Make 3.82, as shipped with Fedora 14, fixes some holes in the parser which in
turn breaks behaviour of some Makefiles. Most notably eglibc's.
Signed-off-by: Joshua Lock <josh@linux.intel.com>
package sstate requires a lock under sysroots/, which however may not be
created when sstate_setscene functions are executed and then causes failures.
here we make sure 'sysroots' created before do_package_setscene is executed.
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Without doing this I can't reuse prebuilts generated from one directory in another
build directory which is really not comfortable for developers. Due to difference
on PWD variable, base hash becomes different for every sstate package.
So here PWD is added to the hase base whitelist, which makes most prebuilts reusable
now in a different build directory. However there still remains one issue about
do_populate_sysroot sstate package. Absolute path is encoded in this task which
then blocks reuse on them. Perhaps we need some redesign for do_populate_sysroot into
two stages, with all absolute path related operations moved to the 2nd stage. This
way then the 1st stage can be reused.
Partially fix [BUGID #522]
Signed-off-by Kevin Tian <kevin.tian@intel.com>
When contacts called for the 1st time, adding new contact always failed but
worked after restart. The reason is e-addressbook-factory used to handle
addressbook leads a segmentation fault. Root cause is that factory would
create the data base if not exist, but use it without proper
initialization.
[BUGID #455] got fixed by this.
Signed-off-by: Zhai Edwin <edwin.zhai@intel.com>
The fetcher has special handling for file:// mirror urls, being efficient and
just providing an updated path. Unfortunately the sstate fetching code wasn't
able to handle this. This patch detects this and injects a symlink to ensure
everything works. It also fixes some datastore references to be correct and
ensures the sstate download directory exists if it doesn't already.
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
If sstate was used to accelerate a build, the pseudo directory might not have
been created leading to subsequent task failures.
Also, sstate packages were not being installed under pseudo context meaning
file permissions could have been lost.
Fix these problems by creating a FAKEROOTDIRS variable which bitbake ensures
exists before running tasks and running the appropriate setscene tasks under
fakeroot context.
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
If a sstate package exists for the package task but not for the rpm packaging
task, the output from the package task will be used. The directory pointed
to by dvar will not exist under this scenario.
Since the directory is never used by the packaging process remove the
check, substituting the pkgd variable which is always present and used.
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
The message "WARNING: Function do_build doesn't exist" doesn't look professional,
so fix the underlying problem even if this warning is harmless.
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
When installing a sstate package, directories tracked by plaindirs were being installed
to the incorrect location. With the current implementation this was limited to
the do_package task.
This patch ensures plaindirs tracked files are created in the correct location, fixing
the bug where these files would go missing.
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
The commit '0e0d14ed1e88ebff876abb3d00b08627a3879e8d' introduced a new patch
nonstripbinaries.patch, the filename format caused the patch to not apply.
Simple fix made by correcting the path.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
This is used to fix the following build failure:
checking for oggpackB_read... no
| configure: error: newer libogg version (1.1 or later) required
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
The format of the syslog.conf files is different between busybox and sysklogd.
Use the alternatives method to ensure we get the correct config file for
any specific configurations.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Fixes [BUGID #501]
While building the poky-image-sato target, gst-ffmpeg failed when
it attempted to run altivec tests on the host. We can live without
this optimization, so the easiest fix is to simply disable it.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Fixes [BUGID #500]
While the tuning for the mpc8315e is 603e, the PACKAGE_EXTRA_ARCHES
was ppce300. This created a mismatch and resulted in rootfs assembly
issues due to missing locales.
We align both at 603 and can revist a better tuning in the future.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
remove nonrootinstall patch, and use a patch named nonstripbinaries to modi
fy the /at-3.1.10.2/Makefile.in, so it can also handles stripping of the bi
naries
Fix [BUG 246]
Signed-off-by: Mei Lei <lei.mei@intel.com>
Duplicate the patch from sato-icon-theme which enables the path to the
icon-name-mapping tool to be specified.
Signed-off-by: Joshua Lock <josh@linux.intel.com>
This commit fix [BUGID #482]
Two issues cause bug 482:
- firsty, there are two version of syslog: sysklogd and busybox.
the busybox one is directly installed as /etc/init.d/syslog,
and the sysklogd one is installed by update-alternative. the
update-alternative will thus fail because the /etc/init.d/syslog
(busybox one) already exist and not a link. so the correct way
is to install busybox one by update-alternative, the layout will be:
/etc/init.d/syslog.busybox
/etc/init.d/syslog.sysklogd
/etc/init.d/syslog -> syslog.busybox or
/etc/init.d/syslog -> syslog.sysklogd
- secondly, sysklogd default conf is not comply with poky. Its dir /var/adm/
does not exist. Check the debian /etc/syslog.conf and find it is more
sophiscated and suitable, so port /etc/syslog.conf from debian.
Signed-off-by: Yu Ke <ke.yu@intel.com>
Fixes [BUGID #465]
wrs_meta: add logbuf size feature
A 16k kernel log buffer size is too small for emenlow and atom-pc -
boot messages are routinely lost with the current setting. Increase
the size to something more reasonable - 64k seems to be a good
compromise.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
emenlow: add cmdline option for reboot quirk
The default reboot options don't work for emenlow i.e. kbd + triple
results in a reboot hang. According to the emenlow chipset manual,
port CF9 is supported, so let's use that.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
update-alternatives.bbclass assumes /usr/bin/sed if ALTERNATIVE_LINK is not
specified, however sed has changed its default installation directory to
/bin/sed. This causes below errors at the first boot of lsb image:
update-alternatives: Error: cannot register alternative sed to /usr/bin/sed
since it is already registered to /bin/sed
This fixes [BUGID #478]
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Richard Purdie
Joshua Lock
Kevin Tian
Koen Kooi
Zhai Edwin
Mark Hatle
Dexuan Cui
Bruce Ashfield
Mei Lei
Qing He
Yu Ke
Saul Wold