| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
| |
Adds development shell support for out-of-tree kernel modules by reproducing
the build environment of the compile task.
(From OE-Core rev: 08e366a386caab547e0dbcad2d4ee08f44814262)
Signed-off-by: Stefan Wiehler <stefan.wiehler@missinglinkelectronics.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
| |
Replace hardcoded '/lib' in kernel modules installation path with
${nonarch_base_libdir}, which is meant exactly for this.
(From OE-Core rev: 22f5ba7154fcbe826d0a3283740903312b2aab46)
Signed-off-by: Amarnath Valluri <amarnath.valluri@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
| |
This is needed for the make_scripts task.
(From OE-Core rev: 554672d5f649eaf6a7d76f79520304415d926a3e)
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>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch is comparatively large and invasive. It does only do one thing, switching the
system to build using recipe specific sysroots and where changes could be isolated from it,
that has been done.
With the current single sysroot approach, its possible for software to find things which
aren't in their dependencies. This leads to a determinism problem and is a growing issue in
several of the market segments where OE makes sense. The way to solve this problem for OE is
to have seperate sysroots for each recipe and these will only contain the dependencies for
that recipe.
Its worth noting that this is not task specific sysroots and that OE's dependencies do vary
enormously by task. This did result in some implementation challenges. There is nothing stopping
the implementation of task specific sysroots at some later point based on this work but
that as deemed a bridge too far right now.
Implementation details:
* Rather than installing the sysroot artefacts into a combined sysroots, they are now placed in
TMPDIR/sysroot-components/PACKAGE_ARCH/PN.
* WORKDIR/recipe-sysroot and WORKDIR/recipe-sysroot-native are built by hardlinking in files
from the sysroot-component trees. These new directories are known as RECIPE_SYSROOT and
RECIPE_SYSROOT_NATIVE.
* This construction is primarily done by a new do_prepare_recipe_sysroot task which runs
before do_configure and consists of a call to the extend_recipe_sysroot function.
* Other tasks need things in the sysroot before/after this, e.g. do_patch needs quilt-native
and do_package_write_deb needs dpkg-native. The code therefore inspects the dependencies
for each task and adds extend_recipe_sysroot as a prefunc if it has populate_sysroot
dependencies.
* We have to do a search/replace 'fixme' operation on the files installed into the sysroot to
change hardcoded paths into the correct ones. We create a fixmepath file in the component
directory which lists the files which need this operation.
* Some files have "postinstall" commands which need to run against them, e.g. gdk-pixbuf each
time a new loader is added. These are handled by adding files in bindir with the name
prefixed by "postinst-" and are run in each sysroot as its created if they're present.
This did mean most sstate postinstalls have to be rewritten but there shouldn't be many of them.
* Since a recipe can have multiple tasks and these tasks can run against each other at the same
time we have to have a lock when we perform write operations against the sysroot. We also have
to maintain manifests of what we install against a task checksum of the dependency. If the
checksum changes, we remove its files and then add the new ones.
* The autotools logic for filtering the view of m4 files is no longer needed (and was the model
for the way extend_recipe_sysroot works).
* For autotools, we used to build a combined m4 macros directory which had both the native and
target m4 files. We can no longer do this so we use the target sysroot as the default and add
the native sysroot as an extra backup include path. If we don't do this, we'd have to build
target pkg-config before we could built anything using pkg-config for example (ditto gettext).
Such dependencies would be painful so we haven't required that.
* PKDDATA_DIR was moved out the sysroot and works as before using sstate to build a hybrid copy
for each machine. The paths therefore changed, the behaviour did not.
* The ccache class had to be reworked to function with rss.
* The TCBOOTSTRAP sysroot for compiler bootstrap is no longer needed but the -initial data
does have to be filtered out from the main recipe sysroots. Putting "-initial" in a normal
recipe name therefore remains a bad idea.
* The logic in insane needed tweaks to deal with the new path layout, as did the debug source
file extraction code in package.bbclass.
* The logic in sstate.bbclass had to be rewritten since it previously only performed search and
replace on extracted sstate and we now need this to happen even if the compiled path was
"correct". This in theory could cause a mild performance issue but since the sysroot data
was the main data that needed this and we'd have to do it there regardless with rss, I've opted
just to change the way the class for everything. The built output used to build the sstate output
is now retained and installed rather than deleted.
* The search and replace logic used in sstate objects also seemed weak/incorrect and didn't hold
up against testing. This has been rewritten too. There are some assumptions made about paths, we
save the 'proper' search and replace operations to fixmepath.cmd but then ignore this. What is
here works but is a little hardcoded and an area for future improvement.
* In order to work with eSDK we need a way to build something that looks like the old style sysroot.
"bitbake build-sysroots" will construct such a sysroot based on everything in the components
directory that matches the current MACHINE. It will allow transition of external tools and can
built target or native variants or both. It also supports a clean task. I'd suggest not relying on
this for anything other than transitional purposes though. To see XXX in that sysroot, you'd have
to have built that in a previous bitbake invocation.
* pseudo is run out of its components directory. This is fine as its statically linked.
* The hacks for wayland to see allarch dependencies in the multilib case are no longer needed
and can be dropped.
* wic needed more extensive changes to work with rss and the fixes are in a separate commit series
* Various oe-selftest tweaks were needed since tests did assume the location to binaries and the
combined sysroot in several cases.
* Most missing dependencies this work found have been sent out as separate patches as they were found
but a few tweaks are still included here.
* A late addition is that extend_recipe_sysroot became multilib aware and able to populate multilib
sysroots. I had hoped not to have to add that complexity but the meta-environment recipe forced my
hand. That implementation can probably be neater but this is on the list of things to cleanup later
at this point.
In summary, the impact people will likely see after this change:
* Recipes may fail with missing dependencies, particularly native tools like gettext-native,
glib-2.0-native and libxml2.0-native. Some hosts have these installed and will mask these errors
* Any recipe/class using SSTATEPOSTINSTFUNCS will need that code rewriting into a postinst
* There was a separate patch series dealing with roots postinst native dependency issues. Any postinst
which expects native tools at rootfs time will need to mark that dependency with PACKAGE_WRITE_DEPS.
There could well be other issues. This has been tested repeatedly against our autobuilders and oe-selftest
and issues found have been fixed. We believe at least OE-Core is in good shape but that doesn't mean
we've found all the issues.
Also, the logging is a bit chatty at the moment. It does help if something goes wrong and goes to the
task logfiles, not the console so I've intentionally left this like that for now. We can turn it down
easily enough in due course.
(From OE-Core rev: 809746f56df4b91af014bf6a3f28997d6698ac78)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Requiring all build systems for external kernel modules to
place Module.symvers directly into ${B}/. is quite an
artificial assumption/requirement. It doesn't have to end
up there, but could easily end up somewhere below ${B}.
Allow to override the location below ${B}
Note that we still don't make use of multiple
Module.symvers in case there are any generated by one
single kernel module build, but this is no change in
behaviour from before and could be added if there really
is a demand for that.
Reported-by: Denys Dmytriyenko <denis@denix.org>
(From OE-Core rev: caa0fa2ddf0c97255b38b1ec8579944ab4821ff1)
Signed-off-by: André Draszik <adraszik@tycoint.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
getVar() now defaults to expanding by default, thus remove the True
option from getVar() calls with a regex search and replace.
Search made with the following regex: getVar ?\(( ?[^,()]*), True\)
(From OE-Core rev: 7c552996597faaee2fbee185b250c0ee30ea3b5f)
Signed-off-by: Joshua Lock <joshua.g.lock@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When compiling multiple external kernel modules, where one
depends on the other, there are two problems at the
moment:
1) we get compile time warnings from the kernel build
system due to missing symbols (from modpost).
2) Any modules generated are missing dependency
information (in the .modinfo elf section) for any
dependencies outside the current source tree and
outside the kernel itself.
This is expected, but the kernel build system has a way to
deal with this - the dependent module is expected to
specify KBUILD_EXTRA_SYMBOLS (as a space-separated list)
to point to any and all Module.symvers of kernel modules
that are dependencies.
While 1) by itself is not really a big issue, 2) prevents
the packaging process from generating cross-source tree
package dependencies.
As a first step to solve the missing dependencies in
packages created, we:
1) install Module.symvers of all external kernel module
builds (into a location that is automatically packaged
into the -dev package)
2) make use of KBUILD_EXTRA_SYMBOLS and pass the location
of all Module.symvers of all kernel-module-* packages
we depend on
This solves both problems mentioned above.
(From OE-Core rev: 88f1bc77c22091fccb00e80839adfdf34534187f)
Signed-off-by: André Draszik <git@andred.net>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Quite a few external kernel modules I've found floating around don't
have a modules_install target, but they do have an install target that
basically differs only in name. To make it easier to build these just
make this a MODULES_INSTALL_TARGET variable that you can set from the
recipe - the alternative would be copy-and-paste the do_install
definition from this class which is potentially fragile.
(From OE-Core rev: effa6ce777540c5557e5cf904b48cc3369ee3f9f)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If we have DEPENDS = "virtual/kernel" is means that the kernel module
depends on the kernel's do_populate_sysroot task. This is not entirely
desireable since that depends on do_install which depends on
do_compile_kernelmodules and so on. In a situation where rm_work in involved
this can cause some pretty length build cycles after the kernel workdir
has been cleaned up by rm_work.
As well as removing this, take the opportunity to clean up duplicated
dependency lines, tweak the dependency of make_scripts for the same
reason and generally try and make things more readable.
(From OE-Core rev: a641247d262971db2d44815c5a668e7e50bdaf4f)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
commit 3b3f7e785e279 [kernel: Rearrange for 1.8] began the process of
moving the kernel source and build artefacts out of sstate control and
into a shared location.
This changed triggered some workflow issues, as well as bugs related
to the kernel source containing build output, and hence being dirty and
breaking kernel rebuilds.
To solve these issues, and to make it clear that the kernel is not under
sstate control, we move the source and build outputs to:
work-shared/MACHINE/kernel-source
work-shared/MACHINE/kernel-build-artifacts
Where kernel-build-artifacts is the kernel build output and
kernel-source is kept "pristine". The build-artifacts contain everything
that is required to build external modules against the kernel source,
and includes the defconfig, the kernel-abiversion, System.map files and
output from "make scripts".
External module builds should either pass O= on the command line, or
set KBUILD_OUTPUT to point to the build-artifacts. module-base.bbclass
takes care of setting KBUILD_OUTPUT, so most existing external module
recipes are transparently adapted to the new source/build layout.
recipes that depend on the kernel source must have a depedency on the
do_shared_workdir task:
do_configure[depends] += "virtual/kernel:do_shared_workdir"
With this dependency added, the STAGING_KERNEL_DIR will be populated and
available to the rest of the build.
(From OE-Core rev: 6a1ff0e7eacef595738f2fed086986fd622ec32a)
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When the sstate hash changes for do_configure task, the do_configure
default implementation triggers the 'clean' to be run. For it to
succeed we need to have KERNEL_SRC defined in EXTRA_OEMAKE. Fixes
following error:
,----
| DEBUG: Executing shell function do_configure
| NOTE: make -e MAKEFLAGS= clean
| make -C M=.../tmp/work/... clean
| make[1]: *** M=.../tmp/work/...: No such file or directory. Stop.
| Makefile:20: recipe for target 'clean' failed
| make: *** [clean] Error 2
| ERROR: oe_runmake failed
`----
(From OE-Core rev: bc0f58f0713ea7db3c4c24a7d321435934d745d7)
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* this way we can reuse the same functionality also for external modules
including module_autoload_foo and module_conf_foo functionality
* MODULE_PACKAGES variable was removed (splited modules are now returned
by do_split_packages
* KERNEL_MODULES_META_PACKAGE is used to append all splitted packages
to RDEPENDS. In kernel.bbclass it's old "kernel-modules" in
module.bbclass it defaults to ${PN} for upgrade path from
single PN with all modules to PN depending on all new kernel-module-*
(From OE-Core rev: 51928b6b5ca0a46a9dcd754483a19af58b95fa18)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously the build path to STAGING_KERNEL_DIR was being embedded into the
package post install scripts. We avoid this behavior by generating a special
depmodwrapper script. This script contains that hard-coded path, ensuring
that re-use of the sstate-cache (and/or packages) will always run through the
wrapper generated by the current build with a checksum that includes
STAGING_KERNEL_DIR.
[ YOCTO #3962 ]
(From OE-Core rev: b18c61bae4d7161c087a004bba3c696006f7a2f6)
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch will allow recipes that provide kernel modules to package
the module or modules in specific packages. That list is contained in
MODULE_PACKAGES, this defaults to to preserve the current behavior.
The package can also define MODULE_FILES to specify files.
[YOCTO #3803]
(From OE-Core rev: c1ff0467bf03a3342846f0d9dde74e34b740798f)
(From OE-Core rev: 977aee43868499ab87a098f3798e90d6978836b9)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The code in module.bbclass was appending the pkg_postinst and
pkg_prerm to all packages that are part of a given recipe, meaning
that the -lic, -dev, -doc, ... packages all got the scriptlet
This change uses only which macthes with the RDEPENDS and FILES
already used in module.bbclass.
The failure was that rootfs creation would fail due to the -lic package
being installed before the kernel and the script would fail.
[YOCTO #3803]
(From OE-Core rev: cf05c4578c99c0cb885cf2706f7f2b39b100aeb8)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
update-modules is obsolete. The bbclass was updated not to use it
anymore.
[YOCTO #3598]
(From OE-Core rev: 6fafbf71adb8b34211c2bc24226f0b1eb206a7d3)
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
If do_make_scripts() executes before do_unpack()/do_patch(), the build
fails because it can't cd into the workdir of a recipe using this
class, so make sure do_make_scripts() doesn't run before the package
has been unpacked and patched.
Fixes [YOCTO #3589].
(From OE-Core rev: 824cf145bcb55bb99a717a2dfd73e43e6b3feea4)
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
| |
It's sometimes useful to have this function available to recipes which
don't wish to use module.bbclass for whatever reason.
(From OE-Core rev: 7632b44e7f487180811d47fbe9c29aa8e58868a2)
Signed-off-by: Phil Blundell <pb@pbcl.net>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The update-modules mechanism is something of a historical relic and it isn't
entirely clear that it has a great deal of value nowadays. Also, it causes a
problem when building a read-only rootfs since update-modules itself refuses
to configure offline.
Allow DISTROs to circumvent this whole thing by declaring (via DISTRO_FEATURES)
that they don't wish to use update-modules. This is backfilled for existing
distributions and will have to be marked as CONSIDERED by those who actually
don't want it.
(From OE-Core rev: 14bf8ed115453077b4d4042b4b70ed6b3bca2a9f)
Signed-off-by: Phil Blundell <pb@pbcl.net>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* depmod already gets executed by pkg_postinst_kernel-image.
* If you build a module using module.bbclass, pkg_postinst returns 1 in
do_rootfs, causing pkg_postinst to run again on first boot. To improve
this situation, I copied pkg_postinst from kernel.bbclass to module.bbclass.
This was rejected by Koen, because he doesn't like the code from
kernel.bblcass, which uses ${STAGING_DIR_KERNEL}. Richard then suggested
that calling depmod during do_rootfs wasn't necessary at all, because
it already gets done by kernel-image.
(From OE-Core rev: c7809c03080925b5e9171df5c9175c7c6420b376)
Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
When external modules are built, files in $STAGING_KERNEL_DIR/scripts/basic will/can get
rebuilt.
This raises a potential race condition. Prevent this by adding a lock around the
do_make_scripts() function. Further, make sure that the kernel has been installed
to the sysroot, prior to executing this new task.
(From OE-Core rev: 8681b82e8b466929205edde7ba479f3ac1a6143e)
Signed-off-by: Anders Darander <anders@chargestorm.se>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
out-of-tree modules
The existing infrastructure uses an external build tree which references the
kernel source in the work dir. If run with rm work, building external modules
will fail.
This patch places a configured source tree in sysroots. Striking a balance
between minimal size and minimal maintenance is difficult. A fully configured
tree is about 500MB after a clean. This version leans on the side of caution and
removes only the obviously unecessary parts of the source tree to conserve
space, resulting in about 170MB. The arch directories would be some additional
pruning we could do. Given examples from the devel package from distributions, I
suspect this size could be reduced to 75MB or so, but at the cost of a much more
complex recipe which is likely to require a great deal more maintenance to keep
current with kernel releases.
Care is also taken to clean the hostprogs in scripts, and the modules are
responsible for building them as needed. Although it is unclear to me if this is
really necessary, especially considering that modules put these bits back as
soon as they compile. If we are not generating an sstate package, I suspect we
can ignore these.
Please try this with your modules and let me know how it does. I tried to take
non linux-yocto kernel recipes into account, but I have only tested with
linux-yocto and the hello-mod recipe so far.
(From OE-Core rev: a9d41062e24a6b99661b3a5256f369b557433607)
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Acked-by: Koen Kooi <koen@dominion.thruhere.net>
Acked-by: Gary Thomas <gary@mlbassoc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This fixes [BUGID #241]
The kernel hostprogs are built for the host architecture. They should not be
deployed to the target, and they should not be included in an sstate package
which might get reused on a host of a different architecture.
As we don't build many out-of-tree modules, this patch takes the approach of
building the hostprogs as part of the module compile process with a
do_compile_prepend() routine in module.bbclass.
We don't have to clean the hostprogs as modules depend on the kernel being
populate_staging, so its done with the staging directory by the time we run.
(From OE-Core rev: e807fc977770cb64a217768672c18437ea8f3057)
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
CC: Gary Thomas <gary@mlbassoc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
| |
|
|
| |
git-svn-id: https://svn.o-hand.com/repos/poky/trunk@4518 311d38ba-8fff-0310-9ca6-ca027cbcb966
|
| |
|
|
| |
git-svn-id: https://svn.o-hand.com/repos/poky/trunk@3422 311d38ba-8fff-0310-9ca6-ca027cbcb966
|
| |
|
|
| |
git-svn-id: https://svn.o-hand.com/repos/poky/trunk@2639 311d38ba-8fff-0310-9ca6-ca027cbcb966
|
| |
|
|
| |
git-svn-id: https://svn.o-hand.com/repos/poky/trunk@1925 311d38ba-8fff-0310-9ca6-ca027cbcb966
|
| |
|
|
| |
git-svn-id: https://svn.o-hand.com/repos/poky/trunk@651 311d38ba-8fff-0310-9ca6-ca027cbcb966
|
|
|
git-svn-id: https://svn.o-hand.com/repos/poky/trunk@530 311d38ba-8fff-0310-9ca6-ca027cbcb966
|
