BRPI0205818B1 - Method, medium and device to cryptographically protect security content - Google Patents

Description

"METHOD, METHOD AND DEVICE FOR CRYPTOGRAPHICALLY PROTECTING SAFETY CONTENT" REFERENCE TO RELATED PATENT APPLICATIONS

This patent application claims the benefit of provisional patent applications serial numbers 60 / 337,617 filed December 4, 2001 and 60 / 339,143 filed December 10, 2001 and is related to commonly assigned copending US Patent Application no. No. 10 / 125,170 filed April 18, 2002. COPYRIGHT NOTICE AND PERMISSION: Part of the disclosure of this patent document may contain material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by any person of the patent document or patent disclosure, as it appears in the patent and patent filings, but otherwise reserves all and guaisguer copyright . The following notice should apply to this document: Copyright © 2001, Microsoft Corp.

FIELD OF THE INVENTION The present invention relates to methods and systems for cryptographically protecting security content. More particularly, the present invention relates to techniques for cryptographically protecting security content sent over a confidential graphics channel, providing both reliability and protection against forgery with respect to content. BACKGROUND OF THE INVENTION: The Internet and many other sources and applications now provide a wide array of streaming and fixed media or other content for listening, viewing, processing, storing and otherwise transmitting. However, there is currently no practical way to capture, record or transmit streaming or fixed media or content in a copy protected manner so that a pirate cannot intercept the stream at any point along the confidential channel, or at any time. as the content is being processed or while the content is being transmitted, and takes possession of a copy or changes the content. This problem has existed with respect to media transmission or recording devices in the past, for example, VCRs for television content, or tape recorders for audio signals, but with at least one fundamental difference. Since digital media content can be recorded virtually without any loss of signal, this poses a "risk" to copyright owners that their works will be freely shared (pirated) without compensation. With VCRs and tape recorders, the device (s) and transmission media attract noise or data corruption within the recording process. With streaming or fixed digital media, there is no reason why virtually less lossy conversions and retransmissions cannot be affected, at least within the limits of the capabilities of the human ear or eye, and there is no reason why unadulterated digital data cannot be stored and freely distributed. Thus, it may be desirable to prevent unrestricted redistribution of digital data because there is little difference between what copyright owners can provide for a fee and what a pirate can provide for free. In addition, with respect to communications that are desired to be confidential, such as e-commerce transactions, it is important for the user to stay in dialogue so that no third party can be aware of the transactions. Thus, with respect to content from a trusted source, there is currently no practical way for data to be safely processed or transmitted on a user's computer without preventing piracy or corruption.

In particular, once content is transmitted over a confidential channel between a host computer system, one or more graphics processing units (GPUs), and a display device, for example, a monitor, there are several opportunities for a hacker or another unauthorized third party camping on a line or signal, and pirating or corrupting the signal. What's more, as user dialogues become more sophisticated through messaging and video teleconferencing services, providing a trusted channel for secure content from wherever it originates becomes the most important advance.

What's more, it is obvious that future generations of operating systems, computing devices, and applications will use more computing power from GPUs for commerce applications, as opposed to drawing much of the computing power from CPUs like today's personal computers (PCs). . In this way, ensuring that content that is sent to GPUs through "reliable graphics" applications will be a key feature of tomorrow's computing devices, and not properly targeted by present computing systems.

This problem of providing a trusted channel for trusted content can be considered to be twofold: (1) one must ensure that trusted content cannot be copied or inspected at any weak point during the trusted channel (reliability) and (2) must Ensure that the confidant channel prevents unauthorized corruption of data on the confidant (protected) channel. In the context of system security, complexity is a risk because proving system security becomes more difficult. As an airport or other security scenario, the more entry and exit points there are in the system, the harder it becomes to ensure security. In this regard, there is currently no means by which the GPU's volume of functionality and presentation driver (s) can be reliable in terms of both reliability and protection capability. Thus, it may be desirable to implement a reliable graphics environment with respect to a computing device that receives content from a trusted source, so that a user of the device can be assured that the content cannot be copied without authorization and cannot be falsified. or amended by a third party.

SUMMARY OF THE INVENTION

In view of the foregoing, the present invention provides methods and systems for cryptographically protecting security content with respect to a graphics subsystem of a computing device. Techniques are implemented to encrypt video memory contents so that unauthorized software cannot gain meaningful access to it, thereby fulfilling the purpose of confidentiality. In addition, a mechanism for counterfeiting detection is provided so that there is a science when data has been altered in some way, thus fulfilling the protection objective. In various embodiments, the invention describes how to encrypt the contents of temporary command overlay and / or storage surfaces, and / or describes how to enable the GPU to operate on encrypted content while preventing its availability to untrusted third parties, devices or software.

The methods of the invention differently include techniques for cryptographically protecting security content with respect to a reliable graphics system having video memory, graphics processing unit (s) (GPU) and a communicatively coupled cryptographic processing device. GPU (s), comprising requesting, by an application or device, the graphics system to perform the processing or transmission of security content, wherein the requirement includes transmitting a session key to the system or application graphics and transmit the security content to the encrypted portion (s) of the video memory which decrypts the content of the encrypted portion (s) of the video memory by the ) GPU (s) communicating with the cryptographic processing device, processing or transmitting the decrypted content by the GPU (s) and outputting the content of the GPU (s). Similar methods of the invention differently include requesting, by an application or device, the graphics system to perform the processing or transmission of security content, wherein the requirement includes transmitting by the application or device a session key to the graphics system for verification. by the cryptographic processing device and transmitting the security content to the encrypted portion (s) of the video memory by decrypting the content of the encrypted portion (s) of the video memory by a mechanism decryption of an input unit of the GPU (s), wherein the decryption mechanism is in communication with the cryptographic processing device, performing the processing or transmission of the decrypted content by the GPU (s) by encrypting the content with an encryption / decryption mechanism of an output unit of the GPU (s) and outputting the decrypted content of the GPU (s) (s).

Other features and embodiments of the present invention are described below. BRIEF DESCRIPTION OF THE DRAWINGS: Methods and systems for cryptographically protecting security content according to the present invention are also described with reference to the accompanying drawings in which: Figure 1A is a block diagram representing an exemplary network environment. having a variety of computing devices in which the present invention may be implemented; Figure 1b is a block diagram representing an exemplary non-limiting computing device in which the present invention may be implemented; Figure 2 is a flowchart illustrating unprotected portions of a graphical confidant channel that are protected according to the invention;

Figures 3A and 3B are block diagrams illustrating exemplary aspects of a first security layer according to the invention;

Figures 4A to 4C are block diagrams illustrating exemplary aspects of a second security layer according to the invention;

Figures 5A to 5B illustrate exemplary arrangement of an AYUV / ARGB format according to the invention;

Figures 6A-6B illustrate exemplary arrangement of a YUY2 format according to the invention;

Figures 7A to 7B illustrate an exemplary arrangement of a compacted flat shape according to the invention;

Figures 8A and 8B are block diagrams illustrating exemplary aspects of a third security layer according to the invention; and Figures 9A and 9B are block diagrams illustrating exemplary encryption techniques that may be applied to the output of a graphics processing unit according to the invention.

DETAILED DESCRIPTION OF THE INVENTION

OVERVIEW

This invention provides system and techniques for extending an operating system, or other intermediary between the content of a trusted source, and a graphics system for processing and / or transmitting the content to enable "reliable graphics" applications such as spoofing, confidential dialogue, and the reproduction of premium content that is resistant to unauthorized duplication. One way to view the invention is that it provides three 'levels' of security: (1) encryption of overlapping surface contents (2) allowing the GPU, or other co-processing device, to operate on the encrypted content without rendering it available to untrusted parties and (3) allow temporary command storage to be encrypted.

As mentioned, in the context of system security, complexity is a risk because it makes it difficult to prove system security. As a result, the invention begins by considering that the GPU functionality volume and display trigger (s) should be considered unreliable. The invention then applies techniques that increase the chances of correct implementation in terms of reliability and protectability by limiting the scope of hardware that can be implemented to meet security criteria.

Some terminology according to the invention has already been discussed above. However, for clarity, some terms will now be emphasized. The term confidential refers to the failure of an unreliable third party, such as a third party device or software, to gain access to information of reliable content. An example of such reliability includes preventing an unreliable third party from gaining access to playback of encrypted premium video content wherever it is along the graphics confidant channel. The term protected refers to preventing an unreliable third party, such as a third party device or software, from gaining access to or altering trusted content information undetected. An example of such protection capability includes preventing an untrusted third party from gaining access or changing the display of a secure dialog that must occur during an e-commerce transaction.

In this regard, the invention considers overlay windows, for example, a user interface on top of a premium content video stream, as well as non-erasable windows, which must be present during an e-commerce dialogue.

EXAMPLE NETWORK ENVIRONMENTS

One skilled in the art may appreciate that a computer or other client or server device may be deployed as part of a computer network, or in a distributed computing environment. With this in mind, the present invention relates to a computer system having any number of memory or storage units, and any number of applications and processes occurring over any number of storage units or volumes, which may imply the storage regime. Reliable graphics of the invention. The present invention may apply to an environment with server computers and client computers arranged in a network environment or distributed computing environments having remote or local storage. The present invention may also be applied to isolated computing devices having programming language functionality, interpretation and execution capabilities for generating, receiving and transmitting information in connection with remote or local services.

Distributed computing facilitates the sharing of computer resources and services through direct exchange between computing devices and systems. These features and services include information exchange, cache storage and disk storage for files. Distributed computing takes advantage of network connectivity, enabling customers to leverage their collective strength to benefit the entire enterprise. In this regard, a variety of devices may have applications, objects or features that may interact to imply reliable graphics channel (s) of the present invention. Fig. 1A provides a schematic diagram of a networked or distributed computing environment. The distributed computing environment comprises computing objects 10a, 10b, etc. and computing objects or devices 110a, 110b, 110c, etc. These objects may comprise programs, methods, data stores, programmable logic, etc. Objects may comprise parts of the same or different device such as PDAs, televisions, MP3 players, televisions, personal computers, etc. Each object may communicate with the other object in the form of communication network 14. This network may itself comprise other computing objects and computing devices that provide services to the system of Fig. 1A. According to one aspect of the invention, each object 10 or 110 may contain an application that must request reliable graphics resources.

It may also be appreciated that an object, such as 110c, may be hosted on another computing device 10 or 110. Thus, while the physical environment described may present the connected devices as computers, such an illustration is merely exemplary and the physical environment may alternatively be depicted or described comprising various digital devices such as PDAs, televisions, MP3 players, etc. software objects such as interfaces, COM objects, and others. There are a variety of systems, components, and network configurations that support distributed computing environments. For example, computer systems can be connected to each other by wired or wireless system, local area networks or widely distributed networks. Today, many of the networks are attached to the Internet, which provides the infrastructure for widely distributed computing and spans many different networks.

In home network environments, there are at least four diverse network transport modes that can each support a single protocol such as power line, data (both wired and wireless), voice (eg, telephone), and entertainment. Most home control devices such as light switches and applications and appliances can use power line for connectivity. Data Services can come in the home as broadband (eg, DSL or Cable modem) and are accessible indoors using either wireless (eg HomeRF or 802.11b) or wired (HomerPNA, Cat 5, even electric power line). Voice traffic may come in either as wired (eg Cat 3) or wireless (eg cell phones (and may be distributed within each Cat 3 wiring). come into the home or via satellite or cable and is typically distributed in the home using coaxial cable.IEEE 1394 and DVI are also emerging as digital interconnectors for media device groups.all of these network environments and others that may emerge as protocol standards can be interconnected to form an intranet that can be connected to the outside world in the form of the Internet In short, a variety of different sources exist for data storage and transmission, and as a result, computing devices will require forms of protection content. throughout the secure data processing channel. The Internet commonly refers to the collection of networks and gateways that use the TCP / IP protocols, which are well known in the art of computer networking. TCP / IP is an acronym for "Transmission Control Protocol / Interface Program". The Internet can be described as a system of geographically distributed remote computer networks interconnected by computers running network protocols that allow users to interact and share information across networks. Because of such widespread information sharing, remote networks such as the Internet have thus often developed far into an open system for whose creators can design software applications to perform specialized operations or services, essentially without restriction.

Thus, the network infrastructure allows a host of network topologies such as client / server, unhierarchical or hybrid architectures. The "client" is a member of a class or group that uses the services of another class or group to which it is not related. Thus, in computing, a client is a process, that is, roughly a series of instructions or tasks, which requires a service provided by another program. The client process uses the required service without having to "know" any operating details about the other program or the service itself. In a client / server architecture, particularly a networked system, a client is usually a computer that accesses network resources shared by another computer, for example, a server. In the example of Fig. 1A, computers 110a, 110b, etc. can be considered as clients and computers 10a, 10b, etc. can be considered as the server where server 10a, 10b, etc. keeps data that is later replicated on client computers 110a, 110b, etc.

A server is typically a remote computer system accessible on a remote network such as the Internet. Client process can be activated on a first computer system, and server process can be activated on a second computer system, communicating with one another in a communications medium, thus providing distributed functionality and allowing multiple clients to take advantage. of the ability to gather information from the server. The client and server communicate with each other using functionality provided by a protocol layer. For example, Hypertext Transfer Protocol (HTTP) is a common protocol that is used in conjunction with World Wild Web (WWW). Typically, a computer network address such as a Universal Resource Locator (URL) or an Internet Protocol (IP) address is used to identify the server or client computers with each other. The network address may be referred to as a Universal Resource Locator address. For example, communication may be provided on a communications medium. In particular, the client and server can be coupled together through TCP / IP connections for high capacity communication.

Thus, Fig. 1A illustrates an exemplary network or distributed environment, with a server communicating with client computers over a network / bus, wherein the present invention may be employed. In more detail, various servers 10a, 10b, etc. are interconnected via a communications network / bus 14, which may be a LAN, WAN, intranet, the Internet, etc., with various remote client or computing devices 110a. 110b, 110c, HOd, HOe, etc., such as a laptop, pocket computer, weak client, networked instrument, or other device, such as a VCR, TV, oven, light, heater, and others according to present invention. It is thus contemplated that the present invention may apply to any computing device with respect to which it is desirable to process, store or transmit security content from a reliable source.

In a network environment where communications network / bus 14 is the Internet, for example, servers 10 may be servers on the network with which clients 110a, 110b, 110c, HOd, HOe, and so on. communicate over any of a variety of protocols known as HTTP. Servers 10 may also serve as clients 110, as may be characteristic of a distributed computing environment. Communications can be wired or wireless where self-propelled. Client devices 110 may or may not communicate over network / communications bus 14, and may have independent communications associated with them. For example, in the case of a TV or VCR, where it may or may not be a network aspect to control it. Each client computer 110 and server computer 10 may be equipped with various application program modules or objects 135 and with connections or access to various types of storage elements or objects by means of which these files may be stored in their part ( s) can be downloaded or migrated. Thus, the present invention may also be used in a computer network environment having client computers 110a, 110b, etc. which can access and interact with a computer network 14 and server computers 10a, 10b, etc. which can interact with client computers 110a, 110b, etc., and other devices 111 and databases 20.

EXAMPLE COMPUTER DEVICE Fig. 1B and the following discussion are intended to provide a brief overview of a suitable computing environment in which the invention may be implemented. It should be understood, however, that handheld, portable and other computing devices and computing objects of all types are contemplated for use with respect to the present invention. Although a general purpose computer is described below, this is an example, and the present invention may be implemented with a weak client having interoperability and network / bus interaction. Thus, the present invention may be implemented in a network hosted services environment where very little or minimal client resources are implied, for example, a networked environment in which the client device merely serves as an interface for the network. / bus, such as an object placed on an instrument. In essence, wherever such data may be stored or from that data may be fetched or transmitted is a desirable or suitable environment for the operation of the cryptographic protection of the security content of the invention.

Although not required, the invention may be implemented through an operating system, application programming interface (API) and / or included within application software that interfaces with trusted content. In various embodiments, the invention also applies to hardware conforming to the interface embodiment, and encryption techniques described below. Software may be described in the general context of computer executable instructions, such as program modules, being executed by one or more computers, such as client workstations, servers, or other devices. In general, program modules include routines, programs, objects, components, data structures, and others that perform particular tasks or implement particular abstract data types. Typically, the functionality of program modules may be combined or distributed as desired in various embodiments. Moreover, those skilled in the art will appreciate that the invention may be practiced with other computer system configurations. Other well-known computing systems, environments and / or configurations that may be suitable for use with the invention include, but are not limited to personal computers (PCs), automated teller machines, server computers, portable devices or laptops, multiple processor systems, microprocessor based systems, consumer programmable electronics, network PCs, instruments, lights, room control elements, mini-computers, main computers, and more. The invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are connected via a communication bus / network or other data transmission medium. In a distributed computing environment, program modules may be located on both local or remote computer storage media including memory storage devices, and client nodes may in turn behave as server nodes. Fig. 1B thus illustrates an example of a suitable computing system environment 100 in which the invention can be implemented, although as made clear above, computing system environment 100 is only an example of a suitable computing environment. and it is not intended to suggest any limitation on the scope of use or functionality of the invention. Neither should the computing environment 100 be construed as having any dependency or requirement related to any or combination of components illustrated in the exemplary operating environment 100.

Referring to Fig. 1B, an exemplary system for implementing the invention includes a general purpose computing device in the form of a computer 110. Computer components 110 may include, but are not limited to, a processing unit 120, a system memory 130 and a system bus 121 that couples various system components including system memory to the processing unit 120. system bus 121 can be any of several types of bus structures including memory bus or controller memory, a peripheral bus, and a local bus using any of a variety of bus architectures. By way of example, and not limitation, such structures include Standard Industrial Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Intensified ISA bus (EISA), Video Electronics Standards Association local bus ( VESA) and Peripheral Component Interconnect (PCI) bus (also known as the Mezzanine bus). Computer 110 typically includes a variety of computer readable media. Computer readable media may be any available media accessible by computer 110 and includes both volatile and nonvolatile media, removable and non-removable media, by way of example, and not limitation, computer readable media may comprise media. computer storage and media. Computer storage media includes both volatile and non-volatile removable and non-removable media implemented in any method or technology for storing information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to RAM, ROM, EEPROM, fast memory or other memory technology, CDROM, digital versatile discs (DVDs) or other optical disc storage, magnetic cassettes, magnetic tape, hard disk storage. magnetic disk or other magnetic storage devices, or other means that can be used to store the desired information that can be accessed by the computer 110. The media typically incorporates computer readable instructions, data structures, program modules or the like. data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information release means. The term "modulated data signal" means a signal that has one or more of its characteristics determined or altered in such a way as to encode the information in the signal. By way of example, and not limitation, the media includes wired media such as wired network or direct wired connection, and wireless media such as acoustic, RF, infrared or other wireless media. Combinations of any of the above should also be included within the scope of computer readable media. System memory 130 includes computer storage media in the form of volatile and / or nonvolatile memory such as read-only memory (ROM) 131 and random access memory (RAM) 132. A basic input / output system 133 ( BIOS), containing the basic routines that help transfer information between elements within computer 110, such as during startup, is typically stored in ROM 131. RAM 132 typically contains data and / or program modules that are immediately accessible and / or currently being operated by a processing unit 12 0. By way of example, and not limitation, Fig. 1B illustrates operating system 134, application programs 135, other program modules 136, and program data 137. Computer 110 may also include other removable / non-removable, volatile / non-volatile computer storage media. By way of example, Fig. 1B illustrates a hard disk drive 141 reading or writing to non-removable, non-volatile magnetic media, a magnetic disk drive 151 reading or writing to a non-volatile removable magnetic disk, and an optical disk drive 155 that reads or writes to a removable, nonvolatile optical disc 156, such as a CD ROM or other optical medium, other removable / non-removable, volatile / nonvolatile computer storage media that may be used in the Exemplary operating environments include, but are not limited to, magnetic tape cassettes, fast memory cards, digital versatile disks, digital video tape, solid state RAM, solid state ROM, and others. Hard disk drive 141 is typically connected to system bus 121 via a non-removable memory interface such as interface 140, and magnetic disk drive 151 and optical disk drive 155 are typically connected to system bus 121 by a removable memory interface such as interface 150.

The drivers and their associated computer storage media discussed above and illustrated in Fig. 1B provide storage of computer readable instructions, data structures, program modules, and other data for computer 110. In Fig. 1B, for example, the hard drive 141 is illustrated as storage operating system 144, application programs 145, other program modules 146, and program data 147. Note that these components may either be the same or different from operating system 134, application programs, and 135, other program modules 136, and program data 137. Operating system 144, application programs 145, other program modules 146, and program data 147 are given different numbers here to illustrate that, at a minimum, they are different copies. A user may enter commands and information within computer 110 through input devices such as a keyboard 162 and pointing device 161, commonly referred to as a mouse, trackball or touch pad. Other input devices (not shown) may include a microphone, joystick, game stand, satellite disc, scanner, or others. These and other input devices are often connected to processing unit 120 via a user input interface 160 that is coupled to system bus 121, but can be connected by other interface and bus structures, such as a parallel port, game port or a universal serial bus (USB). A graphics interface 182, such as Northbridge, may also be connected to system bus 121. Northbridge is a chip group that communicates with the CPU, or host processing unit 120, and assumes responsibility for accelerated graphics port (AGP) communications. ). One or more graphics processing units (GPUs) 184 may communicate with the graphics interface 182. In this regard, GPUs 184 generally include on-chip memory storage, such as record storage, and GPUs 184 communicate with a memory of 186. GPUs 184, however, are an example of a coprocessor and so a variety of coprocessing devices can be included in computer 110. A monitor 191 or other type of display device is also connected to the system bus. 121 through an interface, such as a video interface 190, which may in turn communicate with video memory 186. In general, it is this part of a computing device that is vulnerable, and therefore the present invention provides protection and reliability of the data being processed or transmitted incident to it. In addition to monitor 191, computers may also include other peripheral output devices such as speakers 197 and printer 196, which may be connected through a peripheral output interface 195. Computer 110 may operate in a networked or distributed environment using logical connections to one or more remote computers, such as a remote computer 180. The remote computer 180 may be a personal computer, a server, a router, a network PC, an equalizer, or other common network nodes, and typically includes many or all of the above described elements relating to computer 110, although only one memory storage device 181 has been illustrated in Fig. 1B. The logical connections described in Fig. 1B include a local area network (LAN) 171 and a wide area network (WAN) 173, but may also include other networks / buses. Such networking environments are common in homes, offices, enterprise-wide computer networks, intranets, and the Internet.

When used in a LAN network environment, computer 110 is connected to LAN 171 through a network interface or adapter 170. When used in a WAN network environment, computer 110 typically includes a modem 172 or other means for establishing communications. WAN 173, just like the Internet. Modem 172, which may be internal or external, may be connected to system bus 121 via user input interface 160, or another appropriate mechanism. In a networked environment, the program modules described for computer 110, or portions thereof, may be stored in the remote memory storage device. By way of example, and not limitation, Fig. 1B illustrates remote application programs 185 residing in memory device 181. It will be appreciated that the network connections shown are exemplary and other means of establishing communication links between computers may be be used.

EXEMPLARY DISTRIBUTED COMPUTER STRUCTURES OR ARCHITECTURES Several distributed computing structures have been and have been developed in consideration of the convergence of personal computing and the Internet. Similar individuals and business users are provided with an uninterrupted, network-enabled interoperable interface for computing applications and devices, making the network browser or network-oriented computing activities highly.

For example, the MICROSOFT®1s.Net platform includes servers, building block services such as network-based data storage and downloadable device software. Generally speaking, the .Net platform provides (1) the ability to get the full range of computing devices to work together and have user information automatically updated and synchronized across all of them, (2) increased interactive capability for web sites, allowed by greater use of XML than HTML, (3) online services featuring personalized access and release of products and services to the user from a central starting point to the managerial of various applications, such as email, or (4) centralized data storage, which will increase the effectiveness and ease of access to information, as well as information synchronization between users and devices, (5) the ability to integrate various media such as (6) for creators, the ability to create reusable modules, thereby increasing productivity and reducing the number of programming errors, and (7) also many other cross platform integration features.

While exemplary embodiments herein are generally described with respect to software implemented methods residing in a computing device, one or more parts of the invention may also be implemented via an operating system, application programming interface (API) or "man intermediate" object between a coprocessor and tampered content, so that trusted content services can be executed, supported or accessed through all of .NET's languages and services, and other distributed computing frameworks as well. Furthermore, it may be appreciated that one or more techniques described in accordance with the invention may entail hardware changes, such as GPU changes to suit the techniques. Cryptographically Protecting Security Content The present invention thus provides methods and systems for extending an operating system, or any platform, to enable "reliable graphics" applications such as tamper-resistant, confidential dialogs and to allow reproduction of the contents of a resistant to unauthorized duplication. The problem solved by the present invention is illustrated by Fig. 2 wherein the encrypted premium content 200 is being shown passed or generated by a trusted TS software application. Incident to the use of trusted TS software, 200 content may imply functionality of the GPU 265 prior to transmission (the other use) of 200 content through a 295 transmission device. Such 200 content will be passed from system or host memory 230 to video memory 260 for processing by GPU 265. The dotted line in Fig. 2 illustrates where the security problem lies. As listed in the background, no system today properly ensures secure and confidential release of content through the elements surrounded by the dotted line. From the standpoint of trusted TS software, the first problem is whether or not components within the dotted line can be trusted with content before handling content to the GPU. Assuming that TS trusted software can properly authenticate components within the dotted line, the second problem from the TS trusted software standpoint is that TS trusted software must be reasonably secure that once data is handled within the dotted line, the data will not be changed. The present invention addresses both types of problems in various embodiments described below. In many ways, the methods and systems are implemented (1) by encrypting the contents of the overlay surfaces (2) by allowing the GPU to operate on the encrypted content without making it available to untrusted applications or parts or (3) by encrypting the content of the overlay surfaces. temporary storage of command.

A first aspect of the invention-driven rendering and rendering problem of reliable graphics concerns the provision of a cryptographic processor (sometimes referred to as 'crypto') and indexed key management. US Patent Application commonly referred to as AAA / BBB.CCC (the 'CCC patent application), filed in DD of Month, YYYY, entitled "Methods and Systems for Authentication of Components in Graphics System" teaches the techniques for authenticating the component (s) against a graphics system as described below. It is to be understood, however, that the invention assumes that authentication has occurred and thus any authentication techniques can be leveraged to provide a reliable graphics platform as described in the various embodiments of the invention below.

EXAMPLE AUTHENTICATION

In this regard, secure graphics cards should be able to authenticate themselves as such. In particular, reliable software must be able to distinguish a secure graphics card from a traditional graphics card or an intrusion device such as a scam. In addition, reliable software should be able to reveal cryptographic keys to the graphics card and be able to verify that the key receiver is indeed a secure graphics card. For this purpose, secure graphics cards are equipped with a crypto processor according to the invention which performs the cryptographic authentication tasks and carries the keys.

With respect to hardware, crypto processors are provided according to the invention, which are individualized and certified during manufacture. Each cryptoprocessor contains a unique Kpriv private description key. Although many different encryption and decryption algorithms are contemplated according to the invention and known to those in the cryptographic techniques, for context, the algorithm described herein is RSA and the key length is 1024 bits, both non-limiting model choices, which can be worked on according to the exchange depending on the application and the desired level (s) of security.

In this regard, the cryptoprocessor is permanently attached to the graphics card either by adding it to an existing chip or by adding it as a separate chip to the card. The cryptoprocessor implements a public key crypto algorithm as described in more detail below and hides a unique private key. In this regard, such a cryptoprocessor can perform public key decryption quickly with current silicon technologies. However, the cryptoprocessor may also include an accelerated public key and may implement a symmetric cipher (AES) and some control logic.

In an exemplary non-limiting embodiment, the cryptoprocessor includes the following volatile registers: (1) a 256-bit S register for the session key. The lifetime of this key is typically the reliable software operating time and (2) an array of a plurality of index keys. Each key can be 128 bits long, although other choices may be appropriate. Each key is associated with a particular window and is used by the graphics card to decrypt its contents. The lifetime of each key is governed by trusted software instructions.

As mentioned, the cryptoprocessor of the invention is permanently attached to the graphics card. Thus, a means is required to safely interface with the cryptoprocessor to make use of its functionality. Regarding the interface with the crypto process, the present invention considers at least two techniques: (a) an external interface for the reliable TS software and (b) an interface for the GPU 265. This interface - at least in its aspect cryptographic - should be standardized. That interface may be implementation specific, but must adhere to the general rules outlined below.

With respect to the implementation of the external interface, the external interface uses a private key encryption protocol (PK) for authentication and key transports. Under this protocol, trusted TS software decrypts a session key with the cryptoprocessor public key. The cryptoprocessor receives the resulting cryptobolb and decrypts it with its private key, thus obtaining the session key. Reliable software and the cryptoprocessor now share secrecy. Reliable software can use this session key to send instructions to the cryptoprocessor.

In a non-limiting embodiment, the external interface is exposed through the following functions by the cryptoprocessor: The SetSessionKey () function performs the initial authentication step and key transport. It is the only function that accesses crypto-processor public key functionality. Thus, in one embodiment, the present invention contemplates calling this function once each boot. The following exemplary peseudo code represents a non-limiting implementation of SetSessionKey (): SetSessionKey (cryptoblob) {PKDcrypt (privateKey, cryptoblob, ssesionKey); } Upon successful completion of the operation, the sessionKey record contains a cryptoblob key, such as a 256-bit key. The public key algorithm may be, for example, 1024 bit RSA.

Once the symmetric session key K has been established between trusted software and the cryptoprocessor, this key can be used to ensure all additional communication to and from the cryptoprocessor. Reliable software and the cryptoprocessor can communicate using simple Get and Set methods whose parameters are crypto-protected for reliability and integrity. In particular, the parameter block B of each call can be processed as non-limiting as follows: AES (M | HMAC (M, kl), K2), where: Kl is the first half of K (bits 0 to 127) K2 is the second half of K (bits 128 through 255) AES (M, K) is the result of encrypting message M under the K key with AES in CBC mode.

HMAC (Μ, K) is the result of computing an HMAC during an appropriate verification function in message M with the key K A | B is the result of concatenating A and B.

This format can be used for input parameters and for the parameters of the following functions: Set ([IN] BOOL needsAck, [IN] BITS128 nonce, [IN] ENUM propertylD, [IN] BYTESEQUENCE propertyParamters, [OUT] BYTESEQUENCE ack) where: needsAcl is a boolean value, which allows trusted software to indicate if a confirmation is required.

Nouce is a 128-bit value selected by trusted software. Nonce can be used in confirmation if a confirmation is requested.

PropertylD identifies the property being determined. An exemplary list of supported properties is given below in Table 1.

PropertyParamters is a sequence of parameters, which is specific to each propertylD.

Finally, Ack is the confirmation of the operation. The cripoprocessor produces ack if and only if needsAck has been determined. Ack is composed of nouce followed by a message, which is specific to each propertylD. TABLE 1 - LIST OF PROPERTY IDS TO OBTAIN THE FUNCTION With respect to the property ID Index key, this method writes a new purpose key and tag into the key register identified by the index.

For property ID Output lock, this method determines the output lock flag. When this flag is set, the screen geometry (width, height, color depth, restore rate) as well as the graphics card output (VGA, DVI) cannot be changed. In particular, the graphics card does not execute instructions to change these junctions while the output lock flag is determined.

With respect to property ID L2KeyMgmt, this method adjusts the key renewal frequency under the second protection layer described in accordance with the invention, that is, by encrypting input and output described in more detail below.

Similarly, the Get function is proposed as follows: Get ([IN] BITS128 nonce, [IN] ENUM propertylD, [IN} BYTESEQUENCE propertyParamters, [OUT] BYTESEQUENCE response) where: nonce is a 12-bit value selected by trusted software to be used in the answer.

PropertylD identifies the property being determined. The list of supported properties is given below in Table 2.

PropertyParamters is a sequence of parameters, which is specific to each propertylD. The response contains the result of the operation. The response consists of nonce followed by a message, which is specific to each propertylD.

TABLE 2 - LIST OF PROPERTY IDs FOR SET FUNCTION

With respect to Output Ports, this method returns the setting of the graphics card outputs, eg VGA, DVI, etc.

With respect to the Authentication Code, this method returns the verification of the contents of a second window to the first protection layer described in accordance with the invention, ie overlay encryption.

With respect to the DX-SEC version, this method returns the DX-SEC version supported by the graphics card.

With respect to Safe Surface Account, this method returns the number of safe surfaces supported by the graphics card.

With respect to the Overlay Surface Account, this method returns the number of secure overlay surfaces supported by the graphics card.

With respect to the Primary Type, this method returns 1, and provides future flexibility.

With respect to geometry, this method returns the width, height, restoration rate and color intensity of the primary surface. The Set function can also further include a method that determines the location or size of a protected region overlay, or the location and size of a fraction of a primary (primary) surface that must be decrypted.

Thus, the above SetSessionKey Get and Set function concerns the non-limiting modalities of an external interface. The term "internal interface" refers to the interface between the cryptoprocessor and the rest of the graphics card. The details of this type of interface according to the invention are up to the implementation of each individual graphics card, subject to the following restrictions: (1) the cryptoprocessor must be permanently attached to the graphics card and (2) the connection between the cryptoprocessor and the rest of the graphics card should not be exposed.

In this regard, removing the cryptoprocessor from the graphics card should not be trivial. If the cryptoprocessor is implemented as a separate chip, this is mainly a constraint on the mechanical interface that connects the cryptoprocessor to the graphics card. Typically, the cryptoprocessor should be soldered to the graphics card. Alternatively, the cryptoprocessor may reside on the same chip as the main GPU. The use of standard mechanical interfaces that allow the cryptoprocessor to be removed, for example smart card readers, socket holders and others, is not acceptable.

Also, the physical connection between the cryptoprocessor and the rest of the graphics card should not be accessible and should not be exposed by standard interfaces. For example, a USB connector on this bus is not acceptable according to the invention.

With respect to the key management rules, each index key may only be used according to its associated purpose parameter. In one embodiment, the purpose parameter values have the following meanings: L1STREAM: This switch can only be used with the DX-SEC stream cipher described below with respect to the first security layer provided by the invention, ie the overlay encryption. L2BL0CK: This key can only be used with the ECB mode block cipher of the second security layer provided by the invention, ie encryption inputs and outputs, described below. The block cipher in ECB mode is used to decrypt the texture blocks, which were written by trusted software.

In this regard, no copy of the keys should be kept when the index has been filled with a new value.

FIRST SECURITY LAYER - ENCRYPTED OVERLAYS

Since video memory can be mapped and read by unreliable software operating on the CPU, video memory cannot contain information in plain text form. Video memory subject to this requirement includes video memory used to restore the display. An initial implementation of a system meeting this criterion according to the invention encrypts the contents of an overlay surface. The overlay can then be decrypted on the move by the DAC hardware, or almost before reaching the DAC hardware, while the image is sent to the display. Fig. 3A illustrates an exemplary implementation of such a technique. The encrypted premium content 200, wherever found or generated incident to the trusted software application 210, is received via trusted application 210. An authentication exchange with the cryptoprocessor 220 results, as the procedure (s) of exchange of authentication described above in exemplary embodiments or by any other technique for secure or cryptographic key release or by a route protected by other means. Content 200 passes from system memory 230 to an encrypted overlay surface 240, which overlaps primary surface 270 of video memory 260. Along with cryptoprocessor 220, a decryption portion 250 of GPU 265 decrypts the encryption layer provided by encrypted overlay 240 and outputs the selected pixel content 280 to digital video interface (DVI) / analog to digital converter (DAC) output 290 to output to a transmitting device, such as a monitor. However, the system set forth in Fig. 3A does not meet all the criteria described above, since there is only one overlap. To satisfy the minimum functionality bus required for a reliable environment, such as reliable windows, the invention allows two overlaps in an alternative embodiment. The first "confidential" overlay is basically an overlay that exists on platforms today, primarily for video playback, expanded so that its contents can be encrypted. The second overlay is specifically designed to present the responsive user interface as e-commerce dialogs. This "protected overlay" is always on top and not concealable, ie no color switching and takes precedence over the first overlay. The second overlap may be subject to certain limitations to minimize cost. For example, the second overlay may be provided so that the data is in the same pixel format as the primary and is not capable of being temporarily stretched or multi-stored. In addition, the contents of the protected overlay can be verified by the hardware. Table 3 summarizes the exemplary differences between confidential overlap and protected overlap.

TABLE 3 COMPARISON OF CONFIDENTIAL AND PROTECTED OVERLAYS Fig. 3B shows a system that includes both confidential overlays, for example, overlays 310a, 310b, and 310c of the alternate overlays alternating chain 310, as well as protected overlays 320. When possible, Stream ciphers are used for secure surface encryption because they are faster and simpler to implement than block ciphers (see Appendix A for more details). Stream ciphers encrypt the database at "byte position in stream". Thus, the first security level of the invention initializes the stream cipher with the pixel encryption key in the upper left corner of the surface. The stream cipher is advanced for each pixel contained within the overlay surface regardless of whether the pixel will be displayed. The proposed system contains two stream cipher decryption components 300a and 300b, one for the confidential overlay and one for the protected overlay, respectively. Once decrypted pixel values are available, hardware 280 selects the pixel value of protected overlay 320, confidential overlay 310 (if primary 270 is equal to color key and / or if color keying is allowed) or 270 and sends the pixel value to the hardware display via DVI / DAC 290. It is noted that an opponent can gain access to overlays by any number of means, and thus either make the displayed image invisible or overwrite security content. noisy content, since the data written by the opponent will also be decrypted. Although the invention does not directly defend against these attacks, the invention does not provide an integrity assessment to ensure that expected content is presented to the end user. Thus, if the output is anything other than the input, the trusted user or software 210 may be warned that the content has been falsified.

With respect to a software interface for overlays, in addition to the usual overlay information such as source and destination mooring box, destination color key, etc., confidential overlay 310 maintains that the encryption key index is specified and protected overlay 320 additionally maintains that a memory location is specified where the cyclic redundancy code (CRC) of the decrypted overlay contents must be written. The confidential overlay interface is similar to existing overlays except that the toggle method specifies the encryption key index of the overlay 310a return staging contents, that is, the temporary storage for which the data is being transferred. alternate. The protected overlay interface is simpler and provides a condition for a surface CRC to be sent to a memory location after it is first displayed. The standardized hardware interface controls the location of the overlay, and makes the CRC available to interested software on an interrogation basis. For example, one record may indicate if the CRC is available, and another may allow the CRC to be read. For example, the following exemplary pseudocode may be used with respect to the protected overlay interface: HRESULT UpdateOverlay (LPPOINT ppntUL); where: ppntUL specifies the upper left corner of the protected overlay.

In this regard, software 210 calculates the CRC value expected to be related to integrity.

SECOND LAYER OF SECURITY - ENCRYPTED INPUTS AND OUTPUTS

According to the invention, to augment GPU 265 to control encrypted content as input and output encrypted content as output, encryption and decryption hardware is added to the texture mapping unit (on the input side) and the matching unit. alpha (on the output side), and hardware designers cooperate to follow some rules implementing this functionality. Since stream ciphers do not allow random access to encrypted data, the system uses block ciphers, encrypting the data, for example 128 bits at a time. The texture mapping unit decrypts in a cache line fill and the alpha blending unit decrypts when reading a cache cache from color and encrypts before writing. The encryption keys used in these operations may differ.

The different computational tasks of 3D transmission, such as video decoding, are direct extensions of the paradigm already described. Instead of textures, video macroblocks serve as the encrypted input; and instead of temporary color storage, the output structure being decoded serves as the encrypted salts. If content is protected while it is released in-band in the command stream to the GPU 265, how command temporary storage can be encrypted is also described below.

Fig. 4A depicts the described system which performs a front end operation, considering an encrypted surface 420 as input and emitting an encrypted surface as output, i.e. the front end of the encrypted texture and temporary color storage techniques by means of encryption and decryption component 440. The invention also provides encrypted texture 400a, where texture 400b may be typical texture of video memory 260. Encrypted texture 400a exits to decryption component 450 of GPU 265, which works in conjunction with cryptoprocessor 220 to decrypt texture and apply graphics algorithms such as shading, etc. decrypted data of component 440 with component 430a.

Anticipating the arrangement of a composite page swapping desktop, the system of Fig. 4A already described can protect the entire desktop as long as the DAC hardware can decrypt the primary 270 as well as the overlapping surfaces 310 and 320 above. described. It is observed that the DAC hardware decrypts using a block cipher in place of a stream cipher in such a case. Such a system allows an arbitrary number of confidential surfaces to participate in the desktop, with arbitrary Z-ordering, matching, or even 3D or other effects applied to them without compromising security. Protected overlapping surfaces 320, which should always be on top and whose contents must be verifiable, reside on separate surfaces. The confidential overlay 310 described above remains until it can be emulated in software by the secure page switching desktop, or platform.

In one embodiment, in addition to being able to decrypt the primary 270, the system requires the GPU 265 to be able to encrypt plain text from usual desktop applications as well, such as reliable word processors, so they can also participate on the desktop. Fig. 4B illustrates such a scenario in which primary alternating chain 510, including forward 510b and return 510a, are encrypted. Thus, the primary surfaces may be operated by a desktop composer 430 protected by the encryption / decryption component 440a for its output / input thereon, respectively. In communication with the cryptoprocessor 220, a decryption component 500 decrypts the front surface 510b for output to DVI / DAC 290. This exposes the system to certain types of attacks which are detailed below with respect to ensuring security where some strategies for debating are discussed. defend against these attacks.

Fig. 4C depicts an alternative to Fig. 4B, wherein transcription to the confidential overlap alternating chain 310 is effected. Thus, as an alternative to encrypting the primary 270 according to the invention, the hardware may allow for consumer stream encryption encryption by the confidential overlay hardware 300a which may decrypt the stream encryption data along with cryptoprocessor 220. This presentation mechanism may be less expensive than encrypting the primary with a block cipher, but may not be as scalable or flexible as a planned exchange. Since confidential override 310 uses a stream cipher for encryption, a reasonable operation to support in this context is a 'copy' operation in which the entry is decrypted via decryption component 440b which uses block cipher. from the input surface 510a and re-encrypted by component 440b using the stream cipher overlay. These embodiments, and various combinations thereof, are useful, for example, where one encrypted entry at a time is sufficient as long as any number of plaintext entries can be combined with the encrypted entry to generate encrypted output.

With respect to ensuring safety, there are several measures that can be implemented according to the invention. First, the second security layer described above has the idea that plain text cannot be leaked from GPU 265 once it has been decrypted. For example, no debug log or other facilities exist that allow plain text to be read off the chip by the host central processing unit (CPU). In addition to taking care of hardware design to prevent such leaks, the GPU 265's instruction series is designed so that it is impossible to allow input decryption without also allowing output encryption. In addition, the hardware avoids the leakage of plain text data, either by a fraudulent trigger, by adverse code or by accident.

Also, the hardware may not leak the keys. Once keys are released to the GPU 265 by the cryptographic protocol described in accordance with the authentication exchange, they are only available for the encryption and decryption components.

As discussed above, if the GPU 2 65 is able to encrypt the plaintext for display on primary 270, this feature is considered a system vulnerability since this encryption facility is the only mechanism described in which an adversary can have the text. and the corresponding cipher text available at the same time. By mapping the primary surface so that it can be scanned by the CPU and creating a window to encrypt, the adversary can construct a subset of the ciphertext blocks that correspond to known plaintext blocks. These so-called "dictionary attacks" work best when the number of blocks that are "interesting" is small. For example, for displaying black and white dialog boxes in a 32bpp view, since there are 4 pixels per block, only 16 blocks are required to display such a dialog. One possible avenue for an opponent who discovered the 16 ciphertext blocks would be to fake an end-user dialog by creating content that is meaningful even after decryption by the GPU 265. For this reason, protected overlap is best suited for tough dialogs. counterfeiting because it allows applications to detect when the end user has not seen what was expected. There are two good strategies to frustrate opponents who want to create dictionaries. First, since dictionaries are only good for a given key, changing the key and re-encrypting the content forces the opponent to start over with a new dictionary. Also, for primary encryption, the key does not need to be made available to the software - it can be wrapped around the hardware and the software only needs to be notified that the key has changed. Since the previous key is still available, the software can use the previous key to decrypt and re-encrypt the unchanged parts of the primary. Consequently, cryptoprocessor 220 periodically rolls up the encryption key for primary 270 and such that the previous key is still available, for example, a dual temporary storage action of encryption key indices, and in a manner notify the software that the key has been rolled up.

Another strategy involves coding the location within the image before encryption. For example, the pixel location (x, y) in the image (or some derived value such as image shift) may be unique or-ed (XOR'd) in the pixel data before encrypting it; The operation can be undone later after decryption. As a result, blocks for pixels in different areas of the surface are encrypted differently, and plaintext ciphertext mapping is only significant for a given surface position that is unavailable to a burglar. The present invention also provides predefined arranged formats. Since black screen textures and surfaces require random access, they should be block cipher encrypted. There is good synergy between the typical block size for a block cipher and the typical cache line size for a modern 3D accelerator, for example if the cache line and block size were both 128 bits, then encryption and decryption Effective tools can be implemented in hardware. Even if there are slight differences (eg, 128-bit block size and 256-bit cache line size) the hardware implementation is equally effective.

One problem with encrypted texture data is that a block encryption scheme requires that an adjacent block of bytes be available before it can be encrypted or decrypted; and a cache line fill requires that the pixel data be 'arranged', that is, translation of a position (X, Y) in the image to an address is formed such that the cache line fill gives a region. 2D of pixels. So far, hardware vendors have ostensibly exposed linear surface formats while arranging image data without application knowledge. Once reliable software will be emitting the encrypted texture data, however, the software must have a priori knowledge of the arrangement scheme so that it can encrypt adjacent blocks of data and can preserve 2D location. In response, the invention defines a dictionary of arranged image formats including YUV 4: 4: 4, 4: 2: 2 and 4: 2: 0 as well as RGB formats for use by the application. The performance of these formats may not be quite as high as if the images were arranged in a hardware-specific format, but the encryption is worth a slight performance degradation, ie security for speed.

With respect to AYUV / ARGB (32bpp, compressed), this 32bpp surface format contains an alpha channel in addition to the 8-bit color channels for the luminance (Y) and chromaticity (U and V) samples. Alternatively, it may contain an ARGB 32bpp standard, provided both formats are 32bpp and compressed. The following discussion assumes AYUV. The linear layout is as in Fig. 5A. The pixel offset (X, Y) in the image is as follows: Offset = Y * Pitch + X * 4 Assuming a 128-bit encryption block size and cache line size, 4 pixels can fit into a single block. Interleaving the minimum of significant bits of X and Y before generating the address will result in improved 2D locale in a cache line fill. These blocks are placed linearly according to the format. The resulting image layout is illustrated by Fig. 5B. Each numbered rectangle is a pixel and the bold rectangles are encrypted blocks. Exemplary peseudocode for the arrangement function of the invention for this format which converts a location (x, y) in the image to an offset is as follows: DWORD

SwizzleAYUV (DWORD x, DWORD y, DWORD Pitch) {// pitch is the number of bits per line scanned from macro blocks return BlockOffset + IntraBlockOffset * 4; } With respect to YUY2 (16bpp, compressed), this surface format horizontally sub-samples the U and V chromate samples by a factor of 2. The result is a 16-bit compressed image format by pixel. The linear layout is shown in Fig. 6A, the arrangement form of the invention allocates 4x2 pixel encrypted blocks as shown in Fig. 6B. As for Figs. 5A and 5B, 128-bit blocks are arranged as well. It is observed that with Fig. 6B and the following exemplary array peseudocode which translates the coordinate pairs (x, y) into image compilations, U and V are assumed to have the same X: DWORD coordinates.

SwizzleYUY2Y (DWORD x, DWORD y, const SURFACEDESC & sd) {assert (x <sd.Width); assert (y <sd.Height); DWORD BlockOffset DWORD dwRet = BlockOffset + (IntraBlockOffsetccl); return dwRet; } DWORD

SwizzleYUY2U (DWORD x, DWORD y, const SURFACEDESC & sd) DWORD BlockOffset return BlockOffset + (IntraBlockOffset << l) +1; } DWORD

SwizzleYUY2V (DWORD x, DWORD y, const SURFACEDESC & sd) In this regard, for the peseudocode accompanying the arrangement of Figs. 5A, 5B, 6A and 6B, surface pitch is defined as the number of bytes per scanned line of 128-bit blocks.

With respect to the compressed plane (12bpp), this surface shape sub-samples U and V by a factor of 2 horizontally and vertically. The luminance and chromaticity samples are placed on two separate parts of the surface. The linear layout of the compressed plane (12bpp) is shown in Fig. 7A. Surface pitch is defined as the number of bytes per scanned line of 512-bit blocks in the Y plane. The UV plane pitch is half the Y plane pitch because it has V as much as a sample, but twice as much as color elements. per sample. The resulting arranged image format according to the invention is shown in Fig. 7B.

The exemplary pseudocode for the arrangement function of the invention for this format which translates the (x, y) coordinates of compilations for elements Y, U and V is as follows: DWORD

SwizzlePP12Y (DWORD x, DWORD y, const SURFACEDESC & sd) assert (x <sd.Width);

Dword

SwizzlePP12U (DWORD x, DWORD y, const SURFACEDESC & sd) DWORD

SwizzlePP12V (DWORD x, DWORD y, const SURFACEDESC & sd) THIRD SECURITY LAYER - ENCRYPTED TEMPORARY STORAGE

The facilities of the above described embodiments with respect to the first and second security layers may be extended in accordance with the invention to encrypt the temporary command stores stored on the GPU 265 in addition to the image data on which the GPU 265 is operating. This feature is desirable if application 210 wishes to protect content that is sent to the hardware in-band in command temporary storage. Fig. 9A shows the video deco-diffusion using an encrypted command temporary storage 900, whereby the content is released to the encrypted texture 400a and is decrypted through the decryption component 450 and decoded by the video decoder 430b. Although it is possible for command temporary storage to be encrypted, content is encrypted in video memory as well as in command temporary storage, as shown by the encrypted decoded structure 420a. Encrypting command temporary storage is thus appropriate in situations like this, where macroblocks are in video memory and with motion vectors and other commands sent in the command stream. Atomic coercion for encrypted texture data also applies to encrypted command staging data, while preventing color staging encryption may not be sufficient to protect the content in question. Intermediate temporary stores, for example, temporary Z storage, can also be encrypted to protect the system against plain text attacks. Fig. 9B shows exemplary 3D transmission using encrypted command temporary storage according to the invention. As illustrated, the 3d 810 broadcast commands are encrypted on the route to the 430c video decoder. Texture data 400a is decrypted via decryption component 450 and processed according to commands 810 via video decoder 430c. Incident to this, data in 820 color temporary storage is encrypted by the encryption / decryption component 830. Spoofing detection can be done before command temporary storage is consumed using two steps, or after temporary storage of command has been consumed. In one embodiment, counterfeit detection is enabled after viewing or transmitting the content.

ADDITIONAL ALTERNATIVE MODES - GRAPHIC OUTPUT ENCRYPTION

In each of the above embodiments, although reliability and integrity have been demonstrated and described with respect to the dotted line portion of Fig. 2, reliability and integrity have not been demonstrated with respect to video output, ie theoretically the interface between the graphics card and the transmitting device such as a monitor and / or transmitting device itself is subject to an attack.

Thus, in the above described embodiments, as shown in Fig. 9A, at some point during the process, although the content is protected in video memory and during graphics card processing, the data is clearly sent to the DVI / DAC 290. . As a result, data may be pirated, or altered on the route to the transmitting device, and while within the transmitting device.

Thus, according to an alternative embodiment of the invention which may optionally be combined with other embodiments described herein, a sister cryptoprocessor 220b is provided in the transmission device to complement the functionality performed by cryptoprocessor 220a. In this regard, cryptoprocessor communicatively coupled cryptographic component 910a 220a encrypts the data prior to release to the DVI / DAC 290 component and cryptoprocessor 220b communicatively coupled decryption component 910b decrypts the data as part of the display or transmission that occurs, preventing data piracy. Encryption component 910a can be included alternately with DVI / DAC 290 component. In short, by applying the same encryption and decryption and fundamental administration techniques, content can be protected over the entire secure graphics channel to cryptographically protect the secure release and processing of content.

As mentioned above, while exemplary embodiments of the present invention have been described with respect to various computing devices, hardware, software and network architectures, the underlying concepts may be applied to any computing device or system where it is desirable to protect the computer. content from a trusted source. Thus, techniques for cryptographically protecting security content in accordance with the present invention may be applied in a variety of applications and devices. For example, methods for cryptographically protecting the security content of the invention may be applied to the operating system of a computing device, provided as a separate object on the device, as part of another object, as an object capable of being downloaded from a server. , such as a distributed object, etc. Although exemplary programming languages, peseudocode, names, and examples have been selected herein as representative of various choices, the languages, peseudocode, names, and examples are not intended to be limiting.

The various techniques described herein may be implemented with respect to hardware or software or, where appropriate, a combination of both. Thus, the methods and pairings of the present invention, or certain aspects or parts thereof, may take the form of program code (i.e. instructions) embodied in tangible media such as floppy disks, CD-ROMs, hard disks, or any other machine readable storage medium, where, when program code is loaded and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the invention. In the case of program code execution on programmable computers, the computing device will generally include a processor, processor readable storage medium (including volatile and non-volatile memory and / or storage elements), at least one input device , and at least one output device. One or more programs that may utilize the techniques of the present invention, for example, through the use of a data processing API, operating system, reliable application or others, are preferably implemented in a procedural or object oriented programming language. high level to communicate with a computer system. However, the program (s) can be implemented in the assembly or machine language if desired. In either case, the language may be compiled or interpreted language, and in various embodiments of the invention, conditions imposed on hardware implementations of the GPU 265.

Also the methods and apparatus of the present invention may be practiced by communications incorporated in the form of program code which is transmitted on some transmission medium, such as in electrical installation or cabling, by optical fibers or any other form of transmission, wherein, when program code is received and loaded and executed by a machine, such as an EPROM, a portal array, a programmable logic device (PLD), a client computer, a video recorder or others, or a receiving machine having a Graphics card and encryption capabilities as described in the exemplary embodiments above becomes an apparatus for practicing the invention. When implemented in a general purpose processor, the program code combines with the processor to provide a single apparatus that operates to invoke the functionality of the present invention. Additionally, any storage technique used in connection with the present invention can invariably be a combination of hardware and software.

While the present invention has been described with respect to the preferred embodiments of the various figures, it is to be understood that other similar embodiments may be used or modifications and additions may be made to the embodiment described to perform the same function of the present invention without departing from it. For example, while exemplary network environments of the invention are described in the context of a networked environment, such as unhierarchical networked environment, one skilled in the art will recognize that the present invention is not limited thereto, and that the methods as described in this patent application may apply to any computing device or environment, such as a gaming console, handheld, laptop, etc., wired or wireless, and may be applied to any of several such connected computing devices. by a communications network, and interacting along the network. In addition, it should be emphasized that a variety of computer platforms, including handheld operating systems and other application-specific operating systems are contemplated, especially as the number of wireless network devices continues to proliferate. Still further, the present invention may be implemented within and across a plurality of chips or co-processing devices, such as a device having a plurality of GPUs, and storage may be performed similarly across a plurality of devices. Therefore, the present invention should not be limited to any embodiment only, but otherwise should be interpreted in scope and scope according to the appended claims.

APPENDIX A. FLOW CHANNELS VERSUS BLOCK CHANNELS

This Appendix details the differences between flow ciphers and block ciphers as they relate to the contents of this document.

As a rule, stream ciphers are faster and simpler to implement than block ciphers.

As the name implies, stream ciphers encrypt and decipher a byte stream. To decrypt the Nth byte in the stream, ciphering starts at the beginning and advances one byte at a time to the desired offset within the stream.

In contrast, block ciphers operating in electronic codebook mode can encrypt or decrypt arbitrary blocks in the data, but must encrypt / decrypt a complete block at a time. A typical block size is 16 bytes.

Stream ciphers are used in such a way that the same data is never encrypted twice, ie the key used for encryption and decryption must often be changed. When used for premium video playback, for example, changing keys once per frame is sufficient.

As a final note, there are good quality block ciphers in the public domain.

Claims (41)

1. Method for cryptographically protecting security content with respect to a computing device's reliable graphics system, the reliable graphics system having video memory, at least one graphics processing unit (GPU) and a cryptographic processing device communicatively coupled to at least one GPU, characterized in that it comprises: requesting, by one of an application and a device, the graphics system to perform the processing and rendering of security content, wherein the request includes transmitting by one of an application. application and device, a session key for the graphics system and transmitting security content to at least an encrypted portion of the video memory; decrypting the contents of at least one encrypted portion of video memory by at least one GPU in communication with a cryptographic processing device; perform one of the rendering and rendering on decrypted content for at least one GPU; and output content from at least one GPU. Method according to claim 1, characterized in that: the decrypting step is performed by a decryption mechanism of an input unit of at least one GPU, and further comprising encrypting the content with an encryption mechanism. ia / decrypt an output unit of at least one GPU. 3. Computer readable medium having stored in it a plurality of computer executable instructions characterized in that the plurality of instructions includes: requesting, by one of an application and a device, the graphics system to perform the processing and rendering of the security content, wherein the request includes transmitting for one of an application and device a session key to the graphics system and transmitting security content to at least an encrypted portion of video memory; decrypting the contents of at least one encrypted portion of video memory by at least one GPU in communication with a cryptographic processing device; perform one of the rendering and rendering on decrypted content by at least one GPU; and output content from at least one GPU. 4. A coprocessing device, characterized in that the computer executable modules of the at least one computer readable medium comprise: a device for requesting by one of an application and device, the graphics system for executing one of the processing and rendering of the security content, wherein the device to request includes a device for transmitting between one application and one device, a session key for the graphics system, and a device for transmitting security content to at least one encrypted portion of the memory. video; a device for decrypting the contents of at least one encrypted portion of video memory by at least one GPU in communication with the cryptographic processing device; a device for performing one of the rendering and rendering of content decrypted by at least one GPU; and a device for outputting the content of at least one GPU. A computing device comprising a device for cryptographically protecting security content with respect to a reliable graphics system of a computing device, the reliable graphics system having video memory, at least one graphics processing unit (GPU) and a a cryptographic processing device communicatively coupled to at least one GPU, characterized in that it comprises: a device for requesting one of an application and a device, the graphics system for performing one of processing and rendering of security content, in whereas a requesting device includes a device for transmitting from one application and device, a session key for the graphics system and a device for transmitting security content to at least one encrypted portion of video memory; a device for decrypting the contents of at least one encrypted portion of video memory by at least one GPU in communication with a cryptographic processing device; a device for performing one of the processing and reridering on decrypted content by at least one GPU; and a device for outputting the content of at least one GPU. Computing device according to claim 5, characterized in that if the output of the output device is different from the security content of the requesting device by adjusting for any processing performed on the security content by at least one GPU , one of an application and device is alerted of the difference. A computing device according to claim 5, characterized in that the transmitting device includes a device for transmitting the security content to at least one encrypted overlay surface overlapping at least one primary surface of said memory. video. A computing device according to claim 5, characterized in that the device for decrypting the contents of at least one encrypted portion of video memory includes a device for decrypting a geometric fraction of a primary surface, where pixels different from the geometric fraction are not decrypted. A computing device according to claim 5, characterized in that the cryptographic processor is permanently attached to the graphics card by one of (A) adding the cryptographic processor to an existing chip and (B) adding the cryptographic processor as a separate chip to the graphics card, where the physical connection between the cryptographic processor and the rest of the graphics card is not accessible and not exposed. Computing device according to claim 7, characterized in that the decrypting device includes a device for decrypting at least one overlapping surface encrypted by a GPU decryption mechanism communicatively coupled to a cryptographic processing device. A computing device according to claim 7, characterized in that the decryption device includes one of (A) a device for decrypting at least one digitally to analog conversion hardware moving encrypted overlay surface (DAC). ) of the graphics system as content is output according to the output device output and (B) a device for decrypting at least one moving encrypted overlay surface just before the content reaches the graphics system DAC hardware . A computing device according to claim 7, characterized in that the decrypting device includes a device for decrypting at least one encrypted overlay surface before the content reaches the graphics system DAC hardware by a component having no no return channels to the host system. A computing device according to claim 5, further comprising: a device for re-encrypting content by at least one GPU in communication with the cryptographic processing device prior to output through the output device ; and a device for decrypting re-encrypted content by at least a second cryptographic processing device of an external computing device. Computing device according to claim 5, characterized in that the content is transmitted in digital form to an external device having a second cryptographic processing device and decryption of the decryption device takes place on an external device. Computing device according to claim 13, characterized in that the external computing device is one of (A) a monitor, (B) a decoder apparatus and (C) a signal processing rendering device. digital (DSP). A computing device according to claim 7, characterized in that the transmitting device includes a device for transmitting security content to one of (A) a first encrypted confidential overlay for basic rendering of security content and (B) a second encrypted protected overlay specifically designed to present sensitive user interfaces, (C) a first encrypted region of a primary surface for basic rendering of security content, and (D) a second encrypted region of a primary surface specifically designed for present sensitive user interfaces. A computing device according to claim 5, characterized in that the decryption device includes a device for calculating a cryptographic compilation of the decrypted data, and the computing device further includes: a device for transmitting the cryptographic compilation to one of an application and device to ensure that the displayed pixels are the pixels sent against the request by one of an application and a device by a request device. A computing device according to claim 16, characterized in that the second encrypted protected overlay is always on top and not obscurable and wherein the contents of the second encrypted protected overlay are scanned by at least one GPU. Computing device according to claim 16, characterized in that a decryption device includes (A) a decryption device with a first stream cipher decryption component containing the contents of the first encrypted confidential overlay, (B) a device for decrypting with a second stream cipher decryption component the contents of the second encrypted protected overlay, (C) a device for decrypting with a first stream cipher decryption component the contents of the first surface encrypted region and (D) a device for decrypting with a second stream cipher decryption component the contents of the second primary surface encrypted region. A computing device according to claim 19, characterized in that at least one bit of each pixel on a primary surface is used to determine the ratio on a virtual protected surface to the pixel, wherein the card Graphic Selects an appropriate decryption key for the pixel based on at least one bit. Computing device according to claim 20, characterized in that if at least one bit contains a zero value, then the virtual protected surface associated with at least one bit is interpreted as a non-decrypting region. A computing device according to claim 19, further including a device for selecting, once decrypted pixel values are available, by a pixel selection component of at least one GPU, the pixel value of one of (A) the second encrypted protected overlay, (B) the first encrypted confidential overlay, and (C) a primary surface. A computing device according to claim 16, characterized in that the request for the device to request includes at least one of (A) a source and destination junction box of at least one encrypted overlay surface, ( B) a destination color key of at least one encrypted overlay surface, (C) in the case of the first encrypted confidential overlay, a specification of an encryption key index of the contents of an overlay return buffer for which (D) in the case of the second encrypted protected overlay, a specification of a memory location where at least one of a cyclic redundancy code (CRC), an integrity measure and a compilation value of the contents of decrypted overlays are to be written, (E) a source and destination junction box of at least one surface p (F) a destination color key of said at least one encrypted primary surface. Computing device according to claim 23, characterized in that one of an application and a device calculates at least one of the CRC, integrity measure and compilation value if one of an application and a device is related to the integrity of the content. A computing device according to claim 5, characterized in that at least one command temporary storage sent to a video decoding unit of at least one GPU incident to the request of the requesting device is encrypted by at least one of an application and a device and decrypted through the video decoding unit in communication with the cryptographic processing unit. A computing device according to claim 25, further comprising a device for detecting the tampering of at least one command temporary storage one of (A) using two passages prior to the consumption of the at least one storage. command buffer and (B) after command temporary storage has been consumed. A computing device according to claim 26, characterized in that: the device for decrypting the content of at least one encrypted portion of video memory comprises a decryption mechanism of an input unit of at least a GPU, further comprising: a device for encrypting content with an encryption / decryption mechanism of an output unit of at least one GPU. Computing device according to claim 27, characterized in that the input unit is a texture mapping unit and the output unit is an alpha combination unit, and at least one encrypted part of the Video memory is an encrypted texture surface. A computing device according to claim 27, characterized in that the security content is one of texture data, plain text, and video macroblocations. A computing device according to claim 27, characterized in that the encryption device and a decryption device include a encryption device and a decryption device using block ciphers, respectively. A computing device according to claim 28, characterized in that the decryption mechanism of the decrypt-texture texture mapping unit in a cache line filler and the encryption / decryption mechanism of the Alpha combination encrypts before writing. A computing device according to claim 28, further including a device for decrypting by the alpha combination unit encryption / decryption mechanism when reading a cache line from a color buffer in Video memory. A computing device according to claim 27, further comprising: a device for switching encrypted output content from an encrypted primary surface posterior to a front encrypted primary surface of video memory; a second device for decrypting encrypted output content by a second decryption mechanism of at least one GPU in communication with the cryptographic processing device; and a second device for outputting output content. A computing device according to claim 27, characterized in that the output device includes a device for outputting the encrypted content to a confidential overlapping alternation chain and the computer executable modules additionally include: a device for switching encrypted output content from a later encrypted confidential surface to a front encrypted confidential surface, where encryption via the encryption / decryption mechanism includes a device for encrypting using stream cipher encryption; and a second device for decrypting the encrypted output content by a stream cipher decryption mechanism of at least one GPU in communication with the cryptographic processing device. A computing device according to claim 34, further comprising: a device for encoding a location within the content prior to encryption through the device for encrypting; and a device for decoding the location within content after decryption by the second decrypting device, where the location is externally unavailable preserving the integrity of a plaintext ciphertext mapping. Computing device according to claim 27, characterized in that if the output of the output device is different from the security content of a device to request adjustment for any processing performed on the security content by at least one GPU, one of an application and a device is alerted of the difference. A computing device according to claim 27, characterized in that the computer executable modules additionally include: a device for re-encrypting content by at least one GPU in communication with the cryptographic processing device prior to output via output device; and a device for decrypting re-encrypted content by at least a second cryptographic processing device of an external computing device. Computing device according to claim 37, characterized in that an external computing device is one of (A) a monitor, (B) a decoder apparatus and (C) a signal processing rendering device. digital (DSP). Computing device according to claim 27, characterized in that the encrypted textures and encrypted black screen surfaces released by one of an application and a device are encoded with block ciphers, and one of an application and device arranges block ciphers with a predefined array format that converts a location (x, y) in the content to an offset to at least one of YUV, RGB, YUY2, and compressed planar formats. A computing device according to claim 27, characterized in that at least one command temporary storage sent to a video decoding unit of at least one GPU incident to the request of the requesting device is encrypted by at least one of an application and device is decrypted via the video decoding unit in communication with the cryptographic processing unit. A computing device according to claim 27, characterized in that the computer executable modules further comprise a device for further detecting counterfeit of at least one command buffer one of (A) using two passages prior to consumption. at least one command temporary storage and (B) after the command temporary storage has been consumed.