[Intel-gfx] [PATCH v9 02/39] drm: header for i915 - MEI_HDCP interface
Daniel Vetter
daniel.vetter at ffwll.ch
Mon Dec 17 13:59:41 UTC 2018
On Mon, Dec 17, 2018 at 2:42 PM Winkler, Tomas <tomas.winkler at intel.com> wrote:
>
> >
> > On Mon, Dec 17, 2018 at 12:28 PM Winkler, Tomas
> > <tomas.winkler at intel.com> wrote:
> > >
> > > >
> > > > Header defines the interface for the I915 and MEI_HDCP drivers.
> > > >
> > > > Signed-off-by: Ramalingam C <ramalingam.c at intel.com>
> > > > ---
> > > > include/drm/i915_mei_hdcp_interface.h | 132
> > > > ++++++++++++++++++++++++++++++++++
> > > > 1 file changed, 132 insertions(+)
> > > > create mode 100644 include/drm/i915_mei_hdcp_interface.h
> > > >
> > > > diff --git a/include/drm/i915_mei_hdcp_interface.h
> > > > b/include/drm/i915_mei_hdcp_interface.h
> > > > new file mode 100644
> > > > index 000000000000..e3b7fb32612a
> > > > --- /dev/null
> > > > +++ b/include/drm/i915_mei_hdcp_interface.h
> > >
> > >
> > > This should be just 'i915_hdcp_interface.h'
> >
> > We have other hdcp engines than the mei one. Discrete gpus will definitely get
> > a differnt hdcp2 engine even, can't rely on ME for those. Keeping mei in the
> > names to make that clear makes sense to me, removing it would just cause
> > more confusion I think.
>
> We would like to create a DHCP interface and hold the specific implementation w/o hdcp_mei
mei_hdcp is only part of the hw interface, the other part is setting
the right bits at the right time on the i915 side (afaiui ME has a
priviledge hw channel to hand the key over to the display hw). Only
the overall thing implements hdcp.
> > We already had the entire "should we have a common hdcp interface in drm"
> > discussion when we orginally landed this, and looking at all the intel and non-
> > intel implementations floating around that didn't make sense, due to lack of
> > comonality of how things flow and interact with the hdmi/dp ports.
> > -Daniel
>
> Sorry, was not part of that discussion, though it looks strange to me as the DHCP is a standard no Intel invention.
The stuff going over the wire is a standard, and we have a header for
those bits in drm_hdcp.h. The programming interface exposed by the
hardware isn't standardized. We do share some code in the i915 driver,
because a bunch of things work the same way across hdcd1/mei_hdcp and
dp/hdmi, but that's about it.
-Daniel
>
> >
> > >
> > > > @@ -0,0 +1,132 @@
> > > > +/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
> > > > +/*
> > > > + * Copyright © 2017-2018 Intel Corporation
> > > > + *
> > > > + * Authors:
> > > > + * Ramalingam C <ramalingam.c at intel.com> */
> > > > +
> > > > +#ifndef _I915_MEI_HDCP_INTERFACE_H_ #define
> > > > +_I915_MEI_HDCP_INTERFACE_H_
> > > Drop MEI
> > > > +
> > > > +#include <linux/mutex.h>
> > > > +#include <drm/drm_hdcp.h>
> > > > +
> > > > +/**
> > > > + * enum hdcp_port_type - HDCP port implementation type defined by
> > > > +ME FW
> > > > + * @HDCP_PORT_TYPE_INVALID: Invalid hdcp port type
> > > > + * @HDCP_PORT_TYPE_INTEGRATED: In-Host HDCP2.x port
> > > > + * @HDCP_PORT_TYPE_LSPCON: HDCP2.2 discrete wired Tx port with
> > LSPCON
> > > > + * (HDMI 2.0) solution
> > > > + * @HDCP_PORT_TYPE_CPDP: HDCP2.2 discrete wired Tx port using the
> > > > +CPDP
> > > > (DP 1.3)
> > > > + * solution
> > > > + */
> > > > +enum hdcp_port_type {
> > > > + HDCP_PORT_TYPE_INVALID,
> > > > + HDCP_PORT_TYPE_INTEGRATED,
> > > > + HDCP_PORT_TYPE_LSPCON,
> > > > + HDCP_PORT_TYPE_CPDP
> > > > +};
> > > > +
> > > > +/**
> > > > + * enum hdcp_wired_protocol - HDCP adaptation used on the port
> > > > + * @HDCP_PROTOCOL_INVALID: Invalid HDCP adaptation protocol
> > > > + * @HDCP_PROTOCOL_HDMI: HDMI adaptation of HDCP used on the port
> > > > + * @HDCP_PROTOCOL_DP: DP adaptation of HDCP used on the port */
> > > > +enum hdcp_wired_protocol {
> > > > + HDCP_PROTOCOL_INVALID,
> > > > + HDCP_PROTOCOL_HDMI,
> > > > + HDCP_PROTOCOL_DP
> > > > +};
> > > > +
> > > > +/**
> > > > + * struct hdcp_port_data - intel specific HDCP port data
> > > > + * @port: port index as per I915
> > > > + * @port_type: HDCP port type as per ME FW classification
> > > > + * @protocol: HDCP adaptation as per ME FW
> > > > + * @k: No of streams transmitted on a port. Only on DP MST this is
> > > > +!= 1
> > > > + * @seq_num_m: Count of RepeaterAuth_Stream_Manage msg
> > propagated.
> > > > + * Initialized to 0 on AKE_INIT. Incremented after every successful
> > > > + * transmission of RepeaterAuth_Stream_Manage message. When it
> > > > rolls
> > > > + * over re-Auth has to be triggered.
> > > > + * @streams: struct hdcp2_streamid_type[k]. Defines the type and id for
> > the
> > > > + * streams
> > > > + */
> > > > +struct hdcp_port_data {
> > > > + short int port;
> > > > + u8 port_type;
> > > > + u8 protocol;
> > > > + u16 k;
> > > > + u32 seq_num_m;
> > > > + struct hdcp2_streamid_type *streams; };
> > > > +
> > > > +/**
> > > > + * struct i915_hdcp_component_ops- ops for HDCP2.2 services.
> > > > + * @owner: Module providing the ops
> > > > + * @initiate_hdcp2_session: Initiate a Wired HDCP2.2 Tx Session.
> > > > + * And Prepare AKE_Init.
> > > > + * @verify_receiver_cert_prepare_km: Verify the Receiver Certificate
> > > > + * AKE_Send_Cert and prepare
> > > > + AKE_Stored_Km/AKE_No_Stored_Km
> > > > + * @verify_hprime: Verify AKE_Send_H_prime
> > > > + * @store_pairing_info: Store pairing info received
> > > > + * @initiate_locality_check: Prepare LC_Init
> > > > + * @verify_lprime: Verify lprime
> > > > + * @get_session_key: Prepare SKE_Send_Eks
> > > > + * @repeater_check_flow_prepare_ack: Validate the Downstream
> > topology
> > > > + * and prepare rep_ack
> > > > + * @verify_mprime: Verify mprime
> > > > + * @enable_hdcp_authentication: Mark a port as authenticated.
> > > > + * @close_hdcp_session: Close the Wired HDCP Tx session per port.
> > > > + * This also disables the authenticated state of the port.
> > > > + */
> > > > +struct i915_hdcp_component_ops {
> > > > + /**
> > > > + * @owner: mei_hdcp module
> > > > + */
> > > > + struct module *owner;
> > > > +
> > > > + int (*initiate_hdcp2_session)(struct device *dev,
> > > > + struct hdcp_port_data *data,
> > > > + struct hdcp2_ake_init *ake_data);
> > > > + int (*verify_receiver_cert_prepare_km)(struct device *dev,
> > > > + struct hdcp_port_data *data,
> > > > + struct hdcp2_ake_send_cert
> > > > + *rx_cert,
> > > > + bool *km_stored,
> > > > + struct hdcp2_ake_no_stored_km
> > > > + *ek_pub_km,
> > > > + size_t *msg_sz);
> > > > + int (*verify_hprime)(struct device *dev,
> > > > + struct hdcp_port_data *data,
> > > > + struct hdcp2_ake_send_hprime *rx_hprime);
> > > > + int (*store_pairing_info)(struct device *dev,
> > > > + struct hdcp_port_data *data,
> > > > + struct hdcp2_ake_send_pairing_info
> > > > +
> > > > *pairing_info);
> > > > + int (*initiate_locality_check)(struct device *dev,
> > > > + struct hdcp_port_data *data,
> > > > + struct hdcp2_lc_init *lc_init_data);
> > > > + int (*verify_lprime)(struct device *dev,
> > > > + struct hdcp_port_data *data,
> > > > + struct hdcp2_lc_send_lprime *rx_lprime);
> > > > + int (*get_session_key)(struct device *dev,
> > > > + struct hdcp_port_data *data,
> > > > + struct hdcp2_ske_send_eks *ske_data);
> > > > + int (*repeater_check_flow_prepare_ack)(struct device *dev,
> > > > + struct hdcp_port_data *data,
> > > > + struct
> > > > hdcp2_rep_send_receiverid_list
> > > > +
> > > > *rep_topology,
> > > > + struct
> > > > + hdcp2_rep_send_ack
> > > > +
> > > > *rep_send_ack);
> > > > + int (*verify_mprime)(struct device *dev,
> > > > + struct hdcp_port_data *data,
> > > > + struct hdcp2_rep_stream_ready *stream_ready);
> > > > + int (*enable_hdcp_authentication)(struct device *dev,
> > > > + struct hdcp_port_data *data);
> > > > + int (*close_hdcp_session)(struct device *dev,
> > > > + struct hdcp_port_data *data); };
> > > > +#endif /* _I915_MEI_HDCP_NTERFACE_H_ */
> > > DROP MEI
> > > > 2.7.4
> > >
> >
> >
> > --
> > Daniel Vetter
> > Software Engineer, Intel Corporation
> > +41 (0) 79 365 57 48 - http://blog.ffwll.ch
--
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
More information about the Intel-gfx
mailing list