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 ---------------------------
11 Why: Old mxser driver is obsoleted by the mxser_new. Give it some time yet
13 Who: Jiri Slaby <jirislaby@gmail.com>
15 ---------------------------
17 What: dev->power.power_state
19 Why: Broken design for runtime control over driver power states, confusing
20 driver-internal runtime power management with: mechanisms to support
21 system-wide sleep state transitions; event codes that distinguish
22 different phases of swsusp "sleep" transitions; and userspace policy
23 inputs. This framework was never widely used, and most attempts to
24 use it were broken. Drivers should instead be exposing domain-specific
25 interfaces either to kernel or to userspace.
26 Who: Pavel Machek <pavel@suse.cz>
28 ---------------------------
30 What: old NCR53C9x driver
32 Why: Replaced by the much better esp_scsi driver. Actual low-level
33 driver can be ported over almost trivially.
34 Who: David Miller <davem@davemloft.net>
35 Christoph Hellwig <hch@lst.de>
37 ---------------------------
39 What: Video4Linux API 1 ioctls and video_decoder.h from Video devices.
41 Files: include/linux/video_decoder.h include/linux/videodev.h
42 Check: include/linux/video_decoder.h include/linux/videodev.h
43 Why: V4L1 AP1 was replaced by V4L2 API during migration from 2.4 to 2.6
44 series. The old API have lots of drawbacks and don't provide enough
45 means to work with all video and audio standards. The newer API is
46 already available on the main drivers and should be used instead.
47 Newer drivers should use v4l_compat_translate_ioctl function to handle
48 old calls, replacing to newer ones.
49 Decoder iocts are using internally to allow video drivers to
50 communicate with video decoders. This should also be improved to allow
51 V4L2 calls being translated into compatible internal ioctls.
52 Compatibility ioctls will be provided, for a while, via
54 Who: Mauro Carvalho Chehab <mchehab@infradead.org>
56 ---------------------------
58 What: PCMCIA control ioctl (needed for pcmcia-cs [cardmgr, cardctl])
60 Files: drivers/pcmcia/: pcmcia_ioctl.c
61 Why: With the 16-bit PCMCIA subsystem now behaving (almost) like a
62 normal hotpluggable bus, and with it using the default kernel
63 infrastructure (hotplug, driver core, sysfs) keeping the PCMCIA
64 control ioctl needed by cardmgr and cardctl from pcmcia-cs is
65 unnecessary, and makes further cleanups and integration of the
66 PCMCIA subsystem into the Linux kernel device driver model more
67 difficult. The features provided by cardmgr and cardctl are either
68 handled by the kernel itself now or are available in the new
69 pcmciautils package available at
70 http://kernel.org/pub/linux/utils/kernel/pcmcia/
71 Who: Dominik Brodowski <linux@brodo.de>
73 ---------------------------
77 Option: CONFIG_SYSCTL_SYSCALL
78 Why: The same information is available in a more convenient from
79 /proc/sys, and none of the sysctl variables appear to be
80 important performance wise.
82 Binary sysctls are a long standing source of subtle kernel
83 bugs and security issues.
85 When I looked several months ago all I could find after
86 searching several distributions were 5 user space programs and
87 glibc (which falls back to /proc/sys) using this syscall.
89 The man page for sysctl(2) documents it as unusable for user
92 sysctl(2) is not generally ABI compatible to a 32bit user
93 space application on a 64bit and a 32bit kernel.
95 For the last several months the policy has been no new binary
96 sysctls and no one has put forward an argument to use them.
98 Binary sysctls issues seem to keep happening appearing so
99 properly deprecating them (with a warning to user space) and a
100 2 year grace warning period will mean eventually we can kill
101 them and end the pain.
103 In the mean time individual binary sysctls can be dealt with
104 in a piecewise fashion.
106 Who: Eric Biederman <ebiederm@xmission.com>
108 ---------------------------
110 What: a.out interpreter support for ELF executables
112 Files: fs/binfmt_elf.c
113 Why: Using a.out interpreters for ELF executables was a feature for
114 transition from a.out to ELF. But now it is unlikely to be still
115 needed anymore and removing it would simplify the hairy ELF
117 Who: Andi Kleen <ak@suse.de>
119 ---------------------------
121 What: remove EXPORT_SYMBOL(kernel_thread)
123 Files: arch/*/kernel/*_ksyms.c
125 Why: kernel_thread is a low-level implementation detail. Drivers should
126 use the <linux/kthread.h> API instead which shields them from
127 implementation details and provides a higherlevel interface that
128 prevents bugs and code duplication
129 Who: Christoph Hellwig <hch@lst.de>
131 ---------------------------
133 What: CONFIG_FORCED_INLINING
135 Why: Config option is there to see if gcc is good enough. (in january
136 2006). If it is, the behavior should just be the default. If it's not,
137 the option should just go away entirely.
138 Who: Arjan van de Ven
140 ---------------------------
142 What: eepro100 network driver
144 Why: replaced by the e100 driver
145 Who: Adrian Bunk <bunk@stusta.de>
147 ---------------------------
149 What: Unused EXPORT_SYMBOL/EXPORT_SYMBOL_GPL exports
150 (temporary transition config option provided until then)
151 The transition config option will also be removed at the same time.
153 Why: Unused symbols are both increasing the size of the kernel binary
154 and are often a sign of "wrong API"
155 Who: Arjan van de Ven <arjan@linux.intel.com>
157 ---------------------------
160 When: Soon, provided in-kernel callers have been converted
161 Why: This interface is replaced by vm_ops.fault, but it has been around
162 forever, is used by a lot of drivers, and doesn't cost much to
164 Who: Nick Piggin <npiggin@suse.de>
166 ---------------------------
168 What: PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment
170 Why: The stacking of class devices makes these values misleading and
172 Class devices should not carry any of these properties, and bus
173 devices have SUBSYTEM and DRIVER as a replacement.
174 Who: Kay Sievers <kay.sievers@suse.de>
176 ---------------------------
178 What: ACPI procfs interface
180 Why: ACPI sysfs conversion should be finished by January 2008.
181 ACPI procfs interface will be removed in July 2008 so that
182 there is enough time for the user space to catch up.
183 Who: Zhang Rui <rui.zhang@intel.com>
185 ---------------------------
187 What: /proc/acpi/button
189 Why: /proc/acpi/button has been replaced by events to the input layer
191 Who: Len Brown <len.brown@intel.com>
193 ---------------------------
195 What: /proc/acpi/event
197 Why: /proc/acpi/event has been replaced by events via the input layer
198 and netlink since 2.6.23.
199 Who: Len Brown <len.brown@intel.com>
201 ---------------------------
203 What: 'time' kernel boot parameter
205 Why: replaced by 'printk.time=<value>' so that printk timestamps can be
206 enabled or disabled as needed
207 Who: Randy Dunlap <randy.dunlap@oracle.com>
209 ---------------------------
211 What: drivers depending on OSS_OBSOLETE
212 When: options in 2.6.23, code in 2.6.25
213 Why: obsolete OSS drivers
214 Who: Adrian Bunk <bunk@stusta.de>
216 ---------------------------
218 What: libata spindown skipping and warning
220 Why: Some halt(8) implementations synchronize caches for and spin
221 down libata disks because libata didn't use to spin down disk on
222 system halt (only synchronized caches).
223 Spin down on system halt is now implemented. sysfs node
224 /sys/class/scsi_disk/h:c:i:l/manage_start_stop is present if
225 spin down support is available.
226 Because issuing spin down command to an already spun down disk
227 makes some disks spin up just to spin down again, libata tracks
228 device spindown status to skip the extra spindown command and
230 This is to give userspace tools the time to get updated and will
231 be removed after userspace is reasonably updated.
232 Who: Tejun Heo <htejun@gmail.com>
234 ---------------------------
236 What: The arch/ppc and include/asm-ppc directories
238 Why: The arch/powerpc tree is the merged architecture for ppc32 and ppc64
239 platforms. Currently there are efforts underway to port the remaining
240 arch/ppc platforms to the merged tree. New submissions to the arch/ppc
241 tree have been frozen with the 2.6.22 kernel release and that tree will
242 remain in bug-fix only mode until its scheduled removal. Platforms
243 that are not ported by June 2008 will be removed due to the lack of an
244 interested maintainer.
245 Who: linuxppc-dev@ozlabs.org
247 ---------------------------
249 What: sk98lin network driver
251 Why: In kernel tree version of driver is unmaintained. Sk98lin driver
252 replaced by the skge driver.
253 Who: Stephen Hemminger <shemminger@linux-foundation.org>
255 ---------------------------
257 What: i386/x86_64 bzImage symlinks
260 Why: The i386/x86_64 merge provides a symlink to the old bzImage
261 location so not yet updated user space tools, e.g. package
262 scripts, do not break.
263 Who: Thomas Gleixner <tglx@linutronix.de>
265 ---------------------------
267 ---------------------------
269 What: i2c-i810, i2c-prosavage and i2c-savage4
271 Why: These drivers are superseded by i810fb, intelfb and savagefb.
272 Who: Jean Delvare <khali@linux-fr.org>
274 ---------------------------
276 What: bcm43xx wireless network driver
278 Files: drivers/net/wireless/bcm43xx
279 Why: This driver's functionality has been replaced by the
280 mac80211-based b43 and b43legacy drivers.
281 Who: John W. Linville <linville@tuxdriver.com>
283 ---------------------------
285 What: ieee80211 softmac wireless networking component
286 When: 2.6.26 (or after removal of bcm43xx and port of zd1211rw to mac80211)
287 Files: net/ieee80211/softmac
288 Why: No in-kernel drivers will depend on it any longer.
289 Who: John W. Linville <linville@tuxdriver.com>
291 ---------------------------
293 What: rc80211-simple rate control algorithm for mac80211
295 Files: net/mac80211/rc80211-simple.c
296 Why: This algorithm was provided for reference but always exhibited bad
297 responsiveness and performance and has some serious flaws. It has been
298 replaced by rc80211-pid.
299 Who: Stefano Brivio <stefano.brivio@polimi.it>
301 ---------------------------
304 - include/linux/netfilter_ipv4/ipt_TOS.h ipt_tos.h header files
305 (superseded by xt_TOS/xt_tos target & match)
307 - "forwarding" header files like ipt_mac.h in
308 include/linux/netfilter_ipv4/ and include/linux/netfilter_ipv6/
310 - xt_CONNMARK match revision 0
311 (superseded by xt_CONNMARK match revision 1)
313 - xt_MARK target revisions 0 and 1
314 (superseded by xt_MARK match revision 2)
316 - xt_connmark match revision 0
317 (superseded by xt_connmark match revision 1)
319 - xt_conntrack match revision 0
320 (superseded by xt_conntrack match revision 1)
322 - xt_iprange match revision 0,
323 include/linux/netfilter_ipv4/ipt_iprange.h
324 (superseded by xt_iprange match revision 1)
326 - xt_mark match revision 0
327 (superseded by xt_mark match revision 1)
329 When: January 2009 or Linux 2.7.0, whichever comes first
330 Why: Superseded by newer revisions or modules
331 Who: Jan Engelhardt <jengelh@computergmbh.de>
333 ---------------------------
335 What: b43 support for firmware revision < 410
337 Why: The support code for the old firmware hurts code readability/maintainability
338 and slightly hurts runtime performance. Bugfixes for the old firmware
339 are not provided by Broadcom anymore.
340 Who: Michael Buesch <mb@bu3sch.de>