the kernel itself (and busybox's cifs mount code) are no longer able
to do this in some/most cases and will error out saying:
"CIFS VFS: connecting to DFS root not implemented yet""
Nixos' qemu-vm target is hurt by this, as it wants to mount /nix/store
via cifs very early in the boot process.
this commit just marks the problematic kernels.
An associated commit in nixos will use this info to fix the problem.
spice is a next-generation remote desktop protocol, aimed at virtual
machines.
focus is not just on display/input devices, but clipboard, audio,
video, opengl, smartcards, usb devices as well, no matter if the
virtual machine runs locally or on a remote host.
not everything is implemented yet, and I didn't enable all available
features yet.
Currently, spice is able to make qemu-kvm virtual machines very usable
for workstation guests, with good 2d video support, clipboard sharing,
full resolutions, auto-mouse-grab/ungrab, xinerama / multiple guest
monitors. Good drivers for windows 7 guests are available, as well as
linux Xorg drivers / agents.
Basically, kvm was already the best-performing VM solution (using
virtio drivers), but virtualbox, while slower, had better
desktop-integration support (still wins if you want opengl). Spice
fixes this, making the choice very easy.
Passing install_root=$out isn't a good idea because the install script is going
to pre-pend that prefix to all other paths even though these have the $out
prefix already. The resulting installation is a mess. Instead, we use the
"fake" install prefix "out" and then move all files and directories into the
right place afterward.
adding useful function foldAttr which behave like fold on attr values grouped by name
(without assertions now)
Signed-off-by: Marc Weber <marco-oweber@gmx.de>
With this change, java packages will build with openjdk by default. The
primary driver for this is legal: The build farm is not allowed to
distribute the proprietary Oracle jdk6, and so it is not allowed to
distribute any packages that depend on it. In my view, this is a purely
beneficial change: from the perspective of the build farm, packages will
go from undistributable due to licensing to either distributable or
undistributable due to failed build (if the package doesn't build
properly with openjdk), and from the perspective of the end user it is
very easy to override the jdk on a package-by-package basis or for all
of nixpkgs in the nixpkgs configuration.
This updates the stable version from 21.0.1180.79 to 21.0.1180.81 and introduces
version 22/23 for the beta/dev channels respectively. This needed quite a bit of
patching because beginning in version 22, the seccomp sandbox is considered
legacy (though BPF is still unfinished) and in order to successfully build, we
need to update the patches as well.
I'm merging this right into master for two reasons:
- There are no changes to the derivation if you're building the stable version
(which is the default), except for the upgrade to version 21.0.1180.81.
- Chromium currently has no reverse dependencies that may break due to this
update.
This originally was one single commit (just an update of all channels) until I
discovered the seccomp BPF build failure.
This enables legacy seccomp sandbox by default even on chromium 22, because the
BPF sandbox is still work in progress, please see:
http://crbug.com/139872http://crbug.com/130662
Because the BPF seccomp sandbox is used in case the legacy seccomp mode
initialization fails, we might need to patch this again, as soon as the BPF
sandbox is fully implemented to fall back to legacy seccomp and use BPF by
default.
We now have two patches for "default to seccomp" - one for Chromium 21 and one
for 22 or higher.
Users might want to override the 'src' and 'name' of go from 'hg'.
I make the expression compatible with that.
Aside, I also set GOARM in the wrapper for it to build programs fine on
armv5tel by default.