Data processing device and method based on vector processor
A vector processor and vector technology, applied in the direction of processing data, register devices, machine execution devices, etc. according to predetermined rules, can solve the problems of weakening the fast computing capability of the vector processor, long judgment time, and increasing hardware costs, so as to solve the problem of hardware Effects of Inherent Properties
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0021] The vector read-in unit VL performs a data read-in operation according to a certain fixed hardware transmission capability of the vector read-in unit VL. The certain fixed hardware transmission capability mentioned here refers to that the hardware transmission capability can select a certain fixed value within a certain range, but cannot be a variable value with arbitrary attributes.
[0022] The vector writing unit VS performs a data writing operation according to a certain fixed hardware transmission capability of the vector writing unit VS. The certain fixed hardware transmission capability mentioned here refers to that the hardware transmission capability can select a certain fixed value within a certain range, but cannot be a variable value with arbitrary attributes.
[0023] Due to the hardware properties of the vector processor VP, when selecting the hardware computing capability, only the computing capability value greater than or equal to the actual required va...
Embodiment 2
like image 3 As shown, this embodiment provides a data processing method based on a vector processor, including:
Step S1: utilize the vector read-in unit VL to read in the vector input data to the register of the vector processor VP;
Step S2: using the vector processor VP to perform vector operation on the vector input data to obtain the actual vector output data;
Step S3: use the tail processing unit TP to obtain the recovered data; the recovered data is the data covered by the excess data of the vector actual output data in the internal data storage unit MEM; the excess data and the vector theoretical output data form the actual output data of the vector;
Wherein, step S3 specifically includes:
(1) Use the tail processing unit TP to calculate the start address and length of the recovered data in the internal data storage unit MEM; specifically, it includes:
Utilize the tail processing unit TP to obtain the length of the vector input data and the starting address of...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


