ACPI fan: fix memleak on acpi_fan_add_fs() failure
PATCH 1/1 ACPI fan: fix memleak on acpi_fan_add_fs() failure
--- drivers/acpi/fan.c | 8 ++++++-- 1 files changed, 6 insertions(+), 2 deletions(-)
drivers/acpi/fan.c +6 -2
@@ -283,13 +283,17 @@ static int acpi_fan_add(struct acpi_device *device) result =acpi_fan_add_fs (device);if (result) - goto end; + goto unregister; printk(KERN_INFO PREFIX "%s [%s] (%s)\n",acpi_device_name (device),acpi_device_bid (device), !device->power.state ? "on" : "off"); - end: +unregister: +sysfs_remove_link (&device->dev.kobj,"thermal_cooling" ); +sysfs_remove_link (&cdev->device.kobj,"device" ); +thermal_cooling_device_unregister (cdev); +end:return result; }
Quoted 1 line · Alan Jenkins
Signed-off-by Alan Jenkins <[email protected]>
It sounds reasonable although we don't fail in calling the function of acpi_fan_add_fs.
Quoted 9 lines
--- drivers/acpi/fan.c | 8 ++++++-- 1 files changed, 6 insertions(+), 2 deletions(-)drivers/acpi/fan.c +0
-0
drivers/acpi/fan.c static int acpi_fan_add(struct acpi_device *device) · line 290
Quoted 9 lines
drivers/acpi/fan.c +1
-1@@ -283,13 +283,17 @@ static int acpi_fan_add(struct acpi_device *device) result =acpi_fan_add_fs (device);if (result) - goto end; + goto unregister; printk(KERN_INFO PREFIX "%s [%s] (%s)\n",acpi_device_name (device),acpi_device_bid (device), !device->power.state ? "on" : "off");
At the same time we should return directly if the result is zero. But in this patch it seems that it will also remove the sysfs I/F link when the result is zero.
Thanks.
Quoted 15 lines
drivers/acpi/fan.c +5
-2- end: +unregister: +sysfs_remove_link (&device->dev.kobj,"thermal_cooling" ); +sysfs_remove_link (&cdev->device.kobj,"device" ); +thermal_cooling_device_unregister (cdev); +end:return result; } --To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 7 lines · ykzhao
Quoted 2 lines · Alan Jenkins
Signed-off-by Alan Jenkins <[email protected]>It sounds reasonable although we don't fail in calling the function of
acpi_fan_add_fs.
Agreed, this applies to ENOMEM or other weird internal errors only.
Quoted 24 lines
Quoted 9 lines
--- drivers/acpi/fan.c | 8 ++++++-- 1 files changed, 6 insertions(+), 2 deletions(-)drivers/acpi/fan.c +0
-0drivers/acpi/fan.c static int acpi_fan_add(struct acpi_device *device) · line 291
Quoted 10 lines
drivers/acpi/fan.c +1
-1@@ -283,13 +283,17 @@ static int acpi_fan_add(struct acpi_device *device) result =acpi_fan_add_fs (device);if (result) - goto end; + goto unregister; printk(KERN_INFO PREFIX "%s [%s] (%s)\n",acpi_device_name (device),acpi_device_bid (device), !device->power.state ? "on" : "off");At the same time we should return directly if the result is zero. But in this patch it seems that it will also remove the sysfs I/F link when the result is zero.
Quoted 2 lines
Thanks.
Eek, thanks for pointing this out. I'll send a fixed version Alan