package/x11r7/xapp_sessreg: fix musl build

Fixes
http://autobuild.buildroot.net/results/717/717bd4a2ee4ba62ed9245ec22166a3c63ce1a56d/

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

linux-headers: bump 3.12.x series

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

DEVELOPERS: add Yegor Yefremov for the can-utils package

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

lxc: enable lua if package is selected

lua and python bindings were removed with the same patch:
https://patchwork.ozlabs.org/patch/367361. There is a real issue with
python but not with lua so enable it if lua is selected

Signed-off-by: Fabrice Fontaine <fabrice.fontaine@orange.com>
[Thomas: simplify value passed to --with-lua-pc.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

lxc: bump to version 2.0.6

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

xserver_xorg-server: fix dependency with dbus

If dbus and udev are enabled. Xorg link with both:

$ arm-linux-readelf -d usr/bin/X | grep NEEDED
 0x00000001 (NEEDED)                     Shared library: [libdbus-1.so.3]
 0x00000001 (NEEDED)                     Shared library: [libpthread.so.0]
 0x00000001 (NEEDED)                     Shared library: [libudev.so.1]
[...]

Signed-off-by: Jérôme Pouiller <jezz@sysmic.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/google-breakpad: bump to the latest version

This version add the C++11 support and microdump.

Add the C++11 dependency since it's now mendatory.
Use the same linux-syscall-support version as the one defined by DEPS
file in the Google-breakpad sources.

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Pascal Huerst <pascal.huerst@gmail.com>
Cc: Frank Hunleth <fhunleth@troodon-software.com>
[Thomas: propagate the C++11 related dependency to
BR2_GOOGLE_BREAKPAD_ENABLE.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/google-breakpad: switch to the new git repository

Google-breakpad moved to a new git repository last year and the old svn
repository has been deleted. So, the Buildroot archive mirror is used.

svn: E160013: Unable to connect to a repository at URL 'http://google-breakpad.googlecode.com/svn/trunk'
svn: E160013: '/svn/trunk' path not found
--2016-11-27 10:42:34--  http://sources.buildroot.net/google-breakpad-1373.tar.gz

Swith to the git repository using the git hash corresponding to the
1373 svn revision. But there is an issue with the third parties...
Some of them are not bundled with Google-breakpad or as git submodule.
Even worst, the google-breakpad source code use #include
"third_party/lss" directly, so we can't provide missing third parties
with an external package.

That's why the linux-syscall-support package was added to Buildroot,
it provide the missing linux_syscall_support.h file which is must be
copied to src/third_party/lss/ in the Google-breakpad sources.

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Pascal Huerst <pascal.huerst@gmail.com>
Cc: Frank Hunleth <fhunleth@troodon-software.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

DEVELOPERS: add Romain Naour for linux-syscall-support

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

package/linux-syscall-support: new package

This package provide a header file which will be used by
Google-breakpad package. This avoid using depot-tools script to fetch
all third parties used by Google-breakpad.

Use the version corresponding to the Google-breakpad svn revision 1373:
http://linux-syscall-support.googlecode.com/svn/trunk/lss@24 829466d3-f3f5-3ae4-62ad-de35cf9bba21

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Pascal Huerst <pascal.huerst@gmail.com>
Cc: Frank Hunleth <fhunleth@troodon-software.com>
[Thomas: directly implement installation in the install commands, rather
than using post-install hooks.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

lvm2: ensured TARGET_AR is used during build

The same way TARGET_CC was not being used during the build, TARGET_AR
was not used either. Now TARGET_CONFIGURE_OPTS is injected into
LVM2_MAKE_ENV which should include all the toolchain substitutions.

Signed-off-by: Jordan Yelloz <jordan@yelloz.me>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

pkg-generic: run $(PKG)_PRE_RSYNC_HOOKS before checking for the existence of $(SRCDIR)

This will allow packages to define their pre-rsync hooks which will be
guaranteed to run even if the source is missing.

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

Revert "package/aespipe: fix host compile"

This reverts commit cb33a7d6bd89306635749b497cd004e986bcd7e5, as it
causes build failures on platforms where gcc does not understand the
"-no-pie" option.

Fixes:

  http://autobuild.buildroot.net/results/51be9526300b3d03e2e304dec9cae953f2f59de4/

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

linux: check that a repo and version is defined when using a custom repo

If not set the system will use an empty string which will result in
download errors for 'linux-.tar.gz' packages.

This patch makes it obvious to the user that the variable needs to be
set.

Signed-off-by: Christian Kellermann <christian.kellermann@solectrix.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

rapidjson: bump to 1.1.0

Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

musl: workaround kernel if_ether.h header conflict

Add a patch that defines _LINUX_IF_ETHER_H in netinet/if_ether.h to avoid the
conflicting kernel if_ether.h.

Other musl distros (Alpine, Sabotage, Gentoo) patch[1] the kernel headers
instead. This is not practical in Buildroot, since we allow the user to choose
the kernel version.

This only fixes the normal case of including linux/if_ether.h (directly or
indirectly) after netinet/if_ether.h.

Fixes (tcpreplay):
http://autobuild.buildroot.net/results/a65/a65b27f9e288f94f041ced64313dbf55af47bf36/

[1] https://cgit.gentoo.org/proj/musl.git/diff/sys-kernel/linux-headers/files/libc-compat.h-prevent-redefinition-of-struct-ethhdr.patch?id=8eddda8072add075ebf56cf6d288bc1450d6b5f8

Cc: Rich Felker <dalias@aerifal.cx>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

toolchain: extend musl workaround for recent kernels

Kernels v4.7 and newer[1] require the __USE_MISC macro for libc-compat.h
suppressions to be effective. musl does not define this macros, so add an
unconditional definition of __USE_MISC to libc-compat.h.

In addition, since musl define IFF_LOWER_UP and friends as macros (instead of
enums), define __UAPI_DEF_IF_NET_DEVICE_FLAGS_LOWER_UP_DORMANT_ECHO to
suppress the kernel corresponding definitions to avoid conflict.

Fixes (xl2tp):
http://autobuild.buildroot.net/results/0b9/0b9384c4beaac85f3a17c3245c8a7418c2e2e296/
http://autobuild.buildroot.net/results/657/657bc5687cf01aee38d32d45ba57fb62b2bd56d0/
http://autobuild.buildroot.net/results/bcd/bcd81618de2f745a19c923c66d787b5182bb54aa/

[1] Since commit f0a3fdca79 (uapi glibc compat: fix compilation when
    !__USE_MISC in glibc)

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

python-incremental: new package

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

can-utils: bump version

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

python-sh: new package

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

python-pytablewriter: bump to version 0.14.1

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

python-dominate: bump to version 2.3.1

Use PyPI as download location.

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

python-dataproperty: bump to version 0.13.5

Add new dependency: python-mbstrdecoder.

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

python-dateutil: bump to version 2.6.0

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

python-pytz: bump to version 2016.10

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

python-mbstrdecoder: new package

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

mysql: rename package to oracle-mysql, make a virtual package

This commit turns mysql into a virtual package, after renaming the
original mysql package to oracle-mysql. This way, all existing packages
that "depends on" or "select" BR2_PACKAGE_MYSQL continue to work with no
modification.

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

DEVELOPERS: add entry for mariadb

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

mariadb: new package

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

toolchain-external: remove musl-cross prebuilt toolchain

The upstream link to download musl-cross prebuilt toolchain is dead [1] and
there no new download location. Also the last prebuilt toolchain use musl
1.1.12 version which is not uptodate (currently 1.1.15).

Remove this support and recommend to use Buildroot toolchain instead.

[1] https://googledrive.com/host/0BwnS5DMB0YQ6bDhPZkpOYVFhbk0

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>

tinycbor: disable parallel build

Maybe fixes ([1]):

 make[1]: Entering directory `/home/peko/autobuild/instance-1/output/build/tinycbor-v0.4'
 .config:8: *** missing separator.  Stop.

[1] http://autobuild.buildroot.net/results/8e8fd1e19d6d07212f29e19a305712e715fe781b

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

systemd: enable selinux if libselinux is enabled

systemd currently has selinux disabled unconditionally.  This prevents
/sys/fs/selinux from being mounted automatically at boot which causes
audit to no longer function. This patch fixes the problem by checking if
libselinux is enabled and enabling support for selinux in systemd if so.

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

package/x11r7/xserver_xorg-server: Explicitly disable glamor support when dependencies are not met

Fixes
http://autobuild.buildroot.net/results/b48/b48074089da770463f7ffcc27e385ac59456edc5/

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

package/aespipe: fix host compile

Building host-aespipe fails on

$ cat /etc/debian_version
stretch/sid

at linking stage:

/usr/bin/gcc -L/home/buildroot/br6/output/host/lib -L/home/buildroot/br6/output/host/usr/lib -Wl,-rpath,/home/buildroot/br6/output/host/usr/lib -o aespipe aespipe.o aes-amd64.o md5-amd64.o md5-2x-amd64.o aes-intel64.o sha512.o rmd160.o
/usr/bin/ld: aes-amd64.o: relocation R_X86_64_32S against `.rodata' can not be used when making a shared object; recompile with -fPIC

Fix is also used in Debian:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=837393

[Peter: add comment explaining why]
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/aespipe: bump version to 2.4d

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

linux-headers: bump 4.{4, 8}.x series

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

gvfs: bump to version 1.30.3

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

cjson: fix error with sh4a and powerpc

Fixes:
 - http://autobuild.buildroot.net/results/3d899790acdc5c21733ff6f7f5a1b500e862ea0a

Signed-off-by: Fabrice Fontaine <fabrice.fontaine@orange.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

configs/orangepi: bump kernel and u-boot versions

Update kernel to 4.9 and U-Boot to 2016.09.01.

Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

board/orangepi: drop redundant files

The orangepi board variants use identical post-build and post-image scripts,
so move them up to board/orangepi and add symlinks in the variant
subdirectories, similary to how we handle it for the raspberry pi variants.

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

orangepi: add basic support for orangepi-one board

Board support package includes the following components:
- mainline u-boot 2016.09.01
- mainline kernel 4.9

Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

orangepipc: rename to orangepi-pc

Rename orangepipc files, configs, and directories:
- oragepipc_defconfig -> orangepi_pc_defconfig
- board/orangepi/orangepipc -> board/orangepi/orangepi-pc

The purpose of the patch is to improve readability before
adding other boards of orangepi family.

Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

docker-engine: security bump to 1.12.3

Fixes CVE-2016-8867: https://forums.docker.com/t/docker-1-12-2-security-advisory/23975

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

docker-engine: needs pkg-config

Fixes:
http://autobuild.buildroot.org/results/fcc/fcc32f6cdc46f354a016a9b137d4f51d4054cdb6/
http://autobuild.buildroot.org/results/083/0832b7be86c525522e2caed66246d5aeb365afb2/
http://autobuild.buildroot.org/results/ede/ede0ddf0c5685ece8443a9b1bbd5c63b4d98af33/

pkg-config is used for some of the docker-engine dependencies (at least
libseccomp and libsystemd), so make sure it is available.

The make.sh script used to configure docker-engine uses the PKG_CONFIG
variable, but the libseccomp-golang go module simply uses whatever
pkg-config binary it finds in the path at build time, so also pass
TARGET_MAKE_ENV to ensure our cross-pkg-config is used.

For consistency and robustness against future changes, use both approaches
for the configure and build steps.

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

package/gdb: remove version 7.9

Now that 7.12 has been added and 7.11 is the default version, let's
remove the old 7.9 release.

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/gdb: switch to 7.11 as the default version

7.12 is around, 7.11 has already seen a point release, so it's time to
make 7.11 the default version for gdb.

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/gdb: add support for gdb 7.12

gdb 7.12 was released on October 2016, it's time to include it in
Buildroot, which this commit does.

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/mpd: bump version to 0.19.21

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

wireshark: bump to version 2.2.3

The only security bug fixed concerns windows so it's not a security bump
for buildroot.

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

tinyalsa: add license files

Signed-off-by: Rahul Bedarkar <rahul.bedarkar@imgtec.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

tinyalsa: bump version to 1.1.0

Signed-off-by: Rahul Bedarkar <rahul.bedarkar@imgtec.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/wavpack: bump version to 5.0.0

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

tcpreplay: bump to version 4.1.2

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

fs: set packages permissions even with no system device tables

Currently, when there is no syztem device table (permissions or static
devices) defined, then package permissions are not applied, because they
are guarded by the check on the system device tables being non empty.

Fix that by narrowing the guarding condition.

Note that the dependency on host-makedevs was not conditional; we always
build it even if we don't need it. Making it conditional is not
possible, because we don't know all the packages permissions by the time
the fs infra is parsed (packages from br2-external are parsed after it).

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

configs/linksprite_pcduino: bump u-boot version

Update U-Boot to 2016.11.

Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

configs/mx6: Select NEON and VFP related options

MX6 has a Cortex-A9 core, which supports NEON and VFP.

Add support for them.

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

mx6: udoo: Create a mx6qdl directory for consistency

The udoo neo files are currently placed inside udoo/neo directory.

Place the mx6qdl udoo related files inside udoo/mx6dl for consistency.

[Peter also update path in readme.txt]
Suggested-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

fs/ext2: default to ext2 rev1

Rev0 is very old (E.G. from before Linux was maintained in git), the kernel
prints a scary warning when used:

EXT4-fs warning (device sda): ext4_update_dynamic_rev:746: updating to rev 1
because of new feature flag, running e2fsck is recommended

And rev0 support is broken in u-boot 2016.11:
http://lists.denx.de/pipermail/u-boot/2016-December/275916.html

So default to rev1 instead.

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

configs/linksprite_pcduino: bump kernel and u-boot versions

Update kernel to 4.9 and U-Boot to 2016.09.01.

Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

configs/mx6sx_udoo_neo: Remove BR2_TARGET_GENERIC_GETTY_PORT

As console=ttymxc0 is passed in U-Boot we can use /dev/console to
access the serial console and then BR2_TARGET_GENERIC_GETTY_PORT
can be removed.

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

netsnmp: handle libnl dependency properly

libnl3 include path was hardcoded in the configure script,
causing the following gcc warning:

WARNING: unsafe header/library path used in cross-compilation:
'/usr/include/libnl3'

Let's apply an upstream patch that addresses the issue: now
pkg-config is used to detect libnl include path.

Also:
- add optional libnl build dependency;
- enable libnl support only when libnl package is selected.

Signed-off-by: Julien Floret <julien.floret@6wind.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/ipmitool: bump version to 1.8.18

Some patches are not necessary anymore as upstream fixed the issues.

As we are not patching configure.ac anymore, there is no need to
autoreconf the package.

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/lttng-babeltrace: enable debug info support for the host variant

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Philippe Proulx <eeppeliteloop@gmail.com>
[Thomas: fix typo, adjust on top of master.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/elfutils: add host variant

elfutils (libelf and libdw) can be used by host-lttng-babeltrace.
Enable all library dependencies, but don't build the programs.

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

mx6sx: Add support for mx6sx udoo neo boards

Add support for mx6sx udoo neo boards. For more information about these
boards, please check:
http://www.udoo.org/udoo-neo

The U-Boot and kernel dts patches will be removed when we bump U-Boot
to 2017.01 and kernel to 4.10 version as they have already been sent
upstream.

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

configs/imx6ulpico: Bump to kernel 4.9

Bump to kernel 4.9 and remove the wifi dts patch as it is already
part of 4.9.

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

gdk-pixbuf: security bump to version 2.36.1

Fixes an integer buffer overflow in the jpeg loader (details still
embargoed).

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

python-pyicu: bump to version 1.9.5

Fixes:
http://autobuild.buildroot.net/results/48c/48cfb0fe00fdd52af502f9570d291125fb57b9f0/

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

zlib: use $(HOST_MAKE_ENV) when calling $(MAKE1)

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

zlib: use $(TARGET_MAKE_ENV) when calling $(MAKE1)

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

package/mpd: bump version to 0.19.20

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

policycoreutils: restorecon libglib2 dependency

Updated to match Config.in select of libglib2
package as dependency if restorecon is enabled.

Signed-off-by: Matthew Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

policycoreutils: CPPFLAGS undefine _FILE_OFFSET_BITS

Undefining _FILE_OFFSET_BITS here because of a "bug" with glibc fts.h
large file support.

Similar action was taken previously for CFLAGS.

Resolves:
http://autobuild.buildroot.net/results/4cb/4cb70d776bee01f1284e0c0be5b3fb5f1011353a/
http://autobuild.buildroot.net/results/f43/f438149e7d83475d425e1c1f17550f3cf5f69340/

Signed-off-by: Matthew Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

php: add apache support

Continue work started by Bernd Kuhls in
https://patchwork.ozlabs.org/patch/437544/

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

php: fix build for AArch64

Add a patch from a PHP Github pull request that fixes the build at -O0
on AArch64.

Signed-off-by: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
[Thomas: improved patch commit log.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/enlightenment: bump to version 0.21.5

Bug fix release:
https://www.enlightenment.org/news/e-0.21.5-release

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

icu: fix build failure for musl

musl doesn't have xlocale.h, and stdtod_l is crippled, so disable it.
See thread from OE:
http://lists.openembedded.org/pipermail/openembedded-core/2016-November/128527.html
Fixes:
http://autobuild.buildroot.net/results/6ca/6caebf0f6c4adf66e67bedd0f04155413a9aa7bd/

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

usb_modeswitch: bump to version 2.4.0

Signed-off-by: Rahul Bedarkar <rahul.bedarkar@imgtec.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

nano: bump to version 2.7.2

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

zsh: bump to version 5.3

And drop upstream patch.

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

icu: bump to version 58.2

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

libgtk3: bump to version 3.22.5

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

olimex_a20_olinuxino_lime: bump kernel to 4.9

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

package/dovecot: needs OpenSSL

The latest version bump to 2.27 introduced a bug in the configure
script which occurs when OpenSSL support is missing:
http://lists.busybox.net/pipermail/buildroot/2016-December/179397.html

This patch makes OpenSSL mandatory following the upstream advice:
http://www.dovecot.org/list/dovecot/2016-December/106346.html
"Nobody really should be building without OpenSSL nowadays anyway"

Fixes
http://autobuild.buildroot.net/results/85f/85f2f176c108ab36520f02d975f27c27cddce84b/

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

package/eudev: bump version to 3.2.1

Remove patch 0002 applied upstream:
https://github.com/gentoo/eudev/commit/4f3b8298fb55a7a591831cf71067bc160cd60599

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

e2fsprogs: fix missing OPTS

With the commit 64544178f180f03cedd8fe403d3718406d909c88 many packages
broke because they could not find the ext2fs.pc file.

This is perfectly normal, as the DESTDIR and the install target were not
passed anymore to make.

[Peter: remove redundant E2FSPROGS_INSTALL_STAGING_OPTS]
Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

toolchain: add 4.9.x choice for headers

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

linux-headers: bump 4.{8, 4}.x series

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

linux: bump default to version 4.9

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

linux-headers: bump to 4.9 kernel version

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

package/jpeg: better indentation

Use easier-to-read layout; fix leading spaces.

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

jpeg: fix typo introduced when fixing the indentation

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

jpeg-turbo: add simd support for aarch64

>From the 1.5-b1 release notes:

Added ARM 64-bit (ARMv8) NEON SIMD implementations of the commonly-used
compression algorithms (including the slow integer forward DCT and h2v2 &
h2v1 downsampling algorithms, which are not accelerated in the 32-bit NEON
implementation.) This speeds up the compression of full-color JPEGs by about
75% on average on a Cavium ThunderX processor and by about 2-2.5x on average
on Cortex-A53 and Cortex-A57 cores.

Add it unconditionally for all aarch64 cores, as neon support is required
for all "standard" ARMv8 implementations.  If an ARMv8 implementation w/o
NEON ever shows up, then we will need to add a BR2_AARCH64_CPU_HAS_NEON and
handle it like ARM.

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

jpeg-turbo: add simd support for powerpc cores with altivec

>From the 1.5-b1 release notes:

Added full SIMD acceleration for PowerPC platforms using AltiVec VMX
(128-bit SIMD) instructions.  Although the performance of libjpeg-turbo on
PowerPC was already good, due to the increased number of registers available
to the compiler vs.  x86, it was still possible to speed up compression by
about 3-4x and decompression by about 2-2.5x (relative to libjpeg v6b)
through the use of AltiVec instructions.

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

jpeg: rework jpeg-turbo simd support handling

In preperation for adding more architectures to the jpeg-turbo simd
handling, introduce a hidden BR2_PACKAGE_JPEG_SIMDS_SUPPORT symbol and use
that to enable simd support and default to jpeg-turbo, instead of open
coding it in both places.

While we are at it, reword the help text to be more useful.

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

jpeg-turbo: now checks for pkg-config

The jpeg-turbo configure script now contains PKG_PROG_PKG_CONFIG, even
though it doesn't so far use pkg-config to detect any dependencies (it does
install a .pc file now though):

checking pkg-config is at least version 0.9.0... ./configure: line 13540: \
  /home/peko/source/buildroot/output/host/usr/bin/pkg-config: No such file or directory

Depend on host-pkgconf for consistency.

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

acl: backport upstream patch to fix static build

When building tar package with acl in static build configuration, we
get definition clash error for 'quote' function.

../gnu/libgnu.a(quotearg.o): In function 'quote':
quotearg.c:(.text+0x12fc): multiple definition of 'quote'
../sysroot/usr/lib/libacl.a(quote.o):/home/test/autobuild/run/instance-1/output/build/acl-2.2.52/libmisc/quote.c:27: first defined here

In acl upstream, this is already fixed by renaming quote and unquote
internal functions. This commit backports upstream patch.

Fixes:
  http://autobuild.buildroot.net/results/d64/d64d13745c980d322d3b80c3b324d03eb7f17262

Signed-off-by: Rahul Bedarkar <rahul.bedarkar@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

doc/manual: document the new waf-package infra variables

Document LIBFOO_WAF_OPTS, LIBFOO_BUILD_OPTS, LIBFOO_INSTALL_STAGING_OPTS
and LIBFOO_INSTALL_TARGET_OPTS.

Also document LIBFOO_CONF_OPTS which was missing in waf-package
reference.

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

e2fsprogs: pass LDCONFIG=true also at install time

We correctly set LDCONFIG=true for MAKE_OPTS, but not for the other
steps which don't inherit it, and therefore e2fsprogs tries to use the
host ldconfig during staging and target installation, which causes some
weird error messages at build time (even if they don't abort the build).

Fix that by passing LDCONFIG=true at install time.

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

toolchain/wrapper: use the {str, len} tuples for the unsafe paths

In 61cb120 (toolchain/wrapper: extend paranoid check to -isystem), we
introduced a {str,len} tuple to check the various arguments pased to
gcc, to avoid hard-coding an ever-growing, long list of those args
directly in the condition check.

Now, we're left with a long list of unsafe paths, somehow hidden within
the code, which can use the same mechanism we use for arguments.

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>