conf/machine/include: Cleanup IA tunings to match README

We perform a basic cleanup of the IA32 architecture and related
tunings in order to match the rules and descriptions within the
new tuning README file.

A number of small issues were corrected in the "c3" tuning to
bring it inline with the README.

(From OE-Core rev: ab77d3401908964f3249c761969600b5ec1bfbd0)

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

conf/machine/include/README: Add readme to explain cpu tunings

Add a new README that covers the basic items used with various cpu
tunings.  The goal is to better help people understand the various
settings and where things should or should not be defined.

Corresponding architecture README files will also be generated to
explain the particulars of architectural tunings.

Also remove the default TUNE_PKGARCH setting in bitbake.conf.  This
was done to ensure an error occurs if an invalid tuning is defined.

(From OE-Core rev: e138f9f7e48e0af94c5c88045c4f0581cc68248d)

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

archive-patched-source.bbclass: Archive patched source

This bbclass inherits archiver.bbclass to archive patched source

[YOCTO #1977]

(From OE-Core rev: 8f25bf3881ee568bbe03132d2205da5382fa7dd6)

Signed-off-by: Xiaofeng Yan <xiaofeng.yan@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Remove redundant reference to "task-self-hosted" from self-hosted-image.bb

This recipe already includes "task-self-hosted" in the IMAGE_INSTALL
line:

IMAGE_INSTALL = "task-core-boot task-core-apps-console task-core-ssh-openssh task-self-hosted"

so there's no apparent need to include it again further down.

(From OE-Core rev: bbc7f24d463c11b16f000462528c18bbb86b1e88)

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

qt-4.7.4: Fix build with gcc 4.7

g++ 4.7 is detecting inheritence problems
which we solve by explictly specifying
elements in constructors

(From OE-Core rev: 553ac1bf0ebf2ecf4e45ace4016b50c810ef7b26)

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

qt-mobility: Fix build with gcc 4.7

C getpid needs unistd.h for getting
its signature

(From OE-Core rev: da89bb83298eaf28faebcc8782f207927fbe190a)

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

webkit: Fix build with gcc-4.7

Include unistd.h for all linux and not
just for android.

(From OE-Core rev: c870606c9da2fa90df2cd7c4a198e3bf5340304b)

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

archiver.bbclass: enhance code readability

The modification is as follow:

- Modify some codes with more preferable readability and vague description.
- Use existed functions instead of custom functions.

(From OE-Core rev: 514319c4a15156cd63a4ac3c6ee903f64c98884e)

Signed-off-by: Xiaofeng Yan <xiaofeng.yan@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

eglibc packaging: locale packaging configuration

The PACKAGE_NO_GCONV var manipulations ware happening in the
eglibc-options.inc file, and the eglibc-locale recipe do not
see it. Moving that into the libc-package.bbclass which is
common to eglibc & eglibc-locale recipes.

This fixes bug: [YOCTO #2089]

This avoids this error for poky-tiny
NOTE: package eglibc-locale-2.13-r19: task do_populate_sysroot: Started
ERROR: Error executing a python function in
/opt/poky.git/meta/recipes-core/eglibc/eglibc-locale_2.13.bb:
OSError: [Errno 2] No such file or directory:
'/home/rchatre/concordia/dev/ccd-distro-work/tmp/work/core2-poky-linux/eglibc-locale-2.13-r19/package/usr/lib/gconv'

ERROR: The stack trace of python calls that resulted in this exception/failure
was:
ERROR:   File "package_do_split_gconvs", line 264, in <module>
ERROR:·
ERROR:   File "package_do_split_gconvs", line 45, in package_do_split_gconvs
ERROR:·
ERROR:   File "package.bbclass", line 30, in do_split_packages
ERROR:·
ERROR: The code that was being executed was:
ERROR:      0260:»------»-------bb.note("generation of binary locales disabled.
this may break i18n!")
ERROR:      0261:
ERROR:      0262:
ERROR:      0263:
ERROR:  *** 0264:package_do_split_gconvs(d)
ERROR:      0265:
ERROR: (file: 'package_do_split_gconvs', lineno: 264, function: <module>)
ERROR:      0041:»------»-------»-------d.setVar('RPROVIDES_%s' % pkg,
pkg.replace(bpn, 'glibc'))
ERROR:      0042:
ERROR:      0043:»------do_split_packages(d, gconv_libdir,
file_regex='^(.*)\.so$', output_pattern=bpn+'-gconv-%s', \
ERROR:      0044:»------»-------description='gconv module for character set
%s', hook=calc_gconv_deps, \
ERROR:  *** 0045:»------»-------extra_depends=bpn+'-gconv')
ERROR:      0046:
ERROR:      0047:»------def calc_charmap_deps(fn, pkg, file_regex,
output_pattern, group):
ERROR:      0048:»------»-------deps = []
ERROR:      0049:»------»-------f = open(fn, "r")
ERROR: (file: 'package_do_split_gconvs', lineno: 45, function:
package_do_split_gconvs)
ERROR: Function failed: package_do_split_gconvs
ERROR: Logfile of failure stored in:
/home/rchatre/concordia/dev/ccd-distro-work/tmp/work/core2-poky-linux/eglibc-locale-2.13-r19/temp/log.do_package.31042
NOTE: package eglibc-locale-2.13-r19: task do_package: Failed
ERROR: Task 552 (/opt/poky.git/meta/recipes-core/eglibc/eglibc-locale_2.13.bb,
do_package) failed with exit code '1'

Signed-Off-By: Nitin A Kamble <nitin.a.kamble@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gcc-cross-intermediate: fix do_install for x32

This Fixes [Ycoto #2223] bug.

The gcc osdir is obtained in the do_install by invocation of
command "gcc -print-multi-os-directory". For x32 it returns gcc
osdir for the default abi which is x86_64. Fix this by adding
target abi parameter to the gcc command line to get correct gcc
osdir with invocation of command "gcc -mx32 -print-multi-os-directory"

(From OE-Core rev: e65b6a4282d5c4882d0565b79ccba99db90161ff)

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

package_{deb,ipk,rpm}: apply umask to files generated in do_package_write_{deb,ipk,rpm}

* Explicitly set umask to 022. Otherwise the build system's
  umask may leak into the image.

(From OE-Core rev: d2a54427481856238bdfec9723cf575088320512)

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

copyleft_compliance: also print the reason for including a package

(From OE-Core rev: 20996da46aff03e61de50444ab3a0ab46c057dfd)

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

license.py: fix behaviour of copyleft_compliance

actually if a package has a license in its LICENSE variable
which is not in the whitelist nor in the blacklist and even
if an other license in this variable is in the whitelist,
the package gets excluded and is not taken in account in the
copyleft_compliance.
This patch solves this by excluding a recipe _only_ if the
LICENSE variable includes a pattern from the blacklist and
including a recipe only if it includes a variable from the
whitelist _and_ none from the blacklist.

Example in busybox which has LICENSE="GPLv2 & BSD-4-Clause",
with the actual behaviour (where he blacklist contains only
CLOSED Proprietary) we get :
DEBUG: copyleft: busybox-1.19.4 is excluded: recipe has excluded licenses: BSD-4-Clause
which is not sane because busybox is covered by a copyleft license
which is GPLv2 and should match the default whitelist which is
GPL* LGPL*.

(From OE-Core rev: 987d06447d2eacf2e01f08e29469c00fbb0ef1db)

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

license.bbclass: Check if copyfile succeeded with isfile

A cleaner way to check if copyfile suceeded is to use os.path.isfile.
In this way we can omit warns in some python versions where copyfile
return the same non-0 value even is this action is successfull.

(From OE-Core rev: fa51fe19a1dc1ddc4c9ec879a782953fd6a15117)

Signed-off-by: Andrei Gherzan <andrei@gherzan.ro>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

binutils: Inform binutils that armv5e really is valid!

A comment in the binutils sources indicate that it should support all of the
-march= parameters that gcc supports.  The tune validations noted that
gas failed on -march=armv5e.

It is not yet clear to me if this patch belongs upstream or not.

(From OE-Core rev: 4f2503ebdb5c491b758cbdf6eb7df96d4295c24f)

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

elfutils: disable lzma (and bzip2 for native)

Fix some library dependency issues:
* Disable lzma as xz-native is unstated in DEPENDS
* Disable bzip2 for native as it is in ASSUME_PROVIDED and thus isn't
  available when elfutils-native is normally built, but if it gets
  rebuilt the link will be made; plus we don't need it.

(From OE-Core rev: 1634d736c55f767fe82a46bbb7f83c32006fece9)

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

gdb: build with expat, add missing RRECOMMENDS_gdbserver

* Fixes communication between gdbserver and gdb-cross
  by using the same expat settings for both recipes.
* Adds missing build dependencies for expat/expat-native.
* Adds missing glibc-thread-db runtime recommendation
  to gdbserver, which was set only for gdb.

(From OE-Core rev: 38ee88e3b32b7444d7f1eb64f1b4f69a48fe0458)

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

eglibc: SDK_GLIBC_ADDONS for eglibc-nativesdk.

Use SDK_GLIBC_ADDONS for eglibc-nativesdk so that it is not
dependant on the target eglibc's GLIBC_ADDONS settings.

(From OE-Core rev: e165203a6a7eb1fd47a20c3ede4d5a20ad49487f)

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

gettext: Fix sysroot conflict over owner of macros

gettext-minimal-native andgettext-native are trampling over each other's files. This can
be reproduced with:

bitbake gettext-minimal-native
bitbake gettext-native
bitbake gettext-native -c clean
bitbake shadow-native

which will fail since the aclocal gettext macros will have disappeared. This patch
fixes the problem by giving ownership of them to gettext-minimal-native and ensuring
there is a correct dependency on this.

(From OE-Core rev: 5bcb68a232539cf11a30e3b812b2fbd6d7d76e35)

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

gst-ffmpeg: Fix build failures from sstate relocation

This is a similar situation to:
http://git.yoctoproject.org/cgit.cgi/poky/commit/?id=3c96a5386355969428163ddb60216cc989e00b3d

Builds were failing with a failure in configure:

| checking for i586-poky-linux-gcc... ccache  i586-poky-linux-gcc  -m32   -march=i586 -L/srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-world/build/build/tmp/sysroots/qemux86/usr/lib
| checking whether the C compiler works... no
| configure: error: in `/srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-world/build/build/tmp/work/i586-poky-linux/gst-ffmpeg-0.10.13-r2/gst-ffmpeg-0.10.13':
| configure: error: C compiler cannot create executables
| See `config.log' for more details

config.log shows:

configure:3976: ccache  i586-poky-linux-gcc  -m32   -march=i586 -L/srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-world/build/build/tmp/sysroots/qemux86/usr/lib -O2 -pipe -g -feliminate-unused-debug-types  -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed conftest.c  >&5
/srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-world/build/build/tmp/sysroots/x86_64-linux/usr/bin/i586-poky-linux/../../libexec/i586-poky-linux/gcc/i586-poky-linux/4.6.4/ld: cannot find crt1.o: No such file or directory
/srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-world/build/build/tmp/sysroots/x86_64-linux/usr/bin/i586-poky-linux/../../libexec/i586-poky-linux/gcc/i586-poky-linux/4.6.4/ld: cannot find crti.o: No such file or directory
/srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-world/build/build/tmp/sysroots/x86_64-linux/usr/bin/i586-poky-linux/../../libexec/i586-poky-linux/gcc/i586-poky-linux/4.6.4/ld: cannot find crtbegin.o: No such file or directory
/srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-world/build/build/tmp/sysroots/x86_64-linux/usr/bin/i586-poky-linux/../../libexec/i586-poky-linux/gcc/i586-poky-linux/4.6.4/ld: cannot find -lgcc
/srv/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-world/build/build/tmp/sysroots/x86_64-linux/usr/bin/i586-poky-linux/../../libexec/i586-poky-linux/gcc/i586-poky-linux/4.6.4/ld: cannot find -lgcc_s
collect2: ld returned 1 exit status

The reason being the --sysroot option is missing from the gcc commandline and
its looking in nightly-x86, not nightly-world in this case.

There is no reason to add extra -L options to the compiler, the sysroot already
takes care of this. We can therefore simply remove this incorrect CC line.

(From OE-Core rev: 78299c87eafbea8331fa062f61095dd8de429109)

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

gst-ffmpeg: Fix build with gcc 4.7

A cmp instruction with two constants is invalid, therefore 'g' constraint
is not correct but must be "rm" instead.

(From OE-Core rev: 1161c40a26146629613bd6a17b263175a492b751)

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

apt: Fix a latent bug exposed by gcc 4.7

Further Details are in patch itself

(From OE-Core rev: 0dc952e35da00a753317a4f878b23eab8bb7bc4a)

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

tiny: Update linux-yocto-tiny to 3.2

Migrate linux-yocto-tiny from 3.0 to 3.2. The 3.0 recipe was
based entirely on recipe-space fragments and was only a proof of concept.
The 3.2 linux-yocto meta-data now has a proper tiny KTYPE defined.

By default this recipe supports only the qemux86 machine, which builds the
common-pc support (including networking, sound, USB, VGA and serial consoles,
etc.). New machines can be added and will use the tiny KTYPE, but will need to
add any desired hardware support as the base config is very minimal.

No filesystems are supported by default, only the initramfs.

(From OE-Core rev: 44c556f5300b1d811a1ae71f501320d1a6b0fb2e)

Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

linux-yocto: update tiny meta and configuration for kernel 3.2

Updating the META SRCREV to pickup these commits:

  59f350e meta: Add common-pc-tiny.scc
  0996ca9 tiny: Minimize the tiny config
  d6b57bb meta: common-pc add dependencies to cfg

Which update the configuration for the tiny profile of the kernel
for the 3.2 release.

cc: Darren Hart <dvhart@linux.intel.com>
(From OE-Core rev: df3b8f716cc9203251cd4d4b5e673d7ecab5111b)

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

self-hosted-image: fix indentation: 8 spaces --> Tab

(From OE-Core rev: 80e1cf4deb2a875449876f947a7a8ba29d6def5b)

Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

builder: uncomment the settings for BB_NUMBER_THREADS and PARALLEL_MAKE

We use sed to enable the 2 settingis in conf/local.conf.

(From OE-Core rev: eb985d333eb0f13e142da43b4b2b9ec5ad40ff30)

Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

self-hosted-image: renew the poky src's SRCREV that's installed into the target

(From OE-Core rev: 826918b634deaa2b7201f3068d6f1410c5641d63)

Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

self-hosted-image: improve do_populate_poky_src

1) remove the assumption ${DL_DIR} ends with downloads/.
Thanks Paul Eggleton for pointing this out.

2) remove downloads/git2_* tarballs to speed up the rootfs creation.
This is ok since we still have the git2/.
Thanks Richard Purdie for suggesting this.

(From OE-Core rev: 58ba59f034941bf167e70ae6c08117e5184a20ec)

Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

genext2fs: support large files and filesystems without using large amounts of memory

update_to_1.95.patch was generated by making a diff bewteen the 1.4.1 release
and the latest 1.9.5 version in the cvs repo:
http://genext2fs.cvs.sourceforge.net/viewvc/genext2fs/genext2fs/genext2fs.c?revision=1.95

The patches 0001-0019 come from mailing list of genext2fs-devel
http://sourceforge.net/mailarchive/forum.php?forum_name=genext2fs-devel&max_rows=100&style=flat&viewmonth=201106

(From OE-Core rev: 8f17e499cf91191727c8767e839738cb39c21655)

Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

qemugl: Fix GL apps failure on Ubuntu 11 host with nVidia GLX driver

Previous version of nVidia GLX driver in Ubuntu 10 cause qemu segfault, so we
fall back to Mesa GLX driver if detecting nVidia driver installed. From Ubuntu
11, nVidia GLX driver works well, while previous work around cause GL apps
failure. So this work around is limited in Ubuntu 10 only, and will be removed
in future.

[YOCTO #1886] got fixed.

(From OE-Core rev: b3ccc630e6c12a75111b1f7ca877e17d8d4e1dc7)

Signed-off-by: Zhai Edwin <edwin.zhai@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

buildhistory: make seperate commits for every changed top level entry in the buildhistory dir

This seperates out image changes from package changes making the image diffs a lot easier to read.

(From OE-Core rev: fba198ac7efe476a25c5761878ef2fcee97bf9f1)

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

buildhistory: record all builds

When nothing has changed an empty commit prefixed with "No changes" will get generated so that the commit log of the buildhistory repo provides a complete log of all builds performed, not just those that resulted in changes.

(From OE-Core rev: 2e40558cc33c5c566a9a742c32eda3ea017f8607)

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

buildhistory: remove duplicate entries from dot graph

There are various conditions that lead to duplicate entries in the dot graph which need to get fixed, but this patch is a catchall. A previous attempt to address this only works on rpm which gives a \n seperated output, opkg doesn't.

Another benefit is that the sort order is now know, leading to less spurious diffs in buildhistory commits.

(From OE-Core rev: 479ae1e9b74aa2f04fb5da2f3541c3de0aa9de87)

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

README.hardware: extend USB-ZIP instructions

* Note that on some machines booting just stops after the SYSLINUX
  version banner is displayed
* Add an instruction to show how to get the geometry information for
  the disk
* Specify the number of cylinders when running mkdiskimage as it is
  sometimes unable to detect it automatically
* Create the temporary mountpoints before attempting to use them
* Add an instruction to unmount the disk before removing it

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

README.hardware for mpc8315erdb incorrectly specified a null modem serial cable

I noticed this while beta testing 1.2. but patch is against master
(latest commit: bcd4d14425cadc5bd6296f59f14733e8c39c49dd).

Instructions specified a null modem.  This isn't the case.  A straight
serial cable is required.  I verified it with my own setup and
Freescale's user manual.

Signed-off-by: Robert Cochran<yocto@mindchasers.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

cross-canadian.bbclass: fix rpath for sdk executables

This makes the libraries located in places like this findable:
/usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/usr/lib

Which avoids linking cross canadian sdk executables with host libraries like this:

$ ldd /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/usr/bin/x86_64-oe-linux/x86_64-oe-linux-gdb
        linux-vdso.so.1 =>  (0x00007fffb7fff000)
        libreadline.so.6 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/usr/lib/x86_64-oe-linux/../libreadline.so.6 (0x00007fbfb5511000)
        libdl.so.2 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/lib/libdl.so.2 (0x00007fbfb530c000)
        libncurses.so.5 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/lib/libncurses.so.5 (0x00007fbfb50e9000)
        libtinfo.so.5 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/lib/libtinfo.so.5 (0x00007fbfb4ec2000)
        libz.so.1 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/usr/lib/x86_64-oe-linux/../libz.so.1 (0x00007fbfb4cac000)
        libm.so.6 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/lib/libm.so.6 (0x00007fbfb4a2a000)
        libpthread.so.0 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/lib/libpthread.so.0 (0x00007fbfb480d000)
        libutil.so.1 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/lib/libutil.so.1 (0x00007fbfb4609000)
        libexpat.so.1 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/usr/lib/x86_64-oe-linux/../libexpat.so.1 (0x00007fbfb43e0000)
        libc.so.6 => /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/lib/libc.so.6 (0x00007fbfb4059000)
        /usr/local/oecore-x86_64/sysroots/x86_64-oesdk-linux/lib/ld-linux-x86-64.so.2 => /lib64/ld-linux-x86-64.so.2 (0x0000003f05000000)

[RP: Whitespace tweaks]
(From OE-Core rev: c97f7f4e4ecd6c431712059c34ebc17b68b055ae)

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

tune-cortexa8/9: fix PACKAGE tunes being all armv7at even for non-Thumb ones

All PACKAGE_EXTRA_ARCHS for cortexa8, cortexa8t and cortexa8-neon have typo in
referencing tune-armv7at even for non-Thumb modes. Probably a copy/paste error.
That's not the case for recently-added hard-fp tunes.

Same for cortexa9.

(From OE-Core rev: 4e91c00bb3a171bebdb716451b901f5f099a04bc)

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

scripts/bitbake: add/fix some comments

Add some comments explaining what this script does, fix one grammatical
error in a comment and make the tar-replacement-native comment give the
full reason why it is needed.

(From OE-Core rev: aa946e1d054d3a0b7097339e0fb74ee60bd94a78)

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

scripts/bitbake: allow switching between build directories

The recent addition of the check to ensure the user was in their build
directory disabled the ability to switch between build directories
without re-running the build environment setup script. We can rely
upon checking for conf/bblayers.conf instead, so use this check.

This does allow BUILDDIR (which is normally set by the environment
script) to be unset; however if it is set then it is assumed to be the
correct build directory and will be used in the error message that is
shown when we can't find conf/bblayers.conf.

(From OE-Core rev: 769384decb095fb3c49eb13b8f7f69c978d0bcba)

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

scripts/bitbake: try harder to check if pseudo exists

If pseudodone doesn't exist, we can get STAGING_BINDIR_NATIVE by calling
bitbake -e and use that as the path to check for pseudo before we give
up and try to build it explicitly first.

This is useful for people who share TMPDIR between multiple build
directories.

(From OE-Core rev: 07b0dddab901510208fab44bbc2566d3c3baae93)

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

nasm: fix issue with repeat configure

This Fixes bug: [YOCTO #2176]

This fixes the following issue as mentioned in the bug:
If nasm-native has to be rebuilt (due to a signature change) it will fail:

| checking for a BSD-compatible install... /usr/bin/install -c
| /media/large/tmp/work/x86_64-linux/nasm-native-2.07-r1/nasm-2.07/configure:
line 4261: syntax error near unexpected token `-W'
| /media/large/tmp/work/x86_64-linux/nasm-native-2.07-r1/nasm-2.07/configure:
line 4261: `PA_ADD_CFLAGS(-W)'
| ERROR: oe_runconf failed
NOTE: package nasm-native-2.07-r1: task do_configure: Failed

The failure appears to be caused by do_patch_fixaclocal being run a second time

(From OE-Core rev: 2d26b1a4f83a635f96072589289a389675abb305)

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

pkgconfig.bbclass: corrected dependency for -nativesdk.

For -nativesdk, the dependency should be pkgconfig-nativesdk instead of
pkgconfig.

(From OE-Core rev: 36c0d1fdc71f9c63699049911aeddb07a746ca4a)

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

autotools.bbclass: Drop -nativesdk dependency to lib-cross.

-nativesdk recipes should not have dependency to lib-cross, which is
never used. This unnecessary dependency would result different task hash
values in sstate for different MACHINE settings.

(From OE-Core rev: 313deb802c5411c6c88655057f05a7d8823d999c)

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

crosssdk.bbclass: Set TUNE_PKGARCH to SDK_PKGARCH.

The TUNE_PKGARCH of crosssdk should be set to SDK_ARCH, not the one
from target machine's configuration.

Fixed bug [YOCTO #2206]

(From OE-Core rev: e809b6657c53616a82e73d2f20ec23bc50ccebc9)

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

nativesdk.bbclass: Set PACKAGE_ARCHS to SDK_PACKAGE_ARCHS.

This patch fixed the issue of reading incorrect pkgdata files. Previous
appending '-nativesdk' suffix to PACKAGE_ARCHS would result the
i686-nativesdk recipes reading in x86_64-nativesdk pkgdata files if the
MACHINE is set to qemux86-64.

Fixed bug [YOCTO #2203].

(From OE-Core rev: 951a68731f655f597c5dfa541fc913d399eabfb9)

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bitbake/cooker.py: Bring into sync with upstream bitbake (trivial difference)

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

docbook-utils-native: fix syntax problem in jw.in

Fix runtime error occurred e.g. with docbook-to-man calls:

  grep: character class syntax is [[:space:]], not [:space:]
  grep: character class syntax is [[:space:]], not [:space:]
  jw: There is no frontend called "/docbook/utils-0.6.14/frontends/docbook".

See also:

   <https://qa.mandriva.com/show_bug.cgi?id=61127>

(From OE-Core rev: 627998726ca3ee2ed2510c0f666747f688d06c56)

Signed-off-by: Steffen Sledz <sledz@dresearch-fe.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

virtual/libgl: use mesa-xlib for qemuarm/qemumips/qemuppc

Still need mesa-xlib for emulation of GLX interface on qemuarm/mips/ppc, where
mesa-dri doesn't work for pure qemu emulator.

[YOCTO #2066] fixed.

(From OE-Core rev: 22775b5f1d9c6d9860a579245bf7a48a982ab62f)

Signed-off-by: Zhai Edwin <edwin.zhai@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

adt-installer: Adapted environment file name for powerpc.

Adpated to the new environment file name for powerpc, due to the oe-core
commit c2d96179.

Fixed bug [YOCTO #2055].

(From OE-Core rev: bebcac5e6eaceb7bdafa6432e42fe8073a42c6d2)

Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

arch-armv7a.inc: fix PACKAGE_EXTRA_ARCHS after armv7.inc was removed

(From OE-Core rev: d1ffae623ea9a6be3d2cb9067f64f33cc1fd1e8a)

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

gdk-pixbuf: Fix checksum warning issue

(From OE-Core rev: 6a7ea3cf81f3600fec01bdd4af96ed149398a1a1)

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

sstate-cache-management.sh: fixes and enhancement

Fix a few problems and enhance its functions, it shoud be more useful
than before.

* Search in meta and meta-* for archs, and grep AVAILTUNES for archs,
  (only search meta, and grep DEFAULTTUNE before), add the host arch.
  and also can search in extra layers with --extra-layer.

* Reduce the analyzing time when remove duplicated files. It would cost
  more than 10 minutes to analyze 11,000 files before, now only needs
  about 50 seconds.

* Check the access time rather than create time.

* Need the user's confirm before really remove the file, or use --yes to
  assume yes.

* Add --stamps-dir to keep sstate files which are used by the build
  directory, and remove others this can make the sstate cache dir clean,
  it is faster and should be useful than the --remove-duplicated.

* Add --verbose to explain what is being done.

* Add "-d" which is short for --remove-duplicated

[YOCTO #2198]

(From OE-Core rev: 769a000428e4b2462a4e6d8f179b5816b8ec2417)

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

rootfs_ipk: don't echo opkg-cl output to log.do_rootfs

* log_check can find some files with ERR or Fail substring e.g. in "Source: " field

(From OE-Core rev: 76c83f107d4a4688a879e30821ab70812052bbdc)

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

Hob: In building log page, fixed the issue about 'endpath' not clear when next to start build

Fixed the issue about the building log scrollbar can not auto scroll to page end sometimes

[YOCTO #2098]

(Bitbake rev: 035e146ff92236a3eda71ad71e8389737f91753b)

Signed-off-by: Liming An <limingx.l.an@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

runqueue: wait and deal with those <defunct> sub-processes as soon as they are os.killed() when "Force stop"

When "Force stop" is performed during the build stage, after os.kill() kills the build sub-processes, there are many <defunct> python processes in the system. In Hob, when the user initiates a new build, os.waitpid() in runqueue_process_waitpid() will be called, and the pids of those <defunct> processes will be returned as result[0], then self.build_pids[result[0]] will throw KeyError exception because now for the new build self.build_pids is empty.

This patch is to address the above issue to collect the results and handle the sub-processes as soon as they are killed.

[Yocto #2186]

(Bitbake rev: e9f4ca467e795bbc520d12b0e7a5985b6ff0a20e)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib/bb/ui/crumbs/hoblistmodel: fix sorting of RecipeListModel

Sort in ascending order on the recipe name (A-Z) and unset the default sort
function so that there's no third (unsorted) state.

(Bitbake rev: c6ab6c7caf16c04e2a0c0f2aefd3377f781206c7)

Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib/bb/ui/crumbs/hobpages: change title from HOB to Hob

(Bitbake rev: 2ecbd68abbf564be15707f330f80e811d92ef87d)

Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: tweak font sizes and layout

Switch to proportional font sizes and tweak padding and layout based on
feedback from design team.

(Bitbake rev: 356f190bb3f0ee9a15df900714edcb85593d6989)

Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

libc-packgae.bbclass: Add i686 support in locale_arch_options

* While building for i686 architecture an error was coming that
locale_arch_options does not have support for i686. Add missing support.
* Verified on intel architecture.

(From OE-Core rev: c8afc79b5d3205355ad61d2589221bf8babe8395)

Signed-off-by: Noor Ahsan <noor_ahsan@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

bc: use update-alternatives to make dc play nice with busybox

busybox' default configuration enables dc app, which bc also provides,
setup update-alternatives to resolve the conflict.

(From OE-Core rev: f8456e7a0fd8559497db8292c87fee4fd95eb9c6)

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

gstreamer: Provide easy way to enable runtime debugging

The gstreamer framework has a very useful debugging setup which is
essential for debugging pipelines and plugins.  This patch makes
it simple to enable this (disabled by default).  To enable debugging,
just add this line to local.conf
  GSTREAMER_DEBUG = "--enable-debug"

(From OE-Core rev: 947c00c78732da48a111228e1325ad42cf57a370)

Signed-off-by: Gary Thomas <gary@mlbassoc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

runqemu: set console=ttyS0 when running with nographic option

When passing the nograhic option to the runqemu script, set
console=ttyS0 in the kernel options so the user can view
the kernel boot messages.

This fixes [YOCTO #1475]

(From OE-Core rev: 28f05bf6e5da9cd8f01cff50c317233e3064e3cb)

Signed-off-by: Scott Garman <scott.a.garman@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

qemu.inc: Use '+=' for IMAGE_FSTYPES

As per
http://lists.linuxtogo.org/pipermail/openembedded-core/2012-March/020053.html
a machine conf file should use '+=' to set IMAGE_FSTYPES.

(From OE-Core rev: b04f6504fe049e3e9dd3998377d1fc2d1ef9a13b)

Signed-off-by: Tom Rini <trini@ti.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

knotty: Add back getTasks() call to ensure we don't repeat messges

(Bitbake rev: 8fed4fdf4bbbc9ef036ff96755c0bfe15c3a9dd0)

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

Hob: in build details page, change error icon to denied icon

For make icongraphy consistently to change the original gtk-icon to hob-icon

[YOCTO #2108]

(Bitbake rev: 8c3401bc11c6f3708b8ee9c0175f389399fdff5a)

Signed-off-by: Liming An <limingx.l.an@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: fixed the 'error' icon source picture and adjusted corresponding showed icon size

The error source picture is not right, so fixed it and adjusted its size.

[YOCTO #2097]

(Bitbake rev: ce680f10037b42831179a4598a69fb39688fb238)

Signed-off-by: Liming An <limingx.l.an@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: some color style changes to make HobTabBar more close to visual design

The change includes:
 - remove the colors which are not used
 - change the background of indicators to deep red
 - change the color style of the texts on HobTabBar, i.e., the tab text and the indicator background are gray when the tab is the current tab; the tab text is white and the indicator background is deep red when the tab is not the current tab.

(Bitbake rev: 95acecbf50cee906dca4abd5ce758701e5761668)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: make ORANGE to be the same as the color on the dialog

(Bitbake rev: a50d884ebe1af2b9b2a5a54289181bccb751ddab)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Change "Just bake" to "Build image" as required

[Yocto #2160]

(Bitbake rev: 4ed5b8b862027755d5e9e480fd8a1d0250f7a3b7)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Change icon to error because parsing error is an error

(Bitbake rev: 66fcc8cdf8fb3365cd4a056b94fe3aa3e1aedc86)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: clear the building state after a successful build, as a failed build does

(Bitbake rev: afd51c524006e8e6db1542cc2ebc5decab3faa1b)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Replace "local.conf" in error message

(Bitbake rev: c8dac1eefab79b12243468986c37f391c3421134)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Fix the setting hash calculation

Sometimes even setting are not changed, the hash values differs due to
variable order issue. This commit fixes the issue.

(Bitbake rev: 1fe0996f89952af72cbdc46ca7c6495868d27a56)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Change the format to store image_fstype variable

Use string format to store image_fstype instead of a list.

(Bitbake rev: c91fe7f9d21939fd437dbd79a923499f90fc95c6)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Remember user settings when reparse config

When config reparse is issued (e.x, adding a layer), we will firstly
remember the past user settings by setting them to bitbake server,
and then do the reparse.

(Bitbake rev: 8a1f6953a082d8002585bfd9e8a67d2a7f69d6aa)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Update configuration after parsing

We need to update the parameters stored in Hob side after
configuration is parsed.

(Bitbake rev: 9560e4ccef497e878f2cb8624136ab8f193c061a)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Inherit image_types.bbclass in set_user_config()

(Bitbake rev: f34cb901bd9b3598f4da528cc54f6faa57c9d389)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Improve for init a new build

After a successful build, user can initiate a new build by clicking
"Build a new image" button. The previous solution is simply switch to
the first screen and unset MACHINE. This commit will try to re-init
everything for the new build.

(Bitbake rev: df63e15c41d129fe0cff6b8ab3c97434c4809c83)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

command.py: Move triggerEvent to command async class

The parameters in triggerEvent function is an event object, and it may
cost some time to pass this object through pipe, causing the pipe's
poll() function timeout. Change it to async mode.

(Bitbake rev: 3b5909ebc86a12dedfb30e5446aa81eb58921760)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Improve the matching for runnable machine type

(Bitbake rev: 1b14488bcfb345a3258b15ebfdaa2e1235a5fe87)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Define a variable to store the image name and toolchain

(Bitbake rev: d0b64d901b33c9e5247dbb2b05c89a1d65abbf8b)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Fix adding duplicated meta-hob layer

While adding meta-hob layer, we need to detect whether it already
exists in current layers.

(Bitbake rev: b195823d19bcb71bd5eef19acfe22b2dd670d90a)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Remove un-used signal and functions

(Bitbake rev: 9c632467eddc267ed54b78a40d5b1413287e3032)

Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib/bb/ui/crumbs/hig: fix layers_changed test

Because we sort the treeview to list specific layers at the top, and
therefore implicitly change the sorting of the underlying model, we can't
be certain that the original layer list will equal the new layer list
despite the included layers being the same.

To ensure we can do a simple equality test to determine whether the layers
have been modified first sort the lists to ensure we're comparing based on
contents alone.

(Bitbake rev: ae86cd8b4ef1e43b79230326ccba69e2900d074f)

Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib/bb/ui/crumbs: hob progress bar should not be red when user stops build

If the user explicitly stops the build telling them the build failed is a
misnomer.

(Bitbake rev: 722f4f0e31f9debf5ad20a91da759a8c25151567)

Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib/bb/ui/crumbs/builder: allow user to cancel Layer selection dialogue

Because layer changes may take some time allow the user to cancel out of
the dialogue and prevent Hob from applying any of the changes made.

(Bitbake rev: a2d1c035a5dd3d112a887e4386dfaf9b0c37c104)

Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib/bb/ui/crumbs/hobwidget: change text sizes in HobImageButton

(Bitbake rev: 4301cc1e859a68d93eba3122be099bfaa90fbdfb)

Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib/bb/ui/crumbs/imageconfigurationpage: add extra space between widgets

Add space between the target machine combo and the 'Layers' button.

(Bitbake rev: bdd50d92e90f96bf7ff433bb539cdde4b0d21cdb)

Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

lib/bb/ui/crumbs/hobwidget: add padding to HobImageButton

Add more padding between the icon and the text

(Bitbake rev: a81f6a33cf1fa9fe869ac617f2d14d66e07009ff)

Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: change some words in settings dialog to make them consistent

(Bitbake rev: 1def8d4d8e725bd9fc4330179557604dda27b917)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: change some words on recipes screen and package screen to make them consistent

(Bitbake rev: 63bfa397059823bea6fa453703cf0653fc123f0b)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: change some words on image configuration screen to make them consistent

(Bitbake rev: 6d5bcf80afbd081d41572985b724c615bd3a16b6)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: change some words for build configurations on build details screen

This patch is to change some words of variables to make them consistent with the GUI.

(Bitbake rev: 4af7ab24ff3b170dfa74159c057e7110a615a8d3)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: add Templates and Settings on image details screen

This patch is to add Templates and Settings tool buttons on the image
details screen, which makes things easier and simplier.

In order to fulfill that, the code splits the functions
show_load_template_dialog() and show_adv_settings_dialog() in builder.py
because they will possibly be called from different screens later.

[Yocto #2163]

(Bitbake rev: 29bea7b7076a7b74d36237da86a4eff6605d17ec)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Hob: Set one of deployable images or runnable images as the default toggled item

With this patch, even though there are a lot of images built out, a default image
which is either deployable or runnable is toggled by default. So, for users, one
more action to select an image before running qemu or deploying is not needed any more.

Note: If there are more than one runnable or deployable images (such as ext2, ext3,
jffs2 and btrfs), only the first image is toggled by default for run-qemu or
deployment. If the user wants to run or deploy others, he/she needs to toggle them
manually.

[Yocto #2155]

(Bitbake rev: 4568dfbd5e693cce0e6e947f323eaf08a3176744)

Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

ui/knotty: Add a footer to the build output for interactive terminals as knotty2 UI

On terminals which support it, add summary information to the end of the
build output about the number of tasks currently running and how many tasks
we've run so far.

This provides a summary at a glace of what the current state of the build is
and what the build is currently doing which is lacking in the current UI.

Also disable echo of characters on stdin since this corrupts the disable,
particularly Crtl+C.

The "waiting for X tasks" code can be merged into this code too since
that is only useful on interactive terminals and this improves the
readability of that output too.

Improvements since v0:

* The tasks are ordered in execution order.
* The display is only updated when the list of tasks changes or there
  is output above the footer.
* Running task x oy y and package messages are supressed from the console

This UI can be accessed with "bitbake -u knotty2".

(From Poky rev: e38b4569648f2916c4370871c79e6a6090eb8bc1)

(Bitbake rev: 156189c799d2bb1f69bdaa04b5cd718fe7881425)

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

uihelper: Keep track of pids in execution order

(Bitbake rev: cec2f3eaa45a6fd8d3c53c09aaf010ac6101e7ba)

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

license.bbclass: remove existing license.manifest before appending new data

without this fix, we append license each time we build again the same image,
ending with a large not up to date file.

(From OE-Core rev: 2d49a8f659694b60cdb706e8993cd9550e2002bd)

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

ghostscript: Fix remaining CP_ prallel make races

(From OE-Core rev: 8fe53bdc807184bc41469d8587368b31192e6252)

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

bb.build: use relative log links, not absolute

Using symlinks to absolute paths isn't gaining us anything, and can cause
problems in situations such as automated build systems when they try to
resolve the link, depending on the mechanism used to pull the build artifacts
(e.g. nfs).

(Bitbake rev: f9cf2671c0ecad153db11bca1aebe151213bfb42)

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