Currently, perf-tar*-src-pkg only uses 'git archive', but it is better
to make it work without relying on git.

The file, HEAD, which saves the commit hash, will be included in the
tarball only when the source tree is managed by git. The git tree is
more precisely checked; it has been copied from scripts/setlocalversion.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Change the source format from "1.0" to "3.0 (quilt)" because it works
more cleanly.

All files except .config and debian/ go into the orig tarball.
Add a single patch, debian/patches/config, and delete the ugly
extend-diff-ignore patterns.

The debian tarball will be compressed into *.debian.tar.xz by default.
If you like to use a different compression mode, you can pass the
command line option, DPKG_FLAGS=-Zgzip, for example.

The orig tarball only supports gzip for now. The combination of
gzip and xz is somewhat clumsy, but it is not a practical problem.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>

If '..' belongs to the same filesystem, create a hard link instead of
a copy. In most cases, you can save disk space.

I do not want to use 'mv' because keeping linux.tar.gz is useful to
avoid unneeded rebuilding of the tarball.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>

scripts/Makefile.package does not need to know the value of
KDEB_SOURCENAME because the source name can be taken from
debian/changelog by using dpkg-parsechangelog.

Move the default of KDEB_SOURCENAME (i.e. linux-upstream) to
scripts/package/mkdebian.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>

If you run 'make (src)rpm-pkg', all objects are lost due to 'make clean',
which makes the incremental builds impossible.

Instead of cleaning, pass the exclude list to tar's --exclude-from
option.

Previously, the .config was contained in the source tarball.

With this commit, the source rpm consists of separate linux.tar.gz
and .config.

Remove stale comments. Now, 'make (src)rpm-pkg' works with O= option.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

The build rules of rpm-pkg and srcrpm-pkg are almost the same.
Remove the code duplication.

Change rpm-pkg to build binary packages from the source package generated
by srcrpm-pkg.

This changes the output directory of the srpm generated by 'make rpm-pkg'
because srcrpm-pkg overrides _srcrpmdir.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

If you run 'make deb-pkg', all objects are lost due to 'make clean',
which makes the incremental builds impossible.

Instead of cleaning, pass the exclude list to tar's --exclude-from
option.

Previously, *.diff.gz contained some check-in files such as
.clang-format, .cocciconfig.

With this commit, *.diff.gz will only contain the .config and debian/.
The other source files will go into the .orig tarball.

linux.tar.gz is rebuilt only when the source files that would go into
the tarball are changed.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>

In short, the motivation of this commit is to build a source package
without cleaning the source tree.

The deb-pkg and (src)rpm-pkg targets first run 'make clean' before
creating a source tarball. Otherwise build artifacts such as *.o,
*.a, etc. would be included in the tarball. Yet, the tarball ends up
containing several garbage files since 'make clean' does not clean
everything.

Cleaning the tree every time is annoying since it makes the incremental
build impossible. It is desirable to create a source tarball without
cleaning the tree.

In fact, there are some ways to achieve this.

The easiest solution is 'git archive'. 'make perf-tar*-src-pkg' uses
it, but I do not like it because it works only when the source tree is
managed by git, and all files you want in the tarball must be committed
in advance.

I want to make it work without relying on git. We can do this.

Files that are ignored by git are generated files, so should be excluded
from the source tarball. We can list them out by parsing the .gitignore
files. Of course, .gitignore does not cover all the cases, but it works
well enough.

tar(1) claims to support it:

  --exclude-vcs-ignores

    Exclude files that match patterns read from VCS-specific ignore files.
    Supported files are: .cvsignore, .gitignore, .bzrignore, and .hgignore.

The best scenario would be to use 'tar --exclude-vcs-ignores', but this
option does not work. --exclude-vcs-ignore does not understand any of
the negation (!), preceding slash, following slash, etc.. So, this option
is just useless.

Hence, I wrote this gitignore parser. The previous version [1], written
in Python, was so slow. This version is implemented in C, so it works
much faster.

I imported the code from git (commit: 23c56f7bd5f1), so we get the same
result.

This tool traverses the source tree, parsing all .gitignore files, and
prints file paths that are ignored by git.

The output is similar to 'git ls-files --ignored --directory --others
--exclude-per-directory=.gitignore', except

  [1] Not sorted
  [2] No trailing slash for directories

[2] is intentional because tar's --exclude-from option cannot handle
trailing slashes.

[How to test this tool]

  $ git clean -dfx
  $ make -s -j$(nproc) defconfig all                       # or allmodconifg or whatever
  $ git archive -o ../linux1.tar --prefix=./ HEAD
  $ tar tf ../linux1.tar | LANG=C sort > ../file-list1     # files emitted by 'git archive'
  $ make scripts_package
    HOSTCC  scripts/list-gitignored
  $ scripts/list-gitignored  --prefix=./ -o ../exclude-list
  $ tar cf ../linux2.tar --exclude-from=../exclude-list .
  $ tar tf ../linux2.tar | LANG=C sort > ../file-list2     # files emitted by 'tar'
  $ diff  ../file-list1 ../file-list2 | grep -E '^(<|>)'
  < ./Documentation/devicetree/bindings/.yamllint
  < ./drivers/clk/.kunitconfig
  < ./drivers/gpu/drm/tests/.kunitconfig
  < ./drivers/hid/.kunitconfig
  < ./fs/ext4/.kunitconfig
  < ./fs/fat/.kunitconfig
  < ./kernel/kcsan/.kunitconfig
  < ./lib/kunit/.kunitconfig
  < ./mm/kfence/.kunitconfig
  < ./tools/testing/selftests/arm64/tags/
  < ./tools/testing/selftests/arm64/tags/.gitignore
  < ./tools/testing/selftests/arm64/tags/Makefile
  < ./tools/testing/selftests/arm64/tags/run_tags_test.sh
  < ./tools/testing/selftests/arm64/tags/tags_test.c
  < ./tools/testing/selftests/kvm/.gitignore
  < ./tools/testing/selftests/kvm/Makefile
  < ./tools/testing/selftests/kvm/config
  < ./tools/testing/selftests/kvm/settings

The source tarball contains most of files that are tracked by git. You
see some diffs, but it is just because some .gitignore files are wrong.

  $ git ls-files -i -c --exclude-per-directory=.gitignore
  Documentation/devicetree/bindings/.yamllint
  drivers/clk/.kunitconfig
  drivers/gpu/drm/tests/.kunitconfig
  drivers/hid/.kunitconfig
  fs/ext4/.kunitconfig
  fs/fat/.kunitconfig
  kernel/kcsan/.kunitconfig
  lib/kunit/.kunitconfig
  mm/kfence/.kunitconfig
  tools/testing/selftests/arm64/tags/.gitignore
  tools/testing/selftests/arm64/tags/Makefile
  tools/testing/selftests/arm64/tags/run_tags_test.sh
  tools/testing/selftests/arm64/tags/tags_test.c
  tools/testing/selftests/kvm/.gitignore
  tools/testing/selftests/kvm/Makefile
  tools/testing/selftests/kvm/config
  tools/testing/selftests/kvm/settings

[1]: https://lore.kernel.org/all/20230128173843.765212-1-masahiroy@kernel.org/Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Chimera Linux is a Linux distribution from 2021 that builds its kernels
with Clang.

Google transitioned its data center fleet to run Clang built kernels in
2021, and Meta did so as well in 2022.  Meta talked about this at LPC
2022 at a talk titled Kernel Live Patching at Scale.

These were important milestones for building the kernel with Clang.
Making note of them helps improve confidence in the project.
Reviewed-by: Tom Rix <trix@redhat.com>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Sedat Dilek <sedat.dilek@gmail.com>
Co-developed-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Currently, setlocalversion uses any annotated tag for git-describe.
If we are at a tagged commit, it will not append the commit hash.

  $ git checkout v6.2-rc1^
  $ make -s defconfig kernelrelease
  6.1.0-14595-g292a089d
  $ git tag -a foo -m foo
  $ make -s kernelrelease
  6.1.0

If a local tag 'foo' exists, it pretends to be a released version
'6.1.0', while there are many commits on top of it.

The output should be consistent irrespective of such a local tag.
Pass the correct release tag to --match option of git-describe.

In the mainline kernel, the SUBLEVEL is always '0', which is omitted
from the tag.

  KERNELVERSION      annotated tag
  6.1.0          ->  v6.1            (mainline)
  6.2.0-rc5      ->  v6.2-rc5        (mainline, release candidate)
  6.1.7          ->  v6.1.7          (stable)

To preserve the behavior in linux-next, use the tag derived from
localversion* files if exists. In linux-next, the local version is
specified by the localversion-next file.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Concatenate all components in the last line instead of accumulating
them into the 'res' variable.

No functional change is intended. A preparation for the next change.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

The 'b4' command creates a *.mbx file, and also a *.cover file if the
patch set has a cover-letter. Ignore them.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>

I added $(srctree)/ to some included Makefiles in the following commits:

 - 3204a7fb ("kbuild: prefix $(srctree)/ to some included Makefiles")
 - d8285639 ("kbuild: do not require sub-make for separate output tree builds")

They were a preparation for removing --include-dir flag.

I have never thought --include-dir useful. Rather, it _is_ harmful.

For example, run the following commands:

  $ make -s ARCH=x86 mrproper defconfig
  $ make ARCH=arm O=foo dtbs
  make[1]: Entering directory '/tmp/linux/foo'
    HOSTCC  scripts/basic/fixdep
  Error: kernelrelease not valid - run 'make prepare' to update it
    UPD     include/config/kernel.release
  make[1]: Leaving directory '/tmp/linux/foo'

The first command configures the source tree for x86. The next command
tries to build ARM device trees in the separate foo/ directory - this
must stop because the directory foo/ has not been configured yet.

However, due to --include-dir=$(abs_srctree), the top Makefile includes
the wrong include/config/auto.conf from the source tree and continues
building. Kbuild traverses the directory tree, but of course it does
not work correctly. The Error message is also pointless - 'make prepare'
does not help at all for fixing the issue.

This commit fixes more arch Makefile, and finally removes --include-dir
from the top Makefile.

There are more breakages under drivers/, but I do not volunteer to fix
them all. I just moved --include-dir to drivers/Makefile.

With this commit, the second command will stop with a sensible message.

  $ make -s ARCH=x86 mrproper defconfig
  $ make ARCH=arm O=foo dtbs
  make[1]: Entering directory '/tmp/linux/foo'
    SYNC    include/config/auto.conf.cmd
  ***
  *** The source tree is not clean, please run 'make ARCH=arm mrproper'
  *** in /tmp/linux
  ***
  make[2]: *** [../Makefile:646: outputmakefile] Error 1
  /tmp/linux/Makefile:770: include/config/auto.conf.cmd: No such file or directory
  make[1]: *** [/tmp/linux/Makefile:793: include/config/auto.conf.cmd] Error 2
  make[1]: Leaving directory '/tmp/linux/foo'
  make: *** [Makefile:226: __sub-make] Error 2
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Add missing underscore in CONFIG_DEBUG_INFO_BTF_MODULES.

Fixes: f73edc89 ("kbuild: unify two modpost invocations")
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

When there is a missing input file (vmlinux.o or Module.symvers), you
are likely to get a ton of unresolved symbols.

Currently, Kbuild automatically adds the -w option to allow module builds
to continue with warnings instead of errors.

This may not be what the user expects because it is generally more useful
to catch all possible issues at build time instead of at run time.

Let's not do what the user did not ask.

If you still want to build modules anyway, you can proceed by explicitly
setting KBUILD_MODPOST_WARN=1. Since you may miss a real issue, you need
to be aware of what you are doing.
Suggested-by: William McVicker <willmcvicker@google.com>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Tested-by: Will McVicker <willmcvicker@google.com>

After commit 8d9acfce ("kbuild: Stop using '-Qunused-arguments' with
clang"), the PowerPC vDSO shows the following error with clang-13 and
older when CONFIG_INIT_STACK_ALL_ZERO is enabled:

  clang: error: argument unused during compilation: '-enable-trivial-auto-var-init-zero-knowing-it-will-be-removed-from-clang' [-Werror,-Wunused-command-line-argument]

clang-14 added a change to make sure this flag never triggers
-Wunused-command-line-argument, so it is fixed with newer releases. For
older releases that the kernel still supports building with, just filter
out this flag, as has been done for other flags.

Fixes: f0a42fba ("powerpc/vdso: Improve linker flags")
Fixes: 8d9acfce ("kbuild: Stop using '-Qunused-arguments' with clang")
Link: https://github.com/llvm/llvm-project/commit/ca6d5813d17598cd180995fb3bdfca00f364475fSigned-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

If the source package fails to build, ../linux.orig is left over.
In the next run of 'make deb-pkg', you will get the following error:

  dpkg-source: error: orig directory 'linux.orig' already exists, not overwriting, giving up; use -sA, -sK or -sP to override

You can manually remove ../linux.orig, but it is annoying.

Pass -sP down to dpkg-source.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

${KERNELRELEASE} is used as a part of the installation path.
(INSTALL_DTBS_PATH, MODLIB, etc.)

When KERNELRELEASE is overridden from the command line, it should be
saved in include/config/kernel.release, so that it will be consistently
used for the installation steps.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Print $(KERNELVERSION) in setlocalversion so that the callers get
simpler.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Return earlier if we are not in the correct git repository. This makes
the code more readable.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>

With the --short option given, scm_version() prints "+".
Just append it.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>

.scmversion is used by (src)rpm-pkg and deb-pkg to carry KERNELRELEASE.

In fact, deb-pkg does not rely on it any more because the generated
debian/rules specifies KERNELRELEASE from the command line.

Do likwise for (src)rpm-pkg, and remove this feature.
For the same reason, you do not need to save LOCALVERSION in the
spec file.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>

For each binary Debian package, a directory with the package name is
created in the debian directory. Correct the generated file matches in the
package's clean target, which were renamed without adjusting the target.

Fixes: 1694e94e ("builddeb: match temporary directory name to the package name")
Signed-off-by: Bastian Germann <bage@linutronix.de>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

No need to call chmod three times when it can do everything at once.
Signed-off-by: Sven Joachim <svenjoac@gmx.de>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Commit 80f8be7a ("tomoyo: Omit use of bin2c") removed the last
use of bin2c.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>
Reviewed-by: Sedat Dilek <sedat.dilek@gmail.com>

In the follow-up of commit fb3041d6 ("kbuild: fix SIGPIPE error
message for AR=gcc-ar and AR=llvm-ar"), Kees Cook pointed out that
tools should _not_ catch their own SIGPIPEs [1] [2].

Based on his feedback, LLVM was fixed [3].

However, Python's default behavior is to show noisy bracktrace when
SIGPIPE is sent. So, scripts written in Python are basically in the
same situation as the buggy llvm tools.

Example:

  $ make -s allnoconfig
  $ make -s allmodconfig
  $ scripts/diffconfig .config.old .config | head -n1
  -ALIX n
  Traceback (most recent call last):
    File "/home/masahiro/linux/scripts/diffconfig", line 132, in <module>
      main()
    File "/home/masahiro/linux/scripts/diffconfig", line 130, in main
      print_config("+", config, None, b[config])
    File "/home/masahiro/linux/scripts/diffconfig", line 64, in print_config
      print("+%s %s" % (config, new_value))
  BrokenPipeError: [Errno 32] Broken pipe

Python documentation [4] notes how to make scripts die immediately and
silently:

  """
  Piping output of your program to tools like head(1) will cause a
  SIGPIPE signal to be sent to your process when the receiver of its
  standard output closes early. This results in an exception like
  BrokenPipeError: [Errno 32] Broken pipe. To handle this case,
  wrap your entry point to catch this exception as follows:

    import os
    import sys

    def main():
        try:
            # simulate large output (your code replaces this loop)
            for x in range(10000):
                print("y")
            # flush output here to force SIGPIPE to be triggered
            # while inside this try block.
            sys.stdout.flush()
        except BrokenPipeError:
            # Python flushes standard streams on exit; redirect remaining output
            # to devnull to avoid another BrokenPipeError at shutdown
            devnull = os.open(os.devnull, os.O_WRONLY)
            os.dup2(devnull, sys.stdout.fileno())
            sys.exit(1)  # Python exits with error code 1 on EPIPE

    if __name__ == '__main__':
        main()

  Do not set SIGPIPE’s disposition to SIG_DFL in order to avoid
  BrokenPipeError. Doing that would cause your program to exit
  unexpectedly whenever any socket connection is interrupted while
  your program is still writing to it.
  """

Currently, tools/perf/scripts/python/intel-pt-events.py seems to be the
only script that fixes the issue that way.

tools/perf/scripts/python/compaction-times.py uses another approach
signal.signal(signal.SIGPIPE, signal.SIG_DFL) but the Python
documentation clearly says "Don't do it".

I cannot fix all Python scripts since there are so many.
I fixed some in the scripts/ directory.

[1]: https://lore.kernel.org/all/202211161056.1B9611A@keescook/
[2]: https://github.com/llvm/llvm-project/issues/59037
[3]: https://github.com/llvm/llvm-project/commit/4787efa38066adb51e2c049499d25b3610c0877b
[4]: https://docs.python.org/3/library/signal.html#note-on-sigpipeSigned-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>

This option masks all unused command line argument warnings, which can
hide potential issues, such as an architecture Makefile adding an
unsupported flag to KBUILD_AFLAGS or KBUILD_CFLAGS, which will cause all
as-option and cc-options to silently fail due to -Werror with no
indication as to why in the main kernel build.

Remove this flag so that warnings of this nature can be caught early and
obviously in a build.
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Currently, these warnings are hidden with -Qunused-arguments in
KBUILD_CPPFLAGS. Once that option is removed, these warnings should be
turned into hard errors to make unconditionally added but unsupported
flags for the current compilation mode or target obvious due to a failed
build; otherwise, the warnings might just be ignored if the build log is
not checked.

Link: https://github.com/ClangBuiltLinux/linux/issues/1587Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

When clang's -Qunused-arguments is dropped from KBUILD_CPPFLAGS, it
warns:

  clang-16: error: argument unused during compilation: '-mhard-float' [-Werror,-Wunused-command-line-argument]

Similar to commit 84edc2ef ("selftest/fpu: avoid clang warning"),
just add this flag to GCC builds. Commit 0f0727d9 ("drm/amd/display:
readd -msse2 to prevent Clang from emitting libcalls to undefined SW FP
routines") added '-msse2' to prevent clang from emitting software
floating point routines.
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

When clang's -Qunused-arguments is dropped from KBUILD_CPPFLAGS, it
warns while building objects in the purgatory folder:

  clang-16: error: argument unused during compilation: '-MD' [-Werror,-Wunused-command-line-argument]

'-MMD' is always passed to the preprocessor via c_flags, even when
KBUILD_CFLAGS is overridden in a folder, so clang complains the addition
of '-MD' will be unused. Remove '-MD' to clear up this warning, as it is
unnecessary with '-MMD'.

Additionally, '-c' is also unnecessary, remove it while in the area.
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Reviewed-by: Sven Schnelle <svens@linux.ibm.com>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

When clang's -Qunused-arguments is dropped from KBUILD_CPPFLAGS, it
points out that there is a linking phase flag added to CFLAGS, which
will only be used for compiling

  clang-16: error: argument unused during compilation: '-shared' [-Werror,-Wunused-command-line-argument]

'-shared' is already present in ldflags-y so it can just be dropped.

Fixes: 2b2a2584 ("s390/vdso: Use $(LD) instead of $(CC) to link vDSO")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Reviewed-by: Sven Schnelle <svens@linux.ibm.com>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

When clang's -Qunused-arguments is dropped from KBUILD_CPPFLAGS, it
warns:

  clang-16: error: argument unused during compilation: '-s' [-Werror,-Wunused-command-line-argument]

The compiler's '-s' flag is a linking option (it is passed along to the
linker directly), which means it does nothing when the linker is not
invoked by the compiler. The kernel builds all .o files with '-c', which
stops the compilation pipeline before linking, so '-s' can be safely
dropped from KBUILD_AFLAGS_64.
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Reviewed-by: Sven Schnelle <svens@linux.ibm.com>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

When clang's -Qunused-arguments is dropped from KBUILD_CPPFLAGS, it
warns:

  clang-16: error: argument unused during compilation: '-fno-stack-clash-protection' [-Werror,-Wunused-command-line-argument]

This warning happens because vgettimeofday-32.c gets its base CFLAGS
from the main kernel, which may contain flags that are only supported on
a 64-bit target but not a 32-bit one, which is the case here.
-fstack-clash-protection and its negation are only suppported by the
64-bit powerpc target but that flag is included in an invocation for a
32-bit powerpc target, so clang points out that while the flag is one
that it recognizes, it is not actually used by this compiler job.

To eliminate the warning, remove -fno-stack-clash-protection from
vgettimeofday-32.c's CFLAGS when using clang, as has been done for other
flags previously.
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

When clang's -Qunused-arguments is dropped from KBUILD_CPPFLAGS, there
are several warnings in the PowerPC vDSO:

  clang-16: error: -Wl,-soname=linux-vdso32.so.1: 'linker' input unused [-Werror,-Wunused-command-line-argument]
  clang-16: error: -Wl,--hash-style=both: 'linker' input unused [-Werror,-Wunused-command-line-argument]
  clang-16: error: argument unused during compilation: '-shared' [-Werror,-Wunused-command-line-argument]

  clang-16: error: argument unused during compilation: '-nostdinc' [-Werror,-Wunused-command-line-argument]
  clang-16: error: argument unused during compilation: '-Wa,-maltivec' [-Werror,-Wunused-command-line-argument]

The first group of warnings point out that linker flags were being added
to all invocations of $(CC), even though they will only be used during
the final vDSO link. Move those flags to ldflags-y.

The second group of warnings are compiler or assembler flags that will
be unused during linking. Filter them out from KBUILD_CFLAGS so that
they are not used during linking.

Additionally, '-z noexecstack' was added directly to the ld_and_check
rule in commit 1d53c019 ("powerpc/vdso: link with -z noexecstack")
but now that there is a common ldflags variable, it can be moved there.
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

When clang's -Qunused-arguments is dropped from KBUILD_CPPFLAGS, it
warns:

  clang-16: error: argument unused during compilation: '-s' [-Werror,-Wunused-command-line-argument]

The compiler's '-s' flag is a linking option (it is passed along to the
linker directly), which means it does nothing when the linker is not
invoked by the compiler. The kernel builds all .o files with '-c', which
stops the compilation pipeline before linking, so '-s' can be safely
dropped from ASFLAGS.
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Reviewed-by: Segher Boessenkool <segher@kernel.crashing.org>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

When clang's -Qunused-arguments is dropped from KBUILD_CPPFLAGS, it
points out that KBUILD_AFLAGS contains a linker flag, which will be
unused:

  clang: error: -Wl,-a32: 'linker' input unused [-Werror,-Wunused-command-line-argument]

This was likely supposed to be '-Wa,-a$(BITS)'. However, this change is
unnecessary, as all supported versions of clang and gcc will pass '-a64'
or '-a32' to GNU as based on the value of '-m'; the behavior of the
latest stable release of the oldest supported major version of each
compiler is shown below and each compiler's latest release exhibits the
same behavior (GCC 12.2.0 and Clang 15.0.6).

  $ powerpc64-linux-gcc --version | head -1
  powerpc64-linux-gcc (GCC) 5.5.0

  $ powerpc64-linux-gcc -m64 -### -x assembler-with-cpp -c -o /dev/null /dev/null &| grep 'as '
  .../as -a64 -mppc64 -many -mbig -o /dev/null /tmp/cctwuBzZ.s

  $ powerpc64-linux-gcc -m32 -### -x assembler-with-cpp -c -o /dev/null /dev/null &| grep 'as '
  .../as -a32 -mppc -many -mbig -o /dev/null /tmp/ccaZP4mF.sg

  $ clang --version | head -1
  Ubuntu clang version 11.1.0-++20211011094159+1fdec59bffc1-1~exp1~20211011214622.5

  $ clang --target=powerpc64-linux-gnu -fno-integrated-as -m64 -### \
    -x assembler-with-cpp -c -o /dev/null /dev/null &| grep gnu-as
   "/usr/bin/powerpc64-linux-gnu-as" "-a64" "-mppc64" "-many" "-o" "/dev/null" "/tmp/null-80267c.s"

  $ clang --target=powerpc64-linux-gnu -fno-integrated-as -m64 -### \
    -x assembler-with-cpp -c -o /dev/null /dev/null &| grep gnu-as
   "/usr/bin/powerpc64-linux-gnu-as" "-a32" "-mppc" "-many" "-o" "/dev/null" "/tmp/null-ab8f8d.s"

Remove this flag altogether to avoid future issues.

Fixes: 1421dc6d ("powerpc/kbuild: Use flags variables rather than overriding LD/CC/AS")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

as-instr uses KBUILD_AFLAGS, but as-option uses KBUILD_CFLAGS. This can
cause as-option to fail unexpectedly when CONFIG_WERROR is set, because
clang will emit -Werror,-Wunused-command-line-argument for various -m
and -f flags in KBUILD_CFLAGS for assembler sources.

Callers of as-option and as-instr should be adding flags to
KBUILD_AFLAGS / aflags-y, not KBUILD_CFLAGS / cflags-y. Use
KBUILD_AFLAGS in all macros to clear up the initial problem.

Unfortunately, -Wunused-command-line-argument can still be triggered
with clang by the presence of warning flags or macro definitions because
'-x assembler' is used, instead of '-x assembler-with-cpp', which will
consume these flags. Switch to '-x assembler-with-cpp' in places where
'-x assembler' is used, as the compiler is always used as the driver for
out of line assembler sources in the kernel.

Finally, add -Werror to these macros so that they behave consistently
whether or not CONFIG_WERROR is set.

[nathan: Reworded and expanded on problems in commit message
         Use '-x assembler-with-cpp' in a couple more places]

Link: https://github.com/ClangBuiltLinux/linux/issues/1699Suggested-by: Masahiro Yamada <masahiroy@kernel.org>
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

A future change will switch as-option to use KBUILD_AFLAGS instead of
KBUILD_CFLAGS to allow clang to drop -Qunused-arguments, which may cause
issues if the flag being tested requires a flag previously added to
KBUILD_CFLAGS but not KBUILD_AFLAGS. Use cc-option for cflags additions
so that the flags are tested properly.
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

-Wa,-msoft-float is tested with as-option, which will be a problem for
clang with an upcoming change to move as-option to use KBUILD_AFLAGS
instead of KBUILD_CFLAGS due to a lack of '-mno-abicalls' in
KBUILD_AFLAGS at the point that this check occurs; $(cflags-y) is added
to KBUILD_AFLAGS towards the end of this file.

  clang: error: ignoring '-fno-PIE' option as it cannot be used with implicit usage of -mabicalls and the N64 ABI [-Werror,-Woption-ignored]

This could be resolved by switching to a cc-option check but
'$(cflags-y)' would need to be added so that '-mno-abicalls' is present
for the test. However, this check is no longer necessary, as
-msoft-float is supported by all supported assembler versions (GNU as
2.25+ and LLVM 11+). Eliminate GAS_HAS_SET_HARDFLOAT and all of its
uses, inlining SET_HARDFLOAT where necessary.

Link: https://lore.kernel.org/202209101939.bvk64Fok-lkp@intel.com/Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

as-option tests new options using KBUILD_CFLAGS, which causes problems
when using as-option to update KBUILD_AFLAGS because many compiler
options are not valid assembler options.

This will be fixed in a follow up patch. Before doing so, move the
assembler test for -Wa,-mrelax-relocations=no from using as-option to
cc-option.

Link: https://lore.kernel.org/llvm/CAK7LNATcHt7GcXZ=jMszyH=+M_LC9Qr6yeAGRCBbE6xriLxtUQ@mail.gmail.com/Suggested-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Tested-by: Linux Kernel Functional Testing <lkft@linaro.org>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>