Messages in this thread | | | Date | Wed, 17 Jun 2015 11:01:42 -0600 | From | Sagar Dharia <> | Subject | Re: [PATCH V2 2/6] of/slimbus: OF helper for SLIMbus |
| |
On 6/17/2015 7:09 AM, Mark Brown wrote: > On Tue, Jun 16, 2015 at 07:46:00PM -0600, Sagar Dharia wrote: >> OF helper routine scans the SLIMbus DeviceTree, allocates resources, >> and creates slim_devices according to the hierarchy. > You've not CCed any of the DT maintainers on this, for a completely new > bus it seems like we really ought to get their input. Will do. > >> @@ -0,0 +1,34 @@ >> +SLIM(Serial Low Power Interchip Media Bus) bus >> + >> +SLIMbus is a 2-wire bus, and is used to communicate with peripheral >> +components like audio-codec. >> + >> +Required property for SLIMbus controller node: >> +- compatible - name of SLIMbus controller. >> + >> +No other properties are required in the SLIMbus controller bus node. > It seems better to just say that the controller is a normal device using > the binding for whatever bus it is on and that the binding is for the > bus which is a child node of the controller device? Also, do we need > #address-cells or #size-cells here? It will be required if we use 'reg' property for enumeration address per my understanding (address-cells: 6, size-cells: 0) I thought about using 'reg' and it seemed to work as well. Only reason for using 'enumeration-addr' was to be closer to what HW calls it. Looking at other device-trees, (that 'reg' was used for other buses to represent slave address), I will document the 'reg' property and document it. Thanks Sagar > >> +Required property for SLIMbus child node: >> +enumeration-addr - 6 byte enumeration address of the slave > The idiom for DT seems to be that we define the bus address using the > reg property. Should we not be following that pattern here too? I'd > also expect to see the ability to define a compatible for the slaves.
-- Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation
| |