Documentation/feature-removal-schedule.txt

   1 The following is a list of files and features that are going to be
   2 removed in the kernel source tree.  Every entry should contain what
   3 exactly is going away, why it is happening, and who is going to be doing
   4 the work.  When the feature is removed from the kernel, it should also
   5 be removed from this file.
   6
   7 ---------------------------
   8
   9 What:   x86 floppy disable_hlt
  10 When:   2012
  11 Why:    ancient workaround of dubious utility clutters the
  12         code used by everybody else.
  13 Who:    Len Brown <len.brown@intel.com>
  14
  15 ---------------------------
  16
  17 What:   PRISM54
  18 When:   2.6.34
  19
  20 Why:    prism54 FullMAC PCI / Cardbus devices used to be supported only by the
  21         prism54 wireless driver. After Intersil stopped selling these
  22         devices in preference for the newer more flexible SoftMAC devices
  23         a SoftMAC device driver was required and prism54 did not support
  24         them. The p54pci driver now exists and has been present in the kernel for
  25         a while. This driver supports both SoftMAC devices and FullMAC devices.
  26         The main difference between these devices was the amount of memory which
  27         could be used for the firmware. The SoftMAC devices support a smaller
  28         amount of memory. Because of this the SoftMAC firmware fits into FullMAC
  29         devices's memory. p54pci supports not only PCI / Cardbus but also USB
  30         and SPI. Since p54pci supports all devices prism54 supports
  31         you will have a conflict. I'm not quite sure how distributions are
  32         handling this conflict right now. prism54 was kept around due to
  33         claims users may experience issues when using the SoftMAC driver.
  34         Time has passed users have not reported issues. If you use prism54
  35         and for whatever reason you cannot use p54pci please let us know!
  36         E-mail us at: linux-wireless@vger.kernel.org
  37
  38         For more information see the p54 wiki page:
  39
  40         http://wireless.kernel.org/en/users/Drivers/p54
  41
  42 Who:    Luis R. Rodriguez <lrodriguez@atheros.com>
  43
  44 ---------------------------
  45
  46 What:   IRQF_SAMPLE_RANDOM
  47 Check:  IRQF_SAMPLE_RANDOM
  48 When:   July 2009
  49
  50 Why:    Many of IRQF_SAMPLE_RANDOM users are technically bogus as entropy
  51         sources in the kernel's current entropy model. To resolve this, every
  52         input point to the kernel's entropy pool needs to better document the
  53         type of entropy source it actually is. This will be replaced with
  54         additional add_*_randomness functions in drivers/char/random.c
  55
  56 Who:    Robin Getz <rgetz@blackfin.uclinux.org> & Matt Mackall <mpm@selenic.com>
  57
  58 ---------------------------
  59
  60 What:   Deprecated snapshot ioctls
  61 When:   2.6.36
  62
  63 Why:    The ioctls in kernel/power/user.c were marked as deprecated long time
  64         ago. Now they notify users about that so that they need to replace
  65         their userspace. After some more time, remove them completely.
  66
  67 Who:    Jiri Slaby <jirislaby@gmail.com>
  68
  69 ---------------------------
  70
  71 What:   The ieee80211_regdom module parameter
  72 When:   March 2010 / desktop catchup
  73
  74 Why:    This was inherited by the CONFIG_WIRELESS_OLD_REGULATORY code,
  75         and currently serves as an option for users to define an
  76         ISO / IEC 3166 alpha2 code for the country they are currently
  77         present in. Although there are userspace API replacements for this
  78         through nl80211 distributions haven't yet caught up with implementing
  79         decent alternatives through standard GUIs. Although available as an
  80         option through iw or wpa_supplicant its just a matter of time before
  81         distributions pick up good GUI options for this. The ideal solution
  82         would actually consist of intelligent designs which would do this for
  83         the user automatically even when travelling through different countries.
  84         Until then we leave this module parameter as a compromise.
  85
  86         When userspace improves with reasonable widely-available alternatives for
  87         this we will no longer need this module parameter. This entry hopes that
  88         by the super-futuristically looking date of "March 2010" we will have
  89         such replacements widely available.
  90
  91 Who:    Luis R. Rodriguez <lrodriguez@atheros.com>
  92
  93 ---------------------------
  94
  95 What:   dev->power.power_state
  96 When:   July 2007
  97 Why:    Broken design for runtime control over driver power states, confusing
  98         driver-internal runtime power management with:  mechanisms to support
  99         system-wide sleep state transitions; event codes that distinguish
 100         different phases of swsusp "sleep" transitions; and userspace policy
 101         inputs.  This framework was never widely used, and most attempts to
 102         use it were broken.  Drivers should instead be exposing domain-specific
 103         interfaces either to kernel or to userspace.
 104 Who:    Pavel Machek <pavel@ucw.cz>
 105
 106 ---------------------------
 107
 108 What:   Video4Linux obsolete drivers using V4L1 API
 109 When:   kernel 2.6.39
 110 Files:  drivers/staging/se401/* drivers/staging/usbvideo/*
 111 Check:  drivers/staging/se401/se401.c drivers/staging/usbvideo/usbvideo.c
 112 Why:    There are some drivers still using V4L1 API, despite all efforts we've done
 113         to migrate. Those drivers are for obsolete hardware that the old maintainer
 114         didn't care (or not have the hardware anymore), and that no other developer
 115         could find any hardware to buy. They probably have no practical usage today,
 116         and people with such old hardware could probably keep using an older version
 117         of the kernel. Those drivers will be moved to staging on 2.6.38 and, if nobody
 118         cares enough to port and test them with V4L2 API, they'll be removed on 2.6.39.
 119 Who:    Mauro Carvalho Chehab <mchehab@infradead.org>
 120
 121 ---------------------------
 122
 123 What:   Video4Linux: Remove obsolete ioctl's
 124 When:   kernel 2.6.39
 125 Files:  include/media/videodev2.h
 126 Why:    Some ioctl's were defined wrong on 2.6.2 and 2.6.6, using the wrong
 127         type of R/W arguments. They were fixed, but the old ioctl names are
 128         still there, maintained to avoid breaking binary compatibility:
 129           #define VIDIOC_OVERLAY_OLD    _IOWR('V', 14, int)
 130           #define VIDIOC_S_PARM_OLD     _IOW('V', 22, struct v4l2_streamparm)
 131           #define VIDIOC_S_CTRL_OLD     _IOW('V', 28, struct v4l2_control)
 132           #define VIDIOC_G_AUDIO_OLD    _IOWR('V', 33, struct v4l2_audio)
 133           #define VIDIOC_G_AUDOUT_OLD   _IOWR('V', 49, struct v4l2_audioout)
 134           #define VIDIOC_CROPCAP_OLD    _IOR('V', 58, struct v4l2_cropcap)
 135         There's no sense on preserving those forever, as it is very doubtful
 136         that someone would try to use a such old binary with a modern kernel.
 137         Removing them will allow us to remove some magic done at the V4L ioctl
 138         handler.
 139
 140 Who:    Mauro Carvalho Chehab <mchehab@infradead.org>
 141
 142 ---------------------------
 143
 144 What:   sys_sysctl
 145 When:   September 2010
 146 Option: CONFIG_SYSCTL_SYSCALL
 147 Why:    The same information is available in a more convenient from
 148         /proc/sys, and none of the sysctl variables appear to be
 149         important performance wise.
 150
 151         Binary sysctls are a long standing source of subtle kernel
 152         bugs and security issues.
 153
 154         When I looked several months ago all I could find after
 155         searching several distributions were 5 user space programs and
 156         glibc (which falls back to /proc/sys) using this syscall.
 157
 158         The man page for sysctl(2) documents it as unusable for user
 159         space programs.
 160
 161         sysctl(2) is not generally ABI compatible to a 32bit user
 162         space application on a 64bit and a 32bit kernel.
 163
 164         For the last several months the policy has been no new binary
 165         sysctls and no one has put forward an argument to use them.
 166
 167         Binary sysctls issues seem to keep happening appearing so
 168         properly deprecating them (with a warning to user space) and a
 169         2 year grace warning period will mean eventually we can kill
 170         them and end the pain.
 171
 172         In the mean time individual binary sysctls can be dealt with
 173         in a piecewise fashion.
 174
 175 Who:    Eric Biederman <ebiederm@xmission.com>
 176
 177 ---------------------------
 178
 179 What:   /proc/<pid>/oom_adj
 180 When:   August 2012
 181 Why:    /proc/<pid>/oom_adj allows userspace to influence the oom killer's
 182         badness heuristic used to determine which task to kill when the kernel
 183         is out of memory.
 184
 185         The badness heuristic has since been rewritten since the introduction of
 186         this tunable such that its meaning is deprecated.  The value was
 187         implemented as a bitshift on a score generated by the badness()
 188         function that did not have any precise units of measure.  With the
 189         rewrite, the score is given as a proportion of available memory to the
 190         task allocating pages, so using a bitshift which grows the score
 191         exponentially is, thus, impossible to tune with fine granularity.
 192
 193         A much more powerful interface, /proc/<pid>/oom_score_adj, was
 194         introduced with the oom killer rewrite that allows users to increase or
 195         decrease the badness() score linearly.  This interface will replace
 196         /proc/<pid>/oom_adj.
 197
 198         A warning will be emitted to the kernel log if an application uses this
 199         deprecated interface.  After it is printed once, future warnings will be
 200         suppressed until the kernel is rebooted.
 201
 202 ---------------------------
 203
 204 What:   CS5535/CS5536 obsolete GPIO driver
 205 When:   June 2011
 206 Files:  drivers/staging/cs5535_gpio/*
 207 Check:  drivers/staging/cs5535_gpio/cs5535_gpio.c
 208 Why:    A newer driver replaces this; it is drivers/gpio/cs5535-gpio.c, and
 209         integrates with the Linux GPIO subsystem.  The old driver has been
 210         moved to staging, and will be removed altogether around 2.6.40.
 211         Please test the new driver, and ensure that the functionality you
 212         need and any bugfixes from the old driver are available in the new
 213         one.
 214 Who:    Andres Salomon <dilinger@queued.net>
 215
 216 --------------------------
 217
 218 What:   remove EXPORT_SYMBOL(kernel_thread)
 219 When:   August 2006
 220 Files:  arch/*/kernel/*_ksyms.c
 221 Check:  kernel_thread
 222 Why:    kernel_thread is a low-level implementation detail.  Drivers should
 223         use the <linux/kthread.h> API instead which shields them from
 224         implementation details and provides a higherlevel interface that
 225         prevents bugs and code duplication
 226 Who:    Christoph Hellwig <hch@lst.de>
 227
 228 ---------------------------
 229
 230 What:   Unused EXPORT_SYMBOL/EXPORT_SYMBOL_GPL exports
 231         (temporary transition config option provided until then)
 232         The transition config option will also be removed at the same time.
 233 When:   before 2.6.19
 234 Why:    Unused symbols are both increasing the size of the kernel binary
 235         and are often a sign of "wrong API"
 236 Who:    Arjan van de Ven <arjan@linux.intel.com>
 237
 238 ---------------------------
 239
 240 What:   PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment
 241 When:   October 2008
 242 Why:    The stacking of class devices makes these values misleading and
 243         inconsistent.
 244         Class devices should not carry any of these properties, and bus
 245         devices have SUBSYTEM and DRIVER as a replacement.
 246 Who:    Kay Sievers <kay.sievers@suse.de>
 247
 248 ---------------------------
 249
 250 What:   ACPI procfs interface
 251 When:   July 2008
 252 Why:    ACPI sysfs conversion should be finished by January 2008.
 253         ACPI procfs interface will be removed in July 2008 so that
 254         there is enough time for the user space to catch up.
 255 Who:    Zhang Rui <rui.zhang@intel.com>
 256
 257 ---------------------------
 258
 259 What:   CONFIG_ACPI_PROCFS_POWER
 260 When:   2.6.39
 261 Why:    sysfs I/F for ACPI power devices, including AC and Battery,
 262         has been working in upstream kenrel since 2.6.24, Sep 2007.
 263         In 2.6.37, we make the sysfs I/F always built in and this option
 264         disabled by default.
 265         Remove this option and the ACPI power procfs interface in 2.6.39.
 266 Who:    Zhang Rui <rui.zhang@intel.com>
 267
 268 ---------------------------
 269
 270 What:   /proc/acpi/button
 271 When:   August 2007
 272 Why:    /proc/acpi/button has been replaced by events to the input layer
 273         since 2.6.20.
 274 Who:    Len Brown <len.brown@intel.com>
 275
 276 ---------------------------
 277
 278 What:   /proc/acpi/event
 279 When:   February 2008
 280 Why:    /proc/acpi/event has been replaced by events via the input layer
 281         and netlink since 2.6.23.
 282 Who:    Len Brown <len.brown@intel.com>
 283
 284 ---------------------------
 285
 286 What:   i386/x86_64 bzImage symlinks
 287 When:   April 2010
 288
 289 Why:    The i386/x86_64 merge provides a symlink to the old bzImage
 290         location so not yet updated user space tools, e.g. package
 291         scripts, do not break.
 292 Who:    Thomas Gleixner <tglx@linutronix.de>
 293
 294 ---------------------------
 295
 296 What:   GPIO autorequest on gpio_direction_{input,output}() in gpiolib
 297 When:   February 2010
 298 Why:    All callers should use explicit gpio_request()/gpio_free().
 299         The autorequest mechanism in gpiolib was provided mostly as a
 300         migration aid for legacy GPIO interfaces (for SOC based GPIOs).
 301         Those users have now largely migrated.  Platforms implementing
 302         the GPIO interfaces without using gpiolib will see no changes.
 303 Who:    David Brownell <dbrownell@users.sourceforge.net>
 304 ---------------------------
 305
 306 What:   b43 support for firmware revision < 410
 307 When:   The schedule was July 2008, but it was decided that we are going to keep the
 308         code as long as there are no major maintanance headaches.
 309         So it _could_ be removed _any_ time now, if it conflicts with something new.
 310 Why:    The support code for the old firmware hurts code readability/maintainability
 311         and slightly hurts runtime performance. Bugfixes for the old firmware
 312         are not provided by Broadcom anymore.
 313 Who:    Michael Buesch <mb@bu3sch.de>
 314
 315 ---------------------------
 316
 317 What:   /sys/o2cb symlink
 318 When:   January 2010
 319 Why:    /sys/fs/o2cb is the proper location for this information - /sys/o2cb
 320         exists as a symlink for backwards compatibility for old versions of
 321         ocfs2-tools. 2 years should be sufficient time to phase in new versions
 322         which know to look in /sys/fs/o2cb.
 323 Who:    ocfs2-devel@oss.oracle.com
 324
 325 ---------------------------
 326
 327 What:   Ability for non root users to shm_get hugetlb pages based on mlock
 328         resource limits
 329 When:   2.6.31
 330 Why:    Non root users need to be part of /proc/sys/vm/hugetlb_shm_group or
 331         have CAP_IPC_LOCK to be able to allocate shm segments backed by
 332         huge pages.  The mlock based rlimit check to allow shm hugetlb is
 333         inconsistent with mmap based allocations.  Hence it is being
 334         deprecated.
 335 Who:    Ravikiran Thirumalai <kiran@scalex86.org>
 336
 337 ---------------------------
 338
 339 What:   CONFIG_THERMAL_HWMON
 340 When:   January 2009
 341 Why:    This option was introduced just to allow older lm-sensors userspace
 342         to keep working over the upgrade to 2.6.26. At the scheduled time of
 343         removal fixed lm-sensors (2.x or 3.x) should be readily available.
 344 Who:    Rene Herman <rene.herman@gmail.com>
 345
 346 ---------------------------
 347
 348 What:   Code that is now under CONFIG_WIRELESS_EXT_SYSFS
 349         (in net/core/net-sysfs.c)
 350 When:   After the only user (hal) has seen a release with the patches
 351         for enough time, probably some time in 2010.
 352 Why:    Over 1K .text/.data size reduction, data is available in other
 353         ways (ioctls)
 354 Who:    Johannes Berg <johannes@sipsolutions.net>
 355
 356 ---------------------------
 357
 358 What:   sysfs ui for changing p4-clockmod parameters
 359 When:   September 2009
 360 Why:    See commits 129f8ae9b1b5be94517da76009ea956e89104ce8 and
 361         e088e4c9cdb618675874becb91b2fd581ee707e6.
 362         Removal is subject to fixing any remaining bugs in ACPI which may
 363         cause the thermal throttling not to happen at the right time.
 364 Who:    Dave Jones <davej@redhat.com>, Matthew Garrett <mjg@redhat.com>
 365
 366 -----------------------------
 367
 368 What:   fakephp and associated sysfs files in /sys/bus/pci/slots/
 369 When:   2011
 370 Why:    In 2.6.27, the semantics of /sys/bus/pci/slots was redefined to
 371         represent a machine's physical PCI slots. The change in semantics
 372         had userspace implications, as the hotplug core no longer allowed
 373         drivers to create multiple sysfs files per physical slot (required
 374         for multi-function devices, e.g.). fakephp was seen as a developer's
 375         tool only, and its interface changed. Too late, we learned that
 376         there were some users of the fakephp interface.
 377
 378         In 2.6.30, the original fakephp interface was restored. At the same
 379         time, the PCI core gained the ability that fakephp provided, namely
 380         function-level hot-remove and hot-add.
 381
 382         Since the PCI core now provides the same functionality, exposed in:
 383
 384                 /sys/bus/pci/rescan
 385                 /sys/bus/pci/devices/.../remove
 386                 /sys/bus/pci/devices/.../rescan
 387
 388         there is no functional reason to maintain fakephp as well.
 389
 390         We will keep the existing module so that 'modprobe fakephp' will
 391         present the old /sys/bus/pci/slots/... interface for compatibility,
 392         but users are urged to migrate their applications to the API above.
 393
 394         After a reasonable transition period, we will remove the legacy
 395         fakephp interface.
 396 Who:    Alex Chiang <achiang@hp.com>
 397
 398 ---------------------------
 399
 400 What:   CONFIG_RFKILL_INPUT
 401 When:   2.6.33
 402 Why:    Should be implemented in userspace, policy daemon.
 403 Who:    Johannes Berg <johannes@sipsolutions.net>
 404
 405 ----------------------------
 406
 407 What:   sound-slot/service-* module aliases and related clutters in
 408         sound/sound_core.c
 409 When:   August 2010
 410 Why:    OSS sound_core grabs all legacy minors (0-255) of SOUND_MAJOR
 411         (14) and requests modules using custom sound-slot/service-*
 412         module aliases.  The only benefit of doing this is allowing
 413         use of custom module aliases which might as well be considered
 414         a bug at this point.  This preemptive claiming prevents
 415         alternative OSS implementations.
 416
 417         Till the feature is removed, the kernel will be requesting
 418         both sound-slot/service-* and the standard char-major-* module
 419         aliases and allow turning off the pre-claiming selectively via
 420         CONFIG_SOUND_OSS_CORE_PRECLAIM and soundcore.preclaim_oss
 421         kernel parameter.
 422
 423         After the transition phase is complete, both the custom module
 424         aliases and switches to disable it will go away.  This removal
 425         will also allow making ALSA OSS emulation independent of
 426         sound_core.  The dependency will be broken then too.
 427 Who:    Tejun Heo <tj@kernel.org>
 428
 429 ----------------------------
 430
 431 What:   Support for lcd_switch and display_get in asus-laptop driver
 432 When:   March 2010
 433 Why:    These two features use non-standard interfaces. There are the
 434         only features that really need multiple path to guess what's
 435         the right method name on a specific laptop.
 436
 437         Removing them will allow to remove a lot of code an significantly
 438         clean the drivers.
 439
 440         This will affect the backlight code which won't be able to know
 441         if the backlight is on or off. The platform display file will also be
 442         write only (like the one in eeepc-laptop).
 443
 444         This should'nt affect a lot of user because they usually know
 445         when their display is on or off.
 446
 447 Who:    Corentin Chary <corentin.chary@gmail.com>
 448
 449 ----------------------------
 450
 451 What:   sysfs-class-rfkill state file
 452 When:   Feb 2014
 453 Files:  net/rfkill/core.c
 454 Why:    Documented as obsolete since Feb 2010. This file is limited to 3
 455         states while the rfkill drivers can have 4 states.
 456 Who:    anybody or Florian Mickler <florian@mickler.org>
 457
 458 ----------------------------
 459
 460 What:   sysfs-class-rfkill claim file
 461 When:   Feb 2012
 462 Files:  net/rfkill/core.c
 463 Why:    It is not possible to claim an rfkill driver since 2007. This is
 464         Documented as obsolete since Feb 2010.
 465 Who:    anybody or Florian Mickler <florian@mickler.org>
 466
 467 ----------------------------
 468
 469 What:   capifs
 470 When:   February 2011
 471 Files:  drivers/isdn/capi/capifs.*
 472 Why:    udev fully replaces this special file system that only contains CAPI
 473         NCCI TTY device nodes. User space (pppdcapiplugin) works without
 474         noticing the difference.
 475 Who:    Jan Kiszka <jan.kiszka@web.de>
 476
 477 ----------------------------
 478
 479 What:   KVM paravirt mmu host support
 480 When:   January 2011
 481 Why:    The paravirt mmu host support is slower than non-paravirt mmu, both
 482         on newer and older hardware.  It is already not exposed to the guest,
 483         and kept only for live migration purposes.
 484 Who:    Avi Kivity <avi@redhat.com>
 485
 486 ----------------------------
 487
 488 What:   iwlwifi 50XX module parameters
 489 When:   2.6.40
 490 Why:    The "..50" modules parameters were used to configure 5000 series and
 491         up devices; different set of module parameters also available for 4965
 492         with same functionalities. Consolidate both set into single place
 493         in drivers/net/wireless/iwlwifi/iwl-agn.c
 494
 495 Who:    Wey-Yi Guy <wey-yi.w.guy@intel.com>
 496
 497 ----------------------------
 498
 499 What:   iwl4965 alias support
 500 When:   2.6.40
 501 Why:    Internal alias support has been present in module-init-tools for some
 502         time, the MODULE_ALIAS("iwl4965") boilerplate aliases can be removed
 503         with no impact.
 504
 505 Who:    Wey-Yi Guy <wey-yi.w.guy@intel.com>
 506
 507 ---------------------------
 508
 509 What:   xt_NOTRACK
 510 Files:  net/netfilter/xt_NOTRACK.c
 511 When:   April 2011
 512 Why:    Superseded by xt_CT
 513 Who:    Netfilter developer team <netfilter-devel@vger.kernel.org>
 514
 515 ----------------------------
 516
 517 What:   IRQF_DISABLED
 518 When:   2.6.36
 519 Why:    The flag is a NOOP as we run interrupt handlers with interrupts disabled
 520 Who:    Thomas Gleixner <tglx@linutronix.de>
 521
 522 ----------------------------
 523
 524 What:   The acpi_sleep=s4_nonvs command line option
 525 When:   2.6.37
 526 Files:  arch/x86/kernel/acpi/sleep.c
 527 Why:    superseded by acpi_sleep=nonvs
 528 Who:    Rafael J. Wysocki <rjw@sisk.pl>
 529
 530 ----------------------------
 531
 532 What:   PCI DMA unmap state API
 533 When:   August 2012
 534 Why:    PCI DMA unmap state API (include/linux/pci-dma.h) was replaced
 535         with DMA unmap state API (DMA unmap state API can be used for
 536         any bus).
 537 Who:    FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
 538
 539 ----------------------------
 540
 541 What:   DMA_xxBIT_MASK macros
 542 When:   Jun 2011
 543 Why:    DMA_xxBIT_MASK macros were replaced with DMA_BIT_MASK() macros.
 544 Who:    FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
 545
 546 ----------------------------
 547
 548 What:   namespace cgroup (ns_cgroup)
 549 When:   2.6.38
 550 Why:    The ns_cgroup leads to some problems:
 551         * cgroup creation is out-of-control
 552         * cgroup name can conflict when pids are looping
 553         * it is not possible to have a single process handling
 554         a lot of namespaces without falling in a exponential creation time
 555         * we may want to create a namespace without creating a cgroup
 556
 557         The ns_cgroup is replaced by a compatibility flag 'clone_children',
 558         where a newly created cgroup will copy the parent cgroup values.
 559         The userspace has to manually create a cgroup and add a task to
 560         the 'tasks' file.
 561 Who:    Daniel Lezcano <daniel.lezcano@free.fr>
 562
 563 ----------------------------
 564
 565 What:   iwlwifi disable_hw_scan module parameters
 566 When:   2.6.40
 567 Why:    Hareware scan is the prefer method for iwlwifi devices for
 568         scanning operation. Remove software scan support for all the
 569         iwlwifi devices.
 570
 571 Who:    Wey-Yi Guy <wey-yi.w.guy@intel.com>
 572
 573 ----------------------------
 574
 575 What:   access to nfsd auth cache through sys_nfsservctl or '.' files
 576         in the 'nfsd' filesystem.
 577 When:   2.6.40
 578 Why:    This is a legacy interface which have been replaced by a more
 579         dynamic cache.  Continuing to maintain this interface is an
 580         unnecessary burden.
 581 Who:    NeilBrown <neilb@suse.de>
 582
 583 ----------------------------
 584
 585 What:   i2c_adapter.id
 586 When:   June 2011
 587 Why:    This field is deprecated. I2C device drivers shouldn't change their
 588         behavior based on the underlying I2C adapter. Instead, the I2C
 589         adapter driver should instantiate the I2C devices and provide the
 590         needed platform-specific information.
 591 Who:    Jean Delvare <khali@linux-fr.org>
 592
 593 ----------------------------
 594
 595 What:   cancel_rearming_delayed_work[queue]()
 596 When:   2.6.39
 597
 598 Why:    The functions have been superceded by cancel_delayed_work_sync()
 599         quite some time ago.  The conversion is trivial and there is no
 600         in-kernel user left.
 601 Who:    Tejun Heo <tj@kernel.org>
 602
 603 ----------------------------
 604
 605 What:   Legacy, non-standard chassis intrusion detection interface.
 606 When:   June 2011
 607 Why:    The adm9240, w83792d and w83793 hardware monitoring drivers have
 608         legacy interfaces for chassis intrusion detection. A standard
 609         interface has been added to each driver, so the legacy interface
 610         can be removed.
 611 Who:    Jean Delvare <khali@linux-fr.org>
 612
 613 ----------------------------
 614
 615 What:   noswapaccount kernel command line parameter
 616 When:   2.6.40
 617 Why:    The original implementation of memsw feature enabled by
 618         CONFIG_CGROUP_MEM_RES_CTLR_SWAP could be disabled by the noswapaccount
 619         kernel parameter (introduced in 2.6.29-rc1). Later on, this decision
 620         turned out to be not ideal because we cannot have the feature compiled
 621         in and disabled by default and let only interested to enable it
 622         (e.g. general distribution kernels might need it). Therefore we have
 623         added swapaccount[=0|1] parameter (introduced in 2.6.37) which provides
 624         the both possibilities. If we remove noswapaccount we will have
 625         less command line parameters with the same functionality and we
 626         can also cleanup the parameter handling a bit ().
 627 Who:    Michal Hocko <mhocko@suse.cz>
 628
 629 ----------------------------