2008-02-07 11:13:49 +03:00
|
|
|
#ifndef __RES_COUNTER_H__
|
|
|
|
#define __RES_COUNTER_H__
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Resource Counters
|
|
|
|
* Contain common data types and routines for resource accounting
|
|
|
|
*
|
|
|
|
* Copyright 2007 OpenVZ SWsoft Inc
|
|
|
|
*
|
|
|
|
* Author: Pavel Emelianov <xemul@openvz.org>
|
|
|
|
*
|
2009-01-16 00:50:59 +03:00
|
|
|
* See Documentation/cgroups/resource_counter.txt for more
|
2008-04-29 12:00:18 +04:00
|
|
|
* info about what this counter is.
|
2008-02-07 11:13:49 +03:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/cgroup.h>
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The core object. the cgroup that wishes to account for some
|
|
|
|
* resource may include this counter into its structures and use
|
|
|
|
* the helpers described beyond
|
|
|
|
*/
|
|
|
|
|
|
|
|
struct res_counter {
|
|
|
|
/*
|
|
|
|
* the current resource consumption level
|
|
|
|
*/
|
2008-02-07 11:13:57 +03:00
|
|
|
unsigned long long usage;
|
2008-04-29 12:00:17 +04:00
|
|
|
/*
|
|
|
|
* the maximal value of the usage from the counter creation
|
|
|
|
*/
|
|
|
|
unsigned long long max_usage;
|
2008-02-07 11:13:49 +03:00
|
|
|
/*
|
|
|
|
* the limit that usage cannot exceed
|
|
|
|
*/
|
2008-02-07 11:13:57 +03:00
|
|
|
unsigned long long limit;
|
2009-09-24 02:56:36 +04:00
|
|
|
/*
|
|
|
|
* the limit that usage can be exceed
|
|
|
|
*/
|
|
|
|
unsigned long long soft_limit;
|
2008-02-07 11:13:49 +03:00
|
|
|
/*
|
|
|
|
* the number of unsuccessful attempts to consume the resource
|
|
|
|
*/
|
2008-02-07 11:13:57 +03:00
|
|
|
unsigned long long failcnt;
|
2008-02-07 11:13:49 +03:00
|
|
|
/*
|
|
|
|
* the lock to protect all of the above.
|
|
|
|
* the routines below consider this to be IRQ-safe
|
|
|
|
*/
|
|
|
|
spinlock_t lock;
|
2009-01-08 05:08:05 +03:00
|
|
|
/*
|
|
|
|
* Parent counter, used for hierarchial resource accounting
|
|
|
|
*/
|
|
|
|
struct res_counter *parent;
|
2008-02-07 11:13:49 +03:00
|
|
|
};
|
|
|
|
|
2009-06-18 03:27:20 +04:00
|
|
|
#define RESOURCE_MAX (unsigned long long)LLONG_MAX
|
|
|
|
|
2008-04-29 11:59:58 +04:00
|
|
|
/**
|
2008-02-07 11:13:49 +03:00
|
|
|
* Helpers to interact with userspace
|
2008-04-29 11:59:58 +04:00
|
|
|
* res_counter_read_u64() - returns the value of the specified member.
|
2008-02-07 11:13:49 +03:00
|
|
|
* res_counter_read/_write - put/get the specified fields from the
|
|
|
|
* res_counter struct to/from the user
|
|
|
|
*
|
|
|
|
* @counter: the counter in question
|
|
|
|
* @member: the field to work with (see RES_xxx below)
|
|
|
|
* @buf: the buffer to opeate on,...
|
|
|
|
* @nbytes: its size...
|
|
|
|
* @pos: and the offset.
|
|
|
|
*/
|
|
|
|
|
2008-04-29 11:59:58 +04:00
|
|
|
u64 res_counter_read_u64(struct res_counter *counter, int member);
|
|
|
|
|
2008-02-07 11:13:49 +03:00
|
|
|
ssize_t res_counter_read(struct res_counter *counter, int member,
|
2008-02-07 11:13:57 +03:00
|
|
|
const char __user *buf, size_t nbytes, loff_t *pos,
|
|
|
|
int (*read_strategy)(unsigned long long val, char *s));
|
2008-07-25 12:47:04 +04:00
|
|
|
|
|
|
|
typedef int (*write_strategy_fn)(const char *buf, unsigned long long *val);
|
|
|
|
|
|
|
|
int res_counter_memparse_write_strategy(const char *buf,
|
|
|
|
unsigned long long *res);
|
|
|
|
|
|
|
|
int res_counter_write(struct res_counter *counter, int member,
|
|
|
|
const char *buffer, write_strategy_fn write_strategy);
|
2008-02-07 11:13:49 +03:00
|
|
|
|
|
|
|
/*
|
|
|
|
* the field descriptors. one for each member of res_counter
|
|
|
|
*/
|
|
|
|
|
|
|
|
enum {
|
|
|
|
RES_USAGE,
|
2008-04-29 12:00:17 +04:00
|
|
|
RES_MAX_USAGE,
|
2008-02-07 11:13:49 +03:00
|
|
|
RES_LIMIT,
|
|
|
|
RES_FAILCNT,
|
2009-09-24 02:56:36 +04:00
|
|
|
RES_SOFT_LIMIT,
|
2008-02-07 11:13:49 +03:00
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* helpers for accounting
|
|
|
|
*/
|
|
|
|
|
2009-01-08 05:08:05 +03:00
|
|
|
void res_counter_init(struct res_counter *counter, struct res_counter *parent);
|
2008-02-07 11:13:49 +03:00
|
|
|
|
|
|
|
/*
|
|
|
|
* charge - try to consume more resource.
|
|
|
|
*
|
|
|
|
* @counter: the counter
|
|
|
|
* @val: the amount of the resource. each controller defines its own
|
|
|
|
* units, e.g. numbers, bytes, Kbytes, etc
|
|
|
|
*
|
|
|
|
* returns 0 on success and <0 if the counter->usage will exceed the
|
|
|
|
* counter->limit _locked call expects the counter->lock to be taken
|
net: introduce res_counter_charge_nofail() for socket allocations
There is a case in __sk_mem_schedule(), where an allocation
is beyond the maximum, but yet we are allowed to proceed.
It happens under the following condition:
sk->sk_wmem_queued + size >= sk->sk_sndbuf
The network code won't revert the allocation in this case,
meaning that at some point later it'll try to do it. Since
this is never communicated to the underlying res_counter
code, there is an inbalance in res_counter uncharge operation.
I see two ways of fixing this:
1) storing the information about those allocations somewhere
in memcg, and then deducting from that first, before
we start draining the res_counter,
2) providing a slightly different allocation function for
the res_counter, that matches the original behavior of
the network code more closely.
I decided to go for #2 here, believing it to be more elegant,
since #1 would require us to do basically that, but in a more
obscure way.
Signed-off-by: Glauber Costa <glommer@parallels.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.cz>
CC: Tejun Heo <tj@kernel.org>
CC: Li Zefan <lizf@cn.fujitsu.com>
CC: Laurent Chavey <chavey@google.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-20 08:57:16 +04:00
|
|
|
*
|
|
|
|
* charge_nofail works the same, except that it charges the resource
|
|
|
|
* counter unconditionally, and returns < 0 if the after the current
|
|
|
|
* charge we are over limit.
|
2008-02-07 11:13:49 +03:00
|
|
|
*/
|
|
|
|
|
2008-07-25 12:46:55 +04:00
|
|
|
int __must_check res_counter_charge_locked(struct res_counter *counter,
|
2012-04-25 03:11:35 +04:00
|
|
|
unsigned long val, bool force);
|
2008-07-25 12:46:55 +04:00
|
|
|
int __must_check res_counter_charge(struct res_counter *counter,
|
2009-10-02 02:44:11 +04:00
|
|
|
unsigned long val, struct res_counter **limit_fail_at);
|
2012-05-30 02:07:05 +04:00
|
|
|
int res_counter_charge_nofail(struct res_counter *counter,
|
net: introduce res_counter_charge_nofail() for socket allocations
There is a case in __sk_mem_schedule(), where an allocation
is beyond the maximum, but yet we are allowed to proceed.
It happens under the following condition:
sk->sk_wmem_queued + size >= sk->sk_sndbuf
The network code won't revert the allocation in this case,
meaning that at some point later it'll try to do it. Since
this is never communicated to the underlying res_counter
code, there is an inbalance in res_counter uncharge operation.
I see two ways of fixing this:
1) storing the information about those allocations somewhere
in memcg, and then deducting from that first, before
we start draining the res_counter,
2) providing a slightly different allocation function for
the res_counter, that matches the original behavior of
the network code more closely.
I decided to go for #2 here, believing it to be more elegant,
since #1 would require us to do basically that, but in a more
obscure way.
Signed-off-by: Glauber Costa <glommer@parallels.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.cz>
CC: Tejun Heo <tj@kernel.org>
CC: Li Zefan <lizf@cn.fujitsu.com>
CC: Laurent Chavey <chavey@google.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-20 08:57:16 +04:00
|
|
|
unsigned long val, struct res_counter **limit_fail_at);
|
2008-02-07 11:13:49 +03:00
|
|
|
|
|
|
|
/*
|
|
|
|
* uncharge - tell that some portion of the resource is released
|
|
|
|
*
|
|
|
|
* @counter: the counter
|
|
|
|
* @val: the amount of the resource
|
|
|
|
*
|
|
|
|
* these calls check for usage underflow and show a warning on the console
|
|
|
|
* _locked call expects the counter->lock to be taken
|
|
|
|
*/
|
|
|
|
|
|
|
|
void res_counter_uncharge_locked(struct res_counter *counter, unsigned long val);
|
2009-10-02 02:44:11 +04:00
|
|
|
void res_counter_uncharge(struct res_counter *counter, unsigned long val);
|
2008-02-07 11:13:49 +03:00
|
|
|
|
2012-05-30 02:07:03 +04:00
|
|
|
void res_counter_uncharge_until(struct res_counter *counter,
|
|
|
|
struct res_counter *top,
|
|
|
|
unsigned long val);
|
2011-03-24 02:42:21 +03:00
|
|
|
/**
|
|
|
|
* res_counter_margin - calculate chargeable space of a counter
|
|
|
|
* @cnt: the counter
|
|
|
|
*
|
|
|
|
* Returns the difference between the hard limit and the current usage
|
|
|
|
* of resource counter @cnt.
|
|
|
|
*/
|
|
|
|
static inline unsigned long long res_counter_margin(struct res_counter *cnt)
|
2009-09-24 02:56:36 +04:00
|
|
|
{
|
2011-03-24 02:42:21 +03:00
|
|
|
unsigned long long margin;
|
|
|
|
unsigned long flags;
|
2009-09-24 02:56:36 +04:00
|
|
|
|
2011-03-24 02:42:21 +03:00
|
|
|
spin_lock_irqsave(&cnt->lock, flags);
|
2012-01-20 08:57:15 +04:00
|
|
|
if (cnt->limit > cnt->usage)
|
|
|
|
margin = cnt->limit - cnt->usage;
|
|
|
|
else
|
|
|
|
margin = 0;
|
2011-03-24 02:42:21 +03:00
|
|
|
spin_unlock_irqrestore(&cnt->lock, flags);
|
|
|
|
return margin;
|
2009-09-24 02:56:36 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Get the difference between the usage and the soft limit
|
|
|
|
* @cnt: The counter
|
|
|
|
*
|
|
|
|
* Returns 0 if usage is less than or equal to soft limit
|
|
|
|
* The difference between usage and soft limit, otherwise.
|
|
|
|
*/
|
|
|
|
static inline unsigned long long
|
|
|
|
res_counter_soft_limit_excess(struct res_counter *cnt)
|
|
|
|
{
|
|
|
|
unsigned long long excess;
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&cnt->lock, flags);
|
|
|
|
if (cnt->usage <= cnt->soft_limit)
|
|
|
|
excess = 0;
|
|
|
|
else
|
|
|
|
excess = cnt->usage - cnt->soft_limit;
|
|
|
|
spin_unlock_irqrestore(&cnt->lock, flags);
|
|
|
|
return excess;
|
|
|
|
}
|
|
|
|
|
2008-04-29 12:00:17 +04:00
|
|
|
static inline void res_counter_reset_max(struct res_counter *cnt)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&cnt->lock, flags);
|
|
|
|
cnt->max_usage = cnt->usage;
|
|
|
|
spin_unlock_irqrestore(&cnt->lock, flags);
|
|
|
|
}
|
|
|
|
|
2008-04-29 12:00:21 +04:00
|
|
|
static inline void res_counter_reset_failcnt(struct res_counter *cnt)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&cnt->lock, flags);
|
|
|
|
cnt->failcnt = 0;
|
|
|
|
spin_unlock_irqrestore(&cnt->lock, flags);
|
|
|
|
}
|
2008-07-25 12:47:19 +04:00
|
|
|
|
|
|
|
static inline int res_counter_set_limit(struct res_counter *cnt,
|
|
|
|
unsigned long long limit)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
int ret = -EBUSY;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&cnt->lock, flags);
|
2008-09-06 01:00:18 +04:00
|
|
|
if (cnt->usage <= limit) {
|
2008-07-25 12:47:19 +04:00
|
|
|
cnt->limit = limit;
|
|
|
|
ret = 0;
|
|
|
|
}
|
|
|
|
spin_unlock_irqrestore(&cnt->lock, flags);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2009-09-24 02:56:36 +04:00
|
|
|
static inline int
|
|
|
|
res_counter_set_soft_limit(struct res_counter *cnt,
|
|
|
|
unsigned long long soft_limit)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&cnt->lock, flags);
|
|
|
|
cnt->soft_limit = soft_limit;
|
|
|
|
spin_unlock_irqrestore(&cnt->lock, flags);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-02-07 11:13:49 +03:00
|
|
|
#endif
|