On Sat, May 5, 2018 at 10:48 PM, Wenwen Wang <wang6495@xxxxxxx> wrote:
> In tw_chrdev_ioctl(), the length of the data buffer is firstly copied from
> the userspace pointer 'argp' and saved to the kernel object
> 'data_buffer_length'. Then a security check is performed on it to make sure
> that the length is not more than 'TW_MAX_IOCTL_SECTORS * 512'. Otherwise,
> an error code -EINVAL is returned. If the security check is passed, the
> entire ioctl command is copied again from the 'argp' pointer and saved to
> the kernel object 'tw_ioctl'. Then, various operations are performed on
> 'tw_ioctl' according to the 'cmd'. Given that the 'argp' pointer resides in
> userspace, a malicious userspace process can race to change the buffer
> length between the two copies. This way, the user can bypass the security
> check and inject invalid data buffer length. This can cause potential
> security issues in the following execution.
>
> This patch checks the buffer length obtained in the second copy. An error
> code -EINVAL will be returned if it is not same as the original one in the
> first copy.
>
> Signed-off-by: Wenwen Wang <wang6495@xxxxxxx>
> ---
> drivers/scsi/3w-xxxx.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/drivers/scsi/3w-xxxx.c b/drivers/scsi/3w-xxxx.c
> index 33261b6..ef79194 100644
> --- a/drivers/scsi/3w-xxxx.c
> +++ b/drivers/scsi/3w-xxxx.c
> @@ -919,6 +919,10 @@ static long tw_chrdev_ioctl(struct file *file, unsigned int cmd, unsigned long a
> /* Now copy down the entire ioctl */
> if (copy_from_user(tw_ioctl, argp, data_buffer_length + sizeof(TW_New_Ioctl) - 1))
> goto out2;
> + if (tw_ioctl->data_buffer_length != data_buffer_length) {
> + retval = -EINVAL;
> + goto out2;
> + }
>
> passthru = (TW_Passthru *)&tw_ioctl->firmware_command;
>
> --
> 2.7.4
>
I would drop this patch and check for !capable(CAP_SYS_ADMIN) in
tw_chrdev_open() instead.
-Adam
