- Add python-sphinx and python-sphinx-rtd-theme to Build-Depends-Indep
- Install files from both HTML output directories into the package
- Exclude RST sources from the package
Currently the linux-source-<version> package may or may not include
these files, depending on the order things are built in.
As autotools always modifies the source directory, copy (rsync) the
source to the build directory and build in-place there.
Currently we build-depend on the native python (via asciidoc), and on
the host python (via python-dev). As these are not coinstallable it
is impossible to perform a complete cross-build. Until that's resolved,
this will allow cross-building of most of the package with the
combination of the 'cross' and 'nopython' profiles.
(This also sidesteps the issue of perf wanting a multilib compiler.)
The current cross-compiler packages don't set the Multi-Arch field, so
specify that the cross-compiler package must be native, rather than any
architecture.
flex doesn't support multi-arch, and this would require splitting it
(#611230, #761449). Force use of the native package for now.
openssl doesn't support multi-arch but probably easily could (#827028).
Force use of the native package for now.
We need the native libssl-dev while building the kernel itself and the
host libssl-dev while building tools for linux-kbuild.
Document the state of cross-building in README.source.
linux-kernel-headers was removed from unstable in 2007, so presumably
was last released in etch.
libkxl0-dev still depends on it, so don't remove the Provides yet.
Currently the postinst script distinguishes installation vs upgrade by
checking whether its second argument is empty. However, it will only
be empty if the package was *never* installed before, or was purged.
If the package has been installed, then removed, and is now being
installed again, it's seen as an upgrade.
Unfortunately dpkg only tells the preinst script whether or not this
is an upgrade. So create a flag file there, consume it in postinst
and remove it in postrm.
Use the new linux-update-symlinks command for symlink updates.
Drop support for minimal_swap, no_symlinks, use_hard_links and *_hook
parameters in /etc/kernel-img.conf.
Thanks to Aurelien Jarno and Santiago Vila for pointing out the
ischroot command and that it's in an essential package (i.e. doesn't
add a dependency).
This seems to have been a transitional feature for the switch
from kernel-image to linux-image. mkvmlinuz still uses it but
no hook scripts appear to depend on it.
The list is getting hard to read as we add build profiles. It's also easier
to merge and cherry-pick changes when each dependency is on a separate line.
* Drop redundant gitignore.patch from linux-tools
* Rename linux-tools' debian/templates/control.main.in to
debian/templates/control.tools.in
* Combine changelogs, putting all entries for each upstream release
cycle in chronological order
* Combine rules and gencontrol.py code
Also add a helper executable that implements the check for Hyper-V
(like ConditionVirtualization=microsoft in the systemd units).
(cherry picked from commit 50747480c882dae7d3c8f7b4d0df5cf3276c4ee5)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQIVAwUAVsT0Jue/yOyVhhEJAQo96w/+IZBasj9Vf3KovHNaOXaW1rZL6GRVNzpB
OSNqINfAEtvOR8RuwAcA/zgHy7NNDS9CkDQgjGdJKIhdXXem7mG7Oo3iEwb5osL7
HhUPmeTmXqsa5C6GWzNtERrtX6N8XtBV3inR8iSS46XOm+RHq5UnffqZB35DqUbq
G0h5hiZSXXIXbn0MYJWSPAekAmTXdcevGUoe8bIBD+wH/X4oPINmQLkH6PL5iK7B
KHbf7dJ+JnmbsKJdBpfZvPoa6B1a9FL3HLKkOQjmYF0YjIzZe14JHnvjZNULBmj9
eqmwpfPxAmR5XQgUBu1TPx37FB/nJdMUjZQNz01k0Ybv3l6lKocc7YwX3kViitqe
CiWqbq7LPwN1QNGI6Sf7UB/1XjWJPzWsYoZ0AjDH3ihIPrxmYCdjJstFXr0W3qEG
4e3pJv8gqHxP5SSmE5PQQcS8uiisHmADHZc7C02mxWGDLpqb3gmtgrGqQRFjkk55
tpn8jz2fGGt5VQkhpSBE8wWG4fHRFwiLaAsCVv9jE36pLNq5M9U9VnilCPeoTkti
JcDfPYBjma6RVADIU6tkNnjxJcHFXAqEb5Jn0Wq89GGaS1egAlch4K/XYS+OnIH9
m2Fg4P+TdMePDYxh40tNTIwwHcHLffNfe9K+Ien/DmffuQ87Jk4hfKPW/ZXDO3jO
qdrmLXK2BEg=
=syys
-----END PGP SIGNATURE-----
Merge tag 'debian/4.4.2-1'
Drop rt featureset changes.
Drop patches that are already in 4.5-rc4.
Keep the ABI version at 'trunk'.
Karsten did a test-build on armhf and found failures due to missing modules
and directories.
* The wireless drivers from staging need to be optional; this requires
support for wildcards in optional-inclusion lines
* The check for a containing directory must only be done for non-optional
inclusion lines
Bump the kernel-wedge version requirement and add the optional-include
suffixes.
It would be nice to include the Device Tree model in the reportbug
output. On DT based platforms, /proc/cpuinfo only includes quite
generic information.
Please note that the strange "echo ... $(cat ..)" construct is
intentional. 'cat /proc/device-tree/model' leads to a strange
character at the end because there's no newline and using echo
gets rid of it.
- Use system_failure_message() to generate the error message
- Exit using "die", not passing up the exit code (which will be 0 in case the
process was terminated by signal!)
- Remove redundant path lookup since system() already does that
depmod has been stable for a very long time so there's no reason to
have a configuration variable and debconf question to handle failure.
While we're at it, fix the decoding of $? to handle the exec failure
case ($? == -1).
use source package name in templates so it's easier for derivatives like
src:linux-grsec to not conflict with binaries from src:linux.
Original patch by Ben Hutchings.
Currently we have xz-utils in build-depends-indep as we explicitly
use xz to compress files in linux-source. We also have a build-
dependency on lzma (now a virtual package provided by xz-utils),
but only on armel which was the first architecture for which we
set CONFIG_KERNEL_XZ=y.
However, since 3.7 we've set CONFIG_KERNEL_XZ for all architectures so
xz-utils is needed everywhere. Currently we get away with this
because dpkg-dev depends on xz-utils, but this could change in future.
Build-depend on xz-utils without qualification.
liblockdep is in fairly bad shape - one patch needed to make it build
at all, another to make it build in a separate directory successfully,
and a third to fix the soname. And the install rule doesn't install
its public headers. But maybe it will be useful.
linux-tools-4.1 already exists in experimental. As linux-perf-4.1
installs the same files, it needs to conflict and replace it.
This won't be needed for 4.2 onward.
svn path=/dists/trunk/linux-tools/; revision=22900
Unfortunately it is not sufficient to provide virtual packages
that exactly match the ABI name. For example, 'reportbug kernel'
doesn't find the virtual package, and neither do dpkg commands
such as 'dpkg -l linux-image-$(uname -r)'.
svn path=/dists/sid/linux/; revision=22034
* debian/bin/gencontrol.py: Export internal ABI.
* debian/lib/python/debian_linux/debian.py: Export three-part version.
* debian/rules.real
- Provide more variables.
- Make udeb generation easier.
- Use internal ABI for files.
* debian/templates/control.*: Provide name with internal ABI.
* debian/templates/image.plain.*: Use internal ABI.
svn path=/dists/sid/linux/; revision=22017
- genorig: Include new directory for usbip UAPI header
- debian/control: Update Build-Depends for usbip switching from libsysfs to libudev
svn path=/dists/trunk/linux-tools/; revision=21641
The template file has been moved, but POTFILES.in hasn't been updated
following that move. This way the po files can be updated using:
debconf-updatepo --podir=debian/templates/po
svn path=/dists/sid/linux/; revision=21518
Replace 'powerpc/powernv: Add calls to support little endian host'
with the version committed upstream.
Drop the ABI-preserving changes.
svn path=/dists/trunk/linux/; revision=21452
- Remove versions for debhelper, python, kernel-wedge that are satisfied by stable
- Remove module-init-tools as alternative to kmod, which is in stable
svn path=/dists/sid/linux/; revision=21117
debian/lib/python/debian_linux/debian.py,
debian/lib/python/debian_linux/patches.py,
- Support Python 3.
- Use six if necessary.
* debian/templates/control.main.in, debian/templates/control.source.in:
Depend on python-six.
svn path=/dists/trunk/linux/; revision=20946
The transition from libunwind soversion 7 to 8 is blocked and will
likely remain blocked for some time because libunwind7 is linked into
all ia64 binaries.
For amd64 and i386, downgrade the build-dependency to libunwind7-dev.
For architectures that are supported by libunwind8-dev, remove the
build-dependency and ensure we don't accidentally link with libunwind8
by setting NO_LIBUNWIND=1 when building perf.
svn path=/dists/sid/linux-tools/; revision=20755
Ben Hutchings
Bastian Blank
Mattia Dongili
Martin Michlmayr
Yves-Alexis Perez
Salvatore Bonaccorso
Ben Hutchings
Aurelien Jarno
Maximilian Attems