nasm: updated to 2.10.07

(From OE-Core rev: 1880bcddd2532b22542cc6eccbadf889a1749db9)

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

byacc: updated to 20121003

(From OE-Core rev: 264abcace9d00361618bd98a71d4bd477e840e88)

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

rootfs_rpm.bbclass: export INSTALL_PLATFORM_EXTRA_RPM

Export this variable because package_rpm.bbclass needs it.

[YOCTO #3682]

(From OE-Core rev: 64b01c6c945346b661f294fc8a49bf04a72da3c0)

Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

kernel.bbclass: kernel_do_install fails with kernel <2.6.30

 * kernels <2.6.30 don't have $kerneldir/tools directory
 * so we got this error:
 * sed: can't read ...SNIP...linux/2.6.30-r0/image/usr/src/kernel/tools/perf/Makefile: No such file or directory
 * ERROR: Function failed: do_install

(From OE-Core rev: 4dd39a23dfe069ad34a925e00407b4dde978cb01)

Signed-off-by: Marco Cavallini <m.cavallini@koansoftware.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bluez-hcidump: updated to upstream version 2.5

(From OE-Core rev: 0a2cd2d63ebbafff935e923091ac82aaf4142244)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gmp: add patch from upstream to unbreak AArch64

(From OE-Core rev: 561564d7baf2484365a03360ae27bb80ec4ab257)

Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

python-smartpm: add dependency on python-pprint

smart's "config" command uses the Python pprint module. If it is not
present, "config" fails with this error:

 #############
 # smart config --help
 error: Invalid command 'config'
 #############

This patch adds a dependency on python-pprint which fixes the error.
Tested on qemux86/core-image-sato-sdk.

[YOCTO #3643]

(From OE-Core rev: f5f9cfcccd1526572673ffbc5435bf3221498409)

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gettext-minimal: update files from gettext 0.18.2

(From OE-Core rev: 3f5706348c8c586359ac7fbccadc4bbd55cd8bc9)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gettext: update to upstream version 0.18.2

remove_gets.patch dropped as upstream now contains
fix for the problem.

(From OE-Core rev: 9e936c390006602c27df4c42ea2d9055881db098)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

telepathy-mission-control: upgrade to 5.14.0

(From OE-Core rev: ecfdc5e12cc9d068eb8226020bdf81bb5f790444)

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

telepathy-idle: upgrade to 0.1.14

(From OE-Core rev: dfd69a669c57200bc3846709483db4c82c87d300)

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

libtelepathy: python-native is now used during build

libtelepathy generated makefiles contain an
assignment like this: PYTHON=python
This could lead to potential build problems
in Yocto environment because python-native
is not used during the build.

Fixes [YOCTO #3679]

(From OE-Core rev: 9e7f5dcb51c547ccec014952ee7c82f9a222d187)

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

libpcre: change BUILD_CPPFLAGS and BUILD_CFLAGS values

Set LINK_SIZE in BUILD_CFLAGS given that the autotools bbclass use it to set
CFLAGS_FOR_BUILD, required for the libpcre build.

(From OE-Core rev: 4f1a2bf43e286e630c30738a268df76355e19212)

Signed-off-by: Giuseppe Condorelli <giuseppe.condorelli@st.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

guile: add explicit dependency to avoid parallel build issue

Add explicit dependency for libpath.h on dynl.x which fixes a
potential parallel build issue.

[YOCTO #3558]

(From OE-Core rev: 0cfec10c4c7b0597f0e0c8f85539d901861a2f83)

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

sstate.bbclass:specify function dirs to avoid race

Specify dirs in which the shell function "sstate_create_package" and
"sstate_unpack_package" are executed and don't use ${B} as default dirs
to avoid possible race with task do_rootfs at deb image creation time.

[YOCTO #3674]

(From OE-Core rev: ccef1cf783669a4683eda9d4b44dbe6bcf426259)

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

site/common: deactivate a runtime check for ipv6-support in python >=2.7.1 that fails when cross-compiling

In configure there is a runtime check to determine buggy getaddrinfo and as we
are cross-compiling this check goes wrong. If ipv6 is enabled in python, the
failed test is reported as fatal error with the message
Fatal: You must get working getaddrinfo() function.

Setting ac_cv_buggy_getaddrinfo=no deactivates the runtime check and allows to
compile python with ipv6 enabled.

This commit was derived from the commit 700b75e7661062aa93cf81205b78c8bf7609922d
in the Classic OpenEmbedded Development Tree.

(From OE-Core rev: 95cc8015a867235bed6d6cde08ab87f5351a66c7)

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

runqemu-internal: add 'debugshell' as a default kernel option

Add 'debugshell' as a default kernel option for ramfs booting.

If rootfs.img cannot be found under /media, init-live.sh loops
forever without showing any information. Silently looping forever
and blocking users is inappropriate.

Now that the 'debugshell' feature has been implemented in init-live.sh,
It's reasonable to add it to the kernel option when booting a ramfs-based
image. In this way, the system doesn't loop forever and instead drops
to a shell after a default timeout (30 seconds).

(From OE-Core rev: 0ffdfabc5ab5f6303aff11f8ea511875f1eaf0f5)

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

wpa-supplicant: Include wpa_supplicant.service file in package for dbus to work

The dbus service files include a reference to wpa_supplicant.service, but if it
does not exist the dbus-daemon warns:

[system] Activating via systemd: service name='fi.w1.wpa_supplicant1' unit='wpa_supplicant.service'
[system] Activation via systemd failed for unit 'wpa_supplicant.service': Unit wpa_supplicant.service failed to load: No such file or directory.  See system logs and 'systemctl status wpa_supplicant.service' for details.

If this happens, wpa_supplicant is not run automatically.

(From OE-Core rev: 600c07c4391dc44638e9c920c3c402b6588d1259)

Signed-off-by: Stan Hu <stanhu@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

buildhistory-diff: prepend to sys.path

This ensures that regardless of what else is in our search path, we always use
the bitbake corresponding to the running buildhistory-diff command.

(From OE-Core rev: 246990e8a3da01020f65d151971883740eacb378)

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bash: fix mkbuiltins build failure

Same patch for non-gplv3 version:

http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=1fd9a16d2a4594a4e9179dc7353ac51ce32eb712

[YOCTO #3646]

(From OE-Core rev: d00acdbfa7d10804ff832009888c441fda51e412)

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

.gitignore: Git ignore pull requests and built manuals

We ignore the in-tree built manuals, and
the in-tree pull requests.

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

perl: Remove bashism from perl-tests.inc

The perl-tests.inc file where using the constructs:

 cp {a,b} to/

but this is not supported in every shell so we cannot use it. This
patch fixes it.

(From OE-Core rev: 8dd608dac1e5658197078d002dc1b3dc8ac83184)

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

linux-libc-headers: make 3.7 the default and remove older libc-header recipes

Switch the default libc-headers to the 3.7 version. At the same time, remove
older versions of the headers to keep things simple and clear. All userspace
and kernel combinations should build and boot against this single lib-headers
version.

(From OE-Core rev: e7c9706d6a6777326a62e73bffdbb0f940792ff4)

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

linux-libc-headers: add 3.7 version

(From OE-Core rev: 7f1672f9513f05a98197632fe24279ecf8a58505)

Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-yocto/3.4: update to v3.4.24

Updating the 3.4 recipes to v3.4.24 stable version.

(From OE-Core rev: c3e97905f423e2bab2a0aadf5099ef445f0dd138)

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

linux-yocto/3.4: integrate v3.4.21, v3.4.22, v3.4.23

Updating the SRCREVs to integrate the korg -stable updates.

(From OE-Core rev: 88856144bf97630f4c43fd7c984343caf287a435)

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

libart-lgpl: add AArch64 support

(From OE-Core rev: 928f03017886eb5f4869c559de3d61351fe801a3)

Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

xf86-video-intel: upgrade to 2.20.17

(From OE-Core rev: 42850ed397f8a8dca21a5b95ca0dd0ab61e646b6)

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ghostscript: add AArch64 support

(From OE-Core rev: e3065f681720abd79b4f4cf545460f249276f3e8)

Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

qt4: blacklist untrusted SSL certificates

- this blacklist wrong certificates
https://bugreports.qt-project.org/browse/QTBUG-24654
https://bugreports.qt-project.org/browse/QTBUG-28937
- these patches will be in the next 4.8.5 release

(From OE-Core rev: aafcf34aa8be3525ada517b770e43ad05de5a4b6)

Signed-off-by: Eric Bénard <eric@eukrea.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libnl: Fix building using flex

The build system has not been fully changed to use flex in 3.2.16 so
we need to finish the conversion. In 3.2.17 this bug has been fixed.

(From OE-Core rev: 98c1c4d995416d6355354867d10281c10e9fc9aa)

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

bitbake: build.py: Fix traceback when there are no dependencies

A recipe with no dependencies results in a traceback (e.g. all in
ASSUME_PROVIDED). This shouldn't happen and this patch fixes it.

(Bitbake rev: dee7decf87dfa8cb966fe40846d27f3e6ab1846b)

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

gsettings-desktop-schemas: add missing elements from meta-gnome

- add postinst using gsettings.bbclass
- modify the recipe according to the meta-gnome variant

(From OE-Core rev: 337add4a13275b906a98476321f48d75141c5399)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

copyleft_compliance: force the symlink creation

(From OE-Core rev: 74b82fd4cc3e5eec611800935f1c89db3bac9971)

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Gilbert Coville <gilbert_coville@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

copyleft_compliance: grab the scm mirror tarballs

If the mirror tarball is available, it's a better choice for this than the
repository directory.

(From OE-Core rev: 8183962869363731acf57f2f531da737512f0f2f)

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Gilbert Coville <gilbert_coville@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

git: updated to 1.8.0.2

(From OE-Core rev: 6b437bf245e21fea2a792ebc3919ef4e42b0283f)

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gcc: restore FILESPATH

restore FILESPATH to previous setting because the new one doesn't allow the
normal use of FILESEXTRAPATHS_prepend in the case of append recipe.
In the current implementation, in fact, the FILESPATH is statically set into
the gcc-4.7.inc and so the base_set_filespath is not invoked to manage
extrapaths.

(From OE-Core rev: acdec40f31ecd2eecf8c173b632b8fb01ca59b3d)

Signed-off-by: Giuseppe Condorelli <giuseppe.condorelli@st.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oprofile: updated to 0.9.8

(From OE-Core rev: e8150e3941a471bf7e6d375595c15d54b809b15c)

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

tcl: updated to version 8.5.13

(From OE-Core rev: 771a8851966d256b40f49fcd4352ea2c704fd985)

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

nspr: added gnu-configize for AArch64 support

(From OE-Core rev: 8643c340c738bb487edad8033ebae760c65b6e35)

Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libtasn1: remove help2man dependency

Remove manpage creation. It wasn't working because of help2man
missing when libtasn1 is being built. This attempt to create
manpages without help2man turns from no-op to hard error with
automake-1.13.

(From OE-Core rev: 0a9797c4c0409e26562b80f168409c3f86bc261c)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bison: remove help2man dependency

Remove manpage creation. It wasn't working because of help2man
missing when bison is being built. This attempt to create
manpages without help2man turns from no-op to hard error with
automake-1.13.

(From OE-Core rev: a2f4b4437c6d542349d2027330dc16392ed131ae)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libtool: remove help2man dependency

Remove manpage creation. It wasn't working because of help2man
missing when libtool is being built. This attempt to create
manpages without help2man turns from no-op to hard error with
automake-1.13.
Removed prefix-manpage-fix.patch as only code it touched is being
removed by new dont-depend-on-help2man.patch

(From OE-Core rev: 51a97cfb96ee2e5eeef978678fe6acdd498b9555)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libgdata: remove recipe

Part of the sato trimming process

(From OE-Core rev: cff5f9494f5bbb0df00ba8733d5356d381dde14b)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

relocate_sdk.py: new interpreter string was not '\0' terminated

The problem: SDK binaries were not properly relocated when the SDK
was installed into a path that had a length less than the default one.
Apparently, there were two problems here: the padding was done wrong
(the size of one program header table entry was used instead of the
program section size) and the new padded string was not used at all.

[YOCTO #3655]

(From OE-Core rev: 3815030c5f31c11495893c1ae28d56c1aff31d97)

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

util-linux: moved 'reset' into own subpackage

The 'reset' tool from util-linux is a shell script calling 'tput' which is
provided by 'ncurses'.  To avoid a 'util-linux' -> 'ncurses' dependency,
'reset' is put into an own subpackage which can be installed on demand.

Because most (all) systems are using 'busybux' which provides an
alternative 'reset' implementation, this should not cause too much
regressions.

(From OE-Core rev: 0ed77fe790a76a530b173eefaba648b1c22a4a3e)

Signed-off-by: Enrico Scholz <enrico.scholz@informatik.tu-chemnitz.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

procps: initialized environment correctly (backported from classic)

Without this patch, various procps tool can create "Unknown HZ value"
messages because internal initialization happens in a wrong, random
order.

It has been applied to classic already.

(From OE-Core rev: a746cf196792d1c0646f5e97e87055e0974d753e)

Signed-off-by: Enrico Scholz <enrico.scholz@informatik.tu-chemnitz.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

image.bbclass: quote TARGET_VENDOR

The buildsystem does not install locale files anymore, when
TARGET_VENDOR is empty.  This regression was introduced by
ff01518c6667e1b6e87df4e5a435a701d0f38fa7

(From OE-Core rev: 7d64232f370817c0536412589bd4ed612f214e18)

Signed-off-by: Enrico Scholz <enrico.scholz@informatik.tu-chemnitz.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

xserver-xorg: add AArch64 support

Launchpad bug: https://bugs.launchpad.net/linaro-aarch64/+bug/1092128

(From OE-Core rev: aa2d9049a9a0f73123bf3d7e9221552ff49dab8a)

Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

base-files: Remove the redundant "/"

I'm working on my virt-test, when I check this file find it has
redundant "/"

(From OE-Core rev: f67c0af055b7817f1bb931c98f22f1e3b540b10f)

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

scons: updated to 2.2.0

The license file was updated with a "2012" in the "Copyright" line,
nothing else changed in the license text.

(From OE-Core rev: c8e8ab6b69082805d62193e382534ee93a21f416)

Signed-off-by: Bogdan Marinescu <bogdan.a.marinescu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

perl-rdepends: be more strict in multiline comment

* fails to parse when http://patchwork.openembedded.org/patch/40997/ is used
Parsing recipes...ERROR: There is a confusing multiline, partially commented expression on line 10 of file
/OE/shr-core/openembedded-core/meta/recipes-devtools/perl/perl-rdepends_5.14.2.inc (| sed "s/\/.*\.pm: */ += /g;s/[\"\']//g;s/;.*/\"/g;s/+=
.*\(require\|use\)\> */+= \"perl-module-/g;s/CPANPLUS::.*/cpanplus/g;s/CPAN::.*/cpan/g;s/::/-/g;s/
[^+\"].*//g;s/_/-/g;s/\.pl\"$/\"/;s/\"\?\$/\"/;s/(//;" | tr [:upper:] [:lower:] \).
Please clarify whether this is all a comment or should be parsed.
ERROR: Unable to parse /OE/shr-core/openembedded-core/meta/recipes-devtools/perl/perl_5.14.2.bb: Exited with "1"
ERROR: Command execution failed: Exited with 1

(From OE-Core rev: daee2d2dc29c14a086122c04fbf97924c5dab3b8)

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

cmake: add AArch64 support

(From OE-Core rev: 73dd5842a945e119c8870929e9d71067235a9247)

Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libxcomposite: update to upstream version 0.4.4

Compatible with automake-1.13.

(From OE-Core rev: f63704c6b4686fde437d247f41e081cfc5c1868e)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

polkit: update SRC_URI and HOMEPAGE

Author of old code.google.com homepage lists polkit under heading
'former projects'. Newer versions are available from page at
freedesktop.org. Freedesktop.org homepage also points to SRC_URI
different from what we've been using. Latest version there is 0.109,
while latest version in old SRC_URI is 0.104.

(From OE-Core rev: 36fda78f7ef80a639bf5d760a611c3f2d0327b63)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

directfb: Rename __no_instrument_function__ macro to avoid conflicting redefinitions

The macro was renamed to __dfb_no_instrument_function__ . This will avoid build
errors while compiling gst-plugins-bad with directfb support.

This happens because gstreamer uses this macro to pass it to __attribute__. So,
if we include directfb.h before, the gstreamer definition will end up:
__attribute__ ((__attribute__((no_instrument_function)))).

(From OE-Core rev: 3e87b2b796e2de50adcdf5ed9d777765ae8e3cee)

Signed-off-by: Andrei Gherzan <andrei@gherzan.ro>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

prelink: update to git head

Compatible with automake-1.13.

(From OE-Core rev: 3d8f35a121d54b7a1b825b266a553752e22bf708)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

scripts/sstate-sysroot-cruft.sh: add simple script to find files in sysroots not tracked by sstate

(From OE-Core rev: 18bf0b4e14a74a91dff1a282c946cab37678b2fe)

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

scripts/sstate-diff-machines.sh: add simple script to compare sstate checksums between MACHINEs

* takes tmpdir, machines and targets from command arguments or env variables

(From OE-Core rev: 1e4bdd6147c73547d2451705bbb874918621cbfc)

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

sed: upgrade to 4.2.2

(From OE-Core rev: 16500f9d5fe3446511212cff14d44bce969cdea0)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

icu: update to upstream version 50.1.1

This seems to fix HarfBuzz compile problems.

(From OE-Core rev: 6e1e924f882c1a5f5e5932c0952f6a2c906dd51c)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

package.bbclass: Skip testing "packages" a second time.

Since the "packages" variable is already tested a few lines above in
the same routine:

    if len(packages) < 1:
        bb.debug(1, "No packages to build, skipping do_package")
        return

there is no point testing it again, and the error message doesn't even
refer to "packages", anyway.

(From OE-Core rev: 371fbacff47828cfff99f1152f9e269e27c474da)

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

vte: add missing unpackaged files

(From OE-Core rev: 791d88336fb8312041c30ea744d03d8d150d2b8f)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libnl: Update to 3.2.16

(From OE-Core rev: bf039ff80c2a416ddb793b676ab2940428695b3e)

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

cairo: Adds libxext in X11DEPENDS.

* cairo was giving redefinition error of struct _XLinearGradient,
struct _XCircle and struct _XRadialGradient.
* cario now depends on libxext. Adding that dependency resolved
above errors.
* Verified it on P2020RDB.
* Bump PR.

(From OE-Core rev: 409a739dd89ea256a49c69452dc29439fc3ee64e)

Signed-off-by: Noor Ahsan <noor_ahsan@mentor.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib_package.bbclass: Correct comment referring to bin directories.

  FILES_${PN}-bin = "${bindir}/* ${sbindir}/*"

so the comment should match that.

(From OE-Core rev: 81bc0b96a085742f7dc9e8e90bb32529efbaf920)

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

mpfr: update to upstream version 3.1.1

mpfr_fix_for_automake-1.12.patch part of upstream now.

(From OE-Core rev: 49dd3fb76c59b5baeec4820575db4e8e9c17d417)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

chkconfig-alternatives-native-1.3.59: add recipe

This only builds alternatives, so avoids the libnewt and libpopt dependencies.

This uses 1.3.59 sources from git, with a couple commits of mine to add
--sysroot (and OPKG_OFFLINE_ROOT) support.

LIC_FILES_CHKSUM changed from 1.3.58 to 1.3.59, but only due to minor
formatting differences, not content/terms.

(From OE-Core rev: 7895540e02a563c107934572fdbf23fdf1661c91)

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

chkconfig: package the update-alternatives implementation

(From OE-Core rev: c18b1ce2d98460e5ee6ec6319071bc1fa1016ede)

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

chkconfig: obey sysconfdir, base_libdir

(From OE-Core rev: afd9f47080b36d609297e2eb448bc12af8000d29)

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

chkconfig: don't inherit autotools

This buildsystem is not autoconf/automake, but make.

While we're at it, also make the install not hardcode the path to
/usr/share/locale.

(From OE-Core rev: 0e6f76d9d5929f8c23916827dccc8389419f0163)

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

opkg-native: obey virtual/update-alternatives-native

(From OE-Core rev: 1f9d979ca60eca2277f53f24dc59f9ab18e81195)

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

update-alternatives.bbclass: use absolute paths for link targets

This improves compatibility, as both the debian update-alternatives and the
chkconfig alternatives require absolute paths.

(From OE-Core rev: 2b8bf64f1f6999b65ddd3efc2c6ddb7e70fead53)

Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

init-live.sh: avoid duplicate mount points for the same filesystem

A live system had duplicate mount points for the same filesystem
in the output of 'mount' command. That was because we didn't handle
the temporary mounts properly before switch_root.

This patch fixes this problem by moving the mount points of some
filesystems to the corresponding directories in the real root filesystem.

[YOCTO #3155]

(From OE-Core rev: 050a745413f8bb26212b84da391e0f7665f77728)

Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gmp: update to upstream version 5.1.0

New version of configure.patch added for gmp-5.1.0. Old patch
moved to gmp-4.2.1 specific directory.
gmp_fix_for_automake-1.12.patch part of upstŕeam so removed
gmp-5 version, gmp-4.2.1 version remains.

(From OE-Core rev: ebd56f6a12c55982467b34b9c4feae03358738fe)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gmp: update SRC_URI and HOMEPAGE

Fetch from gmplib.org instead of gnu mirror.
List gmplib.org as homepage instead of old broken URL.

(From OE-Core rev: 64dbe42e45fb5b02c7b06bcf40ae93aa37704ce4)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pcmanfm: upgrade to 1.1.0

pcmanfm_fix_for_automake_1.12.patch: removed
- problem not reprodicible anymore

owl-window-menu.patch: removed
- not needed anymore

(From OE-Core rev: 14fb53c398d7ce5d33c221655643ec51c3a92a6f)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libfm: upgrade to 1.1.0

libfm_fix_for_automake-1.12.patch: removed
- problem not reprodicible anymore

configure_fix.patch: removed
- not used

fix-make-parallelism-issue.patch: added
- assures that the ${includedir}/libfm symlink
is created

(From OE-Core rev: 013d96a601bfb9cc8a2ad51f456e0201666d7bc3)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

automake: update to upstream version 1.12.6

python-libdir.patch and py-compile-only-optimized-byte-code.patch
updated to apply to new codebase.

(From OE-Core rev: f2a530be5941ddf708720b920d0213d7d3928cd1)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

kernel.bbclass: include PE in KERNEL_IMAGE_BASE_NAME

* without this you it's harder to see which kernel is newer version
  e.g. after branch change LOCALCOUNT in SRCPV is reset to 0, so it needs PE bump for upgrade path
  but then it's clear why gitr1+HASH has newer DATETIME then gitrN+HASH, so include PE in name
* it will sort alphabetically wrong in deploy dir for a while until old images are removed

(From OE-Core rev: 3b5fe6d2cd58b5db70c3d319ffc66b1cbff1f43a)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

rpm: replace /usr/lib with ${libdir}

Fix the following warning:
WARNING: QA Issue: rpm: Files/directories were installed but not shipped
  /usr/lib64/rpm/bin/api-sanity-checker.pl

(From OE-Core rev: 625c93784940c0dcda4570258acf2183af7ec5f8)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

sysstat: fix sa_lib_dir

Fix the following warning:
WARNING: QA Issue: sysstat: Files/directories were installed but not shipped
  /usr/lib
  /usr/lib/sa
  /usr/lib/sa/sadc
  /usr/lib/sa/sa2
  /usr/lib/sa/sa1
  /usr/lib/sa/.debug
  /usr/lib/sa/.debug/sadc

(From OE-Core rev: 0da1e769a57c96299dce93b34ec67c2809da8998)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

eglibc-locale: enable multilib

(From OE-Core rev: e5890474ea6ae44f16e66795aa8e3274ec576aa7)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

package.bbclass: don't prepend MLPREFIX to LOCALEBASEPN

- all the recipes that overwrite LOCALEBASEPN must consider
also the MLPREFIX
- if the LOCALEBASEPN variable is not overwritten then it will
have the correct prefix (LOCALEBASEPN ??= "${PN}")

(From OE-Core rev: 855d64a3091c5358e2e1a7f4247929515b068708)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libpam: enable multilib

Fix the following warning:
WARNING: QA Issue: lib32-libpam: Files/directories were installed but not shipped
  /usr/sbin/unix_chkpwd
  /usr/sbin/pam_tally2
  /usr/sbin/mkhomedir_helper
  /usr/sbin/unix_update
  /usr/sbin/pam_timestamp_check
  /usr/sbin/pam_tally

(From OE-Core rev: 1dbb0bc048ceaf4cd2879362ae4a0f1398809349)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

perl: update dependency creating script

Base on dependency create script provided by Saul, I update it with:
1 remove document strings in perl module files that may import fake
  dependencies.

2 not all the 'use' or 'require' clauses start at the begin of line, and
  some of them don't end with ';', such as:

use vars qw(
            $AUTOLOAD
    ...

3 filter some strings that they are not real dependencies: some of them
  are wrong result of script, some are conditional use/require, and some
  are cpan extra modules that we don't need.

(From OE-Core rev: 35291c97edc05756788d206457c56dcc41edb637)

Signed-off-by: Kang Kai <kai.kang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

perl: update RPROVIDES and popuate_package script

File warnings/register.pm has been moved to package perl, update
RPROVIDES_perl to provides perl-module-warnings-register for backward
compatibility.

Update populate_package script for:
1 Way to deal directory 'Module/' is not right, it creates packages
  whose names end with '-pm'. Update to drop these packages.
2 Deal directory 'auto/' to put same name module file(.pm) and library
  file(.so) into one package. That because .pm file requires the same
  name .so file at runtime.

[Part of Yocto 3338]

(From OE-Core rev: e8e03deb1eb8fbbf297ad4b07f7e1642102ae308)

Signed-off-by: Kang Kai <kai.kang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

perl: add sub-package perl-tests

In perl source code, run "make test" will run perl test cases. Put these
cases into a seprated sub-package then we can run the perl test on
target.

Add following module to nonxs_ext to install extra modules that required
by perl test cases:
CPAN/Meta CPAN/Meta/YAML Dumpvalue Env File/CheckTree HTTP/Tiny I18N/Collate
JSON/PP Perl/OSType Version/Requirements

Commands to run test:
cd /opt/perl-tests/t && ./TEST

[Yocto 3296]

(From OE-Core rev: 96fb96ddd00881947875518df832fa74faec8a45)

Signed-off-by: Kang Kai <kai.kang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

autoconf: update RDEDENDS

The only file Glob.so provided by perl-module-file will be moved to
package perl-module-file-glob. So remove perl-module-file from RDEDENDS.

(From OE-Core rev: c76ae248334606d10ab730b1b2e9f626ff3c5809)

Signed-off-by: Kang Kai <kai.kang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Added ability to parse python sources to create-recipe

Hi,

Added python source parsing abilities to create-recipe.

(From OE-Core rev: 417357cbcd3ecc7a2e43f87c618c8fdfe04a9d33)

Signed-off-by: David Nyström <david.nystrom@enea.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

pango: have postinstalls run at do_rootfs time

Since pango-native is built anyway and all the modules are in the native
sysroot, create the cache file by scanning those files instead of the
target files. The latter will fail because the shared objects wouldn't
be from the same ELF class.

[YOCTO #3600]

(From OE-Core rev: 75334cbee1f064b38f167269917e333989802e7d)

Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

freetype: update to 2.4.11 which includes fixes for CVE-2012-{5668, 5669, 5670}

Multiple security issues were reported by Mateusz Jurczyk of Google
security team. These have been fixed in freetype 2.4.11. Details are as
follows.

* CVE-2012-5668: NULL Pointer Dereference in bdf_free_font
Bug: https://savannah.nongnu.org/bugs/?37905
Patch:
http://git.savannah.gnu.org/cgit/freetype/freetype2.git/commit/?id=9b6b5754b57c12b820e01305eb69b8863a161e5a

* CVE-2012-5669: Out-of-bounds read in _bdf_parse_glyphs
Bug: https://savannah.nongnu.org/bugs/?37906
Patch:
http://git.savannah.gnu.org/cgit/freetype/freetype2.git/commit/?id=07bdb6e289c7954e2a533039dc93c1c136099d2d

* CVE-2012-5670: Out-of-bounds write in _bdf_parse_glyphs
Bug: https://savannah.nongnu.org/bugs/?37907
Patch:
http://git.savannah.gnu.org/cgit/freetype/freetype2.git/commit/?id=7f2e4f4f553f6836be7683f66226afac3fa979b8

For original e-mail and CVE assignment, see the following URLs:

http://www.openwall.com/lists/oss-security/2012/12/25/1
http://www.openwall.com/lists/oss-security/2012/12/25/2

(From OE-Core rev: b693f6d3d48b281fbbf71fd56996c85e23c3a9c9)

Signed-off-by: Eren Türkay <eren@hambedded.org>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

webkit-gtk: fix build with bison-2.6+

* bison was upgraded to 2.7 in oe-core

(From OE-Core rev: f116dc32240e2f2d3c6e9bc8bc3320862f5116d3)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

iproute2: DEPENDS on iptables

Make sure iproute rebuilds when iptables is updated

(From OE-Core rev: 50b37d4bd90e22c6d129836f896c4a6de66c003f)

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

sqlite: update to upstream version 3.7.15.1

(From OE-Core rev: 936e45390d5eb46ebb04d27559e5dd5bf41b8fb5)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

curl: update to upstream version 7.28.1

(From OE-Core rev: 2cb1285195439faa48571acc5346d25b4de214b4)

Signed-off-by: Marko Lindqvist <cazfi74@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

xf86-video-omap: skip package if opengl isn't in DISTRO_FEATURES

(From OE-Core rev: 4c2434271cfc41e910969266ced9e5f06ee0c732)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

xorg-driver: enable multilib

[YOCTO #3438]

(From OE-Core rev: 87be2314fe0e67a616b4f713106b4ae46931f5cb)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

xserver-xorg: enable multilib

[YOCTO #3438]

(From OE-Core rev: 4c49e76b793bd80422d7b9566d218ca7d087905b)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

multilib: fix allarch/kernel/module-base multilib issues

- skip the non-packagegroup allarch recipes in multilib_virtclass_handler
- extend PROVIDES/RPROVIDES for allarch recipes which are not packagegroups
- use variants from MULTILIB_GLOBAL_VARIANTS (lib32 lib64 libx32) to create
additional pkgdata files for multilib allarch: ${pkgdatadir}/${variant}-${PN}
and ${pkgdatadir}/runtime/${variant}-${pkg}
- use variants from MULTILIB_VARIANTS to create additional pkgdata files
for multilib kernel/module-base recipes
- add a sanity check to determine if the current multilib is in
MULTILIB_GLOBAL_VARIANTS

[YOCTO #2918]
[YOCTO #3440]
[YOCTO #3565]
[YOCTO #3568]

(From OE-Core rev: bc4da2573dfb59ea2fc4af359701818df20f7663)

Signed-off-by: Constantin Musca <constantinx.musca@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>