Abstract: A method and apparatus are provided for calibrating digital thermal sensors. A processor chip with a plurality of digital thermal sensors receives an analog voltage. A test circuit coupled to the processor chip receives a clock signal and a register coupled to the test circuit outputs a value on each clock cycle to a digital thermal sensor in the plurality of digital thermal sensors. The digital thermal sensor transitions an output state in response to the value of the register received in the digital thermal sensor equaling a temperature threshold of the digital thermal sensor. The value of the register at the point of transition is used to calibrate the digital thermal sensor. An incrementer increments the value of the register on each clock cycle in response to the value of the register received in the digital thermal sensor failing to equal the temperature threshold of the digital thermal sensor.

Abstract: A method, system, and computer program product are provided for distributing net power accurately. A workload is simulated operating on an integrated circuit. Net switching activity is determined for a set of nets and a set of subnets in the integrated circuit. Net switching data is generated based on the net switching activity. A net power value is calculated for each individual net and each individual subnet using the net switching data and a net capacitance for each individual net or subnet. Each calculated net power value is assigned to one of a set of source devices that drives the individual net or subnet, wherein the net power is distributed accurately. A net power assignment list is generated based on the assigning of each net power value to one of the set of source devices that drives the individual net or subnet.

Abstract: A method, system, and computer program product are provided for distributing net power accurately. A workload is simulated operating on an integrated circuit. Net switching activity is determined for a set of nets and a set of subnets in the integrated circuit. Net switching data is generated based on the net switching activity. A net power value is calculated for each individual net and each individual subnet using the net switching data and a net capacitance for each individual net or subnet. Each calculated net power value is assigned to one of a set of source devices that drives the individual net or subnet, wherein the net power is distributed accurately. A net power assignment list is generated based on the assigning of each net power value to one of the set of source devices that drives the individual net or subnet.

Abstract: Systems and methods for positioning thermal sensors within an integrated circuit in a manner that provides useful thermal measurements corresponding to different parts of the integrated circuit. In one embodiment, an integrated circuit includes multiple, duplicate functional blocks. A separate thermal sensor is coupled to each of the duplicate functional blocks, preferably in the same relative location on each of the duplicate functional blocks, and preferably at a hotspot. One embodiment also includes thermal sensors on one or more functional blocks of other types in the integrated circuit. One embodiment includes a thermal sensor positioned at a cool spot, such as at the edge of the integrated circuit chip. Each of the thermal sensors may have ports to enable power and ground connections or data connections between the sensors and external components or devices.

Abstract: A method and apparatus are provided for calibrating digital thermal sensors. A processor chip with a plurality of digital thermal sensors receives an analog voltage. A test circuit coupled to the processor chip receives a clock signal and a register coupled to the test circuit outputs a value on each clock cycle to a digital thermal sensor in the plurality of digital thermal sensors. The digital thermal sensor transitions an output state in response to the value of the register received in the digital thermal sensor equaling a temperature threshold of the digital thermal sensor. The value of the register at the point of transition is used to calibrate the digital thermal sensor. An incrementer increments the value of the register on each clock cycle in response to the value of the register received in the digital thermal sensor failing to equal the temperature threshold of the digital thermal sensor.

Abstract: A method, system, and computer program product for identifying failures in multi-core processors, utilizing logic built-in self test (LBIST) technology. Multi-core processors, having LBIST and pseudo-random pattern generator (PRPG) circuitry, are tested. Controlled by the LBIST control logic, PRPG inputs a test pattern into scan chains within the cores of each device. A new test pattern is generated and executed during the scan shift phase of each LBIST loop. Logic output generated by each scan chain in the core is compared to other core logic output. Failures within the multi-core processors are determined by whether the logic output generated from a core, within a latch sequence, does not match the logic output of the other cores. If logic output, from a core within a latch sequence, does not match, then the latch number, loop number, and latch values are recorded as failed.

Abstract: Systems and methods for sensing temperatures of multiple functional blocks within a digital device and controlling the operation of these functional blocks in a manner that selectively reduces temperatures associated with some of the functional blocks, but not others. One embodiment comprises an integrated circuit having multiple functional blocks (such as processor cores) and a set of thermal sensors coupled to sense the temperatures of the functional blocks. The integrated circuit includes control circuitry configured to receive signals from the thermal sensors, detect thermal events in the functional blocks and to individually adjust operation of the functional blocks to reduce the temperatures causing the thermal events. In one embodiment, the control circuitry includes a detection/control circuit coupled to each of the functional blocks and a thermal management unit configured to evaluate detected thermal events and to determine actions to be taken in response to the thermal events.

Abstract: An improved method and system for integrated circuit device physical design and layout. The physical layout of the integrated circuit device is optimally stored in a database to provide improved analysis capabilities of the integrated circuit device's characteristics. The method and system evaluates local interactions between functional blocks and decoupling cells on a given floor plan of a chip using this optimized database in order to reduce memory and processor utilization. Local noise is projected by using dI/dt and capacitance estimates. Areas of highest noise concern are identified, and floor plan mitigation actions are taken by tuning the placement of neighboring decoupling cells and their properties. Upon several iterative cycles, a near optimal solution for a given floor plan of the total chip is achieved.

Abstract: An improved method and system for integrated circuit device physical design and layout. The physical layout of the integrated circuit device is optimally stored in a database to provide improved analysis capabilities of the integrated circuit device's characteristics. The method and system evaluates local interactions between functional blocks and decoupling cells on a given floor plan of a chip using this optimized database in order to reduce memory and processor utilization. Local noise is projected by using dI/dt and capacitance estimates. Areas of highest noise concern are identified, and floor plan mitigation actions are taken by tuning the placement of neighboring decoupling cells and their properties. Upon several iterative cycles, a near optimal solution for a given floor plan of the total chip is achieved.