tjh.dev
Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
1The following is a list of files and features that are going to be
2removed in the kernel source tree. Every entry should contain what
3exactly is going away, why it is happening, and who is going to be doing
4the work. When the feature is removed from the kernel, it should also
5be removed from this file.
6
7---------------------------
8
9What: devfs
10When: July 2005
11Files: fs/devfs/*, include/linux/devfs_fs*.h and assorted devfs
12 function calls throughout the kernel tree
13Why: It has been unmaintained for a number of years, has unfixable
14 races, contains a naming policy within the kernel that is
15 against the LSB, and can be replaced by using udev.
16Who: Greg Kroah-Hartman <greg@kroah.com>
17
18---------------------------
19
20What: RAW driver (CONFIG_RAW_DRIVER)
21When: December 2005
22Why: declared obsolete since kernel 2.6.3
23 O_DIRECT can be used instead
24Who: Adrian Bunk <bunk@stusta.de>
25
26---------------------------
27
28What: drivers that were depending on OBSOLETE_OSS_DRIVER
29 (config options already removed)
30When: before 2.6.19
31Why: OSS drivers with ALSA replacements
32Who: Adrian Bunk <bunk@stusta.de>
33
34---------------------------
35
36What: RCU API moves to EXPORT_SYMBOL_GPL
37When: April 2006
38Files: include/linux/rcupdate.h, kernel/rcupdate.c
39Why: Outside of Linux, the only implementations of anything even
40 vaguely resembling RCU that I am aware of are in DYNIX/ptx,
41 VM/XA, Tornado, and K42. I do not expect anyone to port binary
42 drivers or kernel modules from any of these, since the first two
43 are owned by IBM and the last two are open-source research OSes.
44 So these will move to GPL after a grace period to allow
45 people, who might be using implementations that I am not aware
46 of, to adjust to this upcoming change.
47Who: Paul E. McKenney <paulmck@us.ibm.com>
48
49---------------------------
50
51What: raw1394: requests of type RAW1394_REQ_ISO_SEND, RAW1394_REQ_ISO_LISTEN
52When: November 2005
53Why: Deprecated in favour of the new ioctl-based rawiso interface, which is
54 more efficient. You should really be using libraw1394 for raw1394
55 access anyway.
56Who: Jody McIntyre <scjody@steamballoon.com>
57
58---------------------------
59
60What: sbp2: module parameter "force_inquiry_hack"
61When: July 2006
62Why: Superceded by parameter "workarounds". Both parameters are meant to be
63 used ad-hoc and for single devices only, i.e. not in modprobe.conf,
64 therefore the impact of this feature replacement should be low.
65Who: Stefan Richter <stefanr@s5r6.in-berlin.de>
66
67---------------------------
68
69What: Video4Linux API 1 ioctls and video_decoder.h from Video devices.
70When: July 2006
71Why: V4L1 AP1 was replaced by V4L2 API. during migration from 2.4 to 2.6
72 series. The old API have lots of drawbacks and don't provide enough
73 means to work with all video and audio standards. The newer API is
74 already available on the main drivers and should be used instead.
75 Newer drivers should use v4l_compat_translate_ioctl function to handle
76 old calls, replacing to newer ones.
77 Decoder iocts are using internally to allow video drivers to
78 communicate with video decoders. This should also be improved to allow
79 V4L2 calls being translated into compatible internal ioctls.
80Who: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
81
82---------------------------
83
84What: remove EXPORT_SYMBOL(insert_resource)
85When: April 2006
86Files: kernel/resource.c
87Why: No modular usage in the kernel.
88Who: Adrian Bunk <bunk@stusta.de>
89
90---------------------------
91
92What: PCMCIA control ioctl (needed for pcmcia-cs [cardmgr, cardctl])
93When: November 2005
94Files: drivers/pcmcia/: pcmcia_ioctl.c
95Why: With the 16-bit PCMCIA subsystem now behaving (almost) like a
96 normal hotpluggable bus, and with it using the default kernel
97 infrastructure (hotplug, driver core, sysfs) keeping the PCMCIA
98 control ioctl needed by cardmgr and cardctl from pcmcia-cs is
99 unnecessary, and makes further cleanups and integration of the
100 PCMCIA subsystem into the Linux kernel device driver model more
101 difficult. The features provided by cardmgr and cardctl are either
102 handled by the kernel itself now or are available in the new
103 pcmciautils package available at
104 http://kernel.org/pub/linux/utils/kernel/pcmcia/
105Who: Dominik Brodowski <linux@brodo.de>
106
107---------------------------
108
109What: ip_queue and ip6_queue (old ipv4-only and ipv6-only netfilter queue)
110When: December 2005
111Why: This interface has been obsoleted by the new layer3-independent
112 "nfnetlink_queue". The Kernel interface is compatible, so the old
113 ip[6]tables "QUEUE" targets still work and will transparently handle
114 all packets into nfnetlink queue number 0. Userspace users will have
115 to link against API-compatible library on top of libnfnetlink_queue
116 instead of the current 'libipq'.
117Who: Harald Welte <laforge@netfilter.org>
118
119---------------------------
120
121What: remove EXPORT_SYMBOL(kernel_thread)
122When: August 2006
123Files: arch/*/kernel/*_ksyms.c
124Why: kernel_thread is a low-level implementation detail. Drivers should
125 use the <linux/kthread.h> API instead which shields them from
126 implementation details and provides a higherlevel interface that
127 prevents bugs and code duplication
128Who: Christoph Hellwig <hch@lst.de>
129
130---------------------------
131
132What: CONFIG_FORCED_INLINING
133When: June 2006
134Why: Config option is there to see if gcc is good enough. (in january
135 2006). If it is, the behavior should just be the default. If it's not,
136 the option should just go away entirely.
137Who: Arjan van de Ven
138
139---------------------------
140
141What: START_ARRAY ioctl for md
142When: July 2006
143Files: drivers/md/md.c
144Why: Not reliable by design - can fail when most needed.
145 Alternatives exist
146Who: NeilBrown <neilb@suse.de>
147
148---------------------------
149
150What: au1x00_uart driver
151When: January 2006
152Why: The 8250 serial driver now has the ability to deal with the differences
153 between the standard 8250 family of UARTs and their slightly strange
154 brother on Alchemy SOCs. The loss of features is not considered an
155 issue.
156Who: Ralf Baechle <ralf@linux-mips.org>
157
158---------------------------
159
160What: eepro100 network driver
161When: January 2007
162Why: replaced by the e100 driver
163Who: Adrian Bunk <bunk@stusta.de>
164
165---------------------------
166
167What: pci_module_init(driver)
168When: January 2007
169Why: Is replaced by pci_register_driver(pci_driver).
170Who: Richard Knutsson <ricknu-0@student.ltu.se> and Greg Kroah-Hartman <gregkh@suse.de>
171
172---------------------------
173
174What: Usage of invalid timevals in setitimer
175When: March 2007
176Why: POSIX requires to validate timevals in the setitimer call. This
177 was never done by Linux. The invalid (e.g. negative timevals) were
178 silently converted to more or less random timeouts and intervals.
179 Until the removal a per boot limited number of warnings is printed
180 and the timevals are sanitized.
181
182Who: Thomas Gleixner <tglx@linutronix.de>
183
184---------------------------
185
186What: I2C interface of the it87 driver
187When: January 2007
188Why: The ISA interface is faster and should be always available. The I2C
189 probing is also known to cause trouble in at least one case (see
190 bug #5889.)
191Who: Jean Delvare <khali@linux-fr.org>
192
193---------------------------
194
195What: remove EXPORT_SYMBOL(tasklist_lock)
196When: August 2006
197Files: kernel/fork.c
198Why: tasklist_lock protects the kernel internal task list. Modules have
199 no business looking at it, and all instances in drivers have been due
200 to use of too-lowlevel APIs. Having this symbol exported prevents
201 moving to more scalable locking schemes for the task list.
202Who: Christoph Hellwig <hch@lst.de>
203
204---------------------------
205
206What: mount/umount uevents
207When: February 2007
208Why: These events are not correct, and do not properly let userspace know
209 when a file system has been mounted or unmounted. Userspace should
210 poll the /proc/mounts file instead to detect this properly.
211Who: Greg Kroah-Hartman <gregkh@suse.de>
212
213---------------------------
214
215What: Support for NEC DDB5074 and DDB5476 evaluation boards.
216When: June 2006
217Why: Board specific code doesn't build anymore since ~2.6.0 and no
218 users have complained indicating there is no more need for these
219 boards. This should really be considered a last call.
220Who: Ralf Baechle <ralf@linux-mips.org>
221
222---------------------------
223
224What: USB driver API moves to EXPORT_SYMBOL_GPL
225When: Febuary 2008
226Files: include/linux/usb.h, drivers/usb/core/driver.c
227Why: The USB subsystem has changed a lot over time, and it has been
228 possible to create userspace USB drivers using usbfs/libusb/gadgetfs
229 that operate as fast as the USB bus allows. Because of this, the USB
230 subsystem will not be allowing closed source kernel drivers to
231 register with it, after this grace period is over. If anyone needs
232 any help in converting their closed source drivers over to use the
233 userspace filesystems, please contact the
234 linux-usb-devel@lists.sourceforge.net mailing list, and the developers
235 there will be glad to help you out.
236Who: Greg Kroah-Hartman <gregkh@suse.de>
237
238---------------------------
239
240What: find_trylock_page
241When: January 2007
242Why: The interface no longer has any callers left in the kernel. It
243 is an odd interface (compared with other find_*_page functions), in
244 that it does not take a refcount to the page, only the page lock.
245 It should be replaced with find_get_page or find_lock_page if possible.
246 This feature removal can be reevaluated if users of the interface
247 cannot cleanly use something else.
248Who: Nick Piggin <npiggin@suse.de>
249
250---------------------------