| Age | Commit message (Collapse) | Author |
|---|
|
If the datastore has a 'hash' method, it will use that, otherwise it uses its
own implementation. Datastore hashing will be added to bitbake master at some
point, to prepare for the move away from stamps.
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
|
|
|
|
Acked-by: Graeme Gregory <dp@xora.org.uk>
Acked-by: Philip Balister <philip@balister.org>
Signed-off-by: Koen Kooi <koen@openembedded.org>
|
|
breaks the build (from poky)
|
|
|
|
|
|
Doing a '-c clean' operation on a staged package with very much files
(e.g. glibc) took several minutes because
* every removed file was reported
* an 'rm' instance was spawned for every file
This patch uses the native 'os.unlink()' method for removing files and
reports only the removed root directory instead of the single files.
Based upon maillist discussion, reporting happens with 'debug' level
instead of 'note' one, and only error conditions due to non-existing
files will be ignored. Other (e.g. permission denied) errors will now
abort the build while they were silently ignored previously.
Signed-off-by: Enrico Scholz <enrico.scholz@sigma-chemnitz.de>
Signed-off-by: Enrico Scholz <enrico.scholz@sigma-chemnitz.de>
Acked-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
Acked-by: Tom Rini <tom_rini@mentor.com>
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
Acked-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
Acked-by: Roman I Khimov <khimov@altell.ru>
|
|
Recipes are using update-rc.d to do postinst and postinst scripts are being
ran on build hosts where native version is needed to run properly. It is
needed only at image build stage, but it's easier to have it in DEPENDS,
build time for it is almost zero anyway.
Signed-off-by: Roman I Khimov <khimov@altell.ru>
|
|
* show note, when there are checksums only in checksums.ini (prepare for
script for moving all to recipes)
* parse checksums.ini only when there is no checksum in recipe (could be
faster, but for more checked items in SRC_URI it is parsed repeatedly)
* if one checksum doesn't match then count and show both (md5 as well as
sha256) - usefull for copy&paste checksums for new recipe.
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
|
|
There are scripts that can and should work when being ran on build
host (for example, simple update-rc.d), but there are also many which
can't and won't ever (for example, anything adding users/groups).
The second group sometimes doesn't get "unpacked" flag because
scripts throw errors in the middle and return something nice from
the last command.
It can be considered as a bug in pre/postinst script (as it should
either explicitly check for "${D}" or just do "set -e" at start),
but it is common enough.
There is also another aspect to this as in general we can't be sure
that everything is OK wrt preinst/postinst if script commands throw
errors.
Running preinst/postinst scripts on host with "-e" should solve
that.
Signed-off-by: Roman I Khimov <khimov@altell.ru>
Acked-by: Chris Larson <clarson@kergoth.com>
|
|
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
Signed-off-by: Vladimir Sorokin <sorokin@altell.ru>
Signed-off-by: Roman I Khimov <khimov@altell.ru>
|
|
|
|
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
This changes insane.bbclass to use s.endswith("str") rather than s[-3:] ==
"str". The primary motivation for this is that we have tests that check for
'la' and 'pc' files when we really want '.la' and '.pc' to be what we check
for. Also, our -dev symlink test does if test1: if test2 and test3: action,
when it could just be if test1 and test2 and test3.
Signed-off-by: Tom Rini <tom_rini@mentor.com>
Acked-by: Khem Raj <raj.khem@gmail.com>
Acked-by: Marcin Juszkiewicz <marcin@juszkiewicz.com.pl>
|
|
|
|
|
|
|
|
from env
|
|
|
|
|
|
|
|
|
|
Based on work done by Matthew Dombroski.
We want to generate a toolchain file for cmake compilations. This
adds a task that is performed before the configure task.
Additional defines can be made in the toolchain file directly in
the recipe file by using a do_generate_toolchain_file_append().
This is especially interesting for recipes that need to define
values for additionnal libraries.
Signed-off-by: Valentin Longchamp <valentin.longchamp@epfl.ch>
|
|
The problem occured when unzip-native is not yet staged, and ZIP
archive unpacking already started resulting in failed do_unpack task.
Added NEED_UNZIP_FOR_UNPACK variable to use in recipes which do_unpack
with unzip utility but doesn't have '.zip' in SRC_URI (f.e. .EXE
windows self-extraction binaries).
Signed-off-by: Ihar Hrachyshka <ihar.hrachyshka@gmail.com>
Signed-off-by: Koen Kooi <koen@openembedded.org>
|
|
This code has never been used, is only half complete and obsolete. If
anyone ever wnats to refer to it, its in the history but keeping
useless code ist just confusing so delete it.
Acked-by: Chris Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
|
|
Converted most of python site-packages references to PYTHON_SITEPACKAGES_DIR.
Signed-off-by: Roman I Khimov <khimov@altell.ru>
|
|
Allows to inherit that in packages containing Python extensions but still tightly
control packaging of those extensions. Mainly for the cases where you want
those python modules/extensions/parts to be packaged separately from the main
package in order not to introduce python dependency where you don't need it.
Technically, you can do it with distutils-common-base, but it might require
more FILES_* work than with python-dir.
Introduce PYTHON_SITEPACKAGES_DIR along the way, site-packages directory is
referenced frequently enough within class files, but some recipes also pack this
directory as
${libdir}/python2.5/site-packages
${libdir}/python2.6/site-packages
${libdir}/python*/site-packages
/usr/lib/python*/site-packages
${libdir}/*/site-packages
all of which are not perfect.
Signed-off-by: Roman I Khimov <khimov@altell.ru>
|
|
This reverts commit e7c6b93906b8909e659d7e680a1457472bc99bda.
Sorry for breakage.
Signed-off-by: Roman I Khimov <khimov@altell.ru>
|
|
|
|
In commit 89b7e433719f43f1c36c76cb8856d559014e99bc metadata_scm.bbclass
functions were factored out from base.bbclass, but the new file was not
included.
This prevented METADATA_SCM, METADATA_REVISION, METADATA_BRANCH to be
set; for instance this caused oestats-client stop working.
Signed-off-by: Antonio Ospite <ospite@studenti.unina.it>
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
Signed-off-by: Roman I Khimov <khimov@altell.ru>
|
|
Signed-off-by: Roman I Khimov <khimov@altell.ru>
|
|
Makes perl modules rebuild on perl version change, since binary compatibility
could be broken with that (although perl itself tries to maintain it within
patchlevel range, like 5.8.7 -> 5.8.8).
This is conditional on PACKAGE_ARCH, since there is nothing to rebuild for
pure perl modules (and those should use 'PACKAGE_ARCH = "all"') even if they
break with perl version change.
This also should ease transition to perl 5.10.1.
Signed-off-by: Roman I Khimov <khimov@altell.ru>
|
|
Using an 'rm -rf $dir' ordering is more portable.
Signed-off-by: Enrico Scholz <enrico.scholz@informatik.tu-chemnitz.de>
Acked-by: Roman I Khimov <khimov@altell.ru>
Signed-off-by: Holger Hans Peter Freyther <zecke@selfish.org>
|
|
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
Provides oe_popen, which is a subprocess.Popen wrapper that automatically
provides our exported variables in the environment, including the PATH, and
oe_system, which is just a wrapper that acts like system.
Signed-off-by: Chris Larson <clarson@mvista.com>
Acked-by: Tom Rini <tom_rini@mentor.com>
|
|
By default cmake guesses whether install shared libaries with exec
permissions or not based on host distribution, so this behaviour
changes from host to host. OpenEmbedded assumes shared libraries to
be with exec permissions (reverse way brakes stripping, for example),
so let's force this in cmake configure invocation.
Signed-off-by: Roman I Khimov <khimov@altell.ru>
Acked-by: Ash Charles <ashcharles@gmail.com>
Acked-by: Khem Raj <raj.khem@gmail.com>
|
|
Obsolete, not used.
Signed-off-by: Roman I Khimov <khimov@altell.ru>
Acked-by: Graeme Gregory <dp@xora.org.uk>
Acked-by: Koen Kooi <koen@openembedded.org>
Acked-by: Frans Meulenbroeks <fransmeulenbroeks@gmail.com>
|
|
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
Acked-by: Richard Purdie <rpurdie@linux.intel.com>
Acked-by: Marcin Juszkiewicz <marcin@juszkiewicz.com.pl>
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
Acked-by: Richard Purdie <rpurdie@linux.intel.com>
Acked-by: Marcin Juszkiewicz <marcin@juszkiewicz.com.pl>
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
packagedstaging_fastpath() was only copying the contents of CROSS_DIR to
PSTAGE_TMPDIR resulting in the folders contents being packaged and then
installed incorrectly at the top level of CROSS_DIR rather than in HOST_ARCH
specific sub directories.
This patch fixes that issue by copying the directory and its contents rather
than just the directory contents.
Signed-off-by: Joshua Lock <josh@linux.intel.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
|
|
|
|
that we don't have other threads try and validate or install based on them
being valid. This is a potential issue for certain BB_STAMP_POLICY (such as
the default) values and enough concurrent threads.
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
Signed-off-by: Adrian Alonso <aalonso00@gmail.com>
Signed-off-by: Stefan Schmidt <stefan@datenfreihafen.org>
|
Chris Larson
Koen Kooi
koen.kooi@gmail.com
Enrico Scholz
Roman I Khimov
Martin Jansa
Vladimir Sorokin
Xerxes Rånby
Tom Rini
Michael 'Mickey' Lauer
Valentin Longchamp
Ihar Hrachyshka
Richard Purdie
Antonio Ospite
Enrico Scholz
Christopher Larson
Joshua Lock
Adrian Alonso