Re: [PATCH] of: get of_root directly in place of of_find_node_by_path("/")

[Tyrel Datwyler <tyreld@linux.vnet.ibm.com>]

On 03/07/2018 12:38 PM, Rob Herring wrote:
> On Fri, Mar 02, 2018 at 08:46:44PM -0500, Tyrel Datwyler wrote:
>> There are a couple places in drivers/of where the root node is found
>> with a of_find_node_by_path("/") call. This call just returns
>> of_node_get(of_root) when the the path "/" is passed as a parameter.
>> So, lets fixup these instances to just do that instead.
>>
>> Signed-off-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
>> ---
>>  drivers/of/base.c     | 2 +-
>>  drivers/of/platform.c | 4 ++--
>>  2 files changed, 3 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/of/base.c b/drivers/of/base.c
>> index ad28de9..94c1b4d 100644
>> --- a/drivers/of/base.c
>> +++ b/drivers/of/base.c
>> @@ -455,7 +455,7 @@ int of_machine_is_compatible(const char *compat)
>>  	struct device_node *root;
>>  	int rc = 0;
>>  
>> -	root = of_find_node_by_path("/");
>> +	root = of_node_get(of_root);
> 
> The former seems more readable to me. I'll wait and see if Frank has any 
> comments on this change.

Fair enough. I find them equally readable now that we have a well named/defined global, namely "of_root",  pointing at the device tree root. My primary goal was saving a function call, but I wasn't sure if anybody really cares enough. A quick check shows ~77 uses of "of_find_node_by_path("/")" in the kernel. Figured I'd start here and see what the consensus was.

-Tyrel

> 
> Rob
>