The code for determining a sensort type for temp1 was based on me misreading
the datasheet, the tested registers do not determine what kind of sensor
is actually reported as temp1. Instead the enable / disable certain additional
digital sensor readouts throug PECI / AMDSI, etc. Independent of these settings
temp1 is still reading an external BJT or thermistor as configured by the
Temperature Sensor Type Register (0x6b).
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

Systems with NCT6775F reported temperature sensor labels for systems with
NCT6776F. This patch fixes the problem.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

NCT6776F and NCT6775F support debouncing the fan RPM signal. This can
help improve the stability of th RPM signal for some fans (Arctic cooling
fans for example).
This patch adds a module parameter fan_debounce, which when set to 1 enables
debounce for all fans that the chip supports.
Signed-off-by: NIan Dobson <i.dobson@planet-ian.com>
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

Since the fan speed value can be above 0xff, we can no longer use
that value to determine if the fan speed reading is valid. This
makes it difficult to manipulate the stored fan speed register value.

If we store rpm instead of the fan speed register value, we do not
need to correct it if the fan divisor value is changed, and the above
mentioned problem no longer exists.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

Some of the chips supported by this driver have 13 bit or 16 bit fan count
registers. This patch improves support for those registers, specifically for
NCT6775F. With the changes in this patch, fan speed is reported correctly even
if the fan divider is set to a low value, which results in a fan speed reading
above 0xff.

With this patch, the width of fan count registers is no longer used to determine
if the chip has fan divider register(s) or not. A dedicated flag is used instead
to determine if this is the case.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

This patch adds support for NCT6775F and NCT6776F to the w83627ehf driver.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Tested-by: Ian Dobson <i.dobson@planet-ian.com> (NCT6776F)
Tested-by: Zachary Marzec <zmarzec@gmail.com> (ASUS P8P67 PRO/NCT6776F)
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

If SmartFan IV mode was configured at startup, it should be possible
to re-enable it later on.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

For newer chips, several registers are banked and thus need to be 16 bit.
Also, register addresses change.

To prepare for those chips, convert affected register arrays to 16 bit,
and change access to those registers to array pointers in struct w83627ehf_data.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

Add support for 4th temperature sensor on W83677HG-B.
Display temperature labels on W83677HG-B to report temperature sources.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

Assume that each register is banked, and set the bank for each access.
Cache the bank number so it only needs to be set if it changes.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

This cleanup fixes most of the checkpatch warnings and errors in the w83627ehf
driver. Remaining warnings and errors are left untouched on purpose to avoid
making the code less readable.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

This patch unifies temperature register access, and replaces simple_strtoXXX
with strict_strtoXXX throughout the driver.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NIan Dobson <i.dobson@planet-ian.com>

Some fan control chips support a configuration register to set the number of
tachometer pulses per fan revolution. Add an ABI attribute to support this
configuration register.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJean Delvare <khali@linux-fr.org>

LTC4151 is High Voltage I2C Current and Voltage Monitor from Linear
Technology.
Signed-off-by: NPer Dalen <per.dalen@appeartv.com>
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

LINEAR16 data format is unsigned, not signed. Impact is that affected
attributes report negative values in the upper half of the supported
value range.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

PMBus only has one set of limit registers for up to three temperature sensors.
Thus, changing a limit for one of the temperature sensors affects limits for
other temperature sensors in the same page (and potentially multiple pages
depending on the chip implementation). To handle this situation, re-read all
temperature limit registers when updating sensor data. This way, all affected
temperature limits are updated whenever the limit for a single sensor is
changed.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

Assumption so far was that PMBus devices would support TEMP2 and TEMP3 registers
only on page 0, and that only the TEMP1 register would be used/supported
on other pages.

Turns out that is not correct. UCD92xx devices support TEMP1 and TEMP2 on
page 0, and TEMP2 on other pages. So it is necessary to change the core code
such that it does not make a page based assumptions about temperature register
support.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

So far, it seemed like fans would either all be in page 0, or that
there would be one page per fan.

Turns out this was a wrong assumption. There is at least one PMBus
fan controller which supports three pages with four fans each.
Update code to handle this situation.
Reported-by: NGreg Schnorr <gschnorr@cisco.com>
Tested-by: NGreg Schnorr <gschnorr@cisco.com>
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NGreg Schnorr <gschnorr@cisco.com>

Some PMBus chips do not support the VOUT_MODE register. To make matters worse,
such chips may not return an error when reading the register either, but instead
return 0xff.

Check if the register exists using pmbus_check_byte_register() before reading
its value. In addition, validate the returned value and ignore it if it is 0xff
(which is invalid).
Reported-by: NGreg Schnorr <gschnorr@cisco.com>
Tested-by: NGreg Schnorr <gschnorr@cisco.com>
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NGreg Schnorr <gschnorr@cisco.com>

Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJonathan Cameron <jic23@cam.ac.uk>

Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJonathan Cameron <jic23@cam.ac.uk>

Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJonathan Cameron <jic23@cam.ac.uk>

Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJonathan Cameron <jic23@cam.ac.uk>

This driver adds support for hardware monitoring features of various PMBus
devices.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJonathan Cameron <jic23@cam.ac.uk>

EMC6D103S is similar to EMC6D103, only it does not support registers 62[5:7],
6D[0:7], and 6E[0:7]. Register respective sysfs attributes and update affected
registers for all other chips only.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJean Delvare <khali@linux-fr.org>

So far, the driver updates temp_auto_hyst and temp_auto_off if temp_auto_min changes.
This may cause problems if temp_auto_off was not configured before configuring
temp_auto_min, because one of the variables used (off_desired) is not initialized
in this case.

To simplify the code, and for consistency with other drivers, no longer update the delta
between "min" and "off" until the user sets the "off" value again.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJean Delvare <khali@linux-fr.org>

Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJean Delvare <khali@linux-fr.org>

ADT7463 and ADT7468 optionally support VID5 instead of the standard +12V
measurement input. Use a boolean to identify the chip configuration instead of
hardcoding it several times.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Acked-by: NJean Delvare <khali@linux-fr.org>

This patch adds support for hardware monitoring of Lineage Compact Power Line
Power Entry Modules.
Reviewed-by: NTom Grennan <tom.grennan@ericsson.com>
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

There is a move to deprecate bus-specific PM operations and move to
using dev_pm_ops instead in order to reduce the amount of boilerplate
code in buses and facilitiate updates to the PM core. Do this move for
the lis3lv02d SPI driver.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

This patch initializes register CONFIG3 to a reasonable default PWM frequency
of 25kHz, to prevent audible sound in fan.
Signed-off-by: NRoland Stigge <stigge@antcom.de>
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

2-Channel Temperature Monitor with Dual PWM Fan-Speed Controller
Signed-off-by: NRoland Stigge <stigge@antcom.de>
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>

* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-2.6-mn10300:
  MN10300: atomic_read() should ensure it emits a load
  MN10300: The SMP_ICACHE_INV_FLUSH_RANGE IPI command does not exist
  MN10300: Proper use of macros get_user() in the case of incremented pointers

* 'upstream' of git://git.linux-mips.org/pub/scm/upstream-linus: (26 commits)
  MIPS: Alchemy: Fix reset for MTX-1 and XXS1500
  MIPS: MTX-1: Make au1000_eth probe all PHY addresses
  MIPS: Jz4740: Add HAVE_CLK
  MIPS: Move idle task creation to work queue
  MIPS, Perf-events: Use unsigned delta for right shift in event update
  MIPS, Perf-events: Work with the new callchain interface
  MIPS, Perf-events: Fix event check in validate_event()
  MIPS, Perf-events: Work with the new PMU interface
  MIPS, Perf-events: Work with irq_work
  MIPS: Fix always CONFIG_LOONGSON_UART_BASE=y
  MIPS: Loongson: Fix potentially wrong string handling
  MIPS: Fix GCC-4.6 'set but not used' warning in arch/mips/mm/init.c
  MIPS: Fix GCC-4.6 'set but not used' warning in ieee754int.h
  MIPS: Remove unused code from arch/mips/kernel/syscall.c
  MIPS: Fix GCC-4.6 'set but not used' warning in signal*.c
  MIPS: MSP: Fix MSP71xx bpci interrupt handler return value
  MIPS: Select R4K timer lib for all MSP platforms
  MIPS: Loongson: Remove ad-hoc cmdline default
  MIPS: Clear the correct flag in sysmips(MIPS_FIXADE, ...).
  MIPS: Add an unreachable return statement to satisfy buggy GCCs.
  ...

Merge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86: ce4100: Set pci ops via callback instead of module init
  x86/mm: Fix pgd_lock deadlock
  x86/mm: Handle mm_fault_error() in kernel space
  x86: Don't check for BIOS corruption in first 64K when there's no need to

This reverts the parent commit.  I hate doing that, but it's generating
some discussion ("half of it is right"), and since I am planning on
doing the 2.6.38 release later today we can punt it to stable if
required. Let's not rock the boat right now.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

oom_kill_process() starts with victim_points == 0.  This means that
(most likely) any child has more points and can be killed erroneously.

Also, "children has a different mm" doesn't match the reality, we should
check child->mm != t->mm.  This check is not exactly correct if t->mm ==
NULL but this doesn't really matter, oom_kill_task() will kill them
anyway.

Note: "Kill all processes sharing p->mm" in oom_kill_task() is wrong
too.
Signed-off-by: NOleg Nesterov <oleg@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>