links: fix static link with directfb

The AC_HAVE_FUNCS autoconf macro uses gcc to check for function existence.
This breaks when the library depends on libstdc++, like directfb. Add this
dependency externally.

Fixes:
http://autobuild.buildroot.net/results/548/548019b71c7db649f20d2c479bb8fb3612162d4a/
http://autobuild.buildroot.net/results/c94/c9495f54db388bde9d64325bb57fc4e58741f7a7/
http://autobuild.buildroot.net/results/9b8/9b883d42bff3633f16447d76d2eb4ddf68013788/

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

mongodb: bump to version 3.3.4

Signed-off-by: Sergio Prado <sergio.prado@e-labworks.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

wine: Bump to version 1.8.2

Bumping to stable version 1.8.2

Signed-off-by: André Hentschel <nerv@dawncrow.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux: bump default to version 4.5.1

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux-headers: bump 3.14.x and 4.{4, 5}.x series

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

samba4: security bump to version 4.4.2

Fixes:

CVE-2016-2118 - A man in the middle can intercept any DCERPC traffic
between a client and a server in order toimpersonate the client and get
the same privileges as the authenticated user account.

CVE-2016-2115 - The protection of DCERPC communication over ncacn_np
(which is the default for most the file server related protocols) is
inherited from the underlying SMB connection. Samba doesn't enforce SMB
signing for this kind of SMB connections by default, which makes man in
the middle attacks possible.

CVE-2016-2114 - Due to a bug Samba doesn't enforce required smb signing,
even if explicitly configured.

CVE-2016-2113 - Man in the middle attacks are possible for client
triggered LDAP connections (with ldaps://) and ncacn_http connections
(with https://).

CVE-2016-2112 - A man in the middle is able to downgrade LDAP
connections to no integrity protection. It's possible to attack client
and server with this.

CVE-2016-2111 - When Samba is configured as Domain Controller it allows
remote attackers to spoof the computer name of a secure channel's
endpoints, and obtain sensitive session information, by running a
crafted application and leveraging the ability to sniff network traffic.

CVE-2016-2110 - The feature negotiation of NTLMSSP is not downgrade
protected. A man in the middle is able to clear even required flags,
especially NTLMSSP_NEGOTIATE_SIGN and NTLMSSP_NEGOTIATE_SEAL.

CVE-2015-5370 - Errors in Samba DCE-RPC code can lead to denial of
service (crashes and high cpu consumption) and man in the middle
attacks.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

barebox: bump to version 2016.04.0

Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

graphite2: bump to version 1.3.8

And upstream switches tarball name yet again, so adjust SOURCE.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

sqlite: bump to version 3.12.1

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

weston: don't forcibly disable simple EGL clients

Only do so when there's no EGL.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

weston: check for both EGL and GLES

The conditional is wrong, it was check for BR2_PACKAGE_MESA3D_OPENGL_ES
(correct, means GLESv2 support) and BR2_PACKAGE_MESA3D_OPENGL_GLES
(incorrect, there's no such symbol).
So now check for BR2_PACKAGE_MESA3D_OPENGL_EGL (EGL support) plus
BR2_PACKAGE_MESA3D_OPENGL_ES (GLESv2 support), both are required.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

xorriso: Unify help description

Signed-off-by: Benoît Allard <benoit.allard@greenbone.net>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/exim: bump version to 4.87

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

mesa3d: needs host-bison and host-flex

Fixes (host-bison):
http://autobuild.buildroot.net/results/e91/e919dbe6660a6c8c087969a6646e97dd058ce2c7/
Fixes (host-flex):
http://autobuild.buildroot.net/results/66d/66d45e34a38f15e391500ea7f7036606bbaf2655/

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

iproute2: bump to version 4.5.0

Drop 0003-iproute2-fix-building-with-musl.patch since it's upstream.
Add gentoo patch to allow for non-iptables builds.
Add new musl build fix.
Tweak 0001-Avoid-in6_addr-redefinition.patch since the first chunk is no
longer required.

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

libqmi: bump to version 1.14.2

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

package/Makefile.in: adjust LDFLAGS for elf2flt

So far, our LDFLAGS for the BR2_BINFMT_FLAT case were only used on
Blackfin. However, passing -elf2flt in LDFLAGS is wrong. Indeed,
LDFLAGS are not linker flags, but flags passed to the compiler when
linking.

If you pass -elf2flt to the compiler when linking, it is understood as
"-e lf2flt", i.e "the entry point is named lf2flt", which isn't
exactly the intention. We in fact need to pass -Wl,-elf2flt in LDFLAGS
as well, so that the compiler passes -elf2flt down to the linker.

For some reason, this option does not cause an issue with the Blackfin
toolchain, but it does with either a Buildroot toolchain for Cortex-M
or an OSELAS toolchain for Cortex-M. We have verified that passing
-Wl,-elf2flt continues to work with the Blackfin toolchain.

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

toolchain-buildroot: don't show musl on noMMU platforms

While musl has recently gained noMMU support for the sh2 platform, we
don't support this yet. So for the time being, let's not show musl as
an available C library on noMMU platforms. This is for example
important on ARM noMMU: ARM is supported by musl, but not its noMMU
variants.

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

toolchain-buildroot: update glibc comment for noMMU

glibc is not available for noMMU platforms, so it doesn't make sense
to show the comment about glibc requiring dynamic libraries on noMMU
platforms.

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

musl: build broken on Thumb, use ARM mode

Building the musl C library on Thumb (Thumb1, not Thumb2), fails with:

 {standard input}:20: Error: only lo regs allowed with immediate -- `mov fp,#0'
 {standard input}:21: Error: only lo regs allowed with immediate -- `mov lr,#0'
 {standard input}:25: Error: unshifted register required -- `and ip,a1,#-16'

Since there are no cores that we support that are Thumb1 only, use the
same solution as the one used by glibc: build the C library in ARM
mode.

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

uclibc: context functions are not Thumb-compatible

The context functions in uClibc unconditionally use the classic ARM
instruction set.

On platforms that do support the ARM instruction set, there is no problem -
However, on platforms that only support the Thumb instruction set, the
context functions cannot be built since the assembler code is not
Thumb-ready. Therefore, these functions must be disabled on such
platforms. All Thumb1 platforms support ARM instructions, so this is
only relevant for Thumb2-only platforms (i.e., Cortex-M).

Note that some packages require the context functions, so these will
fail to build on these platforms. It is worth mentioning that musl
also doesn't provide the context functions, and those are rarely
used. Affected packages will be handled in later patches.

[Peter: slightly reworded]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

uclibc: use upstreamed patch for ARM Thumb simplification

The uClibc patches to simplify the ARM Thumb configuration options
have been merged, but instead of being 5 separate patches, they have
been merged as a single patch.

This commit updates the Buildroot uClibc package to use the patch that
was actually upstreamed.

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

sawman: completely remove the package

In commit 11573f5a14d1c2726ce9623de3a5f118085a420f ("sawman: remove
package"), we removed the sawman package, but forgot to remove the
hash file of that package. This commit finalizes the removal by
getting rid of the hash file.

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

libsrtp: security bump to version 1.5.4

Fixes CVE-2015-6360: Prevent potential DoS attack due to lack of bounds
checking on RTP header CSRC count and extension header length.

Also, add a hash file.

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

evemu: bump version to 2.4.0

See [1] for details.

[1] https://lists.freedesktop.org/archives/input-tools/2016-April/001274.html

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

harfbuzz: bump to version 1.2.5

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

mesa: add vc4 driver support

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux-headers: bump 3.18.x and 4.1.x series

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux: Add custom Subversion repository support for the package

Signed-off-by: Pauli Sundberg <susundberg@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

boot/uboot: Add custom Subversion repository support for the package

Signed-off-by: Pauli Sundberg <susundberg@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux: use zImage by default on ARM

Since quite some time, the kernel and bootloader communities consider
zImage as the default format for kernel images on ARM, replacing
uImage. The load address information in uImage is no longer needed,
since the kernel is position-independent in terms of physical address,
except on a few old platforms. For most people, using zImage is simply
better/simpler, so let's switch to zImage as the default image format
on ARM.

All defconfigs are updated: 46 defconfigs no longer need to select
explicitly zImage because it's the default, and 16 defconfigs now need
to explicitly select uImage because that's no longer the default.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Matt Weber <matt@thewebers.ws>
Acked-by: Julien Boibessot <julien.boibessot@armadeus.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

network-manager: bump version to 1.0.12

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

squid: bump version to 3.5.16

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

lvm2: bump version to 2.02.149

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/libshout: bump version to 2.4.1

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

mosh: requires wchar

Fixes:
http://autobuild.buildroot.net/results/dc6/dc60afdf2b6035cd72567cf37b837db087c79bea/
http://autobuild.buildroot.net/results/624/624a1fcae9633940714ae4e8ad9864e9611e93ec/
http://autobuild.buildroot.net/results/e95/e95c8cff40c87f167f82e37430ebd1f37c8c242e/

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/{mesa3d, mesa3d-headers}: bump version to 11.2.0

Rebased 0002-Fix-runtime-error-with-uClibc.patch

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

weston: egl requires gles as well

>From configure.ac lines 83-89:

AC_ARG_ENABLE(egl, [  --disable-egl],,
enable_egl=yes)
AM_CONDITIONAL(ENABLE_EGL, test x$enable_egl = xyes)
if test x$enable_egl = xyes; then
AC_DEFINE([ENABLE_EGL], [1], [Build Weston with
EGL support])
PKG_CHECK_MODULES(EGL, [egl glesv2])
PKG_CHECK_MODULES([EGL_TESTS], [egl glesv2 wayland-client
wayland-egl])

It also requires wayland-egl which is only provided by mesa3d for now,
hence make it conditional on mesa3d egl+gles.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/x11r7/xserver_xorg-server: bump version to 1.18.3

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/x11r7/xproto_inputproto: bump version to 2.3.2

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

git: bump version to 2.8.1

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

librsvg: bump to version 2.40.15

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

feh: bump to version 2.14.2

Signed-off-by: Petr Vorel <petr.vorel@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

wilc1000-firmware: new package

Firmware for Atmel Wilc1000 wireless device.

Signed-off-by: Ludovic Desroches <ludovic.desroches@atmel.com>
[Thomas:
  - add missing hash file
  - simplify installation step]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

php: security bump to version 5.6.20

Fixes (no CVEs yet):
Buffer over-write in finfo_open with malformed magic file.
Invalid memory write in phar on filename with \0 in name.
Parsing of tar file with duplicate filenames causes memory leak.
php_snmp_error() Format String Vulnerability.
Integer Overflow in php_raw_url_encode.

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

linux-headers: bump 3.2.x series

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

fastd: bump to v18

Release notes at http://fastd.readthedocs.org/en/v18/releases/v18.html

First patch rebased, second patch removed (gone upstream). Using tarball
and hashes instead of git now.

Signed-off-by: Alexander Dahl <post@lespocky.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libuecc: bump to v7

>From v6 to v7 just few changes were made for performance improvement and
compatibility.

Switched from git download to source tarballs and hashes.

Signed-off-by: Alexander Dahl <post@lespocky.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

alsa-utils: bump to version 1.1.1

Patches upstream so drop them.

The 'bat' binary utility was renamed to 'alsabat' probably to avoid some
clash, keep the old .config symbol to avoid pointless legacy.

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

alsa-lib: bump to version 1.1.1

Drop 0001-add-missing-include.patch since it's upstream.

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

libgtk3: bump to version 3.20.2

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

nload: new package

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

glibmm: bump to version 2.48.1

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

gnupg2: bump to version 2.0.30

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

configs: rename Zynq defconfig files

These three are Zynq boards and the build procedure is almost the
same.  Having the SoC name prefix "zynq_" would be more consistent.
Also, this is the way in which the Linux Device Trees and the U-Boot
configuration files do.

This commit renames as follows:

   zedboard_defconfig     -> zynq_zed_defconfig
   microzed_defconfig     -> zynq_microzed_defconfig
   xilinx_zc706_defconfig -> zynq_zc706_defconfig

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libdrm: add vc4 (Raspberry Pi) option

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

norm: fix build failure with old headers

The version check for the nfq_get_payload prototype is wrong since it's
done against linux headers when in fact it changed with the 1.0.0
release of libnetfilter_queue, leading to build failure in toolchains
that have old headers when libnetfilter_queue is present. Fixes:
http://autobuild.buildroot.net/results/c0e/c0e897a6db888d4a7e0302cdc1e0cbec7d1e88f3/

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

cannelloni: needs gcc >= 4.8

Uses C++11, fixes:
http://autobuild.buildroot.net/results/4a9/4a9999555c20e495d4981deafc2960940e67febe/

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

package/systemd: Respect BR2_TARGET_GENERIC_GETTY_BAUDRATE

Currently systemd getty services ignore baudrates set in buildroot in
favour of a hardcoded 115200. This patch SEDs out that hardcoded value with
what is selected.

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

uclibc: enable linuxthreads for xtensa

linuxthreads is the only threading option for noMMU xtensa, enable it.

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

gcc: add xtensa uClinux support

Backported from: r228450
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libwebsockets: bump to version 1.7.4

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libwebsockets: fix build with nothread toolchains

>From upstream changelog:
> If LWS_MAX_SMP=1, then there is no code related to pthreads compiled in the
  library. If more than 1, a small amount of pthread mutex code is built into
  the library.

Set LWS_MAX_SMP=1 for toolchains without thread support, otherwise leave it
unset so upstreams CMakeLists.txt will take care of it.

Fixes:
http://autobuild.buildroot.net/results/dbf/dbf45a52b9aef602fd62ad1ca10957bafe656923/
http://autobuild.buildroot.net/results/26c/26c2ac28f44f8f1b227caa715af60bcd92a5f740/
http://autobuild.buildroot.net/results/d65/d65dcbfb96b249e01b034c01a52884b522c778b3/
http://autobuild.buildroot.net/results/e55/e550ebfdd1e8f459ca7b12f2708a704eaec52474/
.. and more.

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libwebsockets: add patch to fix musl build

Add patch from upstream [1] to fix musl build:
fatal error: sys/cdefs.h: No such file or directory

Fixes:
http://autobuild.buildroot.net/results/564/564675be0d4fbfa113a0f2e52874e0dce2d5ba5e/
http://autobuild.buildroot.net/results/c41/c41d87c0cf22f3b88e5f80ad6bf5e81900603c1d/

Backported from: 9eec5e4470397632af035c429295a41b74bd0585

[1] https://github.com/warmcat/libwebsockets/commit/9eec5e4470397632af035c429295a41b74bd0585

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

boost: fix getchar issue with uClibc and gcc <= 4.8.2

getchar() is defined as a macro in uClibc. This hits gcc bug 58952 [1] for all
gcc versions <= 4.8.2 and building boost/test fails:

./boost/test/impl/unit_test_main.ipp: In function 'int boost::unit_test::unit_test_main(boost::unit_test::init_unit_test_func, int, char**)':
./boost/test/impl/unit_test_main.ipp:194:18: error: expected unqualified-id before '(' token

To allow building boost/test with uClibc based toolchains with gcc <= 4.8.2 use
parenthesis for std::getchar.

Fixes:
http://autobuild.buildroot.net/results/506/506b3b1b4087318a4247456782ac9e4f128d5db6/
http://autobuild.buildroot.net/results/5f2/5f2a8fb5cb823160542ee721591867b34bd25de6/
http://autobuild.buildroot.net/results/e3e/e3e6002ac5b40f7f03d32ba9da99f6497f01b42f/
.. and more

Upstream status: Pending
https://github.com/boostorg/test/pull/97

[1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58952

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/dovecot: bump version to 2.2.23

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

nss: security bump to version 3.23

Fixes:
CVE-2016-1950 - Fixed a heap-based buffer overflow related to the
parsing of certain ASN.1 structures. An attacker could create a
specially-crafted certificate which, when parsed by NSS, would cause a
crash or execution of arbitrary code with the permissions of the user.

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

gst1-libav: add license info

Now that we use unbundled ffmpeg the licensing becomes much easier.
Relevant code is under ext/*, which according to COPYING and comments is
GPLv2+, except for ext/libswscale/gstffmpegscale.c which is LGPLv2+ -
however that's not built/used when using system ffmpeg.

The code under gst-libs/ is the bundled ffmpeg itself.

Tests are under LGPLv2+ as well but we don't build them since we
disabled gst-check support in the gstreamer1 package.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

gst1-libav: switch to unbundled

It avoids the crazy in handling all of the ffmpeg options here as well,
and potentially avoids target code duplication, hence resulting in a
size reduction.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
[Thomas: extend help text as suggested by Arnout.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

cannelloni: new package

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
[Thomas: fix license, it's GPLv2, not GPLv2+.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

xorriso: Add host variant

Signed-off-by: Benoît Allard <benoit.allard@greenbone.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

whois: bump to version 5.2.12

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

nmap: bump to version 7.12

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

package/tvheadend: use wizard mode on first run

The format of the ACL database in tvheadend has changed, and generating
a default user is a little bit more involved than just dumping a file in
the correct locations: filenames are now md5sum (of something?) and the
usernames and passwords now have their own DB.

However, tvheadend has a wizard mode, where it is possible to configure
the basic features, of which creating an admin user.

We remove our canned ACL database, and change the startup script to
start in wizard mode on first run. We also switch to using our infra to
set the permissions.

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

gst1-libav: pass --cpu config option to bundled ffmpeg

Like the proper ffmpeg package does, fixes:
http://autobuild.buildroot.net/results/a46/a462dae8df2450bc0f72cbed6125e106454bde0f/

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

tcl: enable postgresql support

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

tcl: enable mysql support

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

tcl: fix sqlite usage

- Remove the bundled sqlite unconditionally, as we always want to use
  the external sqlite.
- Remove the --with-system-sqlite option as is only valid for the
  bundled sqlite configure script which we will no longer use.
- Do not remove TDBC when BR2_PACKAGE_SQLITE is not selected as it may
  be used for other TDBC drivers such as MySQL or PostgreSQL.

Fixes:
  http://autobuild.buildroot.net/results/022/02296f8624d3406a63d3a179f53862f245c56dc1/

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

tcl: switch to * instead of [0-9]

Those [0-9] expressions are needlessly complex, and they actually no
longer work now that sqlite3.11.0 is bundled internally (11 is two
digits, which was accounted for).

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

aer-inject: new package

aer-inject allows to inject PCIE AER errors on the software level into
a running Linux kernel. This is intended for validation of the PCIE
driver error recovery handler and PCIE AER core handler.

Signed-off-by: Tiago Brusamarello <tiago.brusamarello@datacom.ind.br>
[Thomas:
 - tweak commit log
 - rewrap Config.in help text
 - remove useless AER_INJECT_MAKE_OPTS variable, use
   TARGET_CONFIGURE_OPTS directly
 - add missing newline at end of .mk file.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

toolchain-external: fix getent installation with Codescape MIPS toolchains

When a toolchain is glibc based, the getent package assumes that
$(STAGING_DIR)/usr/bin contains the getent program. Unfortunately, the
Codescape MIPS toolchains do not conform with this:
$(STAGING_DIR)/usr/{bin,sbin} are empty, and instead three directories
are provided: bin-o32, bin-n32 and bin-n64 (ditto for sbin), one for
each supported MIPS ABI.

Since this is a toolchain-specific oddity, we handle it by adding a
post-install fixup hook that creates $(STAGING_DIR)/usr/{bin,sbin} as
symbolic link to the appropriate directory.

Fixes:

   http://autobuild.buildroot.org/results/9c0ee836021553319f166f9de88750535aee0a58/

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

qemu: bump version to 2.5.1

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

python-m2crypto: remove

It only provided the host variant, which was only used by crda (no
longer necessary), and wasn't available as a host selection.

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

crda: enable host-python3 build support

Add an upstream-submitted (but not accepted) patch in order to allow
crda to be built with python2 as well as python3.
This drops m2crypto usage (python2-only) in favour of pycrypto which can
be built against both major versions.

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

python-pycrypto: bump to version 2.6.1

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

python-pycrypto: add host variant

Will be used by the crda python3-enabling patch.

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

weston: handle libxml2 dependency

It's used for JUnit XML output support.

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

weston: handle systemd dependency

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

weston: handle webp dependency

It doesn't have an enable/disable switch so it's just the dependency.

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

weston: handle lcms dependecy

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

weston: handle dbus dependency

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

weston: drop --disable-xwayland-test

It's in the rpi backend block which makes no sense, and it depends on
xwayland being enabled which we currently don't support so remove it.

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

weston: add weston-launch support

The weston-launch supplemental group is for users to be able to
use/launch weston.
For a full weston(-launch) experience users should be at least members
of weston-launch, video (for framebuffer permissions) and optionally
audio.

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

weston: handle optional libva support

It's used for the recorder.

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

weston: add DRM compoitor support

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

weston: set default backend for fbdev

When it's enabled set the default weston backend to fbdev, otherwise the
drm backend is the default even though we are not building it.

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

weston: add EGL support

Also drop simple-egl-clients from the rpi compositor since it's a
duplicate of the global one.

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

powerpc-utils: bump to v1.3.1, change upstream location and license

This patch makes below changes to powerpc-utils package:
  - Update to latest upstream version (v1.3.1)
  - Update License (from CPL to GPLv2)
  - Update source link (from SF to github)
  - Disable librtas by default
  - Finally make necessary adjustment to compile the source
    (run autogen.sh before ./configure as we don't have configure in new tarball).

RTAS:
  This package contains few tools (like nvram, ppc64_cpu, etc) which are not
  dependent on RTAS support. Traditionally we always had RTAS support (at least
  on IBM Power system). But now a days we do have environments like PowerNV host
  where we do not have RTAS support. Instead we use OPAL for runtime service [1].
  Hence lets disable RTAS by default. If someone wants to build powerpc-utils
  with RTAS they can enable it.

[1] https://github.com/open-power/skiboot.git

Signed-off-by: Vasant Hegde <hegdevasant@linux.vnet.ibm.com>
[Thomas: re-add hash file, fix license, it's GPLv2+, improve commit title.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/linphone: alsa support in mediastreamer needs alsa_lib_mixer

Fixes
http://autobuild.buildroot.net/results/dd6/dd6cbff641028b25303c3b462e2058b56e53f1dd/

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

uclibc: NPTL thread implementation only available on MMU platforms

As indicated by the uClibc-ng source code, the NPTL thread
implementation is only available on MMU platforms, so we replicate
this dependency in Buildroot so that the appropriate thread
implementation is chosen by default on ARM noMMU.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>