Jackie Huang [Fri, 14 Oct 2016 00:50:11 +0000 (08:50 +0800)]
ppp: fix building with linux-4.8
Fix a build error when using the linux-4.8 headers that results in:
In file included from pppoe.h:87:0,
from plugin.c:29:
../usr/include/netinet/in.h:211:8: note: originally defined here
struct in6_addr
^~~~~~~~
In file included from ../usr/include/linux/if_pppol2tp.h:20:0,
from ../usr/include/linux/if_pppox.h:26,
from plugin.c:52:
../usr/include/linux/in6.h:49:8: error: redefinition of 'struct sockaddr_in6'
struct sockaddr_in6 {
^~~~~~~~~~~~
Signed-off-by: Jackie Huang <jackie.huang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Fri, 14 Oct 2016 12:57:51 +0000 (13:57 +0100)]
populate_sdk_ext: explicitly set DL_DIR
The eSDK generation assumes that DL_DIR is downloads/ under the build directory,
and puts files such as a freshly buily uninative tarball in there expecting
bitbake will find it later.
Whilst ${TOPDIR}/downloads/ is in fact the default value for DL_DIR in
bitbake.conf, and any instances of DL_DIR are removed from the original
local.conf, there is still the possibility that other layers could contain a
site.conf that assigns DL_DIR.
If this happens the errors are quite mysterious as it fails to find the
uninative tarball and so the hashes all change, and eSDK building fails.
Ensure that this cannot happen by explicitly assigning the DL_DIR that we
require, instead of assuming that the default value will be used.
[ YOCTO #10439 ]
Signed-off-by: Ross Burton <ross.burton@intel.com>
Jianxun Zhang [Thu, 13 Oct 2016 20:16:34 +0000 (13:16 -0700)]
sqlite3: Revert ad601c7962 from 3.14.1 amalgamation package
It turns out this change between 3.12.2 and 3.13 introduces
a 2% increase of build time based on statistic data in
bz10367.
The added patch is forged by diffing the new sqlite3.c
generated from reverting the change in raw source of sqlite3
project, and then manually migrate the delta to a sqlite3.c
from the 3.14.1 tarball package because what recipes reference
is actually a generated C code (amalgamation) release package
and we cannot apply the real change to 3.14.1 cleanly due to
so many changes happened.
Fixes [YOCTO #10367]
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Andre McCurdy [Thu, 13 Oct 2016 18:49:33 +0000 (11:49 -0700)]
linux-yocto.inc: ensure do_kernel_link_images runs before do_strip
If the do_kernel_link_images task is enabled, then it needs to run
before do_strip. The addtask statement for do_strip makes that
explicit. For consistency, make it explicit in the addtask statement
for do_kernel_link_images too.
Signed-off-by: Andre McCurdy <armccurdy@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Martin Jansa [Wed, 12 Oct 2016 12:06:12 +0000 (14:06 +0200)]
live-vm-common.bbclass: Allow to use different kernel image than KERNEL_IMAGETYPE for /vmlinuz
* syslinux config hardcodes kernel image as /vmlinuz add warning message
when the selected image doesn't exist and allow to select different image
with VM_DEFAULT_KERNEL variable (qemuboot.bbclass is using QB_DEFAULT_KERNEL)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
To produce certain types of images wic uses do_bootimg results
to assemble final image. For example, it copies BOOT/EFI directory
produced by do_bootimg to boot partition for every EFI image.
The tricky part of this is that do_bootimg task is not always run,
so we can't always make do_image_wic depend on do_bootimg. We only
need to do it if do_bootimg present in task graph.
Thank to Cristopher Larson for this fix.
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Robert Yang [Thu, 22 Sep 2016 09:36:55 +0000 (02:36 -0700)]
xcb-proto: create .pyc files for python2
Its Makefile's do_install creates .pyc files for python3, now also
create them for python2 so that they will be recorded by manifest, and
can be cleaned correctly.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Robert Yang [Wed, 12 Oct 2016 06:31:29 +0000 (23:31 -0700)]
source-highlight: use with-boost-libdir
Without with-boost-libdir, it checks lib64 libx32 lib lib64 in
sysroot/usr/ and set it as boost lib path if it exist, this is
unreliable when build with multilib.
Fixed when switch baselib from lib64 to lib, for example, when we build
qemux86-64 + multilib setting, the baselib is lib64, and when disable
multilib, it would be lib, then we may see the error when
do_configure since lib64 may exist during build source-highlight:
configure: error: Could not find a version of the Boost::Regex library!
Juro Bystricky [Tue, 11 Oct 2016 16:56:24 +0000 (09:56 -0700)]
insane.bbclass: Additional "mips" and "mipsel" machine definitions
Add "mips" and "mipsel" to "machdata" table.
Although there is a way to add entries to the "machdata" table
from a BSP without modifying the insane.bbclass directly, MIPS is
already supported in poky and as such the relevant entries should be
present in insane.bbclass.
Signed-off-by: Juro Bystricky <juro.bystricky@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Tue, 11 Oct 2016 12:30:42 +0000 (13:30 +0100)]
cmake: improve CMAKE_SYSTEM_PROCESSOR assignment in nativesdk
The previous string manipulations would result in the wrong string being used
for machines such as intel-corei7-64 as the sysroot was split at the first
hyphen (so would result in corei7 instead of corei7-64).
Change the logic so that it looks for processor-distro-os and uses the whole of
the processor field.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Mon, 10 Oct 2016 11:41:34 +0000 (12:41 +0100)]
cmake: also set CMAKE_FIND_ROOT_PATH_MODE_PACKAGE
The other CMAKE_FIND_ROOT_PATH_MODE_* variables were already set to ONLY, but
PACKAGE was left out. Fix this so that cmake doesn't look on the host for cmake
modules when it should only be looking in the target sysroot.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Previous attempts to constrain execution of `resize` to only TTYs did
not properly handle situations when `tty` would return the string "not a
tty". The symptom is "/etc/profile: line 34: test: too many arguments".
Fix this by utilizing the exit code of `tty`. Also use `case` instead of
`cut` to eliminate a subshell.
Signed-off-by: Richard Tollerton <rich.tollerton@ni.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Thu, 13 Oct 2016 19:36:35 +0000 (20:36 +0100)]
pixbufcache: handle gdk-pixbuf not being present
It's possible - albeit unlikely - that gdk-pixbuf isn't present in the sysroot
when a recipe inheriting this class is and the sysroot is finalised.
One example would be if the sstate archive has librsvg but not gdk-pixbuf:
librsvg will be extracted from the sstate but gdk-pixbuf will be built to "fill
in the gap". In this situation the setscene completion hook installed by
pixbufcache.bbclass will attempt to execute gdk-pixbuf-query-loaders, but that
binary hasn't been installed by gdk-pixbuf yet.
Also add gdk-pixbuf-native to DEPENDS in native builds to ensure that the
binaries we expect will be present, as it's possible to build loaders without
linking to GdkPixbuf.
[ YOCTO #10420 ]
Signed-off-by: Ross Burton <ross.burton@intel.com>
Joshua Lock [Wed, 12 Oct 2016 20:46:41 +0000 (21:46 +0100)]
Remove RM_OLD_IMAGE, it's no longer useful
Since the move to put image deployment under sstate control in d54339d4b1a7e884de636f6325ca60409ebd95ff old images are automatically
removed before a new image is deployed (the default behaviour of the
sstate logic).
RM_OLD_IMAGE is therefore no longer required to provide this
behaviour, remove the variable and its users.
Signed-off-by: Joshua Lock <joshua.g.lock@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Paul Eggleton [Tue, 11 Oct 2016 21:33:47 +0000 (10:33 +1300)]
classes/externalsrc: re-run do_configure when configure files change
If the user modifies files such as CMakeLists.txt in the case of cmake,
we want do_configure to re-run so that those changes can take effect. In
order to accomplish that, have a variable CONFIGURE_FILES which
specifies a list of files that will be put into do_configure's checksum
(either full paths, or just filenames which will be searched for in the
entire source tree). CONFIGURE_FILES then just needs to be set
appropriately depending on what do_configure is doing; for now I've set
this for autotools and cmake which are the most common cases.
Fixes [YOCTO #7617].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Todor Minchev [Tue, 4 Oct 2016 15:47:11 +0000 (08:47 -0700)]
sdk-installer: Fix unclear SDK installer message
When the host and the SDK architectures are incompatible the SDK
installer outputs an incomplete error message "Error: Installation
machine not supported!". This commit adds a more verbose error
message e.g "Error: Incompatible SDK installer! Your host is i686
and this SDK was built for x86_64 hosts."
Signed-off-by: Todor Minchev <todor.minchev@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Markus Lehtonen [Tue, 11 Oct 2016 12:43:29 +0000 (15:43 +0300)]
sysklogd and busybox: ignore return code from init script stop
The init script will return '1' if we try to stop the service and it is
not currently running. The prerm scriptlet must not fail because of this
because it will cause package deinstallation of upgrade fail if opkg
package manager is used.
[YOCTO #10299]
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We need to ignore the return code from the init script 'stop' command in
the preinst and prerm scriptlets. Otherwise package upgrade or
deinstallation (at least when opkg is used) is likely to fail if the
daemon is not running. That is because an init script possibly returns
'1' if you try to stop a service that is not running which, in turn,
causes the scriptlet to fail which, in turn, causes the package
(de-)installation to fail.
[YOCTO #10299]
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Tue, 11 Oct 2016 12:19:44 +0000 (13:19 +0100)]
insane: display names instead of ELF machine numbers
The 'arch' QA test currently simply outputs the ELF machine field as a number
which isn't helpful. Display this as a human-readable name to make it clearer
to the user what the problem is.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jussi Kukkonen [Tue, 11 Oct 2016 10:58:02 +0000 (13:58 +0300)]
vte: Build without vala by default
vala.bbclass DEPENDS on vala and vala-native: Drop the inherit so
that these dependencies can be added on-demand based on vte
PACKAGECONFIG. Add relevant items from vala class into the recipe.
Add copy of vapigen.m4 so building without vala actually succeeds.
Make building without vala the default PACKAGECONFIG.
Fixes [YOCTO #10386].
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Saul Wold [Mon, 10 Oct 2016 18:32:05 +0000 (11:32 -0700)]
archiver: fix gcc-source handling
The source archiver was not handling the gcc-source target correctly, since it uses the
work-shared directory, we don't want to unpack and patch it twice, just as the comments
say, but the code was not there to check for the gcc-source target.
[YOCTO #10265]
Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jackie Huang [Sun, 9 Oct 2016 00:56:07 +0000 (08:56 +0800)]
net-tools: fix building with linux-4.8
Fix a build error when using the linux-4.8 headers that results in:
In file included from
.../sysroots/qemuarm64/usr/include/linux/if_tunnel.h:6:0,
from iptunnel.c:39:
.../qemuarm64/usr/include/linux/ip.h:85:8: error: redefinition of
'struct iphdr'
struct iphdr {
^~~~~
In file included from iptunnel.c:29:0:
.../qemuarm64/usr/include/netinet/ip.h:44:8: note: originally defined here
struct iphdr
^~~~~ Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com> Signed-off-by: Jackie Huang <jackie.huang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ed Bartosh [Mon, 10 Oct 2016 07:50:41 +0000 (10:50 +0300)]
canned-wks: use GPT partition table
According to UEFI specification all EFI platforms must support
GUID Partition Table(GPT) disk layout. Here is a list of advantages
of using GPT disk layout over the legacy MBR partitioning:
- Logical Block Addresses (LBAs) are 64 bits (rather than 32 bits).
- Supports many partitions (rather than just four primary partitions).
- Provides both a primary and backup partition table for redundancy.
- Uses version number and size fields for future expansion.
- Uses CRC32 fields for improved data integrity.
- Defines a GUID for uniquely identifying each partition.
- Uses a GUID and attributes to define partition content type.
- Each partition contains a 36 character human readable name.
Used GPT partitioning in all EFI kickstart files.
Tested result images on NUC, MinnowBoard MAX and MinnowBoard Turbot.
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ed Bartosh [Mon, 10 Oct 2016 07:40:51 +0000 (10:40 +0300)]
systemd-bootdisk.wks: update kernel command line
Used ttyS0 console.
Removed usage of ttyPCH0 (FRI2 leftover)
Decreased bootloader timeout to 5 seconds
Removed 'vmalloc=256MB snd-hda-intel.enable_msi=0' as it's not
needed for any of reference BSPs.
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Joshua Lock [Thu, 6 Oct 2016 10:06:54 +0000 (11:06 +0100)]
testimage: disable build tests for qemumips and qemumips64
It's not uncommon for qemumips[64] builds on the Yocto Project
autobuilder to fail during Sanity Tests after a very long timeout
period. This is due to the MIPS emulation in QEMU being slow and
some of the build tests taking a very long time on MIPS machines.
This patch works around this slowness by disabling the more
complex build tests for QEMU MIPS machines.
[YOCTO #10340]
Signed-off-by: Joshua Lock <joshua.g.lock@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ed Bartosh [Fri, 7 Oct 2016 06:30:09 +0000 (09:30 +0300)]
mkgummidisk.wks: update kernel command line
Used ttyS0 console.
Removed usage of ttyPCH0 (FRI2 leftover)
Decreased bootloader timeout to 5 seconds
Removed 'vmalloc=256MB snd-hda-intel.enable_msi=0' as it's not
needed for any of reference BSPs.
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We fail to compile simple CPP programs because CPP cannot
find relevant header files, looking for them in a non-existing place.
To fix this, we create a symlink of the name CPP expects and point it to
the corresponding existing directory.
[YOCTO#10354]
[YOCTO#10380]
Signed-off-by: Juro Bystricky <juro.bystricky@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Fri, 7 Oct 2016 21:27:44 +0000 (17:27 -0400)]
linux-yocto/4.8: Enable R8169 driver since its needed by supported platforms
The Realtek 8169 driver is needed by the Minnowboard MAX, one of the
boards we currently support, since some of our images do not contain
modules by default, the network isnt working on genericx86, genericx86-64
and intel-core2-32 (meta-intel) when no modules are installed.
This patch fixes network on images not containing modules when
using the previously mentioned MACHINES for this board.
Signed-off-by: Alejandro Hernandez <alejandro.hernandez@linux.intel.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Markus Lehtonen [Thu, 6 Oct 2016 13:37:04 +0000 (16:37 +0300)]
update-rc.d.bbclass: check that init script is executable before running it
Check that the init script that is going to be called in the prerm()
script really exists and is executable. There might be a packaging bug
or the script might've been removed already earlier in prerm().
[YOCTO #10299]
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Thu, 6 Oct 2016 21:43:04 +0000 (10:43 +1300)]
classes/populate_sdk_ext: add symlinks and unfsd to support Eclipse plugin
The Yocto Project Eclipse plugin requires that runqemu and unfsd are
accessible within the SDK, and indeed the standard SDK has these. This
turns out to be fairly easy to do - we just need to add unfsd and symlink
it, runqemu and a few other scripts into the SDK's bin directory.
Fixes [YOCTO #10214].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fabio Berton [Tue, 4 Oct 2016 13:54:11 +0000 (10:54 -0300)]
u-boot: Add support to use uboot-extlinux-config class
Use uboot-extlinux-config class to create extlinux.conf file and then
install inside /boot/extlinux directory and also put file to deploy
dir. This file will be only create if UBOOT_EXTLINUX is set to 1.
You can use DEPLOYDIR/extlinux.conf file to install into final image
using wic setting:
Fabio Berton [Tue, 4 Oct 2016 13:54:10 +0000 (10:54 -0300)]
classes/uboot-extlinux-config: Add class
This class allow the extlinux.conf generation for U-Boot use.
The U-Boot support for it is given to allow the Generic Distribution
Configuration specification use by OpenEmbedded-based products.
This class can be inherited by u-boot recipes to create extlinux.conf
and boot using menu options.
U-boot with extlinux support is machine dependent, so to use this class
you need to set UBOOT_EXTLINUX to 1 in machine configuration file and
also set root= kernel cmdline UBOOT_EXTLINUX_ROOT. This variable is used
to pass root kernel cmdline, e.g:
UBOOT_EXTLINUX_ROOT = "root=/dev/mmcblk2p2"
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>
Khem Raj [Thu, 6 Oct 2016 15:35:24 +0000 (08:35 -0700)]
security_flags: Diable PIE for mesa-gl
This creeped in along with rest of the changes in c999b3d88dfcffbe0fb66406fb0bff1fb66f34bc
even after it was reported a build failure in mesa-gl
This is also showing up on arm architecture now
| /usr/src/debug/glibc/2.24-r0/git/csu/elf-init.c:87: undefined reference to `__init_array_end'
| /usr/src/debug/glibc/2.24-r0/git/csu/elf-init.c:87: undefined reference to `__init_array_start'
| /a/builder/mnt/build/tmp-glibc/sysroots/x86_64-linux/usr/libexec/arm-oe-linux-gnueabi/gcc/arm-oe-linux-gnueabi/6.2.0/ld: .libs/mesa_dri_drivers.so: hidden symbol `__init_array_end' isn't defined
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ed Bartosh [Thu, 6 Oct 2016 15:57:53 +0000 (18:57 +0300)]
mkefidisk.wks: update kernel command line
Used ttyS0 console.
Removed usage of ttyPCH0 (FRI2 leftover)
Decreased bootloader timeout to 5 seconds
Removed 'vmalloc=256MB snd-hda-intel.enable_msi=0' as it's not needed
for any of reference BSPs.
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ed Bartosh [Thu, 6 Oct 2016 10:33:36 +0000 (13:33 +0300)]
wic: generate PARTUUID for MDOS partitions
Added generation of partition UUIDs for MSDOS partitions.
UUID for MSDOS partitions is <disk identifier>-<partition number>,
where disk identifier is a random 4 bytes long number. It's usually
generated when MBR/partition table is initialized.
As UUID is used to point to the root partition in bootloader config
we need to generate it before the MBR is initialized.
After MBR is created we need to rewrite system identifier to match
it with what is used in bootloader config. This will be implemented
in the next commit.
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Hatle [Thu, 6 Oct 2016 16:59:38 +0000 (11:59 -0500)]
linuxloader.bbclass: Adjust mips to cover all mips/mips64
[YOCTO #10389]
Use a glob (*) to match all mips (not previously matched). This will ensure
that the linuxloader is properly returned for mips, mipsel, mips64,
mips64el and their n32 variants.
See: https://sourceware.org/glibc/wiki/ABIList#mips for the official list
of loaders.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The loader is located at a new place for multiarch.
For more details, check https://wiki.debian.org/Multiarch
and https://sourceware.org/glibc/wiki/ABIList#mips
Signed-off-by: Zubair Lutfullah Kakakhel <Zubair.Kakakhel@imgtec.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Stephano Cetola [Mon, 3 Oct 2016 23:32:45 +0000 (16:32 -0700)]
utils.bbclass: add function to check for git config user
If attempting to patch a git repo without a proper git config setup,
an error will occur saying user.name/user.email are needed by git
am/apply. After some code was removed from kernel-yocto, it was
simple enough to reproduce this error by creating a kernel patch and
using a container to build.
This patch abstracts out functionality that existed in buildhistory
for use in other classes. It also adds a call to this functionality
to the kernel-yocto class.
bavery [Thu, 6 Oct 2016 00:43:06 +0000 (17:43 -0700)]
uninative: users can override download site
The default download site for the uninative tarball is
http://downloads.yoctoproject.org/releases/uninative/<version>. There
are scenarios in which the user may need to force the download to be
somewhere else. This patch allows the UNINATIVE_URL to be set in the
local.conf.
Signed-off-by: bavery <brian.avery@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Stephano Cetola [Wed, 5 Oct 2016 17:07:17 +0000 (10:07 -0700)]
devtool: modify command fails to ignore source files
With recent changes to recipeutils, the list of local files returned
by get_recipe_local_files could possibly include source files. This
only happens when the recipe contains a SRC_URI using subdir= to put
files in the source tree. These files should be ignored when
populating the list of local files for oe-local-files directory.
package_manager.py: Allow multiple regexps in PACKAGE_EXCLUDE_COMPLEMENTARY
The PACKAGE_EXCLUDE_COMPLEMENTARY variable can currently only contain
one regular expression. This makes it hard to add to it from different
configuration files and recipes.
Allowing it to contain multiple, whitespace separated regular
expressions should be backwards compatible as it is assumed that
whitespace is not used in package names and thus is not used in any
existing instances of the variable.
After this change, the following three examples should be equivalent:
package_manager.py: Allow a leading - in PACKAGE_EXCLUDE_COMPLEMENTARY
This allows a regular expression specified in
PACKAGE_EXCLUDE_COMPLEMENTARY to have a leading dash. Without this,
the dash was treated by oe-pkgdata-util as the beginning of a command
line argument. E.g., if PACKAGE_EXCLUDE_COMPLEMENTARY = "-foo$", it
resulted in an error like:
ERROR: <imagename>-1.0-r0 do_populate_sdk: Could not compute
complementary packages list. Command '<topdir>/scripts/oe-pkgdata-util -p
<builddir>/tmp/sysroots/<machine>/pkgdata glob
<workdir>/installed_pkgs.txt *-dev *-dbg -x -foo$' returned 2:
ERROR: argument -x/--exclude: expected one argument
usage: oe-pkgdata-util glob [-h] [-x EXCLUDE] pkglistfile glob [glob ...]
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Markus Lehtonen [Wed, 5 Oct 2016 13:29:49 +0000 (16:29 +0300)]
oeqa.buildperf: measure apparent size instead of real disk usage
This change aligns disk usage measurements of the eSDK test with the old
build-perf-test.sh script. And thus, also makes the results between the
old and the new script comparable.
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jussi Kukkonen [Tue, 4 Oct 2016 13:16:42 +0000 (16:16 +0300)]
gtk+3: Backport treeview focus fix
Treeview did not grab focus properly on mouse click, leading to e.g.
multifile selection with click/shift-click not working in the
filechooser. Backport a fix.
Fixes [YOCTO #10273].
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Wed, 5 Oct 2016 03:03:48 +0000 (23:03 -0400)]
linux-yocto/4.8: fix BUG_ON() in workingset_node_shadows_dec() triggers
Paul Gotmaker pointed out that a last minute merge to the 4.8 kernel
has the potential to hard hang a kernel when VM debugging is enabled:
https://lkml.org/lkml/2016/10/4/1
He also pointed out the fix for it in commit 21f54dda
[Using BUG_ON() as an assert() is _never_ acceptable].
While that fix will loop through -stable into 4.8.1, that will
likely be too late for our release. So I've cherry picked the
change to make it available.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 5 Oct 2016 07:55:11 +0000 (08:55 +0100)]
pigz: Update SRC_URI
Upstream have released a new tarball and removed the old one. Revert to
the Yocto Project source mirror instead, preserving the upstream version
check.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This patch brings the last bit from meta-mentor for the perf
to build successfully with minnowmax BSP. The meta-mentor
commit for the same is:
http://git.yoctoproject.org/cgit/cgit.cgi/meta-mentor/commit/meta-mentor-staging?id=a8db95c0d4081cf96915e0c3c4063a44f55e21cc
The previous fix:
http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/meta/recipes-kernel/perf?id=ef942d6025e1a339642b10ec1e29055f4ee6bd46
was incomplete and was not submitted upstream. And due to that this change is required.
When built on minnowmax ( machine name: intel-corei7-64),
an error is noticed during the do_compile:
/home/sujith/codebench-linux-install-2015.12-133-i686-pc-linux-gnu/codebench/bin/i686-pc-linux-gnu-ld:
Relocatable linking with relocations from format elf64-x86-64
(/home/sujith/MEL/dogwood/build-minnowmax/tmp/work/intel_corei7_64-mel-linux/perf/1.0-r9/perf-1.0/fd/array.o)
to format elf32-i386 (/home/sujith/MEL/dogwood/build-minnowmax/tmp/work/intel_corei7_64-mel-linux/perf/1.0-r9/perf-1.0/fd/libapi-in.o)
is not supported
This change help fix the issue.
Signed-off-by: Sujith Haridasan <Sujith_Haridasan@mentor.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The following patch comes from the realization that at least ARM code
for atomics is quite broken and nobody has cared for a decade.
A quick dive shows that only snd_atomic_{read,write}_{begin,end}
appear to be used widely. These are implemented using wmb/rmb.
Only other use of atomic functions is in pcm_meter.c.
The #SND_PCM_TYPE_METER plugin type appears rarely, if ever, used.
I presume these days anyone who wants a meter/scope will do in pulseaudio
layer instead of alsa.
It would seem better fit to have pcm_meter in alsa-plugins instead
of alsa-lib, but I guess that would be an ABI break...
So instead, I'm proposing here
1. Removal of all hand-crafted atomics from iatomic.h apart from barriers,
which are used in snd_atomic_{read,write}_{begin,end}.
2. Using __sync_synchronize as the default fallback for barriers. This
has been available since gcc 4.1, so it shouldn't be a problem.
3. Defining the few atomics used by pcm_meter.c withing pcm_meter.c
itself, using gcc atomic builtins[1].
4. Since gcc atomic builtins are available only since gcc 4.7, add a check for
that in gcc configure.in, and don't build pcm meter plugin if using
older gcc.
The last point has the impact, that if there actually is someone who 1)
uses the meter plugin 2) wants to upgrade to 2014 alsa-lib 3) but
does not want to use a 2012+ gcc - that someone will be inconvenienced.
Finally remove the unneeded configure check for cpu type. We can
trust the gcc to set right flags for us.
devtool: deploy-target: Avoid unnecessary dependency on awk on the target
Relying on that awk is installed on the target just to extract the
fourth column (i.e., the free volume size) from `df -P` is an
unnecessary dependency for devtool deploy-target. As it is already
using sed to mangle the output from `df -P`, this can easily be
modified to only extract the free volume size.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
parselogs.py: Add disabling eDP error to x86_common whitelist
The NUC6 firmware tells the kernel to try and initialize an embedded
DisplayPort it does not have, causing this warning. Its harmless, so
just whitelist it.
Fixes [YOCTO #9434].
Signed-off-by: California Sullivan <california.l.sullivan@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Aníbal Limón [Tue, 4 Oct 2016 16:45:53 +0000 (11:45 -0500)]
classes/sstate.bbclass: Enable thread lock when checkstatus
The checkstatus function fires an event to notify bitbake UI about
the progress of the task, this function is implemented using ThreadPool
and is causing event lose when multiple threads tries to fire an event
(writes over socket/fd).
[YOCTO #10330]
Signed-off-by: Aníbal Limón <anibal.limon@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jussi Kukkonen [Tue, 4 Oct 2016 07:17:27 +0000 (10:17 +0300)]
xf86-video-intel: Upgrade to recent git
Upgrade from the latest snapshot to a recent git revision.
Without this xvideo does not work on skylake: Backporting the
specific fixes turned out to be too complex.
Remove patches that are in upstream already, rebase
disable-x11-dri3.patch.
Fixes [YOCTO #10041]
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jussi Kukkonen [Tue, 4 Oct 2016 11:27:26 +0000 (14:27 +0300)]
matchbox-panel-2: Fix small systray icon drawing
Add patch to pack systray icons so that their drawing area is the
size they expect (otherwise GtkStatusIcon based systray items can
end up drawing "tiled", looking like 1.5 icons instead of a single
icon).
Fixes [YOCTO #9995]
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jussi Kukkonen [Tue, 4 Oct 2016 11:27:25 +0000 (14:27 +0300)]
Revert "connman-gnome: StatusIcon adapts to size changes"
The aim of the original commit was to make connman-gnome load the icons
at the exact size of the systray. There are two problems with this:
* There are not enough icon sizes provided to make the scaling
look good at most sizes (including current panel size)
* Both connman-gnome and mb-panel have bugs in the icon size update
code and using scaling to exact size makes these much more visible
(See bug 9995 for example).
The problems the original commit tried to fix can be worked around
with better packing in matchbox-panel-2.
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Tue, 4 Oct 2016 09:31:16 +0000 (22:31 +1300)]
devtool: add: build nodejs-native if npm is needed and not available
If the user runs devtool add on an npm:// URL (or source tree that uses
node.js), and npm is not available, just build nodejs-native instead of
telling the user they need to do it; if that fails because there isn't
any such recipe (which would be the default, since it's not in OE-Core)
then produce a slightly more readable error message hinting at what the
user needs to do.
Note that this forces the use of nodejs-native rather than npm on the
host - this makes sense for two reasons: (1) we need it to be compatible
with nodejs for the target, and (2) we have to have a recipe for that
anyway, so allowing you to avoid having a recipe for the native version
isn't really beneficial.
There's a bit of a hack in here in order to allow this - for node.js
sources that aren't fetched via npm we don't know that they are that
until we've fetched and unpacked them, by which time we're inside
recipetool and have an active tinfoil instance that will prevent bitbake
being run. To avoid this being an issue, we allow recipetool to get to
the point where we know we need npm and then exit with a specific exit
code, at which point devtool can try to build it and then if that
succeeds, it will re-execute recipetool. This is definitely not ideal,
but it can't really be refactored and done properly until we do the
tinfoil2 refactoring; in the mean time though we still want to be
helpful to the user.
Fixes [YOCTO #10337].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
