 | 1 | initial version |
I have seen delayed ACKs lots of times (from multiple stacks, including Windows and Linux), sometimes even in faulty ways when a push flag should have triggered an instant ACK. So I'm not sure how you come to the conclusion that it isn't used? Whenever a single data packet without PSH flag arrives, a 200ms ACK delay happens, waiting for more data to arrive. But maybe you have a different definition of "TCP delayed ACK"?
