Ross Burton [Tue, 12 Apr 2016 14:51:54 +0000 (15:51 +0100)]
rsync: remove upstream's rebuild logic
Upstream some well intended but broken logic to reimplement the rebuild
functionality of automake. However this isn't out-of-tree safe and quite basic,
which means if it ever does execute (say, configure.ac or aclocal.m4 is touched)
then the build fails.
As we delete ${B} and re-run autoreconf on every build this is redundant, so
just delete it all.
[ YOCTO #9445 ]
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Tue, 12 Apr 2016 14:51:52 +0000 (15:51 +0100)]
rsync: don't install acinclude.m4
This acinclude.m4 was actually a stale copy of upstream's generated aclocal.m4.
This generates correctly now, so there isn't a need to install this by hand
anymore.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Tue, 12 Apr 2016 14:51:50 +0000 (15:51 +0100)]
Revert "wic/utils/partitionedfs.py: assemble .wic images as sparse files"
It turns out that dd's conv=sparse doesn't look at the file extents, but simply
checks if a "block" is all zero. If the block of zero was meaningful it gets
lost and if the image is subsequently written to media using a sparse-aware
writer then the block of zeros won't be written at all.
Richard Purdie [Tue, 12 Apr 2016 14:44:08 +0000 (15:44 +0100)]
archiver: Ensure sstate-inputdir directory is created
Without this, do_archive_sources will fail for the gcc tasks where there is
common source and the other tasks are not added as dependencies. The failure
happens when trying to restore these tasks from sstate.
To fix this ensure the sources directory is created as expected by the
sstate code.
[YOCTO #9433]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jussi Kukkonen [Mon, 11 Apr 2016 11:07:52 +0000 (14:07 +0300)]
glib-2.0: Put glib-compile-schemas back in -utils
Commit cc97d576 moved a bunch of development tools to the -dev
package. glib-compile-schemas is actually used in postinst by
gsettings.bbclass so it needs to be available on target at package
install time: Move the tool back to glib-2.0-utils which
gsettings.bbclass depends on.
Fixes [YOCTO #9431].
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Mon, 11 Apr 2016 17:33:16 +0000 (18:33 +0100)]
oeqa/selftest/sstatetests: split 32/64 build host from no-op action tests
As more variables were being added it made sense to split this test into two:
one quick test that verifies that the host being 32-bit or 64-bit doesn't impact
the hashes, and another which compares stamps for two builds with several
variables changed and classes inherited.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Mon, 11 Apr 2016 20:43:34 +0000 (21:43 +0100)]
util-linux: take ownership of hwclock if installed
Previously util-linux had a lower priority for hwclock than busybox but the
reasoning was lost in the mists of time, with just this enigmatic comment
remaining:
There seems to be problem, atleast on nslu2, with these, until they are
fixed the busybox ones have higher priority.
Chasing the comment back through history it first appeared in the following
oe-classic commit:
ac_cv_sizeof_off_t was previously in the site cache files, which was breaking
large file support and required a workaround in each recipe that actually wanted
to use large files.
Now that the entry has been removed from the site cache, we can remove the
workarounds.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sat, 9 Apr 2016 22:16:47 +0000 (23:16 +0100)]
archiver: Fix ASSUME_PROVIDED issues
Currently the dependencies injected by this class for recipes in ASSUME_PROVIDED
e.g. tar-native or chrpath-native are ignored. We need to use their "replacement"
names, e.g. tar-replacement-native.
This avoids broken archives and sstate failures with these recipes.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
build-appliance-image: Support for VirtualBox guest additions
Add support for VirtualBox guest additions.
The additions are built inside the Build Appliance VM.
For this to be possible, the environment for out-of-tree kernel module builds
must be present and set up properly.
A README file with detailed steps on how to build the guest additions is placed
in the home directory of the user "builder" as well.
The main purpose of the guest additions is to allow sharing folders between
the host and Build Appliance VM.
[YOCTO #8073]
Signed-off-by: Juro Bystricky <juro.bystricky@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
local.conf.sample: Make it possible to override EXTRA_IMAGE_FEATURES
It should be possible to override EXTRA_IMAGE_FEATURES in auto.conf,
which is read before local.conf.
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>
oeqa/selftest/wic: add test case for sparse images
Add a testcase to build a directdisk image and check that the
used disk size is less than the apparent size, as wic now
assembles images as sparse files.
Signed-off-by: Joshua Lock <joshua.lock@collabora.co.uk> Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If we create hdddirect in the first time, and run bitbake to create
vmimg in the second time, then the previouse created
${IMAGE_LINK_NAME}.hdddirect may can not be found since it contains the
data string which are different. Use IMAGE_LINK_NAME to fix the problem.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Fri, 8 Apr 2016 07:15:19 +0000 (00:15 -0700)]
image_types.bbclass: set nodesize for btrfs
The default value is 16K which is too big to create image for
core-image-minimal:
rootfs.btrfs is too small to make a usable filesystem
Minimum size for each btrfs device is 41943040.
Use 4K as ext2/3/4 to fix the problem.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Asselstine [Fri, 8 Apr 2016 21:46:27 +0000 (17:46 -0400)]
useradd_base.bbclass: prevent variable expansion in $opts
Many user/group operations will involve hashes which will include '$'
followed by a number or even possibly an env. variable name. Passing
$opts to flock requires that we take additional precautions to prevent
the unexpected expansion of these instances.
This was found by an image which used usermod operations to set the
password hash for root. The image could not be logged-in to and
examining /etc/shadow clearly showed that $0 and other $* variables
had been expanded unexpectedly. This change returnes the behavior to
what existed prior to commit 2ebf697b46c42cee8bfa6d2e6087397f8cce385c
[useradd_base.bbclass: replace retry logic with flock].
Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Asselstine [Fri, 8 Apr 2016 21:46:26 +0000 (17:46 -0400)]
extrausers.bbclass: drop retry count for perform_user/group* calls
Commit 2ebf697b46c42cee8bfa6d2e6087397f8cce385c [useradd_base.bbclass:
replace retry logic with flock] dropped the 3rd (retry) parameter for
these functions. These are simply being ignored now but we should
remove the retry count to avoid confusion.
Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 6 Apr 2016 12:20:29 +0000 (13:20 +0100)]
rpm: brace expansion is a bashism
The constuct rm *.{a,la} is a bashism and as tasks are run under /bin/sh this
may not work as expected. Expand the {a,la} to two separate calls, and remove
the architecture-specific macro deletion as they are not installed in the first
place anymore.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Andre McCurdy [Thu, 7 Apr 2016 00:42:05 +0000 (17:42 -0700)]
openssl.inc: minor packaging cleanup
libcrypto.so was explicitly added to FILES_${PN}-dev as part of moving
libcrypto from libdir -> base_libdir to support dhclient [1].
However, the line has been unnecessary since ${base_libdir}/lib*.so
files started to be included in FILES_${PN}-dev by default [2] (and
it's still unnecessary now, after moving libcrypto from back to libdir
to support ntp [3]).
Robert Yang [Thu, 7 Apr 2016 07:34:32 +0000 (00:34 -0700)]
apr-util: fix path in rules.mk for nativesdk
Fixed when build nativesdk-apr-util:
| gawk: fatal: can't open source file `/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/share/build-1/make_exports.awk' for reading (No such file or directory)
The ${S} should be ${B}.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Thu, 7 Apr 2016 07:34:31 +0000 (00:34 -0700)]
bdwgc: installed-vs-shipped for nativesdk
Fixed:
ERROR: nativesdk-bdwgc-7.4.2-r0 do_package: QA Issue: nativesdk-bdwgc: Files/directories were installed but not shipped in any package:
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/share
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/share/gc
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/share/gc/porting.html
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/share/gc/gcinterface.html
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/share/gc/gcdescr.html
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/share/gc/README.solaris2
[snip]
This was caused by hardcode of datadir.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Thu, 7 Apr 2016 07:34:30 +0000 (00:34 -0700)]
libsolv: fix installed-vs-shipped for nativesdk
Fixed:
ERROR: nativesdk-libsolv-0.6.19-r0 do_package: QA Issue: nativesdk-libsolv: Files/directories were installed but not shipped in any package:
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man3
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man1
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man3/libsolv-constantids.3
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man3/libsolv.3
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man3/libsolv-history.3
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man3/libsolv-pool.3
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man3/libsolv-bindings.3
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man1/mergesolv.1
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man1/testsolv.1
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man1/installcheck.1
/opt/poky/2.0+snapshot/sysroots/x86_64-pokysdk-linux/usr/man/man1/dumpsolv.1
It checks ${CMAKE_INSTALL_PREFIX}/share/man when configure, but it may
not exist when crosscompile, for example, when CMAKE_INSTALL_PREFIX="/",
it is OK, but when CMAKE_INSTALL_PREFIX="/some/path/else", then it
doesn't exist, and the man dir would be set to "/usr/man" which is
incorrect.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ed Bartosh [Wed, 6 Apr 2016 17:00:42 +0000 (18:00 +0100)]
toaster: add DL_DIR and SSTATE_DIR to oe toasterconf
Add the DL_DIR and SSTATE_DIR to the toasterconf with defaults set
We now support per project sharing of the DL_DIR and SSTATE_DIR
in toaster.
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ed Bartosh [Thu, 7 Apr 2016 13:11:46 +0000 (14:11 +0100)]
toaster.bbclass: strip task from the target
Current code in toaster_buildhistory_dump assumes that bitbake
target doesn't contain task name. It uses target as a part of
path to the files with data that it analizes. It fails to find
files if target contains task name. Stripping task from the
target should solve this.
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Elliot Smith <elliot.smith@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jussi Kukkonen [Thu, 7 Apr 2016 11:49:44 +0000 (14:49 +0300)]
x11-common: Add PACKAGECONFIG for screen blanking
when "blank" is not in PACKAGECONFIG, explicitly set dpms and
screensaver off. The intention here is to make it easier to
create builds that should not blank the screen (just remove
"blank" from x11-common PACKAGECONFIG).
Also remove a unneeded "find -exec rm" from do_install.
Partly fixes [YOCTO #7278].
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Randy Witt [Thu, 7 Apr 2016 23:34:53 +0000 (16:34 -0700)]
devtool: Create unlocked-sigs.inc containing items in the workspace
When a recipe is added to the workspace, the signatures for the tasks
will change. This means that bitbake must be told to allow the
signatures to be different if they are in locked-sigs.inc.
This is done by creating an unlocked-sigs.inc file which contains all
the recipes in the workspace each time devtool reads the workspace.
So not only will necessary things get added, previously added items will
be removed by virtue of them no longer being in the workspace.
This also makes sure that the extensible sdk picks up unlocked-sigs.inc
as part of the configuration.
[YOCTO #9195]
Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Randy Witt [Thu, 7 Apr 2016 23:34:52 +0000 (16:34 -0700)]
sstatesig.py: Add a method to "unlock" recipes
In order to support workflows using devtool where a user might want to
modify tasks that exist in locked-sigs.inc, there must be a way to unlock
recipes.
This patch adds that support by allowing the user to add recipes to
SIGGEN_UNLOCKED_RECIPES. Recipes that exist in that variable will have
all their tasks unlocked, as well as any tasks that depend on that
recipe.
For example if foo->bar->baz, if you unlock baz, it will also unlock bar
so that foo can be rebuilt without explicitly specifying bar as being
unlocked.
[YOCTO #9195]
Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With the extensible sdk we want there to be an error if a task tries to
run without signatures that match locked-sigs.inc. This patch enables
that error.
[YOCTO #9195]
Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Randy Witt [Thu, 7 Apr 2016 23:34:49 +0000 (16:34 -0700)]
sstatesig.py: Split single locked sigs check into multiple checks
Add the SIGGEN_LOCKEDSIGS_TASKSIG_CHECK and
SIGGEN_LOCKEDSIGS_SSTATE_EXISTS_CHECK variables to replace
SIGGEN_LOCKEDSIGS_CHECK_LEVEL.
SIGGEN_LOCKEDSIGS_TASKSIG_CHECK will no control whether there is a
warning or error if a task's hash in the locked signature file doesn't match
the computed hash from the current metadata.
SIGGEN_LOCKEDSIGS_SSTATE_EXISTS_CHECK will control whther there is a
warning or error if a task that supports sstate is in the locked
signature file, but no sstate exists for the task.
Previously you could only have warning/errors for both controlled by
SIGGEN_LOCKEDSIGS_CHECK_LEVEL. This was an issue in the extensible sdk,
because we know sstate won't exist for certain items in the reverse
dependencies list for tasks. However, we still want to error if task
signatures don't match.
[YOCTO #9195]
Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
image_types: use compress framework to produce checksums for images
Existing compress framework for producing various compressed versions
of images is powerfull enough to be extended for other uses, e.g. to
convert types of images.
It is possible to use it also to produce image checksums at the time
of image generation. This commit adds support for all supported
at the moment coreutils hashing algorithms: md5, sha1, sha224,
sha256, sha384 and sha512
Usage:
IMAGE_FSTYPES_append = " hddimg.sha256sum"
Signed-off-by: Alexander D. Kanevskiy <kad@kad.name> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexandru Moise <alexandru.moise@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Wed, 6 Apr 2016 02:31:27 +0000 (19:31 -0700)]
create-pull-request: fix for newer git
Fixed when git > 2.1.0:
$ ./scripts/create-pull-request -r HEAD^ -u contrib -b rbt/git
fatal: Not a valid revision: rbt/git
ERROR: git request-pull reported an error
This is because newer git requires both local and remote branch named as
rbt/git, but usually, we only named the remote branch as rbt/foo, and
foo for local branch.
Add a option '-l' to fix the problem, default is HEAD.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
It hasn't been used since June 2015 when upgraded the recipe, this patch
was used for crosscompiling, now the crosscompile works well without it,
so remove it.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jussi Kukkonen [Wed, 6 Apr 2016 08:19:11 +0000 (11:19 +0300)]
ffmpeg: Make configure options explicit
Currently ffmpeg builds libraries that are packaged but never appear
in PACKAGES. Add PACKAGECONFIGs for all libraries, and make sure
PACKAGES_DYNAMIC lists all dynamic packages.
Default choice (build all except libavresample) does not change.
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Maxin B. John [Wed, 6 Apr 2016 09:20:31 +0000 (12:20 +0300)]
useradd.bbclass: remove user/group created by the package in clean* task
At present, if a recipe is built which creates users/groups via
useradd.bbclass, those users/groups are not removed from sysroot
when the recipe/package is cleaned using clean/cleansstate/cleanall
or when a recipe is rebuild and 'unstaged' from the the sysroot.
The "userdel_sysroot_sstate()" provides that functionality.
[YOCTO #9262]
Signed-off-by: Maxin B. John <maxin.john@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
scripts/oe-selftest: avoid the creation of coverage file when coverage not installed
Coverage subprocessing file is being created even when coverage
is not installed, which causes errors of "module not found" to
be send to the oe-selftest output.
This patch adds indent to the block of code creating this coverage
file, so it can only be executed when coverage is actually
installed.
[Yocto #9334]
Signed-off-by: Humberto Ibarra <humberto.ibarra.lopez@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
scripts/oe-selftest: remove coverage file if any coverage option is given
Coverage temporal file for sub-processing is being removed only
when the --coverage option was found. This is wrong since the
file is created when any coverage option (source, include or omit)
is given, even if --coverage is not one of them.
This patch makes sure to remove the file if any coverage option was
given.
Signed-off-by: Humberto Ibarra <humberto.ibarra.lopez@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There is a message that warns the user about enabling subprocessing
for coverage to work. After the fix for Yocto #8930,
this task is done automatically, so the warning is not needed
anymore.
Signed-off-by: Humberto Ibarra <humberto.ibarra.lopez@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Alex Franco [Tue, 5 Apr 2016 16:21:34 +0000 (11:21 -0500)]
patch.bbclass: remove useless path assignment
A path variable is assigned the value of PATH from the
environment, before the PATH is set on the environemnt
from the value of PATH in the bb datastore. This seems
to be an unnecessary leftover.
[YOCTO #8543]
Signed-off-by: Alex Franco <alejandro.franco@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Tue, 5 Apr 2016 16:10:43 +0000 (17:10 +0100)]
package: do_split_packages: expand variables in extra_depends
If a recipe passes for example extra_depends="${PN}-dev" to do_split_packages()
then it isn't expanded, so the check for the multilib prefix doesn't work.
Solve this centrally by expanding extra_depends inside do_split_packages().
[ YOCTO #9381 ]
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Tue, 5 Apr 2016 14:46:21 +0000 (15:46 +0100)]
sanity: Increase minimum git version to 1.8.3.1
The kernel tools assume git > 1.7.9.5, I'm unsure of the exact version but
the oldest in our infrastructure is 1.8.3.1. The git fetcher also currently
has nasty workarounds for git < 1.7.9.2.
Moving to 1.8.3.1 as our minimum version seems sane at this point as the oldest
we're testing/supporting.
[YOCTO #6162]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Markus Lehtonen [Tue, 5 Apr 2016 13:21:49 +0000 (16:21 +0300)]
externalsrc: avoid race in temporary git index file
Use a unique tempfile as the temporary git index file when determining
the git hash of the source tree. A fixed filename was obviously causing
races (with the git index.lock file) under oe-selftest where multiple
bitbake instances were run against the same source tree at the same
time.
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bill Randle [Mon, 4 Apr 2016 16:40:37 +0000 (09:40 -0700)]
bdwgc: use github repo for source location
The Yocto Autobuilder has been unable to fetch the release package
from the developer's website at www.hboehm.info, so change recipe
to fetch from the developer's github repo instead.
Signed-off-by: Bill Randle <william.c.randle@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Mon, 4 Apr 2016 13:41:42 +0000 (15:41 +0200)]
package_manager.py: better error handling in opkg's package listing
opkg does not return a non-zero exit code even if it found
errors. When that happens, parsing the output leads to strange
follow-up errors.
To avoid this we need to check explicitly for non-empty
stderr. Reporting only that on a failure also leads to shorter error
messages (stdout may be very large).
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Mon, 4 Apr 2016 13:41:41 +0000 (15:41 +0200)]
systemd: make systemd-serialgetty optional
Some distros may prefer to use the upstream systemd support for
starting getty on serial ports. This is now possible by adding
"serial-getty-generator" to PACKAGECONFIG.
The default is unchanged, i.e. systemd's own serial-getty@.service
file does not get packaged and instead systemd-serialgetty is pulled
into images via RRECOMMENDS.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Mon, 4 Apr 2016 13:41:40 +0000 (15:41 +0200)]
ncurses: reorder PACKAGES
Having ncurses-terminfo-base before ncurses-terminfo is currently irrelevant
because the current file lists are completely disjunct. However, when building
"stateless" via a .bbappend, the content of curses-terminfo-base also needs
to live under /usr/share and then it becomes important that files
belonging to ncurses-terminfo-base are checked first.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 4 Apr 2016 15:16:42 +0000 (16:16 +0100)]
buildtools-tarball: Add texinfo (for makeinfo)
Initially I was reluctant to do this however makeinfo is a dependency
of the buildsystem and only adds around 400kb to the buildtools-tarball
so it likely makes sense to add it. This allows people to use
the project on older environments.
Need to enable nativesdk-texinfo but this seems straightforward.
[YOCTO #8990]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Mon, 4 Apr 2016 13:28:55 +0000 (14:28 +0100)]
cogl: fix G-I .typelib installation
By default gobject-introspection puts .typelib files into $PN, but the cogl-1.0
recipe doesn't use PN. Installing cogl-1.0 is impossible as libcogl
conflicts/replaces with cogl-1.0 due to a previous upgrade.
Fix this by putting the .typelib files into the right subpackages, and
explicitly setting FILES_${PN} to "" so that nothing accidently ends up in PN.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Mon, 4 Apr 2016 05:02:20 +0000 (17:02 +1200)]
classes/buildhistory: fix filtering of depends-nokernel.dot
For images we produce a number of filtered dependency .dot files for
readability, the first of which is depends-nokernel.dot which filters
out just the kernel itself (not kernel modules). Unfortunately the
filter specifications hadn't been updated for the dash-to-underscore
removal or the 4.x kernel upgrade, thus the filtering wasn't actually
doing anything.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Mon, 4 Apr 2016 05:02:19 +0000 (17:02 +1200)]
classes/buildhistory: optimise getting package size list
Invoking oe-pkgdata-util in turn for every package in the list was slow
with a large image. Modify oe-pkgdata-util's read-value command to take
an option to read the list of packages from a file, as well as prefix
the value with the package name; we can then use this to get all of the
package sizes at once. This reduces the time to gather this information
from minutes to just a second or two.
Fixes [YOCTO #7339].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sun, 3 Apr 2016 09:09:49 +0000 (10:09 +0100)]
populate_sdk_ext: Exclude BBTASKDEPDATA from task signature
Otherwise the task hash changes between server and worker context
leading to changing task checksums. The dependency data here is
tracked by other pieces of the signature.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Sat, 2 Apr 2016 00:09:26 +0000 (20:09 -0400)]
kernel-yocto: enforce SRC_URI specified branch
During the simplication and cleanup of branches and kernel meta data
handling, the ability to force build a branch that didn't match the
meta data was dropped.
There are valid uses cases when a different branch should be built
(testing, development, etc), so we restore the capability with this
change.
If after the kernel meta data is processed the current branch does
not match the SRC_URI specified branch, a warning is generated
about the impending branch switch and that the user should double
check that they are building what they expect.
WARNING: After meta data application, the kernel tree branch is standard/base. The
WARNING: SRC_URI specified branch standard/gt. The branch will be forced to standard/gt,
WARNING: but this means the board meta data (.scc files) do not match the SRC_URI specification.
WARNING: The meta data and branch standard/gt should be inspected to ensure the proper
WARNING: kernel is being built.
Reported-by: Steve Sakoman" <steve.sakoman@intel.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Sat, 2 Apr 2016 00:09:25 +0000 (20:09 -0400)]
linux-yocto/4.4: UVC: Add support for R200 depth camera
Integrating the following commit:
[
Add support for Intel R200 depth camera in uvc driver.
This includes adding new uvc GUIDs for the new pixel formats,
adding new V4L pixel format definition to user api headers,
and updating the uvc driver GUID-to-4cc tables with the new formats.
Bruce Ashfield [Sat, 2 Apr 2016 00:09:24 +0000 (20:09 -0400)]
linux-yocto/4.4: fix PAT for 32bit x86
Integrating the following mainline destined commits to fix PAT issues with
32bit x86 and qemu:
bc22b90fb583 x86/pat: Document the PAT initialization sequence c534b1ec1991 x86/xen, pat: Remove PAT table init code from Xen cd47692e804f x86/mtrr: Fix PAT init handling when MTRR is disabled 039434bdc165 x86/mtrr: Fix Xorg crashes in Qemu sessions c08196e8064c x86/mm/pat: Replace cpu_has_pat with boot_cpu_has() 6928fce8c766 x86/mm/pat: Add pat_disable() interface 3163c8d5468d x86/mm/pat: Add support of non-default PAT MSR setting
With this change applied, we once again have working graphics and no special
work arounds.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Sat, 2 Apr 2016 00:09:21 +0000 (20:09 -0400)]
linux-yocto/4.1: ahci: backport AHCI runtime PM
enabling AHCI runtime PM support via the following commits:
f5b08e95bc03 ahci: Add runtime PM support for the host controller e05e05973f39 ahci: Add functions to manage runtime PM of AHCI ports 558571930e2d ahci: Convert driver to use modern PM hooks 4a2c2074f209 ahci: Cache host controller version 21641413456c scsi: Drop runtime PM usage count after host is added 254ab69f652d scsi: Set request queue runtime PM status back to active on resume 61c02cb167ed block: Add blk_set_runtime_active() 7bcfbac98f3c ahci: Store irq number in struct ahci_host_priv a5a093b7f6e4 ahci: Move interrupt enablement code to a separate function
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Sat, 2 Apr 2016 00:09:20 +0000 (20:09 -0400)]
linux-yocto/4.4: gpio-pca953x: add PCAL9535 interrupt support
Reverting two previous gpio commits to allow the application of proper
upstream ready patches to add this support:
08943f2bbd50 gpio-pca953x: add "drive" property a517d5b72e76 gpio: pca953x: provide GPIO base based on _UID c91063f44368 gpio-pca953x: add PCAL9535 interrupt support 58f3c9f0ac6d Revert "gpio-pca953x: add "drive" property" 7abbd5fec15d Revert "gpio: pca953x: provide GPIO base based on _UID"
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
wic/isoimage-isohybrid.py: change cpio generated uid&gid to root
By default cpio preserves the uid&guid's of the original user which
leads to host contamination and boot failures because commands like
mount from initramfs expect to be run by root and the original host
user might not even exist on the target.
Signed-off-by: Ioan-Adrian Ratiu <adrian.ratiu@ni.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
