similar to: greb + htb debug

The HTB scheduler code is a mess, this patch set does some basic house cleaning. The first four should cause no code change, but the last two need more testing. -- Stephen Hemminger <shemminger@osdl.org> "And in the Packet there writ down that doome" - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to

The HTB qdisc has a compile time option, HTB_HYSTERESIS, that trades accuracy of traffic classification for CPU time. These patches change hysteresis to be a runtime option under the control of "tc". The effects of HYSTERESIS on HTB''s accuracy are significant (see chapter 7, section 7.3.1, pp 69-70 in Jesper Brouer''s thesis: http://www.adsl-optimizer.dk/thesis/ ),

Hello, We know, VQs are used in Gred to offer different schedulers, in order to make clear what the VQs are, I have read the sch_gred.c in net/sched/. In the function gred_enqueue(), at first it will find the VQ(q) according to the lower four bits of the skb->tc_index, then some parameters of the q is modified, such as q->packetsin and q->bytesin. At last the packet was put into the

Hi, I have made some changes to the file sch_gred.c to modify the GRED queueing discipline to support the following features: 1) The first virtual queue should get absolute priority while dequeueing (not caring if the others get starved) 2) While in equalise mode and with RIO mode enabled, the packets in the first virtual queue should not be counted for calculating the qave. I want to confirm if

Hi, I am tring to setup AF on my router using GRED/HTB. I plan to mark and set the DSCP with iptables and the ingress qdisc to set the tc_index for the proper GRED vq(1-12). My question is: Is there a way to allow for priority between the af classes and within the af classes? also would it be possible to to the following: HTB(6Mb) | GRED(1)grio HTB1 (af1) (1MB) GRED(1-3)grio |

ALL: <<diffserv-gred_10-06>> I have attached the current script that I am using. $TC qdisc add dev $EDEV parent 2:20 gred setup DPs 3 default 2 grio $TC qdisc change dev $EDEV parent 2:20 gred DP 1 limit $lim min $minTh max $maxTh avpkt $avgPL burst $bursty bandwidth $netBand probability 0.02 prio 2 $TC qdisc change dev $EDEV parent 2:20 gred DP 2 limit $lim min $minTh max

Hi, I am trying to get assured forwarding/expedited forwarding with gred and htb working. Below is the script I am using. The following steps are what I thing is how the script works. My problem is that if I remove the HTB qdisc from the script and have the GREDS parent as the dsmark it works, but when I add the htb as a parent of GRED and DSmark the parent of htb it does not work? Any

Hi, just wondering if this type of configuration would be possible for using AF and EF classes. HTB (6.0M) --GRED 5 (AF1) --HTB(1M) [is it possible to have this with a 1M rate and a 6M ceiling?] --GRED [3 greds using grio] VQ1 -3 --GRED 4 (AF2) --HTB(1M) --GRED VQ1 -3 --GRED 3 (AF3) --HTB(1M)

I ''ve been trynig for a long time now to make HTB and GRED to work togethen. The problem beeing that GRED doesn''t use handles (instead it uses DP:1 DP:2 etc) i can not preperly filter traffic to them. Tomas Graf suggested to use the tc_index index of u32 classifier so dear Sant .. i am currently with the following scripts that works! But when i try to add an HTB before the

Hi, I used HTB3.6 on Linux to control traffic.It seems that HTB uses DRR(Dificit Round-Robin) algorithm.I heard that it is difficult for DRR to get a delay bounded and WFQ is better than DRR.Is that right?So,how to implement WFQ in Linux? Thanks a lot! Best Regards swcims swcims@163.com 2004-08-03 _______________________________________________ LARTC mailing list

Hi i have configured my server with cbq and gred. The problem is that the cbq doesnt drop packets. The cbq is configured at a rate of 500Kbit "qdisc cbq 4: rate 500000bit (bounded,isolated) prio no-transmit Sent 100103048 bytes 73933 pkts (dropped 0, overlimits 0) borrowed 0 overactions 0 avgidle 24000 undertime 0 " The traffic that being generated is more than 5Mbit with the

All: I am tasked with creating an environment in which DiffServ and IP precedence is handled as close to RFC specification as possible. I have a few problems with implementing this based on current queueing structure. Here are a few of my questions. 1.) Is there any way to get DiffServ AF drop levels without using DSMARK and GRED? Ex. AF11 Gred DP 1 AF12 Gred DP 2 AF13 Gred DP 3 2.)

Hi Andreas! I mainly understand what you mean, but best for me would be if you could make an example from the script I have. I know I did something wrong into it, but I still can''t see the error. If you could fix the error from my script that would be much appreciated. Looking forward to hear from you, Daniel Phlander lartc-request@mailman.ds9a.nl wrote: Send LARTC mailing list

Hello, I think the BUG_TRAP() in the htb_dequeue_tree() is wrong. First it checks if the class pointer "cl" is NULL, which is obviously right. But I do not understand why we also check whenever the queue length of the leaf queue is zero "cl->un.leaf.q->q.qlen". I would have put that in the expression of the "if" statements that comes afterwards. A queue

Hello all, Please could someone help me with this, I have been trying for days to get this to work. I would like to have BE, AF & EF classes with HTB qdisc. I can''t find any scripts online where this has been done. If I enable this filter: #tc filter add dev eth0 parent 2:0 protocol ip prio 1 \ #tcindex mask 0xf0 shift 4 pass_on then BE & AF classification seems to be fine but

Hi Andreas! I mainly understand what you mean, I tried to fix something on the script, I don''t know if I did it well. Can you take another look on it please and if is wrong to make the corrections directly on it so that I see where the mistake is... With this script I want to make limits for IP class 85.120.48.0/25 for international traffic in 256 KBps classes and for metropolitan

Anyone using it? I''ve tried but after about 5 min I always get kernel panic. My setup is based on nano.txt. I works well but only if CONFIG_IP_ROUTE_MULTIPATH_CACHED=n. I just wanted to play with the new mpath feature of ip. But enabling CONFIG_IP_ROUTE_MULTIPATH_CACHED always resuts in kernel panic. I''m not using any kernel patches from http://www.ssi.bg/~ja/ - are they needed

hi, Is there a way to calculate Gred parameters, given a desired delay (e.g 100ms)? Thanx Ntanzi ____________________________________________________________________________________ Do you Yahoo!? Everyone is raving about the all-new Yahoo! Mail beta. http://new.mail.yahoo.com

Hi, In the GRED queueing discipline of linux traffic control, we can define around 16 Virtual Queues with different RED parameters for each one of these. As far as I think, while applying RED on each of the virtual queues, with the grio option set, quave(the average queue length) of a virtual queue with priority= n is given by the sum of the virtual queues with priority greater than n ... I have

Has anyone ever managed to use GRED and HTB ? Been googling all over the net for hours and did not manage to find any proper implementation. In my case all AF traffic goes into VQ 3 In opalsoft afhtb example traffic goes into VQ 1 & 2 any script will do just fine for me Thanks in advance , Antonio _______________________________________________ LARTC mailing list / LARTC@mailman.ds9a.nl