Re: [PATCH] netdev/phy: skip disabled mdio-mux nodes

Re: [PATCH] netdev/phy: skip disabled mdio-mux nodes

[David Daney]

On 08/07/2012 09:51 AM, Timur Tabi wrote:
> The mdio-mux driver scans all child mdio nodes, without regard to whether
> the node is actually used.  Some device trees include all possible
> mdio-mux nodes and rely on the boot loader to disable those that are not
> present, based on some run-time configuration.  Those nodes need to be
> skipped.
>
> Signed-off-by: Timur Tabi<timur@freescale.com>
> ---
>   drivers/net/phy/mdio-mux.c |    9 +++++++++
>   1 files changed, 9 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/net/phy/mdio-mux.c b/drivers/net/phy/mdio-mux.c
> index 5c12018..d0c231e 100644
> --- a/drivers/net/phy/mdio-mux.c
> +++ b/drivers/net/phy/mdio-mux.c
> @@ -135,6 +135,15 @@ int mdio_mux_init(struct device *dev,
>   	for_each_child_of_node(dev->of_node, child_bus_node) {
>   		u32 v;
>
> +		/*
> +		 * Some device trees include all possible mdio-mux nodes and
> +		 * rely on the boot loader to disable those that are not
> +		 * present, based on some run-time configuration.  Those nodes
> +		 * need to be skipped.
> +		 */
> +		if (!of_device_is_available(child_bus_node))
> +			continue;


Although this will get the job done, I don't think it is the cleanest 
approach.

Would it be better to create a new iterator 
(for_each_available_child_of_node perhaps) that skips the unavailable 
nodes?  This seems like a general problem that is not restricted to mdio 
multiplexers.

David Daney

Re: [PATCH] netdev/phy: skip disabled mdio-mux nodes

[Timur Tabi]

David Daney wrote:
> Although this will get the job done, I don't think it is the cleanest 
> approach.
> 
> Would it be better to create a new iterator 
> (for_each_available_child_of_node perhaps) that skips the unavailable 
> nodes?  This seems like a general problem that is not restricted to mdio 
> multiplexers.

You're probably right, but this is how it's normally done.  If you're
going to scan child nodes manually, you have to skip disabled nodes.

If someone else wants to implement for_each_available_child_of_node(), I'm
all for it, but such a patch would not be merged until 3.7.  I was hoping
to have my patch applied to 3.6, since it fixes a real bug.

-- 
Timur Tabi
Linux kernel developer at Freescale