- Added mpfr consolidated patches file to mpfr in gcc. mpfr is built internally for use
in the toolchain.
- Confirmed working by running./make toolchain which ran successfully
confirmed from the _build.toolchain.log file that the patches were successfully
implemented for gcc pass 1, gcc pass L and gcc pass 2
- Full toolchain build successfully completed.
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org>
Reviewed-by: Peter Müller <peter.mueller@ipfire.org>
Historically, the MD5 checksums in our LFS files serve as a protection
against broken downloads, or accidentally corrupted source files.
While the sources are nowadays downloaded via HTTPS, it make sense to
beef up integrity protection for them, since transparently intercepting
TLS is believed to be feasible for more powerful actors, and the state
of the public PKI ecosystem is clearly not helping.
Therefore, this patch switches from MD5 to BLAKE2, updating all LFS
files as well as make.sh to deal with this checksum algorithm. BLAKE2 is
notably faster (and more secure) than SHA2, so the performance penalty
introduced by this patch is negligible, if noticeable at all.
In preparation of this patch, the toolchain files currently used have
been supplied with BLAKE2 checksums as well on
https://source.ipfire.org/.
Cc: Michael Tremer <michael.tremer@ipfire.org>
Signed-off-by: Peter Müller <peter.mueller@ipfire.org>
Acked-by: Michael Tremer <michael.tremeripfire.org>
GCC can use ZSTD to compress debugging/LTO information in binary
objects. However, on riscv64, compiling zstd requires libatomic which is
not available at this point.
In order to make the build work, we explicitely disable ZSTD in GCC and
build ZSTD after libatomic is available.
Although ZSTD offers great compression, we won't have any disadvantages
through this change since we do not ship any debugging information and
at this point in time to not use LTO.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Bumping across one of our scripts with very long trailing whitespaces, I
thought it might be a good idea to clean these up. Doing so, some
missing or inconsistent licence headers were fixed.
There is no need in shipping all these files en bloc, as their
functionality won't change.
Signed-off-by: Peter Müller <peter.mueller@ipfire.org>
Binutils and GCC were misconfigured and used host libraries to build
toolchain programs. That resulted in that those programs were correctly
linked, but could not be executed, because the runtime linker did not
search in the host system.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The stage 2 compiler was looking for libraries outside the bootstrapped
toolchain environment which causes that linked programs cannot be
executied because the runtime linker only looks for libraries inside the
toolchain environment.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This patch removes support for i586 according to the decision being
taken over a year ago.
It removes the architecture from the build system and removes all
required hacks and other quirks that have been necessary before.
There is no need to ship any changed files to the remaining
architectures as the removed code branches have not been used.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
we have no supported armv5tel board left so we can switch to the higher
arch. This now can use the vpu (still in softfp calling convention to
not break existing installations.)
this fix many compile problems, also boost is now working again.
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
gcc-7 not support -fcf-protection so filter it from CFLAGS.
also filter -mtune in first pass because it should optimized for the
actual host.
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
This will allow us to run multiple builds on the same
system at the same time (or at least have them on disk).
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The build fails with various reasons and a full bootstrap
always succeeds. This takes a very long time so we try to
avoid it where ever possible.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The build environment is using a number of variables which
occasionally conflicted with some other build systems.
This patch cleans that up by renaming some variables and
later unexporting them in the lfs files.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
GCC does not need to be bootstrapped in the second pass
any more since the toolchain is not built hardened
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The toolchain will be built without hardening which makes
the entire bootstrapping process way more complicated than
necessary and sometimes fail on some host distribution.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
