Randy MacLeod [Fri, 17 May 2019 20:10:21 +0000 (16:10 -0400)]
valgrind: add 'file' to ptest depends
to avoid:
./run_openmp_test: line 12: /usr/bin/file: No such file or directory
SKIP: drd/tests/omp_matinv
when using core-image-minimal to run valgrind ptests.
Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Randy MacLeod [Fri, 17 May 2019 20:10:20 +0000 (16:10 -0400)]
valgrind: Make ptest timestamps copasetic
Avoid ptest failures due to .diffs such as
cachegrind/tests/ann1.post.diff:
@ Source file 'a.c' is more recent than input file 'cgout-test'.
@ Annotations may not be correct.
by using:
touch a.c -r cgout-test
Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Fri, 17 May 2019 23:37:43 +0000 (16:37 -0700)]
insane: add sanity checks to SRC_URI
The SRC_URI almost definitely shouldn't be using ${PN}, and GitHub */archive/*
tarballs are dynamically generated so the checksums will change over time.
Detect both of these, and emit a QA warning if found.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
python-numpy: Avoid installing copy of f2py script
Both python-numpy and python3-numpy try to install the file:
${D}/${bindir}/f2py
which causes an error during do_rootfs since the package manager complains about that.
The numpy packages create basically copies of f2py script with the following names:
f2py (on both)
f2py2 (on python 2)
f2py2.7 (on python 2)
f2py3 (on python 3)
f2py3.7 (on python 3)
Remove the copy of f2py from python-numpy in favor of using the one from python3
to avoid installation issues at do_rootfs().
Signed-off-by: Alejandro Enedino Hernandez Samaniego <alejandr@xilinx.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Tom Rini [Sat, 18 May 2019 18:34:13 +0000 (14:34 -0400)]
vim: Rework things so vim adds features not vim-tiny removes
Now that we have a common inc file for both vim and vim-tiny move a few
bits of logic out of vim.inc and in to vim_$VER.bb so that vim adds
things rather than vim-tiny removes them.
Suggested-by: Martin Jansa <martin.jansa@gmail.com> Signed-off-by: Tom Rini <trini@konsulko.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Previously vim-tiny would include the vim recipe wholesale;
this meant that they had to be upgraded in a single transaction
which devtool cannot cope with. This thwarted the latest AUH
run in particular.
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mariano López [Sun, 19 May 2019 21:21:18 +0000 (16:21 -0500)]
util-linux: Add missing ptest dependencies
There are some missing dependencies for the util-linux-ptest package
that causes inconsistencies in the package tests run in different images.
The kernel module in RRECOMMENDS is not build at this time, it needs
more testing and check if the configuration change can be part of the
yocto-kernel-cache repository.
Signed-off-by: Mariano López <just.another.mariano@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When dealing with node modules which use "node-gyp" [1] to build native
addons to node. Some temporary build files stay in the image: object
files, static library files, dependency files, ...
This commit does not keep only the required files, but remove the files
which can leads to QA issues (staticdev with static library files).
[1]: https://github.com/nodejs/node-gyp
Signed-off-by: Jean-Marie LEMETAYER <jean-marie.lemetayer@savoirfairelinux.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When dealing with node modules which have declared "bin" files [1], npm
will create a link in '/usr/bin' with a relative link to
'../lib/node_modules/<module bin file>'.
In order to keep valid links, and to keep it as simple as possible, the
path '/usr/lib/node_modules/' is used as install directory for npm. And
a symbolic link is created to have a valid '/usr/lib/node/' path, needed
for node.
When using npm packages with exotic names, like "JSONSelect" (with
uppercase) or "@angular/cli" (with at sign and slash), there are three
different names:
- the recipe name ("jsonselect" or "angular-cli")
- the npm module name ("JSONSelect" or "@angular/cli")
- the npm pack name ("JSONSelect" or "angular-cli")
Signed-off-by: Marco Felsch <m.felsch@pengutronix.de> Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
qemux86: use a Core 2 Duo CPU instead of the original circa-1993 Pentium
This matches what the qemux86_64 is currently using, and
will allow testing the instructions added in the meantime;
particularly various SSE extensions are now enabled.
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Liwei Song [Thu, 16 May 2019 02:41:40 +0000 (22:41 -0400)]
mdadm: install the systemd service through Makefile
The original mdadm service may include variable like BINDIR,
It should use the real value of it, since the install method
"install-systemd" in Makefile can easily do such work, so don't
install it in bb file manually, use "make install-systemd" to
install all the service of mdadm.
Signed-off-by: Liwei Song <liwei.song@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Haiqing Bai [Thu, 16 May 2019 01:50:12 +0000 (09:50 +0800)]
sysstat: Add PACKAGECONFIG[cron] for '--enable-install-cron' option
If 'cron' is added into PACKAGECONFIG, for sysvinit, the related
cron scripts will be installed; for systemd, the services of
data collect and summary will be installed.
Signed-off-by: Haiqing Bai <Haiqing.Bai@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Tim Orling [Thu, 16 May 2019 00:29:19 +0000 (17:29 -0700)]
libtest-needs-perl: upgrade 0.002005 -> 0.002006
Change SRC_URI to https://cpan.metacpan.org
License-Update: Lines moved further down in README.
Upstream release notes:
0.002006 - 2019-04-04
- added additional references to related modules in pod
- various internal cleanups
- fix backwards compatibility with earlier versions of version.pm
- fix output from test script when tests fail
- added note about RELEASE_TESTING to test message on forced failures
Signed-off-by: Tim Orling <timothy.t.orling@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Fri, 10 May 2019 22:26:53 +0000 (15:26 -0700)]
epiphany: Do not bypass initialization of variable with __attribute__((cleanup))
This is reported with clang e.g.
../../../../../../../workspace/sources/epiphany/lib/ephy-web-app-utils.c:391:5: error: cannot jump from this goto statement to its label
goto out;
^
../../../../../../../workspace/sources/epiphany/lib/ephy-web-app-utils.c:398:20: note: jump bypasses initialization of variable with __attribute__((cleanup))
g_autofree char *app_file = g_build_filename (profile_dir, ".app", NULL);
^
../../../../../../../workspace/sources/epiphany/lib/ephy-web-app-utils.c:385:5: error: cannot jump from this goto statement to its label
goto out;
^
../../../../../../../workspace/sources/epiphany/lib/ephy-web-app-utils.c:398:20: note: jump bypasses initialization of variable with __attribute__((cleanup)) g_autofree char *app_file = g_build_filename (profile_dir, ".app", NULL);
^
Ensure that the initialization is deterministic for goto to work reliably
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ming Liu [Tue, 14 May 2019 18:32:13 +0000 (20:32 +0200)]
dhcp: fix a NSUPDATE compiling issue
This patch fixes a following issue:
| omapip/isclib.c: In function 'dns_client_init':
| omapip/isclib.c:356:18: error: 'dhcp_context_t {aka struct dhcp_context}' has no member named 'dnsclient'
| if (dhcp_gbl_ctx.dnsclient == NULL) {
| ^
| omapip/isclib.c:363:24: error: 'dhcp_context_t {aka struct dhcp_context}' has no member named 'dnsclient'
| &dhcp_gbl_ctx.dnsclient,
| ^
| omapip/isclib.c:364:24: error: 'dhcp_context_t {aka struct dhcp_context}' has no member named 'use_local4'
| (dhcp_gbl_ctx.use_local4 ?
| ^
| omapip/isclib.c:365:25: error: 'dhcp_context_t {aka struct dhcp_context}' has no member named 'local4_sockaddr'
| &dhcp_gbl_ctx.local4_sockaddr
| ^
| omapip/isclib.c:367:24: error: 'dhcp_context_t {aka struct dhcp_context}' has no member named 'use_local6'
| (dhcp_gbl_ctx.use_local6 ?
| ^
| omapip/isclib.c:368:25: error: 'dhcp_context_t {aka struct dhcp_context}' has no member named 'local6_sockaddr'
| &dhcp_gbl_ctx.local6_sockaddr
Signed-off-by: Ming Liu <liu.ming50@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Tue, 14 May 2019 16:32:12 +0000 (09:32 -0700)]
glib-2.0: fix last failing ptest
Upstream glib don't really actually test the test suite with modern glibc and
all the locales present so we're finding a number of bugs. Backport another fix
from upstream to fix the test data with modern glibc.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Tue, 14 May 2019 11:22:14 +0000 (12:22 +0100)]
ptest: Add RDEPENDS frpm PN-ptest to PN package
Many different ptests are breaking as they assume that ${PN}-ptest
depends on ${PN}. It doesn't currently but should. If we fix this, many
different ptests start passing when they previously failed.
It does depend on fixing an issue in the dbus-test recipe which is done
in the preceeding patch (mentioned in case this gets backported).
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The dbus-test package is empty, move its dependencies to the ${PN}-ptest
package. Also ensure that it doesn't depend on the empty ${PN} package
which is about to start causing image failures in the following commit.
In this case the correct dependency is dbus itself.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Randy MacLeod [Tue, 14 May 2019 03:57:16 +0000 (23:57 -0400)]
valgrind: fix call/cachegrind ptests
Use the *_annotate executables from ${bindir}
Actually install cachegrind/tests/a.c since it is used by
the call/cachegrind/tests/ann[12].vgtest files.
With this change and the previous commits, the ptest results
on qemux86-64 when invoked with:
runqemu qemux86 kvm nographic slirp qemuparams="-m 2048"
are:
Randy MacLeod [Tue, 14 May 2019 03:57:15 +0000 (23:57 -0400)]
valgrind: adjust test filters and expected output
Skip two filters in filter_xml_frames since they
are intended to filter filesystem paths under '/usr' that
vary from platform to platform. In the ptest case
for Yocto's valgrind, the ptest executables are placed under:
/usr/lib/valgrind/ptest
and if these frames are filtered out, then several 'drd' tests fail
the comparision between expected and actual output.
Also adjust the std_list expected output to agree with that
produced when the --yocto-ptest option to vg_regtest is used.
Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Randy MacLeod [Tue, 14 May 2019 03:57:14 +0000 (23:57 -0400)]
valgrind: update the ptest subdirs list
Split the list into one directory per line.
Sort the list alphabetically.
Add the gdbserver_tests directory to get the 'gone' executable
which is needed by:
memcheck/tests/gone_abrt_xml.vgtest
Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Tue, 14 May 2019 03:57:13 +0000 (23:57 -0400)]
valgrind: Include debugging symbols in ptests
About half the ptests will fail if the executables deployed
as part of the ptest package are stripped. We therefore need
to add a dependency on the dbg symbols package and silence the
QA test which would complain about this since we really do want
it in this case.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 15 May 2019 14:45:14 +0000 (15:45 +0100)]
core-image-sato-sdk-ptest: Tweak size to stay within 4GB limit
Adding the valgrind debug symbol information caused the genericx86-64 image to
overflow the 4GB boundary. Tweak the sizes to avoid autobuilder failures yet
leave enough space all the tests still run successfully.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Martin Jansa [Tue, 14 May 2019 07:30:30 +0000 (07:30 +0000)]
perf: make sure that the tools/include/uapi/asm-generic directory exists
* before trying to copy unistd.h into it
* older kernels don't have uapi in tools/include and do_configure fails
with:
DEBUG: Executing shell function do_configure
cp: cannot create regular file '.../perf/1.0-r9/perf-1.0/tools/include/uapi/asm-generic/unistd.h':
No such file or directory
WARNING: exit code 1 from a shell command.
* tools/include/uapi was added in kernel 4.8 with
https://github.com/torvalds/linux/commit/c4b6014e8bb0c8d47fe5c71ebc604f31091e5d3f
tools: Add copy of perf_event.h to tools/include/linux/
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Originally a web browser was added to build-appliance-image to
enable Hob's use cases that involve opening a URI; with Hob long gone,
a web browser is no longer necessary to have in build-appliance-image.
This will also address the out-of-resources problem when
build-appliance-image builds itself in a VM, as a test case.
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changelog:
- Fix issue with DHCP client and handling renewing state
- Fix issue with DHCP client and handling rebinding state
- Fix issue with DHCP client and recommended retry timeouts
- Fix issue with Generic Netlink and family discovery
Signed-off-by: Oleksandr Kravchuk <open.source@oleksandr-kravchuk.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Hongxu Jia [Mon, 13 May 2019 15:12:46 +0000 (11:12 -0400)]
oeqa/targetcontrol.py: fix qemuparams not work in runqemu with launch_cmd
As runqemu with launch_cmd means directly run the command, don't need set
rootfs or env vars.
Since commit [a847dd7202 runqemu: Let qemuparams override default settings]
applied in oe-core, if launch_cmd contains "qemuparams='***'", it does not
work, which is overridden by latter qemuparams="-serial tcp:127.0.0.1" in
QemuRunner.launch();
So we set qemuparams as a parameter in runqemu, the fix makes it work
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sun, 12 May 2019 18:49:25 +0000 (19:49 +0100)]
oeqa/target/ssh: Replace suggogatepass with ignoring errors
We continued to see encoding problems with ssh commands run in oeqa. After much
research the conclusion was we should use ignore the errors since some occasional bad
locale encoding is better than the unicode decoding issues we were seeing which crashed
large parts of tests.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Breno Leitao [Mon, 13 May 2019 14:26:52 +0000 (15:26 +0100)]
weston-init: Fix WESTON_USER typo
Commit 837c786d600ba69('weston-init: Add support for non-root start') added a
typo that uses WEST_USER instead of WESTON_USER variable when chwon'ing the
XDG_RUNTIME_DIR directory. Since WEST_USER is not defined, it will `chown
:$WESTON_USER file`, which will work, but that is not 100% correctly and should
be fixed.
This patch basically fix the typo and now the file will be chown'ed to the
WESTON_USER user.
Signed-off-by: Breno Leitao <leitao@debian.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Kai Kang [Tue, 23 Apr 2019 10:12:16 +0000 (06:12 -0400)]
webkitgtk: fix compile error for arm64
It removes function JSC::AssemblerBuffer::data() for ARM64 in commit
https://trac.webkit.org/changeset/236589/webkit. But it is required by
Cortex A53 from https://trac.webkit.org/changeset/175514/webkit and
fails to compile for arm64:
| .../tmp/work/aarch64-poky-linux/webkitgtk/2.24.0-r0/webkitgtk-2.24.0/Source/JavaScriptCore/assembler/ARM64Assembler.h:3769:100: error: 'class JSC::AssemblerBuffer' has no member named 'data'
| if (UNLIKELY((*reinterpret_cast_ptr<int32_t*>(reinterpret_cast_ptr<char*>(m_buffer.data()) + m_buffer.codeSize() - sizeof(int32_t)) & 0x0a000000) == 0x08000000))
Not set WTF_CPU_ARM64_CORTEXA53 for arm64 to fix the failure.
Signed-off-by: Kai Kang <kai.kang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
- Use autotools-brokensep to replace autotools to workaround failure
caused by out of tree
...
| rm -f lib/alloca.h-t lib/alloca.h && \
| { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \
| cat ../groff-1.22.4/lib/alloca.in.h; \
| } > lib/alloca.h-t && \
| mv -f lib/alloca.h-t lib/alloca.h
| /bin/sh: line 4: lib/alloca.h-t: No such file or directory
| Makefile:10407: recipe for target 'lib/alloca.h' failed
...
- Add `--without-doc' to not use target groff to generate doc at build time,
since upstream commit [cfe916e Support of configure option to build the
documentation.]
- Remove groff depends groff-native, and add DEPENDS bison-native
- Add 0001-fix-shebang-for-taget.patch
- Add 0001-support-musl.patch
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Hongxu Jia [Sun, 12 May 2019 08:16:27 +0000 (16:16 +0800)]
subversion: upgrade 1.11.1 -> 1.12.0
- Backport a patch to fix build failure while APR 1.7.0
...
checking for apr_int64_t Python/C API format string...
configure: error: failed to recognize APR_INT64_T_FMT on this platform
...
- Rebase disable_macos.patch and serfmacro.patch
License-update: no change, declare two new added file
* in build/ac-macros/ax_boost_base.m4
* in build/ac-macros/ax_boost_unit_test_framework.m4
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
