Commit 33806f06 authored by Shaohua Li's avatar Shaohua Li Committed by Linus Torvalds
Browse files

swap: make each swap partition have one address_space 



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: default avatarShaohua Li <shli@fusionio.com>
Cc: Hugh Dickins <hughd@google.com>
Acked-by: default avatarRik van Riel <riel@redhat.com>
Acked-by: default avatarMinchan Kim <minchan@kernel.org>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 9800339b

fs/proc/meminfo.c

+2 −2
Original line number Diff line number Diff line
@@ -40,7 +40,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v) 
		* sysctl_overcommit_ratio / 100) + total_swap_pages;



	cached = global_page_state(NR_FILE_PAGES) -

			total_swapcache_pages - i.bufferram;

			total_swapcache_pages() - i.bufferram;

	if (cached < 0)

		cached = 0;
@@ -109,7 +109,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v) 
		K(i.freeram),

		K(i.bufferram),

		K(cached),

		K(total_swapcache_pages),

		K(total_swapcache_pages()),

		K(pages[LRU_ACTIVE_ANON]   + pages[LRU_ACTIVE_FILE]),

		K(pages[LRU_INACTIVE_ANON] + pages[LRU_INACTIVE_FILE]),

		K(pages[LRU_ACTIVE_ANON]),

include/linux/swap.h

+5 −4
Original line number Diff line number Diff line
@@ -8,7 +8,7 @@ 
#include <linux/memcontrol.h>

#include <linux/sched.h>

#include <linux/node.h>



#include <linux/fs.h>

#include <linux/atomic.h>

#include <asm/page.h>
@@ -330,8 +330,9 @@ int generic_swapfile_activate(struct swap_info_struct *, struct file *, 
		sector_t *);



/* linux/mm/swap_state.c */

extern struct address_space swapper_space;

#define total_swapcache_pages  swapper_space.nrpages

extern struct address_space swapper_spaces[];

#define swap_address_space(entry) (&swapper_spaces[swp_type(entry)])

extern unsigned long total_swapcache_pages(void);

extern void show_swap_cache_info(void);

extern int add_to_swap(struct page *);

extern int add_to_swap_cache(struct page *, swp_entry_t, gfp_t);
@@ -382,7 +383,7 @@ mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent, bool swapout) 


#define nr_swap_pages				0L

#define total_swap_pages			0L

#define total_swapcache_pages			0UL

#define total_swapcache_pages()			0UL



#define si_swapinfo(val) \

	do { (val)->freeswap = (val)->totalswap = 0; } while (0)

mm/memcontrol.c

+2 −2
Original line number Diff line number Diff line
@@ -6307,7 +6307,7 @@ static struct page *mc_handle_swap_pte(struct vm_area_struct *vma, 
	 * Because lookup_swap_cache() updates some statistics counter,

	 * we call find_get_page() with swapper_space directly.

	 */

	page = find_get_page(&swapper_space, ent.val);

	page = find_get_page(swap_address_space(ent), ent.val);

	if (do_swap_account)

		entry->val = ent.val;
@@ -6348,7 +6348,7 @@ static struct page *mc_handle_file_pte(struct vm_area_struct *vma, 
		swp_entry_t swap = radix_to_swp_entry(page);

		if (do_swap_account)

			*entry = swap;

		page = find_get_page(&swapper_space, swap.val);

		page = find_get_page(swap_address_space(swap), swap.val);

	}

#endif

	return page;

mm/mincore.c

+3 −2
Original line number Diff line number Diff line
@@ -75,7 +75,7 @@ static unsigned char mincore_page(struct address_space *mapping, pgoff_t pgoff) 
	/* shmem/tmpfs may return swap: account for swapcache page too. */

	if (radix_tree_exceptional_entry(page)) {

		swp_entry_t swap = radix_to_swp_entry(page);

		page = find_get_page(&swapper_space, swap.val);

		page = find_get_page(swap_address_space(swap), swap.val);

	}

#endif

	if (page) {
@@ -135,7 +135,8 @@ static void mincore_pte_range(struct vm_area_struct *vma, pmd_t *pmd, 
			} else {

#ifdef CONFIG_SWAP

				pgoff = entry.val;

				*vec = mincore_page(&swapper_space, pgoff);

				*vec = mincore_page(swap_address_space(entry),

					pgoff);

#else

				WARN_ON(1);

				*vec = 1;

mm/swap.c

+7 −2
Original line number Diff line number Diff line
@@ -855,9 +855,14 @@ EXPORT_SYMBOL(pagevec_lookup_tag); 
void __init swap_setup(void)

{

	unsigned long megs = totalram_pages >> (20 - PAGE_SHIFT);



#ifdef CONFIG_SWAP

	bdi_init(swapper_space.backing_dev_info);

	int i;



	bdi_init(swapper_spaces[0].backing_dev_info);

	for (i = 0; i < MAX_SWAPFILES; i++) {

		spin_lock_init(&swapper_spaces[i].tree_lock);

		INIT_LIST_HEAD(&swapper_spaces[i].i_mmap_nonlinear);

	}

#endif



	/* Use a smaller cluster for small-memory machines */

CRA Git | Maintained and supported by SUSTech CRA and CCSE