Putting the kernel modules into the same output path as the main
VirtualBox derivation causes all of VirtualBox to be rebuilt on every
single kernel update.
The build process of VirtualBox already outputs the kernel module source
along with the generated files for the configuration of the main
VirtualBox package. We put this into a different output called "modsrc"
which we re-use from linuxPackages.virtualbox, which is now only
containing the resulting kernel modules without the main user space
implementation.
This not only has the advantage of decluttering the Nix expression for
the user space portions but also gets rid of the need to nuke references
and the need to patch out "depmod -a".
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This moves libsystemd.so and libudev.so into systemd.lib, and gets rid
of libudev (which just contained a copy of libudev.so and the udev
headers). It thus reduces the closure size of all packages that
(indirectly) depend on libsystemd, of which there are quite a few (for
instance, PulseAudio and dbus). For example, it reduces the closure of
Blender from 430.8 to 400.8 MiB.
This removes locales, bash completion and crap like that. This cuts
6.5 MiB from the NixOS system closure (which unfortunately contains
two copies of util-linux, because of the need to break a dependency
cycle with systemd).
Fixes build against dpdk 16.06
Tested build against linux, linux_latest, linux_3_18, linux_4_1,
linux_4_6, linux_grsec_nixos, linux_chromiumos_3_18.
While this is pre-release, the delta since 10.10.1.0 seems to contain
primarily fixes or internal improvements.
Also cleanup build inputs while we're at it.
Looks to be incompatible with the PaX constification plugin:
> /tmp/nix-build-wireguard-unstable-2016-08-08.drv-0/WireGuard-experimental-0.0.20160808/src/device.c:329:29: error: constified variable 'link_ops' placed into writable section ".data..read_mostly"
static struct rtnl_link_ops link_ops __read_mostly = {
https://hydra.nixos.org/build/39671573/log/raw
See also https://github.com/NixOS/nixpkgs/issues/18209
This prevents systemd and by extension a zillion other packages from
having Python 2.7 in their closure. For example, the closure of
systemd dropped from 133 MiB to 85 MiB.
With the default kernel and thus with the build I have tested in
74ec94bfa2, we get an error during
modules_install:
make[2]: execvp: /nix/store/.../bin/bash: Argument list too long
I haven't noticed this build until I actually tried booting using this
kernel because make didn't fail here.
The reason this happens within Nix and probably didn't yet surface in
other distros is that programs only have a limited amount of memory
available for storing the environment and the arguments.
Environment variables however are quite common on Nix and thus we
stumble on problems like this way earlier - in this case Linux 4.8 - but
I have noticed this in 4.7-next as well already.
The fix is far from perfect and suffers performance overhead because we
now run grep for every *.mod file instead of passing all *.mod files
into one single invocation of grep.
But comparing the performance overhead (around 1s on my machine) with
the overall build time of the kernel I think the overhead really is
neglicible.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Tested by only building the linux_testing attribute, but haven't yet
tested it in production.
I've also fixed the extraMeta.branch attribute.
Verified-with-PGP: ABAF 11C6 5A29 70B1 30AB E3C4 79BE 3E43 0041 1886
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
