| Age | Commit message (Collapse) | Author |
|---|
|
Some machines change TARGET_OS, cross-canadian resets this which
is not what we want in this specific case. This fixes spe toolchains
for example.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Currently libdir is coming from cross-canadian which is incorrect. We
need to reset this to target_libdir so that the toolchains contain the
correct value.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Firstly, when multilib builds were being used, the same output files
were used in each case, being overwritten each time due to the fact that
REAL_MULTIMACH_TARGET_SYS has been expanded. Instead of immediate
expansion, just use assignment. This overrides the problematic define
in toolchain-scripts.bbclass but allows the multilib code to work.
Secondly, the target sysroot was being defined incorrectly which
this patch fixes. This was breaking the toolchains.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
These patches are the start of a new idea, a way of allowing a single set of
cross/native sstate to work over mutliple distros, even old ones.
The assumption is that our own C library is basically up to date. We build
and share a small tarball (~2MB) of a prebuilt copy of this along with a
patchelf binary (which sadly is C++ based so libstdc++ is in there). This
tarball can be generated from our usual SDK generation process through
the supplied recipe, uninative-tarball.
At the start of the build, if its not been extracted into the sysroot, this
tarball is extracted there and configured for the specified path.
When we install binaries from a "uninative" sstate feed, we change the
dynamic loader to point at this dynamic loader and C librbary. This works
exactly the same way as our relocatable SDK does. The only real difference
is a switch to use patchelf, so even if the interpreter section is too small,
it can still adjust the binary.
Right now this implements a working proof of concept. If you build the tarball
and place it at the head of the tree (in COREBASE), you can run a build from
sstate and successfully build packages and construct images.
There is some improvement needed, its hardcoded for x86_64 right now, its trivial
to add 32 bit support too. The tarball isn't fetched right now, there is just a
harcoded path assumption and there is no error handling. I haven't figured
out the best delivery mechanism for that yet. BuildStarted is probably not
the right event to hook on either.
I've merged this to illustrate how with a small change, we might make the
native/cross sstate much more reusable and hence improve the accessibility
of lower overhead builds. With this change, its possible the Yocto Project may
be able to support a configured sstate mirror out the box. This also has
positive implications for our developer workflow/SDK improvements.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Currently we have the horrible situation where meta-environment packages
the toolchain environment files and they get included in the SDK but are
broken, then, the SDK code overwrites them with good versions. This is
suboptimal.
This change fixes the code in meta-environment to create working
files and adds in the multilib support from populate_sdk_base, then
we remove the code in that base bbclass and rely on the packages
being installed if/as/when needed. This removes the duplication
and the broken versions of the files, hopefully making all well.
[YOCTO #6608]
[YOCTO #6613]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Due to the use of the cross-canadian inherit, TARGET_OS can be reset to
values other than the correct one, leading to generation of the wrong
files.
By expanding REAL_MULTIMACH_TARGET_SYS before the inherit, the correct
OS value is preserved and we generate the right environment files.
[YOCTO #6608]
[YOCTO #6613]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
export GIT_SSL_CAINFO so git can find the certs
Signed-off-by: Fahad Usman <fahad.usman@gmail.com>
Signed-off-by: Christopher Larson <kergoth@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
nativesdk-ca-certificates is needed in order to support oe/yocto builds with
buildtools-tarball on old hosts, as we provide libcurl, and that needs to be
able to find the certs, and there's no standard path or bundle path, so we
can't rely on the host.
Signed-off-by: Fahad Usman <fahad.usman@gmail.com>
Signed-off-by: Christopher Larson <kergoth@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
As the do_package_write() task is listed as deleted, remove the few
remaining references.
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
GNU Wget cannot be upgrated to a newer that 1.12 version on supported
Centos distro. GNU Wget 1.12 and earlier uses a server-provided filename
instead of the original URL to determine the destination filename of a
download.
This means the files downloaded when fetching cannot be properly used:
$ wget http://www.balabit.com/downloads/files/syslog-ng/open-source-edition/3.4.2/source/eventlog_0.2.13.tar.gz
$ ls
eventlog_0.2.13.tar.gz?AWSAccessKeyId=AKIAICTJ5MANGPMOH7JA&Expires=1400838672&Signature=TjakOBpOvHtEKKDgF14iVinWpY0=
This in turn lead to build errors like:
WARNING: Failed to fetch URL http://www.balabit.com/downloads/files/syslog-ng/open-source-edition/3.4.2/source/eventlog_0.2.13.tar.gz, attempting MIRRORS if available
ERROR: Fetcher failure for URL: 'http://www.balabit.com/downloads/files/syslog-ng/open-source-edition/3.4.2/source/eventlog_0.2.13.tar.gz'. The fetch command returned success for url http://www.balabit.com/downloads/files/syslog-ng/open-source-edition/3.4.2/source/eventlog_0.2.13.tar.gz but /path/to/downloads/eventlog_0.2.13.tar.gz doesn't exist?!
ERROR: Function failed: Fetcher failure for URL: 'http://www.balabit.com/downloads/files/syslog-ng/open-source-edition/3.4.2/source/eventlog_0.2.13.tar.gz'. Unable to fetch URL from any source.
ERROR: Logfile of failure stored in: /path/to/tmp/work/ppce500v2-enea-linux-gnuspe/eventlog/0.2.13-r0/temp/log.do_fetch.28302
ERROR: Task 4 (/path/to/poky/meta-openembedded/meta-oe/recipes-support/eventlog/eventlog_0.2.13.bb, do_fetch) failed with exit code '1'
[YOCTO #6549]
Signed-off-by: Tudor Florea <tudor.florea@enea.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This allows them to co-exist together in the native sysroot, with one
set of cross tools per target architecture.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Other recipes dependencies and even some comments need to be updated
for the removal of unfs-server and the replacement with unfs3. The
unfs3 is a complete drop in replacement providing all the prior
functionality of NFSv2 but also adding NFSv3.
[YOCTO #5639]
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
These recipes all had a long DESCRIPTION but no SUMMARY; since the
SUMMARY is often displayed alone by package managers and the default
value ("${PN} version ${PV}") isn't particularly useful, we should
always try to set SUMMARY.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
A lot of our recipes had short one-line DESCRIPTION values and no
SUMMARY value set. In this case it's much better to just set SUMMARY
since DESCRIPTION is defaulted from SUMMARY anyway and then the SUMMARY
is at least useful. I also took the opportunity to fix up a lot of the
new SUMMARY values, making them concisely explain the function of the
recipe / package where possible.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
As per discussion on the mailing list [1], remove this largely
unmaintained external toolchain support in favour of the maintained
version in meta-sourcery [2].
Also correct the example and documentation.conf entries for TCMODE to
match up with this change.
[1] http://lists.openembedded.org/pipermail/openembedded-core/2013-December/087133.html
[2] https://github.com/MentorEmbedded/meta-sourcery/
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
GMAE (GNOME Mobile And Embedded) doesn't really exist anymore and these packages
were a subset of it anyway. Remove as they don't give the user anything useful
now that you can generate a SDK from an arbitrary image.
Signed-off-by: Ross Burton <ross.burton@intel.com>
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>
|
|
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>
|
|
When using the tar executable in the buildtools, tar will execute
gzip. If this happens before zlib-native is built, then the gzip
on the host will be used and can fail if the libz in the buildtools
is not compatible. Adding pigz to the build tools avoids this host
contamination.
Signed-off-by: Konrad Scherer <Konrad.Scherer@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Currently, the package name contains just the TRANSLATED_TARGET_ARCH.
When compiling the toolchain for two machines, belonging to the same
architecture (for example: qemuarm and beagleboard), this package gets
overwritten and adt-installer repo will contain just one
meta-environment package... This leads to situations like installing the
toolchain for qemuarm and end up with meta-environment package with
beagleboard cpu options.
[YOCTO #4783]
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This is needed by some recently added automated QA tests so we should add it
to the buildtools tarball.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When building a qemu image inside the environment created by the
buildtools-tarball, the qemu image cannot be started, as the runqemu
script uses the tunctl binary which cannot be found inside the sysroot
directory of the buildtools-tarball.
The buildtools-tarball is inherently a tool set instead of a fully
functional SDK, so leaving the OECORE_NATIVE_SYSROOT variable in the
environment will mess things up.
However, we do need a line of 'OECORE_NATIVE_SYSROOT=xxx' in the environment
setup script so that the SDK can be extracted and relocated correctly.
Thus, instead of exporting OECORE_NATIVE_SYSROOT, we use a comment here.
[YOCTO #4939]
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Addess the issue with multiple .bb providers
ERROR: Multiple .bb files are due to be built which each provide virtual/libc (/srv/hdd/releases/dylan/meta/recipes-core/eglibc/eglibc_2.17.bb /srv/hdd/releases/dylan/meta/recipes-core/meta/external-sourcery-toolchain.bb).
This usually means one provides something the other doesn't and should.
ERROR: Multiple .bb files are due to be built which each provide virtual/arm-none-linux-gnueabi-libc-for-gcc (/srv/hdd/releases/dylan/meta/recipes-core/eglibc/eglibc_2.17.bb /srv/hdd/releases/dylan/meta/recipes-core/meta/external-sourcery-toolchain.bb).
This usually means one provides something the other doesn't and should.
ERROR: Multiple .bb files are due to be built which each provide virtual/libiconv (/srv/hdd/releases/dylan/meta/recipes-core/eglibc/eglibc_2.17.bb /srv/hdd/releases/dylan/meta/recipes-core/meta/external-sourcery-toolchain.bb).
This usually means one provides something the other doesn't and should.
Thanks to Kergoth (Chris Larson) and Lpapp (Lazslo)
[YOCTO #4908]
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
The new testimage class needs the mmap python-module in the buildtools-tarball
on machines that use the tarball to provide the correct version of python.
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
python-git is needed by buildhistory-diff and the OE layer index update
script, and on machines where buildtools-tarball is necessary given we
are providing python we must provide this module as well.
Fixes [YOCTO #4747].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Recently it was discovered that many Fedora hosts have a broken version of
make 3.82. Add make to the buildtools-tarball, as well ad modify make to
support building a special nativesdk version.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
The previous method of shrinking down the scale of the environment-setup
file would not work properly when the target is configured with multilibs.
In addition, the configured machine name and similar settings could leak
into the SDK naming. This was resolved by clearing the SDK generated
files and generating our own custom files.
Note, the name of the environment-setup now is suffixed with the SDK_SYS.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
We've been seeing issues where some TERM values don't work well with the
tarball. These show up as escape codes coming from simple python calls such
as printing the python version. Adding the terminfo goes part way to addressing
the problem, the remainder is ensuring the relocation code correctly searches
the right path for it (which will be in a follow on patch).
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Needed for the unittest part of bitbake-selftest can work correctly.
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We need unittest so bitbake-selftest can work, tarfile from the compression module
for opkg-utils and json for webkit builds.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Improve the DESCRIPTION for the recipe and add a SUMMMARY.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The external-python-tarball was intended to provide python for those users who
didn't have python 2.6. It has been unloved for a while so this refreshes
it a bit. Specifically:
* Include git and tar since these can have version issues on some systems
* Rename to buildtools-tarball to better reflect its purpose
* Remove the site-config file from the tarball, its pointless
* Prune down the environment file to just cover PATH and OECORE_NATIVE
(which is needed to correctly install the tarball)
* Add missing grp module (used by do_package) from the unixadmin package
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There are some left over tab characters in the python functions. This
removes them and resolves python 3 errors.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Update TOOLCHAIN_OUTPUTNAME to rename the output package which is the
similiar style to name packages meta-toolchain-*.
[YOCTO #3996]
Signed-off-by: Kang Kai <kai.kang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There are various usages of ALLOW_EMPTY with no packages specified. This
is not recommended syntax, nor is it likely to be supported in the future.
This patch improves the references in OE-Core, either removing them if they're
pointless (e.g. when PACKAGES="") or specifying which package it applies to.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
external-python-tarball only can be built when inherit package_ipk now.
Update it to reuse existed populate sdk code that it could be built for
rpm and deb too.
Remove var DEPENDS, SDK_* and flags of do_populate_sdk because they are
already defined in populate_sdk bbclasses.
[Yocto 3006]
Signed-off-by: Kang Kai <kai.kang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Currently, do_rootfs has a dependency on all the do_package output being present
due to its usage of the pkgdata directories. This means that if you run:
bitbake xxxx-image -c rootfs
you end up having to fetch and unpack all the do_package data which is usually
large and inefficient. It also means rm_work has to leave all the do_package
data lying around so rootfs works.
This patch splits the actual creation of the pkgdata directory off into a separate
task, "packagedata" which happens immediately after do_package. We can then remap
the dependencies so this task is depended upon, not do_package. Sstate can then be
programmed not to require do_package at the appropriate times.
Whilst this patch doesn't do so, it opens the possibility of rm_work wiping
out the do_package output from WORKDIR as long as it also removed the do_package
stamp (both normal and setscene variants) and allowing more space savings
with rm_work which has been regularly requested.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Having to include some dummy class is suboptimal and we no longer need to
do this. Also move this check to populate_sdk_base since we then don't
need to include it in toolchains specifically.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
"Package group" is a much more appropriate name for these than task,
since we use the word task to describe units of work executed by
BitBake.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
As discussed on the mailing lists, using a suffix to package names is
hard and has lead to many recipes having to do PKGSUFFIX games. Its
looking extremely hard to scale nativesdk much further without hacking
many recipes.
By comparison, using a prefix like multilib does works much better and
doesn't involve "hacking" as many recipes. This change converts nativesdk
to use a prefix using the existing multilib infrastructure.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
PACKAGE_CLASSES list
This is needed because ipk tools and variables (specically
IPKGCONF_TARGET and IPKGCONF_SDK) are used from the rootfs_ipk.bbclass
I tried to inherit directly but it still fails to expand those.
[YOCTO #2814]
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
changes
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>
|
|
If the usr/lib directory doesn't exist, the toolchain can fail to even try to
find crti.o in a completely different directory. This causes a failure for the
case where baselib is lib64.
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Rather than hardcoding the multilib path in a map, and hardcoding dest sysroot
symlink creation in a hook, now we just use -print-sysroot for both, and pass
the appropriate multilib args to the toolchain for particular tunes.
Signed-off-by: Christopher Larson <chris_larson@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Christopher Larson <kergoth@gmail.com>
|
Richard Purdie
Fahad Usman
Robert P. J. Day
Tudor Florea
Laurentiu Palcu
Jason Wessel
Paul Eggleton
Ross Burton
Konrad Scherer
Chen Qi
Saul Wold
Mark Hatle
Kang Kai
Jessica Zhang
Christopher Larson
Christopher Larson