1 ==================== Changes in man-pages-3.50 ====================
3 Released: 2013-03-15, 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 Andrey Vagin <avagin@openvz.org>
14 Bernhard Kuemel <bernhard@bksys.at>
15 Elie De Brauwer <eliedebrauwer@gmail.com>
16 Erik Saule <erik.saule@bmi.osu.edu>
17 Florian Weimer <fweimer@bfk.de>
18 Friedrich Delgado Friedrichs <friedel@nomaden.org>
19 Jonathan Nieder <jrnieder@gmail.com>
20 Jose Luis Domingo Lopez <debian@24x7linux.com>
21 Mark R Bannister <mark@proseconsulting.co.uk>
22 Michael Kerrisk <mtk.manpages@gmail.com>
23 Sam Varshavchik <mrsam@courier-mta.com>
24 Simon Paillard <spaillard@debian.org>
26 Apologies if I missed anyone!
29 New and rewritten pages
30 -----------------------
32 canonicalize_file_name.3
34 Rewrite page, adding much more detail
42 Global fix: s/END_LICENSE/LICENSE_END/
46 Global fix: s/bitmask/bit mask/
49 Changes to individual pages
50 ---------------------------
54 netgroup description incorrectly refers to initgroups
62 Port access permission bits (ioperm()) are turned off in the child
66 'timeout' is a minimum duration that the call will wait, not a maximum
70 Note that iopl() level of 3 is needed to access ports
72 'num' is *bits* not bytes!
74 Linux 2.6.8 lifted the port limit to 65,536
75 See http://article.gmane.org/gmane.linux.kernel/202624/
76 From: Stas Sergeev <stsp <at> aknet.ru>
77 Subject: [patch][rfc] Larger IO bitmap
78 Date: 2004-05-07 19:55:03 GMT
80 ioperm() operates on the calling *thread* (not process)
82 Clarify meaning of 'turn_on' argument
83 Plus form formatting fixes.
85 Clarify that default state of permission bits in child is off
87 NOTES: add mention of /proc/ioports
93 CAP_SYS_RAWIO is required to *raise* the I/O privilege level
95 Clarify that the two least significant bits of 'level' are what matter
101 Add version information for all (other) syscalls
106 Michael Kerrisk [Jonathan Nieder]
107 ERRORS: Add ENOSYS for lutimes()
108 See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=620746
111 Michael Kerrisk [Erik Saule]
112 Suggest use of the ECHO flag as an alternative
113 See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=644261
117 Document GNU extensions for EACCES and ENOENT errors
120 Michael Kerrisk [Friedrich Delgado Friedrichs]
122 See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=575077
125 Michael Kerrisk [Bernhard Kuemel]
126 Mention that noncanonical mode does not do input processing
127 See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=643854
131 Document write and document the ioctl interface of /dev/random
132 The update consists out of two parts:
133 - a minor thing which just documents what happens if a write to
134 /dev/(u)random is performed, which is used in the example
135 script but not explicitly mentioned.
136 - the other (biggest) part is the documentation of the ioctl()
137 interface which /dev/(u)random exposes. This ioctl() lives in
138 drivers/char/random.c and the primitives can be found in
139 include/linux/random.h
141 One comment remains, there used to be an RNDGETPOOL ioctl() which
142 disappeared in v2.6.9. I found two patches on the net:
143 - http://www.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.8.1/2.6.8.1-mm4/broken-out/dev-random-remove-rndgetpool-ioctl.patch
144 - https://lkml.org/lkml/2004/3/25/168
146 But as far as I can tell the first one got applied but the 2nd
147 one seems more correct. The result is that even today one can
148 still find traces of the RNDGETPOOL ioctl() in the header files.
149 Is this there for historical reasons or because it might break
150 userspace, even though using it will just give an EINVAL.
153 Jose Luis Domingo Lopez
154 Document 'rootfstype' option
155 See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=182014
159 Add various pieces under CAP_SYS_RAWIO
160 Info obtained by grepping the kernel source.
162 Add CAP_SYS_RESOURCE /proc/PID/oom_score_adj case
166 Add a note about broadcast messages to multiple groups
169 Michael Kerrisk [Florian Weimer]
170 Define _GNU_SOURCE to obtain the definition of 'struct ucred'
171 See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=572210