[PATCH] clk: mvebu: ap-cpu: fix missing clk_put() in ap_cpu_clock_probe()

[PATCH] clk: mvebu: ap-cpu: fix missing clk_put() in ap_cpu_clock_probe()

[Wentao Liang]

The function ap_cpu_clock_probe() calls of_clk_get() to obtain a
reference to the parent clock for each CPU cluster, but it never
releases it with clk_put().  The returned clk is used only to read
the parent's name via __clk_get_name(), and the reference is leaked
on every successful cluster initialization as well as on the error
path when devm_clk_hw_register() fails.

Add the missing clk_put() after the name has been extracted and
before returning on error to fix the leak.

Fixes: af9617b419f7 ("clk: mvebu: ap-cpu-clk: Fix a memory leak in error handling paths")
Cc: stable@vger.kernel.org
Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
---
 drivers/clk/mvebu/ap-cpu-clk.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/clk/mvebu/ap-cpu-clk.c b/drivers/clk/mvebu/ap-cpu-clk.c
index 1e44ace7d951..a8175908e353 100644
--- a/drivers/clk/mvebu/ap-cpu-clk.c
+++ b/drivers/clk/mvebu/ap-cpu-clk.c
@@ -328,9 +328,11 @@ static int ap_cpu_clock_probe(struct platform_device *pdev)
 		ret = devm_clk_hw_register(dev, &ap_cpu_clk[cluster_index].hw);
 		if (ret) {
 			of_node_put(dn);
+			clk_put(parent);
 			return ret;
 		}
 		ap_cpu_data->hws[cluster_index] = &ap_cpu_clk[cluster_index].hw;
+		clk_put(parent);
 	}
 
 	ap_cpu_data->num = cluster_index + 1;
-- 
2.34.1

Re: [PATCH] clk: mvebu: ap-cpu: fix missing clk_put() in ap_cpu_clock_probe()

[Andrew Lunn]

On Thu, Jun 04, 2026 at 02:51:15AM +0000, Wentao Liang wrote:
> The function ap_cpu_clock_probe() calls of_clk_get() to obtain a
> reference to the parent clock for each CPU cluster, but it never
> releases it with clk_put().  The returned clk is used only to read
> the parent's name via __clk_get_name(), and the reference is leaked
> on every successful cluster initialization as well as on the error
> path when devm_clk_hw_register() fails.

Is there potentially a different interpretation?

If the parent clock was to disappear, the clocks to the CPU would
disappear, resulting in a dead system. By keeping a reference on the
parent, it cannot disappear?

Also, do we care that much about the error case? If we fail to setup
the CPU clock, how much longer is the CPU going to run before it dies
a horrible death?

> Add the missing clk_put() after the name has been extracted and
> before returning on error to fix the leak.
> 
> Fixes: af9617b419f7 ("clk: mvebu: ap-cpu-clk: Fix a memory leak in error handling paths")
> Cc: stable@vger.kernel.org

Does this actually bother people? It needs to in order to be sent to
stable:

https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html

     Andrew

Re: [PATCH] clk: mvebu: ap-cpu: fix missing clk_put() in ap_cpu_clock_probe()

[Brian Masney]

On Thu, Jun 04, 2026 at 02:51:15AM +0000, Wentao Liang wrote:
> The function ap_cpu_clock_probe() calls of_clk_get() to obtain a
> reference to the parent clock for each CPU cluster, but it never
> releases it with clk_put().  The returned clk is used only to read
> the parent's name via __clk_get_name(), and the reference is leaked
> on every successful cluster initialization as well as on the error
> path when devm_clk_hw_register() fails.
> 
> Add the missing clk_put() after the name has been extracted and
> before returning on error to fix the leak.
> 
> Fixes: af9617b419f7 ("clk: mvebu: ap-cpu-clk: Fix a memory leak in error handling paths")
> Cc: stable@vger.kernel.org
> Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>

This calls:

                parent = of_clk_get(np, cluster_index);
                if (IS_ERR(parent)) {
			...
                }
                parent_name =  __clk_get_name(parent);

Can this all be replaced with a call to of_clk_get_parent_name() ?

Brian