From cac193ff175e248fbc94aaed775c4b59d4c3a248 Mon Sep 17 00:00:00 2001 From: Michael Kerrisk Date: Wed, 29 Oct 2008 20:52:10 -0500 Subject: [PATCH] Start of man-pages-3.13: updating Changes and Changes.old --- Changes | 458 +----------------------------------------------- Changes.old | 490 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 492 insertions(+), 456 deletions(-) diff --git a/Changes b/Changes index 5c76a8761..a91855f01 100644 --- a/Changes +++ b/Changes @@ -1,6 +1,6 @@ -==================== Changes in man-pages-3.12 ==================== +==================== Changes in man-pages-3.13 ==================== -Released: 2008-10-29, Bucaramanga +Released: ????-??-??, Bucaramanga Contributors @@ -16,473 +16,19 @@ Apologies if I missed anyone! New and rewritten pages ----------------------- -pthread_attr_init.3 - Michael Kerrisk - New page for pthread_attr_init(3) and pthread_attr_destroy(3) - -pthread_attr_setdetachstate.3 - Michael Kerrisk - New page for pthread_attr_setdetachstate(3) and - pthread_attr_getdetachstate(3) - -pthread_attr_setguardsize.3 - Michael Kerrisk - New page for pthread_attr_setguardsize(3) and - pthread_attr_getguardsize(3) - -pthread_attr_setscope.3 - Michael Kerrisk - New page for pthread_attr_setscope(3) and pthread_attr_getscope(3) - -pthread_attr_setstack.3 - Michael Kerrisk - New page for pthread_attr_setstack(3) and pthread_attr_getstack(3) - -pthread_attr_setstackaddr.3 - Michael Kerrisk - New page for pthread_attr_setstackaddr(3) and - pthread_attr_getstackaddr(3) - -pthread_attr_setstacksize.3 - Michael Kerrisk - New page for pthread_attr_setstacksize(3) and - pthread_attr_getstacksize(3) - -pthread_create.3 - Michael Kerrisk - New page describing pthread_create(3) - -pthread_detach.3 - Michael Kerrisk - New page for pthread_detach(3) - -pthread_equal.3 - Michael Kerrisk - New page for pthread_equal(3) - -pthread_exit.3 - Michael Kerrisk - New page describing pthread_exit(3) - -pthread_getattr_np.3 - Michael Kerrisk - New page for pthread_getattr_np(3) - -pthread_join.3 - Michael Kerrisk - New page for pthread_join(3) - -pthread_self.3 - Michael Kerrisk - New page for pthread_self(3) - -pthread_tryjoin_np.3 - Michael Kerrisk - New page for pthread_tryjoin_np(3) and pthread_timedjoin_np(3) - Newly documented interfaces in existing pages --------------------------------------------- -dup.2 - Michael Kerrisk - Add description of dup3() - dup3() was added in kernel 2.6.27. - -epoll_create.2 - Michael Kerrisk - Add description of new epoll_create1() - The new epoll_create1() system call appeared in Linux 2.6.27. -eventfd.2 - Michael Kerrisk - Describe eventfd2() and EFD_NONBLOCK and EFD_CLOEXEC - Linux 2.6.27 added eventfd(), which supports a flags argument - that eventfd() did not provide. The flags so far implemented - are EFD_NONBLOCK and EFD_CLOEXEC, - -inotify_init.2 - Michael Kerrisk - Add description of inotify_init1() - The inotify_init1() system call was added in Linux 2.6.27. - -pipe.2 - Michael Kerrisk - Add description of new pipe2() syscall - pipe2() was added in 2.6.27. Describe the O_NONBLOCK and - O_CLOEXEC flags. - -signalfd.2 - Michael Kerrisk - Describe signalfd4() and SFD_NONBLOCK and SFD_CLOEXEC - Linux 2.6.27 added signalfd4(), which supports a flags argument - that signalfd() did not provide. The flags so far implemented - are SFD_NONBLOCK and SFD_CLOEXEC. - New and changed links --------------------- -dup3.2 - Michael Kerrisk - New link to dup.2 - dup.2 now contains the description of the new dup3() syscall. - -epoll_create1.2 - Michael Kerrisk - New link to epoll_create.2 - epoll_create.2 now includes a description of the new - epoll_create1() system call. - -eventfd2.2 - Michael Kerrisk - New link to eventfd.2 - The eventfd.2 page has some details on the eventfd2() system call, - which was new in Linux 2.6.27. - -inotify_init1.2 - Michael Kerrisk - New link to inotify_init.2 - inotify_init.2 now includes a description of the new - inotify_init1() system call. - -pipe2.2 - Michael Kerrisk - New link to pipe.2 - pipe(2) now contains a description of the new pipe2() syscall. - -pthread_attr_destroy.3 - Michael Kerrisk - New link to new pthread_attr_init.3 - -pthread_attr_getdetachstate.3 - Michael Kerrisk - New link to new pthread_attr_setdetachstate.3 - -pthread_attr_getguardsize.3 - Michael Kerrisk - New link to new pthread_attr_setguardsize.3 - -pthread_attr_getscope.3 - Michael Kerrisk - New link to new pthread_attr_setscope.3 - -pthread_attr_getstack.3 - Michael Kerrisk - New link to new pthread_attr_setstack.3 - -pthread_attr_getstackaddr.3 - Michael Kerrisk - New link to new pthread_attr_setstackaddr.3 - -pthread_attr_getstacksize.3 - Michael Kerrisk - New link to new pthread_attr_setstacksize.3 - -pthread_timedjoin_np.3 - Michael Kerrisk - New link to new pthread_tryjoin_np.3 - -signalfd4.2 - Michael Kerrisk - New link to signalfd.2 - signalfd.2 now includes text describing signalfd4() system call, - new in Linux 2.6.27. - Global changes -------------- -eventfd.2, getdents.2, mprotect.2, signalfd.2, timerfd_create.2, -wait.2, backtrace.3, clock_getcpuclockid.3, end.3, fmemopen.3, -fopencookie.3, getdate.3, getgrouplist.3, getprotoent_r.3, -getservent_r.3, gnu_get_libc_version.3, inet.3, inet_pton.3, -makecontext.3, matherr.3, offsetof.3, pthread_attr_init.3, -pthread_create.3, pthread_getattr_np.3, sem_wait.3, strtol.3, core.5 - Michael Kerrisk - Add ".SS Program source" to EXAMPLE - Add ".SS Program source" to clearly distinguish shell session and - descriptive text from actual program code. - -eventfd.2, execve.2, getdents.2, ioprio_set.2, mprotect.2, -signalfd.2, timerfd_create.2, wait.2, backtrace.3, -clock_getcpuclockid.3, end.3, fmemopen.3, fopencookie.3, frexp.3, -getdate.3, getgrouplist.3, getprotoent_r.3, getservent_r.3, -gnu_get_libc_version.3, inet.3, inet_pton.3, makecontext.3, -malloc.3, matherr.3, offsetof.3, pthread_attr_init.3, -pthread_create.3, pthread_getattr_np.3, sem_wait.3, strftime.3, -strtok.3, strtol.3, core.5, proc.5, cpuset.7, mq_overview.7 - Michael Kerrisk - Format user input in shell sessions in boldface - -frexp.3, strftime.3, strtok.3 - Michael Kerrisk - Relocate shell session above example program - Move the shell session text that demonstrates the use of - the example program so that it precedes the actual - example program. This makes the page consistent with the - majority of other pages. - Changes to individual pages --------------------------- -epoll_create.2 - Michael Kerrisk - Say more about unused epoll_create() 'size' arg - Supply a little more explanation about why the 'size' argument - of epoll_create() is nowadays ignored. - -eventfd.2 - Michael Kerrisk - Remove crufty text relating to flags argument - Remove sentence saying that glibc adds a flags argument - to the syscall; that was only relevant for the older - eventfd() system call. -getdents.2 - Christoph Hellwig - Fix text relating to DT_UNKNOWN and 'd_type' support - Some file systems provide partial support for 'dt_type', - returning DT_UNKNOWN for cases they don't support. - Update the discussion of 'd_type' and DT_UNKNOWN to - support this. - -getpeername.2, getsockname.2 - Michael Kerrisk - SEE ALSO: add ip(7) and unix(7) - -getsockopt.2 - Michael Kerrisk - EINVAL can also occur if 'optval' is invalid - In some cases, EINVAL can occur if 'optval' is invalid. - Note this, and point reader to an example in ip(7). - In response to: - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=216092 - -inotify_init.2, pipe.2, timerfd_create.2 - Michael Kerrisk - Clarify *_NONBLOCK description - Make it clear that the NONBLOCK flag sets an attribute in the new - open file description. - -sched_yield.2 - Michael Kerrisk - Rewrite description in terms of threads - The text formerly described the operation of sched_yield() in - terms of processes. It should be in terms of threads. - Michael Kerrisk - Add NOTES text on appropriate use of sched_yield() - Strategic calls to sched_yield() can be used to improve - performance, but unnecessary use should be avoided. - -sigaction.2 - Michael Kerrisk - Clarify that sa_mask affects the *per-thread* signal mask - The page didn't previously clearly explain the scope of the - signal mask that is affected by sa_mask. - -signalfd.2 - Michael Kerrisk - Remove crufty text relating to flags argument - Remove sentence saying that glibc adds a flags argument - to the syscall; that was only relevant for the older - signalfd() system call. - -sigprocmask.2 - Michael Kerrisk - Clarify that sigprocmask() operates on a per-thread mask - The first sentence of the page was vague on the scope of the - attribute changed by sigprocmask(). Reword to make this - clearer and add a sentence in NOTES to explicitly state that - the signal mask is a per-thread attribute. - -socket.2 - Michael Kerrisk - Document SOCK_NONBLOCK and SOCK_CLOEXEC flags - These flags, specified in the 'type' argument, are supported - since Linux 2.6.27. - -socketpair.2 - Michael Kerrisk - Refer to socket(2) for SOCK_CLOEXEC and SOCK_NONBLOCK - Refer the reader to socket(2) for a description of the SOCK_CLOEXEC - and SOCK_NONBLOCK flags, which are supported by socketpair() since - Linux 2.6.27. - -syscalls.2 - Michael Kerrisk - Add new 2.6.27 system calls - Add pipe2(), dup3(), epoll_create1(), inotify_init1(), - eventfd2(), signalfd4(). - -timerfd_create.2 - Michael Kerrisk - Document timerfd_create() TFD_CLOEXEC and TFD_NONBLOCK - TFD_CLOEXEC and TFD_NONBLOCK are supported since LInux 2.6.27. - -vfork.2 - Michael Kerrisk - Clarify meaning of "child releases the parent's memory" - The man page was not explicit about how the memory used by - the child is released back to the parent. - -ctime.3 - Michael Kerrisk - ctime_r() and localtime_r() need not set 'timezone' and 'daylight' - The man page already noted that these functions need not set - 'tzname', but things could be clearer: it tzset() is not called, - then the other two variables also are not set. - - Also, clarify that ctime() does set 'timezone' and 'daylight'. - -dlopen.3 - Michael Kerrisk - LD_LIBRARY_PATH is inspected once, at program start-up - Make it clear that LD_LIBRARY_PATH is inspected *once*, at - program start-up. (Verified from source and by experiment.) - -fmemopen.3 - Michael Kerrisk - Document binary mode (mode 'b') - Glibc 2.9 adds support to fmemopen() for binary mode opens. - Binary mode is specified by inclusion of the letter 'b' in - the 'mode' argument. - -getaddrinfo.3 - Michael Kerrisk - Clarify error descriptions with some examples - Clarify the description of some errors by giving examples - that produce the errors. (Text added for EAI_SERVICE and - EAI_SOCKTYPE.) - - Also, add an error case for EAI_BADFLAGS. - -gethostbyname.3 - Michael Kerrisk - Rationalize text on POSIX.1-2001 obsolete interfaces - POSIX.1 marks gethostbyname(), gethostbyaddr(), and 'h_errno' - as obsolete. The man page explained this, but with some - duplication. Remove the duplication, and otherwise tidy up - discussion of this point. - -popen.3 - Michael Kerrisk - Change one-line description in NAME - s%process I/O%pipe stream to or from a process% - Michael Kerrisk - Document 'e' (close-on-exec) flag - glibc 2.9 implements the 'e' flag in 'type', which sets the - close-on-exec flag on the underlying file descriptor. - -raise.3 - Michael Kerrisk - SEE ALSO: add pthread_kill(3) - -readdir.3 - Christoph Hellwig - Fix text relating to DT_UNKNOWN and 'd_type' support - (This mirrors the previous change to getdents.2) - Some file systems provide partial support for 'dt_type', - returning DT_UNKNOWN for cases they don't support. - Update the discussion of 'd_type' and DT_UNKNOWN to - support this. - -strcpy.3 - Jason Spiro - Strengthen warning about checking against buffer overruns - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=413940 - -tty_ioctl.4 - Michael Kerrisk - Explain capability requirements for TIOCCONS - Explain capability requirements for TIOCCONS, and describe - changes in 2.6.10 relating to capabilities. - Michael Kerrisk - Explain capability requirements for various ioctls - For TIOCSLCKTRMIOS, TIOCSCTTY, TIOCEXCL, explain the exact - capability that is required (the text formerly just said "root" - in each case). - -proc.5 - Michael Kerrisk - Document /proc/sys/kernel/threads-max - Defines the system-wide limit on the number of threads (tasks). - -utmp.5 - Pierre Cazenave - It is just "other" who should not have write perms on utmp - The page was vague before, saying that utmp should not be - writable by any user. This isn't true: it can be, and - typically is, writable by user and group. - -epoll.7 - Michael Kerrisk - Mention epoll_create1() as part of epoll API - epoll_create1() was added in Linux 2.6.27, and extends the - functionality of epoll_create(). - -inotify.7 - Michael Kerrisk - Mention inotify_init1() in overview of API - Discuss the new inotify_init1() system call in the overview of - the inotify API. - -ip.7 - Michael Kerrisk - Detail EINVAL error for IP_ADD_MEMBERSHIP socket option - In response to: - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=216092 - -iso_8859-7.7 - Lefteris Dimitroulakis - Add Drachma, Euro, and Greek Ypogegrammeni - Also, amend description of characters 0241 and 0242. - -man-pages.7 - Michael Kerrisk - Example shell sessions should have user input boldfaced - -pthreads.7 - Michael Kerrisk - Describe return value from pthreads functions - Describe the usual success (0) and failure (non-zero) returns, - and note that POSIX.1-2001 specifies that pthreads functions - can never fail with the error EINTR. - -signal.7 - Michael Kerrisk - Timeouts make socket interfaces non-restartable - If setsockopt() is used to set a timeout on a socket(), - then the various socket interfaces are not automatically - restarted, even if SA_RESTART is specified when - establishing the signal handler. Analogous behavior occurs - for the "stop signals" case. - -socket.7 - Michael Kerrisk - SEE ALSO: add unix(7) - -ld.so.8 - Michael Kerrisk - Document LD_USE_LOAD_BIAS - Drawing heavily on Jakub Jelinek's description in - http://sources.redhat.com/ml/libc-hacker/2003-11/msg00127.html - (Subject: [PATCH] Support LD_USE_LOAD_BIAS) - --inhibit-rpath is ignored for setuid/setgid ld.so - The --inhibit-rpath option is ignored if ld.so is setuid/setgid - (not if the executable is setuid/setgid). - Michael Kerrisk - Since glibc 2.4, setuid/setgid programs ignore LD_ORIGIN_PATH - Michael Kerrisk - Fix description of LD_PROFILE and LD_PROFILE_OUTPUT - Clarify that LD_PROFILE is pathname or a soname, - and identify name of profiling output file. - Fix description of LD_PROFILE_OUTPUT, which wasn't even close to - the truth. (But why did it remain unfixed for so many years?) - Michael Kerrisk - Since glibc 2.3.4, setuid/setgid programs ignore LD_DYNAMIC_WEAK - Michael Kerrisk - Since version 2.3.5, setuid/setgid programs ignore LD_SHOW_AUXV - Michael Kerrisk - Reorder lists of LD_* environment variables alphabetically - Michael Kerrisk - Since glibc 2.3.4, setuid/setgid programs ignore LD_DEBUG diff --git a/Changes.old b/Changes.old index bf7ebbac7..fb1927016 100644 --- a/Changes.old +++ b/Changes.old @@ -15443,3 +15443,493 @@ signal.7 Include text describing semantics of fork() and execve() for signal dispositions, signal mask, and pending signal set. + + +==================== Changes in man-pages-3.12 ==================== + +Released: 2008-10-29, Bucaramanga + + +Contributors +------------ + +The following people contributed notes, ideas, or patches that have +been incorporated in changes in this release: + + +Apologies if I missed anyone! + + +New and rewritten pages +----------------------- + +pthread_attr_init.3 + Michael Kerrisk + New page for pthread_attr_init(3) and pthread_attr_destroy(3) + +pthread_attr_setdetachstate.3 + Michael Kerrisk + New page for pthread_attr_setdetachstate(3) and + pthread_attr_getdetachstate(3) + +pthread_attr_setguardsize.3 + Michael Kerrisk + New page for pthread_attr_setguardsize(3) and + pthread_attr_getguardsize(3) + +pthread_attr_setscope.3 + Michael Kerrisk + New page for pthread_attr_setscope(3) and pthread_attr_getscope(3) + +pthread_attr_setstack.3 + Michael Kerrisk + New page for pthread_attr_setstack(3) and pthread_attr_getstack(3) + +pthread_attr_setstackaddr.3 + Michael Kerrisk + New page for pthread_attr_setstackaddr(3) and + pthread_attr_getstackaddr(3) + +pthread_attr_setstacksize.3 + Michael Kerrisk + New page for pthread_attr_setstacksize(3) and + pthread_attr_getstacksize(3) + +pthread_create.3 + Michael Kerrisk + New page describing pthread_create(3) + +pthread_detach.3 + Michael Kerrisk + New page for pthread_detach(3) + +pthread_equal.3 + Michael Kerrisk + New page for pthread_equal(3) + +pthread_exit.3 + Michael Kerrisk + New page describing pthread_exit(3) + +pthread_getattr_np.3 + Michael Kerrisk + New page for pthread_getattr_np(3) + +pthread_join.3 + Michael Kerrisk + New page for pthread_join(3) + +pthread_self.3 + Michael Kerrisk + New page for pthread_self(3) + +pthread_tryjoin_np.3 + Michael Kerrisk + New page for pthread_tryjoin_np(3) and pthread_timedjoin_np(3) + + +Newly documented interfaces in existing pages +--------------------------------------------- + +dup.2 + Michael Kerrisk + Add description of dup3() + dup3() was added in kernel 2.6.27. + +epoll_create.2 + Michael Kerrisk + Add description of new epoll_create1() + The new epoll_create1() system call appeared in Linux 2.6.27. +eventfd.2 + Michael Kerrisk + Describe eventfd2() and EFD_NONBLOCK and EFD_CLOEXEC + Linux 2.6.27 added eventfd(), which supports a flags argument + that eventfd() did not provide. The flags so far implemented + are EFD_NONBLOCK and EFD_CLOEXEC, + +inotify_init.2 + Michael Kerrisk + Add description of inotify_init1() + The inotify_init1() system call was added in Linux 2.6.27. + +pipe.2 + Michael Kerrisk + Add description of new pipe2() syscall + pipe2() was added in 2.6.27. Describe the O_NONBLOCK and + O_CLOEXEC flags. + +signalfd.2 + Michael Kerrisk + Describe signalfd4() and SFD_NONBLOCK and SFD_CLOEXEC + Linux 2.6.27 added signalfd4(), which supports a flags argument + that signalfd() did not provide. The flags so far implemented + are SFD_NONBLOCK and SFD_CLOEXEC. + + +New and changed links +--------------------- + +dup3.2 + Michael Kerrisk + New link to dup.2 + dup.2 now contains the description of the new dup3() syscall. + +epoll_create1.2 + Michael Kerrisk + New link to epoll_create.2 + epoll_create.2 now includes a description of the new + epoll_create1() system call. + +eventfd2.2 + Michael Kerrisk + New link to eventfd.2 + The eventfd.2 page has some details on the eventfd2() system call, + which was new in Linux 2.6.27. + +inotify_init1.2 + Michael Kerrisk + New link to inotify_init.2 + inotify_init.2 now includes a description of the new + inotify_init1() system call. + +pipe2.2 + Michael Kerrisk + New link to pipe.2 + pipe(2) now contains a description of the new pipe2() syscall. + +pthread_attr_destroy.3 + Michael Kerrisk + New link to new pthread_attr_init.3 + +pthread_attr_getdetachstate.3 + Michael Kerrisk + New link to new pthread_attr_setdetachstate.3 + +pthread_attr_getguardsize.3 + Michael Kerrisk + New link to new pthread_attr_setguardsize.3 + +pthread_attr_getscope.3 + Michael Kerrisk + New link to new pthread_attr_setscope.3 + +pthread_attr_getstack.3 + Michael Kerrisk + New link to new pthread_attr_setstack.3 + +pthread_attr_getstackaddr.3 + Michael Kerrisk + New link to new pthread_attr_setstackaddr.3 + +pthread_attr_getstacksize.3 + Michael Kerrisk + New link to new pthread_attr_setstacksize.3 + +pthread_timedjoin_np.3 + Michael Kerrisk + New link to new pthread_tryjoin_np.3 + +signalfd4.2 + Michael Kerrisk + New link to signalfd.2 + signalfd.2 now includes text describing signalfd4() system call, + new in Linux 2.6.27. + + +Global changes +-------------- + +eventfd.2, getdents.2, mprotect.2, signalfd.2, timerfd_create.2, +wait.2, backtrace.3, clock_getcpuclockid.3, end.3, fmemopen.3, +fopencookie.3, getdate.3, getgrouplist.3, getprotoent_r.3, +getservent_r.3, gnu_get_libc_version.3, inet.3, inet_pton.3, +makecontext.3, matherr.3, offsetof.3, pthread_attr_init.3, +pthread_create.3, pthread_getattr_np.3, sem_wait.3, strtol.3, core.5 + Michael Kerrisk + Add ".SS Program source" to EXAMPLE + Add ".SS Program source" to clearly distinguish shell session and + descriptive text from actual program code. + +eventfd.2, execve.2, getdents.2, ioprio_set.2, mprotect.2, +signalfd.2, timerfd_create.2, wait.2, backtrace.3, +clock_getcpuclockid.3, end.3, fmemopen.3, fopencookie.3, frexp.3, +getdate.3, getgrouplist.3, getprotoent_r.3, getservent_r.3, +gnu_get_libc_version.3, inet.3, inet_pton.3, makecontext.3, +malloc.3, matherr.3, offsetof.3, pthread_attr_init.3, +pthread_create.3, pthread_getattr_np.3, sem_wait.3, strftime.3, +strtok.3, strtol.3, core.5, proc.5, cpuset.7, mq_overview.7 + Michael Kerrisk + Format user input in shell sessions in boldface + +frexp.3, strftime.3, strtok.3 + Michael Kerrisk + Relocate shell session above example program + Move the shell session text that demonstrates the use of + the example program so that it precedes the actual + example program. This makes the page consistent with the + majority of other pages. + + +Changes to individual pages +--------------------------- + +epoll_create.2 + Michael Kerrisk + Say more about unused epoll_create() 'size' arg + Supply a little more explanation about why the 'size' argument + of epoll_create() is nowadays ignored. + +eventfd.2 + Michael Kerrisk + Remove crufty text relating to flags argument + Remove sentence saying that glibc adds a flags argument + to the syscall; that was only relevant for the older + eventfd() system call. +getdents.2 + Christoph Hellwig + Fix text relating to DT_UNKNOWN and 'd_type' support + Some file systems provide partial support for 'dt_type', + returning DT_UNKNOWN for cases they don't support. + Update the discussion of 'd_type' and DT_UNKNOWN to + support this. + +getpeername.2, getsockname.2 + Michael Kerrisk + SEE ALSO: add ip(7) and unix(7) + +getsockopt.2 + Michael Kerrisk + EINVAL can also occur if 'optval' is invalid + In some cases, EINVAL can occur if 'optval' is invalid. + Note this, and point reader to an example in ip(7). + In response to: + http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=216092 + +inotify_init.2, pipe.2, timerfd_create.2 + Michael Kerrisk + Clarify *_NONBLOCK description + Make it clear that the NONBLOCK flag sets an attribute in the new + open file description. + +sched_yield.2 + Michael Kerrisk + Rewrite description in terms of threads + The text formerly described the operation of sched_yield() in + terms of processes. It should be in terms of threads. + Michael Kerrisk + Add NOTES text on appropriate use of sched_yield() + Strategic calls to sched_yield() can be used to improve + performance, but unnecessary use should be avoided. + +sigaction.2 + Michael Kerrisk + Clarify that sa_mask affects the *per-thread* signal mask + The page didn't previously clearly explain the scope of the + signal mask that is affected by sa_mask. + +signalfd.2 + Michael Kerrisk + Remove crufty text relating to flags argument + Remove sentence saying that glibc adds a flags argument + to the syscall; that was only relevant for the older + signalfd() system call. + +sigprocmask.2 + Michael Kerrisk + Clarify that sigprocmask() operates on a per-thread mask + The first sentence of the page was vague on the scope of the + attribute changed by sigprocmask(). Reword to make this + clearer and add a sentence in NOTES to explicitly state that + the signal mask is a per-thread attribute. + +socket.2 + Michael Kerrisk + Document SOCK_NONBLOCK and SOCK_CLOEXEC flags + These flags, specified in the 'type' argument, are supported + since Linux 2.6.27. + +socketpair.2 + Michael Kerrisk + Refer to socket(2) for SOCK_CLOEXEC and SOCK_NONBLOCK + Refer the reader to socket(2) for a description of the SOCK_CLOEXEC + and SOCK_NONBLOCK flags, which are supported by socketpair() since + Linux 2.6.27. + +syscalls.2 + Michael Kerrisk + Add new 2.6.27 system calls + Add pipe2(), dup3(), epoll_create1(), inotify_init1(), + eventfd2(), signalfd4(). + +timerfd_create.2 + Michael Kerrisk + Document timerfd_create() TFD_CLOEXEC and TFD_NONBLOCK + TFD_CLOEXEC and TFD_NONBLOCK are supported since LInux 2.6.27. + +vfork.2 + Michael Kerrisk + Clarify meaning of "child releases the parent's memory" + The man page was not explicit about how the memory used by + the child is released back to the parent. + +ctime.3 + Michael Kerrisk + ctime_r() and localtime_r() need not set 'timezone' and 'daylight' + The man page already noted that these functions need not set + 'tzname', but things could be clearer: it tzset() is not called, + then the other two variables also are not set. + + Also, clarify that ctime() does set 'timezone' and 'daylight'. + +dlopen.3 + Michael Kerrisk + LD_LIBRARY_PATH is inspected once, at program start-up + Make it clear that LD_LIBRARY_PATH is inspected *once*, at + program start-up. (Verified from source and by experiment.) + +fmemopen.3 + Michael Kerrisk + Document binary mode (mode 'b') + Glibc 2.9 adds support to fmemopen() for binary mode opens. + Binary mode is specified by inclusion of the letter 'b' in + the 'mode' argument. + +getaddrinfo.3 + Michael Kerrisk + Clarify error descriptions with some examples + Clarify the description of some errors by giving examples + that produce the errors. (Text added for EAI_SERVICE and + EAI_SOCKTYPE.) + + Also, add an error case for EAI_BADFLAGS. + +gethostbyname.3 + Michael Kerrisk + Rationalize text on POSIX.1-2001 obsolete interfaces + POSIX.1 marks gethostbyname(), gethostbyaddr(), and 'h_errno' + as obsolete. The man page explained this, but with some + duplication. Remove the duplication, and otherwise tidy up + discussion of this point. + +popen.3 + Michael Kerrisk + Change one-line description in NAME + s%process I/O%pipe stream to or from a process% + Michael Kerrisk + Document 'e' (close-on-exec) flag + glibc 2.9 implements the 'e' flag in 'type', which sets the + close-on-exec flag on the underlying file descriptor. + +raise.3 + Michael Kerrisk + SEE ALSO: add pthread_kill(3) + +readdir.3 + Christoph Hellwig + Fix text relating to DT_UNKNOWN and 'd_type' support + (This mirrors the previous change to getdents.2) + Some file systems provide partial support for 'dt_type', + returning DT_UNKNOWN for cases they don't support. + Update the discussion of 'd_type' and DT_UNKNOWN to + support this. + +strcpy.3 + Jason Spiro + Strengthen warning about checking against buffer overruns + http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=413940 + +tty_ioctl.4 + Michael Kerrisk + Explain capability requirements for TIOCCONS + Explain capability requirements for TIOCCONS, and describe + changes in 2.6.10 relating to capabilities. + Michael Kerrisk + Explain capability requirements for various ioctls + For TIOCSLCKTRMIOS, TIOCSCTTY, TIOCEXCL, explain the exact + capability that is required (the text formerly just said "root" + in each case). + +proc.5 + Michael Kerrisk + Document /proc/sys/kernel/threads-max + Defines the system-wide limit on the number of threads (tasks). + +utmp.5 + Pierre Cazenave + It is just "other" who should not have write perms on utmp + The page was vague before, saying that utmp should not be + writable by any user. This isn't true: it can be, and + typically is, writable by user and group. + +epoll.7 + Michael Kerrisk + Mention epoll_create1() as part of epoll API + epoll_create1() was added in Linux 2.6.27, and extends the + functionality of epoll_create(). + +inotify.7 + Michael Kerrisk + Mention inotify_init1() in overview of API + Discuss the new inotify_init1() system call in the overview of + the inotify API. + +ip.7 + Michael Kerrisk + Detail EINVAL error for IP_ADD_MEMBERSHIP socket option + In response to: + http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=216092 + +iso_8859-7.7 + Lefteris Dimitroulakis + Add Drachma, Euro, and Greek Ypogegrammeni + Also, amend description of characters 0241 and 0242. + +man-pages.7 + Michael Kerrisk + Example shell sessions should have user input boldfaced + +pthreads.7 + Michael Kerrisk + Describe return value from pthreads functions + Describe the usual success (0) and failure (non-zero) returns, + and note that POSIX.1-2001 specifies that pthreads functions + can never fail with the error EINTR. + +signal.7 + Michael Kerrisk + Timeouts make socket interfaces non-restartable + If setsockopt() is used to set a timeout on a socket(), + then the various socket interfaces are not automatically + restarted, even if SA_RESTART is specified when + establishing the signal handler. Analogous behavior occurs + for the "stop signals" case. + +socket.7 + Michael Kerrisk + SEE ALSO: add unix(7) + +ld.so.8 + Michael Kerrisk + Document LD_USE_LOAD_BIAS + Drawing heavily on Jakub Jelinek's description in + http://sources.redhat.com/ml/libc-hacker/2003-11/msg00127.html + (Subject: [PATCH] Support LD_USE_LOAD_BIAS) + --inhibit-rpath is ignored for setuid/setgid ld.so + The --inhibit-rpath option is ignored if ld.so is setuid/setgid + (not if the executable is setuid/setgid). + Michael Kerrisk + Since glibc 2.4, setuid/setgid programs ignore LD_ORIGIN_PATH + Michael Kerrisk + Fix description of LD_PROFILE and LD_PROFILE_OUTPUT + Clarify that LD_PROFILE is pathname or a soname, + and identify name of profiling output file. + Fix description of LD_PROFILE_OUTPUT, which wasn't even close to + the truth. (But why did it remain unfixed for so many years?) + Michael Kerrisk + Since glibc 2.3.4, setuid/setgid programs ignore LD_DYNAMIC_WEAK + Michael Kerrisk + Since version 2.3.5, setuid/setgid programs ignore LD_SHOW_AUXV + Michael Kerrisk + Reorder lists of LD_* environment variables alphabetically + Michael Kerrisk + Since glibc 2.3.4, setuid/setgid programs ignore LD_DEBUG