linux
This is needed when cross-compiling for iOS (Aarch64 + Darwin). I also
changed the syntax of the Linux stdenv for visual consistency, though
that has no effect on semantics as the os is already guaranteed to be
Linux.
2017-09-21 14:40:40 -04:00
John Ericson
e826a6a247 stdenv: Move some logic from cross adapter to stdenv proper
Eventually the adapter will be removed.  Moved is

 - Name suffix from hostPlatform

 - configurePlatforms

   To not cause more breakage, the default is currently [], but
   eventually it will be [ "build" "host" ], as the cross adapter makes
   it today.
2017-07-13 19:05:36 -04:00
John Ericson
a302d7360f top-level: {build,host,target}Platform are defined in the stdenv instead
See  for a discussion of this
2017-07-07 12:55:02 -04:00
John Ericson
e0492d58a6 cross stdenv adaptor: Disable checkPhase by default
Before gcc-cross-wrapper did this unconditionally
2017-06-22 17:52:28 -04:00
John Ericson
594d264205 cross stdenv adaptor: Support --host --build --target across the board
Packages get --host and --target by default, but can explicitly request
any subset to be passed as needed. See docs for more info.

rustc: Avoid hash breakage by using the old (ignored)
dontSetConfigureCross when not cross building
2017-06-22 17:52:28 -04:00
John Ericson
9bfd03eff7 cross stdenv adaptor: Remove ccCross binutils attrs and binutils extra buildDepends
It now has the correct wrapped tools and nothing else is needed.
2017-06-22 17:52:28 -04:00
John Ericson
295315cc87 binutilsCross: Remove and use binutils instead always
See previous commit for what was done to `binutils` to make this
possible.

There were some uses of `forcedNativePackages` added. The
combination of overrides with that attribute is highly spooky: it's
often important that if an overridden package comes from it, the
replaced arguments for that package come from it. Long term this
package set and all the spookiness should be gone and irrelevant:

  "Move along, nothing to see here!"

No hashes should be changed with this commit
2017-04-25 21:36:19 -04:00
John Ericson
49c99b70cf cross-stdenv: Only prune most overrides in the final stage
Before all overrides were also pruned in the previous stage, now
only gcc and binutils are, because they alone care about about the
target platform. The rest of the overrides don't, so it's better to
preserve them in order to avoid spurious rebuilds.
2017-04-24 16:31:53 -04:00
Eelco Dolstra
9d6a55aefd
~/.nixpkgs -> ~/.config/nixpkgs
The former is still respected as a fallback for config.nix for
backwards compatibility (but not for overlays because they're a new
feature).
2017-02-01 16:07:55 +01:00
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
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
John Ericson
5b7bdb2861 Remove more dead code: stdenvDarwinNaked no longer exists 2016-06-20 23:53:09 -07:00
John Ericson
47d95a7571 Remove dead code in stdenvCross adaptor 2016-06-20 23:53:09 -07:00
Vladimír Čunát
8b292a1b35 stdenv on mingw: fix 64-bin DLL detection in some cases
In particular, this makes 64-bit libpng create DLL now,
and thus depending packages won't fail anymore (e.g. freetype).
2016-04-23 10:52:01 +02:00
Eelco Dolstra
13a1c7b8c1 useOldCXXAbi: Change into a setup hook
Stdenv adapters considered weird.
2016-04-01 13:36:59 +02:00
Nikolay Amiantov
9134f9358a stdenvAdapters.useOldCXXAbi: add new adapter 2016-03-31 19:52:33 +03:00
Vladimír Čunát
5ad448df3a enableDebugging: use more suitable gcc flags
-ggdb is supposed to provide better symbol information when used with gdb
-Og is the recommended optimization level for debugging purposes
2015-08-10 09:11:30 +02:00
Eelco Dolstra
a58356c953 useGoldLinker: Fix appending to $NIX_CFLAGS_LINK 2015-05-04 14:00:45 +02:00
Ambroz Bizjak
0a8ca604b4 Fix cross compiling regression due to gcc->cc renaming. 2015-02-05 21:25:25 +01:00
Eric Seidel
48f63c2f2e rename gcc-wrapper to cc-wrapper.
also makes cc-wrapper compatible with clang in the darwin fork.
2015-01-14 20:26:56 -08:00
Shea Levy
bbd2be6075 Restore accidentally-dropped stdenv adapter 2014-12-26 12:38:18 -05:00
John Wiegley
28b6fb61e6 Change occurrences of gcc to the more general cc
This is done for the sake of Yosemite, which does not have gcc, and yet
this change is also compatible with Linux.
2014-12-26 11:06:21 -06:00
Eelco Dolstra
a0f3faf34e Merge remote-tracking branch 'origin/glibc-2.20' into staging 2014-11-16 22:17:36 +01:00
Eelco Dolstra
7a05d33386 useGoldLinker: Don't set dontStrip
Probably caused by a bad cut&paste...
2014-11-12 15:10:54 +01:00
Eelco Dolstra
d135422009 Remove obsolete useKlibc stdenv adapter 2014-10-29 13:43:00 +01:00
Eelco Dolstra
5cc92eb0d8 Remove obsolete useDietLibC stdenv adapter 2014-10-29 13:43:00 +01:00
Eelco Dolstra
ffbdcbfe29 gcc-wrapper: Wrap ld.gold
This ensures that gcc's ‘-fuse-ld=gold’ flag works.
2014-10-10 14:55:17 +02:00
Eelco Dolstra
811de3bfaa Merge remote-tracking branch 'origin/staging' into darwin-clang-stdenv
Conflicts:
	pkgs/tools/security/gnupg/default.nix
2014-10-09 10:57:57 +02:00
Eelco Dolstra
14d6454b9a stdenv-linux: Turn allowedRequisites back on 2014-10-06 22:11:05 +02:00
Joel Taylor
afa1db198b fix even more clang builds 2014-09-09 13:54:57 -07:00
Eelco Dolstra
5ba24cc8ea Typo/comment 2014-04-22 18:42:44 +02:00
Eelco Dolstra
497997cc38 Move generation of coverage reports from nixos/lib/testing to releaseTools
Also, turn some stdenv adapters into setup hooks.
2014-03-03 13:57:08 +01:00
Eelco Dolstra
80647127a3 Turn the coverage analysis stdenv adapters into setup hooks
Stdenv adapters are kinda weird and un-idiomatic (especially when they
don't actually change stdenv).  It's more idiomatic to say

  buildInputs = [ makeCoverageAnalysisReport ];
2014-02-05 19:18:33 +01:00
Eelco Dolstra
bea2b3c517 addCoverageInstrumentation: Set NIX_CFLAGS_COMPILE as an attribute
This allows it to show up in nix-shell.
2014-02-05 19:18:33 +01:00
Eelco Dolstra
c3d84d15ce Apply makeOverridable to stdenv
This removes the need for hacks like stdenv.regenerate.  It also
ensures that overrideGCC is now stackable (so ‘stdenv = useGoldLinker
clangStdenv’ works).
2014-02-05 19:18:33 +01:00
Eelco Dolstra
7703f04b75 Add a stdenv adapter ‘useGoldLinker’ to force use of Gold 2014-02-05 19:18:33 +01:00
Eelco Dolstra
e2d505b24e More renames 2012-12-28 19:42:10 +01:00
Eelco Dolstra
dcdef5a0fe More random cleanup 2012-12-28 19:37:42 +01:00
Eelco Dolstra
22ba03c123 Random cleanup: use "or" in adapters.nix 2012-12-28 19:35:35 +01:00
Eelco Dolstra
ab3eeabfed Rename buildNativeInputs -> nativeBuildInputs
Likewise for propagatedBuildNativeInputs, etc.  "buildNativeInputs"
sounds like an imperative rather than a noun phrase.
2012-12-28 19:20:09 +01:00