| blob | 1bea46a54b1ca252b5390c422f291dafe41a7c32 |
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: 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>
15 ---------------------------
17 What: CONFIG_APM_CPU_IDLE, and its ability to call APM BIOS in idle
18 When: 2012
19 Why: This optional sub-feature of APM is of dubious reliability,
20 and ancient APM laptops are likely better served by calling HLT.
21 Deleting CONFIG_APM_CPU_IDLE allows x86 to stop exporting
22 the pm_idle function pointer to modules.
23 Who: Len Brown <len.brown@intel.com>
25 ----------------------------
27 What: x86_32 "no-hlt" cmdline param
28 When: 2012
29 Why: remove a branch from idle path, simplify code used by everybody.
30 This option disabled the use of HLT in idle and machine_halt()
31 for hardware that was flakey 15-years ago. Today we have
32 "idle=poll" that removed HLT from idle, and so if such a machine
33 is still running the upstream kernel, "idle=poll" is likely sufficient.
34 Who: Len Brown <len.brown@intel.com>
36 ----------------------------
38 What: x86 "idle=mwait" cmdline param
39 When: 2012
40 Why: simplify x86 idle code
41 Who: Len Brown <len.brown@intel.com>
43 ----------------------------
45 What: PRISM54
46 When: 2.6.34
48 Why: prism54 FullMAC PCI / Cardbus devices used to be supported only by the
49 prism54 wireless driver. After Intersil stopped selling these
50 devices in preference for the newer more flexible SoftMAC devices
51 a SoftMAC device driver was required and prism54 did not support
52 them. The p54pci driver now exists and has been present in the kernel for
53 a while. This driver supports both SoftMAC devices and FullMAC devices.
54 The main difference between these devices was the amount of memory which
55 could be used for the firmware. The SoftMAC devices support a smaller
56 amount of memory. Because of this the SoftMAC firmware fits into FullMAC
57 devices's memory. p54pci supports not only PCI / Cardbus but also USB
58 and SPI. Since p54pci supports all devices prism54 supports
59 you will have a conflict. I'm not quite sure how distributions are
60 handling this conflict right now. prism54 was kept around due to
61 claims users may experience issues when using the SoftMAC driver.
62 Time has passed users have not reported issues. If you use prism54
63 and for whatever reason you cannot use p54pci please let us know!
64 E-mail us at: linux-wireless@vger.kernel.org
66 For more information see the p54 wiki page:
68 http://wireless.kernel.org/en/users/Drivers/p54
70 Who: Luis R. Rodriguez <lrodriguez@atheros.com>
72 ---------------------------
74 What: IRQF_SAMPLE_RANDOM
75 Check: IRQF_SAMPLE_RANDOM
76 When: July 2009
78 Why: Many of IRQF_SAMPLE_RANDOM users are technically bogus as entropy
79 sources in the kernel's current entropy model. To resolve this, every
80 input point to the kernel's entropy pool needs to better document the
81 type of entropy source it actually is. This will be replaced with
82 additional add_*_randomness functions in drivers/char/random.c
84 Who: Robin Getz <rgetz@blackfin.uclinux.org> & Matt Mackall <mpm@selenic.com>
86 ---------------------------
88 What: The ieee80211_regdom module parameter
89 When: March 2010 / desktop catchup
91 Why: This was inherited by the CONFIG_WIRELESS_OLD_REGULATORY code,
92 and currently serves as an option for users to define an
93 ISO / IEC 3166 alpha2 code for the country they are currently
94 present in. Although there are userspace API replacements for this
95 through nl80211 distributions haven't yet caught up with implementing
96 decent alternatives through standard GUIs. Although available as an
97 option through iw or wpa_supplicant its just a matter of time before
98 distributions pick up good GUI options for this. The ideal solution
99 would actually consist of intelligent designs which would do this for
100 the user automatically even when travelling through different countries.
101 Until then we leave this module parameter as a compromise.
103 When userspace improves with reasonable widely-available alternatives for
104 this we will no longer need this module parameter. This entry hopes that
105 by the super-futuristically looking date of "March 2010" we will have
106 such replacements widely available.
108 Who: Luis R. Rodriguez <lrodriguez@atheros.com>
110 ---------------------------
112 What: dev->power.power_state
113 When: July 2007
114 Why: Broken design for runtime control over driver power states, confusing
115 driver-internal runtime power management with: mechanisms to support
116 system-wide sleep state transitions; event codes that distinguish
117 different phases of swsusp "sleep" transitions; and userspace policy
118 inputs. This framework was never widely used, and most attempts to
119 use it were broken. Drivers should instead be exposing domain-specific
120 interfaces either to kernel or to userspace.
121 Who: Pavel Machek <pavel@ucw.cz>
123 ---------------------------
125 What: /proc/<pid>/oom_adj
126 When: August 2012
127 Why: /proc/<pid>/oom_adj allows userspace to influence the oom killer's
128 badness heuristic used to determine which task to kill when the kernel
129 is out of memory.
131 The badness heuristic has since been rewritten since the introduction of
132 this tunable such that its meaning is deprecated. The value was
133 implemented as a bitshift on a score generated by the badness()
134 function that did not have any precise units of measure. With the
135 rewrite, the score is given as a proportion of available memory to the
136 task allocating pages, so using a bitshift which grows the score
137 exponentially is, thus, impossible to tune with fine granularity.
139 A much more powerful interface, /proc/<pid>/oom_score_adj, was
140 introduced with the oom killer rewrite that allows users to increase or
141 decrease the badness score linearly. This interface will replace
142 /proc/<pid>/oom_adj.
144 A warning will be emitted to the kernel log if an application uses this
145 deprecated interface. After it is printed once, future warnings will be
146 suppressed until the kernel is rebooted.
148 ---------------------------
150 What: remove EXPORT_SYMBOL(kernel_thread)
151 When: August 2006
152 Files: arch/*/kernel/*_ksyms.c
153 Check: kernel_thread
154 Why: kernel_thread is a low-level implementation detail. Drivers should
155 use the <linux/kthread.h> API instead which shields them from
156 implementation details and provides a higherlevel interface that
157 prevents bugs and code duplication
158 Who: Christoph Hellwig <hch@lst.de>
160 ---------------------------
162 What: Unused EXPORT_SYMBOL/EXPORT_SYMBOL_GPL exports
163 (temporary transition config option provided until then)
164 The transition config option will also be removed at the same time.
165 When: before 2.6.19
166 Why: Unused symbols are both increasing the size of the kernel binary
167 and are often a sign of "wrong API"
168 Who: Arjan van de Ven <arjan@linux.intel.com>
170 ---------------------------
172 What: PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment
173 When: October 2008
174 Why: The stacking of class devices makes these values misleading and
175 inconsistent.
176 Class devices should not carry any of these properties, and bus
177 devices have SUBSYTEM and DRIVER as a replacement.
178 Who: Kay Sievers <kay.sievers@suse.de>
180 ---------------------------
182 What: ACPI procfs interface
183 When: July 2008
184 Why: ACPI sysfs conversion should be finished by January 2008.
185 ACPI procfs interface will be removed in July 2008 so that
186 there is enough time for the user space to catch up.
187 Who: Zhang Rui <rui.zhang@intel.com>
189 ---------------------------
191 What: CONFIG_ACPI_PROCFS_POWER
192 When: 2.6.39
193 Why: sysfs I/F for ACPI power devices, including AC and Battery,
194 has been working in upstream kernel since 2.6.24, Sep 2007.
195 In 2.6.37, we make the sysfs I/F always built in and this option
196 disabled by default.
197 Remove this option and the ACPI power procfs interface in 2.6.39.
198 Who: Zhang Rui <rui.zhang@intel.com>
200 ---------------------------
202 What: /proc/acpi/event
203 When: February 2008
204 Why: /proc/acpi/event has been replaced by events via the input layer
205 and netlink since 2.6.23.
206 Who: Len Brown <len.brown@intel.com>
208 ---------------------------
210 What: i386/x86_64 bzImage symlinks
211 When: April 2010
213 Why: The i386/x86_64 merge provides a symlink to the old bzImage
214 location so not yet updated user space tools, e.g. package
215 scripts, do not break.
216 Who: Thomas Gleixner <tglx@linutronix.de>
218 ---------------------------
220 What: GPIO autorequest on gpio_direction_{input,output}() in gpiolib
221 When: February 2010
222 Why: All callers should use explicit gpio_request()/gpio_free().
223 The autorequest mechanism in gpiolib was provided mostly as a
224 migration aid for legacy GPIO interfaces (for SOC based GPIOs).
225 Those users have now largely migrated. Platforms implementing
226 the GPIO interfaces without using gpiolib will see no changes.
227 Who: David Brownell <dbrownell@users.sourceforge.net>
228 ---------------------------
230 What: b43 support for firmware revision < 410
231 When: The schedule was July 2008, but it was decided that we are going to keep the
232 code as long as there are no major maintanance headaches.
233 So it _could_ be removed _any_ time now, if it conflicts with something new.
234 Why: The support code for the old firmware hurts code readability/maintainability
235 and slightly hurts runtime performance. Bugfixes for the old firmware
236 are not provided by Broadcom anymore.
237 Who: Michael Buesch <m@bues.ch>
239 ---------------------------
241 What: Ability for non root users to shm_get hugetlb pages based on mlock
242 resource limits
243 When: 2.6.31
244 Why: Non root users need to be part of /proc/sys/vm/hugetlb_shm_group or
245 have CAP_IPC_LOCK to be able to allocate shm segments backed by
246 huge pages. The mlock based rlimit check to allow shm hugetlb is
247 inconsistent with mmap based allocations. Hence it is being
248 deprecated.
249 Who: Ravikiran Thirumalai <kiran@scalex86.org>
251 ---------------------------
253 What: Code that is now under CONFIG_WIRELESS_EXT_SYSFS
254 (in net/core/net-sysfs.c)
255 When: 3.5
256 Why: Over 1K .text/.data size reduction, data is available in other
257 ways (ioctls)
258 Who: Johannes Berg <johannes@sipsolutions.net>
260 ---------------------------
262 What: sysfs ui for changing p4-clockmod parameters
263 When: September 2009
264 Why: See commits 129f8ae9b1b5be94517da76009ea956e89104ce8 and
265 e088e4c9cdb618675874becb91b2fd581ee707e6.
266 Removal is subject to fixing any remaining bugs in ACPI which may
267 cause the thermal throttling not to happen at the right time.
268 Who: Dave Jones <davej@redhat.com>, Matthew Garrett <mjg@redhat.com>
270 -----------------------------
272 What: fakephp and associated sysfs files in /sys/bus/pci/slots/
273 When: 2011
274 Why: In 2.6.27, the semantics of /sys/bus/pci/slots was redefined to
275 represent a machine's physical PCI slots. The change in semantics
276 had userspace implications, as the hotplug core no longer allowed
277 drivers to create multiple sysfs files per physical slot (required
278 for multi-function devices, e.g.). fakephp was seen as a developer's
279 tool only, and its interface changed. Too late, we learned that
280 there were some users of the fakephp interface.
282 In 2.6.30, the original fakephp interface was restored. At the same
283 time, the PCI core gained the ability that fakephp provided, namely
284 function-level hot-remove and hot-add.
286 Since the PCI core now provides the same functionality, exposed in:
288 /sys/bus/pci/rescan
289 /sys/bus/pci/devices/.../remove
290 /sys/bus/pci/devices/.../rescan
292 there is no functional reason to maintain fakephp as well.
294 We will keep the existing module so that 'modprobe fakephp' will
295 present the old /sys/bus/pci/slots/... interface for compatibility,
296 but users are urged to migrate their applications to the API above.
298 After a reasonable transition period, we will remove the legacy
299 fakephp interface.
300 Who: Alex Chiang <achiang@hp.com>
302 ---------------------------
304 What: CONFIG_RFKILL_INPUT
305 When: 2.6.33
306 Why: Should be implemented in userspace, policy daemon.
307 Who: Johannes Berg <johannes@sipsolutions.net>
309 ----------------------------
311 What: sound-slot/service-* module aliases and related clutters in
312 sound/sound_core.c
313 When: August 2010
314 Why: OSS sound_core grabs all legacy minors (0-255) of SOUND_MAJOR
315 (14) and requests modules using custom sound-slot/service-*
316 module aliases. The only benefit of doing this is allowing
317 use of custom module aliases which might as well be considered
318 a bug at this point. This preemptive claiming prevents
319 alternative OSS implementations.
321 Till the feature is removed, the kernel will be requesting
322 both sound-slot/service-* and the standard char-major-* module
323 aliases and allow turning off the pre-claiming selectively via
324 CONFIG_SOUND_OSS_CORE_PRECLAIM and soundcore.preclaim_oss
325 kernel parameter.
327 After the transition phase is complete, both the custom module
328 aliases and switches to disable it will go away. This removal
329 will also allow making ALSA OSS emulation independent of
330 sound_core. The dependency will be broken then too.
331 Who: Tejun Heo <tj@kernel.org>
333 ----------------------------
335 What: sysfs-class-rfkill state file
336 When: Feb 2014
337 Files: net/rfkill/core.c
338 Why: Documented as obsolete since Feb 2010. This file is limited to 3
339 states while the rfkill drivers can have 4 states.
340 Who: anybody or Florian Mickler <florian@mickler.org>
342 ----------------------------
344 What: sysfs-class-rfkill claim file
345 When: Feb 2012
346 Files: net/rfkill/core.c
347 Why: It is not possible to claim an rfkill driver since 2007. This is
348 Documented as obsolete since Feb 2010.
349 Who: anybody or Florian Mickler <florian@mickler.org>
351 ----------------------------
353 What: iwlwifi 50XX module parameters
354 When: 3.0
355 Why: The "..50" modules parameters were used to configure 5000 series and
356 up devices; different set of module parameters also available for 4965
357 with same functionalities. Consolidate both set into single place
358 in drivers/net/wireless/iwlwifi/iwl-agn.c
360 Who: Wey-Yi Guy <wey-yi.w.guy@intel.com>
362 ----------------------------
364 What: iwl4965 alias support
365 When: 3.0
366 Why: Internal alias support has been present in module-init-tools for some
367 time, the MODULE_ALIAS("iwl4965") boilerplate aliases can be removed
368 with no impact.
370 Who: Wey-Yi Guy <wey-yi.w.guy@intel.com>
372 ---------------------------
374 What: xt_NOTRACK
375 Files: net/netfilter/xt_NOTRACK.c
376 When: April 2011
377 Why: Superseded by xt_CT
378 Who: Netfilter developer team <netfilter-devel@vger.kernel.org>
380 ----------------------------
382 What: IRQF_DISABLED
383 When: 2.6.36
384 Why: The flag is a NOOP as we run interrupt handlers with interrupts disabled
385 Who: Thomas Gleixner <tglx@linutronix.de>
387 ----------------------------
389 What: PCI DMA unmap state API
390 When: August 2012
391 Why: PCI DMA unmap state API (include/linux/pci-dma.h) was replaced
392 with DMA unmap state API (DMA unmap state API can be used for
393 any bus).
394 Who: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
396 ----------------------------
398 What: iwlwifi disable_hw_scan module parameters
399 When: 3.0
400 Why: Hareware scan is the prefer method for iwlwifi devices for
401 scanning operation. Remove software scan support for all the
402 iwlwifi devices.
404 Who: Wey-Yi Guy <wey-yi.w.guy@intel.com>
406 ----------------------------
408 What: Legacy, non-standard chassis intrusion detection interface.
409 When: June 2011
410 Why: The adm9240, w83792d and w83793 hardware monitoring drivers have
411 legacy interfaces for chassis intrusion detection. A standard
412 interface has been added to each driver, so the legacy interface
413 can be removed.
414 Who: Jean Delvare <khali@linux-fr.org>
416 ----------------------------
418 What: xt_connlimit rev 0
419 When: 2012
420 Who: Jan Engelhardt <jengelh@medozas.de>
421 Files: net/netfilter/xt_connlimit.c
423 ----------------------------
425 What: ipt_addrtype match include file
426 When: 2012
427 Why: superseded by xt_addrtype
428 Who: Florian Westphal <fw@strlen.de>
429 Files: include/linux/netfilter_ipv4/ipt_addrtype.h
431 ----------------------------
433 What: i2c_driver.attach_adapter
434 i2c_driver.detach_adapter
435 When: September 2011
436 Why: These legacy callbacks should no longer be used as i2c-core offers
437 a variety of preferable alternative ways to instantiate I2C devices.
438 Who: Jean Delvare <khali@linux-fr.org>
440 ----------------------------
442 What: Opening a radio device node will no longer automatically switch the
443 tuner mode from tv to radio.
444 When: 3.3
445 Why: Just opening a V4L device should not change the state of the hardware
446 like that. It's very unexpected and against the V4L spec. Instead, you
447 switch to radio mode by calling VIDIOC_S_FREQUENCY. This is the second
448 and last step of the move to consistent handling of tv and radio tuners.
449 Who: Hans Verkuil <hans.verkuil@cisco.com>
451 ----------------------------
453 What: g_file_storage driver
454 When: 3.8
455 Why: This driver has been superseded by g_mass_storage.
456 Who: Alan Stern <stern@rowland.harvard.edu>
458 ----------------------------
460 What: threeg and interface sysfs files in /sys/devices/platform/acer-wmi
461 When: 2012
462 Why: In 3.0, we can now autodetect internal 3G device and already have
463 the threeg rfkill device. So, we plan to remove threeg sysfs support
464 for it's no longer necessary.
466 We also plan to remove interface sysfs file that exposed which ACPI-WMI
467 interface that was used by acer-wmi driver. It will replaced by
468 information log when acer-wmi initial.
469 Who: Lee, Chun-Yi <jlee@novell.com>
471 ---------------------------
473 What: /sys/devices/platform/_UDC_/udc/_UDC_/is_dualspeed file and
474 is_dualspeed line in /sys/devices/platform/ci13xxx_*/udc/device file.
475 When: 3.8
476 Why: The is_dualspeed file is superseded by maximum_speed in the same
477 directory and is_dualspeed line in device file is superseded by
478 max_speed line in the same file.
480 The maximum_speed/max_speed specifies maximum speed supported by UDC.
481 To check if dualspeeed is supported, check if the value is >= 3.
482 Various possible speeds are defined in <linux/usb/ch9.h>.
483 Who: Michal Nazarewicz <mina86@mina86.com>
485 ----------------------------
487 What: The XFS nodelaylog mount option
488 When: 3.3
489 Why: The delaylog mode that has been the default since 2.6.39 has proven
490 stable, and the old code is in the way of additional improvements in
491 the log code.
492 Who: Christoph Hellwig <hch@lst.de>
494 ----------------------------
496 What: iwlagn alias support
497 When: 3.5
498 Why: The iwlagn module has been renamed iwlwifi. The alias will be around
499 for backward compatibility for several cycles and then dropped.
500 Who: Don Fry <donald.h.fry@intel.com>
502 ----------------------------
504 What: pci_scan_bus_parented()
505 When: 3.5
506 Why: The pci_scan_bus_parented() interface creates a new root bus. The
507 bus is created with default resources (ioport_resource and
508 iomem_resource) that are always wrong, so we rely on arch code to
509 correct them later. Callers of pci_scan_bus_parented() should
510 convert to using pci_scan_root_bus() so they can supply a list of
511 bus resources when the bus is created.
512 Who: Bjorn Helgaas <bhelgaas@google.com>