Standard Microsystems Corporation

Raphael Weiss

Alan David Berenbaum

Richard Wahler

Drew Jon Dutton

John D Virzi

Randy B Goldberg

A system including a serialized secondary bus architecture. The system may include an LPC bus, an I/O controller, a serialized secondary bus, and at least one slave device. The LPC bus may be connected to the I/O controller, and the at least one slave device may be connected to the I/O controller via the serialized secondary bus. The serialized secondary bus has a reduced pin count relative to the LPC bus. The I/O controller may receive bus transactions from the LPC bus. The I/O controller may translate and forward LPC bus transactions to the at least one device over the secondary bus. The I/O controller may include a processing unit. The processing unit may initiate bus transactions intended for the at least one slave device. The I/O controller may also include a bus arbitration unit. The bus arbitration unit may arbitrate ownership of the secondary bus between the processing unit and the LPC bus.

Serialized secondary bus architecture

System and method for authenticating data or program code via a configurable signature. Configuration information is retrieved from a protected first memory, e.g., an on-chip register, where the configuration information specifies a plurality of non-contiguous memory locations that store the signature, e.g., in an on-chip memory trailer. The signature is retrieved from the plurality of non-contiguous memory locations based on the configuration information, where the signature is useable to verify security for a system. The signature corresponds to specified data and/or program code stored in a second memory, e.g., in off-chip ROM. The specified data and/or program code may be copied from the second memory to a third memory, and a signature for the specified data and/or program code calculated based on the configuration information. The calculated signature may be compared with the retrieved signature to verify the specified data and/or program code.

Configurable signature for authenticating data or program code

System and method for implementing one time programmable (OTP) memory using embedded flash memory. A system-on-chip (SoC) includes a cleared flash memory array that includes an OTP block, including an OTP write inhibit field that is initially deasserted, a flash memory controller, and a controller. Data are written to the OTP block, including setting the OTP write inhibit field to signify prohibition of subsequent writes to the OTP block. The SoC is power cycled, and, in response, at least a portion of the OTP block is latched in a volatile memory, including asserting an OTP write inhibit bit based on the OTP write inhibit field, after which the OTP block is not writeable. In response to each subsequent power cycling, the controller is held in reset, the latching is performed, the controller is released from reset, and the flash array, now write protected, is configured to be controlled by the controller.

Implementation of one time programmable memory with embedded flash memory in a system-on-chip

In-band event polling mechanism. A master device may initiate a polling transaction to poll at least a subset of a plurality of slave devices for event information. In response to the polling transaction, at least one of the subset of slave devices may transmit event information to the master device. The event information may correspond to at least one of a plurality of asynchronous event types. If the event type associated with the received event information is an event notification for an embedded processor of the master device, the master device may forward the event information to the embedded processor. Otherwise, if the event type associated with the received event information is an event notification for a device external to the master device (e.g., a host processor), the master device may translate the event information to a protocol associated with the event type and forward the event information to the external device.

In-band event polling

System and method for performing pre-boot security verification in a system that includes a host processor and memory, an embedded microcontroller with an auxiliary memory, e.g., an on-chip ROM, or memory controlled to prohibit user-tampering with the contents of the memory, and one or more pre-boot security components coupled to the embedded microcontroller. Upon power-up, but before host processor boot-up, the embedded microcontroller accesses the auxiliary memory and executes the program instructions to verify system security using the one or more pre-boot security components. The one or more pre-boot security components includes at least one identity verification component, e.g., a smart card, or a biometric sensor, e.g., a fingerprint sensor, a retinal scanner, and/or a voiceprint sensor, etc., and/or at least one system verification component, e.g., TPM, to query the system for system state information, and verify that the system has not been compromised.

Enhancing security of a system via access by an embedded controller to a secure storage device

System and method for protecting data in a system including a main processor, an embedded controller, and a memory. In response to a power-on-reset (POR), access to the memory is enabled, e.g., access by the embedded controller. First data is read from the memory (e.g., by the embedded controller) in response to the enabling, where the first data are usable to perform security operations for the system prior to boot-up of the main processor. The first data are used, e.g., by the embedded controller, to perform one or more security operations for the system, then access to the memory, e.g., by the embedded controller, is disabled, where after the disabling the memory is not accessible, e.g., until the next POR initiates enablement.

Memory protection for embedded controllers

A power state broadcast mechanism. A master device may broadcast a message through the use of a protocol to each of one or more slave devices to inform the slave devices of the power state of a computer system. The broadcast message may include a protocol header indicating the start of the broadcast transaction, a function type parameter indicating the type of broadcast transaction, and power state data indicating the power state of the computer system. Each of the slave devices may read the protocol header to detect the start of a broadcast transaction, and the function type parameter to determine the type of broadcast transaction. If the function type parameter indicates a power state broadcast transaction, each of the slave devices may read the power state data included in the broadcast message and determine whether to adjust the current power state of the slave device.

Method, system, and apparatus for a plurality of slave devices determining whether to adjust their power state based on broadcasted power state data

An address assignment mechanism. A computer system may include one or more types of slave devices. Each slave device includes an internal device ID. Slave devices of the same type include the same internal device ID. The master device may broadcast a message through the use of a protocol to each of the slave devices to initiate an address assignment operation. Each of the slave devices determines whether the broadcast device ID included in the broadcast message matches the internal device ID associated with the slave device. If the broadcast device ID matches the internal device ID, the linear bus address included in the broadcast message is assigned to the slave device. The bit size of the linear bus address may be smaller than that of the broadcast device ID. After the address assignment operation, the master device may communicate with the slave device using the assigned linear bus address rather than the device ID.

Address assignment through device ID broadcast

A resistor/capacitor identification detection (RCID) circuit may provide system level identification of hardware (e.g. circuit board ID) through a single pin interface, by identifying up to a specified number of more than two quantized RC time constant states by measuring the discharge and charge times of an external RC circuit coupled to the single pin. The RCID circuit may initiate the discharge followed by a charging of the external RC circuit. The signal developed at the signal pin may be provided to the input of a threshold detector, with the threshold set at a specified percentage of a supply voltage used for operating the RCID circuit. The digitized output of the threshold detector may be used to gate a counter, after having been filtered through an input glitch rejection filter. A resolution of the counter may be determined by a high frequency clock used for clocking the counter. The numeric values of the charge and discharge times may be stored in data registers comprised in the RCID circuit.

Growing community of inventors

Plainview, NY, United States of America

Raphael Weiss