On Wednesday 22 July 2015 12:16 PM, Krzysztof Kozlowski wrote:

On 22.07.2015 15:27, Vaibhav Hiremath wrote:

quoted

On Wednesday 22 July 2015 02:22 AM, Stephen Boyd wrote:

quoted

On 07/21/2015 12:36 PM, Vaibhav Hiremath wrote:

quoted

On Wednesday 22 July 2015 12:40 AM, Stephen Boyd wrote:

quoted

On 07/21/2015 04:07 AM, Vaibhav Hiremath wrote:

quoted

+

<snip>

quoted

quoted

quoted

quoted

+static int pm800_clk_remove(struct platform_device *pdev)
+{
+    struct pm800_clk *pm800_clks = platform_get_drvdata(pdev);
+    int i;
+
+    of_clk_del_provider(pm800_clks[0].clk_np);
+    /* Drop the reference obtained in pm800_clk_parse_dt */
+    of_node_put(pm800_clks[0].clk_np);

This is odd. Why are we keeping the reference in the driver?

Honestly I do not have any good answer here. I have to admit that it is
getting carry forwarded from legacy driver.

Well we shouldn't do things if we don't know why we're doing them.
Krzysztof?

I am really busy now so I am not following closely other discussions. I
assume you are referring to clk-s2mps11.c. The of_node_put() matches
of_get_child_by_name() when parsing DT.

So why not of_node_put() just after parsing DT? Well, the result of
of_get_child_by_name() is stored in state container for entire device
life-cycle so we can use it in of_clk_del_provider().

That was the idea behind it. If it looks incorrect I would be happy to
see a patch :) .

About to respond, I digged more on kobject stuff and sequence in
of/dynamic.c and

I think you are right, we need of_node_put, as a result of
of_get_child_by_name().

Stephen,
Please let me know if you think otherwise.

Thanks,
Vaibhav