python-imaging: fix PN -> BPN in do_install for multilibs

(From OE-Core rev: 9dbff379575e6f5fdc07412906848134b2739598)

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

libatomics-ops: fix PN -> BPN in do_install for multilibs

(From OE-Core rev: eca6309556086f8c6b0e8525749477c8210f83b0)

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

package_rpm: Ensure PV manipulations are correct

The previous change to this function fixed one manipuation at
the expense of the other, depending on ow expanded the versioned
dependency string was when the AUTOINC substitution was made.

This update ensures we cover both cases and ensures the classextended
version works as as well as the normal case.

(From OE-Core rev: 0f8447a7d99d2645b932eac1c24149f0c2d3791e)

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

icecc: improve interaction with sstate

* exclude ICECC_PARALLEL_MAKE (like PARALLEL_MAKE is)
* add ICECC_DISBLED to be able to disable whole icecc functionality
  while keeping icecc enabled. This is useful when you want multiple
  builders sharing same sstate-cache, but only some of them using icecc.

* inheriting icecc changes all checksums because of do_*_prepends calls,
  but because icecc should not influence binary output of task we should
  get same checksums for tasks build with and without icecc

* ICECC_DISABLED when set (to any non-empty value will disable icecc env
  modifications but while keeping same sstate checksum

(From OE-Core rev: f46cf86d0ec3b80ce526b4dad1b1eac5995969ec)

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>

elfutils: Fix size passed to snprintf for invalid sh_name case

Following build error will appear on Fedora18.
nm.c: In function 'show_symbols_sysv':
nm.c:756:27: error: argument to 'sizeof' in 'snprintf' call is the same expression as the destination; did you mean to provide an explicit length? [-Werror=sizeof-pointer-memaccess]
    snprintf (name, sizeof name, "[invalid sh_name %#" PRIx32 "]",
                           ^

(From OE-Core rev: 61010972284cc53cfc7958a9f6422e3fe74fcc0e)

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

binutils: fix ineffectual zero of cache and array bounds issue

binutils build fails on Fedora18+:
1. binutils-2.23.1/bfd/elf32-xtensa.c:6078:36: error: argument to 'sizeof' in 'memset' call is the same expression as the destination; did you mean to dereference it? [-Werror=sizeof-pointer-memaccess]
       memset (sec_cache, 0, sizeof (sec_cache));
                                    ^
2. binutils-2.23.1/bfd/elf32-xtensa.c:6120:32: error: argument to 'sizeof' in 'memset' call is the same expression as the destination; did you mean to dereference it? [-Werror=sizeof-pointer-memaccess]
   memset (sec_cache, 0, sizeof (sec_cache));
                                ^
3. binutils-2.23.1/opcodes/arc-dis.c:430:13: error: argument to 'sizeof' in '__builtin_strncat' call is the same expression as the destination; did you mean to provide an explicit length? [-Werror=sizeof-pointer-memaccess]
       sizeof (state->commentBuffer));
              ^
4. binutils-2.23.1/opcodes/rl78-dis.c:230:13: error: array subscript is above array bounds [-Werror=array-bounds]
      if (oper->use_es && indirect_type (oper->type))
              ^

(From OE-Core rev: 5445e12e5a32cc5c51ce8a29f2800692ed831115)

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

xf86-video-*: delete references to mibstore.h

This patch removes any references to mibstore.h, which has been removed
from xserver-org, from the following drivers:
 * xf86-video-vmware
 * xf86-video-fbdev
 * xf86-video-vesa

(From OE-Core rev: c2a2ac579876695c162ccca0ca3673f4d06ce60e)

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>

xf86-video-intel: upgrade to 2.21.3

(From OE-Core rev: 54a60a31a99394f1707d55f027f2cb12dc50845c)

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>

xserver-xorg: upgrade to 1.14.0

Aditionally:
 * adjust the aarch64.patch because the logic changed and the
   lnx_video.c changes are no longer necessary;
 * created patch to fix compilation issue when not using xinerama;

(From OE-Core rev: 1204a04c28546aa131d295fe7791f5de3694ad11)

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>

inputproto: upgrade to 2.3

(From OE-Core rev: f120c0da20004908378cf3cab2d6c61b39c6cda5)

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>

neard: Update reference commit sha1 and install script

This patch installs neard daemon in /usr/lib/neard.

(From OE-Core rev: 415e5bb60bfffe1f3b2acbb3136a45f63bfa91b9)

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

qemu-native:fix do_compile failed on SLED 11.2

1, When build qemu-native on SLED 11.2, there is an error:
...
| In file included from /usr/include/bits/sigcontext.h:28,
|    from /usr/include/signal.h:339,
|    from /buildarea2/tmp/work/i686-linux/qemu-native/1.4.0-r0/
qemu-1.4.0/include/qemu-common.h:42,
|    from fsdev/virtfs-proxy-helper.c:23:
| /usr/include/asm/sigcontext.h:28: error: expected specifier-
qualifier-list before '__u64'
| /usr/include/asm/sigcontext.h:191: error: expected specifier-
qualifier-list before '__u64'
...

2, The virtfs-proxy-helper.c includes <sys/capability.h> and
qemu-common.h in sequence. The header include map is:
(`-->' presents `include')
...
"virtfs-proxy-helper.c" --> <sys/capability.h>
...
"virtfs-proxy-helper.c" --> "qemu-common.h" --> <signal.h> -->
<bits/sigcontext.h> --> <asm/sigcontext.h> --> <linux/types.h> -->
<asm/types.h> --> <asm-generic/types.h> --> <asm-generic/int-ll64.h>
...

3, The bug is found on SLED 11.2 x86. In libcap header file
/usr/include/sys/capability.h, it does evil stuff like this:
...
  25 /*
  26  * Make sure we can be included from userland by preventing
  27  * capability.h from including other kernel headers
  28  */
  29 #define _LINUX_TYPES_H
  30 #define _LINUX_FS_H
  31 #define __LINUX_COMPILER_H
  32 #define __user
  33
  34 typedef unsigned int __u32;
  35 typedef __u32 __le32;
...
This completely prevents including /usr/include/linux/types.h.
The above `<asm/sigcontext.h> --> <linux/types.h>' is prevented,
and '__u64' is defined in <asm-generic/int-ll64.h>.

4, Modify virtfs-proxy-helper.c to include <sys/capability.h>
last to workaround the issue.

http://www.linuxtv.org/pipermail/vdr/2009-August/021194.html
http://patchwork.linuxtv.org/patch/12748/

[YOCTO #4001]

(From OE-Core rev: 1267bb2fd91f205d35e805aa019d25ab7a921b14)

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

watchdog: Upgrade 5.12 -> 5.13

Part of ping failure patch is already applied
in 5.13 hence refreshed the patch.
Tested on x86_64 box

(From OE-Core rev: 1834400555d0e2bad73ee472d5991464fec70d53)

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>

psplash : update to latest SRCREV afd4e228c606a9998feae44a3fed4474803240b7

(From OE-Core rev: a0235944f5b5a0cfda40cc8e61f49638cb402868)

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

package.bbclass: add debug-without-src PACKAGE_DEBUG_SPLIT_STYLE

* same as original and default version, but does not package source files in PN-dbg

(From OE-Core rev: 3c8452c3abae74a42989c0fbd5ba303788528750)

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>

systemd: remove libsystemd-daemon linkage in libudev

libudev was statically linking to libsystemd-shared, which was linking to
libsystemd-daemon (via libtool).  This is a spurious dependency so backport a
commit from upstream to remove it.

(From OE-Core rev: 3348e113d1e77bbe3b78abb7cddf866250421e06)

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>

systemd: set the location of the kill binary

Otherwise systemd uses AC_PROG_PATH and finds it in the sysroot, which won't
work on the target.

(From OE-Core rev: 084eb0e530426fe57fb4d04680ea1da421e1e4fb)

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>

systemd: remove util-linux-mount dependency

There's been a lot of action in busybox to fix mount's behaviour for systemd, so
remove the dependency on util-linux.

(From OE-Core rev: c6606864472784c40f61e1ac5ebba26b2be362d5)

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>

systemd: busybox's swapon now supports -p

Now that busybox's swapon supports -p, systemd doesn't need to depend on
util-linux.

(From OE-Core rev: 043969d963650048d9aba4da17a3b63e95463d64)

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>

busybox: enable swapon -p (91 byte increase)

systemd needs "swapon -p", so as the impact is minimal enable it unversally.

(From OE-Core rev: acc5e9b4f43a382e56c0af1fae492cd0aaa80230)

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>

systemd-compat-units: add missing systemd-systemctl-native dependency

This package has postinsts that run on the build host, but wasn't depending on
systemd-systemctl-native.

(From OE-Core rev: b4b83eb092e2c1ed2bca64ab43150228e4f6cc80)

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>

xkeyboard-config: remove long-gone build dependencies

(From OE-Core rev: 3e2a5722e25385e3b1b1ec44981af37c2074bc45)

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>

avahi: explicitly disable systemd if we don't want it enabled

(From OE-Core rev: 6126e7801542803aee1756aab41ba10a79bee092)

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>

xf86-video-intel: add explicit dependency on udev

Otherwise configure will auto-detect, and not be reliable.

(From OE-Core rev: 6c0c3dae3ec69c811ff8cd18a634fc4832a2d29f)

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>

polkit: explicitly disable systemd if we're using consolekit

(From OE-Core rev: 3a92104e1f7061296c293ac261e336e17934f8c7)

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>

connman: explicitly disable systemd if we don't want it enabled

Otherwise configure will use pkg-config and may find systemd, even though we
don't want it.

(From OE-Core rev: 87fe717009fb2bcd8e5dbdf0236fe715effca89f)

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>

build-appliance-image: fix lost net connectivity after restart

The issue is that an ethernet config file is removed without
proper checks in place (rm -f), which triggers an error and
makes the connman startup script to fail.

The file is now removed with proper checks in place.

Fixes [YOCTO #4003].

(From OE-Core rev: d9349e6e28cef107ad8bd00767ceba4d8aa464dd)

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>

build-appliance-image: ethernet connection type is now bridged

For network/internet connectivity, Build Appliance uses
a NAT virtual network adapter. While using NAT, inside
Build Appliance network proxies won't provide a network
connection to the outside world.

Basically, connectivity is not provided if network proxies
are needed in order to acces the internet.

The solution is to use bridged ethernet connection type.

Fixes [YOCTO #3998].

(From OE-Core rev: ef7ac8d9c0cf25834e75b4e5b987589f9ff19dc3)

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>

build-appliance-image: upgrade to newer poky revision.

Upgraded to poky commit: d823759b4594143d522eae0b2a2498436a6dcb1e

- Includes fixes and enhancements to Hob;
- Includes fixes and improvements to bitbake;
- Includes the switch to Linux kernel 3.8;
- Includes fixes to network connectivity.

Fixes [YOCTO #3843].

(From OE-Core rev: 6cdd70ddc16f72103e889676a185144ff582deca)

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>

libxi: upgrade to 1.7

(From OE-Core rev: c1fd149ecddb82fdf8ddc18c53f417a4318598e9)

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>

libical: add recipe back in oe-core

version: 0.48.
reason: libical is needed by bluez.

(From OE-Core rev: ea5ae40d4b124eef946ab87d867640937be8af17)

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>

package_rpm: Ensure package dependencies have correct version numbers

If a recipe has versioned dependencies on another package within
the same recipe, there are potentially races where the version
remapping may not happen correctly.

This issue triggered with neard in multilib builds since it
uses a "-" character in its PV which is illegal in an rpm version
field. The remapping to "+" was not occuring.

It only triggers in the multilib case since in this case, expansion
of the datastore happens at slightly different points.

The correct fix is to search for PV, not PKGV but substitute the
PV value.

(From OE-Core rev: bd03014c8fd080e04dd0a96a6b4b9211568c1cf1)

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

external-python-tarball: rename output package

Update TOOLCHAIN_OUTPUTNAME to rename the output package which is the
similiar style to name packages meta-toolchain-*.

[YOCTO #3996]

(From OE-Core rev: e40c36ff8610ebb9ce16a4d1c14c7cf94f87cd6a)

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

lib/oe/lsb: enable getting distro ID when lsb_release is not installed

If lsb_release is not installed (as it may not be on headless/minimal
installations on distros whose LSB package has a long list of
dependencies) we need to gather the information directly from files in
/etc.

Fixes [YOCTO #4012].

(From OE-Core rev: 106a7bcdb5ad9956f1d78f508408bfbcf7ff5120)

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

gst-plugins-gl: allow empty packages

The patch to Poky (git://git.yoctoproject.org/poky) commit
bd73fa69f532ca3d81ec2452a79e7a333b3543c2 titled: "recipes: Fix
ALLOW_EMPTY with no package specified" breaks this recipe. This patch
specifies the allowed empty packages.

(From OE-Core rev: 375ed2eaef10dd7d8f46badca6d779708724e5f6)

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

gst-plugins-package: return few ALLOW_EMPTY settings

(From OE-Core rev: 26c2eed11d9293a7a3b8e2177be76d2457d69efd)

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

linux-yocto/3.8: update EDF configuration

The sched_deadline configuration fragmements were not optimal, so tweaking
the meta data to make an enable-only configuration fragment avaiable and
updating the options within that fragment.

(From OE-Core rev: 43d2ae5695844d29ccad2c8a2760b803352d1f67)

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

linux-yocto/3.8: bump kver to v3.8.1

Updating the linux-yocto_3.8 SRCREVs to reflect the integration of the
korg 3.8.1 -stable release.

(From OE-Core rev: 75e247ed9a8a731480bef3ab0753fbad200469be)

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

linux-yocto/3.4: update to v3.4.34 and LTSI refresh

Updating the 3.4 SRCREVs to incorporate the -stable release of 3.4.34 and
the latest LTSI fixes up to [36a9caf 2 more fixes added].

(From OE-Core rev: e5602b0414fca0183d5aa90d9b462ac8e8401cbb)

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

libtirpc: Fix build on uclibc

Last upgrade of libtirpc broke the uclibc specific
patch. So lets refresh it and make it compile for
uclibc again

(From OE-Core rev: 4f46023fc24354c8144e755daf29c83926f3e1f6)

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>

wpa-supplicant: avoid host contamination by libnl3

Removes hardcoded include path -I/usr/include/libnl3. OE's include
path gets injected by do_configure.

(From OE-Core rev: 1c3beda0015da9a0fec2581af7645c9ea122c7e3)

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

wpa-supplicant: reorder do_configure

Copy from WORKDIR first, then modify. Improves consistency
between successive invocations of do_configure.

(From OE-Core rev: 82205345d777f7f34d43e4f81df0fa3bfe409fc4)

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

wpa-supplicant: change S to point to the top-level directory

This makes it possible to apply patches to ../src.

(From OE-Core rev: 48dbcc8c45d165e67f58a8307dde7594a28cf9bd)

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

gcc-4.7: Fix incorrect warning with -Wcast-qual

This patch backport the fix where -Wcast-qual
reports a bogus warning with 4.7.2, There is
no workaround in code that can be done to avoid
it hence the fix to gcc is backported.

(From OE-Core rev: 5d8ef98225773a6ec88b5e1a76ce01d76163b3a5)

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>

dpkg depends on ncurses

Dpkg compilation depends on the ncurses, so add
this dependency.

(From OE-Core rev: e33df55e69c5c517bcbc85452292740fe3b669da)

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

rootfs_xxx.bbclass: fix bitbake -S hashes mismatch error

The bitbake -S <image> has errors, e.g.:

NOTE: Reparsing files to collect dependency data
ERROR: Bitbake's cached basehash does not match the one we just generated
(/buildarea/lyang1/poky/meta/recipes-core/images/core-image-minimal.bb.do_rootfs)!
ERROR: The mismatched hashes were 8e44f2296f2d6cdf2ccb0236910ef59b and
46357d5e36713646b47b0fe2233e3c8c

This is caused by the BUILDNAME which is defined in cooker.py:

if not self.configuration.data.getVar("BUILDNAME"):
    self.configuration.data.setVar("BUILDNAME", time.strftime('%Y%m%d%H%M'))

The time changes during the first parsing and reparsing, so there is the
error, we already have:

do_rootfs[vardepsexclude] += "BUILDNAME"

but what we need is
rootfs_xxx_do_rootfs[vardepsexclude] +=  "BUILDNAME"

The do_rootfs doesn't depend on BUILDNAME directly, but indirectly, the
depends tree is:

do_rootfs -> rootfs_xxx_do_rootfs -> BUILDNAME

So add BUILDNAME to rootfs_xxx_do_rootfs[vardepsexclude] would fix the
problem.

[YOCTO #3299]

(From OE-Core rev: 9d039f482c84dde264564fde93f39764113cc0d3)

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

musicbrainz: remove

This is in meta-multimedia now, so remove from oe-core.

(From OE-Core rev: 62a2f7e9fad6cd6863670ba2376f73c0e929f075)

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-layers: fix duplicated help info

There are two "help" lines when run bitbake help under python 2.6.6
(perhaps all python 2.6):

$ bitbake-layers help
[snip]
help
    display general help or help on a specified command
[snip]
help
    display general help or help on a specified command

This is because the cmd.py in python 2.6.6 also has a do_help() function
as our bitbake-layers, and the cmd.Cmd.get_names() doesn't consider the
overload, this has been fixed in python 2.7, use set() to remove the
duplicated would fix the problem.

[YOCTO #3428]

(Bitbake rev: 324ed96e28ec31cff8cef1824d20d40f9a5d46ad)

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

base.bbclass: don't backfill features that already exist

It's too easy to cause rebuilds because the DISTRO_FEATURES have changed in
meaningless ways (such as re-ordering or duplicate items).  Help stop this by
checking if the feature to be back-filled is already present.

(From OE-Core rev: 63c7192119d54b92d908441109ed4e4fff761cba)

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

tune-cortexa*: Fix TUNE_FEATURES and PACKAGE_EXTRA_ARCHS for thf-neon

* all cortexa*thf-neon except cortexa8 were missing thumb feature from
  TUNE_FEATURES_tune-armv7athf-neon
* all cortexa*thf-neon except cortexa8 included cortexa9t2-vfp instead
  of cortexa9t2hf-vfp
* PACKAGE_EXTRA_ARCHS_tune-cortexa8thf-neon was including from armv7a
  -PACKAGE_EXTRA_ARCHS_tune-armv7ahf-neon
  +PACKAGE_EXTRA_ARCHS_tune-armv7athf-neon
* please do more testing for this, I'm sending this commit mostly
  because I've noticed that new a7 and a15 differ from a8 more then I've
  expected and I don't have any a7/a15 MACHINEs, feel free to extend
  http://git.openembedded.org/openembedded-core-contrib/log/?h=jansa/tune2-test
  to add and test fake a7/a15 configs

(From OE-Core rev: a207ce735b602b751467eb43e09b958e664a8e81)

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

gconf: make polkit an option, and disable by default

(From OE-Core rev: 9ae273bb17ffc88af390fcfbfc74d4a10ea05cf2)

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

apt-native: fix checksums

(From OE-Core rev: c5632d833a1be4de1dcda18a67019e134844a8ee)

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

package_deb.bbclass: set DPKG_ARCH_arm to "armel"

- the system should be using "armel" by default, not "arm"

[YOCTO #3741]

(From OE-Core rev: 74c671a609172a3801e53fe93e2853c4f0eb93b1)

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

consolekit: make policykit an option, and disable by default

Make PolicyKit usage a PACKAGECONFIG option, and disable it by default.  Also
migrate pam to use PACKAGECONFIG, and add systemd too.

ConsoleKit doesn't currently have an option to force this on or off, so rely on
a build-dependency for now.

(From OE-Core rev: ead060e668b0073f304a696f3fba9dbe24869f90)

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

libsoup-2.4: make libsoup-gnome optional

* Weakens dependency on libgnome-keyring and sqlite3
* Defaults to no if x11 is in DISTRO_FEATURES

(From OE-Core rev: 8a5d0a2be41c1a0471456a1098270632521c0c9f)

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

strace: fix kernel release string parsing

* Strace just aborts with two-digit kernel versions (e.g. 3.0-foo).
  Backport a patch from strace Git.

(From OE-Core rev: bff8456be448f0ae7ddae917ed3e9893f056376d)

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

perf: rename perf_3.4 to perf

There's nothing kernel-version-specific about the perf_3.4 recipe, so
it's actually misnamed and misleading now that it also gets used with
the 3.8 kernel.

Since the recipe isn't tied to a specific PV, and simply uses
whatever's in STAGING_KERNEL_DIR, there's no reason to add anything
else either to the bare PN, so just use that as the recipe name.

(From OE-Core rev: 9a249e074f0329ad54848b84536e5b7cb117ee2c)

Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

perf: rename perf.inc to perf-features.inc

The contents of perf.inc are really specific to perf features and
shouldn't use the generic perf.inc name, which implies common recipe
code.  It's always confusing to open up this file and find out that's
not what it is.

(From OE-Core rev: 4a98bd02ab8918e639353829b221b0c4b6c58165)

Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

adt-installer: remove patches and .pc directories from the shipped tarball

The patches and .pc directories should not be shipped since the patches
were already applied. So, remove these 2 directories.

[YOCTO #3983]

(From OE-Core rev: c18a4bded48ed1f8fd39e1741beb2a3452ae43e3)

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

dropbear: update patch upstream status

Blank password option patch has now been accepted upstream.

(From OE-Core rev: cfcd31e1ccee27fd46c830c01541c77298a13af4)

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

mkefidisk.sh: Cope with translations and modelnames with 'Disk' in it.

The script greps for 'Disk', which doesn't work when your crazy Dutch distro has parted call it 'Schijf', so force LANG=C.

The second problem is that 'Disk' might be a substring in the Model entry:

[root@Angstrom-F16-vm-rpm contrib] # parted /dev/sdc  unit mb print
Model: SanDisk SDDR-113 (scsi)
Disk /dev/sdc: 3905MB

(From OE-Core rev: 2e404930f6fc7d818d2f429793e84bce77163afd)

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

soc-family: fix SOC_FAMILY override order

* the current order has SOC_FAMILY settings, which are generic
  settings for a group of devices, overriding the machine specific
  settings.  For example:

  KERNEL_DEVICETREE_ti33x = "xxxx"
  KERNEL_DEVICETREE_beaglebone = "yyyy"

  Should yield "yyyy" when building for the beaglebone because
  that is a more specific device than ti33x.  However, without this
  change the result is that the value is set to "xxxx" meaning the
  more generic setting overrides the more specific setting.

(From OE-Core rev: 0b836b9d79255a5b2f358fe718c67638f52ecf72)

Signed-off-by: Chase Maupin <Chase.Maupin@ti.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

core-image.bbclass: support read-only rootfs

The hook function is appended to ROOTFS_POSTPROCESS_COMMAND
properly to support the 'read-only-rootfs' image feature.

[YOCTO #3406]

(From OE-Core rev: 05182ee354a9f773c8ab2f4cd7f1fdf552e7fcda)

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>

image.bbclass: add a hook funtion to support readonly rootfs

read_only_rootfs_hook: this funtion is invoked only if 'read-only-rootfs'
is in IMAGE_FEATURES. It tweaks some basic files and makes populate-
volatile.sh script run at rootfs time.

For now, it has real effect only if the init system is sysvinit.

[YOCTO #3406]

(From OE-Core rev: cae51a169e37b54becc08162aaa643acd53ffe2f)

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>

libgpg-error: upgrade to 1.11

License checksum change due to:
- copyright year modification

(From OE-Core rev: bb55cc3dc96a391bd9bc561c08c1cca3b11296b6)

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>

sudo : upgrade to 1.8.6p7

(From OE-Core rev: 8ea6517cac8df48bb9f17d5078360748bdda7b38)

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

Resolve sysvinit and util-linux conflicting man pages.

[ CQID: WIND00404316 ]

Use alternatives mechanism to prevent sysvlinux and util-linux
man pages from causing conflicts.

(From OE-Core rev: af97a79d017722d0571d552e390f2881efe228e2)

Signed-off-by: Jeffrey C Honig <jeffrey.honig@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

kernel-module-split: append space to RDEPENDS

* it was generating invalid RDEPENDS when KERNEL_MODULES_META_PACKAGE had
  RDEPENDS set already without trailing space

(From OE-Core rev: 5ff26f61b6860e56c255b2b6a2b0215be75b1db9)

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>

bash: Use new _ptest functions

(From OE-Core rev: 1839f969bc5d6e1555f5e5d218c0265f817fe300)

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

glib: Use new _ptest functions

(From OE-Core rev: eda5676f7927a3e1f19789a1d5890a3dd2b77725)

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

dbus: Use new _ptest functions

(From OE-Core rev: 60e9487b3d6b8a317436a7394d6635dd28aa1390)

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

ptest: Move ptest checks from recipes to ptest.bbclass

This patch cleans up ptest implementation in recipes by moving ptest
specific code parts into dedicated *_ptest functions.

(From OE-Core rev: 1e4d267a3cbd2135a5de10519814087050b766b7)

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

rpm: save output of postinstall scripts when image first boot

When image feature "debug-tweaks" is enabled, save output of post
install script to log file which can be configured when image first
boot.

[YOCTO #3223]

(From OE-Core rev: 94a58c153958002b117fcb5eeaef3d22be71a0f6)

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>

dpkg: save output of postinstall scripts when image firstboot

When image feature "debug-tweaks" enabled, save output of postinstall
scripts to log file /var/log/postinstall.log when image first boot. And
the log file can be configured.

It also needs image feature "package-management" enabled. If not,
package run-postinsts will be installed and then all packages will be
configured by it. Command 'dpkg --configure' outputs nothing.

[YOCTO #3223]

(From OE-Core rev: 684b94317f5b78f6c9c993f84438f8fa7e59fd5a)

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>

opkg: save output of postinst scripts when image first boot

When image feature "debug-tweaks" enabled, save output of postinstall
scripts to log file /var/log/postinstall.log when image first boot. And the
log file can be configured.

It also needs image feature "package-management" enabled. If not,
package run-postinsts will be installed and then all packages will be
configured by it. Command 'opkg configure' outputs nothing.

[YOCTO #3223]

(From OE-Core rev: 378504c1c0ab1e0e09950b37effc04221b928236)

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>

Near Field communication recipe for neard 0.9

The Linux NFC project aims to provide a full NFC support for Linux.
It is based on the neard NFC user space stack running on top of the
Linux kernel NFC subsystem.
The code generated using this recipe was tested on a ARM11 device, with
a kernel 3.6, using, for the NFC hardware, a USB dongle with the PN533
chipset (SCL3711)

(From OE-Core rev: b2a74ae70725be7efc0226901fd560d3b3b48607)

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

pulseaudio: respect the systemd DISTRO_FEATURE

(From OE-Core rev: 7376f4685899708fced56f212467e78325c99095)

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>

pulseaudio: rationalise existing configuration using PACKAGECONFIG

(From OE-Core rev: 4287440721bc692e8039ee775e6ee272e1349b3e)

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

bitbake-whatchanged: print what is about to happen

* Contents:
  - Summary
  - Usage
  - Implementation summary
  - Output
  - TODO

* Summary:
  This is used for printing what is about to happen between the current and last
  builds, for example:

  $ bitbake core-image-sato
  # Edit some recipes
  $ bitbake-whatchanged core-image-sato

  The changes will be printed.

* Usage:
  bitbake-whatchanged [[opts] recipe]

* Implementation summary:
  - Use the "STAMPS_DIR=<path> bitbake -S recipe" to generate the new
    stamps, compare these stamps to the one in the old stamps dir (tmp/stamps),
    so we will get what are changed.

  - When the "-v" (verbose) is not specified:
    > Figure out the newly added tasks
    > Figure out the PV (including PE) and PR changed tasks
    > The left tasks are the ones that the "Dependencies" changed tasks

  - When "-v" is specified:
    > Figure out the newly added tasks
    > Use bb.siggen.compare_sigfiles to figure out the details

* Output, for example (core-image-sato with different git tags)
  and with recipes upgraded):
  - without "-v":
    Figuring out the STAMPS_DIR ...
    Generating the new stamps ... (need several minutes)
    === Newly added tasks: (5 tasks)
      core-image-sato: do_configure do_populate_lic do_install do_rootfs do_compile
      # Note: This is because the "bitbake -S" always generate the sigdata for
      # do_compile, do_rootfs and other task, we may need fix this from "bitbake -S"

    === PV changed: (130 tasks)
      alsa-utils: 1.0.25 -> 1.0.26
      cross-localedef-native: 2.16 -> 2.17
      eglibc-initial: 2.16 -> 2.17
      [snip]

    === Dependencies changed: (3593 tasks)
      busybox: do_package do_package_write do_build do_packagedata do_populate_sysroot do_install do_compile do_package_write_rpm do_configure do_populate_lic
      atk-native: do_compile do_package_write_rpm do_package do_configure do_populate_sysroot do_install do_populate_lic do_patch do_packagedata do_build do_package_write do_unpack
      [snip]

    === Summary: (3728 changed, 1134 unchanged)
    Newly added: 5
    PV changed: 130
    PR changed: 0
    Dependencies changed: 3593

    Removing the newly generated stamps dir ...

  - with "-v":
    === Newly added tasks: (5 tasks)
      core-image-sato: do_configure do_populate_lic do_install do_rootfs do_compile

    === The verbose changes of glib-2.0-native.do_do_install:
    Hash for dependent task virtual:native:glib-2.0_2.34.3.bb.do_compile changed from bab8b8dd95be1b83dcec93f755b1812b to 70f746df7809acaa52de204b0685abb4

    [snip]

    === Summary: (3728 changed, 1134 unchanged)
    Newly added: 5
    Dependencies changed: 3723

    Removing the newly generated stamps dir ...

* TODO
  - It seems that the "bitbake -S core-image-sato" has bugs, it would always
    report errors, but doesn't fatal errors

  - The gcc-cross' stamps are in tmp/stamps/work-shared, but the
    "bitbake -S" doesn't put the stamps in work-shared.

  - The "bitbake -S" always generates the sigdata for image recipe's do_compile,
    do_install and other tasks, we may need fix this from "bitbake -S".

  - Print the ones which can be installed from the sstate.

[YOCTO #1659]

(From OE-Core rev: 8783fcc23ccbd829ecb0dc59cf71ee44376094cc)

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

bitbake.conf: add STAMPS_DIR for constructing STAMP

Add STAMPS_DIR for constructing STAMP, the defination of STAMP is:

STAMP = "${TMPDIR}/stamps/${MULTIMACH_TARGET_SYS}/${PN}/${EXTENDPE}${PV}-${PR}"

We can only change the TMPDIR if we want to change the STAMP's location,
but the bb_cache.dat would be regenerated if TMPDIR changes, so add
STAMPS_DIR for constructing it, and add it to the BB_ENV_EXTRAWHITE,
this is very usefull for the "bitbake -S", since then it can be run by:

STAMPS_DIR=<path> bitbake -S <recipe>

which will avoid putting the stamps to ${TMPDIR}/stamps.

BTW, break the too long BB_ENV_EXTRAWHITE into several lines.

[YOCTO #1659]

(From OE-Core rev: ce732c04b3ac06633e20efa8799c4189abfd41b3)

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

buildhistory.bbclass: add support for SRCREV logging

- create "latest_srcrev" for each recipe with the following format:
${BB_FILENAME},${SRC_URI},${SRCREV},${FROM_AUTOREV}

[YOCTO #3041]

(From OE-Core rev: f9acd756746e61b82b58d300f3ab13f3b086787d)

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>

connman: Wired inteface provisioned via qemu

- connman-conf package re-implemented
- connman 1.12 introduces provisioning for
wired interfaces also;
- wired interface settings are read from
kernel cmdline if present;
- after that are passed to connman as a
config file
- for BA, this is not needed, as BA will
have a network infrastructure to work with.

Fixes [YOCTO #3227];
Fixes [YOCTO #3804];
Fixes [YOCTO #3843].

(From OE-Core rev: 5711e0cf88fac0de8a5317b16a287b335c827c8f)

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>

connman: upgrade to 1.12

- 0002-storage.c-If-there-is-no-d_type-support-use-fstatat.patch no longer needed;
fix included in package source code

(From OE-Core rev: bd94ab11bbb6501013ed27ecb22685664816986c)

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>

sanity.bbclass:check if necessary to add march to BUILD_CFLAGS

1, There are a set of GCC built-in functions for atomic memory access. The
definition given in the Intel documentation allows only for the use of the
types int, long, long long as well as their unsigned counterparts. GCC will
allow any integral scalar or pointer type that is 1, 2, 4, 8 or 16 bytes in
length, suffix `_n' where n is the size of the data type.Such as:
__sync_fetch_and_add_n
__sync_fetch_and_sub_n
__sync_fetch_and_or_n
__sync_fetch_and_and_n
__sync_fetch_and_xor_n
__sync_fetch_and_nand_n
The above builtins are intended to be compatible with those described in the
Intel Itanium Processor-specific Application Binary Interface, section 7.4.

2, The glib-2.0-native and qemu-native invoke the above builtin function with
suffix `_4', and glib-2.0-native uses __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4 to
test the existance.

3, Not all above builtin functions are supported by all target processors.Such
as i386 does not support the functions with suffix `_4', but i486 or later
support.

4, Prior to GCC 4.5, on the Intel's processor, the default arch is i386 unless
GCC is built with the --with-arch switch. Since GCC 4.5 the default arch is
implied by the target.

5, If your host GCC is older than 4.5 and it is built without the --with-arch
switch, when you use the GCC to compile target, you should specify -march to
tell GCC what the target's arch is, otherwise i386 is used as default.

Above all, when use older GCC to compile glib-2.0-native or glib-2.0-native,
and the GCC incorrectly uses i386 as default, the above builtin function
with suffix `_4' is not referenced. We should have a check in sanity.bbclass
to tell the user if necessary to add march to BUILD_CFLAGS in this situation.

http://gcc.gnu.org/onlinedocs/gcc-4.7.2/gcc/_005f_005fsync-Builtins.html#_005f_005fsync-Builtins
http://gcc.gnu.org/ml/gcc-help/2009-06/msg00037.html
http://gcc.gnu.org/gcc-4.5/changes.html
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47460
http://llvm.org/bugs/show_bug.cgi?id=11174
http://download.intel.com/design/itanium/downloads/245370.pdf

[YOCTO #3563]

(From OE-Core rev: 38042ed8586b3abe427af33debc2402caeca52cb)

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

iproute2: Fix build failure on ppc64

Backport three patches from upstream for ppc64:

http://git.kernel.org/cgit/linux/kernel/git/shemminger/iproute2.git
ae70d96 ipntable: more fixes for ppc64
a55a8fd fix dependency on sizeof(__u64) == sizeof(unsigned long long)
a7c2882 ip: fix ipv6 ntable on ppc64

(From OE-Core rev: 041f784c06403e1d418be677fd15ea159c3bf90d)

Signed-off-by: Ting Liu <b28495@freescale.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

tune-cortexa7: add tunes for ARM Cortex-A7

http://www.arm.com/products/processors/cortex-a/cortex-a7.php

(From OE-Core rev: d6d1752c1f75fa97049d9e46b82e71d1ebbeb05f)

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

tune-cortexa15: add tunes for ARM Cortex-A15

http://www.arm.com/products/processors/cortex-a/cortex-a15.php

(From OE-Core rev: 513180af7f67b614dd0dbd8d9d77c057c45190b8)

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

shadow: add build dependency on shadow-native

The postinst requires patched pwconv and grpconv from shadow-native,
without these the rootfs creation uses the host binaries and fails.

(From OE-Core rev: 4baf25420bafb1f3c063efb053d653bd37e01a62)

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

systemd: set the location of the kill binary

Otherwise systemd uses AC_PROG_PATH and finds it in the sysroot, which won't
work on the target.

(From OE-Core rev: 54760c783089c98cc6eda1933393ac3cb49bcaed)

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>

systemd: remove --with-distro option, it's not used

(From OE-Core rev: 18f4d510798d571127f24139b3121e444b673b38)

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>

systemd: remove unused cached configure variables

(From OE-Core rev: 4a6be4d202526403aaf9130e615773bf7cb7852c)

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>

systemd: make gcrypt support (for signing the journal) optional

(From OE-Core rev: 9ca041ccc6af1378a1072bdde678c0e97faedde6)

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>

populate-volatile.sh: remove trailing whitespaces

Remove trailing whitespaces.

(From OE-Core rev: 082dca6f73858581863f103b467ff73cebe2c0da)

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>

populate-volatile.sh: improve the handling of link config items

Previously, if there's a link config item in the config file like

    l root root 1777 /tmp /var/tmp

and /tmp has existed, the symlink will not be created correctly.

Another example is the /run directory. If /run directory has been
created by some recipe or script before populate-volatile.sh runs,
the symlink of /run to /var/run will not be created correctly.

This patch ensures that the system creates symlinks exactly as the
config file tells it.

[YOCTO #3404]
[YOCTO #3406]

(From OE-Core rev: a54287d8e3d75a727c8ed5654a822bda256b0849)

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>

populate-volatile.sh: add ROOT_DIR variable to support running at rootfs time

For populate-volatile.sh script to run correctly both at rootfs time and
at system boot time, it needs to be aware of which situation it is now in.

We use the ROOT_DIR variable to indicate whether it is run at rootfs time or
not. ROOT_DIR being "/" indicates that this script is run at system boot time,
otherwise, it is run at rootfs time.

Also, we ignore failures when running this script at rootfs time.
For example, if ${ROOT_DIR}/var/dir1 is symlink to /var/volatile/dir1, it's
possible that the link is a dead link. So if we're going to create some file
under ${ROOT_DIR}/var/dir1, it will fail. But the failure does no harm,
because this script will always run at system boot time to set up the correct
files and directories.

[YOCTO #3406]

(From OE-Core rev: 45396e3edcce4a33fcbef6456f31811f30c26c63)

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>

bitbake: Revert "cooker: parse using bb.compat.Pool"

Reverting the pool changes, terminate does not work reliably on
bb.compat.Pool :(

[YOCTO #3978]

This reverts commit 8af519a49a3374bd9004864ef31ca8aa328e9f34.

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

bitbake: Revert "cooker: now that we use a Pool, raise the exceptions"

Reverting the pool changes, terminate does not work reliably on
bb.compat.Pool :(

This reverts commit 9ca8e513dc142dd7aefadeb6db2ccdb00f38b39f.

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

libpcap: fix description variable

* The variable name was truncated.

(From OE-Core rev: 2217176a9e2b9b9ac3b1e42e0feecb2c84d60f49)

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

qt4.inc: package keyboard drivers

* Fixes the following packaging error, if compiled with
  -plugin-kbd-linuxinput:

| WARNING: For recipe qt4-embedded, the following files/directories were installed but not shipped in any package:
| WARNING:   /usr/lib/qtopia/plugins/kbddrivers/.debug
| WARNING:   /usr/lib/qtopia/plugins/kbddrivers/.debug/libqlinuxinputkbddriver.so

* No PR bump, because -plugin-kbd-linuxinput is not enabled
  by default.

(From OE-Core rev: f477717a6e9b5820b6e6e505d64adb5d3a64abc7)

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

avahi: move systemd support where it belongs

We don't want the systemd inherit in avahi-ui since that causes various
warnings since PACKAGES is rewritten.

(From OE-Core rev: 02492d43cd7095acb3cf29843a50444c28b4d994)

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