In twl_chrdev_ioctl(), the ioctl driver command is firstly copied from the
userspace pointer 'argp' and saved to the kernel object 'driver_command'.
Then a security check is performed on the data buffer size indicated by
'driver_command', which is 'driver_command.buffer_length'. 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 size between the two copies. This way, the user can
bypass the security check and inject invalid data buffer size. This can
cause potential security issues in the following execution.
This patch checks the buffer size 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 <[email protected]>
---
drivers/scsi/3w-sas.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/drivers/scsi/3w-sas.c b/drivers/scsi/3w-sas.c
index cf9f2a0..ea41969 100644
--- a/drivers/scsi/3w-sas.c
+++ b/drivers/scsi/3w-sas.c
@@ -757,6 +757,11 @@ static long twl_chrdev_ioctl(struct file *file, unsigned int cmd, unsigned long
/* Now copy down the entire ioctl */
if (copy_from_user(tw_ioctl, argp, driver_command.buffer_length + sizeof(TW_Ioctl_Buf_Apache) - 1))
goto out3;
+ if (tw_ioctl->driver_command.buffer_length !=
+ driver_command.buffer_length) {
+ retval = -EINVAL;
+ goto out3;
+ }
/* See which ioctl we are doing */
switch (cmd) {
--
2.7.4
On Sat, May 5, 2018 at 10:50 PM, Wenwen Wang <[email protected]> wrote:
> In twl_chrdev_ioctl(), the ioctl driver command is firstly copied from the
> userspace pointer 'argp' and saved to the kernel object 'driver_command'.
> Then a security check is performed on the data buffer size indicated by
> 'driver_command', which is 'driver_command.buffer_length'. 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 size between the two copies. This way, the user can
> bypass the security check and inject invalid data buffer size. This can
> cause potential security issues in the following execution.
>
> This patch checks the buffer size 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 <[email protected]>
> ---
> drivers/scsi/3w-sas.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/drivers/scsi/3w-sas.c b/drivers/scsi/3w-sas.c
> index cf9f2a0..ea41969 100644
> --- a/drivers/scsi/3w-sas.c
> +++ b/drivers/scsi/3w-sas.c
> @@ -757,6 +757,11 @@ static long twl_chrdev_ioctl(struct file *file, unsigned int cmd, unsigned long
> /* Now copy down the entire ioctl */
> if (copy_from_user(tw_ioctl, argp, driver_command.buffer_length + sizeof(TW_Ioctl_Buf_Apache) - 1))
> goto out3;
> + if (tw_ioctl->driver_command.buffer_length !=
> + driver_command.buffer_length) {
> + retval = -EINVAL;
> + goto out3;
> + }
>
> /* See which ioctl we are doing */
> switch (cmd) {
> --
> 2.7.4
>
1. Returning -EINVAL after the copy_from_user() doesn't prevent any
invalid copy down to kernel mode from happening.
2. twl_chrdev_open() checks for capable(CAP_SYS_ADMIN):
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/scsi/3w-sas.c#n834
I don't see the point in this patch.
-Adam