Install startfiles for libc variants into the most specific combination
(suffixed sysroot, if applicable + suffixed multi-os dir, if
applicable). Install headers once in every suffixed sysroot (although it
seems that GCC picks up headers from top-level sysroot, GCC manual
claims that sysroot suffix affects headers search path).
In uClibc, this requires a better sanitization of the directory: it
creates symlinks from {sysroot}/usr/lib/{multi_os_dir} to
{sysroot}/lib/{multi_os_dir} and to do so, it counts the number of path
components in the libdir. This breaks if one of such components is `..'
- symlinks contain an extra `../..' then. Since such sanitization had to
be implemented anyway, use it in other places to print more sensible
directory names.
Also, fix the description of configure --host/--target per musl's
configure help message (and its actual code).
Signed-off-by: Alexey Neyman <stilor@att.net>
On some arches (e.g. MIPS) the options like -mabi do not work if
specified more than once (see the comment in 100-gcc.sh). Therefore,
we need to determine which of the options produced by <arch>.sh can
be passed to multilib builds and which must be removed (i.e., which
options vary among the multilibs).
This presents a chicken-and-egg problem. GCC developers, in their
infinite wisdom, do not allow arbitrary multilib specification to be
supplied to GCC's configure. Instead, the target (and sometimes some
extra options) determine the set of multilibs - which may include
different CPUs, different ABIs, different endianness, different FPUs,
different floating-point ABIs, ... That is, we don't know which parts
vary until we build GCC and ask it.
So, the solution implemented here is:
- For multilib builds, start with empty CT_ARCH_TARGET_CFLAGS/LDFLAGS.
- For multilib builds, require core pass 1. Pass 1 does not build any
target binaries, so at that point, our target options have not been
used yet.
- Provide an API to modify the environment variables for the steps that
follow the current one.
- As a part of multilib-related housekeeping, determine the variable
part of multilibs and filter out these options; pass the rest into
CT_TARGET_CFLAGS/LDFLAGS.
This still does not handle extra dependencies between GCC options (like
-ma implying -mcpu=X -mtune=Y, etc.) but I feel that would complicate
matters too much. Let's leave this until there's a compelling case for
it.
Also, query GCC's sysroot suffix for targets that use it (SuperH,
for example) - the default multilib may not work if the command line
specifies the default option explicitly (%sysroot_suffix_spec is not
aware of multilib defaults).
Signed-off-by: Alexey Neyman <stilor@att.net>
For 4 different folders:
${CT_PREFIX_DIR}
${CT_SYSROOT_DIR}
${CT_SYSROOT_DIR}/usr
${CT_PREFIX_DIR}/${CT_TARGET}
.. symlinks from 'lib32' and 'lib64' to 'lib' were created.
This was untidy and incorrect for multilib (the bitness of
the libraries in 'lib32' and 'lib64' will not be the same)
We can not know which folders this toolchain configuration
will require at this time so let them be created on-demand
instead.
Changed by Alexey Neyman: original change removed too much; we
still need to create the default directories because the os
directories are based off them (e.g. `lib/../lib64').
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
Signed-off-by: Alexey Neyman <stilor@att.net>
Written by Bryan Hundven.
Modified by Alexey Neyman to actually add the option to gcc.in.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
Signed-off-by: Alexey Neyman <stilor@att.net>
By default, it is 'auto' - which means, it is enabled if there are
multilibs directories detected in the installation location for libgcc.
Thus, it is not detected for pass-1 GCC: the installation location is
empty at this point.
Signed-off-by: Alexey Neyman <stilor@att.net>
Pass CFLAGS_FOR_TARGET, CXXFLAGS_FOR_TARGET and LDFLAGS_FOR_TARGET to
gcc configure in do_gcc_core_backend as they may be used to build
libstdc++ for bare-metal target.
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
As crosstools-ng only support GCC >= 4.8 we do not need libelf for gcc. GCC dropped this dependency with 4.6.
Signed-off-by: Matthias Weisser <m.weisser.m@gmail.com>
In a5057713a0
...I forgot to add a line continuation at the break in the 'if'
statement.
Reported-by: asavah <asavah@avh.od.ua>
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
I was noticing that $extra_user_env was inconsistently used in
100-gcc.sh. I don't feel comfortable having just any make flag or
environment variable passed to make from a config file. If a specific
option needs to be passed to make for gcc, then a specific kconfig
option should be added for that make flag/option/env.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
The two gcc backend functions are getting very close to being duplicated
code. To help in the process of merging the two backends, this change
syncronizes the two functions so they are easier to diff.
This commit has no functional changes.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
This commit updates the build scripts to match the new usage of
CT_GetCustom from the previous change.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
Instead of checking if thread support is enabled during the build, move
the check to kconfig-time. Since if threading support is not availble,
libgomp should not be available either.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
Because >= gcc-5.x does not require cloog, it should not be forced on
the command line arguments for configure if graphite is enabled.
Make CLooG optionally added, if it is needed (aka: <= gcc-4.9).
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
Now that versions of gcc that required PPL are no longer supported
( >= gcc-4.5.x AND <= gcc-4.7.x )
...we no longer require PPL or CLooG/PPL.
This commit:
* Removes PPL
* Removes CLooG/PPL
* Updates the documentation
* Updates build script for CLooG and GCC
* Removes PPL and CLooG/PPL from scripts/addToolVersion.sh and
scripts/showSamples.sh
* Adds ISL to scripts/addToolVersion.sh and scripts/showSamples.sh
I know that sounds like a lot for one commit, but it was all kind of
inter-tangled.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
As per #222, in crosstool-NG >= 1.23.0, we will only support:
[upstream supported gcc versions] - 1
As of this writing, these versions are:
* 5.2.0
* 4.9.3
* 4.8.5 (the -1, since development on 4.8.x is now closed)
I plan to keep 4.8.5 around because of some architectures having issues
with over-optimization or just faulty optimization in the 4.9.x and
possibly newer versions.
I also cleaned up a requirement for glibc to depend on >= gcc-4.6.x for
>= glibc-2.20, but since the lowest gcc we support after this change is
>= 4.8.5, this condition can go away.
Patches for older gcc versions are removed in the next commit.
This closes#222
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
We check for apps:
* make
* sed
* grep
* awk
* libtool/libtoolize
* install
* patch
* and more
...during configure. Our scripts should be consistent about using the
variables that define where the found tool was found.
Of course, we do hard-link these tools in buildtools, but that should be
a backup for the components we are building. Our scripts should always
use the tools we find.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
The Xtensa processor architecture is a configurable, extensible,
and synthesizable 32-bit RISC processor core. Processor and SOC vendors
can select from various processor options and even create customized
instructions in addition to a base ISA to tailor the processor for
a particular application.
Because of the configurability, the build process requires one additional
step for gcc, binutils, and gdb to update the default configuration.
These configurations are packed into an 'overlay' tar image, and are
simply untarred on top of the default configuration during the build.
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Having *.la in the installation directory breaks ltrace: in ltrace,
libtool somehow considers libsupc++ to be an "accessory library" and
does not add -lsupc++ to the link flags. Neither Ubuntu, nor RedHat
include *.la files into their packages for libstdc++.
Signed-off-by: Alexey Neyman <stilor@att.net>
- New configurations:
- CC_GCC_TARGET_FINAL:
Use the default targets "all" and "install" for the final compiler for
bare metal.
- Adding parameter "build_step" to function do_gcc_core_backend:
do_gcc_core_backend is used for the core compiler and in case of bare metal
for the final compiler, too. To have better control over the parameters for
the final compiler "build_step" is used.
- Used for proper logging.
- Use CT_CC_GCC_CORE_EXTRA_CONFIG_ARRAY or CT_CC_GCC_EXTRA_CONFIG_ARRAY.
- If CT_CC_GCC_TARGET_FINAL is set and the final compiler is build then the
make targets for the final compiler are used ("all", "install").
Signed-off-by: Jasmin Jessich <jasmin@anw.at>
CT_CC_GCC_HAS_LIBQUADMATH and CT_CC_GCC_LIBQUADMATH (--en/disable-libssp,
--en/disable-libquadmath, --en/disable-libquadmath-support) from function
do_gcc_backend.
Signed-off-by: Jasmin Jessich <jasmin@anw.at>
Configure for for core GCC did not use Core gcc extra config.
Use now config variable CT_CC_GCC_CORE_EXTRA_CONFIG_ARRAY.
Signed-off-by: Jasmin Jessich <jasmin@anw.at>
This option is old. GCC 4.3.x old. It isn't supported anymore, and just
confuses me. I'm not planning to support 4.3.x, or really anything older
then 4.7. So this option is gone to the wind.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Update 100-gcc.sh to use the new config option names.
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
Reviewed-by: Bryan Hundven <bryanhundven@gmail.com>
Reviewed-by: Yann Diorcet <diorcetyann@gmail.com>
This commit moves gcc.sh to 100-gcc.sh to accomodate for other
cross-compilers that crosstool-ng might be able to build.
The first, to come soon, is llvm/clang.
Signed-off-by: Ray Donnelly <mingw.android@gmail.com>
Reviewed-by: Bryan Hundven <bryanhundven@gmail.com>
Reviewed-by: Yann Diorcet <diorcetyann@gmail.com>
Without this canadion cross builds create invalid symlinks:
When the code in do_cc_core_backend is called there is no
${CT_TARGET}-gcc in the install directory. Therefore ext is empty and
we create a link to ${CT_TARGET}-gcc. The final compiler
step then installs ${CT_TARGET}-gcc.exe and creates a working
${CT_TARGET}-cc.exe symlink but we still keep the invalid link
as well.
Signed-off-by: Johannes Pfau <johannespfau@gmail.com>
Prirotize http downloads before ftp downloads.
By having http download first, those using proxy will work with the
current download mechnism.
This tells me that that mechnism needs to be updated.
(proxy support and/or kconfig toggles)
closes#3
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
Canadian Cross compile for baremetal fails with error;
checking for the value of EOF... configure: error: computing EOF failed
which is due to libstdc++ configure not being able to find stdio.h
Having all modes of the core compiler copyheaders from CT_HEADERS_DIR
(in combination with previous patch for newlib to add a do_libc_start_files
function to copy into the CT_HEADERS_DIR) resolves this.
Signed-off-by: David Holsgrove <david.holsgrove@xilinx.com>
Yes, I missed the backslash which messed up the linaro stuff.
The more I look at this code, I feel it needs to be refactored a bit. So
I'll come back to this in the future and clean it up.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
It's not my day.
linaro_version is a filter. If it is not a linaro toolchain, it will
just be CT_{CC,GDB}_VERSION. If it is a linaro toolchain, CT_{CC,GDB}_VERSION
will be prefixed with 'linaro-' and will not match linaro_version, as
linaro_version will just have the part after 'linaro-'.
This *really* fixes the issue :sigh:
Thanks again to @elsonwei for being right the first time!
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
linaro_version and linaro_series are defined but not set if we are not
configured for linaro builds.
Therefore we need to default them to "" (null string).
As reported by @elsonwei
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
This change updates the download locations to default to the official
download site.
For gcc and gdb, also separate out the linaro download locations so that
if you are downloading the linaro variant, it skips trying to download
from the official gcc mirror.
This commit closes#3
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
Reported-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
libsaniotizer requires a few headers that are not in uClibc, for
example. Also, it is only available for native threads (NPTL under
glibc.) Finally, it is only available starting with gcc-4.8.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
The previous patch (cset b61a1b1, cc/gcc: avoid passing --enable-multilib)
only fixed the core backend, and missed the final backend.
This patch does the same as b61a1b1, but for the final backend.
Signed-off-by: Cody P Schafer <dev@codyps.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
The final bare-metal compiler is built using the core backend.
Currently the core uses the CC_CORE_EXTRA_CONFIG_ARRAY variable.
While this works as supposed to, this can leave the user puzzled
in the menuconfig, since all he can see is the core options, not
the final options.
Only show the core options if any of the core passes are needed,
and use the final options in the core-backend if we're issuing
the bare-metal compiler.
Signed-off-by: Cody P Schafer <dev@codyps.com>
[yann.morin.1998@free.fr: hide core options if no core pass needed;
use final option in core backend if issuing the bare-metal compiler]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Message-Id: <22181e546ba746202489.1399688067@localhost>
Patchwork-Id: 347586
Some versions of gcc have a broken --enable-multilib flag. As multilib is the
default, only pass the --disable-multilib flag
Signed-off-by: Cody P Schafer <dev@codyps.com>
[yann.morin.1998@free.fr: make it an if-block; duplicate commit log as comment]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Message-Id: <5c970c1ceb22528fe28a.1399687923@localhost>
Patchwork-Id: 347585
We now know exactly what pass to build, so build only what is required.
Reported-by: Trevor Woerner <trevor.woerner@linaro.org>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Decimal floats need support form the C library, namely support
for fenv, which is missing in uClibc for any architecture but
x86/32.
Add an option (a choice) to enable or disable decimal floats.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Since gcc 4.8 C++ is also used as implementation language (see gcc
release notes).
Signed-off-by: "Daniel Dittmann" <ddittmann@gmx.net>
Message-Id: <acc7d11bc77b30f21c5b.1388863298@bernalk.machteam>
Patchwork-Id: 306883
