Abstract:
Eine Speicherverwaltungsschaltung beinhaltet einen DMA-Kanal (Direct Memory Access). Der DMA-Kanal beinhaltet eine Logik, die konfiguriert ist, um einen Puffer von Daten zu empfangen, die unter Verwendung von DMA geschrieben werden sollen. Der DMA-Kanal beinhaltet weiterhin eine Logik zum Durchführen einer Bitmanipulation in Echtzeit während eines DMA-Schreibzyklus des ersten Puffers von Daten.
Abstract:
A special mode key match comparison module has N-storage elements and a special mode key match comparator. The N-storage elements accumulate a serial data stream, and then determine whether a digital device should operate in a normal user mode, in a public programming mode, or in a particular private test mode. To reduce the possibility of accidentally decoding a false test or programming mode, the data stream has a sufficiently large number of N-bits to substantially reduce the probability of a false decode. To further reduce the possibility of accidentally decoding a programming or test mode, the special mode key match comparison module may be reset if less or more than N-clocks are detected during the accumulation of the N-bit serial data stream. The special mode key match data patterns may represent a normal user mode, a public programming mode, and particular private manufacturer test modes.
Abstract:
A data pointer for generating an indirect addressing mode address within a single cycle for a selected one of a plurality of multiple indirect addressing modes. The data pointer is used with a processor architecture scheme which allows for encoding of multiple addressing modes. A data pointer register is coupled to the processor architecture scheme for storing a current address of an operand to be used in a simple indirect addressing mode. An incrementer is coupled to the data pointer register for incrementing the current address of an operand to be used in a simple indirect data addressing mode by a set number thereby generating an address of an operand to be used in an indirect addressing mode with auto preincrement. An adder is coupled to the data pointer register for combining the current address of an operand to be used in a simple indirect data addressing mode with an offset number thereby generating an address of an operand to be used in an indirect addressing mode with offset. A multiplexer circuit is coupled to an output of the data pointer register, to an output of the incrementer, and to an output of the adder for selecting a desired indirect addressing mode address.
Abstract:
A system for allowing a two word instruction to be executed in a single cycle thereby allowing a processor system to increase memory space without reducing performance. A first address bus is coupled to the linearized program memory for sending addresses of instructions to be fetched to a linearized program memory. A pointer is coupled to the first address bus for storing an address location of a current instruction in the linearized program memory to be fetched and for placing the address location of the current instruction to be fetched on the first address bus. A second address bus is provided and has one end coupled to the output of the program memory and a second end coupled to the first address bus. The second address bus is used for placing an address of an operand of a second word of the two word instruction onto the first address bus after an address of an operand of a first word of the two word instruction has been placed on the first address bus. This allows the addresses of the first word and the second word to be combined to provide the full address value of the two word instruction in a single cycle.
Abstract:
A method for hot swapping an operating system kernel in a system, wherein the system comprises a central processing unit, one or more system components and a program memory including an old kernel and a new kernel, the method comprising: - configuring, by the old kernel, the system to boot from the new kernel by branching into a suitable entry point in the new kernel; - identifying, by the new kernel, one or more system components which require a reinitialization or reset; - further executing the new kernel in order to reinitialize or reset the one or more system components.
Abstract:
Peripheral functions of an integrated circuit device may be pooled and dynamically mapped to available external input-output connections of the integrated circuit device by using a set of configuration registers. To provide system robustness, the configuration registers may implement various levels of write protection, error correction and monitoring circuitry. One or more peripheral output functions may be mapped to one or more external output connections. Not more than one output function may be active at the same time on the same output connection. Outputs and inputs may be mapped to the same external input- output connection with or without the output being controllable for placement into an inactive state, e.g., high impedance or open collector. When the input is required to receive external data over the external input-output connection, the output may be placed into the inactive state.
Abstract:
A digital device having selectable modes for USB communications buffer management in a USB interface of the digital device. These modes may comprise (1) no ping-pong buffer support, (2) ping-pong buffer support for some endpoints, e.g., support for OUT endpoint 0 only, and (3) ping-pong buffer support for all endpoints. In the no pingpong buffer support mode, no hardware is required for automatic ping-pong buffer management. The Buffer Descriptor Tables may comprise a maximum of 128 memory locations, e.g., 16 IN endpoints and 16 OUT endpoints, each with at least one buffer descriptor, and each comprising four (4) memory locations. In the ping-pong buffer support for OUT endpoint 0 only mode, the buffer descriptor Tables may comprise a maximum of 132 memory locations, e.g., 16 OUT endpoints with an EVEN and an ODD endpoint 0, 16 IN endpoints, each with at least one descriptor, e.g., memory locations. This mode assures that endpoint 0 setup transfers may be serviced without delay while only requiring a minimal number of memory locations for the remainder of the buffer descriptors. In the ping-pong buffer support for all endpoints mode, automatic ping-pong buffer management may be provided for all endpoints. The Buffer Descriptor Tables may comprise a maximum of 256 memory locations, e.g., 16 IN endpoints and 16 OUT endpoints, an EVEN and ODD set for each, each with one descriptor, e.g., four (4) memory locations. This mode assures that all endpoint transfers may be serviced substantially without delay.
Abstract:
A microcontroller device has a housing with a plurality of external pins. a first microcontroller with a first central processing unit (CPU), a first system bus coupled with the first CPU, first memory coupled with the first system bus, and a first plurality of peripheral devices coupled with the first system bus, and a second microcontroller with a second central processing unit (CPU), a second system bus coupled with the second CPU, second memory coupled with the second system bus, and a second plurality of peripheral devices coupled with the second system bus, wherein first and second microcontroller communicate only via a dedicated interface.
Abstract:
A microcontroller device has a housing with a plurality of external pins having a plurality of input/output pins, a first microcontroller with a first central processing unit (CPU), a first system bus coupled with the first CPU, first memory coupled with the first system bus, and a first plurality of peripheral devices coupled with the first system bus, a second microcontroller with a second central processing unit (CPU), a second system bus coupled with the second CPU, second memory coupled with the second system bus, and a second plurality of peripheral devices coupled with the second system bus, and a pad ownership multiplexer unit being controllable to assign control of the input/output pins to either the first microcontroller or the second microcontroller, wherein the number of external pins is less than the sum of a data bus width of the first and second microcontroller.
Abstract:
A matching capacitor array is implemented on a single, monolithic integrated circuit. The array fastures a matrix of bottom electrodes and a plurality of continuous top electrode strips, where each continuous top electrode strip spans numerous bottom electrodes. The conductive contacts for each continuous top electrode strip are removed from the capacitor interface to the terminal ends of each of the continuous top electrode strips. The invention seeks to match or control parasitic and fringe capacitance, rather than to eliminate or minimize such capacitances. By creating a matched array, the parasitic and fringe capacitances of each matching capacitor unit cell are incorporated into the total capacitance of the unit cell.