out-of-bounds access in sound/soc/sof/topology.c

[Pierre-Louis Bossart]

Thanks Sergey for this email.

On 4/15/22 04:23, Sergey Senozhatsky wrote:
> Hi,
> 
> I'm running 5.10.111 LTS, so if this has been fixed already then we definitely
> want to cherry pick the fix for -stable.
> 
> 
> Anonymous union in this struct is of zero size
> 
> /* generic control data */
> struct sof_ipc_ctrl_data {
>         struct sof_ipc_reply rhdr;
>         uint32_t comp_id;
> 
>         /* control access and data type */
>         uint32_t type;          /**< enum sof_ipc_ctrl_type */
>         uint32_t cmd;           /**< enum sof_ipc_ctrl_cmd */
>         uint32_t index;         /**< control index for comps > 1 control */
> 
>         /* control data - can either be appended or DMAed from host */
>         struct sof_ipc_host_buffer buffer;
>         uint32_t num_elems;     /**< in array elems or bytes for data type */
>         uint32_t elems_remaining;       /**< elems remaining if sent in parts */
> 
>         uint32_t msg_index;     /**< for large messages sent in parts */
> 
>         /* reserved for future use */
>         uint32_t reserved[6];
> 
>         /* control data - add new types if needed */
>         union {
>                 /* channel values can be used by volume type controls */
>                 struct sof_ipc_ctrl_value_chan chanv[0];
>                 /* component values used by routing controls like mux, mixer */
>                 struct sof_ipc_ctrl_value_comp compv[0];
>                 /* data can be used by binary controls */
>                 struct sof_abi_hdr data[0];
>         };
> } __packed;
> 
> sof_ipc_ctrl_value_chan and sof_ipc_ctrl_value_comp are of the same
> size - 8 bytes, while sof_abi_hdr is much larger - _at least_ 32 bytes
> (`__u32 data[0]` in sof_abi_hdr suggest that there should be more
> payload after header). But they all contribute 0 to sizeof(sof_ipc_ctrl_data).
> 
> Now control data allocations looks as follows
> 
> 	scontrol->size = struct_size(scontrol->control_data, chanv,
> 				     le32_to_cpu(mc->num_channels));
> 	scontrol->control_data = kzalloc(scontrol->size, GFP_KERNEL);
> 
> Which is sizeof(sof_ipc_ctrl_data) + mc->num_channels * sizeof(sof_ipc_ctrl_value_chan)
> 
> For some reason it uses sizeof(sof_ipc_ctrl_value_chan), which is not
> the largest member of the union.
> 
> And this is where the problem is: in order to make control->data.FOO loads
> and stores legal we need mc->num_channels to be of at least 4. So that
> 
>    sizeof(sof_ipc_ctrl_data) + mc->num_channels * sizeof(sof_ipc_ctrl_value_chan)
> 
>                 92           +        4         *            8
> 
> will be the same as
> 
>    sizeof(sof_ipc_ctrl_data) + sizeof(sof_abi_hdr).
> 
>                 92           +           32
> 
> Otherwise scontrol->control_data->data.FOO will access nearby/foreign
> slab object.
> 
> And there is at least one such memory access. In sof_get_control_data().
> 
> 	wdata[i].pdata = wdata[i].control->control_data->data;
> 	*size += wdata[i].pdata->size;
> 
> pdata->size is at offset 8, but if, say, mc->num_channels == 1 then
> we allocate only 8 bytes for pdata, so pdata->size is 4 bytes outside
> of allocated slab object.
> 
> Thoughts?

The SOF contributors who wrote that code are all on an extended Easter week-end or vacation so we'll need more time to provide a definitive answer.

I am far from an expert on the topology, but I note that the 'data' field is only used for binary controls, where we use the maximum possible size for a control, without any arithmetic involving channels. It makes sense to me, the binary data does not have any defined structure, it's just a bunch of bytes provided as is to the firmware.

static int sof_ipc3_control_load_bytes(struct snd_sof_dev *sdev, struct snd_sof_control *scontrol)
{
	struct sof_ipc_ctrl_data *cdata;
	int ret;

	scontrol->ipc_control_data = kzalloc(scontrol->max_size, GFP_KERNEL);
	if (!scontrol->ipc_control_data)
		return -ENOMEM;


In other cases, such as volumes and enums, the 'data' field doesn't seem to be used but we do use the channel information for volume and enums.

static int sof_ipc3_control_load_volume(struct snd_sof_dev *sdev, struct snd_sof_control *scontrol)
{
	struct sof_ipc_ctrl_data *cdata;
	int i;

	/* init the volume get/put data */
	scontrol->size = struct_size(cdata, chanv, scontrol->num_channels);

	scontrol->ipc_control_data = kzalloc(scontrol->size, GFP_KERNEL);


static int sof_ipc3_control_load_enum(struct snd_sof_dev *sdev, struct snd_sof_control *scontrol)
{
	struct sof_ipc_ctrl_data *cdata;

	/* init the enum get/put data */
	scontrol->size = struct_size(cdata, chanv, scontrol->num_channels);

	scontrol->ipc_control_data = kzalloc(scontrol->size, GFP_KERNEL);
	if (!scontrol->ipc_control_data)


Given that we have two ways of allocating the memory, I am not sure there is a problem, but I could be wrong.

I checked the v5.10.111 code and I see the same code, with the max_size being used for sof_control_load_bytes() and no channel-based arithmetic.

Can I ask how you found out about this problem, is this the result of a warning/error reported by a software tool or based on your reviews of the code?

Thanks
-Pierre