252 lines
12 KiB
XML
252 lines
12 KiB
XML
<?xml version="1.0" encoding="UTF-8"?>
|
|
<protocol name="weston_content_protection">
|
|
|
|
<copyright>
|
|
Copyright 2016 The Chromium Authors.
|
|
Copyright 2018-2019 Collabora, Ltd.
|
|
Copyright © 2018-2019 Intel Corporation.
|
|
|
|
Permission is hereby granted, free of charge, to any person obtaining a
|
|
copy of this software and associated documentation files (the "Software"),
|
|
to deal in the Software without restriction, including without limitation
|
|
the rights to use, copy, modify, merge, publish, distribute, sublicense,
|
|
and/or sell copies of the Software, and to permit persons to whom the
|
|
Software is furnished to do so, subject to the following conditions:
|
|
|
|
The above copyright notice and this permission notice (including the next
|
|
paragraph) shall be included in all copies or substantial portions of the
|
|
Software.
|
|
|
|
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
|
THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
|
|
FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
|
|
DEALINGS IN THE SOFTWARE.
|
|
</copyright>
|
|
|
|
<description summary="Protocol for providing secure output">
|
|
This protocol specifies a set of interfaces used to provide
|
|
content-protection for e.g. HDCP, and protect surface contents on the
|
|
secured outputs and prevent from appearing in screenshots or from being
|
|
visible on non-secure outputs.
|
|
|
|
A secure-output is defined as an output that is secured by some
|
|
content-protection mechanism e.g. HDCP, and meets at least the minimum
|
|
required content-protection level requested by a client.
|
|
|
|
The term content-protection is defined in terms of HDCP type 0 and
|
|
HDCP type 1, but this may be extended in future.
|
|
|
|
This protocol is not intended for implementing Digital Rights Management on
|
|
general (e.g. Desktop) systems, and would only be useful for closed systems.
|
|
As the server is the one responsible for implementing
|
|
the content-protection, the client can only trust the content-protection as
|
|
much they can trust the server.
|
|
|
|
In order to protect the content and prevent surface contents from appearing
|
|
in screenshots or from being visible on non-secure outputs, a client must
|
|
first bind the global interface "weston_content_protection" which, if a
|
|
compositor supports secure output, is exposed by the registry.
|
|
Using the bound global object, the client uses the "get_protection" request
|
|
to instantiate an interface extension for a wl_surface object.
|
|
This extended interface will then allow surfaces to request for
|
|
content-protection, and also to censor the visibility of the surface on
|
|
non-secure outputs. Client applications should not wait for the protection
|
|
to change, as it might never change in case the content-protection cannot be
|
|
achieved. Alternatively, clients can use a timeout and start showing the
|
|
content in lower quality.
|
|
|
|
Censored visibility is defined as the compositor censoring the protected
|
|
content on non-secure outputs. Censoring may include artificially reducing
|
|
image quality or replacing the protected content completely with
|
|
placeholder graphics.
|
|
|
|
Censored visibility is controlled by protection mode, set by the client.
|
|
In "relax" mode, the compositor may show protected content on non-secure
|
|
outputs. It will be up to the client to adapt to secure and non-secure
|
|
presentation. In "enforce" mode, the compositor will censor the parts of
|
|
protected content that would otherwise show on non-secure outputs.
|
|
</description>
|
|
|
|
<interface name="weston_content_protection" version="1">
|
|
<description summary="content protection global interface">
|
|
The global interface weston_content_protection is used for exposing the
|
|
content protection capabilities to a client. It provides a way for clients
|
|
to request their wl_surface contents to not be displayed on an output
|
|
below their required level of content-protection.
|
|
Using this interface clients can request for a weston_protected_surface
|
|
which is an extension to the wl_surface to provide content-protection, and
|
|
set the censored-visibility on the non-secured-outputs.
|
|
</description>
|
|
|
|
<request name="destroy" type="destructor">
|
|
<description summary="unbind from the content protection interface">
|
|
Informs the server that the client will not be using this
|
|
protocol object anymore. This does not affect any other objects,
|
|
protected_surface objects included.
|
|
</description>
|
|
</request>
|
|
|
|
<enum name="error">
|
|
<entry name="surface_exists" value="0"
|
|
summary="the surface already has a protected surface associated"/>
|
|
</enum>
|
|
|
|
<request name="get_protection">
|
|
<description summary="extend surface interface for protection">
|
|
Instantiate an interface extension for the given wl_surface to
|
|
provide surface protection. If the given wl_surface already has
|
|
a weston_protected_surface associated, the surface_exists protocol
|
|
error is raised.
|
|
</description>
|
|
<arg name="id" type="new_id" interface="weston_protected_surface"
|
|
summary="new object id for protected surface"/>
|
|
<arg name="surface" type="object" interface="wl_surface"
|
|
summary="the surface"/>
|
|
</request>
|
|
</interface>
|
|
|
|
<interface name="weston_protected_surface" version="1">
|
|
<description summary="content protection interface to a wl_surface">
|
|
An additional interface to a wl_surface object, which allows a client to
|
|
request the minimum level of content-protection, request to change the
|
|
visibility of their contents, and receive notifications about changes in
|
|
content-protection.
|
|
|
|
A protected surface has a 'status' associated with it, that indicates
|
|
what type of protection it is currently providing, specified by
|
|
content-type. Updates to this status are sent to the client
|
|
via the 'status' event. Before the first status event is sent, the client
|
|
should assume that the status is 'unprotected'.
|
|
|
|
A client can request a content protection level to be the minimum for an
|
|
output to be considered secure, using the 'set_type' request.
|
|
It is responsibility of the client to monitor the actual
|
|
content-protection level achieved via the 'status' event, and make
|
|
decisions as to what content to show based on this.
|
|
|
|
The server should make its best effort to achieve the desired
|
|
content-protection level on all of the outputs the client's contents are
|
|
being displayed on. Any changes to the content protection status should be
|
|
reported to the client, even if they are below the requested
|
|
content-protection level. If the client's contents are being displayed on
|
|
multiple outputs, the lowest content protection level achieved should be
|
|
reported.
|
|
|
|
A client can also request that its content only be displayed on outputs
|
|
that are considered secure. The 'enforce/relax' requests can achieve this.
|
|
In enforce mode, the content is censored for non-secure outputs.
|
|
The implementation of censored-visibility is compositor-defined.
|
|
In relax mode there are no such limitation. On an attempt to show the
|
|
client on unsecured output, compositor would keep on showing the content
|
|
and send the 'status' event to the client. Client can take a call to
|
|
downgrade the content.
|
|
|
|
If the wl_surface associated with the protected_surface is destroyed,
|
|
the protected_surface becomes inert.
|
|
</description>
|
|
|
|
<enum name="error">
|
|
<entry name="invalid_type" value="0"
|
|
summary="provided type was not valid"/>
|
|
</enum>
|
|
|
|
<enum name="type">
|
|
<description summary="content types">
|
|
Description of a particular type of content protection.
|
|
|
|
A server may not necessarily support all of these types.
|
|
|
|
Note that there is no ordering between enum members unless specified.
|
|
Over time, different types of content protection may be added, which
|
|
may be considered less secure than what is already here.
|
|
</description>
|
|
<entry name="unprotected" value="0" summary="no protection required"/>
|
|
<entry name="hdcp_0" value="1" summary="HDCP type 0"/>
|
|
<entry name="hdcp_1" value="2"
|
|
summary="HDCP type 1. This is a more secure than HDCP type 0."/>
|
|
</enum>
|
|
|
|
<request name="destroy" type="destructor">
|
|
<description summary="remove security from the surface">
|
|
If the protected_surface is destroyed, the wl_surface desired protection
|
|
level returns to unprotected, as if set_type request was sent with type
|
|
as 'unprotected'.
|
|
</description>
|
|
</request>
|
|
|
|
<request name="set_type">
|
|
<description summary="set the acceptable level of content protection">
|
|
Informs the server about the type of content. The level of
|
|
content-protection depends upon the content-type set by the client
|
|
through this request. Initially, this is set to 'unprotected'.
|
|
|
|
If the requested value is not a valid content_type enum value, the
|
|
'invalid_type' protocol error is raised. It is not an error to request
|
|
a valid protection type the compositor does not implement or cannot
|
|
achieve.
|
|
|
|
The requested content protection is double-buffered, see
|
|
wl_surface.commit.
|
|
</description>
|
|
<arg name="type" type="uint" enum="type"
|
|
summary="the desired type of content protection"/>
|
|
</request>
|
|
|
|
<request name="enforce">
|
|
<description summary="enforce censored-visibility constrain">
|
|
Censor the visibility of the wl_surface contents on non-secure outputs.
|
|
See weston_protected_surface for the description.
|
|
|
|
The force constrain mode is double-buffered, see wl_surface.commit
|
|
</description>
|
|
</request>
|
|
|
|
<request name="relax">
|
|
<description summary="relax the censored-visibility constrain">
|
|
Do not enforce censored-visibility of the wl_surface contents on
|
|
non-secure-outputs. See weston_protected_surface for the description.
|
|
|
|
The relax mode is selected by default, if no explicit request is made
|
|
for enforcing the censored-visibility.
|
|
|
|
The relax mode is double-buffered, see wl_surface.commit
|
|
</description>
|
|
</request>
|
|
|
|
<event name="status">
|
|
<description summary="security status changed">
|
|
This event is sent to the client to inform about the actual protection
|
|
level for its surface in the relax mode.
|
|
|
|
The 'type' argument indicates what that current level of content
|
|
protection that the server has currently established.
|
|
|
|
The 'status' event is first sent, when a weston_protected_surface is
|
|
created.
|
|
|
|
Until this event is sent for the first time, the client should assume
|
|
that its contents are not secure, and the type is 'unprotected'.
|
|
|
|
Possible reasons the content protection status can change is due to
|
|
change in censored-visibility mode from enforced to relaxed, a new
|
|
connector being added, movement of window to another output, or,
|
|
the client attaching a buffer too large for what the server may secure.
|
|
However, it is not limited to these reasons.
|
|
|
|
A client may want to listen to this event and lower the resolution of
|
|
their content until it can successfully be shown securely.
|
|
|
|
In case of "enforce" mode, the client will not get any status event.
|
|
If the mode is then changed to "relax", the client will receive the
|
|
status event.
|
|
</description>
|
|
<arg name="type" type="uint" enum="type"
|
|
summary="the current content protection level"/>
|
|
</event>
|
|
</interface>
|
|
|
|
</protocol>
|