GCC recently added option -fmin-function-alignment, which should appear
in GCC 14. Unlike -falign-functions, this option causes all functions to
be aligned at the specified value, including the cold ones.

In particular, when an arm64 kernel is built with
DYNAMIC_FTRACE_WITH_CALL_OPS=y, the 8-byte function alignment is
required for correct functionality. This was done by -falign-functions=8
and having workarounds in the kernel to force the compiler to follow
this alignment. The new -fmin-function-alignment option directly
guarantees it.

Detect availability of -fmin-function-alignment and use it instead of
-falign-functions when present. Introduce CC_HAS_SANE_FUNCTION_ALIGNMENT
and enable __cold to work as expected when it is set.
Signed-off-by: Petr Pavlu <petr.pavlu@suse.com>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

There are two entries for CONFIG_ALPHA_GAMMA, with the second one
7 lines below. Merge them together.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

There are two entries for CONFIG_ALPHA_EV4, on line 337 and line 368.
Merge them together.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

For the same rationale as commit 54b8ae66 ("kbuild: change
*FLAGS_<basetarget>.o to take the path relative to $(obj)").
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>

Commit 54b8ae66 ("kbuild: change *FLAGS_<basetarget>.o to take the
path relative to $(obj)") changed the syntax of per-file compiler flags.

The situation is the same for the following variables:

  OBJECT_FILES_NON_STANDARD_<basetarget>.o
  GCOV_PROFILE_<basetarget>.o
  KASAN_SANITIZE_<basetarget>.o
  KMSAN_SANITIZE_<basetarget>.o
  KMSAN_ENABLE_CHECKS_<basetarget>.o
  UBSAN_SANITIZE_<basetarget>.o
  KCOV_INSTRUMENT_<basetarget>.o
  KCSAN_SANITIZE_<basetarget>.o
  KCSAN_INSTRUMENT_BARRIERS_<basetarget>.o

The <basetarget> is the filename of the target with its directory and
suffix stripped.

This syntax comes into a trouble when two files with the same basename
appear in one Makefile, for example:

  obj-y += dir1/foo.o
  obj-y += dir2/foo.o
  OBJECT_FILES_NON_STANDARD_foo.o := y

OBJECT_FILES_NON_STANDARD_foo.o is applied to both dir1/foo.o and
dir2/foo.o. This syntax is not flexbile enough to handle cases where
one of them is a standard object, but the other is not.

It is more sensible to use the relative path to the Makefile, like this:

  obj-y += dir1/foo.o
  OBJECT_FILES_NON_STANDARD_dir1/foo.o := y
  obj-y += dir2/foo.o
  OBJECT_FILES_NON_STANDARD_dir2/foo.o := y

To maintain the current behavior, I made adjustments to the following two
Makefiles:

 - arch/x86/entry/vdso/Makefile, which compiles vclock_gettime.o, vgetcpu.o,
   and their vdso32 variants.

 - arch/x86/kvm/Makefile, which compiles vmx/vmenter.o and svm/vmenter.o
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>
Acked-by: Sean Christopherson <seanjc@google.com>

When a checklist is opened, the cursor is rendered in a wrong position
(after the last list element on the screen). You can observe it by
opening any checklist in menuconfig.

Added wmove() to set the cursor in the proper position, just like in
menubox.c. Removed wnoutrefresh(dialog) because dialog window has
already been updated in print_buttons(). Replaced wnoutrefresh(list) and
doupdate() calls with one wrefresh(list) call.
Signed-off-by: Matthew Bystrin <dev.mbstr@gmail.com>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

This reverts the following two commits:

  - a555bdd0 ("Kbuild: enable TRIM_UNUSED_KSYMS again, with some guarding")
  - 5cf0fd59 ("Kbuild: disable TRIM_UNUSED_KSYMS option")

Commit 5e9e95cc ("kbuild: implement CONFIG_TRIM_UNUSED_KSYMS without
recursion") solved the build time issue.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Select GENERIC_IRQ_PROBE, as the other architectures do.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Brian Cain <bcain@quicinc.com>

Because FRAME_POINTER is defined in lib/Kconfig.debug, the arch Kconfig
should select it.

Add 'select FRAME_POINTER' to HEXAGON. ARCH_WANT_FRAME_POINTERS must
also be selected to avoid the unmet dependency warning.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Brian Cain <bcain@quicinc.com>

'def_bool X' is a shorthand for 'bool' plus 'default X'.

'def_bool' is redundant where 'bool' is already present, so 'def_bool X'
can be replaced with 'default X', or removed if X is 'n'.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Use helper macros in hashtable.h for generic hashtable implementation.

We can git rid of the hash head index of for_all_symbols().
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

for_all_symbols() iterates in the symbol hash table. The order of
iteration depends on the hash table implementation.

If you use it for printing errors, they are shown in random order.

For example, the order of following test input and the corresponding
error do not match:
 - scripts/kconfig/tests/err_recursive_dep/Kconfig
 - scripts/kconfig/tests/err_recursive_dep/expected_stderr
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Do not feed back the choice type to choice values.

Each choice value should explicitly specify 'bool' or 'tristate',
as all the Kconfig files already do. If the type were missing,
"config symbol defined without type" would be shown.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Currently, a linked list is used to keep track of all the Kconfig
files that have ever been parsed. Every time the "source" statement
is encountered, the linked list is traversed to check if the file has
been opened before. This prevents the same file from being recorded
in include/config/auto.conf.cmd again.

Given 1500+ Kconfig files parsed, a hashtable is now a more optimal
data structure.

By the way, you may wonder why we check this in the first place.
It matters only when the same file is included multiple times.
In old days, such a use case was forbidden, but commit f094f8a1
("kconfig: allow multiple inclusion of the same file") provided a bit
more flexibility. Of course, it is almost hypothetical...
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Remove the 'static' qualifier from strhash() so that it can be accessed
from other files. Move it to util.c, which is a more appropriate location.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

To use ARRAY_SIZE from other files, move it to its own header,
just like include/linux/array_size.h.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

This is similar to include/linux/hashtable.h, but the implementation
has been simplified.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Import more macros and inline functions from include/linux/list.h
and include/linux/types.h.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Update the existing macros and inline functions based on
include/linux/list.h.

The variable name '_new' can be reverted to 'new' because this header
is no longer included from the C++ file, scripts/kconfig/qconf.cc.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

The struct list_head is often embedded in other structures, while other
code is used in C functions.

By separating struct list_head into its own header, other headers are no
longer required to include the entire list.h.

This is similar to the kernel space, where struct list_head is defined
in <linux/types.h> instead of <linux/list.h>.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Currently, file_lookup() returns a pointer to (struct file), but the
callers use only file->name.

Make it return the ->name member directly.

This adjustment encapsulates struct file and file_list as internal
implementation.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Call malloc() just once to allocate needed memory.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

struct file has two link nodes, 'next' and 'parent'.

The former is used to link files in the 'file_list' linked list,
which manages the list of Kconfig files seen so far.

The latter is used to link files in the 'current_file' linked list,
which manages the inclusion ("source") tree.

The latter should be tracked together with the lexer state.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Replace the remaining current_file->name in the lexer context.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Currently, cur_filename is updated at the first token of each statement.
However, this seems unnecessary based on my understanding; the parser
can use the same variable as the lexer tracks.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

The 'file->name' and 'name' are the same in this function.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

struct property is linked to struct file for diagnostic purposes.
It is always used to retrieve the file name through prop->file->name.

Associate struct property with the file name directly.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

struct menu is linked to struct file for diagnostic purposes.
It is always used to retrieve the file name through menu->file->name.

Associate struct menu with the file name directly.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Now zconf_curname() and zconf_lineno() are so simple that they just
return cur_filename, cur_lineno, respectively.

Remove these functions, and then use cur_filename and cur_lineno
directly.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Replace current_pos with separate variables representing the file name
and the line number, respectively.

No functional change is intended.

By the way, you might wonder why the "<none>" fallback exists in
zconf_curname(). menu_add_symbol() saves the current file and the line
number. It is intended to be called only during the yyparse() time.
However, menu_finalize() calls it, where there is no file being parsed.
This is a long-standing hack that should be fixed later. I left a FIXME
comment.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

These are needed only for the parse stage. Move the prototypes into
a separate header to make sure they are not used after that.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

This allows preprocess.c to free up all of its resources when the parse
stage is finished. It also ensures conf_write_autoconf_cmd() produces
consistent results even if called multiple times for any reason.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Currently, include/config/autoconf.cmd saves included Kconfig files in
reverse order. While this is not a big deal, it is inconsistent with
other *.cmd files generated by fixdep.

Output the included Kconfig files in the included order.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

sym_find("n") is equivalent to &symbol_no.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

Commit 1a7a8c6f ("kconfig: allow long lines in config file") added
a self-implemented getline() for better portability.

However, getline() is standardized [1] and already used in other programs
such as scripts/kallsyms.c.

Use getline() provided by libc.

[1]: https://pubs.opengroup.org/onlinepubs/9699919799/functions/getdelim.htmlSigned-off-by: Masahiro Yamada <masahiroy@kernel.org>

There is no definition, no caller for lookup_file().
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

yyerror() reports the line number of the next line.

This +1 adjustment was introduced more than 20 years ago [1]. At that
time, the line number was decremented then incremented back and forth.

The line number management was refactored in a more maintainable way.
Such compensation is no longer needed.

[1]: https://git.kernel.org/pub/scm/linux/kernel/git/history/history.git/commit/?id=d4f8a4530eb07a1385fd17b0e62a7dce97486f49Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

A macro placed at the end of a file with no newline causes an infinite
loop.

[Test Kconfig]
  $(info,hello)
  \ No newline at end of file

I realized that flex-provided input() returns 0 instead of EOF when it
reaches the end of a file.

Fixes: 104daea1 ("kconfig: reference environment variables directly and remove 'option env='")
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

This reverts commit 6ef41e22.

If this is still needed, we can bring it back.

However, I'd like to understand why 'new-kernel-pkg --remove' is
needed for uninstallation, while 'new-kernel-pkg --install' was not
called during the installation.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>

This reverts commit 27c3bffd.

If this is still needed, we can bring it back.

However, I'd like to understand why 'update-bootloader --remove' is
needed for uninstallation, while 'update-bootloader --add' was not
called during the installation.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>