Commit Graph

1785 Commits

Author SHA1 Message Date
John Ericson
05d26adb0a mkDerivation mesonFlags: Fix arm cpu families
In my haste to unbreak eval, I screwed up and got the bit-widths,
backwards.
2020-04-28 22:56:20 -04:00
John Ericson
cf858e6d57 mkDerivation mesonFlags: Fix cross file logic to handle more cases
Otherwise eval breaks
2020-04-28 16:51:44 -04:00
John Ericson
8245230753 meson: Make target-agnostic
The cross file is added in the `mkDerivation`. It isn't nice putting
build tool-specific stuff here, but our current architecture gives us
little alternative.
2020-04-28 10:55:33 -04:00
Jan Tojnar
b4a6714571
Merge branch 'staging-next' into staging 2020-04-13 18:54:59 +02:00
worldofpeace
5384d72885 setup.sh: add dontPatch
Fixes #85038
2020-04-12 07:04:35 -04:00
Michael Reilly
84cf00f980
treewide: Per RFC45, remove all unquoted URLs 2020-04-10 17:54:53 +01:00
Matthew Bauer
e0fb0df64f generic/setup.sh: allow clobbering env-vars file
If the option ‘noclobber’ is set in Bash, we get an error when we
clobber an already existing env-vars. This is an okay error to ignore,
so just >| instead. Note that >| is NOT a Bashism[[1]].

Fixes #79651

[1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_07_02
2020-04-09 22:39:16 -04:00
Matthew Bauer
e45f16b9f5 stdenv/linux: remove powerpc64le bootstrap tools
These files never existed, so best to not leave the reference. If
someone want to step up to maintain this, that would be fine. I don’t
have the hardware to test these out. In addition, someone tried to use
the bootstrap-tools currently built by Hydra and found that they were
broken in some unclear way.
2020-03-31 15:03:26 -04:00
Graham Christensen
6b361c3a31
armv{5tel,6l,7l}: bootstrap-files: quote URLs for rfc-45
Co-authored-by: Cole Helbling <cole.e.helbling@outlook.com>
2020-03-16 19:04:39 -04:00
Graham Christensen
11b54adeac
armv{5tel,6l,7l}: update bootstrap tools after #80497 2020-03-16 18:11:59 -04:00
Frederik Rietdijk
0eb0ddc4db Merge staging-next into master 2020-03-08 08:11:01 +01:00
Ben Wolsieffer
05fc2a3524 unpack-bootstrap-tools: remove obsolete linker script patching
The linker scripts no longer contain store paths, so this does nothing. More
importantly, libpthread.so is not longer a linker script on ARM, so the patching
would corrupt it.
2020-03-06 06:58:20 +01:00
Ben Wolsieffer
ee39e49b68 stdenv: update ARM bootstrap tarballs
Fixes building glibc 2.30, which requires at least GCC 6.

Picked from the following cross-trunk evaluation:
https://hydra.nixos.org/eval/1568864 based on nixpkgs
commit ac241fb7a5.

armv5tel job: https://hydra.nixos.org/build/111556418
armv6l job: https://hydra.nixos.org/build/111556521
armv7l job: https://hydra.nixos.org/build/111556551
2020-03-06 06:58:20 +01:00
Ben Wolsieffer
4478766b64 make-bootstrap-tools: add libssp to allow stack protector to work 2020-02-18 22:40:30 -05:00
Vladimír Čunát
cc1ae9f21b
make-bootstrap-tools: fixup after glibc update
There's a generated header that got comment about the source header
from glibc.dev, which added unwanted runtime dependency.  Tested:
nix build -f pkgs/top-level/release.nix stdenvBootstrapTools.{aarch64,i686,x86_64}-linux.test
2020-02-10 17:31:35 +01:00
Frederik Rietdijk
03755ed59a Merge master into staging-next 2020-02-09 09:17:51 +01:00
Chuck
5d4821141b stdenv: Remove isArm (use isAarch32 instead)
isArm has been deprecated for three releases.  All references have been
removed.  Tree-wide substitution was performed in #37401 21 months ago.
2020-02-05 10:56:14 -08:00
Vladimír Čunát
cf242659fe
stdenv bootstrap tools: switch one file to official URL 2020-02-05 13:59:18 +01:00
Vladimír Čunát
48a997cd76
Merge #66528: glibc: 2.27 -> 2.30 (into staging)
Includes update of stdenv bootstap tools (for three main platforms)
and many package fixes with new glibc.
2020-02-05 13:41:09 +01:00
Vladimír Čunát
53349b9c27
stdenv bootstrap tools: regenerate "officialy"
From Hydra's binaries for c5aabb0 (19.09).
This time I re-tested full stdenv bootstrap on these three platforms.
https://github.com/NixOS/nixpkgs/pull/66528#issuecomment-582351637
2020-02-05 13:04:13 +01:00
Frederik Rietdijk
e4c28f78e7 Merge staging-next into staging 2020-02-02 15:33:13 +01:00
Maximilian Bosch
ea8ae88f04
Merge branch 'staging' into glibc230 2020-02-01 17:42:03 +01:00
Daiderd Jordan
68513e4071
cmake: move CMAKE_OSX_ARCHITECTURES out of the stdenv
This was initially introduced in 92188d9d17,
not clear how relevant this still is but i686 isn't supported anymore so
disable it explicitly.
2020-01-31 21:52:13 +01:00
Daiderd Jordan
6567823996
stdenv: introduce appleSdkVersion and macosVersionMin 2020-01-31 21:52:11 +01:00
Daiderd Jordan
a826b49c97
stdenv: make darwin builds reproducable
Fixes #21629

Passing these extra linker flags removes both the semi-random uuid
included in most binaries as well as making the sdk version consistent
instead of based on the current os version.

    Load command 8
         cmd LC_UUID
     cmdsize 24
        uuid 70FAF921-5DC8-371C-B814-4F121FADFDF4

    Load command 9
          cmd LC_VERSION_MIN_MACOSX
      cmdsize 16
      version 10.12
          sdk 10.13

The -macosx_version_min flag isn't strictly necessary since that's
already handled by MACOSX_DEPLOYMENT_TARGET.
2020-01-31 21:52:07 +01:00
Luka Blaskovic
c206bf12ab stdenv, darwin: fix setupHook location 2020-01-30 18:35:30 -05:00
Maximilian Bosch
eddfcc32b4
Merge branch 'staging' into glibc230 2020-01-23 11:31:13 +01:00
Andreas Rammhold
cb007e69a1 stdenv: make symlinks that refer to the same output relative
While looking at the graph of all the outputs in my personal binary
cache it became obvious that we have a lot of self references within the
package set. That isn't an isuse by itself. However it increases the
size of the binary cache for every (reproducible) build of a package
that carries references to itself. You can no longer deduplicate the
outputs since they are all unique. One of the ways to get rid of (a few)
references is to rewrite all the symlinks that are currently used to be
relative symlinks. Two build of something that didn't really change but
carries a self-reference can the be store as the same NAR file again.

I quickly hacked together this change to see if that would yield and
success. My bash scripting skills are probably not great but so far it
seem to somewhat work.
2020-01-15 09:26:40 +01:00
Frederik Rietdijk
bc18cc72dd
Merge pull request #77610 from LnL7/darwin-stdenv-python3
darwin-stdenv: bootstrap with python3
2020-01-15 09:24:57 +01:00
Maximilian Bosch
2d36eac780 stdenv: rebuild aarch64 bootstrap files with gcc8
The old bootstrapping files would break with glibc 2.30 as at least gcc6
is needed.
2020-01-14 08:52:25 +00:00
Vladimír Čunát
2aea16c4d6 glibc: depend on libidn2 (and libunistring, transitively)
It's a bit hacky, but ATM I can't see any better way for glibc >= 2.28.

Signed-off-by: Luka Blaskovic <lblasc@tvbeat.com>
2020-01-14 08:27:00 +00:00
Vladimír Čunát
0c75f51304 stdenv bootstrap: disable dejagnu tests
There's a failure - I don't know why.  The checks from "normal" dejagnu
still succeed, so I assume this commit is quite a safe work-around.
2020-01-14 08:27:00 +00:00
Luka Blaskovic
e80a85a057 bootstrap-files: bump WIP 2020-01-14 08:26:58 +00:00
Daiderd Jordan
6328518e98
stdenv: bootstrap darwin with python3
- Replaced python override from the final stdenv, instead we
  propagate our bootstrap python to stage4 and override both
  CF and xnu to use it.

- Removed CF argument from python interpreters, this is redundant
  since it's not overidden anymore.

- Inherit CF from stage4, making it the same as the stdenv.
2020-01-13 11:34:36 +01:00
Dmitry Kalinkin
d05ee9c8ff
pkgsStatic: set BUILD_SHARED_LIBS=OFF for cmake 2020-01-03 11:31:17 -05:00
John Ericson
cfd013813e
Merge pull request #74090 from obsidiansystems/ghcjs-cross-without-cc
stdenv, haskell: bonafied GHCJS cross compilation without stdenv.cc
2019-12-30 16:40:43 -08:00
John Ericson
c1fb358a78 Merge remote-tracking branch 'obsidian/ghcjs-cross-without-cc-common' into ghcjs-cross-without-cc 2019-12-30 18:16:22 -05:00
John Ericson
6078f094c6 pkgs/stdenv/booter.nix: Add comment explaining hasCC trickery 2019-12-30 18:09:45 -05:00
Robert Scott
e1e3df423a allowInsecureDefaultPredicate: fix to use getName
this allows correct operation with packages only having pname and version
specified, resolving issue #73737
2019-12-15 18:01:19 +00:00
John Ericson
c4508df25d Merge remote-tracking branch 'obsidian/ghcjs-cross-without-cc-19.09' into ghcjs-cross-without-cc 2019-11-26 12:58:13 -05:00
John Ericson
8737963735 Fix lib tests
js-ghcjs didn't fit in an existing categor.
2019-11-26 12:57:16 -05:00
John Ericson
6bc456c91c Merge remote-tracking branch 'upstream/master' into ghcjs-cross-without-cc 2019-11-25 00:23:07 +00:00
John Ericson
c739c420db Add support for cross compiling to js-ghcjs
This platform doesn't have a C compiler, and so relies and the changes
in the previous commit to work.
2019-11-25 00:12:38 +00:00
John Ericson
63bd851e95 stdenv: Introduce hasCC attribute
Before, we'd always use `cc = null`, and check for that. The problem is
this breaks for cross compilation to platforms that don't support a C
compiler.

It's a very subtle issue. One might think there is no problem because we
have `stdenvNoCC`, and presumably one would only build derivations that
use that. The problem is that one still wants to use tools at build-time
that are themselves built with a C compiler, and those are gotten via
"splicing". The runtime version of those deps will explode, but the
build time / `buildPackages` versions of those deps will be fine, and
splicing attempts to work this by using `builtins.tryEval` to filter out
any broken "higher priority" packages (runtime is the default and
highest priority) so that both `foo` and `foo.nativeDrv` works.

However, `tryEval` only catches certain evaluation failures (e.g.
exceptions), and not arbitrary failures (such as `cc.attr` when `cc` is
null). This means `tryEval` fails to let us use our build time deps, and
everything comes apart.

The right solution is, as usually, to get rid of splicing. Or, baring
that, to make it so `foo` never works and one has to explicitly do
`foo.*`. But that is a much larger change, and certaily one unsuitable
to be backported to stable.

Given that, we instead make an exception-throwing `cc` attribute, and
create a `hasCC` attribute for those derivations which wish to
condtionally use a C compiler: instead of doing `stdenv.cc or null ==
null` or something similar, one does `stdenv.hasCC`. This allows quering
without "tripping" the exception, while also allowing `tryEval` to work.

No platform without a C compiler is yet wired up by default. That will
be done in a following commit.
2019-11-25 00:12:38 +00:00
Lily Ballard
d45d6205de setup.sh: rewrite stripHash
Rewrite the `stripHash` helper function with 2 differences:

* Paths starting with `--` will no longer produce an error.
* Use Bash string manipulation instead of shelling out to `grep` and
  `cut`. This should be faster.
2019-11-12 14:38:41 +01:00
John Ericson
5b6da99ea0 stdenv: Don't unset propagated*DepFiles before main phases
A bunch of stdenv-internal variables were deleted in
1601a7fcce, but these are needed in the
fixup phase, whereas the rest are just needed for the initial work
(findInputs, etc) before the user phases.

CC @matthewbauer
2019-11-05 16:32:22 -05:00
Matthew Bauer
9ffedfef81
Merge pull request #69603 from matthewbauer/unset-unused-setup-var
Unset unused variables in setup.sh
2019-11-05 14:43:11 -05:00
Frederik Rietdijk
c4e30cf98c Merge staging-next into staging 2019-11-05 14:18:08 +01:00
Dmitry Kalinkin
cf8a2d0225
Revert "stdenv/check-meta: getEnv if the attribute is unset (#72376)" (#72752)
This reverts commit 71184f8e15.
2019-11-03 20:38:35 -05:00
zimbatm
71184f8e15
stdenv/check-meta: getEnv if the attribute is unset (#72376)
There were two issues:

* builtins.getEnv was called deep into the nixpkgs tree making it hard
  to discover. This is solved by moving the call into
  pkgs/top-level/impure.nix
* when the config was explicitly set by the user to false, it would
  still try and load the environment variable. This meant that it was
  not possible to guarantee the same outcome on two different systems.
2019-11-03 17:40:43 +00:00