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: old static regulatory information and ieee80211_regdom module parameter
11 Why: The old regulatory infrastructure has been replaced with a new one
12 which does not require statically defined regulatory domains. We do
13 not want to keep static regulatory domains in the kernel due to the
14 the dynamic nature of regulatory law and localization. We kept around
15 the old static definitions for the regulatory domains of:
19 and used by default the US when CONFIG_WIRELESS_OLD_REGULATORY was
20 set. We also kept around the ieee80211_regdom module parameter in case
21 some applications were relying on it. Changing regulatory domains
22 can now be done instead by using nl80211, as is done with iw.
23 Who: Luis R. Rodriguez <lrodriguez@atheros.com>
25 ---------------------------
27 What: dev->power.power_state
29 Why: Broken design for runtime control over driver power states, confusing
30 driver-internal runtime power management with: mechanisms to support
31 system-wide sleep state transitions; event codes that distinguish
32 different phases of swsusp "sleep" transitions; and userspace policy
33 inputs. This framework was never widely used, and most attempts to
34 use it were broken. Drivers should instead be exposing domain-specific
35 interfaces either to kernel or to userspace.
36 Who: Pavel Machek <pavel@suse.cz>
38 ---------------------------
40 What: Video4Linux API 1 ioctls and video_decoder.h from Video devices.
42 Files: include/linux/video_decoder.h include/linux/videodev.h
43 Check: include/linux/video_decoder.h include/linux/videodev.h
44 Why: V4L1 AP1 was replaced by V4L2 API during migration from 2.4 to 2.6
45 series. The old API have lots of drawbacks and don't provide enough
46 means to work with all video and audio standards. The newer API is
47 already available on the main drivers and should be used instead.
48 Newer drivers should use v4l_compat_translate_ioctl function to handle
49 old calls, replacing to newer ones.
50 Decoder iocts are using internally to allow video drivers to
51 communicate with video decoders. This should also be improved to allow
52 V4L2 calls being translated into compatible internal ioctls.
53 Compatibility ioctls will be provided, for a while, via
55 Who: Mauro Carvalho Chehab <mchehab@infradead.org>
57 ---------------------------
59 What: PCMCIA control ioctl (needed for pcmcia-cs [cardmgr, cardctl])
61 Files: drivers/pcmcia/: pcmcia_ioctl.c
62 Why: With the 16-bit PCMCIA subsystem now behaving (almost) like a
63 normal hotpluggable bus, and with it using the default kernel
64 infrastructure (hotplug, driver core, sysfs) keeping the PCMCIA
65 control ioctl needed by cardmgr and cardctl from pcmcia-cs is
66 unnecessary, and makes further cleanups and integration of the
67 PCMCIA subsystem into the Linux kernel device driver model more
68 difficult. The features provided by cardmgr and cardctl are either
69 handled by the kernel itself now or are available in the new
70 pcmciautils package available at
71 http://kernel.org/pub/linux/utils/kernel/pcmcia/
72 Who: Dominik Brodowski <linux@brodo.de>
74 ---------------------------
78 Option: CONFIG_SYSCTL_SYSCALL
79 Why: The same information is available in a more convenient from
80 /proc/sys, and none of the sysctl variables appear to be
81 important performance wise.
83 Binary sysctls are a long standing source of subtle kernel
84 bugs and security issues.
86 When I looked several months ago all I could find after
87 searching several distributions were 5 user space programs and
88 glibc (which falls back to /proc/sys) using this syscall.
90 The man page for sysctl(2) documents it as unusable for user
93 sysctl(2) is not generally ABI compatible to a 32bit user
94 space application on a 64bit and a 32bit kernel.
96 For the last several months the policy has been no new binary
97 sysctls and no one has put forward an argument to use them.
99 Binary sysctls issues seem to keep happening appearing so
100 properly deprecating them (with a warning to user space) and a
101 2 year grace warning period will mean eventually we can kill
102 them and end the pain.
104 In the mean time individual binary sysctls can be dealt with
105 in a piecewise fashion.
107 Who: Eric Biederman <ebiederm@xmission.com>
109 ---------------------------
111 What: remove EXPORT_SYMBOL(kernel_thread)
113 Files: arch/*/kernel/*_ksyms.c
115 Why: kernel_thread is a low-level implementation detail. Drivers should
116 use the <linux/kthread.h> API instead which shields them from
117 implementation details and provides a higherlevel interface that
118 prevents bugs and code duplication
119 Who: Christoph Hellwig <hch@lst.de>
121 ---------------------------
123 What: Unused EXPORT_SYMBOL/EXPORT_SYMBOL_GPL exports
124 (temporary transition config option provided until then)
125 The transition config option will also be removed at the same time.
127 Why: Unused symbols are both increasing the size of the kernel binary
128 and are often a sign of "wrong API"
129 Who: Arjan van de Ven <arjan@linux.intel.com>
131 ---------------------------
133 What: PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment
135 Why: The stacking of class devices makes these values misleading and
137 Class devices should not carry any of these properties, and bus
138 devices have SUBSYTEM and DRIVER as a replacement.
139 Who: Kay Sievers <kay.sievers@suse.de>
141 ---------------------------
143 What: ACPI procfs interface
145 Why: ACPI sysfs conversion should be finished by January 2008.
146 ACPI procfs interface will be removed in July 2008 so that
147 there is enough time for the user space to catch up.
148 Who: Zhang Rui <rui.zhang@intel.com>
150 ---------------------------
152 What: /proc/acpi/button
154 Why: /proc/acpi/button has been replaced by events to the input layer
156 Who: Len Brown <len.brown@intel.com>
158 ---------------------------
160 What: /proc/acpi/event
162 Why: /proc/acpi/event has been replaced by events via the input layer
163 and netlink since 2.6.23.
164 Who: Len Brown <len.brown@intel.com>
166 ---------------------------
168 What: libata spindown skipping and warning
170 Why: Some halt(8) implementations synchronize caches for and spin
171 down libata disks because libata didn't use to spin down disk on
172 system halt (only synchronized caches).
173 Spin down on system halt is now implemented. sysfs node
174 /sys/class/scsi_disk/h:c:i:l/manage_start_stop is present if
175 spin down support is available.
176 Because issuing spin down command to an already spun down disk
177 makes some disks spin up just to spin down again, libata tracks
178 device spindown status to skip the extra spindown command and
180 This is to give userspace tools the time to get updated and will
181 be removed after userspace is reasonably updated.
182 Who: Tejun Heo <htejun@gmail.com>
184 ---------------------------
186 What: i386/x86_64 bzImage symlinks
189 Why: The i386/x86_64 merge provides a symlink to the old bzImage
190 location so not yet updated user space tools, e.g. package
191 scripts, do not break.
192 Who: Thomas Gleixner <tglx@linutronix.de>
194 ---------------------------
197 - include/linux/netfilter_ipv4/ipt_TOS.h ipt_tos.h header files
198 (superseded by xt_TOS/xt_tos target & match)
200 - "forwarding" header files like ipt_mac.h in
201 include/linux/netfilter_ipv4/ and include/linux/netfilter_ipv6/
203 - xt_CONNMARK match revision 0
204 (superseded by xt_CONNMARK match revision 1)
206 - xt_MARK target revisions 0 and 1
207 (superseded by xt_MARK match revision 2)
209 - xt_connmark match revision 0
210 (superseded by xt_connmark match revision 1)
212 - xt_conntrack match revision 0
213 (superseded by xt_conntrack match revision 1)
215 - xt_iprange match revision 0,
216 include/linux/netfilter_ipv4/ipt_iprange.h
217 (superseded by xt_iprange match revision 1)
219 - xt_mark match revision 0
220 (superseded by xt_mark match revision 1)
222 - xt_recent: the old ipt_recent proc dir
223 (superseded by /proc/net/xt_recent)
225 When: January 2009 or Linux 2.7.0, whichever comes first
226 Why: Superseded by newer revisions or modules
227 Who: Jan Engelhardt <jengelh@computergmbh.de>
229 ---------------------------
231 What: b43 support for firmware revision < 410
232 When: The schedule was July 2008, but it was decided that we are going to keep the
233 code as long as there are no major maintanance headaches.
234 So it _could_ be removed _any_ time now, if it conflicts with something new.
235 Why: The support code for the old firmware hurts code readability/maintainability
236 and slightly hurts runtime performance. Bugfixes for the old firmware
237 are not provided by Broadcom anymore.
238 Who: Michael Buesch <mb@bu3sch.de>
240 ---------------------------
242 What: usedac i386 kernel parameter
244 Why: replaced by allowdac and no dac combination
245 Who: Glauber Costa <gcosta@redhat.com>
247 ---------------------------
249 What: remove HID compat support
251 Why: needed only as a temporary solution until distros fix themselves up
252 Who: Jiri Slaby <jirislaby@gmail.com>
254 ---------------------------
256 What: print_fn_descriptor_symbol()
258 Why: The %pF vsprintf format provides the same functionality in a
259 simpler way. print_fn_descriptor_symbol() is deprecated but
260 still present to give out-of-tree modules time to change.
261 Who: Bjorn Helgaas <bjorn.helgaas@hp.com>
263 ---------------------------
265 What: /sys/o2cb symlink
267 Why: /sys/fs/o2cb is the proper location for this information - /sys/o2cb
268 exists as a symlink for backwards compatibility for old versions of
269 ocfs2-tools. 2 years should be sufficient time to phase in new versions
270 which know to look in /sys/fs/o2cb.
271 Who: ocfs2-devel@oss.oracle.com
273 ---------------------------
275 What: SCTP_GET_PEER_ADDRS_NUM_OLD, SCTP_GET_PEER_ADDRS_OLD,
276 SCTP_GET_LOCAL_ADDRS_NUM_OLD, SCTP_GET_LOCAL_ADDRS_OLD
278 Why: A newer version of the options have been introduced in 2005 that
279 removes the limitions of the old API. The sctp library has been
280 converted to use these new options at the same time. Any user
281 space app that directly uses the old options should convert to using
283 Who: Vlad Yasevich <vladislav.yasevich@hp.com>
285 ---------------------------
287 What: CONFIG_THERMAL_HWMON
289 Why: This option was introduced just to allow older lm-sensors userspace
290 to keep working over the upgrade to 2.6.26. At the scheduled time of
291 removal fixed lm-sensors (2.x or 3.x) should be readily available.
292 Who: Rene Herman <rene.herman@gmail.com>
294 ---------------------------
296 What: Code that is now under CONFIG_WIRELESS_EXT_SYSFS
297 (in net/core/net-sysfs.c)
298 When: After the only user (hal) has seen a release with the patches
299 for enough time, probably some time in 2010.
300 Why: Over 1K .text/.data size reduction, data is available in other
302 Who: Johannes Berg <johannes@sipsolutions.net>
304 ---------------------------
306 What: CONFIG_NF_CT_ACCT
308 Why: Accounting can now be enabled/disabled without kernel recompilation.
309 Currently used only to set a default value for a feature that is also
310 controlled by a kernel/module/sysfs/sysctl parameter.
311 Who: Krzysztof Piotr Oledzki <ole@ans.pl>
313 ---------------------------
315 What: i2c_attach_client(), i2c_detach_client(), i2c_driver->detach_client()
316 When: 2.6.29 (ideally) or 2.6.30 (more likely)
317 Why: Deprecated by the new (standard) device driver binding model. Use
318 i2c_driver->probe() and ->remove() instead.
319 Who: Jean Delvare <khali@linux-fr.org>
321 ---------------------------
323 What: fscher and fscpos drivers
325 Why: Deprecated by the new fschmd driver.
326 Who: Hans de Goede <hdegoede@redhat.com>
327 Jean Delvare <khali@linux-fr.org>
329 ---------------------------
331 What: SELinux "compat_net" functionality
332 When: 2.6.30 at the earliest
333 Why: In 2.6.18 the Secmark concept was introduced to replace the "compat_net"
334 network access control functionality of SELinux. Secmark offers both
335 better performance and greater flexibility than the "compat_net"
336 mechanism. Now that the major Linux distributions have moved to
337 Secmark, it is time to deprecate the older mechanism and start the
338 process of removing the old code.
339 Who: Paul Moore <paul.moore@hp.com>