Bruce Ashfield [Mon, 28 Apr 2014 16:09:33 +0000 (12:09 -0400)]
linux-yocto-custom: provide defconfig guidance
As reported by Andrea, defconfigs created with savedefconfig should specify
their expansion mode as "--alldefconfig' for custom yocto recipes. To ensure
that this is documented, we can add a comment in the skeleton recipe.
Reported-by: Andrea Adami <andrea.adami@gmail.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Bruce Ashfield [Thu, 24 Apr 2014 18:09:39 +0000 (14:09 -0400)]
kernel-yocto: quote kconfig mode check
We allow inheriting recipes to control the kconfig mode used by merge_config.sh
via the KCONFIG_MODE variable. An error crept into the variable reference, and
since it is not quoted, the true condition always runs.
The result is that operations without an explicit kconfig mode cannot trigger
allnoconfig for defconfig builds, which can result in some options being
dropped from the final .config.
Quoting the reference allows it to evaluate properly.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
gcc 4.9 now generates slim objects when LTO is used
These slim objects only contain intermediate language
representation for LTO. Use -ffat-lto-objects to create files which
contain additionally the object code. When linking some .so file, the option
'-flto' is missed, so the error happens.
We add '-ffat-lto-objects' flag to make gcc 4.9 behaves the same as gcc 4.8.
Richard Purdie [Sun, 4 May 2014 16:36:10 +0000 (17:36 +0100)]
sstatetests: Use TUNE_ARCH instead of TARGET_ARCH
The scripts extact variable values from bitbake -e. Unfortunately
TARGET_ARCH is unset from that environment so we use TUNE_ARCH instead
which will have the value we need.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Tudor Florea [Fri, 2 May 2014 22:21:23 +0000 (00:21 +0200)]
curl: Backport a fix for a build issue
mkhelp: generate code for --disable-manual as well
This allows configure --disable-manual to run and build without having
to regenerate the src/tool_hugehelp.c file which otherwise is necessary
since we ship tarballs with that file present.
Signed-off-by: Tudor Florea <tudor.florea@enea.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sat, 3 May 2014 10:14:47 +0000 (11:14 +0100)]
gcc-common: Ensure checksums don't change to match old behaviour
There is a fix about to go into bitbake to ensure that datastores
being accessed with a name other than "d" are correctly reflected
in checksums. This will cause this function to add in a number of
dependencies we don't want.
These do need to be properly unravelled in due course but would
only really affect multilib builds. For now therefore just exclude
the variables as per the old behaviour.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Max Eliaser [Thu, 1 May 2014 20:00:28 +0000 (13:00 -0700)]
Add texinfo.bbclass; recipes that use texinfo utils at build-time inherit it.
The class itself currently does nothing. The idea is to mark all recipes that
make use of the texinfo utilities. In the future, this class could be used to
suppress the generation/formatting of documentation for performance,
explicitly track dependencies on these utilities, and eliminate Yocto's
current dependency on the host system's texinfo utilities.
Signed-off-by: Max Eliaser <max.eliaser@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Thu, 1 May 2014 15:40:19 +0000 (16:40 +0100)]
gcc-common: Only apply fpu settings to target gcc
Within the OE build environment, we supply the correct fpu settings. These
only need to be spelt out for the on-target gcc.
Doing this means the checksums for the core compiler don't depend on the fpu
settings. We exclude the compiler tunes for similar reasons, it doesn't need
to influence the compiler build.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Thu, 1 May 2014 14:41:12 +0000 (15:41 +0100)]
cross: Update PKG_CONFIG variables after recent changes
The PKG_CONFIG definitions should now more closely match those from native.bbclass.
We can't use ${libdir} here since it contains the target arch prefix.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Otavio Salvador [Thu, 1 May 2014 19:38:04 +0000 (16:38 -0300)]
sanity: Bump BitBake minimal version to 1.23.0
The metadata started using the bb.utils.contains_any which has been
add in the BitBake 1.23.0 release, set this as the minimal version to
ensure parsing compatibility.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 30 Apr 2014 12:32:04 +0000 (13:32 +0100)]
oeqa: add proper handling for command errors where needed
For use outside of tests themselves, we want a better error than
AssertionError, so create one and allow us to request it when calling
runCmd(). This enables us to avoid tracebacks during master image
operations if the power control command fails.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 30 Apr 2014 12:32:03 +0000 (13:32 +0100)]
scripts/contrib/dialog-power-control: add a trivial power prompt script
If you want to do automated hardware testing but don't have a
controllable power strip this script can be useful so that you know when
you need to cycle the power.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 30 Apr 2014 12:32:02 +0000 (13:32 +0100)]
scripts/contrib/serdevtry: add script to handle transient serial terminals
When running automated tests (or just generally interacting with)
boards whose serial console devices are on the board itself and thus
disappear when powered down or practically disconnected, such as the
BeagleBone white, some terminal programs (e.g. picocom) will exit when
the device disappears and need to be restarted after the serial device
returns. This script handles this automatically for such terminal
programs.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 30 Apr 2014 12:32:01 +0000 (13:32 +0100)]
classes/testimage: if start fails, don't try to stop
If we couldn't start the target, it doesn't make sense to try and stop
it here since logically it shouldn't now be in any kind of "started"
state. (It's the start function's job to clean up after itself if it
fails - to that end, fix up the QemuTarget class so that it does.)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 30 Apr 2014 12:32:00 +0000 (13:32 +0100)]
oeqa/controllers/masterimage: more robust master image startup
Instead of powering up the target when the object is constructed, wait
until deploy is called. Then there are basically two different
scenarios:
a) The device is booted into the master image already, in which case
we can just use it
b) The device is booted into another image or can't be contacted, in
which case we need to power cycle it. Here we also now wait until it
has booted up instead of trying to contact it immediately.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
and for a serial console server: "telnet 10.11.12.13 7003" or some conmux
command.
Signed-off-by: Stefan Stanacar <stefanx.stanacar@intel.com> Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Stefan Stanacar [Wed, 30 Apr 2014 12:31:57 +0000 (13:31 +0100)]
oeqa/controllers/masterimage: add a base class for hw targets
Right now GummibootTarget is the only hardware TEST_TARGET with deployment,
but we will add more, so let's make an abstract base class, that will
do the common thing for all the hw targets.
Signed-off-by: Stefan Stanacar <stefanx.stanacar@intel.com> Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Wed, 30 Apr 2014 12:56:33 +0000 (13:56 +0100)]
sqlite3: restore upstream version numbering
Regardless of the numbering used in source archive filenames, the
upstream version number is in dotted form in documentation, release
notes and the website, so we need to be using that here since that is
what people will expect to see.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Removed the following patch(es):
* sysroot-arg.patch (changes included in release)
Correct the fccache.c license checksum and the line numbers because the
license snippet has moved and, also, at the previous upgrade the lines
were wrong (shifted a couple of lines).
Richard Purdie [Sun, 27 Apr 2014 00:22:10 +0000 (01:22 +0100)]
gcc: Drop ARCH_FLAGS_FOR_TARGET usage
As far as I can tell this variable is now completely unneeded. It would
only ever get used in target builds and these are now correctly done
in the target environment namespace, not any of our cross environments.
As such, CC and other variables contain the correct compilers and other
tune options and these are correctly picked up when building libgcc,
libstdc++ and others.
I tried to figure out where else these would make any sense and couldn't
find anything. Builds appear fine without them so lets drop the complexity
including the patch adding in this flag to gcc.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sun, 27 Apr 2014 00:24:31 +0000 (01:24 +0100)]
gcc-common/gcc-configure-common: Move gnu-configize to its own shared task
This command modifies ${S} and can race against other tasks running do_configure and
having the scripts disappear from under them. To avoid this move to its own
task and work on the shared work directory as a common task.
It needs to be a python task to avoid lots of shell exported variables as
dependencies.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sun, 27 Apr 2014 00:23:46 +0000 (01:23 +0100)]
gcc-target: Limit compile to host targets, don't build runtimes.
Currently the gcc builds are building copies of the target libraries
that we never use (it isn't installed in do_install). This is a rather
pointless waste of cpu time.
Instead just compile the host targets. Comparing the package output of
this compared to a previous build shows that the unwind.h header is
missing since its provided by gcc. Fix this simply by copying it in.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Irina Patru [Mon, 28 Apr 2014 11:08:46 +0000 (14:08 +0300)]
distrodata.bbclass: Fix checkpkg functionality
Currently it wasn't working because *COMMAND variables were removed
from fetcher.
Now checkpkg sets the command internally and sends it as a parameter
to _runwget() function from wget fetch.
Signed-off-by: Irina Patru <irina.patru@intel.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Update ltp version to 20140422
Removed two patches that already exist in the current version:
1. regen-makefile.patch
2. ffsb-remove-hardcoded-configure.patch
Signed-off-by: Maxin B. John <maxin.john@enea.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Stefan Stanacar [Fri, 25 Apr 2014 11:35:27 +0000 (14:35 +0300)]
oeqa/utils: sshcontrol: realtime logging of output
Log the output of the command as it runs not when it finished, else
tail -f tmp/work/minnow-poky-linux/core-image-sato/1.0-r0/testimage/ssh_target_log
isn't as useful as it could be.
Signed-off-by: Stefan Stanacar <stefanx.stanacar@intel.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
Martin Jansa [Fri, 25 Apr 2014 16:40:27 +0000 (18:40 +0200)]
feature-arm-thumb.inc, arch-armv4.inc: Add "arm" to TUNE_FEATURES
* it will be inherited by most DEFAULTTUNEs, except few exceptions which
support only thumb and not arm
* respect missing "arm" in TUNE_FEATURES in feature-arm-thumb.inc, so
when recipe asks for "arm" and MACHINE supports only "thumb" ignore
recipe and try to build with "thumb"
* show warning when overriding ARM_INSTRUCTION_SET set by recipe from tune
config
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Saul Wold <sgw@linux.intel.com>
