On Tue, 9 Sep 2025 14:32:56 +1000
Wilfred Mallawa <wilfred.opensou...@gmail.com> wrote:
> From: Wilfred Mallawa <wilfred.mall...@wdc.com>
>
> This is to support uni-directional transports such as SPDM over Storage.
> As specified by the DMTF DSP0286.
>
> Also update spdm_socket_rsp() to use the new send()/receive() functions. For
> the case of spdm_socket_receive(), this allows us to do error checking
> in one place with the addition of spdm_socket_command_valid().
>
> Signed-off-by: Wilfred Mallawa <wilfred.mall...@wdc.com>
One trivial comment below. Either way
Reviewed-by: Jonathan Cameron <jonathan.came...@huawei.com>
> +uint32_t spdm_socket_rsp(const int socket, uint32_t transport_type,
> + void *req, uint32_t req_len,
> + void *rsp, uint32_t rsp_len)
> +{
> + bool result;
> +
> + result = spdm_socket_send(socket, SPDM_SOCKET_COMMAND_NORMAL,
> + transport_type, req, req_len);
> if (!result) {
> return 0;
> }
>
> - assert(command != 0);
> -
> + rsp_len = spdm_socket_receive(socket, transport_type, rsp, rsp_len);
> return rsp_len;
return spdm_socket_receive(socket, transport_type, rsp, rsp_len);
Both shorter and avoids rsp_len changing meaning (slightly).
> }
>
