Re: [PATCH] KEYS: trusted: fix a couple error pointer dereferences

On Fri, Apr 16, 2021 at 06:15:58PM +0530, Sumit Garg wrote:

Hi Serge,

On Wed, 14 Apr 2021 at 19:37, Serge E. Hallyn [off-list ref] wrote:

quoted

On Wed, Apr 14, 2021 at 09:08:58AM +0300, Dan Carpenter wrote:

quoted

If registering "reg_shm_out" fails, then it is an error pointer and the
error handling will call tee_shm_free(reg_shm_out) which leads to an
error pointer dereference and an Oops.

I've re-arranged it so we only free things that have been allocated
successfully.

Fixes: 6dd95e650c8a ("KEYS: trusted: Introduce TEE based Trusted Keys")
Signed-off-by: Dan Carpenter <redacted>
---
 security/keys/trusted-keys/trusted_tee.c | 24 ++++++++++--------------
 1 file changed, 10 insertions(+), 14 deletions(-)

diff --git a/security/keys/trusted-keys/trusted_tee.c b/security/keys/trusted-keys/trusted_tee.c
index 2ce66c199e1d..45f96f6ed673 100644
--- a/security/keys/trusted-keys/trusted_tee.c
+++ b/security/keys/trusted-keys/trusted_tee.c

@@ -65,7 +65,7 @@ static int trusted_tee_seal(struct trusted_key_payload *p, char *datablob)
      int ret;
      struct tee_ioctl_invoke_arg inv_arg;
      struct tee_param param[4];
-     struct tee_shm *reg_shm_in = NULL, *reg_shm_out = NULL;
+     struct tee_shm *reg_shm_in, *reg_shm_out;

I don't have this file (trusted_tee.c) in my tree and there's no lore
link here to previous what threads this depends on.  Based on the
context I can't verify that reg_shm_in will always be initialized
before you get to the free_shm_in label.

You can find trusted_tee.c here [1].

[1] https://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git/tree/security/keys/trusted-keys/trusted_tee.c

Thanks.  Looks good then :)