Martin Jansa [Wed, 18 Sep 2013 23:58:32 +0000 (01:58 +0200)]
gcc-4.8: temporary disable thumb
* temporary work around for build issue on armv4t:
| cp/decl.o: In function `bad_specifiers':
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7171:(.text.unlikely+0x24): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7173:(.text.unlikely+0x32): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7176:(.text.unlikely+0x3e): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7180:(.text.unlikely+0x4c): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7182:(.text.unlikely+0x5a): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7185:(.text.unlikely+0x66): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7189:(.text.unlikely+0x74): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7191:(.text.unlikely+0x82): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7194:(.text.unlikely+0x8e): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7198:(.text.unlikely+0x9c): relocation truncated to fit: R_ARM_THM_CALL against symbol `error(char const*, ...)' defined in .glue_7 section in linker stubs
| gcc-4.8.0-r0/gcc-4.8.0/gcc/cp/decl.c:7200:(.text.unlikely+0xaa): additional relocation overflows omitted from the output
| collect2: error: ld returned 1 exit status
| make[2]: *** [cc1plus] Error 1
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Wed, 18 Sep 2013 03:27:42 +0000 (03:27 +0000)]
linux-yocto/3.10: mips configuration changes
Updating the 3.10 SRCREV to import the following MIPS configuration changes
4f689aa meta: remove ftrace/ftrace-disable feature 3058d81 mips: have the mips BSPs disable function tracing instead of ftrace 935f43f meta: add ftrace/ftrace-function-tracer-disable feature 0d72a03 mti-malta64: Default to support o32 and n32 userspace binaries
The first three changes improve the ftrace disabling fragments, to allow tracepoints
and ftrace to be enabled, while only disabling dynamic ftrace. This allows tools
that required tracepoints (like lttng) to be built against MIPS.
The mti-malta64 change adds n32 and o32 support to the default configuration to
support a broader range of userspace binaries.
[YOCTO #5215]
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>
Bruce Ashfield [Wed, 18 Sep 2013 03:16:03 +0000 (03:16 +0000)]
linux-yocto/3.8: add haswell-wc board config and branch
Bumping the meta branch SRCREV to import the following board support:
meta: add haswell-wc bsp for Intel Haswell Platform (Walnut Canyon CRB) scc and config files
Signed-off-by Ong Boon Leong <boon.leong.ong@intel.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 18 Sep 2013 16:48:46 +0000 (17:48 +0100)]
qemurunner: disable grabs in automated testing
Use the new QEMU_DONT_GRAB environment variable to disable grabs,
finally/hopefully solving the random hangs that the autobuilder has been hitting
for a while.
[ YOCTO #5131 ]
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 18 Sep 2013 16:48:45 +0000 (17:48 +0100)]
qemu: add option to disable X grabs
When the mouse pointer enters the qemu window it takes a pointer grab. This
doesn't sound too dangerous at first but it turns out that SDL will infinitely
busy-loop if it can't get the grab (e.g. if the screen is locked) and the
average autobuilder setup's X server will have locked the screen a few minutes
after boot.
The result is that on many autobuilders apparently random qemu instances (the
top-most one under the pointer) will hang during boot.
To resolve this add an option (via an environment variable) to never attempt a
grab. The default behaviour remains to grab so that everyone else doesn't see
any change.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Otavio Salvador [Wed, 18 Sep 2013 03:34:18 +0000 (00:34 -0300)]
u-boot-fw-utils: Skip package if UBOOT_MACHINE is unset
The UBOOT_MACHINE variable needs to be set so the firmware utils can
know about some configuration settings which are board dependent. This
patch ensures the package is skipped in case UBOOT_MACHINE is unset
thus avoid its build in 'bitbake world' builds for incompatible
machines.
Fixes [YOCTO: #5223]
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Otavio Salvador [Tue, 17 Sep 2013 04:18:20 +0000 (01:18 -0300)]
perf: Ensure we use CFLAGS and LDFLAGS settings from kernel build system
The kernel build system does the right thing here and we should stop
overriding it. This code has been added based on a change from
'meta-metro' layer, revision 9d698004137c1a888d40d6a4808d94afa22387e7,
without any information about what problem it fixes so I am reverting
it.
Using the CFLAGS and LDFLAGS makes it impossible for kernel build
system to append to it, thus making the build fail in various ways as:
| CC /.../perf/1.0-r8/perf-1.0/perf.o
| In file included from builtin.h:4:0,
| from perf.c:9:
| util/util.h:74:24: fatal error: lk/debugfs.h: No such file or directory
| #include <lk/debugfs.h>
| ^
| compilation terminated.
The unset is done in do_compile and do_install otherwise it /rebuild/
perf as it detects the compiler options has change.
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Tue, 17 Sep 2013 13:57:46 +0000 (14:57 +0100)]
cmake-native: fix dependencies
When building cmake for native, we don't use the system libraries and
thus cmake builds its own internal version of libarchive; this requires
zlib, bzip2, and e2fsprogs. We can add to DEPENDS for the two former
libraries and patch out the latter.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 11 Sep 2013 11:33:48 +0000 (11:33 +0000)]
mesa: add virtual/mesa provider
As there are two alternative mesa recipes (mesa and mesa-gl), there needs to be
a virtual provider that recipes that explicitly need Mesa (such as xserver-xorg)
can depend on.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Hans Beckerus [Sat, 14 Sep 2013 09:01:38 +0000 (09:01 +0000)]
libtool: fix resolve of lt_sysroot
This patch updates libtool.m4 (and its output) to resolve a problem
with variable 'lt_sysroot' not being properly updated if the option
'--with[-libtool]-sysroot' is not provided when running the 'configure'
script for a package.
According to the help text ouput from 'configure':
--with-libtool-sysroot=DIR Search for dependent libraries within DIR
(or the compiler's sysrooot if not specified).
Due to mixed up cases in a switch statement, when checking if the option
was specified or not, wrong actions were taken resulting in an incorrect
sysroot and failures to properly locate e.g. .la files when using the
populated SDK toolchain.
For current upstream status see:
http://lists.gnu.org/archive/html/bug-libtool/2013-09/msg00005.html
Signed-off-by: Hans Beckerus <hans.beckerus at gmail.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Tue, 17 Sep 2013 02:35:02 +0000 (02:35 +0000)]
python-smartpm: truncate the filename to meet NAME_MAX
There is a "File name too long" error when len(TMPDIR) = 410, this is
because the function getLocalPath() converts the filepath into the
filename, so there would be the error when len(filename)
> NAME_MAX, truncate the filename to meet NAME_MAX will fix the problem.
[YOCTO #5201]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Mon, 16 Sep 2013 18:06:34 +0000 (18:06 +0000)]
mesa-gl: add GL-only Mesa recipe
Some machines have hardware-specific GL drivers that do EGL and GLES (many ARM
boards). Others have their own EGL/GLES drivers and provide a Mesa DRI driver
(EMGD). Previously adding Mesa, for software GL/GLX rendering in the first case
and hardware GLX in the second, involved bbappends and changing Mesa to be
machine-specific.
By adding a just-GL Mesa the machine definition can combine it with the hardware
drivers cleanly.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Sat, 14 Sep 2013 02:48:06 +0000 (02:48 +0000)]
linux-libc-headers: do_install: fix "Argument list too long" error
There would be an "Argument list too long" error when the TMPDIR is in a
deep dir, for example, when "len(readlink -f TMPDIR) >= 350 (our
supported value is 410)". Use "$(foreach ,,$(shell echo))" to fix it.
There was already a patch which tried to fix this issue, so squash the
current change into the previous one as Bruce suggested.
[YOCTO #5138]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ming Liu [Mon, 16 Sep 2013 09:50:58 +0000 (09:50 +0000)]
openssh: use volatiles for managing /var/run/sshd
It fixes the following failure:
"fatal: Missing privilege separation directory: /var/run/sshd"
when sshd is started through xinetd.
Signed-off-by: Ming Liu <ming.liu@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
But there is a typo in:
http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=78ac027f2ac6c6663fff7080eabbd3d09c1241bb
It missed to remove an extra "--with-wctype-functions" in EXTRA_OECONF
LSB 4.1 Library Check: fix unable to find library libqt-mt.so.3
The libqt-mt is tested by lsb-dist-checker and lsb-test-desktop,
and it locates in meta-qt3 layer.
So if meta-qt3 is not added, there should be a warning to call
attention; if added, it will add libqt-mt to RDEPENDS.
Jason Wessel [Sat, 14 Sep 2013 00:08:58 +0000 (00:08 +0000)]
cdrtools-native: Update from 3.00 to 3.01a17
The update is needed to support generation of EFI boot images that
work with optical media. Specifically the "-eltorito-platform efi"
capability for mkisofs is needed.
[YOCTO #4100]
Signed-off-by: Jason Wessel <jason.wessel@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jason Wessel [Sat, 14 Sep 2013 00:08:57 +0000 (00:08 +0000)]
bootimage.bbclass: Move fat image creation into a function
In order to call the fat image creation multiple times it needs to be
in its own function. A future commit will make use of the new
function to additionally create EFI image files for use with an ISO.
[YOCTO #4100]
[YOCTO #1913]
Signed-off-by: Jason Wessel <jason.wessel@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Sun, 15 Sep 2013 09:13:12 +0000 (09:13 +0000)]
coreutils: set acpaths to avoid "Argument list too long" error
There would be an error when the TMPDIR is long/deep, for example when
len(TMPDIR) = 350 while our supported longest value is 410:
[snip]
aclocal: error: cannot open xxx
autoreconf: aclocal failed with exit status: 1
ERROR: autoreconf execution failed.
[snip]
Let aclocal use the relative path for the m4 file rather than the
absolute would fix the problem.
Another fix is that we can modify autotools.bbclass to let it use the
relative path rather than the absolute, but I don't think that we have
to do that based on the following 2 thoughts:
* The coreutils is the only recipe which has this issue as far as we
know when len(TMPDIR) <= 410, because it has the most amount of m4
files (more than 400 ones).
* That would impact all the recipes which use autotools.bbclass, and we
are not sure about the side effect, for example, it would break the
build there is a sub-configure.
[YOCTO #2766]
Signed-off-by: Wenzong Fan <wenzong.fan@windriver.com> Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Randy MacLeod [Sun, 15 Sep 2013 16:21:47 +0000 (16:21 +0000)]
vala: add bison-native depends for vala-native
vala-native builds were working because most systems have
bison installed on the host. Add an explicit dependency on
bison-native since bison isn't a sanity checked host tool.
Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some of the external USB devices has internal USB hub, which
make them look like "fixed" rather than "removable". And USB
autosuspend does not work with some of these devices resulting
in inoperable pointing device.
Now the code detect these false "fixed" devices by looking at their
parents. If any of their parent is "removable", then USB autosuspend
is not enabled for that device, which keeps the pointing device
functional.
Fixes bug:
[YOCTO #5166]
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Tue, 17 Sep 2013 09:22:17 +0000 (10:22 +0100)]
libxml2: remove patch for CVE-2012-2871
This CVE patch is actually against Chromium as they ship an internal fork of
libxml2 and breaks ABI. The real issue has been resolved in libxslt 1.1.27, and
we're shipping 1.1.28.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Fri, 13 Sep 2013 16:54:58 +0000 (12:54 -0400)]
linux-yocto/3.10: genericx86 meta data factoring
Updating the 3.10 meta branch to import the following commits from Darren
Hart:
285f93b meta/common-pc-64: Add USB 3.0 support 75072e4 meta/common-pc*: Refactor common-pc-64 to reuse common-pc drivers da06bde meta/common-pc: Split out CPU and Drivers config fragments 5f55e40 meta/common-pc: Cleanup common-pc.cfg and common-pc-gfx.cfg 0a3f784 meta/common-pc: Add common Wifi drivers f4b9f5e meta/common-pc: Add common Realtek and Atheros Ethernet drivers 4d9d3eb meta/common-pc: Build Ethernet and Wifi drivers as modules 13141ee meta/common-pc: Refactor Ethernet and Wifi options
These are in support of a new genericx86 multi-target BSP.
Signed-off-by: Darren Hart <dvhart@linux.intel.com> Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
boot-directdisk: Allow for EFI-only boot direct disk images
Condition building PCBIOS legacy images on MACHINE_FEATURES containing "pcbios"
or not containing "efi". This ensures existing BSPs will continue to get the
old PCBIOS legacy-only images. New BSPs can add "efi", "pcbios", or both. The
images created likewise support one or the other or both.
Signed-off-by: Joรฃo Henrique Ferreira de Freitas <joaohf@gmail.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
elfutils-native: Fix build on distros with gcc 4.8
The patch redhat-portability.diff causes this issue
so lets revert the portion which was using %a instead of %m
thats recommended anyway, redhat patch seems to be targetting
old compilers.
LSB Command Check: fix install_initd and remove_initd not found
The install_initd and remove_initd are linked to /sbin/chkconfig for lsb
core test, but chkconfig has been moved from /sbin to /usr/sbin in order
to fix QA warning about unsafe references in binaries.
(In commit e486242db83297701803866bea971a2f1a1135fe)
Let install_initd and remove_initd link to /usr/sbin/chkconfig could fix
this issue.
[YOCTO #5152]
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Fri, 13 Sep 2013 11:18:14 +0000 (12:18 +0100)]
linux-libc-headers: Add big warning about antisocial behaviour
I'm getting concerned with the number of people forking this recipe
and not understanding what they're doing. I'm therefore proposing
adding in a suitable warning to people thinking of copying it.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Stefan Stanacar [Fri, 13 Sep 2013 12:48:18 +0000 (15:48 +0300)]
lib/oeqa/runtime: timeout increases
Increase the timeout for smart commands as under load for qemumips
it's still to small. Also give ping more time fixing a potential
timeout for sato systemd.
Signed-off-by: Stefan Stanacar <stefanx.stanacar@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Fri, 13 Sep 2013 12:35:31 +0000 (13:35 +0100)]
bitbake.conf/package: Collapse PKGDATA_DIR into a single machine specific directory
Currently we have a hierarchy of pkgdata directories and the code has to put together
a search path and look through each in turn until it finds the data it needs.
This has lead to a number of hardcoded paths and file globing which
is unpredictable and undesirable. Worse, certain tricks that should be
easy like a GL specific package architecture become problematic with the
curretn search paths.
With the modern sstate code, we can do better and construct a single pkgdata
directory for each machine in just the same way as we do for the sysroot. This
is already tried and well tested. With such a single directory, all the code that
iterated through multiple pkgdata directories and simply be removed and give
a significant simplification of the code. Even existing build directories adapt
to the change well since the package contents doesn't change, just the location
they're installed to and the stamp for them.
The only complication is the we need a different shlibs directory for each
multilib. These are only used by package.bbclass and the simple fix is to
add MLPREFIX to the shlib directory name. This means the multilib packages will
repackage and the sstate checksum will change but an existing build directory
will adapt to the changes safely.
It is close to release however I believe the benefits this patch give us
are worth consideration for inclusion and give us more options for dealing
with problems like the GL one. It also sets the ground work well for
shlibs improvements in 1.6.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Thu, 12 Sep 2013 11:04:52 +0000 (12:04 +0100)]
bitbake.conf: include machine name in DEPLOY_DIR_IMAGE
This allows a clean seperation between image outputs from different
machines, and makes it possible to have convenience symlinks to make
the output ready to deploy.
This did require some surgery in runqemu; if explicit paths to the image
and kernel are not supplied then DEPLOY_DIR_IMAGE needs to be determined
from bitbake or set in the environment. However the script does try to
avoid requiring it unless it really is needed. Corresponding changes
were made in the automated testing code as well.
Based on an RFC patch by Koen Kooi <koen@dominion.thruhere.net>
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Thu, 12 Sep 2013 09:33:36 +0000 (05:33 -0400)]
icu-native: do_install: Segmentation fault
There was a "Segmentation fault" error when build icu-native when the
TMPDIR is in a deep directory (for example, when len(readlink -f $TMPDIR
== 410)), use LARGE_BUFFER_MAX_SIZE for cmd rather than
SMALL_BUFFER_MAX_SIZE would fix the problem, this should be a misplay
because other cmd uses LARGE_BUFFER_MAX_SIZE.
[YOCTO #5171]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If not every combination of BB_NUMBER_THREADS and PARALLEL_MAKE have
been tested by bb-matrix.sh, e.g., by using BB_RANGE="04 08 10 12 16"
and PM_RANGE="04 08 10 12 16", then the graph that gnuplot generates by
default looks very jagged due to the missing data points. By using
splines to interpolate the missing data the graph looks a lot better.
This should not change graphs where all data points are available in any
way, only improve sparse graphs.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 9 Sep 2013 14:14:39 +0000 (15:14 +0100)]
bitbake.conf: Stop providing ${P} and ${PF} by default
For a long time we've provided PN-PV and PN-PV-PR by tweaking PROVIDES. This looks
nice at first glance however it turns out to be a bit problematic. Taking make as an
example where there are two versions, 3.81 and 3.82, what should "bitbake make-3.81" do?
Currently it builds make-3.81 and make-3.82 and breaks in interesting ways. Is that
a bitbake bug? Well, it certainly shouldn't try and run the build. Why is it building
3.82 though? Its due to finding a dependency on "make-dev" and then trying to figure
out what provides it? The answer is "make" and the default version of "make" is 3.82.
So arguably, finding "make-3.81" should infer PREFERRED_VERSION_make = "3.81". Doing
so resolved the above problem since now "make" resolves to "make-3.81".
So what about if we have Recipe A:
DEPENDS = "make-3.81"
and Recipe B:
DEPENDS = "make-3.82"
That is clearly an error, easy. So finally what about if we have Recipe A:
DEPENDS = "make-3.81"
and Recipe B:
DEPENDS = "make"
The first recipe infers the PREFERRED_VERSION_make = "3.81" and then forces that
version on everything else. Is that desired? Probably not in most cases, at least not
silently.
As mitigation, we could print a WARNING about this happening. The final part of the problem
is that we can ony figure this out within bitbake itself. That means we'd have to teach bitbake
about the PN-PV format of PROVIDES which is breaking the separation between bitbake and the
metadata. We can't win :(.
Nobody that I know of is using or relying on this functionality so perhaps we should
just remove it instead which is what this patch does. Opinions?
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
bb-matrix: Clean before, rather than after, building
This makes sure the the first build starts from a clean state. Otherwise
one could have the first build affected by any leftover state from
a previous build.
This also leaves a working state behind after the final build.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Roy.Li [Tue, 13 Aug 2013 08:20:18 +0000 (16:20 +0800)]
sstate.bbclass: fix parallel building issue
sstate_package creates hardlink from sysroot to SSTATE_BUILDDIR, then
sstate_create_package will store SSTATE_BUILDDIR into a archive file by
tar, but once other packages install the same file into sysroot, the
creating the archive file will fail with below error:
DEBUG: Executing shell function sstate_create_package
tar: x86_64-linux/usr/share/aclocal/xorg-macros.m4: file changed as we read it
This kind of error is harmless, use --ignore-failed-read to ignore it.
The error in tar occurs when the timestamp of the file changes and this
can happen when the number of symlinks change. The file will be included
in the archive.
[YOCTO #5122]
Signed-off-by: Roy.Li <rongqing.li@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 11 Sep 2013 13:47:11 +0000 (14:47 +0100)]
clutter: fix compilation when building without X11 backend
If Clutter is building the Wayland backend but not the X11 backend, the Cogl
support doesn't get enabled so the Wayland backend (which uses it) fails to
compile.
Backport a fix from upstream to fix this situation.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 11 Sep 2013 13:48:19 +0000 (14:48 +0100)]
x11vnc: add dependencies for xrandr/xfixes/xdamage/xtst
Inspired by a patch by Martin Jansa but with these libraries required, as a VNC
server without them is suboptimal. Don't add an option for Xinerama as our X
server always disables it.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 11 Sep 2013 22:24:57 +0000 (23:24 +0100)]
xf86-video-vmware: respect opengl DISTRO_FEATURE
Conditionalise the GL dependencies on the opengl DISTRO_FEATURE so this driver
can build without the opengl feature active, as the configure script will
enable/disable the GL sub-driver depending on the presence of these libraries.
This is an interim patch to fix the autobuilder, a more complete patch heading
upstream to make the GL sub-driver deterministic will follow.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We can't build both recipes in the world build as there is a collision of package name
and PR values. Specificly the libasound-module-bluez which is the same in both goes
backwards from r5 (bluez4) -> r0 (bluez5) and the subpackage_metadata check fails:
ERROR: Recipe lib32-bluez5 is trying to change PR from 'r0' to 'r5'. This will cause do_package_write_* failures since the incorrect data will be used and they will be unable to find the right workdir.
[YOCTO #5165]
Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 11 Sep 2013 20:30:26 +0000 (21:30 +0100)]
xf86-video-vesa: change depends to refect reality
This recipe appears to have been a copy-and-paste from -intel, because it has
dependencies and configure options that xf86-video-vesa just doesn't have, such
as virtual/libgl and XVMC.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 11 Sep 2013 20:30:25 +0000 (21:30 +0100)]
ia32-base: only depend on GL if opengl DISTRO_FEATURE enabled
As Mesa refuses to compile if the "opengl" DISTRO_FEATURE isn't enabled,
mesa-driver-i9xx and the GLX X module have to be conditional in the ia32 machine
defintion too.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Stefan Stanacar [Wed, 11 Sep 2013 18:40:21 +0000 (21:40 +0300)]
lib/oeqa/runtime: smart: limit channel add to useful ones
Don't add inappropiate channels on the target.
This happens when building two different machines in the
same dir and then running the tests for each machine.
Signed-off-by: Stefan Stanacar <stefanx.stanacar@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Roy Li [Wed, 11 Sep 2013 09:17:32 +0000 (17:17 +0800)]
ptest-runner: trivial fixes and refine
1. ptest files may be installed under /usr/lib64/ for 64bit filesystem
or under /usr/lib/ for 64bit multilib filesystem, so we should check both
directories
2. If a soft link is linking to a directory under the same directory, we
only run once.
[YOCTO #5125]
[YOCTO #5126]
Signed-off-by: Roy Li <rongqing.li@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ting Liu [Wed, 11 Sep 2013 03:11:15 +0000 (11:11 +0800)]
libcroco: disable Bsymbolic if it is not supported on some hosts
When trying to build on my Centos 5.5 machine, got below error:
| checking for gtkdoc-mkpdf... no
| checking whether to build gtk-doc documentation... no
| checking for CROCO... yes
| checking if gcc supports "-Wl,-Bsymbolic-functions" flag...
| configure: error: -Bsymbolic requested but not supported by ld. Use --disable-Bsymbolic to disable
| Configure failed. The contents of all config.log files follows to aid debugging
|
| configure: exit 1
| ERROR: oe_runconf failed
Set --enable-Bsymbolic=auto to disable it when it is not suppported.
Signed-off-by: Ting Liu <b28495@freescale.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 11 Sep 2013 09:42:44 +0000 (10:42 +0100)]
image/populate_sdk: Ensure symlinks in target sysroot are relative
In the target sysroot of an SDK we can have target system absolute links
which don't make sense. This adds a script which fixes them up to become relative
paths instead.
[YOCTO #5020]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
init-live.sh: distinguish between a read-only image and a read-write image
The iso and hddimg share a common concept of 'live image', and they
use the same initramfs and thus the same init. However, that init
script in initramfs made a wrong assumption that the rootfs image
was read-only by itself. This is apparently not true for hddimg.
To make things work as expected, this init script should at least
distinguish between a read-only rootfs image and a read-write one.
This patch adds this ability to the init script. After this change,
the init script would be able to check whether the rootfs image is
read-only or not. If the rootfs image is read-write, the image will
be mounted and then booted directly. No union mounts will be attempted
in this case.
[YOCTO #5164]
Signed-off-by: Chen Qi <Qi.Chen@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
By using os.path.dirname(systemd_unitdir) we get the correct /lib directory instead of
dealing with possibly multilib directories. This address a QA Error for shipped/not
installed /lib with multilib and x32.
Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
systemd: Remove base_libdir if installed only for systemd_unitdir
If a recipe installs systemd_unitdir and it is a non-systemd build than
systemd.bbclass deletes systemd_unitdir (/lib/systemd/) but not
base_libdir (/lib). In this case if base_libdir is empty than following
QA Issue is reported.
ERROR: QA Issue: openssh: Files/directories were installed but not shipped
/lib
If base_libdir was installed due to systemd_unitdir installation than for
non-systemd build it should also be removed.
Signed-off-by: Muhammad Shakeel <muhammad_shakeel@mentor.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
