dropbear: bump version to 2016.74

According to https://matt.ucc.asn.au/dropbear/CHANGES there were some
severe security issues fixed.

Signed-off-by: Alexander Dahl <post@lespocky.de>
Acked-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libcurl: bump version to 7.50.0

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

php: bump version to 7.0.9

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

gamin: improve PTHREAD_MUTEX_RECURSIVE_NP patch to fix build issue

In the gamin package, patch
0003-fix-missing-PTHREAD_MUTEX_RECURSIVE_NP.patch was introduced to fix
the build with musl. Indeed, while musl defines "linux", it does not
define PTHREAD_MUTEX_RECURSIVE_NP, but only PTHREAD_MUTEX_RECURSIVE. So
the check was simplified to only verify if PTHREAD_MUTEX_RECURSIVE_NP is
defined.

However, this doesn't work well with uClibc linuxthreads. In uClibc,
PTHREAD_MUTEX_RECURSIVE_NP and PTHREAD_MUTEX_RECURSIVE are not
pre-processor defines, but enum values. For this reason, even if
PTHREAD_MUTEX_RECURSIVE_NP actually exists, #if
defined(PTHREAD_MUTEX_RECURSIVE_NP) is false. So, the gamin code falls
back to using PTHREAD_MUTEX_RECURSIVE.

Except that for uClibc linuxthreads, PTHREAD_MUTEX_RECURSIVE is defined
only if __USE_UNIX98 is defined. For the NPTL implementation,
PTHREAD_MUTEX_RECURSIVE is defined either if __USE_UNIX98 or
__USE_XOPEN2K8 are defined. This strange difference has been reported to
uClibc-ng upstream [1].

However, regardless of this uClibc behavior, using #if defined to check
for the availability of PTHREAD_MUTEX_RECURSIVE_NP is not good. This
commit therefore switches to using a proper AC_CHECK_DECL() autoconf
test, which works regardless of whether the value is #define'd or
defined as an enum value.

This fixes the build of gamin on linuxthreads platforms, such as
Microblaze or m68k.

Fixes:

  http://autobuild.buildroot.net/results/887df97196d7777efbf18a7bee91aa45c1a98700/ (Microblaze)
  http://autobuild.buildroot.net/results/eb4389474e1b30b5c395a07a857da13a66763bdb/ (m68k)

[1] http://mailman.uclibc-ng.org/pipermail/devel/2016-July/001087.html

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/elementary: add path to eolian_cxx

When efl package is build with Eolian support, eolian_cxx tool is
build for the host.

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

efl: add missing select of zlib

The efl package has zlib in its dependencies, but does not select it,
which causes a dependency check error with the following defconfig:

BR2_arm=y
BR2_TOOLCHAIN_EXTERNAL=y
BR2_TOOLCHAIN_EXTERNAL_CUSTOM=y
BR2_TOOLCHAIN_EXTERNAL_DOWNLOAD=y
BR2_TOOLCHAIN_EXTERNAL_URL="http://autobuild.buildroot.org/toolchains/tarballs/br-arm-full-2016.05-1162-g94c7298.tar.bz2"
BR2_TOOLCHAIN_EXTERNAL_GCC_4_8=y
BR2_TOOLCHAIN_EXTERNAL_HEADERS_3_10=y
BR2_TOOLCHAIN_EXTERNAL_LOCALE=y
BR2_TOOLCHAIN_EXTERNAL_INET_RPC=y
BR2_TOOLCHAIN_EXTERNAL_CXX=y
BR2_INIT_NONE=y
BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_EUDEV=y
BR2_SYSTEM_BIN_SH_NONE=y
BR2_PACKAGE_EFL=y

This commit fixes this by selecting the zlib package at the Config.in
level.

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

package/efl: enable Eolian languages bindings

Eolian languages bindings needs C++11, so we needs at least a gcc 4.8
for the host and target variant.
The C++11 support with gcc 4.7 is not sufficient.

Build eolian_cxx for the host only if Eolian support for the target is
selected.

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

alljoyn-tcl-base: new package

AllJoyn Base Services are common services used by many devices,
providing a set of interfaces for different devices to interact and
interoperate with one another.

Thin Client Library is designed to bring the benefits of the AllJoyn
distributed programming environment to embedded systems.

https://allseenalliance.org

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

alljoyn-tcl: 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.

AllJoyn Thin Core Library (AJTCL) is designed to bring the benefits
of the AllJoyn distributed programming environment to embedded
systems.

https://allseenalliance.org

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

alljoyn-base: new package

AllJoyn Base Services are common services used by many devices,
providing a set of interfaces for different devices to interact and
interoperate with one another.

https://allseenalliance.org

Signed-off-by: Fabrice Fontaine <fabrice.fontaine@orange.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
[Thomas:
 - Add dependency on BR2_TOOLCHAIN_HAS_SYNC_4, inherited from the
   alljoyn package.
 - select BR2_PACKAGE_ALLJOYN_BASE_CONTROLPANEL when neither
   BR2_PACKAGE_ALLJOYN_BASE_NOTIFICATION nor
   BR2_PACKAGE_ALLJOYN_BASE_ONBOARDING are selected. This ensures that
   the package will at least build and install one thing.
 - rename the Config.in option prompts from "alljoyn-<foo>" to just
   "<foo>"
 - rework the build command to use a loop rather than duplicate code
 - rework the install command to also use a loop, and to not discard
   errors.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

kmod: bump version to 23

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

domoticz: remove BR2_PACKAGE_BOOST_ARCH_SUPPORTS

The option was globally removed with
https://git.busybox.net/buildroot/commit/package/boost?id=668ce456448d671f30bf98c4d4819a88b0bf9f4e

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

configs: add defconfig for TS-4800

The TS-4800 is supported by mainline Linux as of 4.5 and by U-boot
as of v2016-07.

The package requires the custom ts4800-mbrboot routine.

A post-image script is provided to generate an image that can be
directly written to an SD card.

More details on the board here:
  http://wiki.embeddedarm.com/wiki/TS-4800

Signed-off-by: Damien Riegel <damien.riegel@savoirfairelinux.com>
Signed-off-by: Patrick Keroulas <patrick.keroulas@savoirfairelinux.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libamcodec: needs dynamic library

libamcodec uses <dlfcn.h>, so it requires dynamic library support.

Fixes:

  http://autobuild.buildroot.net/results/b7653c299a60a67b360c98d59d6707ab468ce6f1/

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

nginx-upload: depends on BR2_PACKAGE_NGINX_HTTP

Fixes:

  http://autobuild.buildroot.net/results/64b1bde02346ca67b670f0b095bd3acad707e8cd

Signed-off-by: Bimal Jacob <bimal.jacob@rockwellcollins.com>
Signed-off-by: Matt Weber <matthew.weber@rockwellcollins.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Thomas: use depends on rather than select.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

bind: bump version to 9.10.4-P2

Security fixes: CVE-2016-2775

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

system: do not set hostname and issue for custom skeleton

We expect the custom skeleton to be fully filled with the necessary
files, now. There is definitely no reason we should handle setting the
hostname and the issue file in there. A user using a custom skeleton
should be fully responsible for providing a functional skeleton.

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: do not handle network settings for custom skeleton

We expect the custom skeleton to be fully filled with the necessary
files, now. There is definitely no reason we should handle network
settings in there. A user using a custom skeleton should be fully
responsible for providing a functional skeleton.

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: move the rootfs skeleton choice

In the following commits, we'll be switching more options to be
conditional on the default or custom skeleton.

So, it makes sense that those options come after the choice of a
skeleton.

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: provide no default for custom skeleton path

Being custom means that our default one is not suitable to start with.
So there is no reason to offer it as the default path.

Add a check that it is not empty.

Add a separating empty line, for good measure, too.

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

toolchain: use the <PKG>_TARGET_FINALIZE_HOOKS

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

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/busybox: add comment about BR2_PACKAGE_BUSYBOX_SHOW_OTHERS

BR2_PACKAGE_BUSYBOX_SHOW_OTHERS is a bit special. When Busybox is
enabled, it is a Busybox option. When Busybox is not enabled, it is a
stand-alone option, forcibly enabled.

So we can safely 'select' it without ensuring (via a 'depends on' or
another 'select') that Busybox is enabled.

However, the name of this option does not express the fact that it is
safe to select it without checking Busybox, which can lead to a bit of
time-consuming head-scratching.

To avoid future puzzlement from an unsuspecting observer, add a a big
fat comment that this option can be selected without any dependency on
Busybox.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Romain Naour <romain.naour@openwide.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
[Thomas: slightly improve the wording of one of the comment.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

fcgiwrap: Add a patch to link with libsystemd, not libsystemd-daemon.

Fixes:

  http://autobuild.buildroot.org/results/9b5e70e42e53e7752245f7c43bd01f208fcf76d0/

Signed-off-by: Thomas Claveirole <thomas.claveirole@green-communications.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

fcgiwrap: Bump to 99c942c90063c73734e56bacaa65f947772d9186.

fcgiwrap was not updated since April 2015, but the last GitHub release
is from February 2013.  Therefore, bump fcgiwrap to the latest commit
as it features fixes and improvements, but is unlikely to be part of a
release soon.

Update as well:
* the license file (latest commit features a COPYING file),
* the Config.in URL, as the previous "official" URL now returns an
  HTTP 502 error.

Signed-off-by: Thomas Claveirole <thomas.claveirole@green-communications.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/jemalloc: needs threads

Fixes:
    http://autobuild.buildroot.org/results/73c437fb932a30bae17653173e8ce35cc91757ef/

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Eric Le Bihan <eric.le.bihan.dev@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/mesa3d: add optional support for libsha1

mesa3d supports several sha1 implementations:
https://cgit.freedesktop.org/mesa/mesa/tree/configure.ac?h=12.0#n1167

This patch adds support for libsha1, inspired by the
xserver_xorg-server package:
https://github.com/buildroot/buildroot/blob/master/package/x11r7/xserver_xorg-server/xserver_xorg-server.mk#L247

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

uboot-tools: fix conditional build of FIT support

The removal of the unconditional compilation and linking of some .o files were
missed when porting the patch 0003 to the version 2016.07, leading to errors
like these when FIT support is enabled:

tools/fit_common.o: In function `fit_verify_header':
fit_common.c:(.text+0x0): multiple definition of `fit_verify_header'
tools/fit_common.o:fit_common.c:(.text+0x0): first defined here
tools/fit_common.o: In function `fit_check_image_types':
fit_common.c:(.text+0x10): multiple definition of `fit_check_image_types'

When FIT support is disabled, the build does not fail but it includes FIT
support.

Fix the patch 0003 to be equivalent to the original one.

Fixes:
http://autobuild.buildroot.net/results/7494a5b22e3df1c05cdcc47670deaf54d9e04133
http://autobuild.buildroot.net/results/f98f30bbf9d4cdde37a96b9310873d83dd649c54

Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
Reported-by: Masahiro Yamada <yamada.masahiro@socionext.com>
CC: Fabio Estevam <festevam@gmail.com>
CC: Carlos Santos <casantos@datacom.ind.br>
Tested-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

lvm2: bump version to 2.02.161

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

lftp: bump version to 4.7.3

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

rrdtool: bump version to 1.6.0

 - Remove patch
   0001-Add-configure-option-to-disable-documentation.patch as this is
   now include in the new version.

 - Add dependency on gettext when needed, since it uses libintl if
   gettext functionality is not provided by the C library.

 - Keep autoreconf/gettextize, which are needed for the link against
   libintl to work properly.

Signed-off-by: Roland Franke <fli4l@franke-prem.de>
[Thomas:
 - add dependency on gettext when needed
 - add comment about why autoreconf/gettextize are kept
 - cleanup commit log.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

directfb: depends on NPTL threading

Fixes following autobuild failure:
http://autobuild.buildroot.net/results/162db0d62075a16c7abae1229676e2cdb29fe953/

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
[Thomas: as noticed yb Romain, propagate the new dependency to the
Config.in comment.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

ti-sgx-km: fix ENV typo

Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/alljoyn: fix sparc64 build issue

Some of the serial port highest speed are not defined on
sparc/sparc64, so alljoyn should not use them. This commit adds a
patch to fix that.

Inspired by commit c5e96d8935016456bea342db170ae6a139a8470f.

Fixes:

  http://autobuild.buildroot.net/results/e5b30b5e0d86b44b97410d434b771e1fb7b18de0

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

package/kodi: Add suboption to control MySQL support

In a Buildroot configuration with mysql enabled it might be useful to
be able to build Kodi without mysql support because Kodi may not need
it at runtime.

Kodi can store its internal databases not only locally using sqlite
but also on a mysql server, this allows several local Kodi machines to
share the same databases. When using only one Kodi instance, mysql
support is not needed, and not building mysql support helps reducing
build time.

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

package/kodi: add optional support for libvdpau

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
[Thomas: drop select of libX11 and libXext, since the libvdpau package
already selects them.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/kodi: if libudev is available, we don't need libusb

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

package/kodi: add support for libudev

Most of the times this is not needed because libdrm & mesa3d already
pull in the dependency to udev. Let's add it anyway to make sure that
udev is really used and to document that Kodi makes use of udev.

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

package/kodi: Add suboption to control unrar support

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
[Thomas: tweak Config.in description.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/kodi: Add suboption to control UPnP support

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

package/kodi: re-structure dependencies

Use one package per line to ease review of Kodi 17 version bump, where
10+ packages will be removed. This patch contains no code changes.

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

package/kodi: lib/cximage-6.0: fix compilation with gcc6

This patch can be removed when Kodi 17.0-Krypton is released because
cximage was removed in the git master branch.

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

package/ti-gfx: list supported SoCs in Config.in help text

The ti-gfx package supports a set of SGX implementations which are
used in some TI SoCs, newer parts are not supported by this package.
This patch adds a list of supported TI SoCs to the help text.

Signed-off-by: Lothar Felten <lothar.felten@gmail.com>
[Thomas: update with the new name of packages: ti-sgx-{km,um}.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/libsemanage: add host-audit as a dependency of the host variant

refpolicy requires host-libsemanage. host-libsemanage will fail to
compile without host-audit. This commit adds host-audit as a
requirement for host-libsemanage.

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

package/audit: add support for host variant

host-libsemanage requires host-audit to compile, therefore this patch
adds a host variant for the audit package.

Signed-off-by: Adam Duskett <Aduskett@gmail.com>
[Thomas: remove libcap-ng usage for now, rework commit log.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

lighttpd: bump to version 1.4.40

Drop upstream patch.

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

package/autofs: requires NPTL

autofs uses pthread_condattr_setclock() which is available only with
toolchains with NPTL threads support.

Fixes:
http://autobuild.buildroot.net/results/152/152d917b0078aeb3715cd562cdecf7d0debf924b

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

package/pppd: disable for musl toolchains

For the same reason as for ipsec-tools [1], disable pppd for musl
toolchains.
It also include <sys/cdefs.h> which is not provided by musl [2].

This fixes a compilation error with musl libc because of undeclared
__P and missing sys/cdefs.h include.

Propagate the dependency to network-manager and rp-pppoe
packages.

[1] 33499484e457018203023cdfdd5286478f2ae381
[2] http://wiki.musl-libc.org/wiki/FAQ#Q:_I.27m_trying_to_compile_something_against_musl_and_I_get_error_messages_about_sys.2Fcdefs.h

Signed-off-by: Romain Naour <romain.naour@gmail.com>
[Thomas: add Config.in comment in network-manager.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/ti-sgx-um: fix Config.in

The continuation line was fogotten. However, it fits on a single line.

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

dtach: bump to version v0.9

As part of the v0.9 release, development moved from sourceforge to
GitHub.

Signed-off-by: Frank Hunleth <fhunleth@troodon-software.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

legacy: properly handle systemd compatibility libs

The legacy symbol for systemd compatibility libraries was not selecting
the BR2_LEGACY symbol, so the buld-time check would not kick in.

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

package/alljoyn: needs BR2_TOOLCHAIN_HAS_SYNC_4

After applying a patch like [1], the build failed with the following
error:
liballjoyn_c.a(Message.o): In function `alljoyn_message_destroy':
Message.cc:(.text+0x74): undefined reference to `__sync_sub_and_fetch_4'
liballjoyn.so: undefined reference to `__sync_bool_compare_and_swap_4'
liballjoyn.so: undefined reference to `__sync_add_and_fetch_4'

So, add dependency on BR2_TOOLCHAIN_HAS_SYNC_4 which also disable the
alljoyn build on sparc.

[1] https://git.buildroot.org/buildroot/commit/?id=c5e96d8935016456bea342db170ae6a139a8470f

Fixes:
http://autobuild.buildroot.net/results/cea/cea62a7ed7f1eda6f817d8e27fd645942def25ec

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

dbus: disable bundled initscript

Otherwise it will install $(TARGET_DIR)/etc/rc.d/init.d/messagebus when
building on a redhat/fedora host.
Regardless of that we provide our own initscript.

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

package/ti-sgx-um: new package

This package adds the userspace libraries for the SGX graphics
accelerator of the following Texas Instruments SoCs: AM335x, AM437x,
AM4430, AM5430 It also adds a config file for the libraries and a
System-V init script.

Signed-off-by: Lothar Felten <lothar.felten@gmail.com>
[Thomas:
 - Add Config.in comment about the ti-sgx-km dependency.
 - Add dependency on glibc since the package consists of pre-built
   libraries that can only work with glibc.
 - Add Config.in comment about thread and glibc dependencies
 - Use tabs for indentation everywhere.
 - Use git:// instead of http:// to clone, since http:// doesn't work.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/ti-sgx-km: new package

This package adds the kernel module for the SGX graphics accelerator of the
following Texas Instuments SoCs: AM335x, AM437x, AM4430, AM5430

Signed-off-by: Lothar Felten <lothar.felten@gmail.com>
[Thomas:
 - Add Config.in comment about Linux kernel dependency
 - Extend Config.in help text to indicate that a TI specific kernel is
   needed
 - Fetch using git:// since fetching from http://, since fetching over
   http:// doesn't work.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/rt-tests: disable for too old toolchains

rt-tests doesn't build with some "old" toolchains due to missing local
syscall number definition for the new kernel deadline scheduler on some
architectures. Upsteam added __NR_sched_setattr and __NR_sched_setattr
definition only for common architectures (x86, x86_64, arm) and tile if
not already defined [1].

Instead of adding missing syscall number, avoid building rt-tests
with toolchain compiled with too old kernel headers (i.e
!BR2_TOOLCHAIN_HEADERS_AT_LEAST_3_14)

At runtime, a 3.14 kernel must be used otherwise these syscall are not
available.

[1]
https://kernel.googlesource.com/pub/scm/utils/rt-tests/rt-tests.git/+/ef2dee4232b7cae594ef7ecb7230c4bd18cea3f3

Fixes:
http://autobuild.buildroot.net/results/6fa/6facaeb10588bdf3ff029b882a8ae6ffba815cdc

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

libamcodec: new package

Signed-off-by: Dagg Stompler <daggs@gmx.com>
[Thomas:
 - use the AML_LIBS_STAGING_DIR variable instead of PREFIX/M_PREFIX to
   indicate the compiler sysroot, which allows to use a foreach loop
   in the build commands.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

xapp_xinit: point to target's "mcookie"

The X11 package builds the "mcookie" executable, when selected, into
the directory output/host/usr/bin/mcookie. The xapp_init's "configure"
attempts to find the "mcookie" executable using the host's path. If
the host has an installed "mcookie" application in their /bin folder,
it can influence the build. The following patch forces the expected
mcookie location.

Related:
http://thread.gmane.org/gmane.comp.lib.uclibc.buildroot/50310

Signed-off-by: Yugendra Sai Babu Nadupuru <yugendra.sai.babu.nadupuru@rockwellcollins.com>
Signed-off-by: Matt Weber <matthew.weber@rockwellcollins.com>
[Thomas:
 - remove conditional on the mcookie package being enabled, since
   xapp_xinit depends on the X.org server, and the X.org server
   selects mcookie.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

gpu-amd-bin-mx51: add missing hash

Add hash for amd-gpu-bin-mx51-11.09.01.bin (the framebuffer variant of
the driver).

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

docs/website: adjust presentation of Mind sponsoring

Make the font size of the Mind block the same as all other blocks (it
was smaller) and re-adjust the min height of all blocks so that they
look the same.

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

docs/website: remove FOSDEM sponsors from "current sponsors"

Google and Mind are already credited in the "Past Sponsors" section
for their contribution to the FOSDEM 2016 meeting.

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

docs/website: add OpenWide/Smile as sponsor

As mentionned in
http://lists.busybox.net/pipermail/buildroot/2016-July/167050.html,
OpenWide/Smile has sponsored the Buildroot Summer Camp, by allowing
Romain Naour to participate, and by paying the food expenses for the
event.

Signed-off-by: Romain Naour <romain.naour@smile.fr>
Cc: Jérémy Rosen <jeremy.rosen@smile.fr>
[Thomas:
 - renamed the logo file
 - remove useless white area in the logo, so that it appears bigger on
   the website]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/vde2: sort alphabetically _VDE2_CONF_OPTS items

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

package/vde2: put the \ just one space after the line of _VDE2_CONF_OPTS items

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

package/vde2: remove trailing \ from last item of _VDE2_CONF_OPTS

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

mtd: add option to install mtd integrity test

The mtd tests have proven very useful in testing both flash stability
and JFFS2 changes. Adding an option to install the integrity test.

Signed-off-by: Yugendra Sai Babu Nadupuru <yugendra.sai.babu.nadupuru@rockwellcollins.com>
Signed-off-by: Matt Weber <matthew.weber@rockwellcollins.com>
[Thomas: rename Config.in option, misc improvements in .mk file, add
patch to fix build with uClibc/musl.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/duma: fix build with C++14

Fixes:
    http://autobuild.buildroot.org/results/f7f/f7fc0092ec43a5a4bc38b5e6878842df77f32d3a/
    http://autobuild.buildroot.org/results/7ae/7ae0d5e4821c70077fc1731f578e8222e9d6691b/
    http://autobuild.buildroot.org/results/438/438032a6d6a38c9929cd26dffb8b3c5d1aed9106/

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

package/xcb-proto: add upstream patch for python3.5 build failure v2

Commit 37f0412 (xcb-proto: add upstream patch for python3.5 build
failure) added an upstream patch to fix building with python3.5, but
missed another one.

Add it now.

Fixes:
    http://autobuild.buildroot.org/results/b43/b43dc7013feddccaff913cb304d3ecb073ef6112/
    http://autobuild.buildroot.org/results/7e0/7e08d7ac2eb40033fd928d7c30d9660825964fa3/
    http://autobuild.buildroot.org/results/ca3/ca35393db46d6b08d3e0e9eaa55ff575c8cf36d7/
    [...]

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

package/acpica: disable -Werror

Fixes:
    http://autobuild.buildroot.org/results/d17/d1728c29e90d978405d24bb97f5c8687dc5a6165/
    http://autobuild.buildroot.org/results/2ec/2ecdf0e87a76781b66098d034768fc8de657faaa/

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

package/dovecot-pigeonhole: bump version to 0.4.15

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

boot: add ts4800-mbrboot package

This MBR routine is based on TS' original bootloader.  It is loaded by
the bootrom stored in the companion FPGA, and chainloads an executable
located at the beginning of the first non-fs (0xda) partition.

Signed-off-by: Patrick Keroulas <patrick.keroulas@savoirfairelinux.com>
[Thomas:
 - add hash file.
 - install image in INSTALL_IMAGES_CMDS instead of
   INSTALL_TARGET_CMDS.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

nginx-nasxi: new package

Naxsi is a third party nginx module reads a small subset of simple rules
containing a list of known patterns involved in website vulnerabilities.
This module behaves like a DROP-by-default firewall for nginx.

Signed-off-by: Adam Duskett <aduskett@codeblue.com>
[Thomas:
 - include Config.in file directly from package/Config.in and not from
   package/nginx/Config.
 - improve Config.in help text with more details
 - rename the package prompt from ngx_http_naxsi_module to nginx-naxsi
 - remove NGINX_NAXSI_SOURCE, and fix the definition of
   NGINX_NAXSI_SITE
 - change license from GPLv3 to GPLv2+ with OpenSSL exception
 - cange license file from LICENSE to naxsi_src/naxsi_json.c. The
   LICENSE file exists in the latest Git master of the project, but
   not in the 0.54 tag that we're packaging.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

nginx-upload: new package

A module for nginx web server for handling file uploads using
multipart/form-data encoding (RFC 1867).

Signed-off-by: Bimal Jacob <bimal.jacob@rockwellcollins.com>
Signed-off-by: Matt Weber <matthew.weber@rockwellcollins.com>
[Thomas:
 - rename from nginx-upload-module to nginx-upload. Indeed, we don't
   want all nginx modules to be suffixed with "-module"
 - include the module Config.in file directly from package/Config.in,
   like we do for Python, Perl, Lua or TCL modules.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/audit: Bump to version 2.6.5

- Removed 0001-lib-gen_tables.c-define-EHWPOISON-when-not-available.patch
  as this was applied upstream.

- Removed 0002-Fix-usage-of-audit_status.feature_bitmap.patch as this was
  applied upstream.

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

libgudev: remove incorrect Config.in help text information

libgudev indeed used to be provided directly by eudev, but it's no
longer the case since eudev 3.1.3, and this has been already taken
into account in Buildroot in commit
6cfa5de33e71b46e9ae668550dc77aacabbde663. We simply forgot to update
the comment.

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

package/eudev: libgudev is no longer provided by eudev

libgudev is now provided by the the eponym package libgudev.

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

package/eudev: fix configure options

--enable-libkmod does not exist, but there's --enable-kmod.

Also, add --enable-blkid to the list of options, since we already depend
on it (from util-linux).

Also do not align \ for line continuations.

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

package/eudev: use the description on their home page

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

package/eudev: explicitly disable rules generator if not wanted

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

package/weston: Add demo clients install option

Signed-off-by: Sergi Granell <xerpi.g.12@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/rtl8821au: set CONFIG_RTL8821AU properly

Due to a recent update in the code, CONFIG_RTL8821AU=m is the expected
kernel configuration environment variable to be set, not
CONFIG_RTL8812AU_8821AU=m as was previously set.

Presently the package does effectively nothing.

Signed-off-by: Christian Stewart <christian@paral.in>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/jemalloc: indentation cleanup

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

package/efl: disable libevas jp2k loader

When the support for JP2K is requested, the configure script is looking
for libopenjpeg1.pc which is not provided by the openjpeg package since
it's too recent (currently v2.1).

Mark BR2_PACKAGE_EFL_JP2K as broken until a new version of the efl
stack support the new openjpeg library.

Fixes:
http://autobuild.buildroot.net/results/ff1/ff1dbbc6cffdb5b2550b3613295151ff9fa964d6

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

package/libfastjson: can use threads instead of atomics

When atomic intrisics are missing, libfastjson falls back to using
pthread mutexes to manage atomicity. Of course, this is much less
efficient than atomics, but it does the job.

Propagate the new dependency to rsyslog, the sole user of libfastjson.

Note: rsyslog already depends on threads for itself, but we believe it
is better to have the exact same dependency propagated.

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

package/libfastjson: fix build without threads

libfastjson is supposed to use the __sync_*4 atomic intrinsics, but alas
it is not using them because their ./configure decides they are not
available: it uses AC_TRY_RUN to check for them, and the default is to
decide they are not available, because of cross-compilation.

Besides, one of the source files was not including the generated
config.h, so even after fixing ./configure there was still a build
error.

The first patch is a backport from upstream to fix the missing
inclusion.

The second patch is switching AC_TRY_RUN over to AC_LINK_IFELSE, as the
only thing we're interested in is to check for the presence of the
atomic intrisics, and linking is enough for that.

Fixes:
    http://autobuild.buildroot.org/results/192/1923d0b570adba494f83747a9610ea6ec35f5223/
    http://autobuild.buildroot.org/results/23a/23ac0e742ed3a70ae4d038f8c9eadc23e708f671/
    and many others...

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

package/upmpdcli: needs MMU

... because it uses fork().

Fixes:
    http://autobuild.buildroot.org/results/9ae/9ae5c773e62bbe84f2d0495f074d659fa3dcd55b/
    http://autobuild.buildroot.org/results/8af/8afb9f96bb827b723b859ff73cdf674ca6434e7c/
    http://autobuild.buildroot.org/results/abe/abe9be3b78f26f56035ec805a86ab33b1d145a02/

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Jörg Krause <joerg.krause@embedded.rocks>
[Thomas: propagate to the Config.in comment.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

sdl2_{gfx,image,ttf}: use lower-case package prompt in Config.in

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

sdl2_ttf: new package

Signed-off-by: Peter Thompson <peter.macleod.thompson@gmail.com>
[Thomas:
 - point <pkg>_LICENSE_FILES to a file that actually exists
 - add host-pkgconf to the dependencies
 - remove unneeded CONF_OPTS options
 - pass FREETYPE_CONFIG in the environment to point to freetype-config]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

sdl2_image: new package

Signed-off-by: Peter Thompson <peter.macleod.thompson@gmail.com>
[Thomas:
 - remove all sub-options to select the various image formats. Between
   no formats enabled and all formats enabled, the size difference of
   the library is ~30 KB, so it really isn't worth having all those
   sub-options:
   -rwxr-xr-x 1 thomas thomas  22444 juil. 15 15:51 libSDL2_image-2.0.so.0.0.1
   -rwxr-xr-x 1 thomas thomas  59216 juil. 15 15:52 libSDL2_image-2.0.so.0.0.1
   consequently, we simply enable unconditionally all image formats
   that don't have any dependency, and use "automatic" dependencies
   for the 4 formats that require an external library
 - remove the host package variant, as it isn't used anywhere.
 - remove --with-sdl2-prefix and --with-sdl2-exec-prefix, and instead
   add a dependency on host-pkgconf so that pkg-config is used
 - remove --disable-static, this is handled by the autotools-package
   infrastructure already
 - point <pkg>_LICENSE_FILES to an existing file.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

sdl2_gfx: new package

Signed-off-by: Peter Thompson <peter.macleod.thompson@gmail.com>
[Thomas:
 - Rewrap Config.in help text.
 - Remove unused SDL2_GFX_VERSION_MAJOR variable
 - Add host-pkgconf to the dependencies, since pkg-config is used to
   find sdl2. This fixes the build, otherwise -I/usr/include/sdl2 was
   added to the CFLAGS when sdl2-config was found on the host machine.
 - Remove --with-sdl2-prefix, since this option doesn't exist, and
   using pkg-config is a better solution (--with-sdl-prefix exists,
   but is not useful when pkg-config is available).
 - Remove --enable-static, packages are not supposed to pass such an
   option, it's not automatically by the autotools package
   infrastructure.
 - Pass --disable-sdltest instead of --disable-sdl2test since the
   latter doesn't exist, while the former does.
 - Pass SDL2_GFX_AUTORECONF = YES since the configure/Makefile.in are
   not up-to-date and therefore it tries to use aclocal at build time.
 - Adjust _LICENSE_FILES, since there is no file named 'LICENSE', use
   the smallest source file instead, since it contains the license
   text.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

qt: add option for enabling the accessibility support

This adds an entry in the configuration menu in order to enable or
disable the accessibility support.

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

package/xserver_xorg-server: add version 1.17.4

Some old binary blobs drivers only implement an old VIDEODRV ABI. This
is the case for the AMD Catalyst driver for example.

Such a situation already exists with the nvidia-tegra23, that only
support the VIDEODRV ABI 14.

Since VIDEODRV ABIs are not backward compatible [0], lets introduce an
older Xserver version that supports such an old ABI.

0. https://www.x.org/wiki/XorgModuleABIVersions/

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

package/rt-tests: disable for musl toolchains

The cyclictest program doesn't build with musl toolchains due to
several issues with a different implementation of the "struct sigevent"
from Glibc/uClibc and musl (similar to [1]).

The pi_stress program doesn't build with musl toolchains with the
following error since musl doesn't provide _POSIX_THREAD_PRIO_INHERIT
define even if musl have Priority Inheritance mutexes [2]:

src/pi_tests/pi_stress.c:78:2: error: #error "Can't run this test without PI Mutex support"
 #error "Can't run this test without PI Mutex support"

Since the fix for cyclictest is not trivial, disable rt-tests for musl
toolchains.

Fixes:
http://autobuild.buildroot.net/results/c27/c272ef7df6190cbb688e6db0b1ee210e6fbb472c

[1] https://bugs.mysql.com/bug.php?id=80322
[2] http://www.openwall.com/lists/musl/2016/03/18/6

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

package/rt-tests: bump to 1.0

Rebase patch 01 and 02 on v1.0
Remove upstream patches 03 and 04.

Switch to release tarball (.xz) and use BR2_KERNEL_MIRROR

Remove HASPYTHON since it has been removed from patch 01 and clear
PYLIB if PYTHON package is not selected.

Remove NPTL handling since the patch upstreamed by Alexey Brodkin
has been reverted [1]

Add hash file

[1] https://git.kernel.org/cgit/utils/rt-tests/rt-tests.git/commit/?id=6f3c1ba9e8403cff6ca12351c43bde68e0fbca25

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

package/rt-tests: switch to the upstream repository

rt-tests is not maintained from Clark Williams's repository anymore.

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

olimex_a20_olinuxino_lime: bump u-boot to 2016.07

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/libgpg-error: disable tests

Tests are not usefull on an embedded device.

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

python-prompt-toolkit: new package

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

python-wcwidth: new package

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

libassuan: bump version to 2.4.3

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

libgcrypt: bump version to 1.7.2

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