Patent application number | Description | Published |
20090231935 | MEMORY WITH WRITE PORT CONFIGURED FOR DOUBLE PUMP WRITE - A memory with a write port configured for double-pump writes. The memory includes a first and second memory locations each having one or more bit cells, and one or more bit lines each coupled to corresponding ones of the bit cells. A write port is coupled to each of the bit lines. Selection circuitry, responsive to a first clock edge, latches first data from a first data path through the write port, and responsive to a second clock edge, latches second data from a second data path through the write port. A first pulse is generated during a first phase of the clock signal to cause writing of the first data into the first memory location. A second pulse is generated during a second phase of the clock signal to cause writing of the second data into the second memory location. | 09-17-2009 |
20100169611 | BRANCH MISPREDICTION RECOVERY MECHANISM FOR MICROPROCESSORS - A system and method for reducing branch misprediction penalty. In response to detecting a mispredicted branch instruction, circuitry within a microprocessor identifies a predetermined condition prior to retirement of the branch instruction. Upon identifying this condition, the entire corresponding pipeline is flushed prior to retirement of the branch instruction, and instruction fetch is started at a corresponding address of an oldest instruction in the pipeline immediately prior to the flushing of the pipeline. The correct outcome is stored prior to the pipeline flush. In order to distinguish the mispredicted branch from other instructions, identification information may be stored alongside the correct outcome. One example of the predetermined condition being satisfied is in response to a timer reaching a predetermined threshold value, wherein the timer begins incrementing in response to the mispredicted branch detection and resets at retirement of the mispredicted branch. | 07-01-2010 |
20100250966 | PROCESSOR AND METHOD FOR IMPLEMENTING INSTRUCTION SUPPORT FOR HASH ALGORITHMS - A processor including instruction support for implementing hash algorithms may issue, for execution, programmer-selectable hash instructions from a defined instruction set architecture (ISA). The processor may include a cryptographic unit that may receive instructions for execution. The instructions include hash instructions defined within the ISA. In addition, the hash instructions may be executable by the cryptographic unit to implement a hash that is compliant with one or more respective hash algorithm specifications. In response to receiving a particular hash instruction defined within the ISA, the cryptographic unit may retrieve a set of input data blocks from a predetermined set of architectural registers of the processor, and generate a hash value of the set of input data blocks according to a hash algorithm that corresponds to the particular hash instruction. | 09-30-2010 |
20100274961 | PHYSICALLY-INDEXED LOGICAL MAP TABLE - Techniques and systems are described herein to maintain a mapping of logical to physical registers—for example, in the context of a multithreaded processor that supports renaming. A mapping unit may have a plurality of entries, each of which stores rename information for a dedicated one of a set of physical registers available to the processor for renaming. This physically-indexed mapping unit may support multiple threads, and may comprise a content-addressable memory (CAM) in certain embodiments. The mapping unit may support various combinations of read operations (to determine if a logical register is mapped to a physical register), write operations (to create or modify one or more entries containing mapping information), thread flush operations, and commit operations. More than one of such operations may be performed substantially simultaneously in certain embodiments. | 10-28-2010 |
20100274993 | LOGICAL MAP TABLE FOR DETECTING DEPENDENCY CONDITIONS - Techniques and structures are described which allow the detection of certain dependency conditions, including evil twin conditions, during the execution of computer instructions. Information used to detect dependencies may be stored in a logical map table, which may include a content-addressable memory. The logical map table may maintain a logical register to physical register mapping, including entries dedicated to physical registers available as rename registers. In one embodiment, each entry in the logical map table includes a first value usable to indicate whether only a portion of the physical register is valid and whether the physical register includes the most recent update to the logical register being renamed. Use of this first value may allow precise detection of dependency conditions, including evil twin conditions, upon an instruction reading from at least two portions of a logical register having an entry in the logical map table whose first value is set. | 10-28-2010 |
20100274994 | PROCESSOR OPERATING MODE FOR MITIGATING DEPENDENCY CONDITIONS - Various techniques for mitigating dependencies between groups of instructions are disclosed. In one embodiment, such dependencies include “evil twin” conditions, in which a first floating-point instruction has as a destination a first portion of a logical floating-point register (e.g., a single-precision write), and in which a second, subsequent floating-point instruction has as a source the first portion and a second portion of the same logical floating-point register (e.g., a double-precision read). The disclosed techniques may be applicable in a multithreaded processor implementing register renaming. In one embodiment, a processor may enter an operating mode in which detection of evil twin “producers” (e.g., single-precision writes) causes the instruction sequence to be modified to break potential dependencies. Modification of the instruction sequence may continue until one or more exit criteria are reached (e.g., committing a predetermined number of single-precision writes). This operating mode may be employed on a per-thread basis. | 10-28-2010 |
20100299499 | DYNAMIC ALLOCATION OF RESOURCES IN A THREADED, HETEROGENEOUS PROCESSOR - Systems and methods for efficient dynamic utilization of shared resources in a processor. A processor comprises a front end pipeline, an execution pipeline, and a commit pipeline, wherein each pipeline comprises a shared resource with entries configured to be allocated for use in each clock cycle by each of a plurality of threads supported by the processor. To avoid starvation of any active thread, the processor further comprises circuitry configured to ensure each active thread is able to allocate at least a predetermined quota of entries of each shared resource. Each pipe stage of a total pipeline for the processor may include at least one dynamically allocated shared resource configured not to starve any active thread. Dynamic allocation of shared resources between a plurality of threads may yield higher performance over static allocation. In addition, dynamic allocation may require relatively little overhead for activation/deactivation of threads. | 11-25-2010 |
20100306510 | SINGLE CYCLE DATA MOVEMENT BETWEEN GENERAL PURPOSE AND FLOATING-POINT REGISTERS - Systems and methods for providing single cycle movement of data between a floating-point register file (FRF) and a general purpose or integer register file (RF) of a microprocessor system are provided. The system may include an integer execution unit operative to execute instructions with single cycle latency, a floating-point execution unit, a working register file (WRF), an FRF, and an IRF. To achieve the single cycle movement functionality, the integer execution unit may physically own the WRF, IRF, and FRF, and may monitor and control any dependencies between them. Thus, since the integer execution unit has direct read access to both the IRF and the FRF, data may be moved between the two register files using the single cycle operation of the integer execution unit, without the need to store and load the data from memory. | 12-02-2010 |
20100318998 | System and Method for Out-of-Order Resource Allocation and Deallocation in a Threaded Machine - A system and method for managing the dynamic sharing of processor resources between threads in a multi-threaded processor are disclosed. Out-of-order allocation and deallocation may be employed to efficiently use the various resources of the processor. Each element of an allocate vector may indicate whether a corresponding resource is available for allocation. A search of the allocate vector may be performed to identify resources available for allocation. Upon allocation of a resource, a thread identifier associated with the thread to which the resource is allocated may be associated with the allocate vector entry corresponding to the allocated resource. Multiple instances of a particular resource type may be allocated or deallocated in a single processor execution cycle. Each element of a deallocate vector may indicate whether a corresponding resource is ready for deallocation. Examples of resources that may be dynamically shared between threads are reorder buffers, load buffers and store buffers. | 12-16-2010 |
20100325188 | PROCESSOR AND METHOD FOR IMPLEMENTING INSTRUCTION SUPPORT FOR MULTIPLICATION OF LARGE OPERANDS - A processor including instruction support for implementing large-operand multiplication may issue, for execution, programmer-selectable instructions from a defined instruction set architecture (ISA). The processor may include an instruction execution unit comprising a hardware multiplier datapath circuit, where the hardware multiplier datapath circuit is configured to multiply operands having a maximum number of bits M. In response to receiving a single instance of a large-operand multiplication instruction defined within the ISA, wherein at least one of the operands of the large-operand multiplication instruction includes more than the maximum number of bits M, the instruction execution unit is configured to multiply operands of the large-operand multiplication instruction within the hardware multiplier datapath circuit to determine a result of the large-operand multiplication instruction without execution of programmer-selected instructions within the ISA other than the large-operand multiplication instruction. | 12-23-2010 |
20100325394 | System and Method for Balancing Instruction Loads Between Multiple Execution Units Using Assignment History - A system and method for balancing instruction loads between multiple execution units are disclosed. One or more execution units may be represented by a slot configured to accept instructions on behalf of the execution unit(s). A decode unit may assign instructions to a particular slot for subsequent scheduling for execution. Slot assignments may be made based on an instruction's type and/or on a history of previous slot assignments. A cumulative slot assignment history may be maintained in a bias counter, the value of which reflects the bias of previous slot assignments. Slot assignments may be determined based on the value of the bias counter, in order to balance the instruction load across all slots, and all execution units. The bias counter may reflect slot assignments made only within a desired historical window. A separate data structure may store data reflecting the actual slot assignments made during the desired historical window. | 12-23-2010 |
20100332787 | System and Method to Manage Address Translation Requests - A system and method for servicing translation lookaside buffer (TLB) misses may manage separate input and output pipelines within a memory management unit. A pending request queue (PRQ) in the input pipeline may include an instruction-related portion storing entries for instruction TLB (ITLB) misses and a data-related portion storing entries for potential or actual data TLB (DTLB) misses. A DTLB PRQ entry may be allocated to each load/store instruction selected from the pick queue. The system may select an ITLB- or DTLB-related entry for servicing dependent on prior PRQ entry selection(s). A corresponding entry may be held in a translation table entry return queue (TTERQ) in the output pipeline until a matching address translation is received from system memory. PRQ and/or TTERQ entries may be deallocated when a corresponding TLB miss is serviced. PRQ and/or TTERQ entries associated with a thread may be deallocated in response to a thread flush. | 12-30-2010 |
20100332804 | UNIFIED HIGH-FREQUENCY OUT-OF-ORDER PICK QUEUE WITH SUPPORT FOR SPECULATIVE INSTRUCTIONS - Systems and methods for efficient picking of instructions for out-of-order issue and execution in a processor. In one embodiment, a processor comprises a unified pick queue that is dynamically allocated. Each entry is configured to store age and dependency information relative to other decoded instructions. Also, each entry stores a picked field, which when asserted indicates the decoded instruction has already been picked for out-of-order issue and execution. When asserted, a trigger field indicates a result of a corresponding decoded instruction will be available a predetermined number of clock cycles afterward. A younger instruction dependent on a result of an older instruction is ready to be picked before the result of the older instruction is available. In this case, the older instruction has asserted picked and trigger fields. | 12-30-2010 |
20100332806 | DEPENDENCY MATRIX FOR THE DETERMINATION OF LOAD DEPENDENCIES - Systems and methods for identification of dependent instructions on speculative load operations in a processor. A processor allocates entries of a unified pick queue for decoded and renamed instructions. Each entry of a corresponding dependency matrix is configured to store a dependency bit for each other instruction in the pick queue. The processor speculates that loads will hit in the data cache, hit in the TLB and not have a read after write (RAW) hazard. For each unresolved load, the pick queue tracks dependent instructions via dependency vectors based upon the dependency matrix. If a load speculation is found to be incorrect, dependent instructions in the pick queue are reset to allow for subsequent picking, and dependent instructions in flight are canceled. On completion of a load miss, dependent operations are re-issued. On resolution of a TLB miss or RAW hazard, the original load is replayed and dependent operations are issued again from the pick queue. | 12-30-2010 |
20100333098 | DYNAMIC TAG ALLOCATION IN A MULTITHREADED OUT-OF-ORDER PROCESSOR - Various techniques for dynamically allocating instruction tags and using those tags are disclosed. These techniques may apply to processors supporting out-of-order execution and to architectures that supports multiple threads. A group of instructions may be assigned a tag value from a pool of available tag values. A tag value may be usable to determine the program order of a group of instructions relative to other instructions in a thread. After the group of instructions has been (or is about to be) committed, the tag value may be freed so that it can be re-used on a second group of instructions. Tag values are dynamically allocated between threads; accordingly, a particular tag value or range of tag values is not dedicated to a particular thread. | 12-30-2010 |
20110029978 | DYNAMIC MITIGATION OF THREAD HOGS ON A THREADED PROCESSOR - Systems and methods for efficient thread arbitration in a processor. A processor comprises a multi-threaded resource. The resource may include an array 8of entries which may be allocated by threads. A thread arbitration table corresponding to a given thread stores a high and a low threshold value in each table entry. A thread history shift register (HSR) indexes the table, wherein each bit of the HSR indicates whether the given thread is a thread hog. When the given thread has more allocated entries in the array than the high threshold of the table entry, the given thread is stalled from further allocating array entries. Similarly, when the given thread has fewer allocated entries in the array than the low threshold of the selected table entry, the given thread is permitted to allocate entries. In this manner, threads that hog dynamic resources can be mitigated such that more resources are available to other threads that are not thread hogs. This can result in a significant increase in overall processor performance. | 02-03-2011 |
20110078414 | MULTIPORTED REGISTER FILE FOR MULTITHREADED PROCESSORS AND PROCESSORS EMPLOYING REGISTER WINDOWS - A processor includes an instruction fetch unit configured to issue instructions for execution, where the instructions are selected from a number of threads, where each given instruction has a corresponding thread identifier, and where at least some of the instructions specify operand(s) via register identifiers. A register file stores operands usable by the instructions, and may include several banks, each corresponding to a register identifiers and including several entries corresponding to the several threads, wherein the entries are configured to store data values. In response to receiving a request to read a particular register identifier for a given thread identifier, the register file may be configured to decode the given thread identifier to retrieve entries from the banks that correspond to the given thread identifier. The register file may further select, from among the retrieved entries, a data value corresponding to the particular register identifier to be output. | 03-31-2011 |
20110078697 | OPTIMAL DEALLOCATION OF INSTRUCTIONS FROM A UNIFIED PICK QUEUE - Systems and methods for efficient out-of-order dynamic deallocation of entries within a shared storage resource in a processor. A processor comprises a unified pick queue that includes an array configured to dynamically allocate any entry of a plurality of entries for a decoded and renamed instruction. This instruction may correspond to any available active threads supported by the processor. The processor includes circuitry configured to determine whether an instruction corresponding to an allocated entry of the plurality of entries is dependent on a speculative instruction and whether the instruction has a fixed instruction execution latency. In response to determining the instruction is not dependent on a speculative instruction, the instruction has a fixed instruction execution latency, and said latency has transpired, the circuitry may deallocate the instruction from the allocated entry. | 03-31-2011 |
20110087866 | PERCEPTRON-BASED BRANCH PREDICTION MECHANISM FOR PREDICTING CONDITIONAL BRANCH INSTRUCTIONS ON A MULTITHREADED PROCESSOR - A multithreaded microprocessor includes an instruction fetch unit including a perceptron-based conditional branch prediction unit configured to provide, for each of one or more concurrently executing threads, a direction branch prediction. The conditional branch prediction unit includes a plurality of storages each including a plurality of entries. Each entry may be configured to store one or more prediction values. Each prediction value of a given storage may correspond to at least one conditional branch instruction in a cache line. The conditional branch prediction unit may generate a separate index value for accessing each storage by generating a first index value for accessing a first storage by combining one or more portions of a received instruction fetch address, and generating each other index value for accessing the other storages by combining the first index value with a different portion of direction branch history information. | 04-14-2011 |
20110087895 | APPARATUS AND METHOD FOR LOCAL OPERAND BYPASSING FOR CRYPTOGRAPHIC INSTRUCTIONS - A processor may include a hardware instruction fetch unit configured to issue instructions for execution, and a hardware functional unit configured to receive instructions for execution, where the instructions include cryptographic instruction(s) and non-cryptographic instruction(s). The functional unit may include a cryptographic execution pipeline configured to execute the cryptographic instructions with a corresponding cryptographic execution latency, and a non-cryptographic execution pipeline configured to execute the non-cryptographic instructions with a corresponding non-cryptographic execution latency that is longer than the cryptographic execution latency. The functional unit may further include a local bypass network configured to bypass results produced by the cryptographic execution pipeline to dependent cryptographic instructions executing within the cryptographic execution pipeline, such that each instruction within a sequence of dependent cryptographic instructions is executable with the cryptographic execution latency, and where the results of the cryptographic execution pipeline are not bypassed to any other functional unit within the processor. | 04-14-2011 |
20110138153 | MECHANISM FOR SELECTING INSTRUCTIONS FOR EXECUTION IN A MULTITHREADED PROCESSOR - In one embodiment, a multithreaded processor includes a plurality of buffers, each configured to store instructions corresponding to a respective thread. The multithreaded processor also includes a pick unit coupled to the plurality of buffers. The pick unit may pick from at least one of the buffers in a given cycle, a valid instruction based upon a thread selection algorithm. The pick unit may further cancel, in the given cycle, the picking of the valid instruction in response to receiving a cancel indication. | 06-09-2011 |
20110276783 | THREAD FAIRNESS ON A MULTI-THREADED PROCESSOR WITH MULTI-CYCLE CRYPTOGRAPHIC OPERATIONS - Systems and methods for efficient execution of operations in a multi-threaded processor. Each thread may include a blocking instruction. A blocking instruction blocks other threads from utilizing hardware resources for an appreciable amount of time. One example of a blocking type instruction is a Montgomery multiplication cryptographic instruction. Each thread can operate in a thread-based mode that allows the insertion of stall cycles during the execution of blocking instructions, during which other threads may utilize the previously blocked hardware resources. At times when multiple threads are scheduled to execute blocking instructions, the thread-based mode may be changed to increase throughput for these multiple threads. For example, the mode may be changed to disallow the insertion of stall cycles. Therefore, the time for sequential operation of the blocking instructions corresponding to the multiple threads may be reduced. | 11-10-2011 |
20120233441 | MULTI-THREADED INSTRUCTION BUFFER DESIGN - An instruction buffer for a processor configured to execute multiple threads is disclosed. The instruction buffer is configured to receive instructions from a fetch unit and provide instructions to a selection unit. The instruction buffer includes one or more memory arrays comprising a plurality of entries configured to store instructions and/or other information (e.g., program counter addresses). One or more indicators are maintained by the processor and correspond to the plurality of threads. The one or more indicators are usable such that for instructions received by the instruction buffer, one or more of the plurality entries of a memory array can be determined as a write destination for the received instructions, and for instructions to be read from the instruction buffer (and sent to a selection unit), one or more entries can be determined as the correct source location from which to read. | 09-13-2012 |
20130290675 | MITIGATION OF THREAD HOGS ON A THREADED PROCESSOR - Systems and methods for efficient thread arbitration in a threaded processor with dynamic resource allocation. A processor includes a resource shared by multiple threads. The resource includes an array with multiple entries, each of which may be allocated for use by any thread. Control logic detects a load miss to memory, wherein the miss is associated with a latency greater than a given threshold. The load instruction or an immediately younger instruction is selected for replay for an associated thread. A pipeline flush and replay for the associated thread begins with the selected instruction. Instructions younger than the load instruction are held at a given pipeline stage until the load instruction completes. During replay, this hold prevents resources from being allocated to the associated thread while the load instruction is being serviced. | 10-31-2013 |
20130297910 | MITIGATION OF THREAD HOGS ON A THREADED PROCESSOR USING A GENERAL LOAD/STORE TIMEOUT COUNTER - Systems and methods for efficient thread arbitration in a threaded processor with dynamic resource allocation. A processor includes a resource shared by multiple threads. The resource includes entries which may be allocated for use by any thread. Control logic detects long latency instructions. Long latency instructions have a latency greater than a given threshold. One example is a load instruction that has a read-after-write (RAW) data dependency on a store instruction that misses a last-level data cache. The long latency instruction or an immediately younger instruction is selected for replay for an associated thread. A pipeline flush and replay for the associated thread begins with the selected instruction. Instructions younger than the long latency instruction are held at a given pipeline stage until the long latency instruction completes. During replay, this hold prevents resources from being allocated to the associated thread while the long latency instruction is being serviced. | 11-07-2013 |