From 88596f05a1d96efec7e7863d41f0e89f68327674 Mon Sep 17 00:00:00 2001 From: Michael Kerrisk Date: Fri, 5 Dec 2008 22:17:41 -0500 Subject: [PATCH] Changes: Update for 3.15 Signed-off-by: Michael Kerrisk --- Changes | 420 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 420 insertions(+) diff --git a/Changes b/Changes index 80aa1febf..6c5a9777e 100644 --- a/Changes +++ b/Changes @@ -9,6 +9,19 @@ Contributors The following people contributed notes, ideas, or patches that have been incorporated in changes in this release: +Andre Majorel +Andries E. Brouwer +Chris Heath +Drake Wilson +Mats Wichmann +Mel Gorman +Michael Kerrisk +Mike Fedyk +Pavel Machek +Petr Baudis +Phil Endecott +Rob Landley +Sam Varshavchik Apologies if I missed anyone! @@ -16,19 +29,426 @@ Apologies if I missed anyone! New and rewritten pages ----------------------- +makedev.3 + Michael Kerrisk + New page for makedev(), major(), and minor() macros + +pthread_cleanup_push_defer_np.3 + Michael Kerrisk + New page for pthread_cleanup_push_defer_np(3) and + pthread_cleanup_pop_restore_np(3) + Newly documented interfaces in existing pages --------------------------------------------- +accept.2 + Michael Kerrisk + Document accept4() system call, new in Linux 2.6.28 + +fmemopen.3 + Petr Baudis + Add description of open_wmemstream(3) + +tcp.7 + Michael Kerrisk + Document MSG_TRUNC flag for TCP sockets New and changed links --------------------- +accept4.2 + Michael Kerrisk + New link to accept.2 + accept.2 now documents the new accept4() system call. + +open_wmemstream.3 + Petr Baudis + New link to fmemopen.3 + fmemopen.3 now documents open_wmemstream(). +pthread_cleanup_pop_restore_np.3 + Michael Kerrisk + New link to new pthread_cleanup_push_defer_np.3 + Global changes -------------- +accept.2 +listen.2 +recv.2 +getpeername.2 +getsockname.2 +shutdown.2 +socketpair.2 + Michael Kerrisk + Global fix: SEE ALSO: add socket(7) + +bind.2 +rcmd.3 +capabilities.7 +ip.7 + Michael Kerrisk + Global fix: s/reserved port/privileged port/ + Some pages used one term, some pages the other term; + make some consistency. + +connect.2 +getpeername.2 +getsockname.2 + Michael Kerrisk + Use consistent argument names + Most other sockets pages are using the names 'addr' + and 'addrlen'; make these pages do the same. + +getpeername.2 +getsockname.2 +getsockopt.2 +recv.2 +send.2 +shutdown.2 +sockatmark.3 +socket.7 +udplite.7 + Michael Kerrisk + SYNOPSIS: Rename socket file descriptor argument to 'sockfd' + Many sockets man pages use the name 'sockfd' already. + For consistency, changes the others to do so as well. + +gnu_dev_major.3 +gnu_dev_makedev.3 +gnu_dev_minor.3 +major.3 +minor.3 + Michael Kerrisk + New links to new makedev(3) page + Changes to individual pages --------------------------- +_exit.2 + Michael Kerrisk + Since glibc 2.3, the exit() wrapper function invokes exit_group(2) + This information is useful to users of strace(1). + +accept.2 + Michael Kerrisk + Clarify details when returned address is truncated + If the returned address is truncated, the 'addrlen' argument + indicates the actual size of the address, rather than a count + of the number of bytes in the truncated buffer. + + Also clarify that if 'addr' argument is NULL, then 'addrlen' + should is unused, and should also be NULL. + Michael Kerrisk + Reorder ERRORS list + Some errors were listed under a separate "may" heading. + There's probably no real need to do this; integrate + those errors into the main list. + +exit_group.2 + Michael Kerrisk + Note that since glibc 2.3, exit(2) invokes exit_group() + +futex.2 + Michael Kerrisk + Mention that glibc provides no wrapper function for futex() + +get_thread_area.2 + Michael Kerrisk + Note that glibc provides no wrapper for this system call + +getdomainname.2 + Michael Kerrisk + Substantial rewrite + Expand description of setdomainname() and getdomainname(). + Note that getdomainname() is implemented as a library function + in glibc. + Note limits on size of domain name. + Reorganize ERRORS list. + +gethostname.2 + Michael Kerrisk + Various parts rewritten + Write a paragraph describing sethostname(). + + Clarify differences between glibc's gethostbyname() and + the kernel gethostbyname() system calls. + +gethostname.2 + Michael Kerrisk + Note that HOST_NAME_MAX is 64 on Linux + Also note that in pre-1.0 days, the limit on hostnames + was 8 bytes. + +getpeername.2 + Michael Kerrisk + Note that returned address may be truncated if buffer is too small + +getsid.2 + Michael Kerrisk + Simplified version information and moved to a new VERSIONS section + +getsockname.2 + Michael Kerrisk + Note that returned address is truncated if buffer is too small + +mknod.2 + Michael Kerrisk + Refer reader to makedev(3) to build a device ID + +mmap.2 + Michael Kerrisk + Loosen language around how 'addr' hint is interpreted + Mel Gorman reported that in Linux 2.6.27, 'addr' is rounded + down to a page boundary. + + Before kernel 2.6.26, if 'addr' was taken as a hint, it was + rounded up to the next page boundary. Since Linux 2.6.24, + it is rounded down. Therefore, loosen the description of + this point to say that the address is rounded to "a nearby + page boundary". + +open.2 + Michael Kerrisk + EFBIG error is now EOVERFLOW (since Linux 2.6.24) + When a 32-bit app opens a file whose size is too big to be + represented in 31-bits, POSIX.1 specifies the error EOVERFLOW. + Linux used to give EFBIG for this case, but 2.6.24 fixed this. + + Also, add some text to describe the error scenario in + more detail. + +pread.2 + Michael Kerrisk + Note that glibc emulation for these calls uses lseek(2) + (This makes it clearer that the emulated calls are not atomic.) + +recv.2 +send.2 + Michael Kerrisk + Make names of "address" and "address length" args more consistent + Make the names of these arguments more consistent with other + sockets man pages. + +recv.2 + Michael Kerrisk + Clarify details when returned address is truncated + If the recvfrom() returned address is truncated, the 'fromlen' + argument indicates the actual size of the address, rather than + a count of the number of bytes in the truncated buffer. + + Also clarify that the 'from' argument can be NULL, in which + case 'fromlen' should is unused, and should also be NULL. + Michael Kerrisk + Internet datagram and netlink sockets support MSG_TRUNC for recv(2) + Internet datagram (since Linux 2.4.27/2.6.8), + and netlink (since Linux 2.6.22) sockets support + the MSG_TRUNC flag for recv(2). + +select.2 + Michael Kerrisk + Rewrote text describing feature test macros requirement for pselect() + +select_tut.2 + Michael Kerrisk + Fix SHUT_FD* macros in example program + Add "do {} while (0)" + +set_thread_area.2 + Michael Kerrisk + Note that glibc provides no wrapper for this system call + +setfsgid.2 +setfsuid.2 + Michael Kerrisk + Simplify version information and move to a VERSIONS section + +setsid.2 + Michael Kerrisk + Rework RETURN VALUE section; add an ERRORS section + +setup.2 + Michael Kerrisk + Relocate some CONFORMING TO text to VERSIONS and NOTES + +stat.2 + Michael Kerrisk + Document EOVERFLOW error + Michael Kerrisk + Refer reader to major() and minor() to decompose a device ID + +syscalls.2 + Michael Kerrisk + Fix version numbers for a few system calls + Some 2.6 system calls were wrongly mentioned as also being + backported into a 2.4.x kernel. + +uname.2 + Michael Kerrisk + DESCRIPTION: Point reader at NOTES for further info on field lengths + +atan.3 + Andries E. Brouwer + Fix return value description + The correct range for the return value is [-pi/2,pi/2]. + (mtk's fix in the last change to the return value text was + a botch-up of a (correct) suggestion by Nicolas Francois.) + +atexit.3 + Michael Kerrisk + atexit() and on_exit(3) register functions on the same list + Michael Kerrisk + Terminating registered function using longjmp() is undefined + According to POSIX.1, using longjmp() to terminate execution of + a function registered using atexit() produces undefined results. + Michael Kerrisk + Calling exit(3) more than once produces undefined results + If an exit handler itself calls exit(3), the results are + undefined (see the POSIX.1-2001 specification of exit(3)). + Michael Kerrisk + The same exit handler may be registered multiple times + Michael Kerrisk + Calling _exit(2) terminates processing of exit handlers + Michael Kerrisk + Terminating registered function using longjmp() is undefined + According to POSIX.1, using longjmp() to terminate execution of + a function registered using atexit() produces undefined results. + +bindresvport.3 + Mats Wichmann + SYNOPSIS: s/\*\*/*/ in prototype + Michael Kerrisk + Fix errors regarding port used, plus other rewrites + Glibc's bindresvport() takes no notice of sin->sin_port: + it always returns an arbitrary reserved port in the + anonymous range (512-1023). (Reported by Mats Wichmann.) + + Also: + * Add EADDRINUSE and EACCES errors. + * Mention use of getsockname(2). + * Other minor rewrites and reorderings of the text. + * Explicitly note that glib's bindresvport() ignores + sin->sin_port. + * Change license There's now virtually no text remaining from + the 1.70 version of this page. + + Reviewed-by: Mats Wichmann + Reviewed-by: Petr Baudis + +dlopen.3 + Petr Baudis + Describe confusing dladdr() behavior + dladdr() will act unexpectedly if called from non-pic code on a + compile-time-generated function pointer. + +fmemopen.3 + Michael Kerrisk + Add VERSIONS section + Petr Baudis + SEE OPEN: Add fopencookie(3) + fopencookie(3) is used to implement fmemopen(). + +fopen.3 + Petr Baudis + SEE ALSO: Add fmemopen(3) and fopencookie(3) + +fopencookie.3 + Petr Baudis + fopencookie() needs _GNU_SOURCE feature test macro + +getaddrinfo.3 + Petr Baudis + Document results ordering and /etc/gai.conf + This patch documents the order of the getaddrinfo(3) results + (RFC 3484), how should the application deal with that, + mentions the extremely common cause of having multiple + results per query (both IPv4 and IPv6 addresses available) + and mentions /etc/gai.conf. + + (mtk: Minor tweaks, and note glibc version for /etc/gai.conf) + +isatty.3 + Michael Kerrisk + Complete rewrite of this page, with rather more detail + +memmem.3 + Michael Kerrisk + Remove sentence saying that libc 5.0.9 is still widely used + That was a *long* time ago. + +on_exit.3 + Michael Kerrisk + Document handling of registrations on fork(2) and execve(2) + Treatment in these cases is the same as for atexit(3). + Michael Kerrisk + Arg given to registered function is status from *last* call to exit() + It's a subtle point, but if a registered function itself + calls exit(3), then subsequent functions that were registered + with on_exit(3) will see the exit status given to the more + recent exit(3) call. + Michael Kerrisk + Note that same function may be registered multiple times + +setlocale.3 +locale.7 + Michael Kerrisk + Clean up the description of LANGUAGE environment variable + Clean up the $LANGUAGE description, by removing bogus comments + from setlocale(3) and expanding the mention in locale(7). + + Maybe you will decide that a more detailed description + should be left to the gettext(3) documentation, but I + actually care about the invisible part of the patch more + since the comments have put me off the track initially + ($LANGUAGE has nothing to do with setlocale(3) and is + completely isolated to gettext, as obvious from the + glibc sources). + +proc.5 + Michael Kerrisk + /proc/stat: s/minor/disk_idx/ in description of /proc/stat + See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=225619 + +capabilities.7 + Drake Wilson + Various minor fixes as per Debian bug 471029 + The relevant pieces of + http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=471029 are: + + - Delete duplicate subentry for KEYCTL_CHOWN/KEYCTL_SETPERM + operations in the CAP_SYS_ADMIN entry. (It feels like that + capability entry should be converted to a list, but I've + left it in semicolon-delimited form for now.) + + - Remove text about ENFILE from the text about the + /proc/sys/fs/file-max limit in the CAP_SYS_ADMIN entry, since + this is already described in the man pages for the relevant + ofile-creating system calls. + + - Correct or clarify a few other bits of grammar and such; + see the diff file itself for details. + +socket.7 + Michael Kerrisk + SEE ALSO: add tcp(7) and udp(7) + +tcp.7 + Michael Kerrisk + Relocate out-of-band data discussion + Move to a new subsection entitled "Sockets API". + Michael Kerrisk + Note that MSG_PEEK can be used on out-of-band data + +time.7 + Michael Kerrisk + SEE ALSO: add clock_gettime(3) + +unix.7 + Michael Kerrisk + Unix domain sockets don't support the recv() MSG_TRUNC flag + Michael Kerrisk + Retitled subsection "(Un)supported features" to "Sockets API" + This is consistent with the recent change in tcp(7). +