man/man3/stpncpy.3: EXAMPLES: Remove 'SRC' BEGIN/END comments

This example is incomplete, and won't build.

Fixes: 10f9e349c888 (2026-08-03; "man/man3/stpncpy.3: EXAMPLES: Use a more realistic example")
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/stpncpy.3: EXAMPLES: Use a more realistic example

Cc: Keith Bostic <keith@bostic.com>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Sam James <sam@gentoo.org>
Cc: Joseph Myers <josmyers@redhat.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: Nevin Liber <nevin@cplusplusguy.com>
Cc: Collin Funk <collin.funk1@gmail.com>
Cc: JeanHeyd Meneide <phdofthehouse@gmail.com>
Cc: Christopher Bazley <chris.bazley.wg14@gmail.com>
Cc: Bruno Haible <bruno@clisp.org>
Cc: Paul Eggert <eggert@cs.ucla.edu>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Iker Pedrosa <ipedrosa@redhat.com>
Cc: Evgeny Grin <k2k@drgrin.dev>
Cc: Kees Cook <keescook@chromium.org>
Cc: Steve Summit <scs@eskimo.com>
Cc: "Maciej W. Rozycki" <macro@orcam.me.uk>
Cc: Arsen Arsenović <arsen@aarsen.me>
Cc: bug-gnulib@gnu.org
Cc: libc-alpha@sourceware.org
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/stpncpy.3: CAVEATS: Remove unreal recommendation

This is not a use case that happens enough to recomment anything.
If one encounters such a use case, they're probably deep enough
that they know what to do.  Casual users should keep it simple.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/memccpy.3: CAVEATS: Document that this function is not appropriate for strings

Cc: Keith Bostic <keith@bostic.com>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Sam James <sam@gentoo.org>
Cc: Joseph Myers <josmyers@redhat.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: Nevin Liber <nevin@cplusplusguy.com>
Cc: Collin Funk <collin.funk1@gmail.com>
Cc: JeanHeyd Meneide <phdofthehouse@gmail.com>
Cc: Christopher Bazley <chris.bazley.wg14@gmail.com>
Cc: Bruno Haible <bruno@clisp.org>
Cc: Paul Eggert <eggert@cs.ucla.edu>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Iker Pedrosa <ipedrosa@redhat.com>
Cc: Evgeny Grin <k2k@drgrin.dev>
Cc: Kees Cook <keescook@chromium.org>
Cc: Steve Summit <scs@eskimo.com>
Cc: "Maciej W. Rozycki" <macro@orcam.me.uk>
Cc: Arsen Arsenović <arsen@aarsen.me>
Cc: bug-gnulib@gnu.org
Cc: libc-alpha@sourceware.org
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/strncat.3: CAVEATS: Clarify that this function doesn't truncate the input nonstring

Cc: Keith Bostic <keith@bostic.com>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Sam James <sam@gentoo.org>
Cc: Joseph Myers <josmyers@redhat.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: Nevin Liber <nevin@cplusplusguy.com>
Cc: Collin Funk <collin.funk1@gmail.com>
Cc: JeanHeyd Meneide <phdofthehouse@gmail.com>
Cc: Christopher Bazley <chris.bazley.wg14@gmail.com>
Cc: Bruno Haible <bruno@clisp.org>
Cc: Paul Eggert <eggert@cs.ucla.edu>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Iker Pedrosa <ipedrosa@redhat.com>
Cc: Evgeny Grin <k2k@drgrin.dev>
Cc: Kees Cook <keescook@chromium.org>
Cc: Steve Summit <scs@eskimo.com>
Cc: "Maciej W. Rozycki" <macro@orcam.me.uk>
Cc: Arsen Arsenović <arsen@aarsen.me>
Cc: bug-gnulib@gnu.org
Cc: libc-alpha@sourceware.org
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Remove references to 'Shlemiel the painter'

Compilers are able to optimize this out.

Programmers should aim for simplicity, and leave this optimization to
the compiler.

Cc: Keith Bostic <keith@bostic.com>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Sam James <sam@gentoo.org>
Cc: Joseph Myers <josmyers@redhat.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: Nevin Liber <nevin@cplusplusguy.com>
Cc: Collin Funk <collin.funk1@gmail.com>
Cc: JeanHeyd Meneide <phdofthehouse@gmail.com>
Cc: Christopher Bazley <chris.bazley.wg14@gmail.com>
Cc: Bruno Haible <bruno@clisp.org>
Cc: Paul Eggert <eggert@cs.ucla.edu>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Iker Pedrosa <ipedrosa@redhat.com>
Cc: Evgeny Grin <k2k@drgrin.dev>
Cc: Kees Cook <keescook@chromium.org>
Cc: Steve Summit <scs@eskimo.com>
Cc: "Maciej W. Rozycki" <macro@orcam.me.uk>
Cc: Arsen Arsenović <arsen@aarsen.me>
Cc: bug-gnulib@gnu.org
Cc: libc-alpha@sourceware.org
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/standards.7: Specify SVr1 for System V release 1

Cc: Keith Bostic <keith@bostic.com>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Sam James <sam@gentoo.org>
Cc: Joseph Myers <josmyers@redhat.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: Nevin Liber <nevin@cplusplusguy.com>
Cc: Collin Funk <collin.funk1@gmail.com>
Cc: JeanHeyd Meneide <phdofthehouse@gmail.com>
Cc: Christopher Bazley <chris.bazley.wg14@gmail.com>
Cc: Bruno Haible <bruno@clisp.org>
Cc: Paul Eggert <eggert@cs.ucla.edu>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Iker Pedrosa <ipedrosa@redhat.com>
Cc: Evgeny Grin <k2k@drgrin.dev>
Cc: Kees Cook <keescook@chromium.org>
Cc: Steve Summit <scs@eskimo.com>
Cc: "Maciej W. Rozycki" <macro@orcam.me.uk>
Cc: Arsen Arsenović <arsen@aarsen.me>
Cc: bug-gnulib@gnu.org
Cc: libc-alpha@sourceware.org
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3head/memory.h.3: STANDARDS, HISTORY: Document some history about <string.h> and <memory.h>

This text is an adaptation of text suggested by Keith and Branden.

Suggested-by: Keith Bostic <keith@bostic.com>
Suggested-by: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Sam James <sam@gentoo.org>
Cc: Joseph Myers <josmyers@redhat.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: Nevin Liber <nevin@cplusplusguy.com>
Cc: Collin Funk <collin.funk1@gmail.com>
Cc: JeanHeyd Meneide <phdofthehouse@gmail.com>
Cc: Christopher Bazley <chris.bazley.wg14@gmail.com>
Cc: Bruno Haible <bruno@clisp.org>
Cc: Paul Eggert <eggert@cs.ucla.edu>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Iker Pedrosa <ipedrosa@redhat.com>
Cc: Evgeny Grin <k2k@drgrin.dev>
Cc: Kees Cook <keescook@chromium.org>
Cc: Steve Summit <scs@eskimo.com>
Cc: "Maciej W. Rozycki" <macro@orcam.me.uk>
Cc: Arsen Arsenović <arsen@aarsen.me>
Cc: bug-gnulib@gnu.org
Cc: libc-alpha@sourceware.org
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man*/{string.3,memory.h.3head}: Move functions to a new page memory.h(3head)

See previous commit.

Cc: Keith Bostic <keith@bostic.com>
Cc: Sam James <sam@gentoo.org>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Joseph Myers <josmyers@redhat.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: Nevin Liber <nevin@cplusplusguy.com>
Cc: Collin Funk <collin.funk1@gmail.com>
Cc: JeanHeyd Meneide <phdofthehouse@gmail.com>
Cc: Christopher Bazley <chris.bazley.wg14@gmail.com>
Cc: Bruno Haible <bruno@clisp.org>
Cc: Paul Eggert <eggert@cs.ucla.edu>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Iker Pedrosa <ipedrosa@redhat.com>
Cc: Evgeny Grin <k2k@drgrin.dev>
Cc: Kees Cook <keescook@chromium.org>
Cc: Steve Summit <scs@eskimo.com>
Cc: "Maciej W. Rozycki" <macro@orcam.me.uk>
Cc: Arsen Arsenović <arsen@aarsen.me>
Cc: bug-gnulib@gnu.org
Cc: libc-alpha@sourceware.org
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/: SYNOPSIS: Document {mem,strn}*() functions as provided by <memory.h>

They've been provided there since forever, and that's a more appropriate
header.

The standard mixes functions for handling strings, functions for
handling bytes, and other hybrids, all in a single header file:
<string.h>.

This has historically caused confusion, for example leading to believe
that strncpy(3) is appropriate to handle strings.

It turns out that mem*() functions haven't been always in <string.h>.
These functions were first introduced in 1983 in System V, and were
added in a separate <memory.h> file.  Very soon, they were added to
4.3BSD in 1986 for compatibility to System V, also in <memory.h>.

Most modern-day POSIX-compatible libc implementations still provide this
header file for backwards compatibility reasons:

$ find ~/src/bsd/freebsd/main/ | grep /include/memory.h
/home/alx/src/bsd/freebsd/main/include/memory.h
$ find ~/src/bsd/netbsd/trunk/ | grep /include/memory.h
/home/alx/src/bsd/netbsd/trunk/include/memory.h
$ find ~/src/bsd/openbsd/master/ | grep /include/memory.h
/home/alx/src/bsd/openbsd/master/include/memory.h
$ find ~/src/gnu/glibc/master/ | grep /include/memory.h
/home/alx/src/gnu/glibc/master/include/memory.h
$ find ~/src/musl/libc/master/ | grep /include/memory.h
/home/alx/src/musl/libc/master/include/memory.h

It seems that it was C89 that moved the mem*() functions to <string.h>,
even though the Rationale document for C89 doesn't mention it at all.

This was a huge mistake --we can see the consequences in the many
misuses of these functions, and programmers that believe they are
appropriate for uses for which they are unappropriate-- that we try
to undo now.

Let's document these as provided by <memory.h>, which is still true;
all POSIX-ish libc implementations I've consulted provide the <memory.h>
header with mem*() functions, except two: Illumos gate, and gnulib.
Let's ignore them, or rather, hope they will follow suit eventually.

This will signal to programmers that this is the header file they should
be using.  The C Committee is discussing the possibility of
standardizing the good old <memory.h> header file, so let's give it a
little bump.

Let's also move the strn*() functions there, since modern libc
implementations also provide them there.  And they certainly don't
belong in <string.h>, so <memory.h> is more adequate.

After this change, <string.h> is documented only in the SYNOPSIS of
functions that exclusively handle (and require) strings.

Link: <https://lore.kernel.org/linux-man/am0RdPxvbIYUKAL-@devuan/T/#t>
Cc: Keith Bostic <keith@bostic.com>
Cc: Sam James <sam@gentoo.org>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Joseph Myers <josmyers@redhat.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: Nevin Liber <nevin@cplusplusguy.com>
Cc: Collin Funk <collin.funk1@gmail.com>
Cc: JeanHeyd Meneide <phdofthehouse@gmail.com>
Cc: Christopher Bazley <chris.bazley.wg14@gmail.com>
Cc: Bruno Haible <bruno@clisp.org>
Cc: Paul Eggert <eggert@cs.ucla.edu>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Iker Pedrosa <ipedrosa@redhat.com>
Cc: Evgeny Grin <k2k@drgrin.dev>
Cc: Kees Cook <keescook@chromium.org>
Cc: Steve Summit <scs@eskimo.com>
Cc: "Maciej W. Rozycki" <macro@orcam.me.uk>
Cc: Arsen Arsenović <arsen@aarsen.me>
Cc: bug-gnulib@gnu.org
Cc: libc-alpha@sourceware.org
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man8/ldconfig.8: Add tunables options

Message-ID: <d85795998b72ab686a683f13d08f24d98d8c1a51.1785522144.git.dj@redhat.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man8/ld.so.8: Note that ld.so.cache includes tunables

Message-ID: <bcaddcd0fb8f60276973ae28f091c158332e20c6.1785522144.git.dj@redhat.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man5/ld.so.conf.5: Document 'include' syntax

Message-ID: <ae9c2cc67047902f68f64d05ca454da0e6d008b3.1785522144.git.dj@redhat.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man4/console_codes.4: Document more xterm codes

Signed-off-by: Jan Engelhardt <jengelh@inai.de>
Message-ID: <20260718153631.55879-1-jengelh@inai.de>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/[b]string.3: Rewrite and merge

Let's have a single page with an organized list of functions.

Cc: "Serge E. Hallyn" <serge@hallyn.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Douglas McIlroy <douglas.mcilroy@dartmouth.edu>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/mem*.3: NAME: Explain the names

Reported-by: "Serge E. Hallyn" <serge@hallyn.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Douglas McIlroy <douglas.mcilroy@dartmouth.edu>
Acked-by: Serge Hallyn <sergeh@kernel.org>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/strpbrk.3: BUGS: Clarify the NAME

Cc: Mark Harris <mark.hsj@gmail.com>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Douglas McIlroy <douglas.mcilroy@dartmouth.edu>
Acked-by: Serge Hallyn <sergeh@kernel.org>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/str*.3: NAME: Explain the names

Reported-by: "Serge E. Hallyn" <serge@hallyn.com>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: Douglas McIlroy <douglas.mcilroy@dartmouth.edu>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man4/console_codes.4: Mention caveats for xterm(1)'s ESC ] 50 command

Fonts are only settable if xterm(1) was launched with the right options
(or X11 resources) that would allow this operation.

Signed-off-by: Jan Engelhardt <jengelh@inai.de>
Message-ID: <20260710221035.26529-1-jengelh@inai.de>
[alx: wfix]
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/landlock.7: wfix

Use the current wording of the equivalent paragraph in
"linux.git/Documentation/userspace-api/landlock.rst".

Signed-off-by: Štěpán Němec <stepnem@smrk.net>
Message-ID: <20260713114855Z.2240445002-stepnem@smrk.net>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/memfd_secret.2: tfix

Fixes: ac5edfeb1dd6 (2021-08-27; "memfd_secret.2: New page describing memfd_secret() system call")
Cc: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Štěpán Němec <stepnem@smrk.net>
Message-ID: <20260713121111.372487-1-stepnem@smrk.net>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man5/proc_pid_fdinfo.5: Document that 'ino' and 'mnt_id' are decimal numbers

Signed-off-by: Maja Kądziołka <maya@compilercrim.es>
Cc: Kalesh Singh <kaleshsingh@google.com>
Message-ID: <7eeaa92f7a7cc14024a6474d4cc117e875185f6d.1784211190.git.maya@compilercrim.es>
[alx: split commit]
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man5/proc_pid_fdinfo.5: Document the 'ino' field

Also expands the tabs into spaces where relevant, in order to ensure
proper alignment between the fields when rendered by man.

Signed-off-by: Maja Kądziołka <maya@compilercrim.es>
Cc: Kalesh Singh <kaleshsingh@google.com>
Message-ID: <7eeaa92f7a7cc14024a6474d4cc117e875185f6d.1784211190.git.maya@compilercrim.es>
[alx: split commit]
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man?/intro.[01]: Split part of intro(1) into a new intro(0)

Let's keep intro(1) for introducing just the man1 section, and move the
more general introduction of the entire manual to a new intro(0).

Acked-by: Douglas McIlroy <douglas.mcilroy@dartmouth.edu>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man0: Add symlink

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man8/ld.so.8: ffix

Fixes: cce97466d5ed (2023-10-18; "ld.so.8: Describe the --list-diagnostics loader option")
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: srcfix

Add missing closing EE for EX.

Reported-by: DJ Delorie <dj@redhat.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man9/strscpy.9: NAME: Clarify both etymology and actual semantics

The function is called strscpy() with an 's' for 'safe', but it
should actually be called strtcpy() with a 't' for 'truncation'.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/strlcat.3: Add link page

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/strlcpy.3: Add page

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man9: Add symlink

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man9/strscpy.9: Add page

This function is well-known, and while it isn't available in
user space, it's good to have documentation for it, so that
users that want to talk about it can refer to this manual page.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man8/ld*.8: Refer to ld.so.conf(5)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man5/ld.so.conf.5: Add page

Cc: DJ Delorie <dj@redhat.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/pipe.7: Note kernel commit of the soft limit triggered cap increase

7543e84442d7 did reference the kernel commit in its commit
message, but the convention to include it in the man page
source as well is helpful to wider audience.

Fixes: 7543e84442d7 (2024-08-29; "pipe.7: Document change to default pipe size when soft limit is exceeded")
Signed-off-by: Štěpán Němec <stepnem@smrk.net>
Message-ID: <20260709153422.210312-2-stepnem@smrk.net>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man4/console_codes.4: Rephrase description for xterm's ESC ] 10 command

The term "dynamic" text color is too unspecific.
Since xterm has more codes for tuning colors, all colors
could be considered "dynamic".

Signed-off-by: Jan Engelhardt <jengelh@inai.de>
Message-ID: <20260708121458.46465-2-jengelh@inai.de>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man4/console_codes.4: Rephrase description for xterm's ESC ] 4 command

The \e]4 code is capable of modifying xterm's entire 256-slot palette,
not just the ANSI sub-portion thereof.

Signed-off-by: Jan Engelhardt <jengelh@inai.de>
Message-ID: <20260708121458.46465-1-jengelh@inai.de>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man4/console_codes.4: wsfix

The space character is used as a visual aid in the escape code
sequences.  A few lines had spacing inconsistent with the rest of the
document, so make them consistent.

Signed-off-by: Jan Engelhardt <jengelh@inai.de>
Message-ID: <20260708104303.39708-1-jengelh@inai.de>
[alx: tweak commit message]
Signed-off-by: Alejandro Colomar <alx@kernel.org>

share/mk/: check-catman-grep: Ignore diagnostics in landlock.7

An example is too long, and we can't make it shorter without reducing
readability.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/landlock.7: Document LANDLOCK_ACCESS_FS_RESOLVE_UNIX (ABI v9)

Document the new LANDLOCK_ACCESS_FS_RESOLVE_UNIX filesystem access right,
which controls lookups of pathname UNIX domain sockets.  Restricts both
connect(2) and sendmsg(2) with an explicit recipient address to UNIX
sockets created outside the Landlock domain (same semantics as
LANDLOCK_SCOPE_* flags).  Denied attempts return EACCES.

Available since Linux 7.1 (Landlock ABI version 9).

Signed-off-by: Günther Noack <gnoack3000@gmail.com>
Message-ID: <20260515165753.8830-4-gnoack3000@gmail.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Make sashimi of termios(3)

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Update old termios(3) references

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man?/termios.?: Move termios(3) to man7

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/cfsetspeed.3: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/termios.3, man/man3/cf{g,s}et{,i,o}speed.3: Split cf{g,s}et{,i,o}speed() from termios(3)

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/termios.3, man/man3/tcflow.3: Split tcflow() from termios(3)

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/termios.3, man/man3/tcflush.3: Split tcflush() from termios(3)

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/termios.3, man/man3/tcdrain.3: Split tcdrain() from termios(3)

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/tcsendbreak.3: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/termios.3, man/man3/tcsendbreak.3: Split tcsendbreak() from termios(3)

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/cfmakeraw.3: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/termios.3, man/man3/cfmakeraw.3: Split cfmakeraw() from termios(3)

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/tcsetattr.3: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/termios.3, man/man3/tc[gs]etattr.3: Split tc[gs]etattr() from termios(3)

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3type/termios.3type: Tweak after movement

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3{,type}/termios.3{,type}: Move 'Canonical and noncanonical mode' to termios(3type)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3type/termios.3type: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/termios.3, man/man3type/termios.3type: Split struct termios from termios(3)

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/openat2.2: Document OPENAT2_REGULAR flag

Signed-off-by: Dorjoy Chowdhury <dorjoychy111@gmail.com>
Message-ID: <20260426111707.36541-3-dorjoychy111@gmail.com>
[alx: wfix]
Cc: Florian Weimer <fweimer@redhat.com>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

.gitignore: Remove checkpatch stuff

I forgot to remove this in the previous commit.

Fixes: e2d5d5ca65f9 (2026-07-01; "*: lint-c-checkpatch: Remove support for checkpatch")
Signed-off-by: Alejandro Colomar <alx@kernel.org>

*: lint-c-checkpatch: Remove support for checkpatch

It is not packaged by any distros, and the effort of separating it from
the kernel repository and packaging it seems not worth it, because most
of the interesting stuff it finds is also found by other linters.

This solves an issue in the output of 'make help-list-build-depends',
which is supposed to list output as Debian package/program, but
checkpatch was not following that.

Closes: <https://bugzilla.kernel.org/show_bug.cgi?id=221712>
Reported-by: Jason Yundt <jason@jasonyundt.email>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/termios.3: setserial(8) rate hacking is dangerous

The use of setserial(8) to modify the 38400 baud rate was an early
hack to support baud rates beyond those standardized by POSIX.  As it
is, it is a real trap for applications that actually want to use 38400
bps if a previous user has not cleaned up the setting.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Message-ID: <20260629135910.143781-3-hpa@zytor.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: ftm's are not really additive

Some times (e.g., _ISOC11_SOURCE and gets(3)), these are not additive.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: _XOPEN_SOURCE the same as _POSIX_C_SOURCE

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: Several improvements and simplifications

man/man7/feature_test_macros.7: _REENTRANT, _THREAD_SAFE: Mark as [[deprecated]]

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: _BSD_SOURCE, _SVID_SOURCE: Use consistent [[deprecated]] notation

The glibc version is unimportant here.  It's still there in the
description, though.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: _LARGEFILE_SOURCE: Mark as [[deprecated]]

Also, clarify that these functions are now part of SUSv2.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: _LARGEFILE64_SOURCE: Mark as [[deprecated]]

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: _ISOCxx_SOURCE: Group and improve the macros for ISO C versions

Also, document that compiler flags should be used intead.

Cc: Joseph Myers <josmyers@redhat.com>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/: Clarify the change in C95 in standards(7), not feature_test_macros(7)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: _XOPEN_SOURCE: Simplify specification

Make the meaning of the numbers clear: they refer to the "Issue"
that corresponds to the SUS version.

Also, make it clear that it's POSIX plus the XSI extension.  That's
the essential difference between _POSIX_C_SOURCE and _XOPEN_SOURCE.

Also, simplify, by not documenting ranges.  Users should ask for the
right number.  If they don't, the behavior is really undefined; really!
If a standard came out and claimed to be SUSv3.3, with a value 567, and
provided slightly different definitions, programs would start breaking
if they relied on >=500 <600 being the same as 500.

Also, requesting one version of the XSI extensions and a different
version of the POSIX base specification --or doing any other weird
combinations-- is undefined, so let's not document that at all.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: _XOPEN_SOURCE_EXTENDED: This deprecated macro enables SUSv1

The fact that it's implied by _XOPEN_SOURCE >= 500 is obvious, because
SUSv2 is essentially a superset of SUSv1 (I don't remember if there was
any removal).

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: _POSIX_SOURCE: Simplify the documentation of this deprecated ftm

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man7/feature_test_macros.7: _POSIX_C_SOURCE: The first standards were cummulative, but later ones are exclusive

The modern POSIX standards supersede previous ones, and in some cases
remove definitions.  Definitions from previous standards are not
guaranteed to be provided by more recent ones.  I've rewritten this
section to reflect this, and I've also simplified the wording and
formatting to make this more clear.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

README: Document README.d/*

Reported-by: Jason Yundt <jason@jasonyundt.email>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/IPV6_V6ONLY.2const: $2 is IPPROTO_IPV6, not IPPROTO_IP

Fixes: 07d5d5b62a8f (2025-12-02; "man/man2const/{IPPROTO_IPV6,IPV6_V6ONLY}.2const: Split IPV6_V6ONLY from IPPROTO_IPV6(2const)")
Reported-by: Phosit <phosit@autistici.org>
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: lseek64(3) and off64_t are deprecated

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/getdents.2: Use loff_t in linux_dirent64

off64_t is a glibc type, but struct linux_dirent64 is a kernel type.
loff_t is the kernel equivalent of off64_t.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Remove some references to deprecated interfaces

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3type/off_t.3type: off64_t is deprecated

... because _LARGEFILE64_SOURCE is deprecated.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man3/lseek64.3: This function is deprecated

... because _LARGEFILE64_SOURCE is deprecated.

Reduce the page to just document that the function is deprecated,
and point out that lseek(2) should be used instead.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/_llseek.2: SYNOPSIS: There *is* a wrapper

lseek(2) wraps this system call, if _FILE_OFFSET_BITS=64 is defined.

lseek64(3) also wraps this system call, but it is deprecated.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: srcfix

Fixes: 78a404cb8878 (2026-06-06; "man/man2/{clone,clone2,__clone2}.2: Split [__]clone2() from clone(2)")
Fixes: 2f06a0f7ef3c (2026-06-06; "man3/{llseek,lseek64}.3: Split llseek() from lseek64(3)")
Fixes: 13204ad03485 (2026-06-06; "man3/llseek.3: Tweak after split")
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Re-organize lseek()-like APIs

It was a bit of a mess.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Refer to _llseek(2), not llseek(2)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/[_]llseek.2: Change which is the page and which is the link page

The system call is __NR__llseek, and thus _llseek.  Make _llseek.2 be
the actual page.  Since the system call is called sys_llseek within the
kernel, keep llseek.2 as a link page.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/llseek.2: Update reference to llseek(3)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man3/llseek.3: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man3/{llseek,lseek64}.3: Split llseek() from lseek64(3)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/{clone,clone2,__clone2}.2: Split [__]clone2() from clone(2)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Update references to _exit(3) and _Exit(3)

I missed this in the previous commit, due to the formatting.  I used
a regex that didn't find these.

Fixes: 2d6c48b1 (2026-06-05; "man/: Update references to exit(2) and _exit(3)")
Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Update references to exit(2) and _exit(3)

See the previous commit, which splitted _exit(2) into exit(2) and
_exit(3)/_Exit(3).

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man?/{_exit,_Exit,exit}.?: Split _exit(3)/_Exit(3) and exit(2)

These are two different APIs with different name and different behavior.
They need different pages, to minimize confusion.

There's the raw system call, whose name is exit, not _exit.  It's
SYS_exit, and not SYS__exit.  See for example that the macro for
_llseek(2) is SYS__llseek.  Let's rename this page to exit(2).  There
was already a link page exit(2), so use that one, and remove _exit(2).

Then, there's the libc wrapper, whose name is _exit().  That should be
documented in man3 since it has no syscall of the same name, and
SYS_exit which resembles the name has different behavior.  _Exit() is
another library name, and should also be in man3.  These are now moved
to _exit(3) and _Exit(3) --a link page--.

The text for the two new pages is a reorganization of the old page,
moving the relevant text to each of these.

The next commit will update references to these pages from other pages.

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/: Make sashimi of quotactl(2)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/quotactl.2: RETURN VALUE: quotactl_fd() returns the same as quotactl()

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/Q_XQUOTASYNC.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/quotactl.2, man/man2const/Q_XQUOTASYNC.2const: Split Q_XQUOTASYNC from quotactl(2)

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2const/Q_XQUOTARM.2const: Tweak after split

Signed-off-by: Alejandro Colomar <alx@kernel.org>

man/man2/quotactl.2, man/man2const/Q_XQUOTARM.2const: Split Q_XQUOTARM from quotactl(2)

Signed-off-by: Alejandro Colomar <alx@kernel.org>