Abstract: A system includes two or more agents and a distributed arbitration scheme for the bus to which the agents are connected. Thus, an arbiter corresponding to each agent is provided. The arbiters are reset using a first reset signal, while the agents are reset using a separate reset signal or signals. The arbiters are concurrently released from reset when the first reset signal is deasserted, and may have a consistent reset state to provide for synchronization of the arbiters. The agents may be independently released from reset by the separate reset signals. Accordingly, the arbiters may be synchronized and may remain synchronized even if the corresponding agents are released from reset at different times, or are temporarily held in reset for any reason.

Abstract: A computer system may include multiple processing nodes, one or more of which may be coupled to separate memories which may form a distributed memory system. The processing nodes may include caches, and the computer system may maintain coherency between the caches and the distributed memory system. Particularly, the computer system may implement a flexible probe command/response routing scheme. The scheme may employ an indication within the probe command which identifies a receiving node to receive the probe responses. For example, probe commands indicating that the target or the source of transaction should receive probe responses corresponding to the transaction may be included. Probe commands may specify the source of the transaction as the receiving node for read transactions (such that dirty data is delivered to the source node from the node storing the dirty data).

Abstract: A high-speed bank select multiplexer latch may be coupled to a pair of differential output nodes and configured to capture and retain an output on the pair of differential output nodes responsive to two or more pairs of differential data inputs being active. A first subcircuit including a first N-channel transistor and a second N-channel transistor is configured to receive at least a first input signal and a second input signal and to drive a first output on a first output node responsive to either of the first input signal or the second input signal being active. Additionally, a second subcircuit including a third N-channel transistor and a fourth N-channel transistor is configured to receive at least a third input signal and a fourth input signal and to drive a second output on a second output node responsive to either of the third input signal or the fourth input signal being active.

Abstract: A system and method for bus arbitration. A computer system includes one or more buses for transferring data. Access to each bus is controlled by an arbitration unit. Various bus agents (i.e. peripherals) are coupled to the bus. Some bus agents are designated as normal-priority agents, while other bus agents are designated as high-priority bus agents. A high-priority bus agent may be a peripheral that is a latency-sensitive device. The arbitration unit may grant bus access to a normal-priority bus agent based on an arbitration scheme. When a high-priority bus agent requests access to the bus, the arbitration unit may cause the termination of access by the normal-priority bus agent. The high-priority bus agent is then granted access to the bus. When the high-priority bus agent has completed its use of the bus, the arbitration unit allows the normal-priority bus agent to regain access to the bus.

Abstract: A memory controller may include a request queue for receiving transaction information (e.g. the address of the transaction) and a channel control circuit. A control circuit for the request queue may issue addresses from the request queue to the channel control circuit out of order, and thus the memory operations may be completed out of order. The request queue may shift entries corresponding to transactions younger than a completing transaction to delete the completing transaction's information from the request queue. However, a data buffer for storing the data corresponding to transactions may not be shifted. Each queue entry in the request queue may store a data buffer pointer indicative of the data buffer entry assigned to the corresponding transaction. The data buffer pointer may be used to communicate between the channel control circuit, the request queue, and the control circuit. In one implementation, the request queue may implement associative comparisons of information in each queue entry (e.g.

Abstract: A method and apparatus for fault handling in computer systems. In one embodiment, a first register is used to store an address which points to the top of a stack. The address stored in the first register may be updated during the execution of an instruction. A second register may be used to store an address previously first register. The contents of the second register may be kept unchanged until the retirement of the instruction that is currently executing. If a fault occurs during execution of the instruction, a microcode fault handler may perform routines that may clear the fault or those conditions which led to the fault. The microcode fault handler may also copy the contents of the second register back into the first register. Execution of the instruction may be restarted from the operation just prior to when the fault occurred. The program from which the instruction originated may then continue to run.

Abstract: A memory controller provides programmable flexibility, via one or more configuration registers, for the configuration of the memory. The memory may be optimized for a given application by programming the configuration registers. For example, in one embodiment, the portion of the address of a memory transaction used to select a storage location for access in response to the memory transaction may be programmable. In an implementation designed for DRAM, a first portion may be programmably selected to form the row address and a second portion may be programmable selected to form the column address. Additional embodiments may further include programmable selection of the portion of the address used to select a bank. Still further, interleave modes among memory sections assigned to different chip selects and among two or more channels to memory may be programmable, in some implementations.

Abstract: A processor employs a store to load forward (STLF) predictor which may indicate, for dispatching loads, a dependency on a store. The dependency is indicated for a store which, during a previous execution, interfered with the execution of the load. Since a dependency is indicated on the store, the load is prevented from scheduling and/or executing prior to the store. The STLF predictor is trained with information for a particular load and store in response to executing the load and store and detecting the interference. Additionally, the STLF predictor may be untrained (e.g. information for a particular load and store may be deleted) if a load is indicated by the STLF predictor as dependent upon a particular store and the dependency does not actually occur. In one implementation, the STLF predictor records at least a portion of the PC of a store which interferes with the load in a first table indexed by the load PC.

Abstract: A voltage regulator may include one or more features for generating high PSRR. For example, source follower devices may be included in the voltage regulator for providing current sources for the output voltage nodes. The source followers may be sensitive to power supply noise at the gate terminal. Filters are included on the gate terminals to filter the power supply noise, thus reducing the noise at the gate terminals. As another example, the voltage regulator may employ current sources on the output voltage nodes which produce current inversely proportional to the current drawn by the load. In one embodiment, the voltage regulator may include a power control circuit used to provide overvoltage protection during power up. The power control circuit provides a voltage during power up, and ceases providing the voltage after a time interval so that the circuit may operate.

Abstract: A scheduler issues memory operations without regard to whether or not resources are available to handle each possible execution outcome of that memory operation. The scheduler also retains the memory operation after issuance. If a condition occurs which prevents correct execution of the memory operation, the memory operation is retried. The scheduler subsequently reschedules and reissues the memory operation in response to the retry. Additionally, the scheduler may receive a retry type indicating the reason for retry. Certain retry types may indicate a delayed reissuance of the memory operation until the occurrence of a subsequent event. In response to such retry types, the scheduler monitors for the subsequent event and delays reissuance until the event is detected. The scheduler may include a physical address buffer to detect a load memory operation which incorrectly issued prior to an older store memory operation upon which it is dependent for the memory operation.

Abstract: A storage circuit for an integrated circuit is configured to couple to a first power supply voltage (e.g. a Vdd power supply voltage used by other circuitry within the integrated circuit) in response to a deassertion of a hold signal and configured to couple to a second power supply in response to an assertion of the hold signal. The second power supply voltage may be the hold signal voltage or another power supply voltage separate from the Vdd power supply voltage. The hold signal may be asserted and the Vdd power supply voltage may be removed. Leakage current in circuits powered only by the Vdd power supply voltage may be eliminated, while the storage circuit may retain its stored value. A system including the integrated circuit and a method for managing power in the system.

Abstract: A clock divider circuit producing 0° and 90° outputs with a 50% duty cycle is provided. In one embodiment, the clock divider circuit may include a pair of cross-coupled circuits. The clock divider circuit may produce a first output clock signal and a second output clock signal that is phase shifted a positive 90° with respect to the first output clock signal. The operation of the circuit may be responsive only to the input clock signal. In other words, the circuit may not require a reset signal to operate in a deterministic fashion.

Abstract: A charge pump includes at least one switching transistor for switching current on or off in response to an up or down signal, a pair of transistors (one coupled to the source and the other to the drain of the switching transistor) each having its source and drain shorted and coupled to receive a complement of the signal on the gate terminal of the switching transistor on their gate terminals, and a fourth transistor coupled to the drain of the switching transistor and a power supply. The pair of transistors are activated concurrent with the deactivation of the switching transistor. The fourth transistor may provide for active shutoff of a current transistor being switched by the switching transistor, by actively charging the source of the current transistor to a voltage which is not exceeded by the gate terminal of the current transistor.

Abstract: A method and apparatus for detecting errors in a First-In-First-Out buffer (FIFO). A FIFO includes verification bits associated with data entries. In addition, the FIFO includes an expected value bit for comparison on reads. Upon reset, the verification bits are initialized to an alternating sequence of binary values and the expected value bit is initialized to a predetermined binary value. On a write to a FIFO entry, the corresponding verification bit is toggled. On a read from an entry, the corresponding verification bit is compared to the expected value. If the verification bit does not match the expected value, an error is detected.

Abstract: A method and apparatus for improving the performance of a superscalar, superpipelined processor by identifying and processing instructions for performing addressing operations is provided. The invention heuristically determines instructions likely to perform addressing operations and assigns those instructions to specialized pipes in a pipeline structure. The invention can assign such instructions to both an execute pipe and a load/store pipe to avoid the occurrence of “bubbles” in the event execution of the instruction requires the calculation capability of the execute pipe. The invention can also examine a sequence of instructions to identify an instruction for performing a calculation where the result of the calculation is used by a succeeding load or store instruction. In this case, the invention controls the pipeline to assure the result of the calculation is available for the succeeding load or store instruction even if both instructions are being processed concurrently.

Abstract: A cache is coupled to receive an access which includes a cache allocate indication. If the access is a miss in the cache, the cache either allocates a cache block storage location to store the cache block addressed by the access or does not allocate a cache block storage location in response to the cache allocate indication. In one implementation, the cache is coupled to an interconnect with one or more agents. In such an implementation, the cache accesses may be performed in response to transactions on the interconnect, and the transactions include the cache allocate indication. Thus, the source of a cache access specifies whether or not to allocate a cache block storage location in response to a miss by the cache access. The source may use a variety of mechanisms for generating the cache allocate indication.

Abstract: An agent, in response to a write to a shared block, is configured to initiate a read exclusive transaction on an interface on which the agent communicates. Additionally, the agent is configured to indicate, to a responding agent or agents on the interface, that a data transfer is not required from the responding agent or agents in response to the read exclusive transaction. In one embodiment, the agent indicates to the responding agents that a data transfer is not required in a response phase of the transaction. Specifically, the agent may respond in such a way that the agent indicates that it will provide the data (i.e. that the agent will provide the data to itself). For example, the agent may respond with an exclusive ownership indication. On the interface for such an embodiment, an exclusive ownership response may require that the agent having exclusive access respond with the data.

Abstract: A processor is described which includes a stride detect table. The stride detect table includes one or more entries, each entry used to track a potential stride pattern. Additionally, each entry includes a confidence counter. The confidence counter may be incremented each time another address in the pattern is detected, and thus may be indicative of the strength of the pattern (e.g., the likelihood of the pattern repeating). At a first threshold of the confidence counter, prefetching of the next address in the pattern (the most recent address plus the stride) may be initiated. At a second, greater threshold, a more aggressive prefetching may be initiated (e.g. the most recent address plus twice the stride). In some implementations, the prefetch mechanism including the stride detect table may replace a prefetch buffer and prefetch logic in the memory controller.

Abstract: A processor supports a processing mode in which the default address size is greater than 32 bits and the default operand size is 32 bits. The default address size may be nominally indicated as 64 bits, although various embodiments of the processor may implement any address size which exceeds 32 bits, up to and including 64 bits, in the processing mode. The processing mode may be established by placing an enable indication in a control register into an enabled state and by setting a first operating mode indication and a second operating mode indication in a segment descriptor to predefined states. Additionally, an instruction prefix may be coded into an instruction to override the default address and/or operand size. Thus, an address size of 32 bits may be used when desired, and an operand size of 64 bits may be used when desired.

Abstract: A cache includes an error circuit for detecting errors in the replacement data. If an error is detected, the cache may update the replacement data to eliminate the error. For example, a predetermined, fixed value may be used for the update of the replacement data. Each of the cache entries corresponding to the replacement data may be represented in the fixed value. In one embodiment, the error circuit may detect errors in the replacement data using only the replacement data (e.g. no parity or ECC information may be used). In this manner, errors may be detected even in the presence of multiple bit errors which may not be detectable using parity/ECC checking.