gdb: add patches to fix musl build on ppc(64)(le) and sh

This commit small patches to gdb that fix redefined structures build
problems that occur when building on PowerPC, PowerPC64, PowerPC64le and
SuperH with the musl C library.

The PowerPC(64)(le) patch was taken from the Yocto Project, the SuperH
patch was inspired from the PowerPC patch. The issue has also been
reported upstream to the musl developers:

  http://www.openwall.com/lists/musl/2017/06/03/1

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
Changes since v2:
 - None
Changes since v1:
 - None

gdb: not available on ARM noMMU and OpenRISC

Neither gdb nor gdbserver have support for ARM noMMU and OpenRISC, so
let's disable gdb on those architectures.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
Changes since v2:
 - Fixed BR2_openrisc to BR2_or1k in Config.in.host
Changes since v1:
 - Fixed BR2_openrisc to BR2_or1k

gdb: introduce BR2_PACKAGE_GDB_ARCH_SUPPORTS

Before adding more architecture dependencies to gdb, let's introduce a
BR2_PACKAGE_GDB_ARCH_SUPPORTS, to avoid duplicating this list of
dependencies between the Config.in comment and the main Config.in
option.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
Changes since v2:
 - None
Changes since v1:
 - None

package/lcdproc: override libpng-config path

Like for freetype-config, override libpng-config path otherwise
the one from the host system may be used when installed.
Make sure that the one from STAGING_DIR is used.

configure:7480: checking for libpng-config
configure:7498: found /usr/bin/libpng-config
configure:7510: result: /usr/bin/libpng-config
configure:7529: checking whether libpng is present and sane
[...]
nios2-linux-gnu-gcc: ERROR: unsafe header/library path used in cross-compilation: '-I/usr/include/libpng12'

Fixes
http://autobuild.buildroot.net/results/b68/b68f54404a7d412d639ebb6d052412eea6ee3bf1/lcdproc-0.5.7/config.log

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

package/lcdproc: override freetype-config path

freetype-config from the host system may be used when installed.
Make sure that the one from STAGING_DIR is used.

checking if freetype support has been enabled... yes
checking for freetype-config... /usr/bin/freetype-config
checking for FreeType -- version >= 7.0.1... cross compiling; assuming OK... yes

Fixes:
http://autobuild.buildroot.net/results/c7c/c7c52cd56b985fd73434d5997e9a6ab3f7d7e71e

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

linuxptp: new package

Add the Linux PTP Project package.
http://linuxptp.sourceforge.net/

The SysV and systemd init scripts start the daemon in slave-only mode on eth0
and synchronize the system clock to PTP.

Signed-off-by: Petr Kulhavy <brain@jikos.cz>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/qt: enable static build for BR2_PACKAGE_QT_MOUSE_TSLIB

The dependency on !BR2_STATIC_LIBS was here because of tslib not
supporting static builds. But since

   commit 05fadfea1 ("package/tslib: enable statically linked building")

it does and we can remove this dependecy.

Signed-off-by: Martin Kepplinger <martin.kepplinger@ginzinger.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/directfb: enable static build for BR2_PACKAGE_DIRECTFB_TSLIB

The dependency on !BR2_STATIC_LIBS was here because of tslib not
supporting static builds. But since

   commit 05fadfea1 ("package/tslib: enable statically linked building")

it does and we can remove this dependecy.

Signed-off-by: Martin Kepplinger <martin.kepplinger@ginzinger.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/qt5base: enable static build for BR2_PACKAGE_QT5BASE_TSLIB

The dependency on !BR2_STATIC_LIBS was here because of tslib not
supporting static builds. But since

   commit 05fadfea1 ("package/tslib: enable statically linked building")

it does and we can remove this dependecy.

Signed-off-by: Martin Kepplinger <martin.kepplinger@ginzinger.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

intltool: add patch to fix compatibility with Perl 5.26

When Perl 5.26 is installed on the host, building some packages like
avahi fail, because of intltool-update:

```
Unescaped left brace in regex is illegal here in regex; marked by <-- HERE in m/^(.*)\${ <-- HERE ?([A-Z_]+)}?(.*)$/ at $BUILDROOT/host/usr/bin/intltool-update line 1065.
```

Apparently, this has been a warning before Perl 5.26, but now it is an
error.

Fetch from:
https://github.com/Alexpux/MSYS2-packages/blob/master/intltool/perl-5.22-compatibility.patch

Reported upstream:
https://bugs.launchpad.net/intltool/+bug/1696658

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

syslinux: add patches to build with the latest gnu-efi release

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

syslinux: add missing dependency on host-util-linux

If util-linux is not installed system-wide on the host, the build
fails with:

/usr/bin/gcc -Wp,-MT,isohybrid.o,-MMD,./.isohybrid.o.d -O2 -I/home/thomas/projets/buildroot/output/host/usr/include -W -Wall -Wstrict-prototypes  -Os -fomit-frame-pointer -D_FILE_OFFSET_BITS=64 -I/home/thomas/projets/buildroot/output/build/syslinux-6.03/utils -c -o isohybrid.o /home/thomas/projets/buildroot/output/build/syslinux-6.03/utils/isohybrid.c
/home/thomas/projets/buildroot/output/build/syslinux-6.03/utils/isohybrid.c:40:23: fatal error: uuid/uuid.h: No such file or directory
 #include <uuid/uuid.h>
                       ^
compilation terminated.

Therefore, this commit adds a dependency on host-util-linux, which
will ensure that libuuid is available. The resulting isohybrid tool is
really installed, and linked with libuuid:

$ readelf -d output/host/usr/bin/isohybrid

Dynamic section at offset 0x3e00 contains 26 entries:
  Tag        Type                         Name/Value
 0x0000000000000001 (NEEDED)             Shared library: [libuuid.so.1]
 0x0000000000000001 (NEEDED)             Shared library: [libc.so.6]
 0x000000000000000f (RPATH)              Library rpath: [/home/thomas/projets/buildroot/output/host/usr/lib]

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

package/python-ipython-genutils: New package

Add 'ipython_genutils'[1] package to Buildroot. Needed by 'traitlets'.

[1] https://pypi.python.org/pypi/ipython_genutils

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Reviewed-by: Yegor Yefremov <yegorslists@googlemail.com>
[Thomas: take into account comments made by Yegor:
 - fix alphabetic ordering in the DEVELOPERS file
 - fix comment in hash file
 - add license file]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/python-simplegeneric: Switch setup type to 'setuptools'

Specifying setup type to be 'distutils' results in this package being
installed as .egg, a zipped file, which inadvertantly makes
'simplegeneric' to be dependent on BR2_PACKAGE_PYTHON*_ZLIB. In order
to avoid that specify setup type to be 'setuptools' which makes use of
--single-version-externally-managed and therefore installs the package
as .pyc and a .egg-info folder.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Reviewed-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

DEVELOPERS: Add missing closing angle bracket to my e-mail

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Reviewed-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/trace-cmd: bump version to 2.6.1, add hash

Removed patch applied upstream.

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

package/mcelog: bump version to 152, add hash

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

package/rpi-bt-firmware: add hash

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

package/netsurf-buildsystem: add hash

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

package/libsvgtiny: add hash

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

package/kvm-unit-tests: add hash

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

package/opkg: add hash

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

package/dbus-triggerd: add hash

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

package/qt-webkit-kiosk: add hash

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

package/sunxi-mali: add hash

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

package/kvmtool: add hash

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

package/am33x-cm3: add hash

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

package/uemacs: add hash

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

package/psplash: add hash

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

package/mmc-utils: add hash

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

package/ltrace: add hash

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

package/libyuv: add hash

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

package/libuci: add hash

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

package/libsoundtouch: add hash

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

package/libsilk: add hash

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

package/libilbc: add hash

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

package/libg7221: add hash

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

package/libcodec2: add hash

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

package/libbroadvoice: add hash

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

package/i2c-tools: add hash

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

package/flashbench: add hash

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

package/f2fs-tools: add hash

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

package/edid-decode: add hash

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

package/dtv-scan-tables: add hash

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

package/aer-inject: add hash

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

pkg-download: enable hash check for git downloads

For a while already, our git downloaded tarballs are reproducible, so
we can actually check the hash for them.

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

arm-trusted-firmware: exclude from hash when downloading from git

Since we will enable hash checks for git downloads soon, the hash check
for the custom git download should be disabled.

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

mxs-bootlets: exclude from hash when downloading from git

Since we will enable hash checks for git downloads soon, the hash check
for the custom git download should be disabled.

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

at91bootstrap3: exclude from hash when downloading from git

Since we will enable hash checks for git downloads soon, the hash check
for the custom git download should be disabled.

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

barebox: exclude from hash check except for latest version

Instead of excluding the hash specifically for the custom tarball and
custom version cases, exclude it always except in the one case where
we do have a hash: the latest version.

This simplifies the code a little because soon we will add hash checks
for git as well, so we also need an exclusion in that case.

It is not needed to exclude it twice for barebox and barebox-aux,
because they use the same source tarball and barebox-aux can only be
enabled if barebox is enabled. So simplify even further by pulling
the exclusion logic out of inner-barebox-package.

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

uboot: exclude from hash check except for latest version

Instead of excluding the hash specifically for the custom tarball and
custom version cases, exclude it always except in the one case where
we do have a hash: the latest version.

This simplifies the code a little because soon we will add hash checks
for git as well, so we also need an exclusion in that case.

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

linux-headers: rework hash exclusion

Although we currently don't have a .hash file for linux-headers, there
already are exclusions for the BR2_KERNEL_HEADERS_AS_KERNEL case (copied
from linux.mk).  However, there is no exclusion for the
BR2_KERNEL_HEADERS_VERSION case.

For the BR2_KERNEL_HEADERS_AS_KERNEL case, the exclusion is actually
not needed. Indeed, KERNEL_HEADERS_SOURCE is computed to be the same
value as LINUX_SOURCE, and linux.mk already adds LINUX_SOURCE to
BR_NO_CHECK_HASH_FOR.

For the other cases, we should exclude the BR2_KERNEL_HEADERS_VERSION
case because there the user supplies the version so it can't be
included in the .hash file.

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

linux: exclude from hash check except for latest version

We don't have a hash for any linux version. We currently also don't
have a hash for the latest version, but if we ever add a hash, it
will only be for the latest version.

This simplifies the code a little because soon we will add hash checks
for git as well, so we also need an exclusion in that case. It's much
easier to always exclude except in the single case where we do have a
hash.

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

package/libv4l: toolchain w/ locale is not mandatory

libv4l package builds fine even if locale is not enabled.
This have been tested with stm32f469_disco_defconfig
enabling build of libv4l package:
+#v4l utilities
+BR2_TOOLCHAIN_BUILDROOT_CXX=y
+BR2_PACKAGE_LIBV4L=y
+BR2_PACKAGE_LIBV4L_UTILS=y

Signed-off-by: Hugues Fruchet <hugues.fruchet@st.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/libv4l: allow build of v4l2 utilities on noMMU platforms

This commit removes the dependencies on MMU and dynamic libraries.

In order to build v4l2 utilities such as compliancy tools like
v4l2-compliance or cec-compliance, v4-utils can now be built without
dynamic libraries support.

In that case the v4l-utils parts which depends on dynamic library
support are not built:
 - libv4l & libv4lconvert libraries
 - libv4l plugins
 - rds-ctl utility
 - contrib test utilities

The rest of the utilities are built.

Signed-off-by: Hugues Fruchet <hugues.fruchet@st.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

erlang-p1-xml: Add host dependencies

This should fix the error reported by autobuild:
http://autobuild.buildroot.net/results/3b4/3b474323cb50d0d5d09611fa068a9d1bd3bd4bb5/

Signed-off-by: Johan Oudinet <johan.oudinet@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/systemd: fix paths to runtime utilities

systemd's configure is looking in $PATH to find utilities that will be
needed at runtime.

Usually, those utilties, when installed on the host, will be found in
the same path they would be present on the target. For example,
/usr/bin/mount on the host would also be /usr/bin/mount on the target,
and all is find.

Except when we need to install a host variant of util-linux, which will
install mount in $(HOST_DIR), in which case systemd's configure would
find that one.

Of course, it is also very well possible that those utilities are not
installed on the host in the same location they would be on the target,
in case a user has manually installed some of those (e.g. in /usr/local/
or in /opt/)

Forcibly set the path to those utilities, as they are expected to be on
the target.

For kexec, we can set it even though we do not depend on it (yet).
systemd will appropriately test it at runtime.

For quota, we point to non-existing files, so as to catch errors at
runtime. It is to be noted that quotacheck is optional, while quotaon
does not seem to be (a service file is always installed, that uses it).

Note: utilties listed in the order they appear in configure.ac

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

package/qt5/qt5webkit: needs host gcc >= 4.8 for icu support

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

package/qt5/qt5base: needs host gcc >= 4.8 for icu support

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

package/beecrypt: needs host gcc >= 4.8 for icu support

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

package/python-pyicu: needs host gcc >= 4.8 for icu support

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

package/php: fix icu support

Propagate icu reverse dependencies.

Php lacks "-std=c++11", provided by icu-config, when compiling icu
support leading to a build failure:

/home/peko/autobuild/instance-1/output/host/usr/sh4-buildroot-linux-uclibc/sysroot/usr/include/unicode/umachine.h:347:13:
 error: 'char16_t' does not name a type
     typedef char16_t UChar;

Fixes
http://autobuild.buildroot.net/results/238/238d0e4bb205c0e7288903c6498e6c1787d86a1b/

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

package/tor: bump version to 0.3.0.8

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

package/php: bump version to 7.1.6

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

package/mesa3d: add optional support for libunwind

Added by commit
https://cgit.freedesktop.org/mesa/mesa/commit/?h=17.1&id=70c272004f727457e852ba5f2498754b07a7d995

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

package/{mesa3d, mesa3d-headers}: bump version to 17.1.2

Patch 0005 tweaked to apply on this release.

Patch 0006 already included in this release:
  https://cgit.freedesktop.org/mesa/mesa/commit/?id=e75001811e3b66986b4ede165a0fdde703d4f05b

zlib is now a mandatory dependency.
https://cgit.freedesktop.org/mesa/mesa/commit/?h=17.1&id=85a9b1b562b6a73b9494b3fad25172da3dc90fc2

[Bernd: added in v2]
- bump to mesa3d version 17.1.2
- squashed mesa3d-headers bump for bisectability
- renumbered patch 0007
- added zlib link in patch message
- Vulkan Intel driver does not depend on the i965 dri driver anymore
  https://cgit.freedesktop.org/mesa/mesa/commit/configure.ac?h=17.1&id=9aebdb5d082ec640fe8a14704201952bddb50a88
- Gallium vc4 driver depends on NEON support
  https://cgit.freedesktop.org/mesa/mesa/commit/?h=17.1&id=4d30024238efa829cabc72c1601beeee18c3dbf2

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

package/jsoncpp: bump version to 1.8.0

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

package/iprutils: needs zlib

Dependency was introduced by
https://sourceforge.net/p/iprdd/iprutils/ci/1f24b814e56939fd09a2319323e604594784fef3/

Fixes
http://autobuild.buildroot.net/results/48e/48e1c7fb920cbc81eee1cf2f43d1551a0a9ce31e/

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

python-chardet: bump to version 3.0.4

Add setuptools runtime dependency, that is needed for
the chardetect script.

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

package/kodi-pvr-stalker: fix build with gcc 7.x

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

package/oracle-mysql: fix build with gcc 7.x

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

package/qt5base: link with -latomic when needed

Qt 5.8 needs atomic operations, which are provided in libatomic when it
exists, like on sparc with gcc >= 4.8.

Fixes:
    http://autobuild.buildroot.org/results/49b/49bc9345b9849c9c3c53ace290c534ff7bb98683/
    http://autobuild.buildroot.org/results/9f2/9f213406954be51dfcad76ebdce8b73850842180/

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Julien Corjon <corjon.j@ecagroup.com>
Cc: Peter Seiderer <ps.report@gmx.net>
Tested-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/qt5base: install arch-specific configuration

For some architectures, it is necessary to link with additional
libraries, such as -latomic for sparc.

Add a bit of support to make that easy; actual use will be introduced in
the following patch.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Julien Corjon <corjon.j@ecagroup.com>
Cc: Peter Seiderer <ps.report@gmx.net>
Tested-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

util-linux: wrap help text at column 72

Respect the <tab><2 spaces><62 chars> rule.

Signed-off-by: Carlos Santos <casantos@datacom.ind.br>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

configs/stm32f4xx: use kernel version 4.11

Use kernel 4.11 for stm32f4 discovery board
(stm32f429i-disco and stm32f469i-disco).

Signed-off-by: Christophe Priouzeau <christophe.priouzeau@st.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/lcdproc: add libpng optional dependency

Fixes:
http://autobuild.buildroot.net/results/b68/b68f54404a7d412d639ebb6d052412eea6ee3bf1

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

package/sunxi-mali: fix conflicting data types between GLES2/gl2.h and GLES/gl.h

When OpenGL ES 1.1 and OpenGL ES 2.0 are used at the same time, the
build fail since GLintptr and GLsizeiptr data type are not the same in
GLES2/gl2.h and GLES/gl.h.

Also sync the data type of GLbyte, GLubyte, GLfloat, GLclampf and
GLfixed.

Fixes:
http://autobuild.buildroot.net/results/258/25898b45cefde9661d8ac87dd84bc883bb5283d1

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

package/socat: Fix TABDLY, CSIZE and CRDLY shifts for PowerPC

As can be seen in linux/arch/powerpc/include/uapi/asm/termbits.h are
defined as (note that those are octal numbers):

\#define TABDLY 00006000
\#define CSIZE 00001400
\#define CRDLY 00030000

which gives shifts of 10, 8 and 12. Adjust socat.mk accordingly to
reflect that difference.

Signed-off-by: Mark Hinds <zoronic@gmail.com>
Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

autofs: bump to version 5.1.3

Drop upstream patches.

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

swupdate: switch to local config librt dependency

Upstream rejected[1] the patch added in commit 6a1aa4e2daf0 (swupdate: fix
build with older glibc). Instead of carrying this patch going forward, make
librt an extra library dependency in the Buildroot local swupdate .config.

This solution is somewhat less optimal than the patch, because it always adds
librt regardless of whether SURICATTA_HAWKBIT is enabled. In theory we could
add a test for (BR2_PACKAGE_JSON_C && BR2_PACKAGE_LIBCURL), which are the
SURICATTA_HAWKBIT dependencies, and edit .config at build time. But this adds
much complexity for little gain.

[1] https://groups.google.com/forum/?hl=en#!topic/swupdate/77QWIka27es

Cc: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

support/scripts: fix shebang with bash

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

support/dependencies: check PATH does not carry a \n

... otherwise it fails spectacularly as soon as PATH is referenced in a
package rule (i.e. very soon, fortunately):

    >>> host-lzip 1.18 Downloading
    /bin/bash: -c: line 0: unexpected EOF while looking for matching `"'
    /bin/bash: -c: line 1: syntax error: unexpected end of file

Fixes # 9886.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Ciro Santilli <ciro.santilli@gmail.com>
[Thomas: fix typo in message, use tabs for indentation.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

sngrep: fix ncurses wchar detection

This commit adds an upstream patch to the sngrep package that fixes the
detection of wchar capable ncurses.

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

uclibc: add upstream patch to fix gdbm compile issue

This patches fixes

gdbmopen.c:35:40: error: unknown type name 'blksize_t'
 compute_directory_size (GDBM_FILE dbf, blksize_t block_size,

Fixes
  http://autobuild.buildroot.net/results/592/59278fb8681d474883e6551df5173edd806b16fa/

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

package/bdwgc: re-add fix for pthread detection on uclinux platforms

Fixes
http://autobuild.buildroot.net/results/d63/d63ee2d03421cc671455be553ab35ada7ff78267/
http://autobuild.buildroot.net/results/d8f/d8f97fd92cd971a0d125f983765ef27c5ac87acb/

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

gauche: fix build with uClibc

As gauche bundles bdwgc, it hit the same uclibc build issue fixed in commit
9b04ae71c5e4c (package/bdwgc: fix uclibc build after bump to 7.6.0). Apply the
upstream bdwgc fix for that.

Fixes:
http://autobuild.buildroot.net/results/ebc/ebc70579a94d260fecf05fe4150eebcc59d3b5fb/
http://autobuild.buildroot.net/results/e1b/e1b2ad89787c345cf8817f6d8436ff52a151c5a6/
http://autobuild.buildroot.net/results/f39/f396b996a59db79f7a0926aec49c488c3a2797a0/

Cc: Bernd Kuhls <bernd.kuhls@t-online.de>
Cc: Hiroshi Kawashima <kei-k@ca2.so-net.ne.jp>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

.gitlab-ci.yml: sync with latest defconfig additions

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

board/orangepi: add basic support for orangepi-zero board

This patch adds basic support for orangepi-zero board.
Board support package includes the following components:

- mainline kernel v4.11
- mainline u-boot v2017.03
- extra kernel config options: spidev, spi nor flash, wifi
- dts patches to enable SPI NOR and spidev

More details about this board are available here:
- http://linux-sunxi.org/Orange_Pi_Zero

Note that at the moment networking is not supported right out of the box.
Ethernet submission to mainline kernel is a work in progress, see the
latest dwmac-sun8i branches at https://github.com/montjoie/linux

On-board SDIO WiFi chip XR819 is supported by off-the-tree wireless
driver available at https://github.com/fifteenhex/xradio

Changes from v1:

* Send patch using proper subject
* Update xr819 kernel dts patch: add compatible property

[Peter: fix s/PC/Zero/ typo]
Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux: fix name of config variable

In 4c10eedc1 (systemd: enable required kernel features), we added
setting a few required kernel features to ensure systemd works.

However, there was a typo for one of the variables: CONFIG_TMPFS_XATTR
was written as CONFIG_TMPFS_POSIX_XATTR, which does not exist (and never
ever existed, at least not since 2.6.12)...

Reported-by: Michael Heinemann <posted@heine.so>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Cc: Eric Le Bihan <eric.le.bihan.dev@free.fr>
Cc: Michael Heinemann <posted@heine.so>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

x11r7: xf86-input-tslib: bump to version 1.0.0

As the update from 0.0.X to 1.0.0 suggests, this is a major update. The driver
uses X11's touch API now, instead of motion with button emulation. Furthermore
it supports multitouch when a recent version of tslib is installed.

It is not backwards compatible for very old versions of the X server. See
https://github.com/merge/xf86-input-tslib/releases for some details.

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

configs/qemu: bump to the latest kernel version

Xtensa patches no longer required, the make target name changed
to Image. The Qemu binary for OpenRisc was renamed upstream.
I removed the x86->x86_64 symlink, independent files preferred.

Defconfig               Kernel  Qemu            Network Status
--------------------------------------------------------------
aarch64_virt            4.11.3   2.9.0           YES     OK
arm_versatile           4.11.3   2.9.0           YES     OK
arm_versatile_nommu     4.4.70   2.9.0           YES     OK
arm_vexpress            4.11.3   2.9.0           YES     OK
m68k_mcf5208            4.11.3   2.9.0           YES     OK
m68k_q800               4.11.3   q800-v2.4.0     NO (2)  OK
microblazebe            4.11.3   2.9.0           YES     OK
microblazeel            4.11.3   2.9.0           YES     OK
mips32r2el_malta        4.11.3   2.9.0           YES     OK
mips32r2_malta          4.11.3   2.9.0           YES     OK
mips32r6el_malta        4.11.3   2.9.0           YES     OK
mips32r6_malta          4.11.3   2.9.0           YES     OK
mips64el_malta          4.11.3   2.9.0           YES     OK
mips64_malta            4.11.3   2.9.0           YES     OK
mips64r6el_malta        4.11.3   2.9.0           YES     OK
mips64r6_malta          4.11.3   2.9.0           YES     OK
nios2-10m50             4.11.3   2.9.0           NO      OK
or1k                    4.11.3   2.9.0           NO      OK (5)
ppc_g3beige             4.11.3   2.9.0           YES     OK
ppc_mpc8544ds           4.11.3   2.9.0           YES     OK
ppc_virtex_ml507        4.9.6    2.9.0           NO      OK (3)
ppc64_pseries           4.11.3   2.9.0           YES     OK
sh4                     4.9.6    2.9.0           YES     OK (4)
sh4eb                   4.9.6    2.9.0           NO (1)  OK (4)
sparc_ss10              4.11.3   2.9.0           YES     OK
sparc64_sun4u           4.11.3   2.9.0           YES     OK
sparc_sun4u             4.11.3   2.9.0           YES     OK
x86                     4.11.3   2.9.0           YES     OK
x86_64                  4.11.3   2.9.0           YES     OK
xtensa_lx60             4.11.3   2.9.0           YES     OK
xtensa_lx60_nommu       4.11.3   2.9.0           YES     OK

(1) - Probably an endian issue with 8139 emulation/driver
(2) - There's a network interface, but enabling it in qemu fails
(3) - Kernel oops with 4.11.3 on boot
(4) - System is extremely slow with 4.11.3, needs further investigation
(5) - Qemu binary got renamed to qemu-system-or1k

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

swupdate: fix build with older glibc

Add -lrt that is needed for clock_gettime() with glibc before 2.17.

Fixes:
http://autobuild.buildroot.net/results/034/0345814c0b0023e124992974c0600bdda0824dbf/

Cc: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

libuv: bump to version 1.12.0

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

libupnpp: needs gcc >= 4.9

After bumping libupnpp from version 0.15.1 to 0.15.2 compilation fails
for toolchains using GCC 4.7:

```
libupnpp/control/discovery.cxx: In constructor 'UPnPClient::UPnPDeviceDirectory::UPnPDeviceDirectory(time_t)':
libupnpp/control/discovery.cxx:338:5: error: 'yield' is not a member of 'std::this_thread'
```

Before version 0.15.2, libupnpp used sched_yield() which was replaced by
`std::this_thread::yield()` in the new version.

Looking at the `<thread>` header file of the PowerPC toolchain [1] from
the autobuilder shows that `yield()` is only enabled if
`_GLIBCXX_USE_SCHED_YIELD` is defined:

```
    /// yield
    inline void
    yield() noexcept
    { __gthread_yield(); }
```

Note, that `__gthread_yield()` inlines to a call to `sched_yield()`:

```
static inline int
__gthread_yield (void)
{
  return __gthrw_(sched_yield) ();
}
```

This macro is only defined if GCC was built with `--enable-libstdcxx-time`,
which is obviously not the case for the affected toolchains. For GCC 4.7 and
below this option defaults to no, which means `sched_yield()` is not enabled
though it is be available.

The issue was discussed on stackoverflow [2]. In GCC 4.9.0 the issue has
been fixed, by automatically enabling `sched_yield()` on platforms that are
known to support them.

Therefore, we update the dependency to GCC 4.9.

Fixes:
http://autobuild.buildroot.net/results/506/50670d8119b4fbd26585d3d9bd646adb1d904dbe/
http://autobuild.buildroot.net/results/8b0/8b021648a073bde25dd6936ed1ea1e39977bf89b/
http://autobuild.buildroot.net/results/b6d/b6d7437e1cf8cf0b3726395401ac4ebd9af2833b/
http://autobuild.buildroot.net/results/0bf/0bf84f61a72a65a75c5fd0a787680f483cfead26/
.. and more.

[1] http://autobuild.buildroot.org/toolchains/tarballs/powerpc-ctng_e500v2-linux-gnuspe.tar.xz
[2] https://stackoverflow.com/a/12961816

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

support/scripts: fix shebang with python

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

package/xf86-input-tslib: enable static build

since tslib now supports static builds the dependency on !BR2_STATIC_LIBS
can be removed from xf86-input-tslib too.

Signed-off-by: Martin Kepplinger <martin.kepplinger@ginzinger.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/tslib: enable statically linked building

tslib can be configured to be built statically linked. So let's add this option.

tslib is quite configurable and each module can be selected to be built into
the static build. This configuration includes the hardware independent filter
modules (obviously), and the Linux evdev input module.

This results in a configuration that fully supports multitouch in any
combination.

Signed-off-by: Martin Kepplinger <martin.kepplinger@ginzinger.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux-headers: bump 4.{4, 9, 11}.x series

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

linux: bump default version to 4.11.4

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

kmod: add patch to fix build failure with old gcc versions

When building with old gcc versions (such as gcc 4.3), the current
version of kmod fails with:

shared/util.c:52: error: expected identifier or '(' before 'do'
shared/util.c:52: error: expected identifier or '(' before 'while'

This is due to a use of the internal assert_cc() macro outside of a
function, while this macro is defined for old gcc versions in a way that
cannot be used outside functions.

This wasn't noticed by the autobuilders because we don't have such old
target compilers. On the host side, we have some autobuilder instances
with such old gcc versions, but host-kmod is only used by the linux
package, which is never tested by the autobuilders.

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