Tags

This is the GPIO bulk changes for the v3.20 series:
- GPIOLIB core changes:
  - Create and use of_mm_gpiochip_remove() for removing
    memory-mapped OF GPIO chips
  - GPIO MMIO library suppports bgpio_set_multiple for
    switching several lines at once, a feature merged in
    the last cycle.
- New drivers:
  - New driver for the APM X-gene standby GPIO controller
  - New driver for the Fujitsu MB86S7x GPIO controller
- Cleanups:
  - Moved rcar driver to use gpiolib irqchip
  - Moxart converted to the GPIO MMIO library
  - GE driver converted to GPIO MMIO library
  - Move sx150x to irqdomain
  - Move max732x to irqdomain
  - Move vx855 to use managed resources
  - Move dwapb to use managed resources
  - Clean tc3589x from platform data
  - Clean stmpe driver to use device tree only probe
- New subtypes:
  - sx1506 support in the sx150x driver
  - Quark 1000 SoC support in the SCH driver
  - Support X86 in the Xilinx driver
  - Support PXA1928 in the PXA driver
- Extended drivers:
  - max732x supports device tree probe
  - sx150x supports device tree probe
- Various minor cleanups and bug fixes

media updates for v3.20-rc1

regulator: Updates for v3.20

This has not been a busy release for the regulator framework, though we
do have the first parts of some ongoing work from Bjorn Andersson to
allow us to support more complex modern systems with dynamic
configuration of regulators in suspend and idle states.

 - Support for device-specific properties on regulator nodes when using
   simplified DT parsing in the core from Krzysztof Kozlowski.
 - Restructuring of the load tracking code, intended to support future
   improvements in this area for more complex system designs.
 - New drivers for Maxim MAX77843 and Mediatek MT6397.
 - Lots of smaller fixes and improvements.

ASoC: Final updates for v3.20

A few more updates for v3.20 that have accumilated over the second half
of last week.  One new (relatively simple) driver for the Maxim
max98357a and some other driver specific fixes and enhancements.  I did
apply a few patches that haven't been in -next just now before sending
this, all fixes except for one simple device ID addition patch.

Linux 3.19

This is the 3.18.6 stable release

This is the 3.14.32 stable release

This is the 3.10.68 stable release

sound fixes for 3.19

Hopefully the final pull request for 3.19: this ended up with a
slightly higher volume than wished, but I put them all as they are
either stable or 3.19 regression fixes.

Most of commits are from ASoC, and have been stewed for a while in
linux-next.  The only change in the common code is the regression
fixes for ASoC AC97 stuff wrt device registrations.  The rest are
device-specific, mostly small fixes in various ASoC drivers and
ak411x on ice1724 boards.

ASoC: Fix for Intel firmware name

Another one liner that arrived after the earlier pull request.  There's
a trivial conflict with my -next branch, I'll send a pull request with
that resolution and some other new stuff before Monday.

ASoC: AC'97 fixes

These are rather too large for this late in the release cycle but
they're clear, well understood and have been tested to fix a regression
which was introduced for v3.19.  The details are all in Lars' changelog
and they've been cooking in -next for a while, to a large extent out
of conservatism about the size.

ASoC: Fixes for v3.19

A few last minute fixes for v3.19, all driver specific.  None of them
stand out particularly - it's all the standard people who are affected
will care stuff.

The Samsung fix is a DT only fix for the audio controller, it's being
merged via the ASoC tree due to process messups (the submitter sent it
at the end of a tangentally related series rather than separately to the
ARM folks) in order to make sure that it gets to people sooner.

regulator: Fix !REGULATOR builds of systems using system suspend calls

The system suspend calls (used to synchronize between system suspend of
the CPU and it's PMIC) are called from board code but not stubbed when
regulator is disabled causing build failures in such cases.  This patch
fixes those build failures.

ASoC: Updates for v3.20

More updates for v3.20:

 - Lots of refactoring from Lars-Peter Clausen, moving drivers to more
   data driven initialization and rationalizing a lot of DAPM usage.
 - Much improved handling of CDCLK clocks on Samsung I2S controllers.
 - Lots of driver specific cleanups and feature improvements.
 - CODEC support for TI PCM514x and TLV320AIC3104 devices.
 - Board support for Tegra systems with Realtek RT5677.

Last round of updates for net-next:
 * revert a patch that caused a regression with mesh userspace (Bob)
 * fix a number of suspend/resume related races
   (from Emmanuel, Luca and myself - we'll look at backporting later)
 * add software implementations for new ciphers (Jouni)
 * add a new ACPI ID for Broadcom's rfkill (Mika)
 * allow using netns FD for wireless (Vadim)
 * some other cleanups (various)

Yet more GPIO fixes for the v3.19 series:
- Two fixes stabilizing that which was never stable before:
  removal of GPIO chips, now let's stop leaking memory.
- Make sure OMAP IRQs are usable when the irqchip API
  is used orthogonally to the gpiochip API.
- Provide a default GPIO base for the mcp23s08 driver.

This is the 3.4.106 stable release

Linux 3.19-rc7

Second round of IIO new drivers, functionality and cleanups for the 3.20 cycle.

New drivers and device support:
* Capella cm3232 ambient light sensor driver.
* Freescale MMA9553 pedometer and activity monitor. Note this involved a
  refactor of the mma9551 driver to pull out shared elements.
* Samsung sensor hub (as used in the galaxy gear 2 watch) Core support
  and initial drivers (gyro and accelerometer) more to follow.
  An additional fix was applied on top of this for a build issue
  thrown up by the autobuilders on some platforms.
* Cosmic Circuits 10001 ADC driver
* Qualcomm SPMI PMIC voltage ADC driver (current adc merged a while back).
* Add binding for AK8963 (in capitals) as unfortunately there are bios'
  out there not using lower case.

New functionality
* Add newe operating mode to the core to allow for non triggered software
  buffers.  This is mostly semantics as previously drivers just claimed they
  had a hardware buffer (when they didn't).
* Add distance channel type.
* Add energy channel type.
* Add velocity channel type and IIO_MOD_ROOT_SUM_SQUARED (i.e. speed when
  our channel type is velocity).
* Add _debounce_count and _debounce_time filter attributes.  Only really
  make sense for counting types of measurements.  First use is for avoiding
  miss detection of steps prior to walking.
* Add change event type.   This replaces the briefly present INSTANCE type
  (which hadn't gained any users).  It is more generic as it allows for
  events say every 10 steps rather than every step.
* Add _calibweight attributes to the ABI (and core support).  Used by
  activity monitors to estimate energy use.  Can imagine there will be
  other uses for this one.

Driver new functionality
* mma9551 gains runtime pm support.
* hid-sensors gain PM support.

Cleanup
* Change calibheight unit to m from cm.  As there are no prior users
  and this was inconsistent with other distance units, it makes sense
  to fix it before hte mma9553 driver which uses it.
* mpu6050 cleanups and devm_ use.
* as3935 switch over to PM ops.
* Fix a few format strings for signed vs unsigned.
* Fix tcs3414 missindentation
* Typo in industrialio-event
* Stop requiring IIO_TRIGGER for IIO_KFIFO_BUF as we have drivers that don't
  need it. No one is quite sure why that dependency was there and it
  seems to not matter.