This commit finally bumps ARC tools to the most recent arc-2018.03
release version.
ARC GNU tools of version arc-2018.03 bring some quite significant
changes like:
* Binutils v2.29.51 with additional ARC patches
* GCC 7.3.1 with additional ARC patches
* GDB 8.0.50 with ARC patches
More information on this release could be found here:
https://github.com/foss-for-synopsys-dwc-arc-processors/toolchain/releases/tag/arc-2018.03-release
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: arc-buildroot@synopsys.com
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Fixed issues are listed in the 2.27 branch NEWS file:
CVE-2017-18269: An SSE2-based memmove implementation for the i386
architecture could corrupt memory. Reported by Max Horn.
CVE-2018-11236: Very long pathname arguments to realpath function could
result in an integer overflow and buffer overflow. Reported by Alexey
Izbyshev.
CVE-2018-11237: The mempcpy implementation for the Intel Xeon Phi
architecture could write beyond the target buffer, resulting in a buffer
overflow. Reported by Andreas Schwab.
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This commit bumps ARC toolchain to arc-2018.03-rc2, which
includes significant changes since arc-2018.03-rc1.
We want to test how new toolchain-rc2 builds packages,
so we can make fixes before release of toolcain.
This makes us closer to toolchain release which will be in a few weeks.
Please note that it is a release candidate
and it might contain some breakages,
please don't use it for production builds.
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: arc-buildroot@synopsys.com
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 7aaabe8fc0)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This commit bumps ARC toolchain to arc-2018.03-rc2, which
includes significant changes since arc-2018.03-rc1.
We want to test how new toolchain-rc2 builds packages,
so we can make fixes before release of toolcain.
This makes us closer to toolchain release which will be in a few weeks.
Please note that it is a release candidate
and it might contain some breakages,
please don't use it for production builds.
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: arc-buildroot@synopsys.com
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
libnsl from Glibc is deprecated and should not be used anymore.
Remove libnsl.so.* from GLIBC_LIBS_LIB.
libnsl is now an separate library that can be packaged later if
necessary [1].
Note: libnsl from Glibc doesn't build with gcc 8 due new warning [2].
[1] https://github.com/thkukuk/libnsl.git
[2] http://patchwork.sourceware.org/patch/26437
This reverts commit 398747f5fa.
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Cc: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
This commit bumps ARC toolchain to arc-2018.03-rc1.
We want to test how new toolchain-rc1 builds packages,
so we can make fixes before release of toolcain.
ARC GNU tools of version arc-2018.03-rc1 bring some quite significant changes like:
* Binutils v2.29.51 with additional ARC patches
* GCC 7.3.1 with additional ARC patches
Please note that it is a release candidate
and it might contain some breakages,
please don't use it for production builds.
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Alexey Brodkin <abrodkin@synopsys.com>
Cc: arc-buildroot@synopsys.com
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Since upstream commit 1faaf7035cabda101e1d6653bff7a539f201db91
("plural.c: improve reproducibility"), glibc now requires bison to be
available on the host for its build process. This is needed starting
with glibc 2.27.
Fixes:
http://autobuild.buildroot.net/results/ca4d883793c1674d3a052edd5e56897f79683448/
Signed-off-by: Ryan Coe <bluemrp9@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
GLIBC_SRC_SUBDIR was needed when Buildroot supported
eglibc which stored all sources in a sub-directory.
It was not removed by the commit removing eglibc support [1].
[1] 500de2598a
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
List of fixes from the 2.26 branch NEWS files:
CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
suffered from a one-byte overflow during ~ operator processing (either
on the stack or the heap, depending on the length of the user name).
Reported by Tim Rühsen.
CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
would sometimes fail to free memory allocated during ~ operator
processing, leading to a memory leak and, potentially, to a denial
of service.
CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
without GLOB_NOESCAPE, could write past the end of a buffer while
unescaping user names. Reported by Tim Rühsen.
CVE-2017-17426: The malloc function, when called with an object size near
the value SIZE_MAX, would return a pointer to a buffer which is too small,
instead of NULL. This was a regression introduced with the new malloc
thread cache in glibc 2.26. Reported by Iain Buclaw.
Cc: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Currently, we do a full git clone of the repository, which takes quite
some time, especially on slow networks.
This was done like that because the initial patch was using the official
repository as the source of the download, and that repository did not
offer remotely-generated tarballs.
But now we've switched to using a mirror on github, which does provide
such a tarball, which provides faster downloads.
Use that.
However, the tarball from github differs from the one we were generating
locally, because the paths inside are different. WE used to create a
archive with paths starting with glibc-glibc-2.26-73-g4b692dfb95[...],
while github does away with the git-describe prefix, and generates paths
that start with just glibc-4b692dffb95[...]. The content are exactly
identicall (checked with a diff), though.
Update the hash accordingly.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Waldemar Brodkorb <wbx@openadk.org>
Cc: Romain Naour <romain.naour@gmail.com>
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This commit finally bumps ARC tools to the most recent arc-2017.09 release version.
ARC GNU tools of version arc-2017.09 bring some quite significant changes like:
* Binutils v2.29 with additional ARC patches
* GCC 7.1.1 with additional ARC patches
* glibc 2.26 with additional ARC patches
More information on this release could be found here:
https://github.com/foss-for-synopsys-dwc-arc-processors/toolchain/releases/tag/arc-2017.09-release
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
Cc: Alexey Brodkin <abrodkin@synopsys.com>
Cc: arc-buildroot@synopsys.com
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
glibc upstream has ruled against doing regular point-releases, but they
do have a lot of interesting and important fixes for regressions and
security.
Backporting each patch, or cherry-picking individual patches is off
limits for us, so we just switch to using the currently-latest HEAD of
the maintenance branch instead.
The version number is obtained with:
$ git describe --match 'glibc-*' --abbrev=40 origin/release/2.26/master
The alternative options were:
- download the tarball from the git tree
--> does not work; not an option
- download the 2.26 tarball, and bundle the individual patches in
Buildroot
--> maintenance of patches is a burden; not an option
- download the 2.26 tarball, maintain the list of patches to download from
the git tree
--> not an option for the same reason
So we end up just doing a git clone. The git tree is today about ten
times the size of the tarball, so a rough estimate makes it at about ten
times the download time.
Also upstream doesn't officially provide an https download location [1].
There is one but it's not reliable, sometimes the connection time out and
end-up with a corrupted git repo:
fatal: unable to access 'https://sourceware.org/git/glibc.git/': Failed to connect to sourceware.org port 443: Connection timed out
So switch to using a git mirror from github which is updated once a day [2].
This allow at the same time to clone the git repository faster.
Note: The glibc 2.26 patches are not kept for the arc toolchain since they
are fixing an issue with the new float128 support introduced in x86, x86_64
and powerpc64le.
[1] https://sourceware.org/git/?p=glibc.git;a=summary
[2] https://github.com/bminor/glibc.git
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Romain Naour <romain.naour@openwide.fr>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Evgeniy Didin <didin@synopsys.com>
CC: Alexey Brodkin <abrodkin@synopsys.com>
[Romain: bump 4b692dffb95ac4812b161eb6a16113d7e824982e]
Signed-off-by: Romain Naour <romain.naour@gmail.com>
[yann.morin.1998@free.fr: update comment to never decide on the mirror]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Finally there's working ARC port of glibc thanks to Vineet and Cuper!
This port is based on pretty recent glibc's master branch and ARC
changes are being reviewed now in glibc's mailing list.
Thus we again have to use sources from our GitHub but as soon as there's
a glibc release with our patches applied we'll switch to upstream releases
and will drop our glibc GitHub repo alltogether.
Note now we cut tags in glibc repo simultaneously with tags
in Binutils and GCC repos and so to make sure everything works in the best
way we plan to update glibc tag together with Binutils and GCC.
Also note as of today ARCompact (AKA ARCv1 ISA) is not supported in glibc
but we plan to fix it soonish so for now we make glibc intentionally
dependent on archs38.
Also note we are not creating directory "2.26" because all patches for glibc
ver 2.26 applies to arc glibc port.
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
CC: Alexey Brodkin <abrodkin@synopsys.com>
CC: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
CC: Waldemar Brodkorb <wbx@openadk.org>
CC: Romain Naour <romain.naour@gmail.com>
Cc: Cupertino Miranda <cmiranda@synopsys.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Anton Kolesov <akolesov@synopsys.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Fixes:
/tmp/output/build/glibc-2.25/build/libc_pic.a(dl-error.os): In function `__GI__dl_signal_error':
dl-error.c:(.text+0x0): multiple definition of `_dl_signal_error'
/tmp/output/build/glibc-2.25/build/elf/dl-allobjs.os:(.text+0x134d8): first defined here
/tmp/output/build/glibc-2.25/build/libc_pic.a(dl-error.os): In function `__GI__dl_catch_error':
dl-error.c:(.text+0x190): multiple definition of `_dl_catch_error'
/tmp/output/build/glibc-2.25/build/elf/dl-allobjs.os:(.text+0x136fc): first defined here
/tmp/output/build/glibc-2.25/build/libc_pic.a(init-first.os):(.data+0x0): multiple definition of `__libc_multiple_libcs'
/tmp/output/build/glibc-2.25/build/elf/dl-allobjs.os:(.bss+0x6c): first defined here
/tmp/output/build/glibc-2.25/build/libc_pic.a(_itoa.os): In function `_itoa':
_itoa.c:(.text+0xd4): multiple definition of `_itoa'
/tmp/output/build/glibc-2.25/build/elf/dl-allobjs.os:(.text+0x133c8): first defined here
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
This fixes the following build issues:
In file included from ../sysdeps/x86_64/multiarch/varshift.c:19:0,
from ../sysdeps/i386/i686/multiarch/varshift.c:1:
../sysdeps/x86_64/multiarch/varshift.h: In function '__m128i_shift_right':
../sysdeps/x86_64/multiarch/varshift.h:26:1: error: SSE vector return without SSE enabled changes the ABI [-Werror=psabi]
{
[...]
.../lib/gcc/i686-buildroot-linux-gnu/7.1.0/include/tmmintrin.h:136:1: error: inlining failed in call to always_inline '_mm_shuffle_epi8': target specific option mismatch
The patch has been taken from glibc's patchwork at
https://patchwork.sourceware.org/patch/21003/.
Fixes bug #10156.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Fixes the following build failure, which occurs on all architectures:
/tmp/cc6APNmN.s: Assembler messages:
/tmp/cc6APNmN.s: Error: `loc1@GLIBC_2.17' can't be versioned to common symbol 'loc1'
/tmp/cc6APNmN.s: Error: `loc2@GLIBC_2.17' can't be versioned to common symbol 'loc2'
/tmp/cc6APNmN.s: Error: `locs@GLIBC_2.17' can't be versioned to common symbol 'locs'
../o-iterator.mk:9: recipe for target '/tmp/output/build/glibc-2.25/build/misc/regexp.os' failed
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
glibc contains a vulnerability that allows specially crafted LD_LIBRARY_PATH
values to manipulate the heap/stack, causing them to alias, potentially
resulting in arbitrary code execution. Please note that additional
hardening changes have been made to glibc to prevent manipulation of stack
and heap memory but these issues are not directly exploitable, as such they
have not been given a CVE.
https://www.qualys.com/2017/06/19/stack-clash/stack-clash.txt
Patches are identical to upstream, except that the ChangeLog modifications
have been stripped.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
We do not support uClibc-ng/musl C library version choice support,
do the same for GNU C Library.
No legacy handling required as only version choice is removed.
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
[Thomas: move 3.2 kernel headers dependency to the libc choice in
toolchain/toolchain-buildroot/Config.in file, and added a Config.in
comment about it.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The input to copy_toolchain_lib_root is not one library, not a list of
libraries, but a library name pattern with glob wildcards.
This pattern is then passed to 'find' to get the actual list of libraries
matching the pattern. Reflect this using an appropriate variable name.
Note: if the root of the buildroot tree contains a file matching one of
these library patterns, the copying of libraries from staging to target will
not be correct. It is not impossible to fix that, e.g. using 'set -f', but
maybe it's not worth it.
Signed-off-by: Thomas De Schampheleire <thomas.de_schampheleire@nokia.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We want to use SPDX identifier for license string as much as possible.
SPDX short identifier for BSD-3c is BSD-3-Clause.
This change is done using following command.
find . -name "*.mk" | xargs sed -ri '/LICENSE( )?[\+:]?=/s/BSD-3c/BSD-3-Clause/g'
Signed-off-by: Rahul Bedarkar <rahulbedarkar89@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We want to use SPDX identifier for license string as much as possible.
SPDX short identifier for LGPLv2.1/LGPLv2.1+ is LGPL-2.1/LGPL-2.1+.
This change is done using following command.
find . -name "*.mk" | xargs sed -ri '/LICENSE( )?[\+:]?=/s/LGPLv2.1(\+)?/LGPL-2.1\1/g'
Signed-off-by: Rahul Bedarkar <rahulbedarkar89@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We want to use SPDX identifier for license strings as much as possible.
SPDX short identifier for GPLv2/GPLv2+ is GPL-2.0/GPL-2.0+.
This change is done by using following command.
find . -name "*.mk" | xargs sed -ri '/LICENSE( )?[\+:]?=/s/\<GPLv2\>/GPL-2.0/g'
Signed-off-by: Rahul Bedarkar <rahulbedarkar89@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We don't want a dozen glibc versions and there's no particular reason to
keep this old version around so drop it.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
[Thomas: add entry to Config.in.legacy.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
libanl.so is needed for asynchronous network address and service
translation, declared in netdb.h
Signed-off-by: Jesper Bækdahl <jbb@gamblify.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
On some architectures (namely x86-64), glibc may provide a libmvec
library since glibc 2.22, which programs built with gcc OpenMP support
might get linked to.
In order for these programs to work on the target, we need to copy
this library to the target filesystem.
This commit takes care of this for the glibc package (used for the
internal toolchain backend). Note that libraries listed in
GLIBC_LIBS_LIB are silently ignored if they don't exist. Therefore, we
don't need to have any condition on the architecture or glibc version.
For more details on libmvec, see
https://sourceware.org/glibc/wiki/libmvec.
Fixes bug #9111.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Backport two upstream patches to fix a build failure on MIPS and SPARC
architectures when using recent binutils versions. It appears that on
these architectures pt-vfork.S includes vfork.S (through the alpha
version of pt-vfork.S) and that the __vfork aliases are not
conditionalized on IS_IN (libc) like on other architectures. Therefore
the aliases are also wrongly included in libpthread.so.
The build failure looks like this:
/br/output/build/glibc-2.22/build/nptl/libpthread.so:(*IND*+0x0):
multiple definition of `vfork@GLIBC_2.0'
/br/output/build/glibc-2.22/build/nptl/libpthread.so::(.text+0xeea0):
first defined here
This problem was reported in Debian here:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=827477
Upstream patches here:
https://sourceware.org/git/?p=glibc.git;a=commit;h=43c2948756bb6e144c7b871e827bba37d61ad3a3https://sourceware.org/git/?p=glibc.git;a=commit;h=b87c1ec3fa398646f042a68f0ce0f7d09c1348c7
I have tweaked them a little bit in order to not patch the ChangeLog
file, otherwise it will fail.
Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The BR2_TOOLCHAIN_BUILDROOT_GLIBC option is already not selectable on
BR2_sparc, so there is no need to have an additional "depends on
!BR2_sparc" on each of the glibc versions.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Following the removal of eglibc support, this commit replaces all
occurences of "(e)glibc" by just "glibc". Most of the occurences are in
package Config.in comments.
In addition, when the form "an (e)glibc ..." was used, it is replaced by
"a glibc ...".
[Peter: add new efi* packages, s/uclibc/uClibc as suggested by Romain,
systemd / liquid-dsp tweaks as suggested by Yann]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
In commit 500de2598a, support for eglibc
was removed, but forgot to remove the corresponding hashes from the hash
file. This commit fixes this mistake.
Reported-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Don't enable SSP support on external toolchains just because they use
glibc or musl. Instead of that, make the external toolchains explictily
declare if they support SSP or not. And also add a check to detect SSP
support when using custom external toolchains.
For internal toolchains we always enable SSP support for glibc and musl.
Fixes:
http://autobuild.buildroot.net/results/ac7c9b3ad2e52abfe6b79a80045e4218eeb87175/
Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
[Thomas:
- remove uClibc-specific SSP check, since there is now a generic
check being done.
- send potential compilation errors caused by the SSP check to
oblivion, in order to avoid causing confusion for the user.
- add autobuilder reference.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The eglibc support has been marked deprecated since 2015.08, so it's
time to remove it.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
glibc-2.22 does not build with gcc 6.x. The first issues can be fixed
by:
- Using the same 0004-gcc6.patch as we're using with glibc 2.23 since
Buildroot commit ab8de336eb.
- Backport glibc commit 5542236837c5c41435f8282ec92799f480c36f18 to fix
strchr() build failure.
But then, there are some more failures caused by the fact that numerous
glibc files were not using modern prototypes in the function
definitions, causing build failures such as:
../sysdeps/unix/sysv/linux/dl-openat64.c:26:1: error: 'openat64' defined as variadic function without prototype [-Werror]
openat64 (dfd, file, oflag)
^~~~~~~~
In file included from ../include/fcntl.h:2:0,
from ../sysdeps/unix/sysv/linux/dl-openat64.c:21:
../io/fcntl.h:214:12: note: previous declaration of 'openat64' was here
extern int openat64 (int __fd, const char *__file, int __oflag, ...)
^~~~~~~~
This could be fixed by backporting glibc commit
9dd346ff431fc761f1b748bd4da8bb59f7652094, but this is a large commit,
touching a significant number of files. We probably don't want to take
this path, and instead making glibc 2.22 as not available with gcc 6.x
seems like the right course of action.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The toolchain-external logic is roughly:
- populate the staging dir by rsyncing the entire ${ARCH_LIB_DIR} and
usr/${ARCH_LIB_DIR} from sysroot.
- populate the target dir by explictly copying some libraries from sysroot
into target/lib and some other libraries in target/usr/lib, the split
being hardcoded into buildroot regardless of the location in the sysroot.
This means that a library libfoo could be located in:
staging/lib/libfoo.so
target/usr/lib/libfoo.so
When debugging an application that links against this library, gdb will
fruitlessly search for 'usr/lib/libfoo.so' in staging, and then suggest to
use 'set solib-search-path' which is a hack, really.
To solve the problem, we need to make sure that libraries from the toolchain
are installed in the same relative location in staging and target.
Achieve this by:
- replacing the convoluted search for libraries using for+find in sysroot
with a simple find in staging.
- determining DESTDIR for each library individually based on the location in
staging.
- treating LIB_EXTERNAL_LIBS and USR_LIB_EXTERNAL_LIBS equivalently
These changes also allow for the removal of most arguments to
copy_toolchain_lib_root in the method itself and their callers.
Test procedure:
- set configuration for a given toolchain
- make clean toolchain
- find output/target | sort > /tmp/out-before
- apply patch
- make clean toolchain
- find output/target | sort > /tmp/out-after
- diff -u /tmp/out-before /tmp/out-after
The only changes should be some libraries moving from lib to usr/lib or vice
versa. Notable examples being libstdc++ and libatomic.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
[Thomas:
- use -L instead of -follow in the find invocation, as suggested by
Arnout.
- move the BR2_STATIC_LIBS condition as a make condition rather than
a shell condition, as suggested by Arnout.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
