Ed Bartosh [Fri, 2 Dec 2016 23:48:06 +0000 (01:48 +0200)]
selftest: wic: fix test_qemu
Setting WKS_FILE variable in qemux86-64 made wic test to
use wrong wks file to produce an image and resulted in
test_qemu failure.
Used conditional assignment in qemux86-64 and explicitly
set WKS_FILE in wic testing suite to make the suite to use
wic-image-minimal.wsk. This should fix test_qemu failure.
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Thu, 1 Dec 2016 08:12:38 +0000 (00:12 -0800)]
kernel.bbclass: let do_deploy depend on do_packagedata
The do_deploy runs depmodwrapper which requires
pkgdata/kernel-depmod/kernel-abiversion, which is generated by
PACKAGEFUNCS. So we need let do_deploy depend on do_packagedata.
This can fix the errors which happens sometimes when kernel upgrades:
DEBUG: Executing shell function do_deploy
Error: Kernel version 4.8.8-WR9.0.0.1_standard does not match kernel-abiversion (4.8.8-WR9.0.0.0_standard)
And we only see this error when kernel upgrades and rebuild, but doesn't
see it in a normal build, this is becuase depmodwrapper doesn't exit
error when kernel-depmod/kernel-abiversion doesn't exit, it just prints
an error which should go into log.do_deploy:
if [ ! -r /path/to/sysroots/qemux86-64/pkgdata/kernel-depmod/kernel-abiversion ]; then
echo "Unable to read: /path/to/sysroots/qemux86-64/pkgdata/kernel-depmod/kernel-abiversion" >&2
else
[foo]
fi
We can see that there is no "exit 1", I guess it was designed to let it
can run without kernel-abiversion
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
André Draszik [Thu, 1 Dec 2016 11:23:22 +0000 (11:23 +0000)]
arch-mips: sort new MACHINEOVERRIDES by priority
While I couldn't find explicit documentation, it appears
that the list of MACHINEOVERRIDES should be sorted from
less specific to more specific left to right, so that
more specific overrides take precedence.
Signed-off-by: André Draszik <adraszik@tycoint.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Thu, 1 Dec 2016 09:40:46 +0000 (10:40 +0100)]
libarchive: enable non-recursive extract/list
Required for meta-swupd performance enhancements: in meta-swupd, the
so called "mega" image contains a rootfs with all files that can
potentially be installed on a device. Other virtual image recipes need
a subset of those files or directories, and a partial extraction from
a single tar archive is faster than letting all virtual image recipes
share access to a directory under a single pseudo instance.
It may be necessary to extract a directory with all of its attributes
without the content of the directory, hence this patch. Upstream
agreed to consider merging such a patch (see
https://groups.google.com/forum/#!topic/libarchive-discuss/JO3hqSaAVfs)
but has been slow in actually commenting on it, so for now it has
to be carried as distro patch.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Thu, 1 Dec 2016 05:56:01 +0000 (21:56 -0800)]
ffmpeg: Add packageconfig for vdpau accelaration
Fixes dangling dependency QA issue
WARNING: ffmpeg-3.2.1-r0 do_package_qa: QA Issue: libavutil rdepends on libvdpau, but it isn't a build dependency, missing libvdpau in DEPENDS or PACKAGECONFIG? [build-deps]
Jussi Kukkonen [Fri, 4 Nov 2016 13:16:49 +0000 (15:16 +0200)]
run-postinsts: Use opkg/dpkg to configure when possible
Currently run-postinsts script has code to run postinst scripts
via opkg/dpkg configure but that code is never used. The advantage
of using package managers instead of just executing the scripts is
to keep the package manager DB updated.
Fix the script so that the package managers are used when appropriate.
Also use $localstatedir for the opkg runtime file location.
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Andre McCurdy [Tue, 29 Nov 2016 18:36:46 +0000 (10:36 -0800)]
tune-cortexa*.inc: squash whitespace within TUNE_FEATURES strings
TUNE_FEATURES is include in BUILDCFG_VARS, so any whitespace is
visible to the user during the build process. Remove the extra
whitespace added during the 2.1 development cycle:
Signed-off-by: Andre McCurdy <armccurdy@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Wed, 30 Nov 2016 09:50:11 +0000 (10:50 +0100)]
pybootchartgui: support reading reduced /proc logs
Pre-processing /proc data during the build considerably reduces the
amount of data written to disk: 176KB instead of 4.7MB for a 20
minuted build. Parsing also becomes faster.
buildstats.bbclass only writes the reduced logs now, but support for
the full /proc files is kept around as reference.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Wed, 30 Nov 2016 09:50:10 +0000 (10:50 +0100)]
buildstats: reduce amount of data stored for system utilization
Pre-processing /proc data during the build considerably reduces the
amount of data written to disk: 176KB instead of 4.7MB for a 20
minuted build. Parsing also becomes faster.
The disk monitor log added another 16KB in that example build. The
overall buildstat was 20MB, so the overhead for monitoring system
utilization is small enough that it can be enabled by default.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Wed, 30 Nov 2016 09:50:09 +0000 (10:50 +0100)]
pybootchartgui: simplify drawing of memory usage
The internal representation after parsing now matches exactly
what the drawing code needs, thus speeding up drawing a bit.
However, the main motivation is to store exactly that required
information in a more compact file.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Wed, 30 Nov 2016 09:50:08 +0000 (10:50 +0100)]
pybootchartgui: render disk space usage
This adds a new, separate chart showing the amount of disk space used
over time for each volume monitored during the build. The hight of the
graph entries represents the delta between current usage and minimal
usage during the build.
That's more useful than showing just the current usage, because then a
graph showing changes in the order of MBs in a volume that is several
GB large would be just flat.
The legend shows the maximum of those deltas, i.e. maximum amount of
space needed for the build. Minor caveat: sampling of disk space usage
starts a bit later than the initial task, so the displayed value may
be slightly lower than the actual amount of space needed because
sampling does not record the actual initial state.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Wed, 30 Nov 2016 09:50:06 +0000 (10:50 +0100)]
buildstats: record disk space usage
Hooks into the new monitordisk.py event and records the used space for
each volume. That is probably the only relevant value when it comes to
visualizing the build and recording more would only increase disk
usage.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Wed, 30 Nov 2016 09:50:04 +0000 (10:50 +0100)]
pybootchartgui: show system utilization
This enables rendering of the original bootchart charts for CPU, disk
and memory usage. It depends on the /proc samples recorded by the
updated buildstats.bbclass. Currently, empty charts CPU and disk usage
charts are drawn if that data is not present; the memory chart already
gets skipped when there's no data, which will also have to be added
for the other two.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Wed, 30 Nov 2016 09:50:03 +0000 (10:50 +0100)]
pybootchartgui/draw.py: fix drawing of samples not starting at zero
The code did not handle x scaling correctly when drawing starts at
some time larger than zero, i.e. it worked for normal bootchart data,
but not for the system statistics recorded by buildstats.bbclass.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Wed, 30 Nov 2016 09:50:02 +0000 (10:50 +0100)]
pybootchartgui/draw.py: allow moving process chart up and down
Substracting curr_y when determining the hight of the process chart is
wrong because the height is independent of the position where the
chart is about to be drawn. It happens to work at the moment because
curr_y is always 10 when render_processes_chart() gets called. But it
leads to a negative height when other charts are drawn above it, and
then the grid gets drawn on top of those other charts.
Substracting some constant is relevant because otherwise the box is
slightly larger than the process bars. Not sure exactly where that
comes from (text height?); leg_s seems a suitable constant and happens
to be 10, so everything still gets rendered exactly as before.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Wed, 30 Nov 2016 09:50:01 +0000 (10:50 +0100)]
buildstats: add system state sampling
/proc/[diskstats|meminfo|stat] get sampled and written to the same
proc_<filename>.log files as during normal bootchat logging. This will
allow rendering the CPU, disk and memory usage charts.
Right now sampling happens once a second, triggered by the heartbeat
event.That produces quite a bit of data for long builds, which will be
addressed in a separate commit by storing the data in a more compact
form.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jair Gonzalez [Wed, 30 Nov 2016 20:43:02 +0000 (14:43 -0600)]
parselogs: Whitelist GPT warnings as the device is fully functional
The warning occurs when the GPT image is not the same size than the
media into which it's being flashed, causing the backup GPT table
not being at the end of the disk. However, this is expected as the
image is created before having the information about the destination
media. The error is harmless, so it will be whitelisted.
The iwlwifi module of any given kernel has a minimum and maximum
supported firmware version. The kernel begins by attempting to load the
maximum version, and decrements until it is successful. The 4.8 kernel's
maximum supported firmware version is 24, but thus far only 22 has been
released, meaning we get errors for 24 and 23.
Filter out iwlwifi firmware load error messages, as they are not
necessarily indicative of real problems.
Signed-off-by: California Sullivan <california.l.sullivan@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When creating a patch set with cover letter using the
send-pull-request script, both the "In-Reply-To" and "References"
headers are appended twice in patch 2 and subsequent.
That's because git-format-patch already inserted them and then
git-send-email repeats that. Suppressing mail threading in
git-send-email with --no-thread avoids the problem and is the
right solution because it works regardless whether git-send-email is
called once or twicee.
Repeating these headers is a violation of RFC 2822 and can confuse
mail programs. For example, Patchwork does not detect a patch series
problem when there are these extra headers.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Fri, 18 Nov 2016 15:23:22 +0000 (16:23 +0100)]
bitbake: codeparser.py: support deeply nested tokens
For shell constructs like
echo hello & wait $!
the process_tokens() method ended up with a situation where "token"
in the "name, value = token" assignment was a list of tuples
and not the expected tuple, causing the assignment to fail.
There were already two for loops (one in _parse_shell(), one in
process_tokens()) which iterated over token lists. Apparently the
actual nesting can also be deeper.
Now there is just one such loop in process_token_list() which calls
itself recursively when it detects that a list entry is another list.
As a side effect (improvement?!) of the loop removal in
_parse_shell(), the local function definitions in process_tokens() get
executed less often.
Michael Wood [Thu, 24 Nov 2016 11:20:04 +0000 (11:20 +0000)]
bitbake: toaster: buildinfohelper Simplify layer event to toaster layer function
Simplify the layer event information to layer version object in toaster
function. Previously this attempted many different methods of trying to
obtain the correct layer from toaster by manipulating the data from the
event or the data from the known layers to try and match them together.
We speed up and simplify this process by making better use of django's
orm methods and by working down the most likely matching methods in order
of accuracy.
Michael Wood [Thu, 24 Nov 2016 11:20:03 +0000 (11:20 +0000)]
bitbake: toaster: bldcontrol Move CustomImageRecipe file creation into own function
Move the custom image file creation (i.e. create the layer file
structure, conf and recipe file) into it's own function and remove the
creation of the BRLayer as this is done at schedule_build just like all
the other layers.
Fix a bug where the toaster-custom-images layer was always being appened
to the layer list if the directory exists.
Michael Wood [Thu, 24 Nov 2016 11:20:02 +0000 (11:20 +0000)]
bitbake: toaster: orm models Handle CustomImageRecipe BRLayer here
The schedule_build function on the project object is where the BRLayers
are created for the build. Instead of creating the BRLayer for the
CustomImageRecipe in the localhostbbcontroller create it here so that
all that mechanism is in one place.
This fixes the unidentified layers issue by making the
toaster-custom-images layer a local layer. By doing this we also fix the
git assumptions made for the local layers which stop recipes and other
meta data being associated with them. This also removed some of the
special casing previously needed when we didn't have the concept of a
local (non git) layer.
Also rename created flag var to a have a different var for each returned
value so that the same value isn't used multiple times.
Sujith H [Thu, 24 Nov 2016 11:19:58 +0000 (11:19 +0000)]
bitbake: toaster: localhostbecontroller accept custom init script for build
When passed variable CUSTOM_BUILD_INIT_SCRIPT to toaster
setting, it would be nice to use it. Else toaster
can use oe-init script. This gives an oppurtunity to
use customized build init scritps.
Reyna, David [Thu, 24 Nov 2016 11:19:56 +0000 (11:19 +0000)]
bitbake: toaster: orm gen_layerdeps Protect against circular Layer dependencies
Limit the recursion (to say 20 levels) when processing layer dependencies
so that circular dependecies do not cause infinite decent and an
out-of-memory failure. The duplicate found layers are already immediately
filtered in the code.
Michael Wood [Thu, 24 Nov 2016 11:19:55 +0000 (11:19 +0000)]
bitbake: toaster: customrecipejs Consume click event on 'a' link if disabled
Consume the click event on the download recipe link if it's disabled. To
prevent the link from sending user to an error page.
See http://getbootstrap.com/css/#forms-disabled-fieldsets and a link
caveat.
Michael Wood [Thu, 24 Nov 2016 11:19:52 +0000 (11:19 +0000)]
bitbake: toaster: runbuilds Write the pidfile in python rather than shell script
Write the pid file out in the start up of this management command. This
ensures this has happened instead of relying on the shell command having
been run which may or may not be the case. This also makes it simpler for
testing.
Couple of clean ups of runbuilds as identified by pyflake
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The copyright for the software has been transferred to Unicode Inc from IBM,
but the terms are same.
libiculx and libicule are no longer produced as they depend on an external
package icu-le-hb (previous versions had an option of using an internal
implementation which now has been dropped). I have verified that icu
dependencies in oe-core and meta-oe still build.
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Yi Zhao [Mon, 28 Nov 2016 09:55:40 +0000 (17:55 +0800)]
libxml2: Security fix CVE-2016-5131
CVE-2016-5131 libxml2: Use-after-free vulnerability in libxml2 through
2.9.4, as used in Google Chrome before 52.0.2743.82, allows remote
attackers to cause a denial of service or possibly have unspecified
other impact via vectors related to the XPointer range-to function.
Signed-off-by: Yi Zhao <yi.zhao@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Martin Vuille [Sun, 27 Nov 2016 20:02:37 +0000 (15:02 -0500)]
terminal.py: Pass string instead of bytes to ExecutionError to avoid exception
Based on run() in bitbake/lib/bb/process.py, ExecutionError() expects strings
not bytes. Passing bytes results in a "TypeError: Can't convert 'bytes' object
to str implicitly" exception.
Signed-off-by: Martin Vuille <jpmv27@yahoo.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Zhixiong Chi <zhixiong.chi@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Zhixiong Chi [Mon, 28 Nov 2016 08:12:04 +0000 (16:12 +0800)]
tiff: Security fix CVE-2016-9540
tools/tiffcp.c in libtiff 4.0.6 has an out-of-bounds write on tiled
images with odd tile width versus image width. Reported as MSVR 35103,
aka "cpStripToTile heap-buffer-overflow."
Signed-off-by: Zhixiong Chi <zhixiong.chi@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Aníbal Limón [Thu, 24 Nov 2016 21:06:04 +0000 (15:06 -0600)]
perl: Upgrade to 5.24.0
Configuration changes,
Simple changes was made to bump version and api version, related to
floating point handling now the configuration needs the inf, mantisa
and nan bytes.
The new version comes with the support of API calls like memmem and
{new,free,use}locale also structure for handle siginfo supported by
glibc and musl.
Finally use64bit{int, all} was disable because the previous
configure_args don't come with them and cases some tests to fail
related to bignum's and shared memory respectively. This doesn't
means that perl couldn't use 64-bit data types, it means that don't
stores by default into a 64 bit that is good for embedded space
purposes.
Modules changes,
Some core modules are now deprecated in order to use the core ones
like version-vpp and version-regex inside module-extutils-makemaker.
For full review see perl-rdepends.inc file.
Patches removed, comes with the upgrade now:
- perl/perl-remove-nm-from-libswanted.patch
- perl/perl-fix-CVE-2015-8607.patch
- perl/perl-fix-CVE-2016-2381.patch
Test,
The upgrade was test using ptest the suite is fixed now.
The pod2man and pod2text installation required now for some tests.
Buildhistory was use to review the changes and only diff changes
related to modules commented above.
Mark Asselstine [Fri, 25 Nov 2016 04:17:14 +0000 (23:17 -0500)]
systemd.bbclass: don't block on service restart
The current class works fine when a recipe uses SYSTEMD_AUTO_ENABLE
'enable' and has no on device pkg_postinst(), ie when the postinst is
run as part of rootfs creation. However, when there is a component of
pkg_postinst() that is run on device the 'systemctl restart' is run as
part of the run_postinsts.service at boot. This results in the boot
spinning indefinitely with:
[ *** ] A start job is running for Run pending postinsts (7s / no limit)
The issue could potentially be that the packages service has an
'After' clause which comes later in the boot, beyond
run_postinsts.service, creating a chicken before the egg
scenario. Even service files without an 'After' clause cause this
situation however. Despite this not being the cause of the issue this
fix will prevent this scenario from happenning.
Using strace we are able to find that during boot, when
run_postinsts.service is running attempting to start or restart any
service will result in the call get stuck on poll(). Since the
run_postinsts.service does not monitor the outcome of the call to
restart we can work around this by using '--no-block'.
Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Jose Perez Carranza <jose.perez.carranza@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Patrick Ohly [Thu, 24 Nov 2016 09:51:13 +0000 (10:51 +0100)]
qemuboot.bbclass: do not hook into do_rootfs
Writing qemuboot.conf in write_qemuboot_conf() does not modify the
rootfs and thus conceptually shouldn't be executed as part of rootfs
creation.
Running it as separate task is cleaner and fixes the problem of
missing qemuboot.conf files for meta-swupd virtual images; those
images replace do_rootfs and ROOTFS_POSTPROCESS_COMMANDs don't run at
all.
The task gets added such that it runs roughly at the same time as
before. Probably it doesn't actually need to depend on do_rootfs, but
this way we don't write a useless qemuboot.conf in cases where
do_rootfs fails.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: David Vincent <freesilicon@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Carlos Alberto Lopez Perez <clopez@igalia.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Samuli Piippo [Wed, 23 Nov 2016 14:36:37 +0000 (16:36 +0200)]
mesa: make sure GLES3 headers are installed
Since there is no libgles3-mesa package that would pull in the headers,
add dependency to libgles2-mesa-dev. Now there no need to manually add
GLES3 headers to image or toolchain.
Ross Burton [Thu, 24 Nov 2016 20:37:48 +0000 (20:37 +0000)]
insane: rewrite the expanded-d test as a QAPKGTEST
Instead of being executed for every file in every package, this is now just
called for each package. It is also now correctly called for packages which
don't have any content but do have postinst scripts.
Ross Burton [Thu, 24 Nov 2016 20:26:44 +0000 (20:26 +0000)]
insane: add QAPKGTEST, a package-wide equivilant to QAPATHTEST
QAPATHTEST defines a function that is executed for every file in every package.
For tests which just need to look at the datastore this is massive overkill.
Add QAPKGTEST, which is invoked for each package in the recipe.
Ross Burton [Thu, 24 Nov 2016 11:24:26 +0000 (11:24 +0000)]
insane: fix expanded-d test
This test should be looking for the expanded value of ${D} in the expanded value
of pkg_postinst and so on, but one of the getVar() calls was passing
expand=False so the test would never be true.
Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 23 Nov 2016 13:33:39 +0000 (13:33 +0000)]
gnomebase: remove gnome-common-native DEPENDS
gnome-common is deprecated upstream and an increasingly smaller fraction of
GNOME packages still use it. To avoid adding it redundantly to DEPENDS, remove
it from the class.
Ross Burton [Wed, 23 Nov 2016 13:29:45 +0000 (13:29 +0000)]
meta: add explicit gnome-common dependencies
As gnome-common is deprecated it is being removed from GNOME packages, so to
avoid redundantly pulling it in as a build dependency to every package it will
be removed from gnomebase.bbclass
Add it explicitly to the recipes that still need it so these still build.
projects / thirdparty / openembedded / openembedded-core-contrib.git / log
