diff options
Diffstat (limited to 'mm')
| -rw-r--r-- | mm/Makefile | 2 | ||||
| -rw-r--r-- | mm/allocpercpu.c | 15 | ||||
| -rw-r--r-- | mm/filemap.c | 23 | ||||
| -rw-r--r-- | mm/hugetlb.c | 56 | ||||
| -rw-r--r-- | mm/memcontrol.c | 365 | ||||
| -rw-r--r-- | mm/memory.c | 13 | ||||
| -rw-r--r-- | mm/mempolicy.c | 7 | ||||
| -rw-r--r-- | mm/migrate.c | 19 | ||||
| -rw-r--r-- | mm/oom_kill.c | 2 | ||||
| -rw-r--r-- | mm/page_alloc.c | 21 | ||||
| -rw-r--r-- | mm/rmap.c | 4 | ||||
| -rw-r--r-- | mm/shmem.c | 9 | ||||
| -rw-r--r-- | mm/slab.c | 9 | ||||
| -rw-r--r-- | mm/slub.c | 217 | ||||
| -rw-r--r-- | mm/swap.c | 2 | ||||
| -rw-r--r-- | mm/truncate.c | 3 | ||||
| -rw-r--r-- | mm/vmscan.c | 9 |
17 files changed, 364 insertions, 412 deletions
diff --git a/mm/Makefile b/mm/Makefile index 9f117ba..a5b0dd9 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -32,5 +32,5 @@ obj-$(CONFIG_FS_XIP) += filemap_xip.o obj-$(CONFIG_MIGRATION) += migrate.o obj-$(CONFIG_SMP) += allocpercpu.o obj-$(CONFIG_QUICKLIST) += quicklist.o -obj-$(CONFIG_CGROUP_MEM_CONT) += memcontrol.o +obj-$(CONFIG_CGROUP_MEM_RES_CTLR) += memcontrol.o diff --git a/mm/allocpercpu.c b/mm/allocpercpu.c index 7e58322..b0012e2 100644 --- a/mm/allocpercpu.c +++ b/mm/allocpercpu.c @@ -6,6 +6,10 @@ #include <linux/mm.h> #include <linux/module.h> +#ifndef cache_line_size +#define cache_line_size() L1_CACHE_BYTES +#endif + /** * percpu_depopulate - depopulate per-cpu data for given cpu * @__pdata: per-cpu data to depopulate @@ -52,6 +56,11 @@ void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu) struct percpu_data *pdata = __percpu_disguise(__pdata); int node = cpu_to_node(cpu); + /* + * We should make sure each CPU gets private memory. + */ + size = roundup(size, cache_line_size()); + BUG_ON(pdata->ptrs[cpu]); if (node_online(node)) pdata->ptrs[cpu] = kmalloc_node(size, gfp|__GFP_ZERO, node); @@ -98,7 +107,11 @@ EXPORT_SYMBOL_GPL(__percpu_populate_mask); */ void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask) { - void *pdata = kzalloc(nr_cpu_ids * sizeof(void *), gfp); + /* + * We allocate whole cache lines to avoid false sharing + */ + size_t sz = roundup(nr_cpu_ids * sizeof(void *), cache_line_size()); + void *pdata = kzalloc(sz, gfp); void *__pdata = __percpu_disguise(pdata); if (unlikely(!pdata)) diff --git a/mm/filemap.c b/mm/filemap.c index 5c74b68..df343d1e 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -28,7 +28,6 @@ #include <linux/backing-dev.h> #include <linux/pagevec.h> #include <linux/blkdev.h> -#include <linux/backing-dev.h> #include <linux/security.h> #include <linux/syscalls.h> #include <linux/cpuset.h> @@ -1743,21 +1742,27 @@ size_t iov_iter_copy_from_user(struct page *page, } EXPORT_SYMBOL(iov_iter_copy_from_user); -static void __iov_iter_advance_iov(struct iov_iter *i, size_t bytes) +void iov_iter_advance(struct iov_iter *i, size_t bytes) { + BUG_ON(i->count < bytes); + if (likely(i->nr_segs == 1)) { i->iov_offset += bytes; + i->count -= bytes; } else { const struct iovec *iov = i->iov; size_t base = i->iov_offset; /* * The !iov->iov_len check ensures we skip over unlikely - * zero-length segments. + * zero-length segments (without overruning the iovec). */ - while (bytes || !iov->iov_len) { - int copy = min(bytes, iov->iov_len - base); + while (bytes || unlikely(!iov->iov_len && i->count)) { + int copy; + copy = min(bytes, iov->iov_len - base); + BUG_ON(!i->count || i->count < copy); + i->count -= copy; bytes -= copy; base += copy; if (iov->iov_len == base) { @@ -1769,14 +1774,6 @@ static void __iov_iter_advance_iov(struct iov_iter *i, size_t bytes) i->iov_offset = base; } } - -void iov_iter_advance(struct iov_iter *i, size_t bytes) -{ - BUG_ON(i->count < bytes); - - __iov_iter_advance_iov(i, bytes); - i->count -= bytes; -} EXPORT_SYMBOL(iov_iter_advance); /* diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 89e6286..74c1b6b 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -71,7 +71,25 @@ static void enqueue_huge_page(struct page *page) free_huge_pages_node[nid]++; } -static struct page *dequeue_huge_page(struct vm_area_struct *vma, +static struct page *dequeue_huge_page(void) +{ + int nid; + struct page *page = NULL; + + for (nid = 0; nid < MAX_NUMNODES; ++nid) { + if (!list_empty(&hugepage_freelists[nid])) { + page = list_entry(hugepage_freelists[nid].next, + struct page, lru); + list_del(&page->lru); + free_huge_pages--; + free_huge_pages_node[nid]--; + break; + } + } + return page; +} + +static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma, unsigned long address) { int nid; @@ -268,6 +286,12 @@ static struct page *alloc_buddy_huge_page(struct vm_area_struct *vma, spin_lock(&hugetlb_lock); if (page) { + /* + * This page is now managed by the hugetlb allocator and has + * no users -- drop the buddy allocator's reference. + */ + put_page_testzero(page); + VM_BUG_ON(page_count(page)); nid = page_to_nid(page); set_compound_page_dtor(page, free_huge_page); /* @@ -296,8 +320,10 @@ static int gather_surplus_pages(int delta) int needed, allocated; needed = (resv_huge_pages + delta) - free_huge_pages; - if (needed <= 0) + if (needed <= 0) { + resv_huge_pages += delta; return 0; + } allocated = 0; INIT_LIST_HEAD(&surplus_list); @@ -335,9 +361,12 @@ retry: * The surplus_list now contains _at_least_ the number of extra pages * needed to accomodate the reservation. Add the appropriate number * of pages to the hugetlb pool and free the extras back to the buddy - * allocator. + * allocator. Commit the entire reservation here to prevent another + * process from stealing the pages as they are added to the pool but + * before they are reserved. */ needed += allocated; + resv_huge_pages += delta; ret = 0; free: list_for_each_entry_safe(page, tmp, &surplus_list, lru) { @@ -346,13 +375,14 @@ free: enqueue_huge_page(page); else { /* - * Decrement the refcount and free the page using its - * destructor. This must be done with hugetlb_lock + * The page has a reference count of zero already, so + * call free_huge_page directly instead of using + * put_page. This must be done with hugetlb_lock * unlocked which is safe because free_huge_page takes * hugetlb_lock before deciding how to free the page. */ spin_unlock(&hugetlb_lock); - put_page(page); + free_huge_page(page); spin_lock(&hugetlb_lock); } } @@ -371,6 +401,9 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages) struct page *page; unsigned long nr_pages; + /* Uncommit the reservation */ + resv_huge_pages -= unused_resv_pages; + nr_pages = min(unused_resv_pages, surplus_huge_pages); while (nr_pages) { @@ -402,7 +435,7 @@ static struct page *alloc_huge_page_shared(struct vm_area_struct *vma, struct page *page; spin_lock(&hugetlb_lock); - page = dequeue_huge_page(vma, addr); + page = dequeue_huge_page_vma(vma, addr); spin_unlock(&hugetlb_lock); return page ? page : ERR_PTR(-VM_FAULT_OOM); } @@ -417,7 +450,7 @@ static struct page *alloc_huge_page_private(struct vm_area_struct *vma, spin_lock(&hugetlb_lock); if (free_huge_pages > resv_huge_pages) - page = dequeue_huge_page(vma, addr); + page = dequeue_huge_page_vma(vma, addr); spin_unlock(&hugetlb_lock); if (!page) { page = alloc_buddy_huge_page(vma, addr); @@ -570,7 +603,7 @@ static unsigned long set_max_huge_pages(unsigned long count) min_count = max(count, min_count); try_to_free_low(min_count); while (min_count < persistent_huge_pages) { - struct page *page = dequeue_huge_page(NULL, 0); + struct page *page = dequeue_huge_page(); if (!page) break; update_and_free_page(page); @@ -1205,12 +1238,13 @@ static int hugetlb_acct_memory(long delta) if (gather_surplus_pages(delta) < 0) goto out; - if (delta > cpuset_mems_nr(free_huge_pages_node)) + if (delta > cpuset_mems_nr(free_huge_pages_node)) { + return_unused_surplus_pages(delta); goto out; + } } ret = 0; - resv_huge_pages += delta; if (delta < 0) return_unused_surplus_pages((unsigned long) -delta); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 631002d..8b9f6ca 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -137,14 +137,21 @@ struct mem_cgroup { */ struct mem_cgroup_stat stat; }; +static struct mem_cgroup init_mem_cgroup; /* * We use the lower bit of the page->page_cgroup pointer as a bit spin - * lock. We need to ensure that page->page_cgroup is atleast two - * byte aligned (based on comments from Nick Piggin) + * lock. We need to ensure that page->page_cgroup is at least two + * byte aligned (based on comments from Nick Piggin). But since + * bit_spin_lock doesn't actually set that lock bit in a non-debug + * uniprocessor kernel, we should avoid setting it here too. */ #define PAGE_CGROUP_LOCK_BIT 0x0 -#define PAGE_CGROUP_LOCK (1 << PAGE_CGROUP_LOCK_BIT) +#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) +#define PAGE_CGROUP_LOCK (1 << PAGE_CGROUP_LOCK_BIT) +#else +#define PAGE_CGROUP_LOCK 0x0 +#endif /* * A page_cgroup page is associated with every page descriptor. The @@ -154,37 +161,27 @@ struct page_cgroup { struct list_head lru; /* per cgroup LRU list */ struct page *page; struct mem_cgroup *mem_cgroup; - atomic_t ref_cnt; /* Helpful when pages move b/w */ - /* mapped and cached states */ - int flags; + int ref_cnt; /* cached, mapped, migrating */ + int flags; }; #define PAGE_CGROUP_FLAG_CACHE (0x1) /* charged as cache */ #define PAGE_CGROUP_FLAG_ACTIVE (0x2) /* page is active in this cgroup */ -static inline int page_cgroup_nid(struct page_cgroup *pc) +static int page_cgroup_nid(struct page_cgroup *pc) { return page_to_nid(pc->page); } -static inline enum zone_type page_cgroup_zid(struct page_cgroup *pc) +static enum zone_type page_cgroup_zid(struct page_cgroup *pc) { return page_zonenum(pc->page); } -enum { - MEM_CGROUP_TYPE_UNSPEC = 0, - MEM_CGROUP_TYPE_MAPPED, - MEM_CGROUP_TYPE_CACHED, - MEM_CGROUP_TYPE_ALL, - MEM_CGROUP_TYPE_MAX, -}; - enum charge_type { MEM_CGROUP_CHARGE_TYPE_CACHE = 0, MEM_CGROUP_CHARGE_TYPE_MAPPED, }; - /* * Always modified under lru lock. Then, not necessary to preempt_disable() */ @@ -193,23 +190,21 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem, int flags, { int val = (charge)? 1 : -1; struct mem_cgroup_stat *stat = &mem->stat; - VM_BUG_ON(!irqs_disabled()); + VM_BUG_ON(!irqs_disabled()); if (flags & PAGE_CGROUP_FLAG_CACHE) - __mem_cgroup_stat_add_safe(stat, - MEM_CGROUP_STAT_CACHE, val); + __mem_cgroup_stat_add_safe(stat, MEM_CGROUP_STAT_CACHE, val); else __mem_cgroup_stat_add_safe(stat, MEM_CGROUP_STAT_RSS, val); } -static inline struct mem_cgroup_per_zone * +static struct mem_cgroup_per_zone * mem_cgroup_zoneinfo(struct mem_cgroup *mem, int nid, int zid) { - BUG_ON(!mem->info.nodeinfo[nid]); return &mem->info.nodeinfo[nid]->zoneinfo[zid]; } -static inline struct mem_cgroup_per_zone * +static struct mem_cgroup_per_zone * page_cgroup_zoneinfo(struct page_cgroup *pc) { struct mem_cgroup *mem = pc->mem_cgroup; @@ -234,18 +229,14 @@ static unsigned long mem_cgroup_get_all_zonestat(struct mem_cgroup *mem, return total; } -static struct mem_cgroup init_mem_cgroup; - -static inline -struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont) +static struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont) { return container_of(cgroup_subsys_state(cont, mem_cgroup_subsys_id), struct mem_cgroup, css); } -static inline -struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p) +static struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p) { return container_of(task_subsys_state(p, mem_cgroup_subsys_id), struct mem_cgroup, css); @@ -267,81 +258,33 @@ void mm_free_cgroup(struct mm_struct *mm) static inline int page_cgroup_locked(struct page *page) { - return bit_spin_is_locked(PAGE_CGROUP_LOCK_BIT, - &page->page_cgroup); + return bit_spin_is_locked(PAGE_CGROUP_LOCK_BIT, &page->page_cgroup); } -void page_assign_page_cgroup(struct page *page, struct page_cgroup *pc) +static void page_assign_page_cgroup(struct page *page, struct page_cgroup *pc) { - int locked; - - /* - * While resetting the page_cgroup we might not hold the - * page_cgroup lock. free_hot_cold_page() is an example - * of such a scenario - */ - if (pc) - VM_BUG_ON(!page_cgroup_locked(page)); - locked = (page->page_cgroup & PAGE_CGROUP_LOCK); - page->page_cgroup = ((unsigned long)pc | locked); + VM_BUG_ON(!page_cgroup_locked(page)); + page->page_cgroup = ((unsigned long)pc | PAGE_CGROUP_LOCK); } struct page_cgroup *page_get_page_cgroup(struct page *page) { - return (struct page_cgroup *) - (page->page_cgroup & ~PAGE_CGROUP_LOCK); + return (struct page_cgroup *) (page->page_cgroup & ~PAGE_CGROUP_LOCK); } -static void __always_inline lock_page_cgroup(struct page *page) +static void lock_page_cgroup(struct page *page) { bit_spin_lock(PAGE_CGROUP_LOCK_BIT, &page->page_cgroup); - VM_BUG_ON(!page_cgroup_locked(page)); -} - -static void __always_inline unlock_page_cgroup(struct page *page) -{ - bit_spin_unlock(PAGE_CGROUP_LOCK_BIT, &page->page_cgroup); } -/* - * Tie new page_cgroup to struct page under lock_page_cgroup() - * This can fail if the page has been tied to a page_cgroup. - * If success, returns 0. - */ -static int page_cgroup_assign_new_page_cgroup(struct page *page, - struct page_cgroup *pc) +static int try_lock_page_cgroup(struct page *page) { - int ret = 0; - - lock_page_cgroup(page); - if (!page_get_page_cgroup(page)) - page_assign_page_cgroup(page, pc); - else /* A page is tied to other pc. */ - ret = 1; - unlock_page_cgroup(page); - return ret; + return bit_spin_trylock(PAGE_CGROUP_LOCK_BIT, &page->page_cgroup); } -/* - * Clear page->page_cgroup member under lock_page_cgroup(). - * If given "pc" value is different from one page->page_cgroup, - * page->cgroup is not cleared. - * Returns a value of page->page_cgroup at lock taken. - * A can can detect failure of clearing by following - * clear_page_cgroup(page, pc) == pc - */ - -static struct page_cgroup *clear_page_cgroup(struct page *page, - struct page_cgroup *pc) +static void unlock_page_cgroup(struct page *page) { - struct page_cgroup *ret; - /* lock and clear */ - lock_page_cgroup(page); - ret = page_get_page_cgroup(page); - if (likely(ret == pc)) - page_assign_page_cgroup(page, NULL); - unlock_page_cgroup(page); - return ret; + bit_spin_unlock(PAGE_CGROUP_LOCK_BIT, &page->page_cgroup); } static void __mem_cgroup_remove_list(struct page_cgroup *pc) @@ -399,7 +342,7 @@ int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *mem) int ret; task_lock(task); - ret = task->mm && vm_match_cgroup(task->mm, mem); + ret = task->mm && mm_match_cgroup(task->mm, mem); task_unlock(task); return ret; } @@ -407,18 +350,30 @@ int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *mem) /* * This routine assumes that the appropriate zone's lru lock is already held */ -void mem_cgroup_move_lists(struct page_cgroup *pc, bool active) +void mem_cgroup_move_lists(struct page *page, bool active) { + struct page_cgroup *pc; struct mem_cgroup_per_zone *mz; unsigned long flags; - if (!pc) + /* + * We cannot lock_page_cgroup while holding zone's lru_lock, + * because other holders of lock_page_cgroup can be interrupted + * with an attempt to rotate_reclaimable_page. But we cannot + * safely get to page_cgroup without it, so just try_lock it: + * mem_cgroup_isolate_pages allows for page left on wrong list. + */ + if (!try_lock_page_cgroup(page)) return; - mz = page_cgroup_zoneinfo(pc); - spin_lock_irqsave(&mz->lru_lock, flags); - __mem_cgroup_move_lists(pc, active); - spin_unlock_irqrestore(&mz->lru_lock, flags); + pc = page_get_page_cgroup(page); + if (pc) { + mz = page_cgroup_zoneinfo(pc); + spin_lock_irqsave(&mz->lru_lock, flags); + __mem_cgroup_move_lists(pc, active); + spin_unlock_irqrestore(&mz->lru_lock, flags); + } + unlock_page_cgroup(page); } /* @@ -437,6 +392,7 @@ int mem_cgroup_calc_mapped_ratio(struct mem_cgroup *mem) rss = (long)mem_cgroup_read_stat(&mem->stat, MEM_CGROUP_STAT_RSS); return (int)((rss * 100L) / total); } + /* * This function is called from vmscan.c. In page reclaiming loop. balance * between active and inactive list is calculated. For memory controller @@ -500,7 +456,6 @@ long mem_cgroup_calc_reclaim_inactive(struct mem_cgroup *mem, struct mem_cgroup_per_zone *mz = mem_cgroup_zoneinfo(mem, nid, zid); nr_inactive = MEM_CGROUP_ZSTAT(mz, MEM_CGROUP_ZSTAT_INACTIVE); - return (nr_inactive >> priority); } @@ -586,26 +541,21 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm, * with it */ retry: - if (page) { - lock_page_cgroup(page); - pc = page_get_page_cgroup(page); - /* - * The page_cgroup exists and - * the page has already been accounted. - */ - if (pc) { - if (unlikely(!atomic_inc_not_zero(&pc->ref_cnt))) { - /* this page is under being uncharged ? */ - unlock_page_cgroup(page); - cpu_relax(); - goto retry; - } else { - unlock_page_cgroup(page); - goto done; - } - } + lock_page_cgroup(page); + pc = page_get_page_cgroup(page); + /* + * The page_cgroup exists and + * the page has already been accounted. + */ + if (pc) { + VM_BUG_ON(pc->page != page); + VM_BUG_ON(pc->ref_cnt <= 0); + + pc->ref_cnt++; unlock_page_cgroup(page); + goto done; } + unlock_page_cgroup(page); pc = kzalloc(sizeof(struct page_cgroup), gfp_mask); if (pc == NULL) @@ -623,16 +573,11 @@ retry: rcu_read_lock(); mem = rcu_dereference(mm->mem_cgroup); /* - * For every charge from the cgroup, increment reference - * count + * For every charge from the cgroup, increment reference count */ css_get(&mem->css); rcu_read_unlock(); - /* - * If we created the page_cgroup, we should free it on exceeding - * the cgroup limit. - */ while (res_counter_charge(&mem->res, PAGE_SIZE)) { if (!(gfp_mask & __GFP_WAIT)) goto out; @@ -641,12 +586,12 @@ retry: continue; /* - * try_to_free_mem_cgroup_pages() might not give us a full - * picture of reclaim. Some pages are reclaimed and might be - * moved to swap cache or just unmapped from the cgroup. - * Check the limit again to see if the reclaim reduced the - * current usage of the cgroup before giving up - */ + * try_to_free_mem_cgroup_pages() might not give us a full + * picture of reclaim. Some pages are reclaimed and might be + * moved to swap cache or just unmapped from the cgroup. + * Check the limit again to see if the reclaim reduced the + * current usage of the cgroup before giving up + */ if (res_counter_check_under_limit(&mem->res)) continue; @@ -657,14 +602,16 @@ retry: congestion_wait(WRITE, HZ/10); } - atomic_set(&pc->ref_cnt, 1); + pc->ref_cnt = 1; pc->mem_cgroup = mem; pc->page = page; pc->flags = PAGE_CGROUP_FLAG_ACTIVE; if (ctype == MEM_CGROUP_CHARGE_TYPE_CACHE) pc->flags |= PAGE_CGROUP_FLAG_CACHE; - if (!page || page_cgroup_assign_new_page_cgroup(page, pc)) { + lock_page_cgroup(page); + if (page_get_page_cgroup(page)) { + unlock_page_cgroup(page); /* * Another charge has been added to this page already. * We take lock_page_cgroup(page) again and read @@ -673,17 +620,16 @@ retry: res_counter_uncharge(&mem->res, PAGE_SIZE); css_put(&mem->css); kfree(pc); - if (!page) - goto done; goto retry; } + page_assign_page_cgroup(page, pc); mz = page_cgroup_zoneinfo(pc); spin_lock_irqsave(&mz->lru_lock, flags); - /* Update statistics vector */ __mem_cgroup_add_list(pc); spin_unlock_irqrestore(&mz->lru_lock, flags); + unlock_page_cgroup(page); done: return 0; out: @@ -693,70 +639,61 @@ err: return -ENOMEM; } -int mem_cgroup_charge(struct page *page, struct mm_struct *mm, - gfp_t gfp_mask) +int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask) { return mem_cgroup_charge_common(page, mm, gfp_mask, - MEM_CGROUP_CHARGE_TYPE_MAPPED); + MEM_CGROUP_CHARGE_TYPE_MAPPED); } -/* - * See if the cached pages should be charged at all? - */ int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask) { - int ret = 0; if (!mm) mm = &init_mm; - - ret = mem_cgroup_charge_common(page, mm, gfp_mask, + return mem_cgroup_charge_common(page, mm, gfp_mask, MEM_CGROUP_CHARGE_TYPE_CACHE); - return ret; } /* * Uncharging is always a welcome operation, we never complain, simply - * uncharge. This routine should be called with lock_page_cgroup held + * uncharge. */ -void mem_cgroup_uncharge(struct page_cgroup *pc) +void mem_cgroup_uncharge_page(struct page *page) { + struct page_cgroup *pc; struct mem_cgroup *mem; struct mem_cgroup_per_zone *mz; - struct page *page; unsigned long flags; /* * Check if our page_cgroup is valid */ + lock_page_cgroup(page); + pc = page_get_page_cgroup(page); if (!pc) - return; + goto unlock; - if (atomic_dec_and_test(&pc->ref_cnt)) { - page = pc->page; + VM_BUG_ON(pc->page != page); + VM_BUG_ON(pc->ref_cnt <= 0); + + if (--(pc->ref_cnt) == 0) { mz = page_cgroup_zoneinfo(pc); - /* - * get page->cgroup and clear it under lock. - * force_empty can drop page->cgroup without checking refcnt. - */ + spin_lock_irqsave(&mz->lru_lock, flags); + __mem_cgroup_remove_list(pc); + spin_unlock_irqrestore(&mz->lru_lock, flags); + + page_assign_page_cgroup(page, NULL); unlock_page_cgroup(page); - if (clear_page_cgroup(page, pc) == pc) { - mem = pc->mem_cgroup; - css_put(&mem->css); - res_counter_uncharge(&mem->res, PAGE_SIZE); - spin_lock_irqsave(&mz->lru_lock, flags); - __mem_cgroup_remove_list(pc); - spin_unlock_irqrestore(&mz->lru_lock, flags); - kfree(pc); - } - lock_page_cgroup(page); + + mem = pc->mem_cgroup; + res_counter_uncharge(&mem->res, PAGE_SIZE); + css_put(&mem->css); + + kfree(pc); + return; } -} -void mem_cgroup_uncharge_page(struct page *page) -{ - lock_page_cgroup(page); - mem_cgroup_uncharge(page_get_page_cgroup(page)); +unlock: unlock_page_cgroup(page); } @@ -764,63 +701,59 @@ void mem_cgroup_uncharge_page(struct page *page) * Returns non-zero if a page (under migration) has valid page_cgroup member. * Refcnt of page_cgroup is incremented. */ - int mem_cgroup_prepare_migration(struct page *page) { struct page_cgroup *pc; - int ret = 0; + lock_page_cgroup(page); pc = page_get_page_cgroup(page); - if (pc && atomic_inc_not_zero(&pc->ref_cnt)) - ret = 1; + if (pc) + pc->ref_cnt++; unlock_page_cgroup(page); - return ret; + return pc != NULL; } void mem_cgroup_end_migration(struct page *page) { - struct page_cgroup *pc; - - lock_page_cgroup(page); - pc = page_get_page_cgroup(page); - mem_cgroup_uncharge(pc); - unlock_page_cgroup(page); + mem_cgroup_uncharge_page(page); } + /* - * We know both *page* and *newpage* are now not-on-LRU and Pg_locked. + * We know both *page* and *newpage* are now not-on-LRU and PG_locked. * And no race with uncharge() routines because page_cgroup for *page* * has extra one reference by mem_cgroup_prepare_migration. */ - void mem_cgroup_page_migration(struct page *page, struct page *newpage) { struct page_cgroup *pc; - struct mem_cgroup *mem; - unsigned long flags; struct mem_cgroup_per_zone *mz; -retry: + unsigned long flags; + + lock_page_cgroup(page); pc = page_get_page_cgroup(page); - if (!pc) + if (!pc) { + unlock_page_cgroup(page); return; - mem = pc->mem_cgroup; + } + mz = page_cgroup_zoneinfo(pc); - if (clear_page_cgroup(page, pc) != pc) - goto retry; spin_lock_irqsave(&mz->lru_lock, flags); - __mem_cgroup_remove_list(pc); spin_unlock_irqrestore(&mz->lru_lock, flags); + page_assign_page_cgroup(page, NULL); + unlock_page_cgroup(page); + pc->page = newpage; lock_page_cgroup(newpage); page_assign_page_cgroup(newpage, pc); - unlock_page_cgroup(newpage); mz = page_cgroup_zoneinfo(pc); spin_lock_irqsave(&mz->lru_lock, flags); __mem_cgroup_add_list(pc); spin_unlock_irqrestore(&mz->lru_lock, flags); - return; + + unlock_page_cgroup(newpage); } /* @@ -829,14 +762,13 @@ retry: * *And* this routine doesn't reclaim page itself, just removes page_cgroup. */ #define FORCE_UNCHARGE_BATCH (128) -static void -mem_cgroup_force_empty_list(struct mem_cgroup *mem, +static void mem_cgroup_force_empty_list(struct mem_cgroup *mem, struct mem_cgroup_per_zone *mz, int active) { struct page_cgroup *pc; struct page *page; - int count; + int count = FORCE_UNCHARGE_BATCH; unsigned long flags; struct list_head *list; @@ -845,46 +777,36 @@ mem_cgroup_force_empty_list(struct mem_cgroup *mem, else list = &mz->inactive_list; - if (list_empty(list)) - return; -retry: - count = FORCE_UNCHARGE_BATCH; spin_lock_irqsave(&mz->lru_lock, flags); - - while (--count && !list_empty(list)) { + while (!list_empty(list)) { pc = list_entry(list->prev, struct page_cgroup, lru); page = pc->page; - /* Avoid race with charge */ - atomic_set(&pc->ref_cnt, 0); - if (clear_page_cgroup(page, pc) == pc) { - css_put(&mem->css); - res_counter_uncharge(&mem->res, PAGE_SIZE); - __mem_cgroup_remove_list(pc); - kfree(pc); - } else /* being uncharged ? ...do relax */ - break; + get_page(page); + spin_unlock_irqrestore(&mz->lru_lock, flags); + mem_cgroup_uncharge_page(page); + put_page(page); + if (--count <= 0) { + count = FORCE_UNCHARGE_BATCH; + cond_resched(); + } + spin_lock_irqsave(&mz->lru_lock, flags); } spin_unlock_irqrestore(&mz->lru_lock, flags); - if (!list_empty(list)) { - cond_resched(); - goto retry; - } - return; } /* * make mem_cgroup's charge to be 0 if there is no task. * This enables deleting this mem_cgroup. */ - -int mem_cgroup_force_empty(struct mem_cgroup *mem) +static int mem_cgroup_force_empty(struct mem_cgroup *mem) { int ret = -EBUSY; int node, zid; + css_get(&mem->css); /* * page reclaim code (kswapd etc..) will move pages between -` * active_list <-> inactive_list while we don't take a lock. + * active_list <-> inactive_list while we don't take a lock. * So, we have to do loop here until all lists are empty. */ while (mem->res.usage > 0) { @@ -906,9 +828,7 @@ out: return ret; } - - -int mem_cgroup_write_strategy(char *buf, unsigned long long *tmp) +static int mem_cgroup_write_strategy(char *buf, unsigned long long *tmp) { *tmp = memparse(buf, &buf); if (*buf != '\0') @@ -945,8 +865,7 @@ static ssize_t mem_force_empty_write(struct cgroup *cont, size_t nbytes, loff_t *ppos) { struct mem_cgroup *mem = mem_cgroup_from_cont(cont); - int ret; - ret = mem_cgroup_force_empty(mem); + int ret = mem_cgroup_force_empty(mem); if (!ret) ret = nbytes; return ret; @@ -955,7 +874,6 @@ static ssize_t mem_force_empty_write(struct cgroup *cont, /* * Note: This should be removed if cgroup supports write-only file. */ - static ssize_t mem_force_empty_read(struct cgroup *cont, struct cftype *cft, struct file *file, char __user *userbuf, @@ -964,7 +882,6 @@ static ssize_t mem_force_empty_read(struct cgroup *cont, return -EINVAL; } - static const struct mem_cgroup_stat_desc { const char *msg; u64 unit; @@ -1017,8 +934,6 @@ static int mem_control_stat_open(struct inode *unused, struct file *file) return single_open(file, mem_control_stat_show, cont); } - - static struct cftype mem_cgroup_files[] = { { .name = "usage_in_bytes", @@ -1084,9 +999,6 @@ static void free_mem_cgroup_per_zone_info(struct mem_cgroup *mem, int node) kfree(mem->info.nodeinfo[node]); } - -static struct mem_cgroup init_mem_cgroup; - static struct cgroup_subsys_state * mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont) { @@ -1176,7 +1088,6 @@ static void mem_cgroup_move_task(struct cgroup_subsys *ss, out: mmput(mm); - return; } struct cgroup_subsys mem_cgroup_subsys = { diff --git a/mm/memory.c b/mm/memory.c index ce3c9e4..0d14d1e 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1711,7 +1711,7 @@ unlock: } return ret; oom_free_new: - __free_page(new_page); + page_cache_release(new_page); oom: if (old_page) page_cache_release(old_page); @@ -2093,12 +2093,9 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, unlock_page(page); if (write_access) { - /* XXX: We could OR the do_wp_page code with this one? */ - if (do_wp_page(mm, vma, address, - page_table, pmd, ptl, pte) & VM_FAULT_OOM) { - mem_cgroup_uncharge_page(page); - ret = VM_FAULT_OOM; - } + ret |= do_wp_page(mm, vma, address, page_table, pmd, ptl, pte); + if (ret & VM_FAULT_ERROR) + ret &= VM_FAULT_ERROR; goto out; } @@ -2163,7 +2160,7 @@ release: page_cache_release(page); goto unlock; oom_free_page: - __free_page(page); + page_cache_release(page); oom: return VM_FAULT_OOM; } diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 6c7ba1a..3c36011 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1296,7 +1296,9 @@ struct zonelist *huge_zonelist(struct vm_area_struct *vma, unsigned long addr, unsigned nid; nid = interleave_nid(pol, vma, addr, HPAGE_SHIFT); - __mpol_free(pol); /* finished with pol */ + if (unlikely(pol != &default_policy && + pol != current->mempolicy)) + __mpol_free(pol); /* finished with pol */ return NODE_DATA(nid)->node_zonelists + gfp_zone(gfp_flags); } @@ -1360,6 +1362,9 @@ alloc_page_vma(gfp_t gfp, struct vm_area_struct *vma, unsigned long addr) unsigned nid; nid = interleave_nid(pol, vma, addr, PAGE_SHIFT); + if (unlikely(pol != &default_policy && + pol != current->mempolicy)) + __mpol_free(pol); /* finished with pol */ return alloc_page_interleave(gfp, 0, nid); } zl = zonelist_policy(gfp, pol); diff --git a/mm/migrate.c b/mm/migrate.c index a73504f..4e0eccc 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -153,11 +153,6 @@ static void remove_migration_pte(struct vm_area_struct *vma, return; } - if (mem_cgroup_charge(new, mm, GFP_KERNEL)) { - pte_unmap(ptep); - return; - } - ptl = pte_lockptr(mm, pmd); spin_lock(ptl); pte = *ptep; @@ -169,6 +164,20 @@ static void remove_migration_pte(struct vm_area_struct *vma, if (!is_migration_entry(entry) || migration_entry_to_page(entry) != old) goto out; + /* + * Yes, ignore the return value from a GFP_ATOMIC mem_cgroup_charge. + * Failure is not an option here: we're now expected to remove every + * migration pte, and will cause crashes otherwise. Normally this + * is not an issue: mem_cgroup_prepare_migration bumped up the old + * page_cgroup count for safety, that's now attached to the new page, + * so this charge should just be another incrementation of the count, + * to keep in balance with rmap.c's mem_cgroup_uncharging. But if + * there's been a force_empty, those reference counts may no longer + * be reliable, and this charge can actually fail: oh well, we don't + * make the situation any worse by proceeding as if it had succeeded. + */ + mem_cgroup_charge(new, mm, GFP_ATOMIC); + get_page(new); pte = pte_mkold(mk_pte(new, vma->vm_page_prot)); if (is_write_migration_entry(entry)) diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 4194b9d..44b2da1 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -412,7 +412,7 @@ static int oom_kill_process(struct task_struct *p, gfp_t gfp_mask, int order, return oom_kill_task(p); } -#ifdef CONFIG_CGROUP_MEM_CONT +#ifdef CONFIG_CGROUP_MEM_RES_CTLR void mem_cgroup_out_of_memory(struct mem_cgroup *mem, gfp_t gfp_mask) { unsigned long points = 0; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8896e87..402a504 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -19,6 +19,7 @@ #include <linux/swap.h> #include <linux/interrupt.h> #include <linux/pagemap.h> +#include <linux/jiffies.h> #include <linux/bootmem.h> #include <linux/compiler.h> #include <linux/kernel.h> @@ -221,13 +222,19 @@ static inline int bad_range(struct zone *zone, struct page *page) static void bad_page(struct page *page) { - printk(KERN_EMERG "Bad page state in process '%s'\n" - KERN_EMERG "page:%p flags:0x%0*lx mapping:%p mapcount:%d count:%d\n" - KERN_EMERG "Trying to fix it up, but a reboot is needed\n" - KERN_EMERG "Backtrace:\n", + void *pc = page_get_page_cgroup(page); + + printk(KERN_EMERG "Bad page state in process '%s'\n" KERN_EMERG + "page:%p flags:0x%0*lx mapping:%p mapcount:%d count:%d\n", current->comm, page, (int)(2*sizeof(unsigned long)), (unsigned long)page->flags, page->mapping, page_mapcount(page), page_count(page)); + if (pc) { + printk(KERN_EMERG "cgroup:%p\n", pc); + page_reset_bad_cgroup(page); + } + printk(KERN_EMERG "Trying to fix it up, but a reboot is needed\n" + KERN_EMERG "Backtrace:\n"); dump_stack(); page->flags &= ~(1 << PG_lru | 1 << PG_private | @@ -453,6 +460,7 @@ static inline int free_pages_check(struct page *page) { if (unlikely(page_mapcount(page) | (page->mapping != NULL) | + (page_get_page_cgroup(page) != NULL) | (page_count(page) != 0) | (page->flags & ( 1 << PG_lru | @@ -602,6 +610,7 @@ static int prep_new_page(struct page *page, int order, gfp_t gfp_flags) { if (unlikely(page_mapcount(page) | (page->mapping != NULL) | + (page_get_page_cgroup(page) != NULL) | (page_count(page) != 0) | (page->flags & ( 1 << PG_lru | @@ -988,7 +997,6 @@ static void free_hot_cold_page(struct page *page, int cold) if (!PageHighMem(page)) debug_check_no_locks_freed(page_address(page), PAGE_SIZE); - VM_BUG_ON(page_get_page_cgroup(page)); arch_free_page(page, 0); kernel_map_pages(page, 1, 0); @@ -1276,7 +1284,7 @@ static nodemask_t *zlc_setup(struct zonelist *zonelist, int alloc_flags) if (!zlc) return NULL; - if (jiffies - zlc->last_full_zap > 1 * HZ) { + if (time_after(jiffies, zlc->last_full_zap + HZ)) { bitmap_zero(zlc->fullzones, MAX_ZONES_PER_ZONELIST); zlc->last_full_zap = jiffies; } @@ -2527,7 +2535,6 @@ void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone, set_page_links(page, zone, nid, pfn); init_page_count(page); reset_page_mapcount(page); - page_assign_page_cgroup(page, NULL); SetPageReserved(page); /* @@ -321,7 +321,7 @@ static int page_referenced_anon(struct page *page, * counting on behalf of references from different * cgroups */ - if (mem_cont && !vm_match_cgroup(vma->vm_mm, mem_cont)) + if (mem_cont && !mm_match_cgroup(vma->vm_mm, mem_cont)) continue; referenced += page_referenced_one(page, vma, &mapcount); if (!mapcount) @@ -382,7 +382,7 @@ static int page_referenced_file(struct page *page, * counting on behalf of references from different * cgroups */ - if (mem_cont && !vm_match_cgroup(vma->vm_mm, mem_cont)) + if (mem_cont && !mm_match_cgroup(vma->vm_mm, mem_cont)) continue; if ((vma->vm_flags & (VM_LOCKED|VM_MAYSHARE)) == (VM_LOCKED|VM_MAYSHARE)) { @@ -1370,14 +1370,17 @@ repeat: shmem_swp_unmap(entry); spin_unlock(&info->lock); unlock_page(swappage); - page_cache_release(swappage); if (error == -ENOMEM) { /* allow reclaim from this memory cgroup */ - error = mem_cgroup_cache_charge(NULL, + error = mem_cgroup_cache_charge(swappage, current->mm, gfp & ~__GFP_HIGHMEM); - if (error) + if (error) { + page_cache_release(swappage); goto failed; + } + mem_cgroup_uncharge_page(swappage); } + page_cache_release(swappage); goto repeat; } } else if (sgp == SGP_READ && !filepage) { @@ -333,7 +333,7 @@ static __always_inline int index_of(const size_t size) return i; \ else \ i++; -#include "linux/kmalloc_sizes.h" +#include <linux/kmalloc_sizes.h> #undef CACHE __bad_size(); } else @@ -2964,11 +2964,10 @@ static void *cache_alloc_refill(struct kmem_cache *cachep, gfp_t flags) struct array_cache *ac; int node; - node = numa_node_id(); - +retry: check_irq_off(); + node = numa_node_id(); ac = cpu_cache_get(cachep); -retry: batchcount = ac->batchcount; if (!ac->touched && batchcount > BATCHREFILL_LIMIT) { /* @@ -3280,7 +3279,7 @@ retry: if (local_flags & __GFP_WAIT) local_irq_enable(); kmem_flagcheck(cache, flags); - obj = kmem_getpages(cache, flags, -1); + obj = kmem_getpages(cache, local_flags, -1); if (local_flags & __GFP_WAIT) local_irq_disable(); if (obj) { @@ -291,32 +291,16 @@ static inline struct kmem_cache_cpu *get_cpu_slab(struct kmem_cache *s, int cpu) #endif } -/* - * The end pointer in a slab is special. It points to the first object in the - * slab but has bit 0 set to mark it. - * - * Note that SLUB relies on page_mapping returning NULL for pages with bit 0 - * in the mapping set. - */ -static inline int is_end(void *addr) -{ - return (unsigned long)addr & PAGE_MAPPING_ANON; -} - -static void *slab_address(struct page *page) -{ - return page->end - PAGE_MAPPING_ANON; -} - +/* Verify that a pointer has an address that is valid within a slab page */ static inline int check_valid_pointer(struct kmem_cache *s, struct page *page, const void *object) { void *base; - if (object == page->end) + if (!object) return 1; - base = slab_address(page); + base = page_address(page); if (object < base || object >= base + s->objects * s->size || (object - base) % s->size) { return 0; @@ -349,8 +333,7 @@ static inline void set_freepointer(struct kmem_cache *s, void *object, void *fp) /* Scan freelist */ #define for_each_free_object(__p, __s, __free) \ - for (__p = (__free); (__p) != page->end; __p = get_freepointer((__s),\ - __p)) + for (__p = (__free); __p; __p = get_freepointer((__s), __p)) /* Determine object index from a given position */ static inline int slab_index(void *p, struct kmem_cache *s, void *addr) @@ -502,7 +485,7 @@ static void slab_fix(struct kmem_cache *s, char *fmt, ...) static void print_trailer(struct kmem_cache *s, struct page *page, u8 *p) { unsigned int off; /* Offset of last byte */ - u8 *addr = slab_address(page); + u8 *addr = page_address(page); print_tracking(s, p); @@ -637,7 +620,7 @@ static int check_bytes_and_report(struct kmem_cache *s, struct page *page, * A. Free pointer (if we cannot overwrite object on free) * B. Tracking data for SLAB_STORE_USER * C. Padding to reach required alignment boundary or at mininum - * one word if debuggin is on to be able to detect writes + * one word if debugging is on to be able to detect writes * before the word boundary. * * Padding is done using 0x5a (POISON_INUSE) @@ -680,7 +663,7 @@ static int slab_pad_check(struct kmem_cache *s, struct page *page) if (!(s->flags & SLAB_POISON)) return 1; - start = slab_address(page); + start = page_address(page); end = start + (PAGE_SIZE << s->order); length = s->objects * s->size; remainder = end - (start + length); @@ -748,7 +731,7 @@ static int check_object(struct kmem_cache *s, struct page *page, * of the free objects in this slab. May cause * another error because the object count is now wrong. */ - set_freepointer(s, p, page->end); + set_freepointer(s, p, NULL); return 0; } return 1; @@ -782,18 +765,18 @@ static int on_freelist(struct kmem_cache *s, struct page *page, void *search) void *fp = page->freelist; void *object = NULL; - while (fp != page->end && nr <= s->objects) { + while (fp && nr <= s->objects) { if (fp == search) return 1; if (!check_valid_pointer(s, page, fp)) { if (object) { object_err(s, page, object, "Freechain corrupt"); - set_freepointer(s, object, page->end); + set_freepointer(s, object, NULL); break; } else { slab_err(s, page, "Freepointer corrupt"); - page->freelist = page->end; + page->freelist = NULL; page->inuse = s->objects; slab_fix(s, "Freelist cleared"); return 0; @@ -870,7 +853,7 @@ static int alloc_debug_processing(struct kmem_cache *s, struct page *page, if (!check_slab(s, page)) goto bad; - if (object && !on_freelist(s, page, object)) { + if (!on_freelist(s, page, object)) { object_err(s, page, object, "Object already allocated"); goto bad; } @@ -880,7 +863,7 @@ static int alloc_debug_processing(struct kmem_cache *s, struct page *page, goto bad; } - if (object && !check_object(s, page, object, 0)) + if (!check_object(s, page, object, 0)) goto bad; /* Success perform special debug activities for allocs */ @@ -899,7 +882,7 @@ bad: */ slab_fix(s, "Marking all objects used"); page->inuse = s->objects; - page->freelist = page->end; + page->freelist = NULL; } return 0; } @@ -939,7 +922,7 @@ static int free_debug_processing(struct kmem_cache *s, struct page *page, } /* Special debug activities for freeing objects */ - if (!SlabFrozen(page) && page->freelist == page->end) + if (!SlabFrozen(page) && !page->freelist) remove_full(s, page); if (s->flags & SLAB_STORE_USER) set_track(s, object, TRACK_FREE, addr); @@ -1015,30 +998,11 @@ static unsigned long kmem_cache_flags(unsigned long objsize, void (*ctor)(struct kmem_cache *, void *)) { /* - * The page->offset field is only 16 bit wide. This is an offset - * in units of words from the beginning of an object. If the slab - * size is bigger then we cannot move the free pointer behind the - * object anymore. - * - * On 32 bit platforms the limit is 256k. On 64bit platforms - * the limit is 512k. - * - * Debugging or ctor may create a need to move the free - * pointer. Fail if this happens. + * Enable debugging if selected on the kernel commandline. */ - if (objsize >= 65535 * sizeof(void *)) { - BUG_ON(flags & (SLAB_RED_ZONE | SLAB_POISON | - SLAB_STORE_USER | SLAB_DESTROY_BY_RCU)); - BUG_ON(ctor); - } else { - /* - * Enable debugging if selected on the kernel commandline. - */ - if (slub_debug && (!slub_debug_slabs || - strncmp(slub_debug_slabs, name, - strlen(slub_debug_slabs)) == 0)) - flags |= slub_debug; - } + if (slub_debug && (!slub_debug_slabs || + strncmp(slub_debug_slabs, name, strlen(slub_debug_slabs)) == 0)) + flags |= slub_debug; return flags; } @@ -1124,7 +1088,6 @@ static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node) SetSlabDebug(page); start = page_address(page); - page->end = start + 1; if (unlikely(s->flags & SLAB_POISON)) memset(start, POISON_INUSE, PAGE_SIZE << s->order); @@ -1136,7 +1099,7 @@ static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node) last = p; } setup_object(s, page, last); - set_freepointer(s, last, page->end); + set_freepointer(s, last, NULL); page->freelist = start; page->inuse = 0; @@ -1152,7 +1115,7 @@ static void __free_slab(struct kmem_cache *s, struct page *page) void *p; slab_pad_check(s, page); - for_each_object(p, s, slab_address(page)) + for_each_object(p, s, page_address(page)) check_object(s, page, p, 0); ClearSlabDebug(page); } @@ -1162,7 +1125,6 @@ static void __free_slab(struct kmem_cache *s, struct page *page) NR_SLAB_RECLAIMABLE : NR_SLAB_UNRECLAIMABLE, -pages); - page->mapping = NULL; __free_pages(page, s->order); } @@ -1307,7 +1269,7 @@ static struct page *get_any_partial(struct kmem_cache *s, gfp_t flags) * may return off node objects because partial slabs are obtained * from other nodes and filled up. * - * If /sys/slab/xx/defrag_ratio is set to 100 (which makes + * If /sys/kernel/slab/xx/defrag_ratio is set to 100 (which makes * defrag_ratio = 1000) then every (well almost) allocation will * first attempt to defrag slab caches on other nodes. This means * scanning over all nodes to look for partial slabs which may be @@ -1366,7 +1328,7 @@ static void unfreeze_slab(struct kmem_cache *s, struct page *page, int tail) ClearSlabFrozen(page); if (page->inuse) { - if (page->freelist != page->end) { + if (page->freelist) { add_partial(n, page, tail); stat(c, tail ? DEACTIVATE_TO_TAIL : DEACTIVATE_TO_HEAD); } else { @@ -1382,9 +1344,11 @@ static void unfreeze_slab(struct kmem_cache *s, struct page *page, int tail) * Adding an empty slab to the partial slabs in order * to avoid page allocator overhead. This slab needs * to come after the other slabs with objects in - * order to fill them up. That way the size of the - * partial list stays small. kmem_cache_shrink can - * reclaim empty slabs from the partial list. + * so that the others get filled first. That way the + * size of the partial list stays small. + * + * kmem_cache_shrink can reclaim any empty slabs from the + * partial list. */ add_partial(n, page, 1); slab_unlock(page); @@ -1404,18 +1368,14 @@ static void deactivate_slab(struct kmem_cache *s, struct kmem_cache_cpu *c) struct page *page = c->page; int tail = 1; - if (c->freelist) + if (page->freelist) stat(c, DEACTIVATE_REMOTE_FREES); /* - * Merge cpu freelist into freelist. Typically we get here + * Merge cpu freelist into slab freelist. Typically we get here * because both freelists are empty. So this is unlikely * to occur. - * - * We need to use _is_end here because deactivate slab may - * be called for a debug slab. Then c->freelist may contain - * a dummy pointer. */ - while (unlikely(!is_end(c->freelist))) { + while (unlikely(c->freelist)) { void **object; tail = 0; /* Hot objects. Put the slab first */ @@ -1442,6 +1402,7 @@ static inline void flush_slab(struct kmem_cache *s, struct kmem_cache_cpu *c) /* * Flush cpu slab. + * * Called from IPI handler with interrupts disabled. */ static inline void __flush_cpu_slab(struct kmem_cache *s, int cpu) @@ -1500,7 +1461,8 @@ static inline int node_match(struct kmem_cache_cpu *c, int node) * rest of the freelist to the lockless freelist. * * And if we were unable to get a new slab from the partial slab lists then - * we need to allocate a new slab. This is slowest path since we may sleep. + * we need to allocate a new slab. This is the slowest path since it involves + * a call to the page allocator and the setup of a new slab. */ static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node, void *addr, struct kmem_cache_cpu *c) @@ -1514,18 +1476,19 @@ static void *__slab_alloc(struct kmem_cache *s, slab_lock(c->page); if (unlikely(!node_match(c, node))) goto another_slab; + stat(c, ALLOC_REFILL); + load_freelist: object = c->page->freelist; - if (unlikely(object == c->page->end)) + if (unlikely(!object)) goto another_slab; if (unlikely(SlabDebug(c->page))) goto debug; - object = c->page->freelist; c->freelist = object[c->offset]; c->page->inuse = s->objects; - c->page->freelist = c->page->end; + c->page->freelist = NULL; c->node = page_to_nid(c->page); unlock_out: slab_unlock(c->page); @@ -1578,7 +1541,6 @@ new_slab: return NULL; debug: - object = c->page->freelist; if (!alloc_debug_processing(s, c->page, object, addr)) goto another_slab; @@ -1607,7 +1569,7 @@ static __always_inline void *slab_alloc(struct kmem_cache *s, local_irq_save(flags); c = get_cpu_slab(s, smp_processor_id()); - if (unlikely(is_end(c->freelist) || !node_match(c, node))) + if (unlikely(!c->freelist || !node_match(c, node))) object = __slab_alloc(s, gfpflags, node, addr, c); @@ -1659,6 +1621,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page, if (unlikely(SlabDebug(page))) goto debug; + checks_ok: prior = object[offset] = page->freelist; page->freelist = object; @@ -1673,11 +1636,10 @@ checks_ok: goto slab_empty; /* - * Objects left in the slab. If it - * was not on the partial list before + * Objects left in the slab. If it was not on the partial list before * then add it. */ - if (unlikely(prior == page->end)) { + if (unlikely(!prior)) { add_partial(get_node(s, page_to_nid(page)), page, 1); stat(c, FREE_ADD_PARTIAL); } @@ -1687,7 +1649,7 @@ out_unlock: return; slab_empty: - if (prior != page->end) { + if (prior) { /* * Slab still on the partial list. */ @@ -1724,8 +1686,8 @@ static __always_inline void slab_free(struct kmem_cache *s, unsigned long flags; local_irq_save(flags); - debug_check_no_locks_freed(object, s->objsize); c = get_cpu_slab(s, smp_processor_id()); + debug_check_no_locks_freed(object, c->objsize); if (likely(page == c->page && c->node >= 0)) { object[c->offset] = c->freelist; c->freelist = object; @@ -1888,20 +1850,21 @@ static unsigned long calculate_alignment(unsigned long flags, unsigned long align, unsigned long size) { /* - * If the user wants hardware cache aligned objects then - * follow that suggestion if the object is sufficiently - * large. + * If the user wants hardware cache aligned objects then follow that + * suggestion if the object is sufficiently large. * - * The hardware cache alignment cannot override the - * specified alignment though. If that is greater - * then use it. + * The hardware cache alignment cannot override the specified + * alignment though. If that is greater then use it. */ - if ((flags & SLAB_HWCACHE_ALIGN) && - size > cache_line_size() / 2) - return max_t(unsigned long, align, cache_line_size()); + if (flags & SLAB_HWCACHE_ALIGN) { + unsigned long ralign = cache_line_size(); + while (size <= ralign / 2) + ralign /= 2; + align = max(align, ralign); + } if (align < ARCH_SLAB_MINALIGN) - return ARCH_SLAB_MINALIGN; + align = ARCH_SLAB_MINALIGN; return ALIGN(align, sizeof(void *)); } @@ -1910,7 +1873,7 @@ static void init_kmem_cache_cpu(struct kmem_cache *s, struct kmem_cache_cpu *c) { c->page = NULL; - c->freelist = (void *)PAGE_MAPPING_ANON; + c->freelist = NULL; c->node = 0; c->offset = s->offset / sizeof(void *); c->objsize = s->objsize; @@ -2092,6 +2055,7 @@ static struct kmem_cache_node *early_kmem_cache_node_alloc(gfp_t gfpflags, #endif init_kmem_cache_node(n); atomic_long_inc(&n->nr_slabs); + /* * lockdep requires consistent irq usage for each lock * so even though there cannot be a race this early in @@ -2173,6 +2137,14 @@ static int calculate_sizes(struct kmem_cache *s) unsigned long align = s->align; /* + * Round up object size to the next word boundary. We can only + * place the free pointer at word boundaries and this determines + * the possible location of the free pointer. + */ + size = ALIGN(size, sizeof(void *)); + +#ifdef CONFIG_SLUB_DEBUG + /* * Determine if we can poison the object itself. If the user of * the slab may touch the object after free or before allocation * then we should never poison the object itself. @@ -2183,14 +2155,7 @@ static int calculate_sizes(struct kmem_cache *s) else s->flags &= ~__OBJECT_POISON; - /* - * Round up object size to the next word boundary. We can only - * place the free pointer at word boundaries and this determines - * the possible location of the free pointer. - */ - size = ALIGN(size, sizeof(void *)); -#ifdef CONFIG_SLUB_DEBUG /* * If we are Redzoning then check if there is some space between the * end of the object and the free pointer. If not then add an @@ -2343,7 +2308,7 @@ int kmem_ptr_validate(struct kmem_cache *s, const void *object) /* * We could also check if the object is on the slabs freelist. * But this would be too expensive and it seems that the main - * purpose of kmem_ptr_valid is to check if the object belongs + * purpose of kmem_ptr_valid() is to check if the object belongs * to a certain slab. */ return 1; @@ -2630,13 +2595,24 @@ void *__kmalloc(size_t size, gfp_t flags) } EXPORT_SYMBOL(__kmalloc); +static void *kmalloc_large_node(size_t size, gfp_t flags, int node) +{ + struct page *page = alloc_pages_node(node, flags | __GFP_COMP, + get_order(size)); + + if (page) + return page_address(page); + else + return NULL; +} + #ifdef CONFIG_NUMA void *__kmalloc_node(size_t size, gfp_t flags, int node) { struct kmem_cache *s; if (unlikely(size > PAGE_SIZE)) - return kmalloc_large(size, flags); + return kmalloc_large_node(size, flags, node); s = get_slab(size, flags); @@ -2653,19 +2629,17 @@ size_t ksize(const void *object) struct page *page; struct kmem_cache *s; - BUG_ON(!object); if (unlikely(object == ZERO_SIZE_PTR)) return 0; page = virt_to_head_page(object); - BUG_ON(!page); if (unlikely(!PageSlab(page))) return PAGE_SIZE << compound_order(page); s = page->slab; - BUG_ON(!s); +#ifdef CONFIG_SLUB_DEBUG /* * Debugging requires use of the padding between object * and whatever may come after it. @@ -2673,6 +2647,7 @@ size_t ksize(const void *object) if (s->flags & (SLAB_RED_ZONE | SLAB_POISON)) return s->objsize; +#endif /* * If we have the need to store the freelist pointer * back there or track user information then we can @@ -2680,7 +2655,6 @@ size_t ksize(const void *object) */ if (s->flags & (SLAB_DESTROY_BY_RCU | SLAB_STORE_USER)) return s->inuse; - /* * Else we can use all the padding etc for the allocation */ @@ -2957,7 +2931,7 @@ void __init kmem_cache_init(void) /* * Patch up the size_index table if we have strange large alignment * requirements for the kmalloc array. This is only the case for - * mips it seems. The standard arches will not generate any code here. + * MIPS it seems. The standard arches will not generate any code here. * * Largest permitted alignment is 256 bytes due to the way we * handle the index determination for the smaller caches. @@ -2986,7 +2960,6 @@ void __init kmem_cache_init(void) kmem_size = sizeof(struct kmem_cache); #endif - printk(KERN_INFO "SLUB: Genslabs=%d, HWalign=%d, Order=%d-%d, MinObjects=%d," " CPUs=%d, Nodes=%d\n", @@ -3083,12 +3056,15 @@ struct kmem_cache *kmem_cache_create(const char *name, size_t size, */ for_each_online_cpu(cpu) get_cpu_slab(s, cpu)->objsize = s->objsize; + s->inuse = max_t(int, s->inuse, ALIGN(size, sizeof(void *))); up_write(&slub_lock); + if (sysfs_slab_alias(s, name)) goto err; return s; } + s = kmalloc(kmem_size, GFP_KERNEL); if (s) { if (kmem_cache_open(s, GFP_KERNEL, name, @@ -3184,7 +3160,7 @@ void *__kmalloc_node_track_caller(size_t size, gfp_t gfpflags, struct kmem_cache *s; if (unlikely(size > PAGE_SIZE)) - return kmalloc_large(size, gfpflags); + return kmalloc_large_node(size, gfpflags, node); s = get_slab(size, gfpflags); @@ -3199,7 +3175,7 @@ static int validate_slab(struct kmem_cache *s, struct page *page, unsigned long *map) { void *p; - void *addr = slab_address(page); + void *addr = page_address(page); if (!check_slab(s, page) || !on_freelist(s, page, NULL)) @@ -3482,7 +3458,7 @@ static int add_location(struct loc_track *t, struct kmem_cache *s, static void process_slab(struct loc_track *t, struct kmem_cache *s, struct page *page, enum track_item alloc) { - void *addr = slab_address(page); + void *addr = page_address(page); DECLARE_BITMAP(map, s->objects); void *p; @@ -3591,8 +3567,8 @@ enum slab_stat_type { #define SO_CPU (1 << SL_CPU) #define SO_OBJECTS (1 << SL_OBJECTS) -static unsigned long slab_objects(struct kmem_cache *s, - char *buf, unsigned long flags) +static ssize_t show_slab_objects(struct kmem_cache *s, + char *buf, unsigned long flags) { unsigned long total = 0; int cpu; @@ -3602,6 +3578,8 @@ static unsigned long slab_objects(struct kmem_cache *s, unsigned long *per_cpu; nodes = kzalloc(2 * sizeof(unsigned long) * nr_node_ids, GFP_KERNEL); + if (!nodes) + return -ENOMEM; per_cpu = nodes + nr_node_ids; for_each_possible_cpu(cpu) { @@ -3754,25 +3732,25 @@ SLAB_ATTR_RO(aliases); static ssize_t slabs_show(struct kmem_cache *s, char *buf) { - return slab_objects(s, buf, SO_FULL|SO_PARTIAL|SO_CPU); + return show_slab_objects(s, buf, SO_FULL|SO_PARTIAL|SO_CPU); } SLAB_ATTR_RO(slabs); static ssize_t partial_show(struct kmem_cache *s, char *buf) { - return slab_objects(s, buf, SO_PARTIAL); + return show_slab_objects(s, buf, SO_PARTIAL); } SLAB_ATTR_RO(partial); static ssize_t cpu_slabs_show(struct kmem_cache *s, char *buf) { - return slab_objects(s, buf, SO_CPU); + return show_slab_objects(s, buf, SO_CPU); } SLAB_ATTR_RO(cpu_slabs); static ssize_t objects_show(struct kmem_cache *s, char *buf) { - return slab_objects(s, buf, SO_FULL|SO_PARTIAL|SO_CPU|SO_OBJECTS); + return show_slab_objects(s, buf, SO_FULL|SO_PARTIAL|SO_CPU|SO_OBJECTS); } SLAB_ATTR_RO(objects); @@ -3971,7 +3949,6 @@ SLAB_ATTR(remote_node_defrag_ratio); #endif #ifdef CONFIG_SLUB_STATS - static int show_stat(struct kmem_cache *s, char *buf, enum stat_item si) { unsigned long sum = 0; @@ -4155,8 +4132,8 @@ static struct kset *slab_kset; #define ID_STR_LENGTH 64 /* Create a unique string id for a slab cache: - * format - * :[flags-]size:[memory address of kmemcache] + * + * Format :[flags-]size */ static char *create_unique_id(struct kmem_cache *s) { @@ -176,7 +176,7 @@ void activate_page(struct page *page) SetPageActive(page); add_page_to_active_list(zone, page); __count_vm_event(PGACTIVATE); - mem_cgroup_move_lists(page_get_page_cgroup(page), true); + mem_cgroup_move_lists(page, true); } spin_unlock_irq(&zone->lru_lock); } diff --git a/mm/truncate.c b/mm/truncate.c index c35c49e..7d20ce4 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -134,8 +134,7 @@ invalidate_complete_page(struct address_space *mapping, struct page *page) } /** - * truncate_inode_pages - truncate range of pages specified by start and - * end byte offsets + * truncate_inode_pages - truncate range of pages specified by start & end byte offsets * @mapping: mapping to truncate * @lstart: offset from which to truncate * @lend: offset to which to truncate diff --git a/mm/vmscan.c b/mm/vmscan.c index a26dabd..4571158 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -126,7 +126,7 @@ long vm_total_pages; /* The total number of pages which the VM controls */ static LIST_HEAD(shrinker_list); static DECLARE_RWSEM(shrinker_rwsem); -#ifdef CONFIG_CGROUP_MEM_CONT +#ifdef CONFIG_CGROUP_MEM_RES_CTLR #define scan_global_lru(sc) (!(sc)->mem_cgroup) #else #define scan_global_lru(sc) (1) @@ -1128,7 +1128,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone, ClearPageActive(page); list_move(&page->lru, &zone->inactive_list); - mem_cgroup_move_lists(page_get_page_cgroup(page), false); + mem_cgroup_move_lists(page, false); pgmoved++; if (!pagevec_add(&pvec, page)) { __mod_zone_page_state(zone, NR_INACTIVE, pgmoved); @@ -1156,8 +1156,9 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone, VM_BUG_ON(PageLRU(page)); SetPageLRU(page); VM_BUG_ON(!PageActive(page)); + list_move(&page->lru, &zone->active_list); - mem_cgroup_move_lists(page_get_page_cgroup(page), true); + mem_cgroup_move_lists(page, true); pgmoved++; if (!pagevec_add(&pvec, page)) { __mod_zone_page_state(zone, NR_ACTIVE, pgmoved); @@ -1427,7 +1428,7 @@ unsigned long try_to_free_pages(struct zone **zones, int order, gfp_t gfp_mask) return do_try_to_free_pages(zones, gfp_mask, &sc); } -#ifdef CONFIG_CGROUP_MEM_CONT +#ifdef CONFIG_CGROUP_MEM_RES_CTLR unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont, gfp_t gfp_mask) |