AVX-512 Programming – Floating-Point
In previous chapters, you learned how to carry out scalar and packed floating-point operations using the AVX and AVX2 instruction sets. In this chapter, you learn how to perform these operations using the AVX-512 instruction set. The first part of this chapter contains source code examples that illustrate basic AVX-512 programming concepts using scalar floating-point operands. This includes examples that illustrate conditional executions, merge and zero masking, and instruction-level rounding. The second part of this chapter demonstrates how to use the AVX-512 instruction set to carry out packed floating-point calculations using 512-bit wide operands and the ZMM register set.