package/crudini: bump to version 0.9.3

This release hasn't been published to pypi, so we switch to the Github
release archive instead. This new version is Python 3.x compatible.

Signed-off-by: Titouan Christophe <titouan.christophe@railnova.eu>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/lapack: avoid selecting lapack and clapack

lapack and clapack generate the same libraries liblapack.so and
libblas.so. So those two packages can't be selected at the same time.

This is a temporary fix waiting for a solution[2].

So:
- add !BR2_PACKAGE_CLAPACK to lapack/Config.in.

[1]: http://autobuild.buildroot.org/results/375/375078ed8f965ecf92eb9674bd071a518c3ef894//
[2]: http://lists.busybox.net/pipermail/buildroot/2019-August/255894.html

Signed-off-by: Alexandre PAYEN <alexandre.payen@smile.fr>
Cc: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Cc: Romain Naour <romain.naour@smile.fr>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Reviewed-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

system: don't attempt swapon/swapoff in inittab if not available

The default inittab files added by busybox and sysvinit run 'swapon -a'
during init and 'swapoff -a' during shutdown.

But, the swapon/swapoff programs are not guaranteed to be
available. For the busybox versions, it is steered by
CONFIG_SWAPON/CONFIG_SWAPOFF. For the util-linux versions, it is steered by
BR2_PACKAGE_UTIL_LINUX_BINARIES.

In a case where swapon/swapoff is not available but the inittab tries to
execute them, the boot log would be polluted by error messages like:

    swapon: not found

Avoid this by commenting out the swapon/swapoff lines if the swapon/swapoff
binaries are not available.

Signed-off-by: Thomas De Schampheleire <thomas.de_schampheleire@nokia.com>
[Peter: test with -x]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/python-iniparse: bump to version 0.5

This version brings Python 3.x support, so we make the package
available with Python 3.x. Also, iniparse now requires six, which has
been detected thanks to the runtime tests.

Signed-off-by: Titouan Christophe <titouan.christophe@railnova.eu>
[Thomas: drop depends on python || python3]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/elixir: new package

elixir is required to build RabbitMQ starting from v3.7

Signed-off-by: Frank Vanbever <frank.vanbever@essensium.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/erlang: bump to version 22.2

The issue fixed by 0003-Link-with-LDLIBS-instead-of-LIBS-for-DED.patch (ERL-529)
has been fixed since OTP20.3 (a5cbcbdb85) and is no longer required.

OTP provides wrapper scripts (otp_build and configure) which perform tasks
previously handled directly by autotools i.e. autoreconf.

Signed-off-by: Frank Vanbever <frank.vanbever@essensium.com>
[Thomas:
 - drop the save_bootstrap step, which is not needed
 - properly run the autoconf step for both the host and target
   variants]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/libpjsip: add portaudio optional dependency

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/docker-containerd: security bump to 1.2.12

Fixes the following security issues:

- CVE-2019-19921: runc volume mount race condition with shared mounts

- CVE-2019-16884: runc through 1.0.0-rc8, as used in Docker through
  19.03.2-ce and other products, allows AppArmor restriction bypass because
  libcontainer/rootfs_linux.go incorrectly checks mount targets, and thus a
  malicious Docker image can mount over a /proc directory.

For details, see the announcement:
https://github.com/containerd/containerd/releases/tag/v1.2.12

containerd is now a separate CNCF sponsored project, and is no longer
explicitly associated with docker/moby.

Signed-off-by: Christian Stewart <christian@paral.in>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

board/freescale/common/imx: add support for i.MX8

Signed-off-by: Maeva Manuel <maeva.manuel@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/freescale-imx/imx-sc-firmware: bump to version 1.1 to support i.MX8

Signed-off-by: Maeva Manuel <maeva.manuel@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/freescale-imx: add option for i.MX8

The i.MX8 family has both the i.MX8QP and i.MX8QM.

See [0] for an overall presentation of all i.MX8 series, and [1] for
i.MX8 itself, which includes i.MX8QM (i.MX8 Quad Max) and i.MX8QP
(i.MX8 Quad Plus), which both have a quad Cortex-A53, the Plus has a
single Cortex-A72 and the Max has a dual Cortex-A72.

[0] https://www.nxp.com/products/processors-and-microcontrollers/arm-processors/i.mx-applications-processors/i.mx-8-processors:IMX8-SERIES
[1] https://www.nxp.com/products/processors-and-microcontrollers/arm-processors/i.mx-applications-processors/i.mx-8-processors/i.mx-8-family-arm-cortex-a53-cortex-a72-virtualization-vision-3d-graphics-4k-video:i.MX8

Signed-off-by: Maeva Manuel <maeva.manuel@nxp.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

linux: bump CIP RT kernel to version 4.19.98-cip19-rt7

This patch bumps Linux CIP RT version to 4.19.98-cip19-rt7.

Signed-off-by: Angelo Compagnucci <angelo@amarulasolutions.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux: bump CIP kernel to version 4.19.98-cip19

This patch bumps Linux CIP version to 4.19.98-cip19.

Signed-off-by: Angelo Compagnucci <angelo@amarulasolutions.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/ntfs-3g: add upstream security fix for CVE-2019-9755

Fixes CVE-2019-9755: An integer underflow issue exists in ntfs-3g 2017.3.23.
A local attacker could potentially exploit this by running /bin/ntfs-3g with
specially crafted arguments from a specially crafted directory to cause a
heap buffer overflow, resulting in a crash or the ability to execute
arbitrary code.  In installations where /bin/ntfs-3g is a setuid-root
binary, this could lead to a local escalation of privileges.

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

package/libxml2: add upstream security fix for CVE-2020-7595

Fixes CVE-2020-7595: xmlStringLenDecodeEntities in parser.c in libxml2
2.9.10 has an infinite loop in a certain end-of-file situation.

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

package/vorbis-tools: add upstream security fixes for CVE-2014-96{38, 39, 40}

Fixes the following security vulnerabilities:

- CVE-2014-9638: oggenc in vorbis-tools 1.4.0 allows remote attackers to
  cause a denial of service (divide-by-zero error and crash) via a WAV file
  with the number of channels set to zero.

- CVE-2014-9639: Integer overflow in oggenc in vorbis-tools 1.4.0 allows
  remote attackers to cause a denial of service (crash) via a crafted number
  of channels in a WAV file, which triggers an out-of-bounds memory access.

- CVE-2014-9640: oggenc/oggenc.c in vorbis-tools 1.4.0 allows remote
  attackers to cause a denial of service (out-of-bounds read) via a crafted
  raw file.

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

system: also create a lib64 -> lib symlink on mips64n32

In gcc 5.1.0, a change was introduced which causes internal search paths
inside the sysroot to be relative to 'lib64' rather than 'lib'. See [1] [2]
and [3].

For example for dtc:

     LD convert-dtsv0
    /opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/mips64-octeon-linux-gnu/7.3.0/../../../../mips64-octeon-linux-gnu/bin/ld: cannot find crt1.o: No such file or directory
    /opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/mips64-octeon-linux-gnu/7.3.0/../../../../mips64-octeon-linux-gnu/bin/ld: cannot find crti.o: No such file or directory
    collect2: error: ld returned 1 exit status
    make[1]: *** [Makefile:236: convert-dtsv0] Error 1
    make[1]: *** Waiting for unfinished jobs....
    make[1]: Leaving directory '/opt/buildroot/output/build/dtc-1.4.7'
    make: *** [package/pkg-generic.mk:241: /opt/buildroot/output/build/dtc-1.4.7/.stamp_built] Error 2

In this case, crt1.o was searched for in following locations:

    16073 access("/opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/mips64-octeon-linux-gnu/7.3.0/n32/octeon3/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/n32/octeon3/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/mips64-octeon-linux-gnu/7.3.0/../../../../mips64-octeon-linux-gnu/lib/mips64-octeon-linux-gnu/7.3.0/n32/octeon3/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/mips64-octeon-linux-gnu/7.3.0/../../../../mips64-octeon-linux-gnu/lib/../lib32-fp/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/mips64-buildroot-linux-gnu/sysroot/lib64/mips64-octeon-linux-gnu/7.3.0/n32/octeon3/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/mips64-buildroot-linux-gnu/sysroot/lib64/../lib32-fp/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/mips64-buildroot-linux-gnu/sysroot/usr/lib64/mips64-octeon-linux-gnu/7.3.0/n32/octeon3/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/mips64-buildroot-linux-gnu/sysroot/usr/lib64/../lib32-fp/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/mips64-octeon-linux-gnu/7.3.0/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/mips64-octeon-linux-gnu/7.3.0/../../../../mips64-octeon-linux-gnu/lib/mips64-octeon-linux-gnu/7.3.0/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/opt/ext-toolchain/bin/../lib/gcc/mips64-octeon-linux-gnu/7.3.0/../../../../mips64-octeon-linux-gnu/lib/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/mips64-buildroot-linux-gnu/sysroot/lib64/mips64-octeon-linux-gnu/7.3.0/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/mips64-buildroot-linux-gnu/sysroot/lib64/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/mips64-buildroot-linux-gnu/sysroot/usr/lib64/mips64-octeon-linux-gnu/7.3.0/crt1.o", R_OK) = -1 ENOENT (No such file or directory)
    16073 access("/opt/buildroot/output/host/mips64-buildroot-linux-gnu/sysroot/usr/lib64/crt1.o", R_OK) = -1 ENOENT (No such file or directory)

As can be seen above, all attempted paths contain 'lib64' as base,
instead of 'lib' or 'lib32', e.g.

.../sysroot/lib64/../lib32-fp/crt1.o
.../sysroot/lib64/crt1.o

This problem was detected on a gcc 7.x toolchain provided by Marvell as part
of their Octeon SDK. For this toolchain, here are the values of the paths
as detected by the Buildroot toolchain logic, for two different Octeon
processors:

- octeon2 (soft-float) (-mabi=n32 -march=octeon2):
SYSROOT_DIR=/opt/buildroot/output/host/opt/ext-toolchain/mips64-octeon-linux-gnu/sys-root/;
ARCH_SYSROOT_DIR=/opt/buildroot/output/host/opt/ext-toolchain/mips64-octeon-linux-gnu/sys-root/;
ARCH_SUBDIR=;
ARCH_LIB_DIR=lib32/octeon2;
SUPPORT_LIB_DIR=/opt/buildroot/output/host/opt/ext-toolchain/mips64-octeon-linux-gnu/lib32/octeon2/

- octeon3 (hard-float) (-mabi=n32 -march=octeon3):
SYSROOT_DIR=/opt/buildroot/output/host/opt/ext-toolchain/mips64-octeon-linux-gnu/sys-root/;
ARCH_SYSROOT_DIR=/opt/buildroot/output/host/opt/ext-toolchain/mips64-octeon-linux-gnu/sys-root/;
ARCH_SUBDIR=;
ARCH_LIB_DIR=lib32-fp;
SUPPORT_LIB_DIR=/opt/buildroot/output/host/opt/ext-toolchain/mips64-octeon-linux-gnu/lib32-fp/

For both cases (MIPS64n32) Buildroot created a symlink 'lib32->lib', from
SYSTEM_LIB_SYMLINK in system/system.mk. Additionally, the function
create_lib_symlinks in
toolchain/toolchain-external/pkg-toolchain-external.mk will use ARCH_LIB_DIR
and create an additional link $(ARCH_LIB_DIR)->lib.
For the Octeon3 case this thus results in the following symlinks (where the
'lib32' one is normally not needed):

    lib32 -> lib/
    lib32-fp -> lib/

Since the toolchain is searching based on a 'lib64' component, it will fail
to find its internal paths.

To solve the problem, we need to create an additional symlink 'lib64':

    lib64 -> lib/

[1] https://github.com/gcc-mirror/gcc/commit/257ccd463a4f25495b252a3bdf989d057c54678d
[2] https://gcc.gnu.org/ml/gcc-patches/2014-10/msg03377.html
[3] https://gcc.gnu.org/ml/gcc-patches/2014-11/msg00539.html

Signed-off-by: Thomas De Schampheleire <thomas.de_schampheleire@nokia.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/freescale-imx/imx-gpu-g2d: bump to version 6.4.0.p1.0

Make the package available on AArch64 now that it is supported, and
add hashes for the license files.

Signed-off-by: Refik Tuzakli <tuzakli.refik@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/rpm: bump to version 4.15.1

- Drop both patches (already in version)
- Add libgcrypt optional dependency (added in version 4.15.1 with
  https://github.com/rpm-software-management/rpm/commit/037106ecc899bad6d6e6f9d95768699542b871ea)
- Add openmp support (added in version 4.15.1 with
  https://github.com/rpm-software-management/rpm/commit/464d21dc8c176222c6586e2ee503fec6207f0d29)

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

package/imagemagick: fix configure error with --disable-opencl in setups without libltdl

Fixes:
http://autobuild.buildroot.net/results/d9a/d9a84b642357f758c3f84270fb9a109abd7e2684/

configure.ac contains a test using $ax_cv_check_cl_libcl:

if test "$build_modules" != 'no' || test "X$ax_cv_check_cl_libcl" != Xno; then
  AC_MSG_RESULT([-------------------------------------------------------------])
  AC_MSG_CHECKING([for libltdl])

But ax_cv_check_cl_libcl is only assigned a value (yes/no) if
--disable-opencl is NOT passed, as the assignment logic is inside a
conditional:

AC_ARG_ENABLE([opencl],
    [AC_HELP_STRING([--disable-opencl],
                    [do not use OpenCL])],
    [disable_opencl=$enableval],
    [disable_opencl='yes'])

if test "$disable_opencl" = 'yes'; then
  ..
  AC_CACHE_CHECK([for OpenCL library], [ax_cv_check_cl_libcl],

So configure errors out if --disable-opencl is passed on setups where
libltdl isn't available:

checking if libltdl package is complete... no
configure: error: in `/home/naourr/work/instance-0/output-1/build/imagemagick-7.0.8-59':
configure: error: libltdl is required for modules and OpenCL builds

As a workaround, explictly set ax_cv_check_cl_libcl=no to skip this
conditional.

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

package/bluez-alsa: fix location of alsa config files

Since alsa-lib version 1.1.7 [1] the location for add-on config files
has changed.

In fact, the path for the alsa add-on config files has never been
correct set in the package (it should have been
`/usr/share/alsa/alsa.conf.d`).

With alsa-lib version 1.1.7 or later the correct path is
`/etc/alsa/conf.d`.

[1] https://github.com/alsa-project/alsa-lib/commit/93e03bdc2a3dcd5d12516f5de78e14d88a32ff2c

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

package/xen: use host-python3 to build

As noted in https://wiki.xenproject.org/wiki/Xen_Project_4.13_Release_Notes

- Xen 4.13 is now fully Py3 compatible

So swith to that now that python 2.x is EOL.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
[Peter: extend commit message]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/sconeserver: add all blue5_utils dependencies

Commit 61a813339af43024ec05e749a05886fff1bf2a39 forgot to add all the
bluez5_utils dependencies

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

package/cwiid: add all blue5_utils dependencies

Commit 61a813339af43024ec05e749a05886fff1bf2a39 forgot to add all the
bluez5_utils dependencies

Fixes:
 - http://autobuild.buildroot.org/results/a125a27814aba007b06fabe3db2d402684d173db

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

package/ussp-push: add all blue5_utils dependencies

Commit 61a813339af43024ec05e749a05886fff1bf2a39 forgot to add all the
bluez5_utils dependencies

Fixes:
 - http://autobuild.buildroot.org/results/9a1600ab6b6c8587e8e254e1a9653f5316d3c4fd

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

package/whois: add host-pkgconf mandatory dependency

pkg-config is used to check if libidn2 is available since version 5.2.18
and
https://github.com/rfc1036/whois/commit/e20b8bbab3687aa26294bc3c443b4a43292b7636

This can lead to a build failure if a broken libidn2.pc is found on host
so add a mandatory host-pkgconf dependency and drop our manual handling
of HAVE_LIBIDN2 option (which does not exist anymore)

Fixes:
 - http://autobuild.buildroot.org/results/6cd0dd779d72b7162943f0f7dbb4e7d258b463eb

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

package/mariadb: security bump to 10.3.22

Release notes:
https://mariadb.com/kb/en/library/mariadb-10322-release-notes/

Changelog:
https://mariadb.com/kb/en/library/mariadb-10322-changelog/

Fixes the following security vulnerability (10.3.22):
CVE-2020-2574 - Vulnerability in the MySQL Client product of Oracle MySQL
(component: C API). Supported versions that are affected are 5.6.46 and
prior, 5.7.28 and prior and 8.0.18 and prior. Difficult to exploit
vulnerability allows unauthenticated attacker with network access via
multiple protocols to compromise MySQL Client. Successful attacks of this
vulnerability can result in unauthorized ability to cause a hang or
frequently repeatable crash (complete DOS) of MySQL Client.

Fixes the following security vulnerabilities (10.3.19):
CVE-2019-2974 - Vulnerability in the MySQL Server product of Oracle MySQL
(component: Server: Optimizer). Supported versions that are affected are
5.6.45 and prior, 5.7.27 and prior and 8.0.17 and prior. Easily exploitable
vulnerability allows low 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-2019-2938 - Vulnerability in the MySQL Server product of Oracle MySQL
(component: InnoDB). Supported versions that are affected are 5.7.27 and
prior and 8.0.17 and prior. Difficult to exploit 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.

Patch 0002-fix-build-error-with-newer-cmake.patch has been removed as it
has been applied upstream.

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/mesa3d: add v3d driver support

Support for v3d was added in mainline 4.18, and requires a few options
to be set in the kernel, so we list that in the help text.

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
[yann.morin.1998@free.fr: tweak the help text]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/gcc: remove gcc 5.5 or1k fork

Remove the old gcc 5.5 fork for or1k architecture
that start to fail to build with recent version
of Binutils >= 2.32 with the following error:

host-gcc-final-or1k-musl-5.4.0-20170218/build/./gcc/crtbeginS.o: addend should be zero for plt relocations
host/or1k-buildroot-linux-uclibc/bin/ld: final link failed: bad value

https://gitlab.com/kubu93/buildroot/-/jobs/391938988

Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/orc: bump to version 0.4.31

Switch to meson buildsystem and so drop:
 - !BR2_BINFMT_FLAT dependency
 - hooks (not needed thanks to orc-test and tools options)

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

configs/bananapro: bump Linux to 5.4.7

Signed-off-by: Bartosz Bilas <b.bilas@grinn-global.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

board/lemaker/bananapro: fix wlan module

Restore a patch which disables 00B IRQ for the built-in wlan module which
allows working wlan module again.  It turned out that it shouldn't have been
deleted because the bug still exists in the mainline linux kernel, so keep
this patch the same as armbian[1] to have wifi connection working correctly.

[1] https://github.com/armbian/build/blob/master/patch/kernel/sunxi-dev/ARM-dts-sun7i-Disable-OOB-IRQ-for-brcm-wifi-on-Cubietruck-and-Banana-Pro.patch

Signed-off-by: Bartosz Bilas <b.bilas@grinn-global.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/python-pydantic: new package

Data validation and settings management using python 3.6
type hinting.

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

configs/stm32mp157c_dk2_defconfig: bump Linux to 5.4.7

Signed-off-by: Bartosz Bilas <b.bilas@grinn-global.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/lvm2: add upstream patch to fix udev unsupported option warning

Since systemd/udev version 243, there's a new message printed if unsupported
OPTIONS value is used:

  Invalid value for OPTIONS key, ignoring: 'event_timeout=180'

Add a patch to drop this invalid value.

Signed-off-by: Pascal de Bruijn <p.debruijn@unilogic.nl>
[Peter: extend commit message]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/lvm2: bump version to 2.02.185

Signed-off-by: Pascal de Bruijn <p.debruijn@unilogic.nl>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/php-*: remove php dependency for php modules

php modules can't be selected if the php package is not selected, see [1].
So adding the dependency is not necessary.
Also we don't add such dependencies for python modules.

[1] https://git.buildroot.net/buildroot/tree/package/Config.in?h=2019.11#n798

Signed-off-by: Romain Naour <romain.naour@smile.fr>
Cc: Nicolas Carrier <nicolas.carrier@orolia.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

utils/check-package: ignore ACLOCAL_PATH

The aclocal program is provided by the automake package, so it makes
sense to define aclocal-related variables in automake.mk.

Add an exception to check-package to ignore that variable.

Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/weston: add pipewire support

Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/grep: fix egrep/fgrep shebang

egrep/fgrep are wrapper scripts, calling the grep binary with the correct
arguments.

The shell wrappers use the value of SHELL at build time as the shebang value
in these wrapper scripts, which in Buildroot points to /bin/bash.

The target may not have bash available, causing runtime errors.

As a fix, add a post-install hook to change this to /bin/sh.

If the target does not have /bin/sh, simply remove the aliases.

Signed-off-by: Angelo Compagnucci <angelo@amarulasolutions.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/grep: install to /bin to fix busybox aliasing

When the grep package is selected, it should be installed at the same exact
location where busybox installs it too, this way the grep/egrep/fgrep
executables will end up overwriting the busybox provided ones.

Signed-off-by: Angelo Compagnucci <angelo@amarulasolutions.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/mariadb: do not pass --user=mysql to mysql_install_db

We currently install the default database by passing --user=mysql to the
install script. With the upcoming bump to the 10.4 series, this does not
work as intended. An error occurs because of missing PAM modules. We work
around this now by creating the default db as root and calling chown to
change the files to user mysql.

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/mariadb: explicitly define logging directory to mysqld

By default, mariadb creates logging files under the data directory.
This patch updates the startup scripts to log under /var/log/mysql.

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/mariadb: rename MYSQL_PID variable in sysv startup script

For consistency with the to-be-added MYSQL_LOGFILE variable.

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/mariadb: fix potential error message in sysv startup script

In order to check if the initial database needs created, the startup
script calls ls -1 $MYSQL_LIB | wc -l to check the number of files in
the directory. If the directory does not exist, an error is printed.
We fix this by redirecting stderr to /dev/null for the ls call.

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/mariadb: remove mariadb_config from target

We already remove mysql_config from the target since it's only useful in
staging. The same is true for mariadb_config. Thus, we remove it from the
target as well.

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/mariadb: fix permissions of data dir when using systemd

mysql_install_db is currently called in the systemd unit without
--user=mysql that the sysv script uses. This will generate the initial
database files with root permissions. However, mysqld runs as user mysql
so this will cause problems. We fix this by calling chown instead of
passing the user parameter because an upcoming version bump will fail when
ran this way.

Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package: use variables defined by autoconf.mk

autoconf.mk defines AUTOCONF and AUTOHEADER variables, use them in packages
using autoconf.
This is a refactoring which shouldn't impact the final behavior.

Signed-off-by: Nicolas Carrier <nicolas.carrier@orolia.com>
Reviewed-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

remove the hook for creating the m4 directory

Now that we do not override the automagic handling of include
directories witbh aclocal, the missing m4 directories will be
automatically created by aclocal itself.

So we can drop of hooks.

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
[yann.morin.1998@free.fr: slight rewording in the commit log]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

autotools: do not overwrite first include path

The first include path is special in aclocal. For example it is the path
for the --install option. Also, the first include is treated in a
special way if it doesn't exists. This might be the case if there is the
following construct:

  configure.ac: AC_CONFIG_MACRO_DIR([m4])
  Makefile.am: ACLOCAL_AMFLAGS="-I m4"

If the package doesn't have local macros, the m4/ directory might not
exist. aclocal will then just issue a warning instead of aborting the
execution with a fatal error. See discussion here:
  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=565663

Don't use the "-I" option in aclocal. Instead use ACLOCAL_PATH to pass
the system-wide include dirs.

As a side effect this should fix the use of $(ACLOCAL) alone. Up until
now, $(ACLOCAL) didn't include the ACLOCAL_HOST_DIR system include path.

autoreconf will pass the "-I" options to every tool it runs, of which
aclocal, which, as seen above, we don't want. So move the argument down
to each individual tool, except for aclocal.

Signed-off-by: Michael Walle <michael@walle.cc>
Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
[yann.morin.1998@free.fr: slight rewording of the commit log]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/jack2: drop dependency on host-python

host-python was listed in jack2 dependencies, but is actually
not needed. Also tested with an experimental python3 based
Waf package infrastructure, so we can exclude that jack2
"accidentally" uses the host-python2 pulled in by Waf.

Signed-off-by: Titouan Christophe <titouan.christophe@railnova.eu>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/qt5x11extras: update hash source URL

Be more specific for the hash source URL and change from
summary page (.mirrorlist) to direct sha256 download page
(.sha256).

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/qt5serialport: update hash source URL

Be more specific for the hash source URL and change from
summary page (.mirrorlist) to direct sha256 download page
(.sha256).

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/can-utils: bump to version 2020.02.0

Remove the upstreamed patch and change the license information, as the
GPL-2.0+ files have been relicensed as GPL-2.0 since
https://github.com/linux-can/can-utils/commit/59f92965b90efc00b0cb63e3d4edfefd9b24326f

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
[Peter: clarify licensing change]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/network-manager-openvpn: new package

This package provides an OpenVPN plugin for network manager.

Signed-off-by: Alex Michel <alex.michel@wiedemann-group.com>
[Peter: add Config.in, DEVELOPERS entry]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/freescale-imx/imx-gpu-viv: bump to version 6.4.0.p1.0

While at it, we add hashes for the licenses files.

Signed-off-by: Refik Tuzakli <tuzakli.refik@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/python-zc-lockfile: new package

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/xen: bump to version 4.13.0

Bump to the latest version of Xen and remove the no longer required
patch.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/python-redis: Bump to version 3.4.1

Signed-off-by: Grzegorz Blach <grzegorz@blach.pl>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/python-aioblescan: Bump to version 0.2.6

Signed-off-by: Grzegorz Blach <grzegorz@blach.pl>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/weston-imx: add license file hash

Adopt the new trending spacing while at it.

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/mii-diag: add license file hash

The indentation of the other lines are adopted for consistency.

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/kodi-screensaver-pyro: add license file hash

Adopt the new trending spacing while at it.

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/kodi-screensaver-biogenesis: add license file hash

Adopt the new trending spacing while at it.

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/kodi-screensaver-matrixtrails: add license file hash

Adopt the new trending spacing while at it.

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/kodi-audioencoder-vorbis: add license file hash

Adopt the new trending spacing while at it.

Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/freescale-imx/firmware-imx: install video encode decode firwmare to target for imx8x

Signed-off-by: Refik Tuzakli <tuzakli.refik@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/libcodec2: bump version to 0.9.2

Removed patch applied upstream:
https://github.com/drowe67/codec2/commit/375d4c8d782c3bcadbb2240efed39b64ffd28430

Switched _SITE to github as suggested by
http://www.rowetel.com/?page_id=452

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

package/freescale-imx: BR2_PACKAGE_FREESCALE_IMX_HAS_VIV_GPU is also true on i.MX8X

This will allow to select the imx-gpu-viv package on i.MX8X.

Signed-off-by: Refik Tuzakli <tuzakli.refik@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/systemd: use interface name for networkd config file

By using the interface in the filename for the networkd config file,
we have a clear association between the config file and the interface
it applies to.

This is beneficical for systems that have multiple interfaces.

Signed-off-by: Pascal de Bruijn <p.debruijn@unilogic.nl>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gqrx: suppress dependency to not BR2_PACKAGE_PYTHON3

Since Gnuradio 3.8 support python3 dependency to only python2 is no more
mandatory

Signed-off-by: Gwenhael Goavec-Merou <gwenhael.goavec-merou@trabucayre.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gr-osmosdr: add python3 support

Gnuradio 3.8 support python2 and python3 and since gr-osmosdr support both two,
adapts dependencies

Signed-off-by: Gwenhael Goavec-Merou <gwenhael.goavec-merou@trabucayre.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gnuradio: add python3 support

gnuradio 3.8.0.0 support both python2 and python3.

Signed-off-by: Gwenhael Goavec-Merou <gwenhael.goavec-merou@trabucayre.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gnuradio: bump to version 3.8.0.0

Since 2019/08/09 a new version of gnuradio is available.
This patch bump to this version with some modifications:
- suppress the applied patch 0001-socket_pdu_impl.cc-fix-build-with-boost-1.70.0.patch
- backport patch to have boost unit_test_framework optional instead of mandatory:
  https://github.com/gnuradio/gnuradio/pull/2927
- backport patch to fix neon version detection:
  https://github.com/gnuradio/volk/pull/319
- add boost atomic (used by pmt), log4cpp (now mandatory) and gmp as dependencies
- suppress the no more available BR2_PACKAGE_GNURADIO_LOG, since log4cpp is mandatory. Since this
  option is now always true no need to add an entry in legacy.
- suppress all workaround for neon. Now volk is able to detect correctly neon
  version.
- add explictly path to python interpreter

Signed-off-by: Gwenhael Goavec-Merou <gwenhael.goavec-merou@trabucayre.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gr-osmosdr: bump version to support Gnuradio 3.8

Gnuradio 3.8 has made some API changes. The current gr-osmosdr release
does not support these changes, so move to a more recent gr-osmosdr
commit, which has Gnuradio 3.8 support.

Signed-off-by: Gwenhael Goavec-Merou <gwenhael.goavec-merou@trabucayre.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gqrx: backport/adapt patch to suppport Gnuradio 3.8

Gnuradio 3.8 has broken some APIs, and the current gqrx release does
not support these API changes.

So we add compatibility backport patches from the following
pull-request: https://github.com/csete/gqrx/pull/705

Signed-off-by: Gwenhael Goavec-Merou <gwenhael.goavec-merou@trabucayre.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/bluez_utils: drop package

BlueZ 4.x is deprecated since a long time (BlueZ 5.x has been released
in 2012) so drop it.

For cwiid, sconeserver and ussp-push, replace bluez_utils by
bluez5_utils. All other packages already support bluez5_utils so just
drop bluez_utils.

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

{linux, linux-headers}: bump 4.{4, 9, 14, 19}.x / 5.4.x series

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

package/haproxy: bump to version 2.1.2

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/mutt: fix build without makedoc

Cross-compilation of mutt is broken since bump to version 1.13.3 with
commit d42603b3872a0eed5a5a3dda13843fb5835f15a5

Indeed, since upstream commit 9bacbaa9, even with --disable-doc, mutt
tries to generate and install documentation files.
Build fails because mutt wants to generate muttrc.man by building (and
then running) a makedoc binary with the target cross-compiler (which
will obviously fail).

See https://gitlab.com/muttmua/mutt/issues/202

To fix this issue, retrieve two upstream patches.

Fixes:
 - http://autobuild.buildroot.org/results/2b01b7e65c0dc0e1191ba48e02e31503958d82fe

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

support/testing: new opkg test case

- Validates an archive can be installed and removed
- Builds an archives that uses postinst and prerm scripts

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

pru-software-support: bump to latest version 5.6.0

Update to the latest version of the PRU software support.

Formatting for the license file has been updated compared with the
previous package version 5.1.0 but the licenses used appear the same.

Signed-off-by: Ash Charles <ashcharles@gmail.com>
[yann.morin.1998@free.fr: fold the license hash patch into this patch]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

ti-cgt-pru: bump to latest version 2.3.3

Update to the latest version of the PRU code generation toolchain [1].
The license files differ only in whitespace and the version numbers
references compared with the previous release.

[1] https://release-monitoring.org/project/59398/

Signed-off-by: Ash Charles <ashcharles@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/libinput: bump version to 1.15.1

For details see [1].

[1] https://lists.freedesktop.org/archives/wayland-devel/2020-February/041186.html

Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/smartmontools: add libselinux optional dependency

libselinux is an optional dependency since a very long time (2010) and
https://github.com/mirror/smartmontools/commit/ab807ee43bd4821f5e0e010de08a746d88506b17

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

package/rocksdb: new package

The RocksDB library provides a persistent key value store. Keys and
values are arbitrary byte arrays. The keys are ordered within the key
value store according to a user-specified comparator function.

The library is maintained by the Facebook Database Engineering Team, and
is based on LevelDB, by Sanjay Ghemawat and Jeff Dean at Google.

http://rocksdb.org

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

boot/uboot: needs make >= 4.0

Starting with 2020.01, uboot started using the 'undefine' make
directive, which was only introduced with make 4.0.

In the general case, we do not have a way to know if the uboot
selected by the user is older or later than 2020.01, so we have
no way to know before hand if make >= 4.0 is needed or not. As
such, we have no other option than to always require it.

So, use the existing $(BR2_MAKE_HOST_DEPENDENCY) and $(BR2_MAKE),
both of each will ensure that we do use a make that is at least 4.0.

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

package/uboot-tools: needs make >= 4.0

Starting with 2020.01, uboot started using the 'undefine' make
directive, which was only introduced with make 4.0.

So, use the existing $(BR2_MAKE_HOST_DEPENDENCY) and $(BR2_MAKE),
both of each will ensure that we do use a make that is at least 4.0.

Fixes:
    http://autobuild.buildroot.org/results/c97/c976ed5eb7760cba192d22b3f1e7460596fd82dd/

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

package/python-validators: bump version to 0.14.2

Fixes an issue with domain name validation.

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

package/gensio: bump to version 1.4.1

Removed patches that are now upstream.

Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
[Peter: drop autoreconf]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/libmpdclient: bump to version 2.18

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

package/iproute2: bump to version v5.5.0

Signed-off-by: Petr Vorel <petr.vorel@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/ncmpc: bump to version 0.37

Bump gcc requirement to gcc >= 7 as C++17 is mandatory since
https://github.com/MusicPlayerDaemon/ncmpc/commit/85f136dd6b6a8d044b19cbef818ee360292cc06e

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

package/ptm2human: bump version to d0b8b6be98

Also drop upstreamed patch.

Signed-off-by: Julien Olivain <juju@cotds.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/minizip: bump to version 2.9.1

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

package/whois: bump to version 5.5.5

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

package/fail2ban: add conditional python-systemd dependancy

fail2ban needs python-systemd for its systemd backend to
be able to read logs from systemd/journald

Signed-off-by: Pascal de Bruijn <p.debruijn@unilogic.nl>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/jack2: bump to version 1.9.14

Amongst other things, the jack2 waf script has been made
compatible with Python3 in this latest release, which makes
it possible to be built on py3

See also upstream release notes:
- https://github.com/jackaudio/jack2/releases/tag/v1.9.13
- https://github.com/jackaudio/jack2/releases/tag/v1.9.14

Signed-off-by: Titouan Christophe <titouan.christophe@railnova.eu>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/fail2ban: add python3 support

The fail2ban codebase is still native python2, but 2to3 is supported
upstream.

Signed-off-by: Pascal de Bruijn <p.debruijn@unilogic.nl>
[Peter: ensure host-python3 is available]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/kodi: needs extra encodings in libiconv

The Kodi build system also enables extra encodings:
https://github.com/xbmc/xbmc/blob/Leia/tools/depends/target/libiconv/Makefile#L13

Without them addons fail to install:
https://github.com/garbear/kodi-steamlink/issues/34#issuecomment-579392269

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>