This fixes multiple issues, including:
- 'operand out of range' errors (due to defaulting to
medium calls instead of short calls)
- library dependencies that were not being found
Signed-off-by: Mischa Jonker <mjonker@synopsys.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Until now, we were using the default ARM instruction set, as used by
the toolchain: the 32 bits ARM instruction set for the internal
backend, and for external toolchain, whatever default was chosen when
the toolchain was generated.
This commit adds support for the Thumb2 instruction set. To do so, it:
* provides a menuconfig choice between ARM and Thumb2. The choice is
only shown when Thumb2 is supported, i.e on ARMv7-A CPUs.
* passes the --with-mode={arm,thumb} option when building gcc in the
internal backend. This tells the compiler which type of
instructions it should generate.
* passes the m{arm,thumb} option in the external toolchain
wrapper. ARM and Thumb2 code can freely be mixed together, so the
fact that the C library has been built either ARM or Thumb2 and
that the rest of the code is built Thumb2 or ARM is not a problem.
[Peter: fix empty BR2_GCC_TARGET_MODE check]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
When using eglibc, the ld-linux loader is installed as
ld-linux-armhf.so.3, but gcc 4.7.x was not yet updated, and generates
binaries that reference ld-linux.so.3. This was fixed starting with
gcc 4.8.x. In order to be able to use gcc 4.7, we create the
appropriate symbolic link.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The ARM EABIhf support was introduced in gcc 4.6.x, so versions
earlier than this one should not be used when EABIhf is selected.
The ARM VFPv4 support was introduced in gcc 4.5.x, so versions earlier
than this one should not be used when VFPv4 is used.
[Peter: drop OABI dependencies]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Buildroot already has the BR2_GCC_TARGET_{TUNE,ARCH,ABI,CPU} hidden
kconfig strings that allow per-architecture Config.in files to feed
the appropriate values of --with-{tune,arch,abi-cpu} when building
gcc, or the appropriate flags for the external toolchain wrapper.
This commit has two additional options:
BR2_GCC_TARGET_{FPU,FLOAT_ABI}, that allows to define the
--with-{fpu,float} gcc configure options for the internal backend, or
the -m{fpu,float-abi} options for the flags of the external toolchain
wrapper.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
When refactoring the internal toolchain backend logic, the code was
changed to pass the custom configure options given through
BR2_EXTRA_GCC_CONFIG_OPTIONS only for the gcc final pass, with the
idea that we're only interested by user customization for the final
compiler.
However, the beaglebone_defconfig was passing --with-float=hard
--with-fpu=vfpv3-d16 as BR2_EXTRA_GCC_CONFIG_OPTIONS, and since the
refactoring, it was causing build failures of the beaglebone_defconfig
(with messages saying that Busybox is built to use VFP arguments, but
libc/libm are not). This is due to the fact that the gcc intermediate,
which is used to build the C library, wasn't built to generate hard
float, while the final compiler was generating hard float.
So, we get back to the original situation where the options in
BR2_EXTRA_GCC_CONFIG_OPTIONS are passed to all of the compiler
passes. Of course, the specific case of hard float will be fixed by
following patches in this area, but the idea still remains: the three
gcc should have the same options, if those options affected the ABI of
the generated code.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Commit 6b48b48034 ("add a know to enable/disable building a shared
libgcc"), from october 2006, isn't really as to why a
BR2_GCC_SHARED_LIBGCC option was needed. However, now that gcc has
been converted to the package infrastructure, it causes problems
because the host packages are always being passed --enable-shared
--disable-static, so re-adding --disable-shared on top of that break
things.
Moreover, our tests indicate that both a shared *and* a static version
of libgcc are built, and that linking dynamically and statically a
program that uses libgcc_s gives correct results: dynamically linked
against libgcc_s in the first case, statically linked in the second
case.
Therefore, it appears that this option is no longer necessary, and
removing it has the advantage of fixing the builds of
qemu_mips64_malta_defconfig and qemu_sparc_ss10_defconfig, both of
which had BR2_GCC_SHARED_LIBGCC not enabled.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Acked-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Using the newly introduced 'eglibc' package, this commit enables the
option of building a toolchain using the eglibc C library in the
Buildroot toolchain backend.
In details, this commit:
* Creates a choice to select uClibc or eglibc in the Buildroot
toolchain backend (in toolchain/toolchain-buildroot/Config.in), and
removes the fact that the Buildroot toolchain backend forcefully
enables uClibc (toolchain/Config.in).
* Creates a BUILDROOT_LIBC variables, which points to the package
implementing the C library (i.e either 'uclibc' or 'eglibc').
* Modifies the gcc-final and gcc-intermediate makefiles to use the
BUILDROOT_LIBC variable instead of hardcoding the use of uclibc.
* Ensures that TLS support is always enabled when building eglibc.
[Peter: fix commit text to refer to BUILDROOT_LIBC]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
[Peter: update manual to match]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Several sub-directories of the gcc code base are in fact not needed
for the Buildroot build: libjava/, libgo/ and gcc/testsuite/ being the
biggest ones. Avoiding their extraction saves quite a bit of disk
space, and compensates a bit the fact that we now extract three times
the gcc source code.
This requires changing the 100-uclibc-conf.patch to no longer patch
files from the libjava/ directory, since this directory is no longer
extracted.
[Peter: add comment about why this is done]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Mischa Jonker
Peter Korsgaard
Thomas Petazzoni