Hewlett-packard Development Company, L.P.

Hewlett-packard Company

Other

Intel Corporation

Reid James Riedlinger

Dean A Mulla

Donald R Weiss

Douglas John Cutter

Samuel D Naffziger

Tom Grutkowski

Eric J DeHaemer

Ian M Steiner

Arijit Biswas

Terry L Lyon

Jayen J Desai

Thomas Grutkowski

Efraim Rotem

Krishnakanth Venkata Sistla

Alon Naveh

Vivek Garg

John J Wuu

Rakesh Kumar

Steven Frederick Liepe

William Johnson Bowhill

Don Soltis

Satish Shrimali

Lokesh Sharma

Douglas Shelborn Stirrett

Eric Fetzer

Steven Affleck

Rich McGowen, Ii

Steven Ray Afleck

Kuldeep Simha

Brandon Yelton

Christopher Craig Seib

A multicore processor may include multiple processing cores that were previously designated as active cores and at least one processing core that was previously designated as a functional spare. The processor may include an interface to receive, during operation of the processor in an end-user environment, a request to change the designation of at least one of the processing cores. The processor may be to store, into a desired cores configuration data structure in response to the request, data representing a bitmask that reflects the requested change, and to execute a reset sequence. During the reset sequence, the processor may activate, dependent on the bitmask, a processing core previously designated as a functional spare, or may deactivate, dependent on the bitmask, a processing core previously designated as an active core. The processor may include a predetermined maximum number of active cores and a predetermined minimum number of functional spares.

Systems and methods for in-field core failover

A processor may include a core and an uncore area. The power consumed by the core area may be controlled by controlling the Cdyn of the processor such that the Cdyn is within an allowable Cdyn value irrespective of the application being processed by the core area. The power management technique includes measuring digital activity factor (DAF), monitoring architectural and data activity levels, and controlling power consumption by throttling the instructions based on the activity levels. As a result of throttling the instructions, throttling may be implemented in 3droop and thermal design point (TDP). Also, the idle power consumed by the uncore area while the core area is in deep power saving states may be reduced by varying the reference voltage VR and the VP provided to the uncore area. As a result, the idle power consumed by the uncore area may be reduced.

Managing power consumption in a multi-core processor

A system for preventing processor errors in accordance with one exemplary embodiment of the present disclosure has a processor core, a patch, and a controller. The patch configures the processor core to detect occurrences of an event indicative of an imminent error in the processor core. The controller is configured to adjust, in response to a detection of an occurrence of the event by the processor core, a clock signal or a power signal provided to the processor core such that the imminent error is prevented.

System and method for preventing processor errors

A processor comprises a processor core and a controller. The processor core has an execution unit configured to execute instructions and to attempt to perform at least one operation in executing one of the instructions. The processor core is configured to detect a processor error associated with the at least one operation. The controller is configured to change an operating point of the processor core in response to a detection of the processor error such that the processor core operates at a new operating point, and the processor core is configured to retry the at least one operation while the processor core is operating at the new operating point.

System and method for adjusting operating points of a processor based on detected processor errors

One disclosed embodiment may comprise a design method for a dynamic circuit system. The method may include providing a design for a single stage network comprising a pull-down network that is configured to perform a desired logic function according to a plurality of inputs. The method may also include designing a multi-stage network that includes at least two stages, each of the at least two stages including a pull-down network that receives a respective portion of the plurality of inputs and each of the at least two stages cooperating to perform the desired logic function.

Circuit and circuit design method

A processor comprises a chip, a temperature sensing device, a processor core, and a controller. The temperature sensing device, the processor core, and the controller are integrated on the chip. The controller is configured to set, based on the temperature sensing device, the processor core to a plurality of specified operating points to enable testing of the specified operating points. Each of the operating points is defined by a different temperature and frequency combination, and the processor core is configured to run a set of test codes at each of the operating points.

System and method for testing a processor

Systems and methods are provided for searching at least one content addressable memory entry associated with a content addressable memory (CAM). A given content addressable memory entry comprises a plurality of CAM fields. At least one input selector controls access to the plurality of CAM fields, such that retrieval of a subset of the plurality of CAM fields is selectively enabled. A match evaluator compares an enabled subset of CAM fields to a search value.

Content addressable memory selectively addressable in a physical address mode and a virtual address mode

In a preferred embodiment, the invention provides a circuit and method for improving the soft error rate in a dual-port read SRAM cell. A write-only transfer device is connected to a cross-coupled latch, a first wordline, and a first bitline. A first read-only transfer device is connected to a second bitline, a second wordline, and a first pull-down device. A second read-only transfer device is connected to the first bitline, the first wordline, and a second pull-down device. A clear memory transfer device is connected to the cross-coupled latch, a third bitline, and a third pull-down device. This configuration allows a reduction in the size of a dual-port SRAM cell with little or no reduction in the read access time of the cell. The reduction in size also reduces SER by reducing the cross-sectional, p/n junction area exposed to radiation.

Dual-ported read SRAM cell with improved soft error immunity

A system and method are disclosed which provide a cache structure that allows early access to the cache structure's data. A cache design is disclosed that, in response to receiving a memory access request, begins an access to a cache level's data before a determination has been made as to whether a true hit has been achieved for such cache level. That is, a cache design is disclosed that enables cache data to be speculatively accessed before a determination is made as to whether a memory address required to satisfy a received memory access request is truly present in the cache. In a preferred embodiment, the cache is implemented to make a determination as to whether a memory address required to satisfy a received memory access request is truly present in the cache structure (i.e., whether a &ldquo;true&rdquo; cache hit is achieved). Although, such a determination is not made before the cache data begins to be accessed. Rather, in a preferred embodiment, a determination of whether a true cache hit is achieved in the cache structure is performed in parallel with the access of the cache structure's data. Therefore, a preferred embodiment implements a parallel path by beginning the cache data access while a determination is being made as to whether a true cache hit has been achieved. Thus, the cache data is retrieved early from the cache structure and is available in a timely manner for use by a requesting execution unit.

System and method utilizing speculative cache access for improved performance

The present invention provides a precharge circuit that has a first precharged node, a second precharged node, and a latch device. The first precharged node is charged to a high value during a precharge state. In response to a transition from the precharge state to an evaluate state, it either discharges to a low value or remains charged at its high value. The second precharged node has a value in the evaluate state that is based on the value of the first precharged node upon the circuit transitioning to the evaluate state. The latch device is connected to the second precharged node for latching this value in the evaluate state. With the latching device, this value is not affected by the first precharged node once the circuit has sufficiently transitioned to the evaluate state.

Latching annihilation based logic gate

The inventive cache uses a queuing structure which provides out-of-order cache memory access support for multiple accesses, as well as support for managing bank conflicts and address conflicts. The inventive cache can support four data accesses that are hits per clocks, support one access that misses the L cache every clock, and support one instruction access every clock. The responses are interspersed in the pipeline, so that conflicts in the queue are minimized. Non-conflicting accesses are not inhibited, however, conflicting accesses are held up until the conflict clears. The inventive cache provides out-of-order support after the retirement stage of a pipeline.

Cache chain structure to implement high bandwidth low latency cache memory subsystem

A system and method are disclosed which provide a built-in self test (BIST) for a content addressable memory (CAM) structure. In a preferred embodiment, an integrated circuit (chip) comprises a CAM structure that is accessible by a processor to satisfy memory access requests and a BIST implemented within such chip, which enables testing the integrity of the CAM structure. Such a preferred embodiment comprises a BIST that enables testing the integrity of the CAM structure that does not require circuitry for reading memory data out of the CAM structure. A preferred embodiment can also be utilized for testing a random access memory structure. In a preferred embodiment, a CAM BIST comprises logic capable of generating test values (e.g., a test pattern), a shift register that temporarily stores the test values generated by the logic, and compare circuitry that determines whether a test value matches an entry within the CAM structure.

Built-in self test for content addressable memory

The inventive cache manages address conflicts and maintains program order without using a store buffer. The cache utilizes an issue algorithm to insure that accesses issued in the same clock are actually issued in an order that is consistent with program order. This is enabled by performing address comparisons prior to insertion of the accesses into the queue. Additionally, when accesses are separated by one or more clocks, address comparisons are performed, and accesses that would get data from the cache memory array before a prior update has actually updated the cache memory in the array are canceled. This provides a guarantee that program order is maintained, as an access is not allowed to complete until it is assured that the most recent data will be received upon access of the array.

Cache address conflict mechanism without store buffers

A self-timed translation lookaside buffer (TLB) is disclosed that utilizes a two-level match scheme to trigger the evaluation of whether a match is achieved for a received virtual address within the TLB. The first level is referred to as the local match, and the second level is referred to as the global match. An entry of a TLB comprises groups of bits, with each group coupled to a separate local match line. Each of the local match lines of an entry is coupled to a global match line, which is initially set to a high voltage level and discharges to a low voltage level if any of the local match lines indicate a mismatch for their respective group. Accordingly, when the global match lines are evaluated, if the global match line has a high voltage level it indicates that the associated TLB entry matches the virtual address, otherwise the global match line indicates a mismatch for the entry. Multiple global match lines are evaluated to trigger a memory access for a matching entry. More specifically, in a preferred embodiment, a pair of neighboring global match lines are input to a NAND gate, the output of which triggers the evaluation of whether a match is achieved for either entry.

Growing community of inventors

Fort Collins, CO, United States of America

Reid James Riedlinger