The following code fragments each compute the same results. Which would incur the least delay due to pipelining on a computer with a single functional unit? You can assume all memory loads take the same time. Select one: a. leaq 8(%rbx), %rdi leaq (%rdi), %rsi sar $4,%rsi movq 16(%ebc), %rbi movq (%rbi),%rax sal $2,%rax addq %rsi,%rax b. leaq 8(%rbx), %rdi leaq 16(%ebc), %rbi movq (%rdi), %rsi movq (%rbi),%rax sar $4,%rsi sal $2,%rax addq %rsi,%rax c. leaq 8(%rbx), %rdi leaq 16(%ebc), %rbi mova (%rdi), %rsi sar $4,%rsi movq (%rbi),%rax sal $2,%rax addq %rsi,%rax