System and method for performing shaped memory access operations

Abstract

One embodiment of the present invention sets forth a technique that provides an efficient way to retrieve operands from a register file. Specifically, the instruction dispatch unit receives one or more instructions, each of which includes one or more operands. Collectively, the operands are organized into one or more operand groups from which a shaped access may be formed. The operands are retrieved from the register file and stored in a collector. Once all operands are read and collected in the collector, the instruction dispatch unit transmits the instructions and corresponding operands to functional units within the streaming multiprocessor for execution. One advantage of the present invention is that multiple operands are retrieved from the register file in a single register access operation without resource conflict. Performance in retrieving operands from the register file is improved by forming shaped accesses that efficiently retrieve operands exhibiting recognized memory access patterns.

Description

technical field [0001] The present invention relates generally to computer architecture, and more particularly to systems and methods for operand collection in register files. Background technique [0002] A common practice in parallel processing systems is to design processors that execute multiple threads simultaneously. When such threads are all executing the same sequence of instructions (typically with different data for each thread), there is a practical benefit to sharing certain resources among the threads. For example, each thread may execute an instruction that accesses one or more operands to be retrieved from a shared register file bank that each thread accesses different register addresses within. This type of operation can be found on Single Instruction Multiple Thread (SIMT) and Single Instruction Multiple Data (SIMD) processors. [0003] During operation, the processor may execute instructions across multiple threads, where the instruction accesses one or ...

AI Technical Summary

Problems solved by technology

However, the distribution of operands throughout the bank of the register file makes it possible for the processor to retrieve all operands in fewer than a dozen register access operations

In such a situation, the potential efficiencies associated with memory access operations are not realized

Images