Description in uClibc for this optoin is:
--->---
bool "Use faster (but larger) generic string functions"
default y
help
Answer Y to use the (tweaked) glibc generic string functions.
In general, they are faster (but 3-5K larger) than the base
uClibc string functions which are optimized solely for size.
Many people will answer Y.
--->---
Additional 3-5K of size is not that important if resulting libuClibc.so is
of size about 300k.
But benefits in terms of performance could be pretty significant.
For example on ARC in LMbench we see more than 3 times bump in some tests
--->---
libc bcopy unaligned
...
8.39 - 6.76 (UCLIBC_HAS_STRING_GENERIC_OPT=no)
8.39 - 23.86 (UCLIBC_HAS_STRING_GENERIC_OPT=yes)
--->---
Also since the option in question is default in uClibc I don't expect if
enabled back (it was silently disabled in http://git.buildroot.net/buildroot/commit/?id=cfbf8abc33d86a0cf5c1bb3e0817a22009b7f301 on introduction of NPTL in uClibc
by Khem Raj) it to introduce problems.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Anton Kolesov <anton.kolesov@synopsys.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Remove the support for generating mips1/2/3/4 code since it has been
deprecated for more than a year now.
Also remove the unnecessary kludges in packages for it.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Acked-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since commit 33de740170 ("Makefile:
convert "target-generatelocales" to a hook"), the GENERATE_LOCALE
variable is never defined when using uClibc. This means that setting
any value to the BR2_GENERATE_LOCALE option had no effect for uClibc
toolchains.
In this commit, we make sure the uclibc.mk logic re-creates its own
qstripped version of BR2_GENERATE_LOCALE, in a variable called
UCLIBC_GENERATE_LOCALES.
This fixes locale generation with uClibc in the internal toolchain
backend.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Synopsys has recently announced its new ARC HS38 core that is capable of
running Linux -
http://www.synopsys.com/dw/ipdir.php?ds=arc-hs38-processor
ARC HS38 is based on ARCv2 ISA and requires special settings of gcc and
libc.
Also in case of HS38 atomic extensions (LLOCK/SCOND instructions) are
built-in by default, so enabling atomic extensions in Buildroot as well.
This commit adds support of the core in buildroot.
[Peter: string type, so must be in quotes as noted by Yann]
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Anton Kolesov <anton.kolesov@synopsys.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As with other architectures for ARC we need to have an ability to set
specific options in uClibc.
In particular this is required for selection of ARC ISA version.
[Peter: string type, so must be in quotes as noted by Yann]
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Anton Kolesov <akolesov@synopsys.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
It fixes the following build failure for packages that expect long
double support in libm:
Linking C executable winpr-hash
../../libwinpr/utils/libwinpr-utils.so.0.1.0: undefined reference to `powl'
../../libwinpr/utils/libwinpr-utils.so.0.1.0: undefined reference to `fmodl'
../../libwinpr/utils/libwinpr-utils.so.0.1.0: undefined reference to `ceill'
../../libwinpr/utils/libwinpr-utils.so.0.1.0: undefined reference to `log10l'
../../libwinpr/utils/libwinpr-utils.so.0.1.0: undefined reference to `floorl'
Fixes:
http://autobuild.buildroot.net/results/323/3236ddfe8eaf89f05f84db60eb42583cca397464/
Mainline status: patch submitted.
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Backporting an upstream patch to fix a problem in the install step:
make[1]: Entering directory `/br/output/build/uclibc-0.9.31.1'
make[2]: Nothing to be done for `locale_headers'.
CC utils/getconf
STRIP -x -R .note -R .comment ../utils/getconf
CC utils/iconv
../lib/libc.a(iconv.os):(.rodata+0x18): multiple definition of
`__iconv_codesets'
/tmp/ccVmV8Lq.o:(.rodata+0x18): first defined here
collect2: error: ld returned 1 exit status
Upstream commit:
http://git.uclibc.org/uClibc/commit/?id=d46dc8bc88e38251bfa3712efe7abf62933f5419
Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Now with change of BR2_PREFER_STATIC_LIB meaning to "do not build dynamic libs
and build statically linked applications" it's possible to disable support of
shared libs in uClibc as well as builting of shared libc libs.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Cc: Peter Korsgaard <peter@korsgaard.com>
Reviewed-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
uClibc can be built stripped (implemented as link-time option in uClibc).
This could be useful for those who build root fs manually.
In Buildroot there is a global strip stage wich strips most of shared libs
in "target" folder (wexcept kernel modules known to not work properly being
stripped and libpthread required to be non-stripped for correct debugging with
gdb of multi-threaded apps).
So there're few problems with current implementation:
1. uClibc is being stripped 2 times (first on its build stage, second on
global Buildroot strip stage)
2. uClibc libs in "staging" folder are also always stripped except if
"no strip" (BR2_STRIP_none) is explicitly is selected in Buildroot config.
That makes it possible to remote debug uClibc libs on target only if target
rootfs was not stripped (which might not be possible due to huge libs
like Qt)
This patch disables embedded strip in uClibc (still users may modify uClibc
config and explicitly set "DOSTRIP=yes" if really needed).
Interesting that DOSTRIP was not only selected in uClibc config but also was
force set with Buildroot on uClibc configuration step with UCLIBC_STRIP_CONFIG.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Waldemar Brodkorb <wbx@openadk.org>
Reviewed-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
To be consistent with the recent change of FOO_MAKE_OPT into FOO_MAKE_OPTS,
make the same change for FOO_KCONFIG_OPT.
Sed command used:
find * -type f | xargs sed -i 's#_KCONFIG_OPT\>#&S#g'
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
After switching to a two stage gcc solution, there is no longer a need
to do weird things in the uclibc build. We can greatly simplify
UCLIBC_CONFIGURE_CMDS to only do the configuration, and let the
existing UCLIBC_BUILD_CMDS do the build. Note that we have to build
the headers before starting the C library build, otherwise there is a
build failure (probably a uClibc bug).
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Currently, the internal toolchain backend does a three stage gcc
build, with the following sequence of builds:
- build gcc-initial
- configure libc, install headers and start files
- build gcc-intermediate
- build libc
- build gcc-final
However, it turns out that this is not necessary, and only a two stage
gcc build is needed. At some point, it was believed that a three stage
gcc build was needed for NPTL based toolchains with old gcc versions,
but even a gcc 4.4 build with a NPTL toolchain works fine.
So, this commit switches the internal toolchain backend to use a two
stage gcc build: just gcc-initial and gcc-final. It does so by:
* Removing the custom dependency of all C libraries build step to
host-gcc-intermediate. Now the C library packages simply have to
depend on host-gcc-initial as a normal dependency (which they
already do), and that's it.
* Build and install both gcc *and* libgcc in
host-gcc-initial. Previously, only gcc was built and installed in
host-gcc-initial. libgcc was only done in host-gcc-intermediate,
but now we need libgcc to build the C library.
* Pass appropriate environment variables to get SSP (Stack Smashing
Protection) to work properly:
- Tell the compiler that the libc will provide the SSP support, by
passing gcc_cv_libc_provides_ssp=yes. In Buildroot, we have
chosen to use the SSP support from the C library instead of the
SSP support from the compiler (this is not changed by this patch
series, it was already the case).
- Tell glibc to *not* build its own programs with SSP support. The
issue is that if glibc detects that the compiler supports
-fstack-protector, then glibc uses it to build a few things with
SSP. However, at this point, the support is not complete (we
only have host-gcc-initial, and the C library is not completely
built). So, we pass libc_cv_ssp=no to tell the C library to not
use SSP support itself. Note that this is not a big loss: only a
few parts of the C library were built with -fstack-protector,
not the entire library.
* A special change is needed for ARC, because its libgcc depends on
the C library, which breaks building libgcc in
host-gcc-initial. This looks like a bug in the ARC compiler, as it
does not obey the inhibit_libc variable which tells the compiler
build process to *not* enable things that depend on the C
library. So for now, in host-gcc-initial, we simply disable the
build of libgmon.a for ARC. It's going to be built as part of
host-gcc-final, so the final compiler will have gmon support.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Now when new shiny tools are released by Synopsys we're ready for version
update in Buildroot.
Important change in this release is switching to combined "binutils-gdb" repo
in accordance to upstream move.
Following patch now is a part of the most recent relese:
e6ab8cac62
So dropping it.
package/binutils/arc-4.8-R3/0001-arc-Honor-DESTDIR-in-custom-Makefile.patch
Since arc-2014.08 tools are still based on GCC 4.8 following patch is still
relevant so moving to the new folder to matxh ARC gcc bump.
package/gcc/arc-4.8-R3/100-libstdcxx-uclibc-c99.patch ->
package/gcc/arc-2014.08/100-libstdcxx-uclibc-c99.patch
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Anton Kolesov <akolesov@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The configure step of the busybox package performs 'make oldconfig', which
causes the .config file to be updated. Thus, the .config file is more recent
than our stamp file .stamp_kconfig_fixup_done. On a subsequent build, our
dependency rules would kick in, and run the config fixup again, thus
kicking in the package's configure, build and install steps yet once
more, that, ad infinitum.
One solution is to modify kconfig-package to introduce an explicit touch of
the .kconfig_fixup_config_done stamp file, as post-configure hook.
Another solution, implemented by this patch, is to move the oldconfig call
from the package's .mk file to the kconfig-package infrastructure and make
sure it is done as part of the fixup commands. This way, the stamp file will
only be touched once, after the full fixup (including oldconfig) and no
endless rebuilds will occur.
Reported-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Suggested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This patch converts the uclibc package to the new kconfig-package
infrastructure, thus removing code duplication and ensuring a consistent
behavior of kconfig packages.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The uclibc-menuconfig command is currently run based on the user-specified
config file, while the fixup of the config file is done afterwards.
However, it makes more sense to do an initial fixup _before_ running
menuconfig, so that the options presented to the user are consistent with
the final configuration.
This patch changes the dependency of the menuconfig target from .config to
.stamp_config_fixup_done to achieve this.
Suggested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
There is no real reason to run uclibc-menuconfig in non-parallel mode,
even though one can neither expect performance benefits from a parallel
menuconfig.
Nevertheless, $(MAKE) is the default, so this patch removes the unnecessary
non-default $(MAKE1) usage for uclibc-menuconfig.
This is a simplification introduced in preparation of the kconfig-package
infrastructure.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In the sequence:
make uclibc-menuconfig
make uclibc-update-config
the freshly configured settings from the menuconfig are lost during the
update-config step. This is because update-config depends on the configure
step, which starts by copying the config file to the build directory.
Instead, stop depending on the configure step from update-config, and
introduce a new stamp file .stamp_config_fixup_done, which applies any
fixups on the .config file.
An alternative solution would be to add a call to UCLIBC_FIXUP_DOT_CONFIG to
the relevant targets instead of depending on a new stamp file. The advantage
of the stamp file, though, is that we avoid redoing the fixup unnecessarily.
Moreover, in the light of the plan to extract the kconfig-specific bits into
a separate kconfig-package infrastructure, the stamp file rules are more
easily moved into such an infrastructure, while the alternative solution
requires the package .mk file to explicitly call the FIXUP rules which may
more easily be forgotten.
No longer depending on the configure step has the added bonus that
'uclibc-update-config' no longer needs the toolchain to be available, which
makes:
make clean uclibc-menuconfig uclibc-update-config
much faster and user-friendly.
Additionally, make sure that 'make clean uclibc-update-config' works
properly, by depending on .stamp_config_fixup_done so that the config file
is present and fixed.
Fixes bug #7154https://bugs.busybox.net/show_bug.cgi?id=7154
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When executing the sequence 'make clean uclibc-menuconfig', the configured
config file is not taken into account and one starts from the default
settings.
This patch adds an explicit target for the config file and lets the
configure and menuconfig steps depend on it, fixing the problem.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In a subsequent patch, the behavior of UCLIBC_SETUP_DOT_CONFIG will change
so that it only applies fixups to the (already copied) configuration file.
This patch renames this function to UCLIBC_FIXUP_DOT_CONFIG to better match
the future behavior.
Suggested-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
If libuClibc is not installed to staging on attempt to build tests following
errors happen:
============
TEST_LINK termios/ termios
.../host/usr/arc-buildroot-linux-uclibc/sysroot/usr/lib/crt1.o: In function `__start':
.../build/uclibc/libc/sysdeps/linux/arc/crt1.S:53: undefined reference to `__uClibc_main'
termios.o: In function `main':
.../build/uclibc/test/termios/termios.c:13: undefined reference to `printf'
.../build/uclibc/test/termios/termios.c:14: undefined reference to `printf'
.../build/uclibc/test/termios/termios.c:16: undefined reference to `stdout'
.../build/uclibc/test/termios/termios.c:16: undefined reference to `fileno'
.../build/uclibc/test/termios/termios.c:16: undefined reference to `ioctl'
.../build/uclibc/test/termios/termios.c:19: undefined reference to `perror'
.../build/uclibc/test/termios/termios.c:21: undefined reference to `printf'
.../host/usr/lib/gcc/arc-buildroot-linux-uclibc/4.8.0/../../../../arc-buildroot-linux-uclibc/lib/libgcc_s.so.1: undefined reference to `memcpy'
.../host/usr/lib/gcc/arc-buildroot-linux-uclibc/4.8.0/../../../../arc-buildroot-linux-uclibc/lib/libgcc_s.so.1: undefined reference to `malloc'
.../host/usr/lib/gcc/arc-buildroot-linux-uclibc/4.8.0/../../../../arc-buildroot-linux-uclibc/lib/libgcc_s.so.1: undefined reference to `abort'
.../host/usr/lib/gcc/arc-buildroot-linux-uclibc/4.8.0/../../../../arc-buildroot-linux-uclibc/lib/libgcc_s.so.1: undefined reference to `calloc'
.../host/usr/lib/gcc/arc-buildroot-linux-uclibc/4.8.0/../../../../arc-buildroot-linux-uclibc/lib/libgcc_s.so.1: undefined reference to `dl_iterate_phdr'
.../host/usr/lib/gcc/arc-buildroot-linux-uclibc/4.8.0/../../../../arc-buildroot-linux-uclibc/lib/libgcc_s.so.1: undefined reference to `realloc'
.../host/usr/lib/gcc/arc-buildroot-linux-uclibc/4.8.0/../../../../arc-buildroot-linux-uclibc/lib/libgcc_s.so.1: undefined reference to `memset'
.../host/usr/lib/gcc/arc-buildroot-linux-uclibc/4.8.0/../../../../arc-buildroot-linux-uclibc/lib/libgcc_s.so.1: undefined reference to `strlen'
.../host/usr/lib/gcc/arc-buildroot-linux-uclibc/4.8.0/../../../../arc-buildroot-linux-uclibc/lib/libgcc_s.so.1: undefined reference to `free'
collect2: error: ld returned 1 exit status
make[3]: *** [termios] Error 1
============
And the simplest solution is to build tests right before installation in
UCLIBC_INSTALL_TARGET_CMDS which follows UCLIBC_INSTALL_STAGING_CMDS so
all required libs are in place.
Interesting note - if one enables uClibc tests after initial full buildroot
build mentioned errors don't happen. This is because uClibc libc libs are
already installed to staging. That's why I didn't noticed this problem before.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Anton Kolesov <akolesov@synopsys.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
While updating the default config for Sparc, I noticed that networking
is broken. The reason is a uClibc backport patch from uClibc
master. This commit adds a fix on top of it, which was submitted to
upstream uClibc.
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
So far, the legal-info infrastructure was not exporting the legal
info for the C libraries. This is because the legal-info only acts
on packages defined in $(TARGETS).
But the C libraries are never added to $(TARGETS), since there is no
corresponding BR2_PACKAGE_<C-LIBRARY>.
This patch adds such symbols for the 4 C libraries we support in our
internal backend: uClibc, glibc, eglibc and musl.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This commit bumps make to version 4.0. This version now uses fork(),
so it is no longer available on non-MMU platforms, which consequently
works around:
http://autobuild.buildroot.org/results/2e1/2e1b11ccb501f1ab431c04435be52a6538d035cb/
The glob special options seem to no longer be needed, at least make
4.0 builds fine with both an ARM uClibc toolchain and an ARM glibc
toolchain.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Don't try to install ld* tools to staging when it's a static build,
there's no point and it'll be broken once we build a purely-static
uclibc.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This patch removes the custom kconfig helpers UCLIBC_OPT_SET and
UCLIBC_OPT_UNSET with the common KCONFIG_SET_OPT, KCONFIG_ENABLE_OPT and
KCONFIG_DISABLE_OPT.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The BR2_UCLIBC_ARM_TYPE was only used for uClibc 0.9.32 on ARM. Now
that only uClibc 0.9.33 is supported for ARM, there is no point in
keeping BR2_UCLIBC_ARM_TYPE, since the corresponding uClibc options no
longer exist.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This commit removes the 0.9.32.1 version of uClibc, which is very old,
and does not bring any specific advantage over 0.9.33, which has been
around for more than two years now.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This is only required for uClibc 0.9.31 and this veriosn is only used by
AVR32 so could be safely removed together with AVR32.
Even 0.9.32 builds well in parallel (at least on x86) so IMHO there's no need
in >= 0.9.33, just disable for a particular broken version.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Anton Kolesov <akolesov@synopsys.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The use of MAKE1 for uClibc dates back 10 years:
commit 8e5fb3fb4a
Author: Eric Andersen <andersen@codepoet.org>
Date: Sat Dec 11 13:01:10 2004 +0000
Add initial BR2_JLEVEL support, with some exceptions for apps that
have broken 'make -j' support
Since that time there were lots of improvements in uClibc that seem to allow
parallel building of libs finally.
Unfortunately uClibc tests have dependences on previously built files,
that's why tests left with MAKE1.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Anton Kolesov <akolesov@synopsys.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently we configure uClibc to use kernel headers from "staging" folder with
KERNEL_HEADERS="$(STAGING_DIR)/usr/include". This path is added to include
search path of uClibc build system in Rules.mak "CFLAGS += -I$(KERNEL_HEADERS)".
At the same time on uClibc installation to "staging" we point to the same
location "$(STAGING_DIR)/usr" (headers effectively go in "usr/include").
So after every installation to "staging" dependences get touched (even though we
copy the same headers every time) and so we may see lots of sources in uClibc
get rebuilt.
This has 2 consequences:
1. Longer build time - becase even on ordinary buildroot build uClibc is built
twice. On "uclibc building" and on "uclibc installation to target".
2. Symbols in libuClibc built initially (that is later installed in
"staging/sysroot") are situated with different offset compared to second build
(later copied in "target"). This happens because as described above only part
of sources get rebuilt and then on final linkage object files are linked in
different order.
And (2) leads to problems on remote rebugging: gdbserver reports offsets that
correspond to pointless assembly in libuClibc on host.
Here's how it looks like.
Before this patch:
$ cd ~/br2_output/i586/target/lib
$ i586-buildroot-linux-uclibc-readelf -s libuClibc-0.9.33.2.so | grep kill
423: 0000c42c 54 FUNC GLOBAL DEFAULT 7 kill
$ cd ~/br2_output/i586/staging/lib
$ i586-buildroot-linux-uclibc-readelf -s libuClibc-0.9.33.2.so | grep kill
423: 0000b518 54 FUNC GLOBAL DEFAULT 7 kill
After this patch:
$ cd ~/br2_output/i586/target/lib
$ i586-buildroot-linux-uclibc-readelf -s libuClibc-0.9.33.2.so | grep kill
423: 0000b518 54 FUNC GLOBAL DEFAULT 7 kill
$ cd ~/br2_output/i586/staging/lib
$ i586-buildroot-linux-uclibc-readelf -s libuClibc-0.9.33.2.so | grep kill
423: 0000b518 54 FUNC GLOBAL DEFAULT 7 kill
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Noam Camus <noamc@ezchip.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Add two uClibc 0.9.33.2 patches which fix issues relating to high signal
number handling on MIPS.
The first patch (0056) fixes _NSIG (and as a result __SIGRTMAX) to match
glibc. This fixes GDB on MIPS with uClibc, which cannot handle
__SIGRTMAX == 128 and emits the error:
GDB bug: target.c (gdb_signal_from_host): unrecognized real-time signal
This patch is from uClibc commit 2da958760f79 (MIPS: set _NSIG to
128, not 129. This matches glibc.) and applies without conflicts.
The second patch (0057) fixes the wait status macros to correctly
interpret status 0x007f on MIPS (other arches don't have signal 127).
This patch is from uClibc commit 4a96b9486871 (bits/waitstatus.h:
correctly interpret status 0x007f on MIPS) and applies without
conflicts.
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Add uClibc 0.9.33.2 patch to fix arch specific eventfd definitions,
particularly EFD_NONBLOCK. The definition in sys/eventfd.h was 04000
(0x800), however MIPS uses 0200 (0x80). This resulted in QEMU built for
MIPS hanging at various points until input is received due to a blocking
poll of stdin.
The patch is a backport of uClibc commit fd355bc1dbcb (eventfd.h: Use
new "bits/" scheme for arch-specific flags). The only conflict was minor
in Makefile.in.
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The default uClibc version for xtensa is broken when LARGEFILE is disabled.
Add a patch to fix this.
Reported-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since commit c31d5bfdec (uclibc: arc: bump to arc-4.8-R3 release) uClibc
version f37101d94284 is no longer used for ARC. Remove a version specific
patch.
Cc: Anton Kolesov <Anton.Kolesov@synopsys.com>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The toolchain currently doesn't build for nommu ARM and is in need of
serious work.
Problem is there are no emulation targets and real ARM(7TDMI/720T/740T)
hardware that's capable of running linux (enough memory, having a
memory controller...) is VERY rare and uses very old versions to
make it usable.
The ARM nommu focus should go into Cortex M series processors that are
obtainable at reasonable cost on modern hardware that has external
memory controllers.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
gcc support was added in version 4.6:
http://gcc.gnu.org/gcc-4.6/changes.html
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit a5e92d92e7 converted uClibc to the
package infrastructure, but while doing so removed support for the
documented uclibc-update-config make target. This make target copies the
uclibc .config file back to the original (typically version-controlled)
configuration file.
As this was an unintentional removal, this patch re-adds the
uclibc-update-config target.
Reported-by: Dan Moulding <dan.moulding@rackwareinc.com>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This will resolve uClibc build error for configurations without large file
support.
Signed-off-by: Anton Kolesov <Anton.Kolesov@synopsys.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Alexey Brodkin
Gustavo Zacarias
Thomas Petazzoni
Jerzy Grzegorek
Max Filippov
Vicente Olivert Riera
Alexey Brodkin
Thomas De Schampheleire
Waldemar Brodkorb
Yann E. MORIN
James Hogan
Baruch Siach
Bernd Kuhls
Anton Kolesov