Attempting to ${CT_TARGET}-gcc -print-multi-lib will fail
In do_cc_core_backend, for the final compiler in a canadian cross
baremetal, warn that multi-libs cannot be determined
In do_cc_backend, for either final compiler for a canadian cross,
warn that multi-libs cannot be determined
(Plus fixed CT_PREFIX_DIR in do_cc_backend to be ${prefix})
Signed-off-by: "David Holsgrove" <david.holsgrove@xilinx.com>
Message-Id: <CAM=EW8aQDoNx-CkJHjXBoDP4iTDJ8z5hh3=KhO5UTU6rp3Pj=w@mail.gmail.com>
Patchwork-Id: 189053
I took some of the svn functionality from eglibc.
Signed-off-by: Bryan Hundven <bryanhundven@gmail.com>
[yann.morin.1998@free.fr: fix the conditional test in build script]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
That comes from way back when nothing would work as expected, and I would
easily get heated as soon as anything would break. Sigh, those were the
old days.
Apologies.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Whatever the threading model (NPTL, LT...), we build the same
core pass-1 compiler, so there is no need to have a case-esac
construct.
Remove now mis-leading and incorect comment.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Both core pass-1 and -2 compilers are unconditionally built,
so we no longer require a condition variable.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Up until now, all conditions requiring a core pass-1 was when the
threading implementation used was NPTL. So we only built the core
pass-1 when NPTL was used.
Now, things have changed (what? when? Dunno...), and some bare-metal
canadian toolchains fail to build if a core pass-1 is not present.
OTOH, a core pass-1, although not needed for non-NPTL builds, does
no harm at all if it is present.
So, unconditionally build a core pass-1 (but still pass conditional
options to the core backend).
Reported-by: Per Arnold Blaasmo <Per-Arnold.Blaasmo@atmel.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Currently, we rely on an existing external cross-compiler targetting
the target, to build the C library.
This can pause quite a few problems if that compiler is different from
the one we are building, because it could introduce some ABI issues.
This patch removes this dependency, by building the core compilers
as we do for standard cross, and also by building the binutils and
gcc, for running on the build machine.
This means we no longer need to offer the cross-sompiler selection in
the menuconfig.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Bizarrely enough, the core gcc are not enough to be able to build a
canadian cross, and a real, full cross compiler is required so that
the canadian cross can be properly built... WTF?!? Sigh...
Add a build-frontend, as was done for the binutils and the complibs.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Do for the final step the same as for the core step: compute the list
of selected langauages from the frontend, not in the backend.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
As the core backend can be used to also build the bare-metal compiler,
we have to tel it what languages to build.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Add a function that prepares the language configure option.
It is needed in at least two places, some commonalisation is needed. ;-)
Unfortunately, it is no longer possible to print warnings about experimental
languages any more. Anyway, the experimental status is clearly indicated
in the menuconfig. so it should not be a surprise if the build breaks. :-/
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
It's easier to have as much as possible stuff in the same place to
ease backup/restore, and make things easier to follow.
Move the host companion libraries install dir as a sub-dir of the
build-tools install dir (but not directly in it, it would break
for canadian or cross-native).
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
A few noop fix-ups:
- fix the comments in core pass-1
- commonalise settings that can be
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
The core compilers are used to build the C library, so they
should always run on the build machine, not on the host.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
There really is no good reason to install the core compilers in their
own places, one for each pass. We can install them with the other
build tools.
Also, this implies that:
- there are fewer directories to save/restore
- there are fewer symlinks to create for binutils
- the PATH is shorter
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
The build dir are created depending on the host (host for that specific
backend, not host for the toolchain). Only the frontends know what host
this is, so only the frontends can create non-ambiguous dirs.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
The only user of the static core compiler in pass-1 was the newlib
C library. Now that it is build in a later step, we do no longer
need to build a static core compiler in pass-1.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
In case we build a baremetal compiler, use the standard passes:
- core_cc is used to build the C library;
- as such, it is meant to run on build, not host;
- the final compiler is meant to run on host;
As the current final compiler step can not build a baremetal compiler,
call the core backend from the final step.
NB: Currently, newlib is built during the start_files pass, so we have
to have a core compiler by then... Once we can build the baremetal
compiler from the final cc step, then we can move the newlib build to
the proper step, and then get rid of the core pass-1 static compiler...
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Currently, we issue the bare-metal compiler from the pass_1 & pass_2
core compilers, because the final gcc breaks while doing so.
This implies we have to build some libces during the start_files step,
instead of the standard libc step. This is the case for newlib.
By adding a backend/frontend infra to the final gcc, we can abstract
what backend to call: the standard backend for non-bare-metal gcc,
and the core backend for bare-metal.
This patch is just an no-op, it just adds the final backend and
frontend without changing the way bare-metal is built, to come in a
subsequent patch.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
As the core backend is used to generate the bare-metal compiler,
we need to pass it the host CFLAGS.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Tell the core compiler what host it should run on (instead of
hard-coding runing on CT_HOST).
No functional change so far, switching between CT_HOST and CT_BUILD
will come in a following patch.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Currently, the discrimination on the core compilers prefixes depends on
the type of core compiler to build.
This is not correct, and the caller of the core backend should specify
the prefix.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
In case of canadian-cross, the companion libraries are not the same for
the core cc (they run on 'build') as they are for the final cc (they run
on 'host').
Prepare for this differentiation (coming later), while retaining the
current behavior (to use the same compblibs).
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Rename the core backend function to do_cc_core_backend, to
make it explicit it is a backend.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
The core backend is going to have more parameters in the upcoming
patches, so it will be a bit complex to handle.
Introduce an array-variable that is filled by the different code-paths
with the required values.
This makes the code easier to read and maintain.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
The current construct consumes the parameters while we parse them.
Change this to a construct that does not consume the parameters.
This has no impact on gcc, but is done for homogeneity with other
components (eg. glibc).
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
In some cases, it might be desirable to use the system zlib
Eg. because latest gcc seem to be totally borked when it comes
to multilib, and tries to build a multilib host zlib, when it
is *absolutely* *not* needed: we want mulitlib on the target,
not on the host! Sigh... :-(
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
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>
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>
Even if the current process is highly parallel, crosstool-NG spends most
of its time in single-job steps on fast machines (with a 12-CPU system,
I approximate the parallel vs. non-parallel time to be in the order os
1 to 3; that is crostool-NG spends two-thirds of its time running
non-parallel jobs).
Some steps to build gcc can be paralleled, gaining a litle bit of time
on the whole compilation.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
If the user builds a toolchain over an existing one, so, without removing
CT_PREFIX_DIR, the build fails as the symlinks already exist, as does the
build.log.
This can also happen (for build.log) if the user first ran in download-
or extract-only.
Patch (with no SoB) originally from:
Phil Wilshire <phil.wilshire@overturenetworks.com>
Modified by me as it did not apply cleanly.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
The new MIPS-specific options are not valid for other targets.
Also, move the arch-specific setting lower in the extra_config setting.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Add the following MIPS specific options when configuring gcc:
--with(out)-llsc
--with(out)-synci
--with(out)-mips-plt
--with-divide=type
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Add an option to specify the hash type that gcc will ask the linker to use.
It is a provision for the upcoming 4.7, as no version currently supports it.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Add an option to configure gcc with --enable-linker-build-id.
Reported-by: Bryan Hundven <bryanhundven@gmail.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
That option is coming from the original crosstool, and is not entirely
understand here.
Moreover, it breaks with newer gcc-s: 4.6.1 now breaks while configuring
libjava (and probably some other libs as well, untested).
There is an related bug report to the gcc BZ:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49555
If need be, the old behavior can be restored with:
CC_CORE_EXTRA_CONFIG_ARRAY="--enable-symver=gnu"
CC_EXTRA_CONFIG_ARRAY="--enable-symver=gnu"
Reported-by: Bryan Hundven <bryanhundven@gmail.com>
Reviewed-by: Bryan Hundven <bryanhundven@gmail.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
"crosstool-NG-${CT_VERSION}" is currently the default for TOOLCHAIN_PKGVERSION,
and this options is passed as is to --with-pkgversion.
This patch prepends "crosstool-NG ${CT_VERSION}" to TOOLCHAIN_PKGVERSION before
passing it to --with-pkgversion.
Signed-off-by: "Benoît THÉBAUDEAU" <benoit.thebaudeau@advansee.com>
This patch promotes the PKGVERSION and BUGURL options to toolchain level so that
all toolchain components supporting them can benefit from them.
These options are passed to configure through --with-pkgversion and
--with-bugurl.
They are supported by binutils 2.18+, gcc 4.3+, eglibc 2.9+ and gdb 7.0+.
Signed-off-by: "Benoît THÉBAUDEAU" <benoit.thebaudeau@advansee.com>
Make it explicit that a variable is an array bu the name of the variable.
It will be used later when .config gets munged to allow both multiple
arguments and arguments with spaces at the same time to be passed from the
configuration down to the build scripts.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
PPL 0.11+ installs three libs: lippl, libppl_c and libpwl.
libppl_c has a dependency on libpwl (at least for watchdog stuff).
While gcc correctly links with libppl and libppl_c, it does not
pull libpwl in. In case of shared libs, this is not a problem, as
libppl_c has a NEEDED dependency on libpwl. But for static libs,
that does not work. Although libppl_c.la exists and has a correct
dependency on lipwl, somehow gcc misses it. So we have to force
pulling libpwl when needed.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
There was a mishap when cut-n-pasting code from the final
step into the core step: a variable was not renamed.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Enabling plugins in binutils is not enough, and gcc also
needs to be ./configured with --enable-plugins, although
this is not documented anywhere... :-/
Reported-by: karthik duraisami <kdconstant@hotmail.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
The reunification of the glibc/eglibc code paths exposed a nasty
bug in the glibc build: use of PARALLELMFLAGS breaks the build.
See the explanations in that bug report against FC6:
https://bugzilla.redhat.com/show_bug.cgi?format=multiple&id=212111
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
To properly enable LTO with gold, gcc has to install a plugin that gold
uses to handle the LTO information.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
- add a new parameter to do_cc_core: build_statically=[yes|no]
- pass build_statically=yes in core_pass_2 when doing bare_metal
- fix handling the static / static libstdc++ / static complibs stuff
- add a commment to keep both blocks (in core and final) in sync
Signed-off-by: "Bryan Hundven" <bryanhundven@gmail.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
If the global static option is set, then build the final gcc statically.
Signed-off-by: "Bryan Hundven" <bryanhundven@gmail.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
This was intended as a fix for g++ not finding its headers,
but it breaks in othe horrible ways. So just revert it.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
I ran into some minor difficulties looking through the build log for a
particular file: I wasn't interested in seeing it unpacked, but only
when it is built or installed. Adding these two levels allows me to
differentiate between those cases.
[Yann E. MORIN: Those are blind log levels, and are used only to search
in the build-log afterward.]
Signed-off-by: Anthony Foiani <anthony.foiani@gmail.com>
In case we build the C++ compiler, we have to tell gcc where to put the C++
headers, or else it will try to # put it in prefix/tuple/include, which we
make a symlink to sysroot/usr/include during the build, and that we delete
(the symlink!) after the build, but gcc will not look in sysroot/usr/inlcude
for C++ headers by default.
Implements a fix suggested by: Bryan Hundven <bryanhundven@gmail.com>
Reported-by: Anthony Foiani <anthony.foiani@gmail.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
libssp is the run-time Stack-Smashing Protection library.
It can be usefull to have or miss, depends...
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
libgomp is the GNU implementation of the OpenMP API.
It can be usefull to have or miss, depends...
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Force gcc to not link with some companion libraries when
there are not needed (because selected-out).
There is no option to tell gcc *not* to build the Graphite and/or
LTO stuff. They *will* be built if gcc finds the suitable companion
libraries. If we do not provide them, but the host has them, then
gcc *will* find them, and link with them.
Consider the following:
- host has suitable PPL and CLooG (eg. Debian Squeeze)
- user wants to build gcc>=4.4
- user de-selects GRAPHITE
- gcc will find the hosts PPL and CLooG, and will use them
- the user moves the toolchain to an older host that does
not have them (eg. Debian Lenny)
- the toolchain fails, when it was properly setup not to
So, explicitly tell gcc *not* to use unneeded companion libs.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
While GMP and MPFR are required by gcc>=4.3 (to build the frontends),
and MPC is required by gcc>=4.5, the other libs are not. If they are
present then gcc will enable advanced features; if they are missing,
then gcc will (should) simply disable those features.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
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>
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>
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>
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.
While compiling a canadian toolchain for host=mingw32, build=linux,
target=m68k-elf the build fails because in this step of the gcc build
the Host compiler is used in this stage with the build-flags for the
build system. This results in an error where the header <sys/wait.h>
cannot be found.
This problem happens at least in the GCC-4.3.x and GCC-4.4.x range.
This is solved by passing the proper compilers on the Make cmd-line
Signed-off-by: Remy Bohmer <linux@bohmer.net>
Previous addition of the canadian cross compiler did not allow
to build a baremetal only variant, no reason why this is not
allowed
Signed-off-by: Bart vdr. Meulen <bartvdrmeulen@gmail.com>
When building a cross-compiler for a host which depends
on file extensions the symlink for cc was not installed correctly
Signed-off-by: Bart vdr. Meulen <bartvdrmeulen@gmail.com>
[Yann E. MORIN: style fixes, enhancements, code prettying]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
When building for bare-metal the core-gcc compiler is delivered
as final compiler, so the version info and bugurl is useful
in the core compiler as well.
Signed-off-by: Remy Bohmer <linux@bohmer.net>
If threads are disabled in libc, we don't want to enable them in the
final compiler. Doing so pass the configure stage, but fails latter on
a missing <pthread.h>.
Moreover, we don't want to build libgomp if threads are disabled; its
configure script would fails anyway.
Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>
Add implementation for a candadian build option already
present in crosstool in order to build a cross-compiler
where build != host != target
Signed-off-by: Bart van der Meulen <bartvdrmeulen@gmail.com>
Initial patch by Dmitry PLOTNIKOV: http://sourceware.org/ml/crossgcc/2009-03/msg00053.html
It [the toolchain] uses current ct-ng (nightly snapshot 20090324, latest
release 1.3.2 work also), glibc 2.9 (from CVS), binutils 2.19 and latest
snapshot of GCC 4.4.0 (as of March 20, 2009).
We have successfully built linux kernel 2.6.29 and a lot of other stuff
with this toolchain.
Here's the patch that adds GCC 4.4.0 to the ct-ng menu and enables it to
download a 4.4.0 snapshot from ftp.
Patch was adpated by me, mostly to better fit the configuration layout.
/trunk/scripts/build/cc/gcc.sh | 34 22 12 0 ++++++++++++++++++++++------------
/trunk/config/cc/gcc.in | 35 30 5 0 ++++++++++++++++++++++++++++++-----
2 files changed, 52 insertions(+), 17 deletions(-)
Andy JOHNSON wrote:
The Java compiler for GCC versions 4.3.0 and up requires the
Eclipse compiler "ecj1" to be built as well. I added "gcj" to
the list of utilities to make the initial link.
/trunk/scripts/build/cc/gcc.sh | 12 12 0 0 ++++++++++++
/trunk/scripts/crosstool.sh | 2 1 1 0 +-
/trunk/config/cc/gcc.in | 6 6 0 0 ++++++
3 files changed, 19 insertions(+), 1 deletion(-)
This helps those who want to relocate their toolchains later.
Reported by Nye Liu: http://sourceware.org/ml/crossgcc/2008-10/msg00093.html
/trunk/scripts/build/cc/gcc.sh | 2 1 1 0 +-
1 file changed, 1 insertion(+), 1 deletion(-)
