| Age | Commit message (Collapse) | Author |
|---|
|
CCACHE_PATH was added to the toolchain-scripts in commit
1d31ddb856a80ba9da1a64ed5d701dc0f7351ef7 without a detailed
explanation as to why. This commit removes that environment
variable as it's causing failures since the introduction of
host_gcc_version() to oe.utils in
d36f41e5658bbbb6080ee833027879c119edf3e0, as used by
uninative to determine NATIVELSBSTRING.
This causes errors when the host has ccache available in in PATH
- host_gcc_version() fails because ccache has been told to only
look for the compiler in CCACHE_PATH and can't find gcc in order
to check its version. Toolchain compilers aren't detected due to
the binaries being prefixed with the triplet of the target.
Signed-off-by: Joshua Lock <joshua.g.lock@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.
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 ?\(( ?[^,()]*), True\)
Signed-off-by: Joshua Lock <joshua.g.lock@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
We use toolchain_create_sdk_version() in buildtools-tarball but
don't want the extra classes toolchain-scripts pulls in, therefore
split out a separate base class for this function which both
toolchain-scripts and the buildtools-tarball can inherit.
Signed-off-by: Joshua Lock <joshua.g.lock@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
As things stand there are multiple races in the CONFIG_SITE handling
where checksums can change depending on whether site directories
exist or not when parsing happens. This is bad.
Secondly, there is a build race that occurs if you build virtuals
in parallel with the "main" recipe, since the main recipe is parsed
when the virtual is (since it sets variables like BBCLASSEXTEND)
and with the current code, it may look for files and directories
which could be created/destroyed which the loop is executing. This
is also bad.
The aclocal-copy directory should only ever be accessed by the call
from autotools.bbclass. This changes the parameter name to make it
clear and ensures all callers have the right usage, neatly avoiding
all the problems above. Also added better comments.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
CFLAGS/CXXFLAGS in the SDK environment script adds debug-prefix mappings
that include staging area/work directories. Remove them since the SDK
shouldn't be aware of them.
Signed-off-by: Jacob Kroon <jacob.kroon@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
Otherwise architecture-independent pkg-config files such as wayland-protocols
won't be found in the SDK.
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
Some shells (e.g. dash) do not support the source built-in. This
replaces it with the dot operator.
[ YOCTO #9535 ]
Signed-off-by: Stephano Cetola <stephano.cetola@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
in sdk,there are some utils in sysroot/host-os/bin, sysroot/host-os/sbin/,sysroot/host-os/usr/bin need to use, so add these three paths to PATH in env.sh.
Signed-off-by: Jun Zhang <jun.zhang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If this is set when nativesdk-python is not installed, it will break the host
python, so it's better to handle this in python.
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
In the extensible sdk it was originally intended that the native sstate
would always be setscened as part of the sdk installation. However, the
soon to come "minimal" sdk won't do that.
A side effect of that is that pointing PYTHONHOME at the native sysroot
within the "bitbake workspace" won't work. For now only set PYTHONPATH
instead and continue using the python that comes from buildtools.
Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
On Ubuntu-system, When sourcing the env.sh from an exported sdk, and
running a bogus linux command (for example "asd"), a core dump of
python is usually generated.
Unset the command_not_found_handle to fix it.
Signed-off-by: Fang Jia <fang.jia@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
This is needed when we add nativesdk-gcc/binutil to an SDK. Being
present doesn't hurt in other cases.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
correct libc selection
gcc-cross-canadian-<arch> is only built once. It needs to target all the
different libcs, not just the currently selected one. This change ensures
that if another libc is used, the compiler correctly selects the right one.
[YOCTO #8025]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Use MLPREFIX to fix:
| cat: /path/to/sysroots/lib32-qemux86-64/sysroot-providers/virtual_libc: No such file or directory
[YOCTO #7924]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The use of TCLIBC in TOOLCHAIN_NEED_CONFIGSITE_CACHE is problematic
since a multilib may have both uclibc and glibc for different multilibs
yet switching between them doesn't change TCLIBC. This would result
in "lib32-glibc" being attempted when lib32 was actually uclibc.
The fix here is to use the virtual providers which bitbake switches
to point correctly at the right things.
This does mean we need to resolve virtual providers but we can do this using
sysroot-providers.
[YCOTO #7607]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Single quotes prevent expansion of $sdkpathnative$bindir.
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
In multilib and baremetal configurations, this variable can cause a variety of
problems due to the use of TCLIBC. At least allowing it to be overriden
is a start and allows various configurations to avoid the issue.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
To add some flexibility to setting up the paths for the toolchain,
add some parameters. This initial use will be in order to point at
the buildsystem toolchain copied in by copy_buildsystem.py.
Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
It is necessary to have an SDK for developers who build their software with
ccache to speed up the compilation. Without resolving this, unfortunately the
compilation will fail for them.
Signed-off-by: Laszlo Papp <lpapp@kde.org>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
When building the U-Boot the lack of a proper sysroot can trigger
following error:
,----
| arm-poky-linux-gnueabi-ld.bfd: cannot find -lgcc
| make[2]: *** [examples/standalone/hello_world] Error 1
| make[1]: *** [examples/standalone] Error 2
| make: *** [examples] Error 2
`----
Guillaume Fournier has posted a very complete analysis of the
problem[1].
1. https://lists.yoctoproject.org/pipermail/meta-freescale/2014-November/011270.html
The use of KCFLAGS makes the build of U-Boot work out of box, now that
it uses the Linux kernel build system.
Reported-by: Guillaume Fournier <gfournier@brioconcept.com>
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
[YOCTO #6951]
The TOOLCHAIN_CONFIGSITE_SYSROOTCACHE value was defaulting to the nativesdk
path and not the associated target path. Set the value in toolchain-scripts
to the target path.
Be sure to set the MLPREFIX within the meta-environment script as multilibs
are processed.
Update the config_site file name to use -BPN- not PN. Otherwise the
environment processing can't find the correct filename.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
|
|
In a similar way to the previous script which adds support for native
environment scripts, this adds support for target environment scripts
too.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Sometimes we require extra environment settings to be available on the
environment for proper SDK work. This were done, in past, using
'_append' tasks however with the split of the environment in a
canadian package this has been broken.
The easier and more flexible solution is to use environment subscripts
which are sources by the main script. These are now looked at:
$OECORE_NATIVE_SYSROOT/environment-setup.d/*.sh
and sourced.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There was an error in the previous fix for the powerpc toolchain issue, this
should correct it (and simplify the code too).
[YOCTO #6490]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When we build a general toolchain script we should add all the OS
variants to PATH, not just the current one. This is because some can
cross reference each other and if the triplet prefixed ld can't be found
for example, you recieve strange errors.
Doing this resolves bugs exposed during SDK testing.
[YOCTO #6490]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There was still duplicated code in toolchain-scripts, this further cleans
up the functions to remove it. The now unused includedir parameter is also
dropped.
The final scripts do end up reordered slightly and the sysroot is parametrised
for the IDE scripts which is an improvement but should have no functional
difference.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The code was clearly broken and isn't used, even by meta-darwin so we
might as well remove it.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
and the for_installer variant
Having two scripts which do basically the same thing is a nightmare. This merges
them together. It also makes the sysroot location a variable in its own right
which may be more useful for end users wanting to change sysroot.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There was a fix missing from a previous commit:
http://git.yoctoproject.org/cgit.cgi/poky/commit/?id=20a86e16bc54b74fbdb2a3a407d54210ea262925
since there was another section of toolchain-scripts needing updating.
This patch fixes the missing reference and unbreaks the ADT toolchain.
[YOCTO #5340]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The cross-canadian compilers are now build once per architecture but were being
installed into tune specific locations which is incorrect. This adjusts things
so they are make TARGET_ARCH specific. We gain the tune specific parts from the
target sysroot which remains tune specific, the compiler and tools are independent
ot that.
binutils/gcc require sysroot options but since we reset at runtime, these shouldn't
have dependencies in the sstate checksums. They are therefore also excluded.
With these patches, switching machines does not result in a rebuild of *-cross-canadian
and the compiler is correctly located and referenced in the target images.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
when compiling the linux kernel manually. users need to set CROSS_COMPILE separately.
adding the CROSS_COMPILE variable will be nice for using.
Signed-off-by: Zongchun Yu <b40527@freescale.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
build external kernel module using SDK
Signed-off-by: Jessica Zhang <jessica.zhang@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
When relocating the SDK, applications using python will search for
python modules in the default location and will fail to start.
The below errors are thrown by gdb, for example:
Could not find platform independent libraries <prefix>
Could not find platform dependent libraries <exec_prefix>
Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
ImportError: No module named site
In order to overcome this, add the PYTHONHOME variable to the
environment-setup script for both standalone toolchain and
adt-installer. No need to do that for meta-ide-support environment
script since this toolchain does not get relocated.
[YOCTO #3839]
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
some packages use M4 variable from environment and sometimes
its hardcoded to /usr/bin/m4 if not found in environment. Lets
define it such that it is picked from path
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
TARGET_PREFIX already has the dash included.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
will make autotools look for arm-none-linux-gnueabi--ar instead of arm-none-linux-gnueabi-ar.
Signed-off-by: Martin Ertsaas <mertsas@cisco.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
[YOCTO #2761]
This patch fixed the issue that after "bitbake meta-ide-support" and try to start qemu and it will fail. This is due to the meta-ide-support lacking dependency for qemu native and ended up using qemu under /usr/bin. Fix the issue by adding dependency for qemu native in meta-ide-support and also add the path info in the environment-setup script.
Signed-off-by: Jessica Zhang <jessica.zhang@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
The build system passes specific TUNE (fundamental) flag values to various
tools instead of using a shell wrapper or similar. It is important that the
build system and ADT/SDK match the behaviors exactly, or we are likely to
have differences in the way build-system and external components are built
leading to configuration, compilation and/or run-time problems.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
In order to enable basic multilib support for the export of an SDK
image, a number of minor changes had to be made:
The value of MULTIMACH_TARGET_SYS needs to be variable. This way we
can define the value to the appropriate multilib. (Also in some cases
the default PACKAGE_ARCH is set to MACHINE_ARCH which is incorrect for
the SDK.) Add a companion REAL_MULTIMACH_TARGET_SYS, based on code
from meta-environment, to allow for this.
We have to convert the do_populate_sdk into a python call, and then
break up the previous items into three parts.
* Image construction
* Setup of environment files
* Generation of the tarball
Then we can iterate over the multilibs to populate the environment files.
Finally, matching changes were needed in the toolchain-scripts file. And
what I presume is an optimization of immediate evaluation for
TOOLCHAIN_CONFIGSITE_NOCACHE and ..._SYSROOTCACHE needed to be done at
runtime, otherwise the wrong values may be used.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
|
|
Fixed bug [YOCTO #2159] by correctly populate the libc's siteconfig
files.
Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
unecessarily machine specific
basehash changed from 30e97f9eeed1df8488b62b4fb47a3a0c to 72ea9a277e6599ae9052b169b9a94f1b
Variable TOOLCHAIN_CONFIGSITE_SYSROOTCACHE value changed from
/OE/oe-core/tmp-eglibc/sysroots/qemux86-64/usr/share/x86_64-oe-linux_config_site.d
to
/OE/oe-core/tmp-eglibc/sysroots/qemux86-64copy/usr/share/x86_64-oe-linux_config_site.d
Variable TIME value changed from 084543 to 085638
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Koen Kooi <koen@dominion.thruhere.net>
|
|
env case
Signed-off-by: Jessica Zhang <jessica.zhang@intel.com>
|
|
This class currently only works with eglibc. Since
it adds dependencies explicitly on eglibc when using
uclibc this creates problems. So we make sure that
it checks for TCLIBC to determine system C library
in use
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
This is the result of running the following over the metadata:
sed \
-e 's:bb.data.\(setVar([^,()]*,[^,()]*\), *\([^ )]*\) *):\2.\1):g' \
-e 's:bb.data.\(setVarFlag([^,()]*,[^,()]*,[^,()]*\), *\([^) ]*\) *):\2.\1):g' \
-e 's:bb.data.\(getVar([^,()]*\), *\([^(), ]*\) *,\([^)]*\)):\2.\1,\3):g' \
-e 's:bb.data.\(getVarFlag([^,()]*,[^,()]*\), *\([^(), ]*\) *,\([^)]*\)):\2.\1,\3):g' \
-e 's:bb.data.\(getVarFlag([^,()]*,[^,()]*\), *\([^() ]*\) *):\2.\1):g' \
-e 's:bb.data.\(getVar([^,()]*\), *\([^) ]*\) *):\2.\1):g' \
-i `grep -ril bb.data *`
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Jessica Zhang <jessica.zhang@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
[YOCTO #892]
Modify the function toolchain_create_sdk_siteconfig to collect the
cached site config files which are specified by
TOOLCHAIN_NEED_CONFIGSITE_CACHE in runtime.
Also added task dependency to ensure the cached site config files are
generated.
Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
|
|
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
|
Joshua Lock
Richard Purdie
Jacob Kroon
Ross Burton
Stephano Cetola
Jun Zhang
Christopher Larson
Randy Witt
Fang Jia
Robert Yang
Andre McCurdy
Laszlo Papp
Otavio Salvador
Mark Hatle
Zongchun Yu
Jessica Zhang
Laurentiu Palcu
Khem Raj
Martin Ertsaas
Lianhao Lu
Koen Kooi