rack, bbr: cleanup ack throttling

[tuexen]

Summary

Use the variable in the TCPCB, not the one in the stack specific data structure. This simplifies the code and brings the functionality to BBR without any change.

This changes builds upon D46066.

[cc]

sys/netinet/tcp_stacks/rack_bbr_common.c

535

I think the logic here can be split into two functions, such that the refined logic 1 on checking if we should send ACK can be re-used across multiple places.

logic 1:

bool
is_ack_unlimited(struct tcpcb *tp)
{
    /* focus on checking the epoch and
     * if should send ACK, return true; else return false
     */ 
}

logic 2:

void
set_tp_acknow(struct tcpcb *tp)
{
    if (is_ack_unlimited(tp))
        tp->t_flags |= TF_ACKNOW;
    else
        tp->t_flags &= ~TF_ACKNOW;
}

761–762

From my above comment, we can re-use the logic 1. Like this:

if (is_ack_unlimited(tp))
    /* Send challenge ACK. */

[peter.lei_ieee.org]

sys/netinet/tcp_stacks/rack_bbr_common.c

761–762

I think this code block could be replaced by tcp_send_challenge_ack() (and leaving the counter increment)?

[tuexen]

• Rebase after vnetifying sysctl-able variables.
• Address Peter's comment.

[tuexen]

sys/netinet/tcp_stacks/rack_bbr_common.c

535

I would like to reuse tcp_send_challenge_ack() instead of the alternate used here. But that is a followup commit.

761–762

Not needed, we can just use tcp_send_challenge_ack().

[cc]

sys/netinet/tcp_stacks/rack_bbr_common.c

535

I would like to reuse tcp_send_challenge_ack() instead of the alternate used here. But that is a followup commit.

I am OK if you have a plan to replace ctf_ack_war_checks() with tcp_send_challenge_ack() for the re-use.

761–762

Not needed, we can just use tcp_send_challenge_ack().

Thanks for the update.