X86-AVX Programming - Packed Floating-Point
In Chapter 9, you became familiar with the packed floating-point resources of x86-SSE. In this chapter, the packed floating-point capabilities of x86-AVX are explored. The sample programs illustrate how to perform basic packed floating-point arithmetic using 256-bit wide operands. They also exemplify use of the x86-AVX instruction set to carry out arithmetic operations using floating-point arrays and matrices. All of the sample programs in this chapter require a processor and operating system that supports AVX. As a reminder, Appendix C contains a list of several freely-available tools that you can use to ascertain whether your PC supports AVX.