Age | Commit message (Collapse) | Author | Files |
|
directory
Currently we have a hierarchy of pkgdata directories and the code has to put together
a search path and look through each in turn until it finds the data it needs.
This has lead to a number of hardcoded paths and file globing which
is unpredictable and undesirable. Worse, certain tricks that should be
easy like a GL specific package architecture become problematic with the
curretn search paths.
With the modern sstate code, we can do better and construct a single pkgdata
directory for each machine in just the same way as we do for the sysroot. This
is already tried and well tested. With such a single directory, all the code that
iterated through multiple pkgdata directories and simply be removed and give
a significant simplification of the code. Even existing build directories adapt
to the change well since the package contents doesn't change, just the location
they're installed to and the stamp for them.
The only complication is the we need a different shlibs directory for each
multilib. These are only used by package.bbclass and the simple fix is to
add MLPREFIX to the shlib directory name. This means the multilib packages will
repackage and the sstate checksum will change but an existing build directory
will adapt to the changes safely.
It is close to release however I believe the benefits this patch give us
are worth consideration for inclusion and give us more options for dealing
with problems like the GL one. It also sets the ground work well for
shlibs improvements in 1.6.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* usually it's more important to know how much space will each
package take on target device then size of compressed package
* example for libewebkit0 with 4 different architectures, interesting
that om_gta02 .ipk is bigger but it's smaller when installed
before:
MACHINE DEFAULTTUNE SIZE (.ipk file)
om_gta04 cortexa8t-neon 15996 KiB libewebkit0
qemux86_64 x86-64 16992 KiB libewebkit0
spitz xscale 16148 KiB libewebkit0
om_gta02 arm920t 16260 KiB libewebkit0
after:
MACHINE DEFAULTTUNE SIZE (installed)
om_gta04 cortexa8t-neon 60544 KiB libewebkit0
qemux86_64 x86-64 63720 KiB libewebkit0
spitz xscale 60588 KiB libewebkit0
om_gta02 arm920t 56268 KiB libewebkit0
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Check after getting the original package name (e.g. undoing Debian
renaming) if there is a complementary package for that name, e.g. if
the glob is *-dev, then libudev0 -> libudev -> libudev-dev.
Fixes [YOCTO #4136].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Check for .packaged file and exclude packages if this file does not
exist - this avoids attempting to install empty packages during
complementary package installation within do_rootfs that didn't end up
being created (and failing as a result).
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Use a similar mechanism that was previously used to install locales at
rootfs generation time to install other "complementary" packages (e.g.
*-dev packages) - i.e. install all of the explicitly requested packages
and their dependencies, then get a list of the packages that were
installed, and use that list to install the complementary packages. This
has been implemented by using a list of globs which should make it
easier to extend in future.
The previous locale package installation code assumed that the locale
packages did not have any dependencies that were not already installed;
now that we are installing non-locale packages this is no longer
correct. In practice only the rpm backend actually made use of this
assumption, so it needed to be changed to call into the existing package
backend code to do the complementary package installation rather than
calling rpm directly.
This fixes the doc-pkgs IMAGE_FEATURES feature to work correctly, and
also ensures that all dev/dbg packages get installed for
dev-pkgs/dbg-pkgs respectively even if the dependency chains between
those packages was not ensuring that already.
The code has also been adapted to work correctly with the new
SDK-from-image functionality. To that end, an SDKIMAGE_FEATURES variable
has been added to allow specifying what extra image features should go
into the SDK (extra, because by virtue of installing all of the packages
in the image into the target part of the SDK, we already include all of
IMAGE_FEATURES) with a default value of "dev-pkgs dbg-pkgs".
Fixes [YOCTO #2614].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|