When the user doesn't have rights to access the installation directory,
the sdk installation will fail, add sudo to fix this.
[YOCTO #4760]
(From OE-Core rev: 040010d04672c93f18d60308ecf3c26a26ec5fd3)
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When run "autoreconf" in toolchain, there is an error if the host's perl's
version is not the same as the one in the SDK, the error says that the
executable perl mismatches the perl lib's version.
This is because most of the autotools' scripts use the "#!/usr/bin/perl -w"
which is host perl, but the gnu-configize uses "#! /usr/bin/env perl" which
invokes the perl wrapper in the SDK, and the wrapper will set the PERL5LIB to
the SDK which causes the mismatch. We can make all the perl scripts to use the
host perl or the SDK perl to fix this problem.
[YOCTO #4758]
(From OE-Core rev: 487d1fa7b79e89518494986461c157bace842613)
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The perl module for automake has an embedded path in it, this needs
to be relocated.
Older versions of 'file' do not return the "ASCII" text in the output for
a perl module file. Hence, the regex pattern didn't match perl module and
they were not getting relocated at all on older distributions.
1) On CentOS release 6.4, the version of file is 5.04
$ file /usr/lib/perl5/Config_heavy.pl
$ /usr/lib/perl5/Config_heavy.pl: Perl5 module source text
2) On Ubuntu 13.04, the version of file is 5.11
$ file /usr/lib/perl/5.14/Config_heavy.pl
$ /usr/lib/perl/5.14/Config_heavy.pl: C source, ASCII text, with very long lines
[YOCTO #4550]
(From OE-Core rev: 7cfab2c895bf4daa7716fb7509e367bf74f016e4)
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Spaces are not handled properly in some parts of oe-core and it's safer
to abort toolchain installation if path contains spaces. Even though
we fix space handling in the toolchain installation script, there are
various other parts in the toolchain (perl scripts, sysroot path passed to
toolchain binaries, shebang lines) that would need special handling. So,
for now, just bail out if path contains spaces.
The checking for spaces in the path is done after expanding relative
paths to absolute and tilde conversion.
[YOCTO #4488]
(From OE-Core rev: 8c35ba2d3048ce69f74f72cb2676e4bc162cfb63)
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Recently it was discovered that many Fedora hosts have a broken version of
make 3.82. Add make to the buildtools-tarball, as well ad modify make to
support building a special nativesdk version.
(From OE-Core rev: ea972fc4ce2268f01be6beeafd27dd949d800f9d)
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The Savannah bug 30612 describes two different issue. The first,
previously fixed, errors parsing multiple objects in parenthesis.
The second, (this issue), extra white space contained in the
parenthesis.
The fix was backported from the current make git tree:
http://git.savannah.gnu.org/cgit/make.git/commit/?id=b06b8c64a29a5ba3a8daecd829fa2f98d42cb285
(From OE-Core rev: 681b4c2cd9830cd523080aa830748d9c3367c7c9)
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The original fix worked some of the time, but not on all machines. Fix
this by applying the change in a different way. Following the example of
the existing BUILD_CFLAGS.
Below is the commit message from the original change to help explain why
this is needed:
Configure identifies a number of warning flags (WARN_CFLAGS and
WARN_CXXFLAGS) from the $CC value. The cross compiler may be different
from the host compiler and may not support the same set of flags. This
leads to problems such as:
cc1plus: error: unrecognized command line option "-Wno-narrowing"
cc1plus: error: unrecognized command line option "-Wno-overlength-strings"
Work around this problem by removing the warning flags from the
BUILD_CXXFLAGS value, in a way similar to the BUILD_CFLAGS.
(From OE-Core rev: be21c6e8e4f810e826538337dac6e34ed96e1f6f)
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
defs.c dependes on gdk-types.defs and gtk-types.defs. When parallel
compile, it may fails with:
"IOError: [Errno 2] No such file or directory: 'gtk-types.defs'"
Add them to dependences of defs.c to fix this issue.
(From OE-Core rev: edf278eec71552bcd3ac661dce8e8b7489463f6a)
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The installation of __glib.so is invoked by install-data-am target
which are generated by automake. installing libpyglib-2.0-python is
invoked by install-exec-am.
"make install" will firstly install libpyglib-2.0-python, then install
__glib.so, the sequence should not be broken, since _glib.so has
dependence on libpyglib-2.0-python. But when enable parallel install,
the sequence maybe break then installation fails with:
"/usr/bin/ld: cannot find -lpyglib-2.0-python"
Disable parallel install to fix this issue.
(From OE-Core rev: e87a0c81c77d11f892a34c2d14ffbeb4342d32dd)
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fix openjade-native build failure when build directory name contains the
characters ".a".
(From OE-Core rev: f005670ee8f6d02e0b0517a48b47b364f0bddf2d)
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add strace configure options libaio and acl to enable or disable build
with them. Then PACKAGECONFIG can be used to handle dependency.
Set "libaio" as a default feature of PACKAGECONFIG.
(From OE-Core rev: 9b6b7973a5e49fe9cd283c5b4af31e8a7940e7b5)
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Update strace to 4.8:
* Update License file.
* Remove the backport patches which are already in version 4.8.
* Add file git-version-gen from git repo. Without this file configure
fails.
(From OE-Core rev: 3ae2592b40900caaebe2452c875912f82f9e5200)
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Unit test clock.test check the boundary year 2099 which is not
inclusive. Include 2099 to be a valide year number.
(From OE-Core rev: c6386c8661f18f6b9104c74345120a56e2330ea4)
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Remove hardcoded library install path and then install library to
${libdir}.
(From OE-Core rev: ec06d828d8ee623b0064713583344db84496ef1e)
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This backported patch will fix this Xorg issue and, probably, many
others which didn't show up yet.
[YOCTO #4737]
(From OE-Core rev: c659903841da7aa44675d2970d1b92841b91d29e)
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
On some overloaded hosts, shell commands of autogen may can not
finish in 5 secs. This has caused many build failures, so increase
the timeout limit to fix this.
(From OE-Core rev: 92c7a0bec3f3c9fec04a6019aa9cee395ba5a9ab)
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We've been using 1.5.1 for a while now with success, so in line with the
OE-Core policy of avoiding multiple versions of the same recipe let's
drop this version.
(From OE-Core rev: c26fd671466f42d76cbd99e275dc91fc322062c4)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The header files for libopkg belong in the package libopkg-dev rather than
opkg-dev where they are placed by default.
(From OE-Core rev: f79cd4208340b56afc044d8e3abc72d43d395062)
Signed-off-by: Paul Barker <paul@paulbarker.me.uk>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When opkg is built outside the source directory, libopkg headers aren't all
installed as the search path for headers only includes the directory that make
is being ran in, not the source directory. So we fix this by adding
$(srcdir)/*.h to the list of header files to install.
Without this patch, "/usr/include/libopkg" will only include the file config.h.
With this patch, other files such as "opkg.h" are present which are needed to
compile software which uses libopkg.
The new patch has also been submitted upstream as a fix to opkg itself.
(From OE-Core rev: 873689bbabba25e7be5c12317c04519a7bc8d0ef)
Signed-off-by: Paul Barker <paul@paulbarker.me.uk>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Currently if you build nativesdk-python, install it and try and run it,
you see it search in the default nativesdk path for terminfo files
when the readline module is used.
If terminfo files cannot be found, or if the ones found are incorrect,
the system may emit control characters which confuse commandline
processing.
This patch sets the TERMINFO_DIRS variable to ensure the correct locations
are searched for terminfo files, starting with the nativesdk terminfo files
and falling back to the host systtem's.
(From OE-Core rev: 682861166f39fbdcd0c9b923139faab2d40362cf)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This fixes errors in packages using python( build with gcc 4.8)
|
/home/jenkins/oe/shr-core-branches/shr-core/tmp-eglibc/sysroots/qemuarm/usr/include/python2.7/modsupport.h:27:1:
error: 'PyArg_ParseTuple' is an unrecognized format function type
[-Werror=format=]
| PyAPI_FUNC(int) PyArg_ParseTuple(PyObject *, const char *, ...)
Py_FORMAT_PARSETUPLE(PyArg_ParseTuple, 2, 3);
| ^
| cc1: all warnings being treated as errors
| cc1: all warnings being treated as errors
(From OE-Core rev: 5745a482a85c064a1eec960aff104cf8ce588e30)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add a symlink to the native Python so that scripts can just invoke
"nativepython" and get the right one without needing absolute paths (these often
end up too long for the #! parser in the kernel as the buffer is 128 bytes
long).
The name "nativepython" was chosen to match the existing "nativeperl" which
serves the same purpose.
(From OE-Core rev: d99dac66e3e121e96c4d8a29aee846d4b8f38622)
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>
create_wrapper was previously abused to invoke the Perl interpretter directly
instead of relying on #! parsing, which is read into a 128 byte buffer so breaks
with deep build directories.
However we already install a "nativeperl" binary into the native sysroot, so
use sed to change the #! line to /usr/bin/env nativeperl.
(From OE-Core rev: c4f987239fe78783d732df673a5e592bb05a877d)
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>
Patches removed:
0001-doc-Fix-texinfo-table-markup-in-qemu-options.hx.patch
- applied
0002-docs-Fix-generating-qemu-doc.html-with-texinfo-5.patch
- fixed different upstream
Revert-vmware_vga-Add-back-some-info-in-local-state-.patch
- fixed different upstream
arm-bgr.patch
- this should be also fixed upstream, if still issue please resend.
fallback-to-safe-mmap_min_addr.patch
- this should be also fixed upstream, if still issue please resend.
linker-flags.patch
- unneeded with makefile changes
from configure the obsolete --audio-cards-list option has been removed.
(From OE-Core rev: 408f933f3b5523fc26ee818a16bb4162c4343192)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
run-postinsts is generic and will handle rpm delayed postinsts just
fine. No need to use another recipe.
(From OE-Core rev: 1195a188f366fbbb04b9379d7ea3c7c02130cc98)
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This recipe is used only when the package manager is not deployed on
target and there still are delayed postinstalls that need to be run on
target. At the moment, run-postinsts is targeted to dpkg/opkg. RPM uses another
recipe, called rpm-postinsts which, in turn, is used even when the rpm
package is deployed.
This patch intends to make run-postinsts generic and deprecate
rpm-postinsts. Here's why:
* when opkg/dpkg are not deplyed, the meta-data files are useless.
Hence, the awk script is not needed. The postinstall files are already
saved in /etc/(deb|ipk)-postinsts directory and we can use the same
procedure as in rpm's case;
Also, the recipe is moved out from meta/recipes-devel/dpkg to
meta/recipes-devel/run-postinsts.
(From OE-Core rev: 0222e6740ee4670bbf42d246ab3ecad415ffcdee)
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We need to be able to generate a standalone tarball containing tar/git so
add nativesdk versions of the appropriate recipes to allow this to be possible.
Tweak the git perl paths to avoid warnings when building the nativesdk version,
ensure the binaries are wrapped correctly and avoid update-alternatives in
nativesdk-tar.
(From OE-Core rev: c91bb8c76e3bd45690e66f3de79cd3adfe45f600)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Base this on DISTRO_FEATURES, because we don't want it
unconditionally enabled.
(From OE-Core rev: 6956fb01e0c7332e4d8b59cbcb1cd997c54191fc)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
upgrade from 0.9.66 -> 0.9.69
- COPYING file contains now more detailed information.
- reworked a patch because it didn't apply.
(From OE-Core rev: 376612cbf7fa5c68ae3adb6f4c1b4427e69a1871)
Signed-off-by: Andrei Dinu <andrei.adrianx.dinu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add build directory to include directories by -I${B}/include which will
be searched before standard system include directories.
Both libunwind and gcc supply unwind.h, once gcc use libunwind's unwind.h,
the compiling will fail.
This patch is generally not applicable to the upstream as they do not
use libunwind.
(From OE-Core rev: 2b47bce78536a800205b2385bba69038351545e5)
Signed-off-by: Roy.Li <rongqing.li@windriver.com>
Signed-off-by: Jeff Polk <jeff.polk@windriver.com>
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When cross compiling the target compiler, both the cross-compiler
and the host compiler are used. However, the -W options used were
discovered from the cross-compiler and may be incompatible with the
host compiler.
(From OE-Core rev: 1f40202ce3c1282674b6cea39ef709972275f201)
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Previously count actually behaved as end, and did not take start into
account.
(From OE-Core rev: 4fe2b01bfd2831b002e7138dadbc0437df6e9ed6)
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This was a floating dependency that got added in recently and was
caught on the Autobuilder, disable building with LZMA to ensure
consistent builds
(From OE-Core rev: 1e58fc8f6ac8f13d6c86a3ae340d90dd53b3ec27)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Back port two patches to fix the error of complex expressions before @l/@h.
Error sample:
test.s: Assembler messages:
test.s:1: Error: operand out of range (0xffffffffffff8000 is not between 0x0000000000000000 and 0x000000000000ffff)
Fix Bug 4524.
(From OE-Core rev: 9ba812ab1f613d28f9eb3192d2ff1a34dfce33e4)
Signed-off-by: Zhenhua Luo <zhenhua.luo@freescale.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The apt recipe had the correct checksums, the -native did not. Depending
on which downloaded first, the build could succeed or fail.
This patch corrects the checksums for -native to match the apt recipe.
(From OE-Core rev: 86d0708bb59952a139e705a8c396e70dd0084b75)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The argument parser in RPM was patched to look for <binary>.real, but since the
wrapper now fakes the right argv[0] rpm wasn't able to parse any options.
(From OE-Core rev: a67b4cfc41819ed77ed2bc4246228e9d006a4317)
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
dpkg has no DEPENDS on lzma so turn it off explicitly.
[YOCTO #4518]
(From OE-Core rev: 12fbd693f4565e66d10af4e801e7435996d67e76)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
After e8d2a2a3646f964ec61ece62e14788cd7184dd01 [OE-Core], file --version
returns file-5.14 instead of file.real-5.14 so the patch is no longer
necessary and causes the build to fail with the following:
| Cannot use the installed version of file (file-5.14) to
| cross-compile file 5.14
| Please install file 5.14 locally first
(From OE-Core rev: f89f705da9bffd9d10628e90a415db4411d22b4e)
Signed-off-by: Jonathan Liu <net147@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fixes the following failure at do_install building python-smartpm-native
if gettext-native has not already been built and gettext tools are not
installed on the host:
| compiling locale/it/LC_MESSAGES/smart.po -> locale/it/LC_MESSAGES/smart.mo
| sh: msgfmt: command not found
...
| creating $D/usr/share/share/locale/it/LC_MESSAGES
| error: can't copy 'locale/it/LC_MESSAGES/smart.mo': doesn't exist or not a regular file
Note that we need gettext-native in DEPENDS and not "inherit gettext"
here because for native variants, gettext.bbclass instead adds
gettext-minimal-native to DEPENDS and that does not provide the msgfmt
command.
(From OE-Core rev: e8d903e2d5e0c0df18dfd9561c3f8ef340297f1f)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If the increment > 1 and the start > 0 then the calculation for the
minor device number was incorrect.
(From OE-Core rev: 0a878426bee607a7d961ba475a7ec7e89115df35)
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When creating a block of devices (i.e., when count > 0), the wrong
path was used with the call to chown(), effectively trying to change
the owner of some (probably) non-existent file. Thus the created
device nodes were always owned by root.
(From OE-Core rev: e7796880164d6a37c2699a94e1c5391337c5eaa5)
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
