Add texinfo.bbclass; recipes that use texinfo utils at build-time inherit it.

The class itself currently does nothing. The idea is to mark all recipes that
make use of the texinfo utilities. In the future, this class could be used to
suppress the generation/formatting of documentation for performance,
explicitly track dependencies on these utilities, and eliminate Yocto's
current dependency on the host system's texinfo utilities.

(From OE-Core rev: e6fb2f9afe2ba6b676c46d1eb297ca9cc532d405)

Signed-off-by: Max Eliaser <max.eliaser@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gcc-common: Only apply fpu settings to target gcc

Within the OE build environment, we supply the correct fpu settings. These
only need to be spelt out for the on-target gcc.

Doing this means the checksums for the core compiler don't depend on the fpu
settings. We exclude the compiler tunes for similar reasons, it doesn't need
to influence the compiler build.

(From OE-Core rev: ce1f3fd20d81545d6d5dfc68f86f9fddf8ac9bbf)

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

gcc-cross: Drop TARGET_CC_ARCH

Since we no longer build target libs within gcc-cross, we can drop the
TARGET_CC_ARCH flags and hence make it independent of tune.

(From OE-Core rev: 74d8866814aec520822518cc4cb8a942f7069bf7)

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

cross: Update PKG_CONFIG variables after recent changes

The PKG_CONFIG definitions should now more closely match those from native.bbclass.
We can't use ${libdir} here since it contains the target arch prefix.

(From OE-Core rev: aebb71afe54b266de39ec7e1caf84e6426aaafba)

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

layer.conf: Exclude gcc-cross dependencies from checksums

gcc-cross technically depends on the C library however we can use
one toolchain for multiple different tunes within a given architecture.

Manually remove these dependencies so that gcc-cross isn't rebuild with
every tune change.

(From OE-Core rev: 97df4e8e61a7693d548c3145e36f1d5503e91d03)

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

base.bbclass: Silence PREFERRED_PROVIDER warnings for sdk builds

Unfortunately expandKeys() notices overlap between the SDK and TARGET
PREFERRED_PROVIDERS entries when we set things up for the SDK recipes.

To avoid this, delete one of the sets of keys to avoid supurious warnings.

(From OE-Core rev: 9068a65baf4a28d04efad9297b16a2883ac42689)

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

tcmode-default.inc: Add default SDK PREFERRED_PROVIDERS

We should be explicit about the providers for the SDK toolchain so
add entries to do this.

(From OE-Core rev: fba380a2a37f04de04c0626c0a09cfe757c24341)

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

sanity: Bump BitBake minimal version to 1.23.0

The metadata started using the bb.utils.contains_any which has been
add in the BitBake 1.23.0 release, set this as the minimal version to
ensure parsing compatibility.

(From OE-Core rev: dea40c2850a32b5b67e90336f9565889f92dabe8)

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

bitbake: bitbake: Bump version to 1.23.1

(Bitbake rev: c40a4100ddb841d231360344616e59ab98e61fb5)

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

tcmode-default: Update PREFERRED_PROVIDERS

Update PREFERRED_PROVIDERS after cross tooling name change to include PN.

Also drop the now obsolete gcc-intermediate which no longer exists.

(From OE-Core rev: 5cbb5c12a5656915c3c9752187b54ef095d42828)

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

alsa-tools: Enable GTK support for X11, Wayland and DirectFB

(From OE-Core rev: 4c2ffce2ae32447a1e39877b9bcbc2cc11ddd2fa)

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

oeqa: add proper handling for command errors where needed

For use outside of tests themselves, we want a better error than
AssertionError, so create one and allow us to request it when calling
runCmd(). This enables us to avoid tracebacks during master image
operations if the power control command fails.

(From OE-Core rev: 89868383685091b0d3723fb8f29590f3f6610078)

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

scripts/contrib/dialog-power-control: add a trivial power prompt script

If you want to do automated hardware testing but don't have a
controllable power strip this script can be useful so that you know when
you need to cycle the power.

(From OE-Core rev: f71e9fe7c31fa44f5185d9ab64813ba2af57ca2a)

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

scripts/contrib/serdevtry: add script to handle transient serial terminals

When running automated tests (or just generally interacting with)
boards whose serial console devices are on the board itself and thus
disappear when powered down or practically disconnected, such as the
BeagleBone white, some terminal programs (e.g. picocom) will exit when
the device disappears and need to be restarted after the serial device
returns. This script handles this automatically for such terminal
programs.

(From OE-Core rev: 0537269df779532245eb2954e04fc26b3edfed85)

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

classes/testimage: if start fails, don't try to stop

If we couldn't start the target, it doesn't make sense to try and stop
it here since logically it shouldn't now be in any kind of "started"
state. (It's the start function's job to clean up after itself if it
fails - to that end, fix up the QemuTarget class so that it does.)

(From OE-Core rev: 819ebddae6b78120e5e082423793ff988419b5c4)

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

oeqa/controllers/masterimage: more robust master image startup

Instead of powering up the target when the object is constructed, wait
until deploy is called. Then there are basically two different
scenarios:

a) The device is booted into the master image already, in which case
   we can just use it
b) The device is booted into another image or can't be contacted, in
   which case we need to power cycle it. Here we also now wait until it
   has booted up instead of trying to contact it immediately.

(From OE-Core rev: c2257fa50071e4704a8152b5f1d16f899b4bed98)

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

oeqa/controllers/masterimage: add a serial control command

Similar to power control command, this depends on a user's setup to get to
the serial port of a board. For a local connected board this could just be:

TEST_SERIALCONTROL_CMD = "picocom /dev/ttyUSB0 -b 115200"

and for a serial console server: "telnet 10.11.12.13 7003" or some conmux
command.

(From OE-Core rev: 5131094ad8f53b2052a97efd505e7a170d0c915d)

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

oeqa/targetcontrol: restart method shouldn't be abstract

And drop the un-needed and un-used restart methods.
Only qemu ever used this and actually does it safely.

(From OE-Core rev: 1dd1edb5ea551c8a01538b130aa4d0c361eae14d)

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

oeqa/controllers/masterimage: add a base class for hw targets

Right now GummibootTarget is the only hardware TEST_TARGET with deployment,
but we will add more, so let's make an abstract base class, that will
do the common thing for all the hw targets.

(From OE-Core rev: 1d70b1908e1dc5d612b0627022659639e3f384e5)

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

sqlite3: restore upstream version numbering

Regardless of the numbering used in source archive filenames, the
upstream version number is in dotted form in documentation, release
notes and the website, so we need to be using that here since that is
what people will expect to see.

(From OE-Core rev: b8f82a8808b0d217815b30ed06efc8a0be4494a5)

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

lighttpd: upgrade to 1.4.35

(From OE-Core rev: cdbcadee68a47e985d25ba39359f3a3fa0049a8a)

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

gstreamer1.0-plugins-ugly: upgrade to 1.2.4

(From OE-Core rev: 013cfd1840fa40742849ff5f9c9a116fbff3e61f)

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>

gstreamer1.0-plugins-bad: upgrade to 1.2.4

(From OE-Core rev: 10b7d0d3458e89d96d0916ff82f0330eb031f1a9)

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>

gstreamer1.0-plugins-good: upgrade to 1.2.4

(From OE-Core rev: 25f410eca19ba262a1d6ab9eddcc66aaaaf3f9ec)

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>

gstreamer1.0-plugins-base: upgrade to 1.2.4

(From OE-Core rev: 4b5500abc56beb69d79713fb2964773af8ea0a8e)

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>

gstreamer1.0-libav: upgrade to 1.2.4

(From OE-Core rev: b398248253a996c88f05289359bed374111763ac)

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>

gstreamer1.0: upgrade to 1.2.4

(From OE-Core rev: c904939011eeb087815f2f02b2faf3839aaf41b0)

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

telepathy-glib: upgrade to 0.24.0

(From OE-Core rev: e54421af3c670e24e12971714233571b9d996e0c)

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>

fontconfig: upgrade to 2.11.1

Removed the following patch(es):
 * sysroot-arg.patch (changes included in release)

Correct the fccache.c license checksum and the line numbers because the
license snippet has moved and, also, at the previous upgrade the lines
were wrong (shifted a couple of lines).

(From OE-Core rev: cc337329309a6ad94a68b2813211a15040638c00)

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>

xauth: upgrade to 1.0.9

(From OE-Core rev: 152de1889fae9579a7db9f9ffc3a08ea5cc0e22d)

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>

nasm: upgrade to 2.11.02

(From OE-Core rev: a3ddc4041a9a04072e4dbfe23fc5bddc7b0217ce)

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>

kexec-tools: upgrade to 2.0.6

(From OE-Core rev: ba802d21eccfd20584fed784134b6fb72c3deb80)

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>

xtrans: upgrade to 1.3.4

(From OE-Core rev: 90c61560fff8517a79741cef7347d3428e3f1be2)

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>

xrandr: upgrade to 1.4.2

(From OE-Core rev: ed70073dbb735f945ebab4faf6f316f902b92976)

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>

xproto: upgrade to 7.0.26

(From OE-Core rev: f13bb6190f50545ef35f8f3a1a0318ccb01db2db)

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>

gdb: upgrade to 7.7

Removed the following patch(es):
 * gdb-fix-cygwin-check-in-configure-script.patch (changes included in release)

(From OE-Core rev: 916489246b0840463957b85bdd1b2c3fccbfda89)

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>

python-pygobject: upgrade to 2.28.3

(From OE-Core rev: e34ffc9bcdb01f9a1cbbd705c56f7ed0358db267)

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 1.15.1

(From OE-Core rev: 7800dd17e6ff511c9146de2bceea6765b0dc1c2c)

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>

libdrm: upgrade to 2.4.53

(From OE-Core rev: dcd2733594297e3528dee07da9e6f11eede8b7ac)

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-input-synaptics: upgrade to 1.7.4

(From OE-Core rev: a2eb22c1c21cd5427dce2e7477e7e49e42101724)

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>

oprofileui*: Update to latest git

(From OE-Core rev: 9082b6fc73c7e4c9d73b988674bb9c09b693cf12)

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

lsbinitscripts: Update to 9.54

(From OE-Core rev: 80a7ca6e986a77df69f90601195c10aad640e9ab)

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

util-linux: Update to 2.24.2

rebase util-linux-native patch

(From OE-Core rev: 82f40f320364db499ef35bb1ce85f166b35aa35f)

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

man-pages: Update to 3.65

(From OE-Core rev: 17679accf8dcca9cc09ed9b72644366859b2b23e)

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

lib/oe/sdk: Ensure target directory exists before creating the link

(From OE-Core rev: d8096433c27643f39eeb29d34e20328a39981fd6)

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

base.bbclass: Update INCOMPATIBLE_LICENSE changes for gcc updates

Add in the TARGET_ARCH and SDK_ARCH suffixes from the gcc-cross changes.
The -intermediate toolchain parts were canned a while ago so drop them.

(From OE-Core rev: 8a733097bddcf0868fb5f8ca08ea59b2cbc4a3a9)

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

gcc: Drop ARCH_FLAGS_FOR_TARGET usage

As far as I can tell this variable is now completely unneeded. It would
only ever get used in target builds and these are now correctly done
in the target environment namespace, not any of our cross environments.
As such, CC and other variables contain the correct compilers and other
tune options and these are correctly picked up when building libgcc,
libstdc++ and others.

I tried to figure out where else these would make any sense and couldn't
find anything. Builds appear fine without them so lets drop the complexity
including the patch adding in this flag to gcc.

(From OE-Core rev: 5484596f4252e707ff791feedf143a72dbb613f6)

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

gcc-common/gcc-configure-common: Move gnu-configize to its own shared task

This command modifies ${S} and can race against other tasks running do_configure and
having the scripts disappear from under them. To avoid this move to its own
task and work on the shared work directory as a common task.

It needs to be a python task to avoid lots of shell exported variables as
dependencies.

(From OE-Core rev: 55ed9ffbab5118140baac5d0d29dcc919e1f1668)

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

gcc-target: Limit compile to host targets, don't build runtimes.

Currently the gcc builds are building copies of the target libraries
that we never use (it isn't installed in do_install). This is a rather
pointless waste of cpu time.

Instead just compile the host targets. Comparing the package output of
this compared to a previous build shows that the unwind.h header is
missing since its provided by gcc. Fix this simply by copying it in.

(From OE-Core rev: 19448a50c106684ee7e3862f759418c982363a29)

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

binutils/gcc/gdb: Add TARGET_ARCH to PN for all cross recipes

This allows them to co-exist together in the native sysroot, with one
set of cross tools per target architecture.

(From OE-Core rev: a2c5509520d5c3e082f55844e6545d0309565f8f)

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

cross: Drop virtclass-cross OVERRIDE

This has been replaced by the class-cross override.

(From OE-Core rev: 7d6050c57f195589c8429397432c78f68298b672)

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

bitbake: bb.utils, bb.codeparser: Add bb.utils.contains_any

This includes contains_any in the special handling code for sstate. It
does not take into account the equivalence of the values. In current
code, considering 'bb.utils.contains_any("A", "foo bar", ...)':

A = "foo"
A = "bar"
A = "foo bar"

All those will get different signatures.

(Bitbake rev: d1e3345d715e488ec3f5515fb0e1fb39366346bc)

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

feature-arm-thumb.inc: set ARMPKGSFX_THUMB only when thumb is in TUNE_FEATURES

* there is issue for TUNE_PKGARCH missing in PACKAGE_ARCHS for machines
  without thumb enabled, it was reported by Jacob Kroon on IRC

(From OE-Core rev: 1e1b42f687b5cd34623fe2682218958e1947eb92)

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

feature-arm-thumb.inc: Suppress false warning

If a recipe does not explicitly set ARM_INSTRUCTION_SET, then there is no
need to throw a warning:

  WARNING: Recipe 'foobar' selects ARM_INSTRUCTION_SET to be 'None',
           but tune configuration overrides it to 'arm'

(From OE-Core rev: e457d71641af8802e47eb4854072e3cfb957b001)

Signed-off-by: Jacob Kroon <jacob.kroon@mikrodidakt.se>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: codeparser: don't interact with the cache for subshells

Doing so was causing leakage between the execs of the main value and that of
the subshell value, and was causing the cached subshell value to be used for
the overall variable. At the least this could cause execs contamination
between two variables that while differing, run the same subshell. Beyond
that, it's possible we could have been using an incomplete cached value of
a subshell for that of the main value.

Before this, bb_codeparser.dat would change between parses with differing
bbfile parse order. After, it does not change.

The codeparser cache version is bumped, to ensure we don't use potentially
incorrect cached values from previous runs.

This should hopefully resolve the difficult-to-reproduce issues we've seen at
Mentor Graphics where bitbake emits a script to run a task and misses
dependent functions, resulting in 'command not found' failures. This issue has
also been mentioned on the oe devel list, where someone hit a case where
oe_runmake was missing from a do_install task (IIRC). Adding debug information
showed that bitbake's information about the variable dependencies for this
task is inaccurate in the failure cases.

(Bitbake rev: 97537e4786a1e3a329249497498b59b8f5174fc3)

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

bitbake: fetch2: Cleanup file checksum generation, v2

The initial version of this patch was reverted in

0a94e568152de550dedc8135a766beb18bf064ab

However, it was later agreed upon that it was reverted due to
a misunderstanding during code review; the patch is ok.

This version of the patch also removes an unneeded initialization of
the 'checksum' variable outside the scopes it is being used in.

(Bitbake rev: 9ee19a3ca2f8e11a91f0289ea3486310c61d40f2)

Signed-off-by: Jacob Kroon <jacob.kroon@mikrodidakt.se>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: HG Fetch with username and password in url

Following up a previous patch for mercurial fetcher, I just fixed a
problem when calling update on a repository with subrepositories enabled.

Attached a patch that fixes this problem

>From caae519a2bd64bf7c729bb26aff344827def47fb Mon Sep 17 00:00:00 2001
From: Volker Vogelhuber <v.vogelhuber@digitalendoscopy.de>
Date: Tue, 29 Apr 2014 15:29:47 +0200
Subject: [PATCH] bitbake: - fixed authentication issues in case of using sub
 repositories

(Bitbake rev: dfa041c940caad12da50126a559afc8de089eeda)

Signed-off-by: Volker Vogelhuber <v.vogelhuber@digitalendoscopy.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

mc: update to 4.8.12

Add PACKAGECONFIG for two optional plugins (smb & sftp), both require
other packages from meta-openembedded

(From OE-Core rev: 64c44f02f1a320399e2f02aeaf83022bf83b69c8)

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

systemd: Check for HAVE_POSIX_FALLOCATE

The check for the config variable is using the wrong define
resulting in the #else always being used.

(From OE-Core rev: 7e498c4c17e2c1b6928d4c1fd127977bb588991f)

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

distrodata.bbclass: Fix checkpkg functionality

Currently it wasn't working because *COMMAND variables were removed
from fetcher.
Now checkpkg sets the command internally and sends it as a parameter
to _runwget() function from wget fetch.

(From OE-Core rev: b9a51fc1901c378375cca041da27ddbd450c0412)

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>

ltp: update version to 20140422

Update ltp version to 20140422
Removed two patches that already exist in the current version:
1. regen-makefile.patch
2. ffsb-remove-hardcoded-configure.patch

(From OE-Core rev: d72430772400c7733da2339f238ab4ea69e48a8e)

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>

oeqa/utils: sshcontrol: realtime logging of output

Log the output of the command as it runs not when it finished, else
tail -f tmp/work/minnow-poky-linux/core-image-sato/1.0-r0/testimage/ssh_target_log
isn't as useful as it could be.

(From OE-Core rev: be8f766f43d85c364b9706b464ed0a59d0fbf0b7)

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>

tune-cortexr4.inc: Add thumb and arm to TUNE_FEATURES

(From OE-Core rev: 1ebcbc6d77171b81dfe6432ffad6cae3148dcf2e)

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>

feature-arm-thumb.inc, arch-armv4.inc: Add "arm" to TUNE_FEATURES

* it will be inherited by most DEFAULTTUNEs, except few exceptions which
  support only thumb and not arm
* respect missing "arm" in TUNE_FEATURES in feature-arm-thumb.inc, so
  when recipe asks for "arm" and MACHINE supports only "thumb" ignore
  recipe and try to build with "thumb"
* show warning when overriding ARM_INSTRUCTION_SET set by recipe from tune
  config

(From OE-Core rev: 1250d3e009363d20f15bbfaced622c5912a7fb93)

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>

feature-arm-thumb.inc: Replace inner quotes with apostrophes

* so that it's highlighted correctly

(From OE-Core rev: 31a3525504ad7cf9fe0a4f8da27ad11e6311f299)

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

openssl: enable ptest support

Install openssl test suite and run it as ptest.

(From OE-Core rev: c48981d2d24a20978a17866fa478dde21bd96b91)

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>

sstate-cache-management: Minor cleanup of help text

(From OE-Core rev: a039ef9d116f0bab0bbb6d25867666b10057f2ad)

(From OE-Core rev: 2f22c098c3ada8c0952c7711f7a7f5b6421e6aea)

Signed-off-by: Jacob Kroon <jacob.kroon@mikrodidakt.se>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: Revert "fetch2: Cleanup file checksum generation"

This reverts commit 6dfb33fd7204e09815bde7ba5bf21941a73fe965.

It causes an unindeded change in behaviour.

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

bitbake: bitbake-diffsigs: Fix runtime error when no arguments are given

If bitbake-diffsigs is run without arguments it will error:

    Traceback (most recent call last):
      File "./bin/bitbake-diffsigs", line 121, in <module>
        if output:
    NameError: name 'output' is not defined

Fix this by moving the check for output into the inner else-clause.

(Bitbake rev: 7d3545a66863ad7183a7650b2af57eee091c45ae)

Signed-off-by: Jacob Kroon <jacob.kroon@mikrodidakt.se>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: fetch2: Cleanup file checksum generation

Cleanup the fix done in f9416e76e272ba3249abff099f6f3a47fe82e03e.

Instead of adding continue statements we can just move the last
statements into the final else-clause.

(Bitbake rev: 6dfb33fd7204e09815bde7ba5bf21941a73fe965)

Signed-off-by: Jacob Kroon <jacob.kroon@mikrodidakt.se>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: fetch2/git: Stop git from triggering fsync() calls

We only ever clone other repositories, if there were a problem such as power
failure, we'd blow away data and rebuild. As such we don't need fsync(). With
filesystems like ext*, the fsync pushes nearly all the data out to disk
which impacts all running processes.

We therefore set a configuration parameter to disable the fsync() calls.

Also fixup a case where basecmd wasn't being used for no good reason.

(Bitbake rev: 0a26abaf3a1e34d556c9375068dd17c879568d0f)

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

cross: Drop package tasks

Cross packages are not packaged so we can drop all the packaging tasks
for some small performance improvements.

(From OE-Core rev: be07344397b383ab5e56048acb64eb56272fe95e)

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

uclibc: Add libgcc-initial to DEPENDS

Now Separated out libgcc-initial is there

(From OE-Core rev: 161bfc91e92a5080d3d4c27b3bebcd9c5dac01e1)

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

default-distrovars.inc: Add libgcc-initial to the GPLv3 whitelist (same as libgcc)

(From OE-Core rev: 7f951c0185cea5174a8582bf6da4438636990121)

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

Globally replace 'base_contains' calls with 'bb.utils.contains'

The base_contains is kept as a compatibility method and we ought to
not use it in OE-Core so we can remove it from base metadata in
future.

(From meta-yocto rev: 5021839af713cb34bea5888981f2184104b6d35e)

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

Upstream-Status Cleanups

(From OE-Core rev: ea438b58c9a90e4c3147f99d63a9afc66963c5a1)

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

bumps to remove PRINC use in meta-fsl-ppc

(From OE-Core rev: 1a91d791a903ab837a7695cfe47828618b8be6a6)

Signed-off-by: Chunrong Guo <B40290@freescale.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

initscripts: Fix PR reversal

(From OE-Core rev: 343f1c49ed223cd188287d4815f9dd956376c94d)

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

dbus: fix a hard dependency about dbus-ptest

If image contains dbus and ptest is in DISTRO_FEATURES, dbus-ptest package
is installed, regardless of whether ptest-pkgs is in IMAGE_FEATURES. This
issue will increase size for most small images.
This patch fixes this problem.

[YOCTO #5702]

(From OE-Core rev: 0416583f014138656babdf78a574357ae5ff25bd)

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

udev-extraconf: update mount.sh to use /run/media instead of /media

This is done to work around the issue of auto-mounting block devices
(i.e. SD cards) when root filesystem is still in read-only mode and
creating /media/<device> mount-points by udev is not possible. That
is due to udev (/etc/rcS.d/S03udev) getting started earlier than
checkroot (/etc/rcS.d/S10checkroot.sh) gets a chance to re-mount the
rootfs as read-write.

Although, canonical FHS specifies /media/<device> as a mount point
for removable media devices, the latest 2.3 version was released in
2004 and since then FreeDesktop/udisks and other tools adopted the
new /run/media/<user>/<device> location. That was done to overcome
read-only rootfs limitation, since /run is usually a tmpfs mounted
partition, plus avoid name-clash between users.

For our embedded systems environment we assume single-user operation
and hence simplify mount point to just /run/media/<device>. But for
proper per-user mounting to /run/media/<user>/<device>, some sort of
session management is required along with the tool like udisks, that
is out of scope of this simple udev-based auto-mounting.

(From OE-Core rev: acfe3014d41de5e87cdbc58d0396349c6b9c3ffd)

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>

files/device_table-minimal.txt: cleanup

This device list is only used if USE_DEVFS is set to '0' (which is NOT the default since 2013-12-14,
see commit f54fdd6673a136ee1cee1f3263a8a7820de43ca3) and if IMAGE_DEVICE_TABLE and IMAGE_DEVICE_TABLES are not set.

This is, of course a very subjective list but minimal with 62 (special) files.
However, you should be able to boot correctly a lot of embedded linux around.

Changes:
- add sda, sdb, sda[1234], sdb[1234]
- add mmcblk0, mmcblk0p[1234]
- add rtc0, rtc1
- restrict maximum hdaX from 19 to 4 (pata is old)
- add ttyS1
- remove initctl, apm_bios, tty8, ttySA
- sorted entries

(From OE-Core rev: 72c011d272d0ffbd349858f16127058c74276866)

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

kernel: don't populate source symbolic link

/usr/src/kernel/source deployed by kernel-dev package is symbolically
linking to a build-time kernel source folder, which make no sense when
cross-compiling.

Fixed by not populating it at install stage.

(From OE-Core rev: edb85a9589be54a1e9c980aa669a380222a76cf4)

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

bind: add support for read-only rootfs

This patch adds support for read-only rootfs to the bind service.
Basically it just bind mounts several directories so that the bind
service could start correctly without reporting any error.

(From OE-Core rev: 99cc96eaee28bfde89096689b1296d28937ead88)

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>

opkg: Upgrade to v0.2.2

The patches for opkg have been rebased using git so that they apply cleanly onto
the new release.

(From OE-Core rev: 354e25fe51a8e85122898d1849008ace5dbe9461)

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

Globally replace 'base_contains' calls with 'bb.utils.contains'

The base_contains is kept as a compatibility method and we ought to
not use it in OE-Core so we can remove it from base metadata in
future.

(From OE-Core rev: d83b16dbf0862be387f84228710cb165c6d2b03b)

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

Globally replace oe.utils.contains to bb.utils.contains

BitBake has the exact same code as oe.utils.contains so there's no
reason to duplicate it. We now rely on the bb.utils.contains code for
metadata.

(From OE-Core rev: 93499ebc46547f5bf6dcecd5a786ead9f726de28)

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

openssh: enable ptest support

Install openssh test-suite and run it as ptest.

(From OE-Core rev: 4d0ecccae671bffb40c870a6e33d20be869b89bc)

Signed-off-by: Maxin B. John <maxin.john@enea.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

iproute2: upgrade to 3.14

- switched to ${BP} variable;
- removed parallel make restriction;

(From OE-Core rev: c27e1c8961e438f2c354ce3cd7682eaf98231776)

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

harfbuzz: upgrade to 0.9.27

(From OE-Core rev: 76447043a36472b7f673b423d584199d9d3dec8b)

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

bluez5: upgrade to 5.18

(From OE-Core rev: 9713a7ce28669db2456400f280ea19b67401e3f9)

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

connman: upgrade to 1.23

- License file copyright years updated;

(From OE-Core rev: 4c6b7634c21261f13b40036998b8204f8e3dee13)

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

gcc-cross-initial: Separate out libgcc-initial

Its useful to separate out the native (cross) binaries from the target
compilation. We already do this for libgcc, this now takes the same
approach for -initial.

(From OE-Core rev: 8a2aaf8b6bdca2d28c0047093c7f668750d57666)

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

gcc-cross: Improve handling of unwind.h

Rather than building the whole of libgcc to obtain the unwind.h header
file, simply configure it and then install the file. This avoids copying
chunks of data around when we don't need to and building the same thing
twice.

After doing this we need to make sure the target build directory exists
in the libgcc case since it will no longer be created automatically.

(From OE-Core rev: 459e4dc25462771038459567c22e87d4cd38b117)

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

libgcc: Spit out common code into libgcc-common.inc

Prepare the ground for the creation of libgcc-initial by splitting common
libgcc code into a libgcc-common.inc file.

(From OE-Core rev: 7c9f4a22d5c14d768be872beb9fc798163d6d204)

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

gcc: Convert to use hardlinkdir

(From OE-Core rev: 204bc1f39030a3c0dd3eadadabb013aca8bb9cc6)

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

utils: Add hardlinkdir shell function

In a number of places it would be helpful to be able to copy trees of
files using hardlinks. This turns out to be harder than you'd expect
since there is no good single command that does this well and handles
all file types correctly.

Abstracting this into a function therefore makes sense, cpio seems
as good an option as any other.

(From OE-Core rev: 14ef03182e1e5fe7cf6b4112f29e6609631aa78d)

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

build-appliance-image: Update to head revision

(From OE-Core rev: 866d2854139e0b00063dde06bc3a34d635ae2fc8)

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

native.bbclass: Override TARGET_ flags too

TARGET_LDFLAGS is currently defined in bitbake.conf to contain
${TARGET_LINK_HASH_STYLE} which differs between MIPS and other
targets. Since TARGET_LDFLAGS is an exported variable it affects the hash
of every shell task even if it is not used.

We don't want native recipe tasks to have different hashes purely because
they happen to have been built in order to satisfy dependencies for
different MACHINEs since this causes lots of churn in the native sysroot
when switching between MACHINEs.

Making native.bbclass override TARGET_LDFLAGS to use BUILD_LDFLAGS ensures
consistent hashes and is a sensible thing to be doing anyway.

Although they don't appear to have the same detrimental affect on task
hashes TARGET_CPPFLAGS, TARGET_CFLAGS and TARGET_CXXFLAGS should be
overridden too.

(From OE-Core rev: 05a70ac30b37cab0952f1b9df501993a9dec70da)

Signed-off-by: Mike Crowe <mac@mcrowe.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

cryptodev-tests: recipe for cryptodev test suite based on OpenSSL

(From OE-Core rev: c54fa061da6195081cd29817a351a36377b58e53)

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

cryptodev-module: recipe for out-of-tree cryptodev device driver

Depends on cryptodev-linux for providing a header file

(From OE-Core rev: b7587d2ef7642dcc248744ade8f85f815185e78c)

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