2012-01-10 02:51:56 +04:00
/*
* zsmalloc memory allocator
*
* Copyright ( C ) 2011 Nitin Gupta
2014-01-31 03:45:55 +04:00
* Copyright ( C ) 2012 , 2013 Minchan Kim
2012-01-10 02:51:56 +04:00
*
* This code is released using a dual license strategy : BSD / GPL
* You can choose the license that better fits your requirements .
*
* Released under the terms of 3 - clause BSD License
* Released under the terms of GNU General Public License Version 2.0
*/
2012-06-10 04:41:14 +04:00
/*
2013-12-11 06:04:37 +04:00
* This allocator is designed for use with zram . Thus , the allocator is
* supposed to work well under low memory conditions . In particular , it
* never attempts higher order page allocation which is very likely to
* fail under memory pressure . On the other hand , if we just use single
* ( 0 - order ) pages , it would suffer from very high fragmentation - -
* any object of size PAGE_SIZE / 2 or larger would occupy an entire page .
* This was one of the major issues with its predecessor ( xvmalloc ) .
2012-06-10 04:41:14 +04:00
*
* To overcome these issues , zsmalloc allocates a bunch of 0 - order pages
* and links them together using various ' struct page ' fields . These linked
* pages act as a single higher - order page i . e . an object can span 0 - order
* page boundaries . The code refers to these linked pages as a single entity
* called zspage .
*
2013-12-11 06:04:37 +04:00
* For simplicity , zsmalloc can only allocate objects of size up to PAGE_SIZE
* since this satisfies the requirements of all its current users ( in the
* worst case , page is incompressible and is thus stored " as-is " i . e . in
* uncompressed form ) . For allocation requests larger than this size , failure
* is returned ( see zs_malloc ) .
*
* Additionally , zs_malloc ( ) does not return a dereferenceable pointer .
* Instead , it returns an opaque handle ( unsigned long ) which encodes actual
* location of the allocated object . The reason for this indirection is that
* zsmalloc does not keep zspages permanently mapped since that would cause
* issues on 32 - bit systems where the VA region for kernel space mappings
* is very small . So , before using the allocating memory , the object has to
* be mapped using zs_map_object ( ) to get a usable pointer and subsequently
* unmapped using zs_unmap_object ( ) .
*
2012-06-10 04:41:14 +04:00
* Following is how we use various fields and flags of underlying
* struct page ( s ) to form a zspage .
*
* Usage of struct page fields :
* page - > first_page : points to the first component ( 0 - order ) page
* page - > index ( union with page - > freelist ) : offset of the first object
* starting in this page . For the first page , this is
* always 0 , so we use this field ( aka freelist ) to point
* to the first free object in zspage .
* page - > lru : links together all component pages ( except the first page )
* of a zspage
*
* For _first_ page only :
*
* page - > private ( union with page - > first_page ) : refers to the
* component page after the first page
* page - > freelist : points to the first free object in zspage .
* Free objects are linked together using in - place
* metadata .
* page - > objects : maximum number of objects we can store in this
* zspage ( class - > zspage_order * PAGE_SIZE / class - > size )
* page - > lru : links together first pages of various zspages .
* Basically forming list of zspages in a fullness group .
* page - > mapping : class index and fullness group of the zspage
*
* Usage of struct page flags :
* PG_private : identifies the first component page
* PG_private2 : identifies the last component page
*
*/
2012-01-10 02:51:56 +04:00
# ifdef CONFIG_ZSMALLOC_DEBUG
# define DEBUG
# endif
# include <linux/module.h>
# include <linux/kernel.h>
# include <linux/bitops.h>
# include <linux/errno.h>
# include <linux/highmem.h>
# include <linux/string.h>
# include <linux/slab.h>
# include <asm/tlbflush.h>
# include <asm/pgtable.h>
# include <linux/cpumask.h>
# include <linux/cpu.h>
2012-02-13 18:47:49 +04:00
# include <linux/vmalloc.h>
2012-07-18 20:55:55 +04:00
# include <linux/hardirq.h>
2012-08-08 10:12:17 +04:00
# include <linux/spinlock.h>
# include <linux/types.h>
2014-01-31 03:45:50 +04:00
# include <linux/zsmalloc.h>
2014-08-07 03:08:38 +04:00
# include <linux/zpool.h>
2012-08-08 10:12:17 +04:00
/*
* This must be power of 2 and greater than of equal to sizeof ( link_free ) .
* These two conditions ensure that any ' struct link_free ' itself doesn ' t
* span more than 1 page which avoids complex case of mapping 2 pages simply
* to restore link_free pointer values .
*/
# define ZS_ALIGN 8
/*
* A single ' zspage ' is composed of up to 2 ^ N discontiguous 0 - order ( single )
* pages . ZS_MAX_ZSPAGE_ORDER defines upper limit on N .
*/
# define ZS_MAX_ZSPAGE_ORDER 2
# define ZS_MAX_PAGES_PER_ZSPAGE (_AC(1, UL) << ZS_MAX_ZSPAGE_ORDER)
/*
* Object location ( < PFN > , < obj_idx > ) is encoded as
2013-12-11 06:04:37 +04:00
* as single ( unsigned long ) handle value .
2012-08-08 10:12:17 +04:00
*
* Note that object index < obj_idx > is relative to system
* page < PFN > it is stored in , so for each sub - page belonging
* to a zspage , obj_idx starts with 0.
*
* This is made more complicated by various memory models and PAE .
*/
# ifndef MAX_PHYSMEM_BITS
# ifdef CONFIG_HIGHMEM64G
# define MAX_PHYSMEM_BITS 36
# else /* !CONFIG_HIGHMEM64G */
/*
* If this definition of MAX_PHYSMEM_BITS is used , OBJ_INDEX_BITS will just
* be PAGE_SHIFT
*/
# define MAX_PHYSMEM_BITS BITS_PER_LONG
# endif
# endif
# define _PFN_BITS (MAX_PHYSMEM_BITS - PAGE_SHIFT)
# define OBJ_INDEX_BITS (BITS_PER_LONG - _PFN_BITS)
# define OBJ_INDEX_MASK ((_AC(1, UL) << OBJ_INDEX_BITS) - 1)
# define MAX(a, b) ((a) >= (b) ? (a) : (b))
/* ZS_MIN_ALLOC_SIZE must be multiple of ZS_ALIGN */
# define ZS_MIN_ALLOC_SIZE \
MAX ( 32 , ( ZS_MAX_PAGES_PER_ZSPAGE < < PAGE_SHIFT > > OBJ_INDEX_BITS ) )
# define ZS_MAX_ALLOC_SIZE PAGE_SIZE
/*
2014-06-05 03:11:08 +04:00
* On systems with 4 K page size , this gives 255 size classes ! There is a
2012-08-08 10:12:17 +04:00
* trader - off here :
* - Large number of size classes is potentially wasteful as free page are
* spread across these classes
* - Small number of size classes causes large internal fragmentation
* - Probably its better to use specific size classes ( empirically
* determined ) . NOTE : all those class sizes must be set as multiple of
* ZS_ALIGN to make sure link_free itself never has to span 2 pages .
*
* ZS_MIN_ALLOC_SIZE and ZS_SIZE_CLASS_DELTA must be multiple of ZS_ALIGN
* ( reason above )
*/
2013-01-25 21:46:18 +04:00
# define ZS_SIZE_CLASS_DELTA (PAGE_SIZE >> 8)
2012-08-08 10:12:17 +04:00
/*
* We do not maintain any list for completely empty or full pages
*/
enum fullness_group {
ZS_ALMOST_FULL ,
ZS_ALMOST_EMPTY ,
_ZS_NR_FULLNESS_GROUPS ,
ZS_EMPTY ,
ZS_FULL
} ;
2014-12-13 03:57:01 +03:00
/*
* number of size_classes
*/
static int zs_size_classes ;
2012-08-08 10:12:17 +04:00
/*
* We assign a page to ZS_ALMOST_EMPTY fullness group when :
* n < = N / f , where
* n = number of allocated objects
* N = total number of objects zspage can store
2014-10-10 02:29:59 +04:00
* f = fullness_threshold_frac
2012-08-08 10:12:17 +04:00
*
* Similarly , we assign zspage to :
* ZS_ALMOST_FULL when n > N / f
* ZS_EMPTY when n = = 0
* ZS_FULL when n = = N
*
* ( see : fix_fullness_group ( ) )
*/
static const int fullness_threshold_frac = 4 ;
struct size_class {
/*
* Size of objects stored in this class . Must be multiple
* of ZS_ALIGN .
*/
int size ;
unsigned int index ;
/* Number of PAGE_SIZE sized pages to combine to form a 'zspage' */
int pages_per_zspage ;
spinlock_t lock ;
struct page * fullness_list [ _ZS_NR_FULLNESS_GROUPS ] ;
} ;
/*
* Placed within free objects to form a singly linked list .
* For every zspage , first_page - > freelist gives head of this list .
*
* This must be power of 2 and less than or equal to ZS_ALIGN
*/
struct link_free {
/* Handle of next free chunk (encodes <PFN, obj_idx>) */
void * next ;
} ;
struct zs_pool {
2014-12-13 03:57:01 +03:00
struct size_class * * size_class ;
2012-08-08 10:12:17 +04:00
gfp_t flags ; /* allocation flags used when growing pool */
zsmalloc: move pages_allocated to zs_pool
Currently, zram has no feature to limit memory so theoretically zram can
deplete system memory. Users have asked for a limit several times as even
without exhaustion zram makes it hard to control memory usage of the
platform. This patchset adds the feature.
Patch 1 makes zs_get_total_size_bytes faster because it would be used
frequently in later patches for the new feature.
Patch 2 changes zs_get_total_size_bytes's return unit from bytes to page
so that zsmalloc doesn't need unnecessary operation(ie, << PAGE_SHIFT).
Patch 3 adds new feature. I added the feature into zram layer, not
zsmalloc because limiation is zram's requirement, not zsmalloc so any
other user using zsmalloc(ie, zpool) shouldn't affected by unnecessary
branch of zsmalloc. In future, if every users of zsmalloc want the
feature, then, we could move the feature from client side to zsmalloc
easily but vice versa would be painful.
Patch 4 adds news facility to report maximum memory usage of zram so that
this avoids user polling frequently via /sys/block/zram0/ mem_used_total
and ensures transient max are not missed.
This patch (of 4):
pages_allocated has counted in size_class structure and when user of
zsmalloc want to see total_size_bytes, it should gather all of count from
each size_class to report the sum.
It's not bad if user don't see the value often but if user start to see
the value frequently, it would be not a good deal for performance pov.
This patch moves the count from size_class to zs_pool so it could reduce
memory footprint (from [255 * 8byte] to [sizeof(atomic_long_t)]).
Signed-off-by: Minchan Kim <minchan@kernel.org>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: <juno.choi@lge.com>
Cc: <seungho1.park@lge.com>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Seth Jennings <sjennings@variantweb.net>
Reviewed-by: David Horner <ds2horner@gmail.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-10-10 02:29:48 +04:00
atomic_long_t pages_allocated ;
2012-08-08 10:12:17 +04:00
} ;
2012-01-10 02:51:56 +04:00
/*
* A zspage ' s class index and fullness group
* are encoded in its ( first ) page - > mapping
*/
# define CLASS_IDX_BITS 28
# define FULLNESS_BITS 4
# define CLASS_IDX_MASK ((1 << CLASS_IDX_BITS) - 1)
# define FULLNESS_MASK ((1 << FULLNESS_BITS) - 1)
2012-07-18 20:55:56 +04:00
struct mapping_area {
2013-12-11 06:04:36 +04:00
# ifdef CONFIG_PGTABLE_MAPPING
2012-07-18 20:55:56 +04:00
struct vm_struct * vm ; /* vm area for mapping object that span pages */
# else
char * vm_buf ; /* copy buffer for objects that span pages */
# endif
char * vm_addr ; /* address of kmap_atomic()'ed pages */
enum zs_mapmode vm_mm ; /* mapping mode */
} ;
2014-08-07 03:08:38 +04:00
/* zpool driver */
# ifdef CONFIG_ZPOOL
static void * zs_zpool_create ( gfp_t gfp , struct zpool_ops * zpool_ops )
{
return zs_create_pool ( gfp ) ;
}
static void zs_zpool_destroy ( void * pool )
{
zs_destroy_pool ( pool ) ;
}
static int zs_zpool_malloc ( void * pool , size_t size , gfp_t gfp ,
unsigned long * handle )
{
* handle = zs_malloc ( pool , size ) ;
return * handle ? 0 : - 1 ;
}
static void zs_zpool_free ( void * pool , unsigned long handle )
{
zs_free ( pool , handle ) ;
}
static int zs_zpool_shrink ( void * pool , unsigned int pages ,
unsigned int * reclaimed )
{
return - EINVAL ;
}
static void * zs_zpool_map ( void * pool , unsigned long handle ,
enum zpool_mapmode mm )
{
enum zs_mapmode zs_mm ;
switch ( mm ) {
case ZPOOL_MM_RO :
zs_mm = ZS_MM_RO ;
break ;
case ZPOOL_MM_WO :
zs_mm = ZS_MM_WO ;
break ;
case ZPOOL_MM_RW : /* fallthru */
default :
zs_mm = ZS_MM_RW ;
break ;
}
return zs_map_object ( pool , handle , zs_mm ) ;
}
static void zs_zpool_unmap ( void * pool , unsigned long handle )
{
zs_unmap_object ( pool , handle ) ;
}
static u64 zs_zpool_total_size ( void * pool )
{
2014-10-10 02:29:50 +04:00
return zs_get_total_pages ( pool ) < < PAGE_SHIFT ;
2014-08-07 03:08:38 +04:00
}
static struct zpool_driver zs_zpool_driver = {
. type = " zsmalloc " ,
. owner = THIS_MODULE ,
. create = zs_zpool_create ,
. destroy = zs_zpool_destroy ,
. malloc = zs_zpool_malloc ,
. free = zs_zpool_free ,
. shrink = zs_zpool_shrink ,
. map = zs_zpool_map ,
. unmap = zs_zpool_unmap ,
. total_size = zs_zpool_total_size ,
} ;
2014-08-30 02:18:40 +04:00
MODULE_ALIAS ( " zpool-zsmalloc " ) ;
2014-08-07 03:08:38 +04:00
# endif /* CONFIG_ZPOOL */
2012-01-10 02:51:56 +04:00
/* per-cpu VM mapping areas for zspage accesses that cross page boundaries */
static DEFINE_PER_CPU ( struct mapping_area , zs_map_area ) ;
static int is_first_page ( struct page * page )
{
2012-04-25 10:23:09 +04:00
return PagePrivate ( page ) ;
2012-01-10 02:51:56 +04:00
}
static int is_last_page ( struct page * page )
{
2012-04-25 10:23:09 +04:00
return PagePrivate2 ( page ) ;
2012-01-10 02:51:56 +04:00
}
static void get_zspage_mapping ( struct page * page , unsigned int * class_idx ,
enum fullness_group * fullness )
{
unsigned long m ;
BUG_ON ( ! is_first_page ( page ) ) ;
m = ( unsigned long ) page - > mapping ;
* fullness = m & FULLNESS_MASK ;
* class_idx = ( m > > FULLNESS_BITS ) & CLASS_IDX_MASK ;
}
static void set_zspage_mapping ( struct page * page , unsigned int class_idx ,
enum fullness_group fullness )
{
unsigned long m ;
BUG_ON ( ! is_first_page ( page ) ) ;
m = ( ( class_idx & CLASS_IDX_MASK ) < < FULLNESS_BITS ) |
( fullness & FULLNESS_MASK ) ;
page - > mapping = ( struct address_space * ) m ;
}
2013-12-11 06:04:37 +04:00
/*
* zsmalloc divides the pool into various size classes where each
* class maintains a list of zspages where each zspage is divided
* into equal sized chunks . Each allocation falls into one of these
* classes depending on its size . This function returns index of the
* size class which has chunk size big enough to hold the give size .
*/
2012-01-10 02:51:56 +04:00
static int get_size_class_index ( int size )
{
int idx = 0 ;
if ( likely ( size > ZS_MIN_ALLOC_SIZE ) )
idx = DIV_ROUND_UP ( size - ZS_MIN_ALLOC_SIZE ,
ZS_SIZE_CLASS_DELTA ) ;
return idx ;
}
2013-12-11 06:04:37 +04:00
/*
* For each size class , zspages are divided into different groups
* depending on how " full " they are . This was done so that we could
* easily find empty or nearly empty zspages when we try to shrink
* the pool ( not yet implemented ) . This function returns fullness
* status of the given page .
*/
2012-01-10 02:51:56 +04:00
static enum fullness_group get_fullness_group ( struct page * page )
{
int inuse , max_objects ;
enum fullness_group fg ;
BUG_ON ( ! is_first_page ( page ) ) ;
inuse = page - > inuse ;
max_objects = page - > objects ;
if ( inuse = = 0 )
fg = ZS_EMPTY ;
else if ( inuse = = max_objects )
fg = ZS_FULL ;
else if ( inuse < = max_objects / fullness_threshold_frac )
fg = ZS_ALMOST_EMPTY ;
else
fg = ZS_ALMOST_FULL ;
return fg ;
}
2013-12-11 06:04:37 +04:00
/*
* Each size class maintains various freelists and zspages are assigned
* to one of these freelists based on the number of live objects they
* have . This functions inserts the given zspage into the freelist
* identified by < class , fullness_group > .
*/
2012-01-10 02:51:56 +04:00
static void insert_zspage ( struct page * page , struct size_class * class ,
enum fullness_group fullness )
{
struct page * * head ;
BUG_ON ( ! is_first_page ( page ) ) ;
if ( fullness > = _ZS_NR_FULLNESS_GROUPS )
return ;
head = & class - > fullness_list [ fullness ] ;
if ( * head )
list_add_tail ( & page - > lru , & ( * head ) - > lru ) ;
* head = page ;
}
2013-12-11 06:04:37 +04:00
/*
* This function removes the given zspage from the freelist identified
* by < class , fullness_group > .
*/
2012-01-10 02:51:56 +04:00
static void remove_zspage ( struct page * page , struct size_class * class ,
enum fullness_group fullness )
{
struct page * * head ;
BUG_ON ( ! is_first_page ( page ) ) ;
if ( fullness > = _ZS_NR_FULLNESS_GROUPS )
return ;
head = & class - > fullness_list [ fullness ] ;
BUG_ON ( ! * head ) ;
if ( list_empty ( & ( * head ) - > lru ) )
* head = NULL ;
else if ( * head = = page )
* head = ( struct page * ) list_entry ( ( * head ) - > lru . next ,
struct page , lru ) ;
list_del_init ( & page - > lru ) ;
}
2013-12-11 06:04:37 +04:00
/*
* Each size class maintains zspages in different fullness groups depending
* on the number of live objects they contain . When allocating or freeing
* objects , the fullness status of the page can change , say , from ALMOST_FULL
* to ALMOST_EMPTY when freeing an object . This function checks if such
* a status change has occurred for the given page and accordingly moves the
* page from the freelist of the old fullness group to that of the new
* fullness group .
*/
2012-01-10 02:51:56 +04:00
static enum fullness_group fix_fullness_group ( struct zs_pool * pool ,
struct page * page )
{
int class_idx ;
struct size_class * class ;
enum fullness_group currfg , newfg ;
BUG_ON ( ! is_first_page ( page ) ) ;
get_zspage_mapping ( page , & class_idx , & currfg ) ;
newfg = get_fullness_group ( page ) ;
if ( newfg = = currfg )
goto out ;
zsmalloc: merge size_class to reduce fragmentation
zsmalloc has many size_classes to reduce fragmentation and they are in 16
bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096. And,
zsmalloc has constraint that each zspage has 4 pages at maximum.
In this situation, we can see interesting aspect. Let's think about
size_class for 1488, 1472, ..., 1376. To prevent external fragmentation,
they uses 4 pages per zspage and so all they can contain 11 objects at
maximum.
16384 (4096 * 4) = 1488 * 11 + remains
16384 (4096 * 4) = 1472 * 11 + remains
16384 (4096 * 4) = ...
16384 (4096 * 4) = 1376 * 11 + remains
It means that they have same characteristics and classification between
them isn't needed. If we use one size_class for them, we can reduce
fragementation and save some memory since both the 1488 and 1472 sized
classes can only fit 11 objects into 4 pages, and an object that's 1472
bytes can fit into an object that's 1488 bytes, merging these classes to
always use objects that are 1488 bytes will reduce the total number of
size classes. And reducing the total number of size classes reduces
overall fragmentation, because a wider range of compressed pages can fit
into a single size class, leaving less unused objects in each size class.
For this purpose, this patch implement size_class merging. If there is
size_class that have same pages_per_zspage and same number of objects per
zspage with previous size_class, we don't create new size_class. Instead,
we use previous, same characteristic size_class. With this way, above
example sizes (1488, 1472, ..., 1376) use just one size_class so we can
get much more memory utilization.
Below is result of my simple test.
TEST ENV: EXT4 on zram, mount with discard option WORKLOAD: untar kernel
source code, remove directory in descending order in size. (drivers arch
fs sound include net Documentation firmware kernel tools)
Each line represents orig_data_size, compr_data_size, mem_used_total,
fragmentation overhead (mem_used - compr_data_size) and overhead ratio
(overhead to compr_data_size), respectively, after untar and remove
operation is executed.
* untar-nomerge.out
orig_size compr_size used_size overhead overhead_ratio
525.88MB 199.16MB 210.23MB 11.08MB 5.56%
288.32MB 97.43MB 105.63MB 8.20MB 8.41%
177.32MB 61.12MB 69.40MB 8.28MB 13.55%
146.47MB 47.32MB 56.10MB 8.78MB 18.55%
124.16MB 38.85MB 48.41MB 9.55MB 24.58%
103.93MB 31.68MB 40.93MB 9.25MB 29.21%
84.34MB 22.86MB 32.72MB 9.86MB 43.13%
66.87MB 14.83MB 23.83MB 9.00MB 60.70%
60.67MB 11.11MB 18.60MB 7.49MB 67.48%
55.86MB 8.83MB 16.61MB 7.77MB 88.03%
53.32MB 8.01MB 15.32MB 7.31MB 91.24%
* untar-merge.out
orig_size compr_size used_size overhead overhead_ratio
526.23MB 199.18MB 209.81MB 10.64MB 5.34%
288.68MB 97.45MB 104.08MB 6.63MB 6.80%
177.68MB 61.14MB 66.93MB 5.79MB 9.47%
146.83MB 47.34MB 52.79MB 5.45MB 11.51%
124.52MB 38.87MB 44.30MB 5.43MB 13.96%
104.29MB 31.70MB 36.83MB 5.13MB 16.19%
84.70MB 22.88MB 27.92MB 5.04MB 22.04%
67.11MB 14.83MB 19.26MB 4.43MB 29.86%
60.82MB 11.10MB 14.90MB 3.79MB 34.17%
55.90MB 8.82MB 12.61MB 3.79MB 42.97%
53.32MB 8.01MB 11.73MB 3.73MB 46.53%
As you can see above result, merged one has better utilization (overhead
ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: <juno.choi@lge.com>
Cc: "seungho1.park" <seungho1.park@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-13 03:56:44 +03:00
class = pool - > size_class [ class_idx ] ;
2012-01-10 02:51:56 +04:00
remove_zspage ( page , class , currfg ) ;
insert_zspage ( page , class , newfg ) ;
set_zspage_mapping ( page , class_idx , newfg ) ;
out :
return newfg ;
}
/*
* We have to decide on how many pages to link together
* to form a zspage for each size class . This is important
* to reduce wastage due to unusable space left at end of
* each zspage which is given as :
* wastage = Zp - Zp % size_class
* where Zp = zspage size = k * PAGE_SIZE where k = 1 , 2 , . . .
*
* For example , for size class of 3 / 8 * PAGE_SIZE , we should
* link together 3 PAGE_SIZE sized pages to form a zspage
* since then we can perfectly fit in 8 such objects .
*/
2012-05-03 10:40:39 +04:00
static int get_pages_per_zspage ( int class_size )
2012-01-10 02:51:56 +04:00
{
int i , max_usedpc = 0 ;
/* zspage order which gives maximum used size per KB */
int max_usedpc_order = 1 ;
2012-03-05 21:33:21 +04:00
for ( i = 1 ; i < = ZS_MAX_PAGES_PER_ZSPAGE ; i + + ) {
2012-01-10 02:51:56 +04:00
int zspage_size ;
int waste , usedpc ;
zspage_size = i * PAGE_SIZE ;
waste = zspage_size % class_size ;
usedpc = ( zspage_size - waste ) * 100 / zspage_size ;
if ( usedpc > max_usedpc ) {
max_usedpc = usedpc ;
max_usedpc_order = i ;
}
}
return max_usedpc_order ;
}
/*
* A single ' zspage ' is composed of many system pages which are
* linked together using fields in struct page . This function finds
* the first / head page , given any component page of a zspage .
*/
static struct page * get_first_page ( struct page * page )
{
if ( is_first_page ( page ) )
return page ;
else
return page - > first_page ;
}
static struct page * get_next_page ( struct page * page )
{
struct page * next ;
if ( is_last_page ( page ) )
next = NULL ;
else if ( is_first_page ( page ) )
2013-07-12 11:08:13 +04:00
next = ( struct page * ) page_private ( page ) ;
2012-01-10 02:51:56 +04:00
else
next = list_entry ( page - > lru . next , struct page , lru ) ;
return next ;
}
2013-11-22 21:30:41 +04:00
/*
* Encode < page , obj_idx > as a single handle value .
* On hardware platforms with physical memory starting at 0x0 the pfn
* could be 0 so we ensure that the handle will never be 0 by adjusting the
* encoded obj_idx value before encoding .
*/
2012-01-10 02:51:56 +04:00
static void * obj_location_to_handle ( struct page * page , unsigned long obj_idx )
{
unsigned long handle ;
if ( ! page ) {
BUG_ON ( obj_idx ) ;
return NULL ;
}
handle = page_to_pfn ( page ) < < OBJ_INDEX_BITS ;
2013-11-22 21:30:41 +04:00
handle | = ( ( obj_idx + 1 ) & OBJ_INDEX_MASK ) ;
2012-01-10 02:51:56 +04:00
return ( void * ) handle ;
}
2013-11-22 21:30:41 +04:00
/*
* Decode < page , obj_idx > pair from the given object handle . We adjust the
* decoded obj_idx back to its original value since it was adjusted in
* obj_location_to_handle ( ) .
*/
2012-06-08 10:39:25 +04:00
static void obj_handle_to_location ( unsigned long handle , struct page * * page ,
2012-01-10 02:51:56 +04:00
unsigned long * obj_idx )
{
2012-06-08 10:39:25 +04:00
* page = pfn_to_page ( handle > > OBJ_INDEX_BITS ) ;
2013-11-22 21:30:41 +04:00
* obj_idx = ( handle & OBJ_INDEX_MASK ) - 1 ;
2012-01-10 02:51:56 +04:00
}
static unsigned long obj_idx_to_offset ( struct page * page ,
unsigned long obj_idx , int class_size )
{
unsigned long off = 0 ;
if ( ! is_first_page ( page ) )
off = page - > index ;
return off + obj_idx * class_size ;
}
2012-04-02 18:13:56 +04:00
static void reset_page ( struct page * page )
{
clear_bit ( PG_private , & page - > flags ) ;
clear_bit ( PG_private_2 , & page - > flags ) ;
set_page_private ( page , 0 ) ;
page - > mapping = NULL ;
page - > freelist = NULL ;
2013-02-23 04:34:59 +04:00
page_mapcount_reset ( page ) ;
2012-04-02 18:13:56 +04:00
}
2012-01-10 02:51:56 +04:00
static void free_zspage ( struct page * first_page )
{
2012-04-02 18:13:56 +04:00
struct page * nextp , * tmp , * head_extra ;
2012-01-10 02:51:56 +04:00
BUG_ON ( ! is_first_page ( first_page ) ) ;
BUG_ON ( first_page - > inuse ) ;
2012-04-02 18:13:56 +04:00
head_extra = ( struct page * ) page_private ( first_page ) ;
2012-01-10 02:51:56 +04:00
2012-04-02 18:13:56 +04:00
reset_page ( first_page ) ;
2012-01-10 02:51:56 +04:00
__free_page ( first_page ) ;
/* zspage with only 1 system page */
2012-04-02 18:13:56 +04:00
if ( ! head_extra )
2012-01-10 02:51:56 +04:00
return ;
2012-04-02 18:13:56 +04:00
list_for_each_entry_safe ( nextp , tmp , & head_extra - > lru , lru ) {
2012-01-10 02:51:56 +04:00
list_del ( & nextp - > lru ) ;
2012-04-02 18:13:56 +04:00
reset_page ( nextp ) ;
2012-01-10 02:51:56 +04:00
__free_page ( nextp ) ;
}
2012-04-02 18:13:56 +04:00
reset_page ( head_extra ) ;
__free_page ( head_extra ) ;
2012-01-10 02:51:56 +04:00
}
/* Initialize a newly allocated zspage */
static void init_zspage ( struct page * first_page , struct size_class * class )
{
unsigned long off = 0 ;
struct page * page = first_page ;
BUG_ON ( ! is_first_page ( first_page ) ) ;
while ( page ) {
struct page * next_page ;
struct link_free * link ;
2014-10-10 02:30:01 +04:00
unsigned int i = 1 ;
2014-12-13 03:56:58 +03:00
void * vaddr ;
2012-01-10 02:51:56 +04:00
/*
* page - > index stores offset of first object starting
* in the page . For the first page , this is always 0 ,
* so we use first_page - > index ( aka - > freelist ) to store
* head of corresponding zspage ' s freelist .
*/
if ( page ! = first_page )
page - > index = off ;
2014-12-13 03:56:58 +03:00
vaddr = kmap_atomic ( page ) ;
link = ( struct link_free * ) vaddr + off / sizeof ( * link ) ;
2014-10-10 02:30:01 +04:00
while ( ( off + = class - > size ) < PAGE_SIZE ) {
link - > next = obj_location_to_handle ( page , i + + ) ;
link + = class - > size / sizeof ( * link ) ;
2012-01-10 02:51:56 +04:00
}
/*
* We now come to the last ( full or partial ) object on this
* page , which must point to the first object on the next
* page ( if present )
*/
next_page = get_next_page ( page ) ;
link - > next = obj_location_to_handle ( next_page , 0 ) ;
2014-12-13 03:56:58 +03:00
kunmap_atomic ( vaddr ) ;
2012-01-10 02:51:56 +04:00
page = next_page ;
2014-10-10 02:30:01 +04:00
off % = PAGE_SIZE ;
2012-01-10 02:51:56 +04:00
}
}
/*
* Allocate a zspage for the given size class
*/
static struct page * alloc_zspage ( struct size_class * class , gfp_t flags )
{
int i , error ;
2012-06-14 01:03:42 +04:00
struct page * first_page = NULL , * uninitialized_var ( prev_page ) ;
2012-01-10 02:51:56 +04:00
/*
* Allocate individual pages and link them together as :
* 1. first page - > private = first sub - page
* 2. all sub - pages are linked together using page - > lru
* 3. each sub - page is linked to the first page using page - > first_page
*
* For each size class , First / Head pages are linked together using
* page - > lru . Also , we set PG_private to identify the first page
* ( i . e . no other sub - page has this flag set ) and PG_private_2 to
* identify the last page .
*/
error = - ENOMEM ;
2012-05-03 10:40:39 +04:00
for ( i = 0 ; i < class - > pages_per_zspage ; i + + ) {
2012-06-14 01:03:42 +04:00
struct page * page ;
2012-01-10 02:51:56 +04:00
page = alloc_page ( flags ) ;
if ( ! page )
goto cleanup ;
INIT_LIST_HEAD ( & page - > lru ) ;
if ( i = = 0 ) { /* first page */
2012-04-25 10:23:09 +04:00
SetPagePrivate ( page ) ;
2012-01-10 02:51:56 +04:00
set_page_private ( page , 0 ) ;
first_page = page ;
first_page - > inuse = 0 ;
}
if ( i = = 1 )
2013-07-12 11:08:13 +04:00
set_page_private ( first_page , ( unsigned long ) page ) ;
2012-01-10 02:51:56 +04:00
if ( i > = 1 )
page - > first_page = first_page ;
if ( i > = 2 )
list_add ( & page - > lru , & prev_page - > lru ) ;
2012-05-03 10:40:39 +04:00
if ( i = = class - > pages_per_zspage - 1 ) /* last page */
2012-04-25 10:23:09 +04:00
SetPagePrivate2 ( page ) ;
2012-01-10 02:51:56 +04:00
prev_page = page ;
}
init_zspage ( first_page , class ) ;
first_page - > freelist = obj_location_to_handle ( first_page , 0 ) ;
/* Maximum number of objects we can store in this zspage */
2012-05-03 10:40:39 +04:00
first_page - > objects = class - > pages_per_zspage * PAGE_SIZE / class - > size ;
2012-01-10 02:51:56 +04:00
error = 0 ; /* Success */
cleanup :
if ( unlikely ( error ) & & first_page ) {
free_zspage ( first_page ) ;
first_page = NULL ;
}
return first_page ;
}
static struct page * find_get_zspage ( struct size_class * class )
{
int i ;
struct page * page ;
for ( i = 0 ; i < _ZS_NR_FULLNESS_GROUPS ; i + + ) {
page = class - > fullness_list [ i ] ;
if ( page )
break ;
}
return page ;
}
2013-12-11 06:04:36 +04:00
# ifdef CONFIG_PGTABLE_MAPPING
2012-07-18 20:55:56 +04:00
static inline int __zs_cpu_up ( struct mapping_area * area )
{
/*
* Make sure we don ' t leak memory if a cpu UP notification
* and zs_init ( ) race and both call zs_cpu_up ( ) on the same cpu
*/
if ( area - > vm )
return 0 ;
area - > vm = alloc_vm_area ( PAGE_SIZE * 2 , NULL ) ;
if ( ! area - > vm )
return - ENOMEM ;
return 0 ;
}
static inline void __zs_cpu_down ( struct mapping_area * area )
{
if ( area - > vm )
free_vm_area ( area - > vm ) ;
area - > vm = NULL ;
}
static inline void * __zs_map_object ( struct mapping_area * area ,
struct page * pages [ 2 ] , int off , int size )
{
2014-08-07 03:06:58 +04:00
BUG_ON ( map_vm_area ( area - > vm , PAGE_KERNEL , pages ) ) ;
2012-07-18 20:55:56 +04:00
area - > vm_addr = area - > vm - > addr ;
return area - > vm_addr + off ;
}
static inline void __zs_unmap_object ( struct mapping_area * area ,
struct page * pages [ 2 ] , int off , int size )
{
unsigned long addr = ( unsigned long ) area - > vm_addr ;
2013-03-27 04:43:14 +04:00
unmap_kernel_range ( addr , PAGE_SIZE * 2 ) ;
2012-07-18 20:55:56 +04:00
}
2013-12-11 06:04:36 +04:00
# else /* CONFIG_PGTABLE_MAPPING */
2012-07-18 20:55:56 +04:00
static inline int __zs_cpu_up ( struct mapping_area * area )
{
/*
* Make sure we don ' t leak memory if a cpu UP notification
* and zs_init ( ) race and both call zs_cpu_up ( ) on the same cpu
*/
if ( area - > vm_buf )
return 0 ;
2014-12-13 03:57:01 +03:00
area - > vm_buf = kmalloc ( ZS_MAX_ALLOC_SIZE , GFP_KERNEL ) ;
2012-07-18 20:55:56 +04:00
if ( ! area - > vm_buf )
return - ENOMEM ;
return 0 ;
}
static inline void __zs_cpu_down ( struct mapping_area * area )
{
2014-12-13 03:57:01 +03:00
kfree ( area - > vm_buf ) ;
2012-07-18 20:55:56 +04:00
area - > vm_buf = NULL ;
}
static void * __zs_map_object ( struct mapping_area * area ,
struct page * pages [ 2 ] , int off , int size )
2012-07-03 01:15:49 +04:00
{
int sizes [ 2 ] ;
void * addr ;
2012-07-18 20:55:56 +04:00
char * buf = area - > vm_buf ;
2012-07-03 01:15:49 +04:00
2012-07-18 20:55:56 +04:00
/* disable page faults to match kmap_atomic() return conditions */
pagefault_disable ( ) ;
/* no read fastpath */
if ( area - > vm_mm = = ZS_MM_WO )
goto out ;
2012-07-03 01:15:49 +04:00
sizes [ 0 ] = PAGE_SIZE - off ;
sizes [ 1 ] = size - sizes [ 0 ] ;
/* copy object to per-cpu buffer */
addr = kmap_atomic ( pages [ 0 ] ) ;
memcpy ( buf , addr + off , sizes [ 0 ] ) ;
kunmap_atomic ( addr ) ;
addr = kmap_atomic ( pages [ 1 ] ) ;
memcpy ( buf + sizes [ 0 ] , addr , sizes [ 1 ] ) ;
kunmap_atomic ( addr ) ;
2012-07-18 20:55:56 +04:00
out :
return area - > vm_buf ;
2012-07-03 01:15:49 +04:00
}
2012-07-18 20:55:56 +04:00
static void __zs_unmap_object ( struct mapping_area * area ,
struct page * pages [ 2 ] , int off , int size )
2012-07-03 01:15:49 +04:00
{
int sizes [ 2 ] ;
void * addr ;
2012-07-18 20:55:56 +04:00
char * buf = area - > vm_buf ;
2012-07-03 01:15:49 +04:00
2012-07-18 20:55:56 +04:00
/* no write fastpath */
if ( area - > vm_mm = = ZS_MM_RO )
goto out ;
2012-07-03 01:15:49 +04:00
sizes [ 0 ] = PAGE_SIZE - off ;
sizes [ 1 ] = size - sizes [ 0 ] ;
/* copy per-cpu buffer to object */
addr = kmap_atomic ( pages [ 0 ] ) ;
memcpy ( addr + off , buf , sizes [ 0 ] ) ;
kunmap_atomic ( addr ) ;
addr = kmap_atomic ( pages [ 1 ] ) ;
memcpy ( addr , buf + sizes [ 0 ] , sizes [ 1 ] ) ;
kunmap_atomic ( addr ) ;
2012-07-18 20:55:56 +04:00
out :
/* enable page faults to match kunmap_atomic() return conditions */
pagefault_enable ( ) ;
2012-07-03 01:15:49 +04:00
}
2012-01-10 02:51:56 +04:00
2013-12-11 06:04:36 +04:00
# endif /* CONFIG_PGTABLE_MAPPING */
2012-07-18 20:55:56 +04:00
2012-01-10 02:51:56 +04:00
static int zs_cpu_notifier ( struct notifier_block * nb , unsigned long action ,
void * pcpu )
{
2012-07-18 20:55:56 +04:00
int ret , cpu = ( long ) pcpu ;
2012-01-10 02:51:56 +04:00
struct mapping_area * area ;
switch ( action ) {
case CPU_UP_PREPARE :
area = & per_cpu ( zs_map_area , cpu ) ;
2012-07-18 20:55:56 +04:00
ret = __zs_cpu_up ( area ) ;
if ( ret )
return notifier_from_errno ( ret ) ;
2012-01-10 02:51:56 +04:00
break ;
case CPU_DEAD :
case CPU_UP_CANCELED :
area = & per_cpu ( zs_map_area , cpu ) ;
2012-07-18 20:55:56 +04:00
__zs_cpu_down ( area ) ;
2012-01-10 02:51:56 +04:00
break ;
}
return NOTIFY_OK ;
}
static struct notifier_block zs_cpu_nb = {
. notifier_call = zs_cpu_notifier
} ;
2014-12-13 03:56:56 +03:00
static int zs_register_cpu_notifier ( void )
2012-01-10 02:51:56 +04:00
{
2014-12-13 03:56:56 +03:00
int cpu , uninitialized_var ( ret ) ;
2012-01-10 02:51:56 +04:00
2014-03-11 00:39:59 +04:00
cpu_notifier_register_begin ( ) ;
__register_cpu_notifier ( & zs_cpu_nb ) ;
2012-01-10 02:51:56 +04:00
for_each_online_cpu ( cpu ) {
ret = zs_cpu_notifier ( NULL , CPU_UP_PREPARE , ( void * ) ( long ) cpu ) ;
2014-12-13 03:56:56 +03:00
if ( notifier_to_errno ( ret ) )
break ;
2012-01-10 02:51:56 +04:00
}
2014-03-11 00:39:59 +04:00
cpu_notifier_register_done ( ) ;
2014-12-13 03:56:56 +03:00
return notifier_to_errno ( ret ) ;
}
2014-03-11 00:39:59 +04:00
2014-12-19 03:17:40 +03:00
static void zs_unregister_cpu_notifier ( void )
2014-12-13 03:57:01 +03:00
{
2014-12-19 03:17:40 +03:00
int cpu ;
2014-12-13 03:57:01 +03:00
2014-12-19 03:17:40 +03:00
cpu_notifier_register_begin ( ) ;
2014-12-13 03:57:01 +03:00
2014-12-19 03:17:40 +03:00
for_each_online_cpu ( cpu )
zs_cpu_notifier ( NULL , CPU_DEAD , ( void * ) ( long ) cpu ) ;
__unregister_cpu_notifier ( & zs_cpu_nb ) ;
2014-12-13 03:57:01 +03:00
2014-12-19 03:17:40 +03:00
cpu_notifier_register_done ( ) ;
2014-12-13 03:56:56 +03:00
}
2014-12-19 03:17:40 +03:00
static void init_zs_size_classes ( void )
2014-12-13 03:56:56 +03:00
{
2014-12-19 03:17:40 +03:00
int nr ;
2014-08-07 03:08:38 +04:00
2014-12-19 03:17:40 +03:00
nr = ( ZS_MAX_ALLOC_SIZE - ZS_MIN_ALLOC_SIZE ) / ZS_SIZE_CLASS_DELTA + 1 ;
if ( ( ZS_MAX_ALLOC_SIZE - ZS_MIN_ALLOC_SIZE ) % ZS_SIZE_CLASS_DELTA )
nr + = 1 ;
2014-12-13 03:57:01 +03:00
2014-12-19 03:17:40 +03:00
zs_size_classes = nr ;
2012-01-10 02:51:56 +04:00
}
zsmalloc: merge size_class to reduce fragmentation
zsmalloc has many size_classes to reduce fragmentation and they are in 16
bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096. And,
zsmalloc has constraint that each zspage has 4 pages at maximum.
In this situation, we can see interesting aspect. Let's think about
size_class for 1488, 1472, ..., 1376. To prevent external fragmentation,
they uses 4 pages per zspage and so all they can contain 11 objects at
maximum.
16384 (4096 * 4) = 1488 * 11 + remains
16384 (4096 * 4) = 1472 * 11 + remains
16384 (4096 * 4) = ...
16384 (4096 * 4) = 1376 * 11 + remains
It means that they have same characteristics and classification between
them isn't needed. If we use one size_class for them, we can reduce
fragementation and save some memory since both the 1488 and 1472 sized
classes can only fit 11 objects into 4 pages, and an object that's 1472
bytes can fit into an object that's 1488 bytes, merging these classes to
always use objects that are 1488 bytes will reduce the total number of
size classes. And reducing the total number of size classes reduces
overall fragmentation, because a wider range of compressed pages can fit
into a single size class, leaving less unused objects in each size class.
For this purpose, this patch implement size_class merging. If there is
size_class that have same pages_per_zspage and same number of objects per
zspage with previous size_class, we don't create new size_class. Instead,
we use previous, same characteristic size_class. With this way, above
example sizes (1488, 1472, ..., 1376) use just one size_class so we can
get much more memory utilization.
Below is result of my simple test.
TEST ENV: EXT4 on zram, mount with discard option WORKLOAD: untar kernel
source code, remove directory in descending order in size. (drivers arch
fs sound include net Documentation firmware kernel tools)
Each line represents orig_data_size, compr_data_size, mem_used_total,
fragmentation overhead (mem_used - compr_data_size) and overhead ratio
(overhead to compr_data_size), respectively, after untar and remove
operation is executed.
* untar-nomerge.out
orig_size compr_size used_size overhead overhead_ratio
525.88MB 199.16MB 210.23MB 11.08MB 5.56%
288.32MB 97.43MB 105.63MB 8.20MB 8.41%
177.32MB 61.12MB 69.40MB 8.28MB 13.55%
146.47MB 47.32MB 56.10MB 8.78MB 18.55%
124.16MB 38.85MB 48.41MB 9.55MB 24.58%
103.93MB 31.68MB 40.93MB 9.25MB 29.21%
84.34MB 22.86MB 32.72MB 9.86MB 43.13%
66.87MB 14.83MB 23.83MB 9.00MB 60.70%
60.67MB 11.11MB 18.60MB 7.49MB 67.48%
55.86MB 8.83MB 16.61MB 7.77MB 88.03%
53.32MB 8.01MB 15.32MB 7.31MB 91.24%
* untar-merge.out
orig_size compr_size used_size overhead overhead_ratio
526.23MB 199.18MB 209.81MB 10.64MB 5.34%
288.68MB 97.45MB 104.08MB 6.63MB 6.80%
177.68MB 61.14MB 66.93MB 5.79MB 9.47%
146.83MB 47.34MB 52.79MB 5.45MB 11.51%
124.52MB 38.87MB 44.30MB 5.43MB 13.96%
104.29MB 31.70MB 36.83MB 5.13MB 16.19%
84.70MB 22.88MB 27.92MB 5.04MB 22.04%
67.11MB 14.83MB 19.26MB 4.43MB 29.86%
60.82MB 11.10MB 14.90MB 3.79MB 34.17%
55.90MB 8.82MB 12.61MB 3.79MB 42.97%
53.32MB 8.01MB 11.73MB 3.73MB 46.53%
As you can see above result, merged one has better utilization (overhead
ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: <juno.choi@lge.com>
Cc: "seungho1.park" <seungho1.park@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-13 03:56:44 +03:00
static unsigned int get_maxobj_per_zspage ( int size , int pages_per_zspage )
{
return pages_per_zspage * PAGE_SIZE / size ;
}
static bool can_merge ( struct size_class * prev , int size , int pages_per_zspage )
{
if ( prev - > pages_per_zspage ! = pages_per_zspage )
return false ;
if ( get_maxobj_per_zspage ( prev - > size , prev - > pages_per_zspage )
! = get_maxobj_per_zspage ( size , pages_per_zspage ) )
return false ;
return true ;
}
2014-12-19 03:17:40 +03:00
unsigned long zs_get_total_pages ( struct zs_pool * pool )
{
return atomic_long_read ( & pool - > pages_allocated ) ;
}
EXPORT_SYMBOL_GPL ( zs_get_total_pages ) ;
2013-01-05 00:14:00 +04:00
/**
2014-12-19 03:17:40 +03:00
* zs_map_object - get address of allocated object from handle .
* @ pool : pool from which the object was allocated
* @ handle : handle returned from zs_malloc
2013-01-05 00:14:00 +04:00
*
2014-12-19 03:17:40 +03:00
* Before using an object allocated from zs_malloc , it must be mapped using
* this function . When done with the object , it must be unmapped using
* zs_unmap_object .
2013-01-05 00:14:00 +04:00
*
2014-12-19 03:17:40 +03:00
* Only one object can be mapped per cpu at a time . There is no protection
* against nested mappings .
*
* This function returns with preemption and page faults disabled .
2013-01-05 00:14:00 +04:00
*/
2014-12-19 03:17:40 +03:00
void * zs_map_object ( struct zs_pool * pool , unsigned long handle ,
enum zs_mapmode mm )
2012-01-10 02:51:56 +04:00
{
2014-12-19 03:17:40 +03:00
struct page * page ;
unsigned long obj_idx , off ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
unsigned int class_idx ;
enum fullness_group fg ;
struct size_class * class ;
struct mapping_area * area ;
struct page * pages [ 2 ] ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
BUG_ON ( ! handle ) ;
2014-12-13 03:57:01 +03:00
zsmalloc: merge size_class to reduce fragmentation
zsmalloc has many size_classes to reduce fragmentation and they are in 16
bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096. And,
zsmalloc has constraint that each zspage has 4 pages at maximum.
In this situation, we can see interesting aspect. Let's think about
size_class for 1488, 1472, ..., 1376. To prevent external fragmentation,
they uses 4 pages per zspage and so all they can contain 11 objects at
maximum.
16384 (4096 * 4) = 1488 * 11 + remains
16384 (4096 * 4) = 1472 * 11 + remains
16384 (4096 * 4) = ...
16384 (4096 * 4) = 1376 * 11 + remains
It means that they have same characteristics and classification between
them isn't needed. If we use one size_class for them, we can reduce
fragementation and save some memory since both the 1488 and 1472 sized
classes can only fit 11 objects into 4 pages, and an object that's 1472
bytes can fit into an object that's 1488 bytes, merging these classes to
always use objects that are 1488 bytes will reduce the total number of
size classes. And reducing the total number of size classes reduces
overall fragmentation, because a wider range of compressed pages can fit
into a single size class, leaving less unused objects in each size class.
For this purpose, this patch implement size_class merging. If there is
size_class that have same pages_per_zspage and same number of objects per
zspage with previous size_class, we don't create new size_class. Instead,
we use previous, same characteristic size_class. With this way, above
example sizes (1488, 1472, ..., 1376) use just one size_class so we can
get much more memory utilization.
Below is result of my simple test.
TEST ENV: EXT4 on zram, mount with discard option WORKLOAD: untar kernel
source code, remove directory in descending order in size. (drivers arch
fs sound include net Documentation firmware kernel tools)
Each line represents orig_data_size, compr_data_size, mem_used_total,
fragmentation overhead (mem_used - compr_data_size) and overhead ratio
(overhead to compr_data_size), respectively, after untar and remove
operation is executed.
* untar-nomerge.out
orig_size compr_size used_size overhead overhead_ratio
525.88MB 199.16MB 210.23MB 11.08MB 5.56%
288.32MB 97.43MB 105.63MB 8.20MB 8.41%
177.32MB 61.12MB 69.40MB 8.28MB 13.55%
146.47MB 47.32MB 56.10MB 8.78MB 18.55%
124.16MB 38.85MB 48.41MB 9.55MB 24.58%
103.93MB 31.68MB 40.93MB 9.25MB 29.21%
84.34MB 22.86MB 32.72MB 9.86MB 43.13%
66.87MB 14.83MB 23.83MB 9.00MB 60.70%
60.67MB 11.11MB 18.60MB 7.49MB 67.48%
55.86MB 8.83MB 16.61MB 7.77MB 88.03%
53.32MB 8.01MB 15.32MB 7.31MB 91.24%
* untar-merge.out
orig_size compr_size used_size overhead overhead_ratio
526.23MB 199.18MB 209.81MB 10.64MB 5.34%
288.68MB 97.45MB 104.08MB 6.63MB 6.80%
177.68MB 61.14MB 66.93MB 5.79MB 9.47%
146.83MB 47.34MB 52.79MB 5.45MB 11.51%
124.52MB 38.87MB 44.30MB 5.43MB 13.96%
104.29MB 31.70MB 36.83MB 5.13MB 16.19%
84.70MB 22.88MB 27.92MB 5.04MB 22.04%
67.11MB 14.83MB 19.26MB 4.43MB 29.86%
60.82MB 11.10MB 14.90MB 3.79MB 34.17%
55.90MB 8.82MB 12.61MB 3.79MB 42.97%
53.32MB 8.01MB 11.73MB 3.73MB 46.53%
As you can see above result, merged one has better utilization (overhead
ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: <juno.choi@lge.com>
Cc: "seungho1.park" <seungho1.park@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-13 03:56:44 +03:00
/*
2014-12-19 03:17:40 +03:00
* Because we use per - cpu mapping areas shared among the
* pools / users , we can ' t allow mapping in interrupt context
* because it can corrupt another users mappings .
zsmalloc: merge size_class to reduce fragmentation
zsmalloc has many size_classes to reduce fragmentation and they are in 16
bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096. And,
zsmalloc has constraint that each zspage has 4 pages at maximum.
In this situation, we can see interesting aspect. Let's think about
size_class for 1488, 1472, ..., 1376. To prevent external fragmentation,
they uses 4 pages per zspage and so all they can contain 11 objects at
maximum.
16384 (4096 * 4) = 1488 * 11 + remains
16384 (4096 * 4) = 1472 * 11 + remains
16384 (4096 * 4) = ...
16384 (4096 * 4) = 1376 * 11 + remains
It means that they have same characteristics and classification between
them isn't needed. If we use one size_class for them, we can reduce
fragementation and save some memory since both the 1488 and 1472 sized
classes can only fit 11 objects into 4 pages, and an object that's 1472
bytes can fit into an object that's 1488 bytes, merging these classes to
always use objects that are 1488 bytes will reduce the total number of
size classes. And reducing the total number of size classes reduces
overall fragmentation, because a wider range of compressed pages can fit
into a single size class, leaving less unused objects in each size class.
For this purpose, this patch implement size_class merging. If there is
size_class that have same pages_per_zspage and same number of objects per
zspage with previous size_class, we don't create new size_class. Instead,
we use previous, same characteristic size_class. With this way, above
example sizes (1488, 1472, ..., 1376) use just one size_class so we can
get much more memory utilization.
Below is result of my simple test.
TEST ENV: EXT4 on zram, mount with discard option WORKLOAD: untar kernel
source code, remove directory in descending order in size. (drivers arch
fs sound include net Documentation firmware kernel tools)
Each line represents orig_data_size, compr_data_size, mem_used_total,
fragmentation overhead (mem_used - compr_data_size) and overhead ratio
(overhead to compr_data_size), respectively, after untar and remove
operation is executed.
* untar-nomerge.out
orig_size compr_size used_size overhead overhead_ratio
525.88MB 199.16MB 210.23MB 11.08MB 5.56%
288.32MB 97.43MB 105.63MB 8.20MB 8.41%
177.32MB 61.12MB 69.40MB 8.28MB 13.55%
146.47MB 47.32MB 56.10MB 8.78MB 18.55%
124.16MB 38.85MB 48.41MB 9.55MB 24.58%
103.93MB 31.68MB 40.93MB 9.25MB 29.21%
84.34MB 22.86MB 32.72MB 9.86MB 43.13%
66.87MB 14.83MB 23.83MB 9.00MB 60.70%
60.67MB 11.11MB 18.60MB 7.49MB 67.48%
55.86MB 8.83MB 16.61MB 7.77MB 88.03%
53.32MB 8.01MB 15.32MB 7.31MB 91.24%
* untar-merge.out
orig_size compr_size used_size overhead overhead_ratio
526.23MB 199.18MB 209.81MB 10.64MB 5.34%
288.68MB 97.45MB 104.08MB 6.63MB 6.80%
177.68MB 61.14MB 66.93MB 5.79MB 9.47%
146.83MB 47.34MB 52.79MB 5.45MB 11.51%
124.52MB 38.87MB 44.30MB 5.43MB 13.96%
104.29MB 31.70MB 36.83MB 5.13MB 16.19%
84.70MB 22.88MB 27.92MB 5.04MB 22.04%
67.11MB 14.83MB 19.26MB 4.43MB 29.86%
60.82MB 11.10MB 14.90MB 3.79MB 34.17%
55.90MB 8.82MB 12.61MB 3.79MB 42.97%
53.32MB 8.01MB 11.73MB 3.73MB 46.53%
As you can see above result, merged one has better utilization (overhead
ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: <juno.choi@lge.com>
Cc: "seungho1.park" <seungho1.park@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-13 03:56:44 +03:00
*/
2014-12-19 03:17:40 +03:00
BUG_ON ( in_interrupt ( ) ) ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
obj_handle_to_location ( handle , & page , & obj_idx ) ;
get_zspage_mapping ( get_first_page ( page ) , & class_idx , & fg ) ;
class = pool - > size_class [ class_idx ] ;
off = obj_idx_to_offset ( page , obj_idx , class - > size ) ;
2014-12-13 03:57:07 +03:00
2014-12-19 03:17:40 +03:00
area = & get_cpu_var ( zs_map_area ) ;
area - > vm_mm = mm ;
if ( off + class - > size < = PAGE_SIZE ) {
/* this object is contained entirely within a page */
area - > vm_addr = kmap_atomic ( page ) ;
return area - > vm_addr + off ;
2012-01-10 02:51:56 +04:00
}
2014-12-19 03:17:40 +03:00
/* this object spans two pages */
pages [ 0 ] = page ;
pages [ 1 ] = get_next_page ( page ) ;
BUG_ON ( ! pages [ 1 ] ) ;
zsmalloc: merge size_class to reduce fragmentation
zsmalloc has many size_classes to reduce fragmentation and they are in 16
bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096. And,
zsmalloc has constraint that each zspage has 4 pages at maximum.
In this situation, we can see interesting aspect. Let's think about
size_class for 1488, 1472, ..., 1376. To prevent external fragmentation,
they uses 4 pages per zspage and so all they can contain 11 objects at
maximum.
16384 (4096 * 4) = 1488 * 11 + remains
16384 (4096 * 4) = 1472 * 11 + remains
16384 (4096 * 4) = ...
16384 (4096 * 4) = 1376 * 11 + remains
It means that they have same characteristics and classification between
them isn't needed. If we use one size_class for them, we can reduce
fragementation and save some memory since both the 1488 and 1472 sized
classes can only fit 11 objects into 4 pages, and an object that's 1472
bytes can fit into an object that's 1488 bytes, merging these classes to
always use objects that are 1488 bytes will reduce the total number of
size classes. And reducing the total number of size classes reduces
overall fragmentation, because a wider range of compressed pages can fit
into a single size class, leaving less unused objects in each size class.
For this purpose, this patch implement size_class merging. If there is
size_class that have same pages_per_zspage and same number of objects per
zspage with previous size_class, we don't create new size_class. Instead,
we use previous, same characteristic size_class. With this way, above
example sizes (1488, 1472, ..., 1376) use just one size_class so we can
get much more memory utilization.
Below is result of my simple test.
TEST ENV: EXT4 on zram, mount with discard option WORKLOAD: untar kernel
source code, remove directory in descending order in size. (drivers arch
fs sound include net Documentation firmware kernel tools)
Each line represents orig_data_size, compr_data_size, mem_used_total,
fragmentation overhead (mem_used - compr_data_size) and overhead ratio
(overhead to compr_data_size), respectively, after untar and remove
operation is executed.
* untar-nomerge.out
orig_size compr_size used_size overhead overhead_ratio
525.88MB 199.16MB 210.23MB 11.08MB 5.56%
288.32MB 97.43MB 105.63MB 8.20MB 8.41%
177.32MB 61.12MB 69.40MB 8.28MB 13.55%
146.47MB 47.32MB 56.10MB 8.78MB 18.55%
124.16MB 38.85MB 48.41MB 9.55MB 24.58%
103.93MB 31.68MB 40.93MB 9.25MB 29.21%
84.34MB 22.86MB 32.72MB 9.86MB 43.13%
66.87MB 14.83MB 23.83MB 9.00MB 60.70%
60.67MB 11.11MB 18.60MB 7.49MB 67.48%
55.86MB 8.83MB 16.61MB 7.77MB 88.03%
53.32MB 8.01MB 15.32MB 7.31MB 91.24%
* untar-merge.out
orig_size compr_size used_size overhead overhead_ratio
526.23MB 199.18MB 209.81MB 10.64MB 5.34%
288.68MB 97.45MB 104.08MB 6.63MB 6.80%
177.68MB 61.14MB 66.93MB 5.79MB 9.47%
146.83MB 47.34MB 52.79MB 5.45MB 11.51%
124.52MB 38.87MB 44.30MB 5.43MB 13.96%
104.29MB 31.70MB 36.83MB 5.13MB 16.19%
84.70MB 22.88MB 27.92MB 5.04MB 22.04%
67.11MB 14.83MB 19.26MB 4.43MB 29.86%
60.82MB 11.10MB 14.90MB 3.79MB 34.17%
55.90MB 8.82MB 12.61MB 3.79MB 42.97%
53.32MB 8.01MB 11.73MB 3.73MB 46.53%
As you can see above result, merged one has better utilization (overhead
ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: <juno.choi@lge.com>
Cc: "seungho1.park" <seungho1.park@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-13 03:56:44 +03:00
2014-12-19 03:17:40 +03:00
return __zs_map_object ( area , pages , off , class - > size ) ;
2012-01-10 02:51:56 +04:00
}
2014-12-19 03:17:40 +03:00
EXPORT_SYMBOL_GPL ( zs_map_object ) ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
void zs_unmap_object ( struct zs_pool * pool , unsigned long handle )
2012-01-10 02:51:56 +04:00
{
2014-12-19 03:17:40 +03:00
struct page * page ;
unsigned long obj_idx , off ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
unsigned int class_idx ;
enum fullness_group fg ;
struct size_class * class ;
struct mapping_area * area ;
zsmalloc: merge size_class to reduce fragmentation
zsmalloc has many size_classes to reduce fragmentation and they are in 16
bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096. And,
zsmalloc has constraint that each zspage has 4 pages at maximum.
In this situation, we can see interesting aspect. Let's think about
size_class for 1488, 1472, ..., 1376. To prevent external fragmentation,
they uses 4 pages per zspage and so all they can contain 11 objects at
maximum.
16384 (4096 * 4) = 1488 * 11 + remains
16384 (4096 * 4) = 1472 * 11 + remains
16384 (4096 * 4) = ...
16384 (4096 * 4) = 1376 * 11 + remains
It means that they have same characteristics and classification between
them isn't needed. If we use one size_class for them, we can reduce
fragementation and save some memory since both the 1488 and 1472 sized
classes can only fit 11 objects into 4 pages, and an object that's 1472
bytes can fit into an object that's 1488 bytes, merging these classes to
always use objects that are 1488 bytes will reduce the total number of
size classes. And reducing the total number of size classes reduces
overall fragmentation, because a wider range of compressed pages can fit
into a single size class, leaving less unused objects in each size class.
For this purpose, this patch implement size_class merging. If there is
size_class that have same pages_per_zspage and same number of objects per
zspage with previous size_class, we don't create new size_class. Instead,
we use previous, same characteristic size_class. With this way, above
example sizes (1488, 1472, ..., 1376) use just one size_class so we can
get much more memory utilization.
Below is result of my simple test.
TEST ENV: EXT4 on zram, mount with discard option WORKLOAD: untar kernel
source code, remove directory in descending order in size. (drivers arch
fs sound include net Documentation firmware kernel tools)
Each line represents orig_data_size, compr_data_size, mem_used_total,
fragmentation overhead (mem_used - compr_data_size) and overhead ratio
(overhead to compr_data_size), respectively, after untar and remove
operation is executed.
* untar-nomerge.out
orig_size compr_size used_size overhead overhead_ratio
525.88MB 199.16MB 210.23MB 11.08MB 5.56%
288.32MB 97.43MB 105.63MB 8.20MB 8.41%
177.32MB 61.12MB 69.40MB 8.28MB 13.55%
146.47MB 47.32MB 56.10MB 8.78MB 18.55%
124.16MB 38.85MB 48.41MB 9.55MB 24.58%
103.93MB 31.68MB 40.93MB 9.25MB 29.21%
84.34MB 22.86MB 32.72MB 9.86MB 43.13%
66.87MB 14.83MB 23.83MB 9.00MB 60.70%
60.67MB 11.11MB 18.60MB 7.49MB 67.48%
55.86MB 8.83MB 16.61MB 7.77MB 88.03%
53.32MB 8.01MB 15.32MB 7.31MB 91.24%
* untar-merge.out
orig_size compr_size used_size overhead overhead_ratio
526.23MB 199.18MB 209.81MB 10.64MB 5.34%
288.68MB 97.45MB 104.08MB 6.63MB 6.80%
177.68MB 61.14MB 66.93MB 5.79MB 9.47%
146.83MB 47.34MB 52.79MB 5.45MB 11.51%
124.52MB 38.87MB 44.30MB 5.43MB 13.96%
104.29MB 31.70MB 36.83MB 5.13MB 16.19%
84.70MB 22.88MB 27.92MB 5.04MB 22.04%
67.11MB 14.83MB 19.26MB 4.43MB 29.86%
60.82MB 11.10MB 14.90MB 3.79MB 34.17%
55.90MB 8.82MB 12.61MB 3.79MB 42.97%
53.32MB 8.01MB 11.73MB 3.73MB 46.53%
As you can see above result, merged one has better utilization (overhead
ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: <juno.choi@lge.com>
Cc: "seungho1.park" <seungho1.park@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-13 03:56:44 +03:00
2014-12-19 03:17:40 +03:00
BUG_ON ( ! handle ) ;
zsmalloc: merge size_class to reduce fragmentation
zsmalloc has many size_classes to reduce fragmentation and they are in 16
bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096. And,
zsmalloc has constraint that each zspage has 4 pages at maximum.
In this situation, we can see interesting aspect. Let's think about
size_class for 1488, 1472, ..., 1376. To prevent external fragmentation,
they uses 4 pages per zspage and so all they can contain 11 objects at
maximum.
16384 (4096 * 4) = 1488 * 11 + remains
16384 (4096 * 4) = 1472 * 11 + remains
16384 (4096 * 4) = ...
16384 (4096 * 4) = 1376 * 11 + remains
It means that they have same characteristics and classification between
them isn't needed. If we use one size_class for them, we can reduce
fragementation and save some memory since both the 1488 and 1472 sized
classes can only fit 11 objects into 4 pages, and an object that's 1472
bytes can fit into an object that's 1488 bytes, merging these classes to
always use objects that are 1488 bytes will reduce the total number of
size classes. And reducing the total number of size classes reduces
overall fragmentation, because a wider range of compressed pages can fit
into a single size class, leaving less unused objects in each size class.
For this purpose, this patch implement size_class merging. If there is
size_class that have same pages_per_zspage and same number of objects per
zspage with previous size_class, we don't create new size_class. Instead,
we use previous, same characteristic size_class. With this way, above
example sizes (1488, 1472, ..., 1376) use just one size_class so we can
get much more memory utilization.
Below is result of my simple test.
TEST ENV: EXT4 on zram, mount with discard option WORKLOAD: untar kernel
source code, remove directory in descending order in size. (drivers arch
fs sound include net Documentation firmware kernel tools)
Each line represents orig_data_size, compr_data_size, mem_used_total,
fragmentation overhead (mem_used - compr_data_size) and overhead ratio
(overhead to compr_data_size), respectively, after untar and remove
operation is executed.
* untar-nomerge.out
orig_size compr_size used_size overhead overhead_ratio
525.88MB 199.16MB 210.23MB 11.08MB 5.56%
288.32MB 97.43MB 105.63MB 8.20MB 8.41%
177.32MB 61.12MB 69.40MB 8.28MB 13.55%
146.47MB 47.32MB 56.10MB 8.78MB 18.55%
124.16MB 38.85MB 48.41MB 9.55MB 24.58%
103.93MB 31.68MB 40.93MB 9.25MB 29.21%
84.34MB 22.86MB 32.72MB 9.86MB 43.13%
66.87MB 14.83MB 23.83MB 9.00MB 60.70%
60.67MB 11.11MB 18.60MB 7.49MB 67.48%
55.86MB 8.83MB 16.61MB 7.77MB 88.03%
53.32MB 8.01MB 15.32MB 7.31MB 91.24%
* untar-merge.out
orig_size compr_size used_size overhead overhead_ratio
526.23MB 199.18MB 209.81MB 10.64MB 5.34%
288.68MB 97.45MB 104.08MB 6.63MB 6.80%
177.68MB 61.14MB 66.93MB 5.79MB 9.47%
146.83MB 47.34MB 52.79MB 5.45MB 11.51%
124.52MB 38.87MB 44.30MB 5.43MB 13.96%
104.29MB 31.70MB 36.83MB 5.13MB 16.19%
84.70MB 22.88MB 27.92MB 5.04MB 22.04%
67.11MB 14.83MB 19.26MB 4.43MB 29.86%
60.82MB 11.10MB 14.90MB 3.79MB 34.17%
55.90MB 8.82MB 12.61MB 3.79MB 42.97%
53.32MB 8.01MB 11.73MB 3.73MB 46.53%
As you can see above result, merged one has better utilization (overhead
ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: <juno.choi@lge.com>
Cc: "seungho1.park" <seungho1.park@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-13 03:56:44 +03:00
2014-12-19 03:17:40 +03:00
obj_handle_to_location ( handle , & page , & obj_idx ) ;
get_zspage_mapping ( get_first_page ( page ) , & class_idx , & fg ) ;
class = pool - > size_class [ class_idx ] ;
off = obj_idx_to_offset ( page , obj_idx , class - > size ) ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
area = this_cpu_ptr ( & zs_map_area ) ;
if ( off + class - > size < = PAGE_SIZE )
kunmap_atomic ( area - > vm_addr ) ;
else {
struct page * pages [ 2 ] ;
2014-12-13 03:57:01 +03:00
2014-12-19 03:17:40 +03:00
pages [ 0 ] = page ;
pages [ 1 ] = get_next_page ( page ) ;
BUG_ON ( ! pages [ 1 ] ) ;
__zs_unmap_object ( area , pages , off , class - > size ) ;
}
put_cpu_var ( zs_map_area ) ;
2012-01-10 02:51:56 +04:00
}
2014-12-19 03:17:40 +03:00
EXPORT_SYMBOL_GPL ( zs_unmap_object ) ;
2012-01-10 02:51:56 +04:00
/**
* zs_malloc - Allocate block of given size from pool .
* @ pool : pool to allocate from
* @ size : size of block to allocate
*
2012-05-03 10:40:40 +04:00
* On success , handle to the allocated object is returned ,
2012-06-08 10:39:25 +04:00
* otherwise 0.
2012-01-10 02:51:56 +04:00
* Allocation requests with size > ZS_MAX_ALLOC_SIZE will fail .
*/
2012-06-08 10:39:25 +04:00
unsigned long zs_malloc ( struct zs_pool * pool , size_t size )
2012-01-10 02:51:56 +04:00
{
2012-06-08 10:39:25 +04:00
unsigned long obj ;
2012-01-10 02:51:56 +04:00
struct link_free * link ;
struct size_class * class ;
2014-12-13 03:56:58 +03:00
void * vaddr ;
2012-01-10 02:51:56 +04:00
struct page * first_page , * m_page ;
unsigned long m_objidx , m_offset ;
if ( unlikely ( ! size | | size > ZS_MAX_ALLOC_SIZE ) )
2012-06-08 10:39:25 +04:00
return 0 ;
2012-01-10 02:51:56 +04:00
zsmalloc: merge size_class to reduce fragmentation
zsmalloc has many size_classes to reduce fragmentation and they are in 16
bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096. And,
zsmalloc has constraint that each zspage has 4 pages at maximum.
In this situation, we can see interesting aspect. Let's think about
size_class for 1488, 1472, ..., 1376. To prevent external fragmentation,
they uses 4 pages per zspage and so all they can contain 11 objects at
maximum.
16384 (4096 * 4) = 1488 * 11 + remains
16384 (4096 * 4) = 1472 * 11 + remains
16384 (4096 * 4) = ...
16384 (4096 * 4) = 1376 * 11 + remains
It means that they have same characteristics and classification between
them isn't needed. If we use one size_class for them, we can reduce
fragementation and save some memory since both the 1488 and 1472 sized
classes can only fit 11 objects into 4 pages, and an object that's 1472
bytes can fit into an object that's 1488 bytes, merging these classes to
always use objects that are 1488 bytes will reduce the total number of
size classes. And reducing the total number of size classes reduces
overall fragmentation, because a wider range of compressed pages can fit
into a single size class, leaving less unused objects in each size class.
For this purpose, this patch implement size_class merging. If there is
size_class that have same pages_per_zspage and same number of objects per
zspage with previous size_class, we don't create new size_class. Instead,
we use previous, same characteristic size_class. With this way, above
example sizes (1488, 1472, ..., 1376) use just one size_class so we can
get much more memory utilization.
Below is result of my simple test.
TEST ENV: EXT4 on zram, mount with discard option WORKLOAD: untar kernel
source code, remove directory in descending order in size. (drivers arch
fs sound include net Documentation firmware kernel tools)
Each line represents orig_data_size, compr_data_size, mem_used_total,
fragmentation overhead (mem_used - compr_data_size) and overhead ratio
(overhead to compr_data_size), respectively, after untar and remove
operation is executed.
* untar-nomerge.out
orig_size compr_size used_size overhead overhead_ratio
525.88MB 199.16MB 210.23MB 11.08MB 5.56%
288.32MB 97.43MB 105.63MB 8.20MB 8.41%
177.32MB 61.12MB 69.40MB 8.28MB 13.55%
146.47MB 47.32MB 56.10MB 8.78MB 18.55%
124.16MB 38.85MB 48.41MB 9.55MB 24.58%
103.93MB 31.68MB 40.93MB 9.25MB 29.21%
84.34MB 22.86MB 32.72MB 9.86MB 43.13%
66.87MB 14.83MB 23.83MB 9.00MB 60.70%
60.67MB 11.11MB 18.60MB 7.49MB 67.48%
55.86MB 8.83MB 16.61MB 7.77MB 88.03%
53.32MB 8.01MB 15.32MB 7.31MB 91.24%
* untar-merge.out
orig_size compr_size used_size overhead overhead_ratio
526.23MB 199.18MB 209.81MB 10.64MB 5.34%
288.68MB 97.45MB 104.08MB 6.63MB 6.80%
177.68MB 61.14MB 66.93MB 5.79MB 9.47%
146.83MB 47.34MB 52.79MB 5.45MB 11.51%
124.52MB 38.87MB 44.30MB 5.43MB 13.96%
104.29MB 31.70MB 36.83MB 5.13MB 16.19%
84.70MB 22.88MB 27.92MB 5.04MB 22.04%
67.11MB 14.83MB 19.26MB 4.43MB 29.86%
60.82MB 11.10MB 14.90MB 3.79MB 34.17%
55.90MB 8.82MB 12.61MB 3.79MB 42.97%
53.32MB 8.01MB 11.73MB 3.73MB 46.53%
As you can see above result, merged one has better utilization (overhead
ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: <juno.choi@lge.com>
Cc: "seungho1.park" <seungho1.park@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-13 03:56:44 +03:00
class = pool - > size_class [ get_size_class_index ( size ) ] ;
2012-01-10 02:51:56 +04:00
spin_lock ( & class - > lock ) ;
first_page = find_get_zspage ( class ) ;
if ( ! first_page ) {
spin_unlock ( & class - > lock ) ;
first_page = alloc_zspage ( class , pool - > flags ) ;
if ( unlikely ( ! first_page ) )
2012-06-08 10:39:25 +04:00
return 0 ;
2012-01-10 02:51:56 +04:00
set_zspage_mapping ( first_page , class - > index , ZS_EMPTY ) ;
zsmalloc: move pages_allocated to zs_pool
Currently, zram has no feature to limit memory so theoretically zram can
deplete system memory. Users have asked for a limit several times as even
without exhaustion zram makes it hard to control memory usage of the
platform. This patchset adds the feature.
Patch 1 makes zs_get_total_size_bytes faster because it would be used
frequently in later patches for the new feature.
Patch 2 changes zs_get_total_size_bytes's return unit from bytes to page
so that zsmalloc doesn't need unnecessary operation(ie, << PAGE_SHIFT).
Patch 3 adds new feature. I added the feature into zram layer, not
zsmalloc because limiation is zram's requirement, not zsmalloc so any
other user using zsmalloc(ie, zpool) shouldn't affected by unnecessary
branch of zsmalloc. In future, if every users of zsmalloc want the
feature, then, we could move the feature from client side to zsmalloc
easily but vice versa would be painful.
Patch 4 adds news facility to report maximum memory usage of zram so that
this avoids user polling frequently via /sys/block/zram0/ mem_used_total
and ensures transient max are not missed.
This patch (of 4):
pages_allocated has counted in size_class structure and when user of
zsmalloc want to see total_size_bytes, it should gather all of count from
each size_class to report the sum.
It's not bad if user don't see the value often but if user start to see
the value frequently, it would be not a good deal for performance pov.
This patch moves the count from size_class to zs_pool so it could reduce
memory footprint (from [255 * 8byte] to [sizeof(atomic_long_t)]).
Signed-off-by: Minchan Kim <minchan@kernel.org>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: <juno.choi@lge.com>
Cc: <seungho1.park@lge.com>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Seth Jennings <sjennings@variantweb.net>
Reviewed-by: David Horner <ds2horner@gmail.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-10-10 02:29:48 +04:00
atomic_long_add ( class - > pages_per_zspage ,
& pool - > pages_allocated ) ;
2012-01-10 02:51:56 +04:00
spin_lock ( & class - > lock ) ;
}
2012-06-08 10:39:25 +04:00
obj = ( unsigned long ) first_page - > freelist ;
2012-01-10 02:51:56 +04:00
obj_handle_to_location ( obj , & m_page , & m_objidx ) ;
m_offset = obj_idx_to_offset ( m_page , m_objidx , class - > size ) ;
2014-12-13 03:56:58 +03:00
vaddr = kmap_atomic ( m_page ) ;
link = ( struct link_free * ) vaddr + m_offset / sizeof ( * link ) ;
2012-01-10 02:51:56 +04:00
first_page - > freelist = link - > next ;
memset ( link , POISON_INUSE , sizeof ( * link ) ) ;
2014-12-13 03:56:58 +03:00
kunmap_atomic ( vaddr ) ;
2012-01-10 02:51:56 +04:00
first_page - > inuse + + ;
/* Now move the zspage to another fullness group, if required */
fix_fullness_group ( pool , first_page ) ;
spin_unlock ( & class - > lock ) ;
return obj ;
}
EXPORT_SYMBOL_GPL ( zs_malloc ) ;
2012-06-08 10:39:25 +04:00
void zs_free ( struct zs_pool * pool , unsigned long obj )
2012-01-10 02:51:56 +04:00
{
struct link_free * link ;
struct page * first_page , * f_page ;
unsigned long f_objidx , f_offset ;
2014-12-13 03:56:58 +03:00
void * vaddr ;
2012-01-10 02:51:56 +04:00
int class_idx ;
struct size_class * class ;
enum fullness_group fullness ;
if ( unlikely ( ! obj ) )
return ;
obj_handle_to_location ( obj , & f_page , & f_objidx ) ;
first_page = get_first_page ( f_page ) ;
get_zspage_mapping ( first_page , & class_idx , & fullness ) ;
zsmalloc: merge size_class to reduce fragmentation
zsmalloc has many size_classes to reduce fragmentation and they are in 16
bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096. And,
zsmalloc has constraint that each zspage has 4 pages at maximum.
In this situation, we can see interesting aspect. Let's think about
size_class for 1488, 1472, ..., 1376. To prevent external fragmentation,
they uses 4 pages per zspage and so all they can contain 11 objects at
maximum.
16384 (4096 * 4) = 1488 * 11 + remains
16384 (4096 * 4) = 1472 * 11 + remains
16384 (4096 * 4) = ...
16384 (4096 * 4) = 1376 * 11 + remains
It means that they have same characteristics and classification between
them isn't needed. If we use one size_class for them, we can reduce
fragementation and save some memory since both the 1488 and 1472 sized
classes can only fit 11 objects into 4 pages, and an object that's 1472
bytes can fit into an object that's 1488 bytes, merging these classes to
always use objects that are 1488 bytes will reduce the total number of
size classes. And reducing the total number of size classes reduces
overall fragmentation, because a wider range of compressed pages can fit
into a single size class, leaving less unused objects in each size class.
For this purpose, this patch implement size_class merging. If there is
size_class that have same pages_per_zspage and same number of objects per
zspage with previous size_class, we don't create new size_class. Instead,
we use previous, same characteristic size_class. With this way, above
example sizes (1488, 1472, ..., 1376) use just one size_class so we can
get much more memory utilization.
Below is result of my simple test.
TEST ENV: EXT4 on zram, mount with discard option WORKLOAD: untar kernel
source code, remove directory in descending order in size. (drivers arch
fs sound include net Documentation firmware kernel tools)
Each line represents orig_data_size, compr_data_size, mem_used_total,
fragmentation overhead (mem_used - compr_data_size) and overhead ratio
(overhead to compr_data_size), respectively, after untar and remove
operation is executed.
* untar-nomerge.out
orig_size compr_size used_size overhead overhead_ratio
525.88MB 199.16MB 210.23MB 11.08MB 5.56%
288.32MB 97.43MB 105.63MB 8.20MB 8.41%
177.32MB 61.12MB 69.40MB 8.28MB 13.55%
146.47MB 47.32MB 56.10MB 8.78MB 18.55%
124.16MB 38.85MB 48.41MB 9.55MB 24.58%
103.93MB 31.68MB 40.93MB 9.25MB 29.21%
84.34MB 22.86MB 32.72MB 9.86MB 43.13%
66.87MB 14.83MB 23.83MB 9.00MB 60.70%
60.67MB 11.11MB 18.60MB 7.49MB 67.48%
55.86MB 8.83MB 16.61MB 7.77MB 88.03%
53.32MB 8.01MB 15.32MB 7.31MB 91.24%
* untar-merge.out
orig_size compr_size used_size overhead overhead_ratio
526.23MB 199.18MB 209.81MB 10.64MB 5.34%
288.68MB 97.45MB 104.08MB 6.63MB 6.80%
177.68MB 61.14MB 66.93MB 5.79MB 9.47%
146.83MB 47.34MB 52.79MB 5.45MB 11.51%
124.52MB 38.87MB 44.30MB 5.43MB 13.96%
104.29MB 31.70MB 36.83MB 5.13MB 16.19%
84.70MB 22.88MB 27.92MB 5.04MB 22.04%
67.11MB 14.83MB 19.26MB 4.43MB 29.86%
60.82MB 11.10MB 14.90MB 3.79MB 34.17%
55.90MB 8.82MB 12.61MB 3.79MB 42.97%
53.32MB 8.01MB 11.73MB 3.73MB 46.53%
As you can see above result, merged one has better utilization (overhead
ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: <juno.choi@lge.com>
Cc: "seungho1.park" <seungho1.park@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-13 03:56:44 +03:00
class = pool - > size_class [ class_idx ] ;
2012-01-10 02:51:56 +04:00
f_offset = obj_idx_to_offset ( f_page , f_objidx , class - > size ) ;
spin_lock ( & class - > lock ) ;
/* Insert this object in containing zspage's freelist */
2014-12-13 03:56:58 +03:00
vaddr = kmap_atomic ( f_page ) ;
link = ( struct link_free * ) ( vaddr + f_offset ) ;
2012-01-10 02:51:56 +04:00
link - > next = first_page - > freelist ;
2014-12-13 03:56:58 +03:00
kunmap_atomic ( vaddr ) ;
2012-06-08 10:39:25 +04:00
first_page - > freelist = ( void * ) obj ;
2012-01-10 02:51:56 +04:00
first_page - > inuse - - ;
fullness = fix_fullness_group ( pool , first_page ) ;
spin_unlock ( & class - > lock ) ;
zsmalloc: move pages_allocated to zs_pool
Currently, zram has no feature to limit memory so theoretically zram can
deplete system memory. Users have asked for a limit several times as even
without exhaustion zram makes it hard to control memory usage of the
platform. This patchset adds the feature.
Patch 1 makes zs_get_total_size_bytes faster because it would be used
frequently in later patches for the new feature.
Patch 2 changes zs_get_total_size_bytes's return unit from bytes to page
so that zsmalloc doesn't need unnecessary operation(ie, << PAGE_SHIFT).
Patch 3 adds new feature. I added the feature into zram layer, not
zsmalloc because limiation is zram's requirement, not zsmalloc so any
other user using zsmalloc(ie, zpool) shouldn't affected by unnecessary
branch of zsmalloc. In future, if every users of zsmalloc want the
feature, then, we could move the feature from client side to zsmalloc
easily but vice versa would be painful.
Patch 4 adds news facility to report maximum memory usage of zram so that
this avoids user polling frequently via /sys/block/zram0/ mem_used_total
and ensures transient max are not missed.
This patch (of 4):
pages_allocated has counted in size_class structure and when user of
zsmalloc want to see total_size_bytes, it should gather all of count from
each size_class to report the sum.
It's not bad if user don't see the value often but if user start to see
the value frequently, it would be not a good deal for performance pov.
This patch moves the count from size_class to zs_pool so it could reduce
memory footprint (from [255 * 8byte] to [sizeof(atomic_long_t)]).
Signed-off-by: Minchan Kim <minchan@kernel.org>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: <juno.choi@lge.com>
Cc: <seungho1.park@lge.com>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Seth Jennings <sjennings@variantweb.net>
Reviewed-by: David Horner <ds2horner@gmail.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-10-10 02:29:48 +04:00
if ( fullness = = ZS_EMPTY ) {
atomic_long_sub ( class - > pages_per_zspage ,
& pool - > pages_allocated ) ;
2012-01-10 02:51:56 +04:00
free_zspage ( first_page ) ;
zsmalloc: move pages_allocated to zs_pool
Currently, zram has no feature to limit memory so theoretically zram can
deplete system memory. Users have asked for a limit several times as even
without exhaustion zram makes it hard to control memory usage of the
platform. This patchset adds the feature.
Patch 1 makes zs_get_total_size_bytes faster because it would be used
frequently in later patches for the new feature.
Patch 2 changes zs_get_total_size_bytes's return unit from bytes to page
so that zsmalloc doesn't need unnecessary operation(ie, << PAGE_SHIFT).
Patch 3 adds new feature. I added the feature into zram layer, not
zsmalloc because limiation is zram's requirement, not zsmalloc so any
other user using zsmalloc(ie, zpool) shouldn't affected by unnecessary
branch of zsmalloc. In future, if every users of zsmalloc want the
feature, then, we could move the feature from client side to zsmalloc
easily but vice versa would be painful.
Patch 4 adds news facility to report maximum memory usage of zram so that
this avoids user polling frequently via /sys/block/zram0/ mem_used_total
and ensures transient max are not missed.
This patch (of 4):
pages_allocated has counted in size_class structure and when user of
zsmalloc want to see total_size_bytes, it should gather all of count from
each size_class to report the sum.
It's not bad if user don't see the value often but if user start to see
the value frequently, it would be not a good deal for performance pov.
This patch moves the count from size_class to zs_pool so it could reduce
memory footprint (from [255 * 8byte] to [sizeof(atomic_long_t)]).
Signed-off-by: Minchan Kim <minchan@kernel.org>
Reviewed-by: Dan Streetman <ddstreet@ieee.org>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: <juno.choi@lge.com>
Cc: <seungho1.park@lge.com>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Seth Jennings <sjennings@variantweb.net>
Reviewed-by: David Horner <ds2horner@gmail.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-10-10 02:29:48 +04:00
}
2012-01-10 02:51:56 +04:00
}
EXPORT_SYMBOL_GPL ( zs_free ) ;
2012-05-03 10:40:40 +04:00
/**
2014-12-19 03:17:40 +03:00
* zs_create_pool - Creates an allocation pool to work from .
* @ flags : allocation flags used to allocate pool metadata
2012-07-03 01:15:51 +04:00
*
2014-12-19 03:17:40 +03:00
* This function must be called before anything when using
* the zsmalloc allocator .
2012-07-03 01:15:51 +04:00
*
2014-12-19 03:17:40 +03:00
* On success , a pointer to the newly created pool is returned ,
* otherwise NULL .
2013-05-20 23:18:14 +04:00
*/
2014-12-19 03:17:40 +03:00
struct zs_pool * zs_create_pool ( gfp_t flags )
2012-01-10 02:51:56 +04:00
{
2014-12-19 03:17:40 +03:00
int i ;
struct zs_pool * pool ;
struct size_class * prev_class = NULL ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
pool = kzalloc ( sizeof ( * pool ) , GFP_KERNEL ) ;
if ( ! pool )
return NULL ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
pool - > size_class = kcalloc ( zs_size_classes , sizeof ( struct size_class * ) ,
GFP_KERNEL ) ;
if ( ! pool - > size_class ) {
kfree ( pool ) ;
return NULL ;
}
2012-01-10 02:51:56 +04:00
2012-07-18 20:55:55 +04:00
/*
2014-12-19 03:17:40 +03:00
* Iterate reversly , because , size of size_class that we want to use
* for merging should be larger or equal to current size .
2012-07-18 20:55:55 +04:00
*/
2014-12-19 03:17:40 +03:00
for ( i = zs_size_classes - 1 ; i > = 0 ; i - - ) {
int size ;
int pages_per_zspage ;
struct size_class * class ;
2012-07-18 20:55:55 +04:00
2014-12-19 03:17:40 +03:00
size = ZS_MIN_ALLOC_SIZE + i * ZS_SIZE_CLASS_DELTA ;
if ( size > ZS_MAX_ALLOC_SIZE )
size = ZS_MAX_ALLOC_SIZE ;
pages_per_zspage = get_pages_per_zspage ( size ) ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
/*
* size_class is used for normal zsmalloc operation such
* as alloc / free for that size . Although it is natural that we
* have one size_class for each size , there is a chance that we
* can get more memory utilization if we use one size_class for
* many different sizes whose size_class have same
* characteristics . So , we makes size_class point to
* previous size_class if possible .
*/
if ( prev_class ) {
if ( can_merge ( prev_class , size , pages_per_zspage ) ) {
pool - > size_class [ i ] = prev_class ;
continue ;
}
}
class = kzalloc ( sizeof ( struct size_class ) , GFP_KERNEL ) ;
if ( ! class )
goto err ;
class - > size = size ;
class - > index = i ;
class - > pages_per_zspage = pages_per_zspage ;
spin_lock_init ( & class - > lock ) ;
pool - > size_class [ i ] = class ;
prev_class = class ;
2012-01-10 02:51:56 +04:00
}
2014-12-19 03:17:40 +03:00
pool - > flags = flags ;
2012-07-03 01:15:52 +04:00
2014-12-19 03:17:40 +03:00
return pool ;
err :
zs_destroy_pool ( pool ) ;
return NULL ;
2012-01-10 02:51:56 +04:00
}
2014-12-19 03:17:40 +03:00
EXPORT_SYMBOL_GPL ( zs_create_pool ) ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
void zs_destroy_pool ( struct zs_pool * pool )
2012-01-10 02:51:56 +04:00
{
2014-12-19 03:17:40 +03:00
int i ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
for ( i = 0 ; i < zs_size_classes ; i + + ) {
int fg ;
struct size_class * class = pool - > size_class [ i ] ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
if ( ! class )
continue ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
if ( class - > index ! = i )
continue ;
2012-01-10 02:51:56 +04:00
2014-12-19 03:17:40 +03:00
for ( fg = 0 ; fg < _ZS_NR_FULLNESS_GROUPS ; fg + + ) {
if ( class - > fullness_list [ fg ] ) {
pr_info ( " Freeing non-empty class with size %db, fullness group %d \n " ,
class - > size , fg ) ;
}
}
kfree ( class ) ;
}
2012-07-18 20:55:56 +04:00
2014-12-19 03:17:40 +03:00
kfree ( pool - > size_class ) ;
kfree ( pool ) ;
}
EXPORT_SYMBOL_GPL ( zs_destroy_pool ) ;
2012-07-03 01:15:52 +04:00
2014-12-19 03:17:40 +03:00
static int __init zs_init ( void )
{
int ret = zs_register_cpu_notifier ( ) ;
if ( ret ) {
zs_unregister_cpu_notifier ( ) ;
return ret ;
2012-07-18 20:55:56 +04:00
}
2014-12-19 03:17:40 +03:00
init_zs_size_classes ( ) ;
# ifdef CONFIG_ZPOOL
zpool_register_driver ( & zs_zpool_driver ) ;
# endif
return 0 ;
2012-01-10 02:51:56 +04:00
}
2014-12-19 03:17:40 +03:00
static void __exit zs_exit ( void )
2012-01-10 02:51:56 +04:00
{
2014-12-19 03:17:40 +03:00
# ifdef CONFIG_ZPOOL
zpool_unregister_driver ( & zs_zpool_driver ) ;
# endif
zs_unregister_cpu_notifier ( ) ;
2012-01-10 02:51:56 +04:00
}
2012-06-20 05:31:11 +04:00
module_init ( zs_init ) ;
module_exit ( zs_exit ) ;
MODULE_LICENSE ( " Dual BSD/GPL " ) ;
MODULE_AUTHOR ( " Nitin Gupta <ngupta@vflare.org> " ) ;