Re: [PATCH v3 2/2] nvmem: core: skip nodes with compatibles other than "nvmem-cell"

[Srinivas Kandagatla <srinivas.kandagatla@linaro.org>]

On 28/04/2020 12:18, Ahmad Fatoum wrote:
> The nvmem cell binding used to apply to all objects which match
> "^.*@[0-9a-f]+$", without taking a compatible into account, which
> precluded extension of EEPROMs by child nodes other than nvmem.
> 
> A previous commit changed the binding, so that nvmem cells that
> feature a compatible property must have "nvmem-cell" as the value,
> otherwise they are skipped.
> 
> Adjust the driver to observe the new binding change. This change
> does not change behavior for any device tree that was already
> compliant with the nvmem binding.
> 
> Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>

Thanks for the patch, this looks good to me, lets wait for Rob to ack 
the bindings!

--srini
> ---
>   drivers/nvmem/core.c | 4 ++++
>   1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/nvmem/core.c b/drivers/nvmem/core.c
> index 05c6ae4b0b97..eb697f5ad07d 100644
> --- a/drivers/nvmem/core.c
> +++ b/drivers/nvmem/core.c
> @@ -547,6 +547,10 @@ static int nvmem_add_cells_from_of(struct nvmem_device *nvmem)
>   	parent = dev->of_node;
>   
>   	for_each_child_of_node(parent, child) {
> +		if (of_find_property(child, "compatible", NULL) &&
> +		    !of_device_is_compatible(child, "nvmem-cell"))
> +			continue;
> +
>   		addr = of_get_property(child, "reg", &len);
>   		if (!addr || (len < 2 * sizeof(u32))) {
>   			dev_err(dev, "nvmem: invalid reg on %pOF\n", child);
>