openssh: Upgrade 7.2p2 -> 7.3p1

Remove CVE-2015-8325.patch as it's included upstream. Rebase another
patch.

(From OE-Core rev: 4b695379dcf378e8d77deaf7e558e8cbd314683c)

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

wayland-protocols: 1.5 -> 1.7

* xdg-shell unstable v6 (backwards incompatible)
* new unstable protocols xdg-foreign, idle-inhibit

(From OE-Core rev: e3ea73039af5fbde52788188b750383aa5d6c2c8)

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

libinput: Upgrade 1.3.0 -> 1.4.1

(From OE-Core rev: 33800186dbfa3a4b28ece558c9ff1eb68b99d54d)

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

json-glib: Upgrade 1.2.0 -> 1.2.2

(From OE-Core rev: a00280a96cd770e6c26e30eab10cc49b54d4992b)

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

vte: Upgrade 0.44.1 -> 0.44.2

(From OE-Core rev: 68898cf20f70fc7e7f517111ea7c2b901859263e)

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

gtk+3: Upgrade 3.20.6 -> 3.20.9

(From OE-Core rev: e82bdb1e5ddecb347a75098d53f4db2d0b5aa853)

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

gnome-themes-standard: Upgrade 3.18.0 -> 3.20.2

(From OE-Core rev: 1d0dcb0e8bc7b5f11b8249053008d9860a40dc61)

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

fontconfig: Upgrade 2.12.0 -> 2.12.1

License text block moved, checksum remains same.

(From OE-Core rev: dbda47cab8742888189131716415777155105d9d)

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

libgcrypt: upgrade to 1.7.3

(From OE-Core rev: 0a6c2db4d79288fc8c9bebbf7d93bf142d358f7e)

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

rm_work: don't remove timestamps of image tasks

Excluded removal of do_bootimg, do_bootdirectdisk and do_vmimg
timestamps to prevent unneeded rootfs rebuilds.

[YOCTO #10159]

(From OE-Core rev: f214da502ad7eda27460dc6f06e9cd29a114f2d2)

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

python-3.5-manifest.inc: the signal module RDEPENDS on enum

Fixed:
$ python3
>>> import signal
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/path/to/sdk/sysroots/x86_64-pokysdk-linux/usr/lib/python3.5/signal.py", line 4, in <module>
    from enum import IntEnum as _IntEnum
ImportError: No module named 'enum'

(From OE-Core rev: 6306dc8351c19059c4c2a8e75bb5733e64532732)

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

libidn: 1.32 -> 1.33

1)Upgrade libidn from 1.32 to 1.33.
2)Modify LIC_FILES_CHKSUM, since the date in it has been changed, But the LICENSE has not been changed

(From OE-Core rev: fa042b49a3a1a78ae28b19e66b30c279da65963a)

Signed-off-by: Dai Caiyun <daicy.fnst@cn.fujitsu.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

useradd_base: avoid unintended expansion for useradd parameters

Now, useradd dollar sign requires three prepending backslash characters to
avoid unintended expansion. It used to be just one prepending backslash
character before Krogoth. Restore that behaviour.

[YOCTO #10062]

(From OE-Core rev: 9e43a73c7ad576666d53c8c9e0283bc6bb9087a8)

Signed-off-by: Niko Mauno <niko.mauno@vaisala.com>
Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

xorg-proto: remove stale git recipes

These two recipes are old and unmaintained, so remove them to avoid confusion
with the tarball recipes.

(From OE-Core rev: edf5b379b4c111fd9870fb3ae139d88fcd9e752d)

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

inputproto: explicitly disable asciidoc to avoid floating dependency

(From OE-Core rev: d6bb98d0c432d8f4ffaf74f63aca61354565a546)

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

gettext_0.16.1: whitespace changes to align with v0.19.8.1

This further aligns this recipe with the GPLv3 version to make
it easier to  spot differences between the two recipes.

(From OE-Core rev: e25a533e8ca2fc1fa897df252830825cb9a5f028)

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

gettext_0.16.1: align configure options with v0.19.8.1 recipe

It doesn't look like we need any of those features, so
let's disable them explicitly.

(From OE-Core rev: 0a095473eec333f918ef831dea1c2f269a64fc62)

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

gettext_0.16.1: fix lispdir configure option

The option is called --with(out)-lispdir, not --with(out)-lisp

(From OE-Core rev: 422c92d2806f776252c15ec9fe204b204503c4d2)

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

gettext_0.16.1: use musl gettext implementation

gettext uses internal symbols to detect whether the
implementation is compatible with GNU gettext. However,
these symbols don't are not part of the public API, they
are specific to glibc.

While musl implements the GNU gettext *API* version 1 and 2
  http://www.openwall.com/lists/musl/2015/04/16/3
it doesn't implement glibc internals. This means that
gettext fails to detect musl's working implementation.

More recent versions of gettext have changed the way
GNU gettext compatibility is done
  https://lists.gnu.org/archive/html/bug-gettext/2016-04/msg00000.html
  http://git.savannah.gnu.org/cgit/gettext.git/commit/gettext-runtime/m4/gettext.m4?id=b67399b40bc5bf3165b09e6a095ec941d4b30a97
and while we could backport the corresponding patch to
gettext.m4, we avoid doing that so as to avoid any
potential GPL-v3 issues.

So instead we force ./configure to assume that the gettext
implementation of the c-library (musl) is compatible.

As a side-effect, this also reduces image sizes as the
internal gettext implementation isn't built anymore, and
it's otherwise packaged into the main gettext package
which blows up the image as the main gettext package
contains a lot of things.

Similarly, libintl.h isn't generated anymore, as the one
from musl is OK.

(From OE-Core rev: 948f0bd162f0b1b0375db884e99a2338f47e8527)

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

insane: improve package_qa_clean_path

Instead of just removing TMPDIR from the path for display, optionally allow a
package to be passed and remove PKGDEST/package too.

This means that messages that specify a package name can pass that name and the
resulting path will be absolute inside that package.

(From OE-Core rev: 55061a43926baf6ff0e17aed02efd299ebba3c24)

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

grub: fix load module all_video failed

While using oe-core toolchain to strip grub module 'all_video.mod',
it stripped symbol table:
--------------
root@localhost:~# objdump -t all_video.mod

all_video.mod:     file format elf64-x86-64

SYMBOL TABLE:
no symbols
--------------

It caused grub to load module all_video failed.
(This module will be loaded by defalut which configed in grub.cfg)
--------------
grub> insmod all_video
error: no symbol table.
--------------

Tweak strip option to keep symbol .module_license could workaround
the issue.
--------------
root@localhost:~# objdump -t all_video.mod

all_video.mod:     file format elf64-x86-64

SYMBOL TABLE:
0000000000000000 l    d  .text  0000000000000000 .text
0000000000000000 l    d  .data  0000000000000000 .data
0000000000000000 l    d  .module_license        0000000000000000 .module_license
0000000000000000 l    d  .bss   0000000000000000 .bss
0000000000000000 l    d  .moddeps       0000000000000000 .moddeps
0000000000000000 l    d  .modname       0000000000000000 .modname
--------------

(From OE-Core rev: 17e7eb96e5446821ad81977ac9ccac26b05e67a7)

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

bash: fix run-intl ptest failed

1. Filter the extra white space in intl.right

   When the sub-test unicode2.sub of intl.tests executed, it produced
   compact results without extra white space, compared to intl.right,
   it failed.

   So we need to filter the extra white space in intl.right.

   Import this patch for intl.right from bash devel branch:

http://git.savannah.gnu.org/cgit/bash.git/log/?h=devel

   Commit is:

85ec0778f9d778e1820fb8c0e3e996f2d1103b45

2. Change intl.right correspond to the unicode3.sub's output

   In sub-test unicode3.sub of intl.tests have this:

printf %q "$payload"

   The payload variable was assigned by ASCII characters, when using
   '%q' format strings, it means print the associated argument shell-quoted.

   When the strings contain the non-alpha && non-digit && non-punctuation &&
   non-ISO 646 character(7-bit), it would output like this: " $'...', ANSI-C
   style quoted string. We can check the bash source code at:

http://git.savannah.gnu.org/cgit/bash.git/tree/builtins/printf.def#n557
http://git.savannah.gnu.org/cgit/bash.git/tree/lib/sh/strtrans.c#n331

   So we need to change the intl.right contain the correct output of unicode3.sub.

   Import parts of this patch for intl.right from bash devel branch:

http://git.savannah.gnu.org/cgit/bash.git/log/?h=devel

   Commit is:

74b8cbb41398b4453d8ba04d0cdd1b25f9dcb9e3

3. Add the sanity check for locales

   When run the intl.tests, we need the following locales:

en_US & fr_FR & de_DE

   So add the locales check for the intl.tests in run-ptest.

(From OE-Core rev: 640676226bb351420a0a8b2d2a3c120ae42da11e)

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

bootchart2: fixes a BOOTLOG_DEST typo

A flaw was observed in bootchartd that BUILDLOG_DEST should actually be
BOOTLOG_DEST, this seems to be a typo or mix-up which has been fixed in
upstream.

Cherry pick the fix since bootchart2 0.14.8 is still the newest release
so far.

(From OE-Core rev: 299e67291f3d396ba93f4c4a94120228bb9b1d88)

Signed-off-by: Ming Liu <peter.x.liu@external.atlascopco.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

adwaita-icon-theme: 3.18.0 -> 3.20

1) Upgrade adwaita-icon-theme from 3.18.0 to 3.20.
2) Delete DEPENDS, since intltool is not needed.

(From OE-Core rev: c3fa2eca5d2667c668641373948acfb7172ff2e8)

Signed-off-by: Wang Xin <wangxin2015.fnst@cn.fujitsu.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

initscripts: Check for logrotate in dmesg.sh

Autodetect previously hardcoded logrotate location because it can be
installed in multiple places like /usr/bin/logrotate which is very
common besides /usr/sbin

(From OE-Core rev: 277a5975d43125623b5a51ddcb48f9ee2474d0fc)

Signed-off-by: Ovidiu Vancea <ovidiu.vancea@ni.com>
Signed-off-by: Ioan-Adrian Ratiu <adrian.ratiu@ni.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

python3-native: Extend python3-native rproviders

Add the following modules to RPROVIDES:

  - python3-email-native
  - python3-io-native
  - python3-json-native
  - python3-lang-native
  - python3-misc-native
  - python3-netclient-native
  - python3-netserver-native
  - python3-numbers-native
  - python3-pkgutil-native
  - python3-pprint-native
  - python3-re-native
  - python3-shell-native
  - python3-subprocess-native
  - python3-threading-native
  - python3-unittest-native

(From OE-Core rev: 1a62ffd108e6aa7b7e5d0a81819550e8a7afeb60)

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

python3-native: Change code style for rprovides

Use a more readable code style for RPROVIDES and sort recipes
alphabetically.

(From OE-Core rev: 21130e2afc4762ad84c86e377146b99224d16032)

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

grub-efi.bbclass: Fix path in startup.nsh for iso image.

The path in startup.nsh for iso image is corrupted as follows:
fs0:\EFI\BOOT^Hootx64.efi

Using printf will emit correct path which is:
fs0:\EFI\BOOT\bootx64.efi

This happens because of echo command. Switching to printf
like the one used in efi_populate() function.

(From OE-Core rev: 7540b9e68d56e7779b478d2bc09fbbedcf28976b)

Signed-off-by: Pranav Tipnis <pranav.tipnis@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

glibc: Fix scope resolution in glibc to be breadth first.

The ELF specification indicates symbol resolution should be breadth first, not
depth first.

The dl-deps.c: dl_build_locale_scope function is processing in a depth first
mode.  This is causes certain symbols to be incorrectly reported when
LD_TRACE_PRELINKING=1 is enabled.

See glibc BZ #20488 for more information.

(From OE-Core rev: fb72263eaa94e64ddeee457b5b1bc999f0e647da)

Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-yocto/4.4: fix configuration warnings

Integrating the following commits to address configuration warnings for
intel-corei7-64 and intel-core2-32:

  features: Fix dependencies and =m vs =y discrepancies for corei7
  intel-core2-32.cfg: Explicitly disable CONFIG_64BIT

(From OE-Core rev: b2a4e07390834fa41fe35d1124ac2a0cd6692524)

Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-yocto/4.1/4.4: -stable updates and configuration changes

Updating the 4.4 kernel to v4.4.18 and the 4.1 kernel to v4.1.30.

We also tweak the configuration with the following commits to remove
warnings being generated from the 4.4 kernel (due to options being
dropped from the final .config):

  features: Create mfd-intel-lpss feature and use where appropriate
  features/iio: Set IIO_BUFFER_CB to =m instead of =y
  features: Add 6lowpan feature and add it where necessary

Tested on qemux86, qemuppc, qemumips and qemuarm.

(From OE-Core rev: 18c6fb387aa6a15de514030c4a7c04dac9c68869)

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

maintainers.inc: update maintainers for Dengke

* Take recipes from Jussi Kukkonen
* Take recipes from Kai, Wenzong and Yi.

(From meta-yocto rev: 508dfcf39e09661950c408497fa23ee8a8e20f55)

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

bitbake: siggen: Fix file variable typo in compare_sigfiles

(Bitbake rev: deab9a30987b225922490ca186c5307c15d45b82)

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

bitbake: fetch2: preserve current working directory

Fix the methods in all fetchers so they don't change
the current working directory of the calling process, which
could lead to "changed cwd" warnings from bitbake.

(Bitbake rev: 6aa78bf3bd1f75728209e2d01faef31cb8887333)

Signed-off-by: Matt Madison <matt@madison.systems>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: dump_cache.py: use python3 as interpreter

Fixed:
  File "bitbake/contrib/dump_cache.py", line 39
    print("Error, need one argument!", file=sys.stderr)

(Bitbake rev: 435c6fb838b9f38c0477bcc2f07c8ce22999132b)

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

bitbake: toasterui, knotty: don't print taskid followed by taskstring which are now in most cases identical

* unify the format how the task is described
* don't show taskid followed by taskstring as the taskstring is
  different only for setscene tasks (by _setscene suffix)
* the duplicated output was introduced by:
  2c88afb   taskdata/runqueue: Rewrite without use of ID indirection
  as reported and confirmed as a bug here:
  http://lists.openembedded.org/pipermail/openembedded-core/2016-June/123148.html
* show:
  NOTE: Running task 541 of 548 (/OE/build/oe-core/openembedded-core/meta/recipes-core/zlib/zlib_1.2.8.bb:do_package)
  instead of much longer:
  NOTE: Running task 541 of 548 (ID: /OE/build/oe-core/openembedded-core/meta/recipes-core/zlib/zlib_1.2.8.bb:do_package, /OE/build/oe-core/openembedded-core/meta/recipes-core/zlib/zlib_1.2.8.bb:do_package)

  and similarly for failed tasks:
  ERROR: Task (virtual:native:/OE/build/oe-core/openembedded-core/meta/recipes-core/zlib/zlib_1.2.8.bb:do_install) failed with exit code '1'
  instead of much longer:
  ERROR: Task virtual:native:/OE/build/oe-core/openembedded-core/meta/recipes-core/zlib/zlib_1.2.8.bb:do_install (virtual:native:/OE/build/oe-core/openembedded-core/meta/recipes-core/zlib/zlib_1.2.8.bb:do_install) failed with exit code '1'

(Bitbake rev: 696693d45f5eff1226866ed79dbfb67161d8cd3f)

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

bitbake: tests: add unit tests for the usehead url parameter

[YOCTO #9351]

(Bitbake rev: 63031c0236ace10a9d52b9db9bbb892c1b4bf7db)

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: git: Allow local repos to use HEAD

Introduce a new 'usehead' url parameter for git repositories. Specifying
usehead=1 causes bitbake to use whatever commit the repository HEAD is
pointing to. Usage of usehead=1 is only allowed for local git
repositories, i.e. it must always be accompanied with protocol=file url
parameter.

[YOCTO #9351]

(Bitbake rev: 2673fac5a9d06de937101e3fb2ddf1e60ff99abf)

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: bitbake-selftest: enable bitbake logging to stdout

Now you get the bb logger output for failed tests. This helps debugging
problems. Also, all stdout/stderr data for successful tests is silenced
which makes for less cluttered console output.

(Bitbake rev: ea19972a16f7639f944823d1d8a7728105460136)

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: bitbake-selftest: introduce BB_TMPDIR_NOCLEAN

Set this env variable to 'yes' to preserve temporary directories used by
the fetcher tests. Useful for debugging tests.

(Bitbake rev: 04132b261df9def3a0cff14c93c29b26ff906e8b)

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: bitbake-selftest: add help text for env variable(s)

(Bitbake rev: 94c63a5b1e731e64eb8efbc09f2ab6a0ce11df05)

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: bitbake-selftest: utilize unittest.main better

This simplifies the script, and, gives new features. It is now possible
to run single test functions, for example. This is nice when writing new
test cases.

(Bitbake rev: 8c513580b9406b031674f799117eae7410f8e01c)

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Revert "local.conf.sample: Disable ARM and PPC due to prelink test case failures"

This reverts commit 85d30c28277a040420c2b2f25028ae1500da54db.

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

packagefeed-stability.bbclass: cleansstate should remove pkgs from deploy dir

"bitbake recipe -ccleansstate" should remove binary pkgs from deploy dir
as normal cleansstate does without packagefeed-stability.bbclass.

(From OE-Core rev: 0865a5b8b8fbf478fb4b2310f808bcffff84a091)

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

libdrm: 2.4.68 -> 2.4.70

Upgrade libdrm from 2.4.68 to 2.4.70.

(From OE-Core rev: 0f9ce74cb62afdd3a0c700be223d0ae0f88daa05)

Signed-off-by: Dai Caiyun <daicy.fnst@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

glib-2.0: 2.48.1 -> 2.48.2

1) Upgrade glib-2.0 from 2.48.1 to 2.48.2.
2) Modify Enable-more-tests-while-cross-compiling.patch, since the data has changed.

(From OE-Core rev: f5af2742003b06f117ba34683cefd168cc78b5a0)

Signed-off-by: Wang Xin <wangxin2015.fnst@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

kbd: remove PARALLEL_MAKEINST = ""

There isn't anything wrong when looked into its Makefile, I guess that
it had been fixed during ugprade, and I've applied this patch locally
for more than 2 months, there isn't anything wrong.

(From OE-Core rev: 53687cadaab307fc843768d61973ed1630eb28af)

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

kmod: upgrade to 23

(From OE-Core rev: 651a08c9eda35edc31e637268be45cda0a439b6d)

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

diffutils: upgrade to 3.4

(From OE-Core rev: 98a23eaf837692ad7d2a1d04318118c41052f7b0)

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

util-linux: upgrade to 2.28.1

(From OE-Core rev: 76e9ea8e5c74ad7ab78138bd330f70d69931410c)

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

bootchart2: Add ALTERNATIVE configuration for bootchartd

Since busybox also provides the bootchartd command use the
update-alternatives mechanism to address this.

Also let bootchartd-stop-initscript RDEPENDS on bootchart2, since
/sbin/bootchartd is being called in that script.

Ming Liu <peter.x.liu@external.atlascopco.com>

(From OE-Core rev: 4c4f440d3a8eb6171f619bceacf57835d1b9841a)

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

asciidoc-native: add dependency on docbook-xml-dtd4-native

During the compilation phase asciidoc runs a2x for validation
of some xmls which in turn invokes xmllint with --nonet
parameter that requires DTDs to be available locally in order
to succeed otherwise the do_compile fails.
We now add a direct dependency on docbook-xml-dtd4 so the
DTDs are always available locally.

(From OE-Core rev: 14be679c7b8241b2d0872242ed358e5eb4f7acac)

Signed-off-by: Awais Belal <awais_belal@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

openssl: fix add missing dependencies building for test directory

Regarding the last commit about missing dependencies, another issue
was found. The problem was found, while ptest has been built with some
set extra settings. It means, when ptest is going to be built,
it is necessary to rebuild dependencies for test directory too.

(From OE-Core rev: 030142d0410bec85aeacfff6be27d5fed41ce808)

Signed-off-by: Andrej Valek <andrej.valek@siemens.com>
Signed-off-by: Pascal Bach <pascal.bach@siemens.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libffi: fix a typo (mips)

While code elsewhere checks for
MIPS_INSTRUCTION_SET == mips16e in order to decide how
to compile, hence the typo doesn't affect behaviour, the
intention was to set it to 'mips', as is done everywhere
else. Fixing the typo also helps to avoid confusion.

(From OE-Core rev: 45b27564324c754a34a1930437a7167079fe1ee4)

Signed-off-by: André Draszik <git@andred.net>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

image-vm.bbclass: remove old images if RM_OLD_IMAGE is enabled

[YOCTO #10164]

(From OE-Core rev: 3762b42233651832c5909d7a3e873365fc0a9756)

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

oeqa.buildperf: fix crash when creating globalres.log

Fix a bug that was introduced when converting to unittest framework.

(From OE-Core rev: 3bdb7b2e512b2f160360e95ed5b2be3871ec0b4b)

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

oe-build-perf-test: align log message format with testrunner output

The previous attempt on this was a bit erroneous, dropping time stamps
completely although only the timestamp format should've been changed.

(From OE-Core rev: bafcff95e2b5e0b9a8c76ce46a62667bf6f49b00)

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

syslinux.bbclass: ensure creation of output directory

build_syslinux_cfg function creates syslinux configuration file.
The code assumes that the output directory exists, which is not
always the case. For example rm_work task removes rootfs directory
structure and causes build_syslinux_cfg to fail with this error:
Unable to open ../<image>-<version>/syslinux_vm.cfg

Made build_syslinux_cfg depend on output directory to ensure that
directory is created before running the function.

[YOCTO #10159]

(From OE-Core rev: c39b072fa7e96f385da338a727c67e607308d637)

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

busybox: Add parallel make fix

We're seeing regular parallel make failures in applet headers in busybox.
This adds a patch to try and avoid the issue, building upon a fix already
backported from upstream. The patch has been sent to upstream.

[YOCTO #10116]

(From OE-Core rev: 199cef0e8a50b20d0ee6fefd1d4cf3372eba7728)

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

sanity.bbclass: Ensure we expand BUILD_PREFIX

This likely used to work when we expanded python functions and broke when
we stopped. Since it defaults to "", it never caused an issue but
is incorrect usage so fix it.

(From OE-Core rev: bfb395fdea642b306f110b4b8f1046f1992c622c)

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

local.conf.sample: Disable ARM and PPC due to prelink test case failures

Internal prelink test cases reloc8 and reloc9 are failing on both ARM
and PPC systems.  Disable them by removing the prelink from the
IMAGE_CLASSES setting.

(From meta-yocto rev: 85d30c28277a040420c2b2f25028ae1500da54db)

Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Revert "local.conf.sample: Disable prelink by default"

This reverts commit 300f858ba07c938427ccd05a3d7220027a03d461.

Reenable prelink

(From meta-yocto rev: 91705d8ae9f56d1de4f0fdcd6a9654b75921aa8c)

Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

prelink: Move to latest version of prelink

* Uprev rtld emulation to glibc-2.23
* Fix compilation warnings
* Add additional debug scopes
* Change rtld build_local_scope to be breadth-first
* Fix LD_PRELOAD emulation
* Change function reordering to work with latest binutils

(From OE-Core rev: 9d2c82f7d3fc0fdafc5c4fdd1707324bc4cdbf22)

Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gcc: Upgrade to 6.2 RC1

(From OE-Core rev: 41ce4b438795108025c79cd3eec067367d53623e)

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

libunwind: Fix build race conflict with gcc and musl

Building libunwind, then gcc-runtime causes build failures. This is hard
to fix since gcc-runtime wants the internal gcc unwind.h header but libunwind
wants to provide this. There are differences in include behaviour between gcc
and glibc which are by design.

This patch hacks around the issue by looking for a define used during gcc-runtime's
build and skipping to the internal header in that case. The patch is only enabled
on musl and is the best workaround I could come up with to unblock failing builds
on our autobuilder.

[YOCTO #10129]

(From OE-Core rev: cd8b64b0a236b27e5383e2394de65b9bfd4b6677)

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

ref-manual: swabber: remove from documentation

Remove documentation as swabber was removed from oe-core with
this commit: commit a7ddbea345c90646e6b9ddb89f768865caffdf07

(From yocto-docs rev: f3c462b2c6aa20de53c77e5d93cf397ae36cb2bb)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Update SERIAL_CONSOLES_CHECK description

The previous description was not accurate. Looking at the code,
SERIAL_CONSOLES_CHECK does not act like SERIAL_CONSOLES, as it
will not add consoles to enable but only check and disable
consoles defined by SERIAL_CONSOLES. Also, the previous patch
adds aliasing functionality that needed to be documented.

I (Scott Rifenbark) did a bit of word-smithing here from the
original patch.

(From yocto-docs rev: 55d07048e831f0dbc955b74e029fe26ed276675b)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: change SYSLINUX_ROOT to ROOT_VM for DISK_SIGNATURE variable

The SYSLINUX_ROOT variable was renamed to ROOT_VM in krogoth.

(From yocto-docs rev: c4bbe8bc4967dd631b939f6806d65e2862df3424)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Applied review edits to "Viewing Dependencies"

Fixes [YOCTO #10131]

Fixed some small issues here and there.  Also, provided a
second itemized item in the note box turning it into a
notes box.

(From yocto-docs rev: a736c3bb707e81eda7760c642084a5a7c4de2539)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Removed the "pid" stuff from viewing failed tasks

Fixes [YOCTO #10132]

My attempt to be complete on the filenames that have a "pid"
portion were not correct.  I have removed them from the first
paragraph.

(From yocto-docs rev: 8261b93b39df9abc9f9d6ccb4c00dc11330ad516)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Various small corrections to package-related stuff

Fixes [YOCTO #10135]

Some small problems were fixed:

 * Added a cross-reference in the FILES glossary entry to the
   PACKAGES variable.  The two are tied and there was not a
   reference to it.

 * Removed a redundant "/" character in a pathname example in
   the dev-manual.

 * Removed a redundant "/" character in an example pathname
   in the FILES glossary description.

(From yocto-docs rev: 11a397c232696deece7ac5c6dafcadb87d7a5775)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updated the "Viewing Logs from Failed Tasks" section.

Fixes [YOCTO #10132]

Provided a better description and removed a deprecated sentence
near the end.

(From yocto-docs rev: bbe588e19bb9ed58883ae7c770da551de659e982)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updated the section on viewing dependencies

Fixes [YOCTO #10131]

The section was renamed "Viewing Dependencies" for consistency.
The section was moved up to be the third item in the sub-section
list.  The section was extensively re-written to provide more
clarity and options for the user to view dependencies.

(From yocto-docs rev: d521c3aabe6ded105cde6f7b3563c85340f759fd)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Clarify and flesh out debugging using bitbake -e

Fixes [YOCTO #10099]

Renamed the log file section to better describe what the user
is accomplishing.

Renamed and repositioned the variables section to better describe
and emphasize the task.  Also fleshed out the variables section with
more information.

(From yocto-docs rev: 0606fe481416a07bf98fc8ae79a30c1d62e75e6d)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

dev-manual: Added a new "known issue" for running qemu.

Fixes [YOCTO #9285]

Added a new bullet item to note that Using QEMU in usermode
might not work properly when running 64-bit binaries under
32-bit host machines. In particular, "qemumips64" is known
to not work under i686.

(From yocto-docs rev: 896beb3fddd427f8327d4ddd35be253866c90377)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updated PROVIDES and FILES variable descriptions

Fixes [YOCTO #10094]

For PROVIDES, I added information about how the do_package
task goes through PACKAGES and uses the FILES variable
corresponding to each package to assign files to the package.

For FILES, I added a blurb to the existing note how you can
find default values for the FILES* variables.

(From yocto-docs rev: c70f79a608076c5c0490918b87986554bc5d8353)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updated the SERIAL_CONSOLES_CHECK variable description.

Provided a better, more accurate description of this variable.

(From yocto-docs rev: 020f927bc01d662601fb44b19e4c6bc70e5e5ee7)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Changed the BPN variable description.

Fixes [YOCTO #10068]

Removed redundant wording.

(From yocto-docs rev: b6c9c979a01c8070d3d2c23340d3c0f5ef358157)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updates to PARALLEL_MAKE, PARALLEL_MAKEINST, EXTRA_OEMAKE

Fixes [YOCTO #10070]

Updated these three variables with various items to make clear
that PARALLEL_MAKE and PARALLEL_MAKEINST won't work unless
EXTRA_OEMAKE is passed to "make".

(From yocto-docs rev: 4f8b56cc67502cd672e0296cf2f143ecbcde22ac)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Added note to PROVIDES variable description.

Fixes [YOCTO #10069]

Added a note at the end of the variable description to
explain how runtime virual dependencies work.

(From yocto-docs rev: de1d16017c27b6b2502735fc41acd22660f6e7b9)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: New "Fakeroot and Pseudo" section.

Fixes [YOCTO #10060]

I provided a new section in the Technical Details chapter.  Also
some extra explanation was added to both the do_install task and to
the D variable.

(From yocto-docs rev: 565fb11d72bf8c585469bcf65f92b6738e344813)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Added new variable description EXTRANATIVEPATH.

Fixes [YOCTO #10002]

Created new base description for EXTRANATIVEPATH.

(From yocto-docs rev: aafc2de2657203440fe4b0bf3895cf367063bed1)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Applied review changes to INITRAMF_IMAGE variable.

Fixes [YOCTO #10012]

There was a mistake referring to the wrong example recipe.  I
changed "core-image-sato-initramfs.bb" to
"core-image-minimal-initramfs.bb" for the fix.

(From yocto-docs rev: 4d63e1fc5786556dd0dd5ca1435252d43dbd745a)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updated the INITRAMFS_IMAGE_BUNDLE variable description.

Fixes [YOCTO #10013]

I enhanced the description with more detail all around.

(From yocto-docs rev: 319dabecf5abf0884295b991f681bed0e1dbf673)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updated the do_compile task section.

Fixes [YOCTO #9964]

There was a small fix to change oe_runmake task to oe_runmake
function.

(From yocto-docs rev: aa049c9165c67e041c84fab9fabfbe98828c79bb)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updates to INITRAMFS_IMAGE and INITRAMFS_FSTYPES

Fixes [YOCTO #10012]

Applied review comments.

(From yocto-docs rev: 1862d03c1916c34c29c1aea86fbb6ee4c7f34cca)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updated the INITRAMFS_FSTYPES variable.

Fixes [YOCTO #10012]

Added a new paragraph to the end explaining the default behavior.

(From yocto-docs rev: 1591f96fc04f64906145f272d205ec6c44ac70c0)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updated the INITRAMFS_IMAGE variable.

Fixes [YOCTO #10012]

Updated the description completely.  New more detailed information.

(From yocto-docs rev: cb6ce91674ab092324f97ca4e56a0cbcd9140fbe)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ref-manual: Updated the PROVIDES variable description.

Added more information about virtual targets to the end of the
description.

Fixes [YOCTO #10011]

(From yocto-docs rev: ce7ae0c6ad4ad3a0c2422b797556563dc48a9a5b)

Signed-off-by: Scott Rifenbark <srifenbark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

systemd-compat-units: Only enable for systemd in DISTRO_FEATURES

This recipe only makes sense when systemd is enabled and otherwise causes
world build failures.

(From OE-Core rev: 5dca6cc2fcdb2799c19b1697f0647a16ce296290)

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

systemd-compat-units: pkg_postinst() does not work

The test for various files is wrong and will always be
true, even if init.d does not exist.

Exit if init.d does not exist, and correctly test for
file existence otherwise.

(From OE-Core rev: 8183309080aee45746daaff46b0506b09b5bd269)

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

sanity: Require bitbake 1.31.1 for multi-config changes

(From OE-Core rev: ecc8d346223030ee06aa6d83427f757982e948e4)

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

devtool/recipetool/meta: Adapt to bitbake API changes for multi-configuration builds

Unfortunately to implenent multiconfig support in bitbake some APIs
had to change. This updates code in OE to match the changes in bitbake.
Its mostly periperhal changes around devtool/recipetool

[Will need a bitbake version requirement bump which I'll make when merging]

(From OE-Core rev: 041212fa37bb83acac5ce4ceb9b7b77ad172c5c3)

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

bitbake: bitbake: Update version to 1.31.1

(Bitbake rev: 3ff1c66e6f336e5de7dcbc983a97fcd19ddc6b81)

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

bitbake: bitbake: Initial multi-config support

This patch adds the notion of supporting multiple configurations within
a single build. To enable it, set a line in local.conf like:

BBMULTICONFIG = "configA configB configC"

This would tell bitbake that before it parses the base configuration,
it should load conf/configA.conf and so on for each different
configuration. These would contain lines like:

MACHINE = "A"

or other variables which can be set which can be built in the same
build directory (or change TMPDIR not to conflict).

One downside I've already discovered is that if we want to inherit this
file right at the start of parsing, the only place you can put the
configurations is in "cwd", since BBPATH isn't constructed until the
layers are parsed and therefore using it as a preconf file isn't
possible unless its located there.

Execution of these targets takes the form "bitbake
multiconfig:configA:core-image-minimal core-image-sato" so similar to
our virtclass approach for native/nativesdk/multilib using BBCLASSEXTEND.

Implementation wise, the implication is that instead of tasks being
uniquely referenced with "recipename/fn:task" it now needs to be
"configuration:recipename:task".

We already started using "virtual" filenames for recipes when we
implemented BBCLASSEXTEND and this patch adds a new prefix to
these, "multiconfig:<configname>:" and hence avoid changes to a large
part of the codebase thanks to this. databuilder has an internal array
of data stores and uses the right one depending on the supplied virtual
filename.

That trick allows us to use the existing parsing code including the
multithreading mostly unchanged as well as most of the cache code.

For recipecache, we end up with a dict of these accessed by
multiconfig (mc). taskdata and runqueue can only cope with one recipecache
so for taskdata, we pass in each recipecache and have it compute the result
and end up with an array of taskdatas. We can only have one runqueue so there
extensive changes there.

This initial implementation has some drawbacks:

a) There are no inter-multi-configuration dependencies as yet

b) There are no sstate optimisations. This means if the build uses the
same object twice in say two different TMPDIRs, it will either load from
an existing sstate cache at the start or build it twice. We can then in
due course look at ways in which it would only build it once and then
reuse it. This will likely need significant changes to the way sstate
currently works to make that possible.

(Bitbake rev: 5287991691578825c847bac2368e9b51c0ede3f0)

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

bitbake: siggen: properly close files rather than opening them inline

If you don't do this, with Python 3 you get a warning on exit under some
circumstances.

(Bitbake rev: 49502685df3e616023df352823156381b1f79cd3)

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

bitbake: ast/ConfHandler: Add a syntax to clear variable

unset VAR
will clear variable VAR
unset VAR[flag]
will clear flag "flag" from var VAR

(Bitbake rev: bedbd46ece8d1285b5cd2ea07dc64b4875b479aa)

Signed-off-by: Jérémy Rosen <jeremy.rosen@openwide.fr>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake: cache: Build datastores from databuilder object

Rather than passing in a datastore to build on top of, use the data builder
object in the cache and base the parsed recipe from this. This turns
things into proper objects building from one another rather than messy
mixes of static and class functions.

This sets things up so we can support parsing and building multiple
configurations.

(Bitbake rev: fef18b445c0cb6b266cd939b9c78d7cbce38663f)

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

bitbake: cache: Split Cache() into a NoCache() parent object

There are some cases we want to parse recipes without any cache
setup or involvement. Split out the standalone functions into
a NoCache variant which the Cache is based upon, setting the scene
for further cleanup and restructuring.

(Bitbake rev: 120b64ea6a0c0ecae7af0fd15d989934fa4f1c36)

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

bitbake: cache/cooker: Pass databuilder into bb.cache.Cache()

Rather that the current mix of static and class methods, refactor
so that the cache has the databuilder object internally. This becomes
useful for the following patches for multi config support.

It effectively completes some of the object oriented work we've been
working towards in the bitbake core for a while.

(Bitbake rev: 7da062956bf40c1b9ac1aaee222a13f40bba9b19)

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