| blob | 61fb823e5a9581f2db62ea23c27a5730957037e2 |
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.
7 ---------------------------
9 What: PRISM54
10 When: 2.6.34
12 Why: prism54 FullMAC PCI / Cardbus devices used to be supported only by the
13 prism54 wireless driver. After Intersil stopped selling these
14 devices in preference for the newer more flexible SoftMAC devices
15 a SoftMAC device driver was required and prism54 did not support
16 them. The p54pci driver now exists and has been present in the kernel for
17 a while. This driver supports both SoftMAC devices and FullMAC devices.
18 The main difference between these devices was the amount of memory which
19 could be used for the firmware. The SoftMAC devices support a smaller
20 amount of memory. Because of this the SoftMAC firmware fits into FullMAC
21 devices's memory. p54pci supports not only PCI / Cardbus but also USB
22 and SPI. Since p54pci supports all devices prism54 supports
23 you will have a conflict. I'm not quite sure how distributions are
24 handling this conflict right now. prism54 was kept around due to
25 claims users may experience issues when using the SoftMAC driver.
26 Time has passed users have not reported issues. If you use prism54
27 and for whatever reason you cannot use p54pci please let us know!
28 E-mail us at: linux-wireless@vger.kernel.org
30 For more information see the p54 wiki page:
32 http://wireless.kernel.org/en/users/Drivers/p54
34 Who: Luis R. Rodriguez <lrodriguez@atheros.com>
36 ---------------------------
38 What: IRQF_SAMPLE_RANDOM
39 Check: IRQF_SAMPLE_RANDOM
40 When: July 2009
42 Why: Many of IRQF_SAMPLE_RANDOM users are technically bogus as entropy
43 sources in the kernel's current entropy model. To resolve this, every
44 input point to the kernel's entropy pool needs to better document the
45 type of entropy source it actually is. This will be replaced with
46 additional add_*_randomness functions in drivers/char/random.c
48 Who: Robin Getz <rgetz@blackfin.uclinux.org> & Matt Mackall <mpm@selenic.com>
50 ---------------------------
52 What: Deprecated snapshot ioctls
53 When: 2.6.36
55 Why: The ioctls in kernel/power/user.c were marked as deprecated long time
56 ago. Now they notify users about that so that they need to replace
57 their userspace. After some more time, remove them completely.
59 Who: Jiri Slaby <jirislaby@gmail.com>
61 ---------------------------
63 What: The ieee80211_regdom module parameter
64 When: March 2010 / desktop catchup
66 Why: This was inherited by the CONFIG_WIRELESS_OLD_REGULATORY code,
67 and currently serves as an option for users to define an
68 ISO / IEC 3166 alpha2 code for the country they are currently
69 present in. Although there are userspace API replacements for this
70 through nl80211 distributions haven't yet caught up with implementing
71 decent alternatives through standard GUIs. Although available as an
72 option through iw or wpa_supplicant its just a matter of time before
73 distributions pick up good GUI options for this. The ideal solution
74 would actually consist of intelligent designs which would do this for
75 the user automatically even when travelling through different countries.
76 Until then we leave this module parameter as a compromise.
78 When userspace improves with reasonable widely-available alternatives for
79 this we will no longer need this module parameter. This entry hopes that
80 by the super-futuristically looking date of "March 2010" we will have
81 such replacements widely available.
83 Who: Luis R. Rodriguez <lrodriguez@atheros.com>
85 ---------------------------
87 What: dev->power.power_state
88 When: July 2007
89 Why: Broken design for runtime control over driver power states, confusing
90 driver-internal runtime power management with: mechanisms to support
91 system-wide sleep state transitions; event codes that distinguish
92 different phases of swsusp "sleep" transitions; and userspace policy
93 inputs. This framework was never widely used, and most attempts to
94 use it were broken. Drivers should instead be exposing domain-specific
95 interfaces either to kernel or to userspace.
96 Who: Pavel Machek <pavel@ucw.cz>
98 ---------------------------
100 What: sys_sysctl
101 When: September 2010
102 Option: CONFIG_SYSCTL_SYSCALL
103 Why: The same information is available in a more convenient from
104 /proc/sys, and none of the sysctl variables appear to be
105 important performance wise.
107 Binary sysctls are a long standing source of subtle kernel
108 bugs and security issues.
110 When I looked several months ago all I could find after
111 searching several distributions were 5 user space programs and
112 glibc (which falls back to /proc/sys) using this syscall.
114 The man page for sysctl(2) documents it as unusable for user
115 space programs.
117 sysctl(2) is not generally ABI compatible to a 32bit user
118 space application on a 64bit and a 32bit kernel.
120 For the last several months the policy has been no new binary
121 sysctls and no one has put forward an argument to use them.
123 Binary sysctls issues seem to keep happening appearing so
124 properly deprecating them (with a warning to user space) and a
125 2 year grace warning period will mean eventually we can kill
126 them and end the pain.
128 In the mean time individual binary sysctls can be dealt with
129 in a piecewise fashion.
131 Who: Eric Biederman <ebiederm@xmission.com>
133 ---------------------------
135 What: /proc/<pid>/oom_adj
136 When: August 2012
137 Why: /proc/<pid>/oom_adj allows userspace to influence the oom killer's
138 badness heuristic used to determine which task to kill when the kernel
139 is out of memory.
141 The badness heuristic has since been rewritten since the introduction of
142 this tunable such that its meaning is deprecated. The value was
143 implemented as a bitshift on a score generated by the badness()
144 function that did not have any precise units of measure. With the
145 rewrite, the score is given as a proportion of available memory to the
146 task allocating pages, so using a bitshift which grows the score
147 exponentially is, thus, impossible to tune with fine granularity.
149 A much more powerful interface, /proc/<pid>/oom_score_adj, was
150 introduced with the oom killer rewrite that allows users to increase or
151 decrease the badness() score linearly. This interface will replace
152 /proc/<pid>/oom_adj.
154 A warning will be emitted to the kernel log if an application uses this
155 deprecated interface. After it is printed once, future warnings will be
156 suppressed until the kernel is rebooted.
158 ---------------------------
160 What: CS5535/CS5536 obsolete GPIO driver
161 When: June 2011
162 Files: drivers/staging/cs5535_gpio/*
163 Check: drivers/staging/cs5535_gpio/cs5535_gpio.c
164 Why: A newer driver replaces this; it is drivers/gpio/cs5535-gpio.c, and
165 integrates with the Linux GPIO subsystem. The old driver has been
166 moved to staging, and will be removed altogether around 2.6.40.
167 Please test the new driver, and ensure that the functionality you
168 need and any bugfixes from the old driver are available in the new
169 one.
170 Who: Andres Salomon <dilinger@queued.net>
172 --------------------------
174 What: remove EXPORT_SYMBOL(kernel_thread)
175 When: August 2006
176 Files: arch/*/kernel/*_ksyms.c
177 Check: kernel_thread
178 Why: kernel_thread is a low-level implementation detail. Drivers should
179 use the <linux/kthread.h> API instead which shields them from
180 implementation details and provides a higherlevel interface that
181 prevents bugs and code duplication
182 Who: Christoph Hellwig <hch@lst.de>
184 ---------------------------
186 What: Unused EXPORT_SYMBOL/EXPORT_SYMBOL_GPL exports
187 (temporary transition config option provided until then)
188 The transition config option will also be removed at the same time.
189 When: before 2.6.19
190 Why: Unused symbols are both increasing the size of the kernel binary
191 and are often a sign of "wrong API"
192 Who: Arjan van de Ven <arjan@linux.intel.com>
194 ---------------------------
196 What: PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment
197 When: October 2008
198 Why: The stacking of class devices makes these values misleading and
199 inconsistent.
200 Class devices should not carry any of these properties, and bus
201 devices have SUBSYTEM and DRIVER as a replacement.
202 Who: Kay Sievers <kay.sievers@suse.de>
204 ---------------------------
206 What: ACPI procfs interface
207 When: July 2008
208 Why: ACPI sysfs conversion should be finished by January 2008.
209 ACPI procfs interface will be removed in July 2008 so that
210 there is enough time for the user space to catch up.
211 Who: Zhang Rui <rui.zhang@intel.com>
213 ---------------------------
215 What: CONFIG_ACPI_PROCFS_POWER
216 When: 2.6.39
217 Why: sysfs I/F for ACPI power devices, including AC and Battery,
218 has been working in upstream kenrel since 2.6.24, Sep 2007.
219 In 2.6.37, we make the sysfs I/F always built in and this option
220 disabled by default.
221 Remove this option and the ACPI power procfs interface in 2.6.39.
222 Who: Zhang Rui <rui.zhang@intel.com>
224 ---------------------------
226 What: /proc/acpi/button
227 When: August 2007
228 Why: /proc/acpi/button has been replaced by events to the input layer
229 since 2.6.20.
230 Who: Len Brown <len.brown@intel.com>
232 ---------------------------
234 What: /proc/acpi/event
235 When: February 2008
236 Why: /proc/acpi/event has been replaced by events via the input layer
237 and netlink since 2.6.23.
238 Who: Len Brown <len.brown@intel.com>
240 ---------------------------
242 What: i386/x86_64 bzImage symlinks
243 When: April 2010
245 Why: The i386/x86_64 merge provides a symlink to the old bzImage
246 location so not yet updated user space tools, e.g. package
247 scripts, do not break.
248 Who: Thomas Gleixner <tglx@linutronix.de>
250 ---------------------------
252 What: GPIO autorequest on gpio_direction_{input,output}() in gpiolib
253 When: February 2010
254 Why: All callers should use explicit gpio_request()/gpio_free().
255 The autorequest mechanism in gpiolib was provided mostly as a
256 migration aid for legacy GPIO interfaces (for SOC based GPIOs).
257 Those users have now largely migrated. Platforms implementing
258 the GPIO interfaces without using gpiolib will see no changes.
259 Who: David Brownell <dbrownell@users.sourceforge.net>
260 ---------------------------
262 What: b43 support for firmware revision < 410
263 When: The schedule was July 2008, but it was decided that we are going to keep the
264 code as long as there are no major maintanance headaches.
265 So it _could_ be removed _any_ time now, if it conflicts with something new.
266 Why: The support code for the old firmware hurts code readability/maintainability
267 and slightly hurts runtime performance. Bugfixes for the old firmware
268 are not provided by Broadcom anymore.
269 Who: Michael Buesch <mb@bu3sch.de>
271 ---------------------------
273 What: /sys/o2cb symlink
274 When: January 2010
275 Why: /sys/fs/o2cb is the proper location for this information - /sys/o2cb
276 exists as a symlink for backwards compatibility for old versions of
277 ocfs2-tools. 2 years should be sufficient time to phase in new versions
278 which know to look in /sys/fs/o2cb.
279 Who: ocfs2-devel@oss.oracle.com
281 ---------------------------
283 What: Ability for non root users to shm_get hugetlb pages based on mlock
284 resource limits
285 When: 2.6.31
286 Why: Non root users need to be part of /proc/sys/vm/hugetlb_shm_group or
287 have CAP_IPC_LOCK to be able to allocate shm segments backed by
288 huge pages. The mlock based rlimit check to allow shm hugetlb is
289 inconsistent with mmap based allocations. Hence it is being
290 deprecated.
291 Who: Ravikiran Thirumalai <kiran@scalex86.org>
293 ---------------------------
295 What: CONFIG_THERMAL_HWMON
296 When: January 2009
297 Why: This option was introduced just to allow older lm-sensors userspace
298 to keep working over the upgrade to 2.6.26. At the scheduled time of
299 removal fixed lm-sensors (2.x or 3.x) should be readily available.
300 Who: Rene Herman <rene.herman@gmail.com>
302 ---------------------------
304 What: Code that is now under CONFIG_WIRELESS_EXT_SYSFS
305 (in net/core/net-sysfs.c)
306 When: After the only user (hal) has seen a release with the patches
307 for enough time, probably some time in 2010.
308 Why: Over 1K .text/.data size reduction, data is available in other
309 ways (ioctls)
310 Who: Johannes Berg <johannes@sipsolutions.net>
312 ---------------------------
314 What: sysfs ui for changing p4-clockmod parameters
315 When: September 2009
316 Why: See commits 129f8ae9b1b5be94517da76009ea956e89104ce8 and
317 e088e4c9cdb618675874becb91b2fd581ee707e6.
318 Removal is subject to fixing any remaining bugs in ACPI which may
319 cause the thermal throttling not to happen at the right time.
320 Who: Dave Jones <davej@redhat.com>, Matthew Garrett <mjg@redhat.com>
322 -----------------------------
324 What: fakephp and associated sysfs files in /sys/bus/pci/slots/
325 When: 2011
326 Why: In 2.6.27, the semantics of /sys/bus/pci/slots was redefined to
327 represent a machine's physical PCI slots. The change in semantics
328 had userspace implications, as the hotplug core no longer allowed
329 drivers to create multiple sysfs files per physical slot (required
330 for multi-function devices, e.g.). fakephp was seen as a developer's
331 tool only, and its interface changed. Too late, we learned that
332 there were some users of the fakephp interface.
334 In 2.6.30, the original fakephp interface was restored. At the same
335 time, the PCI core gained the ability that fakephp provided, namely
336 function-level hot-remove and hot-add.
338 Since the PCI core now provides the same functionality, exposed in:
340 /sys/bus/pci/rescan
341 /sys/bus/pci/devices/.../remove
342 /sys/bus/pci/devices/.../rescan
344 there is no functional reason to maintain fakephp as well.
346 We will keep the existing module so that 'modprobe fakephp' will
347 present the old /sys/bus/pci/slots/... interface for compatibility,
348 but users are urged to migrate their applications to the API above.
350 After a reasonable transition period, we will remove the legacy
351 fakephp interface.
352 Who: Alex Chiang <achiang@hp.com>
354 ---------------------------
356 What: CONFIG_RFKILL_INPUT
357 When: 2.6.33
358 Why: Should be implemented in userspace, policy daemon.
359 Who: Johannes Berg <johannes@sipsolutions.net>
361 ----------------------------
363 What: sound-slot/service-* module aliases and related clutters in
364 sound/sound_core.c
365 When: August 2010
366 Why: OSS sound_core grabs all legacy minors (0-255) of SOUND_MAJOR
367 (14) and requests modules using custom sound-slot/service-*
368 module aliases. The only benefit of doing this is allowing
369 use of custom module aliases which might as well be considered
370 a bug at this point. This preemptive claiming prevents
371 alternative OSS implementations.
373 Till the feature is removed, the kernel will be requesting
374 both sound-slot/service-* and the standard char-major-* module
375 aliases and allow turning off the pre-claiming selectively via
376 CONFIG_SOUND_OSS_CORE_PRECLAIM and soundcore.preclaim_oss
377 kernel parameter.
379 After the transition phase is complete, both the custom module
380 aliases and switches to disable it will go away. This removal
381 will also allow making ALSA OSS emulation independent of
382 sound_core. The dependency will be broken then too.
383 Who: Tejun Heo <tj@kernel.org>
385 ----------------------------
387 What: Support for lcd_switch and display_get in asus-laptop driver
388 When: March 2010
389 Why: These two features use non-standard interfaces. There are the
390 only features that really need multiple path to guess what's
391 the right method name on a specific laptop.
393 Removing them will allow to remove a lot of code an significantly
394 clean the drivers.
396 This will affect the backlight code which won't be able to know
397 if the backlight is on or off. The platform display file will also be
398 write only (like the one in eeepc-laptop).
400 This should'nt affect a lot of user because they usually know
401 when their display is on or off.
403 Who: Corentin Chary <corentin.chary@gmail.com>
405 ----------------------------
407 What: sysfs-class-rfkill state file
408 When: Feb 2014
409 Files: net/rfkill/core.c
410 Why: Documented as obsolete since Feb 2010. This file is limited to 3
411 states while the rfkill drivers can have 4 states.
412 Who: anybody or Florian Mickler <florian@mickler.org>
414 ----------------------------
416 What: sysfs-class-rfkill claim file
417 When: Feb 2012
418 Files: net/rfkill/core.c
419 Why: It is not possible to claim an rfkill driver since 2007. This is
420 Documented as obsolete since Feb 2010.
421 Who: anybody or Florian Mickler <florian@mickler.org>
423 ----------------------------
425 What: capifs
426 When: February 2011
427 Files: drivers/isdn/capi/capifs.*
428 Why: udev fully replaces this special file system that only contains CAPI
429 NCCI TTY device nodes. User space (pppdcapiplugin) works without
430 noticing the difference.
431 Who: Jan Kiszka <jan.kiszka@web.de>
433 ----------------------------
435 What: KVM paravirt mmu host support
436 When: January 2011
437 Why: The paravirt mmu host support is slower than non-paravirt mmu, both
438 on newer and older hardware. It is already not exposed to the guest,
439 and kept only for live migration purposes.
440 Who: Avi Kivity <avi@redhat.com>
442 ----------------------------
444 What: iwlwifi 50XX module parameters
445 When: 2.6.40
446 Why: The "..50" modules parameters were used to configure 5000 series and
447 up devices; different set of module parameters also available for 4965
448 with same functionalities. Consolidate both set into single place
449 in drivers/net/wireless/iwlwifi/iwl-agn.c
451 Who: Wey-Yi Guy <wey-yi.w.guy@intel.com>
453 ----------------------------
455 What: iwl4965 alias support
456 When: 2.6.40
457 Why: Internal alias support has been present in module-init-tools for some
458 time, the MODULE_ALIAS("iwl4965") boilerplate aliases can be removed
459 with no impact.
461 Who: Wey-Yi Guy <wey-yi.w.guy@intel.com>
463 ---------------------------
465 What: xt_NOTRACK
466 Files: net/netfilter/xt_NOTRACK.c
467 When: April 2011
468 Why: Superseded by xt_CT
469 Who: Netfilter developer team <netfilter-devel@vger.kernel.org>
471 ----------------------------
473 What: IRQF_DISABLED
474 When: 2.6.36
475 Why: The flag is a NOOP as we run interrupt handlers with interrupts disabled
476 Who: Thomas Gleixner <tglx@linutronix.de>
478 ----------------------------
480 What: The acpi_sleep=s4_nonvs command line option
481 When: 2.6.37
482 Files: arch/x86/kernel/acpi/sleep.c
483 Why: superseded by acpi_sleep=nonvs
484 Who: Rafael J. Wysocki <rjw@sisk.pl>
486 ----------------------------
488 What: PCI DMA unmap state API
489 When: August 2012
490 Why: PCI DMA unmap state API (include/linux/pci-dma.h) was replaced
491 with DMA unmap state API (DMA unmap state API can be used for
492 any bus).
493 Who: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
495 ----------------------------
497 What: DMA_xxBIT_MASK macros
498 When: Jun 2011
499 Why: DMA_xxBIT_MASK macros were replaced with DMA_BIT_MASK() macros.
500 Who: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
502 ----------------------------
504 What: namespace cgroup (ns_cgroup)
505 When: 2.6.38
506 Why: The ns_cgroup leads to some problems:
507 * cgroup creation is out-of-control
508 * cgroup name can conflict when pids are looping
509 * it is not possible to have a single process handling
510 a lot of namespaces without falling in a exponential creation time
511 * we may want to create a namespace without creating a cgroup
513 The ns_cgroup is replaced by a compatibility flag 'clone_children',
514 where a newly created cgroup will copy the parent cgroup values.
515 The userspace has to manually create a cgroup and add a task to
516 the 'tasks' file.
517 Who: Daniel Lezcano <daniel.lezcano@free.fr>
519 ----------------------------
521 What: iwlwifi disable_hw_scan module parameters
522 When: 2.6.40
523 Why: Hareware scan is the prefer method for iwlwifi devices for
524 scanning operation. Remove software scan support for all the
525 iwlwifi devices.
527 Who: Wey-Yi Guy <wey-yi.w.guy@intel.com>
529 ----------------------------
531 What: access to nfsd auth cache through sys_nfsservctl or '.' files
532 in the 'nfsd' filesystem.
533 When: 2.6.40
534 Why: This is a legacy interface which have been replaced by a more
535 dynamic cache. Continuing to maintain this interface is an
536 unnecessary burden.
537 Who: NeilBrown <neilb@suse.de>
539 ----------------------------
541 What: i2c_adapter.id
542 When: June 2011
543 Why: This field is deprecated. I2C device drivers shouldn't change their
544 behavior based on the underlying I2C adapter. Instead, the I2C
545 adapter driver should instantiate the I2C devices and provide the
546 needed platform-specific information.
547 Who: Jean Delvare <khali@linux-fr.org>
549 ----------------------------
551 What: cancel_rearming_delayed_work[queue]()
552 When: 2.6.39
554 Why: The functions have been superceded by cancel_delayed_work_sync()
555 quite some time ago. The conversion is trivial and there is no
556 in-kernel user left.
557 Who: Tejun Heo <tj@kernel.org>
559 ----------------------------
561 What: Legacy, non-standard chassis intrusion detection interface.
562 When: June 2011
563 Why: The adm9240, w83792d and w83793 hardware monitoring drivers have
564 legacy interfaces for chassis intrusion detection. A standard
565 interface has been added to each driver, so the legacy interface
566 can be removed.
567 Who: Jean Delvare <khali@linux-fr.org>
569 ----------------------------
571 What: noswapaccount kernel command line parameter
572 When: 2.6.40
573 Why: The original implementation of memsw feature enabled by
574 CONFIG_CGROUP_MEM_RES_CTLR_SWAP could be disabled by the noswapaccount
575 kernel parameter (introduced in 2.6.29-rc1). Later on, this decision
576 turned out to be not ideal because we cannot have the feature compiled
577 in and disabled by default and let only interested to enable it
578 (e.g. general distribution kernels might need it). Therefore we have
579 added swapaccount[=0|1] parameter (introduced in 2.6.37) which provides
580 the both possibilities. If we remove noswapaccount we will have
581 less command line parameters with the same functionality and we
582 can also cleanup the parameter handling a bit ().
583 Who: Michal Hocko <mhocko@suse.cz>
585 ----------------------------