GCC 4.8.0, 4.8.1 and 4.8.2 can generate broken epilogues for the
ABI used by the kernel. Apply the patch that is included for GCC
4.8.3 from http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58854.
The issue was found on Yocto/Dora and the patch should be backported
to this branch. A kernel built with Dora's GCC 4.8.1 misbehaved on:
while true;
do
(for i in `seq 1 100`;
do
echo "Log message... $RANDOM";
done) | logger;
done
busybox's syslogd would from time to read a huge negative value and
then exit, strace would get stuck waiting on a syscall. After this
patch it appears to work better.
(From OE-Core rev: 3004eb3b7ee5fd8dfe9c4e5749b4e125d0bd4b59)
Signed-off-by: Holger Hans Peter Freyther <holger@moiji-mobile.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We've had 4.8 around for a while now, I'm not aware of any issues with
it so we can drop the older 4.7 version.
(From OE-Core rev: 15685dccd37f7636bbdd56af04a9dbc9b7595612)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fix statement indenting and spacing issues that I happened to notice.
(From OE-Core rev: d11e297b007aba625b398c52952ec929c3b02b83)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
A lot of our recipes had short one-line DESCRIPTION values and no
SUMMARY value set. In this case it's much better to just set SUMMARY
since DESCRIPTION is defaulted from SUMMARY anyway and then the SUMMARY
is at least useful. I also took the opportunity to fix up a lot of the
new SUMMARY values, making them concisely explain the function of the
recipe / package where possible.
(From OE-Core rev: b8feee3cf21f70ba4ec3b822d2f596d4fc02a292)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Currently the code has problems differentiating between "gcc-cross" and "gcc-cross-initial"
sstate files. We could add in a ton of special casing but tests show this isn't scaling
well. Using a more unique separator resolves the issue.
The choice of which separator to use is a hard one. We need something which isn't commonly
used in PN, PV, PR, *_OS and *_ARCH which rules out '-', '_' and it needs to work ok with
webservers/http which makes ';' and '%' harder.
The change also sets SSTATE_SWSPEC globally since writing out differently named siginfo
files for the fetch/unpack/patch tasks is a waste of diskspace, the hashes match for
all PN in the majority of cases and if they don't, its not a big issue as the hash is
different. This makes the results from sstate debugging more understandable.
(From OE-Core rev: 6f823a23c5f1d0ffa0a27db1c1bc1907de788505)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
For a shared workdir, any one of the fetch/unpack/patch tasks may run yet the
PN and architecture fields in SSTATE_PKGSPEC may differ. This makes looking up
the appropriate siginfo file near impossible.
I've tried several different ways of resolving this and this is the neatest
solution I could find, its still rather ugly. I believe the usefulness of
better sstate debugging outweighs the ugliness of the code.
This patch also changes the sstate_checkhashes() code to look for siginfo
files rather than the actual sstate packages themselves. This means the
function can be used in other contexts to find info files for tasks that
may not have sstate data. It is assumed that sstate mirrors will have both
files available. This is done to allow bitbake to query whether tasks have
matching signatures in sstate directories or not.
(From OE-Core rev: 068e4289b597699cbff2dfde44ba833af4535281)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We had disabled the sdt from configure, let's also disable it from
confgure.ac to keep them compatible.
BTW, the libstdc++-v3 of gcc-4.7 doesn't use the sdt, so we don't need
to edit libstdc++-v3/configure as gcc-4.8.
NOTE, this commit edit the patch gcc-4.7/disablesdt.patch directly.
[YOCTO #5657]
(From OE-Core rev: 452c6afc3883b0304e06eb1764614348fb6599c8)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We may meet such an error when building gcc/libstdc++-v3:
gcc-4.8.1/libstdc++-v3/libsupc++/unwind-cxx.h:41:21: fatal error:
sys/sdt.h: No such file or directory
We already have a patch to disable the sdt for gcc, we also need disable
it for libstdc++-v3.
BTW, we need edit both configure.ac and configure to make them keep
compatible.
NOTE, this commit edit the patch gcc-4.8/0031-Disable-sdt.patch directly.
[YOCTO #5657]
(From OE-Core rev: 32854af3cc6c0626620e827dc1915f61c51250b8)
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Without this sstate builds can fail with missing dependencies.
(From OE-Core rev: f92ebf78d94cb8f4010f8d444d1d0336c1fb1341)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
gcc 4.8 fortran presents some challenges:
* libquadmath headers need to be in the libexec include dir. It turns out
to be easiest just to manually do this.
* libgfortran configure needs libquadmath to be compiled. This means
a separate recipe is needed (the alternative is gross hacks)
* the libtool uses to link libgfortran doesn't have our improved rpath
handling and puts bogus RPATHS into the libraries. We can avoid this
by tweaking libtool with sed.
This patch resolves those issues. Any user of fortran does need to DEPEND
on libgfortran in order to trigger it to build but this shouldn't be a major
issue.
(From OE-Core rev: a5e7ee5770b9e0cf719c573efffd874440f74289)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Using the contains function results in more optimal sstate checksums
resulting in better cache reuse as we as more consistent code.
(From OE-Core rev: 9c93526756e7cbbff027c88eb972f877bcb1f057)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When fortran was enabled, builds were failing due to a extra files.
For now we can remove these and avoid the build failure.
(From OE-Core rev: 2e60ef7fe63974e443a9ddc25c5eb4249ec37963)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
gcc has cross and target components with a shared workdir. The unpack umask
settings need to match for all of these. We need to use strings in each
case to ensure the sstate code matches them correctly.
This patch tweaks various things to ensure the change adding the unpack umask
change doesn't break the compiler builds.
(From OE-Core rev: 67162438ee9c402b23c32853af9d313949eb6e4a)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
MIPS gcc is not configured with multiarch enabled. This causes
compiler generates local label with $ prefix, which is specified
in default o32 abi. It is not recognized as local symbol by n64
assembler, so we get a lot of unexpected external symbols. We
should configure MIPS gcc with --enable-targets=all, as for other
archs.
(From OE-Core rev: c210393b3e0c54606493e311d7c2040efebc4dcd)
Signed-off-by: Lei Liu <lei.liu2@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
PowerPC toolchains can use the OS "linux" or "linux-gnuspe". This
patch links them together so the one cross-canadian toolchain can support
both.
GCC_FOR_TARGET is set for the GCC recipe as otherwise configure
can pick up an incorrect value.
[YOCTO #5354]
(From OE-Core rev: a1d6331238982b0c5d39b0a18794f6654b00d46a)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The cross-canadian compilers are now build once per architecture but were being
installed into tune specific locations which is incorrect. This adjusts things
so they are make TARGET_ARCH specific. We gain the tune specific parts from the
target sysroot which remains tune specific, the compiler and tools are independent
ot that.
binutils/gcc require sysroot options but since we reset at runtime, these shouldn't
have dependencies in the sstate checksums. They are therefore also excluded.
With these patches, switching machines does not result in a rebuild of *-cross-canadian
and the compiler is correctly located and referenced in the target images.
(From OE-Core rev: f58acab6414fe96d9e07ebbe86b348d2ac2bed5f)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We only care about the end result in this case, not the specific inputs
that went into determining the gcc option. This change updates the code
to reflect that.
(From OE-Core rev: 83055511dab1e2aeefe84262ebe52595848c844b)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If sed-native is built before these programs, hardcoded paths to sed-native
can end up in scripts and other parts of the system which may cause issues
if they are later used from sstate and sed-native is not installed.
To avoid this, this patch changes the global site configuration to
specify that plain "sed" is fine to be used. We need to spell this
out for gcc since it doesn't see the site files since we don't autoreconf
it. We can remove the values from libtool. We tell perl to use "/bin/sed"
since it requires a path and the system sed should be just fine for it.
[YOCTO #4971]
(From OE-Core rev: 2ec171cb188601bf18c6c2895870907024b1c52a)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The --enable-dependency-tracking option was added to workaround build
issues in libatomic. This fixes that build problem properly and removes
the flag since the dependency tracking code appears to be full of races
which are much deeper and harder to fix.
As per the automake manual, dependency tracking is only useful and worth
the build performance cost if you are doing more than one compile of the same
source code which in most cases we are not so this is a good thing anyway.
(From OE-Core rev: a3b665a80abed4c0659925e1cceb1568af023711)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* 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
(From OE-Core rev: 148c7e990831ed708f7fb064ec8f05657a323850)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
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.
(From OE-Core rev: 1b8e4abd2d9c0901d38d89d0f944fe1ffd019379)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Add packaging for libgfortran and libquadmath as well as tweak the packaging
for libmudflap since it was broken.
(From OE-Core rev: 8a726d14a345ef35c6d8d8e369bf3691cee879bf)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
These have been deprecated for a long time, convert the remaining
references to the correct modules and prepare for removal of the
compatibility support from bitbake.
(From OE-Core rev: 6a39835af2b2b3c7797fe05479341d71a3f3aaf6)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* http://gcc.gnu.org/bugzilla//show_bug.cgi?id=57717#c7
* fix the segfault issue of dd on e500v2 targets
Since double-float is disabled for e500v2 targets build due to ICE of gcc-4.8.1,
accordingly %Ld format of sprintf is disabled.
Address Bug 4910 - [p1022ds]urandom: segmentation fault
(From OE-Core rev: a4fefac26d91bc56d5d28e1c9973a189d2509d45)
Signed-off-by: Chunrong Guo <B40290@freescale.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
For some platforms its useful to be able to configure LTO so provide a variable
to allow this to happen.
(From OE-Core rev: e4582a51a2500ad3b418e53170f5fb6b2cbd98a5)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
See the patch headers for more information about the cross-canadian build failures
these patches avoid.
(From OE-Core rev: 2bae60b8a3cb7783c06e35a2962e56110e876957)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* Remove the duplicate EXTRA_OECONF_PATHS that is overwritten
* Merge the do_compile and do_compile_prepend
* Group dependency and configuration variables together
(From OE-Core rev: 430b1d6fcbf60df35036fe5bbf8e55f7fb8f5341)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This also has the advantage of removing the confusing sdk naming which
has been purged everywhere else in favour of cross-canadian.
(From OE-Core rev: cbb63ca9e7e6d397198808e862f812f1012c74a7)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Hardcoding -nativesdk as the sdk package architecture is inflexible. We may have
multiple different target OS and we need a way to be able to separate them. Turning
this into a configurable value allows the flexibility we need to build different
SDKMACHINEs with different OS targets.
The commit should have no behaviour change, just makes things more configurable.
(From OE-Core rev: a2110e86b98d646e136de9ec6b8e668079b0d4f4)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This removes duplication and follows the pattern of the other gcc recipes.
(From OE-Core rev: 3296c896f5a5ef7dd50ab4e00ddbf1c2476462dc)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some SDK platforms have elfutils support, some do not, therefore allow
this to be configured.
(From OE-Core rev: 717e940d2c2beccfda31dda16a2d0d6d9a495042)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The include was useful historically, its not anymore so lets remove it.
This should have no functional change except on any layers directly depending
on it or gcc-cross.inc but even then it would only impact sh4 and is easily
fixed if there was a problem.
(From OE-Core rev: e7e8fe11c34bf05179f3bbaa2fb1af7b7125696a)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
I don't understand why we're building the target libgcc in the canadian-cross build
since it should have been built elsewhere. The compiler configuration isn't correct
to build a working target libgcc in all cases anyway.
To avoid various weird build errors, stop building it.
(From OE-Core rev: b0d1ab6242b6ce2bcdd0e4e3e61600344fcd2907)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
On platforms like windows, executables have extensions. Whilst I'm not proposing
we wholesale support windows extensions, this small tweak allows a cross compiler
targetting mingw to be built which does seem like a good use case.
The patch therefore adds an EXEEXT which the mingw layer can set for the libexec
symlinks.
(From OE-Core rev: 8dcf0d95b654fa6cc56193168aaa744052ad8ffc)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We already use relative links for other gcc libexec links, this changes the sdk
do_install to match elsewhere and use relative symlinks too. This makes things
slightly easier in the SDK installation process and standardises.
(From OE-Core rev: ecfa1141e731224cc5a099f8dfd22878f23359ec)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This disabling of PARALLEL_MAKE has been forward ported for gcc-cross-canadian
since at least 2009-09 and gcc 4.3.3, probably older.
I've tested this with high values of parallel make and it all seems to work and
we usually build gcc with parallel make so it seems unlikely there are issues.
Lets therefore enable it.
(From OE-Core rev: cee90d230899eb1255d586e6ee4fad0d94348cfd)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This should fix the problem seen where gcc ICE
was happening when compiling iperf with older 2.6.x
kernel
Test this patch by reverting below commit in meta-fsl-arm
commit daf582c93a7283fb0af3b25fe2ada48f4c9985c4
Author: Otavio Salvador <otavio@ossystems.com.br>
Date: Tue Jul 2 11:52:51 2013 -0300
perf: Disable FPU tune for i.MX5 SoCs to workaround GCC ICE
(From OE-Core rev: 8ab1d16b6c6d946b625b6872e5d0f155206f4bad)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
CC: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
libtool is picking up libstdc++.so from /usr/lib when trying to link
libasan due to libstdc++.la containing libdir="/usr/lib". If compiling
for x86 and the host has 64-bit /usr/lib/libstdc++.so, the compilation
fails linking libasan with:
/usr/lib/libstdc++.so: could not read symbols: File in wrong format
To resolve this, patch libtool to look for the library in the path the
.la is contained in rather than use the libdir which usually points to
a host path.
[YOCTO #4879]
(From OE-Core rev: ec95dfeea1f17eb232563e105085852058a86c0b)
Signed-off-by: Jonathan Liu <net147@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This fixes the issue where g++ is not able to
find C++ headers when sysroot is '/'
The patch needs to be upstreamed into gcc as well.
[Yocto #4812]
(From OE-Core rev: 18537025ca8777a45cb24f1d9ee781323695607a)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Holger Hans Peter Freyther
Richard Purdie
Paul Eggleton
Robert Yang
Khem Raj
Lei Liu
Martin Jansa
Chunrong Guo
Jonathan Liu