公开(公告)号：WO2003021423A3

公开(公告)日：2003-03-13

申请号：PCT/US2002/027970

申请日：2002-09-04

Applicant: MICROUNITY SYSTEMS ENGINEERING, INC.

Inventor： HANSEN, Craig ,  BATEMAN, Bruce ,  MOUSSOURIS, John

IPC: G06F7/48

Abstract: A vector-matrix multiplier unit fully utilizes a 128x128b data path for operand sizes from 8 to 128b and operand types including signed, unsigned or complex, and fixed-, floating-point, polynomial, or Galois-field while maintaining full internal precision. The present disclosure provides a system and method for improving the performance of general-purpose processor, by implementing a functional unit that computes the product of a matrix operand with a vector operand, producing a vector result. The functional unit fully utilizes the entire resources of a 128b by 128b multipliers regardless of the operand size, as the number of elements of the matrix and vector operands increase as operand size is reduced. The unit performs both fixed-point and floating-point multiplications and additions with the highest-possible intermediate accuracy with modest resources.

公开(公告)号：WO2003021423A2

公开(公告)日：2003-03-13

申请号：PCT/US2002/027970

申请日：2002-09-04

Applicant: MICROUNITY SYSTEMS ENGINEERING, INC.

Inventor： HANSEN, Craig ,  BATEMAN, Bruce ,  MOUSSOURIS, John

IPC: G06F7/00

CPC classification number: G06F7/5338 ,  G06F7/4812 ,  G06F7/49994 ,  G06F7/57 ,  G06F7/724 ,  G06F9/30014 ,  G06F9/30036 ,  G06F17/15 ,  G06F17/16 ,  G06F2207/3812 ,  G06F2207/382 ,  G06F2207/3824 ,  G06F2207/3828

Abstract: The present invention provides a system and method for improving the performance of general-purpose processors by implementing a functional unit that computes the product of a matrix operand with a vector operand, producing a vector result. The functional unit fully utilizes the entire resources of a 128b by 128b multipliers regardsless of the operand size, as the number of elements of the matrix and vector operands increase as operand size is reduced. The unit performs both fixed-point and floating-point multiplications and additions with the highest-possible intermediate accuracy with modest resources.

Abstract translation: 本发明提供了一种用于通过实现一个功能单元来提高通用处理器的性能的系统和方法，所述功能单元使用向量操作数来计算矩阵操作数的乘积，产生向量结果。 随着操作数大小减小，矩阵和向量操作数的元素数量增加，功能单元完全利用128b乘128b乘法器的全部资源，而不考虑操作数大小。 该单元通过适度的资源执行具有最高可能的中间精度的定点和浮点乘法和补充。