On Fri, Feb 5, 2021 at 4:27 PM Saravana Kannan <saravanak@xxxxxxxxxx> wrote:
>
> This allows fw_devlink to recognize clock provider drivers that don't
> use the device-driver model to initialize the device. fw_devlink will
> use this information to make sure consumers of such clock providers
> aren't indefinitely blocked from probing, waiting for the power domain
> device to appear and bind to a driver.
Don't we have cases that are a mixture? IOW, a subset of the clock
provider is initialized early, then the full driver takes over. You'd
want consumers that are not a driver to succeed, but drivers to defer
until the full driver is up.
>
> Signed-off-by: Saravana Kannan <saravanak@xxxxxxxxxx>
> ---
> drivers/clk/clk.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c
> index 8c1d04db990d..27ff90eacb1f 100644
> --- a/drivers/clk/clk.c
> +++ b/drivers/clk/clk.c
> @@ -4555,6 +4555,8 @@ int of_clk_add_provider(struct device_node *np,
> if (ret < 0)
> of_clk_del_provider(np);
>
> + fwnode_dev_initialized(&np->fwnode, true);
> +
> return ret;
> }
> EXPORT_SYMBOL_GPL(of_clk_add_provider);
> @@ -4672,6 +4674,7 @@ void of_clk_del_provider(struct device_node *np)
> list_for_each_entry(cp, &of_clk_providers, link) {
> if (cp->node == np) {
> list_del(&cp->link);
> + fwnode_dev_initialized(&np->fwnode, false);
> of_node_put(cp->node);
> kfree(cp);
> break;
> --
> 2.30.0.478.g8a0d178c01-goog
>
