Ensure that the kernel_version_sanity_check task runs after all source
modifications are complete, including any that are introduced during the
kernel_metadata task. This also avoids any race condition issues when
kernel_version_sanity_check and kernel_metadata tasks are running at the
same time.
On Thu, Sep 29, 2016 at 10:10 AM, Philip Balister wrote:
> This check seems to have gone in recently and doesn't appear to work
> very well.
>
> ERROR: linux-xlnx-3.14-xilinx+gitAUTOINC+2b48a8aeea-r0
> do_kernel_version_sanity_check: Package Version
> (3.14-xilinx+gitAUTOINC+2b48a8aeea) does not m
On 09/29/2016 01:59 PM, Robert Yang wrote:
When the client spawns a pseudo server, it starts out sending diagnostics
to stderr. This can be spammy in some cases with races during startup;
everything resolves, but we get scary-looking diagnostics. So shove
those into a log file.
Signed-off-by:
The last line in the generated /etc/build doesn't end
with a newline anymore, restore it.
Signed-off-by: André Draszik
---
meta/classes/image-buildinfo.bbclass | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/meta/classes/image-buildinfo.bbclass
b/meta/classes/image-buildi
Test creation of mksystemd-bootdisk image.
Signed-off-by: Ed Bartosh
---
meta/lib/oeqa/selftest/wic.py | 7 +++
1 file changed, 7 insertions(+)
diff --git a/meta/lib/oeqa/selftest/wic.py b/meta/lib/oeqa/selftest/wic.py
index e550785..6012abc 100644
--- a/meta/lib/oeqa/selftest/wic.py
+++ b/
Without with-boost-libdir, it checks lib64 libx32 lib lib64 in
sysroot/usr/ and set it as boost lib path if it exist, this is
unreliable when build with multilib.
Fixed when switch baselib from lib64 to lib, for example, when we build
qemux86-64 + multilib setting, the baselib is lib64, and when d
The following changes since commit cdaafc3729700778d95afc2413553d7b41c1317b:
oeqa/sstatetests: Ensure we cover deb packaging backend for sstate test
(2016-09-28 10:15:55 +0100)
are available in the git repository at:
git://git.openembedded.org/openembedded-core-contrib rbt/2fixes
http://
The .pyc files will be generated when "bitbake libxcb", but they can't
be removed when "bitabke xcb-proto -ccleansstate" since they are not
tracked by manifest, create blank .pyc files can fix the problem.
When switch
MACHINE = "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
D
On 29 September 2016 at 10:02, Robert Yang
wrote:
> -EXTRA_OECONF = "--with-boost=${PKG_CONFIG_SYSROOT_DIR}/${prefix}"
> +EXTRA_OECONF = "--with-boost=yes --with-boost-libdir=${PKG_
> CONFIG_SYSROOT_DIR}${libdir}"
>
PKG_CONFIG_SYSROOT_DIR is the wrong variable to use here as that's
explicitly fo
On 29 September 2016 at 10:02, Robert Yang
wrote:
> +do_install_append() {
> +# Create blank .pyc files, so that they will be recorded in
> +# manifest, and can be cleaned correctly.
> +find ${D}${libdir}/xcb-proto/xcbgen -type f -maxdepth 1 -name '*.py'
> -exec touch {}c \;
> +}
>
W
On 09/29/2016 05:10 PM, Burton, Ross wrote:
On 29 September 2016 at 10:02, Robert Yang mailto:liezhi.y...@windriver.com>> wrote:
-EXTRA_OECONF = "--with-boost=${PKG_CONFIG_SYSROOT_DIR}/${prefix}"
+EXTRA_OECONF = "--with-boost=yes
--with-boost-libdir=${PKG_CONFIG_SYSROOT_DIR}${libd
Sorry for slow reply.
This is probably caused by the correction of installation path(from
${STAGING_BINDIR_CROSS}/guile-config to
${SYSROOT_DESTDIR}${bindir_crossscripts}/guile-config). do_populate_sysroot
helps us install things into ${STAGING_BINDIR_CROSS}/guile-config, we shouldn't
do that
The fix for [YOCTO #5935] was applied for mips64 but not for mips64el
Patch it for mips64el
For description of issue, check OE-Core 7a5b6b96
Signed-off-by: Zubair Lutfullah Kakakhel
---
meta/site/mips64el-linux | 1 +
1 file changed, 1 insertion(+)
diff --git a/meta/site/mips64el-linux b/meta/
Test image creation using mksystemd-bootdisk.wks
Signed-off-by: Ed Bartosh
---
meta/lib/oeqa/selftest/wic.py | 8
1 file changed, 8 insertions(+)
diff --git a/meta/lib/oeqa/selftest/wic.py b/meta/lib/oeqa/selftest/wic.py
index 6012abc..b05300c 100644
--- a/meta/lib/oeqa/selftest/wic.py
Improve the buildstats-diff in different ways:
- slightly improved error handling and output
- make it possible to compare other attributes of the buiildstats than just
cputime (block io statistics and wall clock time added)
- support averaging over multiple buildstats
The following changes sin
New class representing buildstats data of a single task.
Signed-off-by: Markus Lehtonen
---
scripts/buildstats-diff | 51 +++--
1 file changed, 32 insertions(+), 19 deletions(-)
diff --git a/scripts/buildstats-diff b/scripts/buildstats-diff
index 7728
Rename these arguments to --min-val and --min-absdiff in preparation for
supporting other "quantities" than just cputime.
Signed-off-by: Markus Lehtonen
---
scripts/buildstats-diff | 24
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/scripts/buildstats-d
Signed-off-by: Markus Lehtonen
---
scripts/buildstats-diff | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/scripts/buildstats-diff b/scripts/buildstats-diff
index 8ee2aaf..d0cd766 100755
--- a/scripts/buildstats-diff
+++ b/scripts/buildstats-diff
@@ -115,11 +115,14 @@ def
These are I/O counter values from /proc//io and represent the
number of bytes read from / written to the storage layer. Default values
for --min-val and --min-absdiff limits are set to 512kB and 128kB,
respectively.
Signed-off-by: Markus Lehtonen
---
scripts/buildstats-diff | 46
Dynamically adjust the width of all fields in task diff output. Makes
it easier to print other units than cputime, too.
Signed-off-by: Markus Lehtonen
---
scripts/buildstats-diff | 60 ++---
1 file changed, 32 insertions(+), 28 deletions(-)
diff --git
A new command line option for choosing which "attribute" of the
buildstats to compare. At first, the already supported 'cputime' is the
only available option. But, refactoring done in this patch should make
it easy to add new attribute types.
Signed-off-by: Markus Lehtonen
---
scripts/buildstats
Two new options, making it possible to compare the number of filesystem
operations of tasks. Defaults for --min-val and --min-absdiff are set to
more or less arbitrary 500 and 50 operations, respectively.
Signed-off-by: Markus Lehtonen
---
scripts/buildstats-diff | 27 ---
Signed-off-by: Markus Lehtonen
---
scripts/buildstats-diff | 38 +-
1 file changed, 21 insertions(+), 17 deletions(-)
diff --git a/scripts/buildstats-diff b/scripts/buildstats-diff
index f26a6c1..3c6cb1e 100755
--- a/scripts/buildstats-diff
+++ b/scripts/build
For comparing the elapsed wall clock time of tests. Default values for
--min-val and --min-absdiff are 5 seconds and 2 seconds.
Signed-off-by: Markus Lehtonen
---
scripts/buildstats-diff | 13 ++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/scripts/buildstats-diff b/
Signed-off-by: Markus Lehtonen
---
scripts/buildstats-diff | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/scripts/buildstats-diff b/scripts/buildstats-diff
index 3c6cb1e..c5c48d3 100755
--- a/scripts/buildstats-diff
+++ b/scripts/buildstats-diff
@@ -26,7 +26,7 @@ from datetim
Makes it possible to average over multiple buildstats. If --multi is
specified (and the given path is a directory) the script will read all
buildstats from the given directory and use averaged values calculated
from them.
All of the buildstats must be from a "similar" build, meaning that no
differ
On 29 September 2016 at 09:38, Ed Bartosh
wrote:
> +def test_mksystemd_bootdisk(self):
> +"""Test creation of mksystemd-bootdisk image"""
> +image = "mksystemd-bootdisk"
> +self.assertEqual(0, runCmd("wic create %s -e core-image-minimal" \
> +
Hi i found some bug in gmock receipe when target architecture is 64bit. I
attach simple fix
How to deploy this fix to jethro/kroghot and master branches ??
--
Pozdrawiam, Mateusz Nowakowski
diff --git a/meta-oe/recipes-test/gmock/gmock/cmake-Add-install-command-for-libraries-and-headers.patch
Add mipsel and mips64el as an option.
Signed-off-by: Zubair Lutfullah Kakakhel
---
scripts/runqemu | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/scripts/runqemu b/scripts/runqemu
index 5095e67..acd4105 100755
--- a/scripts/runqemu
+++ b/scripts/runqemu
@@ -854,7 +854,7
On 29 Sep 2016, at 0:59, Robert Yang wrote:
...t-diagnostics-during-startup-for-pseudo-d.patch | 54
++
...er.c-add-prefix-for-log-and-make-log-cons.patch | 83
++
.../files/0002-Use-correct-file-descriptor.patch | 53
++
meta/recipes-devtools/ps
On Thu, Sep 29, 2016 at 7:54 AM, Mateusz Nowakowski
wrote:
> Hi i found some bug in gmock receipe when target architecture is 64bit. I
> attach simple fix
>
> How to deploy this fix to jethro/kroghot and master branches ??
>
Please send a patch as per
http://www.openembedded.org/wiki/How_to_subm
Two small patches that slightly improve the usability of
build-perf-test-wrapper.sh script.
Markus Lehtonen (2):
build-perf-test-wrapper.sh: check for positional arguments
build-perf-test-wrapper.sh: show defaults for '-a' and '-w'
scripts/contrib/build-perf-test-wrapper.sh | 14
Stricter checking of command line arguments. The script doesn't use any
positional arguments so don't accept any and error out if those are
found.
Signed-off-by: Markus Lehtonen
---
scripts/contrib/build-perf-test-wrapper.sh | 8
1 file changed, 8 insertions(+)
diff --git a/scripts/con
Display default values for '-a' and '-w' command line arguments in the
usage help text.
Signed-off-by: Markus Lehtonen
---
scripts/contrib/build-perf-test-wrapper.sh | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/scripts/contrib/build-perf-test-wrapper.sh
b/scripts/co
StreamHandler was added due missing log information on the console in
oe-selftest with Qemu Runner
Signed-off-by: Francisco Pedraza
---
meta/lib/oeqa/utils/qemurunner.py | 1 +
1 file changed, 1 insertion(+)
diff --git a/meta/lib/oeqa/utils/qemurunner.py
b/meta/lib/oeqa/utils/qemurunner.py
ind
getstatusoutput is a wrapper around subprocess.getstatusouput() in
Py3, which is basically deprecated and behaves almost entirely unlike
run().
[ YOCTO #9342 ]
Signed-off-by: Stephano Cetola
---
meta/lib/oe/utils.py | 11 ---
1 file changed, 4 insertions(+), 7 deletions(-)
diff --git a
This is the beginning of an effort to move toward a unified way of
using the subprocess module. The end goal is somday create a fork
of the subprocess.run command intro'd in python 3.5.
The biggest hurdle in these efforts is in testing some of the more
obscure code paths that are touched when cha
When using subprocess call and check_output, it is better to use arrays
rather than strings when possible to avoid whitespace and quoting
problems.
[ YOCTO #9342 ]
Signed-off-by: Stephano Cetola
---
meta/lib/oe/distro_check.py| 2 +-
meta/lib/oe/package.py | 13 +-
meta/lib
This begins moving away from the deprecated subprocess calls in an
effort to eventually move to some more global abstraction using the run
convenience method provided in python 3.5.
[ YOCTO #9342 ]
Signed-off-by: Stephano Cetola
---
meta/classes/buildstats.bbclass | 6 +-
meta/classes/spdx
On Thu, Sep 29, 2016 at 2:34 PM, Stephano Cetola <
stephano.cet...@linux.intel.com> wrote:
> getstatusoutput is a wrapper around subprocess.getstatusouput() in
> Py3, which is basically deprecated and behaves almost entirely unlike
> run().
>
> [ YOCTO #9342 ]
>
> Signed-off-by: Stephano Cetola
>
This is the beginning of an effort to move toward a unified way of
using the subprocess module. The end goal is somday create a fork
of the subprocess.run command intro'd in python 3.5.
The biggest hurdle in these efforts is in testing some of the more
obscure code paths that are touched when chan
On 09/29, Christopher Larson wrote:
> On Thu, Sep 29, 2016 at 2:34 PM, Stephano Cetola <
> stephano.cet...@linux.intel.com> wrote:
>
> > getstatusoutput is a wrapper around subprocess.getstatusouput() in
> > Py3, which is basically deprecated and behaves almost entirely unlike
> > run().
> >
> > [
When using subprocess call and check_output, it is better to use arrays
rather than strings when possible to avoid whitespace and quoting
problems.
[ YOCTO #9342 ]
Signed-off-by: Stephano Cetola
---
meta/lib/oe/distro_check.py| 2 +-
meta/lib/oe/package.py | 13 +-
meta/lib
This begins moving away from the deprecated subprocess calls in an
effort to eventually move to some more global abstraction using the run
convenience method provided in python 3.5.
[ YOCTO #9342 ]
Signed-off-by: Stephano Cetola
---
meta/classes/buildstats.bbclass | 6 +-
meta/classes/spdx
Add systemd service file for rng-tools.
Signed-off-by: Dengke Du
---
meta/recipes-support/rng-tools/rng-tools/rngd.service | 9 +
meta/recipes-support/rng-tools/rng-tools_5.bb | 14 --
2 files changed, 21 insertions(+), 2 deletions(-)
create mode 100644 meta/recipes
The following changes since commit 5d8a968ecdf40babe452b0ac63d94a7a163710ae:
dev-manual: Applied review changes to GNU debugging section. (2016-09-28
15:02:33 +0100)
are available in the git repository at:
git://git.openembedded.org/openembedded-core-contrib
dengke/add-systemd-service-for-
From: Jackie Huang
We want to provide python libs by default, and some other
popular Linux distributions like redhat/fedora does the same.
Signed-off-by: Jackie Huang
---
meta/recipes-support/boost/boost.inc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/recipes-suppo
47 matches
Mail list logo
