1 ==================== Changes in man-pages-3.76 ====================
3 Released: 2014-12-31, Munich
9 The following people contributed patches/fixes or (noted in brackets
10 in the changelog below) reports, notes, and ideas that have been
11 incorporated in changes in this release:
13 Adam Jiang <jiang.adam@gmail.com>
14 Andrea Balboni <andrea.balboni@unimore.it>
15 Andreas Schwab <schwab@suse.de>
16 Bernhard Walle <bernhard@bwalle.de>
17 Carlos O'Donell <carlos@redhat.com>
18 David Wragg <david@wragg.org>
19 Florian Westphal <fw@strlen.de>
20 Heinrich Schuchardt <xypron.glpk@gmx.de>
21 Huxiaoxiang <huxiaoxiang@huawei.com>
22 Jan Chaloupka <jchaloup@redhat.com>
23 Jonathan Wakely <jwakely@redhat.com>
24 Jonny Grant <jg@jguk.org>
25 Josh Triplett <josh@joshtriplett.org>
26 Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
27 Laurent Georget <laurent.georget@supelec.fr>
28 Manuel López-Ibáñez <lopezibanez@gmail.com>
29 Marko Myllynen <myllynen@redhat.com>
30 Ma Shimiao <mashimiao.fnst@cn.fujitsu.com>
31 Mel Gorman <mgorman@suse.de>
32 Michael Gehring <mg@ebfe.org>
33 Michael Haardt <michael@moria.de>
34 Michael Kerrisk <mtk.manpages@gmail.com>
35 Mike Frysinger <vapier@gentoo.org>
36 Rasmus Villemoes <rv@rasmusvillemoes.dk>
37 Richard Weinberger <richard@nod.at>
38 Rich Felker <dalias@libc.org>
39 Scott Harvey <scott.harvey@stonybrook.edu>
40 Siddhesh Poyarekar <siddhesh.poyarekar@gmail.com>
41 Simon Newton <nomis52@gmail.com>
42 Simon Paillard <spaillard@debian.org>
43 Sven Hoexter <sven@stormbind.net>
44 Tobias Werth <werth@cs.fau.de>
45 Weijie Yang <weijie.yang.kh@gmail.com>
46 Will Newton <will.newton@linaro.org>
47 Yuri Kozlov <yuray@komyakino.ru>
48 刘湃 <liupai_work@163.com>
49 尹杰 <writalnaie@gmail.com>
51 Apologies if I missed anyone!
54 New and rewritten pages
55 -----------------------
59 Add fields in struct timex description
60 This patch updates the man page with the new fields added in
61 struct timex since last edition of the man page.
62 Laurent Georget [Michael Kerrisk]
65 Improve description of ADJ_OFFSET_SINGLESHOT
67 Add brief documentation of ADJ_MICRO and ADJ_NANO
69 Reformat return value list
70 And remove numeric values, since they're not needed
72 Other 'modes' bits are ignored on ADJ_OFFSET_*
73 Other bits in 'modes' are ignored if modes contains
74 ADJ_OFFSET_SINGLESHOT or ADJ_OFFSET_SS_READ.
76 Add nanosecond details
77 Fixes https://bugzilla.kernel.org/show_bug.cgi?id=61171.
79 Document ADJ_OFFSET_SS_READ
81 Reformat 'times' flags as list
82 And remove numeric values, since they're not needed.
84 Note effect of ADJ_NANO for ADJ_SETOFFSET
86 Add comment noting that timex structure contains padding bytes
88 Add more details to description of 'tai' field
90 Note meaning of "PLL" abbreviation
92 Clarify which 'timex' field is used by each 'modes' bit
94 Document timex 'status' bits
96 Clarify treatment of other 'modes' bits for ADJ_OFFSET_*
98 Update RFC number: RFC 5905 obsoletes RFC 1305
100 Briefly document ADJ_SETOFFSET
102 Note PPS (pulse per second) fields in timex structure
106 Add (a lot) more detail on the signal trampoline
107 And rewrite much of the page.
110 Newly documented interfaces in existing pages
111 ---------------------------------------------
115 Document /proc/thread-self
116 /proc/thread-self has been introduced in Linux 3.17 with
117 commit 0097875bd41528922fb3bb5f348c53f17e00e2fd.
118 Sven Hoexter [Michael Kerrisk, Kamezawa Hiroyuki]
119 Document "VmSwap" field of /proc/[pid]/status
121 Document /proc/net/netfilter/nfnetlink_queue
124 Changes to individual pages
125 ---------------------------
129 Mention default path for compiled files
133 Note that 'clock_id' can also be a CPU clock ID
137 Regular files and directories can't be monitored with epoll_ctl()
141 Rename 'd' argument to 'fd' in text
142 In most other manpages file descriptors are called 'fd'.
143 This patches renames attribute 'd' to 'fd'.
147 VERSIONS: Support for madvise() is now configurable
148 Support for this system call now depends on the
149 CONFIG_ADVISE_SYSCALLS configuration option.
153 Enhance rationale discussion for openat() and friends
157 Document the behavior of partial page discard requests
158 It is not obvious from the interface that partial page discard'
159 requests are ignored. It should be spelled out.
160 Michael Kerrisk [Weijie Yang]
161 ERRORS: Since 2.6.16, the kernel correctly deals with the ESPIPE case
163 Support for fadvise64() is now configurable
164 Support for this system call now depends on the
165 CONFIG_ADVISE_SYSCALLS configuration option.
169 Correct description of null-termination in PR_GET_NAME and PR_SET_NAME
170 The size of the process name has always been at most 16 byte
171 _including_ the null terminator. This also means that the
172 name returned by PR_GET_NAME is always null-terminated.
174 PR_SET_NAME silently truncates strings that exceed 16 bytes
178 Add some text explaining why restart_syscall() exists
182 NOTES: Add paragraph on how to discover set of CPUs available on system
184 SEE ALSO: add nproc(1) and lscpu(1)
188 SEE ALSO: add restart_syscall(2)
192 Add note that interrupted semtimedop() returns 'timeout' unchanged
194 Remove information about semtimedop() EAGAIN that is repeated elsewhere
196 Add subsection head for semtimedop()
200 Rewrite some pieces and add some details
201 Among other changes, add an explanation of why setsid() can't
202 be called from a process group leader
206 Since 3.16, support for these system call is configurable
207 Support for these calls is now dependent on the setting of the
208 CONFIG_SGETMASK_SYSCALL option.
214 Add some detail on the sa_restorer field
216 SEE ALSO: add sigreturn(2)
220 Reformat description of 'fd_in' and 'off_in' to improve readability
224 SEE ALSO: add errno(3)
228 SEE ALSO: add errno(3)
236 Add shell session demonstrating use of the example program
239 Michael Kerrisk [Rich Felker]
240 Remove bogus text saying tgid==-1 makes tgkill() equivalent to tkill()
244 Note that SIGABRT is raised as though raise(3) is called
245 Also note that abort is POSIX.1-2008 compliant.
249 ensure buf is suitably aligned in sending example
250 Inspection of the definition of CMSG_FIRSTHDR (both in glibc and
251 the suggested definition in RFC3542) shows that it yields the
252 msg_control field. So when sending, the pointer placed in
253 msg_control should be suitably aligned as a struct cmsghdr.
254 In the sending example, buf was declared as a bare char array,
255 and so is not necessarily suitably aligned.
257 The solution here involves placing buf inside a union, and is
258 based on the sockets/scm_rights_send.c sample from The Linux
259 Programming Interface "dist" source code collection.
263 Before glibc 2.19, exp() did not give ERANGE error on underflow
264 http://sources.redhat.com/bugzilla/show_bug.cgi?id=6787
268 FTW_CHDIR has no effect on the 'fpath' argument passed to fn()
271 Michael Kerrisk [Jonny Grant]
272 Ensure that 'nsecs' is used
275 Michael Kerrisk [Will Newton]
276 Since glibc 2.16, ilogb() does correctly diagnose domain errors
280 Document return value for n==0 case
282 Warn against use of memcmp() for comparing security-critical data
286 Document the O_CLOEXEC flag
288 Place 'flags' constants in alphabetical order
292 Add note on performance characteristics of pow()
294 pthread_setschedparam.3
296 Fix logic error in example program
297 The example program will crash if -A is used, since 'attr'
301 *** Error in `./a.out': free(): invalid pointer: 0xb779c3c4 ***
302 Aborted (core dumped)
304 Small fixes to example program
308 Starting with version 2.21, glibc no longer exports sigvec()
312 Document _SC_NGROUPS_MAX
313 Already documented in getgroups(2), but not in sysconf(3).
317 SEE ALSO: add tset(1)
321 Since glibc 2.18, errno is correctly set to EDOM when (x == -infinity))
325 Make it clear that WRDE_NOCMD prevents command substitution
326 The use of WRDE_NOCMD prevents command substitution. If the flag
327 WRDE_NOCMD is set then no command substitution shall occur and
328 the error WRDE_CMDSUB will be returned if such substitution is
329 requested when processing the words.
331 The manual page as-is makes it seem like the command substitution
332 occurs, and an error is returned *after* the substitution.
333 This patch clarifies that.
337 Describe the formats of values
338 locale(5) describes what a locale should define but doesn't
339 spell out how (in what format). The patch attempts to address
340 this, it also has few trivial additional enhancements.
342 * Reference to locale(7) for category descriptions.
343 * Clarify first_workday in NOTES a bit.
344 * Add upstream BZ reference for two missing LC_ADDRESS fields.
346 Fix miscoded character
350 Add missing no-tld-query option
351 Based on commit [1], the no-tld-query option exists for
352 resolv.conf configuration file. Description of this option
353 is provided in [2]. This patch just copies this option
354 into resolv.conf.5 man page. Plus changes 'a' member
355 into 'an' before 'unqualified name as if it ...'
356 on the third line of [2]. Based on [3], this option
357 was added in glibc 2.14 as solving [4] bug.
359 [1] https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=f87dfb1f11c01f2ccdc40d81e134cd06b32e28e8
360 [2] http://www.daemon-systems.org/man/resolv.conf.5.html man page.
361 [3] https://sourceware.org/git/?p=glibc.git;a=blob;f=NEWS;h=952f32af17e7fb49c4c1a305de673a13075bfaf5;hb=f87dfb1f11c01f2ccdc40d81e134cd06b32e28e8
362 [4] https://sourceware.org/bugzilla/show_bug.cgi?id=12734
366 Cross-reference getgroups(2)
367 Since credentials.7 discusses supplementary GIDs, it should
368 reference getgroups(2).
372 Allow relative paths in example
373 The current example code requires passing an absolute
374 path to the mount to be watched.
376 By passing AT_FDCWD to fanotify_mark it can use both
377 absolute and relative paths.
379 fallocate(2) creates no events
380 fallocate(2) should create FAN_MODIFY events but does not.
381 Heinrich Schuchardt [Michael Kerrisk]
382 fanotify notifies only events generated on the same mount
383 Unfortunately, fanotify does not inform listeners for all paths
384 under which a touched filesystem object is visible, but only the
385 listener using the same path as the process touching the
388 Update BUGS to note bugs still not fixed in 3.17
389 I bumped the Linux version number in the BUGS section to 3.17.
393 fallocate(2) does not trigger inotify events
394 Calling fallocate(2) does not result in inotify events.
398 Improve LOCPATH description
399 LOCPATH is ignored by privileged programs.
401 Add locale archive references.
406 Michael Kerrisk [Laurent Georget]
407 Clarify that SEE ALSO entries may refer to pages from other projects
411 Mention other "slow devices"
412 Reads from eventfd(2), signalfd(2), timerfd(2), inotify(7),
413 and fanotify(7) file descriptors are also slow operations
414 that are restartable.
416 Fix SO_RECVTIMEO/ SO_SENDTIMEO confusion in text
418 Since Linux 3.8, reads on inotify(7) file descriptors are restartable
420 inotify(7) reads no longer show the odd EINTR error after SIGCONT
421 Since kernel 3.7, reads from inotify(7) file descriptors no longer
422 show the (Linux oddity) behavior of failing with EINTR when the
423 process resumes after a stop signal + SIGCONT.
425 SEE ALSO: add sigreturn(2)
428 Michael Kerrisk [Scott Harvey]
429 Fix buglet in code snippet in BUGS section
433 Add --inhibit-cache option
434 The dynamic loader has 6 options, only 5 are documented.
435 This patch documents the sixth option i.e. `--inhibit-cache`.
436 Jonathan Wakely [Siddhesh Poyarekar]
437 Correct documentation of $ORIGIN
438 As noted by Siddhesh:
440 The ld.so man page says:
442 $ORIGIN (or equivalently ${ORIGIN})
443 This expands to the directory containing the
444 application executable. Thus, an application located
445 in somedir/app could be compiled with
447 This is incorrect since it expands to the directory containing
448 the DSO and not the application executable. This seems like
449 deliberate behaviour in dl-object.c, so it needs to be fixed in
452 See http://stackoverflow.com/questions/26280738/what-is-the-equivalent-of-loader-path-for-rpath-specification-on-linux/26281226#26281226