• Home
  • History
  • Annotate
Name Date Size #Lines LOC

..--

.gitignoreD22-Nov-202338 65

Android.mkD22-Nov-2023822 2816

MODULE_LICENSE_GPLD22-Nov-20230

MakefileD22-Nov-20233.4 KiB156126

NOTICED22-Nov-202317.6 KiB341281

README.lastD22-Nov-20231.5 KiB4832

em_cmp.cD22-Nov-20234.6 KiB188144

em_meta.cD22-Nov-202312.6 KiB547434

em_nbyte.cD22-Nov-20233.5 KiB144107

em_u32.cD22-Nov-20234 KiB178134

emp_ematch.lD22-Nov-20233.3 KiB146121

emp_ematch.yD22-Nov-20231.2 KiB10288

f_basic.cD22-Nov-20233.4 KiB148115

f_cgroup.cD22-Nov-20232.6 KiB11585

f_flow.cD22-Nov-20238.6 KiB362315

f_fw.cD22-Nov-20233.8 KiB163134

f_route.cD22-Nov-20234.2 KiB169142

f_rsvp.cD22-Nov-202310.1 KiB404359

f_tcindex.cD22-Nov-20234 KiB182155

f_u32.cD22-Nov-202327.9 KiB1,2951,127

m_action.cD22-Nov-202313.4 KiB644510

m_csum.cD22-Nov-20235.2 KiB247197

m_ematch.cD22-Nov-202310.8 KiB571443

m_ematch.hD22-Nov-20232 KiB11285

m_estimator.cD22-Nov-20231.7 KiB6547

m_gact.cD22-Nov-20235.2 KiB255214

m_ipt.cD22-Nov-202312.8 KiB622492

m_mirred.cD22-Nov-20236 KiB297230

m_nat.cD22-Nov-20234 KiB213167

m_pedit.cD22-Nov-202311.8 KiB609486

m_pedit.hD22-Nov-20232 KiB6341

m_police.cD22-Nov-20239 KiB372322

m_skbedit.cD22-Nov-20234.8 KiB209165

m_xt.cD22-Nov-20237.4 KiB359290

m_xt_old.cD22-Nov-20239.5 KiB437342

p_icmp.cD22-Nov-20231.1 KiB6243

p_ip.cD22-Nov-20233.4 KiB160132

p_tcp.cD22-Nov-2023810 3923

p_udp.cD22-Nov-2023810 3923

q_atm.cD22-Nov-20235.5 KiB259227

q_cbq.cD22-Nov-202314.8 KiB581533

q_choke.cD22-Nov-20235.6 KiB228185

q_drr.cD22-Nov-20232.5 KiB12393

q_dsmark.cD22-Nov-20233.7 KiB175149

q_fifo.cD22-Nov-20232.2 KiB10377

q_gred.cD22-Nov-20238.3 KiB329282

q_hfsc.cD22-Nov-20238.6 KiB411339

q_htb.cD22-Nov-20239.3 KiB336287

q_ingress.cD22-Nov-20231.4 KiB6842

q_mqprio.cD22-Nov-20232.9 KiB132107

q_multiq.cD22-Nov-20232 KiB8548

q_netem.cD22-Nov-202314.9 KiB632527

q_prio.cD22-Nov-20232.8 KiB12694

q_qfq.cD22-Nov-20232.5 KiB12392

q_red.cD22-Nov-20235.8 KiB229190

q_rr.cD22-Nov-20232.8 KiB12094

q_sfb.cD22-Nov-20234.9 KiB201166

q_sfq.cD22-Nov-20237.1 KiB278244

q_tbf.cD22-Nov-20237.1 KiB287249

static-syms.cD22-Nov-202393 76

tc.cD22-Nov-20237.4 KiB358296

tc_cbq.cD22-Nov-20231.5 KiB5840

tc_cbq.hD22-Nov-2023283 107

tc_class.cD22-Nov-20238.3 KiB335286

tc_common.hD22-Nov-2023979 2217

tc_core.cD22-Nov-20234.4 KiB212143

tc_core.hD22-Nov-2023894 3525

tc_estimator.cD22-Nov-20231.1 KiB4530

tc_filter.cD22-Nov-20239.3 KiB379324

tc_monitor.cD22-Nov-20232.3 KiB11185

tc_qdisc.cD22-Nov-20239.1 KiB362305

tc_red.cD22-Nov-20231.8 KiB9965

tc_red.hD22-Nov-2023289 96

tc_stab.cD22-Nov-20233.8 KiB161132

tc_util.cD22-Nov-202311.3 KiB554463

tc_util.hD22-Nov-20233.7 KiB10687

README.last

1Kernel code and interface.
2--------------------------
3
4* Compile time switches
5
6There is only one, but very important, compile time switch.
7It is not settable by "make config", but should be selected
8manually and after a bit of thinking in <include/net/pkt_sched.h>
9
10PSCHED_CLOCK_SOURCE can take three values:
11
12	PSCHED_GETTIMEOFDAY
13	PSCHED_JIFFIES
14	PSCHED_CPU
15
16
17 PSCHED_GETTIMEOFDAY
18
19Default setting is the most conservative PSCHED_GETTIMEOFDAY.
20It is very slow both because of weird slowness of do_gettimeofday()
21and because it forces code to use unnatural "timeval" format,
22where microseconds and seconds fields are separate.
23Besides that, it will misbehave, when delays exceed 2 seconds
24(f.e. very slow links or classes bounded to small slice of bandwidth)
25To resume: as only you will get it working, select correct clock
26source and forget about PSCHED_GETTIMEOFDAY forever.
27
28
29 PSCHED_JIFFIES
30
31Clock is derived from jiffies. On architectures with HZ=100
32granularity of this clock is not enough to make reasonable
33bindings to real time. However, taking into account Linux
34architecture problems, which force us to use artificial
35integrated clock in any case, this switch is not so bad
36for schduling even on high speed networks, though policing
37is not reliable.
38
39
40 PSCHED_CPU
41
42It is available only for alpha and pentiums with correct
43CPU timestamp. It is the fastest way, use it when it is available,
44but remember: not all pentiums have this facility, and
45a lot of them have clock, broken by APM etc. etc.
46
47
48