Stephen Boyd <sboyd@xxxxxxxxxxxxxx> writes:
> On 12/01, Jerome Brunet wrote:
>> On Fri, 2017-12-01 at 08:34 -0800, Stephen Boyd wrote:
>> > On 11/30, Yixun Lan wrote:
>> > > Hi Stephen
>> > >
>> > > On 11/30/17 03:35, Stephen Boyd wrote:
>> > > >
>> > > > Maybe just call the node "bus@ff63c000"?
>> > > >
>> > >
>> > > isn't this just a name? what's the benefits to change?
>> > > personally, I tend to keep it this way, because it's better map to the
>> > > data sheet
>> > >
>> > > we also has 'aobus', 'cbus' scattered there..
>> >
>> > Per the ePAPR node names are supposed to be generic, like disk,
>> > cpu, display-controller, gpu, etc. I've never heard of a hiubus,
>> > so probably it's some vendor specific thing? We have the phandle
>> > anyway so it's not like we're losing much information here.
>>
>> Stephen, there is a lot of busses on platform. We can't just call them all
>> 'bus'.
>> I don't get the problem with this name.
>> We are re-using the name from the datasheet here, no fancy invention. It seems
>> to be quite common.
>>
>
> Ok. I'm not the maintainer of the DTS so no worries from me. I'm
> just pointing out that the ePAPR says that node names should be
> generic, and 'hiubus' doesn't sound generic to me. If it matches
> some datasheet then I suppose that's good, but probably that sort
> of distinction should have gone into the compatible string
> instead of the node name.
Stephen is right, the node-name should be generic (e.g. "bus") but the
label can (should) be more SoC-specific, so it should look like:
hiubus: bus@ff63c000 {
Note that we weren't strict about this for all the rest of the amlogic
SoCs (mostly because I didn't notice ) but we should start doing it
correctly now. I'll also clean up the existing DTs.
Kevin
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
