About TcpRateOps::TcpRateSample & TcpRateLinux::GenerateSample(...)

[Ruihua Chan]

Hello all,

I found that in function const TcpRateOps::TcpRateSample &

TcpRateLinux::GenerateSample ( ) when m_rateSample.m_interval < minRtt it will set the variable m_rateSample.m_priorTime to Seconds(0).

if (m_rateSample.m_interval < minRtt) 

{ 

  NS_LOG_INFO ("Sampling interval is invalid"); 

  m_rateSample.m_interval = Seconds (0); 

  m_rateSample.m_priorTime = Seconds (0); // To make rate sample invalid                 m_rateSampleTrace (m_rateSample); return m_rateSample;

}

However I could not find out similar implementation in linux kernel codes when interval < minRTT. Linux kernel codes as below.

/net/ipv4/tcp_rate.c: void tcp_rate_gen(struct sock *sk, u32 delivered, u32 lost,

          bool is_sack_reneg, struct rate_sample *rs)

if (unlikely(rs->interval_us < tcp_min_rtt(tp))) {
        if (!rs->is_retrans)
            pr_debug("tcp rate: %ld %d %u %u %u\n",
                 rs->interval_us, rs->delivered,
                 inet_csk(sk)->icsk_ca_state,
                 tp->rx_opt.sack_ok, tcp_min_rtt(tp));
        rs->interval_us = -1;
        return;
    }

Could you please help me find out where could I find the similar implementation in linux kernel or tell me how does this new line of code work to make sample invalid.Thanks in advance!

Regards,

Chan