access.2, delete_module.2, eventfd.2, fallocate.2, fcntl.2, getrandom.2, init_module.2, open.2, seccomp.2, timerfd_create.2, openpty.3, pthread_spin_lock.3, shm_open.3, tempnam.3, fifo.7, keyrings.7, pid_namespaces.7, sched.7, thread-keyring.7: wfix (will fail --> fail/fails)

Reported-by: Pedro Alves <palves@redhat.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

getrlimit.2: wfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

capget.2, fanotify_init.2, fcntl.2, ioctl_tty.2, madvise.2, mbind.2, prctl.2, sched_setaffinity.2, hash.3, isgreater.3, posix_fallocate.3, sigvec.3, full.4, sd.4, wavelan.4, core.5, host.conf.5, epoll.7, signal.7, unix.7, user_namespaces.7: wfix (will fail --> fail/fails)

Reported-by: Pedro Alves <palves@redhat.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

reboot.2: wfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

open.2: wfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

times.2: Language fix-up: clarify that "tasks" means "work"

See https://bugzilla.kernel.org/show_bug.cgi?id=197183

Reported-by: Victor Porton <porton@narod.ru>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

times.2: ffix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

veth.4: SHow command used to move a veth device into another network namespace

Reported-by: Tomas Pospisek <tpo@sourcepole.ch>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

veth.4: ffix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

namespaces.7: Add a reference to new veth(4) page

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Add a reference to new veth(4) page

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

veth.4: Elaborate on use of ethtool(8) to find veth peer

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

veth.4: Minor fixes

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

veth.4: Integrate content from Eric Biederman

Cowritten-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

veth.4: SEE ALSO: add ip-link(8)

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

veth.4: New page document veth virtual ethernet devices

Based on a page from Tomáš Pospíšek, with some clean-ups by mtk.

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_init.3: srcfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_init.3: Minor rewording

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_init.3: Add more caveats on use of spin locks

Cowritten-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_init.3: Strengthen note about when spin locks should (not) be used

Reported-by: Zack Weinberg <zackw@panix.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_init.3: Operations on copies of *lock yield undefined behavior

Reported-by: Florian Weimer <fweimer@redhat.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_init.3: tfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_lock.3: tfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_lock.3: wfix

Reported-by: Zack Weinberg <zackw@panix.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_lock.3: Fixes suggested by Carlos O'Donell

Reported-by: Carlos O'Donell <carlos@redhat.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_lock.3: Refer reader to important details in NOTES in pthread_spin_init(3)

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_init.3: Warn about deadlock with spin locks

Reported-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_init.3: Spin locks should be used with real-time scheduling policies

Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_yield.3: srcfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthreads.7: SEE ALSO: add pthread_spin_init(3) and pthread_spin_lock(3)

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_trylock.3, pthread_spin_unlock.3: New links to new pthread_spin_lock3 page

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_lock.3: New page describing functions that lock and unlock spin locks

Add a page descrbing pthread_spin_lock(3), pthread_spin_unlock(3),
and pthread_spin_trylock(3).

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_destroy.3: New link to new pthread_spin_init.3 page

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_spin_init.3: New page describing pthread_spin_init(3) and pthread_spin_destroy(3)

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

stat.2: Use lstat() instead of stat()

It's more logical to use lstat() in the example code,
since one can then experiment with sybolic links, and
also the S_IFLNK case can also occur.

Reported-by: Richard Knutsson <richard.knutsson@abelko.se>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

ioctl_userfaultfd.2: wfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

ioctl_userfaultfd.2: Rework version information for feature bits

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

userfaultfd.2: Minor fixes to Prakash Sangappa's patch

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

ioctl_userfaultfd.2: Minor fixes to Prakash Sangappa's patch

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

ioctl_userfaultfd.2, userfaultfd.2: Add description for UFFD_FEATURE_SIGBUS

Userfaultfd feature UFFD_FEATURE_SIGBUS was merged recently and
should be available in the Linux 4.14 release. This patch is for
the man page changes documenting this API.

Documents the following commit:

commit 2d6d6f5a09a96cc1fec7ed992b825e05f64cb50e
Author: Prakash Sangappa <prakash.sangappa@oracle.com>
Date: Wed Sep 6 16:23:39 2017 -0700

     mm: userfaultfd: add feature to request for a signal delivery

Reviewed-by: Andrea Arcangeli <aarcange@redhat.com>
Reviewed-by: Mike Rapoport <rppt@linux.vnet.ibm.com>
Signed-off-by: Prakash Sangappa <prakash.sangappa@oracle.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_yield.3: tfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

namespaces.7: EXAMPLE: refer also to example in clone(2)

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

posix_fallocate.3: ffix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

netlink.7: ffix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

man-pages.7: Note that "x86-64" is generally preferred over "x86_64"

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

proc.5: wfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2, getcpu.2, shmget.2, syscall.2, dladdr.3, proc.5, vdso.7: Consistently use "x86-64", not "x86_64"

When referring to the architecture, consistently use "x86-64",
not "x86_64". Hitherto, there was a mixture of usages, with
"x86-64" predominant.

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Combine redundant paragraphs describing child_stack==NULL

Combine two redundant paragraphs (one of which I recently
added) describing child_stack==NULL for the raw system call.
Also, make sure this text is in a more obvious place than
its previous location.

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Rework the discussion of the historical CLONE_PID for clarity

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: ffix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Remove a redundant paragraph

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Eliminate some redundant phrasing in discussion of "fn()"

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Minor wording fixes

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Warn that the clone() wrapper modifies child_stack in the parent

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Add NOTES heading

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Note that child_stack can be NULL when using the raw system call

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: Minor wording fix (in preparation for subsequent patch)

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

clone.2: tfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

ld.so.8: Simplify language around conferring capabilities

The statement "conferring permitted or effective capabilities"
to the process is somewhat redundant. Binaries with capabilities
confer capabilities only to those process capability sets, so it's
simpler to just say"confers capabilities to the process".

Reported-by: Yubin Ruan <ablacktshirt@gmail.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

core.5: ffix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

core.5: tfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

core.5: Adds some notes on systemd and core dumps

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

core.5: Dumps are not produced if core_pattern is empty and core_uses_pid is 0

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

core.5: SEE ALSO: add coredumpctl(1)

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

core.5: RLIMIT_CORE is not enforced when piping core dump to a program

Reported-by: Per Böhlin <Per.Bohlin@zenterio.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

madvise.2: Note fork() and execve() semantics for wipe-on-fork setting

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

fork.2: The MADV_WIPEONFORK setting is preserved in child of fork()

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

fork.2: Minor reworking of MADV_WIPEONFORK text

The wipe-on-fork setting is specified on address ranges,
not mappings.

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

getrlimit.2: Make it clear RLIMIT_NPROC is a limit on current number of processes

https://twitter.com/silentbicycle/status/893849097903505409

Reported-by: Scott Vokes <vokes.s@gmail.com>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

getrlimit.2: Minor rewording

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

proc.5: tfix

Remove duplicated word.

Signed-off-by: Jakub Wilk <jwilk@jwilk.net>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

epoll.7: Note existence of kcmp() KCMP_EPOLL_TFD operation

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

shmop.2: Note that return value of shmat() is page-aligned

At the current man page for shmat(2)[1], there is no mentioning
whether the returned memory address of shmat(2) will be page size
aligned or not. As that is quite important to many applications(e.g.,
those that use locks heavily and would like to avoid some locks by
some atomic guarantees provided by the CPU), it would be great to
specify that for Linux.

I walked down the current implementation of shmat(2) in the latest
kernel src and found that shmat(2) does return a page size aligned
memory address:

SYSCALL_DEFINE3(shmat, int, shmid, char __user *, shmaddr, int, shmflg)
 -> do_shmat(...)
 -> do_mmap_pgoff(...)
 -> do_mmap(...)
 -> get_unmapped_area(...)
 -> get_area(...) -> offset_in_page(addr)

there is a `offset_in_page(addr)' assertion at the end and if that is
true a -EINVAL would be returned, by which we can be sure that
shmat(2) will return a page size aligned memory address on success[2].

[1]: http://man7.org/linux/man-pages/man2/shmat.2.html
[2]: there is also a `offset_in_page(2)' in get_unmapped_area(...),
but that doesn't lead to -EINVAL...I am not sure whether the logic of
that code is right.

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

shmop.2: Minor wording fix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

abort.3: SEE ALSO: add assert(3)

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

abort.3: Starting with glibc 2.27, abort() does not attempt to flush streams

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sigaction.2: Mention that libc sets the act.sa_restorer field

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sigaction.2: wfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sigaction.2: wfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sigreturn.2: Minor fixes to description of sa_restorer field

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sigreturn.2: Minor wording fix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sigreturn.2: glibc has a simple wrapper for sigreturn() that returns ENOSYS

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pow10.3: tfix

Signed-off-by: Fangrui Song <i@maskray.me>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sigaction.2: Rework discussion of SA_SIGINFO handler arguments

Expand and rework the text a little, in particular adding
a reference to sigreturn(2) as a source of further
information about the ucontext argument.

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sigreturn.2: Make it a little clearer that a stack frame is created by the kernel

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sysfs.5: tfix

Signed-off-by: Jakub Wilk <jwilk@jwilk.net>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

statfs.2: tfix

Signed-off-by: Jakub Wilk <jwilk@jwilk.net>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

mallinfo.3: Fix the example

Remove reference to non-standard "tlpi_hdr.h" and replace calls to
functions that were declared in this header.

Signed-off-by: Jakub Wilk <jwilk@jwilk.net>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

flock.2: Placing an exclusive lock over NFS requires the file is open for writing

Reported-by: Petr Uzel <petr.uzel@suse.cz>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

flock.2: Minor wording fix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

flock.2: Break NFS details int two separate paragraphs

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

flock.2: Move NFS details to a headed subsection

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

setlocale.3: The standards do not specify all of the locale categories

Reported-by: Křištof Želechovski <giecrilj@stegny.2a.pl>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

prctl.2: wfix

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

splice.2: Since Linux 2.6.31,'fd_in' and 'fd_out' may both refer to pipes

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

mremap.2: Update timestamp: review of new BUGS by Michal Hocko and Mike Kravetz

Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Reviewed-by: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

pthread_yield.3: pthread_yield() is intended for use with real-time scheduling policies

Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

sched_yield.2: sched_yield() is intended for use with real-time scheduling policies

Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

backtrace.3: Minor tweaks to Stefan Puiu's patch

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>

backtrace_symbols_fd() can trigger a call to malloc()

Currently, the backtrace(3) manual page says this about backtrace_symbols_fd():

back‐
       trace_symbols_fd() does not call malloc(3), and so can be
employed in situations where the latter function might fail.

However, I watched a video of a presentation about signal handling and
the speaker was saying that calling backtrace() can trigger a call to
malloc - indirectly. That happens because the backtrace*() functions
are part of libgcc, which gets dynamically loaded whenever needed;
dynamic loading would, in turn, trigger a malloc. The talk can be
found here: http://free-electrons.com/pub/video/2008/ols/ols2008-gilad-ben-yossef-fault-handlers.ogg

I decided to test it out, and it seems that this is still true (at
least on Ubuntu 12.04). I compiled the attached program (I used
CXXFLAGS = '-Wall -g -std=c++0x', the -std= part is not really needed)
and ran it through gdb, putting a breakpoint on the line where
backtrace is called. When that breakpoint is hit, I set a breakpoint
on malloc, continued and voila:

Breakpoint 2, __GI___libc_malloc (bytes=36) at malloc.c:2910
2910    malloc.c: No such file or directory.
(gdb) bt
"/lib/x86_64-linux-gnu/libgcc_s.so.1") at dl-load.c:162
"libgcc_s.so.1", type=2, trace_mode=0, mode=-1879048191,
nsid=<optimized out>) at dl-load.c:2473
errstring=0x7fffffffdd00, mallocedp=0x7fffffffdd0f,
operate=0x7ffff7ded700 <dl_open_worker>, args=0x7fffffffdcb0) at
dl-error.c:178
"libgcc_s.so.1", mode=-2147483647, caller_dlopen=0x7ffff7b260a9,
nsid=-2, argc=1, argv=<optimized out>, env=0x7fffffffe068) at
dl-open.c:639
errstring=0x7fffffffded0, mallocedp=0x7fffffffdeef,
operate=0x7ffff7b4bc20 <do_dlopen>, args=0x7fffffffdeb0) at
dl-error.c:178
operate=0x7ffff7b4bc20 <do_dlopen>) at dl-libc.c:48
out>) at dl-libc.c:165
../sysdeps/x86_64/../ia64/backtrace.c:104

Reviewed-by: Walter Harms <wharms@bfs.de>
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>