Paul Eggleton [Mon, 22 Nov 2010 16:05:10 +0000 (16:05 +0000)]
sstate.bbclass: delete sstate files when doing cleanall
This fixes the cleanall task to actually delete the sstage cache files. This
was not working correctly before due to attempting to use the current task's
hash in the package file name to be deleted, which of course is the hash of
the cleanall task when doing cleanall. The siginfo files are now deleted as
well.
Signed-off-by: Paul Eggleton <paul.eggleton@intel.com>
Richard Purdie [Mon, 20 Dec 2010 16:04:51 +0000 (16:04 +0000)]
bitbake/fetch: Checksum validity fixes
If the checksum check failed, the .md5 stamp file would still have been created
meaning subsequent builds would proceed with the corrupt file. Reorder the calls
to avoid this. Also raise a specific error for the checksum not specified error
case.
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
- control variable BB_STRICT_CHECKSUM
when SRC_URI checksum is missing, this variable decide pass or not
if BB_STRICT_CHECKSUM = "1", bitbake should fatal in this case, otherwise bitbake just pass
Paul Eggleton [Thu, 16 Dec 2010 17:35:31 +0000 (17:35 +0000)]
sanity.bbclass: add check for creation of long filenames
Detect and fail if filesystem in use for TMPDIR or SSTATE_DIR has an
unreasonably short file name length limit (eg. eCryptFS). This can cause
"file name too long" errors during poky builds (e.g. when writing sstate
files for packages with a git revision as the version).
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Paul Eggleton [Thu, 16 Dec 2010 16:25:13 +0000 (16:25 +0000)]
sanity.bbclass: allow minimisation of impact of more invasive sanity checks
Add a mechanism to avoid running more invasive checks every time bitbake is
invoked. Adds a file in conf/ (i.e. under user's build directory) that
tracks a new variable SANITY_VERSION (set in sanity.conf) as well as TMPDIR
and SSTATE_DIR, allowing checks to be run when these are changed.
Fixes [BUGID #562]
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Yu Ke [Thu, 16 Dec 2010 06:10:15 +0000 (14:10 +0800)]
distro_tracking: Fix name inconsistency
in distro_tracking_fields.inc, my name has two version: "Ke Yu <ke.yu@intel.com>",
and "Yu Ke <ke.yu@intel.com>". this patch unify it to be "Yu Ke <ke.yu@intel.com>".
Signed-off-by: Yu Ke <ke.yu@intel.com> Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
Nitin A Kamble [Wed, 8 Dec 2010 00:35:50 +0000 (16:35 -0800)]
distro_tracking: update status for upgraded recipes
distro_tracking: chaning owner of python-pygtk
As the python-pygtk's version is tightly connected to the gtk+ recipe version.
changing the recipe ower to Edwin who is the gtk+ recipe owner.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Saul Wold [Fri, 17 Dec 2010 01:48:11 +0000 (17:48 -0800)]
Revert "gst-plugins: fix for make 3.82"
This reverts commit 06570b84e498e8643aa29d9864f377bd55df1f23.
These versions of gst-plugins-* already have been fixed upstream
for the make 3.82 problems, so this patch detects already applied
and bails
Nitin A Kamble [Wed, 8 Dec 2010 00:35:50 +0000 (16:35 -0800)]
distro_tracking: update status for upgraded recipes
distro_tracking: chaning owner of python-pygtk
As the python-pygtk's version is tightly connected to the gtk+ recipe version.
changing the recipe ower to Edwin who is the gtk+ recipe owner.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Darren Hart [Thu, 16 Dec 2010 16:48:31 +0000 (08:48 -0800)]
README.hardware: Add Atom PC image flashing documentation
Cover the basics of flashing images to physical media for the Atom PC MACHINE,
as well as dealing with BIOSs that can only boot USB media in USB-ZIP mode.
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Nitin A Kamble [Tue, 7 Dec 2010 16:49:26 +0000 (08:49 -0800)]
patch-2.6.1: implement new patch recipe with latest upstream
This commit patch recipe based on latest upstream code. This is GPLv3
code based. Hence the earlier patch-2.5.9 recipe is left intact for
GPLv2 needs.
Patches from 2.5.9 patch recipe are rebased to this new recipe except
these exceptions:
unified-reject-files.diff: dropped
This patch implements this new parameter:
" --unified-reject-files Create unified reject files."
And upstream has implemented very similar parameter like this:
" --reject-format=FORMAT Create 'context' or 'unified' rejects."
Hence this patch is dropped for the 2.6.1 recipe.
global-reject-file.diff: rebased
This patch is rebased to the newer upstream codebase.
install.patch: dropped
Newer upstream code now includes code form this patch.
debian.patch: dropped
This huge (10k lines) patch was specific for 2.5.9 version of GPLv2 patch.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Liping Ke [Wed, 15 Dec 2010 17:45:30 +0000 (01:45 +0800)]
Add extra parameters options for poky-qemu scripts
In order to support qemu user's flexibility requirement, we add extra
parameter options, user can add extra params such as "<-m 256>"
in poky-qemu script command.
Jiajun Xu [Thu, 16 Dec 2010 08:04:00 +0000 (16:04 +0800)]
sanity.bbclass: Check if DISPLAY is set when IMAGETEST is set
Fixes [BUGID: 594]
In imagetest-qemu.bbclass, we depend on DISPLAY to start qemu in
a X desktop. When user wants to run qemu sanity test, we need to
check if the variable is set.
Richard Purdie [Wed, 15 Dec 2010 23:22:16 +0000 (23:22 +0000)]
scripts/poky-qemu: Improve tmp layout assumption
If someone has changed TMPDIR in local.conf to a non-standard location, the
poky-qemu script currently doesn't handle this and assumes if BUILDDIR is set,
$BUILDDIR/tmp will exist.
Its simple to check if this exists and if not, to ask bitbake where the
directory is so this patch changes the code to do that.
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
Kevin Tian [Tue, 14 Dec 2010 11:21:24 +0000 (19:21 +0800)]
runqueue.py: use correct task ID when checking validity of setscene tasks
sstate hash validation is done at initialization of RunQueueExecuteScenequeue.
However the index of 'valid' list returned from the validation doesn't
correspond to setscene task ID. It's just an intermediate namespace between
runqueue and sstate hash func. Use it as setscene task ID fully mess the flow.
Previously this doesn't cause trouble because all setscene tasks are passed. Commit 58396a5d24c62710fd0a9f3780d84ac8a95d8e7c add 'noexec' concept to setscene
tasks which grabs some tasks out of the list and thus trigger this problem
Without this fix there're ~50 recipes (gzip-native, glib, ...) rebuilt weirdly
with a minimal build, even though existing sstate packages could accelerate them.
there's another typo using wrong task ID in a debug message which further hide
this issue
Paul Eggleton [Tue, 14 Dec 2010 18:09:58 +0000 (18:09 +0000)]
poky.conf: adjust variable whitelisting to fix sstate checksums
After some analysis of task dependencies I have adjusted the global whitelist
used to filter out variables that would otherwise cause unwanted changes in
task checksums for sstate packages.
The following changes have been made to BB_HASHBASE_WHITELIST:
* Fix typo: FILESEXTRAPATHS not FILESEXTRPATHS
* Add variables FILE_DIRNAME HOME LOGNAME SHELL TERM USER
* Remove variables DATE and TIME (these are referred to by a number of tasks
where they should affect the checksum, e.g. because they influence PV)
* Remove variable _ (not found in dependencies)
Additionally DATE and SRCDATE are excluded but only explicitly for
patch_do_patch, which uses these only to provide a warning if patches are
out of date.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Paul Eggleton [Fri, 10 Dec 2010 15:30:30 +0000 (15:30 +0000)]
bitbake: add optional expansion to getVarFlag()
Add a parameter to getVarFlag() to auto-expand the value of the flag. This
makes getVarFlag() more consistent with getVar(), and allows expansion of
vardeps and vardepsexclude (which has been done in this commit).
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Paul Eggleton [Fri, 10 Dec 2010 14:29:39 +0000 (14:29 +0000)]
bitbake: allow excluding dependencies of variables
Adds a vardepsexclude flag that can be used to exclude a dependency of
a variable (the opposite of vardeps). This will allow the exclusion of
variables from the hash generation much more selectively than blanket
whitelisting using BB_HASHBASE_WHITELIST.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Dexuan Cui [Tue, 14 Dec 2010 07:18:56 +0000 (15:18 +0800)]
gnome-desktop: add --disable-desktop-docs
autobuilder logs show the below failure:
| xsltproc -o lgpl-vi.omf --stringparam db2omf.basename lgpl --stringparam
db2omf.format 'docbook' --stringparam db2omf.dtd "-//OASIS//DTD DocBook XML
V4.1.2//EN" --stringparam db2omf.lang vi --stringparam db2omf.omf_dir
"/usr/share/omf" --stringparam db2omf.help_dir "/usr/share/gnome/help"
--stringparam db2omf.omf_in ...
| http://www.oasis-open.org/docbook/xml/4.1.2/dbpoolx.mod:632: parser warning :
PEReference: %notation.class; not found
| ...
| unable to parse oc/lgpl.xml
| make[3]: *** [lgpl-oc.omf] Error 1
However this issue can't be reproduced in my Ubuntu 9.04 system. I suppose
it's because on autobuilder the newer libxml has a stricter syntax requirement
about the related xml files.
Actually before commit b8b13dcbb76fdc3c9a2866bb35945dcf949c4b20 upgraded
gnome-desktop to 2.32.1, we also didn't build desktop-docs. The commit removed
no-desktop-docs.patch but didn't notice the Makefile.am depends on
--disable-desktop-docs to decice whether desktop-docs will be built.
So we should continue to make sure desktop-docs is not built.
