[PATCH v4 32/71] ARC: [DeviceTree] Basic support

[Vineet Gupta]

On Wednesday 30 January 2013 04:38 PM, James Hogan wrote:
> Hi Vineet
>
> On 29/01/13 13:55, Vineet Gupta wrote:
>> diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
>> index a353849..7666857 100644
>> --- a/arch/arc/Kconfig
>> +++ b/arch/arc/Kconfig
>> @@ -24,8 +24,11 @@ config ARC
>> +	select OF
>> diff --git a/arch/arc/Makefile b/arch/arc/Makefile
>> index 4d52a3b..29b5fcd 100644
>> --- a/arch/arc/Makefile
>> +++ b/arch/arc/Makefile
>> @@ -83,6 +83,9 @@ head-y		:= arch/arc/kernel/head.o
>> +# w/o this dtb won't embed into kernel binary
>> +core-y		+= arch/arc/boot/dts/
>> diff --git a/arch/arc/boot/dts/Makefile b/arch/arc/boot/dts/Makefile
>> new file mode 100644
>> index 0000000..ea3114c
>> --- /dev/null
>> +++ b/arch/arc/boot/dts/Makefile
>> @@ -0,0 +1,17 @@
>> +ifeq ($(CONFIG_OF),y)
> This doesn't seem necessary since you select OF (and if you didn't you
> could put core-$(CONFIG_OF) in the main Makefile.

Right - taken care of.

>
>> diff --git a/arch/arc/kernel/Makefile b/arch/arc/kernel/Makefile
>> index 6d83431..eeab7c79 100644
>> --- a/arch/arc/kernel/Makefile
>> +++ b/arch/arc/kernel/Makefile
>> @@ -7,6 +7,7 @@
>>  
>>  obj-y	:= arcksyms.o setup.o irq.o time.o reset.o ptrace.o entry.o process.o
>>  obj-y	+= signal.o traps.o sys.o troubleshoot.o stacktrace.o clk.o
>> +obj-y   += devtree.o
> nit: whitespace is inconsistent here.

OK !

>> diff --git a/arch/arc/kernel/setup.c b/arch/arc/kernel/setup.c
>> index 82ac206..27aebd6 100644
>> --- a/arch/arc/kernel/setup.c
>> +++ b/arch/arc/kernel/setup.c
>> @@ -71,6 +76,8 @@ void __init setup_arch(char **cmdline_p)
>>  	strlcpy(boot_command_line, command_line, COMMAND_LINE_SIZE);
>>  	*cmdline_p = command_line;
>>  
>> +	rc = setup_machine_fdt(__dtb_start);
> Should this panic or something if it fails? Actually I see it does in
> your -next branch, so I guess it's fixed in a later patch.

Yes, it does in [62/71] when the return value is a machine description which is
NULL. Since my DT support was incrementally added - this patch only enables
printing the board name from the "early" flattened DT and does no more - and
doesn't care if that succeeds or not.

> So other than the stuff above:
> Reviewed-by: James Hogan <james.hogan at imgtec.com>

Thx,
-Vineet