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. The suggested deprecation period is 3 releases.
7 ---------------------------
9 What: support for i.mx25 in mx2_camera.c
11 Why: it's been broken for a year. Furthermore, i.MX25 video capture
12 HW doesn't have much in common with i.MX27. A separate driver
13 will be needed for it.
14 Who: Javier Martin<javier.martin@vista-silicon.com>
16 ---------------------------
18 What: ddebug_query="query" boot cmdline param
20 Why: obsoleted by dyndbg="query" and module.dyndbg="query"
21 Who: Jim Cromie <jim.cromie@gmail.com>, Jason Baron <jbaron@redhat.com>
23 ---------------------------
25 What: /proc/sys/vm/nr_pdflush_threads
27 Why: Since pdflush is deprecated, the interface exported in /proc/sys/vm/
29 Who: Wanpeng Li <liwp@linux.vnet.ibm.com>
31 ---------------------------
33 What: CONFIG_APM_CPU_IDLE, and its ability to call APM BIOS in idle
35 Why: This optional sub-feature of APM is of dubious reliability,
36 and ancient APM laptops are likely better served by calling HLT.
37 Deleting CONFIG_APM_CPU_IDLE allows x86 to stop exporting
38 the pm_idle function pointer to modules.
39 Who: Len Brown <len.brown@intel.com>
41 ----------------------------
43 What: x86_32 "no-hlt" cmdline param
45 Why: remove a branch from idle path, simplify code used by everybody.
46 This option disabled the use of HLT in idle and machine_halt()
47 for hardware that was flakey 15-years ago. Today we have
48 "idle=poll" that removed HLT from idle, and so if such a machine
49 is still running the upstream kernel, "idle=poll" is likely sufficient.
50 Who: Len Brown <len.brown@intel.com>
52 ----------------------------
54 What: x86 "idle=mwait" cmdline param
56 Why: simplify x86 idle code
57 Who: Len Brown <len.brown@intel.com>
59 ----------------------------
64 Why: prism54 FullMAC PCI / Cardbus devices used to be supported only by the
65 prism54 wireless driver. After Intersil stopped selling these
66 devices in preference for the newer more flexible SoftMAC devices
67 a SoftMAC device driver was required and prism54 did not support
68 them. The p54pci driver now exists and has been present in the kernel for
69 a while. This driver supports both SoftMAC devices and FullMAC devices.
70 The main difference between these devices was the amount of memory which
71 could be used for the firmware. The SoftMAC devices support a smaller
72 amount of memory. Because of this the SoftMAC firmware fits into FullMAC
73 devices's memory. p54pci supports not only PCI / Cardbus but also USB
74 and SPI. Since p54pci supports all devices prism54 supports
75 you will have a conflict. I'm not quite sure how distributions are
76 handling this conflict right now. prism54 was kept around due to
77 claims users may experience issues when using the SoftMAC driver.
78 Time has passed users have not reported issues. If you use prism54
79 and for whatever reason you cannot use p54pci please let us know!
80 E-mail us at: linux-wireless@vger.kernel.org
82 For more information see the p54 wiki page:
84 http://wireless.kernel.org/en/users/Drivers/p54
86 Who: Luis R. Rodriguez <lrodriguez@atheros.com>
88 ---------------------------
90 What: The ieee80211_regdom module parameter
91 When: March 2010 / desktop catchup
93 Why: This was inherited by the CONFIG_WIRELESS_OLD_REGULATORY code,
94 and currently serves as an option for users to define an
95 ISO / IEC 3166 alpha2 code for the country they are currently
96 present in. Although there are userspace API replacements for this
97 through nl80211 distributions haven't yet caught up with implementing
98 decent alternatives through standard GUIs. Although available as an
99 option through iw or wpa_supplicant its just a matter of time before
100 distributions pick up good GUI options for this. The ideal solution
101 would actually consist of intelligent designs which would do this for
102 the user automatically even when travelling through different countries.
103 Until then we leave this module parameter as a compromise.
105 When userspace improves with reasonable widely-available alternatives for
106 this we will no longer need this module parameter. This entry hopes that
107 by the super-futuristically looking date of "March 2010" we will have
108 such replacements widely available.
110 Who: Luis R. Rodriguez <lrodriguez@atheros.com>
112 ---------------------------
114 What: dev->power.power_state
116 Why: Broken design for runtime control over driver power states, confusing
117 driver-internal runtime power management with: mechanisms to support
118 system-wide sleep state transitions; event codes that distinguish
119 different phases of swsusp "sleep" transitions; and userspace policy
120 inputs. This framework was never widely used, and most attempts to
121 use it were broken. Drivers should instead be exposing domain-specific
122 interfaces either to kernel or to userspace.
123 Who: Pavel Machek <pavel@ucw.cz>
125 ---------------------------
127 What: /proc/<pid>/oom_adj
129 Why: /proc/<pid>/oom_adj allows userspace to influence the oom killer's
130 badness heuristic used to determine which task to kill when the kernel
133 The badness heuristic has since been rewritten since the introduction of
134 this tunable such that its meaning is deprecated. The value was
135 implemented as a bitshift on a score generated by the badness()
136 function that did not have any precise units of measure. With the
137 rewrite, the score is given as a proportion of available memory to the
138 task allocating pages, so using a bitshift which grows the score
139 exponentially is, thus, impossible to tune with fine granularity.
141 A much more powerful interface, /proc/<pid>/oom_score_adj, was
142 introduced with the oom killer rewrite that allows users to increase or
143 decrease the badness score linearly. This interface will replace
146 A warning will be emitted to the kernel log if an application uses this
147 deprecated interface. After it is printed once, future warnings will be
148 suppressed until the kernel is rebooted.
150 ---------------------------
152 What: remove EXPORT_SYMBOL(kernel_thread)
154 Files: arch/*/kernel/*_ksyms.c
156 Why: kernel_thread is a low-level implementation detail. Drivers should
157 use the <linux/kthread.h> API instead which shields them from
158 implementation details and provides a higherlevel interface that
159 prevents bugs and code duplication
160 Who: Christoph Hellwig <hch@lst.de>
162 ---------------------------
164 What: Unused EXPORT_SYMBOL/EXPORT_SYMBOL_GPL exports
165 (temporary transition config option provided until then)
166 The transition config option will also be removed at the same time.
168 Why: Unused symbols are both increasing the size of the kernel binary
169 and are often a sign of "wrong API"
170 Who: Arjan van de Ven <arjan@linux.intel.com>
172 ---------------------------
174 What: PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment
176 Why: The stacking of class devices makes these values misleading and
178 Class devices should not carry any of these properties, and bus
179 devices have SUBSYTEM and DRIVER as a replacement.
180 Who: Kay Sievers <kay.sievers@suse.de>
182 ---------------------------
184 What: ACPI procfs interface
186 Why: ACPI sysfs conversion should be finished by January 2008.
187 ACPI procfs interface will be removed in July 2008 so that
188 there is enough time for the user space to catch up.
189 Who: Zhang Rui <rui.zhang@intel.com>
191 ---------------------------
193 What: CONFIG_ACPI_PROCFS_POWER
195 Why: sysfs I/F for ACPI power devices, including AC and Battery,
196 has been working in upstream kernel since 2.6.24, Sep 2007.
197 In 2.6.37, we make the sysfs I/F always built in and this option
199 Remove this option and the ACPI power procfs interface in 2.6.39.
200 Who: Zhang Rui <rui.zhang@intel.com>
202 ---------------------------
204 What: /proc/acpi/event
206 Why: /proc/acpi/event has been replaced by events via the input layer
207 and netlink since 2.6.23.
208 Who: Len Brown <len.brown@intel.com>
210 ---------------------------
212 What: i386/x86_64 bzImage symlinks
215 Why: The i386/x86_64 merge provides a symlink to the old bzImage
216 location so not yet updated user space tools, e.g. package
217 scripts, do not break.
218 Who: Thomas Gleixner <tglx@linutronix.de>
220 ---------------------------
222 What: GPIO autorequest on gpio_direction_{input,output}() in gpiolib
224 Why: All callers should use explicit gpio_request()/gpio_free().
225 The autorequest mechanism in gpiolib was provided mostly as a
226 migration aid for legacy GPIO interfaces (for SOC based GPIOs).
227 Those users have now largely migrated. Platforms implementing
228 the GPIO interfaces without using gpiolib will see no changes.
229 Who: David Brownell <dbrownell@users.sourceforge.net>
230 ---------------------------
232 What: b43 support for firmware revision < 410
233 When: The schedule was July 2008, but it was decided that we are going to keep the
234 code as long as there are no major maintanance headaches.
235 So it _could_ be removed _any_ time now, if it conflicts with something new.
236 Why: The support code for the old firmware hurts code readability/maintainability
237 and slightly hurts runtime performance. Bugfixes for the old firmware
238 are not provided by Broadcom anymore.
239 Who: Michael Buesch <m@bues.ch>
241 ---------------------------
243 What: Ability for non root users to shm_get hugetlb pages based on mlock
246 Why: Non root users need to be part of /proc/sys/vm/hugetlb_shm_group or
247 have CAP_IPC_LOCK to be able to allocate shm segments backed by
248 huge pages. The mlock based rlimit check to allow shm hugetlb is
249 inconsistent with mmap based allocations. Hence it is being
251 Who: Ravikiran Thirumalai <kiran@scalex86.org>
253 ---------------------------
255 What: sysfs ui for changing p4-clockmod parameters
257 Why: See commits 129f8ae9b1b5be94517da76009ea956e89104ce8 and
258 e088e4c9cdb618675874becb91b2fd581ee707e6.
259 Removal is subject to fixing any remaining bugs in ACPI which may
260 cause the thermal throttling not to happen at the right time.
261 Who: Dave Jones <davej@redhat.com>, Matthew Garrett <mjg@redhat.com>
263 -----------------------------
265 What: fakephp and associated sysfs files in /sys/bus/pci/slots/
267 Why: In 2.6.27, the semantics of /sys/bus/pci/slots was redefined to
268 represent a machine's physical PCI slots. The change in semantics
269 had userspace implications, as the hotplug core no longer allowed
270 drivers to create multiple sysfs files per physical slot (required
271 for multi-function devices, e.g.). fakephp was seen as a developer's
272 tool only, and its interface changed. Too late, we learned that
273 there were some users of the fakephp interface.
275 In 2.6.30, the original fakephp interface was restored. At the same
276 time, the PCI core gained the ability that fakephp provided, namely
277 function-level hot-remove and hot-add.
279 Since the PCI core now provides the same functionality, exposed in:
282 /sys/bus/pci/devices/.../remove
283 /sys/bus/pci/devices/.../rescan
285 there is no functional reason to maintain fakephp as well.
287 We will keep the existing module so that 'modprobe fakephp' will
288 present the old /sys/bus/pci/slots/... interface for compatibility,
289 but users are urged to migrate their applications to the API above.
291 After a reasonable transition period, we will remove the legacy
293 Who: Alex Chiang <achiang@hp.com>
295 ---------------------------
297 What: CONFIG_RFKILL_INPUT
299 Why: Should be implemented in userspace, policy daemon.
300 Who: Johannes Berg <johannes@sipsolutions.net>
302 ----------------------------
304 What: sound-slot/service-* module aliases and related clutters in
307 Why: OSS sound_core grabs all legacy minors (0-255) of SOUND_MAJOR
308 (14) and requests modules using custom sound-slot/service-*
309 module aliases. The only benefit of doing this is allowing
310 use of custom module aliases which might as well be considered
311 a bug at this point. This preemptive claiming prevents
312 alternative OSS implementations.
314 Till the feature is removed, the kernel will be requesting
315 both sound-slot/service-* and the standard char-major-* module
316 aliases and allow turning off the pre-claiming selectively via
317 CONFIG_SOUND_OSS_CORE_PRECLAIM and soundcore.preclaim_oss
320 After the transition phase is complete, both the custom module
321 aliases and switches to disable it will go away. This removal
322 will also allow making ALSA OSS emulation independent of
323 sound_core. The dependency will be broken then too.
324 Who: Tejun Heo <tj@kernel.org>
326 ----------------------------
328 What: sysfs-class-rfkill state file
330 Files: net/rfkill/core.c
331 Why: Documented as obsolete since Feb 2010. This file is limited to 3
332 states while the rfkill drivers can have 4 states.
333 Who: anybody or Florian Mickler <florian@mickler.org>
335 ----------------------------
337 What: sysfs-class-rfkill claim file
339 Files: net/rfkill/core.c
340 Why: It is not possible to claim an rfkill driver since 2007. This is
341 Documented as obsolete since Feb 2010.
342 Who: anybody or Florian Mickler <florian@mickler.org>
344 ----------------------------
346 What: iwlwifi 50XX module parameters
348 Why: The "..50" modules parameters were used to configure 5000 series and
349 up devices; different set of module parameters also available for 4965
350 with same functionalities. Consolidate both set into single place
351 in drivers/net/wireless/iwlwifi/iwl-agn.c
353 Who: Wey-Yi Guy <wey-yi.w.guy@intel.com>
355 ----------------------------
357 What: iwl4965 alias support
359 Why: Internal alias support has been present in module-init-tools for some
360 time, the MODULE_ALIAS("iwl4965") boilerplate aliases can be removed
363 Who: Wey-Yi Guy <wey-yi.w.guy@intel.com>
365 ---------------------------
368 Files: net/netfilter/xt_NOTRACK.c
370 Why: Superseded by xt_CT
371 Who: Netfilter developer team <netfilter-devel@vger.kernel.org>
373 ----------------------------
377 Why: The flag is a NOOP as we run interrupt handlers with interrupts disabled
378 Who: Thomas Gleixner <tglx@linutronix.de>
380 ----------------------------
382 What: PCI DMA unmap state API
384 Why: PCI DMA unmap state API (include/linux/pci-dma.h) was replaced
385 with DMA unmap state API (DMA unmap state API can be used for
387 Who: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
389 ----------------------------
391 What: iwlwifi disable_hw_scan module parameters
393 Why: Hareware scan is the prefer method for iwlwifi devices for
394 scanning operation. Remove software scan support for all the
397 Who: Wey-Yi Guy <wey-yi.w.guy@intel.com>
399 ----------------------------
401 What: Legacy, non-standard chassis intrusion detection interface.
403 Why: The adm9240, w83792d and w83793 hardware monitoring drivers have
404 legacy interfaces for chassis intrusion detection. A standard
405 interface has been added to each driver, so the legacy interface
407 Who: Jean Delvare <khali@linux-fr.org>
409 ----------------------------
411 What: i2c_driver.attach_adapter
412 i2c_driver.detach_adapter
414 Why: These legacy callbacks should no longer be used as i2c-core offers
415 a variety of preferable alternative ways to instantiate I2C devices.
416 Who: Jean Delvare <khali@linux-fr.org>
418 ----------------------------
420 What: Opening a radio device node will no longer automatically switch the
421 tuner mode from tv to radio.
423 Why: Just opening a V4L device should not change the state of the hardware
424 like that. It's very unexpected and against the V4L spec. Instead, you
425 switch to radio mode by calling VIDIOC_S_FREQUENCY. This is the second
426 and last step of the move to consistent handling of tv and radio tuners.
427 Who: Hans Verkuil <hans.verkuil@cisco.com>
429 ----------------------------
431 What: CONFIG_CFG80211_WEXT
432 When: as soon as distributions ship new wireless tools, ie. wpa_supplicant 1.0
433 and NetworkManager/connman/etc. that are able to use nl80211
434 Why: Wireless extensions are deprecated, and userland tools are moving to
435 using nl80211. New drivers are no longer using wireless extensions,
436 and while there might still be old drivers, both new drivers and new
437 userland no longer needs them and they can't be used for an feature
438 developed in the past couple of years. As such, compatibility with
439 wireless extensions in new drivers will be removed.
440 Who: Johannes Berg <johannes@sipsolutions.net>
442 ----------------------------
444 What: g_file_storage driver
446 Why: This driver has been superseded by g_mass_storage.
447 Who: Alan Stern <stern@rowland.harvard.edu>
449 ----------------------------
451 What: threeg and interface sysfs files in /sys/devices/platform/acer-wmi
453 Why: In 3.0, we can now autodetect internal 3G device and already have
454 the threeg rfkill device. So, we plan to remove threeg sysfs support
455 for it's no longer necessary.
457 We also plan to remove interface sysfs file that exposed which ACPI-WMI
458 interface that was used by acer-wmi driver. It will replaced by
459 information log when acer-wmi initial.
460 Who: Lee, Chun-Yi <jlee@novell.com>
462 ---------------------------
464 What: /sys/devices/platform/_UDC_/udc/_UDC_/is_dualspeed file and
465 is_dualspeed line in /sys/devices/platform/ci13xxx_*/udc/device file.
467 Why: The is_dualspeed file is superseded by maximum_speed in the same
468 directory and is_dualspeed line in device file is superseded by
469 max_speed line in the same file.
471 The maximum_speed/max_speed specifies maximum speed supported by UDC.
472 To check if dualspeeed is supported, check if the value is >= 3.
473 Various possible speeds are defined in <linux/usb/ch9.h>.
474 Who: Michal Nazarewicz <mina86@mina86.com>
476 ----------------------------
478 What: The XFS nodelaylog mount option
480 Why: The delaylog mode that has been the default since 2.6.39 has proven
481 stable, and the old code is in the way of additional improvements in
483 Who: Christoph Hellwig <hch@lst.de>
485 ----------------------------
487 What: iwlagn alias support
489 Why: The iwlagn module has been renamed iwlwifi. The alias will be around
490 for backward compatibility for several cycles and then dropped.
491 Who: Don Fry <donald.h.fry@intel.com>
493 ----------------------------
495 What: pci_scan_bus_parented()
497 Why: The pci_scan_bus_parented() interface creates a new root bus. The
498 bus is created with default resources (ioport_resource and
499 iomem_resource) that are always wrong, so we rely on arch code to
500 correct them later. Callers of pci_scan_bus_parented() should
501 convert to using pci_scan_root_bus() so they can supply a list of
502 bus resources when the bus is created.
503 Who: Bjorn Helgaas <bhelgaas@google.com>
505 ----------------------------
507 What: Low Performance USB Block driver ("CONFIG_BLK_DEV_UB")
509 Why: This driver provides support for USB storage devices like "USB
510 sticks". As of now, it is deactivated in Debian, Fedora and
511 Ubuntu. All current users can switch over to usb-storage
512 (CONFIG_USB_STORAGE) which only drawback is the additional SCSI
514 Who: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
516 ----------------------------
518 What: get_robust_list syscall
520 Why: There appear to be no production users of the get_robust_list syscall,
521 and it runs the risk of leaking address locations, allowing the bypass
522 of ASLR. It was only ever intended for debugging, so it should be
524 Who: Kees Cook <keescook@chromium.org>
526 ----------------------------
528 What: Removing the pn544 raw driver.
530 Why: With the introduction of the NFC HCI and SHDL kernel layers, pn544.c
531 is being replaced by pn544_hci.c which is accessible through the netlink
532 and socket NFC APIs. Moreover, pn544.c is outdated and does not seem to
533 work properly with the latest Android stacks.
534 Having 2 drivers for the same hardware is confusing and as such we
535 should only keep the one following the kernel NFC APIs.
536 Who: Samuel Ortiz <sameo@linux.intel.com>
538 ----------------------------
540 What: setitimer accepts user NULL pointer (value)
542 Why: setitimer is not returning -EFAULT if user pointer is NULL. This
544 Who: Sasikantha Babu <sasikanth.v19@gmail.com>
546 ----------------------------
548 What: remove bogus DV presets V4L2_DV_1080I29_97, V4L2_DV_1080I30 and
551 Why: These HDTV formats do not exist and were added by a confused mind
552 (that was me, to be precise...)
553 Who: Hans Verkuil <hans.verkuil@cisco.com>
555 ----------------------------
557 What: V4L2_CID_HCENTER, V4L2_CID_VCENTER V4L2 controls
559 Why: The V4L2_CID_VCENTER, V4L2_CID_HCENTER controls have been deprecated
560 for about 4 years and they are not used by any mainline driver.
561 There are newer controls (V4L2_CID_PAN*, V4L2_CID_TILT*) that provide
562 similar functionality.
563 Who: Sylwester Nawrocki <sylvester.nawrocki@gmail.com>
565 ----------------------------
567 What: cgroup option updates via remount
569 Why: Remount currently allows changing bound subsystems and
570 release_agent. Rebinding is hardly useful as it only works
571 when the hierarchy is empty and release_agent itself should be
572 replaced with conventional fsnotify.
574 ----------------------------
576 What: xt_recent rev 0
578 Who: Pablo Neira Ayuso <pablo@netfilter.org>
579 Files: net/netfilter/xt_recent.c
581 ----------------------------
583 What: KVM debugfs statistics
585 Why: KVM tracepoints provide mostly equivalent information in a much more
588 ----------------------------
590 What: at91-mci driver ("CONFIG_MMC_AT91")
592 Why: There are two mci drivers: at91-mci and atmel-mci. The PDC support
593 was added to atmel-mci as a first step to support more chips.
594 Then at91-mci was kept only for old IP versions (on at91rm9200 and
595 at91sam9261). The support of these IP versions has just been added
596 to atmel-mci, so atmel-mci can be used for all chips.
597 Who: Ludovic Desroches <ludovic.desroches@atmel.com>
599 ----------------------------
603 Why: Unsupported/unmaintained/unused since 2.6
605 ----------------------------
607 What: V4L2 selections API target rectangle and flags unification, the
608 following definitions will be removed: V4L2_SEL_TGT_CROP_ACTIVE,
609 V4L2_SEL_TGT_COMPOSE_ACTIVE, V4L2_SUBDEV_SEL_*, V4L2_SUBDEV_SEL_FLAG_*
610 in favor of common V4L2_SEL_TGT_* and V4L2_SEL_FLAG_* definitions.
611 For more details see include/linux/v4l2-common.h.
613 Why: The regular V4L2 selections and the subdev selection API originally
614 defined distinct names for the target rectangles and flags - V4L2_SEL_*
615 and V4L2_SUBDEV_SEL_*. Although, it turned out that the meaning of these
616 target rectangles is virtually identical and the APIs were consolidated
617 to use single set of names - V4L2_SEL_*. This didn't involve any ABI
618 changes. Alias definitions were created for the original ones to avoid
619 any instabilities in the user space interface. After few cycles these
620 backward compatibility definitions will be removed.
621 Who: Sylwester Nawrocki <sylvester.nawrocki@gmail.com>
623 ----------------------------
625 What: Using V4L2_CAP_VIDEO_CAPTURE and V4L2_CAP_VIDEO_OUTPUT flags
626 to indicate a V4L2 memory-to-memory device capability
628 Why: New drivers should use new V4L2_CAP_VIDEO_M2M capability flag
629 to indicate a V4L2 video memory-to-memory (M2M) device and
630 applications can now identify a M2M video device by checking
631 for V4L2_CAP_VIDEO_M2M, with VIDIOC_QUERYCAP ioctl. Using ORed
632 V4L2_CAP_VIDEO_CAPTURE and V4L2_CAP_VIDEO_OUTPUT flags for M2M
633 devices is ambiguous and may lead, for example, to identifying
634 a M2M device as a video capture or output device.
635 Who: Sylwester Nawrocki <s.nawrocki@samsung.com>
637 ----------------------------
639 What: OMAP private DMA implementation
641 Why: We have a DMA engine implementation; all users should be updated
642 to use this rather than persisting with the old APIs. The old APIs
643 block merging the old DMA engine implementation into the DMA
645 Who: Russell King <linux@arm.linux.org.uk>,
646 Santosh Shilimkar <santosh.shilimkar@ti.com>
648 ----------------------------
650 What: Remove deprecated DV timings capability defines V4L2_IN_CAP_CUSTOM_TIMINGS
651 and V4L2_OUT_CAP_CUSTOM_TIMINGS.
653 Why: These defines have been replaced by V4L2_IN_CAP_TIMINGS and
654 V4L2_OUT_CAP_TIMINGS respectively.
655 Who: Hans Verkuil <hans.verkuil@cisco.com>
657 ----------------------------