bitbake: lib/bb/utils: fix error in edit_metadata() when deleting first line

If you tried to delete the variable on the first line passed to
edit_metadata() this failed because the logic for trimming extra blank
lines didn't expect the list to be empty at that point - fix that bad
assumption.

(Bitbake rev: 8bce6fefdc5c046b916588962a2b429c0f648133)

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

bitbake: wget.py: parse only <a> tags

For two reasons:
1) The important one: we hit the following bug when doing upstream version checks
on some webpages:
https://bugs.launchpad.net/beautifulsoup/+bug/1471755

2) Also, documentation for beautifulsoup states that memory usage and
speed is improved that way.

(Bitbake rev: 7546d4aeb3ba8fda9832081b84d93138dc5e58d6)

Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: toaster: toastergui tests Add generic test for ToasterTables widget

For each of the ToasterTables defined test:
 - Data returned
 - Search
 - Order by
 - Limit
 - Paginate
 - Filter

These tests test that the functions on all tables work, it does not
validate the content of the data in the tables themselves, this needs
to be done in table specific tests.

(Bitbake rev: 649d563f95ffc57d0fe3dbf494e7dbffcc99a1b4)

Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: toaster: tables Fix invalid field name on NewCustomImagesTable

Correct the field name in the NewCustomImagesTable as it is a Recipe
object it's self and not a container.

(Bitbake rev: ebd1c493d8b7fb9ee7b3e40c17165dc9c22ca795)

Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: toaster: tables Add default_orderby field where it was missing or unset

This value is used to set the default ordering of the model that is used
for ToasterTables, it is picked up client side to set the ordering
indicator.

[YOCTO #8695]

(Bitbake rev: f1c91bff810a579b169c46b7710e22f5553b484e)

Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: toaster: CustomImageRecipe add search_allowed_fields to this model

In order to search the model from the UI some fields must be nominated
as searchable.

(Bitbake rev: 2558729b4c248a2fa5a11d877bc42cb05a30602e)

Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: toaster: machines table Fix missing layers information needed for filter

The current layers information wasn't being passed to the template for
the Select/Add button for the Compatible machines filter.

(Bitbake rev: 559102984d3f62a7675899e058166bdf4d552c46)

Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: toaster: tablejs Make sure click handlers consume click event

Avoid the click event from propagating and causing strange side effects
in toaster tables.

[YOCTO #8527]
[YOCTO #8148]

(Bitbake rev: f23e54292a673b2fd3f17d317c21984c753727c7)

Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: toaster: projectpage Make sure build targets are space separated

Make sure the build targets are space separated when building multiple
targets. Also fix error path now that YOCTO #7995 is resolved.

[YOCTO #8450]

(Bitbake rev: e9719eb2e8249f0d10b39bfdd4aef563368b5ffd)

Signed-off-by: Michael Wood <michael.g.wood@intel.com>
Signed-off-by: Elliot Smith <elliot.smith@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libsdl: remove redundant configure_tweak patch

configure_tweak.patch was simply adding arguments to AC_DEFINE which are only
needed by autoheader.  We we now disable autoheader and use upstream's
config.h.in this patch isn't needed anymore.

(From OE-Core rev: 687e5ef86361a16d6c411386939d4ba96a5909ea)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

iw: upgrade to version 4.3

Refreshed the following patches to work with this release:
 1) 0001-iw-version.sh-don-t-use-git-describe-for-versioning.patch
 2) separate-objdir.patch

(From OE-Core rev: 11e6536bbed515a360870e3f0491c93dd8c9e098)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gstreamer1.0-plugins-good: fix PACKAGECONFIG for gudev and add one for v4l2 and libv4l2

* WARN: gstreamer1.0-plugins-good: gstreamer1.0-plugins-good-video4linux2 rdepends on libcap, but it isn't a build dependency?
  WARN: gstreamer1.0-plugins-good: gstreamer1.0-plugins-good-video4linux2 rdepends on libgudev, but it isn't a build dependency?
  WARN: gstreamer1.0-plugins-good: gstreamer1.0-plugins-good-video4linux2 rdepends on libudev, but it isn't a build dependency?
  WARN: gstreamer1.0-plugins-good: gstreamer1.0-plugins-good-video4linux2 rdepends on zlib, but it isn't a build dependency?

(From OE-Core rev: 43fffa763f87b80a2497a51d6bd827d2bf837a11)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gstreamer1.0-plugins-bad: fix dependencies for uvch264 PACKAGECONFIG

* ERROR: gstreamer1.0-plugins-bad: gstreamer1.0-plugins-bad-uvch264 package isn't created when building with minimal dependencies?
* ERROR: gstreamer1.0-plugins-bad: gstreamer1.0-plugins-bad-uvch264-dev package isn't created when building with minimal dependencies?

* it's because it should depend on libgudev not udev:
  configure: *** for plug-ins: uvch264 ***
  checking linux/uvcvideo.h usability... yes
  checking linux/uvcvideo.h presence... yes
  checking for linux/uvcvideo.h... yes
  checking for GST_VIDEO... yes
  checking for G_UDEV... no
  checking for LIBUSB... yes

(From OE-Core rev: a1dd74e6ea7e6e73c9c85c07034587ce70751068)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gudev: Add from meta-oe

libgudev was split from systemd and is used by recipes in oe-core.
udev already provides libgudev as well.

The recipe is from meta-oe (Andreas Müller).

Fixes [YOCTO #8686]

(From OE-Core rev: c6d37fc47f7ee637a069d3225e294d181e588a0e)

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

lsb: fix installed-vs-shipped for mips

Fixed:
lsb-4.1: lsb: Files/directories were installed but not shipped in any package:
  /lib32
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them within do_install.
lsb: 1 installed and not shipped files. [installed-vs-shipped]

(From OE-Core rev: 614398f106089f1aae58870286f50717ac03c662)

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

rpm: fix for N32 MIPS64

It is 'N32 MIPS64', not 'N32 MIPS32' as command file shows:
$ file image/usr/bin/getent
getent: ELF 32-bit MSB executable, MIPS, N32 MIPS64 [snip]

And "rpm -qp --filecolor" was wrong (it was 1, but should be 4), which
caused multilib installation error.

(From OE-Core rev: a598f6ee369c2a55b080ac7cfc058c1d30c7be2e)

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

glibc/0029-fix-getmnt-empty-lines.patch: fix getmntent()

When confronted with an empty line, getmntent() can underrun
a buffer, possibly doing very strange things if it finds
additional space/tab characters. Backport the upstream fix.

(From OE-Core rev: 983a19a65a31b54a6f505181012bd311c28a0ae1)

Signed-off-by: Peter Seebach <peter.seebach@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

init-install-efi: fix script for eMMC installation

Running the install option from bootloader to install image to eMMC will fail
with error:
Formatting /dev/mmcblk01 to vfat...
mkfs.fat 3.0.28 (2015-05-16)
/dev/mmcblk01: No such file or directory

This issue impacts both grub and gummiboot install option to eMMC device.
The installation failure is due to the following:
[1] Unable to partition eMMC as the partition prefix 'p' is not appended
The condition checking failed with the additional /dev/ appended with
the target device name.
[2] The partition uuid for boot, root and swap partition is not captured
for eMMC

This fix updated the condition checking and changed the variables to
reference the boot, root and swap partitions for UUID.

[YOCTO #8710]
(From OE-Core rev: a7d081c3db776c8b0734942df6bf96f811f15bd3)

Signed-off-by: Ng, Mei Yeen <mei.yeen.ng@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

init-install-efi: fix script for gummiboot loader

After running gummiboot loader install option, the installed target
storage device boot parameter for root=PARTUUID is empty causing boot failure.
This issue is only observed with gummiboot and not with GRUB loader.

This fix assign the rootuuid of the rootfs partition for gummiboot loader.

[YOCTO #8709]
(From OE-Core rev: fd5fa06fe45ca1fdf20df4198ae323967244af5b)

Signed-off-by: Ng, Mei Yeen <mei.yeen.ng@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-firmware: rtl8192cx: Add latest available firmware

Add latest available firmware binaries for RTL8192CX chipsets.
These new firmwares have been released in 2012, have been used
by the mainline kernel as preferred firmware since 3.13 and
even backported to stable branches.

(From OE-Core rev: 2dc67b53d1b7c056bbbff2f90ad16ed214b57609)

Signed-off-by: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libsdl2: add missing dependency on libxkbcommon for PACKAGECONFIG[wayland]

libsdl2 depends on libxkbcommon when enable wayland support.

Can verify it with the following steps:
bitbake libxkbcommon -c clean
bitbake libsdl2 -c cleansstate && bitbake libsdl2

config.log:
[snip]
configure:19347: checking for Wayland support
configure:19357: result: no
[snip]

(From OE-Core rev: ec0354bacfed387f91e610ebfdb77e86135c1d28)

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

libxml2: upgrade to 2.9.3

- Drop all the upstreamed patches
- Rework the ansidecl removal so it's contained in a single patch

(From OE-Core rev: 88e68f25e1756988692108d4c15dfa8efc94e5e5)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libxml2: merge pointless bb/inc split

(From OE-Core rev: f24baa73863258dbdac69cc5e235d3ce8d060cd6)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

openssh: redesign ssh-agent.sh regression test case

ssh-agent regression test case should be run by non-root user,
but non-root user will has issue to run other testcase, so
rewrite it on run-ptest

(From OE-Core rev: 5ca6bb9b73bf09c1847ec3e5a7477829bd3d77b5)

Signed-off-by: Roy Li <rongqing.li@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gcr: Require x11 DISTRO_FEATURE

This enables a world build without x11. GTK3DISTROFEATURES is not
enough because gtk+-x11.pc is still required.

Fixes [YOCTO #8611].

(From OE-Core rev: dbdcd87144cc1cd6c5d50c800c7f266aaf25ca17)

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

psplash: update to latest git version

Latest git version contains various changes, including
fix for big endian RGB888 rendering. It also makes the patch:
"0001-psplash-fb-Convert-psplash_fb_plot_pixel-to-a-static.patch"
unnecessary, as this patch has been upstreamed.

[YOCTO#7236]

(From OE-Core rev: f9450bf01a792e29bf83c820aae431483caff508)

Signed-off-by: Juro Bystricky <juro.bystricky@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

sysvinit-inittab: Add wrapper script to verify console exists

Add a wrapper script around getty to check if a given console exists, this
allows for multiple Console ttys to be specified for various boards without
having additional BSP types just for different console types.

[YOCTO #8689]

(From OE-Core rev: b93369a7943949f51057e0a704f5524ab7682fe6)

Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-yocto/4.1: Bluetooth:Fix the connection fail of 6lowpan over BT LE

Backporting the 4.4 fix for 6lowpan:

   Author: Wu Zheng <wu.zheng@intel.com>
   Date:   Fri Nov 20 13:25:57 2015 +0800

       Bluetooth:Fix the connection fail of 6lowpan over BT LE

       When two devices with 6lowpan over BT LE connect each other,
       6lowpan over BT LE channel is set up between the two devices.
       However, the status of channel is not right.
       It always is set to CONNECTED and the channel can't be created.

       The status of channel need to be removed when connection is created.

       (The patch's reference from
       b0c09f94ff1660a1873549b788c998284ea5fb8a)

Signed-off-by: Wu Zheng <wu.zheng@intel.com>
(From OE-Core rev: 2b6a6cad90bfef3afd5a721ec985f71336914fa7)

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-yocto-rt/4.1: update to -rt15

Updating the 4.1-rt kernel to the latest and greatest upstream version.

(From OE-Core rev: 22912a99163402581e8e92dcacbf46f9049f6ff9)

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-yocto/4.1: fsl-mpc8315e-rdb: Enable EEPROM

Updating the 4.1 SRCREVs for:

  Update the mpc8315erdb.dts to contain the eeprom device
  information. Coupled with adding "MISC_DEVICES" into the
  kernel configuration, we have a working eeprom.

Signed-off-by: Jingdong Lu <jingdong.lu@windriver.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
(From OE-Core rev: 47a222ad6f3ee8b14b9eea28b63397077978f42f)

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-yocto/4.1: update to v4.1.13

Updating the 4.1 kernel to the latest stable (and resolving minor
conflicts with -rt).

(From OE-Core rev: 12be8565b16f0fb571f859d3ef256767feaafc7f)

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

uClibc: enable utmp for shadow compatibility

with the enabling of utmpx in busybox and uClibc it was noted that shadow
support for utmpx also needs utmp explicitly enabled in uclibc. this is
a workaround that might be removed once shadow properly supports
--enable-utmpx to check for utmpx configuration instead of utmp like
it does now

[YOCTO #8243]

(From OE-Core rev: 969158d63ba2c8e2e11af41c2a6d4f1aa5b0099f)

Signed-off-by: Bogdan-Alexandru Voiculescu <bogdanx.a.voiculescu@intel.com>
Signed-off-by: Benjamin Esquivel <benjamin.esquivel@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

glibc: Backported a patch to fix glibc's bug(18589)

Also Fix LSB NG cases:
 * /tset/ANSI.os/locale/setlocale/T.setlocale 1 2 4 5 15
 * /tset/ANSI.os/string/strcoll_X/T.strcoll_X 1
 * /tset/LI18NUX2K.L1/base/wcscoll/T.wcscoll 1
 * /tset/LI18NUX2K.L1/utils/localedef/T.localedef 7
 * /tset/LI18NUX2K.L1/utils/sort/T.sort 1 3 17 19 33 35
 * /tset/LI18NUX2K.L1/utils/comm/T.comm 1 2
 * /tset/LI18NUX2K.L1/utils/ls-fh/T.ls-fh 2

This patch is backported from
https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=6c84109cfa26f35c3dfed3acb97d347361bd5849

(From OE-Core rev: e88fe8f4c0ea70fb271d3a11e1a3bfcac4c92626)

Signed-off-by: Li Xin <lixin.fnst@cn.fujitsu.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ncurses: update SRC_URI

ncurses source tarball from "current" directory in
ftp://invisible-island.net gets updated/deleted during the release cycle.

We can't depend on that FTP server to fetch source as the maintainer
doesn't guarantee the presence of all patched versions in the server.
ftp://invisible-island.net/ncurses/5.9/README

Fetch sources from debian git server as it is comparatively reliable.

[YOCTO #8713]

(From OE-Core rev: 4d3f84f84147145cfd786362d9cd754bbb93873e)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

openssl: enable parallel make

openssl 1.0.2d fixes the parallel make problems (commit 8e6bb99), so enable
parallel make again.

[ YOCTO #7347 ]

(From OE-Core rev: ea89857f17a374b6095371ebe2422d2e83735cee)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

busybox: enable resize applet

The /etc/profile script contains a call to resize, which improves
the usability of shells run on the serial console.

  http://git.openembedded.org/openembedded-core/commit/?id=cc6360f4c4d97e0000f9d3545f381224ee99ce7d

Unfortunately the resize applet is not currently enabled in busybox
defconfig, so resize is never called. Fix that.

(From OE-Core rev: 2136fecd662a6eeea02322628021de7982b5d35e)

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

busybox: disable support for mounting NFS file systems on Linux < 2.6.23

The busybox CONFIG_FEATURE_MOUNT_NFS config option is described as:

  Enable mounting of NFS file systems on Linux kernels prior
  to version 2.6.23. Note that in this case mounting of NFS
  over IPv6 will not be possible.

Since OE-core sets OLDEST_KERNEL = "2.6.32", CONFIG_FEATURE_MOUNT_NFS
is not required in the default busybox defconfig.

(From OE-Core rev: b47ff1605940694150e497d8f18d6201e1627415)

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

busybox: update 1.23.2 -> 1.24.1

The busybox defconfig has also been refreshed, with all new apps
and features disabled by default. Update _git recipe version too.

(From OE-Core rev: 4d5f51dd3b2733f15cb2eee40e622ec43f46d91c)

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

busybox: re-order defconfig to align with busybox 1.24.1

No functional changes, simply re-order lines in defconfig so that
the existing options don't move elsewhere in the file when run
though busybox 1.24.1 'make oldconfig'.

(From OE-Core rev: 28f197915d816ea249b036f8a4dfe11d477f94e7)

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

busybox.inc: remove '-e MAKEFLAGS=' from EXTRA_OEMAKE

Busybox Kbuild likes to control its own MAKEFLAGS (it adds -rR, etc),
so avoid over-riding MAKEFLAGS. Relying on 'make -e' is no longer
required so remove that too.

(From OE-Core rev: 1d841e103b229bf70a1c24c1cc22b7bdc8c9bf64)

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

busybox.inc: set CC=${CC} via make command line

Busybox currently relies on 'make -e' to over-ride CC and the make
command line to over-ride LD. Set CC via the make command line to be
consistent with LD and to allow '-e' to be dropped from EXTRA_OEMAKE.

(From OE-Core rev: b7c265e1edd5c82126c1f3915ba5ca9efef57c00)

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

busybox.inc: fix CONFIG_EXTRA_CFLAGS configmangle

With current busybox Kbuild, setting .config to:

  CONFIG_EXTRA_CFLAGS="foo" "bar"

and then running 'make oldconfig' results in .config containing:

  CONFIG_EXTRA_CFLAGS="foo"

ie the CONFIG_EXTRA_CFLAGS configmangle in the busybox.inc doesn't
currently work as intended. Remove the extra \" \" to ensure that
${HOST_CC_ARCH} gets added to CONFIG_EXTRA_CFLAGS.

(From OE-Core rev: 5f822373fadff2415cf1602b9d58fae0dbd33730)

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

busybox.inc: don't set .config CROSS_COMPILER_PREFIX

Setting CROSS_COMPILER_PREFIX via .config is redundant (setting
CROSS_COMPILE via the make command line will always over-ride it).

(From OE-Core rev: db2daff77a9d533dc03ab3aaeffa87e32f834240)

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

busybox: move EXTRA_OEMAKE etc into busybox.inc

EXTRA_OEMAKE options and do_install_ptest() are common to both
busybox recipes, so move into busybox.inc.

(From OE-Core rev: aeeb80be6db8df0548c97b4a40c1561cc724debb)

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

busybox.inc: don't export EXTRA_OEMAKE

EXTRA_OEMAKE is private to OE and shouldn't be exported to
the busybox build.

(From OE-Core rev: 61e1617f9f024ead0b77004c94f2de45c88a4aa8)

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

busybox_git: Enable getopt applet

Keep git recipe in sync with 1.23.2:

  http://git.openembedded.org/openembedded-core/commit/?id=10c2c484d5916ad476ad7717c3629f6684f01e6d

(From OE-Core rev: 33868c680587036d1aa3b56e66599a5b972626c4)

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

harfbuzz: update 1.0.6 -> 1.1.0

  http://cgit.freedesktop.org/harfbuzz/tree/NEWS

Overview of changes leading to 1.1.0
Wednesday, November 18, 2015
====================================

- Implement 'stch' stretch feature for Syriac Abbreviation Mark.
  https://github.com/behdad/harfbuzz/issues/141
- Disable use of decompose_compatibility() callback.
- Implement "shaping" of various Unicode space characters, even
  if the font does not support them.
  https://github.com/behdad/harfbuzz/issues/153
- If font does not support U+2011 NO-BREAK HYPHEN, fallback to
  U+2010 HYPHEN.
- Changes resulting from libFuzzer continuous fuzzing:
  * Reject font tables that need more than 8 edits,
  * Bound buffer growth during shaping to 32x,
  * Fix assertions and other issues at OOM / buffer max-growth.
- Misc fixes and optimizations.
- API changes:
  * All fonts created with hb_font_create() now inherit from
    (ie. have parent) hb_font_get_empty().

(From OE-Core rev: ed76a9e36edc054cbd87b6d1cab0202526959868)

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

sqlite3: update 3.9.0 -> 3.9.2

(From OE-Core rev: dca2138b4f2adbee452a48c2c6561d8ba295c842)

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

readline: apply missing upstream patches

Some readline-6.3 upstream patches was missing. Also ensure that the
upstream patches are applied in the same way as in readline-5.2.

Remove 'readline-dispatch-multikey.patch' and
'readline-cve-2014-2524.patch' since they are already included in
upstream patches 'readline63-002' and 'readline63-003'.

[YOCTO #8451]

(From OE-Core rev: 7b1fde3f65b674f5973800731c0c284f7d415248)

Signed-off-by: Petter Mabäcker <petter@technux.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

readline: prepare for readline6.3 upstream patches

Upstream patches are always preferred to be applied first (before
integration patches). In order to apply readline-6.3 specific upstream
patches in a preferred order we need to apply the integration-patches
at the end in the 6.3 specific recipe (this is already the case
for readline-5.2).

Also take the oppertunity to move 'norpath.patch' to readline-6.3 dir
since this patch is not shared between the 5.2 and 6.3 recipe.

[YOCTO #8451]

(From OE-Core rev: d454d1c06247b658c6b7d12de610eb6ac72cd7d4)

Signed-off-by: Petter Mabäcker <petter@technux.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

dbus: merge .bb and .inc

The split is no longer necessary, and only makes it more complicated to
do changes in external layers.

(From OE-Core rev: aea82066eb3ba67c1d3a170112ff3bb324b43dfb)

Signed-off-by: Andrew Shadura <andrew.shadura@collabora.co.uk>
Signed-off-by: Joshua Lock <joshua.lock@collabora.co.uk>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pulseaudio: Fix HDMI profile selection

On systems with two cards, the correct output profile does not get
selected automatically even in the simple case where there is one
available profile. This scenario is typical at least with HDMI audio
(which is on a separate card).

Fixes [YOCTO #8448]

(From OE-Core rev: 4bd76b8b292a81ce12d316557ffc9c6a79a7b5c1)

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

initscripts: hide the error in case system is not writeable

To avoid reporting errors for /etc/timestamp is not writeable in a
readonly system.

Reported-by: Niklas Soderlund <niklas.soderlund@external.atlascopco.com>
(From OE-Core rev: 463bef8b055f8305b7aac5045fefe4276b1432ad)

Signed-off-by: Ming Liu <peter.x.liu@external.atlascopco.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

nativesdk-buildtools-perl-dummy: fix rebuilding when SDKMACHINE changes

This recipe produces an empty dummy package (in order to satisfy
dependencies on perl so we don't have perl within buildtools-tarball).
Because we were inheriting nativesdk here the recipe was being rebuilt,
but having forced PACKAGE_ARCH to a particular value the packages for
each architecture were stepping on eachother. Since the packages are
empty they can in fact be allarch (even though they won't actually go
into the "all" package feed). It turns out that nheriting nativesdk
wasn't actually necessary either, so drop that.

Fixes [YOCTO #8509].

(From OE-Core rev: 9b1831cf4a2940dca1d23f14dff460ff5a50a520)

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

xf86-video-vmware: Add vmwgfx PACKAGECONFIG option

(From OE-Core rev: 68c41410ababd69f36ab92f421c0524c264a59b8)

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pkgconfig: merge .bb and .inc

The .inc file included version information and wasn't used anywhere as the same
.bb handles all variations, so merge them to reduce complexity.

(From OE-Core rev: 5c877cd5d211220b145bc88d962673a0e8a7c0c4)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pkgconfig: upgrade to version 0.29

0.28 -> 0.29

(From OE-Core rev: 5e93df29eb13a89bef72a8a163d6d0e63a0c3c98)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ofono: upgrade to version 1.17

Removed the patch 0001-backtrace-Disable-for-non-glibc-C-libraries.patch,
as it is included in the new sources.

(From OE-Core rev: 4ffa6d6950f23afa549116ff2bfb30e687888914)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libxml2: remove legacy LDFLAGS += "-ldl" workaround

libxml2 is able to find libdl.so on its own.

(From OE-Core rev: 148a2d80ea4b095a77e5d4edc1ca964708c3f4d1)

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

apr: fix LTFLAGS to make it work with ccache

When ccache is enabled, libtool requires --tag=CC when use ccache,
otherwise when building apr-util with ccache enabled:

| libtool: compile: unable to infer tagged configuration
| libtool: error: specify a tag with '--tag'
| libtool: compile: unable to infer tagged configuration
| make[1]: *** [buckets/apr_buckets.lo] Error 1

(From OE-Core rev: 52c314963e0ab90e6c1ced7bf66b06d4c304358a)

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

iproute2: install bridge tool by default

bridge is a tool used frequently, so install it

(From OE-Core rev: b1f67c7ea87bb9bd0fb642b33e66ea5595789afc)

Signed-off-by: Roy Li <rongqing.li@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lttng-tools: add libgcc to RDEPENDS

The lttng-sessiond program uses pthread_cancel, whose implementation
in glibc requires libgcc_s.so.1 to be present.

(From OE-Core rev: fff5fe9f356098457c4fd56cb95c3e8afde2defb)

Signed-off-by: Nathan Lynch <nathan_lynch@mentor.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lttng-tools: Upgrade to 2.7 release

(From OE-Core rev: dddf51204238a094f633149663a7ed843dc8f55c)

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lttng-tools: Drop unused patch

The runtest.patch is unused, so there is no reason to keep it on the
tree.

(From OE-Core rev: c1fb471dc83206f3940108319fca760b39cbbac7)

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lttng-ust: Upgrade to 2.7 release

(From OE-Core rev: d63f9a33c26dc1536ea4345c76eb49bd1b00f7a7)

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lttng-modules: Upgrade to 2.7 release

(From OE-Core rev: 72174e357d79c016b48eb4b4d0bb455600e9e34e)

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libunistring: upgrade to version 0.9.6

0.9.5 -> 0.9.6

(From OE-Core rev: f002fbaf69caf001b3345952e6da897f55c27a36)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libtasn1: upgrade to 4.7

4.5 -> 4.7

(From OE-Core rev: bbd0c4d6ac2614486042225312149d32f653955b)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

wpa-supplicant: upgrade to 2.5

wpa-supplicant: upgrade to 2.5

1. upgrade to 2.5
2. remove eight patches since they have been applied in 2.5
3. update SRC_URI, HOMEPAGE and BUGTRACKER to use w1.fi instead

(From OE-Core rev: 80af821d1240a1fc2b32379b75801571db562657)

Signed-off-by: Fan Xin <fan.xin@jp.fujitsu.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

mesa: Make gl libraries RRECOMMEND mesa-megadriver

This makes e.g. core-image-weston actually work out-of-the-box
on hardware that uses megadriver drivers.

(From OE-Core rev: 12c3f7b5501c0035f72c115fa2eac7aaad927057)

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

directfb.inc: force bfd linker for armv7a

Workaround for linker errors seen with armv7a + gold:

 | ../arm-rdk-linux-gnueabi-libtool  --tag=CC   --mode=link arm-rdk-linux-gnueabi-gcc  -march=armv7-a -mfloat-abi=hard -mtune=cortex-a15 --sysroot=.../build/tmp/sysroots/eos -I.../build/tmp/sysroots/eos/usr/include/freetype2 -I.../build/tmp/sysroots/eos/usr/include/libpng16 -Wall -Wstrict-prototypes -Wmissing-prototypes -Wno-strict-aliasing -Werror-implicit-function-declaration -O3 -g2 -ffast-math -pipe -O2 -pipe -g -feliminate-unused-debug-types -D_GNU_SOURCE  -std=gnu99 -Werror-implicit-function-declaration  -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -o directfb-csource directfb-csource.o -lpng16 -ldl -lrt -lpthread
 | arm-rdk-linux-gnueabi-libtool: link: arm-rdk-linux-gnueabi-gcc -march=armv7-a -mfloat-abi=hard -mtune=cortex-a15 --sysroot=.../build/tmp/sysroots/eos -I.../build/tmp/sysroots/eos/usr/include/freetype2 -I.../build/tmp/sysroots/eos/usr/include/libpng16 -Wall -Wstrict-prototypes -Wmissing-prototypes -Wno-strict-aliasing -Werror-implicit-function-declaration -O3 -g2 -ffast-math -pipe -O2 -pipe -g -feliminate-unused-debug-types -D_GNU_SOURCE -std=gnu99 -Werror-implicit-function-declaration -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -o directfb-csource directfb-csource.o  .../build/tmp/sysroots/eos/usr/lib/libpng16.so -lz -lm -ldl -lrt -lpthread
 | .../build/tmp/sysroots/x86_64-linux/usr/bin/arm-rdk-linux-gnueabi/../../libexec/arm-rdk-linux-gnueabi/gcc/arm-rdk-linux-gnueabi/5.2.0/ld: error: directfb-csource.o: requires unsupported dynamic reloc R_ARM_MOVW_ABS_NC; recompile with -fPIC
 | collect2: error: ld returned 1 exit status

(From OE-Core rev: c075d6e514b0a4511007844baf96ad44d7513399)

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

libpng12: update to 1.2.54

1.2.53 -> 1.2.54

No updates in License.The license checksums changed
because of update in Copyright date and update of png.h

(From OE-Core rev: 0ed98246d17504e509f189451cebcfeda305ed31)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libpng: update to 1.6.19

1.6.17 -> 1.6.19

No changes in License.The license checksums changed because of update
in Copyright dates and other restructuring in png.h

(From OE-Core rev: abee587c0b165ff42e9754839adedda0d5240a73)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

orc: update to 0.4.24

0.4.23 -> 0.4.24

(From OE-Core rev: 66e0f5d670837cf823da8ead98148134c40c4ae8)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libpcap: update to 1.7.4

Update libpcap from version 1.6.2 to 1.7.4

LICENSE checksum has changed because of the removal of additional
spaces in the file.

(From OE-Core rev: 0256743e7ad6e430aa2380ce51c84b3ce5928a1f)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

apr-util: add missing RDEPENDS for ptest

testxlate test case fails due to dependency on some character-sets.
Install them.

[YOCTO #8171]

(From OE-Core rev: 814b7589ea857806db7e3d2a09cf5f52df5afd76)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

iproute2: update to 4.3.0

4.1.1 -> 4.3.0

(From OE-Core rev: 9d48756fc3961920ede177a0d94b05071e6a0e0f)

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ruby-native: Depend on openssl-native

This dependency is floating otherwise, It races against openssl-native
and when openssl config does not match with openssl on build host the
build fails occasionally

x86_64-linux/usr/include/openssl/ripemd.h:70:4: error: #error RIPEMD is
disabled.
 #  error RIPEMD is disabled.

Change-Id: I5ff6d8f058ff99c64ad4dc7c0377724071003ae6
(From OE-Core rev: d0c8d98077622a700d92384f676770cb4d6d4f46)

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

db: fix race issue for libdb-6.0.la

Fixed:
libtool: link: `os_map.lo' is not a valid libtool object
Makefile:867: recipe for target 'libdb-6.0.la' failed

(From OE-Core rev: 5e0f27bd0d00dd9aa597d98d0a7d5b4f5d726925)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pango: use ptest-gnome

Instead of hand-coding the ptest integration, use the ptest-gnome class.

(From OE-Core rev: 70674d4ee2a24971d1bab048906fa8d0d2ba2e75)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gst-plugins-bad: improve FILES variables

(From OE-Core rev: 6343321af88543467388f0f9c25707f02651f078)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gstreamer1.0-plugins-base: add PACKAGECONFIG for libvisual

(From OE-Core rev: 9ff66393d7b9b1bb048ae4d476384d37c239a33c)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

python3: fix building nativesdk-python3

When the class nativesdk.bbclass is inherited, it redefines TARGET_CC_ARCH,
in the case of python3, this enables debug, causing an error while linking,
since we dont enable debug during configure theres no declaration of some
functions, this patch makes sure we keep debug disabled, fixing the linking errors.

[YOCTO #8467]

(From OE-Core rev: d4723c609f700180ee808ac3fbbe2225043a8353)

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

python3: Upgrade from 3.4.3 to 3.5

python3-native_3.4.3.bb -> python3-native_3.5.0.bb:
- changed version
- changed cheksum
- no license change, just dates

python3_3.4.3.bb -> python3_3.5.0.bb:
- changed version
- changed cheksum
- no license change, just dates

New:
- use_packed_importlib.patch: Fixes importlib on cross-compile environments

Rebased:
- Manifest
- 000-cross-compile.patch
- 020-dont-compile-python-files.patch
- 04-default-is-optimized.patch
- python-3.3-multilib.patch
- distutils3-base.bbclass
- distutils3-native-base.bbclass
- python3native.bbclass

Upstream:
- makerace.patch

Misc:
- pip2 is handled as default on major distros,
modified python3-pip to leave /usr/bin/pip available for pip2
- Fixed importing pip3 from python3 interpreter

(From OE-Core rev: 701ec1977ced1bb08461e6de98b4f63d21cba8a6)

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

python-git: Add missing dependency

Python-git depends on getpass, which is part of python-unixadmin.

If it is missing:

root@qt5022:~# python
Python 2.7.9 (default, Nov 12 2015, 17:41:32)
[GCC 5.2.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import git
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/lib/python2.7/site-packages/git/__init__.py", line 35, in <module>
    from git.config import GitConfigParser
  File "/usr/lib/python2.7/site-packages/git/config.py", line 21, in <module>
    from git.util import LockFile
  File "/usr/lib/python2.7/site-packages/git/util.py", line 14, in <module>
    import getpass
ImportError: No module named getpass
>>>

Reported-by: Dimitrios Katsaros <patcherwork@gmail.com>
(From OE-Core rev: 91165c1b68e49e5d5392068df3412ebb7b0e969b)

Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

guile, mailx, gcc, opensp, gstreamer1.0-libav, libunwind: disable thumb where it fails for qemuarm

(From OE-Core rev: 981626d8cee345d27b7c9d96e941fd6622f47792)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

icu: force arm mode

* otherwise it triggers following ICE:
ERROR: Function failed: do_compile (log file is located at /OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/temp/log.do_compile.21570)
ERROR: Logfile of failure stored in: /OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/temp/log.do_compile.21570
Log data follows:
| DEBUG: SITE files ['endian-little', 'bit-32', 'arm-common', 'common-linux', 'common-glibc', 'arm-linux', 'arm-linux-gnueabi', 'common']
| DEBUG: Executing shell function do_compile
| NOTE: make
| Note: rebuild with "make VERBOSE=1 " to show all compiler parameters.
| make[0]: Making `all' in `stubdata'
| make[1]: Entering directory '/OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/build/stubdata'
| make[1]: Nothing to be done for 'all'.
| make[1]: Leaving directory '/OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/build/stubdata'
| make[0]: Making `all' in `common'
| make[1]: Entering directory '/OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/build/common'
|    arm-oe-linux-gnueabi-gcc    ...  /OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/icu/source/common/ubidiwrt.c
| /OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/icu/source/common/ubidiwrt.c: In function 'ubidi_writeReordered_53':
| /OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/icu/source/common/ubidiwrt.c:643:1: internal compiler error: in patch_jump_insn, at cfgrtl.c:1275
|  }
|  ^
| Please submit a full bug report,
| with preprocessed source if appropriate.
| See <http://gcc.gnu.org/bugs.html> for instructions.
| *** Failed compilation command follows: ----------------------------------------------------------
| arm-oe-linux-gnueabi-gcc -march=armv4t -mthumb -mthumb-interwork -mtune=arm920t --sysroot=/OE/build/shr-core/tmp-eglibc/sysroots/om-gta02 -D_REENTRANT -DU_HAVE_ELF_H=1 -DU_HAVE_ATOMIC=1 -I/OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/icu/source/common -DDEFAULT_ICU_PLUGINS="/usr/lib/icu"  -DU_ATTRIBUTE_DEPRECATED= -DU_COMMON_IMPLEMENTATION -O2 -pipe -g -feliminate-unused-debug-types -std=c99 -Wall -pedantic -Wshadow -Wpointer-arith -Wmissing-prototypes -Wwrite-strings -c -DPIC -fPIC -o ubidiwrt.o /OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/icu/source/common/ubidiwrt.c
| --- ( rebuild with "make VERBOSE=1 all" to show all parameters ) --------
| /OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/icu/source/config/mh-linux:44: recipe for target 'ubidiwrt.o' failed
| make[1]: *** [ubidiwrt.o] Error 1
| make[1]: Leaving directory '/OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/build/common'
| Makefile:141: recipe for target 'all-recursive' failed
| make: *** [all-recursive] Error 2
| ERROR: oe_runmake failed
| WARNING: /OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/temp/run.do_compile.21570:1 exit 1 from
|   exit 1
| ERROR: Function failed: do_compile (log file is located at /OE/build/shr-core/tmp-eglibc/work/arm920tt-oe-linux-gnueabi/icu/53.1-r0/temp/log.do_compile.21570)
NOTE: recipe icu-53.1-r0: task do_compile: Failed
ERROR: Task 6803 (/OE/build/shr-core/openembedded-core/meta/recipes-support/icu/icu_53.1.bb, do_compile) failed with exit code '1'

(From OE-Core rev: 85b2ddf06ec69607c39b9fb4cea4a4486e34ad7e)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

rpcbind: Security Advisory - rpcbind - CVE-2015-7236

rpcbind: Fix memory corruption in PMAP_CALLIT code

Use-after-free vulnerability in xprt_set_caller in rpcb_svc_com.c in
rpcbind 0.2.1 and earlier allows remote attackers to cause a denial of
service (daemon crash) via crafted packets, involving a PMAP_CALLIT
code.

The patch comes from
<http://www.openwall.com/lists/oss-security/2015/09/18/7>, and it hasn't
been in rpcbind upstream yet.

(From OE-Core rev: cc4f62f3627f3804907e8ff9c68d9321979df32b)

Signed-off-by: Li Zhou <li.zhou@windriver.com>
Signed-off-by: Wenzong Fan <wenzong.fan@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

subversion: fix CVE-2015-3187

The svn_repos_trace_node_locations function in Apache Subversion before
1.7.21 and 1.8.x before 1.8.14, when path-based authorization is used,
allows remote authenticated users to obtain sensitive path information
by reading the history of a node that has been moved from a hidden path.

Patch is from:
http://subversion.apache.org/security/CVE-2015-3187-advisory.txt

(From OE-Core rev: 6da25614edcad30fdb4bea8ff47b81ff81cdaed2)

Signed-off-by: Wenzong Fan <wenzong.fan@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

subversion: fix CVE-2015-3184

mod_authz_svn in Apache Subversion 1.7.x before 1.7.21 and 1.8.x before
1.8.14, when using Apache httpd 2.4.x, does not properly restrict
anonymous access, which allows remote anonymous users to read hidden
files via the path name.

Patch is from:
http://subversion.apache.org/security/CVE-2015-3184-advisory.txt

(From OE-Core rev: 29eb921ed074d86fa8d5b205a313eb3177473a63)

Signed-off-by: Wenzong Fan <wenzong.fan@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/sshcontrol: don't source profile

Instead of sourcing /etc/profile to get $PATH including /usr/sbin, just assign
to PATH in the ssh invocation.

The remote /etc/profile may not actually be manipulating PATH as we expect, and
there may be other commands which can interfere with the tests (such as resize
emitting a series of control characters on connection).

(From OE-Core rev: 0f3fb5bbf2fd7db82898fed3281af143387316ff)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oeqa/runtime/multilib: refactor ELF class extraction

Instead of duplicating the same code over and over, split it out to a separate
function.

(From OE-Core rev: 4f870f020bbf908ab87990803f3c278bf4e44843)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oe-selftest: Enable code coverage on unit tests

Enable code coverage through the library 'python coverage'. In case the environment
variable COVERAGE_PROCESS_START is present (one of the requisites for measuring
sub-processes; the second one is including some coverage statements into the
python sitecustomize.py file) it will be taken into account, otherwise it is
exported with value '.coveragerc'. The latter value is a configuration file
(also automatically created) with some default settings. Once tests are
executed, a coverage report is shown on the log and the coverage output data is stored
with name '.coverage.<args>' where '<args>' is the name of the  unit tests executed
or 'all_tests' when running with --run-all-tests. This output data can be latter used
for better reporting using the same tool (coverage).

As briefly indicate before, measuring sub-process implies setting the env variable
COVERAGE_PROCESS_START (done automatically by the oe-selftest code with this patch if
not already set) and creating a sitecustomize.py as explained on [1].
If either one of these is missing, complete coverage will be incomplete.

Current measurements for 'oe-selftest --run-all-tests' indicate that current coverage
is around 42 % taking into account BBLAYERS, bitbake and scripts folders. More details
on [2], indicating the coverage per file/module.

This tasks has been done together with Humberto Ibarra <humberto.ibarra.lopez@linux.intel.com>

[YOCTO #8679]

[1] http://coverage.readthedocs.org/en/latest/subprocess.html
[2] https://bugzilla.yoctoproject.org/attachment.cgi?id=2854

(From OE-Core rev: b3feee2cefbbd98b66dc395b651f47c5028c80a0)

Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

meta/conf/machine: use ' inside quoted values

(From OE-Core rev: 924ccf202a6d89de32fc34a140bf9e35e8e43b4e)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

runqemu-internal: Replace wacom-tablet with tablet for usbdevice

When booting weston-core-image with latest wayland/weston/libinput
mouse/touchpad would not work on qemux86, this fixes the issue

(From OE-Core rev: 4abf18a25ccb1a062a3775be1a70eaf6c155349c)

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

recipetool: make plugin registration function name consistent with devtool

This should have been register_commands rather than register_command;
I used register_commands in devtool so lets change this here to be
consistent with that. (Since this is extensible through layers though we
need to remain compatible with the old name, so fall back to that if the
new function name isn't there.)

(From OE-Core rev: 1047f6592ac81643cd847f104da766dc4a4c81ea)

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

recipetool: add setvar subcommand

Add a recipetool subcommand "setvar" to set a variable in a file. This
uses our existing logic such that it doesn't matter if the variable is
already set in the recipe, if it's set in the recipe or some inc file,
and if the variable is not currently set that the line setting the
variable gets inserted in the right place in the file.

Implements [YOCTO #7676].

(From OE-Core rev: 7c33ef77fa165182d24f0a9ae769e9e630e6bd47)

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

lib/oe/recipeutils: refactor patch_recipe_file() to use edit_metadata()

Use bb.utils.edit_metadata() to replace some of the logic in this
function; this avoids us effectively having two implementations of the
same thing. In the process fix the following issues:

* Insert values before any leading comments for the next variable
  instead of after them
* Insert overridden variables (e.g. RDEPENDS_${PN}) in the correct place
* Properly handle replacing varflag settings (e.g. SRC_URI[md5sum])

(From OE-Core rev: 0f81b83fc5fd908efa7f6b837137830ca65f6ed6)

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

devtool: clarify help text

* Make some minor clarifications to help text
* Drop ArgumentDefaultsHelpFormatter and just put the defaults in the
  text itself where needed (because otherwise you get defaults shown for
  store_true options which is somewhat confusing).

(From OE-Core rev: a90ffea30c4578fd6acda2c5945b816ad33b13f5)

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

devtool: build: enable showing default task in help

Enable access to the configuration object in register_commands() so that
we can read configuration values there; this allows us to show the
task that will be run in the command line help for the build subcommand.

(From OE-Core rev: 142d006de3235a034839ef7bbe147c56fc7af04a)

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