In an unsecured computer environment, a computer application may access any available computing resources with little or no consideration given to whether those resources are secure. There are many reasons, however, that it is desirable to control access to computing resources.
The Trusted Computing Group (TCG) was formed and has adopted an industry standard specification to enhance the security of computing environments. The goal is to deliver an enhanced hardware and operating system (OS)-based trusted computing platform (TCP) for customers to run their applications. With regard to hardware considerations, a Trusted Platform Module (TPM) has been introduced which includes a micro-controller that stores security information. The TPM is the root of trust to create a secured environment that enables the OS and applications to fight against software attacks. TCG requires the TPM identification to be unique and to physically bind to a specific platform such that it can not be easily removed or transferred to another platform. Furthermore, the TPM must show evidence of physical tampering upon inspection.
Manufacturing platforms with the TPM increases the manufacturing costs, in addition, some countries (e.g., Russia and China) do not permit products to be shipped with security devices such as TPM. Accordingly, separate platforms without the TPM need to be manufactured and tracked (e.g., using unique SKU numbers) to be sold in these markets, thereby further increasing costs.
a is a perspective view of the exemplary tamper-evident connector in
b is a perspective view of the exemplary tamper-evident connector in
a is a perspective view of the exemplary tamper-evident connector in
b is a perspective view of the exemplary tamper-evident connector in
Briefly, embodiments of a tamper-evident connector are disclosed. The designs enable the TPM to be manufactured separately as an optional component, thereby reducing the cost of manufacturing separate system boards for different markets, while still meeting the TCG physical binding requirement (i.e., there is visible evidence of tampering if the TPM is removed). After removal, a malformed TPM likely cannot be reused (or is difficult to reuse) in another system thereby maintaining the integrity of the trusted software environment (TSE) if the TPM has already been compromised. However, the removal process does not affect the system board, thereby allowing an authorized administrator to replace the TPM module on the system board if needed.
Although the systems and methods described herein help to enable security measures for running trusted software and accessing trusted resources, it is noted that application of the tamper-evident connector is not limited to computer security. Still other applications of the tamper-evident connector will be readily apparent to those having ordinary skill in the art after becoming familiar with the teachings herein.
It is noted that computing platform 100 may operate as a stand-alone device and/or may operate in a networked computing environment using logical connections to one or more remote resources (not shown). The logical connections may include a local area network (LAN) and/or a wide area network (WAN). Exemplary remote resources include, but are not limited to, a personal computer, a server, a router, a network PC, and a peer device or other network node. Remote resources may include many or all of the elements described for the computing platform 100, such as, e.g., processing capability and memory,
Computing platform 100 may also include one or more resources 130a-c. As used herein, the term “resource” includes any of a wide variety of different types of devices (e.g., PCIe devices) and/or functions (e.g., provided by the device). In an exemplary embodiment, resources 130a-c may be communicatively coupled to the computing platform 100 via one or more peripheral component interconnect (PCI) links 140a-b implementing the PCI-express (PCIe) specification. In such an embodiment, the resources 130a-c may be connected directly to the root complex 150 via one or more PCIe cards 145a-c.
A host bridge and memory controller hub, also referred to generally as a root complex 150, couples the various system components to the processing unit 110. The root complex 150 is a subsystem which detects and initializes resources 130a-c, and manages the links 140a-c so that processor 110 can read/write to the resources 130a-c and/or otherwise control the resources 130a-c.
Computing platform 100 may operate in a protected or trusted operating environment. A trusted operating environment is a protected or secured environment for running trusted software and accessing trusted devices. Trusted software is software that has a reliably established notion of identity, e.g., indicating that the software is from a trusted source. A trusted device is a device accessible via a Trusted Configuration Access Mechanism (TCAM) 160. It is noted that there may he single or multiple TCAMs for each computing platform 100 (or for each partition on a computing platform).
The TCAM 160 is patterned after the Enhanced Configuration Access Mechanism (ECAM) provided for the standard configuration space defined by the PCIe specification (e.g., the ECAM 340 in
The TPM 165 provides protected storage, protected functions, authentication of the computing platform 100, measurement of platform integrity, and attestation of platform integrity. The TPM 165 may be implemented to assert a hardware signal that enables a TCAM 160 for use only if/when the platform integrity has been attested. The PCIe specification defines the TCAM, which then allows access to the trusted configuration registers via memory mapped address space, e.g., in memory 120.
The TPM 165 may be physically attached to the system board 105 by a tamper-evident connector. The tamper-evident connector provides visible evidence of tampering if the TPM 165 is removed from the system board 105 (e.g., in accordance with the TCG physical binding requirement). These and other features will be better understood by the description of exemplary embodiments of the tamper evident connector provided below with reference to
When the rivet 200 is used in a secure computing environment, an electrical connector 230 may be mounted adjacent the pin 210 on a first component (e.g., TPM 240), and a second electrical connector 235 may be mounted adjacent the housing member 220 on a second component (e.g., system board 250). In an exemplary embodiment, the first electrical connector 230 and second electrical connector 235 may be commercially available 20-pin (or any number pin) mating electrical connectors, in any event, the electrical connectors 230 and 235 can be pushed together to form an electrical connection between the TPM 240 and the system board 250, e.g., for transferring security information from the TPM 240 to the system board 250.
Before continuing, it is noted that although shown as separate parts, the pin 210 and housing member 220 may be manufactured as a single part having the functionality of both pin 210 and housing member 220. For example, the rivet 200 may be manufactured so that, it can be shipped with the pin 210 loosely connected to the housing member 220 so that the parts are less likely to get misplaced or otherwise lost. In addition, the electrical connectors 230 and 235 may also be integrated into the rivet 200 and do not need to be provided separately.
a is a perspective view of the exemplary tamper-evident connector in
The housing member 220 may be fit into an opening 252 formed in the system board 250. For example, slots 226 in the expandable portion 224 of the housing member 220 enable the housing member 220 to reduce in size (e.g., a smaller diameter) when it is squeezed to fit through the opening 252. A spring-action naturally returns the expandable portion 224 to a widened state within the opening 252 to at least partially hold the housing member 220 in the system board 250.
When the body portion 214 of the pin 210 slides into the expandable portion 224 of the housing member 220, the presence of pin 210 forces the expandable portion 224 of the housing member 210 to further widen within the opening 252. Optionally, the pin 210 may be wider (or may include “fins” or other devices) at the end to enhance forcing the expandable portion 224 open. This widening action physically, and irreversibly, secures the TPM 240 to the system board 250.
b is a perspective view of the exemplary tamper-evident connector in
The male block structure 310 includes at least one foldable pin (and a plurality of foldable pins 315a-c are shown in
a is a perspective view of the exemplary tamper-evident connector in
b is a perspective view of the exemplary tamper-evident connector in
It is noted that with regard to any of the embodiments of the tamper-evident connector described above, TPM installation (the initial binding process) may be performed by the system integrator during manufacturing by the original design manufacturer (ODM) or at customer sites. The use of tools is not necessary for the initial binding process, making the tamper-evident connector easy to use.
After removal, a malformed TPM likely cannot be reused (or is difficult to reuse) in another system thereby maintaining the integrity of the trusted software environment (TSE) if the TPM has already been compromised. However, the removal process does not affect, the system board, thereby allowing an authorized administrator to replace the TPM module on the system board if needed, e.g., for servicing or replacement.
It is noted that the exemplary embodiments shown in the figures and discussed above are provided for purposes of illustration. In addition to the specific embodiments explicitly set forth herein, other aspects and embodiments will be apparent to those skilled in the art from consideration of the specification disclosed herein. It is intended that the specification and illustrated embodiments be considered as examples only.