On Tue, Nov 28, 2017 at 10:05:17AM +0100, Antoine Tenart wrote:
> The patch fixes the ahash support by only updating the result buffer
> when provided. Otherwise the driver could crash with NULL pointer
> exceptions, because the ahash caller isn't required to supply a result
> buffer on all calls.
>
> Fixes: 1b44c5a60c13 ("crypto: inside-secure - add SafeXcel EIP197 crypto
> engine driver")
> Signed-off-by: Antoine Tenart <[email protected]>
> ---
> drivers/crypto/inside-secure/safexcel_hash.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/crypto/inside-secure/safexcel_hash.c
> b/drivers/crypto/inside-secure/safexcel_hash.c
> index 6135c9f5742c..424f4c5d4d25 100644
> --- a/drivers/crypto/inside-secure/safexcel_hash.c
> +++ b/drivers/crypto/inside-secure/safexcel_hash.c
> @@ -150,7 +150,12 @@ static int safexcel_handle_req_result(struct
> safexcel_crypto_priv *priv, int rin
>
> if (sreq->finish)
> result_sz = crypto_ahash_digestsize(ahash);
> - memcpy(sreq->state, areq->result, result_sz);
> +
> + /* The called isn't required to supply a result buffer. Updated it only
> + * when provided.
> + */
> + if (areq->result)
> + memcpy(sreq->state, areq->result, result_sz);
I don't think you should use whether areq->result is NULL to
determine whether to copy data into it. For example, I could
be making an update call with a non-NULL areq->result and it would
be completely wrong if you overwrote that with the above memcpy.
IOW areq->result should not be touched at all unless you are doing
a finalisation.
Thanks,
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
