Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberUS20020013861 A1
Publication typeApplication
Application numberUS 09/867,064
Publication date31 Jan 2002
Filing date29 May 2001
Priority date28 Dec 1999
Also published asCN1643499A, CN100351798C, DE60030767D1, DE60030767T2, EP1242883A2, EP1242883B1, US6625654, US7111296, US20040098496, WO2001048606A2, WO2001048606A3
Publication number09867064, 867064, US 2002/0013861 A1, US 2002/013861 A1, US 20020013861 A1, US 20020013861A1, US 2002013861 A1, US 2002013861A1, US-A1-20020013861, US-A1-2002013861, US2002/0013861A1, US2002/013861A1, US20020013861 A1, US20020013861A1, US2002013861 A1, US2002013861A1
InventorsMatthew Adiletta, Debra Bernstein, Gilbert Wolrich, William Wheeler
Original AssigneeIntel Corporation
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Method and apparatus for low overhead multithreaded communication in a parallel processing environment
US 20020013861 A1
Abstract
A parallel hardware-based multithreaded processor is described. The processor includes a general purpose processor that coordinates system functions and a plurality of microengines that support multiple program threads. The processor also includes a memory control system that has a first memory controller that sorts memory references based on whether the memory references are directed to an even bank or an odd bank of memory and a second memory controller that optimizes memory references based upon whether the memory references are read references or write references. A program thread communication scheme for packet processing is also described.
Images(21)
Previous page
Next page
Claims(14)
What is claimed is:
1. A method of inter-thread communication in a multi-threaded computer comprises:
storing an inter-thread message in memory, the inter-thread message having a field for an address that indicates a location of data for a next thread to execute; and
writing to a self-destruct register after storing the message to indicate that a thread which stored the message in memory has completed execution, with the self-destruct register being cleared upon reading by the next thread.
2. The method of claim 1 wherein the inter-thread message field for an address provides an address of a register where the data for the next executing thread is stored.
3. The method of claim 1 wherein writing to a self-destruct register further comprises:
setting at least one bit in the self-destruct register which corresponds to the thread which is writing to the self-destruct register.
4. The method of claim 1 wherein writing to a self-destruct register further comprises:
setting one bit in the self-destruct register which corresponds to the thread which is writing to the self-destruct register.
5. The method of claim 1 further comprising:
writing to the self-destruct register by a first thread;
reading from the self-destruct register by a second thread, where the reading further comprises:
reading bits, if any, that are set in the self-destruct register; and
clearing all of the bits of the self-destruct register.
6. The method of claim 5 further comprises:
reading the inter-thread message from the memory by a new thread, where the new thread is a thread other than the first thread; and
executing the new thread.
7. A hardware-based multi-threaded processor comprises:
a general purpose processor that coordinates system functions;
a plurality of microengines that support multiple thread execution;
a scratchpad memory for storing inter-thread messages where execution of a write to the scratchpad memory by a first thread causes an address to be stored as an inter-thread message which indicates a location of data for a new thread; and
a self-destruct register for indicating the execution status of threads where reading of the self-destruct register clears all of the bits of the self-destruct register.
8. The processor of claim 7 wherein the plurality of microengines further comprise:
a register stack wherein execution of the write to the scratchpad memory by the first thread causes a register address of the register referenced by the write to be stored as the inter-thread message, with the register address indicating the register stack location of data for the new thread.
9. The processor of claim 7 wherein execution of a write to the self-destruct register by a thread causes at least one bit to be set in the self-destruct register, the bit being set corresponding to the thread which executed the write to the self-destruct register.
10. The processor of claim 7 wherein execution of a write to the self-destruct register by a thread causes one bit to be set in the self-destruct register, the bit set corresponding to the first thread which executed the write to the self-destruct register.
12. The processor of claim 11 wherein the read from the self-destruct register by the thread causes execution of a new thread for each bit that is set, if any, in the self-destruct register.
13. A computer program product residing on a computer readable medium causing a processor to perform a function comprises instructions causing the processor to:
store an inter-thread message in memory; and
set at least one bit in a self-destruct register.
14. The computer program of claim 13 further comprises instruction causing the processor to:
store a register address as the inter-thread message in memory.
15. The computer program of claim 13 further comprises instructions causing the processor to:
read the contents of the self-destruct register, the read also clearing the self-destruct register; and
execute a new thread according to any bits set in the self-destruct register.
Description
  • [0001]
    This application is a continuation application of and claims priority to U.S. application Ser. No. 09/473,799 filed on Dec. 28, 1999.
  • BACKGROUND
  • [0002]
    This invention relates to network packet processing.
  • [0003]
    Parallel processing is an efficient form of information processing of concurrent events in a computing process. Parallel processing demands concurrent execution of many programs in a computer, in contrast to sequential processing. In the context of a parallel processor, parallelism involves doing more than one thing at the same time. Unlike a serial paradigm where all tasks are performed sequentially at a single station or a pipelined machine where tasks are performed at specialized stations, with parallel processing, a plurality of stations are provided with each capable of performing all tasks. That is, in general all or a plurality of the stations work simultaneously and independently on the same or common elements of a problem. Certain problems are suitable for solution by applying parallel processing.
  • DESCRIPTION OF DRAWINGS
  • [0004]
    [0004]FIG. 1 is a block diagram of a communication system employing a hardware-based multithreaded processor.
  • [0005]
    [0005]FIG. 2 is a detailed block diagram of the hardware-based multithreaded processor of FIG. 1.
  • [0006]
    [0006]FIG. 3 is a block diagram of a microengine functional unit employed in the hardware-based multithreaded processor of FIGS. 1 and 2.
  • [0007]
    [0007]FIG. 4 is a block diagram of a memory controller for enhanced bandwidth operation used in the hardware-based multithreaded processor.
  • [0008]
    [0008]FIG. 5 is a block diagram of a memory controller for latency limited operations used in the hardware-based multithreaded processor.
  • [0009]
    [0009]FIG. 6 is a block diagram of a communication bus interface in the processor of FIG. 1 depicting hardware used in program thread signaling.
  • [0010]
    FIGS. 7A-7B are a pictorial representation and flow chart useful in understanding program thread signaling with a clear on read register.
  • [0011]
    [0011]FIG. 8 is a flow chart of an inter-thread signaling scheme.
  • [0012]
    [0012]FIG. 9 is a flow chart of a program thread status reporting process.
  • DETAILED DESCRIPTION
  • [0013]
    Architecture:
  • [0014]
    Referring to FIG. 1, a communication system 10 includes a parallel, hardware-based multithreaded processor 12. The hardware-based multithreaded processor 12 is coupled to a bus such as a Peripheral Component Interconnect (PCI) bus 14, a memory system 16 and a second bus 18. The system 10 is especially useful for tasks that can be broken into parallel subtasks or functions. Specifically hardware-based multithreaded processor 12 is useful for tasks that are bandwidth oriented rather than latency oriented. The hardware-based multithreaded processor 12 has multiple microengines 22 each with multiple hardware controlled program threads that can be simultaneously active and independently work on a task.
  • [0015]
    The hardware-based multithreaded processor 12 also includes a central controller 20 that assists in loading microcode control for other resources of the hardware-based multithreaded processor 12 and performs other general purpose computer type functions such as handling protocols, exceptions, extra support for packet processing where the microengines pass the packets off for more detailed processing such as in boundary conditions. In one embodiment, the processor 20 is a Strong Arm 7 (Arm is a trademark of ARM Limited, United Kingdom) based architecture. The general purpose microprocessor 20 has an operating system. Through the operating system the processor 20 can call functions to operate on microengines 22 a-22 f. The processor 20 can use any supported operating system preferably a real time operating system. For the core processor implemented as a Strong Arn architecture, operating systems such as, Microsoft NT real-time, VXWorks and μCUS, a freeware operating system available over the Internet, can be used.
  • [0016]
    The hardware-based multithreaded processor 12 also includes a plurality of microengines 22 a-22 f. Microengines 22 a-22 f each maintain a plurality of program counters in hardware and states associated with the program counters. Effectively, a corresponding plurality of sets of program threads can be simultaneously active on each of the microengines 22 a-22 f while only one is actually operating at any one time.
  • [0017]
    In one embodiment, there are six microengines 22 a-22 f, each having capabilities for processing four hardware program threads. The six microengines 22 a-22 f operate with shared resources including memory system 16 and bus interfaces 24 and 28. The memory system 16 includes a Synchronous Dynamic Random Access Memory (SDRAM) controller 26 a and a Static Random Access Memory (SRAM) controller 26 b. SDRAM memory 16 a and SDRAM controller 26 a are typically used for processing large volumes of data, e.g., processing of network payloads from network packets. The SRAM controller 26 b and SRAM memory 16 b are used in a networking implementation for low latency, fast access tasks, e.g., accessing look-up tables, memory for the core processor 20, and so forth.
  • [0018]
    Hardware context swapping enables other contexts with unique program counters to execute in the same microengine. Hardware context swapping also synchronizes completion of tasks. For example, two program threads could request the same shared resource e.g., SRAM. Each one of these separate functional units, e.g., the FBUS interface 28, the SRAM controller 26 a, and the SDRAM controller 26 b, when they complete a requested task from one of the microengine program thread contexts reports back a flag signaling completion of an operation. When the flag is received by the microengine, the microengine can determine which program thread to turn on.
  • [0019]
    As a network processor, e.g., a router, the hardware-based multithreaded processor 12 interfaces to network devices such as a media access controller device e.g., a 10/100BaseT Octal MAC 13 a or a Gigabit Ethernet device 13 b. In general, as a network processor, the hardware-based multithreaded processor 12 can interface to any type of communication device or interface that receives/sends large amounts of data. The network processor can function as a router 10 in a networking application route network packets amongst devices 13 a, 13 b in a parallel manner. With the hardware-based multithreaded processor 12, each network packet can be independently processed.
  • [0020]
    The processor 12 includes a bus interface 28 that couples the processor to the second bus 18. Bus interface 28 in one embodiment couples the processor 12 to the so-called FBUS 18 (FIFO bus). The FBUS interface 28 is responsible for controlling and interfacing the processor 12 to the FBUS 18. The FBUS 18 is a 64-bit wide FIFO bus, used to interface to Media Access Controller (MAC) devices. The processor 12 includes a second interface e.g., a PCI bus interface 24 that couples other system components that reside on the PCI 14 bus to the processor 12.
  • [0021]
    The functional units are coupled to one or more internal buses. The internal buses are dual, 32 bit buses (i.e., one bus for read and one for write). The hardware-based multithreaded processor 12 also is constructed such that the sum of the bandwidths of the internal buses in the processor 12 exceed the bandwidth of external buses coupled to the processor 12. The processor 12 includes an internal core processor bus 32, e.g., an ASB bus (Advanced System Bus) that couples the processor core 20 to the memory controllers 26 a, 26 b and to an ASB translator 30 described below. The ASB bus is a subset of the so called AMBA bus that is used with the Strong Arm processor core. The processor 12 also includes a private bus 34 that couples the microengine units to SRAM controller 26 b, ASB translator 30 and FBUS interface 28. A memory bus 38 couples the memory controller 26 a, 26 b to the bus interfaces 24 and 28 and memory system 16 including flashrom 16 c used for boot operations and so forth.
  • [0022]
    Referring to FIG. 2, each of the microengines 22 a-22 f includes an arbiter that examines flags to determine the available program threads to be operated upon. Any program thread from any of the microengines 22 a-22 f can access the SDRAM controller 26 a, SDRAM controller 26 b or FBUS interface 28. The SDRAM controller 26 a and SDRAM controller 26 b each include a plurality of queues to store outstanding memory reference requests. The queues either maintain order of memory references or arrange memory references to optimize memory bandwidth.
  • [0023]
    If the memory subsystem 16 is flooded with memory requests that are independent in nature, the processor 12 can perform memory reference sorting. Memory reference sorting reduces dead time or a bubble that occurs with accesses to SRAM. Memory reference sorting allows the processor 12 to organize references to memory such that long strings of reads can be followed by long strings of writes.
  • [0024]
    Reference sorting helps maintain parallel hardware context program threads. Reference sorting allows hiding of pre-charges from one SDRAM bank to another. If the memory system 16 b is organized into an odd bank and an even bank, while the processor is operating on the odd bank, the memory controller 26 b can start precharging the even bank. Precharging is possible if memory references alternate between odd and even banks. By ordering memory references to alternate accesses to opposite banks, the processor 12 improves SDRAM bandwidth. Additionally, other optimizations can be used. For example, merging optimizations where operations that can be merged, are merged prior to memory access, open page optimizations where by examining addresses, an opened page of memory is not reopened, chaining which allows for special handling of contiguous memory references and refreshing mechanisms, can be employed.
  • [0025]
    The FBUS interface 28 supports Transmit and Receive flags for each port that a MAC device supports, along with an Interrupt flag indicating when service is warranted. The FBUS interface 28 also includes a controller 28 a that performs header processing of incoming packets from the FBUS 18. The controller 28 a extracts the packet headers and performs a microprogrammable source/destination/protocol hashed lookup (used for address smoothing) in SRAM. If the hash does not successfully resolve, the packet header is sent to the processor core 20 for additional processing. The FBUS interface 28 supports the following internal data transactions:
    FBUS unit (Shared bus SRAM) to/from microengine.
    FBUS unit (via private bus) writes from SDRAM Unit.
    FBUS unit (via Mbus) Reads to SDRAM.
  • [0026]
    The FBUS 18 is a standard industry bus and includes a data bus, e.g., 64 bits wide and sideband control for address and read/write control. The FBUS interface 28 provides the ability to input large amounts of data using a series of input and output FIFO=s 29 a-29 b. From the FIFOs 29 a-29 b, the microengines 22 a-22 f fetch data from or command the SDRAM controller 26 a to move data from a receive FIFO in which data has come from a device on bus 18, into the FBUS interface 28. The data can be sent through memory controller 26 a to SDRAM memory 16 a, via a direct memory access. Similarly, the microengines can move data from the SDRAM 26 a to interface 28, out to FBUS 18, via the FBUS interface 28.
  • [0027]
    Data functions are distributed amongst the microengines. Connectivity to the SRAM 26 a, SDRAM 26 b and FBUS 28 is via command requests. A command request can be a memory request or a FBUS request. For example, a command request can move data from a register located in a microengine 22 a to a shared resource, e.g., an SDRAM location, SRAM location, flash memory or some MAC address. The commands are sent out to each of the functional units and the shared resources. However, the shared resources do not need to maintain local buffering of the data. Rather, the shared resources access distributed data located inside of the microengines. This enables microengines 22 a-22 f, to have local access to data rather than arbitrating for access on a bus and risk contention for the bus. With this feature, there is a 0 cycle stall for waiting for data internal to the microengines 22 a-22 f.
  • [0028]
    The core processor 20 also can access the shared resources. The core processor 20 has a direct communication to the SDRAM controller 26 a to the bus interface 24 and to SRAM controller 26 b via bus 32. To access the microengines 22 a-22 f and transfer registers located at any of the microengines 22 a-22 f, the core processor 20 access the microengines 22 a-22 f via the ASB Translator 30 over bus 34. The ASB Translator 30 performs an address translation between FBUS microengine transfer register locations and core processor addresses (i.e., ASB bus) so that the core processor 20 can access registers belonging to the microengines 22 a-22 c.
  • [0029]
    Although microengines 22 can use the register set to exchange data. A scratchpad memory 27 is also provided to permit microengines to write data out to the memory for other microengines to read. The scratchpad 27 is coupled to bus 34.
  • [0030]
    Microengines:
  • [0031]
    Referring to FIG. 3, an exemplary one of the microengines 22 a-22 f, e.g., microengine 22 f is shown. The microengine includes a control store 70 which, in one implementation, includes a RAM of here 1,024 words of 32 bits. The RAM stores a microprogram that is loadable by the core processor 20. The microengine 22 f also includes controller logic 72. The controller logic includes an instruction decoder 73 and program counter (PC) units 72 a-72 d. The four micro program counters 72 a-72 d are maintained in hardware. The microengine 22 f also includes context event switching logic 74. Context event logic 74 receives messages (e.g., SEQ_#_EVENT_RESPONSE; FBI_EVENT_RESPONSE; SRAM_EVENT_RESPONSE; SDRAM_EVENT_RESPONSE; and ASB_EVENT_RESPONSE) from each one of the shared resources, e.g., SRAM 26 a, SDRAM 26 b, or processor core 20, control and status registers, and so forth. These messages provide information on whether a requested function has completed. Based on whether or not a function requested by a program thread has completed and signaled completion, the program thread needs to wait for that completion signal, and if the program thread is enabled to operate, then the program thread is placed on an available program thread list (not shown). The microengine 22 f can have a maximum of, e.g., 4 program threads available.
  • [0032]
    In addition to event signals that are local to an executing program thread, the microengines 22 employ signaling states that are global. With signaling states, an executing program thread can broadcast a signal state to all microengines 22. Any program thread in the microengines can branch on these signaling states. These signaling states can be used to determine availability of a resource or whether a resource is due for servicing.
  • [0033]
    The context event logic 74 has arbitration for the four (4) program threads. In one embodiment, the arbitration is a round robin mechanism. Other techniques could be used including priority queuing or weighted fair queuing. The microengine 22 f also includes an execution box (EBOX) data path 76 that includes an arithmetic logic unit 76 a and general purpose register set 76 b. The arithmetic logic unit 76 a performs arithmetic and logical functions as well as shift functions. The registers set 76 b has a relatively large number of general purpose registers. In this implementation there are 64 general purpose registers in a first bank, Bank A and 64 in a second bank, Bank B. The general purpose registers are windowed so that they are relatively and absolutely addressable.
  • [0034]
    The microengine 22 f also includes a write transfer register stack 78 and a read transfer stack 80. These registers are also windowed so that they are relatively and absolutely addressable. Write transfer register stack 78 is where write data to a resource is located. Similarly, read register stack 80 is for return data from a shared resource. Subsequent to or concurrent with data arrival, an event signal from the respective shared resource e.g., the SRAM controller 26 a, SDRAM controller 26 b or core processor 20 will be provided to context event arbiter 74 which will then alert the program thread that the data is available or has been sent. Both transfer register banks 78 and 80 are connected to the execution box (EBOX) 76 through a data path. In one implementation, the read transfer register has 64 registers and the write transfer register has 64 registers.
  • [0035]
    Each microengine 22 a-22 f supports multi-threaded execution of four contexts. One reason for this is to allow one program thread to start executing just after another program thread issues a memory reference and must wait until that reference completes before doing more work. This behavior is critical to maintaining efficient hardware execution of the microengines because memory latency is significant. Stated differently, if only a single program thread execution was supported, the microengines would sit idle for a significant number of cycles waiting for references to return and thereby reduce overall computational throughput. Multi-threaded execution allows an microengines to hide memory latency by performing useful independent work across several program threads. Two synchronization mechanisms are supplied in order to allow a program thread to issue an SRAM or SDRAM reference, and then subsequently synchronize to the point in time when that reference completes.
  • [0036]
    One mechanism is Immediate Synchronization. In immediate synchronization, the microengine issues the reference and immediately swaps out of that context. The context will be signaled when the corresponding reference completes. Once signaled, the context will be swapped back in for execution when a context-swap event occurs and it is its turn to run. Thus, from the point of view of a single context's instruction stream, the microword after issuing the mem reference does not get executed until the reference completes.
  • [0037]
    A second mechanism is Delayed Synchronization. In delayed synchronization, the microengine issues the reference, and continues to execute some other useful work independent of the reference. Some time later it could become necessary to synchronize the program thread's execution stream to the completion of the issued reference before further work is performed. At this point a synchronizing microword is executed that will either swap out the current program thread, and swap it back in sometime later when the reference has completed, or continue executing the current program thread because the reference has already completed. Delayed synchronization is implemented using two different signaling schemes:
  • [0038]
    If the memory reference is associated with a transfer register, the signal from which the program thread is triggered is generated when the corresponding transfer register valid bit is set or cleared. For example, an SRAM read which deposits data into transfer register A would be signaled when the valid bit for A is set. If the memory reference is associated with the transfer FIFO or the receive FIFO, instead of a transfer register, then the signal is generated when the reference completes in the SDRAM controller 26 a. Only one signal state per context is held in the microengines scheduler, thus only one outstanding signal can exist in this scheme.
  • [0039]
    Referring to FIG. 4, the SDRAM memory controller 26 a includes memory reference queues 90 where memory reference requests arrive from the various microengines 22 a-22 f. The memory controller 26 a includes an arbiter 91 that selects the next the microengine reference requests to go to any of the functioning units. Given that one of the microengines is providing a reference request, the reference request will come through the address and command queue 90, inside the SDRAM controller 26 a. If the reference request has a bit set called the “optimized MEM bit” the incoming reference request will be sorted into either the even bank queue 90 a or the odd bank queue 90 b. If the memory reference request does not have a memory optimization bit set, the default will be to go into an order queue 90 c. The SDRAM controller 26 is a resource which is shared among the FBUS interface 28, the core processor 20 and the PCI interface 24. The SDRAM controller 26 also maintains a state machine for performing READ-MODIFY-Write atomic operations. The SDRAM controller 26 also performs byte alignment for requests of data from SDRAM.
  • [0040]
    The order queue 90 c maintains the order of reference requests from the microengines. With a series of odd and even banks references it may be required that a signal is returned only upon completion of a sequence of memory references to both the odd and even banks. If the microengine 22 f sorts the memory references into odd bank and even bank references and one of the banks, e.g., the even bank is drained of memory references before the odd bank but the signal is asserted on the last even reference, the memory controller 26 a could conceivably signal back to a microengine that the memory request had completed, even though the odd bank reference had not been serviced. This occurrence could cause a coherency problem. The order queue 90 c allows a microengine to have multiple memory references outstanding of which only its last memory reference needs to signal a completion.
  • [0041]
    The SDRAM controller 26 a also includes a high priority queue 90 d. In the high priority queue 90 d, an incoming memory reference from one of the microengines goes directly to the high priority queue and is operated upon at a higher priority than other memory references in the other queues. All of these queues, the even bank queue 90 a, the odd bank queue 90 b, the order queue 90 c and the high priority queue, are implemented in a single RAM structure that is logically segmented into four different windows, each window having its own head and tail pointer. Since filling and draining operations are only a single input and a single output, they can be placed into the same RAM structure to increase density of RAM structures.
  • [0042]
    The SDRAM controller 26 a also includes core bus interface logic i.e., ASB bus 92. The ASB bus interface logic 92 interfaces the core processor 20 to the SDRAM controller 26 a. If there is incoming data from the core processor 20 via ASB interface 92, the data can be stored into the MEM ASB device 98 and subsequently removed from MEM ASB device 98 through the SDRAM interface 110 to SDRAM memory 16 a. Although not shown, the same queue structure can be provided for the reads. The SDRAM controller 26 a also includes an engine 97 to pull data from the microengines and PCI bus.
  • [0043]
    Additional queues include the PCI address queue 94 and ASB read/write queue 96 that maintain a number of requests. The memory requests are sent to SDRAM interface 110 via multiplexer 106. The multiplexer 106 is controlled by the SDRAM arbiter 91 which detects the fullness of each of the queues and the status of the requests and from that decides priority based on a programmable value stored in a priority service control register 100.
  • [0044]
    Referring to FIG. 5, the memory controller 26 b for the SRAM is shown. The memory controller 26 b includes an address and command queue 120. The memory controller 26 b is optimized based on the type of memory operation, i.e., a read or a write. The address and command queue 120 includes a high priority queue 120 a, a read queue 120 b which is the predominant memory reference function that an SRAM performs, and an order queue 120 c which in general will include all writes to SRAM and reads that are to be non-optimized. Although not shown, the address and command queue 120 could also include a write queue.
  • [0045]
    The SRAM controller 26 b also includes core bus interface logic i.e., ASB bus 122. The ASB bus interface logic 122 interfaces the core processor 20 to the SRAM controller 26 b. The SRAM controller 26 b also includes an engine 127 to pull data from the microengines and PCI bus.
  • [0046]
    The memory requests are sent to SRAM interface 140 via multiplexer 126. The multiplexer 126 is controlled by the SRAM arbiter 131 which detects the fullness of each of the queues and the status of the requests and from that decides priority based on a programmable value stored in a priority service control register 130. Once control to the multiplexer 126 selects a memory reference request, the memory reference request, is sent to a decoder 138 where it is decoded and an address is generated.
  • [0047]
    The SRAM Unit maintains control of the Memory Mapped off-chip SRAM and Expansion ROM. The SRAM controller 26 b can address, e.g., 16 MBytes, with, e.g., 8 MBytes mapped for SRAM 16 b, and 8 MBytes reserved for special functions including: Boot space via flashrom 16 c; and Console port access for MAC devices 13 a, 13 b and access to associated (RMON) counters. The SRAM is used for local look-up tables and queue management functions.
  • [0048]
    The SRAM controller 26 b supports the following transactions:
  • [0049]
    Microengine requests (via private bus) to/from SRAM.
  • [0050]
    Core Processor (via ASB bus) to/from SRAM.
  • [0051]
    The address and command queue 120 also includes a Read Lock Fail Queue 120 d. The Read Lock Fail Queue 120 d is used to hold read memory reference requests that fail because of a lock existing on a portion of memory.
  • [0052]
    Referring to FIG. 6, communication between the microengines 22 and the FBUS interface Logic (FBI) is shown. The FBUS interface 28 in a network application can performs header processing of incoming packets from the FBUS 18. A key function which the FBUS interface performs is extraction of packet headers, and a microprogrammable source/destination/protocol hashed lookup in SRAM . If the hash does not successfully resolve, the packet header is promoted to the core processor 28 for more sophisticated processing.
  • [0053]
    The FBI 28 contains a Transmit FIFO 182, a Receive FIFO 183, a HASH unit 188 and FBI control and status registers 189. These four units communicate with the microengines 22, via a time-multiplexed access to the SRAM bus 38 which is connected to the transfer registers 78, 80 in the microengines. That is, all communications to and from the microengines are via the transfer registers 78, 80. The FBUS interface 28 includes a push state machine 200 for pushing data into the transfer registers during the time cycles which the SRAM is NOT using the SRAM data bus (part of bus 38) and a pull state machine 202 for fetching data from the transfer registers in the respective microengine.
  • [0054]
    The Hashing unit includes a pair of FIFO=s 188 a, 188 b. The hash unit determines that the FBI 28 received an FBI_hash request. The hash unit 188 fetches hash keys from the calling microengine 22. After the keys are fetched and hashed, the indices are delivered back to the calling microengine 22. Up to three hashes are performed under a single FBI_hash request. The busses 34 and 38 are each unidirectional: SDRAM_push/pull_data, and Sbus_push/pull_data. Each of these buses require control signals which will provide read/write controls to the appropriate microengine 22 Transfer registers.
  • [0055]
    Generally, transfer registers require protection from the context controlling them to guarantee read correctness. In particular, if a write transfer register is being used by a thread_1 to provide data to the SDRAM 16 a, thread_1 does not overwrite this register until the signal back from SDRAM controller 26 a indicates that this register has been promoted and may now be re-used. Every write does not require a signal back from the destination indicating that the function has been completed, because if the program thread writes to the same command queue at that destination with multiple requests, the order of the completion is guaranteed within that command queue, thus only the last command requires the signaling back to the program thread. However, if the program thread uses multiple command queues (order and read), then these command requests are broken into separate context tasks, so that ordering is maintained via context swapping. The exception case indicated at the beginning of this paragraph is relative to a certain class of operations using an unsolicited PUSH to transfer registers from the FBI for FBUS status information. In order to protect read/write determinism on the transfer registers, the FBI provides a special Push_protect signal when these special FBI push operations are set up.
  • [0056]
    Any microengine 22 that uses the FBI unsolicited push technique must test the protection flag prior to accessing the FBUS interface/microengine agreed upon transfer registers If the flag is not asserted, then the transfer registers may be accessed by the microengines 22. If the flag is asserted then the context should wait N cycles prior to accessing the registers. This count is determined a priori by the number of transfer registers being pushed, plus a frontend protection window. The microengine tests this flag then moves the data from the read transfer registers to GPR's in contiguous cycles, so the push engine does not collide with the microengine read.
  • [0057]
    Thread signaling for packet processing
  • [0058]
    Special techniques such as inter-thread communications to communicate status, a self destruct register 210 to allow program threads to self assign tasks and a thread_done register 212 to provide a global program thread communication scheme are used for packet processing. The destruct register 210 and a thread_done register 212 can be implemented as control and status registers 189. They are shown in the FBUS interface 28 outside of the block labeled CSR for clarity. Network functions are implemented in the network processor using a plurality of program threads e.g., contexts to process network packets. For example, scheduler program threads could be executed in one of the microprogram engines e.g., 22 a whereas, processing program threads could execute in the remaining engines e.g., 22 b-22 f. The program threads (processing or scheduling program threads) use inter-thread communications to communicate status.
  • [0059]
    Program threads are assigned specific tasks such as receive and transmit scheduling, receive processing, and transmit processing, etc. Task assignment and task completion are communicated between program threads through the inter-thread signaling, registers with specialized read and write characteristics, e.g., the self-destruct register 210 and the thread-done register 212, SRAM 16 b and data stored in the internal scratchpad memory 186 (FIG. 6) resulting from operations such as bit set, and bit clear.
  • [0060]
    The network processor 10 includes a general context communication signaling protocol that allows any context to set a signal that any other context can detect. This allows cooperating program threads to use a semaphore and thus coordinate using micro-code controlled processing.
  • [0061]
    Processing of network packets can use multiple program threads. Typically, for network processing there is a receive scheduler, a transmit scheduler and processing program threads. A scheduler (either receive or transmit) program thread coordinates amounts of work to be done and sequence of work by processing program threads. The scheduler program thread assigns tasks to processing program threads and in some cases processing program threads can assign tasks to other processing program threads. For instance, a scheduler determines which ports need service and assigns and coordinates tasks to processing program threads to overcome inherent memory latency by processing multiple program threads in parallel.
  • [0062]
    In some examples, with slow ports one processing program thread may perform processing on a portion of a packet and a second processing program thread processes the remainder of the packet or in some cases the scheduler uses the next available program thread. With faster ports e.g., Gigabit ports where 64 byte packets are received very fast, the scheduler can assign M packets to the next available program thread. The program threads signal each other as to what part of a packet the program thread has processed and its status.
  • [0063]
    A program thread can be assigned to process the first 64 bytes of a packet. When the program thread is finish, the program thread has data to set signals to wake up the next program thread has been assigned to process the next 64 bytes. The program thread may write a register and an address of the register in a pre-assigned memory location, e.g., scratch register. The program thread sets signals to wake up the next program thread that has been assigned to work on the next bytes of the packet.
  • [0064]
    Referring to FIGS. 7A-7B, the self-destruct register 210 allows one scheduler program thread S, (230 in FIG. 7B) to request services from multiple program threads Pa-Pn that provide the requested service. The first program thread, e.g., Pi that accesses (232 in FIG. 7B) the self_destruct register 210 takes the request. The “self-destruct register” 210 zeros, i.e., clears (234 in FIG. 7B) upon being read by a program thread. Other program threads capable of servicing that request will no longer be presented with an active request. For example, a program context can request a task be assigned to the first context that is ready, by writing to the self-destruct register 210. A context checks for an assignment by reading the “self-destruct register” 210. If the value of the self-destruct register is 0, there is no new task presently available to assign to the program thread. This could indicate that there are no new tasks or that another program thread may have assigned itself to the task and cleared the self-destruct register 210. If the value is non-zero, the contents of the self-destruct register are interpreted to determine the task, and the register is cleared upon reading by the context. Thus, contexts reading this register for assignment wait for the register to be written subsequently with the next task instruction.
  • [0065]
    Referring to FIG. 8, for networking applications typically different program contexts are used to perform specific system tasks. Tasks include receive scheduling, receive processing contexts, transmit arbiter, transmit scheduling, transmit filling and processor core communications.
  • [0066]
    The receive scheduler initiates 242 a receive operation of e.g., 64 or 128 bytes of input data by sending a command to the FBI interface 28 that specifies a port from which to extract the data and the Receive FIFO element to use to buffer that data as well as the microengine context to be notified once the receive data has been fetched.
  • [0067]
    The receive scheduler thread 244 sends a signal to the specified microengine program thread that activates a specified context. The context reads the FBI Receive Control register to obtain the necessary receive information for processing (i.e. port, Receive FIFO element location, byte count, start of packet, end of packet, error status). If a start of packet is indicated the receive scheduler program thread is responsible for determining where in SDRAM to store the data, (i.e., the output queue to insert the packet) and writing the packet data into SDRAM. If it is not the start of a packet, then the receive program thread determines where the earlier data of this packet was stored in order to continue processing 246 of the packet. When the end of packet indication is received 248 (or after the first 64 byte section if receive to transmit latency is optimized) the receive program thread adds the packet to the queue determined by processing the packet header.
  • [0068]
    The program threads also communicate with a shared resource through a bit set and bit clear mechanism that provides a bit vector. This mechanism allows setting and clearing of individual bits and performing a test and set on individual bits to control a shared resource. The bit vector signals the non-emptiness of output queues. When a receive program thread enqueues a packet, the receive scheduler sets 250 a bit. The transmit scheduler can examine the bit vector to determine the state all queues.
  • [0069]
    The bit set and bit clear operations on the bit vector, can occur in either scratchpad RAM or SRAM. If the scheduler is communicating between program threads on the same microengine 22, the bit vector can be stored in the register set because each context can read the other context's registers. For example, an empty or not empty status of each output queue is support by a bit vector in internal scratchpad memory. When a receive program thread enqueues a packet, the receive program thread uses the scratch pad bit-set command to set a bit in the queue status bit vector to indicate the queue now has at least one entry. The transmit arbiter scans 270 the queue bit vector for non empty queues (e.g., bitx set) to determine packets that are ready to be transmitted. When removing 272 a packet from a queue for transmit if the queue empties 274, the transmit arbiter issues 276 a bit-clear command to the corresponding bit of the queue bit vector.
  • [0070]
    Referring to FIG. 9, the thread_done register is also on the FBI 28 and is a register where bits can be set from different program threads. Each program thread can use, e.g., two bits to communicate its status to all other program threads. Also one scheduler program thread can read 292 the status of all of its processing program threads. Upon completion of a receive task, 282 a “receive” program thread writes 284 a completion code into the “thread_done” register. The receive program thread becomes inactive 286 after writing the thread_done register. That receive program thread waits for another signal from the FBI that indicates another receive task has been assigned. Program threads 1-16 have 2 bit fields for “thread_done_1”, and program threads 17-24 have 2 bit fields for “thread_done2”. The 2 bit field allows a program thread to communicate different levels of task completion.
  • [0071]
    For example, the scheduler can use the two bit status “01” to indicate that data was moved to SDRAM, processing of packet is still in progress and pointers were saved; bits 10 can indicate that data was moved to SDRAM, processing of packet is still in progress and pointers were not saved; and bits 11 can indicates packet processing is completed.
  • [0072]
    Thus, the states 296 a can be used by the receiver scheduler program thread to assign 297 a another thread to process a task when data becomes available, whereas, states 296 b can be used by the receive scheduler to assign 297 b the same thread to continue processing when the data is available.
  • [0073]
    The exact interpretation of the message can be fixed by a software convention determined between a scheduler program thread and processing program threads called by the scheduler program thread. That is the status messages can change depending on whether the convention is for receive, as above, transmit, and so forth. In general, the status messages include “busy”, “not busy”, “not busy but waiting.” The status message of Anot busy, but waiting@signals that the current program thread has completed processing of a portion of a packet and is expected to be assigned to perform a subsequent task on the packet when data is made available. It can be used when the program thread is expecting data from a port and has not saved context so it should process the rest of that packet.
  • [0074]
    The scheduler program thread reads the “thread done” register to determine the completion status of tasks it assigned to other program threads. The “thread done” register is implemented as a write one to clear register, allowing the scheduler to clear just the fields it has recognized.
  • Other Embodiments
  • [0075]
    It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.
Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
US5193167 *29 Jun 19909 Mar 1993Digital Equipment CorporationEnsuring data integrity by locked-load and conditional-store operations in a multiprocessor system
US5870597 *25 Jun 19979 Feb 1999Sun Microsystems, Inc.Method for speculative calculation of physical register addresses in an out of order processor
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US6735679 *23 Jun 200011 May 2004Broadcom CorporationApparatus and method for optimizing access to memory
US701702022 Dec 200321 Mar 2006Broadcom CorporationApparatus and method for optimizing access to memory
US727799030 Sep 20042 Oct 2007Sanjeev JainMethod and apparatus providing efficient queue descriptor memory access
US741854321 Dec 200426 Aug 2008Intel CorporationProcessor having content addressable memory with command ordering
US746725628 Dec 200416 Dec 2008Intel CorporationProcessor having content addressable memory for block-based queue structures
US747522914 Feb 20066 Jan 2009Intel CorporationExecuting instruction for processing by ALU accessing different scope of variables using scope index automatically changed upon procedure call and exit
US755563021 Dec 200430 Jun 2009Intel CorporationMethod and apparatus to provide efficient communication between multi-threaded processing elements in a processor unit
US765044810 Jan 200819 Jan 2010Pact Xpp Technologies AgI/O and memory bus system for DFPS and units with two- or multi-dimensional programmable cell architectures
US765786123 Jul 20032 Feb 2010Pact Xpp Technologies AgMethod and device for processing data
US765787720 Jun 20022 Feb 2010Pact Xpp Technologies AgMethod for processing data
US778208714 Aug 200924 Aug 2010Martin VorbachReconfigurable sequencer structure
US78228817 Oct 200526 Oct 2010Martin VorbachProcess for automatic dynamic reloading of data flow processors (DFPs) and units with two- or three-dimensional programmable cell architectures (FPGAs, DPGAs, and the like)
US782296810 Feb 200926 Oct 2010Martin VorbachCircuit having a multidimensional structure of configurable cells that include multi-bit-wide inputs and outputs
US78408423 Aug 200723 Nov 2010Martin VorbachMethod for debugging reconfigurable architectures
US784479630 Aug 200430 Nov 2010Martin VorbachData processing device and method
US785395125 Jul 200514 Dec 2010Intel CorporationLock sequencing to reorder and grant lock requests from multiple program threads
US78999623 Dec 20091 Mar 2011Martin VorbachI/O and memory bus system for DFPs and units with two- or multi-dimensional programmable cell architectures
US792482831 Aug 200412 Apr 2011Netlogic Microsystems, Inc.Advanced processor with mechanism for fast packet queuing operations
US792876314 Jul 201019 Apr 2011Martin VorbachMulti-core processing system
US794160330 Nov 200910 May 2011Netlogic Microsystems, Inc.Method and apparatus for implementing cache coherency of a processor
US796172331 Aug 200414 Jun 2011Netlogic Microsystems, Inc.Advanced processor with mechanism for enforcing ordering between information sent on two independent networks
US7984268 *23 Jul 200419 Jul 2011Netlogic Microsystems, Inc.Advanced processor scheduling in a multithreaded system
US799197720 Dec 20072 Aug 2011Netlogic Microsystems, Inc.Advanced processor translation lookaside buffer management in a multithreaded system
US799682716 Aug 20029 Aug 2011Martin VorbachMethod for the translation of programs for reconfigurable architectures
US801556731 Aug 20046 Sep 2011Netlogic Microsystems, Inc.Advanced processor with mechanism for packet distribution at high line rate
US803722431 Jul 200711 Oct 2011Netlogic Microsystems, Inc.Delegating network processor operations to star topology serial bus interfaces
US805889913 Feb 200915 Nov 2011Martin VorbachLogic cell array and bus system
US806545624 Jan 200822 Nov 2011Netlogic Microsystems, Inc.Delegating network processor operations to star topology serial bus interfaces
US806937315 Jan 200929 Nov 2011Martin VorbachMethod for debugging reconfigurable architectures
US809961823 Oct 200817 Jan 2012Martin VorbachMethods and devices for treating and processing data
US812706118 Feb 200328 Feb 2012Martin VorbachBus systems and reconfiguration methods
US814588124 Oct 200827 Mar 2012Martin VorbachData processing device and method
US815628424 Jul 200310 Apr 2012Martin VorbachData processing method and device
US815631219 Jun 200710 Apr 2012Martin VorbachProcessor chip for reconfigurable data processing, for processing numeric and logic operations and including function and interconnection control units
US817629831 Aug 20048 May 2012Netlogic Microsystems, Inc.Multi-core multi-threaded processing systems with instruction reordering in an in-order pipeline
US819585621 Jul 20105 Jun 2012Martin VorbachI/O and memory bus system for DFPS and units with two- or multi-dimensional programmable cell architectures
US82096537 Oct 200826 Jun 2012Martin VorbachRouter
US823041113 Jun 200024 Jul 2012Martin VorbachMethod for interleaving a program over a plurality of cells
US825050317 Jan 200721 Aug 2012Martin VorbachHardware definition method including determining whether to implement a function as hardware or software
US828110820 Jan 20032 Oct 2012Martin VorbachReconfigurable general purpose processor having time restricted configurations
US828126519 Nov 20092 Oct 2012Martin VorbachMethod and device for processing data
US83018724 May 200530 Oct 2012Martin VorbachPipeline configuration protocol and configuration unit communication
US83102744 Mar 201113 Nov 2012Martin VorbachReconfigurable sequencer structure
US831220021 Jul 201013 Nov 2012Martin VorbachProcessor chip including a plurality of cache elements connected to a plurality of processor cores
US831230130 Sep 200913 Nov 2012Martin VorbachMethods and devices for treating and processing data
US83862542 May 200826 Feb 2013Nuance Communications, Inc.Multi-class constrained maximum likelihood linear regression
US840752524 Oct 201126 Mar 2013Pact Xpp Technologies AgMethod for debugging reconfigurable architectures
US842938519 Sep 200223 Apr 2013Martin VorbachDevice including a field having function cells and information providing cells controlled by the function cells
US84683298 Jun 201218 Jun 2013Martin VorbachPipeline configuration protocol and configuration unit communication
US84715934 Nov 201125 Jun 2013Martin VorbachLogic cell array and bus system
US847881130 Oct 20082 Jul 2013Netlogic Microsystems, Inc.Advanced processor with credit based scheme for optimal packet flow in a multi-processor system on a chip
US84993026 Sep 201130 Jul 2013Netlogic Microsystems, Inc.Advanced processor with mechanism for packet distribution at high line rate
US85437474 Oct 201124 Sep 2013Netlogic Microsystems, Inc.Delegating network processor operations to star topology serial bus interfaces
US86864759 Feb 20111 Apr 2014Pact Xpp Technologies AgReconfigurable elements
US868654930 Sep 20091 Apr 2014Martin VorbachReconfigurable elements
US8713290 *20 Sep 201029 Apr 2014International Business Machines CorporationScaleable status tracking of multiple assist hardware threads
US8719554 *23 Jan 20136 May 2014International Business Machines CorporationScaleable status tracking of multiple assist hardware threads
US872625010 Mar 201013 May 2014Pact Xpp Technologies AgConfigurable logic integrated circuit having a multidimensional structure of configurable elements
US878873221 Aug 201322 Jul 2014Netlogic Microsystems, Inc.Messaging network for processing data using multiple processor cores
US879347420 Sep 201029 Jul 2014International Business Machines CorporationObtaining and releasing hardware threads without hypervisor involvement
US880355225 Sep 201212 Aug 2014Pact Xpp Technologies AgReconfigurable sequencer structure
US881282019 Feb 200919 Aug 2014Pact Xpp Technologies AgData processing device and method
US881950530 Jun 200926 Aug 2014Pact Xpp Technologies AgData processor having disabled cores
US88691217 Jul 201121 Oct 2014Pact Xpp Technologies AgMethod for the translation of programs for reconfigurable architectures
US889844121 Apr 201225 Nov 2014International Business Machines CorporationObtaining and releasing hardware threads without hypervisor involvement
US891459030 Sep 200916 Dec 2014Pact Xpp Technologies AgData processing method and device
US89536286 Jun 201110 Feb 2015Netlogic Microsystems, Inc.Processor with packet ordering device
US903780711 Nov 201019 May 2015Pact Xpp Technologies AgProcessor arrangement on a chip including data processing, memory, and interface elements
US904744028 May 20132 Jun 2015Pact Xpp Technologies AgLogical cell array and bus system
US907560517 Oct 20127 Jul 2015Pact Xpp Technologies AgMethods and devices for treating and processing data
US908847431 Aug 200421 Jul 2015Broadcom CorporationAdvanced processor with interfacing messaging network to a CPU
US90923601 Aug 201128 Jul 2015Broadcom CorporationAdvanced processor translation lookaside buffer management in a multithreaded system
US915242616 Apr 20126 Oct 2015International Business Machines CorporationInitiating assist thread upon asynchronous event for processing simultaneously with controlling thread and updating its running status in status register
US915444320 Oct 20096 Oct 2015Broadcom CorporationAdvanced processor with fast messaging network technology
US92643807 May 201116 Feb 2016Broadcom CorporationMethod and apparatus for implementing cache coherency of a processor
US94893275 Nov 20138 Nov 2016Oracle International CorporationSystem and method for supporting an efficient packet processing model in a network environment
US95442325 Nov 201310 Jan 2017Oracle International CorporationSystem and method for supporting virtualized switch classification tables
US9575900 *19 Feb 201521 Feb 2017Imagination Technologies LimitedDigital signal processing data transfer
US95963248 Feb 200814 Mar 2017Broadcom CorporationSystem and method for parsing and allocating a plurality of packets to processor core threads
US20030233485 *13 Jun 200218 Dec 2003Mircrosoft CorporationEvent queue
US20040128474 *8 Oct 20011 Jul 2004Martin VorbachMethod and device
US20040165609 *22 Dec 200326 Aug 2004Broadcom CorporationApparatus and method for optimizing access to memory
US20040168099 *1 Mar 200426 Aug 2004Martin VorbachUnit for processing numeric and logic operations for use in central processing units (CPUs), multiprocessor systems
US20040243984 *20 Jun 20022 Dec 2004Martin VorbachData processing method
US20040249880 *9 Sep 20029 Dec 2004Martin VorbachReconfigurable system
US20050027793 *31 Aug 20043 Feb 2005Hass David T.Advanced processor with mechanism for packet distribution at high line rate
US20050033889 *31 Aug 200410 Feb 2005Hass David T.Advanced processor with interrupt delivery mechanism for multi-threaded multi-CPU system on a chip
US20050041666 *31 Aug 200424 Feb 2005Hass David T.Advanced processor with mechanism for enforcing ordering between information sent on two independent networks
US20050044308 *31 Aug 200424 Feb 2005Abbas RashidAdvanced processor with interfacing messaging network to a CPU
US20050044323 *31 Aug 200424 Feb 2005Hass David T.Advanced processor with out of order load store scheduling in an in order pipeline
US20050055540 *23 Jul 200410 Mar 2005Hass David T.Advanced processor scheduling in a multithreaded system
US20050086462 *5 Mar 200221 Apr 2005Martin VorbachMethods and devices for treating and/or processing data
US20050086649 *16 Aug 200221 Apr 2005Martin VorbachMethod for the translation of programs for reconfigurable architectures
US20050132344 *20 Jan 200316 Jun 2005Martin VorbachMethod of compilation
US20050223212 *4 May 20056 Oct 2005Martin VorbachPipeline configuration protocol and configuration unit communication
US20060031595 *7 Oct 20059 Feb 2006Martin VorbachProcess for automatic dynamic reloading of data flow processors (DFPs) and units with two- or three-dimensional programmable cell architectures (FPGAs, DPGAs, and the like
US20060067348 *30 Sep 200430 Mar 2006Sanjeev JainSystem and method for efficient memory access of queue control data structures
US20060075211 *21 Mar 20036 Apr 2006Martin VorbachMethod and device for data processing
US20060090062 *20 Jan 200327 Apr 2006Martin VorbachReconfigurable processor
US20060140203 *28 Dec 200429 Jun 2006Sanjeev JainSystem and method for packet queuing
US20060143373 *28 Dec 200429 Jun 2006Sanjeev JainProcessor having content addressable memory for block-based queue structures
US20060155959 *21 Dec 200413 Jul 2006Sanjeev JainMethod and apparatus to provide efficient communication between processing elements in a processor unit
US20060245225 *19 Sep 20022 Nov 2006Martin VorbachReconfigurable elements
US20060248317 *23 Jul 20032 Nov 2006Martin VorbachMethod and device for processing data
US20070011433 *5 Apr 200411 Jan 2007Martin VorbachMethod and device for data processing
US20070022429 *25 Jul 200525 Jan 2007Mark RosenbluthLock sequencing
US20070044103 *25 Jul 200522 Feb 2007Mark RosenbluthInter-thread communication of lock protected data
US20070050603 *24 Jul 20031 Mar 2007Martin VorbachData processing method and device
US20070083730 *17 Jun 200412 Apr 2007Martin VorbachData processing device and method
US20070113046 *30 Aug 200417 May 2007Martin VorbachData processing device and method
US20070124728 *28 Nov 200531 May 2007Mark RosenbluthPassing work between threads
US20070174372 *14 Feb 200626 Jul 2007Feghali Wajdi KProgrammable processing unit having multiple scopes
US20070299993 *5 Mar 200227 Dec 2007Pact Xpp Technologies AgMethod and Device for Treating and Processing Data
US20080004876 *30 Jun 20063 Jan 2008Chuang HeNon-enrolled continuous dictation
US20080010437 *19 Jun 200710 Jan 2008Martin VorbachUnit for processing numeric and logic operations for use in central processing units (CPUS), multiprocessor systems, data-flow processors (DSPS), systolic processors and field programmable gate arrays (FPGAS)
US20080062927 *31 Jul 200713 Mar 2008Raza Microelectronics, Inc.Delegating Network Processor Operations to Star Topology Serial Bus Interfaces
US20080191737 *7 Apr 200814 Aug 2008Martin VorbachReconfigurable sequencer structure
US20080216074 *20 Dec 20074 Sep 2008Hass David TAdvanced processor translation lookaside buffer management in a multithreaded system
US20080222329 *10 Jan 200811 Sep 2008Martin VorbachI/O and memory bus system for DFPs and units with two- or multi-dimensional programmable cell architectures
US20090006895 *3 Aug 20071 Jan 2009Frank MayMethod for debugging reconfigurable architectures
US20090024390 *2 May 200822 Jan 2009Nuance Communications, Inc.Multi-Class Constrained Maximum Likelihood Linear Regression
US20090031104 *6 Feb 200629 Jan 2009Martin VorbachLow Latency Massive Parallel Data Processing Device
US20090037865 *7 Oct 20085 Feb 2009Martin VorbachRouter
US20090055496 *30 Oct 200826 Feb 2009Gaurav GargAdvanced processor with credit based scheme for optimal packet flow in a multi-processor system on a chip
US20090100286 *23 Oct 200816 Apr 2009Martin VorbachMethods and devices for treating and processing data
US20090144485 *6 Feb 20094 Jun 2009Martin VorbachProcess for automatic dynamic reloading of data flow processors (dfps) and units with two- or three-dimensional programmable cell architectures (fpgas, dpgas, and the like)
US20090144522 *24 Oct 20084 Jun 2009Martin VorbachData Processing Device and Method
US20090146690 *10 Feb 200911 Jun 2009Martin VorbachRuntime configurable arithmetic and logic cell
US20090146691 *13 Feb 200911 Jun 2009Martin VorbachLogic cell array and bus system
US20090150725 *15 Jan 200911 Jun 2009Martin VorbachMethod for debugging reconfigurable architectures
US20090153188 *19 Feb 200918 Jun 2009Martin VorbachPROCESS FOR AUTOMATIC DYNAMIC RELOADING OF DATA FLOW PROCESSORS (DFPs) AND UNITS WITH TWO- OR THREE-DIMENSIONAL PROGRAMMABLE CELL ARCHITECTURES (FPGAs, DPGAs AND THE LIKE)
US20090172351 *19 Feb 20092 Jul 2009Martin VorbachData processing device and method
US20090199167 *17 Jan 20076 Aug 2009Martin VorbachHardware Definition Method
US20090201935 *8 Feb 200813 Aug 2009Hass David TSystem and method for parsing and allocating a plurality of packets to processor core threads
US20090210653 *19 Feb 200920 Aug 2009Pact Xpp Technologies AgMethod and device for treating and processing data
US20090300262 *1 Jul 20093 Dec 2009Martin VorbachMethods and devices for treating and/or processing data
US20090300445 *30 Jun 20093 Dec 2009Martin VorbachMethod and system for alternating between programs for execution by cells of an integrated circuit
US20100023796 *30 Sep 200928 Jan 2010Martin VorbachMethods and devices for treating and processing data
US20100070671 *19 Nov 200918 Mar 2010Martin VorbachMethod and device for processing data
US20100077150 *30 Nov 200925 Mar 2010Rmi CorporationAdvanced processor with cache coherency
US20100082863 *3 Dec 20091 Apr 2010Martin VorbachI/O AND MEMORY BUS SYSTEM FOR DFPs AND UNITS WITH TWO- OR MULTI-DIMENSIONAL PROGRAMMABLE CELL ARCHITECTURES
US20100095094 *17 Dec 200915 Apr 2010Martin VorbachMethod for processing data
US20100153654 *30 Sep 200917 Jun 2010Martin VorbachData processing method and device
US20100174868 *22 Mar 20108 Jul 2010Martin VorbachProcessor device having a sequential data processing unit and an arrangement of data processing elements
US20100228918 *10 Mar 20109 Sep 2010Martin VorbachConfigurable logic integrated circuit having a multidimensional structure of configurable elements
US20100241823 *1 Jun 201023 Sep 2010Martin VorbachData processing device and method
US20100287318 *21 Jul 201011 Nov 2010Martin VorbachI/o and memory bus system for dfps and units with two- or multi-dimensional programmable cell architectures
US20100318703 *14 Jun 201016 Dec 2010Netlogic Microsystems, Inc.Delegating network processor operations to star topology serial bus interfaces
US20110006805 *14 Jul 201013 Jan 2011Martin VorbachReconfigurable sequencer structure
US20110010523 *27 Jul 201013 Jan 2011Martin VorbachRuntime configurable arithmetic and logic cell
US20110060942 *11 Nov 201010 Mar 2011Martin VorbachMethods and devices for treating and/or processing data
US20110148460 *4 Mar 201123 Jun 2011Martin VorbachReconfigurable sequencer structure
US20110173389 *8 Mar 201114 Jul 2011Martin VorbachMethods and devices for treating and/or processing data
US20110238948 *16 Nov 201029 Sep 2011Martin VorbachMethod and device for coupling a data processing unit and a data processing array
US20120072707 *20 Sep 201022 Mar 2012International Business Machines CorporationScaleable Status Tracking Of Multiple Assist Hardware Threads
US20130139168 *23 Jan 201330 May 2013International Business Machines CorporationScaleable Status Tracking Of Multiple Assist Hardware Threads
US20150127762 *5 Nov 20137 May 2015Oracle International CorporationSystem and method for supporting optimized buffer utilization for packet processing in a networking device
US20150161058 *19 Feb 201511 Jun 2015Imagination Technologies LimitedDigital Signal Processing Data Transfer
US20150295860 *25 Jun 201515 Oct 2015Broadcom CorporationMethod and system for remotely configuring an ethernet switch using ethernet packets
USRE4436521 Oct 20109 Jul 2013Martin VorbachMethod of self-synchronization of configurable elements of a programmable module
USRE4438324 Apr 200816 Jul 2013Martin VorbachMethod of self-synchronization of configurable elements of a programmable module
USRE4510921 Oct 20102 Sep 2014Pact Xpp Technologies AgMethod of self-synchronization of configurable elements of a programmable module
USRE4522321 Oct 201028 Oct 2014Pact Xpp Technologies AgMethod of self-synchronization of configurable elements of a programmable module
CN102411512A *19 Sep 201111 Apr 2012国际商业机器公司Scaleable status tracking method and system of multiple assist hardware threads
Classifications
U.S. Classification719/313, 709/230, 712/E09.053
International ClassificationG06F9/48, G06F9/38
Cooperative ClassificationG06F9/4843, G06F9/3851
European ClassificationG06F9/48C4, G06F9/38E4
Legal Events
DateCodeEventDescription
16 Jul 2001ASAssignment
Owner name: INTEL CORPORATION, CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ADILETTA, MATTHEW J.;BERNSTEIN, DEBRA;WOLRICH, GILBERT;AND OTHERS;REEL/FRAME:011986/0883
Effective date: 20010709