Trusted cryptographic processor

Description of equivalent WO 2007006011 A2

TRUSTED CRYPTOGRAPHIC PROCESSOR

0002[0001] This application claims the benefit of US Patent Application No. 11/428,505 filed on July 3, 2006; which is a non-provisional of both US Provisional Application No. 60/697,071 filed on My 5, 2005, and US Provisional Application No. 60/697,072 filed on July 5, 2005; which are all assigned to the assigner hereof and hereby expressly incorporated by reference in their entirety for all purposes.

BACKGROUND

0004[0002] This disclosure relates in general to cryptographic processing and, but not by way of limitation, to programmable cryptographic processing.

0005[0003] Cryptographic systems are used to secure information. Information systems have advanced as we progress into the Information Age. Cryptographic systems have not kept pace. Only a single algorithm is supported along a single processing path to process items at the highest security levels.

0006[0004] New developments in cryptographic design often obsolete older systems. Cryptographic systems are inflexible and cannot incorporate new developments once fielded. Design of new cryptographic systems is expensive and time consuming. Often a new cryptographic system must be produced for each deployment to cover different classification levels and security issues.

0007[0005] In modern cryptosystems, there is a need for multi-port (multi-channel) operation, where one cryptosystem can support multiple interfaces on both the plaintext and ciphertext interfaces. Current cryptosystems are designed in an unscalable architecture such that ports are added with a linear rise in circuit size and/or complexity. For more complex cryptographic systems, multiple paths at multiple classifications may also be used. Each path may have a separate cryptographic device, for example. Interfacing various devices make for a complex system. Each different cryptographic device may be different or configured differently to support complex data transport paths.

0008[0006] In high-assurance applications such as cryptosystems, there is typically a need to have redundant functions operating in parallel and continuously monitored to ensure correct operations. This monitoring can be particularly problematic when multiple microprocessors need to operate in a synchronized but independent manner. Regardless of whether the microprocessors share the same clock or have independent clocks, the microprocessors must respond to asynchronous events such as interrupts. Because of the asynchronous environment, the processors may execute instructions out of order from time to time, even when they are executing the same code base. This can result in different outputs from the microprocessors causing external monitoring functions to detect a mismatch and suspend operations. High assurance design principles dictate certain levels of functional and physical separation. The design issue arises because redundant data processing elements must always be ensured of processing the same information in the same order with the same results.

0009[0007] In a secure system, there is often a need to have data path reconfiguration for different system operations. In a high-assurance secure system, this reconfiguration function is typically established by the same redundant system elements that perform the primary functions. Both these types of processes must also be monitored to ensure correct operations. This monitoring can be particularly problematic, for example, when requests for data path reconfiguration occur asynchronously to the redundant decision making logic. Because of the asynchronotis environment, the redundant decision making logic may occasionally come to different outcomes and the monitoring logic needs to provide a recovery mechanism to re- arbitrate for the correct data path before the data path is reconfigured.

0010SUMMARY [0008] In one embodiment, a cryptographic processor for redundantly-processing cryptographic operations is disclosed. The cryptographic processor includes a number of input ports, a first and second cryptographic engines, comparison logic and a plurality of output ports. The number of input ports is configured to accept both plaintext and ciphertext. Each of the number of input ports is coupled to both the first and second cryptographic engines. The comparison logic is configured to determine if the first and second cryptographic engines produce a result that is different. The number of output ports is configured to produce both plaintext and ciphertext, as required.

0011[0009] Further areas of applicability of the present disclosure will become apparent from the detailed description provided hereinafter. It should be understood that the detailed description and specific examples, while indicating various embodiments, are intended for purposes of illustration only and are not intended to necessarily limit the scope of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS [0010] The present disclosure is described in conjunction with the appended figures:

0012[0011] FIG. 1 depicts a block diagram of a prior art high-assurance system that has multiple ports and operates redundantly;

0013[0012] FIG. 2 depicts a block diagram of an embodiment of a multi-port cryptographic system;

0014[0013] FIG. 3 depicts a block diagram of another embodiment of the multi-port cryptographic system;

0015[0014] FIG. 4 depicts a block diagram of an embodiment of a bi-directional single-channel cryptographic system;

0016[0015] FIG. 5 depicts a block diagram of yet another embodiment of the multi-port cryptographic system;

0017[0016] FIG. 6 illustrates a flowchart of an embodiment of a process for processing packets with the cryptographic system; and

0018[0017] FIG. 7 illustrates a flowchart of an embodiment of a process foj arbitrating ports for the cryptographic system.

0019[0018] In the appended figures, similar components and/or features may have the same reference label. Further, various components of the same type may be distinguished by following the reference label by a dash and a second label that distinguishes among the similar components. If only the first reference label is used in the specification, the description is applicable to any one of the similar components having the same first reference label irrespective of the second reference label.

DETAILED DESCRIPTION

0021[0019] The ensuing description provides preferred exemplary embodiment(s) only, and is not intended to limit the scope, applicability or configuration of the disclosure. Rather, the ensuing description of the preferred exemplary embodiment(s) will provide those skilled in the art with an enabling description for implementing a preferred exemplary embodiment. It being understood that various changes may be made in the function and arrangement of elements without departing from the spirit and scope as set forth in the appended claims. [0020] Referring initially to FIG. 1, a block diagram of a prior art high- assurance system 100 is shown that has multiple ports and operates redundantly. Here, each channel is handled by one processing element 104 and each processing element 104 has a dedicated redundant processing element 106. Each channel has a dedicated comparison logic circuit 108 that determines if the outputs of the processing element 104 and redundant processing element 106 match and, if so, forwards the answer onto the output circuit (not shown). In this configuration of n channels, each additional channel increases the size of the resulting circuitry in a linear fashion.

0022[0021] With reference to FIG. 2, a block diagram of an embodiment of a multi-port cryptographic system 200 is shown. This embodiment re-uses resources for a multi-port cryptosystem 200. Here, a single channel processing element 204 and a redundant processing element 206 are shared by all the input channels 1 through n. Each channel has its own comparison logic that is controlled by a task switching controller (not shown).

0023[0022] The embodiment of FIG. 2 has improved efficiency over the prior art in FIG. 1. The single redundant processing element 204, 206 can handle multiple channels. The single redundant processing element 204, 206 switches among multiple channels. Additionally, the redundant processing element 204, 206 is reconfigurable to use different algorithms for each channel. Where different ports are to be serviced with different algorithms, the redundant processing element 204, 206 reconfigures itself for each channel. Each output port (1 through m) in this embodiment has its own comparison logic 208 to detect any errors in the processing element 204 or redundant processing element 206.

0024[0023] FIG. 3 illustrates a block diagram of another embodiment of a multi-port cryptographic system 300. This embodiment shows redundant arbitration logic 312, 314 that multiplexes 316 input channels and demultiplexes 320 output channels. There are n input channels and m output channels in this embodiment (e.g., n=A and m=4). Multiple input ports vie for the single redundant processing elements 204, 206. The processing elements direct their results to one of m outputs under the direction of the redundant arbitration logic 312, 314. The outputs of the redundant processing element 204, 206 are matched with comparison logic 208 to ensure consistent processing. Any mismatch of the data path processing from the processing element 204 and the redundant processing element 206 would generate an alarm. Additionally, error in the redundant arbitration logic 312, 314 would trigger an alarm. The allowable combinations of input ports and their associated outputs can be restrained through the use of the arbitration logic 312, 314. Both redundant arbitration logic 312, 314 open the same port before it is usable, implementing redundancy in this manner. For example, if the arbitration logic 312 opens a first port and the redundant arbitration logic 314 opens a second port, neither will open until the both arbitration logic 312, 314 are in agreement.

0025[0024] The architecture of FIG. 3 can be collapsed in one embodiment to be used in a single channel system that has a single bi-directional plaintext (PT) interface, and a single bidirectional ciphertext (CT) interface. Referring next to FIG. 4, a bi-directional single- channel cryptographic system 400 of this type is shown. Each interface 450, 454 is separated from the cryptographic engines 404 with input and output ports 428, 424 that are individually controlled according to the direction that the data is intended to be routed through the redundant cryptographic engines 404-1 and 404-2. Each interface is bidirectional to include both an input port 428 and an output port 424. The cryptographic engines 404 are programmably capable encryption, decryption, filtering, guarding, hashing, signing, and bypass for each packet. The type of processing and ports 424, 428 used can be preconfigured for a particular path from input port 428 to output port 424 and/or configured on a packet-by- packet basis.

0026[0025] With reference to FIG. 5, a block diagram of an embodiment a multi-port cryptographic system 500 is shown that has bi-directional interfaces. This allows for multiple, physically separate PT or CT interfaces 454, 450 to utilize a single redundant cryptographic engine 404. The number of additional PT or CT interfaces 454, 450 is limited only by the layout of the circuitry. This embodiment has two PT interfaces 454 and one CT interface 450. Additionally, the arbitration logic 312, 314 is shown. Redundancy is used throughout this specification to refer to multiple circuits that generally perform the same function, but may implement that function in the same or different ways. A redundant element mirrors another element to provide a way to check that the mirrored element is behaving properly and vice versa. A comparison is generally performed that may allow for differences in the time the result is produced by the redundant element.

0027[0026] Each cryptographic engine 404 has a dedicated input data bus from the input ports 428. When one of the interfaces 450, 454 has data to be routed through the cryptographic engine 404, a request is sent to redundant port arbitration logic 312, 314 to open the appropriate input port 428. For example, if PT interface one 454-1 has a data packet for the cryptographic engine 404, the PT interface 1 454-1 will request redundant input port 428-1, 428-2 to be opened. The arbitration logic 312, 314 ensures that only one input port 428 is open at a time, and that only one output port 424 is opened at a time. When the cryptographic system 500 receives the data packet, a determination is made to which output port 424 the data packet should be routed to and a request is sent to the arbitration logic 312, 314 to open the appropriate output port 424. This exemplary embodiment employs redundant comparison logic 208, 210 to compare the redundant data packets as the packets leave the cryptographic engines 404 to ensure that both cryptographic engines 404 produce the same output. The data packets are then out the open output port 424. When the entire data packet has been sent out of the output port 424, the arbitration logic 312, 314 is instructed to close all of the output ports 424. At this time, the cryptographic system 500 is ready to receive additional data packets that would cause the arbitration logic 312, 314 to configure the ports 424, 428 appropriately.

0028[0027] The cryptographic system 500 architecture allows multiple interfaces 450, 454 on either side of the red/black boundary to be added by simply adding a set of input and/or output ports 428, 424, and routing the appropriate control signals to the arbitration logic 312, 314. The arbitration logic 312, 314 and redundant cryptographic engine 404 are configured to use any new ports 428, 424 without additional arbitration logic 312, 314 and cryptographic engines 404. Some embodiments allow reprogramming ports 428, 424 to allow changing CT interfaces 450 to PT interfaces 454 and vice-versa. La one embodiment, this can be done during normal operation, while other embodiments allow reassignment only during system configuration.

0029[0028] The arbitration logic 312, 314 accepts requests from single or redundant elements and acts upon them based upon some pre-established priority. If asynchronous requests result in mis-synchronization of port selection into the data path, the cryptographic system 500 forces the arbitration logic 312, 314 to re-arbitrate. This process would continue until two matching tasks are established. When the port selections do not agree for points leaving the data path, an alarm is generated. One advantage to this embodiment is that decisions are made in a fully independent fashion and results independently compared. »

0030[0029] Referring next to FIG. 6, a flowchart of an embodiment of a process 600 for processing packets with the cryptographic system is illustrated. The depicted portion of the process 600 begins in block 608 where a PT or CT packet is received on the appropriate interface 450, 454. The arbitration logic 312, 314 is manipulated to configure the input port 428, the output port 424, and any multiplexing 316 and de-multiplexing 320 in block 612. The packet is redundantly processed in block 616. A task matching process may be used to keep the processing somewhat synchronized.

0031[0030] In block 620, the results are compared. Some embodiments have synchronizers to realign the results before comparison. A determination is made in block 624 to conclude if there is a match between the results. Where there are errors in the match process as determined in block 628, the process 600 ends. Where there is no error found, processing goes from block 628 to block 632 where the CT or PT packet is sent out of the cryptographic system.

0032[0031] Referring next to FIG. 7, a flowchart of an embodiment of a process 700 for arbitrating ports for the cryptographic system is illustrated. The depicted portion of the process 700 may begin in block 704 where paths through the cryptographic system may be preconfigured. For example, a particular input port 428 can be dedicated to a particular output port 424 along with particular cryptographic processing. Certain paths can be also configured to be disallowed, for example plaintext interface 454 to ciphertext interface 428 bypass could be made an invalid path. Codes or status lines for selecting particular paths could be configured. Some embodiments can work in preconfigured or configured on-the-fly modes. For on-the-fly mode, each packet can have a path configured dynamically.

0033[0032] rn block 708, the path for a packet is somehow conveyed to the arbitration logic 312, 314. This path could be preconfigured, conveyed in a metadata header, conveyed in a separate message, and/or signaled with a status line in various embodiments. The input redundant ports 428 indicated by the selected path are opened by the arbitration logic 314 in block 712. Additionally, the packet received on those redundant ports 428 may be checked in various ways to check for improper formatting or corruption. A packet that is received on redundant input ports 428 in a manner inconsistent with the selected path can be rejected.

0034[0033] In block 716, the packet is routed to the redundant cryptographic engines 404 for processing consistent with what was specified. The cryptographic engines 404 are aware of the path and request the output port in block 720. hi other embodiments, the arbitration logic 312, 314 configures the output port 424 without help from the cryptographic engine 404. Back to this embodiment in block 724 where the arbitration logic opens the output port requested by the cryptographic engine 404, if allowed. Certain paths are configured to be unallowable and the specified path indicates the output port. Should the arbitration logic 312, 314 find something inconsistent with the configuration or path is requested, nothing will be opened. Both redundant arbitration logic 312, 314 are in agreement before a particular output port 424 is opened.

0035[0034] The processed packet is transferred through the comparison logic 208, 210 in block 728 to the selected output port 424. In block 732, the arbitration logic 312, 314 closes the input and/or output ports 428, 424. This completes the processing of a particular packet. Control over the ports in a cryptographic system, which reuses cryptographic engines, for multiple packet streams is provided in this manner for one embodiment.

0036[0035] A number of variations and modifications of the disclosed embodiments can also be used. For example, some of the above embodiments discuss working with packetized information. This specification is also applicable to streams of information. Those streams can be packetized outside of the cryptographic system or within. In either event, the single redundant cryptographic engine or channel processing element can switch between processing for two or more input ports. The processed packets and be reassembled into a stream internal to the cryptographic system or outside. Some embodiments of the cryptographic system may accept packetized information and produce a stream or vice versa.

0037[0036] Specific details are given in the above description to provide a thorough understanding of the embodiments. However, it is understood that the embodiments may be practiced without these specific details. For example, circuits may be shown in block diagrams in order not to obscure the embodiments in unnecessary detail. In other instances, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary detail in order to avoid obscuring the embodiments.

0038[0037] Moreover, as disclosed herein, the term storage or machine-readable medium may represent one or more devices for storing data, including read only memory (ROM), random access memory (RAM), magnetic RAM, core memory, magnetic disk storage mediums, optical storage mediums, flash memory devices and/or other machine readable mediums for storing information. The term "machine-readable medium" includes, but is not limited to portable or fixed storage devices, optical storage devices, wireless channels, and/or various other mediums capable of storing, containing or carrying instruction(s) and/or data.

0039[0038] Furthermore, embodiments may be implemented by hardware, software, scripting languages, firmware, middleware, microcode, hardware description languages, and/or any combination thereof. When implemented in software, firmware, middleware, scripting language, and/or microcode, the program code or code segments to perform the necessary tasks may be stored in a machine readable medium such as a storage medium. A code segment or machine-executable instruction may represent a procedure, a function, a subprogram, a program, a routine, a subroutine, a module, a software package, a script, a class, or any combination of instructions, data structures, and/or program statements. A code segment may be coupled to another code segment or a hardware circuit by passing and/or receiving information, data, arguments, parameters, and/or memory contents. Information, arguments, parameters, data, etc. may be passed, forwarded, or transmitted via any suitable means including memory sharing, message passing, token passing, network transmission, etc.

0040[0039] Implementation of the techniques, blocks, steps and means described above may be done in various ways. For example, these techniques, blocks, steps and means may be implemented in hardware, software, or a combination thereof. For a hardware implementation, the processing units may be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described above, and/or a combination thereof.

0041[0040] For a software implementation, the techniques, processes and functions described herein may be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. The software codes may be stored in memory units and executed by processors. The memory unit may be implemented within the processor or external to the processor, in which case the memory unit can be communicatively coupled to the processor using various known techniques.

0042[0041] While the principles of the disclosure have been described above in connection with specific apparatuses and methods, it is to be clearly understood that this description is made only by way of example and not as limitation on the scope of the disclosure.