Changes

   1 ==================== Changes in man-pages-4.15 ====================
   2
   3 Released: 2018-02-02, Palo Alto
   4
   5
   6 Contributors
   7 ------------
   8
   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:
  12
  13 Adam Liddell <ml+kernel.org@aliddell.com>
  14 Andrea Parri <parri.andrea@gmail.com>
  15 Andries E. Brouwer <Andries.Brouwer@cwi.nl>
  16 Elie Roudninski <xademax@gmail.com>
  17 Eric Benton <erbenton@comcast.net>
  18 Florian Weimer <fweimer@redhat.com>
  19 G. Branden Robinson <g.branden.robinson@gmail.com>
  20 Jakub Wilk <jwilk@jwilk.net>
  21 Joel Williamson <jwilliamson@carnegietechnologies.com>
  22 John Hubbard <jhubbard@nvidia.com>
  23 Jorgen Hansen <jhansen@vmware.com>
  24 Keno Fischer <keno@juliacomputing.com>
  25 Michael Kerrisk <mtk.manpages@gmail.com>
  26 Michal Hocko <mhocko@kernel.org>
  27 NeilBrown <neilb@suse.com>
  28 Nikola Forró <nforro@redhat.com>
  29 Nikolay Borisov <nborisov@suse.com>
  30 Pradeep Kumar <pradeepsixer@gmail.com>
  31 QingFeng Hao <haoqf@linux.vnet.ibm.com>
  32 Ricardo Biehl Pasquali <pasqualirb@gmail.com>
  33 roblabla <man-pages@roblab.la>
  34 Roman Gushchin <guro@fb.com>
  35 Shawn Landden <slandden@gmail.com>
  36 Stefan Hajnoczi <stefanha@redhat.com>
  37 Stefan Raspl <raspl@linux.vnet.ibm.com>
  38 Tejun Heo <tj@kernel.org>
  39
  40 Apologies if I missed anyone!
  41
  42
  43 New and rewritten pages
  44 -----------------------
  45
  46 s390_sthyi.2
  47     QingFeng Hao  [Michael Kerrisk]
  48         New page for s390-specific s390_sthyi(2)
  49
  50 network_namespaces.7
  51     Michael Kerrisk
  52         New page describing network namespaces
  53             Based on content moved from namespaces(7)
  54     Michael Kerrisk
  55         Reorganize text
  56             No content changes...
  57     Michael Kerrisk
  58         When a NW namespace is freed, veth devices are destroyed
  59
  60 vsock.7
  61     Stefan Hajnoczi  [Jorgen Hansen, Michael Kerrisk]
  62         Document the VSOCK socket address family
  63
  64
  65 Newly documented interfaces in existing pages
  66 ---------------------------------------------
  67
  68 cgroups.7
  69     Michael Kerrisk  [Tejun Heo]
  70         Document cgroups v2 "thread mode"
  71     Michael Kerrisk  [Tejun Heo]
  72         Document cgroup v2 delegation via the 'nsdelegate' mount option
  73     Michael Kerrisk
  74         Document the cgroup.max.depth and cgroup.max.descendants files
  75     Michael Kerrisk
  76         Document 'release_agent' mount option
  77     Michael Kerrisk  [Roman Gushchin]
  78         Document /sys/kernel/cgroup/delegate
  79     Michael Kerrisk  [Roman Gushchin]
  80         Document /sys/kernel/cgroup/features
  81     Michael Kerrisk  [Roman Gushchin]
  82         Document cgroups v2 cgroup.stat file
  83
  84
  85 Global changes
  86 --------------
  87
  88 Various pages
  89     G. Branden Robinson
  90         Standardize on "nonzero"
  91             Also add this term to the style guide in man-pages(7).
  92
  93
  94 Changes to individual pages
  95 ---------------------------
  96
  97 bpf.2
  98     Nikolay Borisov
  99         Sync list of supported map types with 4.14 kernel
 100
 101 copy_file_range.2
 102     Michael Kerrisk
 103         Library support was added in glibc 2.27
 104     Shawn Landden
 105         glibc provides a user-space emulation where the system call is absent
 106     Florian Weimer
 107         EFBIG errors are possible, similar to write(2)
 108     Michael Kerrisk
 109         ERRORS: add EISDIR
 110     Michael Kerrisk
 111         Order ERRORS alphabetically
 112     Michael Kerrisk
 113         Add comment to code example explaining use of syscall(2)
 114
 115 fcntl.2
 116 read.2
 117 write.2
 118     NeilBrown
 119         Document "Lost locks" as cause for EIO.
 120             If an advisory lock is lost, then read/write requests on any
 121             affected file descriptor can return EIO - for NFSv4 at least.
 122
 123 memfd_create.2
 124     Michael Kerrisk
 125         glibc support for memfd_create() was added in version 2.27
 126
 127 mlock.2
 128     Michael Kerrisk
 129         Make details for MLOCK_ONFAULT a little more explicit
 130     Michael Kerrisk
 131         glibc support for mlock2() is added in version 2.27
 132
 133 mmap.2
 134     John Hubbard  [Michael Hocko]
 135         MAP_FIXED is no longer discouraged
 136             MAP_FIXED has been widely used for a very long time, yet the man
 137             page still claims that "the use of this option is discouraged".
 138     John Hubbard
 139         MAP_FIXED updated documentation
 140                 -- Expand the documentation to discuss the hazards in
 141                    enough detail to allow avoiding them.
 142
 143                 -- Mention the upcoming MAP_FIXED_SAFE flag.
 144
 145                 -- Enhance the alignment requirement slightly.
 146
 147 mount.2
 148     Keno Fischer  [Michael Kerrisk]
 149         Add EINVAL error condition when MS_BINDing MNT_LOCKED submounts
 150
 151 mprotect.2
 152 pkey_alloc.2
 153     Michael Kerrisk
 154         Glibc support for memory protection keys was added in version 2.27
 155
 156 perf_event_open.2
 157     Michael Kerrisk
 158         SEE ALSO: add perf(1)
 159
 160 pkey_alloc.2
 161     Michael Kerrisk
 162         Clarify description of pkey_alloc() 'flags' argument
 163
 164 prctl.2
 165     Michael Kerrisk
 166         Defer to capabilities(7) for discussion of the "keep capabilities" flag
 167
 168 recvmmsg.2
 169 sendmmsg.2
 170     Nikola Forró
 171         Point out that error handling is unreliable
 172
 173 seccomp.2
 174     Michael Kerrisk
 175         Clarify that SECCOMP_RET_TRAP SIGSYS signal is thread-directed
 176
 177 syscalls.2
 178     Michael Kerrisk
 179         Add s390-specific s390_sthyi(2) to syscall list
 180
 181 unshare.2
 182     Michael Kerrisk
 183         Clarify the EUSERS occurred only until kernel 4.8
 184
 185 errno.3
 186     Michael Kerrisk
 187         'errno -s' can be used to search for errors by string in description
 188     Michael Kerrisk
 189         Add Linux error text corresponding to ENOMEM
 190
 191 fgetpwent.3
 192     Michael Kerrisk
 193         Add missing ATTRIBUTES preamble
 194
 195 fts.3
 196     Michael Kerrisk  [Pradeep Kumar]
 197         fts_pathlen = strlen(fts_path) + strlen(fts_name)
 198
 199 fuse.4
 200     Michael Kerrisk
 201         Places errors in alphabetical order (no content changes)
 202
 203 veth.4
 204     Michael Kerrisk
 205         Add network_namespaces(7)
 206
 207 sysfs.5
 208     Michael Kerrisk
 209         Refer to cgroups(7) for information about files in /sys/kernel/cgroup
 210
 211 capabilities.7
 212     Michael Kerrisk
 213         Note which capability sets are affected by SECBIT_NO_SETUID_FIXUP
 214             Note explicitly that SECBIT_NO_SETUID_FIXUP is relevant for
 215             the permitted, effective, and ambient capability sets.
 216     Michael Kerrisk
 217         Deemphasize the ancient prctl(2) PR_SET_KEEPCAPS command
 218             The modern approach is SECBITS_KEEP_CAPS.
 219     Michael Kerrisk
 220         Clarify effect of CAP_SETFCAP
 221             Make it clear that CAP_SETFCAP allows setting arbitrary
 222             capabilities on a file.
 223     Michael Kerrisk
 224         Clarify which capability sets are effected by SECBIT_KEEP_CAPS
 225             This flag has relevance only for the process permitted and
 226             effective sets.
 227     Michael Kerrisk
 228         Rephrase CAP_SETPCAP description
 229             * Mention kernel versions.
 230             * Place current kernel behavior first
 231     Michael Kerrisk
 232         SECBIT_KEEP_CAPS is ignored if SECBIT_NO_SETUID_FIXUP is set
 233     Michael Kerrisk
 234         Ambient set is also cleared when UIDs are set to nonzero value
 235
 236 cgroups.7
 237     Michael Kerrisk
 238         Add a more complete description of cgroup v1 named hierarchies
 239     Michael Kerrisk
 240         Add a section on unmounting cgroup v1 filesystems
 241     Michael Kerrisk
 242         Add subsection describing cgroups v2 subtree delegation
 243     Michael Kerrisk
 244         Mention ENOENT error that can occur when writing to subtree_control file
 245     Michael Kerrisk
 246         Add list of currently available version 2 controllers
 247     Nikolay Borisov
 248         Add information about RDMA controller
 249     Michael Kerrisk
 250         Rewrite the description of cgroup v2 subtree control
 251     Michael Kerrisk  [Tejun Heo]
 252         Note Linux 4.11 changes to cgroup v2 delegation containment rules
 253     Michael Kerrisk
 254         systemd(1) nowadays automatically mounts the cgroup2 filesystem
 255     Michael Kerrisk
 256         Clarify that cgroup.controllers is read-only
 257     Michael Kerrisk
 258         Elaborate a little on problems of splitting threads across cgroups in v1
 259     Michael Kerrisk  [Tejun Heo]
 260         Tweak the description of delegation of cgroup.subtree_control
 261
 262 ip.7
 263     Ricardo Biehl Pasquali
 264         INADDR_* values cannot be assigned directly to 's_addr'
 265     Michael Kerrisk
 266         s/INADDR_ANY/INADDR_LOOPBACK/ in discussion of htonl()
 267             INADDR_LOOPBACK is a better example, since it is not
 268             byte-order neutral.
 269
 270 namespaces.7
 271 network_namespaces.7
 272     Michael Kerrisk
 273         Move content from namespaces(7) to network_namespaces(7)
 274
 275 pid_namespaces.7
 276     Michael Kerrisk
 277         SEE ALSO: add mount_namespaces(7)
 278
 279 sched.7
 280     Michael Kerrisk  [Andrea Parri]
 281         Correctly describe effect of priority changes for RT threads
 282             The placement of a thread in the run queue for its new
 283             priority depends on the direction of movement in priority.
 284             (This appears to contradict POSIX, except in the case of
 285             pthread_setschedprio().)
 286
 287 user_namespaces.7
 288     Michael Kerrisk
 289         Mention NS_GET_OWNER_UID ioctl() operation