From e286781d5f2e9c846e012a39653a166e9d31777d Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Fri, 25 Jul 2008 19:45:30 -0700 Subject: mm: speculative page references If we can be sure that elevating the page_count on a pagecache page will pin it, we can speculatively run this operation, and subsequently check to see if we hit the right page rather than relying on holding a lock or otherwise pinning a reference to the page. This can be done if get_page/put_page behaves consistently throughout the whole tree (ie. if we "get" the page after it has been used for something else, we must be able to free it with a put_page). Actually, there is a period where the count behaves differently: when the page is free or if it is a constituent page of a compound page. We need an atomic_inc_not_zero operation to ensure we don't try to grab the page in either case. This patch introduces the core locking protocol to the pagecache (ie. adds page_cache_get_speculative, and tweaks some update-side code to make it work). Thanks to Hugh for pointing out an improvement to the algorithm setting page_count to zero when we have control of all references, in order to hold off speculative getters. [kamezawa.hiroyu@jp.fujitsu.com: fix migration_entry_wait()] [hugh@veritas.com: fix add_to_page_cache] [akpm@linux-foundation.org: repair a comment] Signed-off-by: Nick Piggin Cc: Jeff Garzik Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Hugh Dickins Cc: "Paul E. McKenney" Reviewed-by: Peter Zijlstra Signed-off-by: Daisuke Nishimura Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: KOSAKI Motohiro Signed-off-by: Hugh Dickins Acked-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/swap_state.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) (limited to 'mm/swap_state.c') diff --git a/mm/swap_state.c b/mm/swap_state.c index d8aadaf2a0ba..3e3381d6c7ee 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -64,7 +64,7 @@ void show_swap_cache_info(void) } /* - * add_to_swap_cache resembles add_to_page_cache on swapper_space, + * add_to_swap_cache resembles add_to_page_cache_locked on swapper_space, * but sets SwapCache flag and private instead of mapping and index. */ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask) @@ -76,19 +76,26 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask) BUG_ON(PagePrivate(page)); error = radix_tree_preload(gfp_mask); if (!error) { + page_cache_get(page); + SetPageSwapCache(page); + set_page_private(page, entry.val); + write_lock_irq(&swapper_space.tree_lock); error = radix_tree_insert(&swapper_space.page_tree, entry.val, page); - if (!error) { - page_cache_get(page); - SetPageSwapCache(page); - set_page_private(page, entry.val); + if (likely(!error)) { total_swapcache_pages++; __inc_zone_page_state(page, NR_FILE_PAGES); INC_CACHE_INFO(add_total); } write_unlock_irq(&swapper_space.tree_lock); radix_tree_preload_end(); + + if (unlikely(error)) { + set_page_private(page, 0UL); + ClearPageSwapCache(page); + page_cache_release(page); + } } return error; } -- cgit v1.2.3 From 19fd6231279be3c3bdd02ed99f9b0eb195978064 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Fri, 25 Jul 2008 19:45:32 -0700 Subject: mm: spinlock tree_lock mapping->tree_lock has no read lockers. convert the lock from an rwlock to a spinlock. Signed-off-by: Nick Piggin Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Hugh Dickins Cc: "Paul E. McKenney" Reviewed-by: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/swap_state.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'mm/swap_state.c') diff --git a/mm/swap_state.c b/mm/swap_state.c index 3e3381d6c7ee..2c217e33d497 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -39,7 +39,7 @@ static struct backing_dev_info swap_backing_dev_info = { struct address_space swapper_space = { .page_tree = RADIX_TREE_INIT(GFP_ATOMIC|__GFP_NOWARN), - .tree_lock = __RW_LOCK_UNLOCKED(swapper_space.tree_lock), + .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, @@ -80,7 +80,7 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask) SetPageSwapCache(page); set_page_private(page, entry.val); - write_lock_irq(&swapper_space.tree_lock); + spin_lock_irq(&swapper_space.tree_lock); error = radix_tree_insert(&swapper_space.page_tree, entry.val, page); if (likely(!error)) { @@ -88,7 +88,7 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask) __inc_zone_page_state(page, NR_FILE_PAGES); INC_CACHE_INFO(add_total); } - write_unlock_irq(&swapper_space.tree_lock); + spin_unlock_irq(&swapper_space.tree_lock); radix_tree_preload_end(); if (unlikely(error)) { @@ -182,9 +182,9 @@ void delete_from_swap_cache(struct page *page) entry.val = page_private(page); - write_lock_irq(&swapper_space.tree_lock); + spin_lock_irq(&swapper_space.tree_lock); __delete_from_swap_cache(page); - write_unlock_irq(&swapper_space.tree_lock); + spin_unlock_irq(&swapper_space.tree_lock); swap_free(entry); page_cache_release(page); -- cgit v1.2.3 From 2c97b7fc0d8c8661981beb9517da342ced3b3bc7 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Fri, 25 Jul 2008 19:46:01 -0700 Subject: mm: print swapcache page count in show_swap_cache_info() Every arch implements its own show_mem() function. Most of them share quite some code, some of them are completely identical. This series implements a generic version of this function and migrates almost all architectures to it. This patch: Most show_mem() implementations calculate the amount of pages within the swapcache every time. Move the output to a more appropriate place and use the anyway available total_swapcache_pages variable. Signed-off-by: Johannes Weiner Cc: Richard Henderson Cc: Ivan Kokshaysky Cc: Haavard Skinnemoen Cc: Bryan Wu Cc: Chris Zankel Cc: Ingo Molnar Cc: Jeff Dike Cc: David S. Miller Cc: Paul Mundt Cc: Heiko Carstens Cc: Martin Schwidefsky Cc: David Howells Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Yoshinori Sato Cc: Ralf Baechle Cc: Greg Ungerer Cc: Geert Uytterhoeven Cc: Roman Zippel Cc: Hirokazu Takata Cc: Mikael Starvik Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/swap_state.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'mm/swap_state.c') diff --git a/mm/swap_state.c b/mm/swap_state.c index 2c217e33d497..b8035b055129 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -56,7 +56,8 @@ static struct { void show_swap_cache_info(void) { - printk("Swap cache: add %lu, delete %lu, find %lu/%lu\n", + 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); printk("Free swap = %lukB\n", nr_swap_pages << (PAGE_SHIFT - 10)); -- cgit v1.2.3 From 529ae9aaa08378cfe2a4350bded76f32cc8ff0ce Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 2 Aug 2008 12:01:03 +0200 Subject: mm: rename page trylock Converting page lock to new locking bitops requires a change of page flag operation naming, so we might as well convert it to something nicer (!TestSetPageLocked_Lock => trylock_page, SetPageLocked => set_page_locked). This also facilitates lockdeping of page lock. Signed-off-by: Nick Piggin Acked-by: KOSAKI Motohiro Acked-by: Peter Zijlstra Acked-by: Andrew Morton Acked-by: Benjamin Herrenschmidt Signed-off-by: Linus Torvalds --- mm/swap_state.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'mm/swap_state.c') diff --git a/mm/swap_state.c b/mm/swap_state.c index b8035b055129..167cf2dc8a03 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -201,7 +201,7 @@ void delete_from_swap_cache(struct page *page) */ static inline void free_swap_cache(struct page *page) { - if (PageSwapCache(page) && !TestSetPageLocked(page)) { + if (PageSwapCache(page) && trylock_page(page)) { remove_exclusive_swap_page(page); unlock_page(page); } @@ -302,9 +302,9 @@ struct page *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, * re-using the just freed swap entry for an existing page. * May fail (-ENOMEM) if radix-tree node allocation failed. */ - SetPageLocked(new_page); + set_page_locked(new_page); err = add_to_swap_cache(new_page, entry, gfp_mask & GFP_KERNEL); - if (!err) { + if (likely(!err)) { /* * Initiate read into locked page and return. */ @@ -312,7 +312,7 @@ struct page *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, swap_readpage(NULL, new_page); return new_page; } - ClearPageLocked(new_page); + clear_page_locked(new_page); swap_free(entry); } while (err != -ENOMEM); -- cgit v1.2.3 From 07279cdfd964acc032de92a527cb11b1f40f35aa Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 20 Aug 2008 14:09:05 -0700 Subject: mm: show free swap as signed Adjust m show_swap_cache_info() to show "Free swap" as a signed long: the signed format is preferable, because during swapoff nr_swap_pages can legitimately go negative, so makes more sense thus (it used to be shown redundantly, once as signed and once as unsigned). Signed-off-by: Hugh Dickins Reviewed-by: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/swap_state.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm/swap_state.c') diff --git a/mm/swap_state.c b/mm/swap_state.c index 167cf2dc8a03..797c3831cbec 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -60,7 +60,7 @@ void show_swap_cache_info(void) 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); - printk("Free swap = %lukB\n", nr_swap_pages << (PAGE_SHIFT - 10)); + printk("Free swap = %ldkB\n", nr_swap_pages << (PAGE_SHIFT - 10)); printk("Total swap = %lukB\n", total_swap_pages << (PAGE_SHIFT - 10)); } -- cgit v1.2.3