linux-toradex.git/scripts/Makefile.extrawarn, branch v5.19-rc3

kbuild: support W=e to make build abort in case of warning

2022-05-07T18:16:59+00:00

When developing new code/feature, CONFIG_WERROR is most
often turned off, especially for people using make W=12 to
get more warnings.

In such case, turning on -Werror temporarily would require
switching on CONFIG_WERROR in the configuration, building,
then switching off CONFIG_WERROR.

For this use case, this patch introduces a new 'e' modifier
to W= as a short hand for KCFLAGS+=-Werror" so that -Werror
got added to the kernel (built-in) and modules' CFLAGS.

Signed-off-by: Yann Droneaud 
Signed-off-by: Masahiro Yamada

Kbuild: add -Wno-shift-negative-value where -Wextra is used

2022-03-13T08:30:31+00:00

As a preparation for moving to -std=gnu11, turn off the
-Wshift-negative-value option. This warning is enabled by gcc when
building with -Wextra for c99 or higher, but not for c89. Since
the kernel already relies on well-defined overflow behavior,
the warning is not helpful and can simply be disabled in
all locations that use -Wextra.

Signed-off-by: Arnd Bergmann 
Acked-by: Jani Nikula 
Reviewed-by: Nathan Chancellor 
Tested-by: Sedat Dilek  # LLVM/Clang v13.0.0 (x86-64)
Signed-off-by: Masahiro Yamada

Makefile.extrawarn: Move -Wunaligned-access to W=1

2022-02-08T04:20:11+00:00

-Wunaligned-access is a new warning in clang that is default enabled for
arm and arm64 under certain circumstances within the clang frontend (see
LLVM commit below). On v5.17-rc2, an ARCH=arm allmodconfig build shows
1284 total/70 unique instances of this warning (most of the instances
are in header files), which is quite noisy.

To keep a normal build green through CONFIG_WERROR, only show this
warning with W=1, which will allow automated build systems to catch new
instances of the warning so that the total number can be driven down to
zero eventually since catching unaligned accesses at compile time would
be generally useful.

Cc: stable@vger.kernel.org
Link: https://github.com/llvm/llvm-project/commit/35737df4dcd28534bd3090157c224c19b501278a
Link: https://github.com/ClangBuiltLinux/linux/issues/1569
Link: https://github.com/ClangBuiltLinux/linux/issues/1576
Signed-off-by: Nathan Chancellor 
Reviewed-by: Nick Desaulniers 
Signed-off-by: Masahiro Yamada

Makefile.extrawarn: remove -Wnested-externs warning

2020-12-08T14:30:05+00:00

The -Wnested-externs warning has become useless with gcc, since
this warns every time that BUILD_BUG_ON() or similar macros
are used.

With clang, the warning option does nothing to start with, so
just remove it entirely.

Suggested-by: Nathan Chancellor 
Signed-off-by: Arnd Bergmann 
Reviewed-by: Nathan Chancellor 
Signed-off-by: Masahiro Yamada

Makefile.extrawarn: move -Wcast-align to W=3

2020-11-24T18:36:10+00:00

This warning behaves differently depending on the architecture
and compiler. Using x86 gcc, we get no output at all because
gcc knows the architecture can handle unaligned accesses.

Using x86 clang, or gcc on an architecture that needs to
manually deal with unaligned accesses, the build log is
completely flooded with these warnings, as they are commonly
invoked by inline functions of networking headers, e.g.

include/linux/skbuff.h:1426:26: warning: cast increases required alignment of target type [-Wcast-align]

The compiler is correct to point this out, as we are dealing
with undefined behavior that does cause problems in practice,
but there is also no good way to rewrite the code in commonly
included headers to a safer method.

Signed-off-by: Arnd Bergmann 
Reviewed-by: Nathan Chancellor 
Signed-off-by: Masahiro Yamada

Makefile.extrawarn: Move sign-compare from W=2 to W=3

2020-08-17T16:52:08+00:00

This -Wsign-compare compiler warning can be very noisy
and most of the suggested conversions are unnecessary.

Make the warning W=3 so it's described under the
"can most likely be ignored" block.

Signed-off-by: Joe Perches 
Signed-off-by: Masahiro Yamada

kbuild: Move -Wtype-limits to W=2

2020-07-10T01:00:56+00:00

-Wtype-limits is included in -Wextra which is added at W=1. It warns
(among other things) that 'comparison of an unsigned variable `< 0` is
always false. This causes noisy warnings, especially when used in
macros, hence it is more suitable for W=2.

Link: https://lore.kernel.org/lkml/CAHk-=wiKCXEWKJ9dWUimGbrVRo_N2RosESUw8E7m9AEtyZcu=w@mail.gmail.com/
Signed-off-by: Rikard Falkeborn 
Suggested-by: Arnd Bergmann 
Acked-by: Andy Shevchenko 
Signed-off-by: Linus Torvalds

kbuild: Enable -Wtautological-compare

2020-04-08T15:13:45+00:00

Currently, we disable -Wtautological-compare, which in turn disables a
bunch of more specific tautological comparison warnings that are useful
for the kernel such as -Wtautological-bitwise-compare. See clang's
documentation below for the other warnings that are suppressed by
-Wtautological-compare. Now that all of the major/noisy warnings have
been fixed, enable -Wtautological-compare so that more issues can be
caught at build time by various continuous integration setups.

-Wtautological-constant-out-of-range-compare is kept disabled under a
normal build but visible at W=1 because there are places in the kernel
where a constant or variable size can change based on the kernel
configuration. These are not fixed in a clean/concise way and the ones
I have audited so far appear to be harmless. It is not a subgroup but
rather just one warning so we do not lose out on much coverage by
default.

Link: https://github.com/ClangBuiltLinux/linux/issues/488
Link: http://releases.llvm.org/10.0.0/tools/clang/docs/DiagnosticsReference.html#wtautological-compare
Link: https://bugs.llvm.org/show_bug.cgi?id=42666
Signed-off-by: Nathan Chancellor 
Signed-off-by: Masahiro Yamada

kbuild: Disable -Wpointer-to-enum-cast

2020-03-14T01:31:08+00:00

Clang's -Wpointer-to-int-cast deviates from GCC in that it warns when
casting to enums. The kernel does this in certain places, such as device
tree matches to set the version of the device being used, which allows
the kernel to avoid using a gigantic union.

https://elixir.bootlin.com/linux/v5.5.8/source/drivers/ata/ahci_brcm.c#L428
https://elixir.bootlin.com/linux/v5.5.8/source/drivers/ata/ahci_brcm.c#L402
https://elixir.bootlin.com/linux/v5.5.8/source/include/linux/mod_devicetable.h#L264

To avoid a ton of false positive warnings, disable this particular part
of the warning, which has been split off into a separate diagnostic so
that the entire warning does not need to be turned off for clang. It
will be visible under W=1 in case people want to go about fixing these
easily and enabling the warning treewide.

Cc: stable@vger.kernel.org
Link: https://github.com/ClangBuiltLinux/linux/issues/887
Link: https://github.com/llvm/llvm-project/commit/2a41b31fcdfcb67ab7038fc2ffb606fd50b83a84
Signed-off-by: Nathan Chancellor 
Signed-off-by: Masahiro Yamada

kbuild: allow Clang to find unused static inline functions for W=1 build

2019-09-09T14:55:43+00:00

GCC and Clang have different policy for -Wunused-function; GCC does not
warn unused static inline functions at all whereas Clang does if they
are defined in source files instead of included headers although it has
been suppressed since commit abb2ea7dfd82 ("compiler, clang: suppress
warning for unused static inline functions").

We often miss to delete unused functions where 'static inline' is used
in *.c files since there is no tool to detect them. Unused code remains
until somebody notices. For example, commit 075ddd75680f ("regulator:
core: remove unused rdev_get_supply()").

Let's remove __maybe_unused from the inline macro to allow Clang to
start finding unused static inline functions. For now, we do this only
for W=1 build since it is not a good idea to sprinkle warnings for the
normal build (e.g. 35 warnings for arch/x86/configs/x86_64_defconfig).

My initial attempt was to add -Wno-unused-function for no W= build
(https://lore.kernel.org/patchwork/patch/1120594/)

Nathan Chancellor pointed out that would weaken Clang's checks since
we would no longer get -Wunused-function without W=1. It is true GCC
would catch unused static non-inline functions, but it would weaken
Clang as a standalone compiler, at least.

Hence, here is a counter implementation. The current problem is, W=...
only controls compiler flags, which are globally effective. There is
no way to address only 'static inline' functions.

This commit defines KBUILD_EXTRA_WARN[123] corresponding to W=[123].
When KBUILD_EXTRA_WARN1 is defined, __maybe_unused is omitted from
the 'inline' macro.

The new macro __inline_maybe_unused makes the code a bit uglier, so I
hope we can remove it entirely after fixing most of the warnings.

If you contribute to code clean-up, please run "make CC=clang W=1"
and check -Wunused-function warnings. You will find lots of unused
functions.

Some of them are false-positives because the call-sites are disabled
by #ifdef. I do not like to abuse the inline keyword for suppressing
unused-function warnings because it is intended to be a hint for the
compiler optimization. I prefer #ifdef around the definition, or
__maybe_unused if #ifdef would make the code too ugly.

Signed-off-by: Masahiro Yamada 
Reviewed-by: Nathan Chancellor 
Tested-by: Nathan Chancellor