Electrical computers and digital processing systems: processing – Processing architecture – Array processor
Reexamination Certificate
2000-09-06
2004-12-07
An, Weng-Al T. (Department: 2126)
Electrical computers and digital processing systems: processing
Processing architecture
Array processor
72, C718S103000
Reexamination Certificate
active
06829697
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to the field of network processor systems. More particularly, it relates to an embedded processor complex containing one or more protocol processor units (PPUs). Multiple coprocessors are used in connection with the PPUs within the processor system through interfaces that interconnect the PPUs and the coprocessors for transmitting data and instructions therebetween.
BACKGROUND OF THE INVENTION
The use of protocol processor units to provide for and to control the programmability of a network processor is well known. Likewise, the use of coprocessors with the PPU in the design of a computer system processing complex architecture is dwell established. Delays in processing events that require real time processing is a problem that directly affect system performance. By assigning a task to a specific coprocessor, rather than requiring the PPU to perform the task, the efficiency and performance of a computer system can be increased. It is important for the PPU to communicate with the coprocessors in an efficient manner. Continuing improvements in this communication are constantly being sought.
SUMMARY OF THE INVENTION
An object of the present invention is the use of Protocol Processor Units (PPU) which contain one or more core language processors (CLPs) each of which has a plurality of threads and instructs special task coprocessors through a logical coprocessor interface.
Another object is the use of multiple logical coprocessor interfaces (from the perspective of a programmer) to access a shared coprocessor resource. In some cases the coprocessor resource is shared among multiple processing threads within a PPU, while in other cases a single coprocessor resource is shared among multiple PPUs.
An additional object of the present invention relates to specific operations which are enabled at the interface between a PPU and its coprocessors. One such operation is the ability to conditionally execute coprocessor instructions. This is especially useful with the counter coprocessor, but may be generally applied to other coprocessors as well. The coprocessor interface has the ability to identify long latency events and short latency events according to the expected response time to a particular coprocessor command. This identification is then used to control the priority for thread execution.
Still another object of the present invention is a coprocessor interface that provides more flexibility and efficiency than other known coprocessor interfaces.
These and other objects are achieved in the manner to be hereinafter described in greater detail.
The operation of an embedded processor complex for controlling the programmability of a network processor is described. The processor complex includes a plurality of protocol processor units (PPUs), each protocol processor unit containing one or more core language processors (CLPs). Each CLP has multiple code threads. Each PPU utilizes a plurality of coprocessors useful for executing specific tasks for the PPUs. The complex uses multiple logical coprocessor interfaces to access a shared coprocessor resources with the CLPs. Specific operating instructions are executed by the CLPs resulting in commands sent to the coprocessors. One aspect of these instructions serves to enable conditional execution of specific coprocessor instructions. The instruction is able to identify long latency events and short latency events according to the expected response time to a particular coprocessor command. This permits the complex to transfer control from one thread to another depending on the length and type of latency event being processed.
REFERENCES:
patent: 4809169 (1989-02-01), Sfarti et al.
patent: 4876643 (1989-10-01), McNeill et al.
patent: 5404469 (1995-04-01), Chung et al.
patent: 5748468 (1998-05-01), Notenboom et al.
patent: 5928322 (1999-07-01), Bitar et al.
patent: 6016542 (2000-01-01), Gottlieb et al.
patent: 6219833 (2001-04-01), Solomon et al.
patent: 6226738 (2001-05-01), Dowling
patent: 6532530 (2003-03-01), Kim et al.
Davis Gordon Taylor
Heddes Marco C.
Leavens Ross Boyd
Rinaldi Mark Anthony
An Weng-Al T.
Driggs Lucas Brubaker & Hogg Co. LPA
Hoang Phuong N.
Lucas James A.
LandOfFree
Multiple logical interfaces to a shared coprocessor resource does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multiple logical interfaces to a shared coprocessor resource, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multiple logical interfaces to a shared coprocessor resource will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3330149