vo-aacenc: disable arm optimizations on thumb2-only variants

Fixes:
http://autobuild.buildroot.org/results/258/258e14227cdd5a6f5cc83ab7fd176eb5b5b8445a/
http://autobuild.buildroot.org/results/634/6345b4c491fca952cd692844e3ac2932ac4836dc/

The ARM assembly optimizations in vo-aacenc requires ARM instructions
support, so disable for ARM variants without ARM instruction support (E.G.
cortex-m).

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

gcc: enable _REENTRANT for bfin when -lpthread is used

See here, why this should be always enabled:
https://lists.gnu.org/archive/html/autoconf-archive-maintainers/2016-06/msg00001.html

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

efibootmgr: fix reverse dependency from efivar

efivar minimum gcc version dependency changed from 4.7 to 4.9 in the
last efivar bump, so this must be reflected in efibootmgr Config.in as
well.

Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/snappy: bump version

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package.snappy: fix static builds with threads

For some toolchains, libstdc++ uses pthread symbols:

    .../lib/libstdc++.a(eh_alloc.o): In function `__gnu_cxx::__mutex::lock()':
    eh_alloc.cc:(.text._ZN9__gnu_cxx7__mutex4lockEv[_ZN9__gnu_cxx7__mutex4lockEv]+0x2):
    undefined reference to `pthread_mutex_lock'

and a lot of other pthread symbols...

However, when doing a static build, there is no way for the linker to know
what library to link in, so the build fails miserably if -pthread is not
specified at link time.

Fixes:
    http://autobuild.buildroot.org/results/15e/15ecaa23f0116e8120b3d62e553c838f0303df35/
    http://autobuild.buildroot.org/results/f0a/f0abe301816e39eb4ae26d3e8cd42d90901d5ac5/

[Peter: only perform workaround if toolchain has threads support, tweak comment]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/libglew: bump to version 2.0.0

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

nettle: bump to version 3.3

Drop upstream patch and autoreconf.

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

libgtk3: bump to version 3.22.1

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

neon: bump to version 0.30.2

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

package/libass: bump version to 0.13.3

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

efibootmgr: bump to version 14

Upstream has changed version numbering scheme since version 13, which
came after version 0.12.

There is now a make install target, so we can drop the manual file
installing. The make install target also installs the new efibootdump
tool.

Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

efivar: bump to version 28

Upstream has changed version numbering scheme since version 25, which
came after version 0.24.

The two patches that we had in package/efivar/ were backports and so can
be dropped with this bump.

This new version also fixes the build failure with gcc 6.x, so:
Fixes:
http://autobuild.buildroot.net/results/ea6/ea6968d5d0fd0d8ee64e045e2aa56c8e8a0b7b14

[Peter: require GCC 4.9+, adjust Config.in nvme.h comment]
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

add mksh to system shell choice

Add mksh as a choice for system shells.

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

mksh: new package

The MirOS Korn Shell is a quite complete posix shell implementation,
is rather small and supports vi mode properly.

[Peter: use mksh.1 as license file, use TARGET_CONFIGURE_OPTS, install -D]
Signed-off-by: Kurt Van Dijck <dev.kurt@vandijck-laurijssen.be>
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/ffmpeg: bump to version 3.1.4

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

orc: bump to version 0.4.26

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

package/freeswitch: bump to version 1.6.11

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

DEVELOPERS: Add entries for imx6ulpico and mx6cubox

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux: bump default to version 4.7.6

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

linux-headers: bump 4.{4, 7}.x series

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

transmission: fix compile error

After update to uClibc-ng 1.0.18 transmission fails to
compile as the special treatment for sys/quota.h isn't required
anymore, because upstream synced with GNU libc to fix a samba4
compile problem.

Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

tmux: bump version to 2.3

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

nmap: bump version to 7.30

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

package/nodejs: bump version to 6.7.0

https://nodejs.org/en/blog/release/v6.7.0/

The patches from 6.2.1 have been copied to 6.7.0 with the following
changes:

- Add 0002-inspector-don-t-build-when-ssl-support-is-disabled.patch
  to disable the new V8 inspector when openssl is not included.

Signed-off-by: Martin Bark <martin@barkynet.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/nodejs: bump version to 0.10.47

https://nodejs.org/en/blog/release/v0.10.47/

Signed-off-by: Martin Bark <martin@barkynet.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

uclibc: fix compile if locales are enabled

Reported via #buildroot, the newly activated locale
tool should only build for the target, not for the host.

Fix from upstream.

Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
Tested-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

lzop: unbreak powerpc build with GCC >= 5

Fixes:
http://autobuild.buildroot.net/results/c60/c60554f7b84d5050b86c1735fb857856ddfa4dff/
http://autobuild.buildroot.net/results/888/888ea1aea200a71d0887754e2afabdc6b260bf2f/
http://autobuild.buildroot.net/results/219/219194d269d7f3cf4c3db120e6d8079085916bd3/

GCC >= 5 changed the behaviour of 'extern inline', causing linking errors
when the powerpc inline functions are used in multiple compilation units:

https://gcc.gnu.org/gcc-5/porting_to.html

Add a patch from openembedded to change it to use static inline (which has
the correct behaviour in both old/new GCC versions) to fix it.

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

lvm2: bump version to 2.02.166

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

package/acpitool: needs threads, shared; unavailable for bfin

acpitool requires threads.

However, it does not check for threads in its configure script, so
forgets to link with -pthreads, so fails to build in static-only
scenarii.

However, upstream has been dead for the past 7+ years now, so there is
not much point trying to fix this. Besides, a system with ACPI is most
probably a biggish system, so shared libs will probably be enabled, so
we just require shared libs.

As for bfin, the cryptic configure failure:
    configure: error: C++ compiler cannot create executables

is due to the order in which AC_PROG_CXX and AC_PROG_CC are called.
Calling AC_PROG_CC first fixes this error message in this specific case.
However, this is a toolchain issue, as this does not occur for other
toolchains.

Still, inverting the call would anyway then cause linking errors:

    /home/ymorin/dev/buildroot/O/host/usr/bin/bfin-linux-g++  -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -Os    -o acpitool main.o acpitool.o ac_adapter.o toshiba.o asus.o thinkpad.o cpu.o battery.o freq.o -lgcc
    acpitool.o: In function `Has_ACPI(char*)':
    acpitool.cpp:(.text+0x1be): undefined reference to `_Unwind_Resume'
    acpitool.o: In function `Get_Kernel_Version(char*, int)':
    acpitool.cpp:(.text+0x2aa): undefined reference to `_Unwind_Resume'
    acpitool.o: In function `Set_Kernel_Version()':
    acpitool.cpp:(.text+0x43e): undefined reference to `_Unwind_Resume'
    acpitool.o: In function `Do_Fan_Info(int)':
    acpitool.cpp:(.text+0x60a): undefined reference to `_Unwind_Resume'
    acpitool.o: In function `Show_WakeUp_Devices(int)':
    acpitool.cpp:(.text+0x770): undefined reference to `_Unwind_Resume'
    acpitool.o:acpitool.cpp:(.text+0x9ca): more undefined references to `_Unwind_Resume' follow
    /home/ymorin/dev/buildroot/O/host/opt/ext-toolchain/bin/../lib/gcc/bfin-buildroot-linux-uclibc/6.1.0/../../../../bfin-buildroot-linux-uclibc/bin/ld: acpitool: hidden symbol `___udivsi3' in /home/ymorin/dev/buildroot/O/host/opt/ext-toolchain/bin/../lib/gcc/bfin-buildroot-linux-uclibc/6.1.0/libgcc.a(_udivsi3.o) is referenced by DSO
    /home/ymorin/dev/buildroot/O/host/opt/ext-toolchain/bin/../lib/gcc/bfin-buildroot-linux-uclibc/6.1.0/../../../../bfin-buildroot-linux-uclibc/bin/ld: final link failed: Bad value
    collect2: error: ld returned 1 exit status

Again, because upstream is dead, there's not much point trying to fix
this... And since there's not many chances that a bfin board has ACPI,
we just disable acpitool for bfin.

Fixes various build failures:
    http://autobuild.buildroot.org/results/6fc/6fc568228b7c5c5be9e35aae73fb09b431896325/
    http://autobuild.buildroot.org/results/f1e/f1eb711123c2d8aea8399b8984a007afec4f65fb/
    http://autobuild.buildroot.org/results/70e/70eb7d7ed93a122b4c7a9890877e721b8d777b00/
    [...]

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

newt: bump to version 0.52.19

Remove upstreamed patches and change download location.

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

uclibc: update to 1.0.18

Update to 1.0.18

This includes all patches, so remove them.

Signed-off-by: Waldemar Brodkorb <wbx@uclibc-ng.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

mpfr: bump to version 3.1.5

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

harfbuzz: bump to version 1.3.2

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

bind: bump version to 9.10.4-P3

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

mpg123: security bump to version 1.23.8

Fixes an out-of-bounds memory read in the ID3v2 parser for tags that
claim an unrealistically small length. This crashes mpg123 or any
application using libmpg123 with activated ID3v2 parsing.

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

python-pyqt5: hide comment when Qt is enabled

Since the main python-pyqt5 option is not visible when BR2_PACKAGE_QT
is enabled, the comment should also not be visibled when
BR2_PACKAGE_QT is enabled.

Reported-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

openssl: security bump to version 1.0.2j

Fixes:
CVE-2016-7052 - Missing CRL sanity check

[Peter: drop CVE 6309 from description as pointed out by Baruch]
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/ustr: really fix ldconfig

In 99067c1 (package/ustr: don't run ldconfig), we tried to fix parallel
build issues, caused by running ldconfig, by not running ldconfig.

We borked that fix, as build issues still crop up from time to time:
ldconfig is still run because our override is not accounted for by
ustr's buildsystem. Instead of overriding ldconifg at build time, we
need to override it at configure time.

Just do that.

Fixes:
    http://autobuild.buildroot.org/results/7b1/7b12f00612da4ad8dffe0c8010b95db881630b17/
    http://autobuild.buildroot.org/results/242/24209ba13f56889357024b0e460d508b7a03d1d2/
    http://autobuild.buildroot.org/results/963/963141370a9b127f473a7fae161656a6b0a4f33b/

(Those are the only three build issues in the more-than-a-month since we
initially tried to fix that.)

To be noted: upstream has accepted my patch, plus they completely
removed calls to ldconfig altogether. Add a comment so we don't forget
to remove our workaround when we bump.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

board: bump armadeus apf28's kernel version to latest longterm (4.4.22)

Signed-off-by: Julien BOIBESSOT <julien.boibessot@armadeus.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

python-u-msgpack: bump to version 2.2

Package archive includes a proper licence file now, so use it instead of
a source file.

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

configs: nitrogen*: bump kernel and u-boot revision

Here are the main U-Boot modifications:
- Add Spansion flash support
- Add secure boot (HAB) support

Here are the main Linux modifications:
- Add HDMI2CSI 75Hz support
- Add KD024FM display support
- Fix PCIe reset gpios in device tree
- Backport xhci fixes from mainline
- Fix SOM PHY address bug
- Enable IPU Prefetch engine for 6QP
- Fix ft5x06 touch bug

Signed-off-by: Gary Bisson <gary.bisson@boundarydevices.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

u-boot: Fix building for ARC700

With newer ARC toolchain obsolete -marcXXX were finally deperecated
and so compiler throws errors now about unknown option.

Solution is as simple as switching to more generic -mcu-XXX
options. Which we do. Unfortunately that change in upstream U-Boot [1]
happened right after v2016.09 was released an so we need to have that fix
for v2016.07 which is mentioned in axs10x defconfigs and for the latest
U-Boot release v2016.09 (which is selected by default if no defconfig is
used).

Once we deprecate either U-Boot version in U-Boot corresponding patch
should be removed essentially.

[1] http://git.denx.de/?p=u-boot.git;a=commit;h=7c8d81605302e7d7fdd3e7d8eb69302bddc64a2c

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
[Thomas: handle the U-Boot bump to 2016.09.01.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

DEVELOPERS: add developer for Acme Systems boards

Signed-off-by: Biagio Montaruli <biagio.hkr@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/kvm-unit-tests: bump version

Most importantly this version includes a change to kvm-unit-tests which
means it will stop looking attempting to include host headers. This
change should fix build breakage on i386.

"ERROR: unsafe header/library path used in cross-compilation: '-I'
'/lib/modules/3.13.0-042stab111.12/build/include'"

The new version also includes more powerpc tests.

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

configs/imx6dlsabreauto: Add ext4 rootfs option

Allow building a ext4 rootfs type.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

configs/imx6dlsabreauto: Use zImage by default

Currently the generated sdcard.img does not boot because U-Boot looks for
a zImage type of kernel instead of uImage.

Remove the generation of uImage, so that the sdcard.img can successfully
boot.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

configs/imx6qsabreauto: Add ext4 rootfs option

Allow building a ext4 rootfs type.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

configs/imx6qsabreauto: Use zImage by default

Currently the generated sdcard.img does not boot because U-Boot looks for
a zImage type of kernel instead of uImage.

Remove the generation of uImage, so that the sdcard.img can successfully
boot.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

gtkmm3: bump to version 3.22.0

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

libsigc: bump to version 2.10.0

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

glibmm: bump to version 2.50.0

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

libgtk3: bump to version 3.22.0

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

adwaita-icon-theme: bump to version 3.22.0

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

weston: bump to version 1.12.0

Also drop upstream patch.
Drop propietary RPI support since it's gone upstream.

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

wayland: bump to version 1.12.0

And drop upstream patch.

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

wayland-protocols: bump to version 1.7

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

configs/imx: Bump to U-Boot 2016.09.01

2016.09.01 is a bugfix release, so bump to this version.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

uboot-tools: bump to version 2016.09.01

2016.09.01 is a bugfix release, so bump to this version.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

boot/uboot: bump to version 2016.09.01

2016.09.01 is a bugfix release, so bump to this version.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

irssi: security bump to version 0.8.20

Fixes two remote crash issues since version 0.8.17

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

gtksourceview: bump to version 3.22.0

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

mpg123: bump to version 1.23.7

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

package/x11r7/xproto_xproto: bump version to 7.0.31

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

package/x11r7/xlib_libXfont: bump version to 1.5.2

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

package/x11r7/xdriver_xf86-video-nouveau: bump version to 1.0.13

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

package/x11r7/xdriver_xf86-video-ati: bump version to 7.7.1

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

package/x11r7/xdriver_xf86-video-amdgpu: bump version to 1.1.2

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

package/x11r7/xdriver_xf86-input-libinput: bump version to 0.19.1

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

toolchain-external: bump version of Linaro AArch64 toolchain

Tested with Qemu 2.6.1 and qemu_aarch64_virt_defconfig.
As reported by Gustavo Zacarias, this defconfig is known to fail with qemu
versions lower than 2.6.0.

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

toolchain-external: bump version of Linaro ARMeb toolchain

The gdb version is 7.11.1 not 7.10 as stated in the link below [1].
See the manifest to check the version really used [2].

[1] https://releases.linaro.org/components/toolchain/binaries/5.3-2016.05
[2] https://releases.linaro.org/components/toolchain/binaries/5.3-2016.05/armeb-linux-gnueabihf/gcc-linaro-5.3.1-2016.05-linux-manifest.txt

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

toolchain-external: bump version of Linaro ARM toolchain

The gdb version is 7.11.1 not 7.10 as stated in the link below [1].
See the manifest to check the version really used [2].

Tested with Qemu 2.4.1 and qemu_arm_vexpress_defconfig.

[1] https://releases.linaro.org/components/toolchain/binaries/5.3-2016.05
[2] https://releases.linaro.org/components/toolchain/binaries/5.3-2016.05/arm-linux-gnueabihf/gcc-linaro-5.3.1-2016.05-linux-manifest.txt

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

xvisor: needs gcc >= 4.9

xvisor needs gcc >= 4.9, as previous versions trigger an ICE on
dwarf2out_frame_debug_adjust_cfa() [1].

Fixes:

 - http://autobuild.buildroot.net/results/fdb/fdb5a568fb38a9d20780cd37eecb71b60b6ca96a/

[1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=60264

Signed-off-by: Eric Le Bihan <eric.le.bihan.dev@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

xvisor: add BR2_PACKAGE_XVISOR_ARCH_SUPPORTS

To prepare the addition of a Config.in comment for xvisor, and avoid the
repetition of the architecture dependencies, this commits adds a new
BR2_PACKAGE_XVISOR_ARCH_SUPPORTS hidden option.

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

samba4: bump to version 4.4.6

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

stunnel: bump to version 5.36

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

btrfs-progs: bump to version 4.7.3

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

linux: bump default to version 4.7.5

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

linux-headers: bump 4.{4, 7}.x series

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

python-simpleaudio: new package

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

package/linux-tools: don't register each tools twice

Because each tool's individual .mk files is included from the top-level
Makefile and we also include them from linux-tools.mk, they get
registered twice, and thus built twice, and thus installed twice.

We did include them from linux-tools.mk to guarantee they would be
included early and each tool had a chance to register itself before we
were to construct the build and install hooks.

However, the ordering is _currently_ guaranteed, in the C locale by the
files names, which we anyway sort using make's $(sort) function, which
always sorts in the C locale.

Beware if we are to ever rename those files in the future...

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

configs/acmesystems_arietta_g25: update kernel and at91bootstrap3

- configs/acmesystems_arietta_g25_128mb_defconfig,
  configs/acmesystems_arietta_g25_256mb_defconfig :
  update default configuration files for both 128MB and 256MB version

- board/acmesystems/arietta-g25/genimage.cfg
  board/acmesystems/arietta-g25/post-image.sh :
  improve 'post-image.sh' and 'genimage.cfg' used to build sdcard.img

- board/acmesystems/arietta-g25/readme.txt :
  update documentation for Arietta G25

Signed-off-by: Biagio Montaruli <biagio.hkr@gmail.com>
Acked-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
Reviewed-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

boot/at91bootstrap3: bump to version 3.8.6

Signed-off-by: Biagio Montaruli <biagio.hkr@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

configs/imx6qsabresd: Add ext4 rootfs option

Allow building a ext4 rootfs type.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

configs/imx6qsabresd: Use zImage by default

Currently the generated sdcard.img does not boot because U-Boot looks for
a zImage type of kernel instead of uImage.

Remove the generation of uImage, so that the sdcard.img can successfully
boot.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

rpi-userland: bump version

- rebased patches (using git) to fix patch conflicts

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

rpi-firmware: bump version

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

raspberrypi3_defconfig: bump to kernel version 4.4.21

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

raspberrypi2_defconfig: bump to kernel version 4.4.21

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

raspberrypi_defconfig: bump to kernel version 4.4.21

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

raspberrypi0_defconfig: bump to kernel version 4.4.21

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

package/openssl: security bump to version 1.0.2i

https://www.openssl.org/news/secadv/20160922.txt

Fixes
SSL_peek() hang on empty record (CVE-2016-6305)
SWEET32 Mitigation (CVE-2016-2183)
OOB write in MDC2_Update() (CVE-2016-6303)
Malformed SHA512 ticket DoS (CVE-2016-6302)
OOB write in BN_bn2dec() (CVE-2016-2182)
OOB read in TS_OBJ_print_bio() (CVE-2016-2180)
Pointer arithmetic undefined behaviour (CVE-2016-2177)
Constant time flag not preserved in DSA signing (CVE-2016-2178)
DTLS buffered message DoS (CVE-2016-2179)
DTLS replay protection DoS (CVE-2016-2181)
Certificate message OOB reads (CVE-2016-6306)
Excessive allocation of memory in tls_get_message_header()
  (CVE-2016-6307)
Excessive allocation of memory in dtls1_preprocess_fragment()
  (CVE-2016-6308)

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

package/ola: add patch to fix build with GCC 6

Fixes:
http://autobuild.buildroot.net/results/2d5/2d55b5d88a06c7b8e6baeb96973009a451e992d9/
http://autobuild.buildroot.net/results/899/89922e61c583cf1d11bd0bafdd5586c35d8f6e15/
http://autobuild.buildroot.net/results/d5b/d5b8fe66ff8d9ea91e87ef6fbe8274f5e24aa7b0/
http://autobuild.buildroot.net/results/89b/89b136e6dced6ca9842a1f23141b0cb999f783da/
.. and many more.

Building OLA with a GCC 6 cross-toolchain fails:

```
/usr/bin/arm-linux-g++ -DHAVE_CONFIG_H -I.   -D_LARGEFILE_SOURCE
-D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -I./include -I./include
-Wall -Wformat -W -isystem
/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include -pthread
-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -Os
-pthread -c -o libs/acn/e131_transmit_test.o
libs/acn/e131_transmit_test.cpp
/usr/bin/arm-linux-g++ -DHAVE_CONFIG_H -I.   -D_LARGEFILE_SOURCE
-D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -I./include -I./include
-Wall -Wformat -W -isystem
/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include -pthread
-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -Os
-pthread -c -o libs/acn/E131TestFramework.o
libs/acn/E131TestFramework.cpp
In file included from
/opt/ext-toolchain/arm-buildroot-linux-gnueabihf/include/c++/6.1.0/ext/string_conversions.h:41:0,
                 from
/opt/ext-toolchain/arm-buildroot-linux-gnueabihf/include/c++/6.1.0/bits/basic_string.h:5402,
                 from
/opt/ext-toolchain/arm-buildroot-linux-gnueabihf/include/c++/6.1.0/string:52,
                 from ./tools/ola_trigger/config.ypp:2:
/opt/ext-toolchain/arm-buildroot-linux-gnueabihf/include/c++/6.1.0/cstdlib:75:25:
fatal error: stdlib.h: No such file or directory
 #include_next <stdlib.h>
                         ^
compilation terminated.
```

The C++ library in GCC 6 now provides its own `<stdlib.h>` header that
wraps the C library header of the same name, so in `<cstdlib>` the
header include

```
#include <stdlib.h>
```

has become

```
#include_next <stdlib.h>
```

`#include_next` is sensitive to the order of directories in the
preprocessor's search path, so if that order is changed with `-isystem`
then the compiler can't find the right header:

```
[1] /usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include
[2] /opt/ext-toolchain/arm-buildroot-linux-gnueabihf/include/c++/6.1.0
[..]
End of search list.
```

`<cstdlib>` is located in [2] whereas `<stdlib.h>` (C library header) is
in [1]. In this case, the `#include_next <stdlib.h>` statement in
`<cstdlib>`, located in [2], is evaluated **after** the search path [1],
so the compiler does not find the right system header.

The problem here is that the OLA build system replaces the `-I` in the CFLAGS
from libprotobuf with `-isystem` to fix some warnings treated as errors
in the libprotobuf header files.

`-isystem` should be used to suppress warnings in system headers only
and the libprotobuf header files are not system files.

The correct fix is to compile with less restrictions and remove
`-Werror` for the build. This is already done by passing
`--disable-fatal-warnings` to OLA.

Fix the reordering of the GCC search paths by removing the replacement of
`-I` with `-isystem`.

Upstream status: https://github.com/OpenLightingProject/ola/pull/1126

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

package/ola: disable fatal warnings for host variant

By default the build system of OLA passes '-Werror' to GCC. The aborts
compilation with GCC 6:

'''
error: ‘template<class> class std::auto_ptr’ is deprecated
[-Werror=deprecated-declarations]
'''

The target variant already passes the option "-disable-fatal-warnings"
to the build system to disable '-Werror'. Do it for the host variant to
to fix build with host GCC 6.

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

package/ola: bump to version 0.10.2

Remove fetching upstream which was needed to fix musl build issue.

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

configs/acmesystems_aria_g25: update and add genimage handling

- configs/acmesystems_aria_g25_128mb_defconfig,
  configs/acmesystems_aria_g25_256mb_defconfig :
  update default configuration files for both 128MB and 256MB version

- board/acmesystems/aria-g25/genimage.cfg
  board/acmesystems/aria-g25/post-image.sh :
  add support for genimage in order to build sdcard.img

- board/acmesystems/aria-g25/readme.txt : update documentation
  for Aria G25 and add notes about how to build sdcard.img

Signed-off-by: Biagio Montaruli <biagio.hkr@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

DEVELOPERS: change developer for gst1-rtsp-server

Piotr Nakraszewicz no longer works at Imgtec, and Abhilash Tuse
volunteered to take care of this package.

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

DEVELOPERS: add entries for recently added/modified code

This commit updates the DEVELOPERS files with packages or defconfigs
recently added, plus additional entries for developers who have modified
packages they are usually taking care of.

Cc: Christian Stewart <christian@paral.in>
Cc: Cyril Bur <cyrilbur@gmail.com>
Cc: Danomi Manchego <danomimanchego123@gmail.com>
Cc: Eric Le Bihan <eric.le.bihan.dev@free.fr>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Cc: Luca Ceresoli <luca@lucaceresoli.net>
Cc: Manuel Vögele <develop@manuel-voegele.de>
Cc: Mathieu Audat <mathieu.audat@savoirfairelinux.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Peter Seiderer <ps.report@gmx.net>
Cc: Piotr Nakraszewicz <piotr.nakraszewicz@imgtec.com>
Cc: Romain Perier <romain.perier@free-electrons.com>
Cc: Theo Debrouwere <t.debrouwere@televic.com>
Cc: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

linux-tools: fix include path

The .mk files in linux-tools are named linux-tool-*.mk, not
linux-ext-*.mk.

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

docs/manual: update the linux tools section

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Romain Naour <romain.naour@openwide.fr>
[Thomas:
 - Adjust the Config.in example to show that we now need to "select
   BR2_PACKAGE_LINUX_TOOLS"
 - Adjust the .mk file example to use $(LINUX_DIR) instead of $(@D)]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

linux/tools: make it a real, separate package

The kernel source tree also contains the sources for various userland
tools, of which cpupower, perf or selftests.

Currently, we have support for building those tools as part of the
kernel build procedure. This looked the correct thing to do so far,
because, well, they *are* part of the kernel source tree and some
really have to be the same version as the kernel that will run.

However, this is causing quite a non-trivial-to-break circular
dependency in some configurations. For example, this defconfig fails to
build (similar to the one reported by Paul):

    BR2_arm=y
    BR2_cortex_a7=y
    BR2_ARM_FPU_NEON_VFPV4=y
    BR2_TOOLCHAIN_EXTERNAL=y
    BR2_INIT_SYSTEMD=y
    BR2_LINUX_KERNEL=y
    BR2_LINUX_KERNEL_CUSTOM_GIT=y
    BR2_LINUX_KERNEL_CUSTOM_REPO_URL="https://github.com/raspberrypi/linux.git"
    BR2_LINUX_KERNEL_CUSTOM_REPO_VERSION="26f3b72a9c049be10e6af196252283e1f6ab9d1f"
    BR2_LINUX_KERNEL_DEFCONFIG="bcm2709"
    BR2_PACKAGE_LINUX_TOOLS_CPUPOWER=y
    BR2_PACKAGE_CRYPTODEV=y
    BR2_PACKAGE_OPENSSL=y
    BR2_PACKAGE_LIBCURL=y

This causes a circular dependency, as explained by Thomas:

 - When libcurl is enabled, systemd depends on it

 - When OpenSSL is enabled, obviously, will use it for SSL support

 - When cryptodev-linux is enabled, OpenSSL will depend on it to use
   crypto accelerators supported in the kernel via cryptodev-linux.

 - cryptodev-linux being a kernel module, it depends on linux

 - linux by itself (the kernel) does not depend on pciutils, but the
   linux tool "cpupower" (managed in linux-tool-cpupower) depends on
   pciutils

 - pciutils depends on udev when available

 - udev is provided by systemd.

And indeed, during the build, we can see that make warns (it's only
reported as a *warning*, not as an actual error):

    [...]
    make[1]: Circular /home/ymorin/dev/buildroot/O/build/openssl-1.0.2h/.stamp_configured
    <- cryptodev-linux dependency dropped.
    >>> openssl 1.0.2h Downloading
    [...]

So the build fails later on, when openssl is actually built:

    eng_cryptodev.c:57:31: fatal error: crypto/cryptodev.h: No such file or directory
    compilation terminated.
    <builtin>: recipe for target 'eng_cryptodev.o' failed

Furthermore, graph-depends also detects the circular dependency, but
treats it as a hard-error:

    Recursion detected for  : cryptodev-linux
    which is a dependency of: openssl
    which is a dependency of: libcurl
    which is a dependency of: systemd
    which is a dependency of: udev
    which is a dependency of: pciutils
    which is a dependency of: linux
    which is a dependency of: cryptodev-linux
    Makefile:738: recipe for target 'graph-depends' failed

Of course, there is no way to break the loop without losing
functionality in either one of the involved packages *and* keep
our infrastructure and packages as-is.

The only solution is to break the loop at the linux-tools level, by
moving them away into their own package, so that the linux package will
no longer have the opportunity to depend on another package via a
dependency of one the tools.

All three linux tools are thus moved away to their own package.

The package infrastructure only knows of three types of packages: those
in package/ , in boot/ , in toolchain/ and the one in linux/ . So we
create that new linux-tools package in package/ so that we don't have to
fiddle with yet another special case in the infra. Still, we want its
configure options to appear in the kernel's sub-menu.

So, we make it a prompt-less package, with only the tools visible as
options of that package, but without the usual dependency on their
master symbol; they only depend on the Linux kernel.

Furthermore, because the kernel is such a huge pile of code, we would
not be very happy to extract it a second time just for the sake of a few
tools. We can't extract only the tools/ sub-directory from the kernel
source either, because some tools have hard-coded path to includes from
the kernel (arch and stuff).

Instead, we just use the linux source tree as our own build tree, and
ensure the linux tree is extracted and patched before linux-tools is
configured and built.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Paul Ashford <paul.ashford@zurria.co.uk>
[Thomas:
 - fix typo #(@D) -> $(@D)
 - fix the inclusion of the per-tool .mk files.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>