autogen-native: fix build failure on overloaded hosts

On some overloaded hosts, shell commands of autogen may can not
finish in 5 secs. This has caused many build failures, so increase
the timeout limit to fix this.

(From OE-Core rev: 92c7a0bec3f3c9fec04a6019aa9cee395ba5a9ab)

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

libpam: Fix for CVE-2010-4708

Change default for user_readenv to 0 and document the
new default for user_readenv.

This fix from:
http://pam.cvs.sourceforge.net/viewvc/pam/Linux-PAM/modules/pam_env
/pam_env.c?r1=1.22&r2=1.23&view=patch
http://pam.cvs.sourceforge.net/viewvc/pam/Linux-PAM/modules/pam_env
/pam_env.8.xml?r1=1.7&r2=1.8&view=patch

(From OE-Core rev: 871ae7a6453b3b66610fd8bbaa770c92be850e19)

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

procps: fix that top will quit after cpu offline

top utiliy fails to read /proc/stat after cpu offline, because Cpu_tot
is still the original cpu numbers when calling cpus_refresh, in which
it is trying to read and sscanf Cpu_tot times /proc/stat.

The patch is from procps-3.2.8-2.fc12.src.rpm

(From OE-Core rev: f24aed8d7e41cce277c6eff4ff5ab07b8e39ffff)

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

core-image-weston: add weston-examples to the image

(From OE-Core rev: ea31d45def0e9fc4ed1c1ef92756d807ff59fe72)

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

systemd-udevd: fix invoking init script failed

root@emenlow-noemgd:~# /etc/init.d/systemd-udevd restart
Stopping udevd
Starting udev
corrupt queue file
root@emenlow-noemgd:~# /etc/init.d/systemd-udevd status
udevd is stopped
root@emenlow-noemgd:~# ps
3805 root      8728 S    /lib/systemd/systemd-udevd

The process name is systemd-udevd rather than udev which is
used in systemd-udevd's init script.

[YOCTO #4746]

(From OE-Core rev: 8f59aa06f3eb955d58600dfc628bd3b5d2aa0f9a)

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

linux-yocto/3.8: fix gcc 4.8 ARM boot issues

Updating the linux-yocto-3.8 SRCREVs to fix a boot issue with ARM boards
when gcc 4.8 is used.

Without the following mainline backports:

  f200475 ARM: 7670/1: fix the memset fix
  8215b0e ARM: 7668/1: fix memset-related crashes caused by recent GCC (4.7.2) optimizations

The following trap will be seen on boot:

    [<c00fc3b8>] (kmem_cache_alloc_trace+0x54/0x210) from [<c039f074>] (con_insert_unipair+0xcc/0x11c)
    [<c039f074>] (con_insert_unipair+0xcc/0x11c) from [<c039fec8>] (con_set_default_unimap+0xfc/0x198)
    [<c039fec8>] (con_set_default_unimap+0xfc/0x198) from [<c07ee258>] (console_map_init+0x44/0x58)
    [<c07ee258>] (console_map_init+0x44/0x58) from [<c07ee738>] (vty_init+0x16c/0x1b0)
    [<c07ee738>] (vty_init+0x16c/0x1b0) from [<c07edb68>] (tty_init+0x108/0x148)
    [<c07edb68>] (tty_init+0x108/0x148) from [<c07eead0>] (chr_dev_init+0xb4/0xd8)
    [<c07eead0>] (chr_dev_init+0xb4/0xd8) from [<c0008a18>] (do_one_initcall+0x11c/0x18c)
    [<c0008a18>] (do_one_initcall+0x11c/0x18c) from [<c07d89d0>] (kernel_init_freeable+0x16c/0x254)
    [<c07d89d0>] (kernel_init_freeable+0x16c/0x254) from [<c05a3810>] (kernel_init+0x18/0x160)
    [<c05a3810>] (kernel_init+0x18/0x160) from [<c000e530>] (ret_from_fork+0x14/0x20)
    Code: e593a000 e35a0000 0a000020 e5943014 (e79a1003)
    ---[ end trace e6c62de166779f86 ]---
    Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b

Moderate stress and board testing shows the fix to hold, and it is good for
broader testing.

[YOCTO #4549]

(From OE-Core rev: 8ef393bc08a20053765ad58c1ebd9ab621cdd704)

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

sato-screenshot: bump SRCREV for linkage fixes

Newer toolchains are stricter with linking.  Patches have been merged upstream
so bump the SRCREV to use them.

fix_ldadd_order was also merged upstream, so delete it.

(From OE-Core rev: 8d25bce5d550fff01b97a36ef8e1e3e155e2dee6)

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

matchbox-panel: bump SRCREV for linkage fixes

Newer toolchains are stricter with linking.  Patches have been merged upstream
so bump the SRCREV to use them.

(From OE-Core rev: 55801d72917a75e10c3a470cbd183adce3e67cfc)

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

gcc-4.8: Backport fix for ICE triggered in pixman

* http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57329

(From OE-Core rev: 255c22994449c3d9eb34c80f0a1159ad6714347e)

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

pseudo: drop version 1.4.5 recipe

We've been using 1.5.1 for a while now with success, so in line with the
OE-Core policy of avoiding multiple versions of the same recipe let's
drop this version.

(From OE-Core rev: c26fd671466f42d76cbd99e275dc91fc322062c4)

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

opkg: put libopkg header files into libopkg-dev

The header files for libopkg belong in the package libopkg-dev rather than
opkg-dev where they are placed by default.

(From OE-Core rev: f79cd4208340b56afc044d8e3abc72d43d395062)

Signed-off-by: Paul Barker <paul@paulbarker.me.uk>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

opkg: fix header installation for libopkg

When opkg is built outside the source directory, libopkg headers aren't all
installed as the search path for headers only includes the directory that make
is being ran in, not the source directory. So we fix this by adding
$(srcdir)/*.h to the list of header files to install.

Without this patch, "/usr/include/libopkg" will only include the file config.h.
With this patch, other files such as "opkg.h" are present which are needed to
compile software which uses libopkg.

The new patch has also been submitted upstream as a fix to opkg itself.

(From OE-Core rev: 873689bbabba25e7be5c12317c04519a7bc8d0ef)

Signed-off-by: Paul Barker <paul@paulbarker.me.uk>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oe-buildenv-internal: Ensure error messages got to stderr

(From OE-Core rev: 631a0ec8cc7e53c2df2fc62c5276f940ed45b39e)

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

alsa-lib: check if wordexp is supported in libc

eglibc could be configured to build without the wordexp feature.  To
ensure that the wordexp feature could be used, the configure script must
check if wordexp() is supported in libc in addition to checking if
wordexp.h exists.

(From OE-Core rev: 3c3fe1d139fc84d7ff125f87a4692fac6dfc04e6)

Signed-off-by: Hong H. Pham <hong.pham@windriver.com>
Signed-off-by: Jesse Zhang <sen.zhang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

alsa-lib: Change function type to "static __inline__"

"extern __inline__ function()" is the inlined version that
can be used in this compilation unit, but there will be another
definition of this function somewhere, so compiler will not emit
any code for the function body. This causes problem in -O0,
where functions are never inlined, the function call is preserved,
but linker can't find the symbol, thus the error happens.

since no packages provide atomic_add and atomic_sub, and -O0
Optimize is hoped to keep for debug, we can change extern to
static to fix this problem.

(From OE-Core rev: d0af30c92fdea6f48afb4ab1fde69f1b636e8203)

Signed-off-by: Roy.Li <rongqing.li@windriver.com>
Signed-off-by: Jesse Zhang <sen.zhang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

python: Add nativesdk wrapper to handle terminfo

Currently if you build nativesdk-python, install it and try and run it,
you see it search in the default nativesdk path for terminfo files
when the readline module is used.

If terminfo files cannot be found, or if the ones found are incorrect,
the system may emit control characters which confuse commandline
processing.

This patch sets the TERMINFO_DIRS variable to ensure the correct locations
are searched for terminfo files, starting with the nativesdk terminfo files
and falling back to the host systtem's.

(From OE-Core rev: 682861166f39fbdcd0c9b923139faab2d40362cf)

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

utils.bbclass: Tweak create_wrapper

Currently the full path is specified to exec however this may incorporate a
build path which in the nativesdk case, is not transformed to a target system
path. This leads to failures due to incorrect paths.

This patch ensures the full build path is not encoded into the wrapper script.

(From OE-Core rev: 4569d7403638e4c1f637e3a6adb8dea620dc8543)

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

classes/externalsrc: enable global inherit and simplify usage

This class can now be inherited globally using INHERIT += rather than
needing to inherit it in the recipe itself. Additionally, instead of
setting S (and optionally B), set EXTERNALSRC (and optionally
EXTERNALSRC_BUILD) to point to the external source and build locations
respectively.

(From OE-Core rev: b1da10579a28f9a5260a0678f6f15ce4b5a2706c)

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

licences: Add SGI license

Add SGI license used, for example, in glmark2

(From OE-Core rev: cf59801be372bda962a94e6a406e97d20744ae45)

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

atk: inherit gnomebase

This recipe was missing a dependency on gnome-common, so lets just inherit
gnomebase and get both it and GNOME SRC_URI handling for free.

(From OE-Core rev: 7edc510839f7cd3f83265d634ca0c8accfb2f8c3)

Signed-off-by: Ross Burton <ross.burton@intel.com>
[sgw - atk: add archive tag to SRC_URI checksums]

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

libcap: do not pass CFLAGS to gcc

During do_configure(), we modify the BUILD_CFLAGS used
but do not remove the default inclusion of CFLAGS
in BUILD_CFLAGS.  This fix removes CFLAGS inclusion
by modifying do_configure().

(From OE-Core rev: 99fb3e7408d9fe1b7754dc3994f79e6d5f5b3593)

Signed-off-by: Joe Slater <jslater@windriver.com>
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

directfb:filter out -fno-omit-frame-pointer option on x86 arch

directfb need -fomit-frame-pointer option of gcc to build some inline
asm code about mmx. But once -fno-omit-frame-pointer was added
into TARGET_CFLAGS. That will cause directfb build error on x86 arch.

(From OE-Core rev: 07f4030909dcc14c4ce4d6d3690a192c0b4040a9)

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

openssh: obey 'tcp-wrappers' PACKAGECONFIG

(From OE-Core rev: 4defab3588a5a5e6e3fcdb8007eade18120094f6)

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

openssh: fix a unaligned memory access issue

Backport patch to fix segment fault due to unaligned memory access

(From OE-Core rev: c2ce8e3bc10aec4cb53faea091ad867bab405bb7)

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

openssl: Add fix for cipher des-ede3-cfb1

Add patch file for one of the ciphers used in openssl, namely the cipher
des-ede3-cfb1. Details of the bug, without this patch, can be found here.
http://rt.openssl.org/Ticket/Display.html?id=2867

(From OE-Core rev: ed61c28b9af2f11f46488332b80752b734a3cdeb)

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

bind: backport six CVE patches

(From OE-Core rev: de1238a589ade1220d51cb4b9277cc17479f6f17)

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

archiver.bbclass: check if package contains a copyleft license

The copyleft filter is only excluding packages that contain a
closed source license. This is because oe.license.is_included()
returns a boolean value that indicates if the license is excluded,
and a string that contains the matched included licenses. If the
string is empty it indicates that no licenses were matched.

Reject packages that do not contain a copyleft license.

[YOCTO 4630]

(From OE-Core rev: 3d0f9ee3d2fcce331d35467d5965ff44b825427f)

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

webkit-gtk: Support build on mips64

(From OE-Core rev: 3dd21bc6f70a2b157936257c9673dee6ce7f90bd)

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

dhcp: drop noattrmode.patch

This change no longer needs to be backported now that a newer version
of dhcp is being used. The patch isn't causing errors because it is not
included in the dhcp recipe's SRC_URI.

(From OE-Core rev: 4292bef50880283659dcd68198fd7948a6f48dd5)

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

dhcp: drop fixincludes.patch

Recent versions of dhcp have swapped out use of __u16 for u_int16_t and
therefor asm/types.h no longer needs to be included. Drop fixincludes.patch
from dhcp recipe.

(From OE-Core rev: 9b8f5818ada97f0803eae7687263246761e9d543)

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

gtk+3: fix repainting under Weston 1.1

Weston 1.1 is stricter with the protocol, so fix the behaviour in GTK+.

(From OE-Core rev: 3eb9bbea6f08171d0598c48ca65256c2afc92676)

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

gtk+3: clean up libtool link creation to avoid errors in configure log

(From OE-Core rev: 22c469fea28a9e1eb05374b362743545bdf28ec2)

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

gtk+3: split into .bb/.inc

I plan on introducing a git master version for automatically testing upstream
development, so this split makes things a lot easier.

(From OE-Core rev: f245b1f37fc326c85d19a8a48784c44d15b08ea7)

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

gtk+3: register GSetting schemas

The schemas were not being registered, which resulted in a crash when using the
file chooser or colour picker widgets.

(From OE-Core rev: 49220852bdcf7ebd2e78ef500b93b6ac9806d5f7)

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

gtk+: remove spurious libgcrypt dependency

This dependency was for the display migration patch set, which was finally
dropped entirely in 2007.

(From OE-Core rev: d6875015e7c4660fabb58727c2ce0febf2f7307f)

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

gtk+3: add dependencies for gtk+3-demo

The demo loads PNG files and sniffs mime types, so ensure that these
dependencies are present so the demo actually works.

(From OE-Core rev: 3d678e0c49840a703db028ab82d4535145cd06d2)

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

gtk+3: respect x11 and wayland DISTRO_FEATURES

(From OE-Core rev: cdb7b25f0df633b5498988b3037a58edd361a869)

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

gtk+3: explicitly disable introspection

If someone is/has been using meta-gir then it's possible for this floating
option to become enabled.

(From OE-Core rev: 0f761f528b07b0063da4a5a2d015e4acc8a4aa89)

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

gtk+3: update to 3.8.2

Drop cross.patch as it was a backport and is integrated upstream.

(From OE-Core rev: 8148ed73914e631ef88fdaf2f45d4e6f63b626ce)

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

at-spi2: add -core and -atk, for GTK+ 3.8

(From OE-Core rev: d92ffafbdced78b775ef6b4a908fa2532383b984)

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

bitbake: bitbake: read BBSERVER environement for remote server

This change allows bitbake to connect to a remote server
by reading the BBSERVER environment variable and effecting
the config. Basically a shortcut to the command line parameters.

(Bitbake rev: 81929f86c57ed0a4ad0cda7aaa820fceabaa61e9)

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: knotty, bitbake: add option to terminate a remote server

I add an option to terminate a remote server gracefully
as not to need a kill command.

(Bitbake rev: 7495f835666a9561c2c7d84da7aaa74e4df55b9a)

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: knotty, xmlrpc: add observer-only mode

I add an observer only mode for the knotty UI and
the XMLRPC server that will allow the UI to register
a callback with a server in order to receive events.

The observer-UI is able to send read-only commands to the
server, and also is able to register as an event handler.

Read-only commands are the commands that do not change
the state of the server and have been marked as such in
the command module.

The observer can switch to a full client if it calls addClient
at any time, and the server has no other client running.

(Bitbake rev: 4de9ee21f1fa4d04937cc7430fb1fc8b7a8f61e2)

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

distrodata.bbclass: fix syntax errors

Thes were added by the recent event handler changes.

(From OE-Core rev: 166313ca2edd05f7d3472ba23ba6f44ae12a7d12)

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

dbus-wait: Upgrade to fix typo

There was a typo, the fix was merged upstream, this patch
updates to the new SRCREV.

(From OE-Core rev: ae089fddda333228d8589b7951fa96232accdb6a)

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

oe-buildenv-internal: Fix exit verses return

This script is sourced so we should return, not exit.

(From OE-Core rev: bde54b4c4f8de696666a8c79c8cb6cf224c246e1)

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

bitbake: utils: Fix multiprocessing pool reference

Interestingly the previous version comparision was broken and we were always using
the compatibility code, masking this bug. Oops.

(Bitbake rev: d48e8bcb24e8fa5d4fd60fd2c9927a95976d8d8b)

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

bitbake: compat/server/utils: Jettison pre python 2.7.3 workarounds

Now we've moved to require python 2.7.3, we can jettison the compatibility
workarounds/hacks for older python versions.

(Bitbake rev: a51c402304f2080a76720f9b31d6dfdbed393bba)

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

bitbake: runqueue: Improve handling of failing setscene tasks with hard dependencies

If a setscene task has a hard dependency on a task like pseudo-native, its
expected that the setscene task will not run unless the dependency is met.

This adds code to ensure that is the case, otherwise a bug would show up
with a usecase like:

bitbake gnome-common
bitbake pseudo-native -c cleansstate
bitbake gnome-common -c clean
bitbake gnome-common

With the double wrapper script environment, we'd not see issues like
this as it would be masked. The problem theoretically affects code like
useradd too as well as anything using a sstate postinstall.

(Bitbake rev: c54e738e2b5dc0d8e6fd8e93b284ed96e7a83051)

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

documentation.conf: Drop variables for class that no longer exists in OE-Core

(From OE-Core rev: d2e0adc9720b66f56439c574520a1d58eaf0426f)

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

bitbake: xmlrpc: removes the event servers controller

XMLRPC API ran a XMLRPC server at port+2 in order
to provide endpoints to register an event server
in which to dump the events.

This is no longer used, so we remove it.

(Bitbake rev: e171a363913a86e56266f4c9d107110c7f5221e3)

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

qt4: set QMAKE_COMPILER to gcc in mkspec

This avoids the following warning when running qmake:
Project WARNING: qmake spec does not announce the compiler family.
Guessed gcc.

(From OE-Core rev: 6c78f629323bea0be44181d60e3ce45be0416093)

Signed-off-by: Jonathan Liu <net147@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

qt4: disable gdb_dwarf_index

* qmake is trying to call native gdb and we don't depend on gdb-native
  (or even provide gdb-native)
* fixes errors like this:
  /bin/sh: gdb: command not found
  /bin/sh: line 0: test: -gt: unary operator expected

(From OE-Core rev: 6456b596a73232429e0d223089f32dc350ea20b0)

Signed-off-by: Jonathan Liu <net147@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

classes/conf: Add eventmasks for event handlers

Now that bitbake supports masking events for event handlers, lets use
this so event handlers are only called for events they care about. This
lets us simplify the code indentation a bit at least as well as mildly
improving the event handling performance.

(From OE-Core rev: bff73743280f9eafebe4591f7368ead91a4eb74d)

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

scripts/buildenv-internal/sanity: Update to python 2.7.3 as a minimum

We're finding bugs in python 2.6 and starting to require unittest
functionality in python 2.7.x. Its time to bump the minimum version
requirement. Anyone without python 2.7.x can use the buildtools-tarball
out to install a standalone python/git/tar setup which will work
with the system.

(From OE-Core rev: 8b90f1becd40a7f857d2fbe30eaffe218a976419)

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

sanity: Drop pseudo test, now unneeded

(From OE-Core rev: 075278db6f7a4783c952c72eebddf51a12f76075)

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

scripts/bitbake: Drop horrible double execution mess

There is no longer anything of value in the bitbake wrapper script since pseudo
is handled by bitbake internally. We can therefore drop it (yay).

(From OE-Core rev: d716d095751086e72fd789721005f0dc6d632997)

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

sanity.conf: Update minimum bitbake version requirement to 1.19.1 for bitbake-worker

This is due to the need to rely on bitbake-worker being present going
forwards.

(From OE-Core rev: ccf4fe860096f4b2a56b51b612b2a1dcb3525e59)

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

scripts/oe-buildenv-internal bitbake: Migrate python version checks

In preparation for removal of the bitbake wrapper script, move the
python version checks to the environment script. There are also
checks within bitbake itself but these may not always function
correctly on every version of python so this is really insurance.

(From OE-Core rev: 07792e4a83ca4f1c8152c228813c7f795fa6a545)

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

scripts/bitbake sanity.bbclass: Migrate tests for git and tar versions

Migrate tests for correct git and tar versions from the wrapper script
to the sanity class.

This sets the scene to allow us to remove the bitbake wrapper script.

(From OE-Core rev: 7b370e23594da5dcb53cd5507ec289c3ef2d9fb5)

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

bitbake.conf: Add extra fakeroot parameters

In order to correctly handle fakeroot at the bitbake level we need some extra
information which we provide with these new variables.

(From OE-Core rev: 003ea0fd1017dde50ced710179d0dc2e835d5185)

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

bitbake: bitbake: Add event mask flag support

Add a flag to event handlers which lists the events a given handler wishes to
process. By default event handlers recieve all events but this means
we can stop running code in many cases if we know it doesn't want the event.

This is part of the fix for YOCTO #3812, but implements filtering only
for class event handlers; the other part (events filter for UIs) will be
the subject of a different patch.

(Bitbake rev: 074003a4e7530a72863b9c685fc5c31b0f08c039)

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

bitbake: bitbake: Update version to 1.19.1

(Bitbake rev: 6acd444a557bba977ae1772371fdadf5f510e3b6)

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

bitbake: bitbake: Bump minimum python version to 2.7.3

(Bitbake rev: a01d153a8baac7136c2797c95357e74dd8872026)

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

bitbake: prserv: Adapt autostart to bitbake-worker

With the change to bitbake-worker we need to ensure the workers know
how to contact the PR service, the magic 0 port and singleton is
no longer enough.

(Bitbake rev: c761751e259bb8e940552a28794b45887b5a72d9)

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

bitbake: runqueue: Spawn a separate worker for fakeroot tasks

(Bitbake rev: 860ec42b220b7ed3f3bbe52c3546bba66644eac8)

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

bitbake: runqueue: Abstract the start and teardown worker functions

We're going to need a fakeroot/pseudo version of the worker so
abstract the code to start the worker process.

(Bitbake rev: b5d0f12f9df3ab211700473ed145ee6fbd9ca8e9)

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

bitbake: runqueue: Move the bitbake-worker execution to a higher level

The worker was being executed by each execution queue so would get
constructed twice for each build. This is wasteful so move execution
to the main runqueue so we only have to start the worker once.

(Bitbake rev: 8117f8480125b121b2b5ac0afc31b108d9e670ae)

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

bitbake: runqueue: Split runqueue to use bitbake-worker

This is a pretty fundamental change to the way bitbake operates. It
splits out the task execution part of runqueue into a completely
separately exec'd process called bitbake-worker.

This means that the separate process has to build its own datastore and
that configuration needs to be passed from the cooker over to the
bitbake worker process.

Known issues:

* Hob is broken with this patch since it writes to the configuration
  and that configuration isn't preserved in bitbake-worker.
* We create a worker for setscene, then a new worker for the main task
  execution. This is wasteful but shouldn't be hard to fix.
* We probably send too much data over to bitbake-worker, need to
  see if we can streamline it.

These are issues which will be followed up in subsequent patches.

This patch sets the groundwork for the removal of the double bitbake
execution for psuedo which will be in a follow on patch.

(Bitbake rev: b2e26f1db28d74f2dd9df8ab4ed3b472503b9a5c)

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

bitbake: fetch2: Fix AUTOINC handling

AUTOINC was meant to appear once at the start of the version string.
The list of names may not be sorted meaning it could get inserted in
the middle. This patch simplifies the code and ensures it appears at
the start.

Include cache version bump to ensure the cache picks up these changes.

(Bitbake rev: ad8bf10d873abb94d987860a3f6d06b134fb8a99)

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

buildtools-tarball: Add nativesdk-ncurses-terminfo-base

We've been seeing issues where some TERM values don't work well with the
tarball. These show up as escape codes coming from simple python calls such
as printing the python version. Adding the terminfo goes part way to addressing
the problem, the remainder is ensuring the relocation code correctly searches
the right path for it (which will be in a follow on patch).

(From OE-Core rev: 697e64c98affc4e0d8aa4828e74331c918b23a0e)

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

python: -Wformat is needed by gcc 4.8

This fixes errors in packages using python( build with gcc 4.8)

|
/home/jenkins/oe/shr-core-branches/shr-core/tmp-eglibc/sysroots/qemuarm/usr/include/python2.7/modsupport.h:27:1:
error: 'PyArg_ParseTuple' is an unrecognized format function type
[-Werror=format=]
|  PyAPI_FUNC(int) PyArg_ParseTuple(PyObject *, const char *, ...)
Py_FORMAT_PARSETUPLE(PyArg_ParseTuple, 2, 3);
|  ^
| cc1: all warnings being treated as errors
| cc1: all warnings being treated as errors

(From OE-Core rev: 5745a482a85c064a1eec960aff104cf8ce588e30)

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

mesa: introduce gallium-llvmpipe PACKAGECONFIG

(From OE-Core rev: d9f840725f3ea67fbd28d0903c99b729df5b1377)

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

mesa: use PACKAGESPLITFUNCS

(From OE-Core rev: ce127b8d20c9e812f2486405b080ab73d163e8de)

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

mesa: introduce openvg PACKAGECONFIG

(From OE-Core rev: 33d7c8a2e5087f84fb365e57e2555edb96f5fba5)

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

mesa: introduce gallium PACKAGECONFIG

(From OE-Core rev: 26276a215d7e09856d5e418edacab63197508050)

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

mesa: remove more .la files

(From OE-Core rev: 5eea21b86ca2ca09d4eb03aa9fcdb4564bdccb5d)

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

mesa: merge mesa-git.inc to mesa_git.bb

(From OE-Core rev: fb595d8a1181f30daa1c80938a335f9d249de913)

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

mesa: move LIC_FILES_CHKSUM to .inc

* we can also include .inc as first statement in recipe like most recipes

(From OE-Core rev: add5a0c17647a6b44461dd8181a412dac2ef263a)

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

mesa: use PACKAGECONFIG instead of DISTRO_FEATURES to define EGL_PLATFORMS

* it makes it easier to change PACKAGECONFIG in .bbappend and get
  consistent EGL_PLATFORMS

(From OE-Core rev: 50dcfeab2b18422f3ab067893a0b68b17706d6e2)

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

mesa: introduce PACKAGECONFIG for dri

* move DEPENDS which were originally applied only for mesa-dri to
  dri PACKAGECONFIG

(From OE-Core rev: f780ec28cbaade55361390a04ac70574bde426de)

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

mesa: merge mesa-common.inc to mesa.inc

(From OE-Core rev: 72afb46920202588948e256738ee80d585519974)

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

cogl: make cogl be able to build on ARM arch

1. Fix asm() register constraints in cogl when building for ARM.
2. Fix cogl to handle Thumb builds.

(From OE-Core rev: cdfea71ff1c4f80ff3a0ade1d7514cbf3c22abde)

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

systemd: update install-quotaon-once.patch

This patch has been merged upstream now, so refresh the patch with a backport
(the patch was changed before being merged).

(From OE-Core rev: cdad18589d08fdf592a50685f3b5fed32f4f4c76)

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

classes/qmake_base: allow parallel make

Remove MAKEFLAGS= from EXTRA_OEMAKE to allow parallel make.

(From OE-Core rev: 1aeecaeee9bb1eee779973fce7f15cc7fad269a0)

Signed-off-by: Jonathan Liu <net147@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

qt: update qmake2 class to export qconfig.pri mkspec

This is necessary when Qt applications want to know variables such as
QT_MAJOR_VERSION, QT_VERSION, QT_MINOR_VERSION, GCC version, etc.

(From OE-Core rev: dcaa40e65d986587fa7c44f4285bbd0883478cc3)

Signed-off-by: Felipe F. Tonello <eu@felipetonello.com>
Signed-off-by: Jonathan Liu <net147@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gnome-doc-utils: drop the usage of create_wrapper

create_wrapper was previously abused to invoke the Python interpretter directly
instead of relying on #! parsing, which is read into a 128 byte buffer so breaks
with deep build directories.

However now that we already install a "nativepython" binary into the native
sysroot, use sed to change the #! line to /usr/bin/env nativepython.

(From OE-Core rev: 470a67d9047ccabe3b624238f31c0381793404be)

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

createrepo: drop the usage of create_wrapper

create_wrapper was previously abused to invoke the Python interpretter directly
instead of relying on #! parsing, which is read into a 128 byte buffer so breaks
with deep build directories.

However now that we already install a "nativepython" binary into the native
sysroot, use sed to change the #! line to /usr/bin/env nativepython.

(From OE-Core rev: 7d91b44e00d6c50b7b5fbef17f2bae8e3a983a4c)

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

python-native: add nativepython symlink

Add a symlink to the native Python so that scripts can just invoke
"nativepython" and get the right one without needing absolute paths (these often
end up too long for the #! parser in the kernel as the buffer is 128 bytes
long).

The name "nativepython" was chosen to match the existing "nativeperl" which
serves the same purpose.

(From OE-Core rev: d99dac66e3e121e96c4d8a29aee846d4b8f38622)

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

dpkg: drop the usage of create_wrapper

create_wrapper was previously abused to invoke the Perl interpretter directly
instead of relying on #! parsing, which is read into a 128 byte buffer so breaks
with deep build directories.

However we already install a "nativeperl" binary into the native sysroot, so
use sed to change the #! line to /usr/bin/env nativeperl.

(From OE-Core rev: c4f987239fe78783d732df673a5e592bb05a877d)

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

scripts/contrib/build-perf-test.sh: fix passing arguments

Don't pass arguments to bitbake as a single one,
because this will break when the bitbake double-exec
is removed.

(From OE-Core rev: db13f10d233873148156880ab709ec76f8d3c329)

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

scripts/contrib/build-perf-test.sh: add branch name and sizes to results

Be more descriptive about the revision we are running on
in the global results file: add branch:commit and git describe fields.
Also add the sizes for tmp dir not only times. (previously these were
only available in the output.log)

(From OE-Core rev: 769a2c8ce797ee3afa39ab0fe9d9206a60cc4ba1)

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

site: add more alignment values for at-spi2-core

x86 and x86_64 values were added in 8c46ec. The x86-64 values were missing an
entry, add MIPS and PowerPC values from myself in qemu, and ARM values from
Martin Jansa.

(From OE-Core rev: a6a12ef5cad0dbb2d773bdccc340f1f767c5a782)

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

runqemu: when tunctl can't be found, say what package builds it

If runqemu is used without actually building any qemu images (i.e. you
downloaded the images) it's likely that qemu-helper-native hasn't been built.
Instead of just saying what command can't be found, tell the user how to solve
their problem.

(From OE-Core rev: 1498c431a161e8b3ddebefb5f03f4f11d5796c1d)

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

nspr: Update to 4.10

Source directory name changed from mozilla/nsprpub -> npsr

(From OE-Core rev: e4b206f28032916e95cc21a893af687409efe83f)

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

libusb: Update tp 0.1.5

Removed obsoleted patches

(From OE-Core rev: ed72e1867d517b99e60be66cb9f2318919c36056)

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

libxkbcommon: Update to 0.3.1

(From OE-Core rev: 6c0080817476a5a4c6f2a05676d323926318c89a)

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

sysstat: Update to 10.1.6

(From OE-Core rev: 2f27ca61d4f5bd3042ea9aefc2d2824e554daa69)

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

lsbinitscripts: Update to 9.47

(From OE-Core rev: 15644f3fae1525c8847a17deed0e681df8d92b8d)

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

libidn: Update to 1.27

COPYING, idna.h and ida.c:
  Year change only 2012 -> 2013

(From OE-Core rev: ca5f7c2c977e18c4e07555f9a2f8c5f9cbe75060)

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