Abstract: A computer system having an interconnection apparatus for connecting processors, peripherals and memories, the system including a plurality of electronic devices, and a multiple long bus structure with impedance elements disposed thereon for providing non-terminal termination points.

Abstract: A network/portable computer (18) includes a modular bay (20) which can receive either a hard drive (22) or a data transceiver (24) for establishing a wireless network connection. When the network/portable computer (18) is in the local area covered by the network (10), the data transceiver (24) is installed to use the network (10) for storage and retrieval of data. When used in a remote area, where a network connection cannot be established, the hard drive (22) is installed in the modular bay (20) to provide a full featured computer.

Abstract: A method and implementing system are provided for determining and retaining an identification number relevant to an electronic system component and/or component configuration. In an exemplary embodiment, existing pull-up resistors within a computer system are connected in a manner to enable associated circuitry to determine a pre-assigned identification number for the computer system. The identification number is stored in an identification number register and accessible for providing the identification number in response to a requests from other devices within the system.

Abstract: Low-latency distributed round-robin arbitration is used to grant requests for access to a shared resource such as a computer system bus. A plurality of circuit board cards that each include two devices such as CPUs, I/O units, and ram and an address controller plugs into an Address Bus in the bus system. Each address controller contains logic implementing the arbitration mechanism with a two-level hierarchy: a single top arbitrator and preferably four leaf arbitrators. Each address controller is coupled to two devices and the logical "OR" of their arbitration request is coupled via an Arbitration Bus to other address controllers on other boards. Each leaf arbitrator has four prioritized request in lines, each such line being coupled to a single address controller serviced by that leaf arbitrator. By default, each leaf arbitrator and the top arbitrator implement a prioritized algorithm.

Abstract: A shared resource lock mechanism is provided which enables processors in a multi-processor system which each share common resources to obtain locks on those resources using a transactions which minimizes the amount of time system resources are unavailable, while also allowing system resources to be available for other processing tasks.

Abstract: Chipset or a device for attachment of the ROM BIOS within the system architecture. Although normally attached to the ISA bus, the ROM BIOS may be attached to an alternate bus (typically a higher-performance bus) located within the system, thereby potentially eliminating the ISA bus from the computer system.

Abstract: A method for processing an atomic transaction in a multi-bus system that includes a local bus and a remote bus. According to the method, a first transaction in an atomic sequence is received from the local bus. The first transaction is terminated on the local bus. The first transaction is performed on the remote bus. A response to the first transaction is received and the response is placed on the local bus.

Abstract: A computer system includes a motherboard to which a selected one of various processor boards can be coupled via an adaptor. The boards include a CPU and a heat transfer member. The adaptor includes core logic such as a Northbridge module and power control circuitry. The adaptor is pin connectable to the motherboard and the various processor boards are each pin connectable to the adaptor. Alternatively, the adaptor and processor boards can be replaced with a processor module including features of the processor boards and the adaptor. The processor module is also pin connectable to the motherboard at the same connection used by the adaptor.

Abstract: A method and apparatus for interfacing buses includes a system interface processor coupled to a first bus and including a command register accessible via a second bus. A request buffer and a response buffer are provided which are accessible via the second bus and coupled to the interface processor. The request buffer can be used to store information to be transmitted from the second bus to the first via the interface processor while the response buffer can be used to store information to be transmitted from the first bus to the second bus via the interface processor. The interface processor may include a status register to indicate the status of the interface controller. The interface controller may also include a command register to receive commands transmitted over the second bus.

Abstract: A PCI bridge is configured to perform delayed read operations in response to a memory read initiated on the PCI bus. Normally, the PCI bridge is configured to discard delayed read data read from main memory following a predetermined discard count time after the PCI master establishing the delayed read operation is retried on the PCI bus. The computer system further includes a secondary bus bridge such as an ISA bridge for providing an interface between the PCI bus and an ISA bus. When an ISA device desires to read data from the main memory, the ISA bridge asserts a flush request signal. The PCI bridge responsively flushes any pending CPU to PCI transactions pending within the PCI bridge. When the flushing operation is complete, the PCI bridge asserts an acknowledge signal. A PCI arbiter for arbitrating ownership of the PCI bus may increase a level of arbitration priority provided to the ISA bridge in response to assertion of the acknowledge signal.

Abstract: A data processing device uses a processor such as a central processing unit and a special-purpose hardware circuit, such as an accelerator for accelerating the software operation using the operation program of the processor by replacing the software operation partially by the hardware. A practical application of this processor arrangement is found in mobile communication terminal devices including a digital cellular portable telephone in which a digital signal processor of a mobile communication terminal device operates in association with an accelerator for accelerating specific signal processings such as waveform equalization. The processor provides input data to the accelerator and the results of operation by the accelerator are output to a register or memory based on a cycle of operation to be read periodically by the processor.

Abstract: A low overhead, efficient, and simple protection check circuit is inserted into a data path between a master requester and a target resource such as a memory or input/output device. The master requester initiates a memory request, a pio access request, or a dma transaction directed to the target resource. For example, the master requester may be a processor accessing a memory, a processor performing programmed I/O (pio). Alternatively, the master requester may be a DMA device performing direct memory access of a memory. The protection check circuit is configured at initialization time by an operating system or a privileged software process, then passively monitors all transactions on the data paths, disallowing accesses that fail the protection check operation.

Abstract: A SCSI bus extender apparatus coupling a primary SCSI bus to a secondary SCSI bus includes a mechanism for detecting and resolving contention between a substantially simultaneous SELECTION operation on the primary bus and a RESELECTION operation on the secondary bus. The inventive method contemplates the bus extender arbitrating for control of the primary bus after a conflict is detected, and releasing control of the secondary bus if control of the primary bus is obtained. A target device on the secondary bus can then rearbitrate for control of the secondary bus. Once the target device controls the secondary bus, it can direct a RESELECTION signal to the bus extender, which responsively directs the signal to an initiator device on the primary bus. If the bus extender is unable to gain control of the primary bus after a conflict is detected, the SELECTION operation is allowed to proceed and the target device reattempts to assert the RESELECTION operation thereafter.

Abstract: An interrupt scheme for a data processor includes an enable field for a non-maskable interrupt (NMI). The field is automatically cleared by the data processor when it services the highest priority interrupt, a RESET. The user can set the field to enable a subsequent NMI but cannot himself clear the NMI. This strategy prevents an NMI from interrupting a RESET service routine.

Abstract: A flexible general input/output function utilizes a programmable logic circuit in conjunction with general purpose input/output pins. A programmable logic circuit receives the input signals from the input terminals. The programmable logic circuit program conditions the input signals and provides conditioned input signals to the remainder of the integrated circuit. An input register receives the conditioned input signals from the programmable logic circuit, and stores values representing the state of respective conditioned input signals. A transition detection circuit detects a specified transition for each of the conditioned input signals it receives and provides an indication of the specified transition. An interrupt circuit is responsive to transition indications provided from the transition detection circuit to generate an interrupt signal associated with the specified transition of a respective conditioned input signal.

Abstract: A superscalar microprocessor employing a data cache configured to perform store accesses in a single clock cycle is provided. The superscalar microprocessor speculatively stores data within a predicted way of the data cache after capturing the data currently being stored in that predicted way. During a subsequent clock cycle, the cache hit information for the store access validates the way prediction. If the way prediction is correct, then the store is complete, utilizing a single clock cycle of data cache bandwidth. Additionally, the way prediction structure implemented within the data cache bypasses the tag comparisons of the data cache to select data bytes for the output. Therefore, the access time of the associative data cache may be substantially similar to a direct-mapped cache access time. The superscalar microprocessor may therefore be capable of high frequency operation.

Abstract: A read channel (18) having a waveform sampler (58) is provided for use in a disk drive mass storage system (30). The disk drive mass storage system (30) includes a disk/head assembly (12), a preamplifier (14), and a control circuitry (11). The read channel (18) is coupled to the control circuitry (11) through a data/parameter path (13). The read channel (18) includes a plurality of circuit modules for processing a waveform data signal received from the preamplifier (14) and to generate a digital data signal in response. The read channel provides the digital data signal to the control circuitry (11) through the data/parameter path (13). The read channel (18) also includes the waveform sampler (58) for receiving and sampling a processed waveform signal from one of the plurality of circuit modules and for generating a digital waveform sampler signal in response. The read channel can provide either the digital data signal or the digital waveform sampling signal as an output onto the data/parameter path (13).

Abstract: A flash memory control apparatus and method which enables updating of data at high speed. The flash memory control apparatus includes a flash memory having a memory region which is divided into a plurality of sectors each including a logical address portion for storing a logical address of the sector, an erasure managing portion for storing information which indicates at least whether or not the sector may be erased, and a data part for storing data; and a control device, coupled to the flash memory, for making access to an arbitrary sector of the flash memory by specifying the logical address of the arbitrary sector. The flash memory control method includes the steps of: (a) dividing a memory region of a flash memory into a plurality of sector; and (b) making access to an arbitrary sector of the flash memory by specifying the logical address of the arbitrary sector.

Abstract: An associative memory device having a high speed and good performance is provided without degrading the simple design of a peripheral circuit of a conventional associative memory. The associative memory device has an N-bit first buffer and an M-bit second buffer in which W-bit data is stored through a data input port, detection device for detecting that the W-bit data is input to the first buffer k times or to the second buffer r times, a switch for alternately switching buffers in which the W-bit data is stored, and a search control for performing a search operation for a memory region by using data in the first or second buffer. By using data in the first or second buffer, during the search operation for the memory region of the associative memory, the W-bit data for the next search operation is input to the second or first buffer.

Abstract: A virtual disk simulator combines memory space from a given process address space in user space to form a virtual disk with contiguous memory space. The virtual disk may be accessed in the same manner as a physical hardware disk. The specific components of the virtual disk simulator are a virtual disk device driver in kernel space and a multi-threaded application running in user space. The virtual disk device driver supports the disk abstraction (e.g. simulation of the virtual disk having what appears to be unlimited disk space) and the multi-threaded application performs actual physical storage management in cooperation with the virtual disk device driver. The virtual disk device driver and the multi-threaded application interact using a special set of communication protocols.

Abstract: A mirrored memory for a data processing system. The memory system includes two device controllers and related disk drives for storing the mirrored data. Each of the disk drives is divided into logical volumes. Each device controller contains a plurality of reading processes and a correspondence table that establishes the reading process to be used in retrieving data from the corresponding disk drive. Each disk controller responds to a read command that identifies the logical volume by using the correspondence table to select the appropriate reading process and by transferring data from appropriate physical storage device containing the designated logical volume.

Abstract: A system for caching texel information in a cache data store, for use in a graphics rendering system which uses interpolative sampling to compute texture color values. The system includes a texel memory storing texel information, a graphics application program for using interpolative sampling to compute dynamic texture values, a first cache data storage for a number of the most-recently-retrieved texels, a second cache data storage for a previously-retrieved adjacent line of texels, cache tag blocks for determining whether the texels needed by the graphics accelerator system are cached in either of the first or second cache data stores, and a memory request generator for retrieving texels from texel memory upon indication of a miss by the cache tag blocks.

Abstract: A peripheral module capable of interaction with a host system comprises a digital signal processor (DSP), a cache controller, and minimal or no resident memory for initial storage of instructions and data. The peripheral module utilizes the memory resources (e.g., ROM and RAM) of a host system. The cache controller interfaces to the DSP and provides an instruction and data stream, upon request, from a resident cache. A bus interface unit arbitrates access to the host system via a host bus for access to host system resources by the cache controller for extracting DSP information (e.g., instructions and data) from the host system for utilization by the DSP within the peripheral module. The cache controller and cache thereby provide the instruction and data stream as required by the DSP for digital signal processing applications and relieve the need for resident storage within the peripheral module.

Abstract: A memory control apparatus interposed between a central processing unit and a memory device to store data includes a channel control unit to control a data transfer to/from the central processing unit; a drive control unit to control a data transfer to/from the memory device; a plurality of cache memories to temporarily store data which is transferred between the central processing unit and the memory device; and a cache memory control unit having a selector for selecting a cache memory to store data which is transferred from the memory device. The memory control apparatus selects a cache memory to which data is to be stored so as to almost equalize usage of the plurality of cache memories, thereby controlling allocation of the cache memories and enabling a cache memory space to be effectively used.

Abstract: A high performance microprocessor bus protocol for improving system throughput. The bus protocol enables overlapping read burst and write burst bus transactions to a cache, and interleaved bus transactions during external fetch cycles for missed cache lines. The bus protocol is implemented in a system comprising a CPU, and a secondary cache. The secondary cache comprises an SRAM array cache, and a cache controller. The CPU contains an instruction pipeline and a primary cache system.

Abstract: In a cache coherency control method of a multi-processor system comprising a plurality of cache systems of identical configuration-after "method", for quickly determining consistency of a data block designated by a cache coherency request issued by other cache system a multi-processor system using the same, systems have identical configuration and each of the cache systems includes a history table for storing an address included in an access request flowing over a shared bus and a history table control circuit. The history table control circuit determines whether an address of a received access request is stored in the history table, and when the address is stored in the history table, suppresses the operation of a cache control circuit for the access request, and alternatively when the address is not stored in the address table, conducts the operation of the cache control circuit for the access request.

Abstract: A method and apparatus employing a dynamic encryption interface between a processor and a memory. The interface of the present invention dynamically encrypts the contents of the memory. A determination is made whether a memory access request is active. If yes, a further determination is made whether the address associated with the memory access request is greater than a POINTER. If yes, encrypting or decrypting the memory access using a first key. If no, encrypting or decrypting the memory access using a second key. Processing then proceeds to the decision block that determines whether or not the memory access request is active. If the memory access request is not active, data is read from a memory location identified by the POINTER. The read data is decrypted using the first key. The data is encrypted using the second key. The encrypted data is written back to the memory location point identified by the POINTER.

Abstract: An empty block table is constructed by 64 words.times.8 bits and has 512 memory positions 000H(A0) to IFFH(A511) one-to-one corresponding to 512 blocks BL0 to BL511 within a flash memory FMi. Empty data [a] of 1 bit is stored to each memory position (Aj). This empty data has value "1" when a block BLj corresponding to this memory position (Aj) is in an empty state at present. The empty data also has value "0" when no block BLj corresponding to this memory position (Aj) is in the empty state at present (when data are included in this block).

Abstract: A memory controller for a special purpose digital video processor. To achieve a speed enhancement when using multiple bank memory such as SDRAM, the memory controller arbitrates requests for access to the memory such that, if possible, sequential memory accesses are directed to alternating memory banks. To facilitate access to contiguous blocks of memory such as are often accessed in video signal processing, the memory controller includes an address generator for generating multiple memory addresses in response to a single memory access request. The memory controller further includes features which permit the use of multiple physical memory configurations. Specifically, the memory controller includes a memory address mapper for translating virtual memory address signals into physical memory address signals for addressing memory; for different memory configurations, the translation is different.

Abstract: A data backup apparatus determines whether a version serial number of data backed up in a network management system matches a version serial number of data stored in a data version serial number management field of an update history management area, upon having the network management system back up update data generated in a data processing apparatus. Upon a match of the version serial numbers of respective data, the data processing apparatus transfers only update data stored in the data update history management area via a communications link to the network management system. Upon a mismatch of the version serial numbers of respective data, the data processing apparatus transfers all data, including data other than update data, stored in an EEPROM of the data processing apparatus, to the network management system.

Abstract: A memory controller and at least one memory module exchange data at high transfer rates by minimizing clock skew. When writing data to the memory module, the memory controller generates a clock signal that travels along a first clock line segment. The data bus carries the write data, and the electrical characteristics of the data bus and first clock line segment are matched such that incident wavefronts of the data bus and clock signal arrive at the memory module in fixed relation to one another. When reading data, the first clock line segment is looped back from the memory module to the memory controller along a second clock line segment, with a copy of the clock signal provided on the second clock line segment. The data bus carries the read data, and the electrical characteristics of the data bus and the first clock line segment are matched such that incident wavefronts of the data bus and clock signal arrive at the memory controller in fixed relationship to one another.

Abstract: A dual word enable method for memory data access includes the steps of: (i) providing a plurality of address data signals for addressing data stored in an array; (ii) issuing a first row access strobe (RAS) signal to decode the addressing data; and (iii) issuing a second row access strobe (RE2) signal for driving the address data into the memory array after determining that data is present in the memory array.

Abstract: Write transactions are conducted by transmitting a first write address from a source device over a first bus on a first clock cycle and transmitting a first data word corresponding to the first write address from the source device over a second bus commencing on a later clock cycle. In order to execute write transactions in this manner, a memory unit is modified to contain a pending write buffer and a memory array. During a write transaction, the address and corresponding data is first stored in the pending write buffer and the data is later transferred into the memory array upon subsequent write transactions. During a read transaction, the read address is compared to the address stored in the pending write buffer. If the read address matches the address stored in the pending write buffer, the corresponding data stored in the pending write buffer is transmitted in response to the read request. If there is no match, corresponding data from the memory array is transmitted.

Abstract: In a computer system including a packet-switched bus, a method for requesting transactions such that memory accesses are initiated quickly. A master transmits a first portion of a transaction request packet having multiple portion. A memory controller receives the first portion of the transaction request, which includes a row address portion of a memory address. The memory controller initiates a memory access by applying a row address strobe signal to the row of the memory location in response to receiving the first portion of the request packet, and the master transmits any remaining portion of the transaction request. After the full memory address has been received, it is determined whether data stored at the memory location is to be read from a source other than the memory location. The memory controller aborts the memory access by inhibiting assertion of a column access strobe signal to the memory location if the data is to be read from a source other than the memory location.

Abstract: Execution memory for use in processing a program unit for a database is allocated by inspecting an execution memory area pool for a previously allocated execution memory area. If a previously allocated execution memory area is found, then the previously allocated execution memory area is established as the execution memory area for executing the program unit. On the other hand, if a previously allocated execution memory area is not found, then a new execution memory area is allocated and configured for executing the program unit.

Abstract: A flexible memory mapper for selectively inverting the state of an address line on an address bus includes a selectable inverter element and a control circuit. The selectable inverter element has a control input coupled to an output of the control circuit, an input, and an output. The control input, the input, and the output are capable of being in two logic levels. When the control input is in a first logic level, the input is passed through uninverted to the output of the selectable inverter element. When the control input is in a second logic level, the input is passed through inverted to the output of the selectable inverter element. The control circuit output feeds one of two logic levels to the control input of the selectable inverter element, outputting the second logic level in response to an input address within a selected address range and a user selectable input bit, and otherwise outputting the first logic level.

Abstract: In a computer system, a peripheral graphics device (PGD) accesses a graphics buffer (GB) wherein the GB physical pages can be contiguous or discontiguous. A request is received to allocate memory for a GB of a predetermined size and handle. The number of pages within the size parameter is determined based on the page size used by the computer system and the buffer size needed. A first memory block is allocated for storing the GB and locked to prevent swapping. A starting virtual address of a CPU page table (CPU/PT) is accessed and mapped to a starting logical address to allow traversal of the CPU/PT by a user application. A second memory block is allocated for building a graphics device page table (GDPT) that can be accessed by a PGD. The logical address of each GB page is sequentially accessed and the corresponding physical address of each GB page is determined from the CPU/PT. The logical and physical addresses of each GB page are stored into the GDPT.

Abstract: A system for allowing multiple addressing modes while maximizing a number of available opcodes and addressable registers. The system has a processor architecture scheme which allows for encoding of multiple addressing modes through use of virtual register addresses. The system has an instruction set having a plurality of instructions. Each instruction has a plurality of bits wherein none of the plurality of bits in each of the plurality of instructions are dedicated bits for implementing different addressing modes. Each of the plurality of instructions are able to implement different addressing modes by addressing the virtual register addresses in the processor architecture scheme. Since no bits are required for implementing different addressing modes, the length of the opcode field and the register address field are determined by the number of opcodes and the number of addressable registers the user wishes to implement.

Abstract: A method and apparatus for optimizing sample fetching in a peripheral component interconnect (PCI) environment. In one embodiment the present invention generates a sample page base address corresponding to a first part of a first address received from a digital signal processor (DSP). The generated sample page base address is then stored in a sample page base address cache which can be accessed by the DSP without accessing a PCI bus. The first part of the first address is compared to a first part of a second address. Provided that the first part of the first address and the first part of the second address are the same, the present invention combines a second portion of the second address sent from the DSP with the generated sample page base address stored in the sample page base address cache. In so doing, the present invention generates a complete address of a sample to be fetched without accessing the PCI bus.

Abstract: A one-chip microprocessor, in which a built-in cache memory unit 20 has parities, and a cache parity generating & checking unit 21 checks parity of data read from the built-in cache memory unit 20, and when parity error is detected, outputs an internal cache parity error signal 50 to an instruction execution unit 23. By this, the instruction execution unit 23 suspends instruction execution and outputs a processor error signal 37. Accordingly, by checking parity errors of data of built-in memory and inputted address/data, instruction execution are immediately suspended to limit malfunction at least, thereby improving reliability. And by storing kinds of bus operation and errors in a register at the time of error generation, restoring possibility of system level is improved.

Abstract: An apparatus for and method of providing rapid communication between separately clocked system elements. A network interface module is used as the overall system control and communication interface to each of the separate system elements. Each of these system elements is controlled by a different and dedicated programmable micro-engine. A separate register located within and addressable by each of the micro-engines provides the basic data transfer path. Access by a micro-engine to the corresponding register is easily accomplished by firmware. The bit serial scan interface between the network interface module and each of the registers is controlled by the network interface module.

Abstract: The present invention relates to multiprocessors which has several microprocessors on a single chip. Efficiency is improved by stripping certain functions that are used less freely from the microprocessor and sharing these functions between several symmetric microprocessors. This method allows each CPU to occupy a smaller area while preserving complete symmetry of capability for software simplification. For example, the shared execution units can include the floating point unit and multimedia execution units.

Abstract: A processor architecture is described which operates with improved computational efficiency using instruction fetching functions that are decoupled from instruction execution functions by a dynamic register file. The instruction fetching function operates in free-running mode which does not stop if a fetched instruction cannot be executed due to data being unavailable or due to other instruction dependencies. Branch instructions are taken in a predicted direction and the results of execution of all instructions are provisionally stored pending validation or invalidation on the basis of the dependencies becoming available later.

Abstract: A microcomputer that is easy to use and connected direct to such memories as dynamic and static RAM's and to other peripheral circuits. The microcomputer has strobe signal output terminals CASH*, CASL* and RAS* for direct connection to a dynamic RAM, and chip select signal output terminals CS0* through CS6* for outputting a chip select signal in parallel with the output from the strobe signal output terminals. The microcomputer further includes address output terminals for outputting a non-multiplexed or multiplexed address signal as needed, and data I/O terminals for selectively outputting the address signal to comply with a multiple-bus interface scheme.

Abstract: An improved PC system that includes a main CPU microprocessor, a file-based operating system, and a DSP microprocessor arranged so that the DSP can execute main CPU operations during time intervals in which the main CPU is otherwise occupied, thereby increasing the bandwidth of the system is provided. This PC system may include multiple CPUs and/or multiple DSPs.

Abstract: An image of a workpiece is taken by a CCD camera, which includes a view field covering a supply range of the workpiece, and is transmitted to an image processor to detect the position and posture of the workpiece. A robot controller controls a robot to approach a position from another position based on the result of the detection. Then, a measurement using a laser sensor is started. Scanning laser beams projected from an emitter are detected by a light detector and the detected data are analyzed by the image processor. When the robot reaches a third position, the measurement is terminated. The laser sensor may be of a split-light-projection type. For the approaching movement, the image taken by the camera is displayed on a display device to perform a jogfeed of the robot on the screen.

Abstract: A superscalar microprocessor defines a hierarchical structure of registers. The top level of the hierarchy includes performance critical registers and pointers to other levels of the hierarchy. A second level of the hierarchy may include special registers. Special registers may include arrays or groups of data. Special registers may be located in a special register file or remotely located throughout the microprocessor. Remote special registers are accessed via a special register bus. Resources throughout the microprocessor are defined as special registers. In this manner, resources throughout the microprocessor are accessed using special register move instructions that are handled in a manner similar to other register moves in instructions. Accordingly, adding and modifying resources within the microprocessor is transparent to the majority of the circuitry of the microprocessor. Thus, the present invention provides a uniform and flexible mechanism of communicating to resources of a microprocessor.

Abstract: The present invention provides a system and method for managing load and store operations necessary for reading from and writing to memory or I/O in a superscalar RISC architecture environment. To perform this task, a load/store unit is provided whose main purpose is to make load requests out-of-order whenever possible to get the load data back for use by an instruction execution unit as quickly as possible. A load operation can only be performed out-of-order if there are no address collisions and no write pendings. An address collision occurs when a read is requested at a memory location where an older instruction will be writing. Write pending refers to the case where an older instruction requests a store operation, but the store address has not yet been calculated. The data cache unit returns 8 bytes of unaligned data. The load/store unit aligns this data properly before it is returned to the instruction execution unit.

Abstract: A processor that executes coded instructions using an instruction scheduling unit receiving the coded instructions and issuing an instruction for execution. A replay signaling device generates a signal indicating when the instruction failed to execute properly within a predetermined time. A replay device within the instruction scheduling unit responsive to the signaling device then reissues the instruction for execution.

Abstract: The invention in several embodiments includes an apparatus and a method for predicting whether store instructions can be safely executed out-of-order. The apparatus, includes at least one execution unit, a reorder buffer adapted to holding a plurality of instructions from an instruction sequence for execution by the execution units, and a memory storage device adapted to holding a collision history table. The collision history table has entries for load instructions of the instruction sequence Each of the entries is adapted to predicting when the associated load instruction is colliding.