There were puzzling failures when you make a force recompile of any gcc
component. The error was in do_configure with cross-compilation badness
being detected in config.log files.
gcc is different in that many of the config.log files are generated during
the do_compile phase. This means this host contamination issue has always
been present but only shows up on a rebuild.
The fix is to force the appropriate configuration variable to "none required"
then gcc won't look in the bad locations.
[YOCTO #2279]
(From OE-Core rev: e7135f14f3ef37f93f5c28b319464f3d6dd9b8da)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If the toolchain is reused from sstate and ccache is installed, build failures
were occuring due to gcc trying to access the original sysroot rather than the
new one, particularly if the old sysroot existed but was not readable by the
current user.
This turns out of the an issue inside gcc to do with preservation of the sysroot
option. See the gcc patch for more details. It only triggers when preprocessed
sources are used which happens when ccache is used.
The same issue occurs with c++ and c++-cpp-output so the same fix is applied there.
[YOCTO #2074]
(From OE-Core rev: d3a3e81869631ba69874f6fc172240e3aac388f5)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This Fixes [Ycoto #2223] bug.
The gcc osdir is obtained in the do_install by invocation of
command "gcc -print-multi-os-directory". For x32 it returns gcc
osdir for the default abi which is x86_64. Fix this by adding
target abi parameter to the gcc command line to get correct gcc
osdir with invocation of command "gcc -mx32 -print-multi-os-directory"
(From OE-Core rev: e65b6a4282d5c4882d0565b79ccba99db90161ff)
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This fixes multilib issues if you try for example to use a BASELIB of /lib32
which wouldn't work without this change since the compiler install location
is taken from gcc -print-multi-os-directory which can still turn out to be
"/lib".
The reason is that a 32 bit gcc has no multilib code enabled and will always
return "." as that value rather than "../${base_libdir}" which our changes
to gcc enable and return in 64 bit mode.
(From OE-Core rev: cc953e2259bf77e9e1f7a1e3d0dbe1509c79477e)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Cleaning up non shipped empty directories.
(From OE-Core rev: e2fd32919e1d96136528c61677bdfeee2b064320)
Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This reverts commit 7893e74311e53882d8f93ecb95a6bd9f5b14651e since it breaks
the builds.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There are several invalid symlinks in gcc-cross-initial,
gcc-cross-intermediate and gcc-cross, these cause the error:(56 errors)
tmp/work/i586-poky-linux/gcc-cross-initial-4.6.3+svnr184847-r23/temp/log.do_populate_sysroot:
log.do_populate_sysroot:grep: /path/to/invalid/symlink: No such file or directory
Avoid creating invalid symlinks would fix this problem.
Use the:
[ ! -e file ] || do_something
But not use:
[ -e file ] && do_something
is because that if the "file" doesn't exist, then the whole statement
would return false, and bitbake treats this an error, so use the "||" to
let it always be true.
[YOCTO #2095]
(From OE-Core rev: 7893e74311e53882d8f93ecb95a6bd9f5b14651e)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
core-image-sato/core-image-minimal builds and boots
on all qemu machines
(From OE-Core rev: 40d7a4b70d342f75e13de7872fb62e2b9d6d40bf)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Using "1" with getVar is bad coding style and "True" is preferred.
This patch is a sed over the meta directory of the form:
sed \
-e 's:\(\.getVar([^,()]*, \)1 *):\1True):g' \
-e 's:\(\.getVarFlag([^,()]*, [^,()]*, \)1 *):\1True):g' \
-i `grep -ril getVar *`
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
dash does not like >& so be explicit and say 2>&1
(From OE-Core rev: 1d262630853e65be9167d904b934b581acf64182)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We have various variables which are either not quoted at all or are half
quoted. This patch fixes the bad exmaples so everything is consistent.
(From OE-Core rev: 30253358f5e76fb7b25be27198b4c125e0dbdf2c)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Corrected the error introduced by commit afb400e of dropping
DATE/SRCDATE out of the vardepsexlude list of do_patch.
(From OE-Core rev: 6546096d068b51cc9cf4557cdb35aee6f62838d5)
Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ensure do_patch have the same sstate hashes for nativesdk variants of
gcc family recipes, since they share the same source directory.
(From OE-Core rev: afb400e9c3ddac604906bde44df5da29a6c134d9)
Signed-off-by: Lianhao Lu <lianhao.lu@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
a) Use the system zlib to avoid building it for each gcc.
b) -initial and -intermediate don't need plugins or lto-plugin
so disable them.
(From OE-Core rev: a124a0883507caf4ffb7e25be666f6781b30abdd)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Dont use -q to grep we pipe to /dev/null anyway all we
care is the return status of grep
(From OE-Core rev: 9c575b5f37ee53eaf26bfdc468d4cbb661db703f)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Skip the staticdev QA test since the static archives belong
with the -dev.
(From OE-Core rev: 3de810a7f745b253d4b63eea353a31514d64a2e3)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
gcc-cross is installed into a package architecture specific directory
and is not meant to be machine specific. This patch replaces MACHINE_ARCH
with PACKAGE_ARCH to ensure this is really the case.
This was found by examining sstate checksums.
(From OE-Core rev: b3ca57418f47b16ed0a63d3291bf7d31e3ca5a45)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changes in WORKDIR were changing the do_patch stamps between gcc-crosssdk
and gcc-cross. This excludes the variable since in this case, we don't
need want those changes to affect the signatures.
(From OE-Core rev: 0de85d2ca3bc9674323302a35995dfddb6044a71)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Keep the *_nonshared.a in the libgcc-dev package as
required for linking, moved the other *.a libraries
into their respective packages (libstdc++ and libssp).
(From OE-Core rev: c2838d949ffd15282c72551b40c5fd81d0db1fa0)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
libgcov.a is moved to it's own libgcov-dev package, instead
of being part of libgcc-dev.
(From OE-Core rev: a82a3f8ec55e68a003420549392d638e565562fc)
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This was duplicated from the main SRC_URI and was causing incorrect
task signatures. This simply removes the duplication.
(From OE-Core rev: cc8d02c582a72d07fd08c59d14ee5720e01a5cd9)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
These tasks don't do anything, this just avoids the overhead of forking
to exec them.
(From OE-Core rev: 6708b3e908ae383922703390ac2d39f40348e1b3)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The fetch/unpack/patch/headerfix tasks are shared and hence their sstate hashes
should also match. Sadly this is not the case since:
a) gcc-runtime applies an additional patch
b) The do_headerfix task was missing from libgcc
c) The do_headerfix task is a shell task and hence depends
on all exported variables which can vary between cross and target
recipes.
To fix this, the patch moves the patch to the common code, adds
the headerfix task to a common include file and disabled shell
dependencies on the do_headerfix task since its clear in this case
we don't need thsoe dependencies since we just call sed.
With this patch applied, all these recipes now share common sstate checksums.
(From OE-Core rev: 2c4569801a710f34a695b8d2a0ee7fc127fb34e4)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some patches were missing Upstream-Status informantion,
this commit adds that in those patches.
(From OE-Core rev: 01c3669856538eb34d329780e03df2dc7fb985f2)
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Currently we have a problem in our cross compiler since we use
/usr/include/c++ to be default gxx-include-dir and then expect
the patch we did to do the relocation w.r.t. sysroot however it
does not quite work so and we end up gxx-include-dirs not respecting
sysroot. A small test case would be
tst-unique4.cc
and it would fails like
tst-unique4.cc:1:18: fatal error: cstdio: No such file or directory
compilation terminated.
weather we use --sysroot or not it does not matter
arm-oe-linux-gnueabi-g++ -S tst-unique4.cc
--sysroot=/home/kraj/work/angstrom/sources/openembedded-core/build/tmp-eglibc/sysroots/qemuarm
failed in same way.
so we redo the GPLUSPLUS_INCLUDE_DIR_with_sysroot.patch based on upstream
submitted patch which tries to relocate the gxx-include-dir and to
achieve the relocation it has to be specified w.r.t to --with-sysroot
directory. e.g.
--with-sysroot=${SYSROOT}
--with-gxx-include-dir=${SYSROOT}/usr/include/c++
if we configure gcc like above then it becomes relocatable when
we run the compiler and specify --sysroot=<blah> then g++ will search
for gxx-headers under <blah>/usr/include/c++
if sysroot is not defined then it will use the default sysroot
and gxx-include-dir will be w.r.t. default sysroot.
Tested on qemuarm
/arm-oe-linux-gnueabi-g++ -S tst-unique4.cc
--sysroot=/home/kraj/work/angstrom/sources/openembedded-core/build/tmp-eglibc/sysroots/qemuarm
-v
...
/home/kraj/work/angstrom/sources/openembedded-core/build/tmp-eglibc/sysroots/qemuarm/usr/include/c++
/home/kraj/work/angstrom/sources/openembedded-core/build/tmp-eglibc/sysroots/qemuarm/usr/include/c++/arm-oe-linux-gnueabi
/home/kraj/work/angstrom/sources/openembedded-core/build/tmp-eglibc/sysroots/qemuarm/usr/include/c++/backward
...
and if I now change --sysroot to something else
/arm-oe-linux-gnueabi-g++ -S tst-unique4.cc
--sysroot=/home/kraj/work/angstrom/sources/openembedded-core/build/tmp-eglibc/sysroots/qemuarm4
-v
...
ignoring nonexistent directory
"/home/kraj/work/angstrom/sources/openembedded-core/build/tmp-eglibc/sysroots/qemuarm4/usr/include/c++"
ignoring nonexistent directory
"/home/kraj/work/angstrom/sources/openembedded-core/build/tmp-eglibc/sysroots/qemuarm4/usr/include/c++/arm-oe-linux-gnueabi"
ignoring nonexistent directory
"/home/kraj/work/angstrom/sources/openembedded-core/build/tmp-eglibc/sysroots/qemuarm4/usr/include/c++/backward"
...
See now its looking for them in 'qemuarm4' sysroot
(From OE-Core rev: 28b772e42a20faebe1b4f415d28b42b7e0a424fb)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This script will be generated into the build directory of gcc-cross
It should be testing gcc and g++. libstdc++ tests are not run since
we build them as part of gcc-runtime but we can test them here by
building them with 'make all' and then running the tests
The script expects passwordless ssh access to target and is used
in form
./arm-oe-linux-gnueabi-testgcc kraj@192.168.7.2
inside the builddir of gcc-cross
(From OE-Core rev: 130b534fdb0b292158981a12d7d5f01c1a14cb2a)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Right now for cross recipes e.g. gcc-cross and binutils-cross
we specify --disable-nls .... --enable-nls on configure cmdline
the --enable-nls coming from gettext bbclass.
So we disable nls for all cross inheriting recipes in gettext
bbclass and then we remove the extra --disable-nls in gcc-cross
and binutils-cross
This patch needs testing. Please help
(From OE-Core rev: d66b379f809b9c75981848fcc71ed5de13382bf7)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Make get_gcc_multiarch_setting more elegant. Use a dictionnary
to store the config options and replace bb.data.getVar with d.getVar.
Remove i686 from the architecture list because it doesn't seem
to be a valid TARGET_ARCH any more in OE.
Configure gdb (gdb and gdb-cross) with --enable-64-bit-bfd if
multiarch DISTRO_FEATURE is present
(From OE-Core rev: 0b42b2fc118bef8ce0d1473b146673182f5d1f78)
Signed-off-by: Julian Pidancet <julian.pidancet@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This patch introduces a distro feature which enables gcc to produce
both 32bit and 64bit code, and enables binutils to operate on both
32bit and 64bit binaries. It differs from multilib toolchains in
that it does not require to compile a version of the libc for each
architecture variant. However, the code produced for the secondary
architecture will not be linkable against the libc.
v2: - Renamed the feature name from "biarch" to "multiarch". The GCC
installation manual claims that the mips-linux can be made a tri-arch
compiler (http://gcc.gnu.org/install/configure.html)
- For x86_64, the compiler is made bi-arch by default, so nothing
has to be done in particular.
- I analyzed the gcc/config.gcc from GCC sources and added in this
patch all the architectures that could be made biarch with the version
of gcc currently used in OE, which are powerpc, and sparc, in addition
to x86. mips and s390 will probably be supported in future versions of
gcc. For x86 and sparc, only the --enable-targets=all option is valid
to make this work (this option doesn't have any other side effects than
making the compiler bi-arch). For powerpc, I used the
--enable-targets=powerpc64 option (although 'all' also works).
Note: - Untested on powerpc and sparc. But I believe it works the same
as with x86.
- gcc in meta-toolchain is also made multiarch.
(From OE-Core rev: 99e295ef30ba02db3966c66619807c037ef5089f)
Signed-off-by: Julian Pidancet <julian.pidancet@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When I was trying self-hosted-image, eglibc's do_install failed in the target:
ERROR: cannot stat bootparam_prot.h:
the cause is: rpcgen doesn't work properly: rpcgen can't exec /lib/cpp since
it doesn't exist.
According to http://tldp.org/LDP/Linux-Filesystem-Hierarchy/html/lib.html:
"if a C preprocessor is installed, /lib/cpp must be a reference to it, for
historical reasons. The usual placement of this binary is /usr/bin/cpp".
Typical distros, like Ubuntu, openSuSE, Fedora and RHEL, all comply with
the rule.
Actually in meta/recipes-devtools/gcc/gcc-package-target.inc, we do try to
package ${base_libdir}/cpp:
FILES_cpp = "\
${bindir}/${TARGET_PREFIX}cpp \
${base_libdir}/cpp \
${libexecdir}/gcc/${TARGET_SYS}/${BINV}/cc1"
But unluckily we didn't create a symbol link in do_install.
This patch adds the symbol link.
(From OE-Core rev: 58fb9fbce2af05a469691390e75aebb2de16ca96)
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Backport fix for PR 47551 fixes the ICE seen on armv7-a/qt4-x11-free
Bump up SRCREV past gcc 4.6.2 release
(From OE-Core rev: dd2fdf9f5a3923c37e4ea2e46e347bb0657c2f5b)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(From OE-Core rev: da7bf75bcdd5759a0f551dcb7a0326aa2f40921c)
Signed-off-by: Michael Brown <Michael_E_Brown@dell.com>
Port patch from base openembedded. Since 4.6 already has fixes for config.gcc,
the fix only requires a one line change to gcc-cross4.inc.
The patch was imported from the OpenEmbedded git server
(git://git.openembedded.org/openembedded) as of commit id
3aa8afe97e9cf1340feb9c4442a6ed88b7e32c96.
gcc-4.5: Fix toolchain builds for SH4/SH3
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This fix is needed for gold to work. Otherwise
connman fails to build since it used hidden weak
symbols.
See
http://gcc.gnu.org/bugzilla/PR32219http://www.cygwin.com/ml/binutils/2008-02/msg00239.html
The fix proposed to gcc had reviews which were not addressed hence the
patch is not yet
applied to gcc upstream.
connman can also have workaround by changing the visibility of these
symbols to be default
__attribute__ ((weak, visibility("hidden")))
to
__attribute__ ((weak, visibility("default")))
in include/plugin.h
(From OE-Core rev: 3cb2b003db7371b3a47d02c08352a262e1e419b4)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We force the C locale when running builds for determinstic error messages. We
therefore have no need to NLS support in binutils cross or gcc cross.
We also don't need the standard base/autotools dependencies for our
toolchain components since we don't autoreconf these.
This patch turns off nls and cleans up some of the dependencies resulting
in a slightly less convoluted set of build dependencies.
(From OE-Core rev: 54a3e2ee37003fc56af0339f857b0b6442790c26)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Depending upon what hash style is in use this uses
the right flag for setting the hash style type. This
fixes the QA errors about missing GNU hash style reported
in gcc-runtime build particularly libgcc
(From OE-Core rev: f8edd9b872bcf14da037bd0b501ccc8c6fcc79bf)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
a) There is a QA warning from a .so being present in a main package.
In the case of the plugin library for gcc, this is allowed.
b) Remove the unwanted libiberty.a file with the strange path. We
don't need/want this and this removes an unpackaged file warning.
(From OE-Core rev: ca36a3edf9cede9fa0d73ba1a9538ab467cb5e3c)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
First, this lib is usefull for coverage analysis-enabled building.
Second, this fixes the warning about unpackaged files in libgcc recipe.
(From OE-Core rev: 2a807a98d8be3f486e703321773db32657c71d9e)
Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Signed-off-by: Koen Kooi <koen@dominion.thruhere.net>
Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
