KQEMU was a linux kernel module for accelerating the QEMU virtual
machine on x86 hardware. Since QEMU 0.11 (and up), there is no support
for KQEMU any more, the focus is now on KVM.
http://wiki.qemu.org/KQemu/Doc
We already enable VNC and SDL support by default and adding spice only
increases the closure size from 513 MB to 518 MB.
Closure size:
du -sch $(nix-store -qR ./result)
This introduces the following changes:
- Validate Robot's SSL server certificates.
- Admin accounts can new be managed with the "admin" property of a
server instance.
- Better and fixed up error reporting.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This breaks some packages (notably liblapack) due to a broken macro
(CTEST_CUSTOM_POST_TEST). See NixOS/nixpkgs#762
This reverts commit ebc424c3ab.
Signed-off-by: Shea Levy <shea@shealevy.com>
The changelog is available at:
http://www.cmake.org/files/v2.8/CMakeChangeLog-2.8.11
I had to create a new patch from scratch, that's why it contains a lot
of differences. CMake developers has rewritten the files
(Modules/Platform/Linux.cmake and Modules/Platform/UnixPaths.cmake) to
comply to their coding style requirements and a lot of elements has
switched from upper to lower case.
Also, the previous patch partially consisted of commenting some
instructions mixed with line removals whereas mine is directlty
removing them in order to avoid useless evaluations and parsing in
resulting files.
I also added a patch that makes dovecot search for
plugins in /var/lib/dovecot/modules. This way, you
can add plugins from several packages without running
into circular dependencies. The module dir needs to
be populated before the dovecot service is started,
for example. This is currently not done in NixOS, so
you need to implement your own service in order to
get the plugins working.
The module patch has not been added to the old 2.1.x
package.
If no config.pulseaudio is explicitely set to false, build with pulse
support, because even if there is no pulse server available, chromium
will fall back to using ALSA.
And we definitely want to avoid that users have to build chromium for
themselves just for the sake of having pulse support. Thanks to @devhell
for actually helping me discovering this (I for myself do always rebuild
Chromium, so I won't notice those kind of things).
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
The commit downgraded the german language dictionary from version 20030222-1 to
20120607 (and broke the build in the process). That change was probably not
intended?
Changes are:
- Ability to set the name of a server (Server.set_name()).
- New sub-command (set-name) for hetznerctl to set server name.
- Show server name in list command.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Changes are:
- Properly detect GPT disklabels and create proper BIOS boot partitions
if necessary.
- Return exit code 100 if reboot is required due to the kernel not
picking up the new partition table.
- Don't include BIOS boot partition in GRUB devices.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
nix lists are not like haskell lists, and tail is an O(n) operation.
This makes recursion using tail less efficient than recursion using
length + elemAt.
Signed-off-by: Shea Levy <shea@shealevy.com>
This fixes two issues:
- Don't _always_ print out NixOS configuration, only when explicitly
requested with -p.
- Add GRUB boot devices to NixOS configuration output.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Slic3r is a G-code generator for 3D printers.
- Math-Clipper and Boost-Geometry-Utils have been bumped to satisfy
Slic3r.
- Slic3r has problems with perl 5.16 due to a locale issue (comma vs
period in floating point numbers). So we use perl 5.14.
- The tests fail, so we skip them. According to the author of Slic3r,
that should be safe:
"Tests failed because of a typo when the 0.9.10b tag was applied.
You can safely ignore the test results, Slic3r will work."
See https://github.com/alexrj/Slic3r/issues/1303
For reference, the errors look like this:
Use of uninitialized value $deg in numeric eq (==) at /tmp/nix-build-perl-slic3r-0.9.10b.drv-0/git-export/t/../lib/Slic3r/TriangleMesh.pm line 328.
# Looks like your test exited with 255 before it could output anything.
This is just a small dependency fix for ExtUtils::CBuilder and
Module::Build to make them build with perl 5.10.
It seems that perl gradually adds CPAN modules into its core. So when
using older perl there typically some more dependencies to take care of.
ExtUtils-CBuilder 0.280202 is not available anymore, so I had to bump it
to 0.280205.
Add missing dependencies for ExtUtilsTypemapsDefault, needed when
building with perl < 5.16. This works for perl 5.16 too.
ExtUtils-ParseXS 3.15 has disappeared from the mirrors, so I had to bump
it to something available from CPAN; version 3.18.
See note from CPAN[1]:
This module [ExtUtilsTypemap] exists merely as a compatibility
wrapper around ExtUtils::Typemaps. In a nutshell, ExtUtils::Typemap
was renamed to ExtUtils::Typemaps because the Typemap directory in
lib/ could collide with the typemap file on case-insensitive file
systems.
The ExtUtils::Typemaps module is part of the ExtUtils::ParseXS
distribution and ships with the standard library of perl starting with
perl version 5.16.
[1] http://search.cpan.org/~smueller/ExtUtils-Typemap-1.00/lib/ExtUtils/Typemap.pm:
For some reason, SANE suddenly stopped recognizing my scanner recently:
| $ scanimage -L
|
| No scanners were identified. If you were expecting something different,
| check that the scanner is plugged in, turned on and detected by the
| sane-find-scanner tool (if appropriate). Please read the documentation
| which came with this software (README, FAQ, manpages).
I was able to remedy this issue by building SANE with the latest version
of the backends package from Git, by adding the following override to
~/.nixpkgs/config.nix:
| {
| packageOverrides = pkgs:
| {
| saneBackends = pkgs.saneBackendsGit;
| };
| }
Easytag has moved to gnome.org and thus this commit also updates and cleans up a
few meta attributes. More information about the move can be found in the
announcement:
https://mail.gnome.org/archives/easytag-list/2012-November/msg00006.html
In order to get it to compile, we need to do a bit of patching, for example the
configure script tries to find libid3tag through pkg-config, but unfortunately
libid3tag doesn't have a *.pc script, so we're patching it out of the configure
script and use NIX_LDFLAGS to inject the library during linking (note the "-lz"
- it's a propagated dependency of libid3tag).
Also added for MP4 support: taglib.
Thanks to @devhell for the notification of the new upstream release.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This is in order to prepare for fetching the build output from the corresponding
target machine to even further speed up deployment.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This should prevent some annoying messages such as:
tar: usr/bin/nix-build: time stamp 2013-07-09 07:45:19 is 0.159248271 s in the f
uture
tar: usr/bin/nix-channel: time stamp 2013-07-09 07:45:19 is 0.159088763 s in the
future
tar: usr/bin/nix-collect-garbage: time stamp 2013-07-09 07:45:19 is 0.15901664 s
in the future
tar: usr/bin/nix-copy-closure: time stamp 2013-07-09 07:45:19 is 0.158948028 s i
n the future
tar: usr/bin/nix-daemon: time stamp 2013-07-09 07:45:19 is 0.158888042 s in the
future
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This fixes a bunch of issues for the NixOps Hetzner backend, because over there,
it's quite difficult to export the references graph without either duplicaing
lots of code or make a bunch of workarounds.
A detailed description about how it works can be found in the
meta.longDescription attribute.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This is for NixOps and the corresponding Hetzner backend and allows for easy
referencing by nix-build using the -A argument.
Basically the Hetzner rescue system uses an older udev version from Debian, so
we need to use shared object major number 0 here.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Thanks to @jcumming for notifying me about this in #nixos:
03:47 < jack_c> aszlig: chromium builds with -Werror by default.
03:47 < jack_c> Putting: werror = "";
03:48 < jack_c> into gypFlags fixes that..
...
03:52 < jack_c> aszlig: agree -Werror is a good linting tool, but it should
probably disabled for distribution.
So, I guess it makes sense in our case, especially because different GCC
versions will issue different warnings.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This should fix support for TLS and in addition, we now _only_ depend on one
library rather than OpenSSL *and* GnuTLS as we no longer need iksemel.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This is a fork of the iksemel library, which is no longer maintained and is
highly broken in regards to TLS support (even in the release versions).
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This allows blivet to be used outside of NixOS and with a different udev SO
major number. Particularily, this is needed for the NixOps Hetzner backend which
is coming soon :-)
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
This is to allow for easy overriding using <some_pkg>.override <overrides> and
might be used by other python modules not directly in pythonPackages.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
