vDSP_maxv

Declaration

void vDSP_maxv(const float *__A, vDSP_Stride __IA, float *__C, vDSP_Length __N);

Parameters

• __A:

  Single-precision real input vector.

• __I:

  Stride for A.

• __C:

  Output scalar.

• __N:

  The number of elements to process. If N is zero (0), this function returns -INFINITY.

Discussion

This function calculates the maximum value of the first N elements of A and writes the result to C:

[Image]

The operation is:

*C = -INFINITY;
for (n = 0; n < N; ++n)
    if (*C < A[n*I])
        *C = A[n*I];

The following code shows an example of using vDSP_maxv.

let stride = vDSP_Stride(1)

let a: [Float] = [-1.5, 2.25, 3.6,
                  0.2, -0.1, -4.3]
let n = vDSP_Length(a.count)

var c: Float = .nan

vDSP_maxv(a,
          stride,
          &c,
          n)

print("max", c) // Prints "max 3.6"