scripts/functions: extract: portable call for old and defective tars

Instead of using -J, --lzma, --use-compress-program or the like
use <compressor> -dc <file> | tar -f -

Signed-off-by: Titus von Boxberg <titus@v9g.de>

debug/gdb: add extra config for gdb cross.

Signed-off-by: Zhenqiang Chen <zhenqiang.chen@linaro.org>
[yann.morin.1998@anciens.enib.fr: prompt rewording, as suggested by M. Hope]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

libc/newlib: add option to optimise for size

Add an option to build with -Os instead of the default -O.

Reviewed-by: Michael Hope
Signed-off-by: Zhenqiang Chen <zhenqiang.chen@linaro.org>

cc/gcc: Apply CT_CC_GCC_DISABLE_PCH to do_cc_core.

Otherwise, users have to input --disable-libstdcxx-pch option
when building bare-metal CANADIAN C++ compiler.

Reviewed-by: Michael Hope
Signed-off-by: Zhenqiang Chen <zhenqiang.chen@linaro.org>

libc/newlib: Add extra config for newlib.

So users can input config like --enable-newlib-register-fini.

Reviewed-by: Michael Hope
Signed-off-by: Zhenqiang Chen <zhenqiang.chen@linaro.org>

cc/gcc: handle NLS option

Add --disable-nls config when option "Enable nls" is not selected.

Reviewed-by: Michael Hope
Signed-off-by: Zhenqiang Chen <zhenqiang.chen@linaro.org>

binutils/binutils: handle NLS option

Add --disable-nls config when option "Enable nls" is not selected.

Reviewed-by: Michael Hope
Signed-off-by: Zhenqiang Chen <zhenqiang.chen@linaro.org>

config/toolchain: Add "Enable nls" in "Toolchain options".

Reviewed-by: Michael Hope
Signed-off-by: Zhenqiang Chen <zhenqiang.chen@linaro.org>

libc/uClibc: use endian string in tests

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

libc/eglibc: use endian string in tests

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

arch/sh: use endian string in tests

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

arch/mips: use endian string in tests

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

scripts/functions: use endian string in tests

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

config: introduce endianness as a string

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

1.13: update version to 1.13.2+hg

Tagging release 1.13.2

1.13: update version to 1.13.2

arch/x86: prescott is an i686

let CT_ARCH_ARCH=prescott result in a i686 tuple prefix

Signed-off-by: Titus von Boxberg <titus@v9g.de>
(transplanted from d063b1ae8c1b30c7006b110d0f9045e6d7c34f13)

complibs/cloog: auto-reconf for 0.15.10 and later

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
(transplanted from 47199f966983bab6a68b2b5da2e1542adbd9388a)

complibs/cloog: fix extraction

In the early days, cloog-ppl was bizarrely packaged: the first tarball
did not contain the version in the name of the extracted directory, so
we had to play tricks.

Nowadays, however, the first component of the path are stripped when
extracting a tarball, which means that the created directory will
always be properly named. So, our old tricks do no longer work, and
worse, they break the build.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
(transplanted from f070d922facf42b0395b7db909e1867d45b245dd)

scripts: add support for building manuals

Add support for building the HTML and PDF manuals for the major
components. Implement for binutils, GCC, GDB, and GLIBC.

Always build all manuals and install a subset. Be explicit about the
subset to reduce the clutter and to avoid getting copies of common
manuals like bfd from all of the sourceware based components. Downside of
being explicit is that you need to update it when a new component
comes along.

Build the manuals as part of the last GCC build, namely 'cc' for glibc
based ones and cc_core_pass_2 for baremetal.

An example of the output is at:
http://people.linaro.org/~michaelh/incoming/crosstool-NG/

Signed-off-by: Michael Hope <michael.hope@linaro.org>
[yann.morin.1998@anciens.enib.fr: depends on ! remove docs; gold manual install]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

docs: Add strategies for assembling root filesystems.

Expand the documentation for using a crosstool-NG-generated toolchain for
building a root filesystem for a target device.

Signed-off-by: "Trevor Woerner" <twoerner@gmail.com>
yann.morin.1998@anciens.enib.fr: some eye-candy]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

scripts: use wget, not curl

It seems wget is more popular than curl.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

scripts: use the floating-point string option in tests

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

libc/uClibc: use the floating-point string option in tests

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

arch/sh: use the floating-point string option in tests

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

arch/x86: prescott is an i686

let CT_ARCH_ARCH=prescott result in a i686 tuple prefix

Signed-off-by: Titus von Boxberg <titus@v9g.de>

complibs/cloog: auto-reconf for 0.15.10 and later

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

complibs/cloog: fix extraction

In the early days, cloog-ppl was bizarrely packaged: the first tarball
did not contain the version in the name of the extracted directory, so
we had to play tricks.

Nowadays, however, the first component of the path are stripped when
extracting a tarball, which means that the created directory will
always be properly named. So, our old tricks do no longer work, and
worse, they break the build.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

complibs/mpfr: update to 3.1.0

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

debug/gdb: fix tic install path, tell ncurses where to find it

ncurses 5.9 wants tic to be either one of:
- $TIC_PATH
- /usr/bin/tic

Of course, se do not want the latter, for it can be incompatible if the
ncurses in the build system is too old (eg. RHEL 5.6, Debian Lenny...).
So, force TIC_PATH to the location of our own tic.

Also, install tic alongside the other build tools, not in a sub-dir
of the toolchain installation dir.

Signed-off-by: Willy Tarreau <w@1wt.eu>
[yann.morin.1998@anciens.enib.fr: install in builtools/bin, move TIC_PATH]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
(transplanted from bd17cca444ae09af0aae3e2390fa068147e2f4e3)

configure: do not use '@' in version string

Using an '@' in the version string breaks the gcc/glibc/gdb manuals.
Use a scheme similar to Debian, with a plus '+' instead of the '@',
as suggested by Michael Hope.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

docs: --target versus --host

In the very beginnings, eons ago, autotools also got confused by this
whole build vs. host vs. target, and got it wrong. Now they fixed it,
but they want to keep backward compatibility, so the --target is still
recongised, although ./configure will complain if you do so. It is
better to use --host.

Signed-off-by: "Trevor Woerner" <twoerner@gmail.com>
[yann.morin.1998@anciens.enib.fr: add build/host clarification]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

libc/glibc: add 2.14.1

Signed-off-by: Willy Tarreau <w@1wt.eu>

debug/gdb: fix tic install path, tell ncurses where to find it

ncurses 5.9 wants tic to be either one of:
- $TIC_PATH
- /usr/bin/tic

Of course, se do not want the latter, for it can be incompatible if the
ncurses in the build system is too old (eg. RHEL 5.6, Debian Lenny...).
So, force TIC_PATH to the location of our own tic.

Also, install tic alongside the other build tools, not in a sub-dir
of the toolchain installation dir.

Signed-off-by: Willy Tarreau <w@1wt.eu>
[yann.morin.1998@anciens.enib.fr: install in builtools/bin, move TIC_PATH]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

cc/gcc: update Linaro GCC revisions to 2011.11

Update Linaro GCC with the latest available revisions.

Signed-off-by: "Benoît Thébaudeau" <benoit.thebaudeau@advansee.com>

configure: fix version string breakage

As we use the directory names as Makefile targets, they can not contain any
column ':', or else make will complain about 'multiple target patterns'.

Replace the offending ':' with a dash '-', as Titus suggested.

Reported-by: Erdem Budak <erdembudak@hotmail.com>
Reported-by: Titus von Boxberg <titus@v9g.de>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

1.13: update version to 1.13.1+hg

Tagging release 1.13.1

1.13: update version to 1.13.1

configure: compute a simpler version string

Compute a simpler, yet still completely descriptive, version string.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

configure: add support for helper script to compute version string

Some projects are using (or planning to use) crosstool-NG, and are storing
it in their VCS, which might not be Mercurial. At the same time, those
projects may want to track development snapshots versions the way we do
with the Hg identity string (hg id).

Provide a way for these project to do so, without having to patch
./configure, and maintain that patch over-and-over again.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

docs: remove dedication file

This does in no way mean that Cédric will be forgotten.
RIP, Cédric...

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

kernel/linux: missed porting the 3.0.8 patchset to 3.0.9

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

kernel/linux: update to latest versions

Add latest versions.
Remove versions that disapeared during the kernel.org debacle.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

arch/arm: ARM supports the softfp convention

ARM compilers can be built for soft float (software only, floats in
core registers), hard float (uses floating point instructions, floats
in FPU registers), or the half-way house softfp (uses floating point
instructions, floats in core registers).

Signed-off-by: Michael Hope <michael.hope@linaro.org>
[yann.morin.1998@anciens.enib.fr: split the original patch]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

arch: add softfp support

Some architectures support a mixed hard/soft floating point, where
the compiler emits hardware floating point instructions, but passes
the operands in core (aka integer) registers.

For example, ARM supports this mode (to come in the next changeset).

Add support for softfp cross compilers to the GCC and GLIBC
configuration. Needed for Ubuntu and other distros that are softfp.

Signed-off-by: Michael Hope <michael.hope@linaro.org>
[yann.morin.1998@anciens.enib.fr: split the original patch]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

scripts: use the hardfloat option to set configure and CFLAGS

When hardfloat is selected, we need to pass that selection down to
./configure and in the CFLAGS.

Signed-off-by: Michael Hope <michael.hope@linaro.org>
[yann.morin.1998@anciens.enib.fr: split the original patch]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

scripts: introduce float type as a string

With the upcoming softfp support, the case..esac test would become
a bit convoluted if it were to test three different booleans.

Introduce a new blind string config option that defaults to the
selected floating point type used.

Signed-off-by: Michael Hope <michael.hope@linaro.org>
[yann.morin.1998@anciens.enib.fr: split the original patch]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

kernel/linux: fix packed attribute in exported headers

UBI headers (and maybe others as well) are broken because the 'packed'
attribute is not sanitised when the headers are exported to userspace.

Apply the fix from upstream:
https://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commitdiff;h=f210735fe2f17a6225432ee3d1239bcf23a8659c

(Also, buildroot does the same, BTW)

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

kernel/linux: fix packed attribute in exported headers

UBI headers (and maybe others as well) are broken because the 'packed'
attribute is not sanitised when the headers are exported to userspace.

Apply the fix from upstream:
https://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commitdiff;h=f210735fe2f17a6225432ee3d1239bcf23a8659c

(Also, buildroot does the same, BTW)

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

kernel/linux: update revisions

Update Linux with the latest available revisions.

Signed-off-by: "Benoît THÉBAUDEAU" <benoit.thebaudeau@advansee.com>

debug/strace: Add strace-4.6

Now that xz support is in crosstool-ng, Add strace-4.6.

Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>

cc/gcc: Add gcc-4.6.2

Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>

1.13: update version to 1.13.0+hg

Tagging release 1.13.0

1.13: create maintenance branch, update version to 1.13.0

1.12: close branch

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

scripts: fix missing space when using CT_EXTRA_FLAGS_FOR_HOST

CT_EXTRA_FLAGS_FOR_HOST needs a preceding space to separate it from
any other options that have already been set.

Signed-off-by: Michael Hope <michael.hope@linaro.org>

samples: set vendor string in all samples

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>