Currently, when the clang toolchain is used, we use the
brp-llvm-compile-lto-to-elf script to post-process any shipped
object files or static libraries to convert the LLVM bitcode they
contain into ELF object code.
With LLVM 18, Clang has introduced support for fat LTO objects
(https://llvm.org/docs/FatLTO.html), which work essentially the
same way as with GCC: If `-ffat-lto-objects` is passed, then the
objects will contain both the ELF code, as well as the LLVM
bitcode in a special section.
This redhat-rpm-config change enables the use of fat LTO and
drops the brp-llvm-compile-lto-to-elf script. Instead, the
brp-strip-lto script used by GCC also strips the LLVM section
name now.
This is part of https://fedoraproject.org/wiki/Changes/LLVM-18.
No conflict because except for %global build_type_safety 0,
the package is still compatibility with GCC 13 (but does not
enforce the type safety levels anymore).
Now that rust-packaging absorbed rust-srpm-macros in Fedora, RHEL rust
will need to provide an srpm-macros subpackage so that no part of
rust-packaging will be pulled into RHEL 10. This will then allow
rust-packaging (minus macros.rust-srpm) to exist in EPEL.
The package redefines %optflags on i686 and s390x, so the
%__build_for_lang_* macros are no longer expanded, triggering
a warning from RPM (“%__build_for_lang_any defined but not
used within scope”). Clang warns about -Wno-complain-wrong-lang
with -Wall, so the warning option is suppressed for
"%toolchain" != "gcc".
Reported-by: Tulio Magno Quites Machado Filho <tuliom@ascii.art.br>
The two intended uses cases for the _pkg_extra_* macros were to
make it easier for packagers to add new compile flags for use
with their package and also to make it easier to do distro wide
experiments with new flags.
However, it was pointed out on the mailing list[1] that you
can't satisfy both of these uses cases at the same time with
just one set of macros. For example, if a packager uses
_pkg_extra_* macros to add flags specific to their own
package, then someone using _pkg_extra_* macros to
apply a new flag distro wide would override the package
specific flag.
I feel like the distro-wide use case is much more important,
so rather than create two sets of new flags, one for each use
case, I think it's best to rename the _pkg_extra_* macros
to _distro_extra_* and document that they are only meant
to be used for adding distro-wide flags and packagers should
not use this macro.
[1] https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/TOG5RHWPS3VYMM52HFGZOUJVRCGZ7VXB/
The double undefine was unnecessary and exposed a bug in ccache. Club
the undefine and define into a single -Wp instead, which works for all
redefinitions of _FORTIFY_SOURCE.
RHEL does not intend to enable frame pointers at this time, and any
change will be done so based on a variety of inputs including the data
from Fedora. Disable them for ELN to set proper expectations.
Signed-off-by: Stephen Gallagher <sgallagh@redhat.com>
Make the _FORTIFY_SOURCE flags configurable so that the command line is
not cluttered with _FORTIFY_SOURCE definitions and undefines. Introduce
a %_fortify_level variable that a package may override by either
undefining or defining to a specific value.
Also bump the default value to 3, to implement the systemwide proposal
for Fedora 38:
https://fedoraproject.org/wiki/Changes/Add_FORTIFY_SOURCE%3D3_to_distribution_build_flags
If `%_include_frame_pointers` is defined, add `-fno-omit-frame-pointer`
and `-mno-omit-leaf-frame-pointer` to the compiler flags to ensure frame
pointers are always included.
This is in preparation for
https://fedoraproject.org/wiki/Changes/fno-omit-frame-pointer
This requires the build system support VALAFLAGS. At least Meson and
Automake do.
Using -g is desired because it allows us to see Vala source files and
line numbers in backtraces when possible, falling back to generated C
code otherwise. Line numbers for generated C are almost always less
useful when debugging crashes.
From https://bugzilla.redhat.com/2083296:
> The issue is that some packages break up the flags at spaces,
> in order to look for specific flags or to add flags of their own.
> The z3 package, for example, has some python code that does this:
>
> def exec_cmd(cmd):
> if isinstance(cmd, str):
> cmd = cmd.split(' ')
> ...
>
> The result is one of the commands in the list consists of a single tab character.
> When that is passed to the compiler, the compiler does not like it at all.
Packages which do not have %%build section but do also
compile and link test programs in %%check would fail because
no package note would have been generated.
We would try to regenerate the file whenever set_build_flags was used.
For packages which call %configure multiple times this would mean a
few times during build, and also in the build and check
preambles. Let's just simplify this and call it only in the build preamble.
https://src.fedoraproject.org/rpms/package-notes/pull-request/1
changes %_generate_package_note_file to be unconditional. This means we'll
always rewrite the file and will not use a stale version. (But the two
fixes are independent, even though they make the most sense together.)
David Abdurachmanov
Zbigniew Jędrzejewski-Szmek
Nikita Popov
Florian Weimer
Jonathan Wright
Nick Clifton
Yaakov Selkowitz
Stephen Gallagher
Charalampos Stratakis
Tom Stellard
Maxwell G
Siddhesh Poyarekar
Miro Hrončok
Davide Cavalca
Siddhesh Poyarekar
Florian Weimer
Michael Catanzaro
Frederic Berat
Dan Horák
Timm Bäder
Robert-André Mauchin