gnutls: security bump to version 3.4.14

Fixes:
GNUTLS-SA-2016-2 - vulnerability that affects certificate verification
when GnuTLS is used in combination with the p11-kit trust module.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

imagemagick: bump version to 7.0.2-4

Option --without-jp2 removed as it doesn't exist anymore.

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

linux-headers: bump 3.18.x and 4.{1, 4}.x series

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

nasm: bump to version 2.12.02

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

git: bump version to 2.9.1

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

cmake: bump version to 3.6.0

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

toolchain: Bump ARC tools to arc-2016.09-eng006

As described at:
4520524ba055706236db9f00dd79f1b2e2e87fde
this commit continues a series of updates of ARC tools.

This engineering build fixes the kernel dwarf stack unwinder feature for
ARC targets.

We still keep GDB as it is of arc-2016.03 release because there're some
issues we'd like to resolve before releasing it to wider audience.

So again note this is next engineering builds of arc-2016.09 series
and it might have all kinds of breakages, please don't use it for
production builds.

Related to:
4520524ba055706236db9f00dd79f1b2e2e87fde

Signed-off-by: Vlad Zakharov <vzakhar@synopsys.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/syslog-ng: fix localstatdir directory

syslog-ng is built with the default buildroot flag localstatdir=/var

This causes three issues:

1) syslog-ng creates two files (syslog-ng.ctl and syslog-ng.persist in /var)
which shouldn't have random files generated in it.

2) SELinux expects it in it's default directory /var/run.

3) This breakes read only file systems because /var/ usually isn't mounted as
a seperate filing system.

Setting the buildflag to the default /var/run
fixes all three of thee of these issues.

Signed-off-by: Adam Duskett <aduskett@codeblue.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

configs: mx53loco: Bump U-Boot version to 2016.07

Bump U-Boot version to 2016.07 and also remove the extra patch
as it is already part of U-Boot mainline now.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

configs: mx51evk: Bump U-Boot version to 2016.07

Bump U-Boot version to 2016.07 and also remove the extra patch
as it is already part of U-Boot mainline now.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

uboot-tools: bump to version 2016.07

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

opusfile: bump to version 0.8

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

boot/uboot: bump to version 2016.07

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

acpica: needs threads

Fixes:
  http://autobuild.buildroot.net/results/53e/53e38f278ad6d1a7d0a48bce74976f59475b1a99/
  http://autobuild.buildroot.net/results/a7c/a7cc6187451f944e6846d3f6b5cd3aac700796dd/

Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/e2fsprogs: fix 0002 patch

Patch 0002-fuse2fs-might-need-librt.patch introduced a new configure.ac
variable that should contain whether -lrt is needed to gt clock_gettime.

However, it forgot to propagate that variable into the Makefiles.

Because e2fprogs is not using automake (just autoconf), the configure
variables are not automatically propagated; it needs to be done
manually.

Fixes:
    http://autobuild.buildroot.org/results/edf/edf4195d33720d51031a84c3bdb6b6b0ed3b4731/
    http://autobuild.buildroot.org/results/987/987efb687870062c509361eaefd0ff80043392f8/
    http://autobuild.buildroot.org/results/f58/f58b5f8f5cd069b63aae4249bc7aefb145789b7d/
    [...]

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

botan: bump to version 1.10.13

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

htop: bump to version 2.0.1

- drop 0001-native-affinity.patch as it was replaced upstream [1]. Autoreconf
  could be dropped

- drop conf env ac_cv_file__proc_stat=yes ac_cv_file__proc_meminfo=yes because
  these tests no longer run when cross-compiling [2]

- upstream created a custom check function to use ncurses*-config [3] but it
  does not allow to override the path to ncurses*-config, leading to a build
  failure as the htop build system would search the path of host tools.
  A new patch 0001-Allow-to-override-ncurses-config-path.patch allows this
  override [4]. It brings back the need to autoreconf

- set path to ncurses5-config in conf env in order to avoid the htop build
  system finding it in the path of host tools

[1] https://github.com/hishamhm/htop/commit/dfad0afb36df9ac104490454c6472625e3ecbe0c
[2] https://github.com/hishamhm/htop/commit/b56195663760d76ea61d9ed15121dd4a13fa68cd
[3] https://github.com/hishamhm/htop/commit/96c929f82b811f054726f1ba3e5fd065976a1204
[4] https://github.com/hishamhm/htop/pull/524/commits/666f12f60f7d7936af932bf513bc0e9c5c5158f9

Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux-headers: Bump to 4.6.4 kernel

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux: Bump default to version 4.6.4

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

python-tomako: bump to version 0.1.0.post1

Add license file.

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

python-mako: bump to version 1.0.4

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

python-pytablewriter: bump to version 0.5.0

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

python-dataproperty: bump to version 0.7.1

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/libv4l: bump version to 1.10.1

Signed-off-by: Ludovic Desroches <ludovic.desroches@atmel.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/tzdata: bump version to 2016f

Signed-off-by: Martin Bark <martin@barkynet.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/zic: bump version to 2016f

Signed-off-by: Martin Bark <martin@barkynet.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/mesa3d: Fix udev dependency check

Sorry for the typo ;)

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/{mesa3d, mesa3d-headers}: bump version to 12.0.1

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

dante: fix sparc compile issue

Fixes following autobuild failure:
http://autobuild.buildroot.net/results/536d95a46ef9527c6ed38b397f7e1b33f4b7b5fe/

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

imx-usb-loader: remove - from prompt

The prompt of all host packages is "host foo", except for imx-usb-loader
that uses "host-foo", and more precisely "host-imx-usb loader", which is
really weird. This commit fixes this inconsistency.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

mxsldr: new package

Signed-off-by: Erik Stromdahl <erik.stromdahl@gmail.com>
[Thomas:
 - Rename prompt from "host-mxsldr" to "host mxsldr".
 - Add BR2_armeb in the list of architectures under which this package
   can be enabled.
 - Remove double quotes around the _SITE value.
 - The license is GPLv2+, not GPLv2.
 - Add a value for the _LICENSE_FILES variable
 - Rename MXSLDR_DEPENDENCIES to HOST_MXSLDR_DEPENDENCIES, since the
   dependencies of the host package are no longer derived from the
   dependencies of the target package.
 - Remove the MXSLDR_MAKE_OPTS variable that contained the PKG_CONFIG
   variable definition: it is already passed in HOST_CONFIGURE_OPTS.
 - Remove hash file, as it is not checked on Git packages for now.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/runc: use BR_PATH to find pkg-config

When BR2_PACKAGE_LIBSECCOMP is set, go will try to find libseccomp
library using pkg-config. But pkg-config is not in the PATH when go
is called during the runc build.

Set the correct PATH in RUNC_MAKE_ENV.

Fixes:
http://autobuild.buildroot.net/results/fea/fea069fec671b32bffd3d133898d71d45abab552

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Christian Stewart <christian@paral.in>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/{mesa3d, mesa3d-headers}: bump version to 12.0.0

Remove MESA3D_REMOVE_OPENGL_PC, not needed anymore after
https://cgit.freedesktop.org/mesa/mesa/commit/?h=12.0&id=5193fe9f4fe9c2d537afb051323724a399abc4e1

Tested using defconfig from
https://git.buildroot.net/buildroot/commit/package/mesa3d/mesa3d.mk?id=e07a98239969794c2257345b17a604e309a5dbb1

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/ffmpeg: add optional support for libvdpau

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/Config.in: fix alphabetical order

Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libupnp: bump to version 1.6.20

Signed-off-by: Fabrice Fontaine <fabrice.fontaine@orange.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/mono: bump version to 4.4.1.0

This patch bumps mono version to the latest released version 4.4.1.0.

Signed-off-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/samba4: security bump to 4.4.5

Fixes CVE-2016-2119
https://www.samba.org/samba/security/CVE-2016-2119.html

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

openmpi: add Fortran support for MIPS64

We include a file containing cached values for Fortran tests that are
performed at the configure step. These tests fail when cross-compiling
and this is known upstream. See:

https://www.open-mpi.org/community/lists/users/2013/01/21186.php

In that thread the upstream maintainer admits that "cross compiling
OpenMPI is a known issue" and the way to workaround this is to
"pre-populate configure's answers to the Fortran tests (so that it
doesn't actually have to run anything)"

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
[Thomas: use a config cache.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

openmpi: add Fortran support for MIPS32

We include a file containing cached values for Fortran tests that are
performed at the configure step. These tests fail when cross-compiling
and this is known upstream. See:

https://www.open-mpi.org/community/lists/users/2013/01/21186.php

In that thread the upstream maintainer admits that "cross compiling
OpenMPI is a known issue" and the way to workaround this is to
"pre-populate configure's answers to the Fortran tests (so that it
doesn't actually have to run anything)"

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
[Thomas: use the config cache mechanism to preseed the cache variables.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

cifs-utils: disable PIE for m68k

PIE is not supported for uClibc-ng/m68k.

Fixes following autobuild failure:
http://autobuild.buildroot.net/results/4246de0280d4e7f8765adefd379c9dd414c316f1/

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

python-dialog3: new package

Existing python-dialog package supports Python 2 only. For Python 3
another package is required.

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

qt: fix typo in Config.in

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

scanpypi: include LICENCE in the list of supported license files

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

hostapd: Install default hostapd.conf on target

For convenience it's nice to have default hostapd.conf installed
on target automatically as a good starting point.

Otherwise user has to create it manually from scratch.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>i
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/libsndfile: bump version to 1.0.27

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/apache: security bump to version 2.4.23

Fixes CVE-2016-4979:
TLS/SSL X.509 client certificate auth bypass with HTTP/2
http://httpd.apache.org/security/vulnerabilities_24.html

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/pulseaudio: remove WebRTC hook for toolchains without C++

This hook is not required anymore since we handle WebRTC dependency
with --enable/disable-webrtc-aec.

Also the hook doesn't have any effect since we use autoreconf because
it should have been a POST_CONFIGURE_HOOKS.

So, we can remove it safely.

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

musl: bump to version 1.1.15

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

shellinabox: needs MMU support

Fixes:

  http://autobuild.buildroot.net/results/609ec63d36e235581fe0388c0e1102362743d5b2/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

4th: build with -fPIC when shared library build is enabled

Fixes:

  http://autobuild.buildroot.net/results/1e288dabbdd0c5131c18e1426ed6ceedce046b9c/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

alljoyn: new package

The AllJoyn framework defines a common way for devices and apps to
communicate with one another regardless of brands, categories,
transports, and OSes. Developers write applications that discover
nearby devices, and communicate with each other directly and through
the cloud, unleashing new possibilities in the Internet of Things.

https://allseenalliance.org

Signed-off-by: Fabrice Fontaine <fabrice.fontaine@orange.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
[Thomas:
 - rewrap Config.in help text
 - remove / at the beginning of the ALLJOYN_DISTDIR variable.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

4th: create /usr/lib and /usr/bin before installation

The 4th build system doesn't create the installation directories by
itself, causing $(HOST_DIR)/usr/lib to be a file containing the 4th
library if $(HOST_DIR)/usr/lib doesn't already exist as a
directory. Since 4th is the first package in the build order due to
alphabetic ordering, it is very likely that it will install a file as
$(HOST_DIR)/usr/lib, breaking the build of follow-up packages.

Fixes:

  http://autobuild.buildroot.net/results/6291046ca0552e11b79a84df4a7844324d78ab97/
  http://autobuild.buildroot.net/results/252642b7d6ec3f77142dcbd158f56ed8e2426c25/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

docker-containerd: needs CGO linking

Fixes:

  http://autobuild.buildroot.net/results/09a23c8926ce32408fbbfa1fb5b3ed7c4f8733e3/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

python-humanize: new package

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/lapack: new package

lapack is a fortran-based linear algebra math library.

Signed-off-by: Benjamin Kamath <bkamath@spaceflight.com>
[Samuel:
 - Update to use BR2_TOOLCHAIN_HAS_FORTRAN symbol + add comment
   when the toolchain does not meet the requirements.
 - Update powerpc/uclibc dependencies to allow build with musl.
 - Bump to 3.6.1.]
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
[Thomas: move comment about installed libraries from .mk file to
Config.in help text.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/jemalloc: filter on suported architectures

jemalloc has a hard-coded list of supported architectures, which it uses
to define the minimum alignment for allocations. It whines at build time
(not at configure time) when it does not know that alignment.

Fix that by making jemalloc depend on the known-supported architectures.

Fixes:
    http://autobuild.buildroot.org/results/674/674b6022f9a403528a758c0785656d2bda79e0a9/

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Eric Le Bihan <eric.le.bihan.dev@free.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
[Thomas:
 - split ARM and AArch64 dependencies on two lines
 - for SuperH, instead of using BR2_sh && !BR2_sh2, explicitly list the
   SH4 variants, since only SH4 is supported by jemalloc.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/python-pillow: fix wrong install step

This patch changes PYTHON_PILLOW_INSTALL_TARGET_CMDS to actually
install pillow in target directory instead of host.

While at it, it also fixes the version for the hash, and uses the more
conventional "define ... endef" construct to define variables in
python-pillow.mk.

Signed-off-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

4th: new package

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
[Thomas:
 - fix the license information, it's GPLv3+, LGPLv3+.
 - add a comment about the COPYING file containing only the LGPLv3 text,
   even though there is some GPLv3+ code.
 - minor tweaks in the .mk file
 - rewrap Config.in help text.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

opus: don't use assembly when on ARM, but without ARM instructions

The ARM assembly code in opus uses full ARM instructions, which won't
work on Thumb-2 only platforms such as ARMv7-M, so we disable the
assembly code in such situations.

Fixes:

  http://autobuild.buildroot.net/results/ffa12ab7abadd76901228fc82da24e81bb6da625/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libsigsegv: mark as not available on m68k

Fixes:

  http://autobuild.buildroot.net/results/059fd862bd3e9456783f2f4ba307df15dbab0d41/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

runc: add missing dependency on host-pkgconf

runc uses pkg-config to detect libseccomp, so we need to depend on it.

Fixes:

  http://autobuild.buildroot.net/results/72f1cf194843b8519a3ebf213cdbf06873809055/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

runc: depends on CGO linking being supported

Fixes:

  http://autobuild.buildroot.net/results/35d72aac0b0cc0ac92bf309d1a957d7903b1411a/

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

flannel: depends on CGO linking being supported

Fixes:

  http://autobuild.buildroot.net/results/3594f33800857cf57302f9b58e30d68f1d9c33b9/

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

go: define a hidden boolean to specify if CGO linking is supported

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

audit: bump to version 2.6.3

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

audit: fix build with older kernel versions

This commit adds a patch to audit that fixes the build with kernel
headers newer than 3.13, but older than 3.19. The patch has been
submitted upstream.

Fixes:

  http://autobuild.buildroot.net/results/2d3b2f8c2b9f7f2e04b88b1dccb83e183b5876b6/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

boot/uboot: fix missing closing parenthesis

There was a typo in commit
b05ff12b6cd38566003b99291a47a56cbd88f396 ("uboot: install multiple spl
images"), leading to a missing closing parenthesis. This commit fixes
this typo.

Fixes bug #9086

Reported-by: Jebodiah Sensai <dkaplan65@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

tekui: add missing dependency on dynamic library

tekui selects a bunch of Lua modules, which are only available when
!BR2_STATIC_LIBS, so add this missing dependency.

Fixes:

   http://autobuild.buildroot.net/results/0c8ae60a7c0dfb64ee12d557b1660a4a1fc84703/

While we're at it, use the lower-case name of the package in the
Config.in comment.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/gdb: add dependency on host-textinfo

Starting with gdb 7.10, gdb wants to re-generate its documentation.
We were trying to avoid that by patching the Makefiles, but it wasn't
working in all situations. So, we simply add a dependency on
host-texinfo in all case.

Fixes:
http://autobuild.buildroot.net/results/f72/f72eac3e2f995e93f0e8f215e68ce4356a696dcb

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

python-pillow: fix typo on tiff package dependency

Fixes:

  http://autobuild.buildroot.net/results/5ac12f08a366fc15ec138c900df0e5f62d0b70bc/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

support/apply-patches: re-instate set -e

As reported by Sébastien Szymanski [1], the apply-patches script
doesn't stop if a tar command can't extract an archive.

Use "set -e" to exit immediately if a command return an error.

Be sure to ignore any expected error: when we check if a patch to be
applied has the same basename as an already applied patch, the grep
would fail when no such patch was already applied. We should not fail
in this case.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Sébastien Szymanski <sebastien.szymanski@armadeus.com>
Cc: Romain Naour <romain.naour@openwide.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Tested-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

freetype: bump version to 2.6.4

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

lvm2: bump version to 2.02.159

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

Config.in.legacy: fix typo

Thanks to Ben Boeckel who noticed it.

Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

shellinabox: new package

Shell In A Box implements a web server that can export arbitrary command
line tools to a web based terminal emulator. This emulator is accessible
to any JavaScript and CSS enabled web browser and does not require any
additional browser plugins.

Signed-off-by: Olivier Singla <olivier.singla@gmail.com>
[Thomas:
 - Use "select" instead of "depends on" for the OpenSSL dependency, and
   use alphabetic ordering.
 - Rewrap Config.in help text.
 - Use github macro
 - Explain why the OpenSSL dependency is mandatory, while
   --enable-ssl/--disable-ssl are available.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

uclibc: add ldd compile fix from upstream

Fixes a build error, when uclibc utils is selected targeting Blackfin.

Reported-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libiio: BR2_PACKAGE_LIBIIO_TESTS requires BR2_PACKAGE_LIBIIO_XML_BACKEND

Signed-off-by: Paul Cercueil <paul.cercueil@analog.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/python-pillow: new package

This patch adds python pillow, the friendly python image library fork,
it includes a backported patch to disable configuration platfom
guessing.

Signed-off-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
[Thomas:
 - Rewrap Config.in help text.
 - As suggested by Bernd Kuhls, bump the version to 3.3.0 and drop the
   patch that has been applied upstream.
 - Rework the optional dependency handling to follow what we do in most
   Buildroot packages.
 - Add licensing information.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/linknx: make it build with the musl C library

Fixes
http://autobuild.buildroot.net/results/c0f/c0f59d3f9eb115072d410ef199f4773f2c380d6d/

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/sphinxbase: add missing dependency on host-bison

Fixes:

  http://autobuild.buildroot.net/results/741f04fce74b4dfc4d4405d14c82c21f67cbfff1/

Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

dante: new package

Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
[Thomas:
 - Add dependency on BR2_USE_MMU
 - Rewrap Config.in help text
 - PID file goes in /var/run/
 - Daemon to start is /usr/sbin/sockd, not /usr/sbin/dante
 - Remove staging installation, as it is not needed
 - Remove --enable-debug, since that's not what BR2_ENABLE_DEBUG is
   meant for
 - Install an example configuration file.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

Revert "apply-patches: catch unexpected failure"

This reverts commit 9cf1ad6cdb37c5be5772121896f4b13ec42c2b0d.

This breaks many packages.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

apply-patches: catch unexpected failure

As reported by Sébastien Szymanski [1], the apply-patches script
doesn't stop if a tar command can't extract an archive.

Use "set -e" to exit immediately if a command return an error.

[1] http://patchwork.ozlabs.org/patch/626196

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Sébastien Szymanski <sebastien.szymanski@armadeus.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

configs: Use HTTPS URL for Github repos

This makes it possible to clone git repositories hosted on Github while
behind a firewall.

Signed-off-by: Paul Cercueil <paul.cercueil@analog.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

uboot: install multiple spl images

For some platforms, there are multiple generated spl images.  Extend
BR2_TARGET_UBOOT_SPL_NAME to allow these multiple images to be installed
after uboot build completes.

For example, the NextThingCo C.H.I.P. uses two binaries from uboot,
spl/sunxi-spl.bin and spl/sunxi-spl-with-ecc.bin.

Signed-off-by: Jason Abele <jason@nextthing.co>
[Maxime:
 - Add foreach loop for general case and mkpimage
 - Use firstword for zynq case]
Signe-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

circus: new package

Reviewed-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
[Thomas: minor tweaks to Config.in.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/liboping: fix build after bump

Fixes:
    http://autobuild.buildroot.org/results/6a9/6a9b4d7b1b3cd72e32579fbaf5a69dbde0fea8e4/

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

squid: bump version to 3.5.20

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libiio: Bump version to 0.7

The two previous patches have been merged upstream, so they have been
deleted here. Another patch (picked from upstream) has been added to fix
the build with thread-less toolchains.

Libiio v0.7 provides two new backends, a USB backend (using libusb-1.0)
and a serial backend (using libserialport).

Additionally, it is now possible to compile libiio with thread-less
toolchains. In that case, thread safety is disabled.

Signed-off-by: Paul Cercueil <paul.cercueil@analog.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/skeleton: remove useless .empty file

We now have a real file in that directory, so we do not need a .empty
placeholder.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

packages: use the <PKG>_TARGET_FINALIZE_HOOKS

Register package-specific target-finalize hooks with the
newly-introduced <PKG>_TARGET_FINALIZE_HOOKS.

This incidentally fixes luarocks, which was registering target-finalize
hooks even when it was not enabled.

To be noted, the skeleton package is not converted, because it is not
optional, we always have it; so its hooks would always be registered
anyway. Besides, the followup patches would render this conversion moot
anyway, since those hooks would be spread across the various skeleton
packages.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

core/pkg-generic: allow packages to declare target-finalize hooks

Currently, packages using target finalize hooks must remember that they
need to register their hook in TARGET_FINALIZE_HOOKS
conditionally (otherwise their hook will be triggered even if the
package is disabled).

In order to avoid this potential mistake, this commit introduces a
per-package target-finalize hook variable, in which packages can
register their target-finalize hooks, with the guarantee that they will
only be triggered if the package is enabled.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
[Thomas: rework commit log.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/perl: use dummy hostname

The hostname does not look like it serves any useful purpose, except
maybe to set set perladmin email.

Which is undoubtfully useless on the target.

A followup commit will make the hostname depend on the default skeleton,
so it won't always be available. We can not rely on it to be set.

Besides, even today it is not guaranteed to be set; a user may well
leave it empty.

Use a dummy hostname.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/rs485conf: new package

Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Reviewed-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

system: systemd only really supports a R/W rootfs

Supporting a R/O rootfs needs a bit more love to be fully operational.
This will come in later patches...

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

system/skeleton: update etc/mtab with a more sensible link

Currently, our /etc/mtab points to /proc/mounts. This was all neat so
far, and was good for a sysv-like init system.

However, the way today is to point it at /proc/self/mounts, the
per-process mount tab.

Additionnally, that's what systemd expects. If /etc/mtab is not a
symlink to ../proc/self/mounts and the rootfs is readonly, systemd would
whine loudly (and a service unit would be marked failed).

Since it works well for sysv-like init systems too, just use that.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/skeleton: respect variables namespace

In makefiles, variables have global scope. We traditionally ensure that
we get no name clashing by prefixing all variables with the package
name.

Currently, this is not the case in the skeleton package, for historical
reasons (code snippets copied over from the common Makefiles). We
currently have a mix of naming for the variables:
  - some are indeed prefixed with SKELETON_
  - some are prefixed with SYSTEM_
  - some are prefixed with both
  - some are not prefixed

Clean up these discrepancies, and prefix all variables with just
SKELETON_ and drop the SYSTEM_ prefix.

Also include SET_ in all variables that do set something (getty) for
consistency across all the variables.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

binutils: fix build of 2.26 when makeinfo is not available

The 2.26 binutils tarball has incorrect timestamps for the .info files,
so binutils build system tries to regenerate them with makeinfo. In
order to avoid depending on host-texinfo, we simply touch the .info
files so that their timestamp is newer than the corresponding source
files.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Tested-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

gdb: add dependency on host-flex and host-bison for Git versions

gdb versions fetched from git (such as the ARC version) will not build
with host-flex and host-bison.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>