Lines Matching defs:cwnd
251 * may be waiting for an ACK if cwnd-bound or using TX zero
339 * cwnd may be very low (even just 1 packet), so we should ACK
370 /* Better not delay acks, sender can have a very low cwnd */
908 /* set sk_pacing_rate to 200 % of current rate (mss * cwnd / srtt) */
911 /* current rate is (cwnd * mss) / srtt
949 * to do with delayed acks, because at cwnd>2 true delack timeout
969 __u32 cwnd = (dst ? dst_metric(dst, RTAX_INITCWND) : 0);
971 if (!cwnd)
972 cwnd = TCP_INIT_CWND;
973 return min_t(__u32, cwnd, tp->snd_cwnd_clamp);
2318 * with the same cwnd?
2560 /* Try to undo cwnd reduction, because D-SACKs acked all retransmitted data */
2601 /* The cwnd reduction in CWR and Recovery uses the PRR algorithm in RFC 6937.
2605 * cwnd reductions across a full RTT.
2608 * slow starts cwnd up to ssthresh to speed up the recovery.
2658 /* Reset cwnd to ssthresh in CWR or Recovery (unless it's undone) */
2667 /* Enter CWR state. Disable cwnd undo since congestion is proven with ECN */
2773 * cwnd/ssthresh really reduced now.
2835 /* Step 2.b. Try send new data (but deferred until cwnd
2932 * is detected. But it does not reduce the cwnd, it is done by the
2952 * A. ECE, hence prohibit cwnd undoing, the reduction is required. */
3017 /* Change state if cwnd is undone or retransmits are lost */
3035 /* MTU probe failure: don't reduce cwnd */
3463 /* If reordering is high then always grow cwnd whenever data is
3465 * and only grow cwnd on in-order delivery (RFC5681). A stretched ACK w/
3466 * new SACK or ECE mark may first advance cwnd here and later reduce
3467 * cwnd in tcp_fastretrans_alert() based on more states.
3477 * cwnd increase and decrease control (tcp_cong_avoid,tcp_*cwnd_reduction).
3492 /* Reduce cwnd if state mandates */
3495 /* Advance cwnd if state allows */
3689 /* ACK advances: there was a loss, so reduce cwnd. Reset
3713 /* Congestion control has updated the cwnd already. So if we're in
3914 /* If data was DSACKed, see if we can undo a cwnd reduction. */
3932 * If data was DSACKed, see if we can undo a cwnd reduction.
5990 * Cut cwnd down to 1 per RFC5681 if SYN or SYN-ACK has been
5992 * initRTO, we only reset cwnd when more than 1 SYN/SYN-ACK