On Fri, 31 Oct 2008, Oliver Neukum wrote:
> this bug is old, the patch should go into 2.6.28 and stable.
> This fixes a memleak in hiddev's open in an error case.
> ---
>
> diff --git a/drivers/hid/usbhid/hiddev.c b/drivers/hid/usbhid/hiddev.c
> index b218cbc..6834d1f 100644
> --- a/drivers/hid/usbhid/hiddev.c
> +++ b/drivers/hid/usbhid/hiddev.c
> @@ -291,8 +291,13 @@ static int hiddev_open(struct inode *inode, struct file *file)
> if (list->hiddev->exist) {
> struct hid_device *hid = hiddev_table[i]->hid;
> res = usbhid_get_power(hid);
> - if (res < 0)
> + if (res < 0) {
> + spin_lock_irqsave(&list->hiddev->list_lock, flags);
> + list_del(&list->node);
> + spin_unlock_irqrestore(&list->hiddev->list_lock, flags);
> + kfree(list->hiddev);
> return -EIO;
> + }
> usbhid_open(hid);
> }
Hmm, this is probably based on not-yet-merged usbhid autosuspend patchset,
right?
Now, you are right that hiddev_open() doesn't handle error condition from
usbhid_open(), and that should be fixed. But that doesn't seem to be
addressed by your patch at all ... ?
Thanks,
--
Jiri Kosina
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
