Adds the static overlay that can be used to build Nixpkgs statically.
Can be used like:
nix build pkgsStatic.hello
Not all packages build, as some rely on dynamic linking.
(cherry picked from commit 6d90a8b894)
This is kind of a mess, but basically:
- static=true, shared=true means to build statically but move it to
the static output
- static=true, shared=false means to build statically and leave it in
the main output
- static=false, shared=true means to not build static at all
Confusingly, the old default was static=true, shared=true even though
static=false? Still can’t figure out what was meant by that.
(cherry picked from commit e999def159)
- makeStaticBinaries don’t work on Darwin (no stable ABI!)
- Need to make sure NIX_CFLAGS_LINK appends
- isStatic is not used anymore
(cherry picked from commit 8726f6a558)
crossOverlays only apply to the packages being built, not the build
packages. It is useful when you don’t care what is used to build your
packages, just what is being built. The idea relies heavily on the
cross compiling infrastructure. Using this implies that we need to
create a cross stdenv.
(cherry picked from commit a3a6ad7a01)
Adds the static overlay that can be used to build Nixpkgs statically.
Can be used like:
nix build pkgsStatic.hello
Not all packages build, as some rely on dynamic linking.
This is kind of a mess, but basically:
- static=true, shared=true means to build statically but move it to
the static output
- static=true, shared=false means to build statically and leave it in
the main output
- static=false, shared=true means to not build static at all
Confusingly, the old default was static=true, shared=true even though
static=false? Still can’t figure out what was meant by that.
Nixpkgs is meant to link everything dynamically. We don’t want to
expose static packages at the top level. If some package needs
statically built binaries, it should use a custom override.
crossOverlays only apply to the packages being built, not the build
packages. It is useful when you don’t care what is used to build your
packages, just what is being built. The idea relies heavily on the
cross compiling infrastructure. Using this implies that we need to
create a cross stdenv.
eabihf is an abi that can be used with ARM architectures that support
the “hard float”. It should probably only be used with ARM32 when you
are absolutely sure your binaries will run on ARM systems with a FPU.
Also, add an example "armhf-embedded" to match the preexisting
arm-embedded system. qmk_firmware needs hard float in a few places, so
add them here to get that to work.
Fixes#51184
This is needed to build linux kernels on darwin. While we can’t
actually use it to load kernel modules, we can use it to create
indexes of modules on macOS.
(cherry picked from commit ebb31480e8a45efbaff4c4a212bdfc0b053adde0)
You can use stdenv.hostPlatform.emulator to get an executable that
runs cross-built binaries. This could be any emulator. For instance,
we use QEMU to emulate Linux targets and Wine to emulate Windows
targets. To work with qemu, we need to support custom targets.
I’ve reworked the cross tests in pkgs/test/cross to use this
functionality.
Also, I’ve used talloc to cross-execute with the emulator. There
appears to be a cross-execute for all waf builds. In the future, it
would be nice to set this for all waf builds.
Adds stdenv.hostPlatform.qemuArch attrbute to get the qemuArch for
each platform.
We need to set dontUseImakeConfigure in a few places to prevent imake
from overriding the default configure phase. This packages all have a
configure script that needs to get run:
- Xaw3d
- R
- tkgate
- ssvnc
A few changes needed here:
- We don’t want to guess what our platform is. We can patch location
and targetdir to not put things in an unpredictable directory.
- Make premake4 a native build input.
- Set the premakefile variable.
Adds a configure phase for packages using premake.
premakeConfigurePhase runs the correct premake version for the premake
you are using (see premake_cmd).
Some trivial builders use the name attr to choose the exec name
produced. For example nixos-install,
nixos-install = makeProg {
name = "nixos-install";
src = ./nixos-install.sh;
nix = config.nix.package.out;
path = makeBinPath [ nixos-enter ];
};
When cross compiling, this puts the prog in,
/bin/nixos-install-powerpc64le-unknown-linux-gnu
When separateDebugInfo = true & !stdenv.hostPlatform.isLinux, we
always gave an error. There is no reason to do this. Instead, just
don’t add separate debug info when we aren’t on Linux.
A few things changed:
- no need for stdenv.hostPlatform == stdenv.buildPlatform conditional - this is already done in make-derivation
- put xvfb_run in installCheckInputs
propagateNativeBuildInputs will end up going in the output derivation.
This case is allowed to end up in references because of that. Sorry
for the disruption!
Fixes#50865
ao has been renamed to libfive. Because there’s already a libfive
package, we can just remove the old ao package. No packages appear to
depend directly on it.
KRB5_CONFIG needs to be aviailable to get the right one when cross
compiling. These things should just use pkgconfig, but it’s fairly
easy to hack around this. krb5-config will not be on path because we
don’t know ahead of time what architecture it will be for.
This is one of the rare times when flex is not just needed at build
time. When cross compiling, we will need 2 versions of flex - one for
the local system and one for the cross system.
Matthew Bauer