Function `_mm_mpsadbw_epu8`

#[target_feature(enable = "sse4.1")]
pub fn _mm_mpsadbw_epu8(a: __m128i, b: __m128i, IMM8: i32) -> __m128i

Subtracts 8-bit unsigned integer values and computes the absolute values of the differences to the corresponding bits in the destination. Then sums of the absolute differences are returned according to the bit fields in the immediate operand.

The following algorithm is performed:

i = IMM8[2] * 4
j = IMM8[1:0] * 4
for k := 0 to 7
    d0 = abs(a[i + k + 0] - b[j + 0])
    d1 = abs(a[i + k + 1] - b[j + 1])
    d2 = abs(a[i + k + 2] - b[j + 2])
    d3 = abs(a[i + k + 3] - b[j + 3])
    r[k] = d0 + d1 + d2 + d3

Arguments:

a - A 128-bit vector of type __m128i.
b - A 128-bit vector of type __m128i.
IMM8 - An 8-bit immediate operand specifying how the absolute differences are to be calculated
- Bit [2] specify the offset for operand a
- Bits [1:0] specify the offset for operand b

Returns:

A __m128i vector containing the sums of the sets of absolute differences between both operands.

Intel's documentation