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