Age | Commit message (Collapse) | Author |
|
Remove package libstdc++-dev and libgcc-dev from gcc-cross as they
already provided by gcc.
Commit 8c42351519ece9ed60659a103e624d2e154b3641 add dev packages
(libstdc++-dev libgcc-dev) to gcc-cross (gcc-package-cross.inc)
which are already part of gcc. The libgcc-dev package both include
the libgcc_s.so file which triggers a 'file is already provided by
package' error on incremental builds. The libstdc++-dev package have
Acked-by: Steffen Sledz <sledz@dresearch-fe.de>
different content and is empty on some archs (armv5te, armv7e).
The empty package triggers a 'Cannot satisfy the following
dependencies for task-sdk-bare' error. If an empty libstdc++-dev
package is generated it triggers a '..../usr/include/c++ not exist'
error as this is part of the package from gcc but not gcc-cross.
Signed-off-by: Stefan Herbrechtsmeier <sherbrec@cit-ec.uni-bielefeld.de>
Acked-by: Steffen Sledz <sledz@dresearch-fe.de>
|
|
People are seeing these errrors from ldconfig
libstdc++.so.6.0.14-gdb.py is not an ELF file - it has the wrong magic
bytes at the start.
this file is moved into gdb's autoload directory
if it exists there then gdb will find it when debugging
and it wont be in the paths where ldconfig looks for
libraries.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Tested-by: Philip Balister <philip@balister.org>
|
|
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Acked-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Koen Kooi <koen@openembedded.org>
|
|
When TARGET_PREFIX and TARGET_SYS didn't match up, the symlinks we made
were invalid. TARGET_PREFIX is what we use when configuring so it's what we
should be using here.
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
This is strictly nios2
nios2 specific patches are updated to match the latest version
of the toolchain as made available by Altera
Signed-off-by: Frans Meulenbroeks <fransmeulenbroeks@gmail.com>
|
|
Rename ARM_FP_MODE to ARM_FP_ABI
Set target fpu to hard for hardfp ARMv7
Use machine override to add base package arch instead of appending to
OVERRIDES directly
Pass --with-float option to gcc configure taking into account
what float ABI is selected through ARM_FP_ABI
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
This holds off the patches that cause the ICE at -Os -g
for now. The other workaround it to use -fno-shrink-wrap
but it has to go into all affected recipes
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Currently following fails
rm-oe-linux-uclibceabi-gcc -march=armv7-a -mtune=cortex-a8 -mfpu=neon
-mfloat-abi=softfp -mthumb-interwork -mno-thumb
--sysroot=/oe/build-minimal-uclibc/minimal-uclibc-dev/sysroots/armv7a-oe-linux-uclibceabi
-DHAVE_CONFIG_H -I. -I.. -DASN1_BUILDING -I./../gl -I./../gl -I./..
-fexpensive-optimizations -fomit-frame-pointer -frename-registers -Os
-pipe -g -MT coding.lo -MD -MP -MF .deps/coding.Tpo -c coding.c -o
coding.o
coding.c: In function 'asn1_length_der':
coding.c:101:1: internal compiler error: in
dwarf2out_frame_debug_expr, at dwarf2out.c:2221
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
{standard input}: Assembler messages:
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
In most cases we had both ../foopkg/foopkg as the first path and
this simply became foopkg (the second path was already correct for
the default strip level which is why the patch applies normally).
In a few cases the second path was '.' which is valid but not as
clear as 'b' (which is what git shows for example) so update
that while we're at it.
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
Upgrade to latest svnrev
Get new linaro patches
Delete patches that are not applied
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Tested-by: Yuri Bushmelev <jay4mail@gmail.com>
|
|
* This enables linaro patch 45177 and a patch on top
to fix an ICE discovered because of this backport
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Fix for http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43999 and is
required to build gcc for Thumb2 only architectures (e.g. Cortex-M3).
Signed-off-by: Enrico Scholz <enrico.scholz@sigma-chemnitz.de>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
This fixes a QA issue in gcc 4.1.2.
The patch is identical to what is done in gcc 4.2.2 and 4.2.4
The issues caused neek compilations to fail
Signed-off-by: Frans Meulenbroeks <fransmeulenbroeks@gmail.com>
Acked-by: Tom Rini <tom_rini@mentor.com> (on irc)
|
|
gcc-4.5-linaro-r99474.patch is disabled because it ends up
bombing x86 cross-gcc build see
http://lists.linaro.org/pipermail/linaro-dev/2011-February/002614.html
gcc-4.5-linaro-r99474.patch is disabled because it causes GCC crash
for XScale see
http://lists.linaro.org/pipermail/linaro-dev/2011-February/002600.html
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Acked-by: Martin Jansa <Martin.Jansa@gmail.com>
|
|
This reverts commit f76026a6c1a00bd0b2f1ae2b045dee995293edc6.
Pushed prematurely
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Please test this patch out in your respective combinations and report
and regressions you see.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Note that this is dealt with differently in the sstate world, so this should
*not* go into the meta-oe/oe-core universe.
Signed-off-by: Chris Larson <chris_larson@mentor.com>
|
|
* After moving toolchains under native sysroot
gcc-cross-kernels stopped working becaue it could
not find the remaining gcc programs correctly.
${STAGING_DIR_NATIVE}${prefix_native}/bin/ is not
same as STAGING_BINDIR for cross recipes
* Fixed Bugzilla/5468
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Bump INC_PR for everyone that I didn't just change.
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
Signed-off-by: Tom Rini <tom_rini@mentor.com>
|
|
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
* Fixes the problem as described in
http://lists.linaro.org/pipermail/linaro-dev/2011-January/001993.html
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Acked-by: Martin Jansa <Martin.Jansa@gmail.com>
|
|
It provokes conflicts later on, so drop it since it serves no real use.
Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
|
|
* Fixes missing leading whitespace character at some locations
Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
CC: Khem Raj <raj.khem@gmail.com>
Acked-by: Khem Raj <raj.khem@gmail.com>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
gcc 4.2.4. I see the same problem as is commented on in the 4.2.4 file.
|
|
This could be fixed in three ways
1. apply http://gcc.gnu.org/ml/gcc-patches/2010-11/msg02245.html to gcc
2. Backout linaro/gcc-4.5-r99425.patch from gcc
3. Apply the workaround to QT 4.7.1 from
http://bazaar.launchpad.net/%7Eubuntu-branches/ubuntu/natty/qt4-x11/natty/revision/136
I chose the first one.
Signed-off-by: Khem Raj <kraj@leto.(none)>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
* port the patch for PR 46883 which fixed samba build issue
* Drop PR 42172 reverting patch as its not required anymore
* Refresh patches to apply cleanly on 4.5.2
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
* This fixes the gcc ICE as seen compiling samba
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Acked-by: Eric Bénard <eric@eukrea.com>
|
|
In some cases the rename_srcdir ended up moving the sourcedir
as a sub-directory under gcc-4.5 which was not intended actually
it is intended to be renamed and not moved. Which was seen recently
where people complained about a patch not being applied to gcc-4.5
even though patch was not changed. The reason was the srcdir was
moved instead of renaming. This could happen if target dir already
existed.
Anyway so we get rid of this dance completely. So nip the evil
in the bud
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
script stub
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Acked-by: Eric Bénard <eric@eukrea.com>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
* This is temporary till a proper fix lands
Signed-off-by: Koen Kooi <koen@openembedded.org>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Signed-off-by: Koen Kooi <koen@openembedded.org>
|
|
Currently libgcc_s.so points to libgcc_s.so.1 in the same
<target>/usr/lib dir but this library is in /lib and target toolchain
fails. We package libgcc_so into gcc package and create the proper
symlink as it used to be
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
This does not work for distros like micro who use
a flattened tree structure for root file system
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
with-libdir-list for SH4"
This fixed building the toolchain though but broke
kernel as it expect gcc to be able to compile no-fpu
code. Which means it will break all gcc 4.x
So we revert it and keep the status quo of 4.5 being broken
but older 4.x gcc can still build for SH based machines.
This reverts commit 35881fcc6feb9670947056ff868e03b0a0900a51.
|
|
target_includedir
This fixes the problem where gcc fails to detect presence of libssp
and does not detect powerpc long double 128 bit support correctly on systems which do
not use common <sysroot>/usr/include paths like we have in micro distro
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Acked-by: Tom Rini <tom_rini@mentor.com>
|
|
Also bump to latest SVN REV
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|