Skip to contentSkip to conversation

ti: fix possible memory leak in _ti_omap4_clkctrl_setup()

PATCH In review Hangyu Hua 2 participants 3 messages Oct 31, 2023 – Nov 1, 2023

Subsystems (2)
lore
Share & tools
lore

Hangyu Hua

author
#0 lore raw

PATCH v3 1/1 ti: fix possible memory leak in _ti_omap4_clkctrl_setup()

kstrndup() and kstrdup_and_replace() in clkctrl_get_name() can perform dynamic memory allocation. So clkctrl_name needs to be freed when provider->clkdm_name is NULL.

Fixes bd46cd0b802d ("clk: ti: clkctrl: check return value of kasprintf()")

---

v2 and v3: fix commit info.

 drivers/clk/ti/clkctrl.c | 1 +
 1 file changed, 1 insertion(+)
drivers/clk/ti/clkctrl.c +1 -0
index 607e34d8e289..cb4aa8a45bb2 100644
@@ -591,6 +591,7 @@ static void __init _ti_omap4_clkctrl_setup(struct device_node *node)
 		provider->clkdm_name = kasprintf(GFP_KERNEL,
 						 "%s_clkdm", clkctrl_name);
 		if (!provider->clkdm_name) {
+			kfree(clkctrl_name);
 			kfree(provider);
 			return;
 		}
Signature · 1 line
2.34.1
#1 lore raw
Quoted 9 lines · Hangyu Hua

kstrndup() and kstrdup_and_replace() in clkctrl_get_name() can perform dynamic memory allocation. So clkctrl_name needs to be freed when provider->clkdm_name is NULL.

Fixes bd46cd0b802d ("clk: ti: clkctrl: check return value of kasprintf()")

---

v2 and v3: fix commit info.

I believe I gave you my Rb tag. It's your responsibility to care it on if you send a new version. Otherwise, please tell what makes you think that tag should not be here?

Signature · 2 lines
With Best Regards,
Andy Shevchenko
author
#2 lore raw
Quoted 9 lines · Andy Shevchenko
Quoted 8 lines · Hangyu Hua

kstrndup() and kstrdup_and_replace() in clkctrl_get_name() can perform dynamic memory allocation. So clkctrl_name needs to be freed when provider->clkdm_name is NULL.

Fixes bd46cd0b802d ("clk: ti: clkctrl: check return value of kasprintf()")

---

Quoted 5 lines
Quoted 1 line

v2 and v3: fix commit info.

I believe I gave you my Rb tag. It's your responsibility to care it on if you send a new version. Otherwise, please tell what makes you think that tag should not be here?

Do you mean "Reviewed-by"? I checked the previous emails and "lore.kernel.org" and didn’t find this tag in any related emails. I am not sure what happened. Can you please resend it?

Thanks,
Hangyu

Quoted 1 line