[prev in list] [next in list] [prev in thread] [next in thread] 

List:       dpdk-users
Subject:    [dpdk-users] Help with RED (Dropper) Implementation
From:       Gokul Bargaje <gokulbargaje.182009 () nitk ! edu ! in>
Date:       2019-09-19 14:20:36
Message-ID: CAH1-=xiP_cMT0qo0+-S4MKsnbV7p4=v-rQOfdSk=vPVARutaTQ () mail ! gmail ! com
[Download RAW message or body]

Hi,

I am trying to understand the implementation of Random Early Detection
(RED) AQM in DPDK. I have read the dpdk/lib/librte_sched/rte_red.c and
rte_red.h files but i don't understand the implementation.

Please help me to understand the following questions.
1. How RED parameters are configured?
2. What is *wq_log2*, *maxp_inv* and How their default values are
determined? (original terms in RED paper are wq and maxp)
3. What is *uint16_t rte_red_log2_1_minus_Wq[RTE_RED_WQ_LOG2_NUM]* and
*uint16_t
rte_red_pow2_frac_inv[16]* and how the size of these arrays is determined?
4. I found that wq is calculated using lookup tables, how this is actually
implemented?

Thanks in advance!

Thanks,
Gokul
[prev in list] [next in list] [prev in thread] [next in thread]