Functions§
- add_
8x4_ ⚠kernel - Performs an 8x4 micro-kernel matrix multiplication using AVX2 and FMA instructions.
- add_
scalar_ ⚠kernel - A scalar micro-kernel for handling matrix multiplication on edge cases.
- gemm⚠
- Computes the matrix-matrix product C = alpha * op(A) * op(B) + beta * C.
- pack⚠
- Packs a panel of a matrix into a contiguous buffer, handling transposition.