[char-misc-next] mei: hdcp: fix mei_hdcp_verify_mprime() input paramter

[char-misc-next] mei: hdcp: fix mei_hdcp_verify_mprime() input paramter

[Tomas Winkler]

wired_cmd_repeater_auth_stream_req_in has a variable
length array at the end. we use struct_size() overflow
macro to determine the size for the allocation and sending
size.

Fixes: c56967d674e3 (mei: hdcp: Replace one-element array with flexible-array member)
Cc: Ramalingam C <ramalingam.c@intel.com>
Cc: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
---
 drivers/misc/mei/hdcp/mei_hdcp.c | 38 ++++++++++++++++++--------------
 1 file changed, 22 insertions(+), 16 deletions(-)

diff --git a/drivers/misc/mei/hdcp/mei_hdcp.c b/drivers/misc/mei/hdcp/mei_hdcp.c
index d1d3e025ca0e..0e8f12e38494 100644
--- a/drivers/misc/mei/hdcp/mei_hdcp.c
+++ b/drivers/misc/mei/hdcp/mei_hdcp.c
@@ -546,38 +546,44 @@ static int mei_hdcp_verify_mprime(struct device *dev,
 				  struct hdcp_port_data *data,
 				  struct hdcp2_rep_stream_ready *stream_ready)
 {
-	struct wired_cmd_repeater_auth_stream_req_in
-					verify_mprime_in = { { 0 } };
+	struct wired_cmd_repeater_auth_stream_req_in *verify_mprime_in;
 	struct wired_cmd_repeater_auth_stream_req_out
 					verify_mprime_out = { { 0 } };
 	struct mei_cl_device *cldev;
 	ssize_t byte;
+	size_t cmd_size;
 
 	if (!dev || !stream_ready || !data)
 		return -EINVAL;
 
 	cldev = to_mei_cl_device(dev);
 
-	verify_mprime_in.header.api_version = HDCP_API_VERSION;
-	verify_mprime_in.header.command_id = WIRED_REPEATER_AUTH_STREAM_REQ;
-	verify_mprime_in.header.status = ME_HDCP_STATUS_SUCCESS;
-	verify_mprime_in.header.buffer_len =
+	cmd_size = struct_size(verify_mprime_in, streams, data->k);
+	if (cmd_size == SIZE_MAX)
+		return -EINVAL;
+
+	verify_mprime_in = kzalloc(cmd_size, GFP_KERNEL);
+
+	verify_mprime_in->header.api_version = HDCP_API_VERSION;
+	verify_mprime_in->header.command_id = WIRED_REPEATER_AUTH_STREAM_REQ;
+	verify_mprime_in->header.status = ME_HDCP_STATUS_SUCCESS;
+	verify_mprime_in->header.buffer_len =
 			WIRED_CMD_BUF_LEN_REPEATER_AUTH_STREAM_REQ_MIN_IN;
 
-	verify_mprime_in.port.integrated_port_type = data->port_type;
-	verify_mprime_in.port.physical_port = (u8)data->fw_ddi;
-	verify_mprime_in.port.attached_transcoder = (u8)data->fw_tc;
+	verify_mprime_in->port.integrated_port_type = data->port_type;
+	verify_mprime_in->port.physical_port = (u8)data->fw_ddi;
+	verify_mprime_in->port.attached_transcoder = (u8)data->fw_tc;
+
+	memcpy(verify_mprime_in->m_prime, stream_ready->m_prime, HDCP_2_2_MPRIME_LEN);
+	drm_hdcp_cpu_to_be24(verify_mprime_in->seq_num_m, data->seq_num_m);
 
-	memcpy(verify_mprime_in.m_prime, stream_ready->m_prime,
-	       HDCP_2_2_MPRIME_LEN);
-	drm_hdcp_cpu_to_be24(verify_mprime_in.seq_num_m, data->seq_num_m);
-	memcpy(verify_mprime_in.streams, data->streams,
+	memcpy(verify_mprime_in->streams, data->streams,
 	       array_size(data->k, sizeof(*data->streams)));
 
-	verify_mprime_in.k = cpu_to_be16(data->k);
+	verify_mprime_in->k = cpu_to_be16(data->k);
 
-	byte = mei_cldev_send(cldev, (u8 *)&verify_mprime_in,
-			      sizeof(verify_mprime_in));
+	byte = mei_cldev_send(cldev, (u8 *)&verify_mprime_in, cmd_size);
+	kfree(verify_mprime_in);
 	if (byte < 0) {
 		dev_dbg(dev, "mei_cldev_send failed. %zd\n", byte);
 		return byte;
-- 
2.25.4

Re: [char-misc-next] mei: hdcp: fix mei_hdcp_verify_mprime() input paramter

[Gustavo A. R. Silva]

On 7/29/20 12:32, Tomas Winkler wrote:
> wired_cmd_repeater_auth_stream_req_in has a variable
> length array at the end. we use struct_size() overflow
> macro to determine the size for the allocation and sending
> size.
> 
> Fixes: c56967d674e3 (mei: hdcp: Replace one-element array with flexible-array member)

This also fixes:

commit 0a1af1b5c18d ("misc/mei/hdcp: Verify M_prime")

which introduced a potential stack overflow back in Feb 2019, hence it should be
fixed in -stable, too.

So, either both commit c56967d674e3 (mei: hdcp: Replace one-element array with flexible-array member)
and this patch should be ported to -stable in order to fix a potential
stack overflow due to commit 0a1af1b5c18d ("misc/mei/hdcp: Verify M_prime"),
or a separate fix that doesn't touch the one-element array streams[1] in struct
wired_cmd_repeater_auth_stream_req_in, but addresses the stack overflow should
be crafted and applied to -stable.

I can write that fix up for -stable if you want to go in that direction. Just let
me know.

> Cc: Ramalingam C <ramalingam.c@intel.com>
> Cc: Gustavo A. R. Silva <gustavoars@kernel.org>
> Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
> ---
>  drivers/misc/mei/hdcp/mei_hdcp.c | 38 ++++++++++++++++++--------------
>  1 file changed, 22 insertions(+), 16 deletions(-)
> 
> diff --git a/drivers/misc/mei/hdcp/mei_hdcp.c b/drivers/misc/mei/hdcp/mei_hdcp.c
> index d1d3e025ca0e..0e8f12e38494 100644
> --- a/drivers/misc/mei/hdcp/mei_hdcp.c
> +++ b/drivers/misc/mei/hdcp/mei_hdcp.c
> @@ -546,38 +546,44 @@ static int mei_hdcp_verify_mprime(struct device *dev,
>  				  struct hdcp_port_data *data,
>  				  struct hdcp2_rep_stream_ready *stream_ready)
>  {
> -	struct wired_cmd_repeater_auth_stream_req_in
> -					verify_mprime_in = { { 0 } };
> +	struct wired_cmd_repeater_auth_stream_req_in *verify_mprime_in;
>  	struct wired_cmd_repeater_auth_stream_req_out
>  					verify_mprime_out = { { 0 } };
>  	struct mei_cl_device *cldev;
>  	ssize_t byte;
> +	size_t cmd_size;
>  
>  	if (!dev || !stream_ready || !data)
>  		return -EINVAL;
>  
>  	cldev = to_mei_cl_device(dev);
>  
> -	verify_mprime_in.header.api_version = HDCP_API_VERSION;
> -	verify_mprime_in.header.command_id = WIRED_REPEATER_AUTH_STREAM_REQ;
> -	verify_mprime_in.header.status = ME_HDCP_STATUS_SUCCESS;
> -	verify_mprime_in.header.buffer_len =
> +	cmd_size = struct_size(verify_mprime_in, streams, data->k);
> +	if (cmd_size == SIZE_MAX)
> +		return -EINVAL;
> +
> +	verify_mprime_in = kzalloc(cmd_size, GFP_KERNEL);
> +
> +	verify_mprime_in->header.api_version = HDCP_API_VERSION;
> +	verify_mprime_in->header.command_id = WIRED_REPEATER_AUTH_STREAM_REQ;
> +	verify_mprime_in->header.status = ME_HDCP_STATUS_SUCCESS;
> +	verify_mprime_in->header.buffer_len =
>  			WIRED_CMD_BUF_LEN_REPEATER_AUTH_STREAM_REQ_MIN_IN;
>  
> -	verify_mprime_in.port.integrated_port_type = data->port_type;
> -	verify_mprime_in.port.physical_port = (u8)data->fw_ddi;
> -	verify_mprime_in.port.attached_transcoder = (u8)data->fw_tc;
> +	verify_mprime_in->port.integrated_port_type = data->port_type;
> +	verify_mprime_in->port.physical_port = (u8)data->fw_ddi;
> +	verify_mprime_in->port.attached_transcoder = (u8)data->fw_tc;
> +
> +	memcpy(verify_mprime_in->m_prime, stream_ready->m_prime, HDCP_2_2_MPRIME_LEN);
> +	drm_hdcp_cpu_to_be24(verify_mprime_in->seq_num_m, data->seq_num_m);
>  
> -	memcpy(verify_mprime_in.m_prime, stream_ready->m_prime,
> -	       HDCP_2_2_MPRIME_LEN);
> -	drm_hdcp_cpu_to_be24(verify_mprime_in.seq_num_m, data->seq_num_m);
> -	memcpy(verify_mprime_in.streams, data->streams,
> +	memcpy(verify_mprime_in->streams, data->streams,
>  	       array_size(data->k, sizeof(*data->streams)));

Please, use flex_array_size() instead of array_size() here, like this:

memcpy(verify_mprime_in->streams, data->streams,
       flex_array_size(verify_mprime_in, streams, data->k));


Thanks
--
Gustavo

>  
> -	verify_mprime_in.k = cpu_to_be16(data->k);
> +	verify_mprime_in->k = cpu_to_be16(data->k);
>  
> -	byte = mei_cldev_send(cldev, (u8 *)&verify_mprime_in,
> -			      sizeof(verify_mprime_in));
> +	byte = mei_cldev_send(cldev, (u8 *)&verify_mprime_in, cmd_size);
> +	kfree(verify_mprime_in);
>  	if (byte < 0) {
>  		dev_dbg(dev, "mei_cldev_send failed. %zd\n", byte);
>  		return byte;
>