diff options
Diffstat (limited to 'mm')
| -rw-r--r-- | mm/Kconfig | 22 | ||||
| -rw-r--r-- | mm/filemap.c | 14 | ||||
| -rw-r--r-- | mm/hugetlb.c | 2 | ||||
| -rw-r--r-- | mm/madvise.c | 9 | ||||
| -rw-r--r-- | mm/memory.c | 7 | ||||
| -rw-r--r-- | mm/mempolicy.c | 14 | ||||
| -rw-r--r-- | mm/mmap.c | 9 | ||||
| -rw-r--r-- | mm/mremap.c | 4 | ||||
| -rw-r--r-- | mm/oom_kill.c | 62 | ||||
| -rw-r--r-- | mm/page-writeback.c | 6 | ||||
| -rw-r--r-- | mm/page_alloc.c | 57 | ||||
| -rw-r--r-- | mm/readahead.c | 1 | ||||
| -rw-r--r-- | mm/rmap.c | 29 | ||||
| -rw-r--r-- | mm/shmem.c | 114 | ||||
| -rw-r--r-- | mm/slab.c | 1185 | ||||
| -rw-r--r-- | mm/sparse.c | 75 | ||||
| -rw-r--r-- | mm/swap_state.c | 6 | ||||
| -rw-r--r-- | mm/swapfile.c | 412 | ||||
| -rw-r--r-- | mm/vmalloc.c | 9 | ||||
| -rw-r--r-- | mm/vmscan.c | 20 |
20 files changed, 1316 insertions, 741 deletions
diff --git a/mm/Kconfig b/mm/Kconfig index cd379936cac..4e9937ac352 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -89,3 +89,25 @@ config NEED_MULTIPLE_NODES config HAVE_MEMORY_PRESENT def_bool y depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM + +# +# SPARSEMEM_EXTREME (which is the default) does some bootmem +# allocations when memory_present() is called. If this can not +# be done on your architecture, select this option. However, +# statically allocating the mem_section[] array can potentially +# consume vast quantities of .bss, so be careful. +# +# This option will also potentially produce smaller runtime code +# with gcc 3.4 and later. +# +config SPARSEMEM_STATIC + def_bool n + +# +# Architectecture platforms which require a two level mem_section in SPARSEMEM +# must select this option. This is usually for architecture platforms with +# an extremely sparse physical address space. +# +config SPARSEMEM_EXTREME + def_bool y + depends on SPARSEMEM && !SPARSEMEM_STATIC diff --git a/mm/filemap.c b/mm/filemap.c index c11418dd94e..88611928e71 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -54,9 +54,8 @@ * * ->i_mmap_lock (vmtruncate) * ->private_lock (__free_pte->__set_page_dirty_buffers) - * ->swap_list_lock - * ->swap_device_lock (exclusive_swap_page, others) - * ->mapping->tree_lock + * ->swap_lock (exclusive_swap_page, others) + * ->mapping->tree_lock * * ->i_sem * ->i_mmap_lock (truncate->unmap_mapping_range) @@ -86,7 +85,7 @@ * ->page_table_lock (anon_vma_prepare and various) * * ->page_table_lock - * ->swap_device_lock (try_to_unmap_one) + * ->swap_lock (try_to_unmap_one) * ->private_lock (try_to_unmap_one) * ->tree_lock (try_to_unmap_one) * ->zone.lru_lock (follow_page->mark_page_accessed) @@ -1505,8 +1504,12 @@ repeat: return -EINVAL; page = filemap_getpage(file, pgoff, nonblock); + + /* XXX: This is wrong, a filesystem I/O error may have happened. Fix that as + * done in shmem_populate calling shmem_getpage */ if (!page && !nonblock) return -ENOMEM; + if (page) { err = install_page(mm, vma, addr, page, prot); if (err) { @@ -1514,6 +1517,9 @@ repeat: return err; } } else { + /* No page was found just because we can't read it in now (being + * here implies nonblock != 0), but the page may exist, so set + * the PTE to fault it in later. */ err = install_file_pte(mm, vma, addr, pgoff, prot); if (err) return err; diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 6bf720bc662..901ac523a1c 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -360,8 +360,6 @@ int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma) ret = -ENOMEM; goto out; } - if (! pte_none(*pte)) - hugetlb_clean_stale_pgtable(pte); idx = ((addr - vma->vm_start) >> HPAGE_SHIFT) + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); diff --git a/mm/madvise.c b/mm/madvise.c index c8c01a12fea..4454936f87d 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -37,7 +37,7 @@ static long madvise_behavior(struct vm_area_struct * vma, if (new_flags == vma->vm_flags) { *prev = vma; - goto success; + goto out; } pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT); @@ -62,6 +62,7 @@ static long madvise_behavior(struct vm_area_struct * vma, goto out; } +success: /* * vm_flags is protected by the mmap_sem held in write mode. */ @@ -70,7 +71,6 @@ static long madvise_behavior(struct vm_area_struct * vma, out: if (error == -ENOMEM) error = -EAGAIN; -success: return error; } @@ -237,8 +237,9 @@ asmlinkage long sys_madvise(unsigned long start, size_t len_in, int behavior) * - different from the way of handling in mlock etc. */ vma = find_vma_prev(current->mm, start, &prev); - if (!vma && prev) - vma = prev->vm_next; + if (vma && start > vma->vm_start) + prev = vma; + for (;;) { /* Still start < end. */ error = -ENOMEM; diff --git a/mm/memory.c b/mm/memory.c index a596c117224..788a6281034 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -562,7 +562,8 @@ static void zap_pte_range(struct mmu_gather *tlb, pmd_t *pmd, page->index > details->last_index)) continue; } - ptent = ptep_get_and_clear(tlb->mm, addr, pte); + ptent = ptep_get_and_clear_full(tlb->mm, addr, pte, + tlb->fullmm); tlb_remove_tlb_entry(tlb, pte, addr); if (unlikely(!page)) continue; @@ -590,7 +591,7 @@ static void zap_pte_range(struct mmu_gather *tlb, pmd_t *pmd, continue; if (!pte_file(ptent)) free_swap_and_cache(pte_to_swp_entry(ptent)); - pte_clear(tlb->mm, addr, pte); + pte_clear_full(tlb->mm, addr, pte, tlb->fullmm); } while (pte++, addr += PAGE_SIZE, addr != end); pte_unmap(pte - 1); } @@ -1955,7 +1956,7 @@ static int do_file_page(struct mm_struct * mm, struct vm_area_struct * vma, * Fall back to the linear mapping if the fs does not support * ->populate: */ - if (!vma->vm_ops || !vma->vm_ops->populate || + if (!vma->vm_ops->populate || (write_access && !(vma->vm_flags & VM_SHARED))) { pte_clear(mm, address, pte); return do_no_page(mm, vma, address, write_access, pte, pmd); diff --git a/mm/mempolicy.c b/mm/mempolicy.c index b4eababc819..afa06e184d8 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -88,7 +88,7 @@ static kmem_cache_t *sn_cache; policied. */ static int policy_zone; -static struct mempolicy default_policy = { +struct mempolicy default_policy = { .refcnt = ATOMIC_INIT(1), /* never free it */ .policy = MPOL_DEFAULT, }; @@ -664,10 +664,10 @@ asmlinkage long compat_sys_mbind(compat_ulong_t start, compat_ulong_t len, #endif /* Return effective policy for a VMA */ -static struct mempolicy * -get_vma_policy(struct vm_area_struct *vma, unsigned long addr) +struct mempolicy * +get_vma_policy(struct task_struct *task, struct vm_area_struct *vma, unsigned long addr) { - struct mempolicy *pol = current->mempolicy; + struct mempolicy *pol = task->mempolicy; if (vma) { if (vma->vm_ops && vma->vm_ops->get_policy) @@ -786,7 +786,7 @@ static struct page *alloc_page_interleave(unsigned int __nocast gfp, unsigned or struct page * alloc_page_vma(unsigned int __nocast gfp, struct vm_area_struct *vma, unsigned long addr) { - struct mempolicy *pol = get_vma_policy(vma, addr); + struct mempolicy *pol = get_vma_policy(current, vma, addr); cpuset_update_current_mems_allowed(); @@ -908,7 +908,7 @@ void __mpol_free(struct mempolicy *p) /* Find first node suitable for an allocation */ int mpol_first_node(struct vm_area_struct *vma, unsigned long addr) { - struct mempolicy *pol = get_vma_policy(vma, addr); + struct mempolicy *pol = get_vma_policy(current, vma, addr); switch (pol->policy) { case MPOL_DEFAULT: @@ -928,7 +928,7 @@ int mpol_first_node(struct vm_area_struct *vma, unsigned long addr) /* Find secondary valid nodes for an allocation */ int mpol_node_valid(int nid, struct vm_area_struct *vma, unsigned long addr) { - struct mempolicy *pol = get_vma_policy(vma, addr); + struct mempolicy *pol = get_vma_policy(current, vma, addr); switch (pol->policy) { case MPOL_PREFERRED: diff --git a/mm/mmap.c b/mm/mmap.c index 404319477e7..12334aecf8a 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -61,7 +61,7 @@ pgprot_t protection_map[16] = { int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */ int sysctl_overcommit_ratio = 50; /* default is 50% */ -int sysctl_max_map_count = DEFAULT_MAX_MAP_COUNT; +int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT; atomic_t vm_committed_space = ATOMIC_INIT(0); /* @@ -203,13 +203,6 @@ static void remove_vm_struct(struct vm_area_struct *vma) kmem_cache_free(vm_area_cachep, vma); } -/* - * sys_brk() for the most part doesn't need the global kernel - * lock, except when an application is doing something nasty - * like trying to un-brk an area that has already been mapped - * to a regular file. in this case, the unmapping will need - * to invoke file system routines that need the global lock. - */ asmlinkage unsigned long sys_brk(unsigned long brk) { unsigned long rlim, retval; diff --git a/mm/mremap.c b/mm/mremap.c index fc45dc9a617..a32fed454bd 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -141,6 +141,10 @@ move_one_page(struct vm_area_struct *vma, unsigned long old_addr, if (dst) { pte_t pte; pte = ptep_clear_flush(vma, old_addr, src); + /* ZERO_PAGE can be dependant on virtual addr */ + if (pfn_valid(pte_pfn(pte)) && + pte_page(pte) == ZERO_PAGE(old_addr)) + pte = pte_wrprotect(mk_pte(ZERO_PAGE(new_addr), new_vma->vm_page_prot)); set_pte_at(mm, new_addr, dst, pte); } else error = -ENOMEM; diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 1e56076672f..5ec8da12cfd 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -6,8 +6,8 @@ * for goading me into coding this file... * * The routines in this file are used to kill a process when - * we're seriously out of memory. This gets called from kswapd() - * in linux/mm/vmscan.c when we really run out of memory. + * we're seriously out of memory. This gets called from __alloc_pages() + * in mm/page_alloc.c when we really run out of memory. * * Since we won't call these routines often (on a well-configured * machine) this file will double as a 'coding guide' and a signpost @@ -20,13 +20,14 @@ #include <linux/swap.h> #include <linux/timex.h> #include <linux/jiffies.h> +#include <linux/cpuset.h> /* #define DEBUG */ /** * oom_badness - calculate a numeric value for how bad this task has been * @p: task struct of which task we should calculate - * @p: current uptime in seconds + * @uptime: current uptime in seconds * * The formula used is relatively simple and documented inline in the * function. The main rationale is that we want to select a good task @@ -57,9 +58,9 @@ unsigned long badness(struct task_struct *p, unsigned long uptime) /* * Processes which fork a lot of child processes are likely - * a good choice. We add the vmsize of the childs if they + * a good choice. We add the vmsize of the children if they * have an own mm. This prevents forking servers to flood the - * machine with an endless amount of childs + * machine with an endless amount of children */ list_for_each(tsk, &p->children) { struct task_struct *chld; @@ -143,28 +144,36 @@ static struct task_struct * select_bad_process(void) struct timespec uptime; do_posix_clock_monotonic_gettime(&uptime); - do_each_thread(g, p) + do_each_thread(g, p) { + unsigned long points; + int releasing; + /* skip the init task with pid == 1 */ - if (p->pid > 1 && p->oomkilladj != OOM_DISABLE) { - unsigned long points; - - /* - * This is in the process of releasing memory so wait it - * to finish before killing some other task by mistake. - */ - if ((unlikely(test_tsk_thread_flag(p, TIF_MEMDIE)) || (p->flags & PF_EXITING)) && - !(p->flags & PF_DEAD)) - return ERR_PTR(-1UL); - if (p->flags & PF_SWAPOFF) - return p; - - points = badness(p, uptime.tv_sec); - if (points > maxpoints || !chosen) { - chosen = p; - maxpoints = points; - } + if (p->pid == 1) + continue; + if (p->oomkilladj == OOM_DISABLE) + continue; + /* If p's nodes don't overlap ours, it won't help to kill p. */ + if (!cpuset_excl_nodes_overlap(p)) + continue; + + /* + * This is in the process of releasing memory so for wait it + * to finish before killing some other task by mistake. + */ + releasing = test_tsk_thread_flag(p, TIF_MEMDIE) || + p->flags & PF_EXITING; + if (releasing && !(p->flags & PF_DEAD)) + return ERR_PTR(-1UL); + if (p->flags & PF_SWAPOFF) + return p; + + points = badness(p, uptime.tv_sec); + if (points > maxpoints || !chosen) { + chosen = p; + maxpoints = points; } - while_each_thread(g, p); + } while_each_thread(g, p); return chosen; } @@ -189,7 +198,8 @@ static void __oom_kill_task(task_t *p) return; } task_unlock(p); - printk(KERN_ERR "Out of Memory: Killed process %d (%s).\n", p->pid, p->comm); + printk(KERN_ERR "Out of Memory: Killed process %d (%s).\n", + p->pid, p->comm); /* * We give our sacrificial lamb high priority and access to diff --git a/mm/page-writeback.c b/mm/page-writeback.c index a6329fa8f86..0166ea15c9e 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -368,10 +368,8 @@ int wakeup_pdflush(long nr_pages) static void wb_timer_fn(unsigned long unused); static void laptop_timer_fn(unsigned long unused); -static struct timer_list wb_timer = - TIMER_INITIALIZER(wb_timer_fn, 0, 0); -static struct timer_list laptop_mode_wb_timer = - TIMER_INITIALIZER(laptop_timer_fn, 0, 0); +static DEFINE_TIMER(wb_timer, wb_timer_fn, 0, 0); +static DEFINE_TIMER(laptop_mode_wb_timer, laptop_timer_fn, 0, 0); /* * Periodic writeback of "old" data. diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8d088371196..3974fd81d27 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -42,13 +42,13 @@ * MCD - HACK: Find somewhere to initialize this EARLY, or make this * initializer cleaner */ -nodemask_t node_online_map = { { [0] = 1UL } }; +nodemask_t node_online_map __read_mostly = { { [0] = 1UL } }; EXPORT_SYMBOL(node_online_map); -nodemask_t node_possible_map = NODE_MASK_ALL; +nodemask_t node_possible_map __read_mostly = NODE_MASK_ALL; EXPORT_SYMBOL(node_possible_map); -struct pglist_data *pgdat_list; -unsigned long totalram_pages; -unsigned long totalhigh_pages; +struct pglist_data *pgdat_list __read_mostly; +unsigned long totalram_pages __read_mostly; +unsigned long totalhigh_pages __read_mostly; long nr_swap_pages; /* @@ -68,7 +68,7 @@ EXPORT_SYMBOL(nr_swap_pages); * Used by page_zone() to look up the address of the struct zone whose * id is encoded in the upper bits of page->flags */ -struct zone *zone_table[1 << ZONETABLE_SHIFT]; +struct zone *zone_table[1 << ZONETABLE_SHIFT] __read_mostly; EXPORT_SYMBOL(zone_table); static char *zone_names[MAX_NR_ZONES] = { "DMA", "Normal", "HighMem" }; @@ -329,7 +329,7 @@ static inline void free_pages_check(const char *function, struct page *page) 1 << PG_writeback ))) bad_page(function, page); if (PageDirty(page)) - ClearPageDirty(page); + __ClearPageDirty(page); } /* @@ -806,11 +806,14 @@ __alloc_pages(unsigned int __nocast gfp_mask, unsigned int order, classzone_idx = zone_idx(zones[0]); restart: - /* Go through the zonelist once, looking for a zone with enough free */ + /* + * Go through the zonelist once, looking for a zone with enough free. + * See also cpuset_zone_allowed() comment in kernel/cpuset.c. + */ for (i = 0; (z = zones[i]) != NULL; i++) { int do_reclaim = should_reclaim_zone(z, gfp_mask); - if (!cpuset_zone_allowed(z)) + if (!cpuset_zone_allowed(z, __GFP_HARDWALL)) continue; /* @@ -845,6 +848,7 @@ zone_reclaim_retry: * * This is the last chance, in general, before the goto nopage. * Ignore cpuset if GFP_ATOMIC (!wait) rather than fail alloc. + * See also cpuset_zone_allowed() comment in kernel/cpuset.c. */ for (i = 0; (z = zones[i]) != NULL; i++) { if (!zone_watermark_ok(z, order, z->pages_min, @@ -852,7 +856,7 @@ zone_reclaim_retry: gfp_mask & __GFP_HIGH)) continue; - if (wait && !cpuset_zone_allowed(z)) + if (wait && !cpuset_zone_allowed(z, gfp_mask)) continue; page = buffered_rmqueue(z, order, gfp_mask); @@ -867,7 +871,7 @@ zone_reclaim_retry: if (!(gfp_mask & __GFP_NOMEMALLOC)) { /* go through the zonelist yet again, ignoring mins */ for (i = 0; (z = zones[i]) != NULL; i++) { - if (!cpuset_zone_allowed(z)) + if (!cpuset_zone_allowed(z, gfp_mask)) continue; page = buffered_rmqueue(z, order, gfp_mask); if (page) @@ -903,7 +907,7 @@ rebalance: gfp_mask & __GFP_HIGH)) continue; - if (!cpuset_zone_allowed(z)) + if (!cpuset_zone_allowed(z, gfp_mask)) continue; page = buffered_rmqueue(z, order, gfp_mask); @@ -922,7 +926,7 @@ rebalance: classzone_idx, 0, 0)) continue; - if (!cpuset_zone_allowed(z)) + if (!cpuset_zone_allowed(z, __GFP_HARDWALL)) continue; page = buffered_rmqueue(z, order, gfp_mask); @@ -1130,19 +1134,20 @@ EXPORT_SYMBOL(nr_pagecache); DEFINE_PER_CPU(long, nr_pagecache_local) = 0; #endif -void __get_page_state(struct page_state *ret, int nr) +void __get_page_state(struct page_state *ret, int nr, cpumask_t *cpumask) { int cpu = 0; memset(ret, 0, sizeof(*ret)); + cpus_and(*cpumask, *cpumask, cpu_online_map); - cpu = first_cpu(cpu_online_map); + cpu = first_cpu(*cpumask); while (cpu < NR_CPUS) { unsigned long *in, *out, off; in = (unsigned long *)&per_cpu(page_states, cpu); - cpu = next_cpu(cpu, cpu_online_map); + cpu = next_cpu(cpu, *cpumask); if (cpu < NR_CPUS) prefetch(&per_cpu(page_states, cpu)); @@ -1153,19 +1158,33 @@ void __get_page_state(struct page_state *ret, int nr) } } +void get_page_state_node(struct page_state *ret, int node) +{ + int nr; + cpumask_t mask = node_to_cpumask(node); + + nr = offsetof(struct page_state, GET_PAGE_STATE_LAST); + nr /= sizeof(unsigned long); + + __get_page_state(ret, nr+1, &mask); +} + void get_page_state(struct page_state *ret) { int nr; + cpumask_t mask = CPU_MASK_ALL; nr = offsetof(struct page_state, GET_PAGE_STATE_LAST); nr /= sizeof(unsigned long); - __get_page_state(ret, nr + 1); + __get_page_state(ret, nr + 1, &mask); } void get_full_page_state(struct page_state *ret) { - __get_page_state(ret, sizeof(*ret) / sizeof(unsigned long)); + cpumask_t mask = CPU_MASK_ALL; + + __get_page_state(ret, sizeof(*ret) / sizeof(unsigned long), &mask); } unsigned long __read_page_state(unsigned long offset) @@ -1909,7 +1928,7 @@ static void __init free_area_init_core(struct pglist_data *pgdat, zone->nr_scan_inactive = 0; zone->nr_active = 0; zone->nr_inactive = 0; - atomic_set(&zone->reclaim_in_progress, -1); + atomic_set(&zone->reclaim_in_progress, 0); if (!size) continue; diff --git a/mm/readahead.c b/mm/readahead.c index b840e7c6ea7..d0b50034e24 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -540,6 +540,7 @@ void handle_ra_miss(struct address_space *mapping, { ra->flags |= RA_FLAG_MISS; ra->flags &= ~RA_FLAG_INCACHE; + ra->cache_hit = 0; } /* diff --git a/mm/rmap.c b/mm/rmap.c index 08ac5c7fa91..450f5241b5a 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -34,9 +34,8 @@ * anon_vma->lock * mm->page_table_lock * zone->lru_lock (in mark_page_accessed) - * swap_list_lock (in swap_free etc's swap_info_get) + * swap_lock (in swap_duplicate, swap_info_get) * mmlist_lock (in mmput, drain_mmlist and others) - * swap_device_lock (in swap_duplicate, swap_info_get) * mapping->private_lock (in __set_page_dirty_buffers) * inode_lock (in set_page_dirty's __mark_inode_dirty) * sb_lock (within inode_lock in fs/fs-writeback.c) @@ -290,8 +289,6 @@ static int page_referenced_one(struct page *page, pte_t *pte; int referenced = 0; - if (!get_mm_counter(mm, rss)) - goto out; address = vma_address(page, vma); if (address == -EFAULT) goto out; @@ -442,22 +439,19 @@ int page_referenced(struct page *page, int is_locked, int ignore_token) void page_add_anon_rmap(struct page *page, struct vm_area_struct *vma, unsigned long address) { - struct anon_vma *anon_vma = vma->anon_vma; - pgoff_t index; - BUG_ON(PageReserved(page)); - BUG_ON(!anon_vma); inc_mm_counter(vma->vm_mm, anon_rss); - anon_vma = (void *) anon_vma + PAGE_MAPPING_ANON; - index = (address - vma->vm_start) >> PAGE_SHIFT; - index += vma->vm_pgoff; - index >>= PAGE_CACHE_SHIFT - PAGE_SHIFT; - if (atomic_inc_and_test(&page->_mapcount)) { - page->index = index; + struct anon_vma *anon_vma = vma->anon_vma; + + BUG_ON(!anon_vma); + anon_vma = (void *) anon_vma + PAGE_MAPPING_ANON; page->mapping = (struct address_space *) anon_vma; + + page->index = linear_page_index(vma, address); + inc_page_state(nr_mapped); } /* else checking page index and mapping is racy */ @@ -518,8 +512,6 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma) pte_t pteval; int ret = SWAP_AGAIN; - if (!get_mm_counter(mm, rss)) - goto out; address = vma_address(page, vma); if (address == -EFAULT) goto out; @@ -532,6 +524,8 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma) * If the page is mlock()d, we cannot swap it out. * If it's recently referenced (perhaps page_referenced * skipped over this mm) then we should reactivate it. + * + * Pages belonging to VM_RESERVED regions should not happen here. */ if ((vma->vm_flags & (VM_LOCKED|VM_RESERVED)) || ptep_clear_flush_young(vma, address, pte)) { @@ -767,8 +761,7 @@ static int try_to_unmap_file(struct page *page) if (vma->vm_flags & (VM_LOCKED|VM_RESERVED)) continue; cursor = (unsigned long) vma->vm_private_data; - while (get_mm_counter(vma->vm_mm, rss) && - cursor < max_nl_cursor && + while ( cursor < max_nl_cursor && cursor < vma->vm_end - vma->vm_start) { try_to_unmap_cluster(cursor, &mapcount, vma); cursor += CLUSTER_SIZE; diff --git a/mm/shmem.c b/mm/shmem.c index 5a81b1ee4f7..1f7aeb210c7 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -45,7 +45,6 @@ #include <linux/swapops.h> #include <linux/mempolicy.h> #include <linux/namei.h> -#include <linux/xattr.h> #include <asm/uaccess.h> #include <asm/div64.h> #include <asm/pgtable.h> @@ -179,10 +178,9 @@ static struct address_space_operations shmem_aops; static struct file_operations shmem_file_operations; static struct inode_operations shmem_inode_operations; static struct inode_operations shmem_dir_inode_operations; -static struct inode_operations shmem_special_inode_operations; static struct vm_operations_struct shmem_vm_ops; -static struct backing_dev_info shmem_backing_dev_info = { +static struct backing_dev_info shmem_backing_dev_info __read_mostly = { .ra_pages = 0, /* No readahead */ .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, .unplug_io_fn = default_unplug_io_fn, @@ -668,6 +666,7 @@ static void shmem_delete_inode(struct inode *inode) struct shmem_inode_info *info = SHMEM_I(inode); if (inode->i_op->truncate == shmem_truncate) { + truncate_inode_pages(inode->i_mapping, 0); shmem_unacct_size(info->flags, inode->i_size); inode->i_size = 0; shmem_truncate(inode); @@ -1195,6 +1194,7 @@ static int shmem_populate(struct vm_area_struct *vma, err = shmem_getpage(inode, pgoff, &page, sgp, NULL); if (err) return err; + /* Page may still be null, but only if nonblock was set. */ if (page) { mark_page_accessed(page); err = install_page(mm, vma, addr, page, prot); @@ -1202,7 +1202,10 @@ static int shmem_populate(struct vm_area_struct *vma, page_cache_release(page); return err; } - } else if (nonblock) { + } else { + /* No page was found just because we can't read it in + * now (being here implies nonblock != 0), but the page + * may exist, so set the PTE to fault it in later. */ err = install_file_pte(mm, vma, addr, pgoff, prot); if (err) return err; @@ -1296,7 +1299,6 @@ shmem_get_inode(struct super_block *sb, int mode, dev_t dev) switch (mode & S_IFMT) { default: - inode->i_op = &shmem_special_inode_operations; init_special_inode(inode, mode, dev); break; case S_IFREG: @@ -1606,6 +1608,15 @@ shmem_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) int error = -ENOSPC; if (inode) { + error = security_inode_init_security(inode, dir, NULL, NULL, + NULL); + if (error) { + if (error != -EOPNOTSUPP) { + iput(inode); + return error; + } + error = 0; + } if (dir->i_mode & S_ISGID) { inode->i_gid = dir->i_gid; if (S_ISDIR(mode)) @@ -1615,7 +1626,6 @@ shmem_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) dir->i_ctime = dir->i_mtime = CURRENT_TIME; d_instantiate(dentry, inode); dget(dentry); /* Extra count - pin the dentry in core */ - error = 0; } return error; } @@ -1745,6 +1755,16 @@ static int shmem_symlink(struct inode *dir, struct dentry *dentry, const char *s if (!inode) return -ENOSPC; + error = security_inode_init_security(inode, dir, NULL, NULL, + NULL); + if (error) { + if (error != -EOPNOTSUPP) { + iput(inode); + return error; + } + error = 0; + } + info = SHMEM_I(inode); inode->i_size = len-1; if (len <= (char *)inode - (char *)info) { @@ -1800,12 +1820,6 @@ static void shmem_put_link(struct dentry *dentry, struct nameidata *nd, void *co static struct inode_operations shmem_symlink_inline_operations = { .readlink = generic_readlink, .follow_link = shmem_follow_link_inline, -#ifdef CONFIG_TMPFS_XATTR - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, - .listxattr = generic_listxattr, - .removexattr = generic_removexattr, -#endif }; static struct inode_operations shmem_symlink_inode_operations = { @@ -1813,12 +1827,6 @@ static struct inode_operations shmem_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = shmem_follow_link, .put_link = shmem_put_link, -#ifdef CONFIG_TMPFS_XATTR - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, - .listxattr = generic_listxattr, - .removexattr = generic_removexattr, -#endif }; static int shmem_parse_options(char *options, int *mode, uid_t *uid, gid_t *gid, unsigned long *blocks, unsigned long *inodes) @@ -1938,12 +1946,6 @@ static void shmem_put_super(struct super_block *sb) sb->s_fs_info = NULL; } -#ifdef CONFIG_TMPFS_XATTR -static struct xattr_handler *shmem_xattr_handlers[]; -#else -#define shmem_xattr_handlers NULL -#endif - static int shmem_fill_super(struct super_block *sb, void *data, int silent) { @@ -1994,7 +1996,6 @@ static int shmem_fill_super(struct super_block *sb, sb->s_blocksize_bits = PAGE_CACHE_SHIFT; sb->s_magic = TMPFS_MAGIC; sb->s_op = &shmem_ops; - sb->s_xattr = shmem_xattr_handlers; inode = shmem_get_inode(sb, S_IFDIR | mode, 0); if (!inode) @@ -2083,12 +2084,6 @@ static struct file_operations shmem_file_operations = { static struct inode_operations shmem_inode_operations = { .truncate = shmem_truncate, .setattr = shmem_notify_change, -#ifdef CONFIG_TMPFS_XATTR - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, - .listxattr = generic_listxattr, - .removexattr = generic_removexattr, -#endif }; static struct inode_operations shmem_dir_inode_operations = { @@ -2102,21 +2097,6 @@ static struct inode_operations shmem_dir_inode_operations = { .rmdir = shmem_rmdir, .mknod = shmem_mknod, .rename = shmem_rename, -#ifdef CONFIG_TMPFS_XATTR - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, - .listxattr = generic_listxattr, - .removexattr = generic_removexattr, -#endif -#endif -}; - -static struct inode_operations shmem_special_inode_operations = { -#ifdef CONFIG_TMPFS_XATTR - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, - .listxattr = generic_listxattr, - .removexattr = generic_removexattr, #endif }; @@ -2142,48 +2122,6 @@ static struct vm_operations_struct shmem_vm_ops = { }; -#ifdef CONFIG_TMPFS_SECURITY - -static size_t shmem_xattr_security_list(struct inode *inode, char *list, size_t list_len, - const char *name, size_t name_len) -{ - return security_inode_listsecurity(inode, list, list_len); -} - -static int shmem_xattr_security_get(struct inode *inode, const char *name, void *buffer, size_t size) -{ - if (strcmp(name, "") == 0) - return -EINVAL; - return security_inode_getsecurity(inode, name, buffer, size); -} - -static int shmem_xattr_security_set(struct inode *inode, const char *name, const void *value, size_t size, int flags) -{ - if (strcmp(name, "") == 0) - return -EINVAL; - return security_inode_setsecurity(inode, name, value, size, flags); -} - -static struct xattr_handler shmem_xattr_security_handler = { - .prefix = XATTR_SECURITY_PREFIX, - .list = shmem_xattr_security_list, - .get = shmem_xattr_security_get, - .set = shmem_xattr_security_set, -}; - -#endif /* CONFIG_TMPFS_SECURITY */ - -#ifdef CONFIG_TMPFS_XATTR - -static struct xattr_handler *shmem_xattr_handlers[] = { -#ifdef CONFIG_TMPFS_SECURITY - &shmem_xattr_security_handler, -#endif - NULL -}; - -#endif /* CONFIG_TMPFS_XATTR */ - static struct super_block *shmem_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, void *data) { diff --git a/mm/slab.c b/mm/slab.c index c9e706db463..05a391059fe 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -75,6 +75,15 @@ * * At present, each engine can be growing a cache. This should be blocked. * + * 15 March 2005. NUMA slab allocator. + * Shai Fultheim <shai@scalex86.org>. + * Shobhit Dayal <shobhit@calsoftinc.com> + * Alok N Kataria <alokk@calsoftinc.com> + * Christoph Lameter <christoph@lameter.com> + * + * Modified the slab allocator to be node aware on NUMA systems. + * Each node has its own list of partial, free and full slabs. + * All object allocations for a node occur from node specific slab lists. */ #include <linux/config.h> @@ -93,6 +102,7 @@ #include <linux/module.h> #include <linux/rcupdate.h> #include <linux/string.h> +#include <linux/nodemask.h> #include <asm/uaccess.h> #include <asm/cacheflush.h> @@ -189,6 +199,7 @@ * is less than 512 (PAGE_SIZE<<3), but greater than 256. */ +typedef unsigned int kmem_bufctl_t; #define BUFCTL_END (((kmem_bufctl_t)(~0U))-0) #define BUFCTL_FREE (((kmem_bufctl_t)(~0U))-1) #define SLAB_LIMIT (((kmem_bufctl_t)(~0U))-2) @@ -211,6 +222,7 @@ struct slab { void *s_mem; /* including colour offset */ unsigned int inuse; /* num of objs active in slab */ kmem_bufctl_t free; + unsigned short nodeid; }; /* @@ -238,7 +250,6 @@ struct slab_rcu { /* * struct array_cache * - * Per cpu structures * Purpose: * - LIFO ordering, to hand out cache-warm objects from _alloc * - reduce the number of linked list operations @@ -253,6 +264,13 @@ struct array_cache { unsigned int limit; unsigned int batchcount; unsigned int touched; + spinlock_t lock; + void *entry[0]; /* + * Must have this definition in here for the proper + * alignment of array_cache. Also simplifies accessing + * the entries. + * [0] is for gcc 2.95. It should really be []. + */ }; /* bootstrap: The caches do not work without cpuarrays anymore, @@ -265,34 +283,83 @@ struct arraycache_init { }; /* - * The slab lists of all objects. - * Hopefully reduce the internal fragmentation - * NUMA: The spinlock could be moved from the kmem_cache_t - * into this structure, too. Figure out what causes - * fewer cross-node spinlock operations. + * The slab lists for all objects. */ struct kmem_list3 { struct list_head slabs_partial; /* partial list first, better asm code */ struct list_head slabs_full; struct list_head slabs_free; unsigned long free_objects; - int free_touched; unsigned long next_reap; - struct array_cache *shared; + int free_touched; + unsigned int free_limit; + spinlock_t list_lock; + struct array_cache *shared; /* shared per node */ + struct array_cache **alien; /* on other nodes */ }; -#define LIST3_INIT(parent) \ - { \ - .slabs_full = LIST_HEAD_INIT(parent.slabs_full), \ - .slabs_partial = LIST_HEAD_INIT(parent.slabs_partial), \ - .slabs_free = LIST_HEAD_INIT(parent.slabs_free) \ +/* + * Need this for bootstrapping a per node allocator. + */ +#define NUM_INIT_LISTS (2 * MAX_NUMNODES + 1) +struct kmem_list3 __initdata initkmem_list3[NUM_INIT_LISTS]; +#define CACHE_CACHE 0 +#define SIZE_AC 1 +#define SIZE_L3 (1 + MAX_NUMNODES) + +/* + * This function may be completely optimized away if + * a constant is passed to it. Mostly the same as + * what is in linux/slab.h except it returns an + * index. + */ +static inline int index_of(const size_t size) +{ + if (__builtin_constant_p(size)) { + int i = 0; + +#define CACHE(x) \ + if (size <=x) \ + return i; \ + else \ + i++; +#include "linux/kmalloc_sizes.h" +#undef CACHE + { + extern void __bad_size(void); + __bad_size(); + } } -#define list3_data(cachep) \ - (&(cachep)->lists) + return 0; +} + +#define INDEX_AC index_of(sizeof(struct arraycache_init)) +#define INDEX_L3 index_of(sizeof(struct kmem_list3)) + +static inline void kmem_list3_init(struct kmem_list3 *parent) +{ + INIT_LIST_HEAD(&parent->slabs_full); + INIT_LIST_HEAD(&parent->slabs_partial); + INIT_LIST_HEAD(&parent->slabs_free); + parent->shared = NULL; + parent->alien = NULL; + spin_lock_init(&parent->list_lock); + parent->free_objects = 0; + parent->free_touched = 0; +} -/* NUMA: per-node */ -#define list3_data_ptr(cachep, ptr) \ - list3_data(cachep) +#define MAKE_LIST(cachep, listp, slab, nodeid) \ + do { \ + INIT_LIST_HEAD(listp); \ + list_splice(&(cachep->nodelists[nodeid]->slab), listp); \ + } while (0) + +#define MAKE_ALL_LISTS(cachep, ptr, nodeid) \ + do { \ + MAKE_LIST((cachep), (&(ptr)->slabs_full), slabs_full, nodeid); \ + MAKE_LIST((cachep), (&(ptr)->slabs_partial), slabs_partial, nodeid); \ + MAKE_LIST((cachep), (&(ptr)->slabs_free), slabs_free, nodeid); \ + } while (0) /* * kmem_cache_t @@ -305,13 +372,12 @@ struct kmem_cache_s { struct array_cache *array[NR_CPUS]; unsigned int batchcount; unsigned int limit; -/* 2) touched by every alloc & free from the backend */ - struct kmem_list3 lists; - /* NUMA: kmem_3list_t *nodelists[MAX_NUMNODES] */ + unsigned int shared; unsigned int objsize; +/* 2) touched by every alloc & free from the backend */ + struct kmem_list3 *nodelists[MAX_NUMNODES]; unsigned int flags; /* constant flags */ unsigned int num; /* # of objs per slab */ - unsigned int free_limit; /* upper limit of objects in the lists */ spinlock_t spinlock; /* 3) cache_grow/shrink */ @@ -348,6 +414,7 @@ struct kmem_cache_s { unsigned long errors; unsigned long max_freeable; unsigned long node_allocs; + unsigned long node_frees; atomic_t allochit; atomic_t allocmiss; atomic_t freehit; @@ -383,6 +450,7 @@ struct kmem_cache_s { } while (0) #define STATS_INC_ERR(x) ((x)->errors++) #define STATS_INC_NODEALLOCS(x) ((x)->node_allocs++) +#define STATS_INC_NODEFREES(x) ((x)->node_frees++) #define STATS_SET_FREEABLE(x, i) \ do { if ((x)->max_freeable < i) \ (x)->max_freeable = i; \ @@ -401,6 +469,7 @@ struct kmem_cache_s { #define STATS_SET_HIGH(x) do { } while (0) #define STATS_INC_ERR(x) do { } while (0) #define STATS_INC_NODEALLOCS(x) do { } while (0) +#define STATS_INC_NODEFREES(x) do { } while (0) #define STATS_SET_FREEABLE(x, i) \ do { } while (0) @@ -533,9 +602,9 @@ static struct arraycache_init initarray_generic = /* internal cache of cache description objs */ static kmem_cache_t cache_cache = { - .lists = LIST3_INIT(cache_cache.lists), .batchcount = 1, .limit = BOOT_CPUCACHE_ENTRIES, + .shared = 1, .objsize = sizeof(kmem_cache_t), .flags = SLAB_NO_REAP, .spinlock = SPIN_LOCK_UNLOCKED, @@ -556,7 +625,6 @@ static struct list_head cache_chain; * SLAB_RECLAIM_ACCOUNT turns this on per-slab */ atomic_t slab_reclaim_pages; -EXPORT_SYMBOL(slab_reclaim_pages); /* * chicken and egg problem: delay the per-cpu array allocation @@ -564,7 +632,8 @@ EXPORT_SYMBOL(slab_reclaim_pages); */ static enum { NONE, - PARTIAL, + PARTIAL_AC, + PARTIAL_L3, FULL } g_cpucache_up; @@ -573,11 +642,7 @@ static DEFINE_PER_CPU(struct work_struct, reap_work); static void free_block(kmem_cache_t* cachep, void** objpp, int len); static void enable_cpucache (kmem_cache_t *cachep); static void cache_reap (void *unused); - -static inline void **ac_entry(struct array_cache *ac) -{ - return (void**)(ac+1); -} +static int __node_shrink(kmem_cache_t *cachep, int node); static inline struct array_cache *ac_data(kmem_cache_t *cachep) { @@ -600,7 +665,7 @@ static inline kmem_cache_t *__find_general_cachep(size_t size, csizep++; /* - * Really subtile: The last entry with cs->cs_size==ULONG_MAX + * Really subtle: The last entry with cs->cs_size==ULONG_MAX * has cs_{dma,}cachep==NULL. Thus no special case * for large kmalloc calls required. */ @@ -675,48 +740,160 @@ static void __devinit start_cpu_timer(int cpu) } } -static struct array_cache *alloc_arraycache(int cpu, int entries, +static struct array_cache *alloc_arraycache(int node, int entries, int batchcount) { int memsize = sizeof(void*)*entries+sizeof(struct array_cache); struct array_cache *nc = NULL; - if (cpu == -1) - nc = kmalloc(memsize, GFP_KERNEL); - else - nc = kmalloc_node(memsize, GFP_KERNEL, cpu_to_node(cpu)); - + nc = kmalloc_node(memsize, GFP_KERNEL, node); if (nc) { nc->avail = 0; nc->limit = entries; nc->batchcount = batchcount; nc->touched = 0; + spin_lock_init(&nc->lock); } return nc; } +#ifdef CONFIG_NUMA +static inline struct array_cache **alloc_alien_cache(int node, int limit) +{ + struct array_cache **ac_ptr; + int memsize = sizeof(void*)*MAX_NUMNODES; + int i; + + if (limit > 1) + limit = 12; + ac_ptr = kmalloc_node(memsize, GFP_KERNEL, node); + if (ac_ptr) { + for_each_node(i) { + if (i == node || !node_online(i)) { + ac_ptr[i] = NULL; + continue; + } + ac_ptr[i] = alloc_arraycache(node, limit, 0xbaadf00d); + if (!ac_ptr[i]) { + for (i--; i <=0; i--) + kfree(ac_ptr[i]); + kfree(ac_ptr); + return NULL; + } + } + } + return ac_ptr; +} + +static inline void free_alien_cache(struct array_cache **ac_ptr) +{ + int i; + + if (!ac_ptr) + return; + + for_each_node(i) + kfree(ac_ptr[i]); + + kfree(ac_ptr); +} + +static inline void __drain_alien_cache(kmem_cache_t *cachep, struct array_cache *ac, int node) +{ + struct kmem_list3 *rl3 = cachep->nodelists[node]; + + if (ac->avail) { + spin_lock(&rl3->list_lock); + free_block(cachep, ac->entry, ac->avail); + ac->avail = 0; + spin_unlock(&rl3->list_lock); + } +} + +static void drain_alien_cache(kmem_cache_t *cachep, struct kmem_list3 *l3) +{ + int i=0; + struct array_cache *ac; + unsigned long flags; + + for_each_online_node(i) { + ac = l3->alien[i]; + if (ac) { + spin_lock_irqsave(&ac->lock, flags); + __drain_alien_cache(cachep, ac, i); + spin_unlock_irqrestore(&ac->lock, flags); + } + } +} +#else +#define alloc_alien_cache(node, limit) do { } while (0) +#define free_alien_cache(ac_ptr) do { } while (0) +#define drain_alien_cache(cachep, l3) do { } while (0) +#endif + static int __devinit cpuup_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { long cpu = (long)hcpu; kmem_cache_t* cachep; + struct kmem_list3 *l3 = NULL; + int node = cpu_to_node(cpu); + int memsize = sizeof(struct kmem_list3); + struct array_cache *nc = NULL; switch (action) { case CPU_UP_PREPARE: down(&cache_chain_sem); + /* we need to do this right in the beginning since + * alloc_arraycache's are going to use this list. + * kmalloc_node allows us to add the slab to the right + * kmem_list3 and not this cpu's kmem_list3 + */ + list_for_each_entry(cachep, &cache_chain, next) { - struct array_cache *nc; + /* setup the size64 kmemlist for cpu before we can + * begin anything. Make sure some other cpu on this + * node has not already allocated this + */ + if (!cachep->nodelists[node]) { + if (!(l3 = kmalloc_node(memsize, + GFP_KERNEL, node))) + goto bad; + kmem_list3_init(l3); + l3->next_reap = jiffies + REAPTIMEOUT_LIST3 + + ((unsigned long)cachep)%REAPTIMEOUT_LIST3; + + cachep->nodelists[node] = l3; + } - nc = alloc_arraycache(cpu, cachep->limit, cachep->batchcount); + spin_lock_irq(&cachep->nodelists[node]->list_lock); + cachep->nodelists[node]->free_limit = + (1 + nr_cpus_node(node)) * + cachep->batchcount + cachep->num; + spin_unlock_irq(&cachep->nodelists[node]->list_lock); + } + + /* Now we can go ahead with allocating the shared array's + & array cache's */ + list_for_each_entry(cachep, &cache_chain, next) { + nc = alloc_arraycache(node, cachep->limit, + cachep->batchcount); if (!nc) goto bad; - - spin_lock_irq(&cachep->spinlock); cachep->array[cpu] = nc; - cachep->free_limit = (1+num_online_cpus())*cachep->batchcount - + cachep->num; - spin_unlock_irq(&cachep->spinlock); + l3 = cachep->nodelists[node]; + BUG_ON(!l3); + if (!l3->shared) { + if (!(nc = alloc_arraycache(node, + cachep->shared*cachep->batchcount, + 0xbaadf00d))) + goto bad; + + /* we are serialised from CPU_DEAD or + CPU_UP_CANCELLED by the cpucontrol lock */ + l3->shared = nc; + } } up(&cache_chain_sem); break; @@ -731,13 +908,51 @@ static int __devinit cpuup_callback(struct notifier_block *nfb, list_for_each_entry(cachep, &cache_chain, next) { struct array_cache *nc; + cpumask_t mask; + mask = node_to_cpumask(node); spin_lock_irq(&cachep->spinlock); /* cpu is dead; no one can alloc from it. */ nc = cachep->array[cpu]; cachep->array[cpu] = NULL; - cachep->free_limit -= cachep->batchcount; - free_block(cachep, ac_entry(nc), nc->avail); + l3 = cachep->nodelists[node]; + + if (!l3) + goto unlock_cache; + + spin_lock(&l3->list_lock); + + /* Free limit for this kmem_list3 */ + l3->free_limit -= cachep->batchcount; + if (nc) + free_block(cachep, nc->entry, nc->avail); + + if (!cpus_empty(mask)) { + spin_unlock(&l3->list_lock); + goto unlock_cache; + } + + if (l3->shared) { + free_block(cachep, l3->shared->entry, + l3->shared->avail); + kfree(l3->shared); + l3->shared = NULL; + } + if (l3->alien) { + drain_alien_cache(cachep, l3); + free_alien_cache(l3->alien); + l3->alien = NULL; + } + + /* free slabs belonging to this node */ + if (__node_shrink(cachep, node)) { + cachep->nodelists[node] = NULL; + spin_unlock(&l3->list_lock); + kfree(l3); + } else { + spin_unlock(&l3->list_lock); + } +unlock_cache: spin_unlock_irq(&cachep->spinlock); kfree(nc); } @@ -753,6 +968,25 @@ bad: static struct notifier_block cpucache_notifier = { &cpuup_callback, NULL, 0 }; +/* + * swap the static kmem_list3 with kmalloced memory + */ +static void init_list(kmem_cache_t *cachep, struct kmem_list3 *list, + int nodeid) +{ + struct kmem_list3 *ptr; + + BUG_ON(cachep->nodelists[nodeid] != list); + ptr = kmalloc_node(sizeof(struct kmem_list3), GFP_KERNEL, nodeid); + BUG_ON(!ptr); + + local_irq_disable(); + memcpy(ptr, list, sizeof(struct kmem_list3)); + MAKE_ALL_LISTS(cachep, ptr, nodeid); + cachep->nodelists[nodeid] = ptr; + local_irq_enable(); +} + /* Initialisation. * Called after the gfp() functions have been enabled, and before smp_init(). */ @@ -761,6 +995,13 @@ void __init kmem_cache_init(void) size_t left_over; struct cache_sizes *sizes; struct cache_names *names; + int i; + + for (i = 0; i < NUM_INIT_LISTS; i++) { + kmem_list3_init(&initkmem_list3[i]); + if (i < MAX_NUMNODES) + cache_cache.nodelists[i] = NULL; + } /* * Fragmentation resistance on low memory - only use bigger @@ -769,21 +1010,24 @@ void __init kmem_cache_init(void) if (num_physpages > (32 << 20) >> PAGE_SHIFT) slab_break_gfp_order = BREAK_GFP_ORDER_HI; - /* Bootstrap is tricky, because several objects are allocated * from caches that do not exist yet: * 1) initialize the cache_cache cache: it contains the kmem_cache_t * structures of all caches, except cache_cache itself: cache_cache * is statically allocated. - * Initially an __init data area is used for the head array, it's - * replaced with a kmalloc allocated array at the end of the bootstrap. + * Initially an __init data area is used for the head array and the + * kmem_list3 structures, it's replaced with a kmalloc allocated + * array at the end of the bootstrap. * 2) Create the first kmalloc cache. - * The kmem_cache_t for the new cache is allocated normally. An __init - * data area is used for the head array. - * 3) Create the remaining kmalloc caches, with minimally sized head arrays. + * The kmem_cache_t for the new cache is allocated normally. + * An __init data area is used for the head array. + * 3) Create the remaining kmalloc caches, with minimally sized + * head arrays. * 4) Replace the __init data head arrays for cache_cache and the first * kmalloc cache with kmalloc allocated arrays. - * 5) Resize the head arrays of the kmalloc caches to their final sizes. + * 5) Replace the __init data for kmem_list3 for cache_cache and + * the other cache's with kmalloc allocated memory. + * 6) Resize the head arrays of the kmalloc caches to their final sizes. */ /* 1) create the cache_cache */ @@ -792,6 +1036,7 @@ void __init kmem_cache_init(void) list_add(&cache_cache.next, &cache_chain); cache_cache.colour_off = cache_line_size(); cache_cache.array[smp_processor_id()] = &initarray_cache.cache; + cache_cache.nodelists[numa_node_id()] = &initkmem_list3[CACHE_CACHE]; cache_cache.objsize = ALIGN(cache_cache.objsize, cache_line_size()); @@ -809,15 +1054,33 @@ void __init kmem_cache_init(void) sizes = malloc_sizes; names = cache_names; + /* Initialize the caches that provide memory for the array cache + * and the kmem_list3 structures first. + * Without this, further allocations will bug + */ + + sizes[INDEX_AC].cs_cachep = kmem_cache_create(names[INDEX_AC].name, + sizes[INDEX_AC].cs_size, ARCH_KMALLOC_MINALIGN, + (ARCH_KMALLOC_FLAGS | SLAB_PANIC), NULL, NULL); + + if (INDEX_AC != INDEX_L3) + sizes[INDEX_L3].cs_cachep = + kmem_cache_create(names[INDEX_L3].name, + sizes[INDEX_L3].cs_size, ARCH_KMALLOC_MINALIGN, + (ARCH_KMALLOC_FLAGS | SLAB_PANIC), NULL, NULL); + while (sizes->cs_size != ULONG_MAX) { - /* For performance, all the general caches are L1 aligned. + /* + * For performance, all the general caches are L1 aligned. * This should be particularly beneficial on SMP boxes, as it * eliminates "false sharing". * Note for systems short on memory removing the alignment will - * allow tighter packing of the smaller caches. */ - sizes->cs_cachep = kmem_cache_create(names->name, - sizes->cs_size, ARCH_KMALLOC_MINALIGN, - (ARCH_KMALLOC_FLAGS | SLAB_PANIC), NULL, NULL); + * allow tighter packing of the smaller caches. + */ + if(!sizes->cs_cachep) + sizes->cs_cachep = kmem_cache_create(names->name, + sizes->cs_size, ARCH_KMALLOC_MINALIGN, + (ARCH_KMALLOC_FLAGS | SLAB_PANIC), NULL, NULL); /* Inc off-slab bufctl limit until the ceiling is hit. */ if (!(OFF_SLAB(sizes->cs_cachep))) { @@ -836,24 +1099,47 @@ void __init kmem_cache_init(void) /* 4) Replace the bootstrap head arrays */ { void * ptr; - + ptr = kmalloc(sizeof(struct arraycache_init), GFP_KERNEL); + local_irq_disable(); BUG_ON(ac_data(&cache_cache) != &initarray_cache.cache); - memcpy(ptr, ac_data(&cache_cache), sizeof(struct arraycache_init)); + memcpy(ptr, ac_data(&cache_cache), + sizeof(struct arraycache_init)); cache_cache.array[smp_processor_id()] = ptr; local_irq_enable(); - + ptr = kmalloc(sizeof(struct arraycache_init), GFP_KERNEL); + local_irq_disable(); - BUG_ON(ac_data(malloc_sizes[0].cs_cachep) != &initarray_generic.cache); - memcpy(ptr, ac_data(malloc_sizes[0].cs_cachep), + BUG_ON(ac_data(malloc_sizes[INDEX_AC].cs_cachep) + != &initarray_generic.cache); + memcpy(ptr, ac_data(malloc_sizes[INDEX_AC].cs_cachep), sizeof(struct arraycache_init)); - malloc_sizes[0].cs_cachep->array[smp_processor_id()] = ptr; + malloc_sizes[INDEX_AC].cs_cachep->array[smp_processor_id()] = + ptr; local_irq_enable(); } + /* 5) Replace the bootstrap kmem_list3's */ + { + int node; + /* Replace the static kmem_list3 structures for the boot cpu */ + init_list(&cache_cache, &initkmem_list3[CACHE_CACHE], + numa_node_id()); + + for_each_online_node(node) { + init_list(malloc_sizes[INDEX_AC].cs_cachep, + &initkmem_list3[SIZE_AC+node], node); + + if (INDEX_AC != INDEX_L3) { + init_list(malloc_sizes[INDEX_L3].cs_cachep, + &initkmem_list3[SIZE_L3+node], + node); + } + } + } - /* 5) resize the head arrays to their final sizes */ + /* 6) resize the head arrays to their final sizes */ { kmem_cache_t *cachep; down(&cache_chain_sem); @@ -869,7 +1155,6 @@ void __init kmem_cache_init(void) * that initializes ac_data for all new cpus */ register_cpu_notifier(&cpucache_notifier); - /* The reap timers are started later, with a module init call: * That part of the kernel is not yet operational. @@ -884,10 +1169,8 @@ static int __init cpucache_init(void) * Register the timers that return unneeded * pages to gfp. */ - for (cpu = 0; cpu < NR_CPUS; cpu++) { - if (cpu_online(cpu)) - start_cpu_timer(cpu); - } + for_each_online_cpu(cpu) + start_cpu_timer(cpu); return 0; } @@ -1166,6 +1449,20 @@ static void slab_destroy (kmem_cache_t *cachep, struct slab *slabp) } } +/* For setting up all the kmem_list3s for cache whose objsize is same + as size of kmem_list3. */ +static inline void set_up_list3s(kmem_cache_t *cachep, int index) +{ + int node; + + for_each_online_node(node) { + cachep->nodelists[node] = &initkmem_list3[index+node]; + cachep->nodelists[node]->next_reap = jiffies + + REAPTIMEOUT_LIST3 + + ((unsigned long)cachep)%REAPTIMEOUT_LIST3; + } +} + /** * kmem_cache_create - Create a cache. * @name: A string which is used in /proc/slabinfo to identify this cache. @@ -1319,7 +1616,7 @@ kmem_cache_create (const char *name, size_t size, size_t align, size += BYTES_PER_WORD; } #if FORCED_DEBUG && defined(CONFIG_DEBUG_PAGEALLOC) - if (size > 128 && cachep->reallen > cache_line_size() && size < PAGE_SIZE) { + if (size >= malloc_sizes[INDEX_L3+1].cs_size && cachep->reallen > cache_line_size() && size < PAGE_SIZE) { cachep->dbghead += PAGE_SIZE - size; size = PAGE_SIZE; } @@ -1421,10 +1718,6 @@ next: cachep->gfpflags |= GFP_DMA; spin_lock_init(&cachep->spinlock); cachep->objsize = size; - /* NUMA */ - INIT_LIST_HEAD(&cachep->lists.slabs_full); - INIT_LIST_HEAD(&cachep->lists.slabs_partial); - INIT_LIST_HEAD(&cachep->lists.slabs_free); if (flags & CFLGS_OFF_SLAB) cachep->slabp_cache = kmem_find_general_cachep(slab_size,0); @@ -1443,11 +1736,43 @@ next: * the cache that's used by kmalloc(24), otherwise * the creation of further caches will BUG(). */ - cachep->array[smp_processor_id()] = &initarray_generic.cache; - g_cpucache_up = PARTIAL; + cachep->array[smp_processor_id()] = + &initarray_generic.cache; + + /* If the cache that's used by + * kmalloc(sizeof(kmem_list3)) is the first cache, + * then we need to set up all its list3s, otherwise + * the creation of further caches will BUG(). + */ + set_up_list3s(cachep, SIZE_AC); + if (INDEX_AC == INDEX_L3) + g_cpucache_up = PARTIAL_L3; + else + g_cpucache_up = PARTIAL_AC; } else { - cachep->array[smp_processor_id()] = kmalloc(sizeof(struct arraycache_init),GFP_KERNEL); + cachep->array[smp_processor_id()] = + kmalloc(sizeof(struct arraycache_init), + GFP_KERNEL); + + if (g_cpucache_up == PARTIAL_AC) { + set_up_list3s(cachep, SIZE_L3); + g_cpucache_up = PARTIAL_L3; + } else { + int node; + for_each_online_node(node) { + + cachep->nodelists[node] = + kmalloc_node(sizeof(struct kmem_list3), + GFP_KERNEL, node); + BUG_ON(!cachep->nodelists[node]); + kmem_list3_init(cachep->nodelists[node]); + } + } } + cachep->nodelists[numa_node_id()]->next_reap = + jiffies + REAPTIMEOUT_LIST3 + + ((unsigned long)cachep)%REAPTIMEOUT_LIST3; + BUG_ON(!ac_data(cachep)); ac_data(cachep)->avail = 0; ac_data(cachep)->limit = BOOT_CPUCACHE_ENTRIES; @@ -1455,13 +1780,8 @@ next: ac_data(cachep)->touched = 0; cachep->batchcount = 1; cachep->limit = BOOT_CPUCACHE_ENTRIES; - cachep->free_limit = (1+num_online_cpus())*cachep->batchcount - + cachep->num; } - cachep->lists.next_reap = jiffies + REAPTIMEOUT_LIST3 + - ((unsigned long)cachep)%REAPTIMEOUT_LIST3; - /* Need the semaphore to access the chain. */ down(&cache_chain_sem); { @@ -1518,13 +1838,23 @@ static void check_spinlock_acquired(kmem_cache_t *cachep) { #ifdef CONFIG_SMP check_irq_off(); - BUG_ON(spin_trylock(&cachep->spinlock)); + assert_spin_locked(&cachep->nodelists[numa_node_id()]->list_lock); +#endif +} + +static inline void check_spinlock_acquired_node(kmem_cache_t *cachep, int node) +{ +#ifdef CONFIG_SMP + check_irq_off(); + assert_spin_locked(&cachep->nodelists[node]->list_lock); #endif } + #else #define check_irq_off() do { } while(0) #define check_irq_on() do { } while(0) #define check_spinlock_acquired(x) do { } while(0) +#define check_spinlock_acquired_node(x, y) do { } while(0) #endif /* @@ -1546,7 +1876,7 @@ static void smp_call_function_all_cpus(void (*func) (void *arg), void *arg) } static void drain_array_locked(kmem_cache_t* cachep, - struct array_cache *ac, int force); + struct array_cache *ac, int force, int node); static void do_drain(void *arg) { @@ -1555,59 +1885,82 @@ static void do_drain(void *arg) check_irq_off(); ac = ac_data(cachep); - spin_lock(&cachep->spinlock); - free_block(cachep, &ac_entry(ac)[0], ac->avail); - spin_unlock(&cachep->spinlock); + spin_lock(&cachep->nodelists[numa_node_id()]->list_lock); + free_block(cachep, ac->entry, ac->avail); + spin_unlock(&cachep->nodelists[numa_node_id()]->list_lock); ac->avail = 0; } static void drain_cpu_caches(kmem_cache_t *cachep) { + struct kmem_list3 *l3; + int node; + smp_call_function_all_cpus(do_drain, cachep); check_irq_on(); spin_lock_irq(&cachep->spinlock); - if (cachep->lists.shared) - drain_array_locked(cachep, cachep->lists.shared, 1); + for_each_online_node(node) { + l3 = cachep->nodelists[node]; + if (l3) { + spin_lock(&l3->list_lock); + drain_array_locked(cachep, l3->shared, 1, node); + spin_unlock(&l3->list_lock); + if (l3->alien) + drain_alien_cache(cachep, l3); + } + } spin_unlock_irq(&cachep->spinlock); } - -/* NUMA shrink all list3s */ -static int __cache_shrink(kmem_cache_t *cachep) +static int __node_shrink(kmem_cache_t *cachep, int node) { struct slab *slabp; + struct kmem_list3 *l3 = cachep->nodelists[node]; int ret; - drain_cpu_caches(cachep); - - check_irq_on(); - spin_lock_irq(&cachep->spinlock); - - for(;;) { + for (;;) { struct list_head *p; - p = cachep->lists.slabs_free.prev; - if (p == &cachep->lists.slabs_free) + p = l3->slabs_free.prev; + if (p == &l3->slabs_free) break; - slabp = list_entry(cachep->lists.slabs_free.prev, struct slab, list); + slabp = list_entry(l3->slabs_free.prev, struct slab, list); #if DEBUG if (slabp->inuse) BUG(); #endif list_del(&slabp->list); - cachep->lists.free_objects -= cachep->num; - spin_unlock_irq(&cachep->spinlock); + l3->free_objects -= cachep->num; + spin_unlock_irq(&l3->list_lock); slab_destroy(cachep, slabp); - spin_lock_irq(&cachep->spinlock); + spin_lock_irq(&l3->list_lock); } - ret = !list_empty(&cachep->lists.slabs_full) || - !list_empty(&cachep->lists.slabs_partial); - spin_unlock_irq(&cachep->spinlock); + ret = !list_empty(&l3->slabs_full) || + !list_empty(&l3->slabs_partial); return ret; } +static int __cache_shrink(kmem_cache_t *cachep) +{ + int ret = 0, i = 0; + struct kmem_list3 *l3; + + drain_cpu_caches(cachep); + + check_irq_on(); + for_each_online_node(i) { + l3 = cachep->nodelists[i]; + if (l3) { + spin_lock_irq(&l3->list_lock); + ret += __node_shrink(cachep, i); + spin_unlock_irq(&l3->list_lock); + } + } + return (ret ? 1 : 0); +} + /** * kmem_cache_shrink - Shrink a cache. * @cachep: The cache to shrink. @@ -1644,6 +1997,7 @@ EXPORT_SYMBOL(kmem_cache_shrink); int kmem_cache_destroy(kmem_cache_t * cachep) { int i; + struct kmem_list3 *l3; if (!cachep || in_interrupt()) BUG(); @@ -1671,15 +2025,17 @@ int kmem_cache_destroy(kmem_cache_t * cachep) if (unlikely(cachep->flags & SLAB_DESTROY_BY_RCU)) synchronize_rcu(); - /* no cpu_online check required here since we clear the percpu - * array on cpu offline and set this to NULL. - */ - for (i = 0; i < NR_CPUS; i++) + for_each_online_cpu(i) kfree(cachep->array[i]); /* NUMA: free the list3 structures */ - kfree(cachep->lists.shared); - cachep->lists.shared = NULL; + for_each_online_node(i) { + if ((l3 = cachep->nodelists[i])) { + kfree(l3->shared); + free_alien_cache(l3->alien); + kfree(l3); + } + } kmem_cache_free(&cache_cache, cachep); unlock_cpu_hotplug(); @@ -1689,8 +2045,8 @@ int kmem_cache_destroy(kmem_cache_t * cachep) EXPORT_SYMBOL(kmem_cache_destroy); /* Get the memory for a slab management obj. */ -static struct slab* alloc_slabmgmt(kmem_cache_t *cachep, - void *objp, int colour_off, unsigned int __nocast local_flags) +static struct slab* alloc_slabmgmt(kmem_cache_t *cachep, void *objp, + int colour_off, unsigned int __nocast local_flags) { struct slab *slabp; @@ -1721,7 +2077,7 @@ static void cache_init_objs(kmem_cache_t *cachep, int i; for (i = 0; i < cachep->num; i++) { - void* objp = slabp->s_mem+cachep->objsize*i; + void *objp = slabp->s_mem+cachep->objsize*i; #if DEBUG /* need to poison the objs? */ if (cachep->flags & SLAB_POISON) @@ -1798,6 +2154,7 @@ static int cache_grow(kmem_cache_t *cachep, unsigned int __nocast flags, int nod size_t offset; unsigned int local_flags; unsigned long ctor_flags; + struct kmem_list3 *l3; /* Be lazy and only check for valid flags here, * keeping it out of the critical path in kmem_cache_alloc(). @@ -1829,6 +2186,7 @@ static int cache_grow(kmem_cache_t *cachep, unsigned int __nocast flags, int nod spin_unlock(&cachep->spinlock); + check_irq_off(); if (local_flags & __GFP_WAIT) local_irq_enable(); @@ -1840,8 +2198,9 @@ static int cache_grow(kmem_cache_t *cachep, unsigned int __nocast flags, int nod */ kmem_flagcheck(cachep, flags); - - /* Get mem for the objs. */ + /* Get mem for the objs. + * Attempt to allocate a physical page from 'nodeid', + */ if (!(objp = kmem_getpages(cachep, flags, nodeid))) goto failed; @@ -1849,6 +2208,7 @@ static int cache_grow(kmem_cache_t *cachep, unsigned int __nocast flags, int nod if (!(slabp = alloc_slabmgmt(cachep, objp, offset, local_flags))) goto opps1; + slabp->nodeid = nodeid; set_slab_attr(cachep, slabp, objp); cache_init_objs(cachep, slabp, ctor_flags); @@ -1856,13 +2216,14 @@ static int cache_grow(kmem_cache_t *cachep, unsigned int __nocast flags, int nod if (local_flags & __GFP_WAIT) local_irq_disable(); check_irq_off(); - spin_lock(&cachep->spinlock); + l3 = cachep->nodelists[nodeid]; + spin_lock(&l3->list_lock); /* Make slab active. */ - list_add_tail(&slabp->list, &(list3_data(cachep)->slabs_free)); + list_add_tail(&slabp->list, &(l3->slabs_free)); STATS_INC_GROWN(cachep); - list3_data(cachep)->free_objects += cachep->num; - spin_unlock(&cachep->spinlock); + l3->free_objects += cachep->num; + spin_unlock(&l3->list_lock); return 1; opps1: kmem_freepages(cachep, objp); @@ -1968,7 +2329,6 @@ static void check_slabp(kmem_cache_t *cachep, struct slab *slabp) kmem_bufctl_t i; int entries = 0; - check_spinlock_acquired(cachep); /* Check slab's freelist to see if this obj is there. */ for (i = slabp->free; i != BUFCTL_END; i = slab_bufctl(slabp)[i]) { entries++; @@ -2011,10 +2371,11 @@ retry: */ batchcount = BATCHREFILL_LIMIT; } - l3 = list3_data(cachep); + l3 = cachep->nodelists[numa_node_id()]; + + BUG_ON(ac->avail > 0 || !l3); + spin_lock(&l3->list_lock); - BUG_ON(ac->avail > 0); - spin_lock(&cachep->spinlock); if (l3->shared) { struct array_cache *shared_array = l3->shared; if (shared_array->avail) { @@ -2022,8 +2383,9 @@ retry: batchcount = shared_array->avail; shared_array->avail -= batchcount; ac->avail = batchcount; - memcpy(ac_entry(ac), &ac_entry(shared_array)[shared_array->avail], - sizeof(void*)*batchcount); + memcpy(ac->entry, + &(shared_array->entry[shared_array->avail]), + sizeof(void*)*batchcount); shared_array->touched = 1; goto alloc_done; } @@ -2050,7 +2412,8 @@ retry: STATS_SET_HIGH(cachep); /* get obj pointer */ - ac_entry(ac)[ac->avail++] = slabp->s_mem + slabp->free*cachep->objsize; + ac->entry[ac->avail++] = slabp->s_mem + + slabp->free*cachep->objsize; slabp->inuse++; next = slab_bufctl(slabp)[slabp->free]; @@ -2072,12 +2435,12 @@ retry: must_grow: l3->free_objects -= ac->avail; alloc_done: - spin_unlock(&cachep->spinlock); + spin_unlock(&l3->list_lock); if (unlikely(!ac->avail)) { int x; - x = cache_grow(cachep, flags, -1); - + x = cache_grow(cachep, flags, numa_node_id()); + // cache_grow can reenable interrupts, then ac could change. ac = ac_data(cachep); if (!x && ac->avail == 0) // no objects in sight? abort @@ -2087,7 +2450,7 @@ alloc_done: goto retry; } ac->touched = 1; - return ac_entry(ac)[--ac->avail]; + return ac->entry[--ac->avail]; } static inline void @@ -2159,43 +2522,116 @@ static inline void *__cache_alloc(kmem_cache_t *cachep, unsigned int __nocast fl if (likely(ac->avail)) { STATS_INC_ALLOCHIT(cachep); ac->touched = 1; - objp = ac_entry(ac)[--ac->avail]; + objp = ac->entry[--ac->avail]; } else { STATS_INC_ALLOCMISS(cachep); objp = cache_alloc_refill(cachep, flags); } local_irq_restore(save_flags); - objp = cache_alloc_debugcheck_after(cachep, flags, objp, __builtin_return_address(0)); + objp = cache_alloc_debugcheck_after(cachep, flags, objp, + __builtin_return_address(0)); + prefetchw(objp); return objp; } -/* - * NUMA: different approach needed if the spinlock is moved into - * the l3 structure +#ifdef CONFIG_NUMA +/* + * A interface to enable slab creation on nodeid */ +static void *__cache_alloc_node(kmem_cache_t *cachep, int flags, int nodeid) +{ + struct list_head *entry; + struct slab *slabp; + struct kmem_list3 *l3; + void *obj; + kmem_bufctl_t next; + int x; + l3 = cachep->nodelists[nodeid]; + BUG_ON(!l3); + +retry: + spin_lock(&l3->list_lock); + entry = l3->slabs_partial.next; + if (entry == &l3->slabs_partial) { + l3->free_touched = 1; + entry = l3->slabs_free.next; + if (entry == &l3->slabs_free) + goto must_grow; + } + + slabp = list_entry(entry, struct slab, list); + check_spinlock_acquired_node(cachep, nodeid); + check_slabp(cachep, slabp); + + STATS_INC_NODEALLOCS(cachep); + STATS_INC_ACTIVE(cachep); + STATS_SET_HIGH(cachep); + + BUG_ON(slabp->inuse == cachep->num); + + /* get obj pointer */ + obj = slabp->s_mem + slabp->free*cachep->objsize; + slabp->inuse++; + next = slab_bufctl(slabp)[slabp->free]; +#if DEBUG + slab_bufctl(slabp)[slabp->free] = BUFCTL_FREE; +#endif + slabp->free = next; + check_slabp(cachep, slabp); + l3->free_objects--; + /* move slabp to correct slabp list: */ + list_del(&slabp->list); + + if (slabp->free == BUFCTL_END) { + list_add(&slabp->list, &l3->slabs_full); + } else { + list_add(&slabp->list, &l3->slabs_partial); + } + + spin_unlock(&l3->list_lock); + goto done; + +must_grow: + spin_unlock(&l3->list_lock); + x = cache_grow(cachep, flags, nodeid); + + if (!x) + return NULL; + + goto retry; +done: + return obj; +} +#endif + +/* + * Caller needs to acquire correct kmem_list's list_lock + */ static void free_block(kmem_cache_t *cachep, void **objpp, int nr_objects) { int i; - - check_spinlock_acquired(cachep); - - /* NUMA: move add into loop */ - cachep->lists.free_objects += nr_objects; + struct kmem_list3 *l3; for (i = 0; i < nr_objects; i++) { void *objp = objpp[i]; struct slab *slabp; unsigned int objnr; + int nodeid = 0; slabp = GET_PAGE_SLAB(virt_to_page(objp)); + nodeid = slabp->nodeid; + l3 = cachep->nodelists[nodeid]; list_del(&slabp->list); objnr = (objp - slabp->s_mem) / cachep->objsize; + check_spinlock_acquired_node(cachep, nodeid); check_slabp(cachep, slabp); + + #if DEBUG if (slab_bufctl(slabp)[objnr] != BUFCTL_FREE) { - printk(KERN_ERR "slab: double free detected in cache '%s', objp %p.\n", - cachep->name, objp); + printk(KERN_ERR "slab: double free detected in cache " + "'%s', objp %p\n", cachep->name, objp); BUG(); } #endif @@ -2203,24 +2639,23 @@ static void free_block(kmem_cache_t *cachep, void **objpp, int nr_objects) slabp->free = objnr; STATS_DEC_ACTIVE(cachep); slabp->inuse--; + l3->free_objects++; check_slabp(cachep, slabp); /* fixup slab chains */ if (slabp->inuse == 0) { - if (cachep->lists.free_objects > cachep->free_limit) { - cachep->lists.free_objects -= cachep->num; + if (l3->free_objects > l3->free_limit) { + l3->free_objects -= cachep->num; slab_destroy(cachep, slabp); } else { - list_add(&slabp->list, - &list3_data_ptr(cachep, objp)->slabs_free); + list_add(&slabp->list, &l3->slabs_free); } } else { /* Unconditionally move a slab to the end of the * partial list on free - maximum time for the * other objects to be freed, too. */ - list_add_tail(&slabp->list, - &list3_data_ptr(cachep, objp)->slabs_partial); + list_add_tail(&slabp->list, &l3->slabs_partial); } } } @@ -2228,36 +2663,38 @@ static void free_block(kmem_cache_t *cachep, void **objpp, int nr_objects) static void cache_flusharray(kmem_cache_t *cachep, struct array_cache *ac) { int batchcount; + struct kmem_list3 *l3; batchcount = ac->batchcount; #if DEBUG BUG_ON(!batchcount || batchcount > ac->avail); #endif check_irq_off(); - spin_lock(&cachep->spinlock); - if (cachep->lists.shared) { - struct array_cache *shared_array = cachep->lists.shared; + l3 = cachep->nodelists[numa_node_id()]; + spin_lock(&l3->list_lock); + if (l3->shared) { + struct array_cache *shared_array = l3->shared; int max = shared_array->limit-shared_array->avail; if (max) { if (batchcount > max) batchcount = max; - memcpy(&ac_entry(shared_array)[shared_array->avail], - &ac_entry(ac)[0], + memcpy(&(shared_array->entry[shared_array->avail]), + ac->entry, sizeof(void*)*batchcount); shared_array->avail += batchcount; goto free_done; } } - free_block(cachep, &ac_entry(ac)[0], batchcount); + free_block(cachep, ac->entry, batchcount); free_done: #if STATS { int i = 0; struct list_head *p; - p = list3_data(cachep)->slabs_free.next; - while (p != &(list3_data(cachep)->slabs_free)) { + p = l3->slabs_free.next; + while (p != &(l3->slabs_free)) { struct slab *slabp; slabp = list_entry(p, struct slab, list); @@ -2269,12 +2706,13 @@ free_done: STATS_SET_FREEABLE(cachep, i); } #endif - spin_unlock(&cachep->spinlock); + spin_unlock(&l3->list_lock); ac->avail -= batchcount; - memmove(&ac_entry(ac)[0], &ac_entry(ac)[batchcount], + memmove(ac->entry, &(ac->entry[batchcount]), sizeof(void*)*ac->avail); } + /* * __cache_free * Release an obj back to its cache. If the obj has a constructed @@ -2289,14 +2727,46 @@ static inline void __cache_free(kmem_cache_t *cachep, void *objp) check_irq_off(); objp = cache_free_debugcheck(cachep, objp, __builtin_return_address(0)); + /* Make sure we are not freeing a object from another + * node to the array cache on this cpu. + */ +#ifdef CONFIG_NUMA + { + struct slab *slabp; + slabp = GET_PAGE_SLAB(virt_to_page(objp)); + if (unlikely(slabp->nodeid != numa_node_id())) { + struct array_cache *alien = NULL; + int nodeid = slabp->nodeid; + struct kmem_list3 *l3 = cachep->nodelists[numa_node_id()]; + + STATS_INC_NODEFREES(cachep); + if (l3->alien && l3->alien[nodeid]) { + alien = l3->alien[nodeid]; + spin_lock(&alien->lock); + if (unlikely(alien->avail == alien->limit)) + __drain_alien_cache(cachep, + alien, nodeid); + alien->entry[alien->avail++] = objp; + spin_unlock(&alien->lock); + } else { + spin_lock(&(cachep->nodelists[nodeid])-> + list_lock); + free_block(cachep, &objp, 1); + spin_unlock(&(cachep->nodelists[nodeid])-> + list_lock); + } + return; + } + } +#endif if (likely(ac->avail < ac->limit)) { STATS_INC_FREEHIT(cachep); - ac_entry(ac)[ac->avail++] = objp; + ac->entry[ac->avail++] = objp; return; } else { STATS_INC_FREEMISS(cachep); cache_flusharray(cachep, ac); - ac_entry(ac)[ac->avail++] = objp; + ac->entry[ac->avail++] = objp; } } @@ -2366,81 +2836,30 @@ out: * Identical to kmem_cache_alloc, except that this function is slow * and can sleep. And it will allocate memory on the given node, which * can improve the performance for cpu bound structures. + * New and improved: it will now make sure that the object gets + * put on the correct node list so that there is no false sharing. */ void *kmem_cache_alloc_node(kmem_cache_t *cachep, int flags, int nodeid) { - int loop; - void *objp; - struct slab *slabp; - kmem_bufctl_t next; - - if (nodeid == -1) - return kmem_cache_alloc(cachep, flags); - - for (loop = 0;;loop++) { - struct list_head *q; - - objp = NULL; - check_irq_on(); - spin_lock_irq(&cachep->spinlock); - /* walk through all partial and empty slab and find one - * from the right node */ - list_for_each(q,&cachep->lists.slabs_partial) { - slabp = list_entry(q, struct slab, list); - - if (page_to_nid(virt_to_page(slabp->s_mem)) == nodeid || - loop > 2) - goto got_slabp; - } - list_for_each(q, &cachep->lists.slabs_free) { - slabp = list_entry(q, struct slab, list); + unsigned long save_flags; + void *ptr; - if (page_to_nid(virt_to_page(slabp->s_mem)) == nodeid || - loop > 2) - goto got_slabp; - } - spin_unlock_irq(&cachep->spinlock); + if (nodeid == numa_node_id() || nodeid == -1) + return __cache_alloc(cachep, flags); - local_irq_disable(); - if (!cache_grow(cachep, flags, nodeid)) { - local_irq_enable(); - return NULL; - } - local_irq_enable(); + if (unlikely(!cachep->nodelists[nodeid])) { + /* Fall back to __cache_alloc if we run into trouble */ + printk(KERN_WARNING "slab: not allocating in inactive node %d for cache %s\n", nodeid, cachep->name); + return __cache_alloc(cachep,flags); } -got_slabp: - /* found one: allocate object */ - check_slabp(cachep, slabp); - check_spinlock_acquired(cachep); - - STATS_INC_ALLOCED(cachep); - STATS_INC_ACTIVE(cachep); - STATS_SET_HIGH(cachep); - STATS_INC_NODEALLOCS(cachep); - - objp = slabp->s_mem + slabp->free*cachep->objsize; - - slabp->inuse++; - next = slab_bufctl(slabp)[slabp->free]; -#if DEBUG - slab_bufctl(slabp)[slabp->free] = BUFCTL_FREE; -#endif - slabp->free = next; - check_slabp(cachep, slabp); - /* move slabp to correct slabp list: */ - list_del(&slabp->list); - if (slabp->free == BUFCTL_END) - list_add(&slabp->list, &cachep->lists.slabs_full); - else - list_add(&slabp->list, &cachep->lists.slabs_partial); - - list3_data(cachep)->free_objects--; - spin_unlock_irq(&cachep->spinlock); + cache_alloc_debugcheck_before(cachep, flags); + local_irq_save(save_flags); + ptr = __cache_alloc_node(cachep, flags, nodeid); + local_irq_restore(save_flags); + ptr = cache_alloc_debugcheck_after(cachep, flags, ptr, __builtin_return_address(0)); - objp = cache_alloc_debugcheck_after(cachep, GFP_KERNEL, objp, - __builtin_return_address(0)); - return objp; + return ptr; } EXPORT_SYMBOL(kmem_cache_alloc_node); @@ -2510,11 +2929,18 @@ void *__alloc_percpu(size_t size, size_t align) if (!pdata) return NULL; - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_possible(i)) - continue; - pdata->ptrs[i] = kmalloc_node(size, GFP_KERNEL, - cpu_to_node(i)); + /* + * Cannot use for_each_online_cpu since a cpu may come online + * and we have no way of figuring out how to fix the array + * that we have allocated then.... + */ + for_each_cpu(i) { + int node = cpu_to_node(i); + + if (node_online(node)) + pdata->ptrs[i] = kmalloc_node(size, GFP_KERNEL, node); + else + pdata->ptrs[i] = kmalloc(size, GFP_KERNEL); if (!pdata->ptrs[i]) goto unwind_oom; @@ -2555,29 +2981,25 @@ void kmem_cache_free(kmem_cache_t *cachep, void *objp) EXPORT_SYMBOL(kmem_cache_free); /** - * kcalloc - allocate memory for an array. The memory is set to zero. - * @n: number of elements. - * @size: element size. + * kzalloc - allocate memory. The memory is set to zero. + * @size: how many bytes of memory are required. * @flags: the type of memory to allocate. */ -void *kcalloc(size_t n, size_t size, unsigned int __nocast flags) +void *kzalloc(size_t size, unsigned int __nocast flags) { - void *ret = NULL; - - if (n != 0 && size > INT_MAX / n) - return ret; - - ret = kmalloc(n * size, flags); + void *ret = kmalloc(size, flags); if (ret) - memset(ret, 0, n * size); + memset(ret, 0, size); return ret; } -EXPORT_SYMBOL(kcalloc); +EXPORT_SYMBOL(kzalloc); /** * kfree - free previously allocated memory * @objp: pointer returned by kmalloc. * + * If @objp is NULL, no operation is performed. + * * Don't free memory not originally allocated by kmalloc() * or you will run into trouble. */ @@ -2610,11 +3032,11 @@ free_percpu(const void *objp) int i; struct percpu_data *p = (struct percpu_data *) (~(unsigned long) objp); - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_possible(i)) - continue; + /* + * We allocate for all cpus so we cannot use for online cpu here. + */ + for_each_cpu(i) kfree(p->ptrs[i]); - } kfree(p); } EXPORT_SYMBOL(free_percpu); @@ -2632,6 +3054,64 @@ const char *kmem_cache_name(kmem_cache_t *cachep) } EXPORT_SYMBOL_GPL(kmem_cache_name); +/* + * This initializes kmem_list3 for all nodes. + */ +static int alloc_kmemlist(kmem_cache_t *cachep) +{ + int node; + struct kmem_list3 *l3; + int err = 0; + + for_each_online_node(node) { + struct array_cache *nc = NULL, *new; + struct array_cache **new_alien = NULL; +#ifdef CONFIG_NUMA + if (!(new_alien = alloc_alien_cache(node, cachep->limit))) + goto fail; +#endif + if (!(new = alloc_arraycache(node, (cachep->shared* + cachep->batchcount), 0xbaadf00d))) + goto fail; + if ((l3 = cachep->nodelists[node])) { + + spin_lock_irq(&l3->list_lock); + + if ((nc = cachep->nodelists[node]->shared)) + free_block(cachep, nc->entry, + nc->avail); + + l3->shared = new; + if (!cachep->nodelists[node]->alien) { + l3->alien = new_alien; + new_alien = NULL; + } + l3->free_limit = (1 + nr_cpus_node(node))* + cachep->batchcount + cachep->num; + spin_unlock_irq(&l3->list_lock); + kfree(nc); + free_alien_cache(new_alien); + continue; + } + if (!(l3 = kmalloc_node(sizeof(struct kmem_list3), + GFP_KERNEL, node))) + goto fail; + + kmem_list3_init(l3); + l3->next_reap = jiffies + REAPTIMEOUT_LIST3 + + ((unsigned long)cachep)%REAPTIMEOUT_LIST3; + l3->shared = new; + l3->alien = new_alien; + l3->free_limit = (1 + nr_cpus_node(node))* + cachep->batchcount + cachep->num; + cachep->nodelists[node] = l3; + } + return err; +fail: + err = -ENOMEM; + return err; +} + struct ccupdate_struct { kmem_cache_t *cachep; struct array_cache *new[NR_CPUS]; @@ -2644,7 +3124,7 @@ static void do_ccupdate_local(void *info) check_irq_off(); old = ac_data(new->cachep); - + new->cachep->array[smp_processor_id()] = new->new[smp_processor_id()]; new->new[smp_processor_id()] = old; } @@ -2654,54 +3134,43 @@ static int do_tune_cpucache(kmem_cache_t *cachep, int limit, int batchcount, int shared) { struct ccupdate_struct new; - struct array_cache *new_shared; - int i; + int i, err; memset(&new.new,0,sizeof(new.new)); - for (i = 0; i < NR_CPUS; i++) { - if (cpu_online(i)) { - new.new[i] = alloc_arraycache(i, limit, batchcount); - if (!new.new[i]) { - for (i--; i >= 0; i--) kfree(new.new[i]); - return -ENOMEM; - } - } else { - new.new[i] = NULL; + for_each_online_cpu(i) { + new.new[i] = alloc_arraycache(cpu_to_node(i), limit, batchcount); + if (!new.new[i]) { + for (i--; i >= 0; i--) kfree(new.new[i]); + return -ENOMEM; } } new.cachep = cachep; smp_call_function_all_cpus(do_ccupdate_local, (void *)&new); - + check_irq_on(); spin_lock_irq(&cachep->spinlock); cachep->batchcount = batchcount; cachep->limit = limit; - cachep->free_limit = (1+num_online_cpus())*cachep->batchcount + cachep->num; + cachep->shared = shared; spin_unlock_irq(&cachep->spinlock); - for (i = 0; i < NR_CPUS; i++) { + for_each_online_cpu(i) { struct array_cache *ccold = new.new[i]; if (!ccold) continue; - spin_lock_irq(&cachep->spinlock); - free_block(cachep, ac_entry(ccold), ccold->avail); - spin_unlock_irq(&cachep->spinlock); + spin_lock_irq(&cachep->nodelists[cpu_to_node(i)]->list_lock); + free_block(cachep, ccold->entry, ccold->avail); + spin_unlock_irq(&cachep->nodelists[cpu_to_node(i)]->list_lock); kfree(ccold); } - new_shared = alloc_arraycache(-1, batchcount*shared, 0xbaadf00d); - if (new_shared) { - struct array_cache *old; - spin_lock_irq(&cachep->spinlock); - old = cachep->lists.shared; - cachep->lists.shared = new_shared; - if (old) - free_block(cachep, ac_entry(old), old->avail); - spin_unlock_irq(&cachep->spinlock); - kfree(old); + err = alloc_kmemlist(cachep); + if (err) { + printk(KERN_ERR "alloc_kmemlist failed for %s, error %d.\n", + cachep->name, -err); + BUG(); } - return 0; } @@ -2759,11 +3228,11 @@ static void enable_cpucache(kmem_cache_t *cachep) } static void drain_array_locked(kmem_cache_t *cachep, - struct array_cache *ac, int force) + struct array_cache *ac, int force, int node) { int tofree; - check_spinlock_acquired(cachep); + check_spinlock_acquired_node(cachep, node); if (ac->touched && !force) { ac->touched = 0; } else if (ac->avail) { @@ -2771,9 +3240,9 @@ static void drain_array_locked(kmem_cache_t *cachep, if (tofree > ac->avail) { tofree = (ac->avail+1)/2; } - free_block(cachep, ac_entry(ac), tofree); + free_block(cachep, ac->entry, tofree); ac->avail -= tofree; - memmove(&ac_entry(ac)[0], &ac_entry(ac)[tofree], + memmove(ac->entry, &(ac->entry[tofree]), sizeof(void*)*ac->avail); } } @@ -2792,6 +3261,7 @@ static void drain_array_locked(kmem_cache_t *cachep, static void cache_reap(void *unused) { struct list_head *walk; + struct kmem_list3 *l3; if (down_trylock(&cache_chain_sem)) { /* Give up. Setup the next iteration. */ @@ -2812,27 +3282,32 @@ static void cache_reap(void *unused) check_irq_on(); - spin_lock_irq(&searchp->spinlock); + l3 = searchp->nodelists[numa_node_id()]; + if (l3->alien) + drain_alien_cache(searchp, l3); + spin_lock_irq(&l3->list_lock); - drain_array_locked(searchp, ac_data(searchp), 0); + drain_array_locked(searchp, ac_data(searchp), 0, + numa_node_id()); - if(time_after(searchp->lists.next_reap, jiffies)) + if (time_after(l3->next_reap, jiffies)) goto next_unlock; - searchp->lists.next_reap = jiffies + REAPTIMEOUT_LIST3; + l3->next_reap = jiffies + REAPTIMEOUT_LIST3; - if (searchp->lists.shared) - drain_array_locked(searchp, searchp->lists.shared, 0); + if (l3->shared) + drain_array_locked(searchp, l3->shared, 0, + numa_node_id()); - if (searchp->lists.free_touched) { - searchp->lists.free_touched = 0; + if (l3->free_touched) { + l3->free_touched = 0; goto next_unlock; } - tofree = (searchp->free_limit+5*searchp->num-1)/(5*searchp->num); + tofree = (l3->free_limit+5*searchp->num-1)/(5*searchp->num); do { - p = list3_data(searchp)->slabs_free.next; - if (p == &(list3_data(searchp)->slabs_free)) + p = l3->slabs_free.next; + if (p == &(l3->slabs_free)) break; slabp = list_entry(p, struct slab, list); @@ -2845,13 +3320,13 @@ static void cache_reap(void *unused) * searchp cannot disappear, we hold * cache_chain_lock */ - searchp->lists.free_objects -= searchp->num; - spin_unlock_irq(&searchp->spinlock); + l3->free_objects -= searchp->num; + spin_unlock_irq(&l3->list_lock); slab_destroy(searchp, slabp); - spin_lock_irq(&searchp->spinlock); + spin_lock_irq(&l3->list_lock); } while(--tofree > 0); next_unlock: - spin_unlock_irq(&searchp->spinlock); + spin_unlock_irq(&l3->list_lock); next: cond_resched(); } @@ -2885,7 +3360,7 @@ static void *s_start(struct seq_file *m, loff_t *pos) seq_puts(m, " : slabdata <active_slabs> <num_slabs> <sharedavail>"); #if STATS seq_puts(m, " : globalstat <listallocs> <maxobjs> <grown> <reaped>" - " <error> <maxfreeable> <freelimit> <nodeallocs>"); + " <error> <maxfreeable> <nodeallocs> <remotefrees>"); seq_puts(m, " : cpustat <allochit> <allocmiss> <freehit> <freemiss>"); #endif seq_putc(m, '\n'); @@ -2920,39 +3395,53 @@ static int s_show(struct seq_file *m, void *p) unsigned long active_objs; unsigned long num_objs; unsigned long active_slabs = 0; - unsigned long num_slabs; - const char *name; + unsigned long num_slabs, free_objects = 0, shared_avail = 0; + const char *name; char *error = NULL; + int node; + struct kmem_list3 *l3; check_irq_on(); spin_lock_irq(&cachep->spinlock); active_objs = 0; num_slabs = 0; - list_for_each(q,&cachep->lists.slabs_full) { - slabp = list_entry(q, struct slab, list); - if (slabp->inuse != cachep->num && !error) - error = "slabs_full accounting error"; - active_objs += cachep->num; - active_slabs++; - } - list_for_each(q,&cachep->lists.slabs_partial) { - slabp = list_entry(q, struct slab, list); - if (slabp->inuse == cachep->num && !error) - error = "slabs_partial inuse accounting error"; - if (!slabp->inuse && !error) - error = "slabs_partial/inuse accounting error"; - active_objs += slabp->inuse; - active_slabs++; - } - list_for_each(q,&cachep->lists.slabs_free) { - slabp = list_entry(q, struct slab, list); - if (slabp->inuse && !error) - error = "slabs_free/inuse accounting error"; - num_slabs++; + for_each_online_node(node) { + l3 = cachep->nodelists[node]; + if (!l3) + continue; + + spin_lock(&l3->list_lock); + + list_for_each(q,&l3->slabs_full) { + slabp = list_entry(q, struct slab, list); + if (slabp->inuse != cachep->num && !error) + error = "slabs_full accounting error"; + active_objs += cachep->num; + active_slabs++; + } + list_for_each(q,&l3->slabs_partial) { + slabp = list_entry(q, struct slab, list); + if (slabp->inuse == cachep->num && !error) + error = "slabs_partial inuse accounting error"; + if (!slabp->inuse && !error) + error = "slabs_partial/inuse accounting error"; + active_objs += slabp->inuse; + active_slabs++; + } + list_for_each(q,&l3->slabs_free) { + slabp = list_entry(q, struct slab, list); + if (slabp->inuse && !error) + error = "slabs_free/inuse accounting error"; + num_slabs++; + } + free_objects += l3->free_objects; + shared_avail += l3->shared->avail; + + spin_unlock(&l3->list_lock); } num_slabs+=active_slabs; num_objs = num_slabs*cachep->num; - if (num_objs - active_objs != cachep->lists.free_objects && !error) + if (num_objs - active_objs != free_objects && !error) error = "free_objects accounting error"; name = cachep->name; @@ -2964,9 +3453,9 @@ static int s_show(struct seq_file *m, void *p) cachep->num, (1<<cachep->gfporder)); seq_printf(m, " : tunables %4u %4u %4u", cachep->limit, cachep->batchcount, - cachep->lists.shared->limit/cachep->batchcount); - seq_printf(m, " : slabdata %6lu %6lu %6u", - active_slabs, num_slabs, cachep->lists.shared->avail); + cachep->shared); + seq_printf(m, " : slabdata %6lu %6lu %6lu", + active_slabs, num_slabs, shared_avail); #if STATS { /* list3 stats */ unsigned long high = cachep->high_mark; @@ -2975,12 +3464,13 @@ static int s_show(struct seq_file *m, void *p) unsigned long reaped = cachep->reaped; unsigned long errors = cachep->errors; unsigned long max_freeable = cachep->max_freeable; - unsigned long free_limit = cachep->free_limit; unsigned long node_allocs = cachep->node_allocs; + unsigned long node_frees = cachep->node_frees; - seq_printf(m, " : globalstat %7lu %6lu %5lu %4lu %4lu %4lu %4lu %4lu", - allocs, high, grown, reaped, errors, - max_freeable, free_limit, node_allocs); + seq_printf(m, " : globalstat %7lu %6lu %5lu %4lu \ + %4lu %4lu %4lu %4lu", + allocs, high, grown, reaped, errors, + max_freeable, node_allocs, node_frees); } /* cpu stats */ { @@ -3059,9 +3549,10 @@ ssize_t slabinfo_write(struct file *file, const char __user *buffer, batchcount < 1 || batchcount > limit || shared < 0) { - res = -EINVAL; + res = 0; } else { - res = do_tune_cpucache(cachep, limit, batchcount, shared); + res = do_tune_cpucache(cachep, limit, + batchcount, shared); } break; } @@ -3073,20 +3564,24 @@ ssize_t slabinfo_write(struct file *file, const char __user *buffer, } #endif +/** + * ksize - get the actual amount of memory allocated for a given object + * @objp: Pointer to the object + * + * kmalloc may internally round up allocations and return more memory + * than requested. ksize() can be used to determine the actual amount of + * memory allocated. The caller may use this additional memory, even though + * a smaller amount of memory was initially specified with the kmalloc call. + * The caller must guarantee that objp points to a valid object previously + * allocated with either kmalloc() or kmem_cache_alloc(). The object + * must not be freed during the duration of the call. + */ unsigned int ksize(const void *objp) { - kmem_cache_t *c; - unsigned long flags; - unsigned int size = 0; - - if (likely(objp != NULL)) { - local_irq_save(flags); - c = GET_PAGE_CACHE(virt_to_page(objp)); - size = kmem_cache_size(c); - local_irq_restore(flags); - } + if (unlikely(objp == NULL)) + return 0; - return size; + return obj_reallen(GET_PAGE_CACHE(virt_to_page(objp))); } diff --git a/mm/sparse.c b/mm/sparse.c index b54e304df4a..347249a4917 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -6,6 +6,7 @@ #include <linux/mmzone.h> #include <linux/bootmem.h> #include <linux/module.h> +#include <linux/spinlock.h> #include <asm/dma.h> /* @@ -13,9 +14,64 @@ * * 1) mem_section - memory sections, mem_map's for valid memory */ -struct mem_section mem_section[NR_MEM_SECTIONS]; +#ifdef CONFIG_SPARSEMEM_EXTREME +struct mem_section *mem_section[NR_SECTION_ROOTS] + ____cacheline_maxaligned_in_smp; +#else +struct mem_section mem_section[NR_SECTION_ROOTS][SECTIONS_PER_ROOT] + ____cacheline_maxaligned_in_smp; +#endif EXPORT_SYMBOL(mem_section); +#ifdef CONFIG_SPARSEMEM_EXTREME +static struct mem_section *sparse_index_alloc(int nid) +{ + struct mem_section *section = NULL; + unsigned long array_size = SECTIONS_PER_ROOT * + sizeof(struct mem_section); + + section = alloc_bootmem_node(NODE_DATA(nid), array_size); + + if (section) + memset(section, 0, array_size); + + return section; +} + +static int sparse_index_init(unsigned long section_nr, int nid) +{ + static spinlock_t index_init_lock = SPIN_LOCK_UNLOCKED; + unsigned long root = SECTION_NR_TO_ROOT(section_nr); + struct mem_section *section; + int ret = 0; + + if (mem_section[root]) + return -EEXIST; + + section = sparse_index_alloc(nid); + /* + * This lock keeps two different sections from + * reallocating for the same index + */ + spin_lock(&index_init_lock); + + if (mem_section[root]) { + ret = -EEXIST; + goto out; + } + + mem_section[root] = section; +out: + spin_unlock(&index_init_lock); + return ret; +} +#else /* !SPARSEMEM_EXTREME */ +static inline int sparse_index_init(unsigned long section_nr, int nid) +{ + return 0; +} +#endif + /* Record a memory area against a node. */ void memory_present(int nid, unsigned long start, unsigned long end) { @@ -24,8 +80,13 @@ void memory_present(int nid, unsigned long start, unsigned long end) start &= PAGE_SECTION_MASK; for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION) { unsigned long section = pfn_to_section_nr(pfn); - if (!mem_section[section].section_mem_map) - mem_section[section].section_mem_map = SECTION_MARKED_PRESENT; + struct mem_section *ms; + + sparse_index_init(section, nid); + + ms = __nr_to_section(section); + if (!ms->section_mem_map) + ms->section_mem_map = SECTION_MARKED_PRESENT; } } @@ -85,6 +146,7 @@ static struct page *sparse_early_mem_map_alloc(unsigned long pnum) { struct page *map; int nid = early_pfn_to_nid(section_nr_to_pfn(pnum)); + struct mem_section *ms = __nr_to_section(pnum); map = alloc_remap(nid, sizeof(struct page) * PAGES_PER_SECTION); if (map) @@ -96,7 +158,7 @@ static struct page *sparse_early_mem_map_alloc(unsigned long pnum) return map; printk(KERN_WARNING "%s: allocation failed\n", __FUNCTION__); - mem_section[pnum].section_mem_map = 0; + ms->section_mem_map = 0; return NULL; } @@ -114,8 +176,9 @@ void sparse_init(void) continue; map = sparse_early_mem_map_alloc(pnum); - if (map) - sparse_init_one_section(&mem_section[pnum], pnum, map); + if (!map) + continue; + sparse_init_one_section(__nr_to_section(pnum), pnum, map); } } diff --git a/mm/swap_state.c b/mm/swap_state.c index 4f251775ef9..029e56eb5e7 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -124,6 +124,7 @@ void __delete_from_swap_cache(struct page *page) BUG_ON(!PageLocked(page)); BUG_ON(!PageSwapCache(page)); BUG_ON(PageWriteback(page)); + BUG_ON(PagePrivate(page)); radix_tree_delete(&swapper_space.page_tree, page->private); page->private = 0; @@ -196,11 +197,6 @@ void delete_from_swap_cache(struct page *page) { swp_entry_t entry; - BUG_ON(!PageSwapCache(page)); - BUG_ON(!PageLocked(page)); - BUG_ON(PageWriteback(page)); - BUG_ON(PagePrivate(page)); - entry.val = page->private; write_lock_irq(&swapper_space.tree_lock); diff --git a/mm/swapfile.c b/mm/swapfile.c index 60cd24a5520..4b6e8bf986b 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -31,7 +31,7 @@ #include <asm/tlbflush.h> #include <linux/swapops.h> -DEFINE_SPINLOCK(swaplock); +DEFINE_SPINLOCK(swap_lock); unsigned int nr_swapfiles; long total_swap_pages; static int swap_overflow; @@ -51,13 +51,11 @@ static DECLARE_MUTEX(swapon_sem); /* * We need this because the bdev->unplug_fn can sleep and we cannot - * hold swap_list_lock while calling the unplug_fn. And swap_list_lock + * hold swap_lock while calling the unplug_fn. And swap_lock * cannot be turned into a semaphore. */ static DECLARE_RWSEM(swap_unplug_sem); -#define SWAPFILE_CLUSTER 256 - void swap_unplug_io_fn(struct backing_dev_info *unused_bdi, struct page *page) { swp_entry_t entry; @@ -84,116 +82,135 @@ void swap_unplug_io_fn(struct backing_dev_info *unused_bdi, struct page *page) up_read(&swap_unplug_sem); } -static inline int scan_swap_map(struct swap_info_struct *si) +#define SWAPFILE_CLUSTER 256 +#define LATENCY_LIMIT 256 + +static inline unsigned long scan_swap_map(struct swap_info_struct *si) { - unsigned long offset; + unsigned long offset, last_in_cluster; + int latency_ration = LATENCY_LIMIT; + /* - * We try to cluster swap pages by allocating them - * sequentially in swap. Once we've allocated - * SWAPFILE_CLUSTER pages this way, however, we resort to - * first-free allocation, starting a new cluster. This - * prevents us from scattering swap pages all over the entire - * swap partition, so that we reduce overall disk seek times - * between swap pages. -- sct */ - if (si->cluster_nr) { - while (si->cluster_next <= si->highest_bit) { - offset = si->cluster_next++; + * We try to cluster swap pages by allocating them sequentially + * in swap. Once we've allocated SWAPFILE_CLUSTER pages this + * way, however, we resort to first-free allocation, starting + * a new cluster. This prevents us from scattering swap pages + * all over the entire swap partition, so that we reduce + * overall disk seek times between swap pages. -- sct + * But we do now try to find an empty cluster. -Andrea + */ + + si->flags += SWP_SCANNING; + if (unlikely(!si->cluster_nr)) { + si->cluster_nr = SWAPFILE_CLUSTER - 1; + if (si->pages - si->inuse_pages < SWAPFILE_CLUSTER) + goto lowest; + spin_unlock(&swap_lock); + + offset = si->lowest_bit; + last_in_cluster = offset + SWAPFILE_CLUSTER - 1; + + /* Locate the first empty (unaligned) cluster */ + for (; last_in_cluster <= si->highest_bit; offset++) { if (si->swap_map[offset]) - continue; - si->cluster_nr--; - goto got_page; - } - } - si->cluster_nr = SWAPFILE_CLUSTER; - - /* try to find an empty (even not aligned) cluster. */ - offset = si->lowest_bit; - check_next_cluster: - if (offset+SWAPFILE_CLUSTER-1 <= si->highest_bit) - { - unsigned long nr; - for (nr = offset; nr < offset+SWAPFILE_CLUSTER; nr++) - if (si->swap_map[nr]) - { - offset = nr+1; - goto check_next_cluster; + last_in_cluster = offset + SWAPFILE_CLUSTER; + else if (offset == last_in_cluster) { + spin_lock(&swap_lock); + si->cluster_next = offset-SWAPFILE_CLUSTER-1; + goto cluster; } - /* We found a completly empty cluster, so start - * using it. - */ - goto got_page; + if (unlikely(--latency_ration < 0)) { + cond_resched(); + latency_ration = LATENCY_LIMIT; + } + } + spin_lock(&swap_lock); + goto lowest; } - /* No luck, so now go finegrined as usual. -Andrea */ - for (offset = si->lowest_bit; offset <= si->highest_bit ; offset++) { - if (si->swap_map[offset]) - continue; - si->lowest_bit = offset+1; - got_page: + + si->cluster_nr--; +cluster: + offset = si->cluster_next; + if (offset > si->highest_bit) +lowest: offset = si->lowest_bit; +checks: if (!(si->flags & SWP_WRITEOK)) + goto no_page; + if (!si->highest_bit) + goto no_page; + if (!si->swap_map[offset]) { if (offset == si->lowest_bit) si->lowest_bit++; if (offset == si->highest_bit) si->highest_bit--; - if (si->lowest_bit > si->highest_bit) { + si->inuse_pages++; + if (si->inuse_pages == si->pages) { si->lowest_bit = si->max; si->highest_bit = 0; } si->swap_map[offset] = 1; - si->inuse_pages++; - nr_swap_pages--; - si->cluster_next = offset+1; + si->cluster_next = offset + 1; + si->flags -= SWP_SCANNING; return offset; } - si->lowest_bit = si->max; - si->highest_bit = 0; + + spin_unlock(&swap_lock); + while (++offset <= si->highest_bit) { + if (!si->swap_map[offset]) { + spin_lock(&swap_lock); + goto checks; + } + if (unlikely(--latency_ration < 0)) { + cond_resched(); + latency_ration = LATENCY_LIMIT; + } + } + spin_lock(&swap_lock); + goto lowest; + +no_page: + si->flags -= SWP_SCANNING; return 0; } swp_entry_t get_swap_page(void) { - struct swap_info_struct * p; - unsigned long offset; - swp_entry_t entry; - int type, wrapped = 0; + struct swap_info_struct *si; + pgoff_t offset; + int type, next; + int wrapped = 0; - entry.val = 0; /* Out of memory */ - swap_list_lock(); - type = swap_list.next; - if (type < 0) - goto out; + spin_lock(&swap_lock); if (nr_swap_pages <= 0) - goto out; + goto noswap; + nr_swap_pages--; + + for (type = swap_list.next; type >= 0 && wrapped < 2; type = next) { + si = swap_info + type; + next = si->next; + if (next < 0 || + (!wrapped && si->prio != swap_info[next].prio)) { + next = swap_list.head; + wrapped++; + } - while (1) { - p = &swap_info[type]; - if ((p->flags & SWP_ACTIVE) == SWP_ACTIVE) { - swap_device_lock(p); - offset = scan_swap_map(p); - swap_device_unlock(p); - if (offset) { - entry = swp_entry(type,offset); - type = swap_info[type].next; - if (type < 0 || - p->prio != swap_info[type].prio) { - swap_list.next = swap_list.head; - } else { - swap_list.next = type; - } - goto out; - } + if (!si->highest_bit) + continue; + if (!(si->flags & SWP_WRITEOK)) + continue; + + swap_list.next = next; + offset = scan_swap_map(si); + if (offset) { + spin_unlock(&swap_lock); + return swp_entry(type, offset); } - type = p->next; - if (!wrapped) { - if (type < 0 || p->prio != swap_info[type].prio) { - type = swap_list.head; - wrapped = 1; - } - } else - if (type < 0) - goto out; /* out of swap space */ + next = swap_list.next; } -out: - swap_list_unlock(); - return entry; + + nr_swap_pages++; +noswap: + spin_unlock(&swap_lock); + return (swp_entry_t) {0}; } static struct swap_info_struct * swap_info_get(swp_entry_t entry) @@ -214,10 +231,7 @@ static struct swap_info_struct * swap_info_get(swp_entry_t entry) goto bad_offset; if (!p->swap_map[offset]) goto bad_free; - swap_list_lock(); - if (p->prio > swap_info[swap_list.next].prio) - swap_list.next = type; - swap_device_lock(p); + spin_lock(&swap_lock); return p; bad_free: @@ -235,12 +249,6 @@ out: return NULL; } -static void swap_info_put(struct swap_info_struct * p) -{ - swap_device_unlock(p); - swap_list_unlock(); -} - static int swap_entry_free(struct swap_info_struct *p, unsigned long offset) { int count = p->swap_map[offset]; @@ -253,6 +261,8 @@ static int swap_entry_free(struct swap_info_struct *p, unsigned long offset) p->lowest_bit = offset; if (offset > p->highest_bit) p->highest_bit = offset; + if (p->prio > swap_info[swap_list.next].prio) + swap_list.next = p - swap_info; nr_swap_pages++; p->inuse_pages--; } @@ -271,7 +281,7 @@ void swap_free(swp_entry_t entry) p = swap_info_get(entry); if (p) { swap_entry_free(p, swp_offset(entry)); - swap_info_put(p); + spin_unlock(&swap_lock); } } @@ -289,7 +299,7 @@ static inline int page_swapcount(struct page *page) if (p) { /* Subtract the 1 for the swap cache itself */ count = p->swap_map[swp_offset(entry)] - 1; - swap_info_put(p); + spin_unlock(&swap_lock); } return count; } @@ -346,7 +356,7 @@ int remove_exclusive_swap_page(struct page *page) } write_unlock_irq(&swapper_space.tree_lock); } - swap_info_put(p); + spin_unlock(&swap_lock); if (retval) { swap_free(entry); @@ -369,7 +379,7 @@ void free_swap_and_cache(swp_entry_t entry) if (p) { if (swap_entry_free(p, swp_offset(entry)) == 1) page = find_trylock_page(&swapper_space, entry.val); - swap_info_put(p); + spin_unlock(&swap_lock); } if (page) { int one_user; @@ -531,17 +541,18 @@ static int unuse_mm(struct mm_struct *mm, * Scan swap_map from current position to next entry still in use. * Recycle to start on reaching the end, returning 0 when empty. */ -static int find_next_to_unuse(struct swap_info_struct *si, int prev) +static unsigned int find_next_to_unuse(struct swap_info_struct *si, + unsigned int prev) { - int max = si->max; - int i = prev; + unsigned int max = si->max; + unsigned int i = prev; int count; /* - * No need for swap_device_lock(si) here: we're just looking + * No need for swap_lock here: we're just looking * for whether an entry is in use, not modifying it; false * hits are okay, and sys_swapoff() has already prevented new - * allocations from this area (while holding swap_list_lock()). + * allocations from this area (while holding swap_lock). */ for (;;) { if (++i >= max) { @@ -577,7 +588,7 @@ static int try_to_unuse(unsigned int type) unsigned short swcount; struct page *page; swp_entry_t entry; - int i = 0; + unsigned int i = 0; int retval = 0; int reset_overflow = 0; int shmem; @@ -731,9 +742,9 @@ static int try_to_unuse(unsigned int type) * report them; but do report if we reset SWAP_MAP_MAX. */ if (*swap_map == SWAP_MAP_MAX) { - swap_device_lock(si); + spin_lock(&swap_lock); *swap_map = 1; - swap_device_unlock(si); + spin_unlock(&swap_lock); reset_overflow = 1; } @@ -797,9 +808,9 @@ static int try_to_unuse(unsigned int type) } /* - * After a successful try_to_unuse, if no swap is now in use, we know we - * can empty the mmlist. swap_list_lock must be held on entry and exit. - * Note that mmlist_lock nests inside swap_list_lock, and an mm must be + * After a successful try_to_unuse, if no swap is now in use, we know + * we can empty the mmlist. swap_lock must be held on entry and exit. + * Note that mmlist_lock nests inside swap_lock, and an mm must be * added to the mmlist just after page_duplicate - before would be racy. */ static void drain_mmlist(void) @@ -832,9 +843,9 @@ sector_t map_swap_page(struct swap_info_struct *sis, pgoff_t offset) offset < (se->start_page + se->nr_pages)) { return se->start_block + (offset - se->start_page); } - lh = se->list.prev; + lh = se->list.next; if (lh == &sis->extent_list) - lh = lh->prev; + lh = lh->next; se = list_entry(lh, struct swap_extent, list); sis->curr_swap_extent = se; BUG_ON(se == start_se); /* It *must* be present */ @@ -854,15 +865,13 @@ static void destroy_swap_extents(struct swap_info_struct *sis) list_del(&se->list); kfree(se); } - sis->nr_extents = 0; } /* * Add a block range (and the corresponding page range) into this swapdev's - * extent list. The extent list is kept sorted in block order. + * extent list. The extent list is kept sorted in page order. * - * This function rather assumes that it is called in ascending sector_t order. - * It doesn't look for extent coalescing opportunities. + * This function rather assumes that it is called in ascending page order. */ static int add_swap_extent(struct swap_info_struct *sis, unsigned long start_page, @@ -872,16 +881,15 @@ add_swap_extent(struct swap_info_struct *sis, unsigned long start_page, struct swap_extent *new_se; struct list_head *lh; - lh = sis->extent_list.next; /* The highest-addressed block */ - while (lh != &sis->extent_list) { + lh = sis->extent_list.prev; /* The highest page extent */ + if (lh != &sis->extent_list) { se = list_entry(lh, struct swap_extent, list); - if (se->start_block + se->nr_pages == start_block && - se->start_page + se->nr_pages == start_page) { + BUG_ON(se->start_page + se->nr_pages != start_page); + if (se->start_block + se->nr_pages == start_block) { /* Merge it */ se->nr_pages += nr_pages; return 0; } - lh = lh->next; } /* @@ -894,16 +902,8 @@ add_swap_extent(struct swap_info_struct *sis, unsigned long start_page, new_se->nr_pages = nr_pages; new_se->start_block = start_block; - lh = sis->extent_list.prev; /* The lowest block */ - while (lh != &sis->extent_list) { - se = list_entry(lh, struct swap_extent, list); - if (se->start_block > start_block) - break; - lh = lh->prev; - } - list_add_tail(&new_se->list, lh); - sis->nr_extents++; - return 0; + list_add_tail(&new_se->list, &sis->extent_list); + return 1; } /* @@ -926,7 +926,7 @@ add_swap_extent(struct swap_info_struct *sis, unsigned long start_page, * requirements, they are simply tossed out - we will never use those blocks * for swapping. * - * For S_ISREG swapfiles we hold i_sem across the life of the swapon. This + * For S_ISREG swapfiles we set S_SWAPFILE across the life of the swapon. This * prevents root from shooting her foot off by ftruncating an in-use swapfile, * which will scribble on the fs. * @@ -937,7 +937,7 @@ add_swap_extent(struct swap_info_struct *sis, unsigned long start_page, * This is extremely effective. The average number of iterations in * map_swap_page() has been measured at about 0.3 per page. - akpm. */ -static int setup_swap_extents(struct swap_info_struct *sis) +static int setup_swap_extents(struct swap_info_struct *sis, sector_t *span) { struct inode *inode; unsigned blocks_per_page; @@ -945,11 +945,15 @@ static int setup_swap_extents(struct swap_info_struct *sis) unsigned blkbits; sector_t probe_block; sector_t last_block; + sector_t lowest_block = -1; + sector_t highest_block = 0; + int nr_extents = 0; int ret; inode = sis->swap_file->f_mapping->host; if (S_ISBLK(inode->i_mode)) { ret = add_swap_extent(sis, 0, sis->max, 0); + *span = sis->pages; goto done; } @@ -994,22 +998,32 @@ static int setup_swap_extents(struct swap_info_struct *sis) } } + first_block >>= (PAGE_SHIFT - blkbits); + if (page_no) { /* exclude the header page */ + if (first_block < lowest_block) + lowest_block = first_block; + if (first_block > highest_block) + highest_block = first_block; + } + /* * We found a PAGE_SIZE-length, PAGE_SIZE-aligned run of blocks */ - ret = add_swap_extent(sis, page_no, 1, - first_block >> (PAGE_SHIFT - blkbits)); - if (ret) + ret = add_swap_extent(sis, page_no, 1, first_block); + if (ret < 0) goto out; + nr_extents += ret; page_no++; probe_block += blocks_per_page; reprobe: continue; } - ret = 0; + ret = nr_extents; + *span = 1 + highest_block - lowest_block; if (page_no == 0) - ret = -EINVAL; + page_no = 1; /* force Empty message */ sis->max = page_no; + sis->pages = page_no - 1; sis->highest_bit = page_no - 1; done: sis->curr_swap_extent = list_entry(sis->extent_list.prev, @@ -1069,7 +1083,7 @@ asmlinkage long sys_swapoff(const char __user * specialfile) mapping = victim->f_mapping; prev = -1; - swap_list_lock(); + spin_lock(&swap_lock); for (type = swap_list.head; type >= 0; type = swap_info[type].next) { p = swap_info + type; if ((p->flags & SWP_ACTIVE) == SWP_ACTIVE) { @@ -1080,14 +1094,14 @@ asmlinkage long sys_swapoff(const char __user * specialfile) } if (type < 0) { err = -EINVAL; - swap_list_unlock(); + spin_unlock(&swap_lock); goto out_dput; } if (!security_vm_enough_memory(p->pages)) vm_unacct_memory(p->pages); else { err = -ENOMEM; - swap_list_unlock(); + spin_unlock(&swap_lock); goto out_dput; } if (prev < 0) { @@ -1102,18 +1116,15 @@ asmlinkage long sys_swapoff(const char __user * specialfile) nr_swap_pages -= p->pages; total_swap_pages -= p->pages; p->flags &= ~SWP_WRITEOK; - swap_list_unlock(); + spin_unlock(&swap_lock); + current->flags |= PF_SWAPOFF; err = try_to_unuse(type); current->flags &= ~PF_SWAPOFF; - /* wait for any unplug function to finish */ - down_write(&swap_unplug_sem); - up_write(&swap_unplug_sem); - if (err) { /* re-insert swap space back into swap_list */ - swap_list_lock(); + spin_lock(&swap_lock); for (prev = -1, i = swap_list.head; i >= 0; prev = i, i = swap_info[i].next) if (p->prio >= swap_info[i].prio) break; @@ -1125,22 +1136,35 @@ asmlinkage long sys_swapoff(const char __user * specialfile) nr_swap_pages += p->pages; total_swap_pages += p->pages; p->flags |= SWP_WRITEOK; - swap_list_unlock(); + spin_unlock(&swap_lock); goto out_dput; } + + /* wait for any unplug function to finish */ + down_write(&swap_unplug_sem); + up_write(&swap_unplug_sem); + + destroy_swap_extents(p); down(&swapon_sem); - swap_list_lock(); + spin_lock(&swap_lock); drain_mmlist(); - swap_device_lock(p); + + /* wait for anyone still in scan_swap_map */ + p->highest_bit = 0; /* cuts scans short */ + while (p->flags >= SWP_SCANNING) { + spin_unlock(&swap_lock); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(1); + spin_lock(&swap_lock); + } + swap_file = p->swap_file; p->swap_file = NULL; p->max = 0; swap_map = p->swap_map; p->swap_map = NULL; p->flags = 0; - destroy_swap_extents(p); - swap_device_unlock(p); - swap_list_unlock(); + spin_unlock(&swap_lock); up(&swapon_sem); vfree(swap_map); inode = mapping->host; @@ -1213,7 +1237,7 @@ static int swap_show(struct seq_file *swap, void *v) file = ptr->swap_file; len = seq_path(swap, file->f_vfsmnt, file->f_dentry, " \t\n\\"); - seq_printf(swap, "%*s%s\t%d\t%ld\t%d\n", + seq_printf(swap, "%*s%s\t%u\t%u\t%d\n", len < 40 ? 40 - len : 1, " ", S_ISBLK(file->f_dentry->d_inode->i_mode) ? "partition" : "file\t", @@ -1272,7 +1296,9 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) static int least_priority; union swap_header *swap_header = NULL; int swap_header_version; - int nr_good_pages = 0; + unsigned int nr_good_pages = 0; + int nr_extents = 0; + sector_t span; unsigned long maxpages = 1; int swapfilesize; unsigned short *swap_map; @@ -1282,7 +1308,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) if (!capable(CAP_SYS_ADMIN)) return -EPERM; - swap_list_lock(); + spin_lock(&swap_lock); p = swap_info; for (type = 0 ; type < nr_swapfiles ; type++,p++) if (!(p->flags & SWP_USED)) @@ -1301,14 +1327,13 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) * swp_entry_t or the architecture definition of a swap pte. */ if (type > swp_type(pte_to_swp_entry(swp_entry_to_pte(swp_entry(~0UL,0))))) { - swap_list_unlock(); + spin_unlock(&swap_lock); goto out; } if (type >= nr_swapfiles) nr_swapfiles = type+1; INIT_LIST_HEAD(&p->extent_list); p->flags = SWP_USED; - p->nr_extents = 0; p->swap_file = NULL; p->old_block_size = 0; p->swap_map = NULL; @@ -1316,7 +1341,6 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) p->highest_bit = 0; p->cluster_nr = 0; p->inuse_pages = 0; - spin_lock_init(&p->sdev_lock); p->next = -1; if (swap_flags & SWAP_FLAG_PREFER) { p->prio = @@ -1324,7 +1348,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) } else { p->prio = --least_priority; } - swap_list_unlock(); + spin_unlock(&swap_lock); name = getname(specialfile); error = PTR_ERR(name); if (IS_ERR(name)) { @@ -1426,6 +1450,8 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) } p->lowest_bit = 1; + p->cluster_next = 1; + /* * Find out how many pages are allowed for a single swap * device. There are two limiting factors: 1) the number of @@ -1446,6 +1472,10 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) p->highest_bit = maxpages - 1; error = -EINVAL; + if (!maxpages) + goto bad_swap; + if (swap_header->info.nr_badpages && S_ISREG(inode->i_mode)) + goto bad_swap; if (swap_header->info.nr_badpages > MAX_SWAP_BADPAGES) goto bad_swap; @@ -1470,35 +1500,40 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) if (error) goto bad_swap; } - + if (swapfilesize && maxpages > swapfilesize) { printk(KERN_WARNING "Swap area shorter than signature indicates\n"); error = -EINVAL; goto bad_swap; } + if (nr_good_pages) { + p->swap_map[0] = SWAP_MAP_BAD; + p->max = maxpages; + p->pages = nr_good_pages; + nr_extents = setup_swap_extents(p, &span); + if (nr_extents < 0) { + error = nr_extents; + goto bad_swap; + } + nr_good_pages = p->pages; + } if (!nr_good_pages) { printk(KERN_WARNING "Empty swap-file\n"); error = -EINVAL; goto bad_swap; } - p->swap_map[0] = SWAP_MAP_BAD; - p->max = maxpages; - p->pages = nr_good_pages; - - error = setup_swap_extents(p); - if (error) - goto bad_swap; down(&swapon_sem); - swap_list_lock(); - swap_device_lock(p); + spin_lock(&swap_lock); p->flags = SWP_ACTIVE; nr_swap_pages += nr_good_pages; total_swap_pages += nr_good_pages; - printk(KERN_INFO "Adding %dk swap on %s. Priority:%d extents:%d\n", - nr_good_pages<<(PAGE_SHIFT-10), name, - p->prio, p->nr_extents); + + printk(KERN_INFO "Adding %uk swap on %s. " + "Priority:%d extents:%d across:%lluk\n", + nr_good_pages<<(PAGE_SHIFT-10), name, p->prio, + nr_extents, (unsigned long long)span<<(PAGE_SHIFT-10)); /* insert swap space into swap_list: */ prev = -1; @@ -1514,8 +1549,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) } else { swap_info[prev].next = p - swap_info; } - swap_device_unlock(p); - swap_list_unlock(); + spin_unlock(&swap_lock); up(&swapon_sem); error = 0; goto out; @@ -1524,16 +1558,16 @@ bad_swap: set_blocksize(bdev, p->old_block_size); bd_release(bdev); } + destroy_swap_extents(p); bad_swap_2: - swap_list_lock(); + spin_lock(&swap_lock); swap_map = p->swap_map; p->swap_file = NULL; p->swap_map = NULL; p->flags = 0; if (!(swap_flags & SWAP_FLAG_PREFER)) ++least_priority; - swap_list_unlock(); - destroy_swap_extents(p); + spin_unlock(&swap_lock); vfree(swap_map); if (swap_file) filp_close(swap_file, NULL); @@ -1557,7 +1591,7 @@ void si_swapinfo(struct sysinfo *val) unsigned int i; unsigned long nr_to_be_unused = 0; - swap_list_lock(); + spin_lock(&swap_lock); for (i = 0; i < nr_swapfiles; i++) { if (!(swap_info[i].flags & SWP_USED) || (swap_info[i].flags & SWP_WRITEOK)) @@ -1566,7 +1600,7 @@ void si_swapinfo(struct sysinfo *val) } val->freeswap = nr_swap_pages + nr_to_be_unused; val->totalswap = total_swap_pages + nr_to_be_unused; - swap_list_unlock(); + spin_unlock(&swap_lock); } /* @@ -1587,7 +1621,7 @@ int swap_duplicate(swp_entry_t entry) p = type + swap_info; offset = swp_offset(entry); - swap_device_lock(p); + spin_lock(&swap_lock); if (offset < p->max && p->swap_map[offset]) { if (p->swap_map[offset] < SWAP_MAP_MAX - 1) { p->swap_map[offset]++; @@ -1599,7 +1633,7 @@ int swap_duplicate(swp_entry_t entry) result = 1; } } - swap_device_unlock(p); + spin_unlock(&swap_lock); out: return result; @@ -1615,7 +1649,7 @@ get_swap_info_struct(unsigned type) } /* - * swap_device_lock prevents swap_map being freed. Don't grab an extra + * swap_lock prevents swap_map being freed. Don't grab an extra * reference on the swaphandle, it doesn't matter if it becomes unused. */ int valid_swaphandles(swp_entry_t entry, unsigned long *offset) @@ -1631,7 +1665,7 @@ int valid_swaphandles(swp_entry_t entry, unsigned long *offset) toff++, i--; *offset = toff; - swap_device_lock(swapdev); + spin_lock(&swap_lock); do { /* Don't read-ahead past the end of the swap area */ if (toff >= swapdev->max) @@ -1644,6 +1678,6 @@ int valid_swaphandles(swp_entry_t entry, unsigned long *offset) toff++; ret++; } while (--i); - swap_device_unlock(swapdev); + spin_unlock(&swap_lock); return ret; } diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 8ff16a1eee6..13c3d82968a 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -158,8 +158,6 @@ int map_vm_area(struct vm_struct *area, pgprot_t prot, struct page ***pages) return err; } -#define IOREMAP_MAX_ORDER (7 + PAGE_SHIFT) /* 128 pages */ - struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags, unsigned long start, unsigned long end) { @@ -334,9 +332,10 @@ void __vunmap(void *addr, int deallocate_pages) * @addr: memory base address * * Free the virtually contiguous memory area starting at @addr, as - * obtained from vmalloc(), vmalloc_32() or __vmalloc(). + * obtained from vmalloc(), vmalloc_32() or __vmalloc(). If @addr is + * NULL, no operation is performed. * - * May not be called in interrupt context. + * Must not be called in interrupt context. */ void vfree(void *addr) { @@ -354,7 +353,7 @@ EXPORT_SYMBOL(vfree); * Free the virtually contiguous memory area starting at @addr, * which was created from the page array passed to vmap(). * - * May not be called in interrupt context. + * Must not be called in interrupt context. */ void vunmap(void *addr) { diff --git a/mm/vmscan.c b/mm/vmscan.c index cfffe5098d5..a740778f688 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -822,6 +822,8 @@ shrink_zone(struct zone *zone, struct scan_control *sc) unsigned long nr_active; unsigned long nr_inactive; + atomic_inc(&zone->reclaim_in_progress); + /* * Add one to `nr_to_scan' just to make sure that the kernel will * slowly sift through the active list. @@ -861,6 +863,8 @@ shrink_zone(struct zone *zone, struct scan_control *sc) } throttle_vm_writeout(); + + atomic_dec(&zone->reclaim_in_progress); } /* @@ -890,7 +894,7 @@ shrink_caches(struct zone **zones, struct scan_control *sc) if (zone->present_pages == 0) continue; - if (!cpuset_zone_allowed(zone)) + if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) continue; zone->temp_priority = sc->priority; @@ -900,9 +904,7 @@ shrink_caches(struct zone **zones, struct scan_control *sc) if (zone->all_unreclaimable && sc->priority != DEF_PRIORITY) continue; /* Let kswapd poll it */ - atomic_inc(&zone->reclaim_in_progress); shrink_zone(zone, sc); - atomic_dec(&zone->reclaim_in_progress); } } @@ -938,7 +940,7 @@ int try_to_free_pages(struct zone **zones, unsigned int gfp_mask) for (i = 0; zones[i] != NULL; i++) { struct zone *zone = zones[i]; - if (!cpuset_zone_allowed(zone)) + if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) continue; zone->temp_priority = DEF_PRIORITY; @@ -984,7 +986,7 @@ out: for (i = 0; zones[i] != 0; i++) { struct zone *zone = zones[i]; - if (!cpuset_zone_allowed(zone)) + if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) continue; zone->prev_priority = zone->temp_priority; @@ -1254,7 +1256,7 @@ void wakeup_kswapd(struct zone *zone, int order) return; if (pgdat->kswapd_max_order < order) pgdat->kswapd_max_order = order; - if (!cpuset_zone_allowed(zone)) + if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) return; if (!waitqueue_active(&zone->zone_pgdat->kswapd_wait)) return; @@ -1358,14 +1360,13 @@ int zone_reclaim(struct zone *zone, unsigned int gfp_mask, unsigned int order) sc.swap_cluster_max = SWAP_CLUSTER_MAX; /* Don't reclaim the zone if there are other reclaimers active */ - if (!atomic_inc_and_test(&zone->reclaim_in_progress)) + if (atomic_read(&zone->reclaim_in_progress) > 0) goto out; shrink_zone(zone, &sc); total_reclaimed = sc.nr_reclaimed; out: - atomic_dec(&zone->reclaim_in_progress); return total_reclaimed; } @@ -1375,6 +1376,9 @@ asmlinkage long sys_set_zone_reclaim(unsigned int node, unsigned int zone, struct zone *z; int i; + if (!capable(CAP_SYS_ADMIN)) + return -EACCES; + if (node >= MAX_NUMNODES || !node_online(node)) return -EINVAL; |