iproute2/tc
Jamal Hadi Salim 43726b750a tc: don't ignore ok as an action branch
This is what used to happen before:

tc filter add dev tap1 parent ffff: protocol 0xfefe prio 10 \
     u32 match u32 0 0 flowid 1:16 \
     action ife decode allow mark ok

tc -s filter ls dev tap1 parent ffff:
filter protocol [65278] pref 10 u32
filter protocol [65278] pref 10 u32 fh 800: ht divisor 1
filter protocol [65278] pref 10 u32 fh 800::800 order 2048 key ht 800
bkt 0 flowid 1:16
  match 00000000/00000000 at 0
        action order 1: ife decode action pipe
         index 2 ref 1 bind 1 installed 4 sec used 4 sec
         type: 0x0
         Metadata: allow mark
        Action statistics:
        Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
        backlog 0b 0p requeues 0

        action order 2: gact action pass
         random type none pass val 0
         index 1 ref 1 bind 1 installed 4 sec used 4 sec
        Action statistics:
        Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
        backlog 0b 0p requeues 0

Note the extra action added at the end..

Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
2016-05-16 11:13:58 -07:00
..
.gitignore Add ignore files to make using git easier 2006-08-08 12:04:38 -07:00
Makefile tc: introduce IFE action 2016-05-16 11:13:26 -07:00
README.last tc: remove extra whitespace 2015-10-23 15:43:28 -07:00
e_bpf.c {f,m}_bpf: allow updates on program arrays 2015-11-29 11:55:16 -08:00
em_canid.c Ematch used to classify CAN frames according to their identifiers 2012-08-20 13:11:55 -07:00
em_cmp.c tc: code cleanup 2016-03-21 11:48:36 -07:00
em_ipset.c tc: code cleanup 2016-03-21 11:48:36 -07:00
em_meta.c tc: code cleanup 2016-03-21 11:48:36 -07:00
em_nbyte.c tc: code cleanup 2016-03-21 11:48:36 -07:00
em_u32.c tc: code cleanup 2016-03-21 11:48:36 -07:00
emp_ematch.l fix build issues with flex ver 2.5 2010-04-22 15:27:42 -07:00
emp_ematch.y tc: remove extra whitespace 2015-10-23 15:43:28 -07:00
f_basic.c tc: code cleanup 2016-03-21 11:48:36 -07:00
f_bpf.c {f,m}_bpf: allow for sharing maps 2015-11-23 16:10:44 -08:00
f_cgroup.c tc: code cleanup 2016-03-21 11:48:36 -07:00
f_flow.c tc: code cleanup 2016-03-21 11:48:36 -07:00
f_flower.c lib/utils: introduce rt_addr_n2a_rta() 2016-03-27 10:37:35 -07:00
f_fw.c tc: code cleanup 2016-03-21 11:48:36 -07:00
f_route.c tc: code cleanup 2016-03-21 11:48:36 -07:00
f_rsvp.c tc: code cleanup 2016-03-21 11:48:36 -07:00
f_tcindex.c tc: code cleanup 2016-03-21 11:48:36 -07:00
f_u32.c tc: code cleanup 2016-03-21 11:48:36 -07:00
m_action.c scrub out whitespace issues 2016-03-27 10:50:14 -07:00
m_bpf.c tc: code cleanup 2016-03-21 11:48:36 -07:00
m_connmark.c tc: don't ignore ok as an action branch 2016-05-16 11:13:58 -07:00
m_csum.c tc: don't ignore ok as an action branch 2016-05-16 11:13:58 -07:00
m_ematch.c tc: code cleanup 2016-03-21 11:48:36 -07:00
m_ematch.h include needed files 2012-12-23 11:49:06 -08:00
m_estimator.c tc: code cleanup 2016-03-21 11:48:36 -07:00
m_gact.c tc: code cleanup 2016-03-21 11:48:36 -07:00
m_ife.c tc: don't ignore ok as an action branch 2016-05-16 11:13:58 -07:00
m_ipt.c tc: code cleanup 2016-03-21 11:48:36 -07:00
m_mirred.c tc: don't ignore ok as an action branch 2016-05-16 11:13:58 -07:00
m_nat.c tc: don't ignore ok as an action branch 2016-05-16 11:13:58 -07:00
m_pedit.c tc: don't ignore ok as an action branch 2016-05-16 11:13:58 -07:00
m_pedit.h Remove trailing whitespace 2006-12-05 10:10:22 -08:00
m_police.c tc: code cleanup 2016-03-21 11:48:36 -07:00
m_simple.c tc: code cleanup 2016-03-21 11:48:36 -07:00
m_skbedit.c tc: don't ignore ok as an action branch 2016-05-16 11:13:58 -07:00
m_vlan.c tc: don't ignore ok as an action branch 2016-05-16 11:13:58 -07:00
m_xt.c tc: code cleanup 2016-03-21 11:48:36 -07:00
m_xt_old.c tc: code cleanup 2016-03-21 11:48:36 -07:00
p_icmp.c tc: code cleanup 2016-03-21 11:48:36 -07:00
p_ip.c tc/p_ip.c: Minor coding style cleanup 2016-03-27 10:34:22 -07:00
p_tcp.c tc: code cleanup 2016-03-21 11:48:36 -07:00
p_udp.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_atm.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_cbq.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_choke.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_clsact.c tc, clsact: add clsact frontend 2016-01-18 11:41:27 -08:00
q_codel.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_drr.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_dsmark.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_fifo.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_fq.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_fq_codel.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_gred.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_hfsc.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_hhf.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_htb.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_ingress.c tc, ingress: clean up ingress handling a bit 2016-01-18 11:41:27 -08:00
q_mqprio.c ip: make local functions static 2013-02-12 11:38:35 -08:00
q_multiq.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_netem.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_pie.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_prio.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_qfq.c qfq: fix parse_opt dead code 2015-10-27 15:46:20 +09:00
q_red.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_rr.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_sfb.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_sfq.c tc: code cleanup 2016-03-21 11:48:36 -07:00
q_tbf.c tc: code cleanup 2016-03-21 11:48:36 -07:00
static-syms.c Fix build when shared libraries are disabled 2013-03-13 08:29:59 -07:00
tc.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_bpf.c iproute2: tc_bpf.c: fix building with musl libc 2016-04-11 22:09:57 +00:00
tc_bpf.h tc, bpf: further improve error reporting 2016-04-11 21:53:58 +00:00
tc_cbq.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_cbq.h (Logical change 1.3) 2004-04-15 20:56:59 +00:00
tc_class.c include: add linked list implementation from kernel 2016-03-27 10:56:11 -07:00
tc_common.h tc: built-in eBPF exec proxy 2015-04-27 16:39:23 -07:00
tc_core.c scrub out whitespace issues 2016-03-27 10:50:14 -07:00
tc_core.h htb: support 64bit rates 2013-11-22 17:36:18 -08:00
tc_estimator.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_exec.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_filter.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_monitor.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_qdisc.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_red.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_red.h (Logical change 1.3) 2004-04-15 20:56:59 +00:00
tc_stab.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_util.c tc: code cleanup 2016-03-21 11:48:36 -07:00
tc_util.h tc: built-in eBPF exec proxy 2015-04-27 16:39:23 -07:00

README.last

Kernel code and interface.
--------------------------

* Compile time switches

There is only one, but very important, compile time switch.
It is not settable by "make config", but should be selected
manually and after a bit of thinking in <include/net/pkt_sched.h>

PSCHED_CLOCK_SOURCE can take three values:

	PSCHED_GETTIMEOFDAY
	PSCHED_JIFFIES
	PSCHED_CPU


 PSCHED_GETTIMEOFDAY

Default setting is the most conservative PSCHED_GETTIMEOFDAY.
It is very slow both because of weird slowness of do_gettimeofday()
and because it forces code to use unnatural "timeval" format,
where microseconds and seconds fields are separate.
Besides that, it will misbehave, when delays exceed 2 seconds
(f.e. very slow links or classes bounded to small slice of bandwidth)
To resume: as only you will get it working, select correct clock
source and forget about PSCHED_GETTIMEOFDAY forever.


 PSCHED_JIFFIES

Clock is derived from jiffies. On architectures with HZ=100
granularity of this clock is not enough to make reasonable
bindings to real time. However, taking into account Linux
architecture problems, which force us to use artificial
integrated clock in any case, this switch is not so bad
for schduling even on high speed networks, though policing
is not reliable.


 PSCHED_CPU

It is available only for alpha and pentiums with correct
CPU timestamp. It is the fastest way, use it when it is available,
but remember: not all pentiums have this facility, and
a lot of them have clock, broken by APM etc. etc.