xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

swap: make each swap partition have one address_space

Browse Source 
When I use several fast SSD to do swap, swapper_space.tree_lock is
heavily contended.  This makes each swap partition have one
address_space to reduce the lock contention.  There is an array of
address_space for swap.  The swap entry type is the index to the array.

In my test with 3 SSD, this increases the swapout throughput 20%.

[akpm@linux-foundation.org: revert unneeded change to  __add_to_swap_cache]
Signed-off-by: Shaohua Li <shli@fusionio.com>
Cc: Hugh Dickins <hughd@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
Shaohua Li
2013-02-22 16:34:37 -08:00
committed by Linus Torvalds
parent 9800339b5e
commit 33806f06da
8 changed files with 67 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

55
mm/swap_state.c
View File

@@ -36,12 +36,12 @@ static struct backing_dev_info swap_backing_dev_info = {
.capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK | BDI_CAP_SWAP_BACKED,
};
struct address_space swapper_space = {
.page_tree = RADIX_TREE_INIT(GFP_ATOMIC|__GFP_NOWARN),
.tree_lock = __SPIN_LOCK_UNLOCKED(swapper_space.tree_lock),
.a_ops = &swap_aops,
.i_mmap_nonlinear = LIST_HEAD_INIT(swapper_space.i_mmap_nonlinear),
.backing_dev_info = &swap_backing_dev_info,
struct address_space swapper_spaces[MAX_SWAPFILES] = {
[0 ... MAX_SWAPFILES - 1] = {
.page_tree = RADIX_TREE_INIT(GFP_ATOMIC|__GFP_NOWARN),
.a_ops = &swap_aops,
.backing_dev_info = &swap_backing_dev_info,
}
};
#define INC_CACHE_INFO(x) do { swap_cache_info.x++; } while (0)
@@ -53,9 +53,19 @@ static struct {
unsigned long find_total;
} swap_cache_info;
unsigned long total_swapcache_pages(void)
{
int i;
unsigned long ret = 0;
for (i = 0; i < MAX_SWAPFILES; i++)
ret += swapper_spaces[i].nrpages;
return ret;
}
void show_swap_cache_info(void)
{
printk("%lu pages in swap cache\n", total_swapcache_pages);
printk("%lu pages in swap cache\n", total_swapcache_pages());
printk("Swap cache stats: add %lu, delete %lu, find %lu/%lu\n",
swap_cache_info.add_total, swap_cache_info.del_total,
swap_cache_info.find_success, swap_cache_info.find_total);
@@ -70,6 +80,7 @@ void show_swap_cache_info(void)
static int __add_to_swap_cache(struct page *page, swp_entry_t entry)
{
int error;
struct address_space *address_space;
VM_BUG_ON(!PageLocked(page));
VM_BUG_ON(PageSwapCache(page));
@@ -79,14 +90,16 @@ static int __add_to_swap_cache(struct page *page, swp_entry_t entry)
SetPageSwapCache(page);
set_page_private(page, entry.val);
spin_lock_irq(&swapper_space.tree_lock);
error = radix_tree_insert(&swapper_space.page_tree, entry.val, page);
address_space = swap_address_space(entry);
spin_lock_irq(&address_space->tree_lock);
error = radix_tree_insert(&address_space->page_tree,
entry.val, page);
if (likely(!error)) {
total_swapcache_pages++;
address_space->nrpages++;
__inc_zone_page_state(page, NR_FILE_PAGES);
INC_CACHE_INFO(add_total);
}
spin_unlock_irq(&swapper_space.tree_lock);
spin_unlock_irq(&address_space->tree_lock);
if (unlikely(error)) {
/*
@@ -122,14 +135,19 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask)
*/
void __delete_from_swap_cache(struct page *page)
{
swp_entry_t entry;
struct address_space *address_space;
VM_BUG_ON(!PageLocked(page));
VM_BUG_ON(!PageSwapCache(page));
VM_BUG_ON(PageWriteback(page));
radix_tree_delete(&swapper_space.page_tree, page_private(page));
entry.val = page_private(page);
address_space = swap_address_space(entry);
radix_tree_delete(&address_space->page_tree, page_private(page));
set_page_private(page, 0);
ClearPageSwapCache(page);
total_swapcache_pages--;
address_space->nrpages--;
__dec_zone_page_state(page, NR_FILE_PAGES);
INC_CACHE_INFO(del_total);
}
@@ -195,12 +213,14 @@ int add_to_swap(struct page *page)
void delete_from_swap_cache(struct page *page)
{
swp_entry_t entry;
struct address_space *address_space;
entry.val = page_private(page);
spin_lock_irq(&swapper_space.tree_lock);
address_space = swap_address_space(entry);
spin_lock_irq(&address_space->tree_lock);
__delete_from_swap_cache(page);
spin_unlock_irq(&swapper_space.tree_lock);
spin_unlock_irq(&address_space->tree_lock);
swapcache_free(entry, page);
page_cache_release(page);
@@ -263,7 +283,7 @@ struct page * lookup_swap_cache(swp_entry_t entry)
{
struct page *page;
page = find_get_page(&swapper_space, entry.val);
page = find_get_page(swap_address_space(entry), entry.val);
if (page)
INC_CACHE_INFO(find_success);
@@ -290,7 +310,8 @@ struct page *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask,
* called after lookup_swap_cache() failed, re-calling
* that would confuse statistics.
*/
found_page = find_get_page(&swapper_space, entry.val);
found_page = find_get_page(swap_address_space(entry),
entry.val);
if (found_page)
break;