International Business Machines Corporation

Compagnie Ibm France

James Francis Mikos

Glen Howard Handlogten

David Alan Norgaard

William John Goetzinger

Lyle Edwin Grosbach

James David Strom

William John Starke

Jeffrey A Stuecheli

Fabrice Jean Verplanken

Jean Louis Calvignac

Michael Steven Siegel

Lakshminarayana Baba Arimilli

Marco C Heddes

Kenneth Michael Valk

David John Krolak

Joseph Franklin Logan

Tolga Ozguner

Kerry Christopher Imming

David Lawrence Pease

Daniel James Sucher

John T Trnka

Gerald G Fagerness

Merwin Herscher Alferness

Ronald D Gillingham

Brian M McKevett

Roger Dale Emeigh

In a data processing environment, a communication interface of a second host data processing system receives, from a first host data processing system, a host command in a first command set. The host command specifies a memory access to a memory coupled to the second host data processing system. The communication interface translates the host command into a command in a different second command set emulating coupling of an attached functional unit to the communication interface. The communication interface presents the second command to a host bus protocol interface of the second host data processing system. Based on receipt of the second command, the host bus protocol interface initiates, on a system fabric of the second host data processing system, a host bus protocol memory access request specifying the memory access.

Multifunction communication interface supporting memory sharing among data processing systems

Aligning received BDIs with received data on a cross-chip link including receiving, from the cross-chip link, a control flit comprising incoming data flit information for a plurality of incoming data flits; adding the incoming data flit information to a control structure; receiving, from the cross-chip link, the plurality of incoming data flits; directing each of the plurality of incoming data flits to virtual channel queues based on the incoming data flit information at a first read pointer in the control structure; receiving a bookend flit comprising a plurality of BDIs for the plurality of data flits; and associating each of the BDIs with the plurality of data flits based on the incoming data flit information at a second read pointer in the control structure.

Aligning received bad data indicators (BDIS) with received data on a cross-chip link

Aligning received bad data indicators (BDIs) with received data on a cross-chip link

A method is provided for address mapping in a network processor. The method includes the steps of (1) determining a port number of a port that receives a data cell; (2) determining a virtual path identifier and a virtual channel identifier for the data cell; and (3) creating a first index based on at least one of the port number, the virtual path identifier and the virtual channel identifier. The method further includes (1) accessing one of a plurality of entries stored in a first on-chip memory using the first index; (2) creating a second index based on the accessed entry of the first on-chip memory; and (3) accessing an entry of a second memory based on the second index. Numerous other aspects are provided.

Methods and apparatus for indexing memory of a network processor

A data communication apparatus includes a plurality of output ports and a scheduler for assigning priorities for outbound data frames. The scheduler includes one or more scheduling queues. Each scheduling queue indicates an order in which data flows are to be serviced. At least one scheduling queue has a respective plurality of output ports assigned to the scheduling queue. That is, the scheduling queue is shared by two or more output ports.

Weighted fair queue serving plural output ports

In a first aspect, a network processor services a plurality of flows including a first flow and a discard flow. The first flow includes a first flow queue and the discard flow includes a discard queue that lists frames to be discarded. An indication is made that the first flow is to be disabled. In response to the indication, all frames included in the first flow queue are transferred to the discard queue. Because the first flow queue is now empty, reconfiguration of the first flow may proceed immediately.

Network processor having fast flow queue disable process

A method and apparatus are provided for implementing hierarchical scheduling of oversubscribed virtual paths with underutilized bandwidth that works for both ATM (cell) and IP (frame) scheduling. A scheduler includes a first calendar for pipes and autonomous flows and a second calendar for pipe flows. A winner of a pipe or an autonomous flow is identified from the first calendar. Responsive to an identified winner pipe, a pipe queue is checked for an associated pipe flow for the winner pipe. Responsive to identifying an empty pipe queue for the winner pipe, a pipe win credit is assigned to the pipe without reattaching the winner pipe to the first calendar. Then a next winner is identified from the first calendar. When a winner pipe flow is identified from the second calendar and the pipe win credit is assigned to the pipe for the winner pipe flow, then the winner pipe flow is serviced without delay.

Method and apparatus for hierarchial scheduling of virtual paths with underutilized bandwidth

In a first aspect, a method is provided for scheduling connections for a network processor. The method includes the steps of, in a cache, scheduling a plurality of connections to be serviced based on quality of service parameters stored in a control structure corresponding to each connection and during a scheduling opportunity (1) identifying one or more of the plurality of connections in the cache to be serviced; (2) selecting one of the connections identified to be serviced; (3) servicing the selected connection; (4) accessing one or more portions of the control structure in the cache; (5) calculating a next service time when the selected connection is to be serviced; and (6) determining whether to schedule the selected connection to be serviced in one of the cache and a calendar based on the next service time. Numerous other aspects are provided.

High-speed scheduler

A scheduler for a network processor includes one or more scheduling queues. Each scheduling queue defines a respective sequence in which flows are to be serviced. A respective empty indicator is associated with each scheduling queue to indicate whether the respective scheduling queue is empty. By referring to the empty indicators, it is possible to avoid wasting operating cycles of the scheduler on searching scheduling queues that are empty.

Empty indicators for weighted fair queues

A scheduler for a network processor includes a scheduling queue in which weighted fair queuing is applied. The scheduling queue has a range R. Flows are attached to the scheduling queue at a distance D from a current pointer for the scheduling queue. The distance D is calculated for each flow according to the formula D=((WF&#xd7;FS)/SF), where WF is a weighting factor applicable to a respective flow; FS is a frame size attributable to the respective flow; and SF is a scaling factor. The scaling factor SF is adjusted depending on a comparison of the distance D to the range R.

Weighted fair queue having adjustable scaling factor

In a first aspect, a network processor includes a scheduler in which a scheduling queue is maintained. A last frame is dispatched from a flow queue maintained in the network processor, thereby emptying the flow queue. Data indicative of the size of the dispatched last frame is stored in association with the scheduler. A new frame corresponding to the emptied flow queue is received, and the flow corresponding to the emptied flow queue is attached to the scheduling queue. The flow is attached to the scheduling queue at a distance D from a current pointer for the scheduling queue. The distance D is determined based at least in part on the stored data indicative of the size of the dispatched last frame.

Growing community of inventors

Rochester, MN, United States of America

James Francis Mikos