fs/iso9660: prepare cleaner kernel/initrd path handling

The iso9660.mk modifies the menu.lst file from Grub to set the correct
initrd/kernel image locations. However, with the upcoming support of
other bootloaders for iso9660 filesystems, we need to modify a bit
this logic.

Instead of relying on the specific details of the grub menu.lst
syntax, we introduce the __KERNEL_PATH__ and __INITRD_PATH__ magic
keywords, which iso9660.mk will replace by the appropriate
values. They can therefore be used where needed in grub menu.lst, and
in similar configuration files of other bootloaders, as will be
supported in the following commits.

Also, in order to be consistent with the soon to be introduced support
for having the root filesystem itself as iso9660, this commit changes
the installation location of the initrd and kernel. Instead of being
/initrd and /kernel, they become /boot/initrd and
/boot/$(LINUX_IMAGE_NAME).

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>

fs/iso9660: get grub splash from $(TARGET_DIR)

Instead of using directly the splash image from the Buildroot source
directory boot/grub, this commit changes the iso9660 logic to use the
splash image installed in $(TARGET_DIR)/boot/grub.

This effectively allows a user to use a custom splash image by
installing it to $(TARGET_DIR) through a rootfs overlay or using a
post-build script.

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

fs/iso9660: change the location of the splash image

In order to simplify the introduction of the support for using ISO9660
as the real root filesystem, this commit changes the location of the
grub splash image. This makes it match where the splash image is
located in $(TARGET_DIR), so that regardless of whether the
initrd/initramfs solution or the real iso9660 filesystem solution are
used, the splash image is installed at the same location.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>

fs/iso9660: rename all variables to use the ROOTFS_ISO9660 prefix

For consistency reasons, this commit renames all internal variables of
iso9660.mk to use the ROOTFS_ISO9660 prefix.

While we're at it, replace a useless ':=' by '='.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>

fs/iso9660: enable Joliet extension

We already enable the Rockridge extension by default when building
ISO9660, so let's also enable the Joliet extension which allows to
support Unicode file names and long file names.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>

fs/iso9660: convert to the filesystem infrastructure

This commit converts the iso9660 logic to the common rootfs
infrastructure. What previously prevented it from being converted is
that the iso9660 logic needed to remove a temporary folder after the
image has been created.

However, since Buildroot typically keeps build artefacts around, this
commit changes the logic to keep this temporary folder around. Thanks
to this change, converting to the common rootfs infrastructure becomes
possible.

In addition, the temporary folder is renamed from $(BUILD_DIR)/iso9660
to the more descriptive $(BUILD_DIR)/rootfs-iso9660.tmp.

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

fs/iso9660: use if ... endif block instead of depends on

In preparation to the addition of numerous additional options to the
iso9660 filesystem logic, use a if ... endif block instead of a
depends on for the only option that currently exists.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>

package/upmpdcli: bump to version 0.11.0

upmpdcli 0.11.0 requires libupnpp 0.11.0 and does not build with the previous
version 0.8.6 of libupnpp.

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

package/libupnpp: bump to version 0.11.0

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

package/libuv: bump to version 1.6.1

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

package/zic: bump version to 2015e

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

package/tzdata: bump version to 2015e

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

package/libgpg-error: Fix compilation with gcc5

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

package/x11r7/xserver_xorg-server: Fix sdksyms.sh to cope with gcc5

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

uclibc: remove Xtensa specific version

Since we now have uClibc-ng, and it is used as the default uClibc
version, there is no real need to keep the Xtensa version: this
version is just a specific Git revision of uClibc master branch, and
uClibc-ng is also based on uClibc master branch.

In addition, since in the choice of uClibc versions, uClibc-ng is
already *before* the Xtensa specific version, uClibc-ng has in fact
already been the default uClibc version for Xtensa since uClibc-ng
support was introduced.

See for example:

  http://autobuild.buildroot.org/results/c85/c854fae31d82e0f11d5a8b5bfc196c29bd7e21db/config

for a build on Xtensa with the internal toolchain, where in fact
uClibc-ng is used instead of the Xtensa specific version.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Waldemar Brodkorb <wbx@openadk.org>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Max Filippov <jcmvbkbc@gmail.com>

uclibc: use uClibc-ng by default

Since the uClibc project is not making any release since about 3
years, and the uClibc-ng project is now alive and delivering regular
releases based on uClibc master plus a number of fixes, let's make
uClibc-ng the default uClibc version used by Buildroot.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Waldemar Brodkorb <wbx@openadk.org>
Acked-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>

package/musl: bump to version 1.1.10

A new regression in uselocale made it into this release. As suggested [1] by the
musl maintainers add a patch from upstream [2] to avoid breaking programs that
call uselocale.

[1]
http://www.musl-libc.org/download.html

[2]
http://git.musl-libc.org/cgit/musl/patch/?id=63f4b9f18f3674124d8bcb119739fec85e6da005

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

package/shairport-sync: bump to version 2.2.4

Amongst other things, the new release fixes the build with musl.

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

bc: patch makeinfo logic

It uses a hardcoded "makeinfo" command invocation instead of the usual
MAKEINFO variable with the missing logic for docs. Fixes:
http://autobuild.buildroot.net/results/b0b/b0b9dced0014a5a8026fa972f95085e8bd16a07b/

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

sysstat: fix hash

Fix brainfail: mixed hash from dev and stable versions.
Fixes:
http://autobuild.buildroot.net/results/bdc/bdce55b88a4aaff82ce0a492a0a4f49ab5b9031b/

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

atf: needs mmu

Fixes:
http://autobuild.buildroot.net/results/950/95069436dd5b2004e59851b8c731d534b86062a5/

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

bc: bump version to 1.06.95

The current version of bc being used (1.06) is from 2000 and contains a
serious bug causing it to segfault when the math library is used, so bump to
the latest alpha release, which is also 9 years old.

Also include two fixes for Debian (https://packages.debian.org/jessie/bc)

- A patch to fix array initialization by Phul Nelson
- A patch to get bc to notice I/O errors by Ian Jackson

[Peter: fixup white space and tweak commit message]
Signed-off-by: Robert Sohn <grepper@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/gdb: Add optional dependencies

To allow reproducable builds of target gdb add some optional dependencies

$ output/host/usr/bin/i586-buildroot-linux-uclibc-readelf -a output/target/usr/bin/gdb | grep NEEDED
 0x00000001 (NEEDED)                     Shared library: [libdl.so.0]
 0x00000001 (NEEDED)                     Shared library: [libncurses.so.5]
 0x00000001 (NEEDED)                     Shared library: [libz.so.1]
 0x00000001 (NEEDED)                     Shared library: [libm.so.0]
 0x00000001 (NEEDED)                     Shared library: [libiconv.so.2]
 0x00000001 (NEEDED)                     Shared library: [libpthread.so.0]
 0x00000001 (NEEDED)                     Shared library: [libutil.so.0]
 0x00000001 (NEEDED)                     Shared library: [libpython2.7.so.1.0]
 0x00000001 (NEEDED)                     Shared library: [libexpat.so.1]
 0x00000001 (NEEDED)                     Shared library: [libc.so.0]

[Thomas: explicitly use --{with,without}-{expat,zlib}. There is no
such option for libiconv, so for libiconv, the code was left as is.]

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

package/xfont_font-screen-cyrillic: add hash file

Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/xfont_font-sony-misc: add hash file

Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/xfont_font-sun-misc: add hash file

Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/xfont_font-winitzki-cyrillic: add hash file

Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/xfont_font-xfree86-type1: add hash file

Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

package/libpthread-stubs: add hash file

Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

openssl: bump to version 1.0.2c

Fixes HMAC ABI breakage from 1.0.2b

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

glog: new package

Signed-off-by: Rahul Bedarkar <rahul.bedarkar@imgtec.com>
Reviewed-by: Abhimanyu Vishwakarma <abhimanyu.v@imgtec.com>
Reviewed-by: Marcin Nowakowski <marcin.nowakowski@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

kyua: new package

Kyua is a testing framework for infrastructure software, originally
designed to equip BSD-based operating systems with a test suite. This
means that Kyua is lightweight and simple, and that Kyua integrates
well with various build systems and continuous integration frameworks.

https://github.com/jmmv/kyua

[Thomas: fix typo in hash file.]

Signed-off-by: Sebastien Bourdelin <sebastien.bourdelin@savoirfairelinux.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

lutok: new package

Lutok is a lightweight C++ API library for Lua.

https://github.com/jmmv/lutok

[Thomas:
 - move lutok inclusion in package/Config.in after including the Lua
   external modules
 - add a comment explaining why we are passing --without-atf
 - remove the logic around -ldl, it seems to build just fine without it.]

Signed-off-by: Sebastien Bourdelin <sebastien.bourdelin@savoirfairelinux.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

atf: new package

ATF, or Automated Testing Framework, is a collection of libraries to
write test programs in C, C++ and POSIX shell.

https://github.com/jmmv/atf

Signed-off-by: Sebastien Bourdelin <sebastien.bourdelin@savoirfairelinux.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

python-pyxml: add PYTHON_PYXML_VERSION variable to url address

Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

python-pyxml: rename the package file

Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

sysstat: bump to version 11.0.5

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

file: bump to version 5.23

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

stunnel: bump to version 5.18

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

php: security bump to version 5.6.10

Fixes:
CVE-2015-3414, CVE-2015-3415, CVE-2015-3416 (via bundled sqlite
upgrade).
CVE-2015-2325, CVE-2015-2326 (via bundled pcre upgrade).

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

openssl: security bump to version 1.0.2b

CVE-2015-1790 - PKCS7 crash with missing EnvelopedContent
CVE-2015-1789 - Exploitable out-of-bounds read in X509_cmp_time
CVE-2015-1788 - Malformed ECParameters causes infinite loop
CVE-2015-1792 - CMS verify infinite loop with unknown hash function
CVE-2015-1791 - Race condition handling NewSessionTicket

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

ffmpeg: fix dependencies

The ffmpeg binary does not require swscale, but ffplay needs it.

ffserver needs fork().

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

support: remove outdated note re. Python 3 support

Kconfiglib now runs as either Python 2 or Python 3.

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

package/pppd: install pon, poff scripts

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

package/linux-fusion: fix build for Linux >= 4.0

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

support: get prompts via official Kconfiglib APIs

These weren't available when gen-manual-lists.py was first written.

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

support: gen-manual-lists.py base directory simplification

Buildroot doesn't use $srctree from what I could tell.

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

support: update Kconfiglib to the latest version

Corresponds to a95f477 in https://github.com/ulfalizer/Kconfiglib.

Fixes:

 - Unset user values when loading a zero-byte .config. (5e54e2c)

 - Ignore indented .config assignments. (f8a7510)

 - Do not require $srctree to be set for non-kernel projects. (d56e9c1)

 - Allow digits in $-references to symbols. (ecacdd5)

 - Add Symbol.is_allnoconfig_y(). (deaa624)

 - Fix small output issue with Comments inside Choices.

Also adds Python 3 support and has a lot of internal cleanup and
optimization.

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

openvmtools: add OPENVMTOOLS_VERSION_MAJOR variable

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

gflags: define NO_THREADS if toolchain doesn't support threads

Fixes:
http://autobuild.buildroot.net/results/a42/a4288ed5050ce0d14e589745bf26c0a95acf645e/

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

package/lxc: bump version to 1.1.2

Signed-off-by: Hoang Tran <hoang.tran@vnoss.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

freetype: bump to version 2.6.0

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

valgrind: backport support for glibc 2.20+

Fixes:
http://autobuild.buildroot.net/results/352/352d087bf5daed8ce2c7e2d61db2e75db9b069b9/
http://autobuild.buildroot.net/results/a2c/a2c46d35b1c1da45f4d136f5a2be926695695bd9/

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

gflags: new package

Signed-off-by: Rahul Bedarkar <rahul.bedarkar@imgtec.com>
Reviewed-by: Abhimanyu Vishwakarma <abhimanyu.v@imgtec.com>
Reviewed-by: Marcin Nowakowski <marcin.nowakowski@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

wireless-regdb: bump to version 2015.06.05

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

whois: bump to version 5.2.9

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

infra/pkg-kconfig: require an non-empty KCONFIG_FILE

Currently, we only check that the variable is defined, which is not
enough since we really want it to be non-empty.

We however can't check it points to an existing file, because the
package might well not be extracted yet, and we may use an internal
defconfig. If that file does not eventually exist, there will be a
failure down the road at build time when we try to copy it...

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
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/python-can: new package

This patch provides controller area network support for Python.

Signed-off-by: Angelo Compagnucci <angelo.compagnucci@gmail.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

moarvm: new package

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
 [use buildroot's libatomic_ops, libuv and host-luajit; add
  dependencies from libuv]
[Thomas: indicate that the hash has been locally calculated.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

scripts/mkusers: allow users with no password value set

The following allows a user definition to specify that a created user
entry should not have a password value set. Original implementation
allowed a user definition to provide a password value of "=" (no quotes)
to generate a crypt-encoded empty string value. In some cases, it may be
desired to have no value specified for a user's password. By using a
value "-" for a password, no value will be set in the shadow value.

An example when this can be used is when logging into a terminal.
Logging into a session with an encoded empty password will prompt a user
to enter a password since it does not know the password is empty. If the
password field blank, a login session will not prompt for a password.

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

package/matchbox-panel: fix WiFi monitoring

matchbox-panel can build a simple applet to monitor the WiFi. For that,
it tries to see if libiwl is available; if so, WiFi monitoring is
enabled, otherwise it is disabled; there is not option to explicitly
enable or disable it.

Currently, matchbox-panel depends on the wireless-tools package.
However, that's not enough, since the library will only be installed if
explicitly enabled in the menuconfig *and* this is not a static build.

Change the dependency to be on the library, not the package.

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

package/matchbox-panel: add support for dnotify

Using dnotify, matchbox-panel can automatically reload its menu.
dnotify requires a kernel >=2.4, which we can consider to be
virtually for granted.

So, forcibly use dnotify.

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

package/matchbox-panel: use ACPI for the battery applet

It is possible to use ACPI to monitor the battery with the battery
mini-applet. If ACPI is not explicitly enabled, then APM is used if
available; if ACPI is not explcitly enabled and APM is not available,
then the battery applet is not built.

However, APM is not really current nowadays, and possibly missing for a
bunch of architecture...

What is a pity is that there is no way to explicitly enable or disable
the applet; it only relies on auto-detection... :-(

So, forcibly use ACPI on platforms that have it, that is x86 (32- and
64-bit) and AArch64.

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

package/matchbox: xlib_libXfixes is an optional dependency

matchbox can use xlib_libXfixes to completely and really hide the
cursor, so build-depend on it if enabled.

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

package/matchbox: xlib-libXcursor is an optional dependency

matchbox selects xlib-libXcursor, but does not build-depend on it.

But xlib-libXcursor is only an optional dependency.

Fix that by removing the select, and properly build-depend on it when
it is enabled.

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

package/matchbox: wrap long lines

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

package/matchbox-common: explicitly disable PDA folders when not selected

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

package/matchbox-keyboard: enforce disabling support for cairo

We usually explicitly enforce disabling optional support when it is
possible.

Do so for Cairo.

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

package/matchbox: drop useless dependencies

xlib-libXft is only needed when building a 'standalone' matchbox, which
means not using matchbox-lib.

But we *are* building matchbox-lib, and we do not support the standalone
mode (and probably won't, as even upstream says it is ugly, by lack of
theming).

Similarly, xlib-libXext is only used for its 'xsync' extension, for
which support is entirely commented-out in matchbox.

So, drop the dependency on xlib-libXft, and make it explicit we're not
building either standalone modes; drop xlib-libXext.

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

package/matchbox: enable support for the Xsession manager

Since this is a non-obvious dependency, just select the appropriate
library.

Also, add a comment to act as a separator between matchbox' options
and the other matchbox packages.

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

package/matchbox: make it a menu

We're about to add a few more options, so it now warrants a menu for
itself.

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

package/matchbox: forcibly disable support for compositing

Building with the experimental support for compositing is broken:

    /home/ymorin/dev/buildroot/O/host/usr/bin/arm-linux-gnueabihf-gcc
    -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
    -Os -o matchbox-window-manager main.o wm.o base_client.o main_client.o
    toolbar_client.o toolbar_client_alt.o dockbar_client.o dialog_client.o
    select_client.o desktop_client.o ewmh.o misc.o client_common.o keys.o
    list.o stack.o composite-engine.o session.o mbtheme.o xml.o
    -lmb -lX11
    -L/home/ymorin/dev/buildroot/O/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/lib
    -lXext -lXcomposite -lXdamage -lXfixes -lXrender
    -L/home/ymorin/dev/buildroot/O/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/lib
    -lX11 -lexpat
    -L/home/ymorin/dev/buildroot/O/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/lib
    -lXfixes
    composite-engine.o: In function `gaussian':
    /home/ymorin/dev/buildroot/O/build/matchbox-1.2/src/composite-engine.c:65:
    undefined reference to `exp'

That's because it forgets to link with -lm.

However, adding "MATCHBOX_CONF_ENV = -lm" does not solve the issue, as
it still does not link with -lm at all.

Furthermore, the package does not autoreconf, since it is missing the
gconf m4 macros, and Buildroot does not have a package for gconf, hence
we can not even patch Makefile.am and autoreconf.

Patching Makefile.in (in addition to .am) could be a solution, but
support for compositing is explicitly marked as being experimental.

So, just forcibly disable it altogether.

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

package/matchbox: forcibly disable gconf

Buildroot does not have a gconf package, so explicitly disable gconf
support (to avoid it ends up picking the host one).

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

package/matchbox-startup-monitor: invert dependency chain

Previously, matchbox-startup-monitor was forcibly enabled whenever
startup-notification was. Now that it has a prompt, it merely defaults
to 'y' in this case, leaving the user the possibility to disable it.
This is not very logical.

Instead, invert the logic, and select startup-notification from
matchbox-startup-monitor.

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

package/matchbox: finally add prompts to all sub-packages

We currently do not have that many packages without a prompt, and when
we do, they are limited to system-related packages that can't be
selected or are forcibly selected under various system-specific
conditions.

This is clearly not the case for the matchbox sub-packages, so just give
them a prompt, even when they are forcibly selected by the main matchbox
package, and can't be selected without it.

That will at least tell the user those packages exist and are to be
built.

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

package/matchbox-fakekey: add missing dependency

matchbox-fakekey build-depends on xlib_libXtst, but does not select it.

Fix that.

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

package/matchbox-common: fix dependencies

matchbox-common can be used by both matchbox-desktop and matchbox-panel,
so it needs to depend on either, not both.

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

package/matchbox: add missing dependencies to matchbox-lib

Most of the matchbox packages build-depend on matchbox-lib, but were not
selecting it from their Config.in.

Fix that.

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

package/matchbox: fix option prompts

Use the package names instead of some fancy human-readable strings. ;-)

Note: the matchbox package is really named matchbox-window-manager by
upstream. However:
  - it already is in a section of the menu title 'X window managers'
  - users are probably only looking for 'matchbox'
  - keeping just 'matchbox' keeps it in sync with the Kconfig variable,
    and allows us to not handle a legacy entry (see a previous commit)
  - matchbox-window-manager is a bit long. ;-)

So, in this case, the short 'matchbox' prompt is used instead of the
real upstream name.

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

package/matchbox-common: move one directory higher

We usually do not have a sub-directory for a family of related
packages, so move matchbox packages one directory higher, so they
all are in packages/

Also re-order packages alphabetically.

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

package/matchbox: get rid of intermediate non-package matchbox

Currently, the matchbox entry in the menu is a non-package entry, that
forcibly selects the matchbox-wm package, which is the real matchbox WM.

So, get rid of the current matchbox option, rename the existing
matchbox-wm to simply matchbox, so we have a real package from the
onset.

Since we're re-using the previous option for the WM package, there is no
need for an entry in the legacy menu.

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

package/matchbox: point to new home

The new home of the matchbox project is with the Yocto project.
The old home was already redirecting to it anyway, but better point
directly to the official location.

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

package/matchbox: remove superfluous conditional includes

Now that the matchbox package no longer forcibly add entries to the
PACKAGES list, and entirely relies on having proper packages, we can
safely include all matchbox sub-packages .mk files, and the pkg-infra
will take care of building only those that are enabled.

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

package/matchbox: make match-startup-monitor a real package

Currently, matchbox-startup-monitor is missing a Kconfig entry, but is
so far solely a build-time dependency of matchbox-panel.

So, make it a real package, and only build-depend on it when enabled.
Fix the condition to build-depend on it.

To keep the existing behaviour, default to 'y' when startup-notification
is enabled.

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

package/matchbox: make matchbox-keyboard a real package

Currently, matchbox-keyboard is treated as if it were an option of the
matchbox package.

But that's not true, since it is a real package all by itself, with its
own archive and its own buildsystem.

So, make it a real package.

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

package/matchbox: make matchbox-fakekey a real package

Currently, matchbox-fakekey is missing a Kconfig entry, but is so far
solely a build-time dependency of matchbox-keyboard.

So, make it a real package, and select it from matchbox-keyboard.

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

package/matchbox: make matchbox-common a real package

Currently, matchbox-common is treated as if it were an option of the
matchbox package.

But that's not true, since it is a real package all by itself, with its
own archive and its own buildsystem.

So, make it a real package.

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

package/matchbox: make matchbox-desktop a real package

Currently, matchbox-desktop is treated as if it were an option of the
matchbox package.

But that's not true, since it is a real package all by itself, with its
own archive and its own buildsystem.

So, make it a real package.

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

package/matchbox: make matchbox-panel a real package

Currently, matchbox-panel is treated as if it were an option of the
matchbox package.

But that's not true, since it is a real package all by itself, with its
own archive and its own buildsystem.

So, make it a real package.

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

package/matchbox: make matchbox-wm a real package

Similarly to what we just did for matchbox-lib, we make matchbox-wm a
real package.

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

package/matchbox: make matchbox-lib a real package

Currently, the matchbox package is using weird, legacy constructs to
build its different parts.

Notably, it adds matchbox-lib to the list of packages to build, even
though it is not a real package since it does not have a Kconfig entry.

Fix that:
  - add a Kconfig entry for matchbox-lib
  - select it from the top-level matchbox package

Note that matchbox-common already depends on matchbox-lib.

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

package/matchbox-wm: drop useless commented options

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

linux: bump default to version 4.0.5

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

linux-headers: bump 3.{10, 14}.x and 4.0.x series

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

stunnel: add STUNNEL_VERSION_MAJOR variable

Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

lbreakout2: add LBREAKOUT2_VERSION_MAJOR variable

Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

arch: tidy up binary formats config

Instead of (black)listing architectures when deciding the binary format,
we can enable the ELF format only when using an MMU and FLAT only when
we're not. This mimics the logic in the Linux kernel for user binaries
support.

For FDPIC, we introduce a Kconfig option to enable its selection, and
have blackfin select it.

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

arch: tidy up mmu config

Instead of blacklisting which architectures support MMUs (mandatorily
or optionally), introduce two Kconfig options that are selected by each
architecture in each case.

This simplifies the logic in BR2_USE_MMU.

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

package/eudev: bump version to 3.1.1

Removed patch applied upstream:
https://github.com/gentoo/eudev/commit/d2aeb8b63dfcb15161c4dd3e80491c844c3f0581#diff-0a5e92f58aded4146365ea36ee8b3c21

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

package/opencv: cleanup spaces at ends of line

Nothing more than removing white spaces before '\' characters.
Of course, 'git diff -w' shows no change at all.

Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>