Documentation/feature-removal-schedule.txt

   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.
   6
   7 ---------------------------
   8
   9 What:   /sys/devices/.../power/state
  10         dev->power.power_state
  11         dpm_runtime_{suspend,resume)()
  12 When:   July 2007
  13 Why:    Broken design for runtime control over driver power states, confusing
  14         driver-internal runtime power management with:  mechanisms to support
  15         system-wide sleep state transitions; event codes that distinguish
  16         different phases of swsusp "sleep" transitions; and userspace policy
  17         inputs.  This framework was never widely used, and most attempts to
  18         use it were broken.  Drivers should instead be exposing domain-specific
  19         interfaces either to kernel or to userspace.
  20 Who:    Pavel Machek <pavel@suse.cz>
  21
  22 ---------------------------
  23
  24 What:   RAW driver (CONFIG_RAW_DRIVER)
  25 When:   December 2005
  26 Why:    declared obsolete since kernel 2.6.3
  27         O_DIRECT can be used instead
  28 Who:    Adrian Bunk <bunk@stusta.de>
  29
  30 ---------------------------
  31
  32 What:   raw1394: requests of type RAW1394_REQ_ISO_SEND, RAW1394_REQ_ISO_LISTEN
  33 When:   June 2007
  34 Why:    Deprecated in favour of the more efficient and robust rawiso interface.
  35         Affected are applications which use the deprecated part of libraw1394
  36         (raw1394_iso_write, raw1394_start_iso_write, raw1394_start_iso_rcv,
  37         raw1394_stop_iso_rcv) or bypass libraw1394.
  38 Who:    Dan Dennedy <dan@dennedy.org>, Stefan Richter <stefanr@s5r6.in-berlin.de>
  39
  40 ---------------------------
  41
  42 What:   Video4Linux API 1 ioctls and video_decoder.h from Video devices.
  43 When:   December 2006
  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 Who:    Mauro Carvalho Chehab <mchehab@brturbo.com.br>
  54
  55 ---------------------------
  56
  57 What:   PCMCIA control ioctl (needed for pcmcia-cs [cardmgr, cardctl])
  58 When:   November 2005
  59 Files:  drivers/pcmcia/: pcmcia_ioctl.c
  60 Why:    With the 16-bit PCMCIA subsystem now behaving (almost) like a
  61         normal hotpluggable bus, and with it using the default kernel
  62         infrastructure (hotplug, driver core, sysfs) keeping the PCMCIA
  63         control ioctl needed by cardmgr and cardctl from pcmcia-cs is
  64         unnecessary, and makes further cleanups and integration of the
  65         PCMCIA subsystem into the Linux kernel device driver model more
  66         difficult. The features provided by cardmgr and cardctl are either
  67         handled by the kernel itself now or are available in the new
  68         pcmciautils package available at
  69         http://kernel.org/pub/linux/utils/kernel/pcmcia/
  70 Who:    Dominik Brodowski <linux@brodo.de>
  71
  72 ---------------------------
  73
  74 What:   remove EXPORT_SYMBOL(kernel_thread)
  75 When:   August 2006
  76 Files:  arch/*/kernel/*_ksyms.c
  77 Why:    kernel_thread is a low-level implementation detail.  Drivers should
  78         use the <linux/kthread.h> API instead which shields them from
  79         implementation details and provides a higherlevel interface that
  80         prevents bugs and code duplication
  81 Who:    Christoph Hellwig <hch@lst.de>
  82
  83 ---------------------------
  84
  85 What:   CONFIG_FORCED_INLINING
  86 When:   June 2006
  87 Why:    Config option is there to see if gcc is good enough. (in january
  88         2006). If it is, the behavior should just be the default. If it's not,
  89         the option should just go away entirely.
  90 Who:    Arjan van de Ven
  91
  92 ---------------------------
  93
  94 What:   eepro100 network driver
  95 When:   January 2007
  96 Why:    replaced by the e100 driver
  97 Who:    Adrian Bunk <bunk@stusta.de>
  98
  99 ---------------------------
 100
 101 What:  drivers depending on OSS_OBSOLETE_DRIVER
 102 When:  options in 2.6.20, code in 2.6.22
 103 Why:   OSS drivers with ALSA replacements
 104 Who:   Adrian Bunk <bunk@stusta.de>
 105
 106 ---------------------------
 107
 108 What:   pci_module_init(driver)
 109 When:   January 2007
 110 Why:    Is replaced by pci_register_driver(pci_driver).
 111 Who:    Richard Knutsson <ricknu-0@student.ltu.se> and Greg Kroah-Hartman <gregkh@suse.de>
 112
 113 ---------------------------
 114
 115 What:   Usage of invalid timevals in setitimer
 116 When:   March 2007
 117 Why:    POSIX requires to validate timevals in the setitimer call. This
 118         was never done by Linux. The invalid (e.g. negative timevals) were
 119         silently converted to more or less random timeouts and intervals.
 120         Until the removal a per boot limited number of warnings is printed
 121         and the timevals are sanitized.
 122
 123 Who:    Thomas Gleixner <tglx@linutronix.de>
 124
 125 ---------------------------
 126
 127 What:   Unused EXPORT_SYMBOL/EXPORT_SYMBOL_GPL exports
 128         (temporary transition config option provided until then)
 129         The transition config option will also be removed at the same time.
 130 When:   before 2.6.19
 131 Why:    Unused symbols are both increasing the size of the kernel binary
 132         and are often a sign of "wrong API"
 133 Who:    Arjan van de Ven <arjan@linux.intel.com>
 134
 135 ---------------------------
 136
 137 What:   USB driver API moves to EXPORT_SYMBOL_GPL
 138 When:   February 2008
 139 Files:  include/linux/usb.h, drivers/usb/core/driver.c
 140 Why:    The USB subsystem has changed a lot over time, and it has been
 141         possible to create userspace USB drivers using usbfs/libusb/gadgetfs
 142         that operate as fast as the USB bus allows.  Because of this, the USB
 143         subsystem will not be allowing closed source kernel drivers to
 144         register with it, after this grace period is over.  If anyone needs
 145         any help in converting their closed source drivers over to use the
 146         userspace filesystems, please contact the
 147         linux-usb-devel@lists.sourceforge.net mailing list, and the developers
 148         there will be glad to help you out.
 149 Who:    Greg Kroah-Hartman <gregkh@suse.de>
 150
 151 ---------------------------
 152
 153 What:   Interrupt only SA_* flags
 154 When:   Januar 2007
 155 Why:    The interrupt related SA_* flags are replaced by IRQF_* to move them
 156         out of the signal namespace.
 157
 158 Who:    Thomas Gleixner <tglx@linutronix.de>
 159
 160 ---------------------------
 161
 162 What:   PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment
 163 When:   October 2008
 164 Why:    The stacking of class devices makes these values misleading and
 165         inconsistent.
 166         Class devices should not carry any of these properties, and bus
 167         devices have SUBSYTEM and DRIVER as a replacement.
 168 Who:    Kay Sievers <kay.sievers@suse.de>
 169
 170 ---------------------------
 171
 172 What:   i2c-isa
 173 When:   December 2006
 174 Why:    i2c-isa is a non-sense and doesn't fit in the device driver
 175         model. Drivers relying on it are better implemented as platform
 176         drivers.
 177 Who:    Jean Delvare <khali@linux-fr.org>
 178
 179 ---------------------------
 180
 181 What:   i2c_adapter.dev
 182         i2c_adapter.list
 183 When:   July 2007
 184 Why:    Superfluous, given i2c_adapter.class_dev:
 185           * The "dev" was a stand-in for the physical device node that legacy
 186             drivers would not have; but now it's almost always present.  Any
 187             remaining legacy drivers must upgrade (they now trigger warnings).
 188           * The "list" duplicates class device children.
 189         The delay in removing this is so upgraded lm_sensors and libsensors
 190         can get deployed.  (Removal causes minor changes in the sysfs layout,
 191         notably the location of the adapter type name and parenting the i2c
 192         client hardware directly from their controller.)
 193 Who:    Jean Delvare <khali@linux-fr.org>,
 194         David Brownell <dbrownell@users.sourceforge.net>
 195
 196 ---------------------------
 197
 198 What:  drivers depending on OBSOLETE_OSS
 199 When:  options in 2.6.22, code in 2.6.24
 200 Why:   OSS drivers with ALSA replacements
 201 Who:   Adrian Bunk <bunk@stusta.de>
 202
 203 ---------------------------
 204
 205 What:   ACPI hooks (X86_SPEEDSTEP_CENTRINO_ACPI) in speedstep-centrino driver
 206 When:   December 2006
 207 Why:    Speedstep-centrino driver with ACPI hooks and acpi-cpufreq driver are
 208         functionally very much similar. They talk to ACPI in same way. Only
 209         difference between them is the way they do frequency transitions.
 210         One uses MSRs and the other one uses IO ports. Functionaliy of
 211         speedstep_centrino with ACPI hooks is now merged into acpi-cpufreq.
 212         That means one common driver will support all Intel Enhanced Speedstep
 213         capable CPUs. That means less confusion over name of
 214         speedstep-centrino driver (with that driver supposed to be used on
 215         non-centrino platforms). That means less duplication of code and
 216         less maintenance effort and no possibility of these two drivers
 217         going out of sync.
 218         Current users of speedstep_centrino with ACPI hooks are requested to
 219         switch over to acpi-cpufreq driver. speedstep-centrino will continue
 220         to work using older non-ACPI static table based scheme even after this
 221         date.
 222
 223 Who:    Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
 224
 225 ---------------------------
 226
 227 What:   /sys/firmware/acpi/namespace
 228 When:   2.6.21
 229 Why:    The ACPI namespace is effectively the symbol list for
 230         the BIOS.  The device names are completely arbitrary
 231         and have no place being exposed to user-space.
 232
 233         For those interested in the BIOS ACPI namespace,
 234         the BIOS can be extracted and disassembled with acpidump
 235         and iasl as documented in the pmtools package here:
 236         http://ftp.kernel.org/pub/linux/kernel/people/lenb/acpi/utils
 237 Who:    Len Brown <len.brown@intel.com>
 238
 239 ---------------------------
 240
 241 What:   ACPI procfs interface
 242 When:   July 2007
 243 Why:    After ACPI sysfs conversion, ACPI attributes will be duplicated
 244         in sysfs and the ACPI procfs interface should be removed.
 245 Who:    Zhang Rui <rui.zhang@intel.com>
 246
 247 ---------------------------
 248
 249 What:   /proc/acpi/button
 250 When:   August 2007
 251 Why:    /proc/acpi/button has been replaced by events to the input layer
 252         since 2.6.20.
 253 Who:    Len Brown <len.brown@intel.com>
 254
 255 ---------------------------
 256
 257 What:   sk98lin network driver
 258 When:   July 2007
 259 Why:    In kernel tree version of driver is unmaintained. Sk98lin driver
 260         replaced by the skge driver.
 261 Who:    Stephen Hemminger <shemminger@osdl.org>
 262
 263 ---------------------------
 264
 265 What:   Compaq touchscreen device emulation
 266 When:   Oct 2007
 267 Files:  drivers/input/tsdev.c
 268 Why:    The code says it was obsolete when it was written in 2001.
 269         tslib is a userspace library which does anything tsdev can do and
 270         much more besides in userspace where this code belongs. There is no
 271         longer any need for tsdev and applications should have converted to
 272         use tslib by now.
 273         The name "tsdev" is also extremely confusing and lots of people have
 274         it loaded when they don't need/use it.
 275 Who:    Richard Purdie <rpurdie@rpsys.net>
 276
 277 ---------------------------
 278
 279 What:   i8xx_tco watchdog driver
 280 When:   in 2.6.22
 281 Why:    the i8xx_tco watchdog driver has been replaced by the iTCO_wdt
 282         watchdog driver.
 283 Who:    Wim Van Sebroeck <wim@iguana.be>
 284
 285 ---------------------------
 286
 287 What:   Multipath cached routing support in ipv4
 288 When:   in 2.6.23
 289 Why:    Code was merged, then submitter immediately disappeared leaving
 290         us with no maintainer and lots of bugs.  The code should not have
 291         been merged in the first place, and many aspects of it's
 292         implementation are blocking more critical core networking
 293         development.  It's marked EXPERIMENTAL and no distribution
 294         enables it because it cause obscure crashes due to unfixable bugs
 295         (interfaces don't return errors so memory allocation can't be
 296         handled, calling contexts of these interfaces make handling
 297         errors impossible too because they get called after we've
 298         totally commited to creating a route object, for example).
 299         This problem has existed for years and no forward progress
 300         has ever been made, and nobody steps up to try and salvage
 301         this code, so we're going to finally just get rid of it.
 302 Who:    David S. Miller <davem@davemloft.net>
 303
 304 ---------------------------