iproute2/tc
8!tgraf ac2fc2df5e Use parse_rtattr_nested
(Logical change 1.129)
2005-01-18 22:11:58 +00:00
..
f_fw.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
f_route.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
f_rsvp.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
f_tcindex.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
f_u32.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
m_action.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
m_estimator.c (Logical change 1.3) 2004-04-15 20:56:59 +00:00
m_gact.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
m_ipt.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
m_mirred.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
m_pedit.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
m_pedit.h Give pedit kind size a name 2005-01-18 00:17:27 +00:00
m_police.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
Makefile Import patch iproute2.121 2005-01-17 23:32:51 +00:00
normal.c Import patch iproute2.121 2005-01-17 23:32:11 +00:00
p_icmp.c Import patch iproute2.121 2005-01-17 23:32:51 +00:00
p_ip.c Import patch iproute2.121 2005-01-17 23:32:51 +00:00
p_tcp.c Import patch iproute2.121 2005-01-17 23:32:51 +00:00
p_udp.c Import patch iproute2.121 2005-01-17 23:32:51 +00:00
pareto.c Change name of tablefactor. 2004-08-30 20:54:46 +00:00
paretonormal.c Import patch iproute2.121 2005-01-17 23:32:11 +00:00
q_atm.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
q_cbq.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
q_dsmark.c Use NLMSG_TAIL 2005-01-18 01:24:18 +00:00
q_fifo.c : prevent tc crashes 2004-09-28 18:35:49 +00:00
q_gred.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
q_hfsc.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
q_htb.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
q_ingress.c : prevent tc crashes 2004-09-28 18:35:49 +00:00
q_netem.c Use NLMSG_TAIL 2005-01-18 01:24:18 +00:00
q_prio.c : prevent tc crashes 2004-09-28 18:35:49 +00:00
q_red.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
q_sfq.c : prevent tc crashes 2004-09-28 18:35:49 +00:00
q_tbf.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
README.last (Logical change 1.3) 2004-04-15 20:56:59 +00:00
tc_cbq.c (Logical change 1.3) 2004-04-15 20:56:59 +00:00
tc_cbq.h (Logical change 1.3) 2004-04-15 20:56:59 +00:00
tc_class.c iproute2: gnet_stats 2004-12-07 23:52:52 +00:00
tc_common.h Auto merged 2004-08-23 20:21:21 +00:00
tc_core.c Adds mpu support to HTB (ATM) queueing discipline 2004-07-30 20:26:15 +00:00
tc_core.h (Logical change 1.3) 2004-04-15 20:56:59 +00:00
tc_estimator.c (Logical change 1.3) 2004-04-15 20:56:59 +00:00
tc_filter.c iproute2: gnet_stats 2004-12-07 23:52:52 +00:00
tc_qdisc.c iproute2: gnet_stats 2004-12-07 23:52:52 +00:00
tc_red.c (Logical change 1.3) 2004-04-15 20:56:59 +00:00
tc_red.h (Logical change 1.3) 2004-04-15 20:56:59 +00:00
tc_util.c Use parse_rtattr_nested 2005-01-18 22:11:58 +00:00
tc_util.h iproute2: gnet_stats 2004-12-07 23:52:52 +00:00
tc.c Fix build warnings on x86_64 2004-10-06 23:17:10 +00:00

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.