PIC32MX Fractional or Q15_16 support for matrix or vector arithmetic (Paul Bjork)

The XC32 v1.30 Libraries seem to be very light on support for Q15_16 arithmetic that is set up for PIC32MX (not MZ). I see a few functions like mips_vec_mul32() that use Q31 fractional format operands (-1<= x <+1), but this is not usable for