[PATCH 3/3 V2] ARM: pxa: Add basic DTS files for PXA/Vpac270 testing machine

[Rob Herring]

On 11/07/2011 03:55 PM, Marek Vasut wrote:
>> On 11/07/2011 03:31 PM, Marek Vasut wrote:
>>> Signed-off-by: Marek Vasut <marek.vasut at gmail.com>
>>> Cc: Arnd Bergmann <arnd at arndb.de>
>>> Cc: Grant Likely <grant.likely at secretlab.ca>
>>> Cc: Rob Herring <robherring2 at gmail.com>
>>> ---
>>
>> Couple of minor things below. With those fixed:
>>
>> Acked-by: Rob Herring <rob.herring at calxeda.com>
>>
>>>  arch/arm/boot/dts/pxa-vpac270.dts |   37 +++++++++++++++++++
>>>  arch/arm/boot/dts/pxa.dtsi        |   71
>>>  +++++++++++++++++++++++++++++++++++++ 2 files changed, 108
>>>  insertions(+), 0 deletions(-)
>>>  create mode 100644 arch/arm/boot/dts/pxa-vpac270.dts
>>>  create mode 100644 arch/arm/boot/dts/pxa.dtsi
>>>
>>> V2: Fix typo -- rename alias -> aliases in the pxa270.dtsi
>>>
>>>     Drop marvell,pxa250 option, instead leave all ports disable and let
>>>     user select, which will be registered. This depends on the user not
>>>     doing something stupid (like registering HWUART, which is available
>>>     only on pxa250) on pxa320.
>>
>> Should be a safe assumption. Only someone capable of implementing board
>> support should create a dts...
>>
>>> diff --git a/arch/arm/boot/dts/pxa-vpac270.dts
>>> b/arch/arm/boot/dts/pxa-vpac270.dts new file mode 100644
>>> index 0000000..48dd9e5
>>> --- /dev/null
>>> +++ b/arch/arm/boot/dts/pxa-vpac270.dts
>>> @@ -0,0 +1,37 @@
>>> +/*
>>> + * pxa-vpac270.dts - Device Tree file for Voipac PXA270 board
>>> + *
>>> + * Copyright (C) 2011 Marek Vasut <marek.vasut at gmail.com>
>>> + *
>>> + * Licensed under GPLv2 or later.
>>> + */
>>> +/dts-v1/;
>>> +/include/ "pxa.dtsi"
>>> +
>>> +/ {
>>> +	model = "Voipac VPAC270";
>>> +	compatible = "voipac,vpac270";
>>> +
>>> +	chosen {
>>> +		bootargs = "console=ttyS0,115200 mem=128M at 0xa0000000";
>>
>> Isn't mem= redundant with the memory node below?
> 
> Yes it is, but ...
> 
> the board has 256MB or RAM, PXA assumes ram base at 0xa0000000, but the other 
> node is at 0x80000000. U-Boot passes both nodes to kernel, kernel gets confused 
> and crashes. Basically, we need to wait until uboot supports relative uImages 
> (should be soon) and then we can drop this and use whole 256MB of RAM.
> 

Okay.

>>
>>> +		linux,stdout-path = &ffuart;
>>> +	};
>>> +
>>> +	memory at a0000000 {
>>> +		reg = <0xa0000000 0x8000000>;
>>> +	};
>>> +
>>> +	pxabus {
>>> +		ffuart: uart at 40100000 {
>>
>> These should be serial at ...
> 
> Thanks, but why ? Is there some rule about it ?

There's generic names defined in ePAPR spec (It's a powerpc spec, but
many parts still apply).

https://www.power.org/resources/downloads/Power_ePAPR_APPROVED_v1.0.pdf

Rob