systemd-compat-units: Use correct run-postinsts script link

OE-Core commit 75a14923da1ba91eddde47f0907345c19c82d6f0 has moved
run-postinsts script execution from S98 to S99 in rcS.d. run-postinsts.service
should check for this script and run it on first boot rather than
S98run-postinsts, which is for opkg/dpkg.

(From OE-Core rev: fe039170236080291c0220476a5809774f82ee5c)

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>

kmod: ptest fixes

 Make kmod-ptest able to compile with separated source and
 build dir. Since kmod test files contain kernel modules for
 many different architectures, strip and arch gets confused
 and throws errors.

(From OE-Core rev: 61cb45869caaed6578a217effaa72d247395d078)

Signed-off-by: Tudor Florea <tudor.florea@enea.com>
Signed-off-by: Björn Stenberg <bjst@enea.com>
Signed-off-by: Maxin B. John <maxin.john@enea.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

dbus: fix QA error building nativesdk-dbus

Fixes the following QA error:

ERROR: QA Issue: nativesdk-dbus: Files/directories were installed but not shipped
  /run
  /run/dbus

(From OE-Core rev: 796b7510853e71f158ad18dcea4cd1a10c7ef294)

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

linux-yocto: enable CONFIG_INPUT_EVDEV as standard policy

We have a number (in fact most) BSPs that require INPUT_EVDEV for basic
functionality. The size increase is minimal, so we'll add it to the
standard configuration for all platforms.

(From OE-Core rev: 965affabbf36fa5b5e466d331479c2295269242e)

Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
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>

license.bbclass: use shutil instead of bb.utils.copyfile

bb.utils.copyfile is for a specific purpose and more complicated than needed
here, so just use shutil.copyfile.

(From OE-Core rev: dc81df215cc94c279991df35125d94770a1bc3d2)

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>

opkg: ignore result of opkg configure

Some packages may return error while running opkg-cl configure, during
first boot. This will fail 'ExecStart' and 'ExecStartPost' will not run.
Without 'ExecStartPost' opkg-configure service will continue to run on
successive boot attempts. 'ExecStartPost' should disable this service
after first boot irrespective of 'ExecStart' status.

(From OE-Core rev: cd6041071ddf76693cda7632379ceddd1d21a7fb)

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>

cross-canadian: Fix TUNE_PKGARCH references

The cross-canadian compilers are now build once per architecture but were being
installed into tune specific locations which is incorrect. This adjusts things
so they are make TARGET_ARCH specific. We gain the tune specific parts from the
target sysroot which remains tune specific, the compiler and tools are independent
ot that.

binutils/gcc require sysroot options but since we reset at runtime, these shouldn't
have dependencies in the sstate checksums. They are therefore also excluded.

With these patches, switching machines does not result in a rebuild of *-cross-canadian
and the compiler is correctly located and referenced in the target images.

(From OE-Core rev: f58acab6414fe96d9e07ebbe86b348d2ac2bed5f)

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

pixman: Fix nativesdk rebuilds for armv7a

MACHINE=X bitbake nativesdk-pixman; MACHINE=Y bitbake nativesdk-pixman

where X is an armv7a machine and Y is not results in pixman rebuilding
due to the neon option. This is incorrect. The ultimate fix will be
not to apply target system overrides in the nativesdk case. Until
we do that we can at least work around the issue as done by this patch.

(From OE-Core rev: 2881747ec270bffc039d10198399e2686407b521)

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

gcc-common: Improve fpu code checksum dependencies

We only care about the end result in this case, not the specific inputs
that went into determining the gcc option. This change updates the code
to reflect that.

(From OE-Core rev: 83055511dab1e2aeefe84262ebe52595848c844b)

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

cross/crosssdl/native/nativesdk: Fix do_packagedata stamps

If you build nativesdk for machine A, then change to B you will see sstate
manifest warnings for the packagedata files. The stamps are machine
specific and should not be, ditto for native.

This patch copies the populate-sysroot extra stamp entries to avoid
these warnings.

(From OE-Core rev: a64de25d6006ec6dd777d8f3820a48244dfbf62b)

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

boot-directdisk: avoid loop in MBR disk signature generation

Instead of looping until the generated signature is non-zero, just
return 'ffffffff' if it's zero. This avoids an infinite loop if the
generated signature is always zero.

(From OE-Core rev: 34c74ba45c04efc2dd998545bf7a31d7b06a0ee0)

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>

yocto-kernel: Use variable-substituted BBLAYERS

The current find_bblayers() code finds and parses the BBLAYERS
variable manually, and therefore doesn't handle variable substitution,
which causes problems if used.

This change makes find_bblayers() use the variable-substituted
BBLAYERS instead.

Fixes [YOCTO #5106]

(From meta-yocto rev: 1629ac04e909143dc2c275c256094cb44c6cc43c)

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

bitbake: lib/bb/monitordisk.py: fix spelling in error text

(Bitbake rev: 78efda1224a99ed3d2ca6befb9fd719d82f6b0ae)

Signed-off-by: Erik Botö <erik.boto@pelagicore.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

README.hardware: update genericx86 hardware support

(From meta-yocto rev: ed421124e5f9c13ddca1a29a9a645b79e024d602)

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

linux-dummy.bb: Resolve missing task issues

Commit 609d5a9ab9e58bb1c2bcc2145399fbc8b701b85a added additional
initramfs functionality and created an additional task on kernel.bbclass
Adding this task was missed on the linux-dummy recipe, which causes task
dependency issues due to image.bbclass depending on
"virtual/kernel:do_bundle_initramfs". This change adds a dummy task
which resolves the dependency issue.

(From OE-Core rev: c0a8c5c07e0dd6f0ae302e9a4dcf7973e73e68e1)

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

makedevs: Do not return error if the fifo exisits

This ensures that makedevs will not cause image creation failures
when it encounters a pipe (fifo) that exists from a previous image.
This handles mode changes and it will correctly fail for dangling
symlinks.

[YOCTO #5288]

(From OE-Core rev: 3a4b0e7973bef43f16058137e64600e2f890b117)

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

dpkg: Add xz to RDEPENDS

dpkg uses xz as a compressor and the binary image is needed on the system for
dpkg to work correctly.

[YOCTO #1881]

(From OE-Core rev: 8f6bc04c9b563f8659ce0e053072deca02da38a5)

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

wic: Add OpenEmbedded-specific implementation

Reuses the mic/livecd infrastructure but heavily subclasses and
modifies it to adapt to the special needs of building images from
existing OpenEmbedded build artifacts.

In addition to the OE-specific mic objects and modifications to the
underlying infrastructure, this adds a mechanism to allow OE kickstart
files to be 'canned' and made available to users via the 'wic list
images' command.

Two initial OE kickstart files have been added as canned .wks files:
directdisk, which implements the same thing as the images created by
directdisk.bbclass, and mkefidisk, which can essentially be used as a
replacement for mkefidisk.sh.  Of course, since creation of these
images are now driven by .wks files rather than being hard-coded into
class files or scripts, they can be easily modified to generate
different variations on those images.  They also don't require root
priveleges, since they don't use mount to create the images.  They
don't however write to media like mkefidisk.sh does, but rather create
images that can be written onto media.

(From OE-Core rev: f87acc5e59d3c2c39ff171b5557977dab4c8f4a6)

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

wic: Add mic w/pykickstart

This is the starting point for the implemention described in [YOCTO
3847] which came to the conclusion that it would make sense to use
kickstart syntax to implement image creation in OpenEmbedded.  I
subsequently realized that there was an existing tool that already
implemented image creation using kickstart syntax, the Tizen/Meego mic
tool.  As such, it made sense to use that as a starting point - this
commit essentially just copies the relevant Python code from the MIC
tool to the scripts/lib dir, where it can be accessed by the
previously created wic tool.

Most of this will be removed or renamed by later commits, since we're
initially focusing on partitioning only.  Care should be taken so that
we can easily add back any additional functionality should we decide
later to expand the tool, though (we may also want to contribute our
local changes to the mic tool to the Tizen project if it makes sense,
and therefore should avoid gratuitous changes to the original code if
possible).

Added the /mic subdir from Tizen mic repo as a starting point:

 git clone git://review.tizen.org/tools/mic.git

 For reference, the top commit:

 commit 20164175ddc234a17b8a12c33d04b012347b1530
 Author: Gui Chen <gui.chen@intel.com>
 Date:   Sun Jun 30 22:32:16 2013 -0400

    bump up to 0.19.2

Also added the /plugins subdir, moved to under the /mic subdir (to
match the default plugin_dir location in mic.conf.in, which was
renamed to yocto-image.conf (moved and renamed by later patches) and
put into /scripts.

(From OE-Core rev: 31f0360f1fd4ebc9dfcaed42d1c50d2448b4632e)

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

wic: Initial code for wic (OpenEmbedded Image Creator)

Initial implementation of the 'wic' command.

The 'wic' command generates partitioned images from existing
OpenEmbedded build artifacts.  Image generation is driven by
partitioning commands contained in an 'Openembedded kickstart' (.wks)
file specified either directly on the command-line or as one of a
selection of canned .wks files (see 'wic list images').  When applied
to a given set of build artifacts, the result is an image or set of
images that can be directly written onto media and used on a
particular system.

'wic' is based loosely on the 'mic' (Meego Image Creator) framework,
but heavily modified to make direct use of OpenEmbedded build
artifacts instead of package installation and configuration, things
already incorporated int the OE artifacts.

The name 'wic' comes from 'oeic' with the 'oe' diphthong promoted to
the letter 'w', because 'oeic' is impossible to remember or pronounce.

This covers the mechanics of invoking and providing help for the
command and sub-commands; it contains hooks for future commits to
connect with the actual functionality, once implemented.

Help is integrated into the 'wic' command - see that for details on
usage.

(From OE-Core rev: 95455ae4251e06d66e60945092b784d2d9ef165c)

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

pseudo: Fix race problems

There is a potential issue with the fastop code in pseudo since a process may
exit and allow some other function to run before the server has processed
the commands run by the process. Issues have been see with unpredictable
file permissions.

To avoid this, we ping the server before exitting which guarantees it has
processed the current command queue.

The patch was written by peter.seebach@windriver.com

[YOCTO #5132]

(From OE-Core rev: a8a1f12c51ffdca011db194894fd7d14c119fb09)

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

bitbake: HOB: Busy cursor when saving configuration

Display a busy cursor while the configuration is saved.

[YOCTO #4846]
(Bitbake rev: fdb2e0120f189e97b8fdb6eb72d77a3100ba3018)

Signed-off-by: Valentin Popa <valentin.popa@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

qemu: Add fix for floating point issues on x86-64

There was a bug in handling of the mxcsr register since cpu flags
were not getting updated after fxrstor operations. This small tweak
fixes that.

[YOCTO #5248]

(From OE-Core rev: 5dc43cdc08e6698afa16ba79f3506a1555bb3710)

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

slang: fix sprintf bug concerning 8-bit characters

Buffer used for copying a "%c" character was getting
out of scope when it was required by the sprintf operation.

[YOCTO #5272]

(From OE-Core rev: c7de71813c8f47438f44749136877442cf73d536)

Signed-off-by: Marius Avram <marius.avram@intel.com>
Signed-off-by: Irina Patru <irina.patru@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

nss: Fix return codes in postinstall

exit 0 was done if $D != NULL, if one or more
shlibsign executions fails.

(From OE-Core rev: 5dc3eb72c4b9b68ab13310383a90fe7779bf92a7)

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>

boot-directdisk: ensure generated MBR disk signature is non-zero

A zero MBR disk signature is generally seen as no signature and
another partitioning program might install a new signature.

(From OE-Core rev: b6cafb1fcd6c168f8f4a4d2d5c74f3b425f156f3)

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>

mkefidisk.sh: Allow using a loopback mounted file

It should be possible to generate a disk to a file using a loopback
device with mkefidisk.sh, which is useful for booting simulators.  To
make this possible the partitions for the loop back need to work
similarly to the mmc devices.  The mkfs.vfat also requires and
additional argument to force it to write to something other then a
real disk.

Example:
  qemu-img create -f raw bigdisk 4G
  dev=`sudo losetup -f`
  sudo losetup $dev bigdisk
  mkefidisk.sh $dev tmp-eglibc/deploy/images/qemux86/core-image-minimal-qemux86.hddimg /dev/sda
  sudo losetup -d $dev

Note:
  Also a bug was fixed in the mkefidisk.sh where if the disk you are
  writing to initially has an invalid label the size of the first
  partition will be computed incorrectly.  For the simulator disk
  creation this is generally always the case, but this can happen with
  real hardware as well.

(From OE-Core rev: 254899824900f2e8c6a34d2ad1b8cbea91acb4ae)

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

kernel.bbclass: Allow INITRAMFS_TASK to trigger copy initramfs code

Activating the INITRAMFS_TASK can cause circular dependencies, but
that is up to the end user to resolve in recipes.  The INITRAMFS_TASK
should also trigger immediate linking of the cpio task in the first
compile pass.  This was a subtle regression introduced by: 609d5a9ab
(kernel.bbclass, image.bbclass: Implement kernel INITRAMFS dependency
and bundling).

This patch restores the previous behavior and only affects the
INITRAMFS_TASK which is not set by any of the default build profiles
in oe-core.

Reviewed-by: Andrea Adami <andrea.adami@gmail.com>
(From OE-Core rev: 36faac868e086e9c23537b107cdd973d7fd980bd)

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

opkg: Use systemd service for first boot configuration

Currently opkg uses a script to configure packages during first time boot.
This script is present in rcS.d and when 'sysvinit' is disabled this
script doesn't execute. For systemd only distros this newly added service
will run the opkg configure during first boot only.

(From OE-Core rev: fdcfcea5b8eae1769a217f8efea9c02f037f63fa)

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>

sato-icon-theme: add more compatibility symlinks

libfm uses "preferences-desktop", so link it to the close-enough
"preferences-sytem".

x11vnc uses "computer", link it to "terminal".

pcmanfm uses "system-file-manager", link it to "file-manager".

[ YOCTO #4062 ]

(From OE-Core rev: d856488179d14d05f1121c171abf4eac82e81fb9)

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>

pcmanfm: hide the desktop preferences

Sato doesn't use the pcmanfm desktop, so hide the desktop preferences launcher.

(From OE-Core rev: 778ef56c4b5c0d450a9b2e1df2b3e963dcfebed9)

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>

psplash: copy image files from workdir instead of next to recipe

During parsing this recipe builds up a list of splash image files,
however it was recording full paths to local files (i.e. the files next
to the recipe) and then in do_compile it was pointing to those instead
of the fetched files in WORKDIR. Fix it to use the fetched files which
has the added benefit of the do_compile signature not changing if the
recipe is moved around.

Fixes [YOCTO #5250].

(From OE-Core rev: f1850f9835651baee8d3a0858d00a5d22efcab19)

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

libtool: Don't patch generated files

We wipe out and regenerate all configure files so there is no point in
patching them.

(From OE-Core rev: 8d7036cdb71e66e619053c2545cfc1fbddf1895b)

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

native: Use a native site file and ensure the correct sed is used

native recipes were not using the site files. There are some things we need
to correct such as not using the path to sed-native. This allows us
to deploy settings globally for all recipes rather than hacking around
then in individual recipes as libtool-native used to for sed.

This fixes a regression introduced by
http://git.yoctoproject.org/cgit.cgi/poky/commit/?id=32edeb391f2107bb66b361cdcd4b8d4447731c33

(From OE-Core rev: db2eb3258844458569b7d4e6ed82c345858a3b99)

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

nativesdk-packagegroup-sdk-host: Add nativesdk-makedevs

Add makedevs to SDK, to allow simple /dev/ population.

(From OE-Core rev: a65261ab27cc93b8a564b628e50d89bf52492110)

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>

systemd-compat-units: do not create runlevel command

It seems strange that runlevel always returns 1, The comment says it is
related to sysvinit, but if we enable systemd, sysvinit will not be
installed. and we have created a link for runlevel to systemctl if
systemd is installed.

(From OE-Core rev: adc11d60fd4b555198d6653cd71eb1372e0b03a0)

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>

makedevs: rectify the exit codes and handle the invalid parameter

It is correct behaviours to output help and version information,
and should return 0;
When input parameter is invalid, print help information and exit.

(From OE-Core rev: 7c61daa08fa51557e0e6785e738646cb5d8de91b)

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>

makedevs: Add device_table-minimal.txt to the nativesdk sysroot

Add an example device_table in the SDK under /usr/share/

(From OE-Core rev: be697e3cfa67fa61aa9ef2c9c0c75f4ed452414e)

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

dropbear: pass SFTPSERVER_PATH explicitly

The default value of SFTPSERVER_PATH is "/usr/libexec/sftp-server" defined in
dropbear-2013.58/option.h, but after commit 406bd38b423[bitbake.conf: change
libexecdir to ${libdir}/${BPN}], sftp-server is provided by openssh package,
and is installed into ${libdir}/openssh, so we pass it explicitly.

(From OE-Core rev: 5f6deb044226885912214532cebb1d871f03c53a)

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

packagegroup-self-hosted: add xz to enable bootstrap

xz is required on the target filesystem since it's needed
to unpack some of the bootstrap packages.

(From OE-Core rev: 3f302e6686e9c35d3fd771b8aed214bf739e59f6)

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

eglinfo: drop machine-specific configuration from oe-core

eglinfo.inc has some board-specific EGLINFO_DEVICE settings but is missing the
dependencies.  Maintaining them for all supported devices in oe-core isn't
scalable and this is trivially supported though a bbappend in each BSP layer.

This was also causing problems compiling eglinfo on meta-yocto-bsp's beagleboard
machine as the EGLINFO_DEVICE setting for beagleboard was telling it to use the
closed GPU drivers, which are not supported in meta-yocto-bsp.

[ YOCTO #5224 ]

(From OE-Core rev: 2c1e147e61875054039ff08c36657c68c73037ee)

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

matchbox-panel: silence some pointless warnings

The battery applet emits a warning if the hardware doesn't have a battery (it
shouldn't) and the parser emits a warning if two separators are used in a row
(which happens if the hardware has a keyboard).  Silence these with a patch from
upstream.

[ YOCTO #4061 ]

(From OE-Core rev: 4e2b86787ed7a44ad88aa1431207d049c2cb444a)

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

boot-directdisk.bbclass: Fix media generation problems with vmdk

The various populate methods need to accept a path as an argument vs
using hard expanded variables.  In the case of the boot-directdisk
class it uses a different path for HDDDIR but it gets eclipsed by the
the class definition at the point in time ${HDDDIR} gets expanded.

The logical fix is to pass the arguments to the functions as opposed
to using globally expanded variables from the class definitions.

This patch changes 3 things:
1) syslinux_hddimg_populate takes an argument for the destination
2) syslinux_iso_populate takes an argument for the destination
3) populate is changed to boot_direct_populate because there
   was a conflict with it overriding the populate in bootimg.bbclass

[YOCTO #3994]

(From OE-Core rev: 63d164b755b984021f7e3cdba7280918ded6e821)

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

syslinux.bbclass, syslinux: Update to syslinux 6.01

A newer version of syslinux is required for an EFI enabled isohybrid.
This is used for the the capability to generate 3 types of ISO images,
all of which can be booted off a USB device or HDD if copied with dd.

1) PC BIOS only ISO
2) EFI only ISO
3) EFI + PC BIOS ISO

The syslinux.bbclass required a minor tweak because a few .c32
libraries require dynamic loading from the created media as of
syslinux 5 and up.  This was a good time to also fix the
duplication of the AUTO_SYSLINUXMENU block.

[YOCTO #4100]

(From OE-Core rev: 17d74fbd09e377e100423e1a73b9d4ce761a21d7)

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

bootimage.bbclass, zisofs-tools-native: add ability to compress ISO images

The mkzftree is needed to allow ISO images to be compressed with
minimal runtime overhead.  Below is an example of the savings on a
core-image-minimal.

Before ls -l:
24117248 core-image-minimal-qemux86-64.iso

Using the mkzftree ls -l:
16777216 core-image-minimal-qemux86-64.iso

(From OE-Core rev: 808d4371939ec48ed62e0c0b6614b96d61b1f7b8)

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Reviewed-by: Darren Hart <dvhart@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

syslinux.bbclass: Fix hard coding of console=tty*

The SYSLINUX_SERIAL variable was hard coded and occasionally needs to
be different for the kernel argument vs the syslinux argument.

In the auto-generated boot mode console=tty0 was hard coded, and this
is not needed at all, and causes problems in some cases if a end user
wanted to change the console=... via the kernel boot argument APPEND
mechanism.  The default can be forced with SYSLINUX_DEFAULT_CONSOLE
for systems that need a special specification to enable the frame
buffer instead of a serial port.

[YOCTO #3944]

(From OE-Core rev: cf2fba810a8a59cff71bf2c12e516e9080146604)

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

grub-efi.bbclass: Add serial and graphics menu options

The syslinux.bbclass already has support for automatically generated
serial and graphics menu choices.  This patch adds the same concept to
the grub-efi menu.  That makes it possible to generate a single image
which can boot on a PCBIOS or EFI firmware with consistent looking
boot options.

[YOCTO #4100]

(From OE-Core rev: 8444199fb598012f54853b010b5e5cce750db89d)

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

bootimage.bbclass, grub-efi.bbclass: Improve EFI & PCBIOS+EFI ISO support

Using the latest mkisofs it is possible to generate 3 different types
of ISO images, which can be used in various scenarios.

1) PCBIOS Only ISO
   - This option remains unchanged by this commit
   - Uses syslinux menus
   - Can be directly copied with dd to a USB device
   - Can be burned to optical media

2) EFI Only ISO
   - Uses grub 2 menus
   - Can be burned to optical media
   - If you want to use this image on a USB device
     extra steps must be taken in order to format the USB
     device with fat32, and copy an EFI loader which will
     in turn load the iso image

3) PCBIOS / EFI ISO
   - This is a hybrid image ISO that will work for case 1 or 2
     as above with the same restrictions and boot menu types
     depending on what type of firmware is installed on
     the hardware or depending on if EFI or "Legacy Boot" is
     enabled on some UEFI firmwares.

The syslinux.bbclass is now always required because that is where the
isohybrid dependencies come from as well as the configuration data for
the isohybrid.  The isohybrid is the secret sauce which allows the ISO
to work as optical media or as a disk image on USB or a HDD/SSD.

[YOCTO #4100]

(From OE-Core rev: a4baf911ab9d306ce5200e7d794ed6a9ccb25f30)

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

grub-efi-native: Add support for EFI ISO images

The iso9660 file system support needs to be added to grub in order to
be able to correctly find the grub.cfg.  The grub commands to locate
the grub.cfg also needs to be encoded into grub's default
configuration.

This change allows the resulting grub binary to work both in the hard
drive / USB boot case or the optical media boot case.

[YOCTO #4100]

(From OE-Core rev: 0bc0762a479b5182a07fccb2b1e9cd5fc15ca485)

Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Reviewed-by: Darren Hart <dvhart@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

testimage: Exclude BB_ORIGENV variable

The BB_ORIGENV variable isn't picklable and causes failures when generating the task
signature for the autorun image mode. We don't want to depend on its contents anyway
so lets exclude it.

(From OE-Core rev: 3f5f9d2681d41e44dade5d3a5a4bff3fd38f5506)

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

builder: register text files with leafpad

Log files will open in leafpad under Build Appliance.

Part of [YOCTO #4727] fix.

(From OE-Core rev: 1645a3c1d288a4de38683129ed146198d8093a47)

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

packagegroup-self-hosted: add leafpad text editor

A text editor is needed to examine log files.

Part of [YOCTO #4727] fix.

(From OE-Core rev: e598642988d1be1812b8df8ecd4f9d53f66068c9)

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

classes/package_rpm: fix bitbake package-index for RPM

The function that "bitbake package-index" relies upon when using the RPM
package backend (package_update_index_rpm()) uses MULTILIB_PREFIX_LIST
to get the list of package architectures to be indexed, but that
variable is only set when populate_sdk_rpm or rootfs_rpm are inherited,
which is not the case for the package-index recipe. Until we're able to
refactor this properly, for minimal impact just use the value of
ALL_MULTILIB_PACKAGE_ARCHS if MULTILIB_PREFIX_LIST does not give us any
architectures (the equivalent function in the ipk backend uses the
former variable).

Having "bitbake package-index" working is important because it's the
only practical way of indexing RPM packages for use as a feed; host
versions of createrepo won't work properly because they won't support
indexing recommends relationships.

Stopgap fix for [YOCTO #5278].

(From OE-Core rev: 9359719c563e1ab0ff10186d1a1b6bde7840dbf3)

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

oe-init-build-env-memres: use shell instead of Python to show the port number

(From OE-Core rev: 024e95696bad8f2ff09e1fda28c96d89d10999a1)

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

pseudo: warn twice if the 32-bit toolchain appears broken

Putting a warning at the top of do_compile is useful but not everyone reads the
file from beginning to end, so use a trap to put the message at the bottom too.

[ YOCTO #4919 ]

(From OE-Core rev: 51950fcbe4b98bdbb8b3dde88a8729e540d9609f)

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

pseudo: use bbnote/bbwarn instead of echo

In the future bbnote/bbwarn will be integrated into bitbake's logging, so use
those functions instead of echo directly.

(From OE-Core rev: 4933d7ae45d88090191c8ea07fd109ed34925e2d)

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

qt4-x11-free: disable QGtkStyle

This currently leads to a floating dependency on GTK+. Since most users
in an embedded context won't need this, just disable it by default.

Fixes [YOCTO #5116].

(From OE-Core rev: 6bef1e02dff6c6482791ab424d7e5dedeb840cf6)

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

gcc/libtool/perl: Fix various path to sed-native problems

If sed-native is built before these programs, hardcoded paths to sed-native
can end up in scripts and other parts of the system which may cause issues
if they are later used from sstate and sed-native is not installed.

To avoid this, this patch changes the global site configuration to
specify that plain "sed" is fine to be used. We need to spell this
out for gcc since it doesn't see the site files since we don't autoreconf
it. We can remove the values from libtool. We tell perl to use "/bin/sed"
since it requires a path and the system sed should be just fine for it.

[YOCTO #4971]

(From OE-Core rev: 2ec171cb188601bf18c6c2895870907024b1c52a)

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

runqemu: Use correct kvm CPU options for qemux86* with kvm

The existing -cpu host option caused kernel panics when people attempted to use
the kvm option. After research and discussion, the best options appear to
be the kvm32/kvm64 cpu types so lets use these instead. These resolve
the kernel issues for me.

[YOCTO #3908]

(From OE-Core rev: bdc6d3be6ffa4ed358153f9c9332b632324f5833)

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

busybox: update mdev.conf

busybox 1.21.1 's mdev has changed the way the device's name is
reported so now we get input/event0 instead of event0.
I think this commit is responsible of this new behaviour :
http://git.busybox.net/busybox/commit/util-linux/mdev.c?id=c3cf1e30a3022453311a7e9fe11d94c7a381640e

Update mdev.conf according to this behaviour so that sound
and input devices are correctly populated (and now
/etc/mdev/find-touchscreen.sh is executed).

Tested on an arm board.

(From OE-Core rev: 61b2950ebbc01f5e4fd7aece05bf371100c0c390)

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

linux-yocto/3.10: common-pc* config updates

Bumping the 3.10 meta branch SRCREV to import the following config changes
for tghe sugarbay and common-pc wifi fragments.

   dad2b7e common-pc-64: add kernel CONFIG options for sugarbay platform
   37c617d common-pc-wifi.cfg: add support for broadcom wifi drivers

[YOCTO #5117]
[YOCTO #5238]

(From OE-Core rev: 180f465b0cab13e17dc92ac5c88d7d9001cd5b11)

Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

adt-installer: add warning if MACHINE is not set in adt-installer.conf

Since the MACHINE can now be specified in adt-installer.conf, in order to
install the proper toolchain and environment script for the given
machine, add a sanity checking at the beginning to make sure MACHINE is
set for the wanted architecture.

Also:
 * uncomment the x86 target variables, in adt-installer.conf, in order
   to have qemux86 MACHINE set, by default, and also the sysroot
   installed (since it doesn't make much sense to install the toolchain
   without a target sysroot);
 * remove 'sudo' when creating relocate_sdk_tmp.py because the file is
   created in adt-installer directory;

[YOCTO #5259]

(From OE-Core rev: 0623a0e1bd7cf0c405a0469e9f16779971c0c3b5)

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>

adt-installer: check if sudo is needed when installing target sysroot

Currently, if SDK is installed in a directory that needs sudo
preivileges, the target sysroot is also installed with sudo, even though
the location is in users's home directory.

This patch reuses the ownership checking code and uses it to check if
sudo is needed for target sysroot too.

[YOCTO #5259]

(From OE-Core rev: 5acfced041477d8c272485196e87ad601a0ce5b4)

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>

package.bbclass: Clear umask when using os.mkdir

We switched to using os.mkdir with the file creation mode specified as the
second parameter. Python masks this with umask behind the scenes which isn't
what we want, we really want the permissions we specify.

To avoid this we zero the umask beforehand and restore afterwards. Other
solutions are possible but would not perform as well which is why
we're using os.mkdir in the first place.

Martin Jansa deserves the credit for debugging where the problem was.

(From OE-Core rev: f91226553e39439bfd17ab2b06c56cb8bf41061b)

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

lib/oeqa/runtime: ping: wait for 5 echo replies

Instead of considering that ping test passed after 1 reply,
wait for at least 5 consecutive replies in 60 seconds (which should
be enough time for connman to reconfigure the interface in systemd
images and help with the fake ssh/tests fails.)

(From OE-Core rev: 6fd19a9df0ad25b2822f12e2c3a97f1b71068d4e)

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>

nativesdk-packagegroup-sdk-host: Add nativesdk-shadow

Add useradd, groupadd et.c. since all target packages
which inherits useradd will have a postinstall hook
dependecy against the yocto specific groupadd.

(From OE-Core rev: 829edcb6e59e2e20dd4165b727b685f286f38031)

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>

shadow: Add nativesdk to BBCLASSEXTEND

This is a second in a series of patches to enable
offline rootfs creation from a package repository.

Some postinstall cmds are Yocto specific and needed to create a
rootfs with pre and post install hooks successfully run,
using only the toolchain tarball + a package repo.

End goal is to create a sandbox where users of a Yocto
based distribution can customize a rootfs from a package feed
with their package manager of choice.

With this patch, I can successfully create packagegroup-core-boot
with only the toolchain tarball(OPKG). More fixes for a few postinstall
hooks outside of packagegroup-core-boot will come next.

(From OE-Core rev: f90e1a45a042468e4e9a0fc91b57c6dba6f7adc9)

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>

relocate_sdk.py: make it work also with python 3

Strings in Python 3, by default, are assumed to contain unicode
characters. In previous versions of python (<3), unicode strings are
explicitly declared with u"abc". If not, than they're automatically
converted to bytes. This doesn't happen anymore in Python 3.

Since we're dealing with binary files, opened in byte mode, make sure
that we explicitly convert all strings to bytes to make both python 2
and 3 happy.

Other changes:
 * add a safety check to make sure relocation did not change the file
   size;
 * a couple of cosmetic changes (wrap long lines so that we don't have
   to scroll to reach the end of them);

(From OE-Core rev: 175f20e27eadc79df16109961f5ce6232705e96f)

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>

distutils: Replacing path to native path only to be done to non-native python packages

distutils: Replacing path to native path only to be done to non-native python packages

distutils: Replacing path to native python by path to python in the
image to support python packages with console-script setup resulted in
a "bad interpreter" error message because coreutils-native is not
a specified dependency of a number of native python packages.

We modify the change to apply specifically to non-native packages.

(From OE-Core rev: 312b6b33dca565153bc2e92d7ff6dd2974db4edb)

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

bind: port a patch to fix a build failure

(From OE-Core rev: 2cc9106da45a14d41a5269d91d7f79b6ccd8597f)

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>

build-appliance: Update SRCREV

(From OE-Core rev: 821037700ae61fff6a63d4cdb68559c5fcc96fde)

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

bitbake: bitbake: Update to version 1.20.0

(Bitbake rev: 2666865a27cedbffc82800b4cab1d6cff5a0222e)

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

wget: remove po.m4 patch

The po.m4 file is deleted by the more recent autotools.bbclass
autotools_do_configure code which handles gettext. There is therefore
no point in patching the file anymore.

(From OE-Core rev: 986c660cdb34e23a9c72f219db9363a6b4e4b7ec)

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

texi2html: remove po.m4 patch

The po.m4 file is deleted by the more recent autotools.bbclass
autotools_do_configure code which handles gettext. There is therefore
no point in patching the file anymore.

(From OE-Core rev: c3265281c3ab0979ec2753eb7580c39b005d5fa9)

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

ptest-runner: compatibile with busybox

Do not use pushd, busybox does not support it if system only has busybox;
Replace find command with ls to avoid some busybox's find unsupported options;

(From OE-Core rev: 1f059654bc81e55896c96c7c8e32e09f8925a761)

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

systemd: fix ptest support

Currently do_install_ptest_base is failing because it cannot find
'test' directory in build dir, ${B}. 'test' directory is present in source
dir, ${S}. Same is true for build-aux/test-driver and test/sys.tar.xz.

Also this fixes:
QA Issue: systemd: Files/directories were installed but not shipped
  /usr/lib/udev
  /usr/lib/udev/rules.d

(From OE-Core rev: bec8a29fb27ccdada0839d0bdd67ae22179a94e8)

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

perl: do not use pushd

pushd is not available when system is using dash as default shell

(From OE-Core rev: ec7e738845f72888b0016340d7da636e5ec46a1b)

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

grub: don't patch generated configure to fix do_patch re-execution

If do_patch re-executes, patching the configure script will fail because
it will have been regenerated during do_configure. We're patching
configure.ac so we don't need to patch configure anyway.

(From OE-Core rev: ae98a95577ba9c8a6a0d9a78d4cea33a92adff35)

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

scripts/runqemu: write temp file into correct location

We want the temporary file to be written in /tmp not the current
directory.

(From OE-Core rev: fcb40c11998030eb5fce89ce5a9ca567870aafa9)

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

documentation.conf: update contents

Documentation.conf is used in WebHob to display
information strings about the collected variables.

This patch brings the file up-to-date with latest
information available from the manual.

(From OE-Core rev: b9743c7b130bda3e50cdba4cf260232f1c24abcd)

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

sysvinit: don't use useradd

It causes shadow to be used in core-image-minimal and increase the
size by 1.5M.  We will add the shutdown user to group via base-passwd
which we depend upon instead.

[YOCTO #5230]

(From OE-Core rev: e33e4c30a4fa5b98903d6a46cdcc5bca9cf50848)

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

base-passwd: add shutdown group

Since using useradd in sysvinit causes core-image-minimal to
gain shadow utilites instead of using busybox, we add the
shutdown group directly.

[YOCTO #5230]

(From OE-Core rev: 47ebf756efeb88727459165e00c16671854e19a9)

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

shadow: Turn shadow-native into a BBCLASSEXTEND

(From OE-Core rev: 44fafdb5eca1fc8766bf649bc78e770e1d18979d)

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

shadow: Move common code to shadow.inc

(From OE-Core rev: ec64632d312a099c7c544af63b25e2defde0038e)

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

webkit-gtk: limit ld memory requirement

Add --no-keep-memory to LDFLAGS.

(From OE-Core rev: 223ee6bb7447dd64e9c407415675261cc1fdb6f4)

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

init-install-efi.sh: Remove unnecessary udev rules file to avoid errors

Fixes [YOCTO #5233]

Modeled after Chen Qi's fix to [YOCTO #3924] from oe-core commit:
6b6db7b4fb7aa17b8e29076decc830149b9d35bc
init-install.sh: remove unnecessary udev rules file to avoid error messages

/etc/udev/scripts/mount.sh is removed by init-install-efi.sh, but the
udev rules file which specifies the invocation of this script is not
removed, thus causing the error message during a live install:

/etc/udev/scripts/mount.sh: No such file or directory

The /etc/udev/rules/automount.rules no longer works once the mount.sh
script is removed. Remove it to avoid the error message.

(From OE-Core rev: 1f5a2b616d902b1158e348bf8c33b6d36e21cadc)

Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Cc: mihaix.lindner@linux.intel.com
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

init-install-efi.sh: Fix root= specification

Fixes [YOCTO #5237]

The current grub.cfg manipulation depends on an existing root=
parameter. If this doesn't exist, the correct root= parameter will not
be added.

Instead, remove any existing root= parameters and add the correct one
explicitly.

(From OE-Core rev: 14b124122c1b7d10b9a3a96fe4617c6fc1c661c5)

Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Cc: mihaix.lindner@linux.intel.com
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

package_deb, apt.conf: fix apt failure in do_rootfs()

The changes in this commit, dc5f6c3898555b59f16bf809ae4c5418656e6ac9,
moved apt config directory from native SYSROOT to WORKDIR.
Unfortunately, Dir::Etc in apt.conf was not changed accordingly and
sources.list file could not be found during do_rootfs().

This commit fixes this issue.

[YOCTO #5241]

(From OE-Core rev: 65a2a1e1d0b0e3dcc34d47f46185f24d753b53dd)

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>

elfutils: remove configure bits from patch

That is a generated file and patching it might fail, the patch already patches configure.ac
which will have the configure file regenerated.

(From OE-Core rev: daa56093de3a518b79480c90ec5124a79189bf7b)

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

libpam: Avoid host contamination issue w. libprelude

Since we dont use prelude in OE, we just disable autodetection of
prelude in the libpam configuration.

Seems like an old bug:
http://lists.openembedded.org/pipermail/openembedded-devel/2012-March/083804.html

(From OE-Core rev: 9096c6a46cf2467c90873c235b4533faf97d6175)

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>

makedevs: add nativesdk to BBCLASSEXTEND

This is a first in a series of patches to add items
to the nativesdk toolchain tarball.
Many of which are Yocto specific and needed to create a
rootfs with pre and post install hooks successfully run,
using only the toolchain tarball.

End goal is to create a sandbox where _users_ can customize a rootfs from a
package feed with their package manager of choice.

(From OE-Core rev: e70019bcdce83ae2a202338518f8725775abaa32)

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>

autoconf: Add fix for sh verses bash issues

Libtool scripts were finding bash was /bin/sh and then using bashisms which
then got into sstate and used on machines where /bin/sh might be dash.

This changes things to search for bash first since its preferred. We then hardcode
bash into the scripts which is more correct.

This does mean we have a dependency on bash but many of our scripts have
that anyway.

(From OE-Core rev: 367a19aaf31bcf997f10d045e7954cc800189052)

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

qemu: don't claim support for IrDA and PCMCIA

QEMU machines don't have virtual IrDA or PCMCIA hardware, so don't claim to
support them.

(From OE-Core rev: 694ca965eea971077e135cda4e54fa1cb0243233)

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>

boost: set up PACKAGECONFIG for boost-python library

Until now, the comment in the boost.inc file suggested to enable
the boost-python library by uncommenting some lines in that file.
Using the new PACKAGECONFIG feature, boost-python can now be added
optionally without need to modify the file or copying those lines
into a bbappend file.
Furthermore, we obtain the python version by inheriting python-dir
instead of fixing the python version in this file.

This commit is motivated by the need in the meta-ros layer, as
discussed in the issue #145 of the meta-ros issue tracker [1].

[1] https://github.com/bmwcarit/meta-ros/pull/145

(From OE-Core rev: 7568bfdd114597956a1da68746f207ec7f93a48d)

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@oss.bmw-carit.de>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

kern-tools: fix multi patch application without headers

Updating the kern-tools SRCREV to import the following change:

    kgit-s2q: always update ORIG_HEAD after applying changes

    In situations where git am fails to apply patches, and git apply is used,
    we must update ORIG_HEAD as well as HEAD. This is required, since if the
    next patch in the queue also fails git am application, it will reset to
    ORIG_HEAD before using git apply. If we haven't updated ORIG_HEAD, we'll
    end up warping back to the top of the branch each time.

This problem can only be seen in very specific situations, in particular if
a generated BSP branches from qemuppc, and has a series of non git "am able"
patches. We fail, since all of the qemuppc patches are not applied due to
the branch head constantly being reset.

(From OE-Core rev: 5126ac0aeb3154d31769dc20a46b6b1a6b2e3d9b)

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>

telepathy-mission-control: do_compile failed (race issue)

There might be an error when parallel build:

[snip]
Traceback (most recent call last):
  File "/path/to/tools/glib-gtypes-generator.py", line 304, in <module>
    GTypesGenerator(dom, argv[1], argv[2])()
  File "/path/to/tools/glib-gtypes-generator.py", line 295, in __call__
    file_set_contents(self.output + '.h', ''.join(self.header))
  File "/path/to/tools/libtpcodegen.py", line 42, in file_set_contents
    os.rename(filename + '.tmp', filename)
OSError: [Errno 2] No such file or directory
[snip]

This is a race issue, the _gen/gtypes.h and _gen/gtypes-body.h may
write(remove/rename) _gen/gtypes.tmp at the same time, then there would
be the error.

There was a similar bug in telepathy-glib which was already fixed, we use the
similar patch to fix it here.

[YOCTO #5184]

(From OE-Core rev: b0f81f460cf96798d79d72da7a3246c321caf654)

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>

libav: make X11 dependencies dependent on DISTRO_FEATURES

This enables building libav when x11 is not in DISTRO_FEATURES.

(From OE-Core rev: 9b7131e855293064e044cd79259eb350a14f319b)

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

classes/testimage: add informational messages about tests and do some cleanup/formatting

Adds info about tests run even if they passed (in case of errors log will
be printed anyway). Also some style change for qemu object.

(From OE-Core rev: 4adcf960be9c7477e314788d42b80753c2aaf572)

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>

image.bbclass: uninstall package only if it's already installed

Remove a package from rootfs only if it's already installed. Also,
if a package is uninstalled, remove it from installed_pkgs.txt.

[YOCTO #5169]

(From OE-Core rev: b86dc63bc87763119fce8286f37f44361da824d0)

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>

sstate.bbclass: ignore the tar failure.

sstate_package creates hardlink from sysroot to SSTATE_BUILDDIR, and
sstate_create_package stores SSTATE_BUILDDIR into a archive file by
tar, these two task can be run simultaneously for different packages,
and make a hardlink for a file will lead to the change of the links
number of file, and if tar is reading this file, it will fail with exit
code 1, and report "file changed as we read it":

    DEBUG: Executing shell function sstate_create_package
    tar: x86_64-linux/usr/share/aclocal/xorg-macros.m4: file changed as we read it

4b3e353a5[sstate.bbclass: fix parallel building issue] tries to use the
tar parameter --ignore-failed-read to fix, but it does not work, and
tar parameter --warning=no-file-changed can close the warning, but can
not change the exit code. so close shell immediate exit, only fail
if tar returns not 1 and 0.

Exit codes of tar:
    http://www.gnu.org/software/tar/manual/html_section/Synopsis.html

(From OE-Core rev: fad604b719e00b03e09da5fdb485e72332275b4a)

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>