On 11/06/2017 04:30 AM, Eric Biggers wrote:
From: Eric Biggers <ebiggers@xxxxxxxxxx>
The "qat-dh" DH implementation assumes that 'key' and 'g' can be copied
into a buffer with size 'p_size'. However it was never checked that
that was actually the case, which most likely allowed users to cause a
buffer underflow via KEYCTL_DH_COMPUTE.
Fix this by updating crypto_dh_decode_key() to verify this precondition
for all DH implementations.
Fixes: c9839143ebbf ("crypto: qat - Add DH support")
Cc: <stable@xxxxxxxxxxxxxxx> # v4.8+
Reviewed-by: Tudor Ambarus <tudor.ambarus@xxxxxxxxxxxxx>
Signed-off-by: Eric Biggers <ebiggers@xxxxxxxxxx>
---
crypto/dh_helper.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/crypto/dh_helper.c b/crypto/dh_helper.c
index 708ae20d2d3c..7f00c771fe8d 100644
--- a/crypto/dh_helper.c
+++ b/crypto/dh_helper.c
@@ -83,6 +83,14 @@ int crypto_dh_decode_key(const char *buf, unsigned int len, struct dh *params)
if (secret.len != crypto_dh_key_len(params))
return -EINVAL;
+ /*
+ * Don't permit the buffer for 'key' or 'g' to be larger than 'p', since
+ * some drivers assume otherwise.
+ */
+ if (params->key_size > params->p_size ||
+ params->g_size > params->p_size)
+ return -EINVAL;
+
/* Don't allocate memory. Set pointers to data within
* the given buffer
*/