Abstract: Reading a value into a register, checking to see if the value is a NULL, and then jumping out of a loop if the value is a NULL is a common task that processors perform. To speed performance of such a task, a novel “blank bit” is added to the flag register of a processor. When a first instruction (arithmetic, logic or load) is executed, the instruction operands are checked to see if any is a NULL character value. Information on the result of the check is stored in the blank bit. Execution of a second instruction uses the information stored in the blank bit to determine whether or not a second operation (for example, a jump) will be performed. By using the first and second instructions in a loop, the number of instructions executed to check for NULLs at the end of strings and arrays is reduced.

Abstract: Reading a value into a register, checking to see if the value is a NULL, and then jumping out of a loop if the value is a NULL is a common task that processors perform. To speed performance of such a task, a novel “blank bit” is added to the flag register of a processor. When a first instruction (arithmetic, logic or load) is executed, the instruction operands are checked to see if any is a NULL character value. Information on the result of the check is stored in the blank bit. Execution of a second instruction uses the information stored in the blank bit to determine whether or not a second operation (for example, a jump) will be performed. By using the first and second instructions in a loop, the number of instructions executed to check for NULLs at the end of strings and arrays is reduced.

Abstract: Reading a value into a register, checking to see if the value is a NULL, and then jumping out of a loop if the value is a NULL is a common task that processors perform. To speed performance of such a task, a novel “blank bit” is added to the flag register of a processor. When a first instruction (arithmetic, logic or load) is executed, the instruction operands are checked to see if any is a NULL character value. Information on the result of the check is stored in the blank bit. Execution of a second instruction uses the information stored in the blank bit to determine whether or not a second operation (for example, a jump) will be performed. By using the first and second instructions in a loop, the number of instructions executed to check for NULLs at the end of strings and arrays is reduced.

Abstract: A memory is organized into blocks. In a write operation, data to be stored is combined with an address-dependent value (ADV) to form a block of information, and this block is encrypted. The block of encrypted information is written into a block of memory identified by the write address of the write operation. In a read operation, the block of encrypted information is read back from the memory and is decrypted to recover the data and the ADV. The address of the memory block from which the block of encrypted information was read is used to check the ADV to confirm that the ADV is related in the proper way to the address of the memory block that stored the encrypted information. If the check fails, the processor is prevented from executing the data, thereby preventing the processor from executing blocks of code that are in incorrect locations in memory.

Abstract: In a tamper detection system, a control system controls power consumption by sensors and power consumption of a state machine that controls operations of the sensors. A first state machine controls which sensor is activated. A second state machine controls the operations performed by the activated sensor and the operations are timed according to a clock signal. Prior to activating a sensor, the second state machine is in a wait state. The clock signal is gated so that the logic state of the clock signal does not change when the second state machine is in a wait state. Power consumption by the state machine is reduced by gating the clock signal so that the clock signal is held to a fixed value. Immediately after the activated sensor performs all operations, the second state machine samples the pass or fail result from the sensor.

Abstract: A memory controller mechanism is operable in a first mode and a second mode. In the first mode, a first memory controller portion of the mechanism can use a first set of data terminals to perform a first external bus access operation (EBAO) and a second memory controller portion of the mechanism can use a second set of data terminals to perform a second EBAO. The first and second EBAO operations may be narrow accesses that occur simultaneously. In the second mode, one of the controllers can use both the first and second sets of data terminals to perform a wider third EBAO. The memory controller mechanism can dynamically switch between first mode and second mode operations. In situations in which one of the sets of data terminals would not otherwise be used, performing wide accesses in the second mode using the one set of data terminals improves bus utilization.

Abstract: A memory controller mechanism is operable in a first mode and a second mode. In the first mode, a first memory controller portion of the mechanism can use a first set of data terminals to perform a first external bus access operation (EBAO) and a second memory controller portion of the mechanism can use a second set of data terminals to perform a second EBAO. The first and second EBAO operations may be narrow accesses that occur simultaneously. In the second mode, one of the controllers can use both the first and second sets of data terminals to perform a wider third EBAO. The memory controller mechanism can dynamically switch between first mode and second mode operation. In situations in which one of the sets of data terminals would not otherwise be used, performing wide accesses in the second mode using the one set of data terminals improves bus utilization.

Abstract: Upon execution of an interrupt return (IRET) instruction when a second interrupt is pending, rather than popping a stack, obtaining processor state information, and then pushing the state information back onto the stack prior to vectoring off to a second interrupt service routine, direct vectoring is employed such that the stack is not pushed or popped but rather the processor vectors directly from the IRET instruction in the first interrupt service routine to the second interrupt service routine. A novel stored interrupt enable (SIE) bit stores whether maskable interrupts were enabled at the time the first interrupt service routine was entered. Execution of IRET automatically checks the SIE. If the SIE indicates interrupts were enabled, then direct vectoring occurs. If the SIE indicates that interrupts were disabled, then the second interrupt remains pending, and an interrupt return operation is performed by popping the stack and restoring the prior processor state.

Abstract: Program code for a processor is stored in a non-volatile memory (for example, flash memory). An individual data bit stored in a memory cell of the non-volatile memory can be changed from an unprogrammed state to a programmed state using a write cycle. An individual bit stored in the memory cannot, however, be changed from the programmed state back to the unprogrammed state without performing an erase cycle on all the bits of a page of memory cells. The processor has an instruction set that includes a multi-bit breakpoint instruction, all the bits of which are the programmed state. Because all the bits of the breakpoint instruction are the programmed state of the memory, the breakpoint instruction can be written over any other instruction that is stored in the memory without having to perform an erase cycle or erase an entire page of program code.

Abstract: In a tamper detection system, a control system controls power consumption by sensors and power consumption of a state machine that controls operations of the sensors. A first state machine controls which sensor is activated. A second state machine controls the operations performed by the activated sensor and the operations are timed according to a clock signal. Prior to activating a sensor, the second state machine is in a wait state. The clock signal is gated so that the logic state of the clock signal does not change when the second state machine is in a wait state. Power consumption by the state machine is reduced by gating the clock signal so that the clock signal is held to a fixed value. Immediately after the activated sensor performs all operations, the second state machine samples the pass or fail result from the sensor.

Abstract: In an auto baud system and method, the baud rates between two communicating devices are synchronized by timing the transmission of a plurality of bits by counting the cycles of a reference clock. The number of cycles counted is then divided by the number of bits counted over and any remaining cycles are distributed evenly across the data being transmitted or received. The interface of the circuit is preferably implemented as a single pin, open drain interface which can be connected to an RS-232 communications link using external hardware.

Abstract: Fast electromagnetic transient (EFT) events of short duration are often not detected by power-on reset circuitry of an integrated circuit (IC). A fault detector circuit involves many fault detectors. The fault detectors are distributed across the IC and may be embodied in spare cells left in a standard cell IC. Each fault detector is initialized with a digital logic value. The fault detector circuit is then controlled such that the digital logic value stored should not change if the IC is operated under normal operating conditions. An EFT event that is undetected by the power-on reset circuit may, however, cause one of the digital logic values stored in one of the fault detectors to switch. If the digital logic value stored in any one of the fault detectors switches, then a fault signal is provided to the power-on reset circuit that in turn resets the IC.

Abstract: A single shared processing path is used as contexts are switched during processing. Each unique context is processed using a corresponding unique pipeline. If a pipeline that is executing under one context stalls, processing is switched in the shared processing path to another pipeline that is executing under second context. New pipelines are enabled for execution by borrowing a clock cycle from the currently executing pipeline. In some cases contexts are assigned various relative priority levels. In one case a context switching microprocessor is used in a communication engine portion of a system-on-a-chip communication system.

Abstract: Program code for a Processor is stored in a non-volatile memory (for example, flash memory). An individual data bit stored in a memory cell of the non-volatile memory can be changed from an unprogrammed state to a programmed state using a write cycle. An individual bit stored in the memory cannot, however, be changed from the programmed state back to the unprogrammed state without performing an erase cycle on all the bits of a page of memory cells. The processor has an instruction set that includes a multi-bit breakpoint instruction, all the bits of which are the programmed state. Because all the bits of the breakpoint instruction are the programmed state of the memory, the breakpoint instruction can be written over any other instruction that is stored in the memory without having to perform an erase cycle or erase an entire page of program code.

Abstract: A memory mapped programmable output generator, capable of producing events such as creating complex waveforms, triggering analog to digital and digital to analog conversions, and generating processor interrupts is disclosed. These events are considered high speed since they are timed relative to a high-speed clock and require minimal processor over head. The event generator may be embodied as either a peripheral to a microcontroller or as a separate circuit. In its preferred embodiment, the output generator is a peripheral device on a microcontroller and uses a dedicated programmable, reloadable timer which is inaccessible to other blocks. Events are loaded in a serial format, where only one event is active at a given time. These events are sequenced through address pointers associated with each event. Once a given event is completed, the output generator loads the next event from a next address pointer.