While multiple hardlinks of binary located in different dirs,
there are also multiple hardlinks of separated debug info file
with the same binary name in same debug dirs. But in each dir,
only one debug file with original name works. Because all of
binary hardlinks have one `.gnu_debuglink' which is added in
`splitdebuginfo'. It caused gdb could not find debugging
symbols.
[Before the patch]
$ find .
./usr/bin/foo
./usr/bin/foo-hd1
./usr/bin/.debug
./usr/bin/.debug/foo
./usr/bin/.debug/foo-hd1
./usr/libexec/foo-hd2
./usr/libexec/.debug
./usr/libexec/.debug/foo-hd2
$ readelf --debug-dump usr/libexec/foo-hd2
Contents of the .gnu_debuglink section:
Separate debug info file: foo
$ gdb usr/libexec/foo-hd2
Reading symbols from usr/libexec/foo-hd2...(no debugging symbols found)...done.
[Before the patch]
[Apply the patch]
$ find .
./usr/bin/foo
./usr/bin/foo-hd1
./usr/bin/.debug
./usr/bin/.debug/foo
./usr/libexec/foo-hd2
./usr/libexec/.debug
./usr/libexec/.debug/foo
$ gdb usr/libexec/foo-hd2
Reading symbols from usr/libexec/foo-hd2...Reading symbols from usr/libexec/.debug/foo...done.
[Apply the patch]
(From OE-Core rev: d4eaf42f7708f8d3a31a04d958bd7420dd7dd6b9)
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
In finalize() we save event handlers, register the ones relevant to the
recipe being finalised, trigger events, and then restore the handlers so
that one recipe's custom handlers (actually implemented within a class
inherited by the recipe) do not affect other recipes. However, if an
exception occurs during parsing, the saved handlers were not being
restored. Use a try...finally block to ensure that the handlers are
always restored.
This issue became apparent since in OpenEmbedded-Core we have recently
introduced a find_intercepts() handler for the
bb.event.RecipePreFinalise event in image-postinst-intercepts.bbclass
that images and old-style SDK recipes will end up inheriting. So far it
doesn't seem that the the error has manifested itself in normal builds,
but when parsing OE-Core recipes in the OE layer index it has:
core-image-rt-* image recipes were parsed which in the default
configuration raise SkipRecipe. The next non-image recipe that is parsed
will trigger a real exception, because the find_intercepts() handler is
still registered and gets fired, but in the context of the new recipe
the POSTINST_INTERCEPTS_PATHS variable is not set, and the code in
find_intercepts() is written with the reasonable assumption that that
isn't possible given that the class itself sets a default, and thus it
fails.
(Bitbake rev: e5f1f8fa201774e0c3c554d59b277baa2128708f)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The following statement was accidently removed. Add it back.
values['ml'] = ['']
This patch fixes the following error.
Exception: KeyError: 'ml'
(From OE-Core rev: 0791c4d39edbedaa493a9fc6aa6e7415f14980bb)
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Update /sbin to $base_sbindir to fix the
below warning when usrmerge enabled in
DISTRO_FEATURES.
WARNING: mingetty-1.08-r3 do_package: mingetty: NOT adding alternative provide /usr/sbin/getty: /usr/sbin/mingetty does not exist
WARNING: mingetty-1.08-r3 do_package: QA Issue: mingetty: Files/directories were installed but not shipped in any package:
/sbin
/usr/sbin
(From OE-Core rev: 35b92053c5bc0f4ae4b92ca75456535ff78797a2)
Signed-off-by: Mingli Yu <Mingli.Yu@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Update /lib to $nonarch_base_libdir to fix
the below error when usrmerge enabled in
DISTRO_FEATURES.
ERROR: mdadm-4.0-r0 do_package_qa: QA Issue: mdadm package is not obeying usrmerge distro feature. /lib should be relocated to /usr. [usrmerge]
(From OE-Core rev: 4b2c86ff0031effd4cfdb477f67bc213d39ae0fd)
Signed-off-by: Mingli Yu <Mingli.Yu@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The actual dependency on native Python and is handled by inheriting
python3native
(From OE-Core rev: 115a6dea664c9b18fd19b79659029afb52b1a660)
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add a script which can be run on new autobuilder workers to check all needed configuration
is present. Designed to be run in a repo where bitbake/oe-core are already present.
This means when we add new autobuilder workers, we can quickly test whether all the needed
funcationality to support the standard yocto project autobuilder (ie. the standard
OE-Core tests) are present.
It uses images prebuilt in a previous release to cut build/testing time.
(From OE-Core rev: c9dd81fe9c88d8d9881686e60c17e6fab4b230f1)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Refresh avoid-check-user-break-cc.patch to fix fuzz warning.
(From OE-Core rev: 6d9ab12ac113f0f33289200ca3c9df24d9fc1cc2)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
skipIfInDataVar will skip a test if a value is in a certain variable.
(From OE-Core rev: 10b935c713748346aea6c36c2f41e0ae6c320821)
Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The name string is too small to contain normal full path names and causes
the following failure.
"ftest06 2 TFAIL : ftest06.c:223: Can't chdir(): errno=ENOENT(2): No such file or directory"
(From OE-Core rev: 24faf8b9bbee82033d5f636f254e59b8d3d31ea5)
Signed-off-by: He Zhe <zhe.he@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Although the relative_symlinks class converts any absolute symlinks
in ${D} into relative symlinks automatically, it's a little clearer
to create relative symlinks directly where possible.
(From OE-Core rev: 959b4d30b5b11e4a098654b0d4469bbdf01b3812)
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Replace usage of oe.utils.getstatusoutput() with direct subprocess calls.
(From OE-Core rev: f1a1b31add242380490023c2ee7eec1b4fbcd85b)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Replace usage of oe.utils.getstatusoutput() with direct subprocess calls.
(From OE-Core rev: 2a06abb258768504a3ad97f61c987709227d7109)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We can't use subprocess.check_output() or subprocess.call() here since the one
who invokes runcmd() needs handle CmdError() exception (error out or ignore
it).
(From OE-Core rev: c3e7739987d804f7865428442479d5bece5ff2dd)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Replace usage of oe.utils.getstatusoutput() with direct subprocess calls.
(From OE-Core rev: 90c730a898f11adb2ecd377cdd913af83123bcb7)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The previous patch has duplicate split calls and one needs to be removed
to avoid failures
(From OE-Core rev: 02c8d048cbab38a48f698504d0f5e912d3d24a36)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
get_multilib_datastore() should be able to handle the original datastore
correctly now so rely upon this rather than custom coding.
(From OE-Core rev: 2ae85af480066e252fca01f3005ecac2ff37a8d4)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We have library functions to handle multilib variables/datastores, lets
use them so we have good common functions.
(From OE-Core rev: 774219567987956fb7bbb50e64eb6cebef1efe5b)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Currently the original datastore returned by this function doesn't
always work as the tune isn't set back to the original. Fix it
to work like all_multilib_tune_list() in utils.bbclass and correct
the data returned.
(From OE-Core rev: 4e1dc858fbf671ef27089a2b9bcdc965fe19d698)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Currently if a tune isn't specified in the table, the loader defaults for the
architecture are used which may or may not match our path specification. This
leads to general confusion.
Change the code to use the linuxloader class which works of architecture, not
tune.
This still isn't perfect as n32/x32 aren't covered but its an improvement
to listing all tunes here.
(From OE-Core rev: 46a6da24b51426bedd9af8a2d63b2992b9d3fa5a)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We could do with one decent general purpose python function to query the
path to the dynamic loader. Convert the shell code into python.
Also correct baremetal to return "None", not musl loaders.
(From OE-Core rev: 73fab4ede12d8ae31be72b5cb4ab29d7ef7dae17)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This does the same thing, but is more efficient in case st_nlinks
is (already) 1.
Depends on bitbake commit 7ae93cf40ab91965147055100432961436bce46c .
(From OE-Core rev: a09f8e32044c8daec2d2fb3ff0e830c21402df6e)
Signed-off-by: Rasmus Villemoes <rv@rasmusvillemoes.dk>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This does the same thing, but is more efficient in case st_nlinks
is (already) 1.
Depends on bitbake commit 7ae93cf40ab91965147055100432961436bce46c .
(From OE-Core rev: 38180b5c1044be13458fb927ad1babae61e4c51f)
Signed-off-by: Rasmus Villemoes <rv@rasmusvillemoes.dk>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We need this for the new break_hardlinks helper function.
(From OE-Core rev: d6ccefd347bf31acf8d31996d796717acb4da74f)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
In spite of a comment suggesting otherwise, os._exit() does not flush
buffered output from file descriptors before exiting the process like
os.exit() does. This means that any un-flushed output is lost in the
daemon process, in particular the traceback from any thrown exceptions,
making debugging exceptions in the daemon quite difficult.
The solution is to flush stdout and stderr before exiting.
(Bitbake rev: fcc8e1ff53696f78dd64b4ee32f3c433b7a47df0)
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Use == instead of = when comparing task outcome to OUTCOME_FAILED.
Prior to this fix the recipe template would cause a TemplateSyntaxError
exception.
(Bitbake rev: a53ffec4ed3d0f9221bca398e20e8f480fb2b325)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Remove existing files before overwriting them
Archive should extract only the latest same-named entry.
Extracted regular file should not be writtent into existing block
device (or any other one).
https://rt.cpan.org/Ticket/Display.html?id=125523
Affects perl <= 5.26.2
(From OE-Core rev: 037b205e89ae4c7d638a5c15d64f60e16facbaa2)
Signed-off-by: Jagadeesh Krishnanjanappa <jkrishnanjanappa@mvista.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
double64_init: Check psf->sf.channels against upper bound
This prevents division by zero later in the code.
While the trivial case to catch this (i.e. sf.channels < 1) has already
been covered, a crafted file may report a number of channels that is
so high (i.e. > INT_MAX/sizeof(double)) that it "somehow" gets
miscalculated to zero (if this makes sense) in the determination of the
blockwidth. Since we only support a limited number of channels anyway,
make sure to check here as well.
CVE-2017-14634
Closes: #318
Affects libsndfile1 = 1.0.28
(From OE-Core rev: eee93149a49274dc3deed7d89754ee4bda240575)
Signed-off-by: Jagadeesh Krishnanjanappa <jkrishnanjanappa@mvista.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
sfe_copy_data_fp: check value of "max" variable for being normal
and check elements of the data[] array for being finite.
Both checks use functions provided by the <math.h> header as declared
by the C99 standard.
Fixes#317
CVE-2017-14245
CVE-2017-14246
Affects libsndfile1 = 1.0.28
(From OE-Core rev: ad842a3a0e6ef78fb9449362753ae3592c775192)
Signed-off-by: Jagadeesh Krishnanjanappa <jkrishnanjanappa@mvista.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
By including PACKAGECONFIG options, the recipe takes responsibility
for defining the default state of these options. Although the recipe
currently aligns with the gnutls defaults (ie both disabled) tracking
new gnutls releases will be a maintenance effort. Unless there's a
clear reason to do otherwise, it seems safer to leave the choice of
which SSL/TLS versions to enable by default up to the gnutls
developers.
(From OE-Core rev: 4c1d03eb226aa838622852b70a87260ab1ac9d91)
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
>From gnutls 3.5.8 onwards, the code in configure.ac has been passing
"basename $i" to sed, rather than "echo $i". Since the full ${srcdir}
path is not being processed, there's no risk of unexpected matches.
478179316b
(From OE-Core rev: bce938174d1207685c67c40e341a36ab1158e6eb)
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* Merge the two tests together as having them separate is pointless
* Test that ldd runs correctly
* Add in a dependency on the "ldd" package being installed instead of
the sdk tools feature
(From OE-Core rev: 80db456387fb63ee74d53a9719ab3997432f4c80)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Using += with an over-ride can be a source of confusion so try to
avoid the construct in core recipes.
(From OE-Core rev: 2588c2aae11b7b480022dc11575295fdc792bf3f)
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The -config script can now be disabled from configure.
Drop backported patch now merged.
(From OE-Core rev: f2c1d7109014ef9d804c3a6967b96143834ff7bd)
Signed-off-by: Changqing Li <changqing.li@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The perl distribution "XML-Parser" relies for configuration
on the tooling of Devel::CheckLib - which is not aware of
sysroot locations nor of reasonable compiler/link definitions
from outside.
This causes
ERROR: libxml-parser-perl-2.44-r0 do_package_qa: QA Issue: package libxml-parser-perl contains bad RPATH ${BUILDDIR}/tmp/work/core2-64-poky-linux/libxml-parser-perl/2.44-r0/recipe-sysroot/usr/lib in file ${BUILDDIR}/tmp/work/core2-64-poky-linux/libxml-parser-perl/2.44-r0/packages-split/libxml-parser-perl/usr/lib/perl/vendor_perl/5.24.4/auto/XML/Parser/Expat/Expat.so
package libxml-parser-perl contains bad RPATH ${BUILDDIR}/tmp/work/core2-64-poky-linux/libxml-parser-perl/2.44-r0/recipe-sysroot/usr/lib in file ${BUILDDIR}/tmp/work/core2-64-poky-linux/libxml-parser-perl/2.44-r0/packages-split/libxml-parser-perl/usr/lib/perl/vendor_perl/5.24.4/auto/XML/Parser/Expat/Expat.so [rpaths]
ERROR: libxml-parser-perl-2.44-r0 do_package_qa: QA run found fatal errors. Please consider fixing them.
ERROR: libxml-parser-perl-2.44-r0 do_package_qa: Function failed: do_package_qa
It's strongly encouraged to the maintainer @toddr to rework the
toolchain for up to date environments.
[RP: Added fix for nativesdk RPATH issues too]
(From OE-Core rev: b103bb9426c0e5e3ef0fe0c34274ad3a06af8b6a)
Signed-off-by: Jens Rehsack <sno@netbsd.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The modern the time, the improvements in ExtUtils::MakeMaker.
Nowadays, .packlist and perllocal.pod aren't touched anymore when appropriate
flags set during configure stage. Controlling the flags globally avoids
dual-life recipes need share patching.
Further: remove prepending ${PERL_ARCHLIB} in PERL5LIB - it's wrong (search
order is site_lib, vendor_lib, core) - and ${PERL_ARCHLIB} contains core
libpath only ...
(From OE-Core rev: 2e61533e7c1b1cfd49dc771e907207f11a15c44f)
Signed-off-by: Jens Rehsack <sno@netbsd.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The crypto API for AEAD ciphers changed in recent kernels, so that
associated data is now part of both source and destination scatter
gathers. The source, destination and associated data buffers need
to be stiched accordingly for the operations to succeed.
(From OE-Core rev: 76da04571b8cb2241b3f46dec4935ff299639b7d)
Signed-off-by: Hongzhi.Song <hongzhi.song@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
