On Wed, 2025-10-01 at 10:46 +0200, Klaus Jensen wrote:
> On Sep 12 12:11, Wilfred Mallawa wrote:
> > From: Wilfred Mallawa <[email protected]>
> > 
> > This header contains the transport encoding for an SPDM message
> > that
> > uses the SPDM over Storage transport as defined by the DMTF
> > DSP0286.
> > 
> > Signed-off-by: Wilfred Mallawa <[email protected]>
> > ---
> >  include/system/spdm-socket.h | 12 ++++++++++++
> >  1 file changed, 12 insertions(+)
> > 
> > diff --git a/include/system/spdm-socket.h b/include/system/spdm-
> > socket.h
> > index 29aa04fd52..80cd7021c1 100644
> > --- a/include/system/spdm-socket.h
> > +++ b/include/system/spdm-socket.h
> > @@ -88,6 +88,18 @@ bool spdm_socket_send(const int socket, uint32_t
> > socket_cmd,
> >   */
> >  void spdm_socket_close(const int socket, uint32_t transport_type);
> >  
> > +/*
> > + * Defines the transport encoding for SPDM, this information shall
> > be passed
> > + * down to the SPDM server, when conforming to the SPDM over
> > Storage standard
> > + * as defined by DSP0286.
> > + */
> > +typedef struct {
> > +    uint8_t security_protocol;              /* Must be 0xE8 for
> > SPDM Commands */
> > +    uint16_t security_protocol_specific;    /* Bit[7:2] SPDM
> > Operation
> > +                                               Bit[0:1] Connection
> > ID */
> > +    uint32_t length;                        /* Length of the SPDM
> > Message*/
> > +} QEMU_PACKED StorageSpdmTransportHeader;
> > +
> >  #define SPDM_SOCKET_COMMAND_NORMAL                0x0001
> >  #define SPDM_SOCKET_STORAGE_CMD_IF_SEND           0x0002
> >  #define SPDM_SOCKET_STORAGE_CMD_IF_RECV           0x0003
> > -- 
> > 2.51.0
> > 
> 
> The above struct is specific to the communication protocol with the
> specific SPDM server, correct? It's not defined by DSP0286 as far as
> I
> can tell?
Yeah that's correct. For example, spdm-utils will use this [1]. Which
is the only server to support SPDM Storage as far as I am aware.

[1] https://github.com/westerndigitalcorporation/spdm-utils/pull/139

Reply via email to