Home » Mailing lists » Devel » [PATCH] BC: resource beancounters (v5) (added userpages reclamation)
| [PATCH] BC: resource beancounters (v5) (added userpages reclamation) [message #7157] |
Thu, 05 October 2006 15:27  |
Kirill Korotaev
Messages: 137
Registered: January 2006
|
Senior Member |
|
|
MAJOR CHANGES:
- phys pages limit hit leads to user pages reclamation
- bc can be changed on arbitrary task
-----------------------------------------------------
Core Resource Beancounters (BC) + kernel/user memory control.
BC allows to account and control consumption
of kernel resources used by group of processes.
Draft UBC description on OpenVZ wiki can be found at
http://wiki.openvz.org/UBC_parameters
The full BC patch set allows to control:
- kernel memory. All the kernel objects allocatable
on user demand should be accounted and limited
for DoS protection.
E.g. page tables, task structs, vmas etc.
- virtual memory pages. BCs allow to
limit a container to some amount of memory and
introduces 2-level OOM killer taking into account
container's consumption.
pages shared between containers are correctly
charged as fractions (tunable).
- network buffers. These includes TCP/IP rcv/snd
buffers, dgram snd buffers, unix, netlinks and
other buffers.
- minor resources accounted/limited by number:
tasks, files, flocks, ptys, siginfo, pinned dcache
mem, sockets, iptentries (for containers with
virtualized networking)
As the first step we want to propose for discussion
the most complicated parts of resource management:
kernel memory and virtual memory.
The patch set to be sent provides core for BC and
management of kernel memory only. Virtual memory
management will be sent in a couple of days.
The patches in these series are:
* diff-atomic-dec-and-lock-irqsave.patch:
introduce atomic_dec_and_lock_irqsave()
* diff-bc-kconfig.patch:
Adds kernel/bc/Kconfig file with UBC options and
includes it into arch Kconfigs
* diff-bc-core.patch:
Contains core functionality and interfaces of BC:
find/create beancounter, initialization,
charge/uncharge of resource, core objects' declarations.
* diff-bc-task.patch:
Contains code responsible for setting BC on task,
it's inheriting and setting host context in interrupts.
* diff-bc-syscalls.patch:
Patch adds system calls for BC management:
1. sys_get_bcid - get current BC id
2. sys_set_bcid - changes BC on task
3. sys_set_bclimit - set limits for resources consumtions
4. sys_get_bcstat - returns limits/usages/fails for BC
* diff-bc-kmem-core.patch:
Introduces BC_KMEMSIZE resource which accounts kernel
objects allocated by task's request.
Objects are accounted via struct page and slab objects.
For the latter ones each slab contains a set of pointers
corresponding object is charged to.
Allocation charge rules:
1. Pages - if allocation is performed with __GFP_BC flag - page
is charged to current's exec_bc.
2. Slabs - kmem_cache may be created with SLAB_BC flag - in this
case each allocation is charged. Caches used by kmalloc are
created with SLAB_BC | SLAB_BC_NOCHARGE flags. In this case
only __GFP_BC allocations are charged.
* diff-bc-kmem-charge.patch:
Adds SLAB_BC and __GFP_BC flags in appropriate places
to cause charging/limiting of specified resources.
* diff-bc-vmpages-core:
Accounting of private pages (mappings), done per-vma.
Charged on mmap().
* diff-bc-rsspages-core:
Phys pages accounting. Charging is done on page touch
(page fault) and limit hit leads to reclamation.
Reclamation is done based on beancounter list of pages.
* diff-bc-userpages-charge:
Hooks across the kernel for user pages accounting.
* bcctl.c:
Tool for BC management. Allows changing BC, setting BC limits,
viewing current usages.
Summary of changes from v4 patch set:
* changed set of resources - kmemsize, privvmpages, physpages
* added event hooks for resources (init, limit hit etc)
* added user pages reclamation (bc_try_to_free_pages)
* removed pages sharing accounting - charge to first user
* task now carries only one BC pointer, simplified
* make set_bcid syscall move arbitrary task into BC
* resources are not recharged when task moves
* each vm_area_struct carries a BC pointer
Summary of changes from v3 patch set:
* Added basic user pages accounting (lockedpages/privvmpages)
* spell in Kconfig
* Makefile reworked
* EXPORT_SYMBOL_GPL
* union w/o name in struct page
* bc_task_charge is void now
* adjust minheld/maxheld splitted
Summary of changes from v2 patch set:
* introduced atomic_dec_and_lock_irqsave()
* bc_adjust_held_minmax comment
* added __must_check for bc_*charge* funcs
* use hash_long() instead of own one
* bc/Kconfig is sourced from init/Kconfig now
* introduced bcid_t type with comment from Alan Cox
* check for barrier <= limit in sys_set_bclimit()
* removed (bc == NULL) checks
* replaced memcpy in beancounter_findcrate with assignment
* moved check 'if (mask & BC_ALLOC)' out of the lock
* removed unnecessary memset()
Summary of changes from v1 patch set:
* CONFIG_BEANCOUNTERS is 'n' by default
* fixed Kconfig includes in arches
* removed hierarchical beancounters to simplify first patchset
* removed unused 'private' pointer
* removed unused EXPORTS
* MAXVALUE redeclared as LONG_MAX
* beancounter_findcreate clarification
* renamed UBC -> BC, ub -> bc etc.
* moved BC inheritance into copy_process
* introduced reset_exec_bc() with proposed BUG_ON
* removed task_bc beancounter (not used yet, for numproc)
* fixed syscalls for sparc
* added sys_get_bcstat(): return info that was in /proc
* cond_syscall instead of #ifdefs
Many thanks to Oleg Nesterov, Alan Cox, Matt Helsley and others
for patch review and comments.
Patch set is applicable to 2.6.18-mm3
Thanks,
Kirill
|
|
|
|
| [PATCH 1/10] BC: introduce atomic_dec_and_lock_irqsave() [message #7159 is a reply to message #7157] |
Thu, 05 October 2006 15:46  |
dev
Messages: 1693
Registered: September 2005
Location: Moscow
|
Senior Member |
|
|
Oleg Nesterov noticed to me that the construction like
(used in beancounter patches and free_uid()):
local_irq_save(flags);
if (atomic_dec_and_lock(&refcnt, &lock))
...
is not that good for preemtible kernels, since with preemption
spin_lock() can schedule() to reduce latency. However, it won't schedule
if interrupts are disabled.
So this patch introduces atomic_dec_and_lock_irqsave() as a logical
counterpart to atomic_dec_and_lock().
Signed-Off-By: Pavel Emelianov <xemul@openvz.org>
Signed-Off-By: Kirill Korotaev <dev@openvz.org>
---
include/linux/spinlock.h | 6 ++++++
kernel/user.c | 5 +----
lib/dec_and_lock.c | 19 +++++++++++++++++++
3 files changed, 26 insertions(+), 4 deletions(-)
--- ./include/linux/spinlock.h.dlirq 2006-08-28 10:17:35.000000000 +0400
+++ ./include/linux/spinlock.h 2006-08-28 11:22:37.000000000 +0400
@@ -266,6 +266,12 @@ extern int _atomic_dec_and_lock(atomic_t
#define atomic_dec_and_lock(atomic, lock) \
__cond_lock(lock, _atomic_dec_and_lock(atomic, lock))
+extern int _atomic_dec_and_lock_irqsave(atomic_t *atomic, spinlock_t *lock,
+ unsigned long *flagsp);
+#define atomic_dec_and_lock_irqsave(atomic, lock, flags) \
+ __cond_lock(lock, \
+ _atomic_dec_and_lock_irqsave(atomic, lock, &flags))
+
/**
* spin_can_lock - would spin_trylock() succeed?
* @lock: the spinlock in question.
--- ./kernel/user.c.dlirq 2006-07-10 12:39:20.000000000 +0400
+++ ./kernel/user.c 2006-08-28 11:08:56.000000000 +0400
@@ -108,15 +108,12 @@ void free_uid(struct user_struct *up)
if (!up)
return;
- local_irq_save(flags);
- if (atomic_dec_and_lock(&up->__count, &uidhash_lock)) {
+ if (atomic_dec_and_lock_irqsave(&up->__count, &uidhash_lock, flags)) {
uid_hash_remove(up);
spin_unlock_irqrestore(&uidhash_lock, flags);
key_put(up->uid_keyring);
key_put(up->session_keyring);
kmem_cache_free(uid_cachep, up);
- } else {
- local_irq_restore(flags);
}
}
--- ./lib/dec_and_lock.c.dlirq 2006-04-21 11:59:36.000000000 +0400
+++ ./lib/dec_and_lock.c 2006-08-28 11:22:08.000000000 +0400
@@ -33,3 +33,22 @@ int _atomic_dec_and_lock(atomic_t *atomi
}
EXPORT_SYMBOL(_atomic_dec_and_lock);
+
+/*
+ * the same, but takes the lock with _irqsave
+ */
+int _atomic_dec_and_lock_irqsave(atomic_t *atomic, spinlock_t *lock,
+ unsigned long *flagsp)
+{
+#ifdef CONFIG_SMP
+ if (atomic_add_unless(atomic, -1, 1))
+ return 0;
+#endif
+ spin_lock_irqsave(lock, *flagsp);
+ if (atomic_dec_and_test(atomic))
+ return 1;
+ spin_unlock_irqrestore(lock, *flagsp);
+ return 0;
+}
+
+EXPORT_SYMBOL(_atomic_dec_and_lock_irqsave);
|
|
|
|
|
|
| [PATCH 3/10] BC: beancounters core (API) [message #7161 is a reply to message #7157] |
Thu, 05 October 2006 15:47 |
Kirill Korotaev
Messages: 137
Registered: January 2006
|
Senior Member |
|
|
Core functionality and interfaces of BCs:
find/create beancounter, initialization,
charge/uncharge of resource, core objects' declarations.
Basic structures:
bc_resource_parm - resource description
beancounter - set of resources, id, lock
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
---
include/bc/beancounter.h | 169 +++++++++++++++++++++++++++++++++++++
include/linux/types.h | 16 +++
init/main.c | 3
kernel/Makefile | 1
kernel/bc/Makefile | 12 ++
kernel/bc/beancounter.c | 214 +++++++++++++++++++++++++++++++++++++++++++++++
6 files changed, 415 insertions(+)
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./include/bc/beancounter.h 2006-10-05 12:05:33.000000000 +0400
@@ -0,0 +1,169 @@
+/*
+ * include/bc/beancounter.h
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#ifndef __BEANCOUNTER_H__
+#define __BEANCOUNTER_H__
+
+#define BC_KMEMSIZE 0
+#define BC_PRIVVMPAGES 1
+#define BC_PHYSPAGES 2
+
+#define BC_RESOURCES 3
+
+struct bc_resource_parm {
+ unsigned long barrier;
+ unsigned long limit;
+ unsigned long held;
+ unsigned long minheld;
+ unsigned long maxheld;
+ unsigned long failcnt;
+};
+
+#ifdef __KERNEL__
+
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/init.h>
+#include <asm/atomic.h>
+
+#define BC_MAXVALUE ((unsigned long)LONG_MAX)
+
+enum bc_severity {
+ BC_BARRIER,
+ BC_LIMIT,
+ BC_FORCE,
+};
+
+#ifdef CONFIG_BEANCOUNTERS
+
+struct beancounter {
+ atomic_t bc_refcount;
+ spinlock_t bc_lock;
+ bcid_t bc_id;
+ struct hlist_node bc_hash;
+
+ struct bc_resource_parm bc_parms[BC_RESOURCES];
+
+ struct list_head bc_page_list;
+ spinlock_t bc_page_lock;
+};
+
+struct bc_resource {
+ char *bcr_name;
+
+ int (*bcr_init)(struct beancounter *bc, gfp_t mask);
+ int (*bcr_change)(struct beancounter *bc,
+ unsigned long new_bar, unsigned long new_lim);
+ void (*bcr_barrier_hit)(struct beancounter *bc);
+ int (*bcr_limit_hit)(struct beancounter *bc, unsigned long val,
+ unsigned long flags);
+ void (*bcr_fini)(struct beancounter *bc);
+};
+
+extern struct bc_resource *bc_resources[BC_RESOURCES];
+
+static inline struct beancounter *bc_get(struct beancounter *bc)
+{
+ atomic_inc(&bc->bc_refcount);
+ return bc;
+}
+extern void bc_put(struct beancounter *bc);
+
+#define BC_LOOKUP 0 /* Just lookup in hash
+ */
+#define BC_ALLOC 1 /* Lookup in hash and try to make
+ * new BC if no one found
+ */
+#define BC_ALLOC_ATOMIC 2 /* When BC_ALLOC is set perform
+ * GFP_ATOMIC allocation
+ */
+
+extern struct beancounter *bc_findcreate(bcid_t bcid, int bc_flags);
+
+static inline void bc_adjust_maxheld(struct bc_resource_parm *parm)
+{
+ if (parm->maxheld < parm->held)
+ parm->maxheld = parm->held;
+}
+
+static inline void bc_adjust_minheld(struct bc_resource_parm *parm)
+{
+ if (parm->minheld > parm->held)
+ parm->minheld = parm->held;
+}
+
+static inline void bc_init_resource(struct beancounter *bc, int res_id,
+ unsigned long bar, unsigned long lim)
+{
+ struct bc_resource_parm *parm;
+
+ parm = &bc->bc_parms[res_id];
+
+ parm->barrier = bar;
+ parm->limit = lim;
+ parm->held = 0;
+ parm->minheld = 0;
+ parm->maxheld = 0;
+ parm->failcnt = 0;
+}
+
+int __must_check bc_charge_locked(struct beancounter *bc, int res_id,
+ unsigned long val, int strict, unsigned long flags);
+static inline int __must_check bc_charge(struct beancounter *bc, int res_id,
+ unsigned long val, int strict)
+{
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&bc->bc_lock, flags);
+ ret = bc_charge_locked(bc, res_id, val, strict, flags);
+ spin_unlock_irqrestore(&bc->bc_lock, flags);
+ return ret;
+}
+
+void bc_uncharge_locked(struct beancounter *bc, int res_id, unsigned long val);
+static inline void bc_uncharge(struct beancounter *bc, int res_id,
+ unsigned long val)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&bc->bc_lock, flags);
+ bc_uncharge_locked(bc, res_id, val);
+ spin_unlock_irqrestore(&bc->bc_lock, flags);
+}
+
+void __init bc_register_resource(int res_id, struct bc_resource *br);
+void __init bc_init_early(void);
+#else /* CONFIG_BEANCOUNTERS */
+static inline int __must_check bc_charge_locked(struct beancounter *bc, int res,
+ unsigned long val, int strict, unsigned long flags)
+{
+ return 0;
+}
+
+static inline int __must_check bc_charge(struct beancounter *bc, int res,
+ unsigned long val, int strict)
+{
+ return 0;
+}
+
+static inline void bc_uncharge_locked(struct beancounter *bc, int res,
+ unsigned long val)
+{
+}
+
+static inline void bc_uncharge(struct beancounter *bc, int res,
+ unsigned long val)
+{
+}
+
+static inline void bc_init_early(void)
+{
+}
+#endif /* CONFIG_BEANCOUNTERS */
+#endif /* __KERNEL__ */
+#endif
--- ./include/linux/types.h.bc_core 2006-10-05 11:42:43.000000000 +0400
+++ ./include/linux/types.h 2006-10-05 11:44:32.000000000 +0400
@@ -40,6 +40,21 @@ typedef __kernel_gid32_t gid_t;
typedef __kernel_uid16_t uid16_t;
typedef __kernel_gid16_t gid16_t;
+/*
+ * Type of beancounter id (CONFIG_BEANCOUNTERS)
+ *
+ * The ancient Unix implementations of this kind of resource management and
+ * security are built around setluid() which sets a uid value that cannot
+ * be changed again and is normally used for security purposes. That
+ * happened to be a uid_t and in simple setups at login uid = luid = euid
+ * would be the norm.
+ *
+ * Thus the Linux one happens to be a uid_t. It could be something else but
+ * for the "container per user" model whatever a container is must be able
+ * to hold all possible uid_t values. Alan Cox.
+ */
+typedef uid_t bcid_t;
+
#ifdef CONFIG_UID16
/* This is defined by include/asm-{arch}/posix_types.h */
typedef __kernel_old_uid_t old_uid_t;
@@ -52,6 +67,7 @@ typedef __kernel_old_gid_t old_gid_t;
#else
typedef __kernel_uid_t uid_t;
typedef __kernel_gid_t gid_t;
+typedef __kernel_uid_t bcid_t;
#endif /* __KERNEL__ */
#if defined(__GNUC__) && !defined(__STRICT_ANSI__)
--- ./init/main.c.bc_core 2006-10-05 11:42:43.000000000 +0400
+++ ./init/main.c 2006-10-05 11:44:32.000000000 +0400
@@ -50,6 +50,8 @@
#include <linux/debug_locks.h>
#include <linux/lockdep.h>
+#include <bc/beancounter.h>
+
#include <asm/io.h>
#include <asm/bugs.h>
#include <asm/setup.h>
@@ -480,6 +482,7 @@ asmlinkage void __init start_kernel(void
char * command_line;
extern struct kernel_param __start___param[], __stop___param[];
+ bc_init_early();
smp_setup_processor_id();
/*
--- ./kernel/Makefile.bc_core 2006-10-05 11:42:43.000000000 +0400
+++ ./kernel/Makefile 2006-10-05 11:44:32.000000000 +0400
@@ -12,6 +12,7 @@ obj-y = sched.o fork.o exec_domain.o
obj-$(CONFIG_STACKTRACE) += stacktrace.o
obj-y += time/
+obj-$(CONFIG_BEANCOUNTERS) += bc/
obj-$(CONFIG_DEBUG_MUTEXES) += mutex-debug.o
obj-$(CONFIG_LOCKDEP) += lockdep.o
ifeq ($(CONFIG_PROC_FS),y)
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./kernel/bc/Makefile 2006-10-05 11:44:32.000000000 +0400
@@ -0,0 +1,12 @@
+#
+# kernel/bc/Makefile
+#
+# Copyright (C) 2006 OpenVZ SWsoft Inc.
+#
+obj-y = beancounter.o
+
+obj-y += sys.o
+obj-y += misc.o
+obj-y += kmem.o
+obj-y += vmpages.o
+obj-y += rsspages.o
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./kernel/bc/beancounter.c 2006-10-05 12:02:19.000000000 +0400
@@ -0,0 +1,214 @@
+/*
+ * kernel/bc/beancounter.c
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#include <linux/sched.h>
+#include <linux/list.h>
+#include <linux/hash.h>
+#include <linux/gfp.h>
+#include <linux/slab.h>
+
+#include <bc/beancounter.h>
+#include <bc/task.h>
+
+#define BC_HASH_BITS (8)
+#define BC_HASH_SIZE (1 << BC_HASH_BITS)
+
+struct bc_resource *bc_resources[BC_RESOURCES];
+
+struct beancounter init_bc;
+
+static struct hlist_head bc_hash[BC_HASH_SIZE];
+static spinlock_t bc_hash_lock;
+static kmem_cache_t *bc_cache;
+
+static void init_beancounter_struct(struct beancounter *bc, bcid_t bcid)
+{
+ bc->bc_id = bcid;
+ spin_lock_init(&bc->bc_lock);
+ atomic_set(&bc->bc_refcount, 1);
+}
+
+struct beancounter *bc_findcreate(bcid_t bcid, int bc_flags)
+{
+ unsigned long flags;
+ struct beancounter *bc;
+ struct beancounter *new_bc;
+ struct hlist_head *head;
+ struct hlist_node *ptr;
+ gfp_t mask;
+ int i;
+
+ head = &bc_hash[hash_long(bcid, BC_HASH_BITS)];
+ bc = NULL;
+ new_bc = NULL;
+ mask = ((bc_flags & BC_ALLOC_ATOMIC) ? GFP_ATOMIC : GFP_KERNEL);
+
+retry:
+ spin_lock_irqsave(&bc_hash_lock, flags);
+ hlist_for_each (ptr, head) {
+ bc = hlist_entry(ptr, struct beancounter, bc_hash);
+ if (bc->bc_id == bcid)
+ break;
+ }
+
+ if (bc != NULL) {
+ bc_get(bc);
+ spin_unlock_irqrestore(&bc_hash_lock, flags);
+
+ if (new_bc != NULL)
+ kmem_cache_free(bc_cache, new_bc);
+ return bc;
+ }
+
+ if (new_bc != NULL) {
+ hlist_add_head(&new_bc->bc_hash, head);
+ spin_unlock_irqrestore(&bc_hash_lock, flags);
+ return new_bc;
+ }
+ spin_unlock_irqrestore(&bc_hash_lock, flags);
+
+ if (!(bc_flags & BC_ALLOC))
+ return NULL;
+
+ new_bc = kmem_cache_alloc(bc_cache, mask);
+ if (new_bc == NULL)
+ return NULL;
+
+ init_beancounter_struct(new_bc, bcid);
+ for (i = 0; i < BC_RESOURCES; i++)
+ if (bc_resources[i]->bcr_init(new_bc, mask))
+ goto out_unroll;
+ goto retry;
+
+out_unroll:
+ for (i--; i >= 0; i--)
+ if (bc_resources[i]->bcr_fini)
+ bc_resources[i]->bcr_fini(new_bc);
+ kmem_cache_free(bc_cache, new_bc);
+ return NULL;
+}
+
+void bc_put(struct beancounter *bc)
+{
+ int i;
+ unsigned long flags;
+
+ if (likely(!atomic_dec_and_lock_irqsave(&bc->bc_refcount,
+ &bc_hash_lock, flags)))
+ return;
+
+ hlist_del(&bc->bc_hash);
+ spin_unl
...
|
|
|
|
| [PATCH 4/10] BC: context inheriting and changing [message #7162 is a reply to message #7157] |
Thu, 05 October 2006 15:50 |
Kirill Korotaev
Messages: 137
Registered: January 2006
|
Senior Member |
|
|
Contains code responsible for setting BC on task,
it's inheriting and setting host context in interrupts.
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
---
include/bc/task.h | 51 ++++++++++++++++++++++++++++++++++++
include/linux/sched.h | 5 +++
kernel/bc/misc.c | 70 ++++++++++++++++++++++++++++++++++++++++++++++++++
kernel/fork.c | 5 +++
kernel/irq/handle.c | 6 ++++
kernel/softirq.c | 5 +++
6 files changed, 142 insertions(+)
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./include/bc/task.h 2006-10-05 12:11:44.000000000 +0400
@@ -0,0 +1,51 @@
+/*
+ * include/bc/beancounter.h
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#ifndef __BC_TASK_H__
+#define __BC_TASK_H__
+
+struct beancounter;
+struct task_struct;
+
+#ifdef CONFIG_BEANCOUNTERS
+#define get_exec_bc() (current->exec_bc)
+
+#define set_exec_bc(bc) ({ \
+ struct task_struct *t; \
+ struct beancounter *old; \
+ t = current; \
+ old = t->exec_bc; \
+ t->exec_bc = bc; \
+ old; \
+ })
+
+#define reset_exec_bc(old, expected) do { \
+ struct task_struct *t; \
+ t = current; \
+ BUG_ON(t->exec_bc != expected);\
+ t->exec_bc = old; \
+ } while (0)
+
+extern struct beancounter init_bc;
+
+void copy_beancounter(struct task_struct *tsk, struct task_struct *parent);
+void free_beancounter(struct task_struct *tsk);
+int bc_task_move(struct task_struct *tsk, struct beancounter *bc);
+#else
+static inline void copy_beancounter(struct task_struct *tsk,
+ struct task_struct *parent)
+{
+}
+
+static inline void free_beancounter(struct task_struct *tsk)
+{
+}
+
+#define set_exec_bc(bc) (NULL)
+#define reset_exec_bc(bc, exp) do { } while (0)
+#endif
+#endif
--- ./include/linux/sched.h.bc_task 2006-10-05 11:42:43.000000000 +0400
+++ ./include/linux/sched.h 2006-10-05 12:11:44.000000000 +0400
@@ -77,6 +77,8 @@ struct sched_param {
#include <linux/futex.h>
#include <linux/rtmutex.h>
+#include <bc/task.h>
+
#include <linux/time.h>
#include <linux/param.h>
#include <linux/resource.h>
@@ -1055,6 +1057,9 @@ struct task_struct {
#ifdef CONFIG_TASK_DELAY_ACCT
struct task_delay_info *delays;
#endif
+#ifdef CONFIG_BEANCOUNTERS
+ struct beancounter *exec_bc;
+#endif
};
static inline pid_t process_group(struct task_struct *tsk)
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./kernel/bc/misc.c 2006-10-05 12:11:44.000000000 +0400
@@ -0,0 +1,70 @@
+/*
+ * kernel/bc/misc.c
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#include <linux/sched.h>
+#include <linux/stop_machine.h>
+
+#include <bc/beancounter.h>
+#include <bc/task.h>
+
+void copy_beancounter(struct task_struct *tsk, struct task_struct *parent)
+{
+ tsk->exec_bc = bc_get(parent->exec_bc);
+}
+
+void free_beancounter(struct task_struct *tsk)
+{
+ bc_put(tsk->exec_bc);
+ tsk->exec_bc = NULL;
+}
+
+struct set_bcid_data {
+ struct beancounter *bc;
+ struct mm_struct *mm;
+};
+
+static int do_set_bcid(void *data)
+{
+ struct set_bcid_data *d;
+ struct mm_struct *mm;
+ struct task_struct *g, *p;
+
+ d = (struct set_bcid_data *)data;
+ mm = d->mm;
+
+ do_each_thread (g, p) {
+ if (p->mm == mm) {
+ bc_put(p->exec_bc);
+ p->exec_bc = bc_get(d->bc);
+ }
+ } while_each_thread (g, p);
+
+ bc_put(mm->mm_bc);
+ mm->mm_bc = bc_get(d->bc);
+ return 0;
+}
+
+int bc_task_move(struct task_struct *tsk, struct beancounter *bc)
+{
+ int err;
+ struct set_bcid_data data;
+ struct mm_struct *mm;
+
+ mm = get_task_mm(tsk);
+ if (mm == NULL)
+ return -EINVAL;
+
+ data.bc = bc;
+ data.mm = mm;
+
+ down_write(&mm->mmap_sem);
+ err = stop_machine_run(do_set_bcid, &data, NR_CPUS);
+ up_write(&mm->mmap_sem);
+
+ mmput(mm);
+ return err;
+}
--- ./kernel/fork.c.bc_task 2006-10-05 11:42:43.000000000 +0400
+++ ./kernel/fork.c 2006-10-05 12:11:44.000000000 +0400
@@ -49,6 +49,8 @@
#include <linux/taskstats_kern.h>
#include <linux/random.h>
+#include <bc/task.h>
+
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/uaccess.h>
@@ -105,6 +107,7 @@ static kmem_cache_t *mm_cachep;
void free_task(struct task_struct *tsk)
{
+ free_beancounter(tsk);
free_thread_info(tsk->thread_info);
rt_mutex_debug_task_free(tsk);
free_task_struct(tsk);
@@ -984,6 +987,8 @@ static struct task_struct *copy_process(
if (!p)
goto fork_out;
+ copy_beancounter(p, current);
+
#ifdef CONFIG_TRACE_IRQFLAGS
DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled);
DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled);
--- ./kernel/irq/handle.c.bc_task 2006-10-05 11:42:43.000000000 +0400
+++ ./kernel/irq/handle.c 2006-10-05 12:11:44.000000000 +0400
@@ -16,6 +16,8 @@
#include <linux/interrupt.h>
#include <linux/kernel_stat.h>
+#include <bc/task.h>
+
#include "internals.h"
/**
@@ -172,6 +174,7 @@ fastcall unsigned int __do_IRQ(unsigned
struct irq_desc *desc = irq_desc + irq;
struct irqaction *action;
unsigned int status;
+ struct beancounter *bc;
kstat_this_cpu.irqs[irq]++;
if (CHECK_IRQ_PER_CPU(desc->status)) {
@@ -228,6 +231,8 @@ fastcall unsigned int __do_IRQ(unsigned
* useful for irq hardware that does not mask cleanly in an
* SMP environment.
*/
+
+ bc = set_exec_bc(&init_bc);
for (;;) {
irqreturn_t action_ret;
@@ -242,6 +247,7 @@ fastcall unsigned int __do_IRQ(unsigned
break;
desc->status &= ~IRQ_PENDING;
}
+ reset_exec_bc(bc, &init_bc);
desc->status &= ~IRQ_INPROGRESS;
out:
--- ./kernel/softirq.c.bc_task 2006-10-05 11:42:43.000000000 +0400
+++ ./kernel/softirq.c 2006-10-05 12:11:44.000000000 +0400
@@ -18,6 +18,8 @@
#include <linux/rcupdate.h>
#include <linux/smp.h>
+#include <bc/task.h>
+
#include <asm/irq.h>
/*
- No shared variables, all the data are CPU local.
@@ -209,6 +211,7 @@ asmlinkage void __do_softirq(void)
__u32 pending;
int max_restart = MAX_SOFTIRQ_RESTART;
int cpu;
+ struct beancounter *bc;
pending = local_softirq_pending();
account_system_vtime(current);
@@ -225,6 +228,7 @@ restart:
h = softirq_vec;
+ bc = set_exec_bc(&init_bc);
do {
if (pending & 1) {
h->action(h);
@@ -233,6 +237,7 @@ restart:
h++;
pending >>= 1;
} while (pending);
+ reset_exec_bc(bc, &init_bc);
local_irq_disable();
|
|
|
|
| [PATCH 5/10] BC: user interface (syscalls) [message #7163 is a reply to message #7157] |
Thu, 05 October 2006 15:50 |
Kirill Korotaev
Messages: 137
Registered: January 2006
|
Senior Member |
|
|
Add the following system calls for BC management:
1. sys_get_bcid - get current BC id
2. sys_set_bcid - change BC on task
3. sys_set_bclimit - set limits for resources consumtions
4. sys_get_bcstat - return br_resource_parm on resource
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
---
arch/i386/kernel/syscall_table.S | 4 +
arch/ia64/kernel/entry.S | 4 +
arch/sparc/kernel/entry.S | 2
arch/sparc/kernel/systbls.S | 6 +
arch/sparc64/kernel/entry.S | 2
arch/sparc64/kernel/systbls.S | 10 ++-
include/asm-i386/unistd.h | 6 +
include/asm-ia64/unistd.h | 6 +
include/asm-powerpc/systbl.h | 4 +
include/asm-powerpc/unistd.h | 6 +
include/asm-sparc/unistd.h | 4 +
include/asm-sparc64/unistd.h | 4 +
include/asm-x86_64/unistd.h | 10 ++-
kernel/bc/sys.c | 130 +++++++++++++++++++++++++++++++++++++++
kernel/sys_ni.c | 6 +
15 files changed, 195 insertions(+), 9 deletions(-)
--- ./arch/i386/kernel/syscall_table.S.bc_syscalls 2006-10-05 11:42:39.000000000 +0400
+++ ./arch/i386/kernel/syscall_table.S 2006-10-05 12:09:09.000000000 +0400
@@ -322,3 +322,7 @@ ENTRY(sys_call_table)
.long sys_kevent_get_events /* 320 */
.long sys_kevent_ctl
.long sys_kevent_wait
+ .long sys_get_bcid
+ .long sys_set_bcid
+ .long sys_set_bclimit /* 325 */
+ .long sys_get_bcstat
--- ./arch/ia64/kernel/entry.S.bc_syscalls 2006-10-05 11:42:39.000000000 +0400
+++ ./arch/ia64/kernel/entry.S 2006-10-05 12:09:09.000000000 +0400
@@ -1610,5 +1610,9 @@ sys_call_table:
data8 sys_sync_file_range // 1300
data8 sys_tee
data8 sys_vmsplice
+ data8 sys_get_bcid
+ data8 sys_set_bcid
+ data8 sys_set_bclimit // 1305
+ data8 sys_get_bcstat
.org sys_call_table + 8*NR_syscalls // guard against failures to increase NR_syscalls
--- ./arch/sparc/kernel/entry.S.bc_syscalls 2006-09-20 14:46:17.000000000 +0400
+++ ./arch/sparc/kernel/entry.S 2006-10-05 12:09:09.000000000 +0400
@@ -37,7 +37,7 @@
#define curptr g6
-#define NR_SYSCALLS 300 /* Each OS is different... */
+#define NR_SYSCALLS 304 /* Each OS is different... */
/* These are just handy. */
#define _SV save %sp, -STACKFRAME_SZ, %sp
--- ./arch/sparc/kernel/systbls.S.bc_syscalls 2006-09-20 14:46:17.000000000 +0400
+++ ./arch/sparc/kernel/systbls.S 2006-10-05 12:09:09.000000000 +0400
@@ -78,7 +78,8 @@ sys_call_table:
/*285*/ .long sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, sys_fstatat64
/*290*/ .long sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat
/*295*/ .long sys_fchmodat, sys_faccessat, sys_pselect6, sys_ppoll, sys_unshare
-/*300*/ .long sys_set_robust_list, sys_get_robust_list
+/*300*/ .long sys_set_robust_list, sys_get_robust_list, sys_get_bcid, sys_set_bcid, sys_set_bclimit
+/*305*/ .long sys_get_bcstat
#ifdef CONFIG_SUNOS_EMUL
/* Now the SunOS syscall table. */
@@ -192,4 +193,7 @@ sunos_sys_table:
.long sunos_nosys, sunos_nosys, sunos_nosys
.long sunos_nosys, sunos_nosys, sunos_nosys
+ .long sunos_nosys, sunos_nosys, sunos_nosys,
+ .long sunos_nosys
+
#endif
--- ./arch/sparc64/kernel/entry.S.bc_syscalls 2006-09-20 14:46:17.000000000 +0400
+++ ./arch/sparc64/kernel/entry.S 2006-10-05 12:09:09.000000000 +0400
@@ -25,7 +25,7 @@
#define curptr g6
-#define NR_SYSCALLS 300 /* Each OS is different... */
+#define NR_SYSCALLS 304 /* Each OS is different... */
.text
.align 32
--- ./arch/sparc64/kernel/systbls.S.bc_syscalls 2006-09-20 14:46:17.000000000 +0400
+++ ./arch/sparc64/kernel/systbls.S 2006-10-05 12:09:09.000000000 +0400
@@ -79,7 +79,8 @@ sys_call_table32:
.word sys_mkdirat, sys_mknodat, sys_fchownat, compat_sys_futimesat, compat_sys_fstatat64
/*290*/ .word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat
.word sys_fchmodat, sys_faccessat, compat_sys_pselect6, compat_sys_ppoll, sys_unshare
-/*300*/ .word compat_sys_set_robust_list, compat_sys_get_robust_list
+/*300*/ .word compat_sys_set_robust_list, compat_sys_get_robust_list, sys_nis_syscall, sys_nis_syscall, sys_nis_syscall
+ .word sys_nis_syscall
#endif /* CONFIG_COMPAT */
@@ -149,7 +150,9 @@ sys_call_table:
.word sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, sys_fstatat64
/*290*/ .word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat
.word sys_fchmodat, sys_faccessat, sys_pselect6, sys_ppoll, sys_unshare
-/*300*/ .word sys_set_robust_list, sys_get_robust_list
+/*300*/ .word sys_set_robust_list, sys_get_robust_list, sys_get_bcid, sys_set_bcid, sys_set_bclimit
+ .word sys_get_bcstat
+
#if defined(CONFIG_SUNOS_EMUL) || defined(CONFIG_SOLARIS_EMUL) || \
defined(CONFIG_SOLARIS_EMUL_MODULE)
@@ -263,4 +266,7 @@ sunos_sys_table:
.word sunos_nosys, sunos_nosys, sunos_nosys
.word sunos_nosys, sunos_nosys, sunos_nosys
.word sunos_nosys, sunos_nosys, sunos_nosys
+
+ .word sunos_nosys, sunos_nosys, sunos_nosys
+ .word sunos_nosys
#endif
--- ./include/asm-i386/unistd.h.bc_syscalls 2006-10-05 11:42:42.000000000 +0400
+++ ./include/asm-i386/unistd.h 2006-10-05 12:09:09.000000000 +0400
@@ -328,10 +328,14 @@
#define __NR_kevent_get_events 320
#define __NR_kevent_ctl 321
#define __NR_kevent_wait 322
+#define __NR_get_bcid 323
+#define __NR_set_bcid 324
+#define __NR_set_bclimit 325
+#define __NR_get_bcstat 326
#ifdef __KERNEL__
-#define NR_syscalls 323
+#define NR_syscalls 327
#include <linux/err.h>
/*
--- ./include/asm-ia64/unistd.h.bc_syscalls 2006-10-05 11:42:42.000000000 +0400
+++ ./include/asm-ia64/unistd.h 2006-10-05 12:09:09.000000000 +0400
@@ -291,11 +291,15 @@
#define __NR_sync_file_range 1300
#define __NR_tee 1301
#define __NR_vmsplice 1302
+#define __NR_get_bcid 1303
+#define __NR_set_bcid 1304
+#define __NR_set_bclimit 1305
+#define __NR_get_bcstat 1306
#ifdef __KERNEL__
-#define NR_syscalls 279 /* length of syscall table */
+#define NR_syscalls 283 /* length of syscall table */
#define __ARCH_WANT_SYS_RT_SIGACTION
--- ./include/asm-powerpc/systbl.h.bc_syscalls 2006-09-20 14:46:39.000000000 +0400
+++ ./include/asm-powerpc/systbl.h 2006-10-05 12:09:09.000000000 +0400
@@ -304,3 +304,7 @@ SYSCALL_SPU(fchmodat)
SYSCALL_SPU(faccessat)
COMPAT_SYS_SPU(get_robust_list)
COMPAT_SYS_SPU(set_robust_list)
+SYSCALL(sys_get_bcid)
+SYSCALL(sys_set_bcid)
+SYSCALL(sys_set_bclimit)
+SYSCALL(sys_get_bcstat)
--- ./include/asm-powerpc/unistd.h.bc_syscalls 2006-10-05 11:42:42.000000000 +0400
+++ ./include/asm-powerpc/unistd.h 2006-10-05 12:09:09.000000000 +0400
@@ -323,10 +323,14 @@
#define __NR_faccessat 298
#define __NR_get_robust_list 299
#define __NR_set_robust_list 300
+#define __NR_get_bcid 301
+#define __NR_set_bcid 302
+#define __NR_set_bclimit 303
+#define __NR_get_bcstat 304
#ifdef __KERNEL__
-#define __NR_syscalls 301
+#define __NR_syscalls 305
#define __NR__exit __NR_exit
#define NR_syscalls __NR_syscalls
--- ./include/asm-sparc/unistd.h.bc_syscalls 2006-10-05 11:42:43.000000000 +0400
+++ ./include/asm-sparc/unistd.h 2006-10-05 12:09:09.000000000 +0400
@@ -318,6 +318,10 @@
#define __NR_unshare 299
#define __NR_set_robust_list 300
#define __NR_get_robust_list 301
+#define __NR_get_bcid 302
+#define __NR_set_bcid 303
+#define __NR_set_bclimit 304
+#define __NR_get_bcstat 305
#ifdef __KERNEL__
/* WARNING: You MAY NOT add syscall numbers larger than 301, since
--- ./include/asm-sparc64/unistd.h.bc_syscalls 2006-10-05 11:42:43.000000000 +0400
+++ ./include/asm-sparc64/unistd.h 2006-10-05 12:09:09.000000000 +0400
@@ -320,6 +320,10 @@
#define __NR_unshare 299
#define __NR_set_robust_list 300
#define __NR_get_robust_list 301
+#define __NR_get_bcid 302
+#define __NR_set_bcid 303
+#define __NR_set_bclimit 304
+#define __NR_get_bcstat 305
#ifdef __KERNEL__
/* WARNING: You MAY NOT add syscall numbers larger than 301, since
--- ./include/asm-x86_64/unistd.h.bc_syscalls 2006-10-05 11:42:43.000000000 +0400
+++ ./include/asm-x86_64/unistd.h 2006-10-05 12:09:09.000000000 +0400
@@ -625,8 +625,16 @@ __SYSCALL(__NR_kevent_get_events, sys_ke
__SYSCALL(__NR_kevent_ctl, sys_kevent_ctl)
#define __NR_kevent_wait 282
__SYSCALL(__NR_kevent_wait, sys_kevent_wait)
+#define __NR_get_bcid 283
+__SYSCALL(__NR_get_bcid, sys_get_bcid)
+#define __NR_set_bcid 284
+__SYSCALL(__NR_set_bcid, sys_set_bcid)
+#define __NR_set_bclimit 285
+__SYSCALL(__NR_set_bclimit, sys_set_bclimit)
+#define __NR_get_bcstat 286
+__SYSCALL(__NR_get_bcstat, sys_get_bcstat)
-#define __NR_syscall_max __NR_kevent_wait
+#define __NR_syscall_max __NR_get_bcstat
#ifdef __KERNEL__
#include <linux/err.h>
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./kernel/bc/sys.c 2006-10-05 12:10:31.000000000 +0400
@@ -0,0 +1,130 @@
+/*
+ * kernel/bc/sys.c
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#include <linux/sched.h>
+
+#include <asm/uaccess.h>
+
+#include <bc/beancounter.h>
+#include <bc/task.h>
+
+asmlinkage long sys_get_bcid(void)
+{
+ struct beancounter *bc;
+
+ bc = get_exec_bc();
+ return bc->bc_id;
+}
+
+asmlinkage long sys_set_bcid(bcid_t id, pid_t pid)
+{
+ int error;
+ struct task_struct *tsk;
+ struct beancounter *bc;
+
+ error = -EPERM;
+ if (!capable(CAP_SETUID))
+ goto out;
+
+ error = -ENOMEM;
+ bc = bc_findcreate(id, BC_ALLOC);
+ if (bc == NULL)
+ goto out;
+
+ read_lock(&tasklist_lock);
+ tsk = find_task_by_pid(pid);
+ if (tsk != NULL)
+ get_task_struct(tsk);
+ read_unlock(&tasklist_lock);
+
+ error = -ESRCH;
+ if (tsk == NULL)
+ goto out_putbc;
+
+ error = bc_task_move(tsk, bc);
+
+ put_task_struct(tsk);
+out_putbc:
+ bc_put(bc);
+out:
+ return error;
+}
+
+asmlinkage long sys_set_bclimit(bcid_t id, unsigned long resource,
+ unsigned long __user *limits)
+{
+ int error;
+ struct beancounter *bc;
+ unsigned long new_limits[2];
+
+ e
...
|
|
|
|
| [PATCH 6/10] BC: kernel memory (core) [message #7164 is a reply to message #7157] |
Thu, 05 October 2006 15:52 |
Kirill Korotaev
Messages: 137
Registered: January 2006
|
Senior Member |
|
|
Introduce BC_KMEMSIZE resource which accounts kernel
objects allocated by task's request.
Reference to BC is kept on struct page or slab object.
For slabs each struct slab contains a set of pointers
corresponding objects are charged to.
Allocation charge rules:
1. Pages - if allocation is performed with __GFP_BC flag - page
is charged to current's exec_bc.
2. Slabs - kmem_cache may be created with SLAB_BC flag - in this
case each allocation is charged. Caches used by kmalloc are
created with SLAB_BC | SLAB_BC_NOCHARGE flags. In this case
only __GFP_BC allocations are charged.
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
---
include/bc/kmem.h | 48 ++++++++++++++++++++
include/linux/gfp.h | 8 ++-
include/linux/mm.h | 1
include/linux/mm_types.h | 3 +
include/linux/slab.h | 4 +
include/linux/vmalloc.h | 1
kernel/bc/kmem.c | 112 +++++++++++++++++++++++++++++++++++++++++++++++
mm/mempool.c | 2
mm/page_alloc.c | 11 ++++
mm/slab.c | 112 +++++++++++++++++++++++++++++++++++++++--------
mm/vmalloc.c | 6 ++
11 files changed, 287 insertions(+), 21 deletions(-)
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./include/bc/kmem.h 2006-10-05 12:13:58.000000000 +0400
@@ -0,0 +1,48 @@
+/*
+ * include/bc/kmem.h
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#ifndef __BC_KMEM_H_
+#define __BC_KMEM_H_
+
+/*
+ * BC_KMEMSIZE accounting
+ */
+
+#include <linux/slab.h>
+#include <linux/gfp.h>
+
+struct page;
+struct beancounter;
+
+#ifdef CONFIG_BEANCOUNTERS
+int __must_check bc_page_charge(struct page *page, int order, gfp_t flags);
+void bc_page_uncharge(struct page *page, int order);
+
+int __must_check bc_slab_charge(kmem_cache_t *cachep, void *obj, gfp_t flags);
+void bc_slab_uncharge(kmem_cache_t *cachep, void *obj);
+#else
+static inline int __must_check bc_page_charge(struct page *page, int order,
+ gfp_t flags)
+{
+ return 0;
+}
+
+static inline void bc_page_uncharge(struct page *page, int order)
+{
+}
+
+static inline int __must_check bc_slab_charge(kmem_cache_t *cachep, void *obj,
+ gfp_t flags)
+{
+ return 0;
+}
+
+static inline void bc_slab_uncharge(kmem_cache_t *cachep, void *obj)
+{
+}
+#endif
+#endif /* __BC_SLAB_H_ */
--- ./include/linux/gfp.h.bc_kmem_core 2006-10-05 11:42:43.000000000 +0400
+++ ./include/linux/gfp.h 2006-10-05 12:12:24.000000000 +0400
@@ -46,15 +46,18 @@ struct vm_area_struct;
#define __GFP_NOMEMALLOC ((__force gfp_t)0x10000u) /* Don't use emergency reserves */
#define __GFP_HARDWALL ((__force gfp_t)0x20000u) /* Enforce hardwall cpuset memory allocs */
#define __GFP_THISNODE ((__force gfp_t)0x40000u)/* No fallback, no policies */
+#define __GFP_BC ((__force gfp_t)0x80000u) /* Charge allocation with BC */
+#define __GFP_BC_LIMIT ((__force gfp_t)0x100000u) /* Charge against BC limit */
-#define __GFP_BITS_SHIFT 20 /* Room for 20 __GFP_FOO bits */
+#define __GFP_BITS_SHIFT 21 /* Room for 21 __GFP_FOO bits */
#define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
/* if you forget to add the bitmask here kernel will crash, period */
#define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \
__GFP_COLD|__GFP_NOWARN|__GFP_REPEAT| \
__GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP| \
- __GFP_NOMEMALLOC|__GFP_HARDWALL|__GFP_THISNODE)
+ __GFP_NOMEMALLOC|__GFP_HARDWALL|__GFP_THISNODE| \
+ __GFP_BC|__GFP_BC_LIMIT)
/* This equals 0, but use constants in case they ever change */
#define GFP_NOWAIT (GFP_ATOMIC & ~__GFP_HIGH)
@@ -63,6 +66,7 @@ struct vm_area_struct;
#define GFP_NOIO (__GFP_WAIT)
#define GFP_NOFS (__GFP_WAIT | __GFP_IO)
#define GFP_KERNEL (__GFP_WAIT | __GFP_IO | __GFP_FS)
+#define GFP_KERNEL_BC (__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_BC)
#define GFP_USER (__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL)
#define GFP_HIGHUSER (__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL | \
__GFP_HIGHMEM)
--- ./include/linux/mm.h.bc_kmem_core 2006-10-05 11:42:43.000000000 +0400
+++ ./include/linux/mm.h 2006-10-05 12:12:24.000000000 +0400
@@ -219,6 +219,7 @@ struct vm_operations_struct {
struct mmu_gather;
struct inode;
+#define page_bc(page) ((page)->bc)
#define page_private(page) ((page)->private)
#define set_page_private(page, v) ((page)->private = (v))
--- ./include/linux/mm_types.h.bc_kmem_core 2006-10-05 11:42:43.000000000 +0400
+++ ./include/linux/mm_types.h 2006-10-05 12:12:24.000000000 +0400
@@ -62,6 +62,9 @@ struct page {
void *virtual; /* Kernel virtual address (NULL if
not kmapped, ie. highmem) */
#endif /* WANT_PAGE_VIRTUAL */
+#ifdef CONFIG_BEANCOUNTERS
+ struct beancounter *bc;
+#endif
#ifdef CONFIG_PAGE_OWNER
int order;
unsigned int gfp_mask;
--- ./include/linux/slab.h.bc_kmem_core 2006-10-05 11:42:43.000000000 +0400
+++ ./include/linux/slab.h 2006-10-05 12:12:24.000000000 +0400
@@ -46,6 +46,8 @@ typedef struct kmem_cache kmem_cache_t;
#define SLAB_PANIC 0x00040000UL /* panic if kmem_cache_create() fails */
#define SLAB_DESTROY_BY_RCU 0x00080000UL /* defer freeing pages to RCU */
#define SLAB_MEM_SPREAD 0x00100000UL /* Spread some memory over cpuset */
+#define SLAB_BC 0x00200000UL /* Account with BC */
+#define SLAB_BC_NOCHARGE 0x00400000UL /* Explicit accounting */
/* flags passed to a constructor func */
#define SLAB_CTOR_CONSTRUCTOR 0x001UL /* if not set, then deconstructor */
@@ -305,6 +307,8 @@ extern kmem_cache_t *fs_cachep;
extern kmem_cache_t *sighand_cachep;
extern kmem_cache_t *bio_cachep;
+struct beancounter;
+struct beancounter **kmem_cache_bcp(kmem_cache_t *cachep, void *obj);
#endif /* __KERNEL__ */
#endif /* _LINUX_SLAB_H */
--- ./include/linux/vmalloc.h.bc_kmem_core 2006-10-05 11:42:43.000000000 +0400
+++ ./include/linux/vmalloc.h 2006-10-05 12:12:24.000000000 +0400
@@ -36,6 +36,7 @@ struct vm_struct {
* Highlevel APIs for driver use
*/
extern void *vmalloc(unsigned long size);
+extern void *vmalloc_bc(unsigned long size);
extern void *vmalloc_user(unsigned long size);
extern void *vmalloc_node(unsigned long size, int node);
extern void *vmalloc_exec(unsigned long size);
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./kernel/bc/kmem.c 2006-10-05 12:13:35.000000000 +0400
@@ -0,0 +1,112 @@
+/*
+ * kernel/bc/kmem.c
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#include <linux/sched.h>
+#include <linux/gfp.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+
+#include <bc/beancounter.h>
+#include <bc/kmem.h>
+#include <bc/task.h>
+
+#define BC_KMEMSIZE_BARRIER (64 * 1024)
+#define BC_KMEMSIZE_LIMIT (64 * 1024)
+
+/*
+ * Slab accounting
+ */
+
+int bc_slab_charge(kmem_cache_t *cachep, void *objp, gfp_t flags)
+{
+ unsigned int size;
+ struct beancounter *bc, **slab_bcp;
+
+ bc = get_exec_bc();
+
+ size = kmem_cache_size(cachep);
+ if (bc_charge(bc, BC_KMEMSIZE, size,
+ (flags & __GFP_BC_LIMIT ? BC_LIMIT : BC_BARRIER)))
+ return -ENOMEM;
+
+ slab_bcp = kmem_cache_bcp(cachep, objp);
+ BUG_ON(*slab_bcp != NULL);
+ *slab_bcp = bc_get(bc);
+ return 0;
+}
+
+void bc_slab_uncharge(kmem_cache_t *cachep, void *objp)
+{
+ unsigned int size;
+ struct beancounter *bc, **slab_bcp;
+
+ slab_bcp = kmem_cache_bcp(cachep, objp);
+ if (*slab_bcp == NULL)
+ return;
+
+ bc = *slab_bcp;
+ size = kmem_cache_size(cachep);
+ bc_uncharge(bc, BC_KMEMSIZE, size);
+ bc_put(bc);
+ *slab_bcp = NULL;
+}
+
+/*
+ * Pages accounting
+ */
+
+int bc_page_charge(struct page *page, int order, gfp_t flags)
+{
+ struct beancounter *bc;
+
+ bc = get_exec_bc();
+
+ if (bc_charge(bc, BC_KMEMSIZE, PAGE_SIZE << order,
+ (flags & __GFP_BC_LIMIT ? BC_LIMIT : BC_BARRIER)))
+ return -ENOMEM;
+
+ BUG_ON(page_bc(page) != NULL);
+ page_bc(page) = bc_get(bc);
+ return 0;
+}
+
+void bc_page_uncharge(struct page *page, int order)
+{
+ struct beancounter *bc;
+
+ bc = page_bc(page);
+ if (bc == NULL)
+ return;
+
+ bc_uncharge(bc, BC_KMEMSIZE, PAGE_SIZE << order);
+ bc_put(bc);
+ page_bc(page) = NULL;
+}
+
+/*
+ * Generic resource info
+ */
+
+static int bc_kmem_init(struct beancounter *bc, gfp_t mask)
+{
+ bc_init_resource(bc, BC_KMEMSIZE,
+ BC_KMEMSIZE_BARRIER, BC_KMEMSIZE_LIMIT);
+ return 0;
+}
+
+struct bc_resource bc_kmem_resource = {
+ .bcr_name = "kmemsize",
+ .bcr_init = bc_kmem_init,
+};
+
+static int __init bc_kmem_init_resource(void)
+{
+ bc_register_resource(BC_KMEMSIZE, &bc_kmem_resource);
+ return 0;
+}
+
+__initcall(bc_kmem_init_resource);
--- ./mm/mempool.c.bc_kmem_core 2006-09-20 14:46:41.000000000 +0400
+++ ./mm/mempool.c 2006-10-05 12:12:24.000000000 +0400
@@ -119,6 +119,7 @@ int mempool_resize(mempool_t *pool, int
unsigned long flags;
BUG_ON(new_min_nr <= 0);
+ gfp_mask &= ~__GFP_BC;
spin_lock_irqsave(&pool->lock, flags);
if (new_min_nr <= pool->min_nr) {
@@ -212,6 +213,7 @@ void * mempool_alloc(mempool_t *pool, gf
gfp_mask |= __GFP_NOMEMALLOC; /* don't allocate emergency reserves */
gfp_mask |= __GFP_NORETRY; /* don't loop in __alloc_pages */
gfp_mask |= __GFP_NOWARN; /* failures are OK */
+ gfp_mask &= ~__GFP_BC; /* do not charge */
gfp_temp = gfp_mask & ~(__GFP_WAIT|__GFP_IO);
--- ./mm/page_alloc.c.bc_kmem_core 2006-10-05 11:42:43.000000000 +0400
+++ ./mm/page_alloc.c 2006-10-05 12:12:24.000000000 +0400
@@ -40,6 +40,8 @@
#include <linux/sort.h>
#include <linux/pfn.h>
+#include <bc/kmem.h>
+
#include <asm/tlbflush.h>
#include <asm/div64.h>
#include "internal.h"
@@ -517,6 +519,8 @@ static void __free_pages_ok(struct page
if (reserved)
return;
+ bc_page_uncharge(page, order);
+
kernel_map_pages(page, 1 << order, 0);
local_irq_save(flags);
__count_vm_events(PGFREE, 1 << order);
@@ -800,6 +804,8 @@ static
...
|
|
|
|
| [PATCH 7/10] BC: kernel memory (marks) [message #7165 is a reply to message #7157] |
Thu, 05 October 2006 15:53 |
dev
Messages: 1693
Registered: September 2005
Location: Moscow
|
Senior Member |
|
|
Mark some kmem caches with SLAB_BC and some allocations
with __GFP_BC to cause charging/limiting of appropriate
kernel resources.
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
---
arch/i386/kernel/ldt.c | 4 ++--
arch/i386/mm/init.c | 4 ++--
arch/i386/mm/pgtable.c | 6 ++++--
drivers/char/tty_io.c | 10 +++++-----
fs/file.c | 8 ++++----
fs/locks.c | 2 +-
fs/namespace.c | 3 ++-
fs/select.c | 7 ++++---
include/asm-i386/thread_info.h | 4 ++--
include/asm-ia64/pgalloc.h | 24 +++++++++++++++++-------
include/asm-x86_64/pgalloc.h | 12 ++++++++----
include/asm-x86_64/thread_info.h | 5 +++--
ipc/msgutil.c | 4 ++--
ipc/sem.c | 7 ++++---
ipc/util.c | 8 ++++----
kernel/fork.c | 15 ++++++++-------
kernel/posix-timers.c | 3 ++-
kernel/signal.c | 2 +-
kernel/user.c | 2 +-
mm/rmap.c | 3 ++-
mm/shmem.c | 3 ++-
mm/slab.c | 9 ++++++---
22 files changed, 86 insertions(+), 59 deletions(-)
--- ./arch/i386/kernel/ldt.c.bc_kmem_charge 2006-10-05 11:42:39.000000000 +0400
+++ ./arch/i386/kernel/ldt.c 2006-10-05 12:14:35.000000000 +0400
@@ -39,9 +39,9 @@ static int alloc_ldt(mm_context_t *pc, i
oldsize = pc->size;
mincount = (mincount+511)&(~511);
if (mincount*LDT_ENTRY_SIZE > PAGE_SIZE)
- newldt = vmalloc(mincount*LDT_ENTRY_SIZE);
+ newldt = vmalloc_bc(mincount*LDT_ENTRY_SIZE);
else
- newldt = kmalloc(mincount*LDT_ENTRY_SIZE, GFP_KERNEL);
+ newldt = kmalloc(mincount*LDT_ENTRY_SIZE, GFP_KERNEL_BC);
if (!newldt)
return -ENOMEM;
--- ./arch/i386/mm/init.c.bc_kmem_charge 2006-10-05 11:42:39.000000000 +0400
+++ ./arch/i386/mm/init.c 2006-10-05 12:14:35.000000000 +0400
@@ -709,7 +709,7 @@ void __init pgtable_cache_init(void)
pmd_cache = kmem_cache_create("pmd",
PTRS_PER_PMD*sizeof(pmd_t),
PTRS_PER_PMD*sizeof(pmd_t),
- 0,
+ SLAB_BC,
pmd_ctor,
NULL);
if (!pmd_cache)
@@ -718,7 +718,7 @@ void __init pgtable_cache_init(void)
pgd_cache = kmem_cache_create("pgd",
PTRS_PER_PGD*sizeof(pgd_t),
PTRS_PER_PGD*sizeof(pgd_t),
- 0,
+ SLAB_BC,
pgd_ctor,
PTRS_PER_PMD == 1 ? pgd_dtor : NULL);
if (!pgd_cache)
--- ./arch/i386/mm/pgtable.c.bc_kmem_charge 2006-10-05 11:42:39.000000000 +0400
+++ ./arch/i386/mm/pgtable.c 2006-10-05 12:14:35.000000000 +0400
@@ -186,9 +186,11 @@ struct page *pte_alloc_one(struct mm_str
struct page *pte;
#ifdef CONFIG_HIGHPTE
- pte = alloc_pages(GFP_KERNEL|__GFP_HIGHMEM|__GFP_REPEAT|__GFP_ZERO , 0);
+ pte = alloc_pages(GFP_KERNEL|__GFP_HIGHMEM|__GFP_REPEAT|__GFP_ZERO |
+ __GFP_BC | __GFP_BC_LIMIT, 0);
#else
- pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
+ pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO|
+ __GFP_BC | __GFP_BC_LIMIT, 0);
#endif
return pte;
}
--- ./drivers/char/tty_io.c.bc_kmem_charge 2006-10-05 11:42:40.000000000 +0400
+++ ./drivers/char/tty_io.c 2006-10-05 12:14:35.000000000 +0400
@@ -165,7 +165,7 @@ static void release_mem(struct tty_struc
static struct tty_struct *alloc_tty_struct(void)
{
- return kzalloc(sizeof(struct tty_struct), GFP_KERNEL);
+ return kzalloc(sizeof(struct tty_struct), GFP_KERNEL_BC);
}
static void tty_buffer_free_all(struct tty_struct *);
@@ -1904,7 +1904,7 @@ static int init_dev(struct tty_driver *d
if (!*tp_loc) {
tp = (struct termios *) kmalloc(sizeof(struct termios),
- GFP_KERNEL);
+ GFP_KERNEL_BC);
if (!tp)
goto free_mem_out;
*tp = driver->init_termios;
@@ -1912,7 +1912,7 @@ static int init_dev(struct tty_driver *d
if (!*ltp_loc) {
ltp = (struct termios *) kmalloc(sizeof(struct termios),
- GFP_KERNEL);
+ GFP_KERNEL_BC);
if (!ltp)
goto free_mem_out;
memset(ltp, 0, sizeof(struct termios));
@@ -1937,7 +1937,7 @@ static int init_dev(struct tty_driver *d
if (!*o_tp_loc) {
o_tp = (struct termios *)
- kmalloc(sizeof(struct termios), GFP_KERNEL);
+ kmalloc(sizeof(struct termios), GFP_KERNEL_BC);
if (!o_tp)
goto free_mem_out;
*o_tp = driver->other->init_termios;
@@ -1945,7 +1945,7 @@ static int init_dev(struct tty_driver *d
if (!*o_ltp_loc) {
o_ltp = (struct termios *)
- kmalloc(sizeof(struct termios), GFP_KERNEL);
+ kmalloc(sizeof(struct termios), GFP_KERNEL_BC);
if (!o_ltp)
goto free_mem_out;
memset(o_ltp, 0, sizeof(struct termios));
--- ./fs/file.c.bc_kmem_charge 2006-10-05 11:42:42.000000000 +0400
+++ ./fs/file.c 2006-10-05 12:14:35.000000000 +0400
@@ -44,9 +44,9 @@ struct file ** alloc_fd_array(int num)
int size = num * sizeof(struct file *);
if (size <= PAGE_SIZE)
- new_fds = (struct file **) kmalloc(size, GFP_KERNEL);
+ new_fds = (struct file **) kmalloc(size, GFP_KERNEL_BC);
else
- new_fds = (struct file **) vmalloc(size);
+ new_fds = (struct file **) vmalloc_bc(size);
return new_fds;
}
@@ -213,9 +213,9 @@ fd_set * alloc_fdset(int num)
int size = num / 8;
if (size <= PAGE_SIZE)
- new_fdset = (fd_set *) kmalloc(size, GFP_KERNEL);
+ new_fdset = (fd_set *) kmalloc(size, GFP_KERNEL_BC);
else
- new_fdset = (fd_set *) vmalloc(size);
+ new_fdset = (fd_set *) vmalloc_bc(size);
return new_fdset;
}
--- ./fs/locks.c.bc_kmem_charge 2006-10-05 11:42:42.000000000 +0400
+++ ./fs/locks.c 2006-10-05 12:14:35.000000000 +0400
@@ -2228,7 +2228,7 @@ EXPORT_SYMBOL(lock_may_write);
static int __init filelock_init(void)
{
filelock_cache = kmem_cache_create("file_lock_cache",
- sizeof(struct file_lock), 0, SLAB_PANIC,
+ sizeof(struct file_lock), 0, SLAB_PANIC | SLAB_BC,
init_once, NULL);
return 0;
}
--- ./fs/namespace.c.bc_kmem_charge 2006-10-05 11:42:42.000000000 +0400
+++ ./fs/namespace.c 2006-10-05 12:14:35.000000000 +0400
@@ -1812,7 +1812,8 @@ void __init mnt_init(unsigned long mempa
init_rwsem(&namespace_sem);
mnt_cache = kmem_cache_create("mnt_cache", sizeof(struct vfsmount),
- 0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL, NULL);
+ 0, SLAB_HWCACHE_ALIGN | SLAB_BC | SLAB_PANIC,
+ NULL, NULL);
mount_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC);
--- ./fs/select.c.bc_kmem_charge 2006-10-05 11:42:42.000000000 +0400
+++ ./fs/select.c 2006-10-05 12:14:35.000000000 +0400
@@ -103,7 +103,8 @@ static struct poll_table_entry *poll_get
if (!table || POLL_TABLE_FULL(table)) {
struct poll_table_page *new_table;
- new_table = (struct poll_table_page *) __get_free_page(GFP_KERNEL);
+ new_table = (struct poll_table_page *)
+ __get_free_page(GFP_KERNEL_BC);
if (!new_table) {
p->error = -ENOMEM;
__set_current_state(TASK_RUNNING);
@@ -339,7 +340,7 @@ static int core_sys_select(int n, fd_set
if (size > sizeof(stack_fds) / 6) {
/* Not enough space in on-stack array; must use kmalloc */
ret = -ENOMEM;
- bits = kmalloc(6 * size, GFP_KERNEL);
+ bits = kmalloc(6 * size, GFP_KERNEL_BC);
if (!bits)
goto out_nofds;
}
@@ -687,7 +688,7 @@ int do_sys_poll(struct pollfd __user *uf
if (!stack_pp)
stack_pp = pp = (struct poll_list *)stack_pps;
else {
- pp = kmalloc(size, GFP_KERNEL);
+ pp = kmalloc(size, GFP_KERNEL_BC);
if (!pp)
goto out_fds;
}
--- ./include/asm-i386/thread_info.h.bc_kmem_charge 2006-09-20 14:46:38.000000000 +0400
+++ ./include/asm-i386/thread_info.h 2006-10-05 12:14:35.000000000 +0400
@@ -99,13 +99,13 @@ static inline struct thread_info *curren
({ \
struct thread_info *ret; \
\
- ret = kmalloc(THREAD_SIZE, GFP_KERNEL); \
+ ret = kmalloc(THREAD_SIZE, GFP_KERNEL_BC); \
if (ret) \
memset(ret, 0, THREAD_SIZE); \
ret; \
})
#else
-#define alloc_thread_info(tsk) kmalloc(THREAD_SIZE, GFP_KERNEL)
+#define alloc_thread_info(tsk) kmalloc(THREAD_SIZE, GFP_KERNEL_BC)
#endif
#define free_thread_info(info) kfree(info)
--- ./include/asm-ia64/pgalloc.h.bc_kmem_charge 2006-09-20 14:46:38.000000000 +0400
+++ ./include/asm-ia64/pgalloc.h 2006-10-05 12:14:35.000000000 +0400
@@ -19,6 +19,8 @@
#include <linux/page-flags.h>
#include <linux/threads.h>
+#include <bc/kmem.h>
+
#include <asm/mmu_context.h>
DECLARE_PER_CPU(unsigned long *, __pgtable_quicklist);
@@ -37,7 +39,7 @@ static inline long pgtable_quicklist_tot
return ql_size;
}
-static inline void *pgtable_quicklist_alloc(void)
+static inline void *pgtable_quicklist_alloc(int charge)
{
unsigned long *ret = NULL;
@@ -45,13 +47,20 @@ static inline void *pgtable_quicklist_al
ret = pgtable_quicklist;
if (likely(ret != NULL)) {
+ if (charge && bc_page_charge(virt_to_page(ret),
+ 0, __GFP_BC_LIMIT)) {
+ ret = NULL;
+ goto out;
+ }
pgtable_quicklist = (unsigned long *)(*ret);
ret[0] = 0;
--pgtable_quicklist_size;
+out:
preempt_enable();
} else {
preempt_enable();
- ret = (unsigned long *)__get_free_page(GFP_KERNEL | __GFP_ZERO);
+ ret = (unsigned long *)__get_free_page(GFP_KERNEL |
+ __GFP_ZERO | __GFP_BC | __GFP_BC_LIMIT);
}
return ret;
@@ -69,6 +78,7 @@ static inline void pgtable_quicklist_fre
#endif
preempt_disable();
+ bc_page_uncharge(virt_to_page(pgtable_entry), 0);
*(unsigned long *)pgtable_entry = (unsigned long)pgtable_quicklist;
pgtable_quicklist = (unsigned long *)pgtable_entry;
++pgtable_quicklist_size;
@@ -77,7 +87,7 @@ static inline void pgtable_quicklist_fre
static inline pgd_t *pgd_alloc(struct mm_struct *mm)
{
- return pgtable_quicklist_alloc();
+ return pgtable_quicklist_alloc(1);
}
static inline void pgd_free(pgd_t * pgd)
@@ -94,7 +104,7 @@ pgd_populate(struct mm_struct *mm, pgd_t
st
...
|
|
|
|
| [PATCH 8/10] BC: private pages (mappings) accounting (core) [message #7166 is a reply to message #7157] |
Thu, 05 October 2006 15:54 |
dev
Messages: 1693
Registered: September 2005
Location: Moscow
|
Senior Member |
|
|
Accounting of private pages.
Accounting is performed on mmap per vma. When mm changes
its BC, vmas are not recharged.
On mlock and mprotect vma status may change from BC point of
view, so recharging is done accordingly.
See definition of private pages at
http://wiki.openvz.org/User_pages_accounting
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
---
include/bc/vmpages.h | 90 ++++++++++++++++++++++++++++++++
include/linux/mm.h | 3 +
include/linux/sched.h | 3 +
kernel/bc/vmpages.c | 138 ++++++++++++++++++++++++++++++++++++++++++++++++++
4 files changed, 234 insertions(+)
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./include/bc/vmpages.h 2006-10-05 12:19:59.000000000 +0400
@@ -0,0 +1,90 @@
+/*
+ * include/bc/vmpages.h
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#ifndef __BC_VMPAGES_H_
+#define __BC_VMPAGES_H_
+
+#include <bc/beancounter.h>
+
+struct vm_area_struct;
+struct mm_struct;
+struct file;
+
+#define BC_NOCHARGE 0
+#define BC_UNCHARGE 1
+#define BC_CHARGE 2
+
+#ifdef CONFIG_BEANCOUNTERS
+#define __vma_set_bc(vma, bc) do { (vma)->vma_bc = bc_get(bc); } while (0)
+#define vma_set_bc(vma) __vma_set_bc(vma, (vma)->vm_mm->mm_bc)
+#define vma_copy_bc(vma) __vma_set_bc(vma, (vma)->vma_bc)
+#define vma_release_bc(vma) do { bc_put((vma)->vma_bc); } while (0)
+
+#define mm_init_beancounter(mm) do { \
+ struct beancounter *bc; \
+ bc = get_exec_bc(); \
+ (mm)->mm_bc = bc_get(bc); \
+ } while (0)
+#define mm_free_beancounter(mm) do { bc_put(mm->mm_bc); } while (0)
+
+int __must_check bc_need_memory_recharge(struct vm_area_struct *vma,
+ struct file *new_file, unsigned long new_flags);
+
+int __must_check __bc_memory_charge(struct mm_struct *mm, unsigned long len,
+ int severity);
+int __must_check bc_memory_charge(struct mm_struct *mm, unsigned long len,
+ struct file *file, unsigned long flags, int severity);
+int __must_check bc_vma_charge(struct vm_area_struct *vma);
+
+void __bc_memory_uncharge(struct mm_struct *mm, unsigned long len);
+void bc_memory_uncharge(struct mm_struct *mm, unsigned long len,
+ struct file *file, unsigned long flags);
+void bc_vma_uncharge(struct vm_area_struct *vma);
+
+#define bc_equal(bc1, bc2) (bc1 == bc2)
+#else
+static inline
+int __must_check bc_need_memory_recharge(struct vm_area_struct *vma,
+ struct file *new_file, unsigned long new_flags)
+{
+ return BC_NOCHARGE;
+}
+static inline int __must_check __bc_memory_charge(struct mm_struct *mm,
+ unsigned long len, int severity)
+{
+ return 0;
+}
+static inline int __must_check bc_memory_charge(struct mm_struct *mm,
+ unsigned long len, struct file *file, unsigned long flags,
+ int severity)
+{
+ return 0;
+}
+static inline int __must_check bc_vma_charge(struct vm_area_struct *vma)
+{
+ return 0;
+}
+static inline void __bc_memory_uncharge(struct mm_struct *mm, unsigned long len)
+{
+}
+static inline void bc_memory_uncharge(struct mm_struct *mm, unsigned long len,
+ struct file *file, unsigned long flags)
+{
+}
+static inline void bc_vma_uncharge(struct vm_area_struct *vma)
+{
+}
+
+#define mm_init_beancounter(mm) do { } while (0)
+#define mm_free_beancounter(mm) do { } while (0)
+#define __vma_set_bc(vma, bc) do { } while (0)
+#define vma_set_bc(vma) do { } while (0)
+#define vma_copy_bc(vma) do { } while (0)
+#define vma_release_bc(vma) do { } while (0)
+#define bc_equal(bc1, bc2) 1
+#endif
+#endif
--- ./include/linux/mm.h.bc_vmpages_core 2006-10-05 12:12:24.000000000 +0400
+++ ./include/linux/mm.h 2006-10-05 12:16:03.000000000 +0400
@@ -112,6 +112,9 @@ struct vm_area_struct {
#ifdef CONFIG_NUMA
struct mempolicy *vm_policy; /* NUMA policy for the VMA */
#endif
+#ifdef CONFIG_BEANCOUNTERS
+ struct beancounter *vma_bc;
+#endif
};
/*
--- ./include/linux/sched.h.bc_vmpages_core 2006-10-05 12:11:44.000000000 +0400
+++ ./include/linux/sched.h 2006-10-05 12:16:03.000000000 +0400
@@ -369,6 +369,9 @@ struct mm_struct {
/* aio bits */
rwlock_t ioctx_list_lock;
struct kioctx *ioctx_list;
+#ifdef CONFIG_BEANCOUNTERS
+ struct beancounter *mm_bc;
+#endif
};
struct sighand_struct {
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./kernel/bc/vmpages.c 2006-10-05 12:20:08.000000000 +0400
@@ -0,0 +1,138 @@
+/*
+ * kernel/bc/vmpages.c
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#include <linux/mm.h>
+
+#include <bc/beancounter.h>
+#include <bc/vmpages.h>
+
+#define BC_PRIVVMPAGES_BARRIER BC_MAXVALUE
+#define BC_PRIVVMPAGES_LIMIT BC_MAXVALUE
+
+/*
+ * Core routines
+ */
+
+/*
+ * bc_vma_private checks whether VMA (file, flags) is private
+ * from BC point of view. private VMAs are charged when they are mapped
+ * thus prventing system from resource exhausting when pages from these VMAs
+ * are touched.
+ */
+static inline int bc_vma_private(struct file *file, unsigned long flags)
+{
+ return (flags & VM_LOCKED) ||
+ ((flags & VM_WRITE) && (file == NULL || !(flags & VM_SHARED)));
+}
+
+/*
+ * Accounting is performed in pages (not in Kbytes)
+ */
+static inline int do_memory_charge(struct beancounter *bc,
+ unsigned long len, int severity)
+{
+ return bc_charge(bc, BC_PRIVVMPAGES, len >> PAGE_SHIFT, severity);
+}
+
+static inline void do_memory_uncharge(struct beancounter *bc, unsigned long len)
+{
+ bc_uncharge(bc, BC_PRIVVMPAGES, len >> PAGE_SHIFT);
+}
+
+/*
+ * API calls
+ */
+
+int __bc_memory_charge(struct mm_struct *mm, unsigned long len, int severity)
+{
+ return do_memory_charge(mm->mm_bc, len, severity);
+}
+
+int bc_memory_charge(struct mm_struct *mm, unsigned long len,
+ struct file *file, unsigned long flags, int severity)
+{
+ int ret;
+
+ ret = 0;
+ if (bc_vma_private(file, flags))
+ ret = do_memory_charge(mm->mm_bc, len, severity);
+ return ret;
+}
+
+int bc_vma_charge(struct vm_area_struct *vma)
+{
+ int ret;
+
+ ret = (bc_vma_private(vma->vm_file, vma->vm_flags) ?
+ do_memory_charge(vma->vm_mm->mm_bc,
+ vma->vm_end - vma->vm_start, BC_BARRIER) : 0);
+ if (ret == 0)
+ vma_set_bc(vma);
+ return ret;
+}
+
+void __bc_memory_uncharge(struct mm_struct *mm, unsigned long len)
+{
+ do_memory_uncharge(mm->mm_bc, len);
+}
+
+void bc_memory_uncharge(struct mm_struct *mm, unsigned long len,
+ struct file *file, unsigned long flags)
+{
+ if (bc_vma_private(file, flags))
+ do_memory_uncharge(mm->mm_bc, len);
+}
+
+void bc_vma_uncharge(struct vm_area_struct *vma)
+{
+ if (bc_vma_private(vma->vm_file, vma->vm_flags))
+ do_memory_uncharge(vma->vma_bc, vma->vm_end - vma->vm_start);
+ vma_release_bc(vma);
+}
+
+
+int bc_need_memory_recharge(struct vm_area_struct *vma, struct file *new_file,
+ unsigned long new_flags)
+{
+ if (bc_vma_private(vma->vm_file, vma->vm_flags)) {
+ if (bc_vma_private(new_file, new_flags))
+ return BC_NOCHARGE;
+
+ /* private -> non-private */
+ return BC_UNCHARGE;
+ } else {
+ if (!bc_vma_private(new_file, new_flags))
+ return BC_NOCHARGE;
+
+ /* non-private -> private */
+ return BC_CHARGE;
+ }
+}
+
+/*
+ * Generic resource info
+ */
+
+static int bc_privvm_init(struct beancounter *bc, gfp_t mask)
+{
+ bc_init_resource(bc, BC_PRIVVMPAGES,
+ BC_PRIVVMPAGES_BARRIER, BC_PRIVVMPAGES_LIMIT);
+ return 0;
+}
+
+struct bc_resource bc_privvm_resource = {
+ .bcr_name = "privvmpages",
+ .bcr_init = bc_privvm_init,
+};
+
+static int __init bc_privvm_init_resource(void)
+{
+ bc_register_resource(BC_PRIVVMPAGES, &bc_privvm_resource);
+ return 0;
+}
+
+__initcall(bc_privvm_init_resource);
|
|
|
|
| [PATCH 9/10] BC: physical pages accounting/reclamation (core) [message #7167 is a reply to message #7157] |
Thu, 05 October 2006 15:55 |
Kirill Korotaev
Messages: 137
Registered: January 2006
|
Senior Member |
|
|
Accounting of physical pages.
Accounting is performed on page mapping into address space.
If page is shared, it is charged to its first user.
If physpages limit is hit, BC pages are reclaimed on the
same basis as try_to_free_pages() does.
Full BC implementation allows fractions of pages accounting
as well: http://wiki.openvz.org/RSS_fractions_accounting
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
---
include/bc/rsspages.h | 46 +++++++
include/linux/mm.h | 1
include/linux/mm_types.h | 5
include/linux/page-flags.h | 4
kernel/bc/rsspages.c | 270 +++++++++++++++++++++++++++++++++++++++++++++
mm/vmscan.c | 190 +++++++++++++++++++++++++++++++
6 files changed, 515 insertions(+), 1 deletion(-)
--- ./include/bc/rsspages.h.bc_rsspages 2006-10-05 12:20:42.000000000 +0400
+++ ./include/bc/rsspages.h 2006-10-05 12:35:57.000000000 +0400
@@ -0,0 +1,46 @@
+/*
+ * include/bc/rsspages.h
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#ifndef __BC_RSSPAGES_H_
+#define __BC_RSSPAGES_H_
+
+#include <linux/compiler.h>
+
+struct page;
+struct vm_area_struct;
+struct page_beancounter;
+
+#ifdef CONFIG_BEANCOUNTERS
+int __must_check bc_rsspage_prepare(struct page *p,
+ struct vm_area_struct *vma, struct page_beancounter **ppb);
+void bc_rsspage_charge(struct page_beancounter *pb);
+void bc_rsspage_release(struct page_beancounter *pb);
+void bc_rsspage_uncharge(struct page *p);
+
+unsigned long bc_try_to_free_pages(struct beancounter *bc);
+unsigned long bc_isolate_pages(unsigned long nr_to_scan,
+ struct beancounter *bc, struct list_head *dst,
+ int active, unsigned long *scanned);
+unsigned long bc_nr_physpages(struct beancounter *bc);
+#else
+static inline int __must_check bc_rsspage_prepare(struct page *p,
+ struct vm_area_struct *vma, struct page_beancounter **ppb)
+{
+ return 0;
+}
+
+static inline void bc_rsspage_charge(struct page_beancounter *pb)
+{
+}
+static inline void bc_rsspage_release(struct page_beancounter *pb)
+{
+}
+static inline void bc_rsspage_uncharge(struct page *p)
+{
+}
+#endif
+#endif
--- ./include/linux/mm.h.bc_rsspages 2006-10-05 12:16:03.000000000 +0400
+++ ./include/linux/mm.h 2006-10-05 12:36:42.000000000 +0400
@@ -223,6 +223,7 @@ struct mmu_gather;
struct inode;
#define page_bc(page) ((page)->bc)
+#define page_pb(page) ((page)->pb)
#define page_private(page) ((page)->private)
#define set_page_private(page, v) ((page)->private = (v))
--- ./include/linux/mm_types.h.bc_rsspages 2006-10-05 12:12:24.000000000 +0400
+++ ./include/linux/mm_types.h 2006-10-05 12:36:21.000000000 +0400
@@ -63,7 +63,10 @@ struct page {
not kmapped, ie. highmem) */
#endif /* WANT_PAGE_VIRTUAL */
#ifdef CONFIG_BEANCOUNTERS
- struct beancounter *bc;
+ union {
+ struct beancounter *bc;
+ struct page_beancounter *pb;
+ };
#endif
#ifdef CONFIG_PAGE_OWNER
int order;
--- ./include/linux/page-flags.h.bc_rsspages 2006-10-05 11:42:43.000000000 +0400
+++ ./include/linux/page-flags.h 2006-10-05 12:20:42.000000000 +0400
@@ -92,6 +92,7 @@
#define PG_buddy 19 /* Page is free, on buddy lists */
#define PG_readahead 20 /* Reminder to do readahead */
+#define PG_charged 21 /* Page is already charged to BC */
#if (BITS_PER_LONG > 32)
@@ -253,6 +254,9 @@ static inline void SetPageUptodate(struc
#define SetPageReadahead(page) set_bit(PG_readahead, &(page)->flags)
#define TestClearPageReadahead(page) test_and_clear_bit(PG_readahead, &(page)->flags)
+#define TestSetPageCharged(page) test_and_set_bit(PG_charged, &(page)->flags)
+#define ClearPageCharged(page) clear_bit(PG_charged, &(page)->flags)
+
struct page; /* forward declaration */
int test_clear_page_dirty(struct page *page);
--- ./kernel/bc/rsspages.c.bc_rsspages 2006-10-05 12:20:42.000000000 +0400
+++ ./kernel/bc/rsspages.c 2006-10-05 12:37:40.000000000 +0400
@@ -0,0 +1,270 @@
+/*
+ * kernel/bc/rsspages.c
+ *
+ * Copyright (C) 2006 OpenVZ SWsoft Inc
+ *
+ */
+
+#include <linux/mm_types.h>
+#include <linux/mm.h>
+#include <linux/page-flags.h>
+#include <linux/hardirq.h>
+#include <linux/kernel.h>
+
+#include <bc/beancounter.h>
+#include <bc/vmpages.h>
+#include <bc/rsspages.h>
+
+#include <asm/bitops.h>
+
+#define BC_PHYSPAGES_BARRIER BC_MAXVALUE
+#define BC_PHYSPAGES_LIMIT BC_MAXVALUE
+
+/*
+ * page_beancounter is a tie between page and beancounter page is
+ * charged to. it is used to reclaim pages faster by walking bc's
+ * page list, not zones' ones
+ *
+ * this tie can also be used to implement fractions accounting mechanism
+ * as it is done in OpenVZ kernels
+ */
+struct page_beancounter {
+ struct page *page;
+ struct beancounter *bc;
+ struct list_head list;
+};
+
+/*
+ * API calls
+ */
+
+/*
+ * bc_rsspage_prepare allocates a tie and charges page to vma's beancounter
+ * this must be called in non-atomic context to give a chance for pages
+ * reclaiming. otherwise hitting limits will cause -ENOMEM returned.
+ */
+int bc_rsspage_prepare(struct page *page, struct vm_area_struct *vma,
+ struct page_beancounter **ppb)
+{
+ struct beancounter *bc;
+ struct page_beancounter *pb;
+
+ pb = kmalloc(sizeof(struct page_beancounter), GFP_KERNEL);
+ if (pb == NULL)
+ goto out_nomem;
+
+ bc = vma->vma_bc;
+ if (bc_charge(bc, BC_PHYSPAGES, 1, BC_LIMIT))
+ goto out_charge;
+
+ pb->page = page;
+ pb->bc = bc;
+ *ppb = pb;
+ return 0;
+
+out_charge:
+ kfree(pb);
+out_nomem:
+ return -ENOMEM;
+}
+
+/*
+ * bc_rsspage_release is a rollback call for bc_rsspage_prepare
+ */
+void bc_rsspage_release(struct page_beancounter *pb)
+{
+ bc_uncharge(pb->bc, BC_PHYSPAGES, 1);
+ kfree(pb);
+}
+
+/*
+ * bc_rsspage_charge actually ties page and beancounter together
+ * and marks page as PG_charged. this is done in not-failing path
+ * to be sure the page IS charged
+ */
+void bc_rsspage_charge(struct page_beancounter *pb)
+{
+ struct page *pg;
+ struct beancounter *bc;
+
+ pg = pb->page;
+ if (TestSetPageCharged(pg)) {
+ bc_rsspage_release(pb);
+ return;
+ }
+
+ bc = bc_get(pb->bc);
+ spin_lock(&bc->bc_page_lock);
+ list_add(&pb->list, &bc->bc_page_list);
+ spin_unlock(&bc->bc_page_lock);
+ page_pb(pg) = pb;
+}
+
+/*
+ * bc_rsspage_uncharge is called when pages is get completely unapped
+ * from all address spaces
+ */
+void bc_rsspage_uncharge(struct page *page)
+{
+ struct page_beancounter *pb;
+ struct beancounter *bc;
+
+ pb = page_pb(page);
+ if (pb == NULL)
+ return;
+
+ ClearPageCharged(page);
+ page_pb(page) = NULL;
+
+ bc = pb->bc;
+ spin_lock(&bc->bc_page_lock);
+ list_del(&pb->list);
+ spin_unlock(&bc->bc_page_lock);
+
+ bc_uncharge(bc, BC_PHYSPAGES, 1);
+ bc_put(bc);
+ kfree(pb);
+}
+
+/*
+ * Page reclamation helper
+ *
+ * this function resembles isolate_lru_pages() but is scans through
+ * bc's page list, not zone's active/inactive ones.
+ */
+
+unsigned long bc_isolate_pages(unsigned long nr_to_scan, struct beancounter *bc,
+ struct list_head *dst, int active, unsigned long *scanned)
+{
+ unsigned long nr_taken = 0;
+ struct page *page;
+ struct page_beancounter *pb;
+ unsigned long scan;
+ struct list_head *src;
+ LIST_HEAD(pb_list);
+ struct zone *z;
+
+ spin_lock(&bc->bc_page_lock);
+ src = &bc->bc_page_list;
+ for (scan = 0; scan < nr_to_scan && !list_empty(src); scan++) {
+ struct list_head *target;
+ pb = list_entry(src->prev, struct page_beancounter, list);
+ page = pb->page;
+ z = page_zone(page);
+
+ list_move(&pb->list, &pb_list);
+
+ spin_lock_irq(&z->lru_lock);
+ if (PageLRU(page)) {
+ if ((active && PageActive(page)) ||
+ (!active && !PageActive(page))) {
+ if (likely(get_page_unless_zero(page))) {
+ ClearPageLRU(page);
+ target = dst;
+ nr_taken++;
+ list_move(&page->lru, dst);
+ }
+ }
+ }
+ spin_unlock_irq(&z->lru_lock);
+ }
+
+ list_splice(&pb_list, src);
+ spin_unlock(&bc->bc_page_lock);
+
+ *scanned = scan;
+ return nr_taken;
+}
+
+unsigned long bc_nr_physpages(struct beancounter *bc)
+{
+ return bc->bc_parms[BC_PHYSPAGES].held;
+}
+
+/*
+ * Generic resource info
+ */
+
+static int bc_phys_init(struct beancounter *bc, gfp_t mask)
+{
+ spin_lock_init(&bc->bc_page_lock);
+ INIT_LIST_HEAD(&bc->bc_page_list);
+
+ bc_init_resource(bc, BC_PHYSPAGES,
+ BC_PHYSPAGES_BARRIER, BC_PHYSPAGES_LIMIT);
+ return 0;
+}
+
+static void bc_phys_barrier_hit(struct beancounter *bc)
+{
+ /*
+ * May wake up kswapd here to start asynchronous reclaiming of pages
+ */
+}
+
+static int bc_phys_limit_hit(struct beancounter *bc, unsigned long val,
+ unsigned long flags)
+{
+ int did_some_progress = 0;
+ struct bc_resource_parm *parm;
+
+ spin_unlock_irqrestore(&bc->bc_lock, flags);
+ might_sleep();
+
+ parm = &bc->bc_parms[BC_PHYSPAGES];
+ while (1) {
+ did_some_progress = bc_try_to_free_pages(bc);
+
+ spin_lock_irq(&bc->bc_lock);
+ if (parm->held + val <= parm->limit) {
+ parm->held += val;
+ bc_adjust_maxheld(parm);
+ return 0;
+ }
+
+ if (!did_some_progress) {
+ parm->failcnt++;
+ return -ENOMEM;
+ }
+ spin_unlock_irq(&bc->bc_lock);
+ }
+}
+
+static int bc_phys_change(struct beancounter *bc,
+ unsigned long barrier, unsigned long limit)
+{
+ int did_some_progress;
+ struct bc_resource_parm *parm;
+
+ parm = &bc->bc_parms[BC_PHYSPAGES];
+ if (limit >= parm->held)
+ return 0;
+
+ while (1) {
+ spin_unlock_irq(&bc->bc_lock);
+
+ did_some_progress = bc_try_to_free_pages(bc);
+
+ spin_lock_irq(&bc->bc_lock);
+ if (parm->held < limit)
+ return 0;
+ if (!did_some_progress)
+ return -ENOMEM;
+ }
+}
+
+struct bc_resource bc_phys_resource = {
+ .bcr_name = "physpages",
+ .bcr_init = bc_ph
...
|
|
|
|
| [PATCH 10/10] BC: user memory accounting (hooks) [message #7168 is a reply to message #7157] |
Thu, 05 October 2006 15:56 |
Kirill Korotaev
Messages: 137
Registered: January 2006
|
Senior Member |
|
|
Install BC hooks in appropriate places for user memory accounting.
Signed-off-by: Pavel Emelianov <xemul@openvz.org>
Signed-off-by: Kirill Korotaev <dev@openvz.org>
---
fs/binfmt_elf.c | 5 +--
fs/exec.c | 14 +++++++-
include/asm-alpha/mman.h | 1
include/asm-generic/mman.h | 1
include/asm-mips/mman.h | 1
include/asm-parisc/mman.h | 1
include/asm-xtensa/mman.h | 1
kernel/fork.c | 11 ++++++
mm/fremap.c | 10 ++++++
mm/memory.c | 40 +++++++++++++++++++++++-
mm/migrate.c | 10 ++++++
mm/mlock.c | 16 +++++++++
mm/mmap.c | 74 ++++++++++++++++++++++++++++++++++++++-------
mm/mprotect.c | 18 ++++++++++
mm/mremap.c | 22 ++++++++++---
mm/rmap.c | 3 +
mm/shmem.c | 15 +++++++++
17 files changed, 219 insertions(+), 24 deletions(-)
--- ./fs/binfmt_elf.c.bc_userpages 2006-10-05 11:42:42.000000000 +0400
+++ ./fs/binfmt_elf.c 2006-10-05 12:44:20.000000000 +0400
@@ -360,7 +360,7 @@ static unsigned long load_elf_interp(str
eppnt = elf_phdata;
for (i = 0; i < interp_elf_ex->e_phnum; i++, eppnt++) {
if (eppnt->p_type == PT_LOAD) {
- int elf_type = MAP_PRIVATE | MAP_DENYWRITE;
+ int elf_type = MAP_PRIVATE|MAP_DENYWRITE|MAP_EXECPRIO;
int elf_prot = 0;
unsigned long vaddr = 0;
unsigned long k, map_addr;
@@ -846,7 +846,8 @@ static int load_elf_binary(struct linux_
if (elf_ppnt->p_flags & PF_X)
elf_prot |= PROT_EXEC;
- elf_flags = MAP_PRIVATE | MAP_DENYWRITE | MAP_EXECUTABLE;
+ elf_flags = MAP_PRIVATE | MAP_DENYWRITE | MAP_EXECUTABLE |
+ MAP_EXECPRIO;
vaddr = elf_ppnt->p_vaddr;
if (loc->elf_ex.e_type == ET_EXEC || load_addr_set) {
--- ./fs/exec.c.bc_userpages 2006-10-05 11:42:42.000000000 +0400
+++ ./fs/exec.c 2006-10-05 12:44:20.000000000 +0400
@@ -50,6 +50,8 @@
#include <linux/cn_proc.h>
#include <linux/audit.h>
+#include <bc/rsspages.h>
+
#include <asm/uaccess.h>
#include <asm/mmu_context.h>
@@ -308,27 +310,35 @@ void install_arg_page(struct vm_area_str
struct mm_struct *mm = vma->vm_mm;
pte_t * pte;
spinlock_t *ptl;
+ struct page_beancounter *pc;
if (unlikely(anon_vma_prepare(vma)))
goto out;
+ if (bc_rsspage_prepare(page, vma, &pc))
+ goto out;
+
flush_dcache_page(page);
pte = get_locked_pte(mm, address, &ptl);
if (!pte)
- goto out;
+ goto out_unch;
if (!pte_none(*pte)) {
pte_unmap_unlock(pte, ptl);
- goto out;
+ goto out_unch;
}
inc_mm_counter(mm, anon_rss);
lru_cache_add_active(page);
set_pte_at(mm, address, pte, pte_mkdirty(pte_mkwrite(mk_pte(
page, vma->vm_page_prot))));
page_add_new_anon_rmap(page, vma, address);
+ bc_rsspage_charge(pc);
pte_unmap_unlock(pte, ptl);
/* no need for flush_tlb */
return;
+
+out_unch:
+ bc_rsspage_release(pc);
out:
__free_page(page);
force_sig(SIGKILL, current);
--- ./include/asm-alpha/mman.h.bc_userpages 2006-09-20 14:46:38.000000000 +0400
+++ ./include/asm-alpha/mman.h 2006-10-05 12:44:20.000000000 +0400
@@ -14,6 +14,7 @@
#define MAP_TYPE 0x0f /* Mask for type of mapping (OSF/1 is _wrong_) */
#define MAP_FIXED 0x100 /* Interpret addr exactly */
#define MAP_ANONYMOUS 0x10 /* don't use a file */
+#define MAP_EXECPRIO 0x20 /* charge with BC_LIMIT severity */
/* not used by linux, but here to make sure we don't clash with OSF/1 defines */
#define _MAP_HASSEMAPHORE 0x0200
--- ./include/asm-generic/mman.h.bc_userpages 2006-09-20 14:46:38.000000000 +0400
+++ ./include/asm-generic/mman.h 2006-10-05 12:44:20.000000000 +0400
@@ -19,6 +19,7 @@
#define MAP_TYPE 0x0f /* Mask for type of mapping */
#define MAP_FIXED 0x10 /* Interpret addr exactly */
#define MAP_ANONYMOUS 0x20 /* don't use a file */
+#define MAP_EXECPRIO 0x2000 /* charge with BC_LIMIT severity */
#define MS_ASYNC 1 /* sync memory asynchronously */
#define MS_INVALIDATE 2 /* invalidate the caches */
--- ./include/asm-mips/mman.h.bc_userpages 2006-09-20 14:46:39.000000000 +0400
+++ ./include/asm-mips/mman.h 2006-10-05 12:44:20.000000000 +0400
@@ -46,6 +46,7 @@
#define MAP_LOCKED 0x8000 /* pages are locked */
#define MAP_POPULATE 0x10000 /* populate (prefault) pagetables */
#define MAP_NONBLOCK 0x20000 /* do not block on IO */
+#define MAP_EXECPRIO 0x40000 /* charge with BC_LIMIT severity */
/*
* Flags for msync
--- ./include/asm-parisc/mman.h.bc_userpages 2006-09-20 14:46:39.000000000 +0400
+++ ./include/asm-parisc/mman.h 2006-10-05 12:44:20.000000000 +0400
@@ -22,6 +22,7 @@
#define MAP_GROWSDOWN 0x8000 /* stack-like segment */
#define MAP_POPULATE 0x10000 /* populate (prefault) pagetables */
#define MAP_NONBLOCK 0x20000 /* do not block on IO */
+#define MAP_EXECPRIO 0x40000 /* charge with BC_LIMIT severity */
#define MS_SYNC 1 /* synchronous memory sync */
#define MS_ASYNC 2 /* sync memory asynchronously */
--- ./include/asm-xtensa/mman.h.bc_userpages 2006-09-20 14:46:40.000000000 +0400
+++ ./include/asm-xtensa/mman.h 2006-10-05 12:44:20.000000000 +0400
@@ -53,6 +53,7 @@
#define MAP_LOCKED 0x8000 /* pages are locked */
#define MAP_POPULATE 0x10000 /* populate (prefault) pagetables */
#define MAP_NONBLOCK 0x20000 /* do not block on IO */
+#define MAP_EXECPRIO 0x40000 /* charge with BC_LIMIT severity */
/*
* Flags for msync
--- ./kernel/fork.c.bc_userpages 2006-10-05 12:14:35.000000000 +0400
+++ ./kernel/fork.c 2006-10-05 12:44:20.000000000 +0400
@@ -50,6 +50,7 @@
#include <linux/random.h>
#include <bc/task.h>
+#include <bc/vmpages.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
@@ -252,6 +253,9 @@ static inline int dup_mmap(struct mm_str
tmp->vm_flags &= ~VM_LOCKED;
tmp->vm_mm = mm;
tmp->vm_next = NULL;
+ vma_set_bc(tmp);
+ if (bc_vma_charge(tmp))
+ goto fail_charge;
anon_vma_link(tmp);
file = tmp->vm_file;
if (file) {
@@ -294,6 +298,10 @@ out:
flush_tlb_mm(oldmm);
up_write(&oldmm->mmap_sem);
return retval;
+
+fail_charge:
+ mpol_free(pol);
+ vma_release_bc(tmp);
fail_nomem_policy:
kmem_cache_free(vm_area_cachep, tmp);
fail_nomem:
@@ -344,6 +352,7 @@ static struct mm_struct * mm_init(struct
mm->cached_hole_size = ~0UL;
if (likely(!mm_alloc_pgd(mm))) {
+ mm_init_beancounter(mm);
mm->def_flags = 0;
return mm;
}
@@ -374,6 +383,7 @@ struct mm_struct * mm_alloc(void)
void fastcall __mmdrop(struct mm_struct *mm)
{
BUG_ON(mm == &init_mm);
+ mm_free_beancounter(mm);
mm_free_pgd(mm);
destroy_context(mm);
free_mm(mm);
@@ -508,6 +518,7 @@ fail_nocontext:
* If init_new_context() failed, we cannot use mmput() to free the mm
* because it calls destroy_context()
*/
+ mm_free_beancounter(mm);
mm_free_pgd(mm);
free_mm(mm);
return NULL;
--- ./mm/fremap.c.bc_userpages 2006-10-05 11:42:43.000000000 +0400
+++ ./mm/fremap.c 2006-10-05 12:44:20.000000000 +0400
@@ -16,6 +16,8 @@
#include <linux/module.h>
#include <linux/syscalls.h>
+#include <bc/rsspages.h>
+
#include <asm/mmu_context.h>
#include <asm/cacheflush.h>
#include <asm/tlbflush.h>
@@ -57,6 +59,10 @@ int install_page(struct mm_struct *mm, s
pte_t *pte;
pte_t pte_val;
spinlock_t *ptl;
+ struct page_beancounter *pc;
+
+ if (bc_rsspage_prepare(page, vma, &pc))
+ goto out_nocharge;
pte = get_locked_pte(mm, addr, &ptl);
if (!pte)
@@ -84,10 +90,14 @@ int install_page(struct mm_struct *mm, s
page_add_file_rmap(page);
update_mmu_cache(vma, addr, pte_val);
lazy_mmu_prot_update(pte_val);
+ bc_rsspage_charge(pc);
err = 0;
unlock:
pte_unmap_unlock(pte, ptl);
out:
+ if (err != 0)
+ bc_rsspage_release(pc);
+out_nocharge:
return err;
}
EXPORT_SYMBOL(install_page);
--- ./mm/memory.c.bc_userpages 2006-10-05 11:42:43.000000000 +0400
+++ ./mm/memory.c 2006-10-05 12:44:20.000000000 +0400
@@ -57,6 +57,8 @@
#include <asm/tlbflush.h>
#include <asm/pgtable.h>
+#include <bc/rsspages.h>
+
#include <linux/swapops.h>
#include <linux/elf.h>
@@ -1483,6 +1485,7 @@ static int do_wp_page(struct mm_struct *
pte_t entry;
int reuse = 0, ret = VM_FAULT_MINOR;
struct page *dirty_page = NULL;
+ struct page_beancounter *pc;
old_page = vm_normal_page(vma, address, orig_pte);
if (!old_page)
@@ -1568,6 +1571,9 @@ gotten:
cow_user_page(new_page, old_page, address);
}
+ if (bc_rsspage_prepare(new_page, vma, &pc))
+ goto oom;
+
/*
* Re-check the pte - we dropped the lock
*/
@@ -1596,11 +1602,14 @@ gotten:
update_mmu_cache(vma, address, entry);
lru_cache_add_active(new_page);
page_add_new_anon_rmap(new_page, vma, address);
+ bc_rsspage_charge(pc);
/* Free the old page.. */
new_page = old_page;
ret |= VM_FAULT_WRITE;
- }
+ } else
+ bc_rsspage_release(pc);
+
if (new_page)
page_cache_release(new_page);
if (old_page)
@@ -1977,6 +1986,7 @@ static int do_swap_page(struct mm_struct
swp_entry_t entry;
pte_t pte;
int ret = VM_FAULT_MINOR;
+ struct page_beancounter *pc;
if (!pte_unmap_same(mm, pmd, page_table, orig_pte))
goto out;
@@ -2009,6 +2019,11 @@ static int do_swap_page(struct mm_struct
count_vm_event(PGMAJFAULT);
}
+ if (bc_rsspage_prepare(page, vma, &pc)) {
+ ret = VM_FAULT_OOM;
+ goto out;
+ }
+
delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
mark_page_accessed(page);
lock_page(page);
@@ -2022,6 +2037,7 @@ static int do_swap_page(struct mm_struct
if (unlikely(!PageUptodate(page))) {
ret = VM_FAULT_SIGBUS;
+ bc_rsspage_uncharge(page);
goto out_nomap;
}
@@ -2037,6 +2053,7 @@ static int do_swap_page(struct mm_struct
flush_icache_page(vma, page);
set_pte_at(mm, address, page_table, pte);
page_add_anon_rmap(page, vma, address);
+ bc_rsspage_charge(pc);
swap_free(entry);
if (vm_swap_f
...
|
|
|
|
| [TOOL] BC: bcctl.c [message #7169 is a reply to message #7157] |
Thu, 05 October 2006 15:59 |
Kirill Korotaev
Messages: 137
Registered: January 2006
|
Senior Member |
|
|
Utility (bcctl) usage and examples:
Usage: bcctl <cmd> [args...]
Commands:
get
gets current bc id
enter <id> [<pid>]
moves task into bc.
if no pid - moves parent task
stat <id> [<resource>]
prints stats for bc's resource.
if no resource - prints all resources
set <id> <resource> <barrier> [<limit>]
sets barrier/limit for bc's resource
if no limit - it is set equal to barrier
Resources:
kmemsize
privvmpages
physpages
Examples:
Move shell into beancounter
$ bcctl enter 1
Look statistics
$ bcctl stat 1
resource held minheld maxheld barrier limit failcnt
kmemsize 824 0 1056 65536 65536 0
privvmpages 100 0 304 2147483647 2147483647 0
physpages 40 0 150 2147483647 2147483647 0
See how reclamation work
$ bcctl set 1 physpages 200
$ memeat
$ bcctl stat 1 physpages
resource held minheld maxheld barrier limit failcnt
physpages 40 0 200 200 200 0
Now turn swap off and run memeater again
$ swapoff -a
$ memeat
Killed
$ bcctl stat 1 physpages
resource held minheld maxheld barrier limit failcnt
physpages 41 0 200 200 200 1
---
#include <sys/syscall.h>
#include <unistd.h>
#include <stdio.h>
#include <stdlib.h>
typedef unsigned int bcid_t;
#include "bc/beancounter.h"
#define get_bcid() (syscall(__NR_get_bcid))
#define set_bcid(id, pid) (syscall(__NR_set_bcid, id, pid))
#define set_bclimit(id, res, l) (syscall(__NR_set_bclimit, id, res, l))
#define get_bcstat(id, res, s) (syscall(__NR_get_bcstat, id, res, s))
static char *res_names[] = {
"kmemsize",
"privvmpages",
"physpages",
NULL,
};
static int bc_get_id(int argc, char **argv)
{
bcid_t id;
id = get_bcid();
printf("%u\n", id);
return 0;
}
static int bc_enter_id(int argc, char **argv)
{
bcid_t id;
pid_t pid;
char *end;
int ret;
if (argc < 1)
return -2;
id = strtol(argv[0], &end, 10);
if (*end != '\0')
return -2;
if (argc > 1) {
pid = strtol(argv[1], &end, 10);
if (*end != '\0')
return -2;
} else
pid = getppid();
ret = set_bcid(id, pid);
if (ret < 0)
perror("Can't set bcid");
return ret;
}
static int bc_set_parms(int argc, char **argv)
{
bcid_t id;
int res;
unsigned long limits[2];
char *end;
int ret;
if (argc < 3)
return -2;
id = strtol(argv[0], &end, 10);
if (*end != '\0')
return -2;
for (res = 0; res_names[res] != NULL; res++)
if (strcmp(res_names[res], argv[1]) == 0)
break;
if (res_names[res] == NULL)
return -2;
limits[0] = strtoul(argv[2], &end, 10);
if (*end != '\0')
return -2;
if (argc > 3) {
limits[1] = strtoul(argv[3], &end, 10);
if (*end != '\0')
return -2;
} else
limits[1] = limits[0];
ret = set_bclimit(id, res, limits);
if (ret < 0)
perror("Can't set limits");
return ret;
}
static void bc_show_stat_hdr(void)
{
printf("%-12s %10s %10s %10s %10s %10s %10s\n",
"resource",
"held",
"minheld",
"maxheld",
"barrier",
"limit",
"failcnt");
}
static int bc_show_stat(bcid_t id, int res)
{
struct bc_resource_parm parm;
int ret;
ret = get_bcstat(id, res, &parm);
if (ret < 0) {
perror("Can't get stat");
return ret;
}
printf("%-12s %10lu %10lu %10lu %10lu %10lu %10lu\n",
res_names[res],
parm.held,
parm.minheld,
parm.maxheld,
parm.barrier,
parm.limit,
parm.failcnt);
return 0;
}
static int bc_get_stat(int argc, char **argv)
{
bcid_t id;
int res, ret;
char *end;
if (argc < 1)
return -2;
id = strtol(argv[0], &end, 10);
if (*end != '\0')
return -2;
if (argc > 1) {
for (res = 0; res_names[res] != NULL; res++)
if (strcmp(res_names[res], argv[1]) == 0)
break;
if (res_names[res] == NULL)
return -2;
bc_show_stat_hdr();
return bc_show_stat(id, res);
}
bc_show_stat_hdr();
for (res = 0; res_names[res] != NULL; res++) {
ret = bc_show_stat(id, res);
if (ret < 0)
return ret;
}
return 0;
}
static struct bc_command {
char *name;
int (*cmd)(int argc, char **argv);
char *args;
char *help;
} bc_commands[] = {
{
.name = "get",
.cmd = bc_get_id,
.args = "",
.help = "gets current bc id",
},
{
.name = "enter",
.cmd = bc_enter_id,
.args = "<id> [<pid>]",
.help = "moves task into bc.\n"
"\t\tif no pid - moves parent task",
},
{
.name = "stat",
.cmd = bc_get_stat,
.args = "<id> [<resource>]",
.help = "prints stats for bc's resource.\n\t\t"
"if no resource - prints all resources",
},
{
.name = "set",
.cmd = bc_set_parms,
.args = "<id> <resource> <barrier> [<limit>]",
.help = "sets barrier/limit for bc's resource\n"
"\t\tif no limit - it is set equal to barrier",
},
{
.name = NULL,
},
};
static void usage(void)
{
int i;
printf("Usage: bcctl <cmd> [args...]\n");
printf("Commands:\n");
for (i = 0; bc_commands[i].name != NULL; i++)
printf("\t%s %s\n\t\t%s\n",
bc_commands[i].name,
bc_commands[i].args,
bc_commands[i].help);
printf("Resources:\n");
for (i = 0; res_names[i] != NULL; i++)
printf("\t%s\n", res_names[i]);
}
int main(int argc, char **argv)
{
int i, ret;
if (argc < 2) {
usage();
return 0;
}
ret = -1;
for (i = 0; bc_commands[i].name != NULL; i++) {
if (strcmp(argv[1], bc_commands[i].name) != 0)
continue;
ret = bc_commands[i].cmd(argc - 2, argv + 2);
}
if (ret < 0)
usage();
return ret;
}
|
|
|
|
Goto Forum:
Current Time: Tue Aug 13 23:22:50 GMT 2024
Total time taken to generate the page: 0.02876 seconds
|
OpenVZ Forum
Members
Search
Help
Register
Login
Home
