2005-04-17 06:20:36 +08:00
|
|
|
/*
|
|
|
|
* net/sched/gen_estimator.c Simple rate estimator.
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public License
|
|
|
|
* as published by the Free Software Foundation; either version
|
|
|
|
* 2 of the License, or (at your option) any later version.
|
|
|
|
*
|
|
|
|
* Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
|
2016-12-05 01:48:16 +08:00
|
|
|
* Eric Dumazet <edumazet@google.com>
|
2005-04-17 06:20:36 +08:00
|
|
|
*
|
|
|
|
* Changes:
|
|
|
|
* Jamal Hadi Salim - moved it to net/core and reshulfed
|
|
|
|
* names to make it usable in general net subsystem.
|
|
|
|
*/
|
|
|
|
|
2016-12-25 03:46:01 +08:00
|
|
|
#include <linux/uaccess.h>
|
2007-10-19 14:40:25 +08:00
|
|
|
#include <linux/bitops.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/jiffies.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/socket.h>
|
|
|
|
#include <linux/sockios.h>
|
|
|
|
#include <linux/in.h>
|
|
|
|
#include <linux/errno.h>
|
|
|
|
#include <linux/interrupt.h>
|
|
|
|
#include <linux/netdevice.h>
|
|
|
|
#include <linux/skbuff.h>
|
|
|
|
#include <linux/rtnetlink.h>
|
|
|
|
#include <linux/init.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
|
|
|
#include <linux/slab.h>
|
2016-12-05 01:48:16 +08:00
|
|
|
#include <linux/seqlock.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
#include <net/sock.h>
|
|
|
|
#include <net/gen_stats.h>
|
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
/* This code is NOT intended to be used for statistics collection,
|
|
|
|
* its purpose is to provide a base for statistical multiplexing
|
|
|
|
* for controlled load service.
|
|
|
|
* If you need only statistics, run a user level daemon which
|
|
|
|
* periodically reads byte counters.
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
struct net_rate_estimator {
|
2009-08-16 17:36:49 +08:00
|
|
|
struct gnet_stats_basic_packed *bstats;
|
2005-04-17 06:20:36 +08:00
|
|
|
spinlock_t *stats_lock;
|
2016-06-07 00:37:16 +08:00
|
|
|
seqcount_t *running;
|
2016-12-05 01:48:16 +08:00
|
|
|
struct gnet_stats_basic_cpu __percpu *cpu_bstats;
|
|
|
|
u8 ewma_log;
|
|
|
|
u8 intvl_log; /* period : (250ms << intvl_log) */
|
|
|
|
|
|
|
|
seqcount_t seq;
|
2015-07-02 21:57:19 +08:00
|
|
|
u32 last_packets;
|
2005-04-17 06:20:36 +08:00
|
|
|
u64 last_bytes;
|
2016-12-05 01:48:16 +08:00
|
|
|
|
|
|
|
u64 avpps;
|
2009-05-19 10:26:37 +08:00
|
|
|
u64 avbps;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
unsigned long next_jiffies;
|
|
|
|
struct timer_list timer;
|
|
|
|
struct rcu_head rcu;
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
static void est_fetch_counters(struct net_rate_estimator *e,
|
|
|
|
struct gnet_stats_basic_packed *b)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2016-12-05 01:48:16 +08:00
|
|
|
if (e->stats_lock)
|
|
|
|
spin_lock(e->stats_lock);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
__gnet_stats_copy_basic(e->running, b, e->cpu_bstats, e->bstats);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
if (e->stats_lock)
|
|
|
|
spin_unlock(e->stats_lock);
|
2016-12-03 00:11:00 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
static void est_timer(unsigned long arg)
|
2008-11-25 07:48:05 +08:00
|
|
|
{
|
2016-12-05 01:48:16 +08:00
|
|
|
struct net_rate_estimator *est = (struct net_rate_estimator *)arg;
|
|
|
|
struct gnet_stats_basic_packed b;
|
|
|
|
u64 rate, brate;
|
2008-11-25 07:48:05 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
est_fetch_counters(est, &b);
|
|
|
|
brate = (b.bytes - est->last_bytes) << (8 - est->ewma_log);
|
|
|
|
brate -= (est->avbps >> est->ewma_log);
|
2008-11-25 07:48:05 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
rate = (u64)(b.packets - est->last_packets) << (8 - est->ewma_log);
|
|
|
|
rate -= (est->avpps >> est->ewma_log);
|
2008-11-25 07:48:05 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
write_seqcount_begin(&est->seq);
|
|
|
|
est->avbps += brate;
|
|
|
|
est->avpps += rate;
|
|
|
|
write_seqcount_end(&est->seq);
|
2008-11-25 07:48:05 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
est->last_bytes = b.bytes;
|
|
|
|
est->last_packets = b.packets;
|
2008-11-25 07:48:05 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
est->next_jiffies += ((HZ/4) << est->intvl_log);
|
2008-11-25 07:48:05 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
if (unlikely(time_after_eq(jiffies, est->next_jiffies))) {
|
|
|
|
/* Ouch... timer was delayed. */
|
|
|
|
est->next_jiffies = jiffies + 1;
|
2008-11-25 07:48:05 +08:00
|
|
|
}
|
2016-12-05 01:48:16 +08:00
|
|
|
mod_timer(&est->timer, est->next_jiffies);
|
2008-11-25 07:48:05 +08:00
|
|
|
}
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/**
|
|
|
|
* gen_new_estimator - create a new rate estimator
|
|
|
|
* @bstats: basic statistics
|
2016-03-20 05:31:38 +08:00
|
|
|
* @cpu_bstats: bstats per cpu
|
2005-04-17 06:20:36 +08:00
|
|
|
* @rate_est: rate estimator statistics
|
|
|
|
* @stats_lock: statistics lock
|
2016-06-07 00:37:16 +08:00
|
|
|
* @running: qdisc running seqcount
|
2005-04-17 06:20:36 +08:00
|
|
|
* @opt: rate estimator configuration TLV
|
|
|
|
*
|
|
|
|
* Creates a new rate estimator with &bstats as source and &rate_est
|
|
|
|
* as destination. A new timer with the interval specified in the
|
|
|
|
* configuration TLV is created. Upon each interval, the latest statistics
|
|
|
|
* will be read from &bstats and the estimated rate will be stored in
|
2014-09-04 22:44:36 +08:00
|
|
|
* &rate_est with the statistics lock grabbed during this period.
|
2007-02-09 22:24:36 +08:00
|
|
|
*
|
2005-04-17 06:20:36 +08:00
|
|
|
* Returns 0 on success or a negative error code.
|
2007-07-17 09:28:32 +08:00
|
|
|
*
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
2009-08-16 17:36:49 +08:00
|
|
|
int gen_new_estimator(struct gnet_stats_basic_packed *bstats,
|
2014-09-29 02:52:56 +08:00
|
|
|
struct gnet_stats_basic_cpu __percpu *cpu_bstats,
|
2016-12-05 01:48:16 +08:00
|
|
|
struct net_rate_estimator __rcu **rate_est,
|
2007-07-17 09:28:32 +08:00
|
|
|
spinlock_t *stats_lock,
|
2016-06-07 00:37:16 +08:00
|
|
|
seqcount_t *running,
|
2008-01-23 14:11:17 +08:00
|
|
|
struct nlattr *opt)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2008-01-23 14:11:17 +08:00
|
|
|
struct gnet_estimator *parm = nla_data(opt);
|
2016-12-05 01:48:16 +08:00
|
|
|
struct net_rate_estimator *old, *est;
|
|
|
|
struct gnet_stats_basic_packed b;
|
|
|
|
int intvl_log;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-01-23 14:11:17 +08:00
|
|
|
if (nla_len(opt) < sizeof(*parm))
|
2005-04-17 06:20:36 +08:00
|
|
|
return -EINVAL;
|
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
/* allowed timer periods are :
|
|
|
|
* -2 : 250ms, -1 : 500ms, 0 : 1 sec
|
|
|
|
* 1 : 2 sec, 2 : 4 sec, 3 : 8 sec
|
|
|
|
*/
|
2005-04-17 06:20:36 +08:00
|
|
|
if (parm->interval < -2 || parm->interval > 3)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2006-04-08 05:52:59 +08:00
|
|
|
est = kzalloc(sizeof(*est), GFP_KERNEL);
|
2016-12-05 01:48:16 +08:00
|
|
|
if (!est)
|
2005-04-17 06:20:36 +08:00
|
|
|
return -ENOBUFS;
|
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
seqcount_init(&est->seq);
|
|
|
|
intvl_log = parm->interval + 2;
|
2005-04-17 06:20:36 +08:00
|
|
|
est->bstats = bstats;
|
|
|
|
est->stats_lock = stats_lock;
|
2016-06-07 00:37:16 +08:00
|
|
|
est->running = running;
|
2005-04-17 06:20:36 +08:00
|
|
|
est->ewma_log = parm->ewma_log;
|
2016-12-05 01:48:16 +08:00
|
|
|
est->intvl_log = intvl_log;
|
2014-09-29 02:52:56 +08:00
|
|
|
est->cpu_bstats = cpu_bstats;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
est_fetch_counters(est, &b);
|
|
|
|
est->last_bytes = b.bytes;
|
|
|
|
est->last_packets = b.packets;
|
|
|
|
old = rcu_dereference_protected(*rate_est, 1);
|
|
|
|
if (old) {
|
|
|
|
del_timer_sync(&old->timer);
|
|
|
|
est->avbps = old->avbps;
|
|
|
|
est->avpps = old->avpps;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
2007-07-17 09:28:32 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
est->next_jiffies = jiffies + ((HZ/4) << intvl_log);
|
|
|
|
setup_timer(&est->timer, est_timer, (unsigned long)est);
|
|
|
|
mod_timer(&est->timer, est->next_jiffies);
|
2008-11-25 07:48:05 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
rcu_assign_pointer(*rate_est, est);
|
|
|
|
if (old)
|
|
|
|
kfree_rcu(old, rcu);
|
2005-04-17 06:20:36 +08:00
|
|
|
return 0;
|
|
|
|
}
|
2008-11-26 13:14:06 +08:00
|
|
|
EXPORT_SYMBOL(gen_new_estimator);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
/**
|
|
|
|
* gen_kill_estimator - remove a rate estimator
|
2016-12-05 01:48:16 +08:00
|
|
|
* @rate_est: rate estimator
|
2005-04-17 06:20:36 +08:00
|
|
|
*
|
2016-12-05 01:48:16 +08:00
|
|
|
* Removes the rate estimator.
|
2007-07-17 09:28:32 +08:00
|
|
|
*
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
2016-12-05 01:48:16 +08:00
|
|
|
void gen_kill_estimator(struct net_rate_estimator __rcu **rate_est)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2016-12-05 01:48:16 +08:00
|
|
|
struct net_rate_estimator *est;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
est = xchg((__force struct net_rate_estimator **)rate_est, NULL);
|
|
|
|
if (est) {
|
|
|
|
del_timer_sync(&est->timer);
|
|
|
|
kfree_rcu(est, rcu);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
}
|
2008-11-26 13:14:06 +08:00
|
|
|
EXPORT_SYMBOL(gen_kill_estimator);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
/**
|
2008-01-21 18:36:02 +08:00
|
|
|
* gen_replace_estimator - replace rate estimator configuration
|
2005-04-17 06:20:36 +08:00
|
|
|
* @bstats: basic statistics
|
2016-03-20 05:31:38 +08:00
|
|
|
* @cpu_bstats: bstats per cpu
|
2005-04-17 06:20:36 +08:00
|
|
|
* @rate_est: rate estimator statistics
|
|
|
|
* @stats_lock: statistics lock
|
2016-06-07 00:37:16 +08:00
|
|
|
* @running: qdisc running seqcount (might be NULL)
|
2005-04-17 06:20:36 +08:00
|
|
|
* @opt: rate estimator configuration TLV
|
|
|
|
*
|
|
|
|
* Replaces the configuration of a rate estimator by calling
|
|
|
|
* gen_kill_estimator() and gen_new_estimator().
|
2007-02-09 22:24:36 +08:00
|
|
|
*
|
2005-04-17 06:20:36 +08:00
|
|
|
* Returns 0 on success or a negative error code.
|
|
|
|
*/
|
2009-08-16 17:36:49 +08:00
|
|
|
int gen_replace_estimator(struct gnet_stats_basic_packed *bstats,
|
2014-09-29 02:52:56 +08:00
|
|
|
struct gnet_stats_basic_cpu __percpu *cpu_bstats,
|
2016-12-05 01:48:16 +08:00
|
|
|
struct net_rate_estimator __rcu **rate_est,
|
2016-06-07 00:37:16 +08:00
|
|
|
spinlock_t *stats_lock,
|
|
|
|
seqcount_t *running, struct nlattr *opt)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2016-12-05 01:48:16 +08:00
|
|
|
return gen_new_estimator(bstats, cpu_bstats, rate_est,
|
|
|
|
stats_lock, running, opt);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
2008-11-26 13:14:06 +08:00
|
|
|
EXPORT_SYMBOL(gen_replace_estimator);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* gen_estimator_active - test if estimator is currently in use
|
2016-12-05 01:48:16 +08:00
|
|
|
* @rate_est: rate estimator
|
2008-11-26 13:14:06 +08:00
|
|
|
*
|
2008-11-27 07:24:32 +08:00
|
|
|
* Returns true if estimator is active, and false if not.
|
2008-11-26 13:14:06 +08:00
|
|
|
*/
|
2016-12-05 01:48:16 +08:00
|
|
|
bool gen_estimator_active(struct net_rate_estimator __rcu **rate_est)
|
2008-11-26 13:14:06 +08:00
|
|
|
{
|
2016-12-05 01:48:16 +08:00
|
|
|
return !!rcu_access_pointer(*rate_est);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(gen_estimator_active);
|
2010-06-09 07:39:10 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
bool gen_estimator_read(struct net_rate_estimator __rcu **rate_est,
|
|
|
|
struct gnet_stats_rate_est64 *sample)
|
|
|
|
{
|
|
|
|
struct net_rate_estimator *est;
|
|
|
|
unsigned seq;
|
|
|
|
|
|
|
|
rcu_read_lock();
|
|
|
|
est = rcu_dereference(*rate_est);
|
|
|
|
if (!est) {
|
|
|
|
rcu_read_unlock();
|
|
|
|
return false;
|
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
do {
|
|
|
|
seq = read_seqcount_begin(&est->seq);
|
|
|
|
sample->bps = est->avbps >> 8;
|
|
|
|
sample->pps = est->avpps >> 8;
|
|
|
|
} while (read_seqcount_retry(&est->seq, seq));
|
2010-06-09 07:39:10 +08:00
|
|
|
|
2016-12-05 01:48:16 +08:00
|
|
|
rcu_read_unlock();
|
|
|
|
return true;
|
2008-11-26 13:14:06 +08:00
|
|
|
}
|
2016-12-05 01:48:16 +08:00
|
|
|
EXPORT_SYMBOL(gen_estimator_read);
|