This fixes the "sliding window" principle:
0. Run packages: build = native; host = foreign; target = foreign;
1. Build packages: build = native; host = native; target = foreign;
2. Vanilla packages: build = native; host = native; target = native;
3. Vanilla packages: build = native; host = native; target = native;
n+3. ...
Each stage's build dependencies are resolved against the previous stage,
and the "foreigns" are shifted accordingly. Vanilla packages alone are
built against themsevles, since there are no more "foreign"s to shift away.
Before, build packages' build dependencies were resolved against
themselves:
0. Run packages: build = native; host = foreign; target = foreign;
1. Build packages: build = native; host = native; target = foreign;
2. Build packages: build = native; host = native; target = foreign;
n+2. ...
This is wrong because that principle is violated by the target
platform staying foreign.
This will change the hashes of many build packages and run packages, but
that is OK. This is an unavoidable cost of fixing cross compiling.
The cross compilation docs have been updated to reflect this fix.
Our bootstrap tools are actually broken right now due to busybox not
working when invoked directly from a store path. (It says e.g.
"0qqqw19y4gmknajw8vg4fvhx9gxdqlhz-busybox: applet not found").
Make this test actually fail in such case, the next commit will fix the
problem with busybox.
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outdated configure scripts that don't recognize aarch64.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.
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.
[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
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.
`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.
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.
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.
- 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
- 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.
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`.
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
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.
`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.
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.
The 3.8 upgrade was premature and appears to be breaking a load of stuff
that I can't look into right now. Will take it back to 3.8 and test more
thoroughly before pushing more broadly. I should learn to change fewer
variables at a time :)
It's a long build and generally painful to split into smaller commits,
so I apologize for lumping many changes into one commit but this is far
easier.
There are still several outdated parts of the darwin stdenv but these
changes should bring us closer to the goal.
Fixes#18461
I broke this in the cleanups I did in 171c7f0, the gcc inside the bootstrap
tarball is not getting built with the correct --with-fpu, --with-float
etc. options.
Revert a revert of a merge that shouldn't have been in master but was intentionally in staging.
Next time I'll do this right after the revert instead of so far down the line...
This reverts commit 9adad8612b.
This will break part of the bootstrap tools tests because the new tools
need some changes in the stdenv, but if I change them all at once, the
stdenv breaks with the old bootstrap tools. So I'm doing this first, then
will make changes to the stdenv once this bundle is built and I can use
it.
I also added some functionality to let me test one set of bootstrap tools
on another nixpkgs tree, which makes testing a lot more pleasant.
- cloog, ppl, cloogppl aren't used by recent GCCs. Kill references to them.
- Use correct versions of isl, as the current GCC depends
on non-default versions of them.
- Also clarify isl dynamic libraries are needed in cross
builds, but not in native builds
- Since aeb3d8c (bzip2: fix cross build on mingw by using autoconf patch),
it seems that the bzip2 binary depends on libbz2 when cross compiling.
So copy libbz2 into the bootstrap tarball as well.
- Curl isn't used in the bootstrap tools since e6f61b4cf3.
This reverts commit debd401b0f.
We must not use a single-binary build for the bootstrap since the common
binary gains a dynamic linkage to gmp (due to 'factor' and 'expr'
handling arbitrary-precision arithmetic).
Our coreutils now uses single-binary-build mode where, by default,
simple shebang scripts are used for all the binaries. That doesn't work
e.g. with the Linux unpacker which only handles standard binaries and
symlinks. Let's use the symlinked mode instead for boostrapping.
This does NOT change any stdenv hashes.
I only tested the case most important to me:
$ nix-build pkgs/top-level/release.nix -A stdenvBootstrapTools.x86_64-linux.test
Close#15803. This avoids the error:
while setting up the build environment: executing
‘/nix/store/7sb42axk5lrxqz45nldrb2pchlys14s1-bash-4.3-p42/bin/bash’:
Argument list too long
Note: I wanted to make it optional based on buildCommand length,
but that seems pointless as I'm sure it's less performant.
Amended by vcunat:
https://github.com/NixOS/nixpkgs/pull/15803#issuecomment-224841225
On Linux, paxctl's setup hook should overwrite the paxmark stub, but the
stub is defined after the setup hooks are sourced, so the stub ends up
overwriting the real function. The result is that paxmark fails to do
anything. The fix is to define the stub before any setup hooks are
sourced. Thanks to @vcunat for figuring this out.
Closes#15492
This changes cygwin stdenv, but I don't think it will hurt much people.
This allows mkDerivation to get "dontRebase=true" to skip the usual cygwin
rebase. This is required, if we are using this stdenv to build DLLs for win32
inside x86_64-cygwin, because /bin/rebase crashes at finding an arch mismatch.
Additionally, we don't need any rebase for libraries built by visual studio and
meant for visual studio (my use case).
I'm using nix in x86_64-cygwin to build libraries with visual studio, both for
x86_64 and x86.
I'm giving this up. Feel free to find some reasonable variant that works
at least on Linux and Darwin. Problems encountered:
- During bootstrap of Darwin stdenv `env -0` and some bash features
don't work.
- Without `env -0` the contents of some multi-line phases is taken as
variable declarations, which wouldn't typically matter, but the PR
wanted to refuse bash-invalid names which would be occasionally
triggered. This commit dowgrades that to a warning with explanation.
It turned out that process substitution fed into a while-cycle
isn't recognized during darwin bootstrap:
http://hydra.nixos.org/build/35382446/nixlog/1/raw
Also fix broken NIX_DEBUG output, noticed by abbradar.
The set/env fix in #14907 wasn't very good, so let's use a null-delimited
approach. Suggested by Aszlig.
In particular, this should fix a mass-breakage on Darwin, though I was
unable to test that.
bash variable names may only contain alphanumeric ASCII-symbols and _,
and must not start with a number. Nix expression attribute names however
might contain nearly every character (in particular spaces and dashes).
Previously, a substitution that was not a valid bash name would be
expanded to an empty string. This commit introduce a check that throws
a (hopefully) helpful error when a wrong name is used in a substitution.
For every *.{exe,dll} in $output/bin/ we try to find all (potential)
transitive dependencies and symlink those DLLs into $output/bin
so they are found on invocation.
(DLLs are first searched in the directory of the running exe file.)
The links are relative, so relocating whole /nix/store won't break them.
The hook is activated on cygwin and when cross-compiling to mingw.
Close#14335.
Since 89036ef76a, when a package doesn't include a configure script,
the build complains with:
grep: : No such file or directory
grep: : No such file or directory
This prevents that.
Commit 2040a9ac57 changed the order of
$PATH elements, causing initialpath to appear after buildInputs. Thus
gnugrep ended up depending on bin/sh from bootstrapTools, rather than
from pkgs.bash. The fix is to provide pkgs.bash via buildInputs rather
than initialPath.
http://hydra.nixos.org/build/33276697
For some reason, the current bootstrap tools fail to build gettext:
init2.c:37: MPFR assertion failed: (64 - 0) == ((64 - 0)/8) * 8 && sizeof(mp_limb_t) == ((64 - 0)/8)
libxml/xpath.c: In function 'xmlXPathCompPathExpr':
libxml/xpath.c:10627:1: internal compiler error: Aborted
xmlXPathCompPathExpr(xmlXPathParserContextPtr ctxt) {
^
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
make[5]: *** [libxml/libxml_rpl_la-xpath.lo] Error 1
I didn't investigate why this is the case but rebuilding the bootstrap
tools seems to help.
I used this old-ish WIP branch https://github.com/dezgeg/nixpkgs/commits/arm-bootstrap
since latest master has even more problems with cross-compiling anything.
(I will eventually push this stuff and make the ARM bootstraps build on hydra.)
The $lib output refers to the terminfo database in $out, which is about
10x larger than the ncurses shared library. Splitting these outputs
saves a small amount of space for any derivations that use the terminfo
database but not the ncurses library, but we do not have evidence that
any such exist.
Otherwise, when building glibc and other packages, the "strip" from
bootstrapTools is used, which doesn't recognise some tags produced by
the newer "ld" from binutils.
I assume there's not much use for it during bootstrapping.
This fixes them as well, as curl was compiled against libnghttp2 but the
lib wasn't copied to the bootstrap tools.
Fixes#12632.
I think it's better to quote this variable in general, because it is
common and even documented to pass space-separated commands in there.
The greps should just fail in that case and `if` won't proceed
which seems fine for such cases, and it's certainly better than
passing additional unintended parameters to grep
(which was happening all the time before).
Doing it in an openssl setup hook only works if packages have openssl
as a build input - it doesn't work if they're using a program linked
against openssl.
Commit 6d928ab684 changed this to not
preserve timestamps. However, that results in non-determinism; in
particular, it gives us a broken $SOURCE_DATE_EPOCH (especially for
everything using fetchFromGitHub). Builds affected by timestamps <
1980 should be fixed in some other way (e.g. changing the timestamp to
some fixed date > 1980).
The ld-wrapper.sh script calls `readlink` in some circumstances. We need
to ensure that this is the `readlink` from the `coreutils` package so
that flag support is as expected.
This is accomplished by explicitly setting PATH at the top of each shell
script.
Without doing this, the following happens with a trivial `main.c`:
```
nix-env -f "<nixpkgs>" -iA pkgs.clang
$ clang main.c -L /nix/../nix/store/2ankvagznq062x1gifpxwkk7fp3xwy63-xnu-2422.115.4/Library -o a.out
readlink: illegal option -- f
usage: readlink [-n] [file ...]
```
The key element is the `..` in the path supplied to the linker via a
`-L` flag. With this patch, the above invocation works correctly on
darwin, whose native `/usr/bin/readlink` does not support the `-f` flag.
The explicit path also ensures that the `grep` called by `cc-wrapper.sh`
is the one from Nix.
Fixes#6447
This is used by some build tools to provide reproducible builds. See
https://reproducible-builds.org/specs/source-date-epoch/
for more info.
Later, we'll want to set this to a more intelligent value (such as the
most recent mtime of any source file).
So far if no configure script is found or no makefile,
the rest of the phase is skipped, *including* post-hooks.
I find that behavior unexpected/unintuitive.
Earlier version of this patch had problems due to me assuming
that $configureScript is always a simple path, but that turned out
to be false in many cases, e.g. perl.
This un-hardcodes the bootstrap tools passed into the Darwin stdenv and
thus allows us to quickly iterate on improving the design of the full
bootstrap process. We can easily change the contents of the bootstrap
tools and evaluate an entire bootstrap all the way up to real packages.
The most complex problems were from dealing with switches reverted in
the meantime (gcc5, gmp6, ncurses6).
It's likely that darwin is (still) broken nontrivially.
