Skip to contentSkip to conversation

platform/x86: fujitsu-laptop: add NULL check on devm_kzalloc() return value

PATCH In review Gustavo A. R. Silva 3 participants 4 messages Jul 6, 2017 – Jul 10, 2017

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 platform/x86: fujitsu-laptop: add NULL check on devm_kzalloc() return value

Check return value from call to devm_kzalloc()
in order to prevent a NULL pointer dereference.

This issue was detected using Coccinelle and the following semantic patch:

@@
expression x;
identifier fld;
@@

* x = devm_kzalloc(...);
... when != x == NULL
x->fld

---
 drivers/platform/x86/fujitsu-laptop.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
drivers/platform/x86/fujitsu-laptop.c +12 -0
index c1a8528..593a350 100644
@@ -695,6 +695,9 @@ static int acpi_fujitsu_laptop_leds_register(struct acpi_device *device)
 	if (call_fext_func(device,
 			   FUNC_LEDS, 0x0, 0x0, 0x0) & LOGOLAMP_POWERON) {
 		led = devm_kzalloc(&device->dev, sizeof(*led), GFP_KERNEL);
+		if (!led)
+			return -ENOMEM;
+
 		led->name = "fujitsu::logolamp";
 		led->brightness_set_blocking = logolamp_set;
 		led->brightness_get = logolamp_get;
@@ -707,6 +710,9 @@ static int acpi_fujitsu_laptop_leds_register(struct acpi_device *device)
 			    FUNC_LEDS, 0x0, 0x0, 0x0) & KEYBOARD_LAMPS) &&
 	    (call_fext_func(device, FUNC_BUTTONS, 0x0, 0x0, 0x0) == 0x0)) {
 		led = devm_kzalloc(&device->dev, sizeof(*led), GFP_KERNEL);
+		if (!led)
+			return -ENOMEM;
+
 		led->name = "fujitsu::kblamps";
 		led->brightness_set_blocking = kblamps_set;
 		led->brightness_get = kblamps_get;
@@ -723,6 +729,9 @@ static int acpi_fujitsu_laptop_leds_register(struct acpi_device *device)
 	 */
 	if (call_fext_func(device, FUNC_BUTTONS, 0x0, 0x0, 0x0) & BIT(24)) {
 		led = devm_kzalloc(&device->dev, sizeof(*led), GFP_KERNEL);
+		if (!led)
+			return -ENOMEM;
+
 		led->name = "fujitsu::radio_led";
 		led->brightness_set_blocking = radio_led_set;
 		led->brightness_get = radio_led_get;
@@ -741,6 +750,9 @@ static int acpi_fujitsu_laptop_leds_register(struct acpi_device *device)
 	    (call_fext_func(device,
 			    FUNC_LEDS, 0x2, ECO_LED, 0x0) != UNSUPPORTED_CMD)) {
 		led = devm_kzalloc(&device->dev, sizeof(*led), GFP_KERNEL);
+		if (!led)
+			return -ENOMEM;
+
 		led->name = "fujitsu::eco_led";
 		led->brightness_set_blocking = eco_led_set;
 		led->brightness_get = eco_led_get;
Signature · 1 line
2.5.0
maintainer Reviewed-by on v1 1/1
#1 lore raw
Quoted 10 lines · Gustavo A. R. Silva

Check return value from call to devm_kzalloc()
in order to prevent a NULL pointer dereference.

This issue was detected using Coccinelle and the following semantic patch:

@@
expression x;
identifier fld;
@@

Quoted 5 lines

* x = devm_kzalloc(...);
... when != x == NULL
x->fld

These checks should be added in the interest of code correctness. devm_kzalloc() can fail (even if it's extremely unlikely in practice) so we should check for this.

Quoted 50 lines
---
 drivers/platform/x86/fujitsu-laptop.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
drivers/platform/x86/fujitsu-laptop.c +12 -0
index c1a8528..593a350 100644
@@ -695,6 +695,9 @@ static int acpi_fujitsu_laptop_leds_register(struct acpi_device *device)
 	if (call_fext_func(device,
 			   FUNC_LEDS, 0x0, 0x0, 0x0) & LOGOLAMP_POWERON) {
 		led = devm_kzalloc(&device->dev, sizeof(*led), GFP_KERNEL);
+		if (!led)
+			return -ENOMEM;
+
 		led->name = "fujitsu::logolamp";
 		led->brightness_set_blocking = logolamp_set;
 		led->brightness_get = logolamp_get;
@@ -707,6 +710,9 @@ static int acpi_fujitsu_laptop_leds_register(struct acpi_device *device)
 			    FUNC_LEDS, 0x0, 0x0, 0x0) & KEYBOARD_LAMPS) &&
 	    (call_fext_func(device, FUNC_BUTTONS, 0x0, 0x0, 0x0) == 0x0)) {
 		led = devm_kzalloc(&device->dev, sizeof(*led), GFP_KERNEL);
+		if (!led)
+			return -ENOMEM;
+
 		led->name = "fujitsu::kblamps";
 		led->brightness_set_blocking = kblamps_set;
 		led->brightness_get = kblamps_get;
@@ -723,6 +729,9 @@ static int acpi_fujitsu_laptop_leds_register(struct acpi_device *device)
 	 */
 	if (call_fext_func(device, FUNC_BUTTONS, 0x0, 0x0, 0x0) & BIT(24)) {
 		led = devm_kzalloc(&device->dev, sizeof(*led), GFP_KERNEL);
+		if (!led)
+			return -ENOMEM;
+
 		led->name = "fujitsu::radio_led";
 		led->brightness_set_blocking = radio_led_set;
 		led->brightness_get = radio_led_get;
@@ -741,6 +750,9 @@ static int acpi_fujitsu_laptop_leds_register(struct acpi_device *device)
 	    (call_fext_func(device,
 			    FUNC_LEDS, 0x2, ECO_LED, 0x0) != UNSUPPORTED_CMD)) {
 		led = devm_kzalloc(&device->dev, sizeof(*led), GFP_KERNEL);
+		if (!led)
+			return -ENOMEM;
+
 		led->name = "fujitsu::eco_led";
 		led->brightness_set_blocking = eco_led_set;
 		led->brightness_get = eco_led_get;
Signature · 1 line
2.5.0
#2 lore raw
Quoted 15 lines · Jonathan Woithe
Quoted 14 lines · Gustavo A. R. Silva

Check return value from call to devm_kzalloc()
in order to prevent a NULL pointer dereference.

This issue was detected using Coccinelle and the following semantic patch:

@@
expression x;
identifier fld;
@@

* x = devm_kzalloc(...);
... when != x == NULL
x->fld

Quoted 7 lines
Quoted 1 line

These checks should be added in the interest of code correctness. devm_kzalloc() can fail (even if it's extremely unlikely in practice) so we should check for this.

Thanks Gustavo and Jonathan,

Queued to testing.

Signature · 2 lines
Darren Hart
VMware Open Source Technology Center
#3 lore raw

Hi Darren, Jonathan,

Quoting Darren Hart <[email protected]>:

Quoted 22 lines · Jonathan Woithe
Quoted 21 lines · Gustavo A. R. Silva
Quoted 10 lines

Check return value from call to devm_kzalloc()
in order to prevent a NULL pointer dereference.

This issue was detected using Coccinelle and the following semantic patch:

@@
expression x;
identifier fld;
@@

Quoted 5 lines

* x = devm_kzalloc(...);
... when != x == NULL
x->fld

These checks should be added in the interest of code correctness. devm_kzalloc() can fail (even if it's extremely unlikely in practice) so we should check for this.

Quoted 6 lines
Quoted 1 line

Thanks Gustavo and Jonathan,

Queued to testing.

Glad to help. :)

Thanks
--
Gustavo A. R. Silva