| blob | 51b369e7fc702939a8cb882bd7884355bb3518c1 |
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: MXSER
10 When: December 2007
11 Why: Old mxser driver is obsoleted by the mxser_new. Give it some time yet
12 and remove it.
13 Who: Jiri Slaby <jirislaby@gmail.com>
15 ---------------------------
17 What: V4L2 VIDIOC_G_MPEGCOMP and VIDIOC_S_MPEGCOMP
18 When: October 2007
19 Why: Broken attempt to set MPEG compression parameters. These ioctls are
20 not able to implement the wide variety of parameters that can be set
21 by hardware MPEG encoders. A new MPEG control mechanism was created
22 in kernel 2.6.18 that replaces these ioctls. See the V4L2 specification
23 (section 1.9: Extended controls) for more information on this topic.
24 Who: Hans Verkuil <hverkuil@xs4all.nl> and
25 Mauro Carvalho Chehab <mchehab@infradead.org>
27 ---------------------------
29 What: /sys/devices/.../power/state
30 dev->power.power_state
31 dpm_runtime_{suspend,resume)()
32 When: July 2007
33 Why: Broken design for runtime control over driver power states, confusing
34 driver-internal runtime power management with: mechanisms to support
35 system-wide sleep state transitions; event codes that distinguish
36 different phases of swsusp "sleep" transitions; and userspace policy
37 inputs. This framework was never widely used, and most attempts to
38 use it were broken. Drivers should instead be exposing domain-specific
39 interfaces either to kernel or to userspace.
40 Who: Pavel Machek <pavel@suse.cz>
42 ---------------------------
44 What: RAW driver (CONFIG_RAW_DRIVER)
45 When: December 2005
46 Why: declared obsolete since kernel 2.6.3
47 O_DIRECT can be used instead
48 Who: Adrian Bunk <bunk@stusta.de>
50 ---------------------------
52 What: old NCR53C9x driver
53 When: October 2007
54 Why: Replaced by the much better esp_scsi driver. Actual low-level
55 driver can be ported over almost trivially.
56 Who: David Miller <davem@davemloft.net>
57 Christoph Hellwig <hch@lst.de>
59 ---------------------------
61 What: Video4Linux API 1 ioctls and video_decoder.h from Video devices.
62 When: December 2006
63 Files: include/linux/video_decoder.h
64 Why: V4L1 AP1 was replaced by V4L2 API. during migration from 2.4 to 2.6
65 series. The old API have lots of drawbacks and don't provide enough
66 means to work with all video and audio standards. The newer API is
67 already available on the main drivers and should be used instead.
68 Newer drivers should use v4l_compat_translate_ioctl function to handle
69 old calls, replacing to newer ones.
70 Decoder iocts are using internally to allow video drivers to
71 communicate with video decoders. This should also be improved to allow
72 V4L2 calls being translated into compatible internal ioctls.
73 Who: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
75 ---------------------------
77 What: PCMCIA control ioctl (needed for pcmcia-cs [cardmgr, cardctl])
78 When: November 2005
79 Files: drivers/pcmcia/: pcmcia_ioctl.c
80 Why: With the 16-bit PCMCIA subsystem now behaving (almost) like a
81 normal hotpluggable bus, and with it using the default kernel
82 infrastructure (hotplug, driver core, sysfs) keeping the PCMCIA
83 control ioctl needed by cardmgr and cardctl from pcmcia-cs is
84 unnecessary, and makes further cleanups and integration of the
85 PCMCIA subsystem into the Linux kernel device driver model more
86 difficult. The features provided by cardmgr and cardctl are either
87 handled by the kernel itself now or are available in the new
88 pcmciautils package available at
89 http://kernel.org/pub/linux/utils/kernel/pcmcia/
90 Who: Dominik Brodowski <linux@brodo.de>
92 ---------------------------
94 What: remove EXPORT_SYMBOL(kernel_thread)
95 When: August 2006
96 Files: arch/*/kernel/*_ksyms.c
97 Funcs: kernel_thread
98 Why: kernel_thread is a low-level implementation detail. Drivers should
99 use the <linux/kthread.h> API instead which shields them from
100 implementation details and provides a higherlevel interface that
101 prevents bugs and code duplication
102 Who: Christoph Hellwig <hch@lst.de>
104 ---------------------------
106 What: CONFIG_FORCED_INLINING
107 When: June 2006
108 Why: Config option is there to see if gcc is good enough. (in january
109 2006). If it is, the behavior should just be the default. If it's not,
110 the option should just go away entirely.
111 Who: Arjan van de Ven
113 ---------------------------
115 What: eepro100 network driver
116 When: January 2007
117 Why: replaced by the e100 driver
118 Who: Adrian Bunk <bunk@stusta.de>
120 ---------------------------
122 What: drivers depending on OSS_OBSOLETE_DRIVER
123 When: options in 2.6.20, code in 2.6.22
124 Why: OSS drivers with ALSA replacements
125 Who: Adrian Bunk <bunk@stusta.de>
127 ---------------------------
129 What: Unused EXPORT_SYMBOL/EXPORT_SYMBOL_GPL exports
130 (temporary transition config option provided until then)
131 The transition config option will also be removed at the same time.
132 When: before 2.6.19
133 Why: Unused symbols are both increasing the size of the kernel binary
134 and are often a sign of "wrong API"
135 Who: Arjan van de Ven <arjan@linux.intel.com>
137 ---------------------------
139 What: USB driver API moves to EXPORT_SYMBOL_GPL
140 When: February 2008
141 Files: include/linux/usb.h, drivers/usb/core/driver.c
142 Why: The USB subsystem has changed a lot over time, and it has been
143 possible to create userspace USB drivers using usbfs/libusb/gadgetfs
144 that operate as fast as the USB bus allows. Because of this, the USB
145 subsystem will not be allowing closed source kernel drivers to
146 register with it, after this grace period is over. If anyone needs
147 any help in converting their closed source drivers over to use the
148 userspace filesystems, please contact the
149 linux-usb-devel@lists.sourceforge.net mailing list, and the developers
150 there will be glad to help you out.
151 Who: Greg Kroah-Hartman <gregkh@suse.de>
153 ---------------------------
155 What: Interrupt only SA_* flags
156 When: September 2007
157 Why: The interrupt related SA_* flags are replaced by IRQF_* to move them
158 out of the signal namespace.
160 Who: Thomas Gleixner <tglx@linutronix.de>
162 ---------------------------
164 What: PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment
165 When: October 2008
166 Why: The stacking of class devices makes these values misleading and
167 inconsistent.
168 Class devices should not carry any of these properties, and bus
169 devices have SUBSYTEM and DRIVER as a replacement.
170 Who: Kay Sievers <kay.sievers@suse.de>
172 ---------------------------
174 What: i2c-isa
175 When: December 2006
176 Why: i2c-isa is a non-sense and doesn't fit in the device driver
177 model. Drivers relying on it are better implemented as platform
178 drivers.
179 Who: Jean Delvare <khali@linux-fr.org>
181 ---------------------------
183 What: i2c_adapter.list
184 When: July 2007
185 Why: Superfluous, this list duplicates the one maintained by the driver
186 core.
187 Who: Jean Delvare <khali@linux-fr.org>,
188 David Brownell <dbrownell@users.sourceforge.net>
190 ---------------------------
192 What: drivers depending on OBSOLETE_OSS
193 When: options in 2.6.22, code in 2.6.24
194 Why: OSS drivers with ALSA replacements
195 Who: Adrian Bunk <bunk@stusta.de>
197 ---------------------------
199 What: ACPI hooks (X86_SPEEDSTEP_CENTRINO_ACPI) in speedstep-centrino driver
200 When: December 2006
201 Why: Speedstep-centrino driver with ACPI hooks and acpi-cpufreq driver are
202 functionally very much similar. They talk to ACPI in same way. Only
203 difference between them is the way they do frequency transitions.
204 One uses MSRs and the other one uses IO ports. Functionaliy of
205 speedstep_centrino with ACPI hooks is now merged into acpi-cpufreq.
206 That means one common driver will support all Intel Enhanced Speedstep
207 capable CPUs. That means less confusion over name of
208 speedstep-centrino driver (with that driver supposed to be used on
209 non-centrino platforms). That means less duplication of code and
210 less maintenance effort and no possibility of these two drivers
211 going out of sync.
212 Current users of speedstep_centrino with ACPI hooks are requested to
213 switch over to acpi-cpufreq driver. speedstep-centrino will continue
214 to work using older non-ACPI static table based scheme even after this
215 date.
217 Who: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
219 ---------------------------
221 What: /sys/firmware/acpi/namespace
222 When: 2.6.21
223 Why: The ACPI namespace is effectively the symbol list for
224 the BIOS. The device names are completely arbitrary
225 and have no place being exposed to user-space.
227 For those interested in the BIOS ACPI namespace,
228 the BIOS can be extracted and disassembled with acpidump
229 and iasl as documented in the pmtools package here:
230 http://ftp.kernel.org/pub/linux/kernel/people/lenb/acpi/utils
231 Who: Len Brown <len.brown@intel.com>
233 ---------------------------
235 What: ACPI procfs interface
236 When: July 2007
237 Why: After ACPI sysfs conversion, ACPI attributes will be duplicated
238 in sysfs and the ACPI procfs interface should be removed.
239 Who: Zhang Rui <rui.zhang@intel.com>
241 ---------------------------
243 What: /proc/acpi/button
244 When: August 2007
245 Why: /proc/acpi/button has been replaced by events to the input layer
246 since 2.6.20.
247 Who: Len Brown <len.brown@intel.com>
249 ---------------------------
251 What: sk98lin network driver
252 When: July 2007
253 Why: In kernel tree version of driver is unmaintained. Sk98lin driver
254 replaced by the skge driver.
255 Who: Stephen Hemminger <shemminger@osdl.org>
257 ---------------------------
259 What: Compaq touchscreen device emulation
260 When: Oct 2007
261 Files: drivers/input/tsdev.c
262 Why: The code says it was obsolete when it was written in 2001.
263 tslib is a userspace library which does anything tsdev can do and
264 much more besides in userspace where this code belongs. There is no
265 longer any need for tsdev and applications should have converted to
266 use tslib by now.
267 The name "tsdev" is also extremely confusing and lots of people have
268 it loaded when they don't need/use it.
269 Who: Richard Purdie <rpurdie@rpsys.net>
271 ---------------------------
273 What: Multipath cached routing support in ipv4
274 When: in 2.6.23
275 Why: Code was merged, then submitter immediately disappeared leaving
276 us with no maintainer and lots of bugs. The code should not have
277 been merged in the first place, and many aspects of it's
278 implementation are blocking more critical core networking
279 development. It's marked EXPERIMENTAL and no distribution
280 enables it because it cause obscure crashes due to unfixable bugs
281 (interfaces don't return errors so memory allocation can't be
282 handled, calling contexts of these interfaces make handling
283 errors impossible too because they get called after we've
284 totally commited to creating a route object, for example).
285 This problem has existed for years and no forward progress
286 has ever been made, and nobody steps up to try and salvage
287 this code, so we're going to finally just get rid of it.
288 Who: David S. Miller <davem@davemloft.net>
290 ---------------------------
292 What: read_dev_chars(), read_conf_data{,_lpm}() (s390 common I/O layer)
293 When: December 2007
294 Why: These functions are a leftover from 2.4 times. They have several
295 problems:
296 - Duplication of checks that are done in the device driver's
297 interrupt handler
298 - common I/O layer can't do device specific error recovery
299 - device driver can't be notified for conditions happening during
300 execution of the function
301 Device drivers should issue the read device characteristics and read
302 configuration data ccws and do the appropriate error handling
303 themselves.
304 Who: Cornelia Huck <cornelia.huck@de.ibm.com>
306 ---------------------------
308 What: i2c-ixp2000, i2c-ixp4xx and scx200_i2c drivers
309 When: September 2007
310 Why: Obsolete. The new i2c-gpio driver replaces all hardware-specific
311 I2C-over-GPIO drivers.
312 Who: Jean Delvare <khali@linux-fr.org>
314 ---------------------------
316 What: drivers depending on OSS_OBSOLETE
317 When: options in 2.6.23, code in 2.6.25
318 Why: obsolete OSS drivers
319 Who: Adrian Bunk <bunk@stusta.de>
321 ---------------------------
323 What: libata spindown skipping and warning
324 When: Dec 2008
325 Why: Some halt(8) implementations synchronize caches for and spin
326 down libata disks because libata didn't use to spin down disk on
327 system halt (only synchronized caches).
328 Spin down on system halt is now implemented. sysfs node
329 /sys/class/scsi_disk/h:c:i:l/manage_start_stop is present if
330 spin down support is available.
331 Because issuing spin down command to an already spun down disk
332 makes some disks spin up just to spin down again, libata tracks
333 device spindown status to skip the extra spindown command and
334 warn about it.
335 This is to give userspace tools the time to get updated and will
336 be removed after userspace is reasonably updated.
337 Who: Tejun Heo <htejun@gmail.com>
339 ---------------------------