2 * net/sched/gen_estimator.c Simple rate estimator.
4 * This program is free software; you can redistribute it and/or
5 * modify it under the terms of the GNU General Public License
6 * as published by the Free Software Foundation; either version
7 * 2 of the License, or (at your option) any later version.
9 * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
12 * Jamal Hadi Salim - moved it to net/core and reshulfed
13 * names to make it usable in general net subsystem.
16 #include <asm/uaccess.h>
17 #include <asm/system.h>
18 #include <linux/bitops.h>
19 #include <linux/module.h>
20 #include <linux/types.h>
21 #include <linux/kernel.h>
22 #include <linux/jiffies.h>
23 #include <linux/string.h>
25 #include <linux/socket.h>
26 #include <linux/sockios.h>
28 #include <linux/errno.h>
29 #include <linux/interrupt.h>
30 #include <linux/netdevice.h>
31 #include <linux/skbuff.h>
32 #include <linux/rtnetlink.h>
33 #include <linux/init.h>
34 #include <linux/rbtree.h>
36 #include <net/gen_stats.h>
39 This code is NOT intended to be used for statistics collection,
40 its purpose is to provide a base for statistical multiplexing
41 for controlled load service.
42 If you need only statistics, run a user level daemon which
43 periodically reads byte counters.
45 Unfortunately, rate estimation is not a very easy task.
46 F.e. I did not find a simple way to estimate the current peak rate
47 and even failed to formulate the problem 8)8)
49 So I preferred not to built an estimator into the scheduler,
50 but run this task separately.
51 Ideally, it should be kernel thread(s), but for now it runs
52 from timers, which puts apparent top bounds on the number of rated
53 flows, has minimal overhead on small, but is enough
54 to handle controlled load service, sets of aggregates.
56 We measure rate over A=(1<<interval) seconds and evaluate EWMA:
58 avrate = avrate*(1-W) + rate*W
60 where W is chosen as negative power of 2: W = 2^(-ewma_log)
62 The resulting time constant is:
69 * avbps is scaled by 2^5, avpps is scaled by 2^10.
70 * both values are reported as 32 bit unsigned values. bps can
71 overflow for fast links : max speed being 34360Mbit/sec
72 * Minimal interval is HZ/4=250msec (it is the greatest common divisor
73 for HZ=100 and HZ=1024 8)), maximal interval
74 is (HZ*2^EST_MAX_INTERVAL)/4 = 8sec. Shorter intervals
75 are too expensive, longer ones can be implemented
76 at user level painlessly.
79 #define EST_MAX_INTERVAL 5
83 struct list_head list
;
84 struct gnet_stats_basic
*bstats
;
85 struct gnet_stats_rate_est
*rate_est
;
86 spinlock_t
*stats_lock
;
92 struct rcu_head e_rcu
;
96 struct gen_estimator_head
98 struct timer_list timer
;
99 struct list_head list
;
102 static struct gen_estimator_head elist
[EST_MAX_INTERVAL
+1];
104 /* Protects against NULL dereference */
105 static DEFINE_RWLOCK(est_lock
);
107 /* Protects against soft lockup during large deletion */
108 static struct rb_root est_root
= RB_ROOT
;
110 static void est_timer(unsigned long arg
)
113 struct gen_estimator
*e
;
116 list_for_each_entry_rcu(e
, &elist
[idx
].list
, list
) {
122 spin_lock(e
->stats_lock
);
123 read_lock(&est_lock
);
124 if (e
->bstats
== NULL
)
127 nbytes
= e
->bstats
->bytes
;
128 npackets
= e
->bstats
->packets
;
129 brate
= (nbytes
- e
->last_bytes
)<<(7 - idx
);
130 e
->last_bytes
= nbytes
;
131 e
->avbps
+= ((s64
)(brate
- e
->avbps
)) >> e
->ewma_log
;
132 e
->rate_est
->bps
= (e
->avbps
+0xF)>>5;
134 rate
= (npackets
- e
->last_packets
)<<(12 - idx
);
135 e
->last_packets
= npackets
;
136 e
->avpps
+= ((long)rate
- (long)e
->avpps
) >> e
->ewma_log
;
137 e
->rate_est
->pps
= (e
->avpps
+0x1FF)>>10;
139 read_unlock(&est_lock
);
140 spin_unlock(e
->stats_lock
);
143 if (!list_empty(&elist
[idx
].list
))
144 mod_timer(&elist
[idx
].timer
, jiffies
+ ((HZ
/4) << idx
));
148 static void gen_add_node(struct gen_estimator
*est
)
150 struct rb_node
**p
= &est_root
.rb_node
, *parent
= NULL
;
153 struct gen_estimator
*e
;
156 e
= rb_entry(parent
, struct gen_estimator
, node
);
158 if (est
->bstats
> e
->bstats
)
159 p
= &parent
->rb_right
;
161 p
= &parent
->rb_left
;
163 rb_link_node(&est
->node
, parent
, p
);
164 rb_insert_color(&est
->node
, &est_root
);
168 struct gen_estimator
*gen_find_node(const struct gnet_stats_basic
*bstats
,
169 const struct gnet_stats_rate_est
*rate_est
)
171 struct rb_node
*p
= est_root
.rb_node
;
174 struct gen_estimator
*e
;
176 e
= rb_entry(p
, struct gen_estimator
, node
);
178 if (bstats
> e
->bstats
)
180 else if (bstats
< e
->bstats
|| rate_est
!= e
->rate_est
)
189 * gen_new_estimator - create a new rate estimator
190 * @bstats: basic statistics
191 * @rate_est: rate estimator statistics
192 * @stats_lock: statistics lock
193 * @opt: rate estimator configuration TLV
195 * Creates a new rate estimator with &bstats as source and &rate_est
196 * as destination. A new timer with the interval specified in the
197 * configuration TLV is created. Upon each interval, the latest statistics
198 * will be read from &bstats and the estimated rate will be stored in
199 * &rate_est with the statistics lock grabed during this period.
201 * Returns 0 on success or a negative error code.
203 * NOTE: Called under rtnl_mutex
205 int gen_new_estimator(struct gnet_stats_basic
*bstats
,
206 struct gnet_stats_rate_est
*rate_est
,
207 spinlock_t
*stats_lock
,
210 struct gen_estimator
*est
;
211 struct gnet_estimator
*parm
= nla_data(opt
);
214 if (nla_len(opt
) < sizeof(*parm
))
217 if (parm
->interval
< -2 || parm
->interval
> 3)
220 est
= kzalloc(sizeof(*est
), GFP_KERNEL
);
224 idx
= parm
->interval
+ 2;
225 est
->bstats
= bstats
;
226 est
->rate_est
= rate_est
;
227 est
->stats_lock
= stats_lock
;
228 est
->ewma_log
= parm
->ewma_log
;
229 est
->last_bytes
= bstats
->bytes
;
230 est
->avbps
= rate_est
->bps
<<5;
231 est
->last_packets
= bstats
->packets
;
232 est
->avpps
= rate_est
->pps
<<10;
234 if (!elist
[idx
].timer
.function
) {
235 INIT_LIST_HEAD(&elist
[idx
].list
);
236 setup_timer(&elist
[idx
].timer
, est_timer
, idx
);
239 if (list_empty(&elist
[idx
].list
))
240 mod_timer(&elist
[idx
].timer
, jiffies
+ ((HZ
/4) << idx
));
242 list_add_rcu(&est
->list
, &elist
[idx
].list
);
247 EXPORT_SYMBOL(gen_new_estimator
);
249 static void __gen_kill_estimator(struct rcu_head
*head
)
251 struct gen_estimator
*e
= container_of(head
,
252 struct gen_estimator
, e_rcu
);
257 * gen_kill_estimator - remove a rate estimator
258 * @bstats: basic statistics
259 * @rate_est: rate estimator statistics
261 * Removes the rate estimator specified by &bstats and &rate_est.
263 * NOTE: Called under rtnl_mutex
265 void gen_kill_estimator(struct gnet_stats_basic
*bstats
,
266 struct gnet_stats_rate_est
*rate_est
)
268 struct gen_estimator
*e
;
270 while ((e
= gen_find_node(bstats
, rate_est
))) {
271 rb_erase(&e
->node
, &est_root
);
273 write_lock_bh(&est_lock
);
275 write_unlock_bh(&est_lock
);
277 list_del_rcu(&e
->list
);
278 call_rcu(&e
->e_rcu
, __gen_kill_estimator
);
281 EXPORT_SYMBOL(gen_kill_estimator
);
284 * gen_replace_estimator - replace rate estimator configuration
285 * @bstats: basic statistics
286 * @rate_est: rate estimator statistics
287 * @stats_lock: statistics lock
288 * @opt: rate estimator configuration TLV
290 * Replaces the configuration of a rate estimator by calling
291 * gen_kill_estimator() and gen_new_estimator().
293 * Returns 0 on success or a negative error code.
295 int gen_replace_estimator(struct gnet_stats_basic
*bstats
,
296 struct gnet_stats_rate_est
*rate_est
,
297 spinlock_t
*stats_lock
, struct nlattr
*opt
)
299 gen_kill_estimator(bstats
, rate_est
);
300 return gen_new_estimator(bstats
, rate_est
, stats_lock
, opt
);
302 EXPORT_SYMBOL(gen_replace_estimator
);
305 * gen_estimator_active - test if estimator is currently in use
306 * @bstats: basic statistics
307 * @rate_est: rate estimator statistics
309 * Returns true if estimator is active, and false if not.
311 bool gen_estimator_active(const struct gnet_stats_basic
*bstats
,
312 const struct gnet_stats_rate_est
*rate_est
)
316 return gen_find_node(bstats
, rate_est
) != NULL
;
318 EXPORT_SYMBOL(gen_estimator_active
);