dest[0] = dest[0] - source[0] dest[1] = dest[1] - source[1]
The subpd instruction subtracts the 2 source values (second operand) from the 2 values of the destination (an XMM register). The source can be an XMM register or a 64 bit memory location. There is also vsubpd on CPUs with AVX instructions which allows using 3 XMM registers or 2 XMM registers and a memory location which can simplify coding and which subtracts 4 pairs of values if you use YMM registers.
subpd xmm0, xmm1 ; subtract 2 pairs from xmm1 and xmm0 ; 2 differences go in xmm0 ; leave the rest of ymm0 as is subpd xmm0, [x] ; subtract 2 pairs from x and xmm0 ; x is an array of doubles ; leave the rest of ymm0 as is subpd xmm0, [x+8*rax] ; subtract 2 pairs from x and xmm0 ; x is an array of doubles ; rax holds the index of the first element ; leave the rest of ymm0 as is vsubpd xmm3, xmm0, xmm15 ; subtract 2 pairs from xmm0 & xmm15 ; store results in xmm3 vsubpd ymm3, ymm0, [x] ; subtract 4 pairs values from ymm0 & x ; x is a double array ; store results in ymm3 vsubpd ymm3, ymm0, [rsi] ; subtract 4 pairs values from ymm0 & [rsi] ; rsi contains the address of a double array ; store results in ymm3