OpenVZ Forum


Home » Mailing lists » Devel » [PATCH] BC: resource beancounters (v4) (added user memory)
[PATCH 12/13] BC: vmrss (core) [message #5934 is a reply to message #5922] Tue, 05 September 2006 15:28 Go to previous messageGo to previous message
dev is currently offline  dev
Messages: 1693
Registered: September 2005
Location: Moscow
Senior Member
This is the core of vmrss accounting.

The main introduced object is page_beancounter.
It ties together page and BCs which use the page.
This allows correctly account fractions of memory shared
between BCs (http://wiki.openvz.org/RSS_fractions_accounting)

Accounting API:
1. bc_alloc_rss_counter() allocates a tie between page and BC
2. bc_free_rss_counter frees it.

(1) and (2) must be done each time a page is about
to be added to someone's rss.

3. When page is touched by BC (i.e. by any task which mm belongs to BC)
page is bc_vmrss_page_add()-ed to that BC. Touching page leads
to subtracting it from unused_prvvmpages and adding to held_pages.
4. When page is unmapped from BC it is bc_vmrss_page_del()-ed from it.

5. When task forks all it's mapped pages must be bc_vmrss_page_dup()-ed.
i.e. page beancounter reference counter must be increased.

6. Some pages (former PGReserved) must be added to rss, but without
having a reference on it. These pages are bc_vmrss_page_add_noref()-ed.

Signed-Off-By: Pavel Emelianov <xemul@sw.ru>
Signed-Off-By: Kirill Korotaev <dev@sw.ru>

---

include/bc/beancounter.h | 3
include/bc/vmpages.h | 4
include/bc/vmrss.h | 72 ++++++
include/linux/mm.h | 6
include/linux/shmem_fs.h | 2
init/main.c | 2
kernel/bc/Kconfig | 9
kernel/bc/Makefile | 1
kernel/bc/beancounter.c | 9
kernel/bc/vmpages.c | 7
kernel/bc/vmrss.c | 508 +++++++++++++++++++++++++++++++++++++++++++++++
mm/shmem.c | 6
12 files changed, 627 insertions(+), 2 deletions(-)

--- ./include/bc/beancounter.h.bcrsscore 2006-09-05 13:44:33.000000000 +0400
+++ ./include/bc/beancounter.h 2006-09-05 13:50:29.000000000 +0400
@@ -68,6 +68,9 @@ struct beancounter {
struct hlist_node hash;

unsigned long unused_privvmpages;
+#ifdef CONFIG_BEANCOUNTERS_RSS
+ unsigned long long rss_pages;
+#endif
/* resources statistics and settings */
struct bc_resource_parm bc_parms[BC_RESOURCES];
};
--- ./include/bc/vmpages.h.bcrsscore 2006-09-05 13:40:21.000000000 +0400
+++ ./include/bc/vmpages.h 2006-09-05 13:46:35.000000000 +0400
@@ -77,6 +77,8 @@ void bc_locked_shm_uncharge(struct shmem
put_beancounter((info)->shm_bc); \
} while (0)

+#define mm_same_bc(mm1, mm2) ((mm1)->mm_bc == (mm2)->mm_bc)
+
void bc_update_privvmpages(struct beancounter *bc);

#else /* CONFIG_BEANCOUNTERS */
@@ -136,6 +138,8 @@ static inline void bc_locked_shm_uncharg
#define shmi_init_bc(info) do { } while (0)
#define shmi_free_bc(info) do { } while (0)

+#define mm_same_bc(mm1, mm2) (1)
+
#endif /* CONFIG_BEANCOUNTERS */
#endif

--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./include/bc/vmrss.h 2006-09-05 13:50:25.000000000 +0400
@@ -0,0 +1,72 @@
+/*
+ * include/ub/vmrss.h
+ *
+ * Copyright (C) 2006 OpenVZ. SWsoft Inc
+ *
+ */
+
+#ifndef __BC_VMRSS_H_
+#define __BC_VMRSS_H_
+
+struct page_beancounter;
+
+struct page;
+struct mm_struct;
+struct vm_area_struct;
+
+/* values that represens page's 'weight' in bc rss accounting */
+#define PB_PAGE_WEIGHT_SHIFT 24
+#define PB_PAGE_WEIGHT (1 << PB_PAGE_WEIGHT_SHIFT)
+/* page obtains one more reference within beancounter */
+#define PB_COPY_SAME ((struct page_beancounter *)-1)
+
+#ifdef CONFIG_BEANCOUNTERS_RSS
+
+struct page_beancounter * __must_check bc_alloc_rss_counter(void);
+struct page_beancounter * __must_check bc_alloc_rss_counter_list(long num,
+ struct page_beancounter *list);
+
+void bc_free_rss_counter(struct page_beancounter *rc);
+
+void bc_vmrss_page_add(struct page *pg, struct mm_struct *mm,
+ struct vm_area_struct *vma, struct page_beancounter **ppb);
+void bc_vmrss_page_del(struct page *pg, struct mm_struct *mm,
+ struct vm_area_struct *vma);
+void bc_vmrss_page_dup(struct page *pg, struct mm_struct *mm,
+ struct vm_area_struct *vma, struct page_beancounter **ppb);
+void bc_vmrss_page_add_noref(struct page *pg, struct mm_struct *mm,
+ struct vm_area_struct *vma);
+
+unsigned long mm_rss_pages(struct mm_struct *mm, unsigned long start,
+ unsigned long end);
+
+void bc_init_rss(void);
+
+#else /* CONFIG_BEANCOUNTERS_RSS */
+
+static inline struct page_beancounter * __must_check bc_alloc_rss_counter(void)
+{
+ return NULL;
+}
+
+static inline struct page_beancounter * __must_check bc_alloc_rss_counter_list(
+ long num, struct page_beancounter *list)
+{
+ return NULL;
+}
+
+static inline void bc_free_rss_counter(struct page_beancounter *rc)
+{
+}
+
+#define bc_vmrss_page_add(pg, mm, vma, pb) do { } while (0)
+#define bc_vmrss_page_del(pg, mm, vma) do { } while (0)
+#define bc_vmrss_page_dup(pg, mm, vma, pb) do { } while (0)
+#define bc_vmrss_page_add_noref(pg, mm, vma) do { } while (0)
+#define mm_rss_pages(mm, start, end) (0)
+
+#define bc_init_rss() do { } while (0)
+
+#endif /* CONFIG_BEANCOUNTERS_RSS */
+
+#endif
--- ./include/linux/mm.h.bcrsscore 2006-09-05 13:06:37.000000000 +0400
+++ ./include/linux/mm.h 2006-09-05 13:47:12.000000000 +0400
@@ -275,11 +275,15 @@ struct page {
unsigned long trace[8];
#endif
#ifdef CONFIG_BEANCOUNTERS
- struct beancounter *page_bc;
+ union {
+ struct beancounter *page_bc;
+ struct page_beancounter *page_pb;
+ };
#endif
};

#define page_bc(page) ((page)->page_bc)
+#define page_pb(page) ((page)->page_pb)
#define page_private(page) ((page)->private)
#define set_page_private(page, v) ((page)->private = (v))

--- ./include/linux/shmem_fs.h.bcrsscore 2006-09-05 12:59:27.000000000 +0400
+++ ./include/linux/shmem_fs.h 2006-09-05 13:50:19.000000000 +0400
@@ -41,4 +41,6 @@ static inline struct shmem_inode_info *S
return container_of(inode, struct shmem_inode_info, vfs_inode);
}

+int is_shmem_mapping(struct address_space *mapping);
+
#endif
--- ./init/main.c.bcrsscore 2006-09-05 12:54:17.000000000 +0400
+++ ./init/main.c 2006-09-05 13:46:35.000000000 +0400
@@ -51,6 +51,7 @@
#include <linux/lockdep.h>

#include <bc/beancounter.h>
+#include <bc/vmrss.h>

#include <asm/io.h>
#include <asm/bugs.h>
@@ -608,6 +609,7 @@ asmlinkage void __init start_kernel(void
check_bugs();

acpi_early_init(); /* before LAPIC and SMP init */
+ bc_init_rss();

/* Do the rest non-__init'ed, we're now alive */
rest_init();
--- ./kernel/bc/Kconfig.bcrsscore 2006-09-05 12:54:14.000000000 +0400
+++ ./kernel/bc/Kconfig 2006-09-05 13:50:35.000000000 +0400
@@ -22,4 +22,13 @@ config BEANCOUNTERS
per-process basis. Per-process accounting doesn't prevent malicious
users from spawning a lot of resource-consuming processes.

+config BEANCOUNTERS_RSS
+ bool "Account physical memory usage"
+ default y
+ depends on BEANCOUNTERS
+ help
+ This allows to estimate per beancounter physical memory usage.
+ Implemented alghorithm accounts shared pages of memory as well,
+ dividing them by number of beancounter which use the page.
+
endmenu
--- ./kernel/bc/Makefile.bcrsscore 2006-09-05 12:59:37.000000000 +0400
+++ ./kernel/bc/Makefile 2006-09-05 13:50:48.000000000 +0400
@@ -9,3 +9,4 @@ obj-y += misc.o
obj-y += sys.o
obj-y += kmem.o
obj-y += vmpages.o
+obj-$(CONFIG_BEANCOUNTERS_RSS) += vmrss.o
--- ./kernel/bc/beancounter.c.bcrsscore 2006-09-05 13:44:53.000000000 +0400
+++ ./kernel/bc/beancounter.c 2006-09-05 13:49:38.000000000 +0400
@@ -11,6 +11,7 @@
#include <linux/hash.h>

#include <bc/beancounter.h>
+#include <bc/vmrss.h>

static kmem_cache_t *bc_cachep;
static struct beancounter default_beancounter;
@@ -112,6 +113,14 @@ void put_beancounter(struct beancounter
printk("BC: %d has %lu of %s held on put", bc->bc_id,
bc->bc_parms[i].held, bc_rnames[i]);

+ if (bc->unused_privvmpages != 0)
+ printk("BC: %d has %lu of unused pages held on put", bc->bc_id,
+ bc->unused_privvmpages);
+#ifdef CONFIG_BEANCOUNTERS_RSS
+ if (bc->rss_pages != 0)
+ printk("BC: %d hash %llu of rss pages held on put", bc->bc_id,
+ bc->rss_pages);
+#endif
hlist_del(&bc->hash);
nr_beancounters--;
spin_unlock_irqrestore(&bc_hash_lock, flags);
--- ./kernel/bc/vmpages.c.bcrsscore 2006-09-05 13:45:34.000000000 +0400
+++ ./kernel/bc/vmpages.c 2006-09-05 13:48:50.000000000 +0400
@@ -11,12 +11,17 @@

#include <bc/beancounter.h>
#include <bc/vmpages.h>
+#include <bc/vmrss.h>

#include <asm/page.h>

void bc_update_privvmpages(struct beancounter *bc)
{
- bc->bc_parms[BC_PRIVVMPAGES].held = bc->unused_privvmpages;
+ bc->bc_parms[BC_PRIVVMPAGES].held = bc->unused_privvmpages
+#ifdef CONFIG_BEANCOUNTERS_RSS
+ + (bc->rss_pages >> PB_PAGE_WEIGHT_SHIFT)
+#endif
+ ;
bc_adjust_minheld(bc, BC_PRIVVMPAGES);
bc_adjust_maxheld(bc, BC_PRIVVMPAGES);
}
--- /dev/null 2006-07-18 14:52:43.075228448 +0400
+++ ./kernel/bc/vmrss.c 2006-09-05 13:51:21.000000000 +0400
@@ -0,0 +1,508 @@
+/*
+ * kernel/bc/vmrss.c
+ *
+ * Copyright (C) 2006 OpenVZ. SWsoft Inc
+ *
+ */
+
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/list.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/shmem_fs.h>
+#include <linux/highmem.h>
+
+#include <bc/beancounter.h>
+#include <bc/vmpages.h>
+#include <bc/vmrss.h>
+
+#include <asm/pgtable.h>
+
+/*
+ * Core object of accounting.
+ * page_beancounter (or rss_counter) ties together page an bc.
+ * Page has associated circular list of such pbs. When page is
+ * shared between bcs then it's size is splitted between all of
+ * them in 2^n-s parts.
+ *
+ * E.g. three bcs will share page like 1/2:1/4:1/4
+ * adding one more reference would produce such a change:
+ * 1/2(bc1) : 1/4(bc2) : 1/4(bc3) ->
+ * (1/4(bc1) + 1/4(bc1)) : 1/4(bc2) : 1/4(bc3) ->
+ * 1/4(bc2) : 1/4(bc3) : 1/4(bc4) : 1/4(bc1)
+ */
+
+#define PB_MAGIC 0x62700001UL
+
+struct page_beancounter {
+ unsigned ...

[ Show the rest of the message ]

Report message to a moderator

[Message index]
 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Acks for 3 pid-namespace patches
Next Topic: [Patch 01/05]- Containers: Documentation on using containers
Goto Forum:
  

[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Fri Jul 18 13:24:31 GMT 2025

Total time taken to generate the page: 0.08164 seconds
Powered by FUDforum Powered by Virtuozzo Containers