[PATCH linux v1 1/2] Documentation: dt-bindings: Document bindings for ASPEED AST2400/AST2500 pwm and fan tach controller device driver
Jaghathiswari Rankappagounder Natarajan
jaghu at google.com
Fri Jan 27 20:32:59 AEDT 2017
On Fri, Jan 13, 2017 at 8:21 AM, Rob Herring <robh at kernel.org> wrote:
> On Mon, Jan 09, 2017 at 01:59:34PM -0800, Jaghathiswari Rankappagounder
> Natarajan wrote:
> > This binding provides interface for adding values related to ASPEED
> > AST2400/2500 PWM and Fan tach controller support.
> > The PWM controller can support upto 8 PWM output ports.
> > The Fan tach controller can support upto 16 tachometer inputs.
> > PWM clock types M, N and 0 are three types just to have three independent
> > PWM sources.
> >
> > Signed-off-by: Jaghathiswari Rankappagounder Natarajan <jaghu at google.com
> >
> > ---
> > .../devicetree/bindings/hwmon/aspeed-pwm-tacho.txt | 153
> +++++++++++++++++++++
>
> Perhaps bindings/pwm/... even though this is more than just PWM.
>
The corresponding hwmon driver is present in /drivers/hwmon/. So would it
make more
sense to include the devicetree document for that driver in
/devicetree/bindings/hwmon/.
In future if there any more hwmon related functionalities are added to this
driver then /devicetree/bindings/hwmon/ would be more relevant.
>
> > 1 file changed, 153 insertions(+)
> > create mode 100644 Documentation/devicetree/bindings/hwmon/aspeed-pwm-
> tacho.txt
> >
> > diff --git a/Documentation/devicetree/bindings/hwmon/aspeed-pwm-tacho.txt
> b/Documentation/devicetree/bindings/hwmon/aspeed-pwm-tacho.txt
> > new file mode 100644
> > index 000000000000..8f346409ee8c
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/hwmon/aspeed-pwm-tacho.txt
> > @@ -0,0 +1,153 @@
> > +ASPEED AST2400/AST2500 PWM and Fan Tacho controller device driver
> > +
> > +The ASPEED PWM controller can support upto 8 PWM outputs. The ASPEED
> Fan Tacho
> > +controller can support upto 16 tachometer inputs. The PWM controller
> supports
> > +3 types of frequency mode PWM for fan speed control. PWM clock types M,
> N and 0
> > +are 3 types of frequency mode PWM just to have 3 independent PWM
> sources.
> > +
> > +Required properties for pwm_tacho node:
> > +- #address-cells : should be 1.
> > +
> > +- #size-cells : should be 1.
> > +
> > +- reg : address and length of the register set for the device.
> > +
> > +- pinctrl-names : a pinctrl state named "default" must be defined.
> > +
> > +- pinctrl-0 : phandle referencing pin configuration of the
> AST2400/AST2500 PWM
> > + ports.
> > +
> > +- compatible : should be "aspeed,aspeed2400-pwm-tacho" for AST2400 or
> > + "aspeed,aspeed2500-pwm-tacho" for AST2500.
> > +
> > +- clocks : a fixed clock providing input clock frequency(PWM
> > + and Fan Tach clock)
> > +
> > +type_values subnode format:
>
> Don't use '_' in node or property names.
>
Done.
>
> > +===========================
> > +Under type_values subnode there can be upto 3 child nodes indicating
> type M/N/O
> > +values. Atleast one child node is required.
> > +
> > +Required properties for the child node(type M/N/O):
> > +- pwm_period : indicates type M/N/O PWM period, as per the
> AST2400/AST2500
> > + datasheet. integer value in the range 0 to 255.
>
>
> > +
> > +- pwm_clock_division_l : indicates type M/N/O PWM clock division L
> value,
> > + as per the AST2400/AST2500 datasheet.
> > + integer value in the range 0 to 15.
> > + 0 here indicates divide 1, 1 indicates divide 2,
> > + 2 indicates divide 4, 3 indicates divide 6, and
> so on
> > + till 15 indicates divide 30.
> > +
> > +- pwm_clock_division_h : indicates type M/N/O PWM clock division H
> value,
> > + as per the AST2400/AST2500 datasheet.
> > + integer value in the range 0 to 15.
> > + 0 here indicates divide 1, 1 indicates divide 2,
> > + 2 indicates divide 4, 3 indicates divide 8, and
> so on
> > + till 15 indicates divide 32768.
>
> Can't you have a single divider value and driver convert to register
> values? Really, you should specify the PWM period in ns/us and calculate
> the divider based on the input clock freq. (i.e. use the clock binding).
> There's already PWM binding to specify the period.
>
> I think you should have a node for the fan using the PWM binding and
> perhaps moving some of these properties to the fan node.
>
There are three different types M, N, O. These three types are just to have
different PWM/Fan Tach settings. Each type can have the following settings:
PWM related:
1) PWM period
2) PWM clock division high
3) PWM clock division low
Fan Tach related:
1) Fan Tach type enable
2) Fan Tach clock division
3) Fan Tach mode selection
4) Fan Tach period
Each PWM port should be assigned a type which can be type M, N or O.
Each Fan Tach channel should be assigned a PWM source, which can be PWM
port A
through H. The Fan Tach channel implicitly gets the type (M, N or O), from
the type assigned to its PWM source.
If the clock type is type M then :
The PWM frequency = 24MHz / (type M clock division L value * type M
clock division H value *
(type M PWM period value + 1))
Because of this formula, I thought I could use properties to specify
type clock division High and Low and
type period values so that the desired PWM frequency be achieved.
> > +
> > +- fan_tach_enable : indicates fan tach enable of type M/N/O as per the
> > + AST2400/AST2500 datasheet. boolean value.
> > +
> > +- fan_tach_clock_division : indicates fan tach clock division as per the
> > + AST2400/AST2500 datasheet.
> > + integer value in the range 0 to 7.
> > + 0 indicates divide 4, 1 indicates divide 16,
> > + 2 indicates divide 64, 3 indicates divide 256
> > + and so on till 7 indicates divide 65536.
> > +
> > +- fan_tach_mode_selection : indicates fan tach mode mode selection as
> per the
> > + AST2400/AST2500 datasheet. integer value in the
> > + range 0 to 2. 0 indicates falling edge, 1
> indicates
> > + rising edge and 2 indicates both edges.
> > +
> > +- fan_tach_period : indicates fan tach period as per the AST2400/AST2500
> > + datasheet. integer value (can be upto 16 bits long).
> > +
> > +pwm_port subnode format:
> > +========================
> > +Under pwm_port subnode there can upto 8 child nodes each indicating
> values
> > +for one of the 8 PWM output ports.
> > +
> > +Required properties for each child node(starting from PWM A through PWM
> H):
> > +- enable : enable PWM #X port, X ranges from A through H. boolean value.
>
> A connection in the PWM binding would imply this.
>
Had to change this property to indicate PWM port number.
>
> > +
> > +- type : indicates type selection value of PWM #X port, X ranges from A
> > + through H. integer value in the range 0 to 2;
> > + 0 indicates type M, 1 indicates type N, 2 indicates type O.
>
> This is meaningless to me.
>
Each PWM port should be assigned a type which can be type M, N or O.
>
> > +- fan_ctrl : set the PWM fan control initial value. integer value
> between
> > + 0(off) and 255(full speed).
> > +
> > +fan_tach_channel subnode format:
> > +================================
> > +Under fan_tach_channel subnode there can be upto 16 child nodes each
> indicating
> > +values for one of the 16 fan tach channels.
> > +
> > +Required properties for each child node(starting from fan tach #0
> through
> > +fan tach #16):
> > +- fan-ctrl-gpios : should specify the tachometer input pin on the
> hardware.
>
> I don't understand the connection here. You have a tach block with 16
> inputs on the SoC, so why the GPIOs? If the GPIOs go to the fan, then a
> fan node should have the GPIO properties.
>
This property specifies the tachometer input GPIO pin on the hardware.
Fan Tachometer function
can only work when GPIO is in ”input mode” .
> > +
> > +- enable : enable fan tach #X, X ranges from 0 through 16. boolean
> value.
>
> 'status' property is the standard way to enable a node or not. Not sure
> if that makes sense here though.
>
Had to change this property to indicate Fan tach channel number.
>
> > +
> > +- pwm_source : indicates PWM source of fan tach #X, X ranges from 0
> through 16.
> > + integer value in the range 0 to 7. 0 indicates PWM port A,
> > + 1 indicates PWM port B and so on till 7 indicates PWM port
> H.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ozlabs.org/pipermail/openbmc/attachments/20170127/246d7ba5/attachment-0001.html>
More information about the openbmc
mailing list