User-space programs should not be using headers extracted from random
kernels, but should in fact use the headers that were used to build
libc; see e.g., this LKML posting by Linus Torvalds on the subject of
Linux API headers at [1].
What is more, the Linux API headers are supposed to be backwards
compatible[2], so there's really no good reason to have more than one
such package, namely the latest one required by a package in our tree.
That is, `kernelHeaders` is not only incorrect but serves no real
purpose: out-of-tree modules use the sources provided by `kernel.dev`;
user space should use `stdenv.cc.libc.linuxHeaders` or the top-level
`linuxHeaders` attribute.
Apart from klibc, nothing in nixpkgs used `linuxPackages.kernelHeaders`,
so the impact of this change is minimal.
[1]: http://lkml.iu.edu/hypermail/linux/kernel/0007.3/0587.html
[2]: https://www.kernel.org/doc/Documentation/kbuild/headers_install.txt
Extracting headers from a grsecurity patched kernel triggers additional
build steps that require gcc plugins. For this to work, we'd need to
add gmp, libmpfr, and libmpc to the build inputs as well as run `make
prepare` before installing the headers (lest the build fail due to
missing files).
Out-of-tree modules use kernel.dev and user space should use the Linux
API headers used to build libc, not headers extracted from random
kernels, so fixing this for grsecurity is pointless.
Note: I ignored the C++ libraries, but it appears we're not currently
using them. Once we do, we'll probably want to put them in a separate
output as well (to prevent non-C++ users from depending on Boost).
This allows them to be found by find-tarballs.nix (which can only see
fetchurl dependencies if they are attributes, not if they're part of a
string context).
Package stopped building after closure-size merge because
buildPerlPackage has docdev in its default outputs.
There is also a decent man page in the source distribution, which I
added to the "man" output.
I supplied meta.outputsToInstall automatically in all
mkDerivation products, but some packages still don't use it.
The reported case: jekyll -> bundlerEnv -> buildEnv -> runCommand.
PR #14326 fixed gradle so that its native code library could find
libstdc++.so.6, but this fix is inoperative now, because this library
is in `"${stdenv.cc.cc.lib}/lib"` now, rather than
`"${stdenv.cc.cc}/lib"`.
Otherwise, cups will be built without systemd support. The upstream
systemd service unit expects cups to have systemd support however. When
cups is built without systemd support, the daemon will fork to the
background. Consequently, systemd will think the service is done and
enter inactive state.
Along with 9de93be6cd448bdce64e7ef5cf80ff8d81592140 and
47330b1732b67e7d5fee283685b48fdcafdc91e6 this patch fixes the nixos
printer test timing out in `waitForUnit("cups.service")`; see
NixOS/nixpkgs#14748
`cp`, `grep`, `chmod`, `sed` executables
no longer found when upgrading from nixos
15.09 to 16.03. Fixed by use of wrapper
script that brings these executables into
`PATH`.
Also fix lpd binaries on 64 bits machines
by use of `callPackage_i686`.
Adds support for camera input devices on OS X.
This required fixing links in output libraries and executables to pull
in the system's CoreFoundation.framework rather than the open source CF
one.
The AVFoundation framework uses a relative path that presumes that
CoreGraphics is a child framework of ApplicationServices. It is not in
the 10.9 SDK.
This patch makes it one by tweaking the framework derivation generator
with special support to address this problem.
Now without some options, reflecting upstream development.
In special, no GTK building (as it was obsoleted on 096)
and no profile selecion, because it will be obsoleted
on next release.
The config.py script doesn't seem to cope very well with symlinks, so
let's pass it the right derivation outputs wherever possible and fall
back to drv.out.
I've disabled the tests because they somehow now seem to cause the build
to fail even though the tests failed *before* the merge of the
closure-size branch, but the whole build didn't fail regardless.
Here is a build from before the closure-size branch merge:
http://hydra.nixos.org/build/34367296
If you have a look at the build log, you already see a bunch of failing
tests (to be exact: the same set of tests that are failing now with the
fix of the preConfigure phase).
Other than that, the build now succeeds on my machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of cmst override
the whole configurePhase, so this hook isn't run at all.
I've also added running the preBuild and postBuild hooks for the sake of
completeness so that it's easier to override the derivation attributes.
This fixes the build of cmst and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of pyqt5 override
the whole configurePhase, so this hook isn't run at all.
This fixes the build of pyqt5 and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of signon override
the whole configurePhase, so this hook isn't run at all.
This fixes the build of signon and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of libkeyfinder
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of libkeyfinder and it now successfully compiles on
my machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of accounts-qt
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of accounts-qt and it now successfully compiles on
my machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of zeal override
the whole configurePhase, so this hook isn't run at all.
This fixes the build of zeal and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of shotcut
override the whole configurePhase, so this hook isn't run at all.
However, this still doesn't fix the package and its already marked as
broken.
The build fails with the following error:
Project ERROR: Unknown module(s) in QT: qml quick webkitwidgets x11extras
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of qgroundcontrol
override the whole configurePhase, so this hook isn't run at all.
Instead of using pushd/popd, I'm making it a bit more readable by just
spawning a subshell with the right working directory.
This fixes the build of qgroundcontrol and it now successfully compiles
on my machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
The configurePhase is overridden here and the preConfigure hook isn't
referenced otherwise in the package, so it's basically dead code.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of fritzing
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of fritzing and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of linssid
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of linssid and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of twmn override
the whole configurePhase, so this hook isn't run at all.
This fixes the build of twmn and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of qtpass override
the whole configurePhase, so this hook isn't run at all.
This fixes the build of qtpass and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of rapcad
override the whole configurePhase, so this hook isn't run at all.
However, this still doesn't fix the build of rapcad, because it's unable
to find GL/glu.h.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of phototonic
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of phototonic and it now successfully compiles on
my machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of keyfinder
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of keyfinder and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of iannix override
the whole configurePhase, so this hook isn't run at all.
This fixes the build of iannix and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of fmit override
the whole configurePhase, so this hook isn't run at all.
This doesn't fix the build of fmit, because it's unable to find
GL/glu.h. However I think this isn't an issue of the fmit package but
rather rooted in propagating the wrong output paths from freeglut.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of dfasma override
the whole configurePhase, so this hook isn't run at all.
This fixes the build of dfasma and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of communi
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of communi and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path.
As the configurePhase is replaced by a qmake path directly from qtbase,
it won't work because the setup-hook from qtbase needs to do some setup
in order for qmake to find its own data files.
So instead of hardcoding that path, we just go for setting the
configureFlags attribute and replacing which with "type -P" so that the
configure script can figure out the right path to qmake on its own.
This fixes the build of libcommuni and it now successfully compiles on
my machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of cutegram
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of cutegram and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of telegram-qml
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of telegram-qml and it now successfully compiles on
my machine.
Built it using the following command, because it wasn't exposed via the
top-level packages attribute:
nix-build -E '
with import ./. {};
qt55.callPackage pkgs/applications/networking/instant-messengers/telegram/telegram-qml {
libqtelegram-aseman-edition = qt55.callPackage pkgs/applications/networking/instant-messengers/telegram/libqtelegram-aseman-edition {};
}
'
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of
libqtelegram-aseman-edition override the whole configurePhase, so this
hook isn't run at all.
This fixes the build of libqtelegram-aseman-edition and it now
successfully compiles on my machine.
Built it using the following command, because it wasn't exposed via the
top-level packages attribute:
nix-build -E '
with import ./. {};
qt55.callPackage pkgs/applications/networking/instant-messengers/telegram/libqtelegram-aseman-edition {}
'
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of cool-retro-term
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of cool-retro-term and it now successfully compiles
on my machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of qmltermwidget
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of qmltermwidget and it now successfully compiles
on my machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of antimony
override the whole configurePhase, so this hook isn't run at all.
This fixes the build of antimony and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of qwt override
the whole configurePhase, so this hook isn't run at all.
This fixes the build of qwt and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
There were two sources of non-determinisim coming into the images. The
first was tar mtimes, the second was pigz/gzip times.
An example image now passes with the --check flag.
Commit 0055c6a introduced a new preConfigure hook that sets the right
qmake path. Unfortunately the mkDerivation attributes of quazip override
the whole configurePhase, so this hook isn't run at all.
This fixes the build of quazip and it now successfully compiles on my
machine.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Fixes#14695
I'm not entirely sure if including `stdenv.cc.cc` in `makeLibraryPath`
is the correct thing to do here. If it's incorrect, please feel free to
ping me.
- Reconfigure OpenGL/CL related packages to propagate the necessary
framework dependencies.
- Disable tests on all versions of c2hs so package sets such as LTS
releases work.
Closes https://github.com/NixOS/nixpkgs/pull/14609.
This update was generated by hackage2nix v20160406-1-ge7059e3 using the following inputs:
- Hackage: 4d5f229e9e
- LTS Haskell: 924977cc28
- Stackage Nightly: 0002f21766
Mainly done because of this in all-packages.nix:
````
cipherscan = callPackage ../tools/security/cipherscan {
openssl = if stdenv.system == "x86_64-linux"
then openssl-chacha
else openssl;
};
````
... and inside cipherscan we want to refer to `openssl.bin`
