Package changes from 3.6:
- CMake exports patch no longer necessary
- Cosmetic purity patch fix
- Build libc++ with private libc++abi headers visible from sources
- Work around bugs in lldb's configure scripts
Intel Integrated Performance Primitives (IPP) speeds up parts of OpenCV
on Intel processors (and compatible). It increases the store path from
220 MiB to 300 MiB, so it defaults to off.
Original patch from Bas van Dijk <v.dijk.bas@gmail.com>.
I tried applying the same change to opencv(2.x). OpenCV 2.x didn't
automatically detect IPP, so I reverted the change.
The error was reported at HaxeFoundation/haxelib#152 and was fixed by
HaxeFoundation/neko#41 in HaxeFoundation/neko@ccc78c2, the latter being
fetchpatch'ed by us now.
This has caused the hxcpp build to fail on i686-linux with an "Invalid
array access" error.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
I also switched the build to depend on openjdk instead of Oracle's proprietary
one. I'm open to suggestions on how to determine the proper value of $JAVA_HOME
in a jdk-agnostic fashion. Right now, I just hard-coded the proper choice for
openjdk.
* pkgs/development/tools/database/sqldeveloper/default.nix:
This effectively reverts 86c283824f
("If cuda headers are presented to nix [...]") and all the following
workarounds that was added due to that commit.
As far as I can tell[1] this hack isn't needed anymore. And moving
includes to $out/usr_include causes pain for cudatoolkit users, so
better get rid of it.
In patches that did more than the $out/usr_include workaround, I only
changed the line back to $out/include instead of re-generating the
patches and fully removing the changed line.
[1]: I build tested blender and caffe, and temporarily added
recurseIntoAttrs to rPackages and haskellPackages so that nox-review
could get proper coverage. However, many of the packages do not build
even before this patch. I also built CUDA samples with cudatoolkit7
that ran fine.
