Nvidia Corporation

Raymond Hoi Man Wong

Samuel Hammond Duncan

Lukito Muliadi

Madhukiran V Swarna

Wishwesh Anil Gandhi

John H Edmondson

Debajit Bhattacharya

Various embodiments include techniques for performing speculative remote memory operation tracking in a multiprocessor computing system. Conventionally, transfers of data between processors and other components of a computing system require memory synchronization operations to determine that the data is valid and coherent before the data is transferred from a destination to a requesting source. Existing techniques for performing these memory synchronization operations are increasingly inefficient as the number of components in a computing system increases, particularly for remote memory operations. The disclosed techniques track remote memory operations and speculatively perform these memory synchronization operations. As a result, a given memory synchronization operation is often complete prior to the corresponding remote memory operation arrives at the destination, leading to improved efficiency and performance of remote memory operations in complex computing systems.

Speculative remote memory operation tracking for efficient memory barrier

The MMU services data requests associated with isochronous (ISO) data (referred to herein as 'ISO requests') with a high priority to meet a fixed latency requirement. Such data includes display data for transmission to the display device or other display devices. Conversely data requests associated with non-isochronous (NISO) data are serviced with a relatively lower priority. Such data requests include requests received from the CPU, video requests and copy requests. The MMU utilizes a buffering mechanism to buffer ISO and NISO requests. The size of the buffer that stores ISO requests controls the amount of memory bandwidth that is allocated to the ISO requests and the amount of memory bandwidth available for NISO requests.

Memory bandwidth reallocation for isochronous traffic

One or more client engines issues write transactions to system memory or peer parallel processor (PP) memory across a peripheral component interconnect express (PCIe) interface. The client engines may issue write transactions faster than the PCIe interface can transport those transactions, causing write transactions to accumulate within the PCIe interface. To prevent the accumulation of write transactions within the PCIe interface, an arbiter throttles write transactions received from the client engines based on the number of write transactions currently being transported across the PCIe interface.

Method and apparatus for equalizing a bandwidth impedance mismatch between a client and an interface

One embodiment of the present invention sets forth a technique for arbitrating between a set of requesters that transmit data transmission requests to the weighted LRU arbiter. Each data transmission request is associated with a specific amount of data to be transmitted over the crossbar unit. Based on the priority state associated with each requester, the weighted LRU arbiter then selects the requester in the set of requesters with the highest priority. The weighted LRU arbiter then decrements the weight associated with the selected requester stored in a corresponding weight store based on the size of the data to be transmitted. If the decremented weight is equal to or less than zero, then the priority associated with the selected requester is set to a lowest priority. If, however, the decremented weight is greater than zero, then the priority associated with the selected requester is not changed.

Fast and highly scalable quota-based weighted arbitration

An input/output unit for a computer system that is interfaced with a memory unit having a plurality of partitions manages completions of read requests in the order that they were made. A read request buffer tracks the order in which the read requests were made so that read data responsive to the read requests can be completed and returned to a requesting client in the order the read requests were made.

Read completion data management

A data packer of an input/output hub of a computer system packs and formats write data that is supplied to it before the write data is written into a memory unit of the computer system. More particularly, the data packer accumulates write data received from lower bandwidth clients for delivery to a high bandwidth memory interface. Also, the data packer aligns the write data, so that when the write data is read out from the write data packer, no further alignment is needed.

Data packer for packing and aligning write data

Read completion buffer space is allocated in accordance with a preset limit. When a read request is received from a client, the sum of a current allocation of the read completion buffer space and a new allocation of the read completion buffer space required by the read request is compared with the preset limit. If the preset limit is not exceeded, read completion buffer space is allocated to the read request. If the preset limit is exceeded, the read request is suspended until sufficient data is read out from the read completion buffer.

Growing community of inventors

Santa Clara, CA, United States of America

Raymond Hoi Man Wong