12.28 __smlsldx intrinsic
This intrinsic inserts an
SMLSLDX instruction into the instruction stream generated by the compiler.
you to exchange the halfwords of the second operand, perform two
16-bit multiplications, adding the difference of the products to
a 64-bit accumulate operand. Overflow cannot occur during the multiplications
or the subtraction. Overflow can occur as a result of the 64-bit
addition, and this overflow is not detected. Instead, the result
wraps round to modulo 264.
unsigned long long __smlsldx(unsigned int
, unsigned int
, unsigned long long
holds the first halfword operands for each multiplication
holds the second halfword operands for each multiplication
holds the accumulate value.
__smlsld intrinsic returns the difference
of the product of each multiplication, added to the accumulate value.
unsigned long long dual_multiply_diff_prods(unsigned int val1, unsigned int val2, unsigned long long val3)
unsigned int res;
res = __smlsld(val1,val2,val3); /* p1 = val1[15:0] × val2[31:16]
p2 = val1[31:16] × val2[15:0]
res[63:0] = p1 - p2 + val3[63:0]