elfutils: fix an incorrect patch for 0.161

The original uclibc-support.patch is not compatible with elfutils-0.161.
It should be corrected through adjusting context.

So regenerate a new patch for elfutils-0.161, rename the patch for
elfutils-0.148, and put them into respective directories.

(From OE-Core rev: 64acb72e7ec63528073d8290137fe74d3382f876)

Signed-off-by: Junling Zheng <zhengjunling@huawei.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

crypto: use bigint in x86-64 perl

on some hosts openssl fails to build with this error:

ghash-x86_64.s: Assembler messages:
ghash-x86_64.s:890: Error: junk '.15473355479995e+19' after expression

backported fix from community.

(From OE-Core rev: 8230f873921d5c16106e3ebf57053a646bc6ad78)

Signed-off-by: Armin Kuster <akuster@mvista.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

elfutils: fix build with -Os

(From OE-Core rev: c46661bf4ffe1a474e4bcfadb4a3f12a46e8cc4a)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

syslinux: add syslinux-nomtools to PACKAGES

The default installed syslinux depends on mtools, we install
syslinux-nomtools too, which has the ext2/3/4 support.

(From OE-Core rev: 649110f5ab435231153e8f309ad2370e76c31e4f)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-firmware: fix the mess of licenses

Each firmware has their own license, it was a kind of a mess to
leave the license as Proprietary and we got a lot warnings like:

WARNING: The license listed Firmware-atheros_firmware was not in the
licenses collected for linux-firmware

To fix this and make it more clear:
- Fix some incorrect license names.
- Add all license files into common-licenses in previous commit.
- Add all licenses and their md5 values in alphabetical order.
- Add pn-*-license to split up per firmware package and pn-license for others.
- Re-order the codes for each split-out firmware as:
  LICENSE_*, FILES_*, RDEPENDS_*

[YOCTO #7539]

(From OE-Core rev: 1ee083da0730408fffdbbf5f29abc299c0e61be9)

Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

common-licenses: Add common license files for linux-firmware

[YOCTO #7539]

(From OE-Core rev: 9021ee03e0f9bd36b0ce63fa29dd010e07241fdb)

Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

kmod: upgrade to 20

The following patches are rebased.
-- Change-to-calling-bswap_-instead-of-htobe-and-be-toh.patch
-- avoid_parallel_tests.patch

The ptest part of kmod is removed because new version of kmod port all tests to
use modules from module-playground instead of copying prebuilt modules to the
repository. So, we cannot use ptest as before. Remove it so that kmod can build
successfully.

(From OE-Core rev: f80d136bdd578468035a88125fa1b84973fd912b)

Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa: Add test for layer append and FILESPATH tracking

This test actually tests a spectrum of bitbake functionality. Three
layers are created, one containing a recipe, one with a bbappend adding
a file to the recipe and another which overwrites the file in another
bbappend.

The correct outcomes in building the recipe are tested, with the file
in the final layer added, removed and then re-added.

This tests bitbake's cache handling as well as restoration from
sstate which happens in the final test phase.

Based on a test case from:
[YOCTO #7019]

(From OE-Core rev: e57437c0e02b148a878c8db91660674069fbd6e2)

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wic: Print bitbake error output

Added bitbake error output to wic error message to make it
more descriptive.

[YOCTO #6911]

(From OE-Core rev: f68576d4c9cecbe7122325d7fc78c9b27ff07a8e)

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wic: use native parted

Used exec_native_cmd instead of find_binary_path to run parted.
Got rid of find_binary_path as it's not used anywhere else.

There are several tools wic is trying to find not only in sysroot,
but also in host root. Parted is a special as on some distros it's
installed in /usr/sbin, which is not in the user's PATH. This makes
wic to fail with error "External command 'parted' not found, exiting."

[YOCTO #7122]

(From OE-Core rev: 76adf38c0d8e0faf04a5ecb3fcfbe831c85bb81f)

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: bitbake-layers: ensure non-extended recipe name is preferred

In show-recipes and show-overlayed, we only list one variant of each
recipe where multiple exist, therefore we should show the main one (e.g.
in OpenEmbedded, we now show openssl and not nativesdk-openssl which
would otherwise sort first.)

Fixes [YOCTO #7514].

(Bitbake rev: c4172fe33f0c63eda7f4af2bd36a5e9738e13595)

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libgpg-error 1.18: simplify tupple handling and add armv8b support

Add a default option to the case statement and remove the duplicates. Also add support for armv8b architectures.

(From OE-Core rev: 8d207e55031c0b93387e728f4312b8cb34ad5b12)

Signed-off-by: Koen Kooi <koen.kooi@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oe-init-build-env: fix for build dirs that have spaces in their path

Enclose expansions of BUILDDIR in quotes in order to correctly handle paths
which include spaces.

(From OE-Core rev: 0c08723cf5ebc1a7df7dc4ca36cffa090286468c)

Signed-off-by: Joshua Lock <joshua.lock@collabora.co.uk>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gdb: Refresh patches against git

Drop dead patches sim-install.patch, uclibc.patch

Change-Id: I5bd160d0959154896096ebc93a8a450d04979c73
(From OE-Core rev: 682ab52d776d74fb5ede949505f72b4c15ae8191)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

kbd: Mark glibc headers under __GLIBC__

Change-Id: I4e91038ea8b05defbb9d574c61eea96ccc064e51
(From OE-Core rev: f4ad72e6e46e481d3fad920be9bb417fd175a991)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

xserver-xf86-config: Preload graphics driver modules for musl

musl does not support BIND_LAZY intentionally, which means
dlopen will always complain about missing symbols which is what
X does when loading graphics drivers, here we preload the needed
drivers for all emulator machines if libc is musl

Change-Id: I908c94c30db8a5e872922e1a677126d82fa17145
(From OE-Core rev: 92be71296e1b7ed1e6387ae381942072aa3bb4f3)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

sudo: Disable hardening for musl

musl does not yet enable ssp

Change-Id: If40ab0a54b17d5528676b1268182b6c7e7ef5761
(From OE-Core rev: a583f81d9bc6ae984699d9aba7a53019ff0afc23)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pulseaudio: Fix build with musl

It reimplements some of standard C library functions but then it assumed
glibc so we need to break that assumption into conditional defines

Change-Id: I92109c2e4c48ab4c3565f64e68d9cbb165823b45
(From OE-Core rev: 87a55935a6d5f038c5a8a4c7f2357dd972d2fa92)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

xserver-xorg: Use __GLIBC__ instread of __linux__ to denote glibc dependent code

Change-Id: Ia899520181feb1c601a76e04a176f4387a331c00
(From OE-Core rev: e9ddb1ba61e1b773e7a0293c0e10f2b329d8bbe1)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

iptables: Fix build on musl

Added needed headers and resuffled existing ones to get it portable
Added defined for missing TCOPTS*

Change-Id: I74977dd052c5569b00631379d7f4bacfb86cf381
(From OE-Core rev: d30fba63286dc8f5ac72ac65fae6af6001e58ec2)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

perl: Fix build for musl

The changes are covered under "${TARGET_OS}" = "linux-musl"

Change-Id: I24a1a8e07abb35c7e3d64b372addfb3bd6f7731c
(From OE-Core rev: 3386ab153608f584e6b7e17c4bd0554e2ab31c66)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libfm-extra: Add own nls.m4

It helps to build when NLS is disabled distro-wide
otherwise it has dependency on gettext to provide it

Change-Id: Ib6920629b5ab1a1938cc5dfa447078f6fca234c7
(From OE-Core rev: 67a76eae50182da4426f35a96802f5d08b38e69b)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

vte: Pass _GNU_SOURCE via CFLAGS

It uses function extentions like  ptsname_r()
so we need to let build know about it

Change-Id: I5a06d08c56529a6b2cf3adf264656ae03b95c1d4
(From OE-Core rev: 1aa0346cf2a2104cfc969bfeb4ac58fa923d5fcb)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

musl: Enable NLS

It supports it, so lets enable it even though
we can compile lot of stuff without NLS but its
harder to get them all compiled without NLS

Change-Id: I49a06c05b004654dabbef980c4e6ad991d581341
(From OE-Core rev: 6cf978dfcc90632680db15b58f8f9edbbd7c083b)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ppp: Add extra include dirs

Just run configure, the configure script is not really
a autoconf generated script so just run it

Include path is needed to be there since
the headers from it are used during build in subcomponents

Change-Id: Ib1f24fd18bc8564e3c74ab834b7cf0fdf955d0e1
(From OE-Core rev: c64c9f7ffc88018404afdde373dfd7ef2bdf7ae8)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libproxy: Include select.h for posix compliance

unearthed with musl

Change-Id: I527139905785f253e41dd6a2f0cf555eee752a3a
(From OE-Core rev: 615ade48bfb8f1d1287e4ec79f4e2f4b610bdb5b)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

groff: add runtime dependency on sed

This is to suppress the following QA warning:
groff-1.22.2: groff requires /bin/sed, but no providers in its RDEPENDS [file-rdeps]

(From OE-Core rev: 1763ff614d11c78bf36d80d27dd6a4f45cbea070)

Signed-off-by: Dmitry Eremin-Solenikov <dmitry_eremin@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake.conf: add sed-native to ASSUME_PROVIDED

There is no reason to build sed for the host, however now:
ERROR: Nothing RPROVIDES 'sed-native' (but virtual:native:/OE/sources/openembedded-core/meta/recipes-extended/groff/groff_1.22.2.bb RDEPENDS on or otherwise requires it)

(From OE-Core rev: 5111407cfe59bd368e2528d953282b74d78057ed)

Signed-off-by: Dmitry Eremin-Solenikov <dmitry_eremin@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

package_manager: RpmPM: Fix scriptlet for rpm 4

Path of rpm post install script is different between rpm 4 and 5

(From OE-Core rev: a880669f767532ce3c39eb160fa9deb9fce29677)

Signed-off-by: Jean-Benoit MARTIN <jean-benoit.martin@open.eurogiciel.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

tiff: remove extra dev and dbg from PACKAGES

There should be only one dev and dbg package

(From OE-Core rev: dd65ee47d04f3f8e51a7abec02d564b9e46626ae)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

sysfsutils: remove libsysfs-dev and libsysfs-staticdev from PACKAGES

There should be only one dev and dbg package

(From OE-Core rev: 0b6f0efe7a96782634f72a68ee14a48205bf9577)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pciutils: remove libpci-dev and libpci-dbg from PACKAGES

There should be only one dev and dbg package

(From OE-Core rev: 91f91ff86f3259caaad290fa4efc288e502d4970)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

opkg: remove libopkg-dev and libopkg-staticdev from PACKAGES

There should be only one dev and dbg package

(From OE-Core rev: ebd0f747feee5bc8f5b2433127cb263a17351f45)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libxmu: remove libxmuu-dev from PACKAGES

There should be only one dev and dbg package

(From OE-Core rev: fbb0aa87fb178fd89c8828d8611b618bf21e5799)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libtool: remove libltdl-dev libltdl-dbg libltdl-staticdev from PACKAGES

There should be only one dev and dbg package

(From OE-Core rev: 5e40bcd93d9ea64f1a6d75e62aabda3dcb33fc01)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

elfutils: remove libdw-dev libasm-dev libelf-dev from PACKAGES

There should be only one dev and dbg package

(From OE-Core rev: 0dacfcc974068569f88adc8e33aea3ee1c7676de)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

cogl: remove extra dev from PACKAGES

There should be only one dev and dbg package.

(From OE-Core rev: e6be75c98edf878f73e530bc6e63b3c50bf365cc)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

augeas: remove extra dev and dbg from PACKAGES

There should be only one dev and dbg package.

(From OE-Core rev: 29136f21d33f211fc93c01c06ef636e2f8cdf909)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

apmd: remove libapm-dev and libapm-staticdev from PACKAGES

There should be only one dev and dbg package.

(From OE-Core rev: 4e6a03256ac527c69022a7136ca5c3cde2a8fc36)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

aspell: remove libpspell-dev from PACKAGES

There should be only one dev and dbg package.

(From OE-Core rev: b2a59adec6804dfeb85f2aef6bd8f7b2d07929b4)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

acl/attr: remove extra dev and staticdev from PACKAGES

There should be only one dev and dbg package.

(From OE-Core rev: 93afa6fcc0f84f305fad97407ef408e7a1e37652)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bzip2: remove libbz2-dev libbz2-staticdev from from PACKAGES

There should be only one dev and dbg package.

(From OE-Core rev: 8a8e8bf374a7c41a9056f446f11889992fec3448)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wic: setup infrastructure for nosetests

cd scripts && nosetests runs 3 simple tests and
produces nice test coverage report for wic codebase.

This infra can be used to test other scripts and modules.

(From OE-Core rev: 510dadc2cd8ff8a09932c101512dba74d4624a07)

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wic: Add argv argument to main

Make it possible to call wic as an API from tests passing
command line parameters as arguments to main.

This is yet another enabler for wic unit testing.

(From OE-Core rev: 8cb9ac3751424d429a38f6f80d733b289b6684ea)

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wic: Use __file__ instead of sys.argv[0]

Using __file__ makes the code work independently of the way
it's loaded. In some cases wic can be imported by another
program without executing it. sys.argv[0] would not contain
path to the wic in such a cases.

This is an enabler for unit testing with nose framework.

(From OE-Core rev: 7166262032648dbbc8b4b476c8770e35b1d4e234)

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wic:code cleanup: No space allowed

Fixed pylint warning
  'No space allowed around keyword argument assignment'

(From OE-Core rev: e07dd9b9c71960fbeded162ed52fbce06de620e9)

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

insane.bbclass: Enhance file-rdeps QA check

Adds symlink-to-sysroot check to QA_WARN to detect symlinks that
point to locations under TMPDIR, which are most likely broken.

Changes filerdepends from set() to dict(), hence methods for adding
or deleting items had to change too. Now it keeps track of key:value
relationship, flags the QA issue; warning the user about which
file/package causes the problem, making it easier to debug.

[YOCTO #7126]

(From OE-Core rev: 9946909dc95c3274a98112cf786d171547b3ed75)

Signed-off-by: Alejandro Hernandez <alejandro.hernandez@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

generate-manifest-2.7.py: Add fractions to ${PN}-numbers

python-cryptography requires fractions module which is currently unpackaged.

(From OE-Core rev: ac09a1830fffb6aded499ade55c039e35202867e)

Signed-off-by: Tim Orling <TicoTimo@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pcmciautils: fix for parallel build

Fixed:
Compiling lex_config.c.
  src/lex_config.l:34:25: fatal error: yacc_config.h: No such file or directory

There was a patch for fixing the paralle issue before, so modify the
patch again.

(From OE-Core rev: e1193feb30fbbfacb430ccbb9f33cd782f7cb87d)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

grub-efi: add PACKAGECONFIG for grub-mount and device-mapper

We had PACKAGECONFIGs for grub_2.00 and grub_git, we also need them for
grub-efi, otherwise there might be failures as:
grub-2.00/util/grub-mount.c:37:23: fatal error: fuse/fuse.h: No such file or directory

(From OE-Core rev: a7fca64d04949e76805e7acc3f617a1cd4a107a7)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

at-spi2-core: Upgrade to 2.16.0

* Add new org.a11y.atspi.Registry D-Bus service file

(From OE-Core rev: e4cbbfb167e22211abf41b90b71cab2fe5b330da)

Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libepoxy: Add as GTK+ 3.16 requirement

Libepoxy is a OpenGL function pointer management library required by
GTK 3.16.

(From OE-Core rev: 9c3ed6ed0e432291109242fb3aebca232944a7ee)

Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

at-spi2-atk: Upgrade to 2.16.0

(From OE-Core rev: dae6fccacfc7dbf3b2e5e1e0cb6de69d01671458)

Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

atk: Upgrade to 2.16.0

(From OE-Core rev: 191fbdecfd1eecb95fb433f6854bb6862561987c)

Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

cairo: Upgrade to 1.14.2

(From OE-Core rev: 38c401eed21c532fc0b3a90aa3cebfc6d570a526)

Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

glib-networking: Upgrade to 2.44.0

(From OE-Core rev: 42855986d425f0635f79808d9d7b7244f4601bff)

Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

git: 2.3.1 -> 2.3.5

(From OE-Core rev: 3cc46912c1933fd484a49c82e5e027030abecbf2)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wget: 1.16.2 -> 1.16.3

(From OE-Core rev: 9b2e9a08f3e77b1df604f3d283cac91e7ff7c602)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

groff: 1.22.2 -> 1.22.3

* Updated groff-1.22.2-correct-man.local-install-path.patch

* Add perl to RDEPENDS_groff since there are a few perl scripts:
  groff/usr/bin/roff2pdf:#! /usr/bin/env perl
  groff/usr/bin/afmtodit:#! /usr/bin/env perl -w
  groff/usr/bin/roff2dvi:#! /usr/bin/env perl
  groff/usr/bin/roff2text:#! /usr/bin/env perl
  groff/usr/bin/glilypond:#! /usr/bin/env perl
  groff/usr/bin/roff2ps:#! /usr/bin/env perl
  groff/usr/bin/gperl:#! /usr/bin/env perl
  groff/usr/bin/mmroff:#! /usr/bin/env perl
  groff/usr/bin/gpinyin:#! /usr/bin/env perl
  groff/usr/bin/roff2html:#! /usr/bin/env perl
  groff/usr/bin/grog:#! /usr/bin/env perl
  groff/usr/bin/gropdf:#! /usr/bin/env perl -w
  groff/usr/bin/pdfmom:#! /usr/bin/env perl -w
  groff/usr/bin/chem:#! /usr/bin/env perl
  groff/usr/bin/roff2x:#! /usr/bin/env perl
  groff/usr/bin/groffer:#! /usr/bin/env perl
  groff/usr/lib64/groff/gpinyin/subs.pl:#! /usr/bin/env perl
  groff/usr/lib64/groff/grog/subs.pl:#! /usr/bin/env perl
  groff/usr/lib64/groff/groffer/subs.pl:#! /usr/bin/env perl
  groff/usr/lib64/groff/groffer/man.pl:#! /usr/bin/env perl
  groff/usr/lib64/groff/groffer/main_subs.pl:#! /usr/bin/env perl

* Fix the dangling /usr/bin/perl or /bin/perl for grog.

(From OE-Core rev: f516c288d37e53856bce273af2cdbf24f3f5a6e1)

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

shadow: split files needed for PAM use into separate package

The rootfs creator automatically removes shadow for read-only
root filesystems, which breaks use of PAM plugins for login and
other process identity management utilities.  Package those programs
and config files separately, so they don't get removed.

(From OE-Core rev: a7d8eaef04c9dd6ede8d4efd8c4b776efbe3c767)

Signed-off-by: Matt Madison <matt@madison.systems>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bluez5: upgrade to 5.29

- Large release with over a month and 475 commits since 5.28;
- Internal GATT library received lots of updates;
- Fix for AVCTP key repeat timeout;
- Added support for the Multi Profile Specification (MPS).

(From OE-Core rev: 6a85e15eef1cfc0333a3b676976e454093f07ce1)

Signed-off-by: Cristian Iorga <cristian.iorga@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/runtime: Boot test for poky-tiny.

Bug 6705.

(From OE-Core rev: 4d3574e7fe5353b4dbbf5266a8b950b57542a743)

Signed-off-by: Lucian Musat <george.l.musat@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/utils: Add runner for poky-tiny qemu.

The connection and commands are done via serial.

(From OE-Core rev: bc6ac3058d7adb6f1bf87e57deaa50347f452e59)

Signed-off-by: Lucian Musat <george.l.musat@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/targetcontrol: Add support for poky-tiny in QemuTarget.

(From OE-Core rev: 8f1a52a3f72506911154769e6ad4a44f32c3112e)

Signed-off-by: Lucian Musat <george.l.musat@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

mesa: Fix building with musl

backport needed patches
add a patch to use _NP macros only on glibc
based linux and not on all linux

Change-Id: Ib43bec9d222856f8e7acb9f0143b406517f049c8
(From OE-Core rev: 7dcd8f532fef71e711bf11a00f0e246a36b6138a)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

at-spi2-core: Add nls.m4 from gettext

This is needed for building it with musl, since
we disable NLS, I think it will fails in same
way with glibc if NLS was disabled there too

Change-Id: I95c6858adfb33361cab466999e0bacb47c1e1fde
(From OE-Core rev: 5dc86b1a45581e2b6ae845b818e2323645c9cff4)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bluez4: Fix encrypt symbol namespace collision

| Makefile:3352: recipe for target 'test/l2test.o' failed
| make[1]: *** [test/l2test.o] Error 1
| make[1]: *** Waiting for unfinished jobs....
| test/rctest.c:82:12: error: 'encrypt' redeclared as different kind of
symbol
|  static int encrypt = 0;
|             ^
| In file included from test/rctest.c:33:0:

Change-Id: Iaf2f644ef3cef6f96c6e4bc421c9e78a0e23e674
(From OE-Core rev: 795a409919774f3a2a4c859799de36f1064efd83)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

quota: Fix build with musl

add more packageconfigs so we can disable
certain features on demand

Add patches to make it more portable
by making posix compliant and libc agnostic code

Change-Id: Ifa1c7f7923322fd22368538310685cd5c440516d
(From OE-Core rev: 6da9a5269782923fe5807b07217ffe19b9fb633e)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

binutils: Fix linking issue for R_ARM_REL32 relocs

This issue is surfaces with gcc 5.0 but its a real issue in gold
errors are like

https://sourceware.org/ml/binutils/2010-12/msg00473.html
is the test case

Change-Id: I44806e9fb75a164745e52f0040b7efc9be624ae6
(From OE-Core rev: 887c8cb41a3857b2381e3d83c2db83c1d78ae48d)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ncurses: Upgrade 4.9 stable to latest patchlevel 20150329

gcc5 barfs on older release

Change-Id: Ib1ad9b1cf7f16eb454da21fd61dc56fe1d9f6783
(From OE-Core rev: 47f67fae98faec17087a827eff141e05b0400560)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libusb-compat: Include sys/types.h in usb.h

Fixes compile errors when usb.h is included in
other packages, it expects sys/types.h to be
there, instead it should include it itself

Change-Id: Ib49477b7d3f3cd97d65c9578bdc33f2ee33f5a26
(From OE-Core rev: 577b5c6ce6753b9ae3587da47caf6ecfac358668)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

util-linux: Upgrade to 2.26

Drop backports

Change-Id: I94d442edaf37ab8e685670f0e14ed60031a995fa
(From OE-Core rev: 302844ab3302050733df5e9c5e9cb92351cd315e)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

quota: Upgrade quota_4.01.bb -> quota_4.02.bb

Change-Id: Idea094dfe4b49eef58b90296c4cd2b3e239eeabb
(From OE-Core rev: be5c87032317edae1b91aa15713411829c2d77f0)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libffi: Use proper compiler define for linux platform

Change-Id: I8253249a2e1c5e9ded0d028545f21cfdc093c3cd
(From OE-Core rev: f7307ea20c8aa4aed4c38a7c15fb9b3b6b836035)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

common-licenses: Add BSD-0-Clause

Some packages like toybox use it

Change-Id: I2ae4fd024e310bce644581e4c63ed2980ee1998f
(From OE-Core rev: ae85c89810488a9eb86ae3bd765260eb5935711b)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libdaemon: Explicitly include unisttd.h instead of sys/unistd.h

Fixes build errors on musl

Change-Id: Ia00bdc9c0578c8d5e1249c9d433079d795e83cb2
(From OE-Core rev: bdd9bd8beaf49133cd62b27255897cdbd2b42b49)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wic: make error message more informative

Included name of the missing program into the error message produced
by exec_native_cmd when program is not found.

Removed mentioning of the host programs as this API is not running
them anymore.

(From OE-Core rev: 7d1599cf6a55d3624d2b90c1f55e14c4df73e570)

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wic: don't use host paths when looking for native tools

exec_native_cmd should not use host paths in $PATH to
avoid finding and using host tools.

[YOCTO: #6204]

(From OE-Core rev: fa263f238bbddb00c9953994fb69cc358170e2ec)

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

xorg-server: Security Advisory - xorg-server - CVE-2015-0255

Updated x11-server packages fix security vulnerability:

Olivier Fourdan from Red Hat has discovered a protocol handling issue in
the way the X server code base handles the XkbSetGeometry request, where
the server trusts the client to send valid string lengths. A malicious
client with string lengths exceeding the request length can cause the server
to copy adjacent memory data into the XKB structs. This data is then
available to the client via the XkbGetGeometry request. This can lead to
information disclosure issues, as well as possibly a denial of service if a
similar request can cause the server to crash (CVE-2015-0255).

(From OE-Core rev: 564e2f9732ac4318bb5923dd1ff771514c9afd2f)

Signed-off-by: Li Zhou <li.zhou@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libpcap.inc: consolidate inherit and PACKAGECONFIG lines

(From OE-Core rev: ce9fa5588acdf56e32bbdf9da512aa4071cf7266)

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libpcap.inc: remove obsolete libnl1 PACKAGECONFIG

There's no libnl1 recipe in oe-core (or any other layer in the layer
index).

Keeping the libnl1 PACKAGECONFIG is likely to cause problems for the
libnl PACKAGECONFIG since libnl and libnl1 both use --with-libnl and
--without-libnl.

(From OE-Core rev: 4c820481147e9c11160b00f7df7ca8c329b23eac)

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/runtime/smart: Added new decorators for existing automated tests.

(From OE-Core rev: d307950f4938ea2f9d480d4f4cd184b76e33df0b)

Signed-off-by: Lucian Musat <george.l.musat@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/runtime/rpm: Added new decorators for existing automated tests.

(From OE-Core rev: 7381c10f2260181eccd3abddf5d0289c969863cb)

Signed-off-by: Lucian Musat <george.l.musat@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/runtime/ping: Added new decorators for existing automated tests.

(From OE-Core rev: 9ca0f5c7837ef42cf3b95ff4091ea6951b4e8021)

Signed-off-by: Lucian Musat <george.l.musat@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/runtime/parselogs: Added new decorators for existing automated tests.

(From OE-Core rev: 8fb4bc7296d8da62ac2db3bf65d202a9c27d743f)

Signed-off-by: Lucian Musat <george.l.musat@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/runtime/ldd: Added new decorators for existing automated tests.

(From OE-Core rev: 6385874553b8bd9302c1b77d0586edd821e999be)

Signed-off-by: Lucian Musat <george.l.musat@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

testimage: sort modules based on dependencies

TEST_SUITES="auto" is useful to run all suitable tests without
having to hard-code the list. However, it did not take test
dependencies into account, which can be an issue for tests
which really depend on some other test to run first.

To fix this, modules get loaded in the order determined by
TESTS_SUITES, but then get re-ordered based on dependencies
derived from @skipUnlessPassed before running them. The original
order is used to break ties when there are no dependencies, so
reordering only occurs when really necessary.

@skipUnlessPassed gets extended such that it makes the test name
a method depends on available for inspection by the test loader
in oetest.py.

Unfortunately Python's unittest offers no API to inspect tests
in a TestSuite, so the code has to rely on implementation details
to find all tests. The worst that can happen when the implementation
changes is that tests are not found and reordering does not happen.

(From OE-Core rev: 6e4543a35836c572b23b9f8162b19d1e038d3ed2)

Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/runtime/systemd.py: skip instead of failing without avahi

The SystemdServiceTests assume that avahi-daemon is installed,
which is not necessarily the case depending on the image being
tested.

Better check this dependency before starting the tests and skip
them if the service is not installed. This has to be done for
each test instead of for the entire module, because other
tests in the module can run without avavi.

(From OE-Core rev: ca23337d517cfdb7119e5fd8bd9a9a663ae135de)

Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/oetest.py: use bb logging

Plain print has two drawbacks:
- the output is handled by different processes, causing the message
  about found tests to appear randomly after the result of the initial
  tests
- the output is not part of the bitbake console output

Affects image testing with testimage.bbclass.

(From OE-Core rev: 4b6f87a82ece230b7296f1577d4092ef559af1d4)

Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gst-ffmpeg: fix internal-libav builds with inherit autotools-brokensep

Building the internal copy of ffmpeg is broken when using S!=B, so
inherit autotools-brokensep

(From OE-Core rev: 4ac5d1011fe6b5bba9511416082891e925cbd03c)

Signed-off-by: Koen Kooi <koen@dominion.thruhere.net>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gst-ffmpeg: remove bogus patch that leads to build failures

'0001-huffyuvdec-check-width-more-completely-avoid-out-of-.patch'
patches the internal copy of ffmpeg with a hunk that generates a compile
failure because AV_PIX_FMT_YUV422P is undefined.

(From OE-Core rev: 3657e0fc2cd48a08035c7a0ab4da7e6666d2d91d)

Signed-off-by: Koen Kooi <koen@dominion.thruhere.net>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gst-ffmpeg: fix libav-9.patch

The patch has a stray '>' in an #incude directive leading to build
failures when used.

(From OE-Core rev: f2a45c0c1b9878319890fe03f26dd4e0db740091)

Signed-off-by: Koen Kooi <koen@dominion.thruhere.net>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

dpkg: add triplet entry to fix build error for armeb

Cross-compling dpkg application for armeb fails with below error
during configure task,

(snip)
 configure:23141: checking dpkg cpu type
 configure:23148: result: armeb
 configure:23150: WARNING: armeb not found in cputable
 configure:23162: checking dpkg operating system type
 configure:23169: result: linux-gnueabi
 configure:23171: WARNING: linux-gnueabi not found in ostable
 configure:23183: checking dpkg architecture name
 configure:23189: error: cannot determine host dpkg architecture
-- CUT --

Add the required combination of "gnueabi-linux-armeb" entry in
triplet list.

(From OE-Core rev: 63eb33bced1fc1e5451988fc5249ab362fb82615)

Signed-off-by: Krishnanjanappa, Jagadeesh <jagadeesh.krishnanjanappa@caviumnetworks.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib/oeqa/selftest/bblayers: improve test for add-layer and remove-layer

Use absolute path instead of relative path for meta-skeleton.
Otherwise the case would fail if the builddir outside the poky source
tree.

(From OE-Core rev: 90edaf9b9e51a24bf67c1c95732fc2d8cb22b098)

Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

shadow: add 'util-linux-sulogin' to RDEPENDS

If shadow is installed, sulogin from busybox cannot work correctly because
it still assumes that /etc/shadow is not there. This leads to the problem
when booting into rescue mode in an image with shadow installed but not
sulogin from util-linux.

To fix this problem, we add 'util-linux-sulogin' to RDEPENDS of shadow.
This runtime dependency is specific to OE, because we have to ensure
that sulogin can work correctly and sulogin from busybox cannot because
FEATURE_SHADOWPASSWORDS is not enabled by default. And we cannot enable
it by default for busybox, because that would lead to utilities in busybox
to assume the existence of /etc/shadow which is not always true in OE.

[YOCTO #6698]

(From OE-Core rev: 9b45b990d2fc870df556f05908dcb48b9ebcfc02)

Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

util-linux: split out util-linux-sulogin

[YOCTO #6698]

(From OE-Core rev: 4bde182ed236243547929dd98763f1c09eddd097)

Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pango: fix postinst

* merge postinst_prologue with the append in the do_split_packages call
  Now we can call correct pango-querymodules binary and respect D
  variable, otherwise we're trying to regenerate it on host which fails:
  pango-module-basic-fc.postinst: line 17: /usr/bin/pango-querymodules:
  No such file or directory

(From OE-Core rev: f5d189b80a1fd7187f47678935f9a584ccff8d0a)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

tzdata: fix postinst

* add quotes around possibly empty tz variable
* use exit instead of return, because we're not in function and postinst
  fails:
  line 9: return: can only `return' from a function or sourced script"

(From OE-Core rev: 87429d37e72b1fd2f329d693b4d1b1229a90e0a2)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libgpg-error: Fix native build on i686

Most modern x86 systems return i686 as the system type.  This
patch handles i486, i586 and i686 correctly.

(From OE-Core rev: 1b24eaa71702c07e525d8a62c08c0983e9917468)

Signed-off-by: Gary Thomas <gary@mlbassoc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

connman: Create connman.service at proper moment

ConnMan commit ac332c5d01b0737c18cb58c8ccc67cf6b0427e1d changes
how the connman.service file is created from .in file. After
that commit, the file is created by Makefile instead of configure.
This means that we need to tweak the service file in compile
time instead of configure time because the generated file will not
be there after the configuration.

This commit can be used even with older ConnMan version as the
connman.service file is there when the compilation happens.

(From OE-Core rev: 32839103727d92a6580f916b6dd8e4439b2347c8)

Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>