genrandconfig is used by the Buildroot autobuilders to generate
semi-random configurations that we build test. As part of this, we use
"make randpackageconfig" to randomize the selection of packages,
together with a KCONFIG_PROBABILITY value, which indicates the
probabibility for each option to be enabled. This probability is
itself randomized, between 1% and 30% for every build.
However, with our increasing number of packages (over 2900), when we
use a 30% probability for options to be enabled, it means a *lot* of
options are enabled, causing very large configurations to be
tested. These configurations are not very realistic, and they take
ages to build on our autobuilders: we have builds that take 4, 5 or
even 7 hours to build.
In order to test a larger number of configurations and therefore a
larger variety of configurations, this commit reduces the maximum
probability to 20%.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
In commit
40bb37bd70 ("utils/getdeveloperlib.py:
use relative paths for files"), the Developer class was changed to use
relative paths, including for its .hasfile() method.
However the check_developers() function of getdeveloperlib.py was not
updated accordingly, and continued to pass absolute paths. This caused
"get-developers -c" to return the entire list of files in Buildroot as
being unmaintained, as none of them were matching the file listed in
the DEVELOPERS file.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
In Python 3.x, check_output() returns a "bytes" array, and not a
string. Its result needs to be decoded to be turned into a
string. Without this fix, "get-developers -c" bails out with:
Traceback (most recent call last):
File "/home/thomas/projets/buildroot/./utils/get-developers", line 105, in <module>
__main__()
File "/home/thomas/projets/buildroot/./utils/get-developers", line 53, in __main__
files = getdeveloperlib.check_developers(devs)
File "/home/thomas/projets/buildroot/utils/getdeveloperlib.py", line 280, in check_developers
files = subprocess.check_output(cmd).strip().split("\n")
TypeError: a bytes-like object is required, not 'str'
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
ee8b680816 ("utils/scanpypi: use python3 explicitly") started to use python3,
thus compatibility can be removed:
from __future__ import print_function
from __future__ import absolute_import
Tested with python3 -m py_compile.
Signed-off-by: Petr Vorel <petr.vorel@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Python 2 is EOL sice 2020 [1], it's still available on distros, but may not
be installed by default (as being replaced by python3).
Thus remove compatibility imports:
from __future__ import print_function
from __future__ import absolute_import
Tested with python3 -m py_compile.
[1] https://www.python.org/doc/sunset-python-2/
Signed-off-by: Petr Vorel <petr.vorel@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Force relative file path resolution of DEVELOPERS file entries to use
forward-slash separators since pattern matching assumes forward slashes.
This is to help permit uses invoking `get-developers` on Platforms where
`os.sep` may not be a forward slash.
Signed-off-by: James Knight <james.d.knight@live.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Explicitly indicate the file encoding to UTF-8 for the DEVELOPERS
document. This prevents Unicode decoding errors when printing E-Mail
entries with Unicode characters on systems using an alternative default
encoding (e.g. 'CP1252').
This corrects the following observed error:
$ ./utils/get-developers outgoing/*
Traceback (most recent call last):
File "utils\get-developers", line 105, in <module>
__main__()
File "utils\get-developers", line 47, in __main__
devs = getdeveloperlib.parse_developers()
File "...\buildroot\utils\getdeveloperlib.py", line 239, in parse_developers
for line in f:
File "...\Python<ver>\lib\encodings\cp1252.py", line 23, in decode
return codecs.charmap_decode(input,self.errors,decoding_table)[0]
UnicodeDecodeError: 'charmap' codec can't decode byte 0x81 in position 6659: character maps to <undefined>
Signed-off-by: James Knight <james.d.knight@live.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Currently, running test-pkg is only done locally on the developers
machine.
In a follow up commit, we'll add the possibility to run test-pkg in a
gitlab-ci pipeline and, to speed things up, with one job per buildable
configuration.
As such, we will need to make sure that test-pkg only prepares the
configurations, and that it does not build them.
Add such a mode, with a new option, --prepare-only
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Romain Naour <romain.naour@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
When the config fragment provided by the user is not usable with a
specific toolchain configuration, the resulting .config file was kept
around.
In a follow up commit, we'll need to know, from outside test-pkg, if a
specific configuration was indeed usable or not.
So, unless if the user actually requested to keep the build directories,
remove the .config file when it contains a configration that would be
skipped.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Romain Naour <romain.naour@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Commit e43c050944 introduced two flake8
errors:
utils/scanpypi:300:26: E231 missing whitespace after ','
utils/scanpypi:302:9: F841 local variable 'setup' is assigned to but never used
The first one is easily fixed. The second one needs a little bit of
explanation. Before commit e43c0509, the return value of
imp.load_module() was used to be able to explicitly call the 'setup'
function in it in case the metadata was not populated. Since that
commit, calling that function is no longer needed, since setup.py is
executed in exactly the same way as when it's run from the command line,
so if that doesn't work, it's completely broken anyway. Therefore, we
can simply discard the return value of imp.load_module().
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
In case the setup.py file of a python package does not directly call the
'setup' method, utils/scanpypi was hoping there be a 'main' function which
would do the work, normally called via a construct like:
if __name__ == '__main__':
main()
However, this construct is nonstandard, and there are packages in PyPI which
call 'setup()' directly from the 'if' statement, without a main() method.
But scanpypi does not actually need to make such assumption: when loading
the module, it can decide the name to be '__main__', just as if setup.py
would be loaded interactively.
Additionally, remove some logic seemingly related to the previous trick of
calling 'main'. There should not be a problem in keeping already loaded
modules in sys.modules, as this is the purpose of sys.modules.
Signed-off-by: Thomas De Schampheleire <thomas.de_schampheleire@nokia.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Even though the directory containing a package's setup.py was added to
sys.path, some setup.py implementations rely on the fact that it is placed
in sys.path[0].
An example package is 'cram' which failed to be added with scanpypi:
Traceback (most recent call last):
File "utils/scanpypi", line 756, in <module>
main()
File "utils/scanpypi", line 703, in main
package.load_setup()
File "utils/scanpypi", line 303, in load_setup
setup = imp.load_module('setup', s_file, s_path, s_desc)
File "/usr/lib/python3.8/imp.py", line 234, in load_module
return load_source(name, filename, file)
File "/usr/lib/python3.8/imp.py", line 171, in load_source
module = _load(spec)
File "<frozen importlib._bootstrap>", line 702, in _load
File "<frozen importlib._bootstrap>", line 671, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 783, in exec_module
File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed
File "/tmp/scanpypi-2pzc5wb_/python-cram/cram-0.7/setup.py", line 44, in <module>
long_description=long_description(),
File "/tmp/scanpypi-2pzc5wb_/python-cram/cram-0.7/setup.py", line 20, in long_description
return open(os.path.join(sys.path[0], 'README.rst')).read()
FileNotFoundError: [Errno 2] No such file or directory: '.../buildroot/utils/README.rst'
The corresponding code from cram's setup.py is:
def long_description():
"""Get the long description from the README"""
return open(os.path.join(sys.path[0], 'README.rst')).read()
Indeed, the Python documentation says:
https://docs.python.org/3.8/library/sys.html#sys.path
"...
As initialized upon program startup, the first item of this list,
path[0], is the directory containing the script that was used to invoke
the Python interpreter.
..."
Fix this by inserting explicitly at index 0 instead of appending to
sys.path.
Signed-off-by: Thomas De Schampheleire <thomas.de_schampheleire@nokia.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Some packages such as python-idna has a LICENSE.md file:
https://github.com/kjd/idna/blob/master/LICENSE.md
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
... similar to BR2_ENABLE_DEBUG.
Signed-off-by: Thomas De Schampheleire <thomas.de_schampheleire@nokia.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Since 810ba387be, some form of these options are enable
by default. Specifically:
- Kept FORTIFY level 2 option as the default is now level 1.
- Removed all SSP options as the default now uses the best
option based on toolchain support.
- Similar to SSP, for RELRO, the default now uses the best
option based on toolchain support.
- Completely drop PIC PIE as it defaults =y
Signed-off-by: Matthew Weber <matthew.weber@collins.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
scanpypi is python3 compatible. In addition, it executes the setup.py
of Python modules to extract the relevant information. Since these are
more and more commonly using python3 constructs, using "python" to run
scanpypi causes problems on systems that have python2 installed as
python, when trying to parse setup.py scripts with python3 constructs.
Fixes part of #13516.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
When a developer has package/pkg-<infra>.mk assigned to him/her in the
DEVELOPERS file, this has 3 implications:
(1) Patches adding new packages using this infrastructure are Cc'ed
to this developer. This is done by the analyze_patch() function,
which matches the regexp r"^\+\$\(eval
\$\((host-)?([^-]*)-package\)\)$" in the patch, i.e where an
added line contains a reference to the infra maintained by the
developer.
(2) Patches touching the package/pkg-<infra>.mk file itself are Cc'ed
to this developer.
(3) Any patch touching a package using this infra are also Cc'ed to
this developer.
Point (3) causes a significant amount of patches to be sent to
developers who have package/pkg-generic.mk and
package/pkg-autotools.mk assigned to them in the DEVELOPERS
file. Basically, all patches touching generic or autotools packages
get CC'ed to such developers, which causes a massive amount of patches
to be received.
So this patch adjusts the getdeveloperlib.py to drop point (3), but
preserves point (1) and (2). Indeed, it makes sense to be Cc'ed on new
package additions (to make a review that they use the package
infrastructure correctly), and it makes sense to be Cc'ed on patches
that touch the infrastructure code itself.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Commit 40bb37bd70 refactored get-developers, and now the 'os' module is
no longer needed, but still imported:
utils/get-developers:6:1: F401 'os' imported but unused
1 F401 'os' imported but unused
Drop it now.
Reported-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Using absolute paths within getdeveloperlib isn't very sensible, it
makes a lot more sense to handle everything as relative paths from the
top-level Buildroot source directory.
parse_developers() is changed to no longer take the base path as
argument: it is automatically calculated based on the location of
utils/getdeveloperlib.py. Then, the rest of the logic is adjusted to
use relative paths, and prepend them with the base "brpath" when
needed.
This commit allows pkg-stats to report correct developers information
even when executed from an out of tree directory.
Before this patch:
$ ~/buildroot/support/scripts/pkg-stats -p ipmitool --json out.json
$ cat out.json | jq '.packages.ipmitool.developers'
[]
$ cat out.json | jq '.defconfigs.stm32f469_disco'
{
"name": "stm32f469_disco",
"path": "configs/stm32f469_disco_defconfig",
"developers": []
}
After this patch:
$ ~/buildroot/support/scripts/pkg-stats -p ipmitool --json out.json
$ cat out.json | jq '.packages.ipmitool.developers'
[
"Floris Bos <bos@je-eigen-domein.nl>",
"Heiko Thiery <heiko.thiery@gmail.com>"
]
$ cat out.json | jq '.defconfigs.stm32f469_disco'
{
"name": "stm32f469_disco",
"path": "configs/stm32f469_disco_defconfig",
"developers": [
"Christophe Priouzeau <christophe.priouzeau@st.com>"
]
}
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Tested-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Instead of open-coding Developers.hasfile() in utils/get-developers,
use it directly.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Tested-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
An 'else' or 'elif' clause inside a make conditional should not be indented
in the same way as the if/endif clause. check-package did not recognize the
else statement and expected an indentation.
For example:
ifdef FOOBAR
interesting
else
more interesting
endif
would, according to check-package, need to become:
ifdef FOOBAR
interesting
else
more interesting
endif
Treat 'else' and 'elif' the same as if-like keywords in the Indent test, but
take into account that 'else' is also valid shell, so we need to correctly
handle line continuation to prevent complaining about the 'else' in:
ifdef FOOBAR
if true; \
... \
else \
... \
fi
endif
We don't add the 'else' and 'elif' statements to start_conditional, because
it would cause incorrect nesting counting in class OverriddenVariable.
Signed-off-by: Thomas De Schampheleire <thomas.de_schampheleire@nokia.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
pkg-stats is not able anymore to set the developers for defconfigs and
packages. This issue is introduced with
ae86067a15. The hasfile() method from
Developer object tries to check an absolute path against a relative path.
Convert the filepath to be checked also into an absolute path.
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Normally the kconfig stings would end up empty and cause a build
error. This patch provides test files to allow testing the creation
of uboot environment and script bin files from user provided txt files.
Cc: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Matthew Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Raw strings need to be used when calling re.compile() otherwise Python
3.x flake8 complains with:
W605 invalid escape sequence '\s'
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Reviewed-by: Titouan Christophe <titouan.christophe@railnova.eu>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Commit 9ea528f84b (package/python-nfc: bump to version 0.13.5) changed the
python-nfc package to download from github, so the package no longer needs
bzr on the host.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
With classpath removed, no packages select these symbols any more - So drop
them and their corresponding logic in dependencies.sh / genrandconfig.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The new .hash convention is to use 2 spaces between fields.
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
The aclocal program is provided by the automake package, so it makes
sense to define aclocal-related variables in automake.mk.
Add an exception to check-package to ignore that variable.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
we don't want create new BR package with perl core module,
because core modules are already included in perl distribution,
and built with the BR package perl.
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Now that the support for per-package directories has been merged, it
is time to get some exposure for it in the autobuilders, so let's
build 1 out of 15 builds with this feature enabled, at least as an
initial step.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
With Python 3.7, genrandconfig fails with:
'str' object has no attribute 'decode'
We are already working on str objects, and there is no need to decode
them, so we drop the call to decode_byte_list() and its definition as
it was only used there.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
That change will alphabetically set list order
of required packages in Config.in file automatically.
Example below:
before: ['python-pyserial', 'python-pyaes', 'python-ecdsa']
after: ['python-ecdsa', 'python-pyaes', 'python-pyserial']
Signed-off-by: Bartosz Bilas <b.bilas@grinn-global.com>
Reviewed-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
License files are sometimes called LICENCE, so support that as well.
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
This patch adds the remaining ssp and relro cases plus both of the
fortify options. The randomization was left consistant between the
options but the order of the conditions placed the most restrictive
hardening options with more priority.
Cc: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Matthew Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
