Unverified Commit e3d52fbe authored by Marc24force's avatar Marc24force Committed by GitHub
Browse files

Typo correction in readme

parent 20288623
# Xilinx Open Hardware - Hardware-Software Co-design forLow-cost AI processing in SpaceProcessors
# Xilinx Open Hardware - Hardware-Software Co-design for Low-cost AI processing in Space Processors
## Project description
This project started as part of my master course Processor Design, later it was extended as my master thesis. With this project I compete in the Xilinx Open Hardware competition in the student category. For the project I have modified the LEON3 space processor and have included a SIMD module with features targeting acceleration in artificail intelligence applications.
This project started as part of my master course Processor Design, later it was extended as my master thesis. With this project I compete in the Xilinx Open Hardware competition in the student category. For the project I have modified the LEON3 space processor and have included a SIMD module with features targeting acceleration in artificial intelligence applications.
The SIMD module, operates over the integer unit registers (SIMD within a register, aka SWAR) and has two stages. In the first stage both input operands are operated against each other at byte granularity, the result is passed to the second stage where reduction operations are implemented. Additionally a mask vector can restrict the bytes to be computed in the first stage and swizzling option is included.
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment