kernel_optimize_test/Documentation/ABI
Linus Torvalds 8d2faea672 This is the bulk of GPIO changes for the v4.3 kernel cycle:
Core changes:
 
 - Root out the wrapper devm_gpiod_get() and gpiod_get() etc
   versions of the descriptor calls that did not use the flags
   argument on the end. This was around for too long and eventually
   Uwe Kleine-König took the time to clean it out and the last
   users are removed along with the macros in this tag. In several
   cases the use of flags simplifies the code. For this reason we
   have (ACKed) patches hitting in DRM, IIO, media, NFC, USB+PHY
   up until we hammer in the nail with removing the macros.
 
 - Add a fat document describing how much ready-made GPIO stuff
   we have i the kernel to discourage people from reinventing
   a square wheel in userspace, as so often happens.
 
 - Create a separate lockdep class for each instance of a GPIO
   IRQ chip instead of using one class for all chips, as the current
   code will not work with systems with several GPIO chips doing
   lockdep debugging.
 
 - Protect against driver unloading also when a GPIO line is only
   used as IRQ for the GPIOLIB_IRQCHIP helpers.
 
 - If the GPIO chip has no designated owner, assign the parent
   device driver owner as owner.
 
 - Consolidation of chained IRQ handler install/remove replacing
   all call sites where irq_set_handler_data() and
   irq_set_chained_handler() were done in succession with a
   combined call to irq_set_chained_handler_and_data(). This
   series was created by Thomas Gleixner after the problem was
   observed by Russell King.
 
 - Tglx also made another series of patches switching
   __irq_set_handler_locked() for irq_set_handler_locked() which
   is way cleaner.
 
 - Tglx and Jiang Liu wrote a good bunch of patches to make use of
   irq_desc_get_xxx() accessors and avoid looking up irq_descs
   from IRQ numbers. The goal is to get rid of the irq number
   from the handlers in the IRQ flow which is nice.
 
 - Rob Herring killed off the set_irq_flags() for all GPIO
   drivers. This was an ARM specific function that is replaced
   with the generic irq_modify_status() where special flags
   are actually needed.
 
 - When an OF node has a pin range for its GPIOs, return
   -EPROBE_DEFER if the pin controller isn't available.
   Pretty logical, yet needed to be fixed.
 
 - If a driver using GPIOLIB_IRQCHIP has its own
   irq_*_resources call back, then call these instead of the
   defaults provided by the GPIOLIB.
 
 - Fix an undocumented ABI hole: named GPIOs were not
   properly documented.
 
 Driver improvements:
 
 - Add get_direction() support to the generic GPIO driver, it's
   strange that we didn't have that before.
 
 - Make it possible to have input-only GPIO chips using the
   generic GPIO driver.
 
 - Clean out platform data support from the Emma Mobile (EM)
   driver
 
 - Finegrained runtime PM support for the RCAR driver.
 
 - Support r8a7795 (R-car H3) in the RCAR driver.
 
 - Support interrupts on GPIOs 16 thru 31 in the DaVinci driver.
 
 - Some consolidation and new support in the MPC8xxx driver,
   we now support MPC5125.
 
 - Preempt-RT-friendly patches: the OMAP, MPC8xxx, drivers uses raw
   spinlocks making it work better with the realime patches.
 
 - Interrupt support for the EXTRAXFS GPIO driver.
 
 - Make the ETRAXFS GPIO driver support also ARTPEC-3.
 
 - Interrupt and wakeup support for the BRCMSTB driver, also for
   wakeup from S5 cold boot.
 
 - Mask MXC IRQs during suspend.
 
 - Improve OMAP2 GPIO set_debounce() to work according to spec.
 
 - The VF610 driver handles IRQs properly.
 
 New drivers:
 
 - ZTE ZX GPIO driver.
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1
 
 iQIcBAABAgAGBQJV52lvAAoJEEEQszewGV1zOVcP/2ZgkfRgl119LZnWShfrEJWq
 UXaNzSaPNgvDzvGaqqi62SZQuhrdIWRKfMPtAuMGbEn5aJx0JC5UAOYjjfkKBqpO
 toqc1w2DScc0JTorY8qgczIBDO1A3ZBAcIvXXpOduy/JaKPoQteRN8WYTynPw48/
 0+97ZODwhyOkfeqmvUClkc9gW4XT68dudb0Lv1nQjsZmd1dHF2PZlwH3aL9sV68j
 GJAqf09xNqZaWWQBhs+J3ptsYjaJfYjo9NOOUf0Y/UgqXO3vB+2S4EmRATaRHS2F
 aHdj03sNZCNSDEa35WwetbLRGxPzSWmfxmBzQQ1baGdcJICn7Yv58EklPKRvwtMo
 ZpUsgiOV4OUIEClPJohs4xbl2HRsOYB3VbcihkXjVAxS6i2/jgA3Tn8ATvUSZ8wq
 TX8D6BfciigRCkT2G+B0TQBmcX1IQsMd1DBUNfw7Dk1TK/vxH4UYWbke422RjKGz
 ORJ+0DfShMCdYjrCVlt7UbFcqE3L5CnrztLQ3oFt0om2JsSWztV9V579G+Dqo9CI
 fE4G3xlsF33UCvXcmnOp6PuU+ZYBodLggkmK4REy2D3LCOnkcKq0U8Fj5RssApZ9
 FdqVYck555ZpcBiN8ihB97WsmU+0XhBjblCbgzr6GxUw8EJ4x8H9nlraA6bluFoP
 9c2qgPxjCq/VWA/F0YOU
 =iQ2P
 -----END PGP SIGNATURE-----

Merge tag 'gpio-v4.3-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio

Pull GPIO updates from Linus Walleij:
 "This is the bulk of GPIO changes for the v4.3 kernel cycle.

  There is quite a lot going on in the GPIO subsystem this merge window,
  so the main matter is decribed below.

  The hits in other subsystems when making the GPIO flags optional are
  all ACKed by their respective subsystem maintainers.

  Core changes:

   - Root out the wrapper devm_gpiod_get() and gpiod_get() etc versions
     of the descriptor calls that did not use the flags argument on the
     end.  This was around for too long and eventually Uwe Kleine-König
     took the time to clean it out and the last users are removed along
     with the macros in this tag.  In several cases the use of flags
     simplifies the code.  For this reason we have (ACKed) patches
     hitting in DRM, IIO, media, NFC, USB+PHY up until we hammer in the
     nail with removing the macros.

   - Add a fat document describing how much ready-made GPIO stuff we
     have i the kernel to discourage people from reinventing a square
     wheel in userspace, as so often happens.

   - Create a separate lockdep class for each instance of a GPIO IRQ
     chip instead of using one class for all chips, as the current code
     will not work with systems with several GPIO chips doing lockdep
     debugging.

   - Protect against driver unloading also when a GPIO line is only used
     as IRQ for the GPIOLIB_IRQCHIP helpers.

   - If the GPIO chip has no designated owner, assign the parent device
     driver owner as owner.

   - Consolidation of chained IRQ handler install/remove replacing all
     call sites where irq_set_handler_data() and
     irq_set_chained_handler() were done in succession with a combined
     call to irq_set_chained_handler_and_data().

     This series was created by Thomas Gleixner after the problem was
     observed by Russell King.

   - Tglx also made another series of patches switching
     __irq_set_handler_locked() for irq_set_handler_locked() which is
     way cleaner.

   - Tglx and Jiang Liu wrote a good bunch of patches to make use of
     irq_desc_get_xxx() accessors and avoid looking up irq_descs from
     IRQ numbers.  The goal is to get rid of the irq number from the
     handlers in the IRQ flow which is nice.

   - Rob Herring killed off the set_irq_flags() for all GPIO drivers.
     This was an ARM specific function that is replaced with the generic
     irq_modify_status() where special flags are actually needed.

   - When an OF node has a pin range for its GPIOs, return -EPROBE_DEFER
     if the pin controller isn't available.  Pretty logical, yet needed
     to be fixed.

   - If a driver using GPIOLIB_IRQCHIP has its own irq_*_resources call
     back, then call these instead of the defaults provided by the
     GPIOLIB.

   - Fix an undocumented ABI hole: named GPIOs were not properly
     documented.

  Driver improvements:

   - Add get_direction() support to the generic GPIO driver, it's
     strange that we didn't have that before.

   - Make it possible to have input-only GPIO chips using the generic
     GPIO driver.

   - Clean out platform data support from the Emma Mobile (EM) driver

   - Finegrained runtime PM support for the RCAR driver.

   - Support r8a7795 (R-car H3) in the RCAR driver.

   - Support interrupts on GPIOs 16 thru 31 in the DaVinci driver.

   - Some consolidation and new support in the MPC8xxx driver, we now
     support MPC5125.

   - Preempt-RT-friendly patches: the OMAP, MPC8xxx, drivers uses raw
     spinlocks making it work better with the realime patches.

   - Interrupt support for the EXTRAXFS GPIO driver.

   - Make the ETRAXFS GPIO driver support also ARTPEC-3.

   - Interrupt and wakeup support for the BRCMSTB driver, also for
     wakeup from S5 cold boot.

   - Mask MXC IRQs during suspend.

   - Improve OMAP2 GPIO set_debounce() to work according to spec.

   - The VF610 driver handles IRQs properly.

  New drivers:

   - ZTE ZX GPIO driver"

* tag 'gpio-v4.3-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio: (87 commits)
  Revert "gpio: extraxfs: fix returnvar.cocci warnings"
  gpio: tc3589x: use static container helper
  gpio: xlp: fix error return code
  gpio: vf610: handle level IRQ's properly
  gpio: max732x: Fix error handling in probe()
  gpio: omap: fix clk_prepare/unprepare usage
  gpio: omap: protect regs access in omap_gpio_irq_handler
  gpio: omap: fix omap2_set_gpio_debounce
  gpio: omap: switch to use platform_get_irq
  gpio: omap: remove wrong irq_domain_remove usage in probe
  gpiolib: add description for gpio irqchip fields in struct gpio_chip
  gpio: extraxfs: fix returnvar.cocci warnings
  gpiolib: irqchip: use different lockdep class for each gpio irqchip
  gpio/grgpio: fix deadlock in grgpio_irq_unmap()
  Documentation: gpio: consumer: describe active low property
  gpio: mxc: fix section mismatch warning
  gpio/mxc: mask gpio interrupts in suspend
  gpio: omap: Fix missing raw locks conversion
  gpio: brcmstb: support wakeup from S5 cold boot
  gpio: brcmstb: Add interrupt and wakeup source support
  ...
2015-09-04 10:07:45 -07:00
..
obsolete
removed
stable Drivers: hv: vmbus: document the VMBus sysfs files 2015-08-05 11:44:29 -07:00
testing This is the bulk of GPIO changes for the v4.3 kernel cycle: 2015-09-04 10:07:45 -07:00
README

This directory attempts to document the ABI between the Linux kernel and
userspace, and the relative stability of these interfaces.  Due to the
everchanging nature of Linux, and the differing maturity levels, these
interfaces should be used by userspace programs in different ways.

We have four different levels of ABI stability, as shown by the four
different subdirectories in this location.  Interfaces may change levels
of stability according to the rules described below.

The different levels of stability are:

  stable/
	This directory documents the interfaces that the developer has
	defined to be stable.  Userspace programs are free to use these
	interfaces with no restrictions, and backward compatibility for
	them will be guaranteed for at least 2 years.  Most interfaces
	(like syscalls) are expected to never change and always be
	available.

  testing/
	This directory documents interfaces that are felt to be stable,
	as the main development of this interface has been completed.
	The interface can be changed to add new features, but the
	current interface will not break by doing this, unless grave
	errors or security problems are found in them.  Userspace
	programs can start to rely on these interfaces, but they must be
	aware of changes that can occur before these interfaces move to
	be marked stable.  Programs that use these interfaces are
	strongly encouraged to add their name to the description of
	these interfaces, so that the kernel developers can easily
	notify them if any changes occur (see the description of the
	layout of the files below for details on how to do this.)

  obsolete/
  	This directory documents interfaces that are still remaining in
	the kernel, but are marked to be removed at some later point in
	time.  The description of the interface will document the reason
	why it is obsolete and when it can be expected to be removed.

  removed/
	This directory contains a list of the old interfaces that have
	been removed from the kernel.

Every file in these directories will contain the following information:

What:		Short description of the interface
Date:		Date created
KernelVersion:	Kernel version this feature first showed up in.
Contact:	Primary contact for this interface (may be a mailing list)
Description:	Long description of the interface and how to use it.
Users:		All users of this interface who wish to be notified when
		it changes.  This is very important for interfaces in
		the "testing" stage, so that kernel developers can work
		with userspace developers to ensure that things do not
		break in ways that are unacceptable.  It is also
		important to get feedback for these interfaces to make
		sure they are working in a proper way and do not need to
		be changed further.


How things move between levels:

Interfaces in stable may move to obsolete, as long as the proper
notification is given.

Interfaces may be removed from obsolete and the kernel as long as the
documented amount of time has gone by.

Interfaces in the testing state can move to the stable state when the
developers feel they are finished.  They cannot be removed from the
kernel tree without going through the obsolete state first.

It's up to the developer to place their interfaces in the category they
wish for it to start out in.


Notable bits of non-ABI, which should not under any circumstances be considered
stable:

- Kconfig.  Userspace should not rely on the presence or absence of any
  particular Kconfig symbol, in /proc/config.gz, in the copy of .config
  commonly installed to /boot, or in any invocation of the kernel build
  process.

- Kernel-internal symbols.  Do not rely on the presence, absence, location, or
  type of any kernel symbol, either in System.map files or the kernel binary
  itself.  See Documentation/stable_api_nonsense.txt.