Add support for enumerating i2c devices directly from the platform
device tree. Direct configuration still looks for i2c-child-devices
property on the controller, and will use it if it exists in order to
ease transition.

A driver and userland utility for a couple of families of the
Microchip Technology / SMSC fan controller chips.

The driver and utility supports the:

o EMC2101 and EMC2101-R
o EMC2103-1, EMC2102-2 and EMC2103-4
o EMC2104
o EMC2106
o EMC2301
o EMC2302
o EMC2303
o EMC2305

The EMC210X family supports 1 or 2 fans with tachometer. Depending on
the chip, it may support 2 additional fans without tach and might
support a high side attachment (i.e. a fan, usually 5v, driven
directly from the chip and not PWM or DAC). All versions of EMC210X
support internal temperature measurements, and depending on the chip,
may support up to 5 additional temperature zones. The tachometers and
temperature measurements are provided to the system via the envsys(4)
framework. Some chip types support GPIO pins and support is provided
via the gpio(4) framework.

The EMC230X family supports 1, 2, 3 or 5 fans with the same number of
tachometers. No temperature zone or GPIO support. The tachometers
are provided to the system via the envsys(4) framework. The fan
support can be provided by PWM signaling or DAC.

The two chip families mostly do PWM signaling for the fan speed, but a
number of them support DAC output, a 0 to 3v or so voltage. When the
chip supports external temperature zones, this is done usually by a
bipolar NPN or PNP transister configured as a diode, but some of the
chip varients support thermistors.

The emcfan(4) kernel driver provides a simple read / write / seek
device in /dev/ to the register set in the chip. The heavy lifting is
done in the userland utility emcfanctl(8) which provides the ability
to read and write to any valid register and provides some basic higher
level commands to control fan behavior. The output is simple text
lines, or JSON.

The kernel driver does not reset or other mess with the chip, aside
from reading registers. It is entirely possible that something else
in any particular system is the major manager of the fan controller
and it would not do for the kernel driver to mess too much with the
attached device. All interactions are intentional via the userland
utility.

It is known that a EMC2301 is present on the Raspberry PI 4 Compute IO
module (not to be confused with the Raspberry PI 4 Compute module
itself) and there is a breakout board from Adafruit with a EMC2101 on
it. The chips themselves are pretty inexpensive from Mouser or
Digi-key and can be soldered using the simpler SMD soldering
techniques. A number of the variants are QFN packages, but the pads
are exposed to the side of the chip. No other external components are
required to use these fan controllers.

branches: 1.3.92;
Install dev/i2c/i2c_io.h and implement the API in the iic(4) driver.
Obsolete the I2C_SCAN option as this can now be done from userland.

Only install headers which are actually used by our userland. This
saves about 2.2MB under /usr/include/dev/. Discussed on tech-kern@
recently.

I HOPE to get the list right. The headers I left in are ones
used for MI tools and those whose usage I discovered by grep over tree sources.
Feel free to put needed includes back in if you encounter anything which
should not be removed from lists.

branches: 1.1.14; 1.1.26;
Rework the way kernel include files are installed. In the new method,
as with user-land programs, include files are installed by each directory
in the tree that has includes to install. (This allows more flexibility
as to what gets installed, makes 'partial installs' easier, and gives us
more options as to which machines' includes get installed at any given
time.) The old SYS_INCLUDES={symlinks,copies} behaviours are _both_
still supported, though at least one bug in the 'symlinks' case is
fixed by this change. Include files can't be build before installation,
so directories that have includes as targets (e.g. dev/pci) have to move
those targets into a different Makefile.

These drivers conditionally support FDT (with an #ifdef). Rather than
blindly assume that i2c_attach_args::ia_cookie is an FDT phandle, consult
self's devhandle and, if the type is DEVHANDLE_TYPE_OF, exctract the phandle
from there.

XXX These drivers should probably become FDT-only (like many others already
are) because the functionality is questionable without information from the
device tree.

Garbage-collect fdtbus_todr_attach(); todr_attach() does all the
necessary work now.

kern/59630

Change todr_chip_handle::cookie -> todr_chip_handle::todr_dev, and
make it a device_t. Upcoming functional changes will require the
device_t associated with a TODR device.

Change todr_chip_handle::bus_cookie -> todr_chip_handle::todr_devaux.
Nothing was using the old field, but I decided to keep it around just
in cause something needs it in the future.

And with these largely mechanical yet semantically meaningful changes,
thus spake the Oracle: "Welcome to NetBSD 11.99.2."

Attach the AC100 RTC as a separate child device node.

XXX This driver doesn't really work without FDT.

branches: 1.7.4; 1.7.14;
Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.4.10;
No need to check cold to add I2C_F_POLL; the i2c API does it for us now.

FDT-ize AC100 driver

branches: 1.2.2;
More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.1.2; 1.1.18; 1.1.20;
Add driver for X-Powers AC100 integrated audio codec and RTC subsystem.
Only RTC functionality implemented for now.

No need to check cold to add I2C_F_POLL; the i2c API does it for us now.

branches: 1.5.2;
More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

act8846_match: filter on ia_addr (device is expected at 0x5a)

branches: 1.3.2; 1.3.18; 1.3.20;
disable ACT_DEBUG, but still print current state at attach

write reg value not target freq in act8846_set_voltage

Add driver for Active-Semi ACT8846 Power Management controller.

branches: 1.1.2; 1.1.18;
Add driver for Active-Semi ACT8846 Power Management controller.

These drivers blindly assume that i2c_attach_args::ia_cookie is an FDT / OF
phandle. Instead, extract the phandle from self's devhandle, both eliminating
the reference to ia_cookie and providing an assertion that it is, in fact,
an FDT / OF phandle.

find sensor names etc. on older G5 like PowerMac7,2

branches: 1.10.4; 1.10.14; 1.10.24;
Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.7.10;
explicitly initialize sensor->state

branches: 1.6.2;
In my quest to make device_compatible_entry (and associated goo)
super-general, it turns out I also made it a little to cumbersome
to use (if my tired fingers are any indication). So, this is a
course-correction -- one string per entry (like of_compat_data,
which it will soon replace), and remove the over-verbose macros.

- Rename iic_compat_match() to iic_compatible_match() and change it
to use the new device_compatible_match() routine. A pointer to
the matching device_compatible_entry is returned if a match is
found.
- Adjust iic_use_direct_match() accordingly.
- i2c drivers now provide device_compatible_entry tables when performing
direct-config matching.
- In the dsrtc driver, take advantage of this new capability to greatly
simplify model selection.

(I'm coming for you next, of_compat_data...)

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

get EEPROM data from uni_n instead of poking around in OF

branches: 1.2.2;
remove some unused code, add comments for credits & explanations

yet another temperature sensor / ADC
this one supports Analog Devices AD7417, found in some G5 macs
depends on OpenFirmware for calibration data

Use device_{get,set}prop_string() for sensor names.

Get the sensor descriptive information from our device properties
dictionary, rather than from a secondary dictionary passed in the
i2c_attach_args (this secondary dictionary is deprecated and going
away).

remove double "with" in comments and usage text. Also fix one typo.

branches: 1.29.6;
fix proplib deprecation

iic_acquire_bus may fail.

branches: 1.27.4;
If we're going to keep a reference on the "props" dictionary from
the i2c_attach_args, we should retain it.

Use iic_compatible_lookup() in admtemp_setflags().

Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use temperature sensors as rnd(9) sources.

Use designated initializers and a consistent termination style in
compat_data[].

Handle failing to acquire the iic bus.
Better handle errors when attaching sysmon sensors.

branches: 1.20.2;
get sensore names from properties, remove OF code

In my quest to make device_compatible_entry (and associated goo)
super-general, it turns out I also made it a little to cumbersome
to use (if my tired fingers are any indication). So, this is a
course-correction -- one string per entry (like of_compat_data,
which it will soon replace), and remove the over-verbose macros.

- Rename iic_compat_match() to iic_compatible_match() and change it
to use the new device_compatible_match() routine. A pointer to
the matching device_compatible_entry is returned if a match is
found.
- Adjust iic_use_direct_match() accordingly.
- i2c drivers now provide device_compatible_entry tables when performing
direct-config matching.
- In the dsrtc driver, take advantage of this new capability to greatly
simplify model selection.

(I'm coming for you next, of_compat_data...)

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.16.2;
get sensor descriptions from OpenFirmware if we know how
this needs to be generalized

add MAX6690 which seems to be a straight up ADM1032 clone

MAX6642 does not have lower limit registers

support MAX6642 - yet another adm1021 kinda-sorta clone

PR/50621: David Binderman: Wrap iic_exec() in a simpler utility function
and fix all the pointer/sizeof botches in the proces.

Add display and setting of chip temperature limits for envsys(4).
Different chips are recognised where possible, and the appropriate limits
are displayed/settable.

Use the destination buffer size when copying strings.
(Didn't matter in practice, because the destination was large enough.)

Direct config: Only match via the list of compatible hardware when we have
such a list available (ia_ncompat > 0). Otherwise fall back to matching
the device name with ia_name.

branches: 1.8.14;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.7.2; 1.7.12;
Initialize sensor state before registering.

XXX Still need to fix i2c/ibmhawk.c

Initialize sensor states to SINVALID

branches: 1.5.8;
various aprint_* fixes.

Add direct config support for a few devices

branches: 1.3.2;
From the specs: address 0xff gives the die revision

struct device * -> device_t, no functional changes intended.

branches: 1.1.6; 1.1.8; 1.1.10; 1.1.12; 1.1.16;
Ported alipm(4) and admtemp(4) from OpenBSD.

Call sysmon_envsys_destroy() if we receive any error from sysmon, rather
than just on some errors.

Avoid double free when attach fails.

branches: 1.11.4;
Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.8.2;
Avoid creating a full temporary softc struct on the stack.

No need to do the poll-during-autoconfiguration dance; the i2c code
does the right thing for us.

Chip matching improvements:
only match for currently known addresses
don't generate messages when matching
Tested on Sun Blade 2500 by martin@.

branches: 1.5.2;
In my quest to make device_compatible_entry (and associated goo)
super-general, it turns out I also made it a little to cumbersome
to use (if my tired fingers are any indication). So, this is a
course-correction -- one string per entry (like of_compat_data,
which it will soon replace), and remove the over-verbose macros.

- Rename iic_compat_match() to iic_compatible_match() and change it
to use the new device_compatible_match() routine. A pointer to
the matching device_compatible_entry is returned if a match is
found.
- Adjust iic_use_direct_match() accordingly.
- i2c drivers now provide device_compatible_entry tables when performing
direct-config matching.
- In the dsrtc driver, take advantage of this new capability to greatly
simplify model selection.

(I'm coming for you next, of_compat_data...)

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.2.16; 1.2.18;
Increase the number of read retries (to 5).

branches: 1.1.2;
Add a driver for the ADM1026 Thermal System Management Fan Controller
i2c chip.

Revert previous for now:
http://mail-index.netbsd.org/source-changes/2020/04/16/msg116278.html

The reasoning turned out to be wrong; __KERNEL_RCSID() in header files
does *not* overwrite RCSID in main source files. The real problem is that
it inserts its RCSID into *every* object files. However, it can be still
useful even if heavily duplicated.

Stop using __KERNEL_RCSID() in header files; it confuses ident(1) by
overwriting RCSID in main source files.

XXX
The first argument of __KERNEL_RCSID() is neglected for ELF. If we wish
to have RCSID of header files in kernel binary, we need something like
__FBSDID() macro in FreeBSD.

branches: 1.3.6;
Chip matching improvements:
only match for currently known addresses
don't generate messages when matching
Tested on Sun Blade 2500 by martin@.

set ADM1026_ADDRMASK as 0x2f, not 0x3f8.

GCC-8 complained about always false condition, which indicated
the bug. patch from jdc@.

branches: 1.1.2; 1.1.18; 1.1.22;
Add a driver for the ADM1026 Thermal System Management Fan Controller
i2c chip.

Remove chip-specific drivers for ADT7463, ADT7467, and ADM1030; these
have been superseded by the dbcool(4) driver.

branches: 1.14.4;
nuke clause 3

branches: 1.13.2;
device_t/softc split for adt7467c(4) and adm1030, and other related
cosmetic changes.

branches: 1.12.2; 1.12.4;
use aprint_*_dev and device_xname

branches: 1.11.14;
Initialize sensor's state
on behalf of xtraeme

Extend the envsys2 API (one more time, sorry) as defined in:

http://mail-index.netbsd.org/tech-kern/2007/11/09/0001.html

sysmon_envsys_create() and sysmon_envsys_destroy() were added to
create/destroy sysmon_envsys objects (and its TAILQ/LIST for sensors/events).

sysmon_envsys_sensor_attach() and sysmon_envsys_sensor_detach() were
added to attach/detach sensors to a specified sysmon_envsys device.

The events framework is now per device and configurable via the
ENVSYS_SETDICTIONARY ioctl or /etc/envsys.conf and envstat(8).

Update all users and documentation to reflect these changes.

branches: 1.9.6; 1.9.8; 1.9.12; 1.9.14;
There's no need to use another envsys_data_t pointer that points to
sme->sme_sensor_data again, just use the edata pointer passed in
the gtredata function.

Imported envsys 2, a brief description of the new features:
(Part 2: drivers)

* Support for detachable sensors.
* Cleaned up the API for simplicity and efficiency.
* Ability to send capacity/critical/warning events to powerd(8).
* Adapted all the code to the new locking order.
* Compatibility with the old envsys API: the ENVSYS_GTREINFO
and ENVSYS_GTREDATA ioctl(2)s are supported.
* Added support for a 'dictionary based communication channel' between
sysmon_power(9) and powerd(8), that means there is no 32 bytes event
size restriction anymore.
* Binary compatibility with old envstat(8) and powerd(8) via COMPAT_40.
* All drivers with the n^2 gtredata bug were fixed, PR kern/36226.

Tested by:

blymn: smsc(4).
bouyer: ipmi(4), mfi(4).
kefren: ug(4).
njoly: viaenv(4), adt7463.c.
riz: owtemp(4).
xtraeme: acpiacad(4), acpibat(4), acpitz(4), aiboost(4), it(4), lm(4).

branches: 1.7.6; 1.7.8;
Remove include of machine/autoconf.h to make this compile.

branches: 1.6.6; 1.6.10;
Use device_private().

branches: 1.5.2; 1.5.4; 1.5.6;
back out previous commit

set I2C_OP_WRITE when passing a write buffer to iic_exec()

branches: 1.3.2; 1.3.4; 1.3.6;
merge ktrace-lwp.

branches: 1.2.6;
Shuffle thermal sensor numbering so temp0 is always the on-chip sensor and temp1
is always CPU temperature.

added drivers for Analog Devices ADT7467 and ADM1030 thermal monitor / fan controller chips found in various Apple laptops and probably other machines.

Remove chip-specific drivers for ADT7463, ADT7467, and ADM1030; these
have been superseded by the dbcool(4) driver.

branches: 1.6.4;
nuke clause 3

branches: 1.5.2;
device_t/softc split for adt7467c(4) and adm1030, and other related
cosmetic changes.

branches: 1.4.14; 1.4.16; 1.4.18;
Extend the envsys2 API (one more time, sorry) as defined in:

http://mail-index.netbsd.org/tech-kern/2007/11/09/0001.html

sysmon_envsys_create() and sysmon_envsys_destroy() were added to
create/destroy sysmon_envsys objects (and its TAILQ/LIST for sensors/events).

sysmon_envsys_sensor_attach() and sysmon_envsys_sensor_detach() were
added to attach/detach sensors to a specified sysmon_envsys device.

The events framework is now per device and configurable via the
ENVSYS_SETDICTIONARY ioctl or /etc/envsys.conf and envstat(8).

Update all users and documentation to reflect these changes.

branches: 1.3.18; 1.3.20; 1.3.24; 1.3.26;
Remove include of machine/autoconf.h to make this compile.

branches: 1.2.18; 1.2.22;
merge ktrace-lwp.

branches: 1.1.6;
added drivers for Analog Devices ADT7467 and ADM1030 thermal monitor / fan controller chips found in various Apple laptops and probably other machines.

Remove chip-specific drivers for ADT7463, ADT7467, and ADM1030; these
have been superseded by the dbcool(4) driver.

Provide correct URL for current data sheet. This device is now owned
by ON Semi.

Note that previous URL for ADT7463 datasheet is no longer operative, and
provide link to an earlier, but still active, revision.

branches: 1.15.4; 1.15.6; 1.15.10;
Split device_t/softc and other related cosmetic changes.

branches: 1.14.14;
Set sensor state to ENVSYS_SVALID for correct values. Otherwise,
envstat(8) will report an unknown state and won't print the sensor
value.

ok by xtraeme.

Extend the envsys2 API (one more time, sorry) as defined in:

http://mail-index.netbsd.org/tech-kern/2007/11/09/0001.html

sysmon_envsys_create() and sysmon_envsys_destroy() were added to
create/destroy sysmon_envsys objects (and its TAILQ/LIST for sensors/events).

sysmon_envsys_sensor_attach() and sysmon_envsys_sensor_detach() were
added to attach/detach sensors to a specified sysmon_envsys device.

The events framework is now per device and configurable via the
ENVSYS_SETDICTIONARY ioctl or /etc/envsys.conf and envstat(8).

Update all users and documentation to reflect these changes.

branches: 1.12.4; 1.12.6; 1.12.10; 1.12.12; 1.12.14;
Add missing STOP conditions at the end of I2C operations.

branches: 1.11.2;
Mark sensor as invalid if no fan is connected.

ok by xtraeme

adt74673_refresh_{fan,temp,volt}: use the edata pointer rather than
using softc's edata array index.

Imported envsys 2, a brief description of the new features:
(Part 2: drivers)

* Support for detachable sensors.
* Cleaned up the API for simplicity and efficiency.
* Ability to send capacity/critical/warning events to powerd(8).
* Adapted all the code to the new locking order.
* Compatibility with the old envsys API: the ENVSYS_GTREINFO
and ENVSYS_GTREDATA ioctl(2)s are supported.
* Added support for a 'dictionary based communication channel' between
sysmon_power(9) and powerd(8), that means there is no 32 bytes event
size restriction anymore.
* Binary compatibility with old envstat(8) and powerd(8) via COMPAT_40.
* All drivers with the n^2 gtredata bug were fixed, PR kern/36226.

Tested by:

blymn: smsc(4).
bouyer: ipmi(4), mfi(4).
kefren: ug(4).
njoly: viaenv(4), adt7463.c.
riz: owtemp(4).
xtraeme: acpiacad(4), acpibat(4), acpitz(4), aiboost(4), it(4), lm(4).

KNF, from Anon Ymous. Also kill extra unneeded initializers and use size_t
for indexes.

Eliminate unnessary floating point.
XXX: someone should KNF this file.

branches: 1.6.2; 1.6.8; 1.6.10; 1.6.16;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.4.6; 1.4.10; 1.4.12; 1.4.14;
PR kern/32885: adt7463 i2c device does not always unlock i2c bus on error.
Apply patch from Nicolas Joly.

branches: 1.3.2;
Use device_private().

branches: 1.2.2; 1.2.4; 1.2.6; 1.2.8; 1.2.10;
* Add missing $NetBSD$ tag and __KERNEL_RCSID.
* Nuke completely __P().

Add support for i2c bus on AMD-8111 used on many Opteron server motherboards,
and ADT 7463c thermal and voltage monitor found on the Tyan S2881 and S2882-D
(and probably other boards as well). We do not currently support any kind
of detection of the i2c address of the thermal monitor; it appears to be
at 0x2D on the S2881 and 0x2E on the S2882-D (kernel config examples
forthcoming).

From PR kern/32463 submitted by Anil Gopinath, anil_public@yahoo.com.

Remove chip-specific drivers for ADT7463, ADT7467, and ADM1030; these
have been superseded by the dbcool(4) driver.

branches: 1.8.4; 1.8.6; 1.8.10;
Split device_t/softc and other related cosmetic changes.

branches: 1.7.14;
Extend the envsys2 API (one more time, sorry) as defined in:

http://mail-index.netbsd.org/tech-kern/2007/11/09/0001.html

sysmon_envsys_create() and sysmon_envsys_destroy() were added to
create/destroy sysmon_envsys objects (and its TAILQ/LIST for sensors/events).

sysmon_envsys_sensor_attach() and sysmon_envsys_sensor_detach() were
added to attach/detach sensors to a specified sysmon_envsys device.

The events framework is now per device and configurable via the
ENVSYS_SETDICTIONARY ioctl or /etc/envsys.conf and envstat(8).

Update all users and documentation to reflect these changes.

branches: 1.6.6; 1.6.8; 1.6.12; 1.6.14;
Imported envsys 2, a brief description of the new features:
(Part 2: drivers)

* Support for detachable sensors.
* Cleaned up the API for simplicity and efficiency.
* Ability to send capacity/critical/warning events to powerd(8).
* Adapted all the code to the new locking order.
* Compatibility with the old envsys API: the ENVSYS_GTREINFO
and ENVSYS_GTREDATA ioctl(2)s are supported.
* Added support for a 'dictionary based communication channel' between
sysmon_power(9) and powerd(8), that means there is no 32 bytes event
size restriction anymore.
* Binary compatibility with old envstat(8) and powerd(8) via COMPAT_40.
* All drivers with the n^2 gtredata bug were fixed, PR kern/36226.

Tested by:

blymn: smsc(4).
bouyer: ipmi(4), mfi(4).
kefren: ug(4).
njoly: viaenv(4), adt7463.c.
riz: owtemp(4).
xtraeme: acpiacad(4), acpibat(4), acpitz(4), aiboost(4), it(4), lm(4).

KNF, from Anon Ymous. Also kill extra unneeded initializers and use size_t
for indexes.

Eliminate unnessary floating point.
XXX: someone should KNF this file.

Fix envsys ranges.

branches: 1.2.2; 1.2.4; 1.2.12; 1.2.18; 1.2.24; 1.2.28; 1.2.30; 1.2.34; 1.2.36; 1.2.42;
* Add missing $NetBSD$ tag and __KERNEL_RCSID.
* Nuke completely __P().

Add support for i2c bus on AMD-8111 used on many Opteron server motherboards,
and ADT 7463c thermal and voltage monitor found on the Tyan S2881 and S2882-D
(and probably other boards as well). We do not currently support any kind
of detection of the i2c address of the thermal monitor; it appears to be
at 0x2D on the S2881 and 0x2E on the S2882-D (kernel config examples
forthcoming).

From PR kern/32463 submitted by Anil Gopinath, anil_public@yahoo.com.

Remove chip-specific drivers for ADT7463, ADT7467, and ADM1030; these
have been superseded by the dbcool(4) driver.

Add verification of proper chip before blindly matching for consistency
with other drivers (notably adt7463).

NOTE: The verify routine in _both_ drivers might better be placed in-line
in the _match() routine. But these drivers will soon be superseded by the
dbcool driver any way.

OK gmcgarry@

Fix typo - no functional changes.

OK macallan@

branches: 1.13.4;
nuke clause 3

branches: 1.12.2;
device_t/softc split for adt7467c(4) and adm1030, and other related
cosmetic changes.

branches: 1.11.2; 1.11.4;
use aprint_*_dev and device_xname

branches: 1.10.14;
Initialize sensor's state
on behalf of xtraeme

Extend the envsys2 API (one more time, sorry) as defined in:

http://mail-index.netbsd.org/tech-kern/2007/11/09/0001.html

sysmon_envsys_create() and sysmon_envsys_destroy() were added to
create/destroy sysmon_envsys objects (and its TAILQ/LIST for sensors/events).

sysmon_envsys_sensor_attach() and sysmon_envsys_sensor_detach() were
added to attach/detach sensors to a specified sysmon_envsys device.

The events framework is now per device and configurable via the
ENVSYS_SETDICTIONARY ioctl or /etc/envsys.conf and envstat(8).

Update all users and documentation to reflect these changes.

branches: 1.8.6; 1.8.8; 1.8.12; 1.8.14;
There's no need to use another envsys_data_t pointer that points to
sme->sme_sensor_data again, just use the edata pointer passed in
the gtredata function.

Imported envsys 2, a brief description of the new features:
(Part 2: drivers)

* Support for detachable sensors.
* Cleaned up the API for simplicity and efficiency.
* Ability to send capacity/critical/warning events to powerd(8).
* Adapted all the code to the new locking order.
* Compatibility with the old envsys API: the ENVSYS_GTREINFO
and ENVSYS_GTREDATA ioctl(2)s are supported.
* Added support for a 'dictionary based communication channel' between
sysmon_power(9) and powerd(8), that means there is no 32 bytes event
size restriction anymore.
* Binary compatibility with old envstat(8) and powerd(8) via COMPAT_40.
* All drivers with the n^2 gtredata bug were fixed, PR kern/36226.

Tested by:

blymn: smsc(4).
bouyer: ipmi(4), mfi(4).
kefren: ug(4).
njoly: viaenv(4), adt7463.c.
riz: owtemp(4).
xtraeme: acpiacad(4), acpibat(4), acpitz(4), aiboost(4), it(4), lm(4).

branches: 1.6.6; 1.6.20; 1.6.22;
Use device_private().

branches: 1.5.2; 1.5.4; 1.5.6;
back out previous commit

set I2C_OP_WRITE when passing a write buffer to iic_exec()

branches: 1.3.2; 1.3.4; 1.3.6;
merge ktrace-lwp.

branches: 1.2.6;
Shuffle thermal sensor numbering so temp0 is always the on-chip sensor and temp1
is always CPU temperature.

added drivers for Analog Devices ADT7467 and ADM1030 thermal monitor / fan controller chips found in various Apple laptops and probably other machines.

Remove chip-specific drivers for ADT7463, ADT7467, and ADM1030; these
have been superseded by the dbcool(4) driver.

Add verification of proper chip before blindly matching for consistency
with other drivers (notably adt7463).

NOTE: The verify routine in _both_ drivers might better be placed in-line
in the _match() routine. But these drivers will soon be superseded by the
dbcool driver any way.

OK gmcgarry@

Fix typo - no functional changes.

OK macallan@

branches: 1.6.4;
nuke clause 3

branches: 1.5.2;
device_t/softc split for adt7467c(4) and adm1030, and other related
cosmetic changes.

branches: 1.4.14; 1.4.16; 1.4.18;
Extend the envsys2 API (one more time, sorry) as defined in:

http://mail-index.netbsd.org/tech-kern/2007/11/09/0001.html

sysmon_envsys_create() and sysmon_envsys_destroy() were added to
create/destroy sysmon_envsys objects (and its TAILQ/LIST for sensors/events).

sysmon_envsys_sensor_attach() and sysmon_envsys_sensor_detach() were
added to attach/detach sensors to a specified sysmon_envsys device.

The events framework is now per device and configurable via the
ENVSYS_SETDICTIONARY ioctl or /etc/envsys.conf and envstat(8).

Update all users and documentation to reflect these changes.

branches: 1.3.4; 1.3.6;
don't include machine/autoconf.h - we don't need it (anymore?)

branches: 1.2.18; 1.2.32; 1.2.42; 1.2.46; 1.2.48;
merge ktrace-lwp.

branches: 1.1.6;
added drivers for Analog Devices ADT7467 and ADM1030 thermal monitor / fan controller chips found in various Apple laptops and probably other machines.

A driver for the Aosong AHT20 temperature and humidity sensor. While
slow for an I2C sensor it is inexpensive and should work well enough
in most indoor conditions. All features of the chip are supported.

A driver for the Aosong AHT20 temperature and humidity sensor. While
slow for an I2C sensor it is inexpensive and should work well enough
in most indoor conditions. All features of the chip are supported.

A driver for the Aosong AHT20 temperature and humidity sensor. While
slow for an I2C sensor it is inexpensive and should work well enough
in most indoor conditions. All features of the chip are supported.

These modules need to depend on iic, not just on i2cexec. (They will
still recursively get i2cexec since the iic depends on it.)

Partial fix for kern/56772

avoid double-free

branches: 1.5.20;
Oops, another one.

Remove now-unused variable.

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.2.2;
fixes from Brad Spenser

PR/52848: Brad Spencer: Two environment sensor drivers: AM2315 and SI70xx
XXX: Please check that my refactoring did not break them!

add RCSID's

PR/52848: Brad Spencer: Two environment sensor drivers: AM2315 and SI70xx
XXX: Please check that my refactoring did not break them!

add RCSID's

PR/52848: Brad Spencer: Two environment sensor drivers: AM2315 and SI70xx
XXX: Please check that my refactoring did not break them!

These drivers blindly assume that i2c_attach_args::ia_cookie is an FDT / OF
phandle. Instead, extract the phandle from self's devhandle, both eliminating
the reference to ia_cookie and providing an assertion that it is, in fact,
an FDT / OF phandle.

Sort includes.

Get drm to build on arm64 again.


Author: Jared McNeill <jmcneill@NetBSD.org>
Committer: Taylor R Campbell <riastradh@NetBSD.org>

branches: 1.6.4; 1.6.14;
Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.3.8;
No need to use I2C_F_POLL here.

branches: 1.2.4;
Broadcast PMF display on/off events when our connector is handling dpms so we can turn off the backlight

Add basic support for Analogix eDP transmitters. This driver doesn't
know how to mode set yet -- it only knows how to read EDID from a panel.

These drivers conditionally support FDT (with an #ifdef). Rather than
blindly assume that i2c_attach_args::ia_cookie is an FDT phandle, consult
self's devhandle and, if the type is DEVHANDLE_TYPE_OF, exctract the phandle
from there.

XXX These drivers should probably become FDT-only (like many others already
are) because the functionality is questionable without information from the
device tree.

Garbage-collect fdtbus_todr_attach(); todr_attach() does all the
necessary work now.

kern/59630

Change todr_chip_handle::cookie -> todr_chip_handle::todr_dev, and
make it a device_t. Upcoming functional changes will require the
device_t associated with a TODR device.

Change todr_chip_handle::bus_cookie -> todr_chip_handle::todr_devaux.
Nothing was using the old field, but I decided to keep it around just
in cause something needs it in the future.

And with these largely mechanical yet semantically meaningful changes,
thus spake the Oracle: "Welcome to NetBSD 11.99.2."

branches: 1.24.2;
Merge thorpej-cfargs2.

branches: 1.23.2; 1.23.8;
Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
actually needed.
- Don't be explicit about what interface attribute is attaching if
the device only has one. (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
situations, making is visibly easier to see when indirect config is
in play, and allowing for future change in semantics. (As of now,
this is just a wrapper around config_match(), but that is an
implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance. This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

branches: 1.22.2;
Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.19.8;
No need to use I2C_F_POLL here.

In as3722_poweroff and as3722_reboot(), check for errors from iic_acquire_bus()
before proceeding with writing to the device.

No need to do the poll-during-autoconfiguration dance; the i2c code
does the right thing for us.

No need to check cold to add I2C_F_POLL; the i2c API does it for us now.

branches: 1.15.2;
In my quest to make device_compatible_entry (and associated goo)
super-general, it turns out I also made it a little to cumbersome
to use (if my tired fingers are any indication). So, this is a
course-correction -- one string per entry (like of_compat_data,
which it will soon replace), and remove the over-verbose macros.

- Rename iic_compat_match() to iic_compatible_match() and change it
to use the new device_compatible_match() routine. A pointer to
the matching device_compatible_entry is returned if a match is
found.
- Adjust iic_use_direct_match() accordingly.
- i2c drivers now provide device_compatible_entry tables when performing
direct-config matching.
- In the dsrtc driver, take advantage of this new capability to greatly
simplify model selection.

(I'm coming for you next, of_compat_data...)

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.12.8; 1.12.10;
Add facility for fdt devices to register reset and poweroff handlers.
Use this to remove as3722-specific code in tegra_machdep.c

- Recognize the AS3722 sd0_v_minus_200mV bit, which is 1 on my Jetson TK1.
- Reduce intended core voltage for Tegra K1 124 by 200mV to 1.2V.

The actual value programmed into the SD0 regulator remains the same (0x50),
a value corresponding to 1.20V.

Add AS3722 SD4 regulator frobbing bits.

Reduces aprint_error counter with TEGRA kernel on Jetson TK1 to 1 from 3.

branches: 1.9.2;
remove unused field in softc

Add SD0 regulator.

Add LDO6 regulator support.

Add support for AS3722 RTC.

Fix typo in string.

This is a driver for the AS3722 PMIC, not the AS3822 LED driver.

branches: 1.4.2; 1.4.4;
Support direct config.

expose pmic reset functionality

add watchdog support

Add basic driver for AMS AS3722 power management IC

branches: 1.2.2; 1.2.18;
expose pmic reset functionality

Add basic driver for AMS AS3722 power management IC

Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.1.2; 1.1.4;
a very preliminary driver for the acclerometer found in later *Books
register definitions from OpenBSD

No need to consult ia->ia_ncompat before calling iic_compatible_lookup().

Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Change the device_compatible_match() function to only perform the match.
Introduce a device_compatible_lookup() function to return an entry based
on the same matching criteria (a'la of_search_compatible()).

Update iic_compatible_match() to reflect the above change, and introduce
iic_compatible_lookup(). This pattern is less awkward to use.

Change the device_compatible_entry struct to have a union of
a scalar value (uintptr_t value) and a pointer value (const void *data),
rather than just "uintptr_t data". This eliminates the need for drivers
to cast this value, and doesn't lose const'ness of pointers assigned
to the field.

Update all of the users of this field to use the correct set of
designated initialisers for each specific case.

branches: 1.36.6;
tsleep -> kpause

branches: 1.35.2;
Oops, missed one more instance of unneeded-I2C_F_POLL.

seeprom_bootstrap_read(): no need to hard-code I2C_F_POLL here because
the i2c code will ensure it's there if the system is cold.

Support more than 128Kbit model.

Add atmel,24c256 compat data

branches: 1.31.4;
provide compatible names for all sizes.

branches: 1.30.2;
Remove the i2c "size" locator and corresponding property / attach arg.
It was a hack meant only for EEPROMs, which have another way to specify
size in the config directive ("flag" paramter), as well as a better way
to detect size based on "compatible" string.

Encode the size information in the compat data.

In my quest to make device_compatible_entry (and associated goo)
super-general, it turns out I also made it a little to cumbersome
to use (if my tired fingers are any indication). So, this is a
course-correction -- one string per entry (like of_compat_data,
which it will soon replace), and remove the over-verbose macros.

- Rename iic_compat_match() to iic_compatible_match() and change it
to use the new device_compatible_match() routine. A pointer to
the matching device_compatible_entry is returned if a match is
found.
- Adjust iic_use_direct_match() accordingly.
- i2c drivers now provide device_compatible_entry tables when performing
direct-config matching.
- In the dsrtc driver, take advantage of this new capability to greatly
simplify model selection.

(I'm coming for you next, of_compat_data...)

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.25.2;
Kill some more extern struct cfdriver declarations.

Down with externs in .c!

match atmel,24c16

Iterate over ia_ncompat array properly.

Also check the compatibles table when the size is below zero.

Fixes an aprint_error() on Jetson TK1.

branches: 1.21.2;
Support direct config.

Direct config: Only match via the list of compatible hardware when we have
such a list available (ia_ncompat > 0). Otherwise fall back to matching
the device name with ia_name.

Many I2C drivers hold spin locks between iic_acquire_bus / iic_release_bus.
Avoid sleeping while owning the I2C bus.

branches: 1.18.4;
Add d_discard to all struct cdevsw instances I could find.

All have been set to "nodiscard"; some should get a real implementation.

branches: 1.17.2;
Change (mostly mechanically) every cdevsw/bdevsw I can find to use
designated initializers.

I have not built every extant kernel so I have probably broken at
least one build; however I've also found and fixed some wrong
cdevsw/bdevsw entries so even if so I think we come out ahead.

Add "i2c-at34c02" AT34C02 EEPROM. This is compatible with the AT24C02
EEPROM, apart from software write protection (not supported in our driver).

Allow the EEPROM size to be specified in the kernel configuration by
using flags.
Add the flags values, and configuration examples to the manual page.

Allow i2c addr wildcard matching. Use with care!

branches: 1.13.2;
Handle direct configuration if ia->ia_name is set, using iic_compat_match().
If ia->ia_name is set, display the name on attach.
Display the size (if known) on attach.
Move the iic_acquire_bus() and iic_release_bus() calls inside the read and
write loops, to avoid holding the bus for the full duration of the read or
write.

branches: 1.12.32; 1.12.42;
Replace device_lookup() with device_lookup_private() to get softc
after device_t/softc split. PR kern/38885 from Jonathan A. Kollasch.

branches: 1.11.2;
device_t/softc split and other related cosmetic changes.

branches: 1.10.2; 1.10.4;
use aprint_*_dev and device_xname

branches: 1.9.8;
use __KERNEL_RCSID()

branches: 1.8.4; 1.8.6; 1.8.8;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.7.6; 1.7.18; 1.7.20; 1.7.24;
Update a few comments based on newer members of the at24xx family.
Add support for the 128kbit, 256kbit and 512kbit parts. Note issues
with the 1024kbit and larger parts, but don't add support for them at
this time. I've not updated with the explosion of part numbers in
this family.

Complete initialiser of the cdevsw structure.

branches: 1.5.8;
Use device_private().

branches: 1.4.4; 1.4.6; 1.4.8; 1.4.10; 1.4.12;
Update some missed _close() and _open() functions from 'struct proc *p'
to 'struct lwp *l'.

merge ktrace-lwp.

branches: 1.2.4; 1.2.18;
PR/24746: Jared Momose: The _write routine in the at24cxx eeprom driver has
a =/== bug.

New generic I2C framework. Supports bit-bang and "intelligent" I2C
interface controllers (of varying intelligence levels).

Contributed by Wasabi Systems, Inc. Primarily written by Steve Woodford,
with some modification by me.

merge ktrace-lwp.

branches: 1.1.4;
New generic I2C framework. Supports bit-bang and "intelligent" I2C
interface controllers (of varying intelligence levels).

Contributed by Wasabi Systems, Inc. Primarily written by Steve Woodford,
with some modification by me.

remove checks for failure after memory allocation calls that cannot fail:

kmem_alloc() with KM_SLEEP
kmem_zalloc() with KM_SLEEP
percpu_alloc()
pserialize_create()
psref_class_create()

all of these paths include an assertion that the allocation has not failed,
so callers should not assert that again.

i2c devices don't actually have a dependency on "iic" (the bus driver)
but on "i2cexec". Create an i2cexec module (i2c_exec.c) to mirror the
dependencies in files.i2c and switch device drivers to depend on "i2cexec"
module instead.

A kernel with a USB stack and no I2C controller / bus attachment would spit
out a bunch of "WARNING: module error: can't find builtin dependency `iic'"
messages at boot. This change solves that problem.

branches: 1.6.12; 1.6.28; 1.6.30;
now that iic is a module, add "iic" dependency to iic drivers

add snr reporting

add digital capture support

add a function to mute/unmute the audio input

branches: 1.2.2; 1.2.6;
don't make module declaration conditional on _MODULE

add driver for the Auvitek AU0828 USB video controllers's analog video
capture functions:

auvitek0 at uhub6 port 2: AU0828
video0 at auvitek0: WinTV HVR-950Q
uaudio0 at auvitek0 port 2 configuration 1 interface 1
uaudio0: vendor 0x2040 product 0x7200, rev 2.00/0.05, addr 2
uaudio0: audio rev 1.00
audio1 at uaudio0: full duplex, playback, capture, independent

add snr reporting

add digital capture support

add snr reporting

add digital capture support

add snr reporting

add digital capture support

add snr reporting

add digital capture support

add snr reporting

branches: 1.1.6;
add driver for the Auvitek AU0828 USB video controllers's analog video
capture functions:

auvitek0 at uhub6 port 2: AU0828
video0 at auvitek0: WinTV HVR-950Q
uaudio0 at auvitek0 port 2 configuration 1 interface 1
uaudio0: vendor 0x2040 product 0x7200, rev 2.00/0.05, addr 2
uaudio0: audio rev 1.00
audio1 at uaudio0: full duplex, playback, capture, independent

add snr reporting

add digital capture support

add a function to mute/unmute the audio input

branches: 1.1.2; 1.1.6;
add driver for the Auvitek AU0828 USB video controllers's analog video
capture functions:

auvitek0 at uhub6 port 2: AU0828
video0 at auvitek0: WinTV HVR-950Q
uaudio0 at auvitek0 port 2 configuration 1 interface 1
uaudio0: vendor 0x2040 product 0x7200, rev 2.00/0.05, addr 2
uaudio0: audio rev 1.00
audio1 at uaudio0: full duplex, playback, capture, independent

These drivers blindly assume that i2c_attach_args::ia_cookie is an FDT / OF
phandle. Instead, extract the phandle from self's devhandle, both eliminating
the reference to ia_cookie and providing an assertion that it is, in fact,
an FDT / OF phandle.

branches: 1.21.2;
Merge thorpej-cfargs2.

branches: 1.20.2; 1.20.8;
Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
actually needed.
- Don't be explicit about what interface attribute is attaching if
the device only has one. (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
situations, making is visibly easier to see when indirect config is
in play, and allowing for future change in semantics. (As of now,
this is just a wrapper around config_match(), but that is an
implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance. This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

branches: 1.19.2;
Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.16.6;
- Don't use I2C_F_POLL.
- G/C the non-FDT entry points; nothing uses them anymore.

branches: 1.15.2;
- Don't read/write the device if iic_acquire_bus() fails.
- axp20x_poweroff(): report the error code if power off fails.
- Don't use polled access in axp20xreg_{get,set}_voltage().

No need to do the poll-during-autoconfiguration dance; the i2c code
does the right thing for us.

branches: 1.13.2;
In my quest to make device_compatible_entry (and associated goo)
super-general, it turns out I also made it a little to cumbersome
to use (if my tired fingers are any indication). So, this is a
course-correction -- one string per entry (like of_compat_data,
which it will soon replace), and remove the over-verbose macros.

- Rename iic_compat_match() to iic_compatible_match() and change it
to use the new device_compatible_match() routine. A pointer to
the matching device_compatible_entry is returned if a match is
found.
- Adjust iic_use_direct_match() accordingly.
- i2c drivers now provide device_compatible_entry tables when performing
direct-config matching.
- In the dsrtc driver, take advantage of this new capability to greatly
simplify model selection.

(I'm coming for you next, of_compat_data...)

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.10.2; 1.10.4;
AXP209 isn't an SMBus device, so don't use iic_smbus_* APIs.

Change aprint_verbose on voltage change to aprint_debug to prevent dmesg
spam when running estd. While here, remove ": " prefix on aprint_*_dev
format strings.

Hook axp20x into fdt regulator api

Add powerdown support and hook it in to FDT

Add direct configuration support.

define control register and associated bits for RTC supercap charger

branches: 1.4.8;
Fix dcdc3 voltage settings (s/dcdc2/dcdc3/g).
While there convert some aprint_normal to aprint_verbose

Implement sensor reports for power supplies (voltage and intensities).
Implement a callback to change DCDC2 and DCDC3 voltages, so that
CPU management can change CPU core voltage when changing frequency.

branches: 1.2.2; 1.2.4;
fix a sizeof typo

AXP20x Power Management Unit driver, pretty basic at the moment -- only acts as a temperature sensor

- Don't use I2C_F_POLL.
- G/C the non-FDT entry points; nothing uses them anymore.

branches: 1.3.2; 1.3.6; 1.3.14;
Hook axp20x into fdt regulator api

Add powerdown support and hook it in to FDT

branches: 1.1.2;
Implement sensor reports for power supplies (voltage and intensities).
Implement a callback to change DCDC2 and DCDC3 voltages, so that
CPU management can change CPU core voltage when changing frequency.

Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.6.16;
In my quest to make device_compatible_entry (and associated goo)
super-general, it turns out I also made it a little to cumbersome
to use (if my tired fingers are any indication). So, this is a
course-correction -- one string per entry (like of_compat_data,
which it will soon replace), and remove the over-verbose macros.

- Rename iic_compat_match() to iic_compatible_match() and change it
to use the new device_compatible_match() routine. A pointer to
the matching device_compatible_entry is returned if a match is
found.
- Adjust iic_use_direct_match() accordingly.
- i2c drivers now provide device_compatible_entry tables when performing
direct-config matching.
- In the dsrtc driver, take advantage of this new capability to greatly
simplify model selection.

(I'm coming for you next, of_compat_data...)

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.3.2; 1.3.4;
add direct config support

Change internal temperature base from -267.7C to -243.7C. The AXP223
datasheet says that the value is between -267.7C and 165.8C in steps of 0.1C,
but this doesn't make sense for a 12-bit field. If we take the maximum value
and subtract 4095 we get -243.7C, effectively adding +24C to the reported
temp. On my A31 board, this changes a chilly 15.3C to a more reasonable 39.3C.

branches: 1.1.2;
add AXP22x Power Management Unit driver

Rename axp806pmic to axppmic, and add support for AXP803.

Fix pasto

Add support for poweroff and the power on-off key (POK).

FDT-ize X-Powers AXP805/806 PMIC driver.

branches: 1.1.2; 1.1.18; 1.1.20;
Add AXP806 and AXP809 power management controller drivers.

FDT-ize X-Powers AXP805/806 PMIC driver.

branches: 1.1.2; 1.1.18; 1.1.20;
Add AXP806 and AXP809 power management controller drivers.

No need to check cold to add I2C_F_POLL; the i2c API does it for us now.

branches: 1.2.2;
More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.1.2; 1.1.18; 1.1.20;
Add AXP806 and AXP809 power management controller drivers.

branches: 1.1.2; 1.1.18;
Add AXP806 and AXP809 power management controller drivers.

These drivers blindly assume that i2c_attach_args::ia_cookie is an FDT / OF
phandle. Instead, extract the phandle from self's devhandle, both eliminating
the reference to ia_cookie and providing an assertion that it is, in fact,
an FDT / OF phandle.

Remove WIP

Add support for AXP15060

Wrap a long line.

Trailing whitespace.

branches: 1.37.8;
Add GPIO support.

PR# kern/57030

branches: 1.36.2;
Merge thorpej-cfargs2.

branches: 1.35.4;
Bail out of axpreg_attach if axpreg_get_voltage returns an error.

uvol isn't updated and shouldn't be used.

branches: 1.34.2; 1.34.4;
Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
actually needed.
- Don't be explicit about what interface attribute is attaching if
the device only has one. (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
situations, making is visibly easier to see when indirect config is
in play, and allowing for future change in semantics. (As of now,
this is just a wrapper around config_match(), but that is an
implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance. This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

branches: 1.33.2;
Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Change the device_compatible_match() function to only perform the match.
Introduce a device_compatible_lookup() function to return an entry based
on the same matching criteria (a'la of_search_compatible()).

Update iic_compatible_match() to reflect the above change, and introduce
iic_compatible_lookup(). This pattern is less awkward to use.

Change the device_compatible_entry struct to have a union of
a scalar value (uintptr_t value) and a pointer value (const void *data),
rather than just "uintptr_t data". This eliminates the need for drivers
to cast this value, and doesn't lose const'ness of pointers assigned
to the field.

Update all of the users of this field to use the correct set of
designated initialisers for each specific case.

branches: 1.29.6;
Don't access the i2c bus in interrupt context. Instead, mask the
interrupt and process it on a work queue.

branches: 1.28.2;
In axppmic_power_poweroff(), check for errors from iic_acquire_bus()
before proceeding with writing to the device.

No need to do the poll-during-autoconfiguration dance; the i2c code
does the right thing for us.

Add support for devices with separate "init" and "default" pinctrl states.

branches: 1.25.2;
No need to check cold to add I2C_F_POLL; the i2c API does it for us now.

Add more axp809 controls

Fix ldo_io[01] enable and disable values for axp809

Enable always-on and boot-on regulators at attach time

Add basic support for AXP809

Adjust regulator voltage at attach time to be between the min and max values specified in the dts

Add support for AXP806

Revert part of previous - do not reserve regulators, it's perfectly normal for devices to share one.

Add support for AXP813 and only allow one consumer per regulator at a time

Add AXP803 current and maximum battery capacity sensors

Use semantically-correct "_lo" in AXP_ADC_RAW macro.

branches: 1.14.2;
In my quest to make device_compatible_entry (and associated goo)
super-general, it turns out I also made it a little to cumbersome
to use (if my tired fingers are any indication). So, this is a
course-correction -- one string per entry (like of_compat_data,
which it will soon replace), and remove the over-verbose macros.

Change device_compatible_match() and iic_compatible_match() to return
the weighted match value and take an optional compatible-entry pointer,
rather than the other way around.

Use the device_compatible_entry mechanism rather than of_compat_data;
all of the OF / FDT data we need is already in the i2c_attach_args.

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

Add battery voltage, charge current, and discharge current sensors for
AXP803.

branches: 1.9.2;
Battery charge state is invalid until we refresh the first time

Use IRQs to update status whenever we can. Reduces the amount of work that
needs to be done whenever sysmon_envsys refreshes sensors.

Trigger shutdown on POKLIRQ instead of POKSIRQ.

POKSIRQ is triggered if POK remains low for less than IRQLEVEL. This makes
it way too easy to accidentally trigger shutdown. POKLIRQ is triggered if
POK hold time is greater than IRQLEVEL, which is much more reasonable for
this use case.

Fix off by one in axp803 dcdc2/3/4 definitions

POKS IRQ register / bit differ between AXP803 and AXP805/806. Handle these
differences.

No need to read battery capacity warning levels each time the sensor is refreshed

Add ACIN and VBUS present sensors

Add battery sensors.

Rename axp806pmic to axppmic, and add support for AXP803.

Split the BMP280 / BME280 driver into common code and create I2C and
SPI attachments.

Use a better choice for the print format string.

Correct the humidity conversion for the BME280

Mention where in the datasheet the compensation algorithms came from.

Use kpause() instead of delay() in the measurement cycle. Try and
derive the proper wait delay for the measurement based upon the over
sampling setting and allow the wait factor multiplier to be adjusted.

Read the datasheet more closely and put in some delays. The chip will
just return junk if the wait is not long enough to allow a measurement
to start.

A driver for the Bosch BMP280 / BME280 temperature, humidity and
atmospheric pressure sensor. This is an inexpensive to moderately
expensive chip available from a large number of places. The driver
supports all aspects of the two chips, except for the repeating read
mode which would allow for sub-second queries, such as fall detection
or perhaps even as an altimeter. This driver also only supports the
I2C interface and not the SPI interface.

The BME280, the one with humidity, is not fully tested at this point,
awaiting upon a breakout board and may not show proper humidity.

Split the BMP280 / BME280 driver into common code and create I2C and
SPI attachments.

A driver for the Bosch BMP280 / BME280 temperature, humidity and
atmospheric pressure sensor. This is an inexpensive to moderately
expensive chip available from a large number of places. The driver
supports all aspects of the two chips, except for the repeating read
mode which would allow for sub-second queries, such as fall detection
or perhaps even as an altimeter. This driver also only supports the
I2C interface and not the SPI interface.

The BME280, the one with humidity, is not fully tested at this point,
awaiting upon a breakout board and may not show proper humidity.

When the I2C attachment code for the BMx280 driver was touched last,
the handles for I2C and SPI were split out into attachment specific
softc structures. This was all good and fine and a nice improvement.
However, the CFATTACH_DECL_ call still included the general softc in
the I2C case and not the I2C attachment specific softc. This would
probably overwrite some of the config information in the kernel, but
might not show itself to be a problem unless you did a detach /
reattach operation. Fix this up.

Re-reactor more common code.

Don't put SPI- or I2C-specific stuff in common code.

Lip-service to direct configuration. More work here is needed to conform
to the device tree bindings.

Split the BMP280 / BME280 driver into common code and create I2C and
SPI attachments.

Split the BMP280 / BME280 driver into common code and create I2C and
SPI attachments.

Use kpause() instead of delay() in the measurement cycle. Try and
derive the proper wait delay for the measurement based upon the over
sampling setting and allow the wait factor multiplier to be adjusted.

A driver for the Bosch BMP280 / BME280 temperature, humidity and
atmospheric pressure sensor. This is an inexpensive to moderately
expensive chip available from a large number of places. The driver
supports all aspects of the two chips, except for the repeating read
mode which would allow for sub-second queries, such as fall detection
or perhaps even as an altimeter. This driver also only supports the
I2C interface and not the SPI interface.

The BME280, the one with humidity, is not fully tested at this point,
awaiting upon a breakout board and may not show proper humidity.

These drivers blindly assume that i2c_attach_args::ia_cookie is an FDT / OF
phandle. Instead, extract the phandle from self's devhandle, both eliminating
the reference to ia_cookie and providing an assertion that it is, in fact,
an FDT / OF phandle.

cwfg: update for dts-5.15 binding changes

branches: 1.4.4; 1.4.14;
Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.1.4; 1.1.8; 1.1.12;
Add driver for CellWise CW2015 Fuel Gauge IC.

Fix mis-placed parentheses. PR kern/54812.

No need to use I2C_F_POLL here.

In cx24227_writereg() / cx24227_readreg(), return the error
from iic_acquire_bus(), not some bogus return value that can
potentially be confused for a real error code.

branches: 1.8.10;
remove checks for failure after memory allocation calls that cannot fail:

kmem_alloc() with KM_SLEEP
kmem_zalloc() with KM_SLEEP
percpu_alloc()
pserialize_create()
psref_class_create()

all of these paths include an assertion that the allocation has not failed,
so callers should not assert that again.

i2c devices don't actually have a dependency on "iic" (the bus driver)
but on "i2cexec". Create an i2cexec module (i2c_exec.c) to mirror the
dependencies in files.i2c and switch device drivers to depend on "i2cexec"
module instead.

A kernel with a USB stack and no I2C controller / bus attachment would spit
out a bunch of "WARNING: module error: can't find builtin dependency `iic'"
messages at boot. This change solves that problem.

branches: 1.6.2; 1.6.14; 1.6.16;
always set *data to appease gcc -03 (some users of cx24227_readreg() don't
check return value for error).

branches: 1.5.2; 1.5.6;
now that iic is a module, add "iic" dependency to iic drivers

hide debug messages under CX24227_DEBUG

add QAM64/256 support

modularize cx24227

Add Conexant/Samsung CX24227/S5H1409 demodulator subroutines.

Add Conexant/Samsung CX24227/S5H1409 demodulator subroutines.

fix polarity reversing tpyo

Use device_{get,set}prop_string() for sensor names.

Get the sensor descriptive information from our device properties
dictionary, rather than from a secondary dictionary passed in the
i2c_attach_args (this secondary dictionary is deprecated and going
away).

Remove unnecessary include of <sys/malloc.h>.

These modules need to depend on iic, not just on i2cexec. (They will
still recursively get i2cexec since the iic depends on it.)

Partial fix for kern/56772

fix various typos in compatibility, mainly in comments.

branches: 1.62.6;
fix proplib deprecation

avoid double-free

branches: 1.60.4;
If we're going to keep a reference on the "props" dictionary from
the i2c_attach_args, we should retain it.

Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Use designated initializers and a consistent termination style in
compat_data[].

branches: 1.56.2;
in sysctl_dbcool_behavior() - actually use the array index when translating
text from sysctl -w *.behavior=<whatever>
now this actually works on my sb2500's two adm1030s

get sensor names from device properties if possible

Change a aprint_verbose into aprint_debug - the message does not add
real value to the dmesg(8) output.

- dbcool_pmf_suspend() set the wrong bit out of the suspended data as
it is stored as a bool and thus when bit 2 was wanted, bit 1 was
restored. found by GCC7's "don't use ~ on bool" checker.
- consolidate the dbcool_pmf_suspend()/dbcool_pmf_resume() guts in to
a shared function dbcool_do_pmf() as they are identical except for
1 vs 2 lines.

branches: 1.52.2;
In my quest to make device_compatible_entry (and associated goo)
super-general, it turns out I also made it a little to cumbersome
to use (if my tired fingers are any indication). So, this is a
course-correction -- one string per entry (like of_compat_data,
which it will soon replace), and remove the over-verbose macros.

Add a kernel panic when we matched but can not handle a string - better
die explicitly instead of crashing a few lines later with strange NULL
derefs.

- Rename iic_compat_match() to iic_compatible_match() and change it
to use the new device_compatible_match() routine. A pointer to
the matching device_compatible_entry is returned if a match is
found.
- Adjust iic_use_direct_match() accordingly.
- i2c drivers now provide device_compatible_entry tables when performing
direct-config matching.
- In the dsrtc driver, take advantage of this new capability to greatly
simplify model selection.

(I'm coming for you next, of_compat_data...)

More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.48.2;
dbcool_read_volt() has a missing {} issue.

add more names for direct config

Fix compile error.

KNF. No functional change.

Add direct configuration support via compatible names.

Update module dependencies for all the existing modules that depend on sysmon components.

i2c devices don't actually have a dependency on "iic" (the bus driver)
but on "i2cexec". Create an i2cexec module (i2c_exec.c) to mirror the
dependencies in files.i2c and switch device drivers to depend on "i2cexec"
module instead.

A kernel with a USB stack and no I2C controller / bus attachment would spit
out a bunch of "WARNING: module error: can't find builtin dependency `iic'"
messages at boot. This change solves that problem.

branches: 1.41.2; 1.41.4;
Merge tls-earlyentropy branch into HEAD.

branches: 1.40.2;
Ensure that the top level sysctl nodes (kern, vfs, net, ...) exist before
the sysctl link sets are processed, and remove redundancy.

Shaves >13kB off of an amd64 GENERIC, not to mention >1k duplicate
lines of code.

Remove unused variables

branches: 1.38.2; 1.38.4;
Add some pre-processor magic to verify that the type of the data item
passed to sysctl_createv() actually matches the declared type for
the item itself.
In the places where the caller specifies a function and a structure
address (typically the 'softc') an explicit (void *) cast is now needed.
Fixes bugs in sys/dev/acpi/asus_acpi.c sys/dev/bluetooth/bcsp.c
sys/kern/vfs_bio.c sys/miscfs/syncfs/sync_subr.c and setting
AcpiGbl_EnableAmlDebugObject.
(mostly passing the address of a uint64_t when typed as CTLTYPE_INT).
I've test built quite a few kernels, but there may be some unfixed MD
fallout. Most likely passing &char[] to char *.
Also add CTLFLAG_UNSIGNED for unsiged decimals - not set yet.

Update handling of sysctl stuff when built as modules

Fix logical errors (&& instead of & in a few places).

branches: 1.35.2; 1.35.6;
now that iic is a module, add "iic" dependency to iic drivers

fix gcc-4.5 warnings

When detaching the device, tear down its sysctl tree.

appease gcc 4.5

modularize

Initialize sensor state before registering.

XXX Still need to fix i2c/ibmhawk.c

branches: 1.29.2;
Fix another instance of the same typo in the same comment! dosn't -->
doesn't

Thanks again, Jukka!

Fix typo in previous - dosn't --> doesn't

Thanks, Jukka!

Add support for SMSC EMC6D103S

From jasper@pointless.net in PR kern/44685

branches: 1.26.2;
Don't pull in the whole uvm(9) API to access only PAGE_SIZE and
some other constants. These are provided by sys/param.h now.

Use SYSCTL_DESCR()

Save initial, boot-time limit values, and restore them upon request
from sysmon_envsys(9).

don't divide by zero. ever.

add some more ADM1031 support:
- additional temperature and fan speed sensors work now
- limits on should work on all temperature sensors
Controlling fan behaviour doesn't work yet, the ADM1031 uses the same
register for both fans which doesn't quite fit with the way the dbcool
driver works.

Add preliminary support for the ADM1031 found in the Sun Blade 2500 and similar
machines. For now we treat it like an ADM1030 - the only difference is the
device ID and a few previously reserved registers.

Use correct format directive in aprint_debug_dev()

Thanks to Kurt Schreiner for reporting this on current-users@

Fix the dbcool_{get,set}_limits() routines to properly deal with min/max
values on temp limits. The actual ranges for temp limits varies if the
chip is running in "extended temperature" mode, so we need to handle this
difference, too.

Verified on my home machine which runs in extended-temp mode and by
njoly@ on a machine which does runs in the older 2s-complement mode.

branches: 1.18.2;
Now that sysmon_envsys(8) has learned how to retrieve and update sensor
limit values from the device driver, remove the high_limit and low_limit
entries from the hw.dbcool<#>.<sensor> sysctl(8) tree. These limits are
now directly accessible via envstat(8). Simplify the remaining code
somewhat.

No objections raised on either tech-kern@ or current-users@

XXX The fan controllers's parameters (including those that are tied to
XXX temperature sensors) are still in sysctl, since there isn't any clean
XXX way to implement the fan controllers in sysmon_envsys yet. It is on
XXX my To-Do list...

Disable DBCOOL_DEBUG by default, and make it compile that way.

Fix handling of 6bit VID register (6th bit takes place of 12VIn sensor).

Move VID out of sysctl tree - make it a separate sensor of type INTEGER.

Clean-up some of the chip-descriptors.

A pointer typedef entails trading too much flexibility to declare const
and non-const types, and the kernel uses both const and non-const
PMF qualifiers and device suspensors, so change the pmf_qual_t and
device_suspensor_t typedefs from "pointers to const" to non-pointer,
non-const types.

branches: 1.14.2;
Expand PMF_FN_* macros.

don't pull a 1.5K softc on the stack to do a chip ident. Idea from mrg.

Since we no longer have individual events for each sensor value limit,
we don't need individual flag bits. Clean up extra bit definitions.
Bump kernel version - welcome to 5.99.13

branches: 1.11.2;
Previous check for ADM1027 caused a kernel panic, since the read routine
can be called _before_ we've identified the chip. Ensure that sc_chip
is initialized to NULL before chip is identified, and if it is NULL or
ADM1027, use the less-efficient send_byte/receive_byte.

While here, don't attempt to release the i2c bus if we failed to acquire
it. This would cause a panic on a LOCKDEBUG kernel.

Thanks to Nicholas Joly for pointing out both of these bugs.

branches: 1.10.2;
The ADM1027 chip doesn't support i2c read_byte protocol, so in this
case we use send_byte followed by receive_byte.

Use pre-scripted i2c protocols rather than re-inventing our own. No
functional change intended.

While here, change some indentation to use tabs vs spaces.

Include ADT7460 in chip table (it's identical to the ADT7463, except for
physical packaging).

Add i2c bus address checking for these chips. All supported chips live
at predefined addresses 0x2c, 0x2d, or 0x2e.

PR/39947 - Juan RP -- dbcool_detach implicitly called
sysmon_envsys_destroy() twice

branches: 1.5.2; 1.5.4; 1.5.6;
Move the {read,write}reg() routine addresses into the dbcool_softc so
that we can use different routines for macppc's ki2c attachment. Fixes
problems reported by Robert Fritzsche on current-users@ list.

branches: 1.4.2;
sysctl doesn't like names that start with digits, nor does it like names
with embedded '.' so change the names to be better-conforming. Reported
by njoly

Increase accuracy in nominal voltage table.

Various clean-ups:
1. Trange sysctl's belong to the temp sensors, not to the fan controllers
2. Trange really describes a slope on many chips, so modify the description
3. Most of the sysctl's are read/write even if the chip's config is
locked; reflect that in the CTLFLAGs
4. Apply correct 'nominal' values for voltage sensors/limits, specify
them in microVolts, and calculate once rather than each time needed
5. Be more consistent in register names - for example, use VCC instead of
SUPPLY_VOLTAGE, to match VCC_LOWLIM & VCC_HIGHLIM
6. Type of dbcool_islocked() should be bool, not int
7. Reduce some unnecessary code indentation
8. Define Vtt and Imon, and add ADT7490 support (excluding PECI sensors)
9. Split the huge (250+ lines) dbcool_setup() function into a few smaller
routines for better readability
10. Update sensor tables for ADT7476 and ADT7468 - these chips have five
voltage sensors, not two
11. Adjust flags for ADT7463 and ADM1027 - these chips can monitor CPU
VID data bits
12. Update man page

branches: 1.1.2;
Add new driver for dbCool(tm) family of Thermal Monitor and Fan Controller.
Supported chips: ADM1027, ADM1030, ADT7463, ADT7466, ADT7467, ADT7468,
ADT7473, ADT7475, and ADT7476. Notably missing is the ADT7490, and fan
controller support on the ADT7466 is still on the to-do list.

Tested by myself and njoly@
Reviewed by garbled@
Commit approved by christos@, bouyer@, cube@, and matt@

Revert previous for now:
http://mail-index.netbsd.org/source-changes/2020/04/16/msg116278.html

The reasoning turned out to be wrong; __KERNEL_RCSID() in header files
does *not* overwrite RCSID in main source files. The real problem is that
it inserts its RCSID into *every* object files. However, it can be still
useful even if heavily duplicated.

Stop using __KERNEL_RCSID() in header files; it confuses ident(1) by
overwriting RCSID in main source files.

XXX
The first argument of __KERNEL_RCSID() is neglected for ELF. If we wish
to have RCSID of header files in kernel binary, we need something like
__FBSDID() macro in FreeBSD.

branches: 1.7.42;
Allow i2c addr wildcard matching. Use with care!

branches: 1.6.4; 1.6.14; 1.6.18;
Add support for SMSC EMC6D103S

From jasper@pointless.net in PR kern/44685

branches: 1.5.2;
add some more ADM1031 support:
- additional temperature and fan speed sensors work now
- limits on should work on all temperature sensors
Controlling fan behaviour doesn't work yet, the ADM1031 uses the same
register for both fans which doesn't quite fit with the way the dbcool
driver works.

Add preliminary support for the ADM1031 found in the Sun Blade 2500 and similar
machines. For now we treat it like an ADM1030 - the only difference is the
device ID and a few previously reserved registers.

branches: 1.3.4; 1.3.6; 1.3.8;
Add i2c bus address checking for these chips. All supported chips live
at predefined addresses 0x2c, 0x2d, or 0x2e.

branches: 1.2.2; 1.2.4; 1.2.6; 1.2.8;
Various clean-ups:
1. Trange sysctl's belong to the temp sensors, not to the fan controllers
2. Trange really describes a slope on many chips, so modify the description
3. Most of the sysctl's are read/write even if the chip's config is
locked; reflect that in the CTLFLAGs
4. Apply correct 'nominal' values for voltage sensors/limits, specify
them in microVolts, and calculate once rather than each time needed
5. Be more consistent in register names - for example, use VCC instead of
SUPPLY_VOLTAGE, to match VCC_LOWLIM & VCC_HIGHLIM
6. Type of dbcool_islocked() should be bool, not int
7. Reduce some unnecessary code indentation
8. Define Vtt and Imon, and add ADT7490 support (excluding PECI sensors)
9. Split the huge (250+ lines) dbcool_setup() function into a few smaller
routines for better readability
10. Update sensor tables for ADT7476 and ADT7468 - these chips have five
voltage sensors, not two
11. Adjust flags for ADT7463 and ADM1027 - these chips can monitor CPU
VID data bits
12. Update man page

branches: 1.1.2;
Add new driver for dbCool(tm) family of Thermal Monitor and Fan Controller.
Supported chips: ADM1027, ADM1030, ADT7463, ADT7466, ADT7467, ADT7468,
ADT7473, ADT7475, and ADT7476. Notably missing is the ADT7490, and fan
controller support on the ADT7466 is still on the to-do list.

Tested by myself and njoly@
Reviewed by garbled@
Commit approved by christos@, bouyer@, cube@, and matt@

Get the sensor descriptive information from our device properties
dictionary, rather than from a secondary dictionary passed in the
i2c_attach_args (this secondary dictionary is deprecated and going
away).

Make this compile

Revert previous for now:
http://mail-index.netbsd.org/source-changes/2020/04/16/msg116278.html

The reasoning turned out to be wrong; __KERNEL_RCSID() in header files
does *not* overwrite RCSID in main source files. The real problem is that
it inserts its RCSID into *every* object files. However, it can be still
useful even if heavily duplicated.

Stop using __KERNEL_RCSID() in header files; it confuses ident(1) by
overwriting RCSID in main source files.

XXX
The first argument of __KERNEL_RCSID() is neglected for ELF. If we wish
to have RCSID of header files in kernel binary, we need something like
__FBSDID() macro in FreeBSD.

branches: 1.14.66;
When detaching the device, tear down its sysctl tree.

Set DBCOOL_DEBUG if global DEBUG is set

branches: 1.12.2;
Save initial, boot-time limit values, and restore them upon request
from sysmon_envsys(9).

branches: 1.11.2;
Now that sysmon_envsys(8) has learned how to retrieve and update sensor
limit values from the device driver, remove the high_limit and low_limit
entries from the hw.dbcool<#>.<sensor> sysctl(8) tree. These limits are
now directly accessible via envstat(8). Simplify the remaining code
somewhat.

No objections raised on either tech-kern@ or current-users@

XXX The fan controllers's parameters (including those that are tied to
XXX temperature sensors) are still in sysctl, since there isn't any clean
XXX way to implement the fan controllers in sysmon_envsys yet. It is on
XXX my To-Do list...

Disable DBCOOL_DEBUG by default, and make it compile that way.

Fix handling of 6bit VID register (6th bit takes place of 12VIn sensor).

Move VID out of sysctl tree - make it a separate sensor of type INTEGER.

Clean-up some of the chip-descriptors.

A pointer typedef entails trading too much flexibility to declare const
and non-const types, and the kernel uses both const and non-const
PMF qualifiers and device suspensors, so change the pmf_qual_t and
device_suspensor_t typedefs from "pointers to const" to non-pointer,
non-const types.

branches: 1.7.2;
Expand PMF_FN_* macros.

don't pull a 1.5K softc on the stack to do a chip ident. Idea from mrg.

struct device * -> device_t, no functional changes intended.

branches: 1.4.2; 1.4.10; 1.4.14;
Move the {read,write}reg() routine addresses into the dbcool_softc so
that we can use different routines for macppc's ki2c attachment. Fixes
problems reported by Robert Fritzsche on current-users@ list.

branches: 1.3.2;
It's not necessary to #include "sysmon_envsys.h" any more.

Various clean-ups:
1. Trange sysctl's belong to the temp sensors, not to the fan controllers
2. Trange really describes a slope on many chips, so modify the description
3. Most of the sysctl's are read/write even if the chip's config is
locked; reflect that in the CTLFLAGs
4. Apply correct 'nominal' values for voltage sensors/limits, specify
them in microVolts, and calculate once rather than each time needed
5. Be more consistent in register names - for example, use VCC instead of
SUPPLY_VOLTAGE, to match VCC_LOWLIM & VCC_HIGHLIM
6. Type of dbcool_islocked() should be bool, not int
7. Reduce some unnecessary code indentation
8. Define Vtt and Imon, and add ADT7490 support (excluding PECI sensors)
9. Split the huge (250+ lines) dbcool_setup() function into a few smaller
routines for better readability
10. Update sensor tables for ADT7476 and ADT7468 - these chips have five
voltage sensors, not two
11. Adjust flags for ADT7463 and ADM1027 - these chips can monitor CPU
VID data bits
12. Update man page

branches: 1.1.2;
Add new driver for dbCool(tm) family of Thermal Monitor and Fan Controller.
Supported chips: ADM1027, ADM1030, ADT7463, ADT7466, ADT7467, ADT7468,
ADT7473, ADT7475, and ADT7476. Notably missing is the ADT7490, and fan
controller support on the ADT7466 is still on the to-do list.

Tested by myself and njoly@
Reviewed by garbled@
Commit approved by christos@, bouyer@, cube@, and matt@

No need to use I2C_F_POLL here.

Rename min/max -> uimin/uimax for better honesty.

These functions are defined on unsigned int. The generic name
min/max should not silently truncate to 32 bits on 64-bit systems.
This is purely a name change -- no functional change intended.

HOWEVER! Some subsystems have

#define min(a, b) ((a) < (b) ? (a) : (b))
#define max(a, b) ((a) > (b) ? (a) : (b))

even though our standard name for that is MIN/MAX. Although these
may invite multiple evaluation bugs, these do _not_ cause integer
truncation.

To avoid `fixing' these cases, I first changed the name in libkern,
and then compile-tested every file where min/max occurred in order to
confirm that it failed -- and thus confirm that nothing shadowed
min/max -- before changing it.

I have left a handful of bootloaders that are too annoying to
compile-test, and some dead code:

cobalt ews4800mips hp300 hppa ia64 luna68k vax
acorn32/if_ie.c (not included in any kernels)
macppc/if_gm.c (superseded by gem(4))

It should be easy to fix the fallout once identified -- this way of
doing things fails safe, and the goal here, after all, is to _avoid_
silent integer truncations, not introduce them.

Maybe one day we can reintroduce min/max as type-generic things that
never silently truncate. But we should avoid doing that for a while,
so that existing code has a chance to be detected by the compiler for
conversion to uimin/uimax without changing the semantics until we can
properly audit it all. (Who knows, maybe in some cases integer
truncation is actually intended!)

branches: 1.7.2;
More cleanup to i2c autoconfiguration:

- Get all of the drivers onto the new match quality constants.
- Introduce a new helper function, iic_use_direct_match(), that has
all of the logic for direct-config matching. If it returns true,
the driver returns the match result (which may be 0). If it returns
false, the driver does indirect-config matching.
- iic_compat_match() now returns a weighted match quality; matches to
lower-indexed "compatible" device property are more-specific matches,
and return a better match quality accordingly.

branches: 1.6.16;
Fix block offsets for ddc_read_edid_block.

add ddc_dev_read_edid{,_block} functions, same as ddc_read_edid{,_block} but takes a device_t instead of i2c_tag_t

branches: 1.4.2;
add ddc_read_edid_block, which is the same as ddc_read_edid but takes an additional block number argument

branches: 1.3.44; 1.3.62;
device_t/softc split and other related cosmetic changes.

branches: 1.2.16; 1.2.18; 1.2.20;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.1.2; 1.1.6; 1.1.8; 1.1.10; 1.1.14; 1.1.20; 1.1.30; 1.1.44; 1.1.46; 1.1.50;
Add basic DDC (VESA Data Display Channel) driver for use by framebuffers.
Reviewed by jmcneill@ and macallan@.

Add a definition for the DDC segment address (30h).

branches: 1.2.4; 1.2.6; 1.2.8; 1.2.12; 1.2.18; 1.2.162;
integrate kauth.

branches: 1.1.2;
Add basic DDC (VESA Data Display Channel) driver for use by framebuffers.
Reviewed by jmcneill@ and macallan@.

add ddc_dev_read_edid{,_block} functions, same as ddc_read_edid{,_block} but takes a device_t instead of i2c_tag_t

branches: 1.2.2;
add ddc_read_edid_block, which is the same as ddc_read_edid but takes an additional block number argument

branches: 1.1.2; 1.1.6; 1.1.8; 1.1.10; 1.1.14; 1.1.20; 1.1.120; 1.1.138;
Add basic DDC (VESA Data Display Channel) driver for use by framebuffers.
Reviewed by jmcneill@ and macallan@.

g/c opt_dsrtc.h reference

Garbage-collect the "base_year_is_2000" property and associated
DSRTC_YEAR_START_2K config option (only used by Iyonix GENERIC).
Instead, set the "start-year" property to 2000 as needed (matches
the property name in the Devicetree RTC bindings used for the same
purpose).

Change todr_chip_handle::cookie -> todr_chip_handle::todr_dev, and
make it a device_t. Upcoming functional changes will require the
device_t associated with a TODR device.

Change todr_chip_handle::bus_cookie -> todr_chip_handle::todr_devaux.
Nothing was using the old field, but I decided to keep it around just
in cause something needs it in the future.

And with these largely mechanical yet semantically meaningful changes,
thus spake the Oracle: "Welcome to NetBSD 11.99.2."

Remove unnecessary NULL-initialization of TODR handle fields.

Don't panic with invalid user data, just return an error.

Fix various typos in comments. Also add missing NetBSD RCS Id in some of these files.

Use DEVICE_COMPAT_EOL.

Since we're using designated initialisers for compat data, we should
use a completely empty initializer for the sentinel.

Change the device_compatible_match() function to only perform the match.
Introduce a device_compatible_lookup() function to return an entry based
on the same matching criteria (a'la of_search_compatible()).

Update iic_compatible_match() to reflect the above change, and introduce
iic_compatible_lookup(). This pattern is less awkward to use.

Change the device_compatible_entry struct to have a union of
a scalar value (uintptr_t value) and a pointer value (const void *data),
rather than just "uintptr_t data". This eliminates the need for drivers
to cast this value, and doesn't lose const'ness of pointers assigned
to the field.

Update all of the users of this field to use the correct set of
designated initialisers for each specific case.

branches: 1.34.8;
No need to use I2C_F_POLL here.

appease -Werror=stack-protector by avoiding variable sized arrays
Now this builds and works on sparc64

add another matchstring for the ds1307 found in Sun Fire V210 ( and probably
V240 )

make DSRTC_FLAG_YEAR_START_2K settable by device property

add options DSRTC_YEAR_START_2K for machines which use 2000 and not 1970
as base to count years from, like Iyonix.
While there apply the offset when writing to the clock as well.