toolchain/buildroot: not available for a few mips cores

Commit 1b974425 (MIPS: add support for M6201 cores) explained that the
new core was not supported by upstream gcc, and as of gcc-8-trunk
that's still the case.

Ditto for 3cfbeb83 (MIPS: add support for P6600 cores).

This means that we currently allow to build an internal tolchain for
those cores, yet we have no suitable gcc version.

Disable the internal backend in this case.

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

connman: security bump to version 1.35

Fixes CVE-2017-12865: stack overflow in dns proxy feature.

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

package/vde2: fix build issue with gcc 7.x

Like for dieharder package [1], add static to inline functions.
https://git.buildroot.net/buildroot/commit/?id=21133ada326c87627f7bdee4493d8086587c3cca

Fixes:
http://autobuild.buildroot.net/results/ed9/ed9abb1aed2ae87971ae119f4bca83cbf9861466
http://autobuild.buildroot.net/results/2a1/2a1f73bfaf7097d42165d5c5f6c24d1aad72c929

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

dnsmasq: fix build without NLS

dnsmasq i18n support requires NLS for the provided gettext utilities.
Locale support, OTOH, is not required.

Fixes:
http://autobuild.buildroot.net/results/d45/d453aff1ce1f5e1d380bd7692a4a0565505027a3/
http://autobuild.buildroot.net/results/fb7/fb7844a239472eb48c18778aa7d515552ccfb4b9/
http://autobuild.buildroot.net/results/d85/d856ec3e6b28bd723176bb892a56dff5262a6b92/

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

package/nvidia-driver: install an egl.pc

A lot of packages expect an egl.pc to decide that EGL is available. So,
provide one.

As suggested by Alexandre, use the one from nvidia-tegra23 as template.

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

package/libphidget: fix upstream source URL

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

package/iucode-tool: fix upstream source URL

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

package/iostat: fix upstream source URL

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

package/dialog: fix upstream source URL

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

package/nss-pam-ldapd: Do not mark .service file executable

Do not mark .service file executable, otherwise systemd
will give us a warning about it.

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

package/transmission: Do not mark .service file executable

Do not mark .service file executable, otherwise systemd
will give us a warning about it.

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

package/minidlna: Do not mark .service file executable

Do not mark .service file executable, otherwise systemd
will give us a warning about it.

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

DEVELOPERS: adopt bcusdk

Signed-off-by: Kurt Van Dijck <dev.kurt@vandijck-laurijssen.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

linux: bump default to version 4.12.9

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

linux-headers: bump 4.{4, 9, 12}.x series

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

mediastreamer: add dependency on host-vim

host-vim is needed to provide the xxd tool, otherwise the build fails
with:

checking for xxd... no
configure: error: "xxd is required (provided by vim package)"

This isn't noticed by the autobuilders, presumably because all of them
have vim installed locally.

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

vim: add host variant

Vim contains a tool called xxd, which is needed by mediastreamer on
the host as part of its build process. Therefore, this commit
introduces a host variant for the vim package, that will be used by
mediastreamer.

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

libv4l: ensure libv4l gets built, fixes mediastreamer build

In commit f837251785e9389f53d695ddb8a094707865171b ("package/libv4l:
allow build of v4l2 utilities on noMMU platforms"), Hugues Fruchet
added a bunch of patches to libv4l in order to allow the v4l utilities
to be built on noMMU platforms.

However, as part of those patches, he entirely disabled the build of
libv4l in static linking configurations, because libv4l uses
dlopen(). Unfortunately, this breaks the build of applications like
mediastreamer in static linking configurations, and generally makes
our libv4l packages a little bit awkward: you can enable it, but it
doesn't install anything (in static linking configurations).

A closer look shows that dlopen() is only used by libv4l for plugin
support, and libv4l only provides one single plugin, and its build is
already conditional. Therefore, this commit adds yet another patch to
libv4l, which re-enables the build of libv4l, but disables the
plugin-related logic when plugin support is disabled (and it was
already automatically disabled in static linking configurations).

While at it, we update the comment in libv4l.mk that lists the patches
that makes autoreconf necessary.

Fixes the build of mediastreamer:

  http://autobuild.buildroot.net/results/af091cfd0508df9395778cdc796f77e95c168410/

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

package/rpcbind: Do not mark .service and .socket files executable

Do not mark .service and .socket files executable, otherwise systemd
will give us a warning about it.

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

qt-webkit-kiosk: bump to version 1.99.6-8-g26e3085

This bumps the package from 1.99.3 to 1.99.6 (stability fixes), plus a
few documentation changes and a fix for the build issue logged at [1].

[1] http://autobuild.buildroot.net/results/9ac/9acb15f955b8af31a3beeb0bd84c4b0db495e354/

Signed-off-by: Jérôme Oufella <jerome.oufella@savoirfairelinux.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

Update for 2017.08-rc3

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

cc-tool: add patch fixing boost.m4 logic for static linking

This commit adds a patch to cc-tool that fixes the boost.m4 logic used
to detect the linker rpath option so that it works properly with
static linking and additional libraries passed in LIBS.

This is the second step to fix static linking of cc-tool on
architectures like SPARC that need to link against libatomic:

  http://autobuild.buildroot.net/results/ed9f2524d0ccef318ff1bc99e5dea980111de989/

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

cc-tool: pass -latomic in LIBS

Just like -lpthread was passed in LIBS, -latomic should also be passed
in LIBS. In order for this to work, we however need to first fix
cc-tool's Makefile.am so that it does not overwrite LIBS.

This is the first part of fixing the build of cc-tool in a static
linking scenario on SPARC, i.e to fix:

  http://autobuild.buildroot.net/results/ed9f2524d0ccef318ff1bc99e5dea980111de989/

The patch has been merged upstream, in
https://github.com/dashesy/cc-tool/commit/553f9c601646af0c6e3f6150d0fa3bc63ff56416.

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

package/zmqpp: link with libatomic when needed

Fixes
http://autobuild.buildroot.net/results/c32/c32b9b8dd00d6f6d3db27fae9d8de758a4f25138/

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

dbus-cpp: fix build with gcc 7.x

This commit adds a patch to dbus-cpp to make it build with gcc 7.x.

Fixes:

  http://autobuild.buildroot.net/results/07a7559c0efeeda16c239e0fa06259d4cd48c71b/

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

mariadb: security bump version to 10.1.26

Release notes: https://mariadb.com/kb/en/mariadb-10126-release-notes/
Changelog: https://mariadb.com/kb/en/mariadb-10126-changelog/

Fixes the following security vulnerabilities:

CVE-2017-3636 - Vulnerability in the MySQL Server component of Oracle MySQL
(subcomponent: Client programs). Supported versions that are affected are
5.5.56 and earlier and 5.6.36 and earlier. Easily exploitable vulnerability
allows low privileged attacker with logon to the infrastructure where MySQL
Server executes to compromise MySQL Server. Successful attacks of this
vulnerability can result in unauthorized update, insert or delete access to
some of MySQL Server accessible data as well as unauthorized read access to
a subset of MySQL Server accessible data and unauthorized ability to cause
a partial denial of service (partial DOS) of MySQL Server.

CVE-2017-3641 - Vulnerability in the MySQL Server component of Oracle MySQL
(subcomponent: Server: DML). Supported versions that are affected are
5.5.56 and earlier, 5.6.36 and earlier and 5.7.18 and earlier. Easily
exploitable vulnerability allows high privileged attacker with network
access via multiple protocols to compromise MySQL Server. Successful
attacks of this vulnerability can result in unauthorized ability to cause
a hang or frequently repeatable crash (complete DOS) of MySQL Server.

CVE-2017-3653 - Vulnerability in the MySQL Server component of Oracle MySQL
(subcomponent: Server: DDL). Supported versions that are affected are
5.5.56 and earlier, 5.6.36 and earlier and 5.7.18 and earlier. Difficult
to exploit vulnerability allows low privileged attacker with network access
via multiple protocols to compromise MySQL Server. Successful attacks of
this vulnerability can result in unauthorized update, insert or delete
access to some of MySQL Server accessible data.

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

gcc: bump 7.x series to 7.2.0

All patches continue to apply with no changes. 7.2.0 is a bugfix
release of the 7.x branch.

The only change that is not a simple bump is that the 7.2.0 tarball is
now available xz-compressed instead of bz2-compressed.

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

libunwind: needs dynamic library support

libunwind configure script explicitly links libunwind against
libgcc_s. libgcc_s is only guaranteed to be available for toolchains
that supports dynamic linking: pure static linking toolchains only
have libgcc.a, not libgcc_s.so.

Therefore, let's make libunwind unavailable on toolchains that lack
dynamic linking support. We could potentially support linking with
libgcc, but switching to libgcc_s was done upstream because libgcc was
lacking some symbols on ARM
(https://lists.nongnu.org/archive/html/libunwind-devel/2014-06/msg00024.html). Even
though recent gcc versions seem to provide such symbols in libgcc.a,
having libunwind available on static linking configurations is not a
useful enough use-case to do the necessary research to find when this
issue was fixed in gcc.

Since libunwind is not used as a mandatory dependency in any package,
adding this !BR2_STATIC_LIBS dependency is trivial and nicely avoids
the problematic situation.

This fixes two different autobuilder failures:

 - Gstreamer 1.x programs failing to link, because libunwind links
   against libgcc_s that isn't available (static linking
   configuration):

   http://autobuild.buildroot.net/results/9d4fbf7167e9afce0eef5c9e0cfd42c966ecba36/

 - Gmrender-resurrect, which fails to link, because GStreamer 1.x uses
   some libunwind functionality, but does not take into account the
   libunwind dependency in its .pc files (static linking
   configuration):

   http://autobuild.buildroot.net/results/0a3a2485c187a000482c178f1e9c64dd716a858f/

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

python3: remove target Python packages from PYTHONPATH

This commit is similar to 350941e31de5b454cad75abe5fb0d3db027bc1a7
("python: remove target Python packages from PYTHONPATH") but for
python3.

We currently have
$(TARGET_DIR)/usr/lib/python$(PYTHON3_VERSION_MAJOR)/site-packages/
inside the PYTHON3_PATH variable, which gets used to define
PYTHONPATH, passed to the host Python interpreter when
building/installing target packages.

However, this is terribly wrong, as it causes the host interpreter to
potentially import target Python packages. This is wrong for several
reasons:

 - Some Python packages might need some Python modules to be installed
   on the host (described in setup_requires in setup.py), but their
   installation currently works because by luck the corresponding
   Python module is installed for the target. Some of those cases were
   happening for real, and fixed by previous patches.

 - Some Python packages include some native code, therefore built for
   a specific CPU architecture. When you point the host Python
   interpreter to native libraries built for the target, you get nice
   build failures, such as the one affecting the python-cffi related
   packages.

This change fixes the following build failures:

   http://autobuild.buildroot.net/results/9005b89407e46b537a54cac6cc0c69dcac4dc5ea/
   (python-cryptography)

   http://autobuild.buildroot.net/results/395682d33d02fdcaa39d3c0326355bd9ea3d6feb/
   (python-pynacl)

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

kvmtool: add patches fixing build warnings with musl

This commit backports three patches that are already upstream in
kvmtool fixing build warnings with musl. Those are not strictly needed
for the build to succeed, they just reduce the amount of warning
noise.

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

kvmtool: fix build with kernel headers >= 4.12

In Linux 4.12, the header <asm/msr-index.h> has been removed from the
set of headers exported to userspace. Therefore, it cannot be used by
kvmtool anymore. This commit takes the simple approach of duplicating
inside kvmtool the MSR_* definitions that were used from this
<asm/msr-index.h> header.

This fixes:

x86/kvm-cpu.c:7:27: fatal error: asm/msr-index.h: No such file or directory
 #include <asm/msr-index.h>

Which is the second part of:

  http://autobuild.buildroot.net/results/4459a909e735343d1cf768d30466bc3c57eca19e/

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

kvmtool: add patch to fix build with musl

This commit backports an upstream patch that fixes the build of
kvmtool with musl:

In file included from builtin-balloon.c:9:0:
include/kvm/kvm.h:22:0: warning: "PAGE_SIZE" redefined
 #define PAGE_SIZE (sysconf(_SC_PAGE_SIZE))

Fixes one part of:

  http://autobuild.buildroot.net/results/4459a909e735343d1cf768d30466bc3c57eca19e/

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

board: remove Armadeus's APF9328 support

i.MX1/L boards require an old Linux kernel version to run, which is not
compatible with latest Buildroot needs.
Will fix:
https://gitlab.com/buildroot.org/buildroot/-/jobs/27873568

Signed-off-by: Julien BOIBESSOT <julien.boibessot@armadeus.com>
[Thomas: update .gitlab-ci.yml file.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

norm: fix build with gcc 7.x

This commit adds a patch to the norm package that fixes the build with
gcc 7.x. Many thanks to Romain Naour for pointing out the solution to
this C++ build problem.

Fixes:

  http://autobuild.buildroot.net/results/c79dc84cdc34d62199099eb4438b1aed3e7459bb/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
[Thomas: add information that the patch has been submitted upstream
and accepted.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/xen: fix build with gcc7

Fixes:

  http://autobuild.buildroot.net/results/f9553c1669bd05acee3ccea76e36c2474e534d8c/

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
[Thomas: improve patch, reformatted with Git and better commit log.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

board/boundarydevices: add mac parameter for BD-SDMAC in bootscript

Our new WiFi 802.11ac BT4.1 module (BD-SDMAC [1]) driver allows to
override the wlan MAC address using a module parameter.

Since its driver is now included in our external repository [2], update
the bootscript so it sets the parameter for that driver too.

As a FYI, $wlmac is based on Ethernet MAC address located in fuses.

[1] https://boundarydevices.com/product/bd_sdmac_wifi/
[2] https://github.com/boundarydevices/buildroot-external-boundary

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

linux: bump default to version 4.12.8

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

linux-headers: bump 4.{4, 9, 12}.x series

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

core/pkg-toolchain-external: quiesce spurious stderr

Since 392b0a26f5 (toolchain-external: default BR2_TOOLCHAIN_EXTERNAL_PATH
to empty), calling 'make clean' or similar can yield a spurious stderr
message:
    dirname: missing operand
    Try 'dirname --help' for more information.

Which is definitely baffling and unsettling...

It turns out that it is pretty trivial to reproduce, and this defconfig
is just enough:

    $ cat my-defconfig
    BR2_TOOLCHAIN_EXTERNAL=y

    $ make BR2_DEFCONFIG=$(pwd)/my-defconfig defconfig

    $ make clean
    dirname: missing operand
    Try 'dirname --help' for more information.
    [--snip--]

This is because the cross-compiler is not found in the PATH (and for
good reasons, I don't have it in the PATH, not even at all).

So, when the cross-compiler is not found in the path, we simply
continue as if all was good, and postpone the check to much later,
when we try to copy the toolchain libs...

So, use a make construct rather than calling to the shell: $(dir ...)
does not whine if passed nothing.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/openocd: add license infos

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

package/openocd: fix uclinux build

Fixes
http://autobuild.buildroot.net/results/cd3/cd3f5b7329f45c2b75ed319caef3c6aa0e94ee2e/

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

package/librsync: fix build error with gcc7

Fixes
http://autobuild.buildroot.net/results/4a1/4a1931565674442c6117b2b202a002dd0ec12a4b/

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

configs/toradex_apalis_imx6: bump to U-Boot 2016.11 and kernel 4.1.41

Since Toradex V2.7 SDK, U-Boot image has changed from u-boot.imx to
SPL + u-boot.img.

Also fix U-Boot build error with GCC6.

Signed-off-by: Sergio Prado <sergio.prado@e-labworks.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

faad2: fix build with musl libc

The getopt.c code declares the strncmp() routine in a non confirming way
under non GNU libc. Patch the code to make the declaration standard
conforming.

Fixes:
http://autobuild.buildroot.net/results/447/4471be349d7ad2e998a4d55afd33aa046a5d1fd2/
http://autobuild.buildroot.net/results/2a9/2a90f4f518884fb50f7ad6ab505dee7565ed869e/
http://autobuild.buildroot.net/results/6b1/6b159b766d791492bab4d897c33ce07845fb7119/

Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

gst1-validate: disable building documentation using sphinx

Fix a build failure in case (a non functional) sphinx documentation
system is installed on the host (reported [1] and fix tested [2]
by grunpferd@netscape.net).

Fixes:

  sphinx-build -b html -d ./doctrees   . ./html
  Error: Source directory doesn't contain a conf.py file.

[1] http://lists.busybox.net/pipermail/buildroot/2017-August/200021.html
[2] http://lists.busybox.net/pipermail/buildroot/2017-August/200267.html

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Cc: grunpferd@netscape.net
[Arnout:
 - use --disable-sphinx-doc instead of a cv variable
 - remove the comment, it speaks for itself]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/gcc: remove legacy leftover for PR60102

Commit 2c8de6c4 (gcc 4.9.1: add patch for PR60102) removed the SPE
condition becasue of said PR, but forgot to remove the associated
comment, which has been tagging along all this time...

Remove it, it is no longer valid and causes confusion.

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>

package/mesa3d: vc4 backend is an EGL provider

The VC4 GPU does not support full GL, it only provides EGL.

Currently, it is possible to build the VC4 backend without EGL support,
but that does not make sense in the slighest.

So, forcibly enable EGL with VC4, like is done for etnaviv and virgl.

Reported-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Bernd Kuhls <bernd.kuhls@t-online.de>
Cc: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libepoxy: add patch to fix segfaults when using in X11 without GLX

This adds a patch cherry-picked from libepoxy 1.4.2, which adds missing
NULL-pointer checks and avoids segmentation faults when using libepoxy
under X11, when the server does not have the GLX extension, or it is
disabled -- and applications can still use EGL.

Signed-off-by: Adrian Perez de Castro <aperez@igalia.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

whois: use github repo as home page

The github repo is more informative than the list Marco's software on
the previous link.

Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

whois: fix build with NLS disabled

whois build uses the xgettext utility to build translation files.
This utility may not be available when NLS is disabled. Omit the pos and
install-pos Makefile targets from the default and the install targets
when NLS is disabled, respectively.

Fixes:
http://autobuild.buildroot.net/results/862/862d8165a59711c970c65d90009a527315a1a6b4/
http://autobuild.buildroot.net/results/b62/b62ec9d6f350e409731d47ffb585344c516944e8/
http://autobuild.buildroot.net/results/be2/be28d06ae004a46c87133d6f07709c6f909288d8/

Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

boot/syslinux: fix i386 bios build with recent binutils

When we use the cross-compiler to build syslinux with a recent binutils
version, it fails with:

/builds/arnout/buildroot/output/host/bin/i586-buildroot-linux-uclibc-ld  -Bsymbolic -pie -E --hash-style=gnu -T
/builds/arnout/buildroot/output/build/syslinux-6.03/core/i386/syslinux.ld -M -o ldlinux.elf ldlinux.o \
--start-group libcom32.a --whole-archive /builds/arnout/buildroot/output/build/syslinux-6.03/bios/com32/lib/libcom32core.a libldlinux.a --end-group -N
--no-omagic \
> ldlinux.map
/builds/arnout/buildroot/output/host/bin/i586-buildroot-linux-uclibc-ld: ldlinux.elf: Not enough room for program headers, try linking with -N
/builds/arnout/buildroot/output/host/bin/i586-buildroot-linux-uclibc-ld: final link failed: Bad value
/builds/arnout/buildroot/output/build/syslinux-6.03/core/Makefile:167: recipe for target 'ldlinux.elf' failed

Backport an upstream patch that reorganises the i386 bios build by
removing some symbols and making others hidden. To simplify the
backport, an additional patch that also touches the link script is
also included - it anyway looks like that patch could be relevant as
well.

Partially fixes: https://gitlab.com/arnout/buildroot/-/jobs/28979377

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

boot/syslinux: renumber patches

Also the last two are regenerated, their context has changed due to
the patches that have been removed earlier.

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

package/mtd: test programs need threads

The test programs depend on pthreads:
http://git.infradead.org/mtd-utils.git/blob/1bfee8660131fca7a18f68e9548a18ca6b3378a0:/configure.ac#l43

Also propagate the threads dependency to BR2_PACKAGE_MTD_INTEGCK.

Fixes
http://autobuild.buildroot.net/results/879/879c79e505f65387a46c4be263dc8783c8ca61bf/

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
[Arnout: add depends on BR2_USE_MMU to comment]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

libspatialindex: fix build on Microblaze

This commit fixes the build of libspatialindex on the Microblaze
architecture by working around a gcc bug, by passing -O0 as the
optimization level.

A patch to configure.ac is needed to not have -O2 be forced by the
libspatialindex build system, and therefore AUTORECONF=YES is now
needed.

Fixes:

  http://autobuild.buildroot.net/results/bbba2a2c97dbec21340c7fd07162a316a411cba4/

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

fs/iso9660: really create initrd temp dir

In case we're using an initrd, we create an empty "root" directory that
will contain only the bootloader stuff, not the actual root filesystem,
because it is in an initrd (standalone or initramfs).

We have to ensure that the directory is empty before assembling the
filesystem (to avoid any file lingering from a previous run, like the
sequence  "make; make"). So we first remove it before we create it, so
that on each build (especially not-from-scratch builds) we get the exact
expected content without any leftover.

However, the macro responsible for that, although defined since 7080eef9,
was never called.

Fix that by registering it as a pre-gen hook.

Note: the directory need not be created, as there are quite a few
"install -D" commands that ensure it is created. Yet, we prefer to
create it explicitly to avoid any confusion.

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

package/fakeroot: fix highly parallel uses

Although the issue can very well occur with low-paralle builds, or even
with non-parallel builds, the conditions are so strict that the ocasion
it breaks is extremely rare, to the point where a failure would go
unnoticed.

Fixes #10141.

Reported-by: Clayton Shotwell <clayton.shotwell@rockwellcollins.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Clayton Shotwell <clayton.shotwell@rockwellcollins.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

elf2flt: add patch to fix ld.real search logic

elf2flt contains a program called ld-elf2flt, which gets installed as
a replacement for ld, with ld renamed to ld.real. This program
therefore calls ld.real internally.

The logic to find ld.real worked fine for:

 - bin/TARGET_ALIAS-ld
 - TARGET_ALIAS/bin/ld

However, it failed badly if bin/SIMPLER_TARGET_ALIAS-ld is used, as
symlink to bin/TARGET_ALIAS-ld. For example:

$ ./output/host/bin/arm-buildroot-uclinux-uclibcgnueabi-ld
/home/thomas/buildroot/buildroot/output/host/bin/../arm-buildroot-uclinux-uclibcgnueabi/bin/ld.real: no input files

$ ./output/host/bin/arm-linux-ld
arm-linux-ld (ld-elf2flt): error trying to exec '/home/thomas/buildroot/buildroot/output/host/bin/ld.real': execvp: No such file or directory

$ ./output/host/arm-buildroot-uclinux-uclibcgnueabi/bin/ld
/home/thomas/buildroot/buildroot/output/host/arm-buildroot-uclinux-uclibcgnueabi/bin/ld.real: no input files

This commit fixes that by adding a patch that adjusts the ld-elf2flt
logic to properly cope with this situation.

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

whois: bump to version 5.2.17

Fixes serious usability issues (Debian changelog):

  * Fixed whois referrals for .com, .net, .jobs, .bz, .cc and .tv,
    broken by an ICANN-mandated output change:
    https://www.icann.org/resources/pages/rdds-labeling-policy-2017-02-01-en

Use snapshot.debian.org for the .dsc reference, since files tend to
disappear from the official Debian ftp site over time.

Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

gpsd: fix build for microblaze architecture

Disable O2 and add O0. The suggested flags in the gcc report
are not enough for gcc >= 6.

Fixes:

  http://autobuild.buildroot.net/results/3686cd3f3e7b6aee84f4377bd2dded1115321fb3/

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
[Thomas: use Git to format the patch, improve patch description.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

e2fsprogs: add patch for recent glibc

Recent glibc have deprecated the implicit include of sys/sysmacros.h
from sys/types.h. That means that the macros major and minor are no
longer defined unless this header is included.

This problem was observed for host-e2fsprogs when building on a host
with recent glibc.

Add an upstream patch that includes sys/sysmacros.h when needed.

Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Cc: Dagg Stompler <daggs@gmx.com>
Cc: Baruch Siach <baruch@tkos.co.il>
Tested-by: Dagg Stompler <daggs@gmx.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

configs/ci20: update kernel and u-boot

Unbreak failing defconfig.
Resulting kernel and rootfs booted on ci20 device.

While at it, drop the useless BR2_TARGET_UBOOT_VERSION variable, this
variable is a hidden variable, so it is useless to mention it in a
defconfig.

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
[Thomas: drop BR2_TARGET_UBOOT_VERSION.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

lua: ensure pkgconfig directory exists before installing files

The lua staging and host installation commands generate a file in
usr/lib/pkgconfig, without first making sure that this directory
exists, which causes build failures if it doesn't. This commit adjusts
those installation commands to create this directory if needed.

Fixes:

  http://autobuild.buildroot.net/results/101c89e1d6aee942a0b1c4e4f3daf8ac2414a56c/

Based on investigation and initial (more complicated) fix provided by
Francois Perrad <francois.perrad@gadz.org>.

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

sysvinit: fix compilation error against musl

This commit adds a patch to the sysvinit package that fixes various
build issues against musl due to missing header includes.

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

socrates_cyclone5_defconfig: U-Boot needs DTC

Probably not noticed before because dtc was already built or it was
installed on the host.

Fixes: https://gitlab.com/arnout/buildroot/-/jobs/28979369
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Cc: Michał Łyszczek <michal.lyszczek@bofc.pl>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

skeleton: Rename skeleton-common to skeleton-init-common

The skeletons are based on the selection of BR2_INIT_*, so add init- to
the package name to make this clearer. The name skeleton-common implies
that it is common to all skeletons, yet it does not apply to
skeleton-custom. It is only common to the skeleton-init-* packages, so
name it the same way.

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

skeleton: Rename skeleton-sysv to skeleton-init-sysv

The skeletons are based on the selection of BR2_INIT_*, so add init- to
the package name to make this clearer. While skeleton-sysv is relatively
clear, skeleton-common and skeleton-none are less clear on their
relationship to BR2_INIT_*. So rename skeleton-sysv to conform to a
clearer pattern.

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

skeleton: Rename skeleton-systemd to skeleton-init-systemd

The skeletons are based on the selection of BR2_INIT_*, so add init- to
the package name to make this clearer. While skeleton-systemd is
relatively clear, skeleton-common and skeleton-none are less clear on
their relationship to BR2_INIT_*. So rename skeleton-systemd to conform
to clearer pattern.

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

skeleton: Rename skeleton-none to skeleton-init-none

The skeletons are based on the selection of BR2_INIT_*, so add init- to
the package name to make this clearer. The name skeleton-none implies no
skeleton at all, not a base skeleton with no init-specific files.

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

libconfuse: bugfix bump to 3.2

This release contains only two real changes to fix two bugs. One is
only relevant for Windows; the other one fixes handling of empty
comment lines.

libconfuse is used by genimage, and we have one defconfig,
linksprite_pcduino, that has empty comment lines in its genimage.cfg.
Thus, it fixes https://gitlab.com/buildroot.org/buildroot/-/jobs/27873642

While we're at it, also change the upstream URL in Config.in to github.
The old URL on nongnu was just a download directory and didn't have
anything newer than 2010, while github has a nice README.md.

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

package/git: security bump to version 2.13.5

Fixes CVE-2017-1000117:
http://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1466490.html

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

Revert "alsa-utils: amidi option needs seq support in alsa-lib"

This reverts commit c69088b8c35177cecdd0f1f385c13f5b2c509f1d, which is
no longer needed now that alsa-lib has been fixed properly.

Indeed, this commit was made because alsa-utils failed to build with:

/home/buildroot/build/instance-0/output/host/usr/xtensa-buildroot-linux-uclibc/sysroot/usr/lib/libasound.a(pcm_dmix.o): In function `snd_pcm_dmix_htimestamp':
pcm_dmix.c:(.text+0x5c0): warning:
/home/buildroot/build/instance-0/output/host/usr/xtensa-buildroot-linux-uclibc/sysroot/usr/lib/libasound.a(rawmidi_symbols.o):(.data.rel+0x4): undefined reference to `_snd_module_rawmidi_virt'
collect2: error: ld returned 1 exit status

Which is exactly the problem that we fixed in alsa-lib by making sure
that rawmidi_symbols.c doesn't reference symbols from rawmidi_virt.c
when SEQ support is not enabled.

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

alsa-lib: add patch to fix incorrect code dependency

This commit adds a patch to alsa-lib that fixes an incorrect code
dependency: some code built when MIDI support is enabled is
unconditionally referencing code that is built when SEQ support is
enabled, both options being independent.

This causes build failures such as:

/home/buildroot/buildroot-test/instance-1/output/host/arm-buildroot-linux-uclibcgnueabi/sysroot/usr/lib/libasound.a(pcm_dmix.o): In function `snd_pcm_dmix_sync_ptr':
pcm_dmix.c:(.text+0x83c): warning:
/home/buildroot/buildroot-test/instance-1/output/host/arm-buildroot-linux-uclibcgnueabi/sysroot/usr/lib/libasound.a(rawmidi_symbols.o):(.data+0x4): undefined reference to `_snd_module_rawmidi_virt'

Fixes:

  http://autobuild.buildroot.net/results/5964e7028c31a8f75b38a27270410f0b3c5c8422/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Arnout: add upstream link to the patch]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

Update for 2017.08-rc2

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

board/beaglebone: skip bonegreen.dtb if not built

The beaglebone_qt5_defconfig uses an older kernel version which does not
provide a device tree for the beagle bone green.
post-image.sh now selects genimage_linux41.cfg if am335x-bonegreen.dtb is
not build and genimage.cfg otherwise.

Signed-off-by: Lothar Felten <lothar.felten@gmail.com>
[Thomas: minor tweaks.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libcurl: fix build on uncommon architectures

Since the bump to 7.55.0, libcurl fails to build on a number of
uncommon architectures (ARC, OpenRISC, etc.). This is due to upstream
commit 73a2fcea0b4adea6ba342cd7ed1149782c214ae3 ("includes: remove
curl/curlbuild.h and curl/curlrules.h"), which makes libcurl rely on
more architecture-specific related defines in include/curl/system.h.

This commit therefore adds a patch that fixes the 32-bit vs. 64-bit
detection for all architecture, using gcc's __SIZEOF_LONG__
definition. It has been tested successfully with test-pkg on all 47
toolchain configurations.

Fixes:

  http://autobuild.buildroot.net/results/bf26c08cf3267214278674472f931603f69951ae/
  (and many similar issues)

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

beaglebone_qt5_defconfig: use gcc 5.x

The linux 4.1 kernel does not compile with gcc 6.x.
The compiler version is now fixed to 5.x until a new kernel is used.

Signed-off-by: Lothar Felten <lothar.felten@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

nanopi_m1{, _plus}_defconfig: U-Boot needs pylibfdt

nanopi_m1 and nanopi_m1_plus are sunxi boards that need pylibfdt in
U-Boot.

Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/27873655
https://gitlab.com/buildroot.org/buildroot/-/jobs/27873656

Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Chakra Divi <2chakrass@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

minnowboard_max_defconfig: grub2 needs WCHAR

Since commit 5ffafd2353f5d67 (grub2: bump up version), grub2 needs a
toolchain with WCHAR support. However, the defconfigs that use grub2
were not updated at the time.

Add BR2_TOOLCHAIN_BUILDROOT_WCHAR=y to minnowboard_max_defconfig.

Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Cc: Erico Nunes <nunes.erico@gmail.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
Cc: Vincent Stehlé <vincent.stehle@intel.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/qt5base: only build libgtk3 support for X11 backend

libgtk3 can conditionally be build with or without the X11 backend
unlike libgtk2. If libgtk3 is enabled and the X11 backend is not enabled
for BR2_PACKAGE_QT5_VERSION_LATEST, QT_EGL_NO_X11 or
MESA_EGL_NO_X11_HEADERS needs to be defined.

The proper way is to patch Qt to use pkg-config to fetch the correct
flags [1], or to add it to arch.conf like is done in commit 7d286be4f9.
For now, however, just disable libgtk3 support unless the X11 backend
is used.

Fixes http://autobuild.buildroot.net/results/6f3c2fcb0a44cd13ef4c73822d4d08f923291099

[1] https://codereview.qt-project.org/#/c/198906/

Signed-off-by: Joshua Henderson <joshua.henderson@microchip.com>
Cc: Julien Corjon <corjon.j@ecagroup.com>
Cc: Brent Sink <brents_3@hotmail.com>
Cc: Naoki Matsumoto <n-matsumoto@melcoinc.co.jp>
[Arnout:
- extend commit message
- improve comment
- also pass -no-gtk if libgtk3 is not selected]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/valgrind: hide comment when arch is not supported

Currently, the comment that "valgrind needs shared libs" is not hidden
when the architecture dependencies are not met, which can confuse some
users (as recently seen on IRC).

Fix that by introducing the traditional _ARCH_SUPPORTS option, and have
the comment and the symbol depend on that.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Brandon Maier <brandon.maier@rockwellcollins.com>
Cc: Jérôme Pouiller <jezz@sysmic.org>
Cc: Peter Seiderer <ps.report@gmx.net>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Reviewed--by: Pedro Ribeiro <pedrib@gmail.com>
Tested-by: Pedro Ribeiro <pedrib@gmail.com>
[Arnout: put _ARCH_SUPPORTS at the top of the file]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

erlang-p1-xml: Fix host dependencies variable name

This should fix the error reported by autobuild:
http://autobuild.buildroot.net/results/63c/63c63b2578e9f6ad71cbea82b987c128305cdab4/

Signed-off-by: Johan Oudinet <johan.oudinet@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

libcurl: security bump to version 7.55.0

Fixes:

 glob: do not parse after a strtoul() overflow range (CVE-2017-1000101)
 tftp: reject file name lengths that don't fit (CVE-2017-1000100)
 file: output the correct buffer to the user (CVE-2017-1000099)

Switch to .tar.xz to save bandwidth.

Add reference to tarball signature.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

linux: Fix URL for release candidate versions

Starting with 4.12-rc1, tarballs are generated by cgit directly from
Linus's tree. This also implies that no .tar.xz can be used for them.

This method also applies to older release candidates.

Signed-off-by: Luis Araneda <luaraneda@gmail.com>
[Arnout: added comment in the code]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

swupdate: Lua 5.1 / LuaJIT are not supported

Before commit 87b6ac1478821351c92f7ca1c3154550e4713b28 support for Lua
was always disabled by the default config file:

```
CONFIG_LUA is not set
```

The commit removed this setting and Lua support is now enabled if a Lua
interpreter is enabled. As the swupdate build system uses pkg-config to
check for the lua library by default (LUAPKG="lua") this throws a lot of
undefined referenced in case LuaJIT is uses as Lua interpreter, e.g.:

```
corelib/lib.a(lua_interface.o): In function `l_info':
lua_interface.c:(.text.l_info+0x14): undefined reference to `luaL_checklstring'
```

In addition, since version 2017.07, Lua 5.1 really isn't supported any
more (before this version, it was only unsupported if the option
CONFIG_HANDLER_IN_LUA was set).  Therefore, remove support for Lua 5.1
and LuaJIT in the swupdate package for now until upstream might fix
this issue.

Reported upstream:
https://groups.google.com/forum/#!topic/swupdate/WAm8npAOd6o

Fixes:
http://autobuild.buildroot.net/results/df2/df2a71efe5af52d7b8721a355c49934b1be197a3/
http://autobuild.buildroot.net/results/400/4006225c8a47eb0b56399c83bd6d00406a0f62c2/
http://autobuild.buildroot.net/results/098/098f9d6cd905359adac4fb15e1d54c5022757325/
http://autobuild.buildroot.net/results/e5f/e5f6f99d96d9c661454335e7f931a03c3ae6a9e2/
http://autobuild.buildroot.net/results/f3a/f3a0abe8d5e35c85da40d20dab260c28506a0d4c/

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

lua: fix pkg-config file

When Lua is linked with additional libraries, these libraries should go
into the pkg-config file as well.

Otherwise, linking swupdate with the lua library fails:

```
/home/buildroot/output/host/arm-buildroot-linux-gnueabihf/sysroot/usr/lib/liblua.so: undefined reference to `dlopen'
/home/buildroot/output/host/arm-buildroot-linux-gnueabihf/sysroot/usr/lib/liblua.so: undefined reference to `dlclose'
/home/buildroot/output/host/arm-buildroot-linux-gnueabihf/sysroot/usr/lib/liblua.so: undefined reference to `dlerror'
/home/buildroot/output/host/arm-buildroot-linux-gnueabihf/sysroot/usr/lib/liblua.so: undefined reference to `dlsym'
```

Fixes http://autobuild.buildroot.net/results/1c3/1c349cc5904868e4def292b9fbfa164828e46156

Signed-off-by: Jörg Krause <joerg.krause@embedded.rocks>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

olimex_a20_olinuxino_lime: bump linux and U-Boot versions

This fixes build errors with GCC 6.

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

webkitgtk: Add patch for properly picking GL flags when building

This solves build failures caused by WebKit trying to include X11 headers
when support for X11 is disabled in Mesa3D. A common situation is when
configuring both GTK+ and WebKitGTK+ only with Wayland support.

Once the fix for https://bugs.webkit.org/show_bug.cgi?id=175125 makes it
into a release, the patch can be dropped.

Signed-off-by: Adrian Perez de Castro <aperez@igalia.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

gdb: add fix for gdb 7.12 and gdb 8.0 build on noMMU platforms

This adds a patch to gdb 7.12 and gdb 8.x, which fixes the build on
noMMU platforms. It is not needed for older versions of gdb, since
it's related to the switch of gdb to C++ in the 7.12 release.

Fixes:

../nat/linux-ptrace.c: In function 'int linux_fork_to_function(gdb_byte*, int (*)(void*))':
../nat/linux-ptrace.c:273:29: error: invalid conversion from 'void*' to 'gdb_byte* {aka unsigned char*}' [-fpermissive]
       child_stack = xmalloc (STACK_SIZE * 4);

The patch has already been merged upstream, as of commit
ffce45d2243e5f52f411e314fc4e1a69f431a81f, and will therefore be part
of future gdb releases.

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

glibc: fix sh4 compile error

Fixes:

/tmp/output/build/glibc-2.25/build/libc_pic.a(dl-error.os): In function `__GI__dl_signal_error':
dl-error.c:(.text+0x0): multiple definition of `_dl_signal_error'
/tmp/output/build/glibc-2.25/build/elf/dl-allobjs.os:(.text+0x134d8): first defined here
/tmp/output/build/glibc-2.25/build/libc_pic.a(dl-error.os): In function `__GI__dl_catch_error':
dl-error.c:(.text+0x190): multiple definition of `_dl_catch_error'
/tmp/output/build/glibc-2.25/build/elf/dl-allobjs.os:(.text+0x136fc): first defined here
/tmp/output/build/glibc-2.25/build/libc_pic.a(init-first.os):(.data+0x0): multiple definition of `__libc_multiple_libcs'
/tmp/output/build/glibc-2.25/build/elf/dl-allobjs.os:(.bss+0x6c): first defined here
/tmp/output/build/glibc-2.25/build/libc_pic.a(_itoa.os): In function `_itoa':
_itoa.c:(.text+0xd4): multiple definition of `_itoa'
/tmp/output/build/glibc-2.25/build/elf/dl-allobjs.os:(.text+0x133c8): first defined here

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

gcc: fix sparcv8 compile issue

Fix compile issue with gcc 7.1.0:
/tmp/ccxalnSf.s: Assembler messages:
/tmp/ccxalnSf.s:12: Error: Architecture mismatch on "rd %tick,%o1".
/tmp/ccxalnSf.s:12: (Requires v9|v9a|v9b|v9c|v9d|v9e|v9v|v9m|m8; requested architecture is v8.)
/tmp/ccxalnSf.s:13: Error: Architecture mismatch on "srlx %o1,32,%o0".
/tmp/ccxalnSf.s:13: (Requires v9|v9a|v9b|v9c|v9d|v9e|v9v|v9m|m8; requested architecture is v8.)
/tmp/ccxalnSf.s:30: Error: Architecture mismatch on "rd %ccr,%g0".
/tmp/ccxalnSf.s:30: (Requires v9|v9a|v9b|v9c|v9d|v9e|v9v|v9m|m8; requested architecture is v8.)
/tmp/ccxalnSf.s:33: Error: Architecture mismatch on "rd %ccr,%g0".
/tmp/ccxalnSf.s:33: (Requires v9|v9a|v9b|v9c|v9d|v9e|v9v|v9m|m8; requested architecture is v8.)
/tmp/ccxalnSf.s:36: Error: Architecture mismatch on "rd %ccr,%g0".
/tmp/ccxalnSf.s:36: (Requires v9|v9a|v9b|v9c|v9d|v9e|v9v|v9m|m8; requested architecture is v8.)
/tmp/ccxalnSf.s:39: Error: Architecture mismatch on "rd %ccr,%g0".
/tmp/ccxalnSf.s:39: (Requires v9|v9a|v9b|v9c|v9d|v9e|v9v|v9m|m8; requested architecture is v8.)
Makefile:684: recipe for target 'os-unix-sysdep.lo' failed

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

.gitlab-ci.yml: use large timeouts for runtime tests

Multiply the timeouts for emulator in the gitlab runners by 10 to avoid
sporadic failures in elastic runners.

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

support/testing: allow to use a multiplier for timeouts

Add a parameter to run-tests to act as a multiplier for all timeouts of
emulator.
It can be used to avoid sporadic failures on slow host machines as well
in elastic runners on the cloud.

Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
[Arnout: rename multiplier to timeout_multiplier everywhere]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

ficl: fix build with gcc 7

gcc 7 with -Os (optimize for size) takes the liberty to remove the code
of inline functions entirely - which is indeed allowed according to the
C99 standard, cfr. this [1] very complete StackOverflow answer. This
leads to undefined function references at link time.

The proper approach is to patch the code to use 'static inline'
instead. However, for various reasons such a patch is not entirely
trivial, and upstream has not moved since 7 years, so instead take the
simpler approach of restoring the gcc original inline behaviour with
-fgnu89-inline to fix this issue.

Fixes http://autobuild.buildroot.org/results/d72/d72a7e9775a10bc884b9953dd2eba3ad32100cca

[1] https://stackoverflow.com/questions/6312597/is-inline-without-static-or-extern-ever-useful-in-c99

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/stella: fix gcc7 support

Fixes
http://autobuild.buildroot.net/results/6ed/6edd7950c402dbf433f3904193c645ebc8531c7e/

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

Revert "package/skeleton-custom: also check for missing directories"

This partially reverts commit cf6d321e9d5c9dec63e42466fec4bf5de5853bd3.

The underlying reason for doing the checks is to ensure that a
merged-usr setting is properly enforced, even for custom skeletons.

Before that patch, a custom skeleton where both /bin and /usr/bin were
missing was accepted; but then the first package that intalled something
in /bin would create it as a directory, thus breaking the merged-usr
situation. Ditto sbin and lib, of course.

cf6d321 was created to detect that situation, and the fix was to require
that both directories do exist in the custom skeleton, so that we new
the check for consistency were OK, in all cases.

However, that broke existing skeletons which where missing both
directories, and some people are shinning about it...

The crux of the problem is that Buildroot should be responsible for
creating the directories or the symliks when they are missing, and only
fail if the existing ones are incorrect, but not impose the burden on
the user.

A situation where the problem arises is when a skeleton is shared
between various builds, some using a merged-usr while other do not.

We fix that by reverting the offending changes change, back to the
previous behaviour. We keep the stderr redirection in stat calls, and
the variable renaming.

A proper fix to create the missing directories can be added later.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

uclibc: allow thread_db to be built with C++

gdb 7.12+ is compiled and linked with C++, therefore the thread_db
headers need to use extern "C", to be usable for C++ code.

Fixes:
 Issue when building gdbserver for m68k-coldfire:

thread-db.o: In function `thread_db_init()':
thread-db.c:(.text+0x5b6): undefined reference to `td_ta_new(ps_prochandle*, td_thragent**)'
thread-db.c:(.text+0x61e): undefined reference to `td_thr_get_info(td_thrhandle const*, td_thrinfo*)'
thread-db.c:(.text+0x632): undefined reference to `td_symbol_list()'
...

Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Tested-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

pkg-cmake.mk: set pkg-config env. vars. in the host configure environment

This change is fixing the bug reported in [1].

Cmake may run pkg-config to find the dependencies when configuring a
package. Because of the value of PATH, and it will use the Buildroot's
pkg-config wrapper, which, by default, is configured (via some
environment variables) to find the target dependencies.

When configuring a host package using cmake, to prevent cmake from
wrongly solving dependencies from the target tree (when the
*-config.cmake files are using pkg-config) instead of looking for them
in the Buildroot's host tree or in the host system itself, we need to
set the environment variables altering the pkg-config behavior in the
cmake configure environment.

So, this change is fixing the cmake host-packages configuration step,
by properly setting the pkg-config environment variables to their values
for finding host dependencies.

Before this patch:

  make O=/opt/br/abo/cmake-host-find-root-path libxml2 host-mariadb{-dirclean,-configure} && echo $?
  [...]
  >>> host-mariadb 10.1.25 Configuring
  (mkdir -p /opt/br/abo/cmake-host-find-root-path/build/host-mariadb-10.1.25/ && cd /opt/br/abo/cmake-host-find-root-path/build/host-mariadb-10.1.25/ && rm -f CMakeCache.txt && PATH="/opt/br/abo/cmake-host-find-root-path/host/bin:/opt/br/abo/cmake-host-find-root-path/host/sbin:/home/sam/.local/bin:/sbin:/usr/sbin:/bin:/usr/bin"  /usr/bin/cmake /opt/br/abo/cmake-host-find-root-path/build/host-mariadb-10.1.25/ -DCMAKE_INSTALL_SO_NO_EXE=0 -DCMAKE_FIND_ROOT_PATH="/opt/br/abo/cmake-host-find-root-path/host" -DCMAKE_FIND_ROOT_PATH_MODE_PROGRAM="BOTH" -DCMAKE_FIND_ROOT_PATH_MODE_LIBRARY="BOTH" -DCMAKE_FIND_ROOT_PATH_MODE_INCLUDE="BOTH" -DCMAKE_INSTALL_PREFIX="/opt/br/abo/cmake-host-find-root-path/host" -DCMAKE_C_FLAGS="-O2 -I/opt/br/abo/cmake-host-find-root-path/host/include" -DCMAKE_CXX_FLAGS="-O2 -I/opt/br/abo/cmake-host-find-root-path/host/include" -DCMAKE_EXE_LINKER_FLAGS="-L/opt/br/abo/cmake-host-find-root-path/host/lib -Wl,-rpath,/opt/br/abo/cmake-host-find-root-path/host/lib"
  -DCMAKE_ASM_COMPILER="/usr/bin/as" -DCMAKE_C_COMPILER="/usr/bin/gcc" -DCMAKE_CXX_COMPILER="/usr/bin/g++"  -DCMAKE_C_COMPILER_ARG1="" -DCMAKE_CXX_COMPILER_ARG1=""  -DCMAKE_COLOR_MAKEFILE=OFF -DBUILD_DOC=OFF -DBUILD_DOCS=OFF -DBUILD_EXAMPLE=OFF -DBUILD_EXAMPLES=OFF -DBUILD_TEST=OFF -DBUILD_TESTS=OFF -DBUILD_TESTING=OFF  -DWITH_SSL=bundled )
  -- Running cmake version 3.8.2
  [...]
  -- Found PkgConfig: /opt/br/abo/cmake-host-find-root-path/host/bin/pkg-config (found version "0.28")
  [...]
  -- Found LibXml2: /opt/br/abo/cmake-host-find-root-path/host/arm-buildroot-linux-gnueabihf/sysroot/usr/lib/libxml2.so (found version "2.9.4")
  [...]
  0

After this patch is applied:

  make O=/opt/br/abo/cmake-host-find-root-path libxml2 host-mariadb{-dirclean,-configure} && echo $?
  [...]
  >>> host-mariadb 10.1.25 Configuring
  (mkdir -p /opt/br/abo/cmake-host-find-root-path/build/host-mariadb-10.1.25/ && cd /opt/br/abo/cmake-host-find-root-path/build/host-mariadb-10.1.25/ && rm -f CMakeCache.txt && PATH="/opt/br/abo/cmake-host-find-root-path/host/bin:/opt/br/abo/cmake-host-find-root-path/host/sbin:/home/sam/.local/bin:/sbin:/usr/sbin:/bin:/usr/bin" PKG_CONFIG="/opt/br/abo/cmake-host-find-root-path/host/bin/pkg-config" PKG_CONFIG_SYSROOT_DIR="/" PKG_CONFIG_LIBDIR="/opt/br/abo/cmake-host-find-root-path/host/lib/pkgconfig:/opt/br/abo/cmake-host-find-root-path/host/share/pkgconfig" PKG_CONFIG_ALLOW_SYSTEM_CFLAGS=1 PKG_CONFIG_ALLOW_SYSTEM_LIBS=1  /usr/bin/cmake /opt/br/abo/cmake-host-find-root-path/build/host-mariadb-10.1.25/ -DCMAKE_INSTALL_SO_NO_EXE=0 -DCMAKE_FIND_ROOT_PATH="/opt/br/abo/cmake-host-find-root-path/host" -DCMAKE_FIND_ROOT_PATH_MODE_PROGRAM="BOTH" -DCMAKE_FIND_ROOT_PATH_MODE_LIBRARY="BOTH" -DCMAKE_FIND_ROOT_PATH_MODE_INCLUDE="BOTH" -DCMAKE_INSTALL_PREFIX="/opt/br/abo/cmake-host-find-roo
 t-path/host" -DCMAKE_C_FLAGS="-O2 -I/opt/br/abo/cmake-host-find-root-path/host/include" -DCMAKE_CXX_FLAGS="-O2 -I/opt/br/abo/cmake-host-find-root-path/host/include" -DCMAKE_EXE_LINKER_FLAGS="-L/opt/br/abo/cmake-host-find-root-path/host/lib -Wl,-rpath,/opt/br/abo/cmake-host-find-root-path/host/lib" -DCMAKE_ASM_COMPILER="/usr/bin/as" -DCMAKE_C_COMPILER="/usr/bin/gcc" -DCMAKE_CXX_COMPILER="/usr/bin/g++"  -DCMAKE_C_COMPILER_ARG1="" -DCMAKE_CXX_COMPILER_ARG1=""  -DCMAKE_COLOR_MAKEFILE=OFF -DBUILD_DOC=OFF -DBUILD_DOCS=OFF -DBUILD_EXAMPLE=OFF -DBUILD_EXAMPLES=OFF -DBUILD_TEST=OFF -DBUILD_TESTS=OFF -DBUILD_TESTING=OFF  -DWITH_SSL=bundled )
  -- Running cmake version 3.8.2
  [...]
  -- Found PkgConfig: /opt/br/abo/cmake-host-find-root-path/host/bin/pkg-config (found version "0.28")
  [...]
  -- Found LibXml2: /usr/lib/libxml2.so (found version "2.9.4")
  [...]
  0

[1] http://lists.busybox.net/pipermail/buildroot/2017-August/199776.html

Reported-by: "Sigalas, Antonios (Nokia - GR/Athens)" <antonios.sigalas@nokia.com>
Cc: "Sigalas, Antonios (Nokia - GR/Athens)" <antonios.sigalas@nokia.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

faad2: security bump to version 2.8.1

Fixes: CVE-2017-9218, CVE-2017-9219, CVE-2017-9220, CVE-2017-9221,
CVE-2017-9222, CVE-2017-9223, CVE-2017-9253, CVE-2017-9254,
CVE-2017-9255, CVE-2017-9256, CVE-2017-9257

http://seclists.org/fulldisclosure/2017/Jun/32

Switch to .tar.bz2 to save some bandwidth.

Add autoreconf since unfortunately upstream tarball does not ship the
configure script.

Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

nvme: remove the udev optional dependency

Upstream removed the udev dependent code in commit 4cbe5888f4bc (nvme-cli:
fabrics: remove libudev dependency).

Cc: Mamatha Inamdar <mamatha4@linux.vnet.ibm.com>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

nvme: fix libuuid test when cross compiling

Use $(LD) to test whether libuuid is installed, so that we check the target,
not the host.

Fixes:
http://autobuild.buildroot.net/results/e26/e263ffbf8947374c2aac73ab42edcbf4d9a3600a/
http://autobuild.buildroot.net/results/53b/53b2d1e0c3cdeafb980389bcfd8a10b6ca48fbc2/

Cc: Mamatha Inamdar <mamatha4@linux.vnet.ibm.com>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>