lkml.org 
[lkml]   [2008]   [Jul]   [4]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
/
DateFri, 4 Jul 2008 17:42:46 -0600
From"Grant Likely" <>
SubjectRe: [PATCH v2 3/5] of-bindings: Add binding documentation for SPI busses and devices
On Fri, Jul 4, 2008 at 5:36 PM, Segher Boessenkool
<segher@kernel.crashing.org> wrote:
>> +    The SPI master node requires the following properties:
>> +    - #address-cells  - number of cells required to define a chip select
>> +                       address on the SPI bus.>> Hrm.  Should this (and "reg" in the child node) be required for SPI
> masters that have only one chip select?

I suppose it could be skipped, but I'd rather have it there for consistency

>>> +    - max-speed       - (required) Maximum SPI clocking speed of device
>> in Hz>> The property name should include something "SPI", it's way too generic
> otherwise.

Good point; I'll change this to 'spi-max-speed'

>> +    - spi,cpol        - (optional) Device requires inverse clock polarity
>> +    - spi,cpha        - (optional) Device requires shifted clock phase>> Don't abbr the property names, there's nothing wrong with longer names.
> The names shouldn't start with "spi," either, "spi" isn't a vendor;
> how about "spi-inverse-clock-polarity" or similar?

Okay, but cpol and cpha are common abbreviations w.r.t. SPI devices.

>> +    - linux,modalias  - (optional, Linux specific) Force binding of SPI
>> device>> +                       to a particular spi_device driver.  Useful for
>> changing>> +                       driver binding between spidev and a kernel SPI
>> driver.>> This is a temporary workaround I hope?

Yeah, I'm kind of ashamed of this one.  I'll drop it.

g.

-- 
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.


\
 
 \ /
  Last update: 2008-07-05 01:45    [from the cache]
©2003-2008