Some work to extend SFQ to be more configurable
Subsystems (4)
Hi there!
Quoted 6 lines · Martin Devera
This is often discussed and is on "TODO" for someone ;)
Quoted 4 lines
SFQ is connection oriented. right?
Would be a good idea to make the queues per ip rather than per tcp flow? So there would be per host fairnes.
I've done some in this direction , probably needs more work, and
it's poorly tested - expect b00ms ;)
This adds a new qdisc for now - esfq which is a 100% clone of
original sfq.
- You can set all sfq parameters: hash table size, queue depths,
queue limits.
- You can choose from 3 hash types: original(classic), dst ip, src
ip.
Things to consider: perturbation with dst and src hashes is not
good IMHO, you can try with perturb 0 if it couses trouble.
Please, see the attached files.
Plaing with it gives interesting results:
higher depth -> makes flows equal slower
small depth -> makes flows equal faster
limit kills big delays when set at about 75-85% of depth.
Needs testings and mesurements - that's why i made it
separate qdisc and not a patch over sfq, i wanted to compare both.
Any feedback good or bad is welcome.
Signature · 2 lines
have fun, alex
You need: iproute2-2.2.4-now-ss001007.tar.gz linux-2.4.18.tar.gz This may work with next versions too. Example Step by Step install ------------------------------------------------------------------ Install kernel: tar zxvf linux-2.4.18.tar.gz cd linux cat linux-2.4.18-esfq.diff | patch -p1 make menuconfig Now you have it in: Networking options ---> QoS and/or fair queueing ---> ESFQ queue Configure and install kernel make dep clean bzImage modules_install cp System.map /boot/ cp arch/i386/boot/bzImage /boot/bzImage Edit lilo.conf and add your new kernel. ------------------------------------------------------------------ Install iproute: tar zxvf iproute2-2.2.4-now-ss001007.tar.gz cd iproute2 cat iproute2-2.2.4-now-ss001007-esfq.diff | patch -p1 make
/* Note: If you get an error in lib/ll_proto.c comment _PF(ECHO,echo) */
cp tc/tc your_favourite_iproute_path/tc
cp ip/ip your_favourite_iproute_path/ip
------------------------------------------------------------------
Usage: ... esfq [ perturb SECS ] [ quantum BYTES ] [ depth FLOWS ] [ divisor HASHBITS ] [ limit PKTS ] [ hash HASHTYPE]
Where:
HASHTYPE := { classic | src | dst }
Examples:
tc qdisc add dev eth0 root esfq limit 128 depth 128 divisor 10 \
hash classic perturb 15
Setups a classic SFQ.
tc qdisc add dev eth0 root esfq limit 64 depth 64 divisor 11 \
hash dst
Setups a dst SFQ with limit and depth of 64 packets and 11bits (2048 rows) hash table. 1:1 with sch_sfq.
You can experiment with the values as you like to find the best which sfq can do for you.
More can be found in:
linux/net/sched/sch_sfq.c
linux/net/sched/sch_esfq.c
Limits:
- limit must be less than depth
- divisor must be less than 15
tc/Makefile +1 -0
@@ -5,6 +5,7 @@ TCMODULES := TCMODULES += q_fifo.o TCMODULES += q_sfq.o +TCMODULES += q_esfq.o TCMODULES += q_red.o TCMODULES += q_prio.o TCMODULES += q_tbf.o
tc/q_esfq.c +169 -0
169 lines not on this page · raw message
include/linux/pkt_sched.h renamed from include/linux/pkt_sched.h.orig +10 -0
@@ -157,6 +157,13 @@/* SFQ section */ +enum +{ +TCA_SFQ_HASH_CLASSIC , +TCA_SFQ_HASH_DST , +TCA_SFQ_HASH_SRC , +}; +struct tc_sfq_qopt {unsigned quantum;/* Bytes per round allocated to flow */
@@ -164,6 +171,7 @@__u32 limit;/* Maximal packets in queue */ unsigned divisor;/* Hash divisor */ unsigned flows;/* Maximal number of flows */ +unsigned hash_kind;/* Hash function to use for flow identification */ };/*
@@ -173,6 +181,8 @@* * The only reason for this is efficiency, it is possible * to change these parameters in compile time. +* +* If you need to play with this values use esfq. */ /* RED section */
net/ipv4/netfilter/ipchains_core.c renamed from net/ipv4/netfilter/ipchains_core.c.orig +1 -0
@@ -723,6 +723,7 @@ src_port, dst_port, count, tcpsyn)) { ret =FW_BLOCK ; +cleanup (chain,0 , slot); goto out; }break ;
net/sched/Makefile renamed from net/sched/Makefile.orig +1 -0
@@ -17,6 +17,7 @@ obj-$(CONFIG_NET_SCH_HPFQ) += sch_hpfq.o obj-$(CONFIG_NET_SCH_HFSC) += sch_hfsc.o obj-$(CONFIG_NET_SCH_SFQ) += sch_sfq.o +obj-$(CONFIG_NET_SCH_ESFQ) += sch_esfq.o obj-$(CONFIG_NET_SCH_RED) += sch_red.o obj-$(CONFIG_NET_SCH_TBF) += sch_tbf.o obj-$(CONFIG_NET_SCH_PRIO) += sch_prio.o
net/sched/Config.in renamed from net/sched/Config.in.orig +1 -0
@@ -11,6 +11,7 @@ tristate ' The simplest PRIO pseudoscheduler' CONFIG_NET_SCH_PRIO tristate ' RED queue' CONFIG_NET_SCH_RED tristate ' SFQ queue' CONFIG_NET_SCH_SFQ +tristate ' ESFQ queue' CONFIG_NET_SCH_ESFQ tristate ' TEQL queue' CONFIG_NET_SCH_TEQL tristate ' TBF queue' CONFIG_NET_SCH_TBF tristate ' GRED queue' CONFIG_NET_SCH_GRED
Documentation/Configure.help renamed from Documentation/Configure.help.orig +18 -0
@@ -9433,6 +9433,24 @@ whenever you want). If you want to compile it as a module, say M here and read <file:Documentation/modules.txt>. +ESFQ queue +CONFIG_NET_SCH_ESFQ + Say Y here if you want to use the Stochastic Fairness Queueing (SFQ) + packet scheduling algorithm for some of your network devices or as a + leaf discipline for the CBQ scheduling algorithm (see the top of + <file:net/sched/sch_esfq.c> for details and references about the SFQ + algorithm). + + This is an enchanced SFQ version which allows you to control the + hardcoded values in the SFQ scheduler: queue depth, hash table size, + queues limit. Also adds control to the hash function used to identify + packet flows. Hash by src or dst ip and original sfq hash. + + This code is also available as a module called sch_esfq.o ( = code + which can be inserted in and removed from the running kernel + whenever you want). If you want to compile it as a module, say M + here and read <file:Documentation/modules.txt>. + TEQL queue CONFIG_NET_SCH_TEQL Say Y here if you want to use the True Link Equalizer (TLE) packet
net/sched/sch_esfq.c new file +591 -0
591 lines not on this page · raw message