Ross Burton [Tue, 26 Mar 2013 16:18:28 +0000 (16:18 +0000)]
udev: move /run volatile entry to udev instead of initscripts
initscripts is generally installed on systemd-using images, but because it
specifies that /run is a symlink to /var/run managed by volatiles it totally
breaks systemd by copying/deleting /run from underneath systemd. Deleting
sockets mid-boot doesn't leave systemd in a happy place.
As this volatile reference of /run was introduced by udev 182, move it's
reference to the udev recipe. This way it will never be present on systemd
images, as systemd manages /run as a tmpfs itself.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Wed, 20 Mar 2013 17:00:06 +0000 (17:00 +0000)]
packagegroup-core-boot: revert to specifying sysvinit as default init manager
Don't follow DISTRO_FEATURES_INITMAN as that makes the packagegroups rebuild if
you switch init manager.
As in hybrid situations there's generally a clear primary and minimal init
manager choice, so change VIRTUAL-RUNTIME_init_manager to set the primary init
manager, and roll your own groups/images for the secondary.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Tue, 5 Mar 2013 14:12:50 +0000 (14:12 +0000)]
systemd: check for systemctl first, and don't force systemd to be installed.
With both sysvinit and systemd features enabled these postinsts may actually run
on a target without systemd, so check that systemctl is present before using it.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Thu, 28 Feb 2013 16:28:28 +0000 (16:28 +0000)]
update-rcd.bbclass: handle both sysvinit and systemd features being present
Run the helper if the sysvinit feature is present, or if the systemd feature is
present but the systemd class hasn't been inherited. We want to run in the
latter case as systemd has sysvinit compatibility, but we don't want to always
run so that pure systemd images don't have redundant sysvinit files.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Thu, 28 Feb 2013 15:58:53 +0000 (15:58 +0000)]
default-providers: change udev selection logic
Change the logic so that the udev provider is the standalone udev, unless the
systemd DISTRO_FEATURE is set. The previous logic was designed to fail if both
sysvinit and systemd were enabled, which we're supporting now.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Paul Eggleton [Tue, 26 Mar 2013 17:47:26 +0000 (17:47 +0000)]
classes/rm_work: rename RM_WORK_WHITELIST to RM_WORK_EXCLUDE
This really functions as a blacklist, not a whitelist, since we are
listing recipes to exclude. To avoid any possibility of confusion, since
this was a recent addition, rename the variable.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Hatle [Tue, 26 Mar 2013 03:35:13 +0000 (22:35 -0500)]
rpm: Fix debugedit buildid processing
[ YOCTO #4089 ]
When constructing a new buildid, the items being hashed need to be
returned to their native endian. In the process we were munging
the sh_type field that we relied on to determine if a section was
loadable or not. The patch avoids this behavior by only modifying
a copy of the local endian data.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Nitin A Kamble [Tue, 26 Mar 2013 16:51:45 +0000 (09:51 -0700)]
grub-efi-native_2.00.bb: fix a build issue
Due to recent change in the oecore layer grub-efi recipe uses separate
builddir and the paths used in the do_deply need to change accordingly.
This change avoids this build issue:
install: cannot stat `/build-fri2/tmp/work/x86_64-linux/grub-efi-i586-native/2.00-r1/grub-2.00/bootia32.efi': No such file or directory
ERROR: Function failed: do_deploy (see /build-fri2/tmp/work/x86_64-linux/grub-efi-i586-native/2.00-r1/temp/log.do_deploy.5875 for further information)
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
having /usr/local/include hardcoded into the makefile is not necessary
as this is automatically included by GCC. It also infects cross-compile
builds with the host systems includes.
Signed-off-by: Jack Mitchell Acked-by: Namhyung Kim Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1362741712-21308-1-git-send-email-ml@communistcode.co.uk Signed-off-by: Arnaldo Carvalho de Melo
This implements the fix I described in Yocto [BUG #3993][perf using
host includes], now upstream.
Integrated-by: Tom Zanussi <tom.zanussi@linux.intel.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
[YOCTO #3993]
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Tue, 26 Mar 2013 14:46:32 +0000 (07:46 -0700)]
linux-yocto/3.8: allow kernel feature _appends to be overriden
It was pointed out that the current way the KERNEL_FEATURES variable
is appended in the base linux-yocto recipe doesn't allow the appended
features to be prevented in a layer without using python code and
a recipe finalize hook.
To allow easier overriding of 'extra' or 'optional' features that are
defined in the linux-yocto recipe, we create a KERNEL_EXTRA_FEATURES
variable. This variable can be set in a layer to define extra features
or cleared to prevent the recipe's extra features from being appended
to the core functionality.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Tue, 26 Mar 2013 12:43:05 +0000 (12:43 +0000)]
qemuimagetest: collect and print runqemu output
If runqemu (or qemu itself) fails we need to know why, so tee out to a
log file and print it when we can't find the qemu process or determine
its IP address.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Tue, 26 Mar 2013 12:43:04 +0000 (12:43 +0000)]
qemuimagetest: fix erroneous ps errors when qemu couldn't be started
The helper script was printing an error to stdout when it couldn't find
any qemu child processes; output this error to stderr instead and
redirect stderr to /dev/null when running from qemuimage-testlib so that
QEMUPID is actually blank if there are no qemu instances found.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Tue, 26 Mar 2013 17:32:40 +0000 (17:32 +0000)]
font-alias: Add depends on font-util-native
Without this, the font-util.m4 macro can be missing leading to an undefined
macro. The recipe will still build but generate an empty font-alias package
since the files are installed into "${D}@XORGROOTFONTDIR@".
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Tue, 26 Mar 2013 17:31:14 +0000 (17:31 +0000)]
opkg-utils: Update to newer git rev for bashism fix
On systems with dash as /bin/sh there were failures in do_package_write_ipk
due to bashisms in opkg-build. This newer revision contains a fix for this.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
After the recent cached path operation improvements, we need to tell
the cache about directory changes. /usr/src/debug was was cached as
not created and this was leading to bugs in file ownership. This
change updates the cache and avoids this issue.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Mon, 25 Mar 2013 18:00:56 +0000 (11:00 -0700)]
docbook-utils-native: Unbreak the build after source and build dir split
docbook-utils-native started to install frontends and backends in
wrong locations after this change. The absolute build and source
directories were being created in install locations and the recipes
using docbook-utils-native were complaining like dbus-java is saying
| jw: There is no frontend called
"/b/kraj/jlinux-next/poky/build/tmp-eglibc/sysroots/x86_64-linux/usr/share/sgml/docbook/utils-0.6.14/frontends/docbook".
| make: *** [CreateInterface.1] Error 2
| ERROR: oe_runmake failed
Rightly so because now they are installed under
/b/kraj/jlinux-next/poky/build/tmp-eglibc/sysroots/x86_64-linux/usr/share/sgml/docbook/utils-0.6.14/b/kraj/jlinux-next/poky/build/tmp-eglibc/work/x86_64-linux/docbook-utils-native/0.6.14-r3/
This patch fixes the install to be like it was but consider
the build and sourcedir split.
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 25 Mar 2013 20:46:50 +0000 (20:46 +0000)]
at: Fix --with/without pam options
The --with/--without pam support in configure appears to have been dropped
leading to builds which can detect host headers, then fail or result in non-
determinstic builds. Add an explicit option to fix this detection.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 25 Mar 2013 17:34:07 +0000 (17:34 +0000)]
package.bbclass: Ensure debug source file is generated correctly
The command for generating debug sources appends to the file, therefore we need
to ensure it does not exist before we run the command else we end up doing
more work that we should have to.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 25 Mar 2013 16:31:45 +0000 (16:31 +0000)]
conf/machine: Clean up MACHINEOVERRIDES handling
OVERRIDES reads from left to right, least to most specific. We were
appending to MACHINEOVERRIDES when we should have been prepending so
the ordering of qemuall verses qemuxxx was incorrect, as was the x86
override and several of the arm overrides. This patch is a batch cleanup
of the various issues to correct the order from least to most specific.
The include order does matter and we needed to tweak some of that in this
patch too.
[YOCTO #4090]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If an error occurs in subprocess.call() we currently don't catch it. In particular
we have issues where debugedit is segfaulting unnoticed. This fixes up
various code paths to catch the errors.
[YOCTO #4089]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Barker [Mon, 25 Mar 2013 13:54:43 +0000 (13:54 +0000)]
gnupg: Add symlink for gpgv
Gnupg 2.0.19 installs 'gpgv2' but apt-get and possibly other utilities expect
to find this as 'gpgv'. A symlink is created in the same way as the link for
'gpg'.
Signed-off-by: Paul Barker <paul@paulbarker.me.uk> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Hatle [Mon, 25 Mar 2013 17:19:52 +0000 (12:19 -0500)]
rpm: Add workaround for debugedit-segv
[ YOCTO #4089 ]
On PPC and MIPS, there appears to be a condition that causes
debugedit to segfault. The segfault is related to a call into
the md5hash algorithm, an address of '0', and a size > 0 is passed
causing the access of the address to segv.
This workaround may prove to be the final fix, but it's currently
unclear what the actual cause of the 0 address is.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Laurentiu Palcu [Mon, 25 Mar 2013 16:08:18 +0000 (18:08 +0200)]
nspr: remove unnecessary files from the main package
This patch does the following:
* removes compile-et.pl and prerr.properties from the main package
because these files are needed only at compile time;
* moves nspr-config script to nspr-dev package where it belongs;
* adds 'perl' RDEPENDS for nspr-dev since the package contains a perl
script for running some tests;
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Mon, 25 Mar 2013 14:03:13 +0000 (14:03 +0000)]
libxcb: fix repeated configures
If configure is re-ran on an existing build tree the string substitutions we
need are done twice, resulting in invalid paths. Anchor the expressions so they
only match a pristine configure.ac.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sat, 23 Mar 2013 17:36:53 +0000 (17:36 +0000)]
package/populate_sdk: Move functions from package_* to populate_sdk_*
This fixes build failures introduced with "classes/buildhistory: implement history
collection for SDK" by moving the functions to files where only the specific image
type which is enabled is inherited. The failures occured when multiple PACKAGE_CLASSES
were enabled.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jesse Zhang [Tue, 19 Mar 2013 02:07:01 +0000 (10:07 +0800)]
Perl packages shouldn't be allarch
Perl packages should know about arch so that they could install to the
correct libdir. Or else they always install to the default libdir in
multilib builds.
Signed-off-by: Jesse Zhang <sen.zhang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Tue, 19 Mar 2013 11:41:38 +0000 (11:41 +0000)]
packagegroup-core-tools-debug: remove rsync
This was added for supporting Anjuta, which is no longer actively
supported; in any case the anjuta-remote-run package RDEPENDS on rsync
so it will be brought in by that if that is added.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Sun, 10 Mar 2013 22:05:18 +0000 (15:05 -0700)]
sudo: Compile mksigname and mksiglist for build host
cross compiling sudo doesnt work well since it uses
mksigname and mksiglist to generate C sources which
are then used in sudo build itself. With this patch
now we make sure those hosttools are compiled for
build machine. It fixes the build failures like
Paul Eggleton [Fri, 22 Mar 2013 19:53:52 +0000 (19:53 +0000)]
classes/buildhistory: implement history collection for SDKs
SDKs are constructed in a similar manner to images, and the contents can
be influenced by a number of different factors, thus tracking the
contents of produced SDKs when buildhistory is enabled can help detect
the same kinds of issues as with images.
This required adding POPULATE_SDK_POST_HOST_COMMAND and
SDK_POSTPROCESS_COMMAND variables so that data collection functions can
be injected at the appropriate points in the SDK construction process,
as well as moving the list_installed_packages and
rootfs_list_installed_depends functions from the rootfs_{rpm,ipk,deb} to
the package_{rpm,ipk,deb} classes so they can also be called during
do_populate_sdk as well as do_rootfs.
Implements [YOCTO #3964].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Hatle [Tue, 19 Mar 2013 20:17:48 +0000 (15:17 -0500)]
vte: Fix conflict between FILES_${PN}-dbg and FILES_vte-dbg
Due to quarks with the way bitbake handles variable key values, the
hard coded FILES_vte-dbg or automatic FILES_${PN}-dbg could replace
each other, leaving the system in an inconsistent state.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Wed, 20 Mar 2013 06:31:05 +0000 (14:31 +0800)]
local.conf.sample.extended: update for disk monitor
The previous disk monitor is based on the disk, and one disk only can
have one action, now it is based on the path, so we remove the additional
comments from local.conf.sample.extended
[YOCTO #3995]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Zhenhua Luo [Wed, 20 Mar 2013 07:52:12 +0000 (15:52 +0800)]
fix march sanity check issue
1. check if gcc_test is really generate before os.remove("gcc_test") to avoid
following error:
ERROR: Execution of event handler 'check_sanity_eventhandler' failed
Traceback (most recent call last):
File "check_sanity_eventhandler(e)", line 4, in check_sanity_eventhandler(e=<bb.event.ConfigParsed object at 0x3151450>)
File "sanity.bbclass", line 107, in check_sanity(sanity_data=<bb.data_smart.DataSmart object at 0x11ba110>)
File "sanity.bbclass", line 22, in check_gcc_march(sanity_data=<bb.data_smart.DataSmart object at 0x11ba110>)
OSError: [Errno 2] No such file or directory: 'gcc_test'
2. set result to False when build failed with -march=native to ensure
-march=native is appended to BUILD_CFLAGS when host gcc really supports this flag,
otherwise following error appears when build native packages.
| cap_text.c:1: error: bad value (native) for -march= switch
| cap_text.c:1: error: bad value (native) for -mtune= switch
Signed-off-by: Zhenhua Luo <zhenhua.luo@freescale.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Darren Hart [Tue, 19 Mar 2013 22:15:06 +0000 (15:15 -0700)]
mkefidisk: Use msdos partition tables
While GPT works fine when writing to actual media, it cannot be reliably
used for distributing disk images as it requires the backup table to be
on the last block on the device, which of course varies from device to
device. Use MSDOS tables instead.
Use mkfs to label the filesystems as msdos tables do not support
partition labeling.
Signed-off-by: Darren Hart <dvhart@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Darren Hart [Tue, 19 Mar 2013 22:15:05 +0000 (15:15 -0700)]
mkefidisk: Remove startup.nsh
Most firmware implementations use the EFI specified
EFI/BOOT/bootia32.efi (and similar) boot paths. Only broken firmware
uses different paths for removable media. In those cases, the user can
add their own startup.nsh.
For the compliant case, selecting "Shell" from the EFI boot menu should
go to the shell.
Signed-off-by: Darren Hart <dvhart@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Darren Hart [Tue, 19 Mar 2013 22:15:04 +0000 (15:15 -0700)]
mkefidisk: Always use rootwait and document kernel parameters used
Without a reliable way of knowing if the target device with be an
asyncronous block device on the target (MMC or USB), err on the side of
caution of always specifcy "rootwait", ensuring the kernel will wait for
the device to appear and not abort if it hasn't appeared in time for
mount.
Document the remaining kernel parameters added by this script on the
same line as rootwait.
Signed-off-by: Darren Hart <dvhart@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Darren Hart [Tue, 19 Mar 2013 22:15:03 +0000 (15:15 -0700)]
mkefidisk: Boot with ro kernel parameter instead of rw
There is no need to boot with "rw". Booting with "ro" will allow for
fsck to be run during boot, and a proper /etc/fstab will still ensure
the rootfs is "rw" by the time the user can interact with the system.
Change the "rw" to "ro" in the kernel parameters specified in the
generated grub.cfg file.
Fixes [YOCTO 4036] mkefidisk.sh hardcodes 'rw' as root mount option
Signed-off-by: Darren Hart <dvhart@linux.intel.com> Cc: Koen Kooi <koen@dominion.thruhere.net> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Darren Hart [Tue, 19 Mar 2013 22:15:02 +0000 (15:15 -0700)]
mkefidisk: Always specify a root= kernel parameter
The current script only replaces an existing root= kernel parameter
which can result images created without a root= paremeter, even though
the script expects a target rootfs parameter.
Rather than replacing the root= parameter, delete the root= parameter if
it exists, then append an appropriate root= parameter.
Fixes [YOCTO 4035] mkefidisk.sh forgets to add root= parameter
Signed-off-by: Darren Hart <dvhart@linux.intel.com> Cc: Koen Kooi <koen@dominion.thruhere.net> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Thu, 21 Mar 2013 09:24:25 +0000 (09:24 +0000)]
consolekit: don't inherit gnome
By inheriting gnome this package was build-depending on
gtk-update-icon-cache-native, gconf, gconf-native, shared-mime-info and
hicolor-icon-theme for no reason. Change this to inherit autotools pkgconfig to
reduce the build-dependencies.
Also, add a missing dbus-glib build-dependency that was previously pulled in
through something in the gnome class.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
package_{ipk, deb, rpm}: drop the TARGET_OS conditional
The tclibc file for uclibc already empties IMAGE_LINGUAS, so there's no point
to this conditional as far as I can tell, and it can cause issues for certain
values of TARGET_OS.
Signed-off-by: Christopher Larson <chris_larson@mentor.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This was removed from package_rpm and package_ipk years ago, and shouldn't be
needed, installing the packages from package_linguas should be sufficient.
Signed-off-by: Christopher Larson <chris_larson@mentor.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Coreutils does not depend on ACL but also does not disable it. Effect is
that from time to time I have a copy in sstate-cache which got built
with ACL and then I have problem while building images or sdk:
| Collected errors:
| * satisfy_dependencies_for: Cannot satisfy the following dependencies for packagegroup-core-standalone-hhvm-sdk-target:
| * libacl1 (>= 2.2.51) *
| * opkg_install_cmd: Cannot install package packagegroup-core-standalone-hhvm-sdk-target.
Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Thu, 21 Mar 2013 15:43:40 +0000 (15:43 +0000)]
perl: avoid splitting out .debug directories as packages
The recursive parameter is set to True, the regex for the first call to
do_split_packages matches any path under ${libdir}/perl/${PV}/auto/, and
the .debug directories contain .so files, so each one was getting picked
up as a package. Change the regex to disallow dots in the path beneath
auto/ and thus avoid the .debug directories.
Fixes [YOCTO #4048].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jukka Rissanen [Fri, 22 Mar 2013 08:11:41 +0000 (10:11 +0200)]
connman-conf: Move the setup script into /usr/lib/connman
The /etc/connman is not suitable for the setup script.
There are other connman related scripts in /usr/lib/connman
so moving the wired setup script there.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Tue, 5 Mar 2013 16:48:11 +0000 (16:48 +0000)]
distro: Add separatebuilddir.inc
Going forward its going to be useful to separate build data from source data
in those autotooled projects which support it. Unfortunately there is a lot
of breakage so for now, this starts the creation of an opt in list which
we can iterate over enable more recipes over time.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
