debug/gdb: install dependable libs in a generic target static libs dir

For now, ncurses is the only dependable target library built for gdb.
But expat is coming, and there's no reason to install each library in
its own place.

So, install ncurses in a generic directory, where other dependable
libraries can be installed as well.

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

complibs: noone is using companion libs on the target; nuke them

As there's no longer any user of the companion libraries on the
target, nuke the build for the target.

Well, at least, there's libelf that's still needed by ltrace, so
we keep it.

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

debug/gdb: companion libs are not used

Although the gdb ./configure advertises for GMP and MPFR, those libraries
are not used by gdb (the ./configure is used across different packages,
hence the check for GMP/MPFR). See:
http://sourceware.org/ml/crossgcc/2010-08/msg00168.html

The same applies to MPC.

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

binutils/binutils: companion libraries are not used

GNU binutils does not use the companion libraries, although
./configure advertises switches for them. This is because the
configure file is shared between different packages, and is
a bit 'lax' about what it checks. See:
http://sourceware.org/ml/crossgcc/2010-08/msg00160.html
http://sourceware.org/ml/crossgcc/2010-08/msg00161.html

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

scripts/log: do not interpret log string as printf format

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

1.8: update version to 1.8.1+hg

Tagging release 1.8.1

1.8: update version to 1.8.1

scripts: in case of failed download, remove partial files

It happens from time to time that the server mis-behaves, and breaks the
connection right in the middle of nowhere, for no good reason, leaving us
with a partial file, on which the extract pass would choke.

Remove partial downloads, to fail early.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
(transplanted from eefacabd9e72f3d578e5dc5908026ed780f9a240)

docs: review REaDME

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

libc/glibc: repair debugging with gdb (glibc 2.9)

Without this patch, crosstool-ng-built glibc-2.9 prevents
debugging any exeutable with gdb.
gdb says:
[Thread debugging using libthread_db enabled]
find_new_threads_callback: cannot get thread info: generic error
See also https://bugzilla.redhat.com/show_bug.cgi?id=487212
for a discussion of the bug and the solution.

internals: call strip in a portable way

strip on CT_HOST darwin does not want --strip-all or -v

binutils/binutils: forward-port the patchset from 2.20 to 2.20.1

Among other things, this should fix the issue reported by Thomas at:
http://sourceware.org/ml/crossgcc/2010-08/msg00115.html

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

libc/mingw: fix space-damage

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

mingw32: enable sysrooted toolchains

When targeting mingw32, gcc expects to find its include files
in "mingw/include" instead of the traditional "usr/include".

[Yann E. MORIN: split the original patch]

libc/mingw: use pre-set headers dir

There is a variable that contains the headers directory,
so there's no need to re-compute it again.

[Yann E. MORIN: split the original patch]

kernel/mingw32: use pre-set headers dir

There is a variable that contains the headers directory,
so there's no need to re-compute it again.

[Yann E. MORIN: split the original patch]

binutils/sstrip: mark as obsolete

sstrip is causing more trouble and grief than tolerable.
It is broken at least on PPC. It does not build on non-ELF
systems (eg. mingw32, MacOS-X...). Plus, it is easy to
install.

Hide it behind OBSOLETE.

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

libc/uClibc: fix downloading pregen locales

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
(transplanted from 223aaf040327e204f0801549e9c0cf6225bdf29d)

libc/uClibc: fix downloading pregen locales

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

scripts: in case of failed download, remove partial files

It happens from time to time that the server mis-behaves, and breaks the
connection right in the middle of nowhere, for no good reason, leaving us
with a partial file, on which the extract pass would choke.

Remove partial downloads, to fail early.

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

scripts: when downloading, use aria2 only when //

Even when // downloads are not enabled, aria2 can
fail on some servers (eg. uclibc.org).

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

scripts: do not download in // by default

Parallel downloads can be a bit harsh on the servers,
and some will fail (eg. uclibc.org) in some cases.

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

docs: move the Mercurial tutorial to the tutorials section

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

docs: split into multiple files

The overview.txt file has evolved into more than just an overview.
Split it into chapters, and include the misc tutorials.

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

scripts: better sanitise the patch names

When generating the fixes-patches, better sanitise the generated
file names, as it seems to raise a few issues for some people.

complibs/cloog: with static ppl, correctly link with libm

On some Fedora boxen (at least FC13), it is also required
to link with libm when static ppl is used.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
(transplanted from 5526fb72a893f89286d6daa29da0161ca988a1d0)

cc/gcc: with static ppl, correctly link with libm

On some Fedora boxen (at least FC13), it is also required
to link with libm when static ppl is used.
(transplanted from bcd33fce4db9c18223b59ecdc96f10bf6dd574de)

complibs/ppl: add latest version 0.11

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

complibs/cloog: with static ppl, correctly link with libm

On some Fedora boxen (at least FC13), it is also required
to link with libm when static ppl is used.

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

cc/gcc: with static ppl, correctly link with libm

On some Fedora boxen (at least FC13), it is also required
to link with libm when static ppl is used.

complibs/mpc: fix MPC 0.8.1 build with MPFR 3.0.0

This is the change introduced by revision 734 of MPC repository.

Author: Paul Zimmermann <Paul.Zimmermann@loria.fr>
Revision log: [acos.c] fixed problem with GMP_RNDA (should be MPFR_RNDA, and code was wrong)

Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>

libc/glibc: let glibc 3.3.6 link correctly

From http://sourceware.org/git/?p=glibc.git;a=commit;f=elf/Makefile;h=7c8a67320e26b8c11108bf0a3410d3aef9cf3486

Author: Ulrich Drepper <drepper@redhat.com>:
Commit log:
* elf/Makefile (ld.so): Adjust the sed script to insert _begin in to
newer linker scripts.

Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>

complibs/ppl: fix PPL build with GMP 5.x.x

From http://www.cs.unipr.it/git/gitweb.cgi?p=ppl/ppl.git;a=commitdiff;h=9c19bc2b318a35016e0189f9552c98910be37f53

Author: Roberto Bagnara <bagnara@cs.unipr.it>
Revision log: Added support for GMP 5.0.

Adapted manually to match and not having to re-generate the configure script.

Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>

1.7: close branch

1.6: close branch

libc/uClibc: add uClibc 0.9.30.3

This version has been released a couple of month ago, but it never reached
crosstool-ng tree. This may be linked to the fact that the current 0.9.30.2,
once patched, has nothing much different from 0.9.30.3, released.

I'm not including any patch with this upgrade, on purpose.

Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>

kernel/linux: : nuke old 2.6.33.[0-6]

Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>

kernel/linux: 2.6.32.17, 2.6.33.7, 2.6.34.2 and 2.6.35 are out !

Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>
[Yann E. MORIN: also add latest 2.6.27.49]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

cc/gcc: 4.5.1 is out!

Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>

scripts: add option to strip all toolchain executables

To reduce filesizes of the toolchain and even improve build times
of projects to be build with this toolchain it is usefull to strip
the delivered toolchain executables. Since it is not likely that we
will debug the toolchain executables itself we do not need the
debug information inside the executables itself.

Signed-off-by: Remy Bohmer <linux@bohmer.net>

1.8: update version to 1.8.0+hg

Tagging release 1.8.0

1.8: create maintenance branch, update version to 1.8.0

docs: update known-issues

samples: update all samples

All samples get rebased, no new option is chosen.

cc/gcc: do not show sjlj choice for bare-metal

In the conversion from choic to tristate, this got lost.

samples: add a short list of all samples

Add an action to list only the sample names.
Use that for auto-completion.

libc: do not offer cross-ldd for bare-metal

Fix CT_SanitizePath

Replace the over-engineered and buggy test in CT_SanitizePath
with a straight forward string pattern match, and also
handle empty PATH elements which are qeuivalent to ".".

Thanks-To: Arnaud Lacombe <lacombar@gmail.com>
Signed-off-by: Johannes Stezenbach <js@sig21.net>

1.7: update version to 1.7.2+hg

Tagging release 1.7.2

1.7: update version to 1.7.2

scripts: remove . from $PATH

Add CT_SanitizePath function which removes entries referring to ., /tmp
and non-existing directories from $PATH, and call it early in the
build script.

If . is in PATH, gcc-4.4.4 build breaks:

[ALL ] checking what assembler to use...
/tmp/build/targets/arm-unknown-linux-uclibcgnueabi/build/gcc-core-static/arm-unknown-linux-uclibcgnueabi/bin/as
...
[ALL ] config.status: creating as

i.e. "as" is supposed to be the arm-unknown-linux-uclibcgnueabi cross assembler,
but config.status creates a local "as" script which is calling the
host assembler.

Signed-off-by: Johannes Stezenbach <js@sig21.net>
[Yann E. MORIN: style fixes + explanations]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
(transplanted from 20dd8cef1c8adff0aa3e78ae6d7acfbc45ed5a83)

complibs: fix using static companion libraries

When building a cross-compiler for a target which uses a file extension for
binaries the symbolic link to cc is not created correctly because the lookup
of the gcc binary is done in a incorrect path

Signed-off-by: Bart vdr. Meulen <bartvdrmeulen@gmail.com>
(transplanted from 3917f2dafed1bb189e39f50e8506b3141926d7e1)

cc/gcc: add option to compile against static libstdc++, for gcc-4.4 and newer

Idea and know-how taken from CodeSourcery build script.

Normal build:
$ ldd arm-unknown-linux-uclibcgnueabi-gcc
linux-gate.so.1 => (0xb77f3000)
libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0xb76e8000)
libc.so.6 => /lib/i686/cmov/libc.so.6 (0xb75a1000)
libm.so.6 => /lib/i686/cmov/libm.so.6 (0xb757a000)
/lib/ld-linux.so.2 (0xb77f4000)
libgcc_s.so.1 => /lib/libgcc_s.so.1 (0xb755c000)

CC_STATIC_LIBSTDCXX=y:
$ ldd arm-unknown-linux-uclibcgnueabi-gcc
linux-gate.so.1 => (0xb7843000)
libc.so.6 => /lib/i686/cmov/libc.so.6 (0xb76e6000)
/lib/ld-linux.so.2 (0xb7844000)

I made CC_STATIC_LIBSTDCXX default=y since I think
it is always desirable.

Signed-off-by: Johannes Stezenbach <js@sig21.net>

scripts: remove . from $PATH

Add CT_SanitizePath function which removes entries referring to ., /tmp
and non-existing directories from $PATH, and call it early in the
build script.

If . is in PATH, gcc-4.4.4 build breaks:

[ALL ] checking what assembler to use...
/tmp/build/targets/arm-unknown-linux-uclibcgnueabi/build/gcc-core-static/arm-unknown-linux-uclibcgnueabi/bin/as
...
[ALL ] config.status: creating as

i.e. "as" is supposed to be the arm-unknown-linux-uclibcgnueabi cross assembler,
but config.status creates a local "as" script which is calling the
host assembler.

Signed-off-by: Johannes Stezenbach <js@sig21.net>
[Yann E. MORIN: style fixes + explanations]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

cc/gcc: add option to enable/disable libmudflap

For some scenarii, libmudflap is not very usefull
or can break the build. Make in an optioon that
defaults to 'N' to be on the safe side.

For the core gcc-s, there is absolutely no need
to build libmidflap.

Idea from: Bernhard Pfund <bernhard@chapter7.ch>

cc/gcc: make sjlj config option a tristate

A tristate fits better here than a choice.

cc/gcc: move options around

config: allow unconditional usage of tristates

libc-uclibc: move options

libc/uClibc: fix space damage

libc/uClibc: fix using pre-generated locale data

Some time, someone updated the locale Makefile to use
newer pre-generated locales data, but did not upload
those.

So we just force using the existing, ageing archive,
dating back 20030818. Sigh...

libc/uClibc: do not systematically use pre-generated locale data

It seems that using pre-generated locale data can be more problematic
than usefull in some circumstances.

Offer a config knob to enable/disable use of the pregen locale data.
Also, do not extract pregen locales data ourselves, it's broken.

configure: fix --mandir

--mandir points to the base dir of the man pages, so
we have to append our man section below given dir.
(transplanted from a8a4d5c9773893fa6978e5f9e1445d1e8185c24a)

scripts: add a cross-ldd-like

Add a cross-ldd that mimicks a native ldd.

libc/uClibc: do not install cross-ldd

I was unable to make the cross-ldd from uClibc to work, and
it is not possible to build it on non-POSIX system.

Besides, we have a generic script that is in the starting-blocks
to replace it, that will work for any C library, and also will
work on non-POSIX systems. Bonus!

docs/known-issues: document the "error compiling FC test program" issue

complibs: fix using static companion libraries

When building a cross-compiler for a target which uses a file extension for
binaries the symbolic link to cc is not created correctly because the lookup
of the gcc binary is done in a incorrect path

Signed-off-by: Bart vdr. Meulen <bartvdrmeulen@gmail.com>

libc/uClibc: fix verbosity level for uClibc >= 0.9.30

uClibc >= 0.9.30 supports three verbosity levels, according to "make help":

V="" - Quiet build (default)
V=1 - Brief build (show defines, ld flags)
V=2 - Very verbose build

I think older versions of uClibc treat V=2 the same as V=1.
For current uClibc, only V=2 shows the full command lines.

Signed-off-by: Johannes Stezenbach <js@sig21.net>

scripts/functions: recognise Cygwin build hosts

scripts: use 'hg archive' to create release tarballs

kernel/mingw32: mark as being EXPERIMENTAL

configure: fix --mandir

--mandir points to the base dir of the man pages, so
we have to append our man section below given dir.

scripts: update config.{guess,sub}

Merge.

complibs/ppl: backout #9f0de857cb05

The patch breaks the build on most machines,
while the purpose of the patch is dubious.
So, simply back it out.

debug/gdb: properly remove build tic

tic is isntalled in buildtools, not in the toolchain's bin dir.
Plus, it may have an extension...

debug/gdb: Fix compilation for Mingw hosts

GDB requires PDcurses instead of ncurses while running on Windows.
So, do not always compile ncurses in case GDB needs to build.

PDcurses is provided by an earlier build step and is not described in
this file.

Signed-off-by: Remy Bohmer <linux@bohmer.net>
[yann.morin.1998@anciense.nib.fr: we already have a way to detect ncurses usage]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

complibs/ppl: fix build for MINGW

Signed-off-by: Bart vdr. Meulen <bartvdrmeulen@gmail.com>
---
patches/ppl/0.10.2/100-update-for-mingw-build.patch | 19 +++++++++++++++++++
1 file changed, 19 insertions(+)
create mode 100644 patches/ppl/0.10.2/100-update-for-mingw-build.patch

samples: add Linux -> Mingw32 sample

Sample to build compiler targetting mingw32 target systems

Signed-off-by: Bart vdr. Meulen <bartvdrmeulen@gmail.com>
---
samples/i686-pc-mingw32/crosstool.config | 302 +++++++++++++++++++++++++++++++
samples/i686-pc-mingw32/reported.by | 3
2 files changed, 305 insertions(+)
create mode 100644 samples/i686-pc-mingw32/crosstool.config
create mode 100644 samples/i686-pc-mingw32/reported.by

kernel/mingw: add target libraries

Add several development libraries to the build of the mingw cross-compiler
to be used on target

Libraries:
PDCurses (port of the ncurses library)
GnuRX (the regex library)
DirectX
OpenGL

Signed-off-by: Bart vdr. Meulen <bartvdrmeulen@gmail.com>
[yann.morin.1998@anciens.enib.fr: don't show DX and RX versions if disabled]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

kernel: add mingw

Add the option to build a cross-compiler for kernel type 'mingw'.
The resulting cross-compiler can be used to build applications on a Linux host
that can be run on a Windows target.

Compiler is build using the mingwrt and w32-api packages aviable from the
MinGW project (http://sourceforge.net/projects/mingw).

The windows headers (w32-api package) are extracting with the kernel_headers
step The libraries and other headers from both packages are build and
installed in the various steps of libc

Signed-off-by: Bart vdr Meulen <bartvdrmeulen@gmail.com>
[yann.morin.1998@anciens.enib.fr: fix kernel headers comment, don't "return 0"]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>

libc: add global threads support option

samples: add an ARMeb uClibc/NPTL experimental sample

cc/gcc: gcc-4.4 does not use MPC

As reported by Johannes Stezenbach, see thread at:
http://sourceware.org/ml/crossgcc/2010-07/msg00017.html

config: no need to push the download mirror in a sub menu

binutils/binutils: add binutils-2.20.1

Signed-off-by: Johannes Stezenbach <js@sig21.net>

kernel/linux: update linux stable releases

Signed-off-by: Johannes Stezenbach <js@sig21.net>

libc/uClibc: fix snapshots

Snapshots are in a subdir named uClibc, not uClibc-snapshot
(or uClibc-YYYYMMDD either).
(transplanted from b13d0db6fa95795bb663bf674372b58221682dd6)

libc/uClibc: enable NPTL for snapshots

Recently, NPTL has matured a lot in uClibc, and more and more
people are interested to at least give it a try. So enable it.

libc/uClibc: fix snapshots

Snapshots are in a subdir named uClibc, not uClibc-snapshot
(or uClibc-YYYYMMDD either).

libc/uClibc: fix munging .config for LT old/new

debug/ltrace: add latest version

scripts/finish: do not try to symlink if sed expr is a no-op

If the sed expression does not transform the
tuple, do not try to create the symlink.
(transplanted from 5d6d3c28a658a65551d4eb5de414fd97b575f314)

scripts/finish: do not try to symlink if sed expr is a no-op

If the sed expression does not transform the
tuple, do not try to create the symlink.

debug/gdb: fix incorrect path

Fix discrepancy between the directory we create,
and the directory into which we install tic.

libc/glibc: fix installed scripts

glibc installs some bash-scripts, but uses the path to the buildtool
bash as interpreter (on the shebang line). This is only a symlink to
the real bash, and thus is not available at runtime.

Fix that by assuming that bash on the target *will* be /bin/bash.

1.7: update version to 1.7.1+hg

Tagging release 1.7.1

1.7: update version to 1.7.1

scripts/wrapper: fix wrong test when checking access to the reall tool

In C, the proper syntax for a bit-wise OR is a single '|', not two.

It worked so far because all was well:
- X_OK == 1
- R_OK||X_OK == 1
- the file we searched for had the x-bit set
-> access( file, R_OK||X_OK ) worked
- inicidentally, the file we searched for also had the r-bit set,
but we were not testing that in fact.
(transplanted from eebcaff6626f09af4f69b7b06f1246a06769d0af)

complibs/gmp: 5.0.0 misses 'EXPERIMENTAL' in the prompt
(transplanted from 95b4a458fd60d3e0911750c2b7c1b6cf36285845)

cc/gcc: fix headers copy for core static
(transplanted from 22ce0c19f06a530d72fd65c89d0777d16c0a97ac)

crosstool-NG.sh: suppress failure message from uname -o

On non-GNU systems (BSD/MacOS) there is no uname -o.
Suppress the failure message on these systems in the
call to set CT_SYS_OS (uname -s actually sets this variable).

yann.morin.1998@anciens.enib.fr: remove 'uname -o' altogether.
(transplanted from acae437fddccd7681c93a3b6c999c0796693d625)

scripts: enhance Makefile for installation

Split the multiple directory copies into autonomous rules.
Fix the autocompletion message.

test-suite: correctly install when not --local

scripts/wrapper: fix wrong test when checking access to the reall tool

In C, the proper syntax for a bit-wise OR is a single '|', not two.

It worked so far because all was well:
- X_OK == 1
- R_OK||X_OK == 1
- the file we searched for had the x-bit set
-> access( file, R_OK||X_OK ) worked
- inicidentally, the file we searched for also had the r-bit set,
but we were not testing that in fact.

cc/gcc: add option do disable PCH

In some cases, using Pre-Compiled Headers breaks the build.
Ass an option to disable building the PCH, as suggested by:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40974

kernel/linux: allow headers from full custom source tree

Accept a local tarball name as the source of the Linux kernel headers,
rather than forcing the user to use either an upstream tarball, or a
local pre-installed headers tree.

kernel/linux: reorder upstream/custom-tree handling

docs/README.freebsd.txt: Short intro for FreeBSD

Provide short README for FreeBSD as a host OS for ct-ng.

docs/README.macos.txt: Provide hints section

Hints for building tool chains.

patches/gdb: Fix canadian building of GDB for powerpc

Adds patch for GDB v6.8, v7.0, v7.0.1 to fix canadian
cross building of GDB for powerpc.

See original patch information here:
http://sources.redhat.com/bugzilla/show_bug.cgi?id=9638

The patch is not required for GDB v7.1 (fixed).

Tested in canadian combination using mingw32 and powerpc toolchains.
Tested to not affect normal cross building of GDB for powerpc target.

Signed-off-by: Martin Lund <mgl@doredevelopment.dk>

complibs/mpc: add latest version

complibs/gmp: 5.0.0 misses 'EXPERIMENTAL' in the prompt

cc/gcc: baremetal requires a two-pass process

Here, we implement a highly ugly hack. I'm not proud of that one...

To build the libstdc++ library, the compiler requires the C library. In
case we build for non-baremetal, this is normally handled by the final
step, later.

But in the case of bare-metal, we never go through the final step (because
it does not work, and it seems complex enough to make it work), so the
baremetal compilers are issued out of the core step.

libc/newlib: build in the 'start files' pass

A few facts:
- building the C library requires a proper core compiler
- core compiler is issued from one of the core passes
- the C library is required to build libstdc++
- newlib is only built for baremetal
- in bare metal, the final compiler is issued from one of the core passes

So we need to build the C library between core pass 1 and core pass 2.
The only place is eithe libc_headers() or libc_start_files(). The most
pertinent seems to be libc_start_files().

So we build newlib from libc_start_files(), and leave libc() empty.

cc/gcc: store core build rules in an array

Using an array makes it easier to add new rules.
Besides, it is easy to expand from build rules to install rules