Speeding up the multiplying, adding, subtracting of 3D matrices

This is what I mean:

julia> C, A, B = (rand(91,91,91) for _ in 1:3); range = 1:30;

julia> function f!(C,A,B,r)
           C[r,r,r] .= A[r,r,r] .* B[r,r,r]
       end
f! (generic function with 1 method)

julia> @btime f!($C, $A, $B, $range);
  67.778 μs (4 allocations: 422.00 KiB)

julia> @views function f!(C,A,B,r)
           C[r,r,r] .= A[r,r,r] .* B[r,r,r]
       end
f! (generic function with 1 method)

julia> @btime f!($C, $A, $B, $range);
  21.476 μs (0 allocations: 0 bytes)

and, please, follow the guidelines here: Please read: make it easier to help you