Richard Purdie [Thu, 10 May 2012 08:23:11 +0000 (09:23 +0100)]
bitbake/runqueue: Drop check_stamp_fn, it's obsolete and should not be used
This function was used by old code such as packaged staging but is thankfully
obsolete now and replaced with better mechanisms. Its time to remove it and
the horrible internal only variables associated with it.
Richard Purdie [Thu, 10 May 2012 08:22:22 +0000 (09:22 +0100)]
bitbake/runqueue: Drop check_stamps function, it is obsolete and unused
This was originally used to check the consistency of the stamps in one function
call. This turns out to be inefficient, unnecessary and if it were necessary,
check_stamp_task() could be called in a loop. The function has been unmodified
for a while and likely contains bugs. Its best simply removed.
- for a normal exit, use WEXITSTATUS, rather than manually shifting
- for exit via signal, set the exit code to 128+N, per shell convention
- if a process was stopped, return and don't handle it, as the process can yet
be continued
This should fix the case where bitbake says a task failed with an exit code of
0 (we assumed failure based on the overall status, but didn't pass all the
information along to task_fail).
Liming An [Tue, 8 May 2012 11:48:06 +0000 (19:48 +0800)]
Hob: add original url show function with the tooltip hyperlink for user
When case about No browser, such as running in 'Build Appliance', user can't open
the hyper link, so add this work around for user. (Checking the browser is avaiable
or not is hard by different system and browser type)
Saul Wold [Wed, 9 May 2012 18:08:57 +0000 (11:08 -0700)]
eglibc: Add patch to fix /var installation location
This address the packaging issues since it's installed in the wrong place
ERROR: For recipe eglibc-nativesdk, the following files/directories were installed but not shipped in any package:
ERROR: /var
ERROR: /var/db
ERROR: /var/db/Makefile
This introduces a sanity check for the toolchain, which verifies
each tuning (including any multilibs), producing meaningful diagnostics
for problems, and also provides some higher-level tuning features.
The TUNEVALID and TUNECONFLICT/TUNECONFLICTS settings were not
implemented. Listed one or two missing features in TUNEVALID,
also (in a previous patch) fixed the references to
features which didn't exist.
This patch also provides a whitelisting mechanism (which is completely
unused) to allow vendors providing prebuilt toolchain components to
restrict tunings to those based on or compatible with a particular ABI.
Otavio Salvador [Wed, 2 May 2012 01:00:01 +0000 (22:00 -0300)]
qt4-graphics-system: set graphic system in profile instead of X11 session
To allow the use of Qt applications started manually to use the
configured graphics system the setting needs to be available in user
environment. This moves the setting to /etc/profile.d ensuring it is
set and available. This allow the removal of x11-common as dependency.
Martin Jansa [Tue, 8 May 2012 22:53:38 +0000 (15:53 -0700)]
kmod: fix upgrade path from module-init-tools
* PROVIDES does not need _${PN}
* CONFLICTS should be RCONFLICTS
* opkg needs RREPLACES to automatically remove older module-init-tools
instead of just reporting it conflicts with kmod
SHR root@gjama ~ $ opkg install kmod
Installing kmod (7+git02629fa02e96763db7460a930239cc93649a52f8-r0.0) to root...
Downloading http://jama.dyndns-home.com/org.openembedded.shr-core//armv4t/kmod_7+git02629fa02e96763db7460a930239cc93649a52f8-r0.0_armv4t.ipk.
Removing package module-init-tools-depmod from root...
update-alternatives: removing //sbin/depmod as no more alternatives exist for it
Removing package module-init-tools from root...
update-alternatives: Linking //sbin/insmod to ../bin/busybox
update-alternatives: Linking //sbin/modprobe to ../bin/busybox
update-alternatives: Linking //sbin/rmmod to ../bin/busybox
update-alternatives: removing //sbin/modinfo as no more alternatives exist for it
update-alternatives: removing //bin/lsmod as no more alternatives exist for it
update-alternatives: Linking //sbin/lsmod to ../bin/busybox
Configuring kmod.
update-alternatives: Linking //sbin/insmod to /sbin/insmod.kmod
update-alternatives: Linking //sbin/modprobe to /sbin/modprobe.kmod
update-alternatives: Linking //sbin/rmmod to /sbin/rmmod.kmod
update-alternatives: Linking //sbin/modinfo to /sbin/modinfo.kmod
update-alternatives: Linking //bin/lsmod to /bin/lsmod.kmod
update-alternatives: Linking //sbin/lsmod to /bin/lsmod.kmod
update-alternatives: Linking //sbin/depmod to /sbin/depmod.kmod
Bruce Ashfield [Wed, 9 May 2012 03:42:24 +0000 (23:42 -0400)]
kern-tools: fix do_patch errors
The linux-yocto repository and scripts can support a mode of
meta data management that merges a base meta branch to every
BSP branch. In this case, the scripts don't have to restore
a checkpoint for the meta data to be globally accessible.
The decision to restore or not is made based on whether or
not the meta branch is part of all branches or not.
The linux-yocto recipes have a sanity check to determine if
the requested SRCREV for meta data matches the head of the
meta branch (via do_validate_branches). If the wrong commit
is at the head, the meta branch is moved aside and the branch
reset to the right commit. This creates two meta branches that
contain the base meta data.
The test for integrated meta data mistakes this for a globally
merged set of meta data and doesn't restore the checkpoint, which
leads to build failures.
The immediate fix is to allow two branches to have the meta
data. The long term fix is to make the detection only consider
if the build branch contains the meta data.
Bruce Ashfield [Tue, 8 May 2012 14:52:30 +0000 (10:52 -0400)]
meta-yocto/linux-yocto-3.0: update branch mappings for new tools
Updates to the kern-tools to be more flexible and support multiple
upstream repository formats means that KMACHINE should really only
map to a machine and that KBRANCH should specify branch information.
To adapt to this change, we split the hardware reference boards into
KMACHINE and KBRANCH mappings.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Mon, 7 May 2012 15:14:10 +0000 (11:14 -0400)]
linux-yocto/3.0: mm/msync: tweak tmpfs patch for syscall msync
Bumping the routerstationpro SRCREV for the following fix:
mm/msync: tweak tmpfs patch for syscall msync
Commit 1c3ae5441 "mm: msync: fix issues of sys_msync on tmpfs"
fixes the problem that sys_msync fails with tmpfs on MIPS CPU which
has feature "cache alias".
But it makes POSIX test cases mlockall/3-6 3-7 fail on MIPS.
Case mlockall/3-6 creates a share memory, and maps it to memory.
fd = shm_open(SHM_NAME, O_RDWR|O_CREAT, S_IRUSR|S_IWUSR);
foo = mmap(NULL, BUF_SIZE, PROT_WRITE, MAP_SHARED, fd, 0);
Then calls mlockall to lock all of the virtual address space:
if (mlockall(MCL_CURRENT) == -1) {
At last tests whether the virtual address spaces are locked:
page_ptr = (void*) ((long)foo - ((long)foo % page_size));
result = msync(page_ptr, page_size, MS_SYNC|MS_INVALIDATE);
It espects msync returns -1 with EBUSY but returns 0.
Case mlockall/3-7 creates a normal file to mmap, and it fails too.
Tweak the patch to:
1 Moved the CONFIG_TMPFS block down in the loop so that the normal
vma flag checking will perform.
2 There may be other VMAs in the list after this VMA which belongs
to tmpfs file, and we should sync them after the tmpfs one. So
remove the break loop clauses.
Signed-off-by: Kang Kai <kai.kang@windriver.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Michel Thebeau <michel.thebeau@windriver.com>
2/2:
kconf_check: fix bad quoting around missing_required.cfg
missing_required.cfg won't have it's path truncated (if applicable), since
the quoting it wrong.
Signed-off-by: Andrea Adami <andrea.adami@gmail.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Thu, 22 Mar 2012 20:00:08 +0000 (16:00 -0400)]
linux-yocto: streamline support for multiple upstream repo types
In order to support repositories of various types (with or without
meta data, branched, pristine, custom, etc) information about the
type of processing that is required was passed to the processing
phases via variables.
The combination of variables involved in coordinating the processing
creates a learning curve and overly complicates recipe extensions.
With minor tweaks to the kern-tools, adding flexibility and keying
off the existence of the meta branch it is possible to remove all
of the variables that were added to support different repository
types.
Bruce Ashfield [Fri, 4 May 2012 14:40:34 +0000 (10:40 -0400)]
linux-yocto/3.2: configuration and pch merge
Updating the 3.2 SRCREVs to import the following meta/config
changes:
6b3d4e0 meta: add mei feature 519abac meta: add usb/uhci-hcd feature a67c5a3 meta/crownbay: use usb features 0855066 meta: add usb/ohci-hcd feature 15f1a99 meta: add usb/ehci-hcd feature 8fa6408 meta: add usb/xhci-hcd feature c724a55 meta: add usb/base feature b55b3a1 sys940x: Cleanup sys940x.scc 93f2e97 sys940x: Use PHYSICAL_START of 0x200000 to boot aaa034b sys940x: Add common standard and preempt-rt features e2b1286 sys940x: Add efi-ext to standard and preempt-rt configs d188c21 sys940x: Move emgd-1.10 data to the standard scc file 72d9369 fri2: Cleanup fri2-$KTYPE.scc files re efi-ext.scc dbcb120 fri2: Use emgd-1.10 feature and branch
Signed-off-by: Darren Hart <dvhart@linux.intel.com> Signed-off-by: Tom Zanussi <tom.zanussi@intel.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Zhai Edwin [Tue, 8 May 2012 14:43:49 +0000 (22:43 +0800)]
pango: Fix modules load failure in multilib environment
Multi-libs of Pango need different modules, thus different config files and
utils. This patch separate config file and utils with different MLPREFIX to
avoid conflict.
Scott Garman [Mon, 7 May 2012 22:49:14 +0000 (15:49 -0700)]
opensp: upgrade to 1.5.2
Removed all patches, they've been integrated upstream. :)
Added --disable-doc-build to prevent creation of docs, which
otherwise fails with the following configure error:
could not find xmlto; set XMLTO or consider --disable-doc-build
Removed configure_prepend step that was deleting the m4/ directory,
since some macros needed for the build are defined there, and I've
not encountered any problems with keeping them there in my testing.
Dongxiao Xu [Tue, 8 May 2012 05:28:30 +0000 (13:28 +0800)]
qt4: move functions from python to shell style
In qt4's do_configure operation, it will refer to some variables that
are derived from 'd', however these variable values may be not correct
in multilib case since the extraction of these variables happens before
the multilib handler.
The fix is to move these python style functions back to shell style.
Richard Purdie [Mon, 7 May 2012 09:50:03 +0000 (09:50 +0000)]
connman: Disable tist plugin on powerpc
This plugin doesn't build on powerpc due to powerpc's terminal ioctl defintions
being incompatible with assumptions being made by this module. Until someone has
need and can test this on powerpc, disabling is the safest option.
Martin Jansa [Sun, 6 May 2012 07:28:23 +0000 (09:28 +0200)]
evolution-data-server: enable deprecated glib API
* we have very old version and deprecated g_atexit is used even in current master
http://git.gnome.org/browse/evolution-data-server/log/libedataserver/e-categories.c
* LIC_FILES_CHKSUM change is only because PCRE was upgraded and now
pcre.h says "Copyright (c) 1997-2012 University of Cambridge" instead
of 2010 and COPYING file
"Please see the file LICENCE in the PCRE distribution for licensing
details." but LICENSE file is not part of glib distribution (but still
BSD in standalone PCRE)
Tom Zanussi [Mon, 30 Apr 2012 19:12:15 +0000 (14:12 -0500)]
yocto-bsp: clarify help with reference to meta-intel
The current yocto-bsp help assumes knowledge that the meta-intel layer
needs to be cloned before it's put into the BBLAYERS. Avoid the
guesswork and state the details explicitly in the help.
Also, the shorter 'usage' string doesn't mention it at all; it would
help to at minimum mention it and refer the user to the detailed help.
Fixes [YOCTO #2330].
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Thu, 3 May 2012 16:40:51 +0000 (17:40 +0100)]
poky.conf: use correct version string for Ubuntu 12.04
Since it is an LTS release, the final version string was not
"Ubuntu 12.04" but "Ubuntu 12.04 LTS", so use this when doing the tested
host distribution check.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Darren Hart [Wed, 2 May 2012 04:14:18 +0000 (21:14 -0700)]
initrdscripts: Update install.sh to work with mmc devices
Fixes [YOCTO #2385]
The installer only searches for hd[ab] sd[ab]. Some newer BSPs have mmcblk
devices that should be used as the install target. These devices also have a
partition prefix (mmcblk0p1 instead of mmcblk01). As they are detected
asynchronously, it is necessary to add the rootwait kernel parameter to avoid
a race condition trying to mount the root device.
As BSPs like the FRI2 and the sys940x have mmc devices and will have a 1.2
release, we should push this to 1.2.1. The changes are perfectly contained and
easily verified.
Test for an mmcblk device and add the p partition prefix if necessary. Add the
rootwait kernel parameter when an mmcblk device is detected. Replace the series
of explicit umount commands with a single umount using a wildcard. This will
find all the partitions and will not try to unmount non-existant devices. Avoid
copy and paste errors by replacing /dev/${device}${pX} references with the
previously assigned rootfs, bootfs, and swap variables.
These changes have been tested on the FRI2 Sato image which installed to
/dev/mmcblk0 as well as the N450 Sato image which installed to /dev/sda. Both
were successful.
Signed-off-by: Darren Hart <dvhart@linux.intel.com> CC: Scott Garman <scott.a.garman@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If build system has those libraries installed
gcc configure will pick them up. We want
consistent builds so we disable them since we
do not (yet) support them
Some versions of the screen utility provided from the host OS vendor
write the socket directory to $HOME/.screen. When using a shared home
directory across many servers, one sets the SCREENDIR environment
variable to avoid collisions in the shared home directory. This
results in problems launching a devshell where it is not entirely
obvious what happened because the SCREENDIR environment variable
got stripped from the environment prior to setting up the screen
in detached mode.
Example:
% bitbake -c devshell busybox
# ...Please connect in another terminal with "screen -r devshell"
% screen -r devshell
There is no screen to be resumed matching devshell.
The temporary work around was to do something like:
sh -c "unset SCREENDIR; screen -r devshell"
This patch adds SCREENDIR to the white list to ensure screen
works properly on systems where a developer needs to use
the SCREENDIR with shared home directories.
Peter Seebach [Wed, 2 May 2012 01:45:10 +0000 (20:45 -0500)]
conf/machine: Clean up configuration values.
This cleans up and/or corrects a few values from machine includes
for consistency with future toolchain sanity checks, and also adds
the TUNEVALID and TUNECONFLICTS to documentation.conf.
Richard Purdie [Fri, 4 May 2012 15:18:55 +0000 (16:18 +0100)]
bitbake: Add start of bitbake regression/self testing
This adds some basic unit testing for the codeparser and data store code. Many of
the actual test cases were taken from work by Chris Larson's OE-Signatures work but with
changes to adapt to the current bitbake APIs we need to test.
I also imported CoW tests written by Holger Freyther from the original bitbake-test
codebase: http://svn.berlios.de/wsvn/bitbake/trunk/bitbake-tests/tests/ and
some tests from the doctests that were removed in commit:
http://git.openembedded.org/bitbake/commit?id=3a11c2807972bbbddffde2fa67fc380d159da467
* Fixes the following error after a system library upgrade
| .../mipsel-oe-linux/4.6.4/cc1: error while loading shared libraries: libppl.so.10: cannot open shared object file: No such file or directory
Paul Eggleton [Fri, 4 May 2012 13:36:01 +0000 (14:36 +0100)]
buildhistory: fix multiple commit of images and packages at the same time
The echo line here was merging multiple lines into one, and the result
was that if both image and package changes had to be comitted then only
the image changes were being committed and the package changes could
potentially be merged into the next package change. Quoting the variable
reference fixes this.
Richard Purdie [Fri, 4 May 2012 14:03:33 +0000 (15:03 +0100)]
bitbake/utils: Convert vercmp_string() to use vercmp internally
Having two different version comparision algorithms in bitbake has never seemed
like a sensible idea. Worryingly, they also return different results to each other.
The vercmp_string API is relatively unused with no users in OE-Core or BitBake
itself for example. This patch converts it to use vercmp internalls, bringing
consitency to the comparisions which is easy now we have other recently added
functions. Yes, this changes behaviour but in this case I'd prefer we were
consistent than having two different comparisions.
Saul Wold [Tue, 1 May 2012 15:13:20 +0000 (08:13 -0700)]
gcc-package-target: add libexec for plugin
Fixes:
ERROR: For recipe gcc, the following files/directories were installed but not shipped in any package:
ERROR: /usr/libexec/gcc/i586-poky-linux/4.7.1/plugin/gengtype
(From OE-Core rev: 0be64cda0a20ca3aab75ee0cc360236d2dc40c95)
Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
kernel.bbclass: move kernel-vmlinux up in PACKAGES
If KERNEL_IMAGETYPE is vmlinux, the expectation is most likely that there will
be no kernel image package, but we still want a vmlinux package for debugging,
so move kernel-vmlinux in front of kernel-image in PACKAGES.
Jason Wessel [Wed, 2 May 2012 11:30:46 +0000 (06:30 -0500)]
runqemu: Fix TAP='TUNSETGROUP: Invalid argument' by falling back to tunctl -u
By default the runqemu script tries to set the group permissions on any
tap device it creates. The TUNSETGROUP ioctl is not implemented on some
popular host enterprise linux distributions.
This patch implements a fallback to using the userid as the owner of
the tap device which is supported by all 2.6 kernels, the default remains
to try and use the groupid first.
gcc 4.7 uncovers a type mismatch in the code. And this commit fixes the issue:
| /srv/home/nitin/builds/build-gcc47/tmp/sysroots/qemux86/usr/include/glib-2.0/glib/gthread.h:58:7: error: 'union' tag used in naming 'struct _GMutex' [-Werror=permissive]
| In file included from ./wtf/Platform.h:1217:0,
| from ./config.h:30,
| from wtf/gobject/GOwnPtr.cpp:19:
| wtf/gobject/GTypedefs.h:55:16: note: 'struct _GMutex' was previously declared here
NOTE: package qt4-x11-free-4.8.1-r41.1: task do_compile: Failed
projects / thirdparty / openembedded / openembedded-core-contrib.git / log
