[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAFA6WYOZwWuMHXGscgK0Lv9Jbt5WHhfhmK+2ZNR6cr2EAbikCw@mail.gmail.com>
Date: Tue, 28 Feb 2023 16:37:18 +0530
From: Sumit Garg <sumit.garg@...aro.org>
To: Rijo Thomas <Rijo-john.Thomas@....com>
Cc: Jens Wiklander <jens.wiklander@...aro.org>,
lm0963 <lm0963hack@...il.com>,
Mythri PK <Mythri.Pandeshwarakrishna@....com>,
Devaraj Rangasamy <Devaraj.Rangasamy@....com>,
security@...nel.org, Herbert Xu <herbert@...dor.apana.org.au>,
op-tee@...ts.trustedfirmware.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/1] tee: amdtee: fix race condition in amdtee_open_session
On Tue, 28 Feb 2023 at 15:11, Rijo Thomas <Rijo-john.Thomas@....com> wrote:
>
> There is a potential race condition in amdtee_open_session that may
> lead to use-after-free. For instance, in amdtee_open_session() after
> sess->sess_mask is set, and before setting:
>
> sess->session_info[i] = session_info;
>
> if amdtee_close_session() closes this same session, then 'sess' data
> structure will be released, causing kernel panic when 'sess' is
> accessed within amdtee_open_session().
>
> The solution is to set the bit sess->sess_mask as the last step in
> amdtee_open_session().
>
> Fixes: 757cc3e9ff1d ("tee: add AMD-TEE driver")
> Cc: stable@...r.kernel.org
> Signed-off-by: Rijo Thomas <Rijo-john.Thomas@....com>
> ---
> drivers/tee/amdtee/core.c | 29 ++++++++++++++---------------
> 1 file changed, 14 insertions(+), 15 deletions(-)
>
Looks sane to me, FWIW:
Acked-by: Sumit Garg <sumit.garg@...aro.org>
-Sumit
> diff --git a/drivers/tee/amdtee/core.c b/drivers/tee/amdtee/core.c
> index 297dc62bca29..372d64756ed6 100644
> --- a/drivers/tee/amdtee/core.c
> +++ b/drivers/tee/amdtee/core.c
> @@ -267,35 +267,34 @@ int amdtee_open_session(struct tee_context *ctx,
> goto out;
> }
>
> + /* Open session with loaded TA */
> + handle_open_session(arg, &session_info, param);
> + if (arg->ret != TEEC_SUCCESS) {
> + pr_err("open_session failed %d\n", arg->ret);
> + handle_unload_ta(ta_handle);
> + kref_put(&sess->refcount, destroy_session);
> + goto out;
> + }
> +
> /* Find an empty session index for the given TA */
> spin_lock(&sess->lock);
> i = find_first_zero_bit(sess->sess_mask, TEE_NUM_SESSIONS);
> - if (i < TEE_NUM_SESSIONS)
> + if (i < TEE_NUM_SESSIONS) {
> + sess->session_info[i] = session_info;
> + set_session_id(ta_handle, i, &arg->session);
> set_bit(i, sess->sess_mask);
> + }
> spin_unlock(&sess->lock);
>
> if (i >= TEE_NUM_SESSIONS) {
> pr_err("reached maximum session count %d\n", TEE_NUM_SESSIONS);
> + handle_close_session(ta_handle, session_info);
> handle_unload_ta(ta_handle);
> kref_put(&sess->refcount, destroy_session);
> rc = -ENOMEM;
> goto out;
> }
>
> - /* Open session with loaded TA */
> - handle_open_session(arg, &session_info, param);
> - if (arg->ret != TEEC_SUCCESS) {
> - pr_err("open_session failed %d\n", arg->ret);
> - spin_lock(&sess->lock);
> - clear_bit(i, sess->sess_mask);
> - spin_unlock(&sess->lock);
> - handle_unload_ta(ta_handle);
> - kref_put(&sess->refcount, destroy_session);
> - goto out;
> - }
> -
> - sess->session_info[i] = session_info;
> - set_session_id(ta_handle, i, &arg->session);
> out:
> free_pages((u64)ta, get_order(ta_size));
> return rc;
> --
> 2.25.1
>
Powered by blists - more mailing lists